Date: 2019-12-25 20:17:15 CET, cola version: 1.3.2
Document is loading...
All available functions which can be applied to this res_list object:
res_list
#> A 'ConsensusPartitionList' object with 24 methods.
#> On a matrix with 21512 rows and 100 columns.
#> Top rows are extracted by 'SD, CV, MAD, ATC' methods.
#> Subgroups are detected by 'hclust, kmeans, skmeans, pam, mclust, NMF' method.
#> Number of partitions are tried for k = 2, 3, 4, 5, 6.
#> Performed in total 30000 partitions by row resampling.
#>
#> Following methods can be applied to this 'ConsensusPartitionList' object:
#> [1] "cola_report" "collect_classes" "collect_plots" "collect_stats"
#> [5] "colnames" "functional_enrichment" "get_anno_col" "get_anno"
#> [9] "get_classes" "get_matrix" "get_membership" "get_stats"
#> [13] "is_best_k" "is_stable_k" "ncol" "nrow"
#> [17] "rownames" "show" "suggest_best_k" "test_to_known_factors"
#> [21] "top_rows_heatmap" "top_rows_overlap"
#>
#> You can get result for a single method by, e.g. object["SD", "hclust"] or object["SD:hclust"]
#> or a subset of methods by object[c("SD", "CV")], c("hclust", "kmeans")]
The call of run_all_consensus_partition_methods() was:
#> run_all_consensus_partition_methods(data = mat, mc.cores = 4, anno = anno)
Dimension of the input matrix:
mat = get_matrix(res_list)
dim(mat)
#> [1] 21512 100
The density distribution for each sample is visualized as in one column in the following heatmap. The clustering is based on the distance which is the Kolmogorov-Smirnov statistic between two distributions.
library(ComplexHeatmap)
densityHeatmap(mat, top_annotation = HeatmapAnnotation(df = get_anno(res_list),
col = get_anno_col(res_list)), ylab = "value", cluster_columns = TRUE, show_column_names = FALSE,
mc.cores = 4)
Folowing table shows the best k (number of partitions) for each combination
of top-value methods and partition methods. Clicking on the method name in
the table goes to the section for a single combination of methods.
The cola vignette explains the definition of the metrics used for determining the best number of partitions.
suggest_best_k(res_list)
| The best k | 1-PAC | Mean silhouette | Concordance | ||
|---|---|---|---|---|---|
| MAD:kmeans | 2 | 1.000 | 0.977 | 0.988 | ** |
| CV:skmeans | 2 | 0.999 | 0.965 | 0.985 | ** |
| MAD:skmeans | 2 | 0.979 | 0.953 | 0.982 | ** |
| SD:skmeans | 2 | 0.958 | 0.944 | 0.977 | ** |
| SD:kmeans | 2 | 0.938 | 0.947 | 0.973 | * |
| ATC:mclust | 4 | 0.936 | 0.937 | 0.974 | * |
| ATC:skmeans | 3 | 0.936 | 0.945 | 0.976 | * |
| CV:NMF | 2 | 0.935 | 0.944 | 0.975 | * |
| ATC:NMF | 2 | 0.918 | 0.918 | 0.968 | * |
| SD:NMF | 2 | 0.917 | 0.934 | 0.973 | * |
| MAD:NMF | 2 | 0.917 | 0.944 | 0.975 | * |
| MAD:mclust | 4 | 0.908 | 0.905 | 0.945 | * |
| CV:mclust | 3 | 0.865 | 0.901 | 0.950 | |
| SD:mclust | 4 | 0.815 | 0.874 | 0.914 | |
| MAD:pam | 3 | 0.727 | 0.839 | 0.924 | |
| CV:kmeans | 2 | 0.704 | 0.885 | 0.937 | |
| ATC:kmeans | 2 | 0.625 | 0.870 | 0.928 | |
| ATC:pam | 2 | 0.624 | 0.902 | 0.944 | |
| SD:pam | 3 | 0.608 | 0.787 | 0.879 | |
| CV:pam | 3 | 0.494 | 0.720 | 0.855 | |
| CV:hclust | 2 | 0.426 | 0.786 | 0.886 | |
| ATC:hclust | 2 | 0.344 | 0.688 | 0.802 | |
| SD:hclust | 2 | 0.239 | 0.675 | 0.821 | |
| MAD:hclust | 2 | 0.219 | 0.661 | 0.817 |
**: 1-PAC > 0.95, *: 1-PAC > 0.9
Cumulative distribution function curves of consensus matrix for all methods.
collect_plots(res_list, fun = plot_ecdf)
Consensus heatmaps for all methods. (What is a consensus heatmap?)
collect_plots(res_list, k = 2, fun = consensus_heatmap, mc.cores = 4)
collect_plots(res_list, k = 3, fun = consensus_heatmap, mc.cores = 4)
collect_plots(res_list, k = 4, fun = consensus_heatmap, mc.cores = 4)
collect_plots(res_list, k = 5, fun = consensus_heatmap, mc.cores = 4)
collect_plots(res_list, k = 6, fun = consensus_heatmap, mc.cores = 4)
Membership heatmaps for all methods. (What is a membership heatmap?)
collect_plots(res_list, k = 2, fun = membership_heatmap, mc.cores = 4)
collect_plots(res_list, k = 3, fun = membership_heatmap, mc.cores = 4)
collect_plots(res_list, k = 4, fun = membership_heatmap, mc.cores = 4)
collect_plots(res_list, k = 5, fun = membership_heatmap, mc.cores = 4)
collect_plots(res_list, k = 6, fun = membership_heatmap, mc.cores = 4)
Signature heatmaps for all methods. (What is a signature heatmap?)
Note in following heatmaps, rows are scaled.
collect_plots(res_list, k = 2, fun = get_signatures, mc.cores = 4)
collect_plots(res_list, k = 3, fun = get_signatures, mc.cores = 4)
collect_plots(res_list, k = 4, fun = get_signatures, mc.cores = 4)
collect_plots(res_list, k = 5, fun = get_signatures, mc.cores = 4)
collect_plots(res_list, k = 6, fun = get_signatures, mc.cores = 4)
The statistics used for measuring the stability of consensus partitioning. (How are they defined?)
get_stats(res_list, k = 2)
#> k 1-PAC mean_silhouette concordance area_increased Rand Jaccard
#> SD:NMF 2 0.917 0.934 0.973 0.499 0.500 0.500
#> CV:NMF 2 0.935 0.944 0.975 0.496 0.508 0.508
#> MAD:NMF 2 0.917 0.944 0.975 0.498 0.502 0.502
#> ATC:NMF 2 0.918 0.918 0.968 0.503 0.497 0.497
#> SD:skmeans 2 0.958 0.944 0.977 0.500 0.500 0.500
#> CV:skmeans 2 0.999 0.965 0.985 0.500 0.500 0.500
#> MAD:skmeans 2 0.979 0.953 0.982 0.499 0.500 0.500
#> ATC:skmeans 2 0.696 0.948 0.973 0.505 0.495 0.495
#> SD:mclust 2 0.500 0.665 0.861 0.331 0.677 0.677
#> CV:mclust 2 0.395 0.736 0.768 0.357 0.495 0.495
#> MAD:mclust 2 0.538 0.906 0.905 0.440 0.496 0.496
#> ATC:mclust 2 0.599 0.822 0.911 0.282 0.802 0.802
#> SD:kmeans 2 0.938 0.947 0.973 0.493 0.508 0.508
#> CV:kmeans 2 0.704 0.885 0.937 0.492 0.505 0.505
#> MAD:kmeans 2 1.000 0.977 0.988 0.492 0.508 0.508
#> ATC:kmeans 2 0.625 0.870 0.928 0.500 0.495 0.495
#> SD:pam 2 0.646 0.818 0.903 0.445 0.535 0.535
#> CV:pam 2 0.344 0.538 0.787 0.463 0.576 0.576
#> MAD:pam 2 0.597 0.733 0.899 0.454 0.529 0.529
#> ATC:pam 2 0.624 0.902 0.944 0.465 0.547 0.547
#> SD:hclust 2 0.239 0.675 0.821 0.438 0.529 0.529
#> CV:hclust 2 0.426 0.786 0.886 0.418 0.547 0.547
#> MAD:hclust 2 0.219 0.661 0.817 0.423 0.540 0.540
#> ATC:hclust 2 0.344 0.688 0.802 0.443 0.553 0.553
get_stats(res_list, k = 3)
#> k 1-PAC mean_silhouette concordance area_increased Rand Jaccard
#> SD:NMF 3 0.432 0.484 0.662 0.307 0.779 0.589
#> CV:NMF 3 0.470 0.630 0.806 0.318 0.784 0.595
#> MAD:NMF 3 0.449 0.504 0.714 0.313 0.727 0.510
#> ATC:NMF 3 0.618 0.710 0.875 0.326 0.661 0.418
#> SD:skmeans 3 0.759 0.855 0.930 0.340 0.728 0.505
#> CV:skmeans 3 0.768 0.834 0.924 0.341 0.731 0.510
#> MAD:skmeans 3 0.820 0.849 0.931 0.341 0.753 0.541
#> ATC:skmeans 3 0.936 0.945 0.976 0.326 0.706 0.475
#> SD:mclust 3 0.520 0.738 0.845 0.845 0.665 0.521
#> CV:mclust 3 0.865 0.901 0.950 0.652 0.795 0.628
#> MAD:mclust 3 0.669 0.846 0.890 0.390 0.894 0.785
#> ATC:mclust 3 0.581 0.885 0.917 1.012 0.602 0.510
#> SD:kmeans 3 0.603 0.811 0.876 0.334 0.704 0.481
#> CV:kmeans 3 0.656 0.834 0.889 0.330 0.697 0.472
#> MAD:kmeans 3 0.545 0.755 0.846 0.336 0.725 0.509
#> ATC:kmeans 3 0.732 0.856 0.929 0.326 0.699 0.466
#> SD:pam 3 0.608 0.787 0.879 0.469 0.687 0.473
#> CV:pam 3 0.494 0.720 0.855 0.401 0.634 0.431
#> MAD:pam 3 0.727 0.839 0.924 0.436 0.666 0.447
#> ATC:pam 3 0.864 0.900 0.949 0.412 0.772 0.593
#> SD:hclust 3 0.207 0.384 0.672 0.366 0.791 0.655
#> CV:hclust 3 0.297 0.657 0.801 0.368 0.890 0.804
#> MAD:hclust 3 0.248 0.351 0.653 0.409 0.835 0.738
#> ATC:hclust 3 0.369 0.598 0.723 0.269 0.906 0.834
get_stats(res_list, k = 4)
#> k 1-PAC mean_silhouette concordance area_increased Rand Jaccard
#> SD:NMF 4 0.602 0.586 0.754 0.1313 0.670 0.311
#> CV:NMF 4 0.531 0.507 0.763 0.1344 0.766 0.446
#> MAD:NMF 4 0.564 0.536 0.742 0.1290 0.671 0.300
#> ATC:NMF 4 0.654 0.703 0.842 0.1026 0.905 0.727
#> SD:skmeans 4 0.760 0.810 0.894 0.1227 0.845 0.576
#> CV:skmeans 4 0.564 0.587 0.791 0.1197 0.828 0.543
#> MAD:skmeans 4 0.719 0.793 0.887 0.1215 0.829 0.544
#> ATC:skmeans 4 0.720 0.699 0.843 0.0956 0.908 0.735
#> SD:mclust 4 0.815 0.874 0.914 0.2114 0.737 0.415
#> CV:mclust 4 0.529 0.282 0.691 0.1710 0.851 0.675
#> MAD:mclust 4 0.908 0.905 0.945 0.2215 0.754 0.442
#> ATC:mclust 4 0.936 0.937 0.974 0.2631 0.806 0.570
#> SD:kmeans 4 0.708 0.777 0.861 0.1295 0.823 0.532
#> CV:kmeans 4 0.604 0.577 0.782 0.1225 0.864 0.632
#> MAD:kmeans 4 0.733 0.778 0.865 0.1317 0.836 0.560
#> ATC:kmeans 4 0.534 0.536 0.741 0.0990 0.897 0.709
#> SD:pam 4 0.626 0.640 0.793 0.0930 0.908 0.748
#> CV:pam 4 0.579 0.747 0.830 0.0954 0.901 0.737
#> MAD:pam 4 0.662 0.741 0.839 0.1101 0.892 0.704
#> ATC:pam 4 0.807 0.822 0.921 0.0943 0.754 0.439
#> SD:hclust 4 0.302 0.492 0.689 0.1453 0.807 0.611
#> CV:hclust 4 0.345 0.588 0.747 0.1706 0.886 0.759
#> MAD:hclust 4 0.337 0.527 0.696 0.1696 0.761 0.568
#> ATC:hclust 4 0.530 0.668 0.803 0.2476 0.787 0.572
get_stats(res_list, k = 5)
#> k 1-PAC mean_silhouette concordance area_increased Rand Jaccard
#> SD:NMF 5 0.550 0.434 0.694 0.0780 0.819 0.445
#> CV:NMF 5 0.538 0.464 0.701 0.0765 0.826 0.450
#> MAD:NMF 5 0.538 0.399 0.675 0.0745 0.836 0.484
#> ATC:NMF 5 0.639 0.592 0.793 0.0831 0.780 0.362
#> SD:skmeans 5 0.635 0.574 0.743 0.0609 0.964 0.860
#> CV:skmeans 5 0.552 0.381 0.648 0.0657 0.896 0.642
#> MAD:skmeans 5 0.616 0.547 0.719 0.0615 0.976 0.904
#> ATC:skmeans 5 0.772 0.804 0.888 0.0726 0.875 0.591
#> SD:mclust 5 0.617 0.761 0.853 0.0288 0.949 0.802
#> CV:mclust 5 0.579 0.499 0.711 0.0934 0.698 0.318
#> MAD:mclust 5 0.692 0.678 0.849 0.0152 0.864 0.550
#> ATC:mclust 5 0.785 0.763 0.872 0.0631 0.964 0.877
#> SD:kmeans 5 0.685 0.658 0.770 0.0579 0.971 0.883
#> CV:kmeans 5 0.609 0.528 0.716 0.0684 0.896 0.644
#> MAD:kmeans 5 0.696 0.665 0.791 0.0609 0.960 0.843
#> ATC:kmeans 5 0.611 0.555 0.767 0.0692 0.778 0.377
#> SD:pam 5 0.602 0.589 0.753 0.0738 0.902 0.688
#> CV:pam 5 0.565 0.476 0.739 0.0776 0.943 0.820
#> MAD:pam 5 0.629 0.684 0.767 0.0615 0.959 0.855
#> ATC:pam 5 0.761 0.562 0.772 0.0750 0.891 0.643
#> SD:hclust 5 0.387 0.540 0.690 0.0763 0.911 0.734
#> CV:hclust 5 0.406 0.453 0.685 0.0911 0.921 0.786
#> MAD:hclust 5 0.434 0.568 0.697 0.0667 0.890 0.672
#> ATC:hclust 5 0.593 0.682 0.815 0.0633 0.946 0.819
get_stats(res_list, k = 6)
#> k 1-PAC mean_silhouette concordance area_increased Rand Jaccard
#> SD:NMF 6 0.643 0.536 0.729 0.0458 0.864 0.452
#> CV:NMF 6 0.618 0.481 0.690 0.0446 0.892 0.535
#> MAD:NMF 6 0.595 0.458 0.646 0.0467 0.844 0.401
#> ATC:NMF 6 0.637 0.474 0.694 0.0402 0.908 0.610
#> SD:skmeans 6 0.627 0.430 0.611 0.0400 0.918 0.664
#> CV:skmeans 6 0.589 0.398 0.638 0.0407 0.867 0.498
#> MAD:skmeans 6 0.614 0.429 0.637 0.0393 0.951 0.789
#> ATC:skmeans 6 0.774 0.640 0.794 0.0462 0.952 0.783
#> SD:mclust 6 0.721 0.753 0.862 0.0532 0.914 0.646
#> CV:mclust 6 0.679 0.661 0.809 0.0703 0.845 0.426
#> MAD:mclust 6 0.745 0.752 0.840 0.0664 0.925 0.690
#> ATC:mclust 6 0.887 0.833 0.924 0.0671 0.871 0.553
#> SD:kmeans 6 0.702 0.547 0.674 0.0408 0.922 0.674
#> CV:kmeans 6 0.667 0.430 0.693 0.0440 0.889 0.563
#> MAD:kmeans 6 0.699 0.462 0.639 0.0390 0.941 0.755
#> ATC:kmeans 6 0.773 0.830 0.868 0.0557 0.872 0.511
#> SD:pam 6 0.628 0.541 0.736 0.0581 0.866 0.509
#> CV:pam 6 0.594 0.404 0.661 0.0511 0.884 0.613
#> MAD:pam 6 0.684 0.515 0.720 0.0584 0.919 0.683
#> ATC:pam 6 0.890 0.871 0.933 0.0531 0.899 0.599
#> SD:hclust 6 0.523 0.514 0.701 0.0610 0.964 0.863
#> CV:hclust 6 0.473 0.528 0.698 0.0463 0.929 0.767
#> MAD:hclust 6 0.576 0.538 0.702 0.0665 0.977 0.909
#> ATC:hclust 6 0.620 0.517 0.729 0.0606 0.973 0.890
Following heatmap plots the partition for each combination of methods and the lightness correspond to the silhouette scores for samples in each method. On top the consensus subgroup is inferred from all methods by taking the mean silhouette scores as weight.
collect_stats(res_list, k = 2)
collect_stats(res_list, k = 3)
collect_stats(res_list, k = 4)
collect_stats(res_list, k = 5)
collect_stats(res_list, k = 6)
Collect partitions from all methods:
collect_classes(res_list, k = 2)
collect_classes(res_list, k = 3)
collect_classes(res_list, k = 4)
collect_classes(res_list, k = 5)
collect_classes(res_list, k = 6)
Overlap of top rows from different top-row methods:
top_rows_overlap(res_list, top_n = 1000, method = "euler")
top_rows_overlap(res_list, top_n = 2000, method = "euler")
top_rows_overlap(res_list, top_n = 3000, method = "euler")
top_rows_overlap(res_list, top_n = 4000, method = "euler")
top_rows_overlap(res_list, top_n = 5000, method = "euler")
Also visualize the correspondance of rankings between different top-row methods:
top_rows_overlap(res_list, top_n = 1000, method = "correspondance")
top_rows_overlap(res_list, top_n = 2000, method = "correspondance")
top_rows_overlap(res_list, top_n = 3000, method = "correspondance")
top_rows_overlap(res_list, top_n = 4000, method = "correspondance")
top_rows_overlap(res_list, top_n = 5000, method = "correspondance")
Heatmaps of the top rows:
top_rows_heatmap(res_list, top_n = 1000)
top_rows_heatmap(res_list, top_n = 2000)
top_rows_heatmap(res_list, top_n = 3000)
top_rows_heatmap(res_list, top_n = 4000)
top_rows_heatmap(res_list, top_n = 5000)
Test correlation between subgroups and known annotations. If the known annotation is numeric, one-way ANOVA test is applied, and if the known annotation is discrete, chi-squared contingency table test is applied.
test_to_known_factors(res_list, k = 2)
#> n disease.state(p) specimen(p) cell.type(p) other(p) k
#> SD:NMF 97 7.18e-05 0.7381 3.31e-13 0.1495 2
#> CV:NMF 98 8.95e-05 0.6496 4.19e-14 0.1842 2
#> MAD:NMF 98 1.37e-04 0.5025 3.14e-14 0.0947 2
#> ATC:NMF 96 2.74e-02 0.0378 1.75e-03 0.2217 2
#> SD:skmeans 97 5.64e-04 0.4151 3.09e-13 0.1020 2
#> CV:skmeans 99 9.78e-04 0.4095 3.76e-12 0.1858 2
#> MAD:skmeans 97 5.64e-04 0.4151 3.09e-13 0.1020 2
#> ATC:skmeans 99 5.34e-02 0.0201 3.41e-02 0.0991 2
#> SD:mclust 68 2.02e-03 0.1868 1.59e-05 0.2470 2
#> CV:mclust 84 2.21e-02 0.1703 2.41e-10 0.4651 2
#> MAD:mclust 99 2.75e-02 0.4512 1.16e-11 0.4408 2
#> ATC:mclust 99 5.43e-02 0.5156 4.66e-04 0.5101 2
#> SD:kmeans 99 3.34e-04 0.3698 3.49e-13 0.1112 2
#> CV:kmeans 98 2.21e-04 0.4030 2.16e-13 0.1459 2
#> MAD:kmeans 100 2.77e-04 0.2984 2.66e-13 0.0975 2
#> ATC:kmeans 99 5.34e-02 0.0201 3.41e-02 0.0991 2
#> SD:pam 94 5.42e-06 0.7345 6.12e-16 0.0454 2
#> CV:pam 66 1.31e-03 0.9926 4.68e-11 0.1244 2
#> MAD:pam 83 1.94e-04 0.6970 4.28e-13 0.1590 2
#> ATC:pam 99 6.38e-05 0.8415 2.96e-13 0.0823 2
#> SD:hclust 83 4.74e-07 1.0000 4.45e-16 0.0554 2
#> CV:hclust 90 6.48e-08 1.0000 1.15e-16 0.0559 2
#> MAD:hclust 84 1.73e-06 1.0000 2.80e-16 0.0431 2
#> ATC:hclust 90 1.98e-02 0.0110 3.85e-01 0.1214 2
test_to_known_factors(res_list, k = 3)
#> n disease.state(p) specimen(p) cell.type(p) other(p) k
#> SD:NMF 52 1.69e-02 0.4368 4.61e-08 0.6141 3
#> CV:NMF 79 7.34e-05 0.1071 9.83e-13 0.4021 3
#> MAD:NMF 61 9.50e-06 0.0726 1.84e-10 0.0259 3
#> ATC:NMF 82 1.44e-06 0.0332 3.74e-11 0.0603 3
#> SD:skmeans 94 3.88e-04 0.4719 3.48e-13 0.4671 3
#> CV:skmeans 91 2.44e-05 0.4304 4.93e-14 0.3416 3
#> MAD:skmeans 92 1.86e-04 0.4102 1.35e-13 0.3788 3
#> ATC:skmeans 99 9.07e-07 0.1940 1.22e-12 0.0406 3
#> SD:mclust 93 1.62e-03 0.0534 3.99e-10 0.1784 3
#> CV:mclust 97 2.50e-04 0.0331 1.46e-11 0.0353 3
#> MAD:mclust 95 1.29e-03 0.0433 2.87e-10 0.1608 3
#> ATC:mclust 98 3.01e-06 0.4748 2.61e-17 0.1297 3
#> SD:kmeans 94 1.44e-05 0.3183 2.74e-13 0.1502 3
#> CV:kmeans 96 1.37e-05 0.1746 1.05e-13 0.1390 3
#> MAD:kmeans 93 1.41e-04 0.2491 1.94e-12 0.2283 3
#> ATC:kmeans 97 9.51e-07 0.1400 2.13e-12 0.0897 3
#> SD:pam 91 2.19e-04 0.3314 4.32e-12 0.0923 3
#> CV:pam 85 6.27e-04 0.1882 4.51e-08 0.0449 3
#> MAD:pam 94 6.47e-04 0.2784 1.32e-10 0.0607 3
#> ATC:pam 96 3.29e-03 0.4576 1.35e-09 0.1412 3
#> SD:hclust 37 3.09e-02 0.7366 3.51e-11 0.7217 3
#> CV:hclust 89 2.43e-06 0.9556 3.86e-15 0.2565 3
#> MAD:hclust 24 1.75e-01 1.0000 6.14e-06 0.7279 3
#> ATC:hclust 81 3.94e-02 0.0731 2.47e-02 0.1825 3
test_to_known_factors(res_list, k = 4)
#> n disease.state(p) specimen(p) cell.type(p) other(p) k
#> SD:NMF 78 1.01e-03 0.9927 5.54e-12 0.6043 4
#> CV:NMF 62 2.76e-03 0.2527 1.41e-11 0.1797 4
#> MAD:NMF 69 1.83e-05 0.5813 3.99e-10 0.3021 4
#> ATC:NMF 87 6.01e-06 0.2059 2.36e-14 0.1471 4
#> SD:skmeans 93 6.25e-04 0.4239 3.06e-15 0.1922 4
#> CV:skmeans 72 1.31e-05 0.0481 5.42e-14 0.0260 4
#> MAD:skmeans 93 6.25e-04 0.4239 3.06e-15 0.1922 4
#> ATC:skmeans 85 1.78e-06 0.5372 5.62e-16 0.1843 4
#> SD:mclust 98 2.89e-04 0.6454 1.37e-15 0.2386 4
#> CV:mclust 46 1.52e-04 0.2732 8.96e-08 0.1054 4
#> MAD:mclust 97 1.01e-04 0.6347 2.67e-16 0.2753 4
#> ATC:mclust 98 2.32e-06 0.1255 1.64e-15 0.0431 4
#> SD:kmeans 94 6.30e-04 0.3085 2.01e-14 0.2307 4
#> CV:kmeans 70 8.41e-05 0.1582 4.56e-12 0.1558 4
#> MAD:kmeans 93 2.66e-04 0.3308 7.45e-16 0.1535 4
#> ATC:kmeans 71 1.03e-03 0.3798 2.62e-09 0.2825 4
#> SD:pam 76 1.62e-04 0.6021 5.19e-08 0.5594 4
#> CV:pam 89 2.31e-03 0.2634 1.49e-06 0.3288 4
#> MAD:pam 85 3.11e-04 0.5061 1.05e-08 0.0238 4
#> ATC:pam 92 1.73e-05 0.1991 2.61e-15 0.0358 4
#> SD:hclust 59 1.90e-03 0.7800 6.17e-10 0.8124 4
#> CV:hclust 77 5.86e-05 0.4603 7.45e-14 0.3050 4
#> MAD:hclust 62 1.61e-04 0.4723 1.18e-10 0.2285 4
#> ATC:hclust 86 5.66e-03 0.0555 1.79e-06 0.4598 4
test_to_known_factors(res_list, k = 5)
#> n disease.state(p) specimen(p) cell.type(p) other(p) k
#> SD:NMF 44 2.78e-03 0.476 8.31e-09 0.4096 5
#> CV:NMF 48 2.44e-04 0.228 6.96e-10 0.3840 5
#> MAD:NMF 38 2.02e-01 0.693 1.89e-08 0.3341 5
#> ATC:NMF 75 6.08e-03 0.966 1.16e-10 0.3964 5
#> SD:skmeans 77 8.93e-06 0.200 2.55e-14 0.0219 5
#> CV:skmeans 40 2.93e-03 0.228 5.02e-06 0.2191 5
#> MAD:skmeans 70 2.27e-05 0.422 1.45e-12 0.1932 5
#> ATC:skmeans 95 2.61e-04 0.558 2.02e-15 0.3617 5
#> SD:mclust 90 3.47e-04 0.650 9.82e-13 0.3464 5
#> CV:mclust 59 8.03e-05 0.074 4.15e-09 0.2242 5
#> MAD:mclust 79 1.15e-04 0.231 5.81e-12 0.1496 5
#> ATC:mclust 89 8.31e-07 0.120 1.82e-17 0.1541 5
#> SD:kmeans 84 7.86e-05 0.443 2.56e-14 0.0673 5
#> CV:kmeans 66 1.02e-04 0.191 1.61e-11 0.2423 5
#> MAD:kmeans 82 9.70e-05 0.187 1.56e-15 0.0497 5
#> ATC:kmeans 69 7.22e-04 0.258 9.31e-12 0.2071 5
#> SD:pam 76 2.44e-02 0.570 8.38e-10 0.1164 5
#> CV:pam 53 5.00e-03 0.413 1.39e-03 0.6961 5
#> MAD:pam 89 4.27e-03 0.663 1.34e-08 0.1969 5
#> ATC:pam 75 1.08e-04 0.334 6.07e-12 0.2391 5
#> SD:hclust 71 4.06e-04 0.815 5.03e-14 0.3725 5
#> CV:hclust 41 4.10e-04 0.341 1.08e-07 0.3823 5
#> MAD:hclust 73 7.71e-05 0.701 1.95e-15 0.0744 5
#> ATC:hclust 85 2.05e-03 0.103 3.40e-08 0.3582 5
test_to_known_factors(res_list, k = 6)
#> n disease.state(p) specimen(p) cell.type(p) other(p) k
#> SD:NMF 63 5.59e-02 0.8694 2.11e-11 0.1213 6
#> CV:NMF 53 7.07e-02 0.7203 8.65e-10 0.1080 6
#> MAD:NMF 51 7.90e-02 0.8392 1.49e-11 0.3598 6
#> ATC:NMF 57 2.17e-03 0.9315 1.43e-10 0.8218 6
#> SD:skmeans 56 5.12e-05 0.7821 2.75e-09 0.2266 6
#> CV:skmeans 45 7.18e-03 0.3311 6.31e-08 0.0441 6
#> MAD:skmeans 54 2.48e-05 0.1813 7.58e-09 0.1809 6
#> ATC:skmeans 72 9.22e-04 0.6769 3.37e-13 0.6042 6
#> SD:mclust 90 1.04e-04 0.3602 4.01e-11 0.2976 6
#> CV:mclust 84 2.66e-04 0.3095 4.48e-10 0.4903 6
#> MAD:mclust 93 3.34e-05 0.2292 3.99e-11 0.2657 6
#> ATC:mclust 92 4.34e-06 0.0833 9.96e-15 0.1335 6
#> SD:kmeans 66 6.63e-02 0.3465 1.49e-10 0.1512 6
#> CV:kmeans 48 2.87e-03 0.0930 5.36e-09 0.1183 6
#> MAD:kmeans 61 2.76e-03 0.3095 5.40e-10 0.2320 6
#> ATC:kmeans 100 4.95e-05 0.6295 3.66e-13 0.3836 6
#> SD:pam 74 8.01e-02 0.7609 1.48e-07 0.2209 6
#> CV:pam 42 7.25e-03 0.7464 6.80e-04 0.4629 6
#> MAD:pam 54 2.94e-02 0.5759 6.07e-05 0.0453 6
#> ATC:pam 96 2.10e-06 0.2116 4.71e-15 0.0833 6
#> SD:hclust 60 3.27e-03 0.7202 3.94e-15 0.3439 6
#> CV:hclust 69 2.07e-04 0.5955 1.65e-16 0.1259 6
#> MAD:hclust 68 1.63e-03 0.6445 4.50e-15 0.0769 6
#> ATC:hclust 51 1.69e-02 0.1213 1.94e-06 0.3816 6
The object with results only for a single top-value method and a single partition method can be extracted as:
res = res_list["SD", "hclust"]
# you can also extract it by
# res = res_list["SD:hclust"]
A summary of res and all the functions that can be applied to it:
res
#> A 'ConsensusPartition' object with k = 2, 3, 4, 5, 6.
#> On a matrix with 21512 rows and 100 columns.
#> Top rows (1000, 2000, 3000, 4000, 5000) are extracted by 'SD' method.
#> Subgroups are detected by 'hclust' method.
#> Performed in total 1250 partitions by row resampling.
#> Best k for subgroups seems to be 2.
#>
#> Following methods can be applied to this 'ConsensusPartition' object:
#> [1] "cola_report" "collect_classes" "collect_plots"
#> [4] "collect_stats" "colnames" "compare_signatures"
#> [7] "consensus_heatmap" "dimension_reduction" "functional_enrichment"
#> [10] "get_anno_col" "get_anno" "get_classes"
#> [13] "get_consensus" "get_matrix" "get_membership"
#> [16] "get_param" "get_signatures" "get_stats"
#> [19] "is_best_k" "is_stable_k" "membership_heatmap"
#> [22] "ncol" "nrow" "plot_ecdf"
#> [25] "rownames" "select_partition_number" "show"
#> [28] "suggest_best_k" "test_to_known_factors"
collect_plots() function collects all the plots made from res for all k (number of partitions)
into one single page to provide an easy and fast comparison between different k.
collect_plots(res)
The plots are:
k and the heatmap of
predicted classes for each k.k.k.k.All the plots in panels can be made by individual functions and they are plotted later in this section.
select_partition_number() produces several plots showing different
statistics for choosing “optimized” k. There are following statistics:
k;k, the area increased is defined as \(A_k - A_{k-1}\).The detailed explanations of these statistics can be found in the cola vignette.
Generally speaking, lower PAC score, higher mean silhouette score or higher
concordance corresponds to better partition. Rand index and Jaccard index
measure how similar the current partition is compared to partition with k-1.
If they are too similar, we won't accept k is better than k-1.
select_partition_number(res)
The numeric values for all these statistics can be obtained by get_stats().
get_stats(res)
#> k 1-PAC mean_silhouette concordance area_increased Rand Jaccard
#> 2 2 0.239 0.675 0.821 0.4382 0.529 0.529
#> 3 3 0.207 0.384 0.672 0.3664 0.791 0.655
#> 4 4 0.302 0.492 0.689 0.1453 0.807 0.611
#> 5 5 0.387 0.540 0.690 0.0763 0.911 0.734
#> 6 6 0.523 0.514 0.701 0.0610 0.964 0.863
suggest_best_k() suggests the best \(k\) based on these statistics. The rules are as follows:
suggest_best_k(res)
#> [1] 2
Following shows the table of the partitions (You need to click the show/hide
code output link to see it). The membership matrix (columns with name p*)
is inferred by
clue::cl_consensus()
function with the SE method. Basically the value in the membership matrix
represents the probability to belong to a certain group. The finall class
label for an item is determined with the group with highest probability it
belongs to.
In get_classes() function, the entropy is calculated from the membership
matrix and the silhouette score is calculated from the consensus matrix.
cbind(get_classes(res, k = 2), get_membership(res, k = 2))
#> class entropy silhouette p1 p2
#> GSM97138 1 0.1843 0.79589 0.972 0.028
#> GSM97145 1 0.1414 0.79524 0.980 0.020
#> GSM97147 2 0.9170 0.66404 0.332 0.668
#> GSM97125 1 0.1414 0.79524 0.980 0.020
#> GSM97127 1 0.1843 0.79214 0.972 0.028
#> GSM97130 1 0.6973 0.68136 0.812 0.188
#> GSM97133 1 0.1184 0.79626 0.984 0.016
#> GSM97134 1 0.9775 0.18757 0.588 0.412
#> GSM97120 1 0.0376 0.79614 0.996 0.004
#> GSM97126 1 0.9795 0.16879 0.584 0.416
#> GSM97112 1 0.0938 0.79842 0.988 0.012
#> GSM97115 1 0.9044 0.46729 0.680 0.320
#> GSM97116 1 0.0672 0.79765 0.992 0.008
#> GSM97117 2 0.9635 0.57084 0.388 0.612
#> GSM97119 1 0.0672 0.79812 0.992 0.008
#> GSM97122 1 0.0672 0.79812 0.992 0.008
#> GSM97135 1 0.0672 0.79812 0.992 0.008
#> GSM97136 2 0.9754 0.45856 0.408 0.592
#> GSM97139 1 0.0000 0.79435 1.000 0.000
#> GSM97146 1 0.0000 0.79435 1.000 0.000
#> GSM97123 2 0.1843 0.78410 0.028 0.972
#> GSM97129 1 0.9775 0.18757 0.588 0.412
#> GSM97143 1 0.7528 0.65733 0.784 0.216
#> GSM97113 1 0.9933 -0.00562 0.548 0.452
#> GSM97056 1 0.0938 0.79755 0.988 0.012
#> GSM97124 1 0.1184 0.79906 0.984 0.016
#> GSM97132 1 0.8267 0.60140 0.740 0.260
#> GSM97144 1 0.7219 0.66757 0.800 0.200
#> GSM97149 1 0.0000 0.79435 1.000 0.000
#> GSM97068 1 0.9983 -0.10058 0.524 0.476
#> GSM97071 2 0.7299 0.77846 0.204 0.796
#> GSM97086 2 0.6712 0.78945 0.176 0.824
#> GSM97103 2 0.2603 0.78832 0.044 0.956
#> GSM97057 1 0.9896 0.05481 0.560 0.440
#> GSM97060 2 0.0000 0.76364 0.000 1.000
#> GSM97075 2 0.8813 0.69977 0.300 0.700
#> GSM97098 2 0.2948 0.79134 0.052 0.948
#> GSM97099 2 0.9087 0.67161 0.324 0.676
#> GSM97101 2 0.9044 0.67688 0.320 0.680
#> GSM97105 2 0.5408 0.79947 0.124 0.876
#> GSM97106 2 0.0000 0.76364 0.000 1.000
#> GSM97121 2 0.9087 0.67507 0.324 0.676
#> GSM97128 2 0.9833 0.44221 0.424 0.576
#> GSM97131 2 0.4562 0.80011 0.096 0.904
#> GSM97137 1 0.3733 0.77199 0.928 0.072
#> GSM97118 1 0.8081 0.61868 0.752 0.248
#> GSM97114 2 0.9635 0.57084 0.388 0.612
#> GSM97142 1 0.0938 0.79842 0.988 0.012
#> GSM97140 2 0.8386 0.73240 0.268 0.732
#> GSM97141 2 0.9358 0.63565 0.352 0.648
#> GSM97055 1 0.8909 0.48777 0.692 0.308
#> GSM97090 1 0.9170 0.43864 0.668 0.332
#> GSM97091 1 0.1184 0.79872 0.984 0.016
#> GSM97148 1 0.0000 0.79435 1.000 0.000
#> GSM97063 1 0.1184 0.79872 0.984 0.016
#> GSM97053 1 0.0000 0.79435 1.000 0.000
#> GSM97066 2 0.2778 0.79060 0.048 0.952
#> GSM97079 2 0.6712 0.78945 0.176 0.824
#> GSM97083 2 0.9833 0.44221 0.424 0.576
#> GSM97084 2 0.7219 0.78012 0.200 0.800
#> GSM97094 2 0.7376 0.77554 0.208 0.792
#> GSM97096 2 0.2948 0.79134 0.052 0.948
#> GSM97097 2 0.7139 0.78118 0.196 0.804
#> GSM97107 2 0.7299 0.77839 0.204 0.796
#> GSM97054 2 0.7299 0.77846 0.204 0.796
#> GSM97062 2 0.6712 0.78945 0.176 0.824
#> GSM97069 2 0.1843 0.78307 0.028 0.972
#> GSM97070 2 0.2778 0.79060 0.048 0.952
#> GSM97073 2 0.3733 0.79732 0.072 0.928
#> GSM97076 2 0.9460 0.55757 0.364 0.636
#> GSM97077 2 0.8443 0.72606 0.272 0.728
#> GSM97095 1 0.9988 -0.11366 0.520 0.480
#> GSM97102 2 0.2603 0.78832 0.044 0.956
#> GSM97109 2 0.6623 0.77711 0.172 0.828
#> GSM97110 2 0.6623 0.77711 0.172 0.828
#> GSM97074 2 0.9686 0.47313 0.396 0.604
#> GSM97085 2 0.9710 0.49580 0.400 0.600
#> GSM97059 2 0.9209 0.65480 0.336 0.664
#> GSM97072 2 0.0000 0.76364 0.000 1.000
#> GSM97078 2 0.9815 0.44974 0.420 0.580
#> GSM97067 2 0.1843 0.78307 0.028 0.972
#> GSM97087 2 0.0376 0.76637 0.004 0.996
#> GSM97111 2 0.9427 0.61998 0.360 0.640
#> GSM97064 2 0.5737 0.80022 0.136 0.864
#> GSM97065 2 0.8861 0.67617 0.304 0.696
#> GSM97081 2 0.5178 0.80395 0.116 0.884
#> GSM97082 2 0.4562 0.79923 0.096 0.904
#> GSM97088 2 0.9775 0.46593 0.412 0.588
#> GSM97100 2 0.7883 0.75665 0.236 0.764
#> GSM97104 2 0.0672 0.76945 0.008 0.992
#> GSM97108 2 0.8555 0.71910 0.280 0.720
#> GSM97050 2 0.5842 0.80097 0.140 0.860
#> GSM97080 2 0.1633 0.78041 0.024 0.976
#> GSM97089 2 0.0672 0.76981 0.008 0.992
#> GSM97092 2 0.0938 0.77312 0.012 0.988
#> GSM97093 2 0.9170 0.62025 0.332 0.668
#> GSM97058 2 0.8327 0.73564 0.264 0.736
#> GSM97051 2 0.4939 0.79952 0.108 0.892
#> GSM97052 2 0.0000 0.76364 0.000 1.000
#> GSM97061 2 0.2043 0.78587 0.032 0.968
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM97138 1 0.456 0.79943 0.860 0.060 0.080
#> GSM97145 1 0.362 0.78540 0.896 0.072 0.032
#> GSM97147 2 0.478 0.50461 0.200 0.796 0.004
#> GSM97125 1 0.350 0.78674 0.900 0.072 0.028
#> GSM97127 1 0.392 0.78020 0.884 0.080 0.036
#> GSM97130 1 0.812 0.59915 0.640 0.224 0.136
#> GSM97133 1 0.386 0.78257 0.888 0.072 0.040
#> GSM97134 2 0.831 0.07345 0.420 0.500 0.080
#> GSM97120 1 0.337 0.78616 0.908 0.052 0.040
#> GSM97126 2 0.832 0.07186 0.424 0.496 0.080
#> GSM97112 1 0.471 0.79569 0.852 0.056 0.092
#> GSM97115 1 0.902 0.27249 0.496 0.364 0.140
#> GSM97116 1 0.293 0.78551 0.924 0.036 0.040
#> GSM97117 2 0.590 0.48977 0.244 0.736 0.020
#> GSM97119 1 0.453 0.79737 0.860 0.052 0.088
#> GSM97122 1 0.453 0.79737 0.860 0.052 0.088
#> GSM97135 1 0.453 0.79737 0.860 0.052 0.088
#> GSM97136 2 0.867 0.39584 0.272 0.580 0.148
#> GSM97139 1 0.326 0.78664 0.912 0.048 0.040
#> GSM97146 1 0.253 0.77393 0.936 0.020 0.044
#> GSM97123 2 0.625 -0.70151 0.000 0.556 0.444
#> GSM97129 2 0.831 0.07345 0.420 0.500 0.080
#> GSM97143 1 0.847 0.58010 0.604 0.252 0.144
#> GSM97113 2 0.773 0.20581 0.436 0.516 0.048
#> GSM97056 1 0.438 0.79128 0.868 0.064 0.068
#> GSM97124 1 0.482 0.79736 0.848 0.064 0.088
#> GSM97132 1 0.872 0.51827 0.576 0.272 0.152
#> GSM97144 1 0.829 0.57622 0.624 0.236 0.140
#> GSM97149 1 0.253 0.77393 0.936 0.020 0.044
#> GSM97068 2 0.813 0.30047 0.356 0.564 0.080
#> GSM97071 2 0.622 0.39330 0.016 0.688 0.296
#> GSM97086 2 0.492 0.41526 0.020 0.816 0.164
#> GSM97103 2 0.675 -0.57993 0.012 0.556 0.432
#> GSM97057 2 0.757 0.16707 0.452 0.508 0.040
#> GSM97060 3 0.620 0.93164 0.000 0.424 0.576
#> GSM97075 2 0.517 0.49013 0.172 0.804 0.024
#> GSM97098 2 0.669 -0.51159 0.012 0.580 0.408
#> GSM97099 2 0.497 0.49439 0.188 0.800 0.012
#> GSM97101 2 0.520 0.49552 0.184 0.796 0.020
#> GSM97105 2 0.287 0.36469 0.008 0.916 0.076
#> GSM97106 3 0.620 0.93164 0.000 0.424 0.576
#> GSM97121 2 0.649 0.48803 0.192 0.744 0.064
#> GSM97128 2 0.923 0.21235 0.152 0.428 0.420
#> GSM97131 2 0.357 0.28642 0.004 0.876 0.120
#> GSM97137 1 0.611 0.73870 0.780 0.140 0.080
#> GSM97118 1 0.879 0.53817 0.572 0.268 0.160
#> GSM97114 2 0.590 0.48977 0.244 0.736 0.020
#> GSM97142 1 0.471 0.79569 0.852 0.056 0.092
#> GSM97140 2 0.362 0.48739 0.136 0.864 0.000
#> GSM97141 2 0.550 0.49954 0.208 0.772 0.020
#> GSM97055 1 0.948 0.32876 0.468 0.336 0.196
#> GSM97090 1 0.909 0.22949 0.480 0.376 0.144
#> GSM97091 1 0.479 0.79487 0.848 0.056 0.096
#> GSM97148 1 0.253 0.77393 0.936 0.020 0.044
#> GSM97063 1 0.479 0.79487 0.848 0.056 0.096
#> GSM97053 1 0.418 0.79971 0.876 0.052 0.072
#> GSM97066 2 0.615 -0.30884 0.004 0.640 0.356
#> GSM97079 2 0.492 0.41526 0.020 0.816 0.164
#> GSM97083 2 0.923 0.21432 0.152 0.432 0.416
#> GSM97084 2 0.594 0.40154 0.020 0.732 0.248
#> GSM97094 2 0.626 0.40057 0.032 0.724 0.244
#> GSM97096 2 0.669 -0.51159 0.012 0.580 0.408
#> GSM97097 2 0.577 0.40275 0.020 0.748 0.232
#> GSM97107 2 0.598 0.40154 0.020 0.728 0.252
#> GSM97054 2 0.622 0.39330 0.016 0.688 0.296
#> GSM97062 2 0.492 0.41526 0.020 0.816 0.164
#> GSM97069 2 0.642 -0.56906 0.004 0.572 0.424
#> GSM97070 2 0.615 -0.30884 0.004 0.640 0.356
#> GSM97073 2 0.581 -0.14119 0.004 0.692 0.304
#> GSM97076 2 0.825 0.41096 0.252 0.620 0.128
#> GSM97077 2 0.392 0.48488 0.140 0.856 0.004
#> GSM97095 2 0.804 0.30703 0.352 0.572 0.076
#> GSM97102 2 0.675 -0.57993 0.012 0.556 0.432
#> GSM97109 2 0.826 -0.02498 0.112 0.604 0.284
#> GSM97110 2 0.826 -0.02498 0.112 0.604 0.284
#> GSM97074 2 0.911 0.39781 0.244 0.548 0.208
#> GSM97085 2 0.910 0.27059 0.140 0.456 0.404
#> GSM97059 2 0.496 0.50261 0.200 0.792 0.008
#> GSM97072 2 0.625 -0.58233 0.000 0.556 0.444
#> GSM97078 2 0.919 0.21578 0.148 0.432 0.420
#> GSM97067 2 0.617 -0.35908 0.004 0.636 0.360
#> GSM97087 3 0.621 0.93227 0.000 0.428 0.572
#> GSM97111 2 0.617 0.47971 0.224 0.740 0.036
#> GSM97064 2 0.684 0.03105 0.056 0.704 0.240
#> GSM97065 2 0.748 0.42081 0.192 0.692 0.116
#> GSM97081 2 0.614 -0.00138 0.032 0.736 0.232
#> GSM97082 3 0.665 0.74343 0.008 0.456 0.536
#> GSM97088 2 0.922 0.23764 0.152 0.440 0.408
#> GSM97100 2 0.392 0.47786 0.112 0.872 0.016
#> GSM97104 3 0.623 0.90486 0.000 0.436 0.564
#> GSM97108 2 0.423 0.49012 0.148 0.844 0.008
#> GSM97050 2 0.321 0.37734 0.028 0.912 0.060
#> GSM97080 2 0.627 -0.65278 0.000 0.544 0.456
#> GSM97089 3 0.622 0.93031 0.000 0.432 0.568
#> GSM97092 3 0.627 0.90614 0.000 0.452 0.548
#> GSM97093 2 0.818 0.41133 0.208 0.640 0.152
#> GSM97058 2 0.420 0.47993 0.136 0.852 0.012
#> GSM97051 2 0.286 0.34059 0.004 0.912 0.084
#> GSM97052 3 0.621 0.93230 0.000 0.428 0.572
#> GSM97061 2 0.624 -0.68452 0.000 0.560 0.440
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM97138 1 0.382 0.7157 0.840 0.040 0.000 0.120
#> GSM97145 1 0.300 0.7162 0.892 0.064 0.000 0.044
#> GSM97147 2 0.450 0.5842 0.200 0.776 0.008 0.016
#> GSM97125 1 0.316 0.7177 0.884 0.064 0.000 0.052
#> GSM97127 1 0.297 0.7116 0.892 0.072 0.000 0.036
#> GSM97130 1 0.722 0.4410 0.576 0.124 0.016 0.284
#> GSM97133 1 0.230 0.7068 0.920 0.064 0.000 0.016
#> GSM97134 2 0.797 0.1496 0.380 0.436 0.020 0.164
#> GSM97120 1 0.191 0.7159 0.940 0.040 0.000 0.020
#> GSM97126 2 0.797 0.1491 0.380 0.436 0.020 0.164
#> GSM97112 1 0.430 0.6783 0.752 0.008 0.000 0.240
#> GSM97115 1 0.815 0.1921 0.468 0.272 0.020 0.240
#> GSM97116 1 0.172 0.7150 0.948 0.020 0.000 0.032
#> GSM97117 2 0.518 0.5658 0.252 0.716 0.020 0.012
#> GSM97119 1 0.423 0.6838 0.760 0.008 0.000 0.232
#> GSM97122 1 0.423 0.6838 0.760 0.008 0.000 0.232
#> GSM97135 1 0.423 0.6838 0.760 0.008 0.000 0.232
#> GSM97136 2 0.943 0.2866 0.220 0.428 0.172 0.180
#> GSM97139 1 0.171 0.7137 0.948 0.036 0.000 0.016
#> GSM97146 1 0.117 0.7002 0.968 0.012 0.000 0.020
#> GSM97123 3 0.537 0.6458 0.000 0.364 0.616 0.020
#> GSM97129 2 0.797 0.1496 0.380 0.436 0.020 0.164
#> GSM97143 1 0.786 0.3654 0.508 0.116 0.040 0.336
#> GSM97113 2 0.647 0.3214 0.456 0.492 0.024 0.028
#> GSM97056 1 0.420 0.6947 0.828 0.040 0.008 0.124
#> GSM97124 1 0.454 0.6867 0.752 0.020 0.000 0.228
#> GSM97132 1 0.794 0.3074 0.488 0.120 0.040 0.352
#> GSM97144 1 0.744 0.4161 0.556 0.132 0.020 0.292
#> GSM97149 1 0.130 0.7002 0.964 0.016 0.000 0.020
#> GSM97068 2 0.742 0.3497 0.328 0.524 0.012 0.136
#> GSM97071 2 0.677 0.1000 0.008 0.504 0.072 0.416
#> GSM97086 2 0.567 0.4374 0.012 0.744 0.124 0.120
#> GSM97103 3 0.544 0.6576 0.000 0.288 0.672 0.040
#> GSM97057 2 0.610 0.2916 0.472 0.492 0.012 0.024
#> GSM97060 3 0.400 0.6959 0.000 0.164 0.812 0.024
#> GSM97075 2 0.469 0.5671 0.176 0.780 0.040 0.004
#> GSM97098 3 0.569 0.6196 0.000 0.336 0.624 0.040
#> GSM97099 2 0.414 0.5747 0.196 0.788 0.016 0.000
#> GSM97101 2 0.449 0.5727 0.192 0.780 0.024 0.004
#> GSM97105 2 0.352 0.4234 0.004 0.852 0.128 0.016
#> GSM97106 3 0.409 0.6973 0.000 0.172 0.804 0.024
#> GSM97121 2 0.636 0.5562 0.192 0.700 0.060 0.048
#> GSM97128 4 0.327 0.8445 0.032 0.076 0.008 0.884
#> GSM97131 2 0.457 0.3278 0.004 0.772 0.200 0.024
#> GSM97137 1 0.575 0.6267 0.724 0.092 0.008 0.176
#> GSM97118 1 0.816 0.2796 0.472 0.128 0.048 0.352
#> GSM97114 2 0.518 0.5658 0.252 0.716 0.020 0.012
#> GSM97142 1 0.430 0.6783 0.752 0.008 0.000 0.240
#> GSM97140 2 0.421 0.5653 0.136 0.824 0.028 0.012
#> GSM97141 2 0.476 0.5779 0.216 0.756 0.020 0.008
#> GSM97055 4 0.849 0.0452 0.352 0.160 0.052 0.436
#> GSM97090 1 0.810 0.1467 0.440 0.284 0.012 0.264
#> GSM97091 1 0.460 0.6625 0.732 0.008 0.004 0.256
#> GSM97148 1 0.130 0.7002 0.964 0.016 0.000 0.020
#> GSM97063 1 0.460 0.6625 0.732 0.008 0.004 0.256
#> GSM97053 1 0.409 0.6915 0.776 0.008 0.000 0.216
#> GSM97066 3 0.672 0.3926 0.004 0.444 0.476 0.076
#> GSM97079 2 0.567 0.4374 0.012 0.744 0.124 0.120
#> GSM97083 4 0.313 0.8431 0.032 0.076 0.004 0.888
#> GSM97084 2 0.679 0.3465 0.012 0.632 0.124 0.232
#> GSM97094 2 0.710 0.3252 0.024 0.616 0.120 0.240
#> GSM97096 3 0.569 0.6196 0.000 0.336 0.624 0.040
#> GSM97097 2 0.676 0.3545 0.012 0.640 0.132 0.216
#> GSM97107 2 0.682 0.3413 0.012 0.628 0.124 0.236
#> GSM97054 2 0.677 0.1000 0.008 0.504 0.072 0.416
#> GSM97062 2 0.567 0.4374 0.012 0.744 0.124 0.120
#> GSM97069 3 0.628 0.5865 0.004 0.332 0.600 0.064
#> GSM97070 3 0.672 0.3926 0.004 0.444 0.476 0.076
#> GSM97073 2 0.694 -0.3359 0.008 0.480 0.428 0.084
#> GSM97076 2 0.897 0.3390 0.252 0.476 0.132 0.140
#> GSM97077 2 0.423 0.5627 0.140 0.820 0.032 0.008
#> GSM97095 2 0.733 0.3687 0.324 0.536 0.012 0.128
#> GSM97102 3 0.544 0.6576 0.000 0.288 0.672 0.040
#> GSM97109 2 0.790 -0.2140 0.108 0.440 0.412 0.040
#> GSM97110 2 0.790 -0.2140 0.108 0.440 0.412 0.040
#> GSM97074 2 0.964 -0.0245 0.196 0.336 0.152 0.316
#> GSM97085 4 0.487 0.8121 0.028 0.132 0.040 0.800
#> GSM97059 2 0.496 0.5816 0.200 0.760 0.020 0.020
#> GSM97072 3 0.579 0.5382 0.000 0.384 0.580 0.036
#> GSM97078 4 0.317 0.8449 0.028 0.076 0.008 0.888
#> GSM97067 3 0.660 0.4302 0.004 0.432 0.496 0.068
#> GSM97087 3 0.415 0.6986 0.000 0.168 0.804 0.028
#> GSM97111 2 0.632 0.5362 0.220 0.684 0.068 0.028
#> GSM97064 2 0.631 0.0185 0.052 0.608 0.328 0.012
#> GSM97065 2 0.850 0.3649 0.200 0.544 0.140 0.116
#> GSM97081 2 0.653 -0.1799 0.020 0.568 0.368 0.044
#> GSM97082 3 0.608 0.6589 0.008 0.192 0.696 0.104
#> GSM97088 4 0.417 0.8304 0.024 0.116 0.024 0.836
#> GSM97100 2 0.422 0.5520 0.108 0.836 0.040 0.016
#> GSM97104 3 0.365 0.7040 0.000 0.152 0.832 0.016
#> GSM97108 2 0.419 0.5693 0.152 0.816 0.024 0.008
#> GSM97050 2 0.416 0.4476 0.028 0.840 0.108 0.024
#> GSM97080 3 0.567 0.6523 0.004 0.296 0.660 0.040
#> GSM97089 3 0.419 0.6994 0.000 0.172 0.800 0.028
#> GSM97092 3 0.464 0.7036 0.000 0.228 0.748 0.024
#> GSM97093 2 0.852 0.4059 0.196 0.528 0.192 0.084
#> GSM97058 2 0.412 0.5568 0.136 0.820 0.044 0.000
#> GSM97051 2 0.384 0.4001 0.004 0.836 0.136 0.024
#> GSM97052 3 0.427 0.7049 0.000 0.188 0.788 0.024
#> GSM97061 3 0.557 0.6368 0.000 0.368 0.604 0.028
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM97138 1 0.347 0.7115 0.836 0.044 0.000 0.004 0.116
#> GSM97145 1 0.260 0.7161 0.896 0.060 0.000 0.004 0.040
#> GSM97147 2 0.336 0.6348 0.164 0.816 0.000 0.000 0.020
#> GSM97125 1 0.275 0.7175 0.888 0.060 0.000 0.004 0.048
#> GSM97127 1 0.257 0.7121 0.896 0.068 0.000 0.004 0.032
#> GSM97130 1 0.702 0.4682 0.556 0.088 0.000 0.112 0.244
#> GSM97133 1 0.197 0.7071 0.924 0.060 0.000 0.004 0.012
#> GSM97134 2 0.755 0.2076 0.360 0.428 0.016 0.040 0.156
#> GSM97120 1 0.184 0.7130 0.936 0.040 0.000 0.008 0.016
#> GSM97126 2 0.758 0.2129 0.352 0.436 0.016 0.044 0.152
#> GSM97112 1 0.348 0.6750 0.752 0.000 0.000 0.000 0.248
#> GSM97115 1 0.781 0.2050 0.452 0.256 0.000 0.104 0.188
#> GSM97116 1 0.169 0.7118 0.944 0.020 0.000 0.008 0.028
#> GSM97117 2 0.446 0.6205 0.212 0.748 0.012 0.016 0.012
#> GSM97119 1 0.342 0.6807 0.760 0.000 0.000 0.000 0.240
#> GSM97122 1 0.342 0.6807 0.760 0.000 0.000 0.000 0.240
#> GSM97135 1 0.342 0.6807 0.760 0.000 0.000 0.000 0.240
#> GSM97136 2 0.917 0.2182 0.208 0.392 0.156 0.072 0.172
#> GSM97139 1 0.165 0.7108 0.944 0.036 0.000 0.008 0.012
#> GSM97146 1 0.131 0.6989 0.960 0.016 0.000 0.012 0.012
#> GSM97123 3 0.496 0.5726 0.000 0.352 0.608 0.040 0.000
#> GSM97129 2 0.755 0.2076 0.360 0.428 0.016 0.040 0.156
#> GSM97143 1 0.734 0.3815 0.504 0.076 0.036 0.052 0.332
#> GSM97113 2 0.537 0.4003 0.420 0.540 0.008 0.024 0.008
#> GSM97056 1 0.420 0.6898 0.808 0.028 0.000 0.060 0.104
#> GSM97124 1 0.403 0.6842 0.744 0.016 0.000 0.004 0.236
#> GSM97132 1 0.767 0.3425 0.472 0.084 0.032 0.076 0.336
#> GSM97144 1 0.720 0.4480 0.536 0.096 0.000 0.120 0.248
#> GSM97149 1 0.151 0.6984 0.952 0.024 0.000 0.012 0.012
#> GSM97068 2 0.705 0.3880 0.304 0.512 0.000 0.064 0.120
#> GSM97071 4 0.622 0.7479 0.000 0.196 0.000 0.544 0.260
#> GSM97086 2 0.492 -0.2177 0.000 0.552 0.004 0.424 0.020
#> GSM97103 3 0.557 0.6886 0.000 0.164 0.700 0.100 0.036
#> GSM97057 2 0.505 0.3760 0.436 0.536 0.000 0.020 0.008
#> GSM97060 3 0.290 0.6755 0.000 0.108 0.864 0.028 0.000
#> GSM97075 2 0.378 0.6361 0.140 0.820 0.020 0.008 0.012
#> GSM97098 3 0.588 0.6641 0.000 0.220 0.656 0.084 0.040
#> GSM97099 2 0.341 0.6359 0.156 0.824 0.008 0.004 0.008
#> GSM97101 2 0.364 0.6350 0.156 0.816 0.012 0.008 0.008
#> GSM97105 2 0.314 0.5011 0.000 0.864 0.076 0.056 0.004
#> GSM97106 3 0.300 0.6757 0.000 0.116 0.856 0.028 0.000
#> GSM97121 2 0.550 0.5672 0.164 0.728 0.032 0.044 0.032
#> GSM97128 5 0.158 0.7137 0.032 0.000 0.000 0.024 0.944
#> GSM97131 2 0.447 0.4200 0.000 0.768 0.148 0.076 0.008
#> GSM97137 1 0.572 0.6270 0.700 0.068 0.000 0.080 0.152
#> GSM97118 1 0.768 0.2985 0.464 0.084 0.040 0.064 0.348
#> GSM97114 2 0.446 0.6205 0.212 0.748 0.012 0.016 0.012
#> GSM97142 1 0.348 0.6750 0.752 0.000 0.000 0.000 0.248
#> GSM97140 2 0.302 0.6219 0.104 0.868 0.004 0.012 0.012
#> GSM97141 2 0.382 0.6336 0.176 0.796 0.012 0.012 0.004
#> GSM97055 5 0.751 0.0904 0.336 0.104 0.052 0.028 0.480
#> GSM97090 1 0.793 0.1721 0.420 0.268 0.000 0.100 0.212
#> GSM97091 1 0.374 0.6595 0.732 0.000 0.004 0.000 0.264
#> GSM97148 1 0.151 0.6984 0.952 0.024 0.000 0.012 0.012
#> GSM97063 1 0.374 0.6595 0.732 0.000 0.004 0.000 0.264
#> GSM97053 1 0.331 0.6885 0.776 0.000 0.000 0.000 0.224
#> GSM97066 3 0.697 0.5459 0.000 0.284 0.504 0.180 0.032
#> GSM97079 2 0.494 -0.2604 0.000 0.536 0.004 0.440 0.020
#> GSM97083 5 0.167 0.7127 0.032 0.000 0.000 0.028 0.940
#> GSM97084 4 0.408 0.8798 0.000 0.228 0.000 0.744 0.028
#> GSM97094 4 0.464 0.8773 0.012 0.220 0.000 0.728 0.040
#> GSM97096 3 0.588 0.6641 0.000 0.220 0.656 0.084 0.040
#> GSM97097 4 0.409 0.8761 0.000 0.228 0.008 0.748 0.016
#> GSM97107 4 0.413 0.8822 0.000 0.224 0.000 0.744 0.032
#> GSM97054 4 0.622 0.7479 0.000 0.196 0.000 0.544 0.260
#> GSM97062 2 0.494 -0.2604 0.000 0.536 0.004 0.440 0.020
#> GSM97069 3 0.608 0.6457 0.000 0.220 0.628 0.128 0.024
#> GSM97070 3 0.697 0.5459 0.000 0.284 0.504 0.180 0.032
#> GSM97073 3 0.733 0.5061 0.000 0.312 0.448 0.196 0.044
#> GSM97076 2 0.940 0.0650 0.228 0.340 0.124 0.216 0.092
#> GSM97077 2 0.281 0.6281 0.108 0.872 0.008 0.000 0.012
#> GSM97095 2 0.693 0.3977 0.304 0.524 0.000 0.060 0.112
#> GSM97102 3 0.557 0.6886 0.000 0.164 0.700 0.100 0.036
#> GSM97109 3 0.778 0.3150 0.076 0.392 0.416 0.076 0.040
#> GSM97110 3 0.778 0.3150 0.076 0.392 0.416 0.076 0.040
#> GSM97074 5 0.972 0.1914 0.180 0.212 0.124 0.176 0.308
#> GSM97085 5 0.338 0.6819 0.020 0.032 0.040 0.032 0.876
#> GSM97059 2 0.368 0.6273 0.168 0.804 0.000 0.008 0.020
#> GSM97072 3 0.605 0.6245 0.000 0.228 0.604 0.160 0.008
#> GSM97078 5 0.158 0.7123 0.028 0.000 0.000 0.028 0.944
#> GSM97067 3 0.688 0.5654 0.000 0.280 0.512 0.180 0.028
#> GSM97087 3 0.312 0.6800 0.000 0.120 0.852 0.024 0.004
#> GSM97111 2 0.576 0.6023 0.188 0.700 0.052 0.024 0.036
#> GSM97064 2 0.479 0.1617 0.024 0.652 0.316 0.008 0.000
#> GSM97065 2 0.884 0.2025 0.176 0.452 0.128 0.168 0.076
#> GSM97081 2 0.646 -0.1268 0.016 0.540 0.356 0.036 0.052
#> GSM97082 3 0.498 0.6375 0.000 0.124 0.748 0.024 0.104
#> GSM97088 5 0.250 0.7068 0.020 0.020 0.016 0.028 0.916
#> GSM97100 2 0.274 0.6053 0.076 0.892 0.008 0.016 0.008
#> GSM97104 3 0.319 0.6860 0.000 0.088 0.864 0.036 0.012
#> GSM97108 2 0.355 0.6307 0.120 0.840 0.012 0.008 0.020
#> GSM97050 2 0.311 0.5345 0.008 0.880 0.064 0.036 0.012
#> GSM97080 3 0.546 0.6843 0.000 0.184 0.696 0.096 0.024
#> GSM97089 3 0.325 0.6794 0.000 0.120 0.848 0.024 0.008
#> GSM97092 3 0.369 0.6809 0.000 0.200 0.780 0.020 0.000
#> GSM97093 2 0.786 0.4470 0.176 0.532 0.180 0.036 0.076
#> GSM97058 2 0.293 0.6276 0.100 0.872 0.020 0.004 0.004
#> GSM97051 2 0.363 0.4860 0.000 0.836 0.092 0.064 0.008
#> GSM97052 3 0.311 0.6861 0.000 0.140 0.840 0.020 0.000
#> GSM97061 3 0.462 0.5977 0.000 0.340 0.636 0.024 0.000
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM97138 1 0.4440 0.6577 0.764 0.060 0.000 0.000 0.112 0.064
#> GSM97145 1 0.2302 0.6740 0.900 0.060 0.000 0.000 0.032 0.008
#> GSM97147 2 0.3519 0.6966 0.136 0.820 0.008 0.008 0.016 0.012
#> GSM97125 1 0.2445 0.6749 0.892 0.060 0.000 0.000 0.040 0.008
#> GSM97127 1 0.2350 0.6719 0.900 0.064 0.000 0.004 0.024 0.008
#> GSM97130 1 0.7203 0.4232 0.528 0.064 0.004 0.124 0.224 0.056
#> GSM97133 1 0.2146 0.6667 0.908 0.060 0.000 0.000 0.008 0.024
#> GSM97134 2 0.7206 0.2036 0.352 0.420 0.004 0.020 0.128 0.076
#> GSM97120 1 0.2291 0.6653 0.904 0.040 0.000 0.000 0.012 0.044
#> GSM97126 2 0.7229 0.2183 0.332 0.432 0.004 0.016 0.124 0.092
#> GSM97112 1 0.3695 0.6221 0.732 0.000 0.000 0.000 0.244 0.024
#> GSM97115 1 0.8001 0.2026 0.424 0.236 0.004 0.104 0.172 0.060
#> GSM97116 1 0.3078 0.6445 0.856 0.028 0.000 0.000 0.032 0.084
#> GSM97117 2 0.4050 0.6732 0.180 0.764 0.008 0.012 0.000 0.036
#> GSM97119 1 0.3645 0.6281 0.740 0.000 0.000 0.000 0.236 0.024
#> GSM97122 1 0.3645 0.6281 0.740 0.000 0.000 0.000 0.236 0.024
#> GSM97135 1 0.3645 0.6281 0.740 0.000 0.000 0.000 0.236 0.024
#> GSM97136 2 0.8871 0.0961 0.188 0.324 0.128 0.012 0.124 0.224
#> GSM97139 1 0.2728 0.6476 0.872 0.040 0.000 0.000 0.008 0.080
#> GSM97146 1 0.3031 0.6198 0.852 0.032 0.000 0.000 0.016 0.100
#> GSM97123 3 0.5077 0.3700 0.000 0.308 0.616 0.036 0.000 0.040
#> GSM97129 2 0.7206 0.2036 0.352 0.420 0.004 0.020 0.128 0.076
#> GSM97143 1 0.6923 0.3281 0.468 0.060 0.012 0.000 0.284 0.176
#> GSM97113 2 0.5207 0.4488 0.352 0.564 0.000 0.000 0.012 0.072
#> GSM97056 1 0.5045 0.6328 0.744 0.016 0.004 0.064 0.100 0.072
#> GSM97124 1 0.3914 0.6356 0.740 0.012 0.000 0.004 0.228 0.016
#> GSM97132 1 0.7387 0.2972 0.444 0.064 0.016 0.012 0.280 0.184
#> GSM97144 1 0.7312 0.4088 0.512 0.072 0.004 0.128 0.232 0.052
#> GSM97149 1 0.3172 0.6184 0.844 0.040 0.000 0.000 0.016 0.100
#> GSM97068 2 0.7401 0.3941 0.264 0.496 0.012 0.076 0.108 0.044
#> GSM97071 4 0.4689 0.7158 0.000 0.040 0.004 0.680 0.256 0.020
#> GSM97086 2 0.4892 0.1418 0.000 0.524 0.004 0.432 0.012 0.028
#> GSM97103 3 0.4825 0.4116 0.000 0.044 0.620 0.016 0.000 0.320
#> GSM97057 2 0.5181 0.4347 0.360 0.560 0.000 0.000 0.012 0.068
#> GSM97060 3 0.1503 0.5788 0.000 0.016 0.944 0.008 0.000 0.032
#> GSM97075 2 0.3741 0.6896 0.116 0.816 0.032 0.004 0.004 0.028
#> GSM97098 3 0.5473 0.4119 0.000 0.100 0.584 0.012 0.004 0.300
#> GSM97099 2 0.3346 0.6928 0.116 0.836 0.016 0.004 0.004 0.024
#> GSM97101 2 0.3670 0.6890 0.128 0.816 0.020 0.008 0.004 0.024
#> GSM97105 2 0.3366 0.6100 0.000 0.844 0.060 0.052 0.000 0.044
#> GSM97106 3 0.1838 0.5796 0.000 0.020 0.928 0.012 0.000 0.040
#> GSM97121 2 0.5418 0.6522 0.152 0.712 0.020 0.032 0.020 0.064
#> GSM97128 5 0.0632 0.8144 0.024 0.000 0.000 0.000 0.976 0.000
#> GSM97131 2 0.5032 0.5196 0.000 0.724 0.132 0.084 0.008 0.052
#> GSM97137 1 0.6263 0.5729 0.648 0.044 0.004 0.100 0.140 0.064
#> GSM97118 1 0.7121 0.2448 0.428 0.064 0.012 0.000 0.292 0.204
#> GSM97114 2 0.4050 0.6732 0.180 0.764 0.008 0.012 0.000 0.036
#> GSM97142 1 0.3695 0.6221 0.732 0.000 0.000 0.000 0.244 0.024
#> GSM97140 2 0.3024 0.6918 0.084 0.868 0.016 0.012 0.008 0.012
#> GSM97141 2 0.3416 0.6912 0.140 0.816 0.008 0.004 0.000 0.032
#> GSM97055 5 0.7886 0.1630 0.284 0.080 0.040 0.012 0.408 0.176
#> GSM97090 1 0.8173 0.1825 0.388 0.244 0.004 0.104 0.196 0.064
#> GSM97091 1 0.3956 0.6054 0.712 0.000 0.000 0.000 0.252 0.036
#> GSM97148 1 0.3172 0.6184 0.844 0.040 0.000 0.000 0.016 0.100
#> GSM97063 1 0.3956 0.6054 0.712 0.000 0.000 0.000 0.252 0.036
#> GSM97053 1 0.3190 0.6427 0.772 0.000 0.000 0.000 0.220 0.008
#> GSM97066 6 0.4482 0.4816 0.000 0.040 0.360 0.000 0.000 0.600
#> GSM97079 2 0.4772 0.0991 0.000 0.512 0.004 0.452 0.012 0.020
#> GSM97083 5 0.0777 0.8134 0.024 0.000 0.000 0.004 0.972 0.000
#> GSM97084 4 0.1370 0.8695 0.000 0.036 0.000 0.948 0.004 0.012
#> GSM97094 4 0.1767 0.8649 0.012 0.036 0.000 0.932 0.020 0.000
#> GSM97096 3 0.5473 0.4119 0.000 0.100 0.584 0.012 0.004 0.300
#> GSM97097 4 0.1080 0.8658 0.000 0.032 0.004 0.960 0.000 0.004
#> GSM97107 4 0.1409 0.8699 0.000 0.032 0.000 0.948 0.008 0.012
#> GSM97054 4 0.4689 0.7158 0.000 0.040 0.004 0.680 0.256 0.020
#> GSM97062 2 0.4772 0.0991 0.000 0.512 0.004 0.452 0.012 0.020
#> GSM97069 3 0.4372 -0.1643 0.000 0.024 0.544 0.000 0.000 0.432
#> GSM97070 6 0.4482 0.4816 0.000 0.040 0.360 0.000 0.000 0.600
#> GSM97073 6 0.4377 0.4825 0.000 0.044 0.312 0.000 0.000 0.644
#> GSM97076 6 0.4382 0.4155 0.164 0.104 0.000 0.000 0.004 0.728
#> GSM97077 2 0.2688 0.6910 0.084 0.880 0.020 0.004 0.008 0.004
#> GSM97095 2 0.7353 0.4062 0.264 0.500 0.012 0.080 0.104 0.040
#> GSM97102 3 0.4825 0.4116 0.000 0.044 0.620 0.016 0.000 0.320
#> GSM97109 3 0.7202 0.1556 0.048 0.284 0.368 0.008 0.004 0.288
#> GSM97110 3 0.7202 0.1556 0.048 0.284 0.368 0.008 0.004 0.288
#> GSM97074 6 0.6218 0.0737 0.148 0.048 0.008 0.000 0.216 0.580
#> GSM97085 5 0.3331 0.7613 0.016 0.008 0.032 0.000 0.840 0.104
#> GSM97059 2 0.3968 0.6946 0.128 0.804 0.012 0.012 0.016 0.028
#> GSM97072 6 0.4535 0.2384 0.000 0.024 0.472 0.004 0.000 0.500
#> GSM97078 5 0.0632 0.8142 0.024 0.000 0.000 0.000 0.976 0.000
#> GSM97067 6 0.4419 0.4460 0.000 0.032 0.384 0.000 0.000 0.584
#> GSM97087 3 0.0922 0.5832 0.000 0.024 0.968 0.004 0.004 0.000
#> GSM97111 2 0.5635 0.6271 0.160 0.680 0.044 0.012 0.008 0.096
#> GSM97064 2 0.4532 0.3031 0.020 0.628 0.336 0.004 0.000 0.012
#> GSM97065 6 0.5713 0.3398 0.116 0.264 0.008 0.012 0.004 0.596
#> GSM97081 2 0.6628 0.0187 0.012 0.480 0.336 0.020 0.016 0.136
#> GSM97082 3 0.3608 0.5160 0.000 0.024 0.828 0.004 0.068 0.076
#> GSM97088 5 0.2765 0.8008 0.024 0.004 0.020 0.008 0.888 0.056
#> GSM97100 2 0.3088 0.6844 0.064 0.872 0.012 0.016 0.008 0.028
#> GSM97104 3 0.2872 0.5217 0.000 0.012 0.832 0.004 0.000 0.152
#> GSM97108 2 0.3354 0.6932 0.096 0.848 0.020 0.008 0.012 0.016
#> GSM97050 2 0.3198 0.6172 0.000 0.860 0.052 0.024 0.008 0.056
#> GSM97080 3 0.4245 0.1648 0.000 0.024 0.644 0.000 0.004 0.328
#> GSM97089 3 0.1036 0.5825 0.000 0.024 0.964 0.004 0.008 0.000
#> GSM97092 3 0.2613 0.5408 0.000 0.140 0.848 0.000 0.000 0.012
#> GSM97093 2 0.7800 0.4408 0.164 0.488 0.196 0.016 0.072 0.064
#> GSM97058 2 0.2781 0.6908 0.084 0.872 0.032 0.004 0.000 0.008
#> GSM97051 2 0.4210 0.5813 0.000 0.796 0.072 0.064 0.008 0.060
#> GSM97052 3 0.1333 0.5808 0.000 0.048 0.944 0.000 0.000 0.008
#> GSM97061 3 0.3840 0.4212 0.000 0.284 0.696 0.000 0.000 0.020
Heatmaps for the consensus matrix. It visualizes the probability of two samples to be in a same group.
consensus_heatmap(res, k = 2)
consensus_heatmap(res, k = 3)
consensus_heatmap(res, k = 4)
consensus_heatmap(res, k = 5)
consensus_heatmap(res, k = 6)
Heatmaps for the membership of samples in all partitions to see how consistent they are:
membership_heatmap(res, k = 2)
membership_heatmap(res, k = 3)
membership_heatmap(res, k = 4)
membership_heatmap(res, k = 5)
membership_heatmap(res, k = 6)
As soon as we have had the classes for columns, we can look for signatures which are significantly different between classes which can be candidate marks for certain classes. Following are the heatmaps for signatures.
Signature heatmaps where rows are scaled:
get_signatures(res, k = 2)
get_signatures(res, k = 3)
get_signatures(res, k = 4)
get_signatures(res, k = 5)
get_signatures(res, k = 6)
Signature heatmaps where rows are not scaled:
get_signatures(res, k = 2, scale_rows = FALSE)
get_signatures(res, k = 3, scale_rows = FALSE)
get_signatures(res, k = 4, scale_rows = FALSE)
get_signatures(res, k = 5, scale_rows = FALSE)
get_signatures(res, k = 6, scale_rows = FALSE)
Compare the overlap of signatures from different k:
compare_signatures(res)
get_signature() returns a data frame invisibly. TO get the list of signatures, the function
call should be assigned to a variable explicitly. In following code, if plot argument is set
to FALSE, no heatmap is plotted while only the differential analysis is performed.
# code only for demonstration
tb = get_signature(res, k = ..., plot = FALSE)
An example of the output of tb is:
#> which_row fdr mean_1 mean_2 scaled_mean_1 scaled_mean_2 km
#> 1 38 0.042760348 8.373488 9.131774 -0.5533452 0.5164555 1
#> 2 40 0.018707592 7.106213 8.469186 -0.6173731 0.5762149 1
#> 3 55 0.019134737 10.221463 11.207825 -0.6159697 0.5749050 1
#> 4 59 0.006059896 5.921854 7.869574 -0.6899429 0.6439467 1
#> 5 60 0.018055526 8.928898 10.211722 -0.6204761 0.5791110 1
#> 6 98 0.009384629 15.714769 14.887706 0.6635654 -0.6193277 2
...
The columns in tb are:
which_row: row indices corresponding to the input matrix.fdr: FDR for the differential test. mean_x: The mean value in group x.scaled_mean_x: The mean value in group x after rows are scaled.km: Row groups if k-means clustering is applied to rows.UMAP plot which shows how samples are separated.
dimension_reduction(res, k = 2, method = "UMAP")
dimension_reduction(res, k = 3, method = "UMAP")
dimension_reduction(res, k = 4, method = "UMAP")
dimension_reduction(res, k = 5, method = "UMAP")
dimension_reduction(res, k = 6, method = "UMAP")
Following heatmap shows how subgroups are split when increasing k:
collect_classes(res)
Test correlation between subgroups and known annotations. If the known annotation is numeric, one-way ANOVA test is applied, and if the known annotation is discrete, chi-squared contingency table test is applied.
test_to_known_factors(res)
#> n disease.state(p) specimen(p) cell.type(p) other(p) k
#> SD:hclust 83 4.74e-07 1.000 4.45e-16 0.0554 2
#> SD:hclust 37 3.09e-02 0.737 3.51e-11 0.7217 3
#> SD:hclust 59 1.90e-03 0.780 6.17e-10 0.8124 4
#> SD:hclust 71 4.06e-04 0.815 5.03e-14 0.3725 5
#> SD:hclust 60 3.27e-03 0.720 3.94e-15 0.3439 6
If matrix rows can be associated to genes, consider to use functional_enrichment(res,
...) to perform function enrichment for the signature genes. See this vignette for more detailed explanations.
The object with results only for a single top-value method and a single partition method can be extracted as:
res = res_list["SD", "kmeans"]
# you can also extract it by
# res = res_list["SD:kmeans"]
A summary of res and all the functions that can be applied to it:
res
#> A 'ConsensusPartition' object with k = 2, 3, 4, 5, 6.
#> On a matrix with 21512 rows and 100 columns.
#> Top rows (1000, 2000, 3000, 4000, 5000) are extracted by 'SD' method.
#> Subgroups are detected by 'kmeans' method.
#> Performed in total 1250 partitions by row resampling.
#> Best k for subgroups seems to be 2.
#>
#> Following methods can be applied to this 'ConsensusPartition' object:
#> [1] "cola_report" "collect_classes" "collect_plots"
#> [4] "collect_stats" "colnames" "compare_signatures"
#> [7] "consensus_heatmap" "dimension_reduction" "functional_enrichment"
#> [10] "get_anno_col" "get_anno" "get_classes"
#> [13] "get_consensus" "get_matrix" "get_membership"
#> [16] "get_param" "get_signatures" "get_stats"
#> [19] "is_best_k" "is_stable_k" "membership_heatmap"
#> [22] "ncol" "nrow" "plot_ecdf"
#> [25] "rownames" "select_partition_number" "show"
#> [28] "suggest_best_k" "test_to_known_factors"
collect_plots() function collects all the plots made from res for all k (number of partitions)
into one single page to provide an easy and fast comparison between different k.
collect_plots(res)
The plots are:
k and the heatmap of
predicted classes for each k.k.k.k.All the plots in panels can be made by individual functions and they are plotted later in this section.
select_partition_number() produces several plots showing different
statistics for choosing “optimized” k. There are following statistics:
k;k, the area increased is defined as \(A_k - A_{k-1}\).The detailed explanations of these statistics can be found in the cola vignette.
Generally speaking, lower PAC score, higher mean silhouette score or higher
concordance corresponds to better partition. Rand index and Jaccard index
measure how similar the current partition is compared to partition with k-1.
If they are too similar, we won't accept k is better than k-1.
select_partition_number(res)
The numeric values for all these statistics can be obtained by get_stats().
get_stats(res)
#> k 1-PAC mean_silhouette concordance area_increased Rand Jaccard
#> 2 2 0.938 0.947 0.973 0.4933 0.508 0.508
#> 3 3 0.603 0.811 0.876 0.3336 0.704 0.481
#> 4 4 0.708 0.777 0.861 0.1295 0.823 0.532
#> 5 5 0.685 0.658 0.770 0.0579 0.971 0.883
#> 6 6 0.702 0.547 0.674 0.0408 0.922 0.674
suggest_best_k() suggests the best \(k\) based on these statistics. The rules are as follows:
suggest_best_k(res)
#> [1] 2
Following shows the table of the partitions (You need to click the show/hide
code output link to see it). The membership matrix (columns with name p*)
is inferred by
clue::cl_consensus()
function with the SE method. Basically the value in the membership matrix
represents the probability to belong to a certain group. The finall class
label for an item is determined with the group with highest probability it
belongs to.
In get_classes() function, the entropy is calculated from the membership
matrix and the silhouette score is calculated from the consensus matrix.
cbind(get_classes(res, k = 2), get_membership(res, k = 2))
#> class entropy silhouette p1 p2
#> GSM97138 1 0.0000 0.980 1.000 0.000
#> GSM97145 1 0.0000 0.980 1.000 0.000
#> GSM97147 1 0.0376 0.977 0.996 0.004
#> GSM97125 1 0.0000 0.980 1.000 0.000
#> GSM97127 1 0.0000 0.980 1.000 0.000
#> GSM97130 1 0.0000 0.980 1.000 0.000
#> GSM97133 1 0.0000 0.980 1.000 0.000
#> GSM97134 1 0.0000 0.980 1.000 0.000
#> GSM97120 1 0.0000 0.980 1.000 0.000
#> GSM97126 1 0.0000 0.980 1.000 0.000
#> GSM97112 1 0.0000 0.980 1.000 0.000
#> GSM97115 1 0.0000 0.980 1.000 0.000
#> GSM97116 1 0.0000 0.980 1.000 0.000
#> GSM97117 2 0.2778 0.947 0.048 0.952
#> GSM97119 1 0.0000 0.980 1.000 0.000
#> GSM97122 1 0.0000 0.980 1.000 0.000
#> GSM97135 1 0.0000 0.980 1.000 0.000
#> GSM97136 2 0.2603 0.948 0.044 0.956
#> GSM97139 1 0.0000 0.980 1.000 0.000
#> GSM97146 1 0.0000 0.980 1.000 0.000
#> GSM97123 2 0.0000 0.967 0.000 1.000
#> GSM97129 2 0.2948 0.945 0.052 0.948
#> GSM97143 1 0.0000 0.980 1.000 0.000
#> GSM97113 2 0.5059 0.889 0.112 0.888
#> GSM97056 1 0.0000 0.980 1.000 0.000
#> GSM97124 1 0.0000 0.980 1.000 0.000
#> GSM97132 1 0.0000 0.980 1.000 0.000
#> GSM97144 1 0.0000 0.980 1.000 0.000
#> GSM97149 1 0.0000 0.980 1.000 0.000
#> GSM97068 2 0.9881 0.288 0.436 0.564
#> GSM97071 2 0.0672 0.966 0.008 0.992
#> GSM97086 2 0.0672 0.966 0.008 0.992
#> GSM97103 2 0.0000 0.967 0.000 1.000
#> GSM97057 2 0.7602 0.752 0.220 0.780
#> GSM97060 2 0.0000 0.967 0.000 1.000
#> GSM97075 2 0.0376 0.967 0.004 0.996
#> GSM97098 2 0.0000 0.967 0.000 1.000
#> GSM97099 2 0.2948 0.945 0.052 0.948
#> GSM97101 2 0.2948 0.945 0.052 0.948
#> GSM97105 2 0.0672 0.966 0.008 0.992
#> GSM97106 2 0.0000 0.967 0.000 1.000
#> GSM97121 2 0.2948 0.945 0.052 0.948
#> GSM97128 1 0.8081 0.691 0.752 0.248
#> GSM97131 2 0.0000 0.967 0.000 1.000
#> GSM97137 1 0.0000 0.980 1.000 0.000
#> GSM97118 1 0.0000 0.980 1.000 0.000
#> GSM97114 2 0.8207 0.695 0.256 0.744
#> GSM97142 1 0.0000 0.980 1.000 0.000
#> GSM97140 2 0.2948 0.945 0.052 0.948
#> GSM97141 2 0.2948 0.945 0.052 0.948
#> GSM97055 1 0.0000 0.980 1.000 0.000
#> GSM97090 1 0.0000 0.980 1.000 0.000
#> GSM97091 1 0.0000 0.980 1.000 0.000
#> GSM97148 1 0.0000 0.980 1.000 0.000
#> GSM97063 1 0.0000 0.980 1.000 0.000
#> GSM97053 1 0.0000 0.980 1.000 0.000
#> GSM97066 2 0.0000 0.967 0.000 1.000
#> GSM97079 2 0.0672 0.966 0.008 0.992
#> GSM97083 1 0.0000 0.980 1.000 0.000
#> GSM97084 2 0.0672 0.966 0.008 0.992
#> GSM97094 1 0.3274 0.926 0.940 0.060
#> GSM97096 2 0.0000 0.967 0.000 1.000
#> GSM97097 2 0.0376 0.967 0.004 0.996
#> GSM97107 1 0.5629 0.846 0.868 0.132
#> GSM97054 2 0.0672 0.966 0.008 0.992
#> GSM97062 2 0.0672 0.966 0.008 0.992
#> GSM97069 2 0.0000 0.967 0.000 1.000
#> GSM97070 2 0.0000 0.967 0.000 1.000
#> GSM97073 2 0.0000 0.967 0.000 1.000
#> GSM97076 1 0.0000 0.980 1.000 0.000
#> GSM97077 2 0.0672 0.966 0.008 0.992
#> GSM97095 1 0.0672 0.974 0.992 0.008
#> GSM97102 2 0.0000 0.967 0.000 1.000
#> GSM97109 2 0.3274 0.939 0.060 0.940
#> GSM97110 2 0.2948 0.945 0.052 0.948
#> GSM97074 1 0.4161 0.910 0.916 0.084
#> GSM97085 2 0.2236 0.946 0.036 0.964
#> GSM97059 1 0.1414 0.964 0.980 0.020
#> GSM97072 2 0.0000 0.967 0.000 1.000
#> GSM97078 1 0.8144 0.687 0.748 0.252
#> GSM97067 2 0.0000 0.967 0.000 1.000
#> GSM97087 2 0.0000 0.967 0.000 1.000
#> GSM97111 2 0.2778 0.947 0.048 0.952
#> GSM97064 2 0.0000 0.967 0.000 1.000
#> GSM97065 2 0.0376 0.967 0.004 0.996
#> GSM97081 2 0.0000 0.967 0.000 1.000
#> GSM97082 2 0.0000 0.967 0.000 1.000
#> GSM97088 2 0.4815 0.878 0.104 0.896
#> GSM97100 2 0.0672 0.966 0.008 0.992
#> GSM97104 2 0.0000 0.967 0.000 1.000
#> GSM97108 2 0.2948 0.945 0.052 0.948
#> GSM97050 2 0.0672 0.966 0.008 0.992
#> GSM97080 2 0.0000 0.967 0.000 1.000
#> GSM97089 2 0.0000 0.967 0.000 1.000
#> GSM97092 2 0.0000 0.967 0.000 1.000
#> GSM97093 2 0.0376 0.967 0.004 0.996
#> GSM97058 2 0.0376 0.967 0.004 0.996
#> GSM97051 2 0.0376 0.967 0.004 0.996
#> GSM97052 2 0.0000 0.967 0.000 1.000
#> GSM97061 2 0.0000 0.967 0.000 1.000
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM97138 1 0.2356 0.9175 0.928 0.072 0.000
#> GSM97145 1 0.1860 0.9203 0.948 0.052 0.000
#> GSM97147 2 0.1031 0.8007 0.024 0.976 0.000
#> GSM97125 1 0.0892 0.9219 0.980 0.020 0.000
#> GSM97127 1 0.2711 0.9148 0.912 0.088 0.000
#> GSM97130 1 0.4842 0.8291 0.776 0.224 0.000
#> GSM97133 1 0.2711 0.9148 0.912 0.088 0.000
#> GSM97134 1 0.4931 0.7561 0.768 0.232 0.000
#> GSM97120 1 0.2711 0.9148 0.912 0.088 0.000
#> GSM97126 1 0.3941 0.8424 0.844 0.156 0.000
#> GSM97112 1 0.0000 0.9208 1.000 0.000 0.000
#> GSM97115 2 0.1031 0.7934 0.024 0.976 0.000
#> GSM97116 1 0.2711 0.9148 0.912 0.088 0.000
#> GSM97117 2 0.5098 0.8034 0.000 0.752 0.248
#> GSM97119 1 0.0000 0.9208 1.000 0.000 0.000
#> GSM97122 1 0.0000 0.9208 1.000 0.000 0.000
#> GSM97135 1 0.0000 0.9208 1.000 0.000 0.000
#> GSM97136 3 0.2063 0.8566 0.008 0.044 0.948
#> GSM97139 1 0.2711 0.9148 0.912 0.088 0.000
#> GSM97146 1 0.2711 0.9148 0.912 0.088 0.000
#> GSM97123 3 0.5529 0.4185 0.000 0.296 0.704
#> GSM97129 2 0.4974 0.8128 0.000 0.764 0.236
#> GSM97143 1 0.0000 0.9208 1.000 0.000 0.000
#> GSM97113 2 0.4291 0.8121 0.000 0.820 0.180
#> GSM97056 1 0.2625 0.9163 0.916 0.084 0.000
#> GSM97124 1 0.0237 0.9212 0.996 0.004 0.000
#> GSM97132 1 0.1964 0.9044 0.944 0.056 0.000
#> GSM97144 1 0.4504 0.8076 0.804 0.196 0.000
#> GSM97149 1 0.2711 0.9148 0.912 0.088 0.000
#> GSM97068 2 0.0000 0.8022 0.000 1.000 0.000
#> GSM97071 2 0.7561 0.0962 0.040 0.516 0.444
#> GSM97086 2 0.2796 0.8331 0.000 0.908 0.092
#> GSM97103 2 0.6305 0.3978 0.000 0.516 0.484
#> GSM97057 2 0.1267 0.8149 0.004 0.972 0.024
#> GSM97060 3 0.0000 0.8819 0.000 0.000 1.000
#> GSM97075 2 0.5098 0.8034 0.000 0.752 0.248
#> GSM97098 3 0.5138 0.5291 0.000 0.252 0.748
#> GSM97099 2 0.4974 0.8113 0.000 0.764 0.236
#> GSM97101 2 0.4605 0.8297 0.000 0.796 0.204
#> GSM97105 2 0.3879 0.8449 0.000 0.848 0.152
#> GSM97106 3 0.3412 0.7606 0.000 0.124 0.876
#> GSM97121 2 0.3686 0.8461 0.000 0.860 0.140
#> GSM97128 3 0.9410 0.3949 0.220 0.276 0.504
#> GSM97131 2 0.4796 0.8211 0.000 0.780 0.220
#> GSM97137 1 0.3340 0.9075 0.880 0.120 0.000
#> GSM97118 1 0.1860 0.9036 0.948 0.052 0.000
#> GSM97114 2 0.5085 0.7799 0.092 0.836 0.072
#> GSM97142 1 0.0000 0.9208 1.000 0.000 0.000
#> GSM97140 2 0.3038 0.8450 0.000 0.896 0.104
#> GSM97141 2 0.4974 0.8113 0.000 0.764 0.236
#> GSM97055 1 0.0592 0.9199 0.988 0.012 0.000
#> GSM97090 2 0.3482 0.7061 0.128 0.872 0.000
#> GSM97091 1 0.0000 0.9208 1.000 0.000 0.000
#> GSM97148 1 0.2711 0.9148 0.912 0.088 0.000
#> GSM97063 1 0.0000 0.9208 1.000 0.000 0.000
#> GSM97053 1 0.0237 0.9212 0.996 0.004 0.000
#> GSM97066 3 0.0000 0.8819 0.000 0.000 1.000
#> GSM97079 2 0.3267 0.8332 0.000 0.884 0.116
#> GSM97083 1 0.5072 0.7834 0.792 0.196 0.012
#> GSM97084 2 0.3030 0.8273 0.004 0.904 0.092
#> GSM97094 2 0.5092 0.6968 0.176 0.804 0.020
#> GSM97096 3 0.0747 0.8757 0.000 0.016 0.984
#> GSM97097 2 0.5138 0.7723 0.000 0.748 0.252
#> GSM97107 2 0.5384 0.6722 0.188 0.788 0.024
#> GSM97054 2 0.2860 0.8325 0.004 0.912 0.084
#> GSM97062 2 0.3038 0.8303 0.000 0.896 0.104
#> GSM97069 3 0.0000 0.8819 0.000 0.000 1.000
#> GSM97070 3 0.0237 0.8825 0.000 0.004 0.996
#> GSM97073 3 0.0000 0.8819 0.000 0.000 1.000
#> GSM97076 1 0.8548 0.4271 0.568 0.312 0.120
#> GSM97077 2 0.3340 0.8459 0.000 0.880 0.120
#> GSM97095 2 0.0747 0.7988 0.016 0.984 0.000
#> GSM97102 3 0.0237 0.8825 0.000 0.004 0.996
#> GSM97109 2 0.5024 0.8183 0.004 0.776 0.220
#> GSM97110 2 0.4974 0.8113 0.000 0.764 0.236
#> GSM97074 3 0.7283 0.5605 0.260 0.068 0.672
#> GSM97085 3 0.2448 0.8218 0.076 0.000 0.924
#> GSM97059 2 0.0424 0.8000 0.008 0.992 0.000
#> GSM97072 3 0.0000 0.8819 0.000 0.000 1.000
#> GSM97078 3 0.9765 0.2448 0.240 0.336 0.424
#> GSM97067 3 0.0000 0.8819 0.000 0.000 1.000
#> GSM97087 3 0.0237 0.8825 0.000 0.004 0.996
#> GSM97111 2 0.5016 0.8096 0.000 0.760 0.240
#> GSM97064 2 0.5138 0.8030 0.000 0.748 0.252
#> GSM97065 2 0.6126 0.5814 0.000 0.600 0.400
#> GSM97081 3 0.1031 0.8711 0.000 0.024 0.976
#> GSM97082 3 0.0237 0.8825 0.000 0.004 0.996
#> GSM97088 3 0.5582 0.7339 0.088 0.100 0.812
#> GSM97100 2 0.2625 0.8396 0.000 0.916 0.084
#> GSM97104 3 0.0237 0.8825 0.000 0.004 0.996
#> GSM97108 2 0.3551 0.8465 0.000 0.868 0.132
#> GSM97050 2 0.4121 0.8422 0.000 0.832 0.168
#> GSM97080 3 0.0237 0.8825 0.000 0.004 0.996
#> GSM97089 3 0.0237 0.8825 0.000 0.004 0.996
#> GSM97092 3 0.0237 0.8825 0.000 0.004 0.996
#> GSM97093 2 0.5016 0.8113 0.000 0.760 0.240
#> GSM97058 2 0.4555 0.8336 0.000 0.800 0.200
#> GSM97051 2 0.3752 0.8435 0.000 0.856 0.144
#> GSM97052 3 0.0237 0.8825 0.000 0.004 0.996
#> GSM97061 3 0.3879 0.7242 0.000 0.152 0.848
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM97138 1 0.1743 0.8497 0.940 0.056 0.000 0.004
#> GSM97145 1 0.1733 0.8550 0.948 0.024 0.000 0.028
#> GSM97147 2 0.0927 0.8955 0.008 0.976 0.000 0.016
#> GSM97125 1 0.1792 0.8547 0.932 0.000 0.000 0.068
#> GSM97127 1 0.1722 0.8497 0.944 0.048 0.000 0.008
#> GSM97130 4 0.4881 0.5991 0.196 0.048 0.000 0.756
#> GSM97133 1 0.1824 0.8454 0.936 0.060 0.000 0.004
#> GSM97134 4 0.3355 0.6202 0.160 0.004 0.000 0.836
#> GSM97120 1 0.1824 0.8454 0.936 0.060 0.000 0.004
#> GSM97126 2 0.7403 -0.1162 0.380 0.452 0.000 0.168
#> GSM97112 1 0.3400 0.8389 0.820 0.000 0.000 0.180
#> GSM97115 4 0.4836 0.6287 0.008 0.320 0.000 0.672
#> GSM97116 1 0.1743 0.8474 0.940 0.056 0.000 0.004
#> GSM97117 2 0.2156 0.9062 0.008 0.928 0.060 0.004
#> GSM97119 1 0.3400 0.8389 0.820 0.000 0.000 0.180
#> GSM97122 1 0.3400 0.8389 0.820 0.000 0.000 0.180
#> GSM97135 1 0.3400 0.8389 0.820 0.000 0.000 0.180
#> GSM97136 3 0.5560 0.5983 0.016 0.276 0.684 0.024
#> GSM97139 1 0.1743 0.8474 0.940 0.056 0.000 0.004
#> GSM97146 1 0.1743 0.8474 0.940 0.056 0.000 0.004
#> GSM97123 3 0.4049 0.7368 0.000 0.212 0.780 0.008
#> GSM97129 2 0.2317 0.9078 0.012 0.928 0.048 0.012
#> GSM97143 1 0.3400 0.8389 0.820 0.000 0.000 0.180
#> GSM97113 2 0.1109 0.8884 0.028 0.968 0.004 0.000
#> GSM97056 1 0.3533 0.7978 0.864 0.056 0.000 0.080
#> GSM97124 1 0.3400 0.8389 0.820 0.000 0.000 0.180
#> GSM97132 4 0.5000 -0.1604 0.496 0.000 0.000 0.504
#> GSM97144 4 0.3479 0.6333 0.148 0.012 0.000 0.840
#> GSM97149 1 0.1824 0.8454 0.936 0.060 0.000 0.004
#> GSM97068 2 0.2831 0.7827 0.004 0.876 0.000 0.120
#> GSM97071 4 0.2111 0.6941 0.000 0.044 0.024 0.932
#> GSM97086 4 0.4331 0.6301 0.000 0.288 0.000 0.712
#> GSM97103 3 0.4678 0.6953 0.000 0.232 0.744 0.024
#> GSM97057 2 0.0804 0.8900 0.012 0.980 0.000 0.008
#> GSM97060 3 0.0188 0.9151 0.000 0.000 0.996 0.004
#> GSM97075 2 0.2234 0.9051 0.008 0.924 0.064 0.004
#> GSM97098 3 0.2987 0.8546 0.000 0.104 0.880 0.016
#> GSM97099 2 0.1994 0.9083 0.008 0.936 0.052 0.004
#> GSM97101 2 0.1706 0.9107 0.016 0.948 0.036 0.000
#> GSM97105 2 0.2032 0.9050 0.000 0.936 0.036 0.028
#> GSM97106 3 0.0707 0.9121 0.000 0.000 0.980 0.020
#> GSM97121 2 0.1706 0.9086 0.000 0.948 0.036 0.016
#> GSM97128 4 0.2450 0.6551 0.072 0.000 0.016 0.912
#> GSM97131 2 0.4579 0.7189 0.000 0.768 0.200 0.032
#> GSM97137 1 0.4150 0.7560 0.824 0.056 0.000 0.120
#> GSM97118 4 0.4977 -0.0831 0.460 0.000 0.000 0.540
#> GSM97114 2 0.1209 0.8851 0.032 0.964 0.000 0.004
#> GSM97142 1 0.3400 0.8389 0.820 0.000 0.000 0.180
#> GSM97140 2 0.1624 0.9076 0.000 0.952 0.028 0.020
#> GSM97141 2 0.1798 0.9104 0.016 0.944 0.040 0.000
#> GSM97055 1 0.4562 0.7982 0.764 0.028 0.000 0.208
#> GSM97090 4 0.4914 0.6364 0.012 0.312 0.000 0.676
#> GSM97091 1 0.3569 0.8293 0.804 0.000 0.000 0.196
#> GSM97148 1 0.1743 0.8474 0.940 0.056 0.000 0.004
#> GSM97063 1 0.3569 0.8293 0.804 0.000 0.000 0.196
#> GSM97053 1 0.3024 0.8465 0.852 0.000 0.000 0.148
#> GSM97066 3 0.1940 0.9014 0.000 0.000 0.924 0.076
#> GSM97079 4 0.4456 0.6363 0.000 0.280 0.004 0.716
#> GSM97083 4 0.2814 0.6294 0.132 0.000 0.000 0.868
#> GSM97084 4 0.4428 0.6387 0.000 0.276 0.004 0.720
#> GSM97094 4 0.3271 0.7144 0.012 0.132 0.000 0.856
#> GSM97096 3 0.1584 0.9022 0.000 0.036 0.952 0.012
#> GSM97097 4 0.7159 0.4967 0.000 0.272 0.180 0.548
#> GSM97107 4 0.3280 0.7134 0.016 0.124 0.000 0.860
#> GSM97054 4 0.4382 0.6249 0.000 0.296 0.000 0.704
#> GSM97062 4 0.4456 0.6363 0.000 0.280 0.004 0.716
#> GSM97069 3 0.1792 0.9034 0.000 0.000 0.932 0.068
#> GSM97070 3 0.1940 0.9014 0.000 0.000 0.924 0.076
#> GSM97073 3 0.1940 0.9014 0.000 0.000 0.924 0.076
#> GSM97076 4 0.7886 -0.0575 0.364 0.228 0.004 0.404
#> GSM97077 2 0.1936 0.9043 0.000 0.940 0.028 0.032
#> GSM97095 4 0.5099 0.5517 0.008 0.380 0.000 0.612
#> GSM97102 3 0.0000 0.9153 0.000 0.000 1.000 0.000
#> GSM97109 2 0.2521 0.9024 0.028 0.924 0.032 0.016
#> GSM97110 2 0.2573 0.9042 0.024 0.920 0.044 0.012
#> GSM97074 4 0.5646 0.4256 0.056 0.000 0.272 0.672
#> GSM97085 3 0.3552 0.8243 0.024 0.000 0.848 0.128
#> GSM97059 2 0.1209 0.8845 0.004 0.964 0.000 0.032
#> GSM97072 3 0.1716 0.9053 0.000 0.000 0.936 0.064
#> GSM97078 4 0.2561 0.6580 0.068 0.004 0.016 0.912
#> GSM97067 3 0.1940 0.9014 0.000 0.000 0.924 0.076
#> GSM97087 3 0.0376 0.9156 0.000 0.004 0.992 0.004
#> GSM97111 2 0.2140 0.9088 0.008 0.932 0.052 0.008
#> GSM97064 2 0.2988 0.8694 0.000 0.876 0.112 0.012
#> GSM97065 2 0.4337 0.8261 0.016 0.836 0.072 0.076
#> GSM97081 3 0.2149 0.8747 0.000 0.088 0.912 0.000
#> GSM97082 3 0.0188 0.9151 0.000 0.000 0.996 0.004
#> GSM97088 4 0.4755 0.5690 0.040 0.000 0.200 0.760
#> GSM97100 2 0.1890 0.8900 0.000 0.936 0.008 0.056
#> GSM97104 3 0.0188 0.9151 0.000 0.000 0.996 0.004
#> GSM97108 2 0.1820 0.9077 0.000 0.944 0.036 0.020
#> GSM97050 2 0.2032 0.9050 0.000 0.936 0.036 0.028
#> GSM97080 3 0.1302 0.9098 0.000 0.000 0.956 0.044
#> GSM97089 3 0.0376 0.9156 0.000 0.004 0.992 0.004
#> GSM97092 3 0.0376 0.9156 0.000 0.004 0.992 0.004
#> GSM97093 2 0.2593 0.8740 0.000 0.892 0.104 0.004
#> GSM97058 2 0.1929 0.9067 0.000 0.940 0.036 0.024
#> GSM97051 2 0.4036 0.8202 0.000 0.836 0.088 0.076
#> GSM97052 3 0.0376 0.9156 0.000 0.004 0.992 0.004
#> GSM97061 3 0.2611 0.8666 0.000 0.096 0.896 0.008
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM97138 1 0.2248 0.68858 0.900 0.012 0.000 0.000 0.088
#> GSM97145 1 0.4268 0.67260 0.708 0.024 0.000 0.000 0.268
#> GSM97147 2 0.1430 0.87523 0.000 0.944 0.000 0.052 0.004
#> GSM97125 1 0.3857 0.66916 0.688 0.000 0.000 0.000 0.312
#> GSM97127 1 0.1830 0.68734 0.924 0.008 0.000 0.000 0.068
#> GSM97130 4 0.5830 0.51839 0.144 0.016 0.000 0.652 0.188
#> GSM97133 1 0.0404 0.67889 0.988 0.012 0.000 0.000 0.000
#> GSM97134 4 0.5329 0.00832 0.028 0.012 0.000 0.488 0.472
#> GSM97120 1 0.0510 0.67732 0.984 0.016 0.000 0.000 0.000
#> GSM97126 2 0.5960 0.07934 0.120 0.528 0.000 0.000 0.352
#> GSM97112 1 0.4227 0.62398 0.580 0.000 0.000 0.000 0.420
#> GSM97115 4 0.5463 0.65581 0.028 0.132 0.000 0.708 0.132
#> GSM97116 1 0.0404 0.67919 0.988 0.012 0.000 0.000 0.000
#> GSM97117 2 0.0992 0.88035 0.000 0.968 0.024 0.000 0.008
#> GSM97119 1 0.4227 0.62398 0.580 0.000 0.000 0.000 0.420
#> GSM97122 1 0.4227 0.62398 0.580 0.000 0.000 0.000 0.420
#> GSM97135 1 0.4219 0.62680 0.584 0.000 0.000 0.000 0.416
#> GSM97136 3 0.6514 0.35824 0.000 0.304 0.476 0.000 0.220
#> GSM97139 1 0.0290 0.67997 0.992 0.008 0.000 0.000 0.000
#> GSM97146 1 0.0404 0.67919 0.988 0.012 0.000 0.000 0.000
#> GSM97123 3 0.4167 0.72221 0.000 0.160 0.788 0.028 0.024
#> GSM97129 2 0.1845 0.87009 0.000 0.928 0.016 0.000 0.056
#> GSM97143 1 0.4249 0.60973 0.568 0.000 0.000 0.000 0.432
#> GSM97113 2 0.0798 0.87966 0.016 0.976 0.000 0.000 0.008
#> GSM97056 1 0.2217 0.61443 0.920 0.012 0.000 0.044 0.024
#> GSM97124 1 0.4235 0.61981 0.576 0.000 0.000 0.000 0.424
#> GSM97132 5 0.6344 0.29696 0.260 0.008 0.000 0.176 0.556
#> GSM97144 4 0.4880 0.53142 0.040 0.012 0.000 0.692 0.256
#> GSM97149 1 0.0510 0.67732 0.984 0.016 0.000 0.000 0.000
#> GSM97068 2 0.4208 0.75422 0.032 0.788 0.000 0.156 0.024
#> GSM97071 4 0.4290 0.57987 0.000 0.016 0.032 0.768 0.184
#> GSM97086 4 0.1732 0.70784 0.000 0.080 0.000 0.920 0.000
#> GSM97103 3 0.6024 0.65419 0.000 0.236 0.640 0.052 0.072
#> GSM97057 2 0.2597 0.86311 0.040 0.896 0.000 0.060 0.004
#> GSM97060 3 0.2585 0.80814 0.000 0.008 0.896 0.024 0.072
#> GSM97075 2 0.0898 0.88105 0.000 0.972 0.020 0.000 0.008
#> GSM97098 3 0.4778 0.71031 0.000 0.188 0.740 0.020 0.052
#> GSM97099 2 0.0854 0.88174 0.004 0.976 0.012 0.000 0.008
#> GSM97101 2 0.0727 0.88259 0.004 0.980 0.012 0.000 0.004
#> GSM97105 2 0.2474 0.86780 0.000 0.896 0.012 0.084 0.008
#> GSM97106 3 0.3278 0.79426 0.000 0.024 0.868 0.056 0.052
#> GSM97121 2 0.1074 0.88372 0.000 0.968 0.012 0.016 0.004
#> GSM97128 5 0.4940 0.15160 0.004 0.008 0.012 0.392 0.584
#> GSM97131 2 0.6179 0.61953 0.000 0.628 0.148 0.196 0.028
#> GSM97137 1 0.4093 0.44684 0.808 0.012 0.000 0.092 0.088
#> GSM97118 5 0.5939 0.41931 0.188 0.008 0.000 0.180 0.624
#> GSM97114 2 0.0992 0.87782 0.024 0.968 0.000 0.000 0.008
#> GSM97142 1 0.4227 0.62398 0.580 0.000 0.000 0.000 0.420
#> GSM97140 2 0.1770 0.87759 0.000 0.936 0.008 0.048 0.008
#> GSM97141 2 0.0854 0.88174 0.004 0.976 0.012 0.000 0.008
#> GSM97055 5 0.5742 -0.42718 0.436 0.056 0.000 0.012 0.496
#> GSM97090 4 0.5444 0.65820 0.028 0.112 0.000 0.708 0.152
#> GSM97091 1 0.4562 0.49321 0.496 0.000 0.000 0.008 0.496
#> GSM97148 1 0.0404 0.67919 0.988 0.012 0.000 0.000 0.000
#> GSM97063 1 0.4549 0.55129 0.528 0.000 0.000 0.008 0.464
#> GSM97053 1 0.4101 0.64910 0.628 0.000 0.000 0.000 0.372
#> GSM97066 3 0.4413 0.73006 0.000 0.000 0.724 0.044 0.232
#> GSM97079 4 0.1831 0.70823 0.000 0.076 0.000 0.920 0.004
#> GSM97083 4 0.4907 0.04879 0.012 0.008 0.000 0.512 0.468
#> GSM97084 4 0.1671 0.70984 0.000 0.076 0.000 0.924 0.000
#> GSM97094 4 0.3351 0.67836 0.004 0.028 0.000 0.836 0.132
#> GSM97096 3 0.3506 0.79036 0.000 0.076 0.852 0.020 0.052
#> GSM97097 4 0.4409 0.57625 0.000 0.068 0.092 0.800 0.040
#> GSM97107 4 0.3474 0.67919 0.008 0.028 0.000 0.832 0.132
#> GSM97054 4 0.2628 0.70900 0.000 0.088 0.000 0.884 0.028
#> GSM97062 4 0.1671 0.70984 0.000 0.076 0.000 0.924 0.000
#> GSM97069 3 0.4168 0.74739 0.000 0.000 0.756 0.044 0.200
#> GSM97070 3 0.4450 0.73816 0.000 0.004 0.736 0.044 0.216
#> GSM97073 3 0.4538 0.73289 0.000 0.004 0.724 0.044 0.228
#> GSM97076 5 0.6896 0.26669 0.024 0.276 0.044 0.084 0.572
#> GSM97077 2 0.2115 0.87307 0.000 0.916 0.008 0.068 0.008
#> GSM97095 4 0.6780 0.41436 0.032 0.304 0.000 0.520 0.144
#> GSM97102 3 0.2967 0.80468 0.000 0.012 0.868 0.016 0.104
#> GSM97109 2 0.1143 0.88083 0.008 0.968 0.008 0.004 0.012
#> GSM97110 2 0.1143 0.88083 0.008 0.968 0.008 0.004 0.012
#> GSM97074 5 0.5217 0.35834 0.004 0.008 0.116 0.156 0.716
#> GSM97085 3 0.4884 0.48351 0.004 0.000 0.572 0.020 0.404
#> GSM97059 2 0.3093 0.84253 0.032 0.872 0.000 0.080 0.016
#> GSM97072 3 0.4709 0.74882 0.000 0.004 0.716 0.056 0.224
#> GSM97078 5 0.5045 -0.05167 0.004 0.008 0.012 0.456 0.520
#> GSM97067 3 0.4538 0.73147 0.000 0.004 0.724 0.044 0.228
#> GSM97087 3 0.1612 0.80786 0.000 0.012 0.948 0.016 0.024
#> GSM97111 2 0.0898 0.88105 0.000 0.972 0.020 0.000 0.008
#> GSM97064 2 0.4514 0.79683 0.000 0.780 0.132 0.064 0.024
#> GSM97065 2 0.4855 0.70913 0.000 0.760 0.060 0.040 0.140
#> GSM97081 3 0.2966 0.76854 0.000 0.136 0.848 0.000 0.016
#> GSM97082 3 0.1329 0.81071 0.000 0.008 0.956 0.004 0.032
#> GSM97088 5 0.6548 0.11335 0.004 0.008 0.140 0.360 0.488
#> GSM97100 2 0.2798 0.83544 0.000 0.852 0.000 0.140 0.008
#> GSM97104 3 0.2802 0.80482 0.000 0.008 0.876 0.016 0.100
#> GSM97108 2 0.1522 0.88086 0.000 0.944 0.012 0.044 0.000
#> GSM97050 2 0.4394 0.81205 0.000 0.788 0.084 0.112 0.016
#> GSM97080 3 0.2886 0.78327 0.000 0.000 0.844 0.008 0.148
#> GSM97089 3 0.1612 0.80786 0.000 0.012 0.948 0.016 0.024
#> GSM97092 3 0.1820 0.80588 0.000 0.020 0.940 0.020 0.020
#> GSM97093 2 0.4117 0.77439 0.000 0.788 0.164 0.028 0.020
#> GSM97058 2 0.2266 0.87426 0.000 0.912 0.016 0.064 0.008
#> GSM97051 2 0.6262 0.60974 0.000 0.608 0.148 0.220 0.024
#> GSM97052 3 0.1913 0.80535 0.000 0.024 0.936 0.020 0.020
#> GSM97061 3 0.2888 0.79091 0.000 0.056 0.888 0.036 0.020
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM97138 5 0.3975 -0.30224 0.452 0.000 0.000 0.000 0.544 0.004
#> GSM97145 5 0.2182 0.67849 0.068 0.020 0.000 0.000 0.904 0.008
#> GSM97147 2 0.2503 0.79257 0.044 0.896 0.004 0.044 0.000 0.012
#> GSM97125 5 0.1471 0.70590 0.064 0.000 0.000 0.000 0.932 0.004
#> GSM97127 5 0.3769 0.00761 0.356 0.000 0.000 0.000 0.640 0.004
#> GSM97130 4 0.5977 0.59867 0.332 0.004 0.000 0.528 0.032 0.104
#> GSM97133 1 0.3864 0.38421 0.520 0.000 0.000 0.000 0.480 0.000
#> GSM97134 4 0.7651 0.42628 0.260 0.004 0.000 0.356 0.184 0.196
#> GSM97120 1 0.3864 0.38421 0.520 0.000 0.000 0.000 0.480 0.000
#> GSM97126 2 0.5641 0.51346 0.052 0.632 0.000 0.000 0.208 0.108
#> GSM97112 5 0.0000 0.76672 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM97115 4 0.6391 0.64558 0.272 0.100 0.000 0.532 0.000 0.096
#> GSM97116 1 0.3864 0.38421 0.520 0.000 0.000 0.000 0.480 0.000
#> GSM97117 2 0.1888 0.79368 0.012 0.916 0.004 0.000 0.000 0.068
#> GSM97119 5 0.0000 0.76672 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM97122 5 0.0000 0.76672 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM97135 5 0.0260 0.76375 0.008 0.000 0.000 0.000 0.992 0.000
#> GSM97136 2 0.7550 -0.15776 0.048 0.368 0.308 0.000 0.044 0.232
#> GSM97139 1 0.3864 0.38421 0.520 0.000 0.000 0.000 0.480 0.000
#> GSM97146 1 0.3864 0.38421 0.520 0.000 0.000 0.000 0.480 0.000
#> GSM97123 3 0.2056 0.68490 0.000 0.080 0.904 0.004 0.000 0.012
#> GSM97129 2 0.2633 0.77833 0.020 0.864 0.004 0.000 0.000 0.112
#> GSM97143 5 0.0603 0.75869 0.004 0.000 0.000 0.000 0.980 0.016
#> GSM97113 2 0.1769 0.79715 0.012 0.924 0.004 0.000 0.000 0.060
#> GSM97056 1 0.4428 0.35277 0.676 0.000 0.000 0.052 0.268 0.004
#> GSM97124 5 0.0508 0.76397 0.012 0.000 0.000 0.000 0.984 0.004
#> GSM97132 5 0.7389 -0.15876 0.272 0.004 0.000 0.132 0.404 0.188
#> GSM97144 4 0.6058 0.65074 0.244 0.012 0.000 0.592 0.044 0.108
#> GSM97149 1 0.3993 0.38252 0.520 0.000 0.000 0.000 0.476 0.004
#> GSM97068 2 0.5227 0.60634 0.144 0.668 0.004 0.168 0.000 0.016
#> GSM97071 4 0.4849 0.61998 0.112 0.000 0.000 0.648 0.000 0.240
#> GSM97086 4 0.1623 0.67546 0.020 0.032 0.004 0.940 0.000 0.004
#> GSM97103 3 0.7485 0.38311 0.056 0.232 0.480 0.080 0.000 0.152
#> GSM97057 2 0.3316 0.78272 0.076 0.848 0.016 0.052 0.000 0.008
#> GSM97060 3 0.2822 0.68394 0.040 0.000 0.852 0.000 0.000 0.108
#> GSM97075 2 0.1913 0.79517 0.012 0.920 0.004 0.004 0.000 0.060
#> GSM97098 3 0.6323 0.49628 0.056 0.200 0.576 0.008 0.000 0.160
#> GSM97099 2 0.2126 0.79040 0.020 0.904 0.004 0.000 0.000 0.072
#> GSM97101 2 0.1410 0.79946 0.008 0.944 0.004 0.000 0.000 0.044
#> GSM97105 2 0.3069 0.78471 0.044 0.868 0.024 0.056 0.000 0.008
#> GSM97106 3 0.2389 0.70753 0.036 0.012 0.908 0.020 0.000 0.024
#> GSM97121 2 0.1173 0.80115 0.016 0.960 0.000 0.016 0.000 0.008
#> GSM97128 1 0.7658 -0.35942 0.308 0.000 0.004 0.232 0.152 0.304
#> GSM97131 2 0.6447 0.56112 0.044 0.572 0.176 0.188 0.000 0.020
#> GSM97137 1 0.4445 0.33532 0.712 0.000 0.000 0.072 0.208 0.008
#> GSM97118 1 0.7487 -0.20017 0.308 0.000 0.000 0.132 0.300 0.260
#> GSM97114 2 0.1895 0.79419 0.016 0.912 0.000 0.000 0.000 0.072
#> GSM97142 5 0.0000 0.76672 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM97140 2 0.2727 0.78952 0.044 0.888 0.020 0.040 0.000 0.008
#> GSM97141 2 0.1668 0.79602 0.008 0.928 0.004 0.000 0.000 0.060
#> GSM97055 5 0.5032 0.50055 0.096 0.052 0.004 0.012 0.740 0.096
#> GSM97090 4 0.6289 0.65437 0.272 0.068 0.000 0.556 0.008 0.096
#> GSM97091 5 0.2009 0.69006 0.068 0.000 0.000 0.000 0.908 0.024
#> GSM97148 1 0.3864 0.38421 0.520 0.000 0.000 0.000 0.480 0.000
#> GSM97063 5 0.1524 0.71516 0.060 0.000 0.000 0.000 0.932 0.008
#> GSM97053 5 0.0777 0.75139 0.024 0.000 0.000 0.000 0.972 0.004
#> GSM97066 6 0.3672 0.47788 0.000 0.000 0.368 0.000 0.000 0.632
#> GSM97079 4 0.1198 0.68612 0.012 0.020 0.004 0.960 0.000 0.004
#> GSM97083 4 0.7520 0.36514 0.304 0.000 0.004 0.320 0.116 0.256
#> GSM97084 4 0.0964 0.70016 0.012 0.016 0.000 0.968 0.000 0.004
#> GSM97094 4 0.2849 0.71141 0.084 0.016 0.000 0.872 0.008 0.020
#> GSM97096 3 0.5540 0.60495 0.056 0.112 0.676 0.008 0.000 0.148
#> GSM97097 4 0.3483 0.58974 0.052 0.016 0.068 0.844 0.000 0.020
#> GSM97107 4 0.2983 0.71206 0.088 0.016 0.000 0.864 0.008 0.024
#> GSM97054 4 0.4098 0.67613 0.092 0.060 0.008 0.800 0.000 0.040
#> GSM97062 4 0.0692 0.69292 0.000 0.020 0.004 0.976 0.000 0.000
#> GSM97069 6 0.3937 0.40601 0.004 0.000 0.424 0.000 0.000 0.572
#> GSM97070 6 0.3737 0.46588 0.000 0.000 0.392 0.000 0.000 0.608
#> GSM97073 6 0.3706 0.46604 0.000 0.000 0.380 0.000 0.000 0.620
#> GSM97076 6 0.4737 0.23004 0.016 0.280 0.004 0.004 0.032 0.664
#> GSM97077 2 0.3207 0.78395 0.044 0.860 0.028 0.060 0.000 0.008
#> GSM97095 4 0.7341 0.43555 0.276 0.288 0.000 0.332 0.000 0.104
#> GSM97102 3 0.4011 0.59900 0.060 0.000 0.736 0.000 0.000 0.204
#> GSM97109 2 0.2812 0.78148 0.032 0.876 0.016 0.004 0.000 0.072
#> GSM97110 2 0.2884 0.78128 0.036 0.872 0.016 0.004 0.000 0.072
#> GSM97074 6 0.4563 0.29442 0.116 0.000 0.016 0.044 0.056 0.768
#> GSM97085 6 0.6577 0.29378 0.120 0.000 0.280 0.012 0.064 0.524
#> GSM97059 2 0.4111 0.74051 0.100 0.788 0.016 0.088 0.000 0.008
#> GSM97072 6 0.4168 0.39439 0.016 0.000 0.400 0.000 0.000 0.584
#> GSM97078 1 0.7539 -0.41371 0.304 0.000 0.004 0.280 0.116 0.296
#> GSM97067 6 0.3737 0.46588 0.000 0.000 0.392 0.000 0.000 0.608
#> GSM97087 3 0.1692 0.70704 0.012 0.008 0.932 0.000 0.000 0.048
#> GSM97111 2 0.1982 0.79186 0.016 0.912 0.004 0.000 0.000 0.068
#> GSM97064 2 0.5509 0.63270 0.044 0.636 0.252 0.056 0.000 0.012
#> GSM97065 2 0.4289 0.32693 0.012 0.540 0.004 0.000 0.000 0.444
#> GSM97081 3 0.4887 0.57030 0.016 0.192 0.688 0.000 0.000 0.104
#> GSM97082 3 0.2520 0.68262 0.012 0.008 0.872 0.000 0.000 0.108
#> GSM97088 6 0.8232 -0.33530 0.304 0.000 0.076 0.212 0.096 0.312
#> GSM97100 2 0.3750 0.76091 0.044 0.816 0.024 0.108 0.000 0.008
#> GSM97104 3 0.3555 0.61725 0.040 0.000 0.776 0.000 0.000 0.184
#> GSM97108 2 0.2263 0.79357 0.044 0.908 0.004 0.036 0.000 0.008
#> GSM97050 2 0.5718 0.66941 0.052 0.656 0.176 0.104 0.000 0.012
#> GSM97080 3 0.4010 0.02704 0.008 0.000 0.584 0.000 0.000 0.408
#> GSM97089 3 0.1873 0.70604 0.020 0.008 0.924 0.000 0.000 0.048
#> GSM97092 3 0.0806 0.71632 0.000 0.020 0.972 0.000 0.000 0.008
#> GSM97093 2 0.4800 0.55957 0.024 0.636 0.312 0.012 0.000 0.016
#> GSM97058 2 0.3348 0.78312 0.048 0.852 0.032 0.060 0.000 0.008
#> GSM97051 2 0.6664 0.48655 0.048 0.504 0.240 0.200 0.000 0.008
#> GSM97052 3 0.1036 0.71549 0.000 0.024 0.964 0.004 0.000 0.008
#> GSM97061 3 0.1768 0.70247 0.004 0.044 0.932 0.008 0.000 0.012
Heatmaps for the consensus matrix. It visualizes the probability of two samples to be in a same group.
consensus_heatmap(res, k = 2)
consensus_heatmap(res, k = 3)
consensus_heatmap(res, k = 4)
consensus_heatmap(res, k = 5)
consensus_heatmap(res, k = 6)
Heatmaps for the membership of samples in all partitions to see how consistent they are:
membership_heatmap(res, k = 2)
membership_heatmap(res, k = 3)
membership_heatmap(res, k = 4)
membership_heatmap(res, k = 5)
membership_heatmap(res, k = 6)
As soon as we have had the classes for columns, we can look for signatures which are significantly different between classes which can be candidate marks for certain classes. Following are the heatmaps for signatures.
Signature heatmaps where rows are scaled:
get_signatures(res, k = 2)
get_signatures(res, k = 3)
get_signatures(res, k = 4)
get_signatures(res, k = 5)
get_signatures(res, k = 6)
Signature heatmaps where rows are not scaled:
get_signatures(res, k = 2, scale_rows = FALSE)
get_signatures(res, k = 3, scale_rows = FALSE)
get_signatures(res, k = 4, scale_rows = FALSE)
get_signatures(res, k = 5, scale_rows = FALSE)
get_signatures(res, k = 6, scale_rows = FALSE)
Compare the overlap of signatures from different k:
compare_signatures(res)
get_signature() returns a data frame invisibly. TO get the list of signatures, the function
call should be assigned to a variable explicitly. In following code, if plot argument is set
to FALSE, no heatmap is plotted while only the differential analysis is performed.
# code only for demonstration
tb = get_signature(res, k = ..., plot = FALSE)
An example of the output of tb is:
#> which_row fdr mean_1 mean_2 scaled_mean_1 scaled_mean_2 km
#> 1 38 0.042760348 8.373488 9.131774 -0.5533452 0.5164555 1
#> 2 40 0.018707592 7.106213 8.469186 -0.6173731 0.5762149 1
#> 3 55 0.019134737 10.221463 11.207825 -0.6159697 0.5749050 1
#> 4 59 0.006059896 5.921854 7.869574 -0.6899429 0.6439467 1
#> 5 60 0.018055526 8.928898 10.211722 -0.6204761 0.5791110 1
#> 6 98 0.009384629 15.714769 14.887706 0.6635654 -0.6193277 2
...
The columns in tb are:
which_row: row indices corresponding to the input matrix.fdr: FDR for the differential test. mean_x: The mean value in group x.scaled_mean_x: The mean value in group x after rows are scaled.km: Row groups if k-means clustering is applied to rows.UMAP plot which shows how samples are separated.
dimension_reduction(res, k = 2, method = "UMAP")
dimension_reduction(res, k = 3, method = "UMAP")
dimension_reduction(res, k = 4, method = "UMAP")
dimension_reduction(res, k = 5, method = "UMAP")
dimension_reduction(res, k = 6, method = "UMAP")
Following heatmap shows how subgroups are split when increasing k:
collect_classes(res)
Test correlation between subgroups and known annotations. If the known annotation is numeric, one-way ANOVA test is applied, and if the known annotation is discrete, chi-squared contingency table test is applied.
test_to_known_factors(res)
#> n disease.state(p) specimen(p) cell.type(p) other(p) k
#> SD:kmeans 99 3.34e-04 0.370 3.49e-13 0.1112 2
#> SD:kmeans 94 1.44e-05 0.318 2.74e-13 0.1502 3
#> SD:kmeans 94 6.30e-04 0.308 2.01e-14 0.2307 4
#> SD:kmeans 84 7.86e-05 0.443 2.56e-14 0.0673 5
#> SD:kmeans 66 6.63e-02 0.347 1.49e-10 0.1512 6
If matrix rows can be associated to genes, consider to use functional_enrichment(res,
...) to perform function enrichment for the signature genes. See this vignette for more detailed explanations.
The object with results only for a single top-value method and a single partition method can be extracted as:
res = res_list["SD", "skmeans"]
# you can also extract it by
# res = res_list["SD:skmeans"]
A summary of res and all the functions that can be applied to it:
res
#> A 'ConsensusPartition' object with k = 2, 3, 4, 5, 6.
#> On a matrix with 21512 rows and 100 columns.
#> Top rows (1000, 2000, 3000, 4000, 5000) are extracted by 'SD' method.
#> Subgroups are detected by 'skmeans' method.
#> Performed in total 1250 partitions by row resampling.
#> Best k for subgroups seems to be 2.
#>
#> Following methods can be applied to this 'ConsensusPartition' object:
#> [1] "cola_report" "collect_classes" "collect_plots"
#> [4] "collect_stats" "colnames" "compare_signatures"
#> [7] "consensus_heatmap" "dimension_reduction" "functional_enrichment"
#> [10] "get_anno_col" "get_anno" "get_classes"
#> [13] "get_consensus" "get_matrix" "get_membership"
#> [16] "get_param" "get_signatures" "get_stats"
#> [19] "is_best_k" "is_stable_k" "membership_heatmap"
#> [22] "ncol" "nrow" "plot_ecdf"
#> [25] "rownames" "select_partition_number" "show"
#> [28] "suggest_best_k" "test_to_known_factors"
collect_plots() function collects all the plots made from res for all k (number of partitions)
into one single page to provide an easy and fast comparison between different k.
collect_plots(res)
The plots are:
k and the heatmap of
predicted classes for each k.k.k.k.All the plots in panels can be made by individual functions and they are plotted later in this section.
select_partition_number() produces several plots showing different
statistics for choosing “optimized” k. There are following statistics:
k;k, the area increased is defined as \(A_k - A_{k-1}\).The detailed explanations of these statistics can be found in the cola vignette.
Generally speaking, lower PAC score, higher mean silhouette score or higher
concordance corresponds to better partition. Rand index and Jaccard index
measure how similar the current partition is compared to partition with k-1.
If they are too similar, we won't accept k is better than k-1.
select_partition_number(res)
The numeric values for all these statistics can be obtained by get_stats().
get_stats(res)
#> k 1-PAC mean_silhouette concordance area_increased Rand Jaccard
#> 2 2 0.958 0.944 0.977 0.5000 0.500 0.500
#> 3 3 0.759 0.855 0.930 0.3396 0.728 0.505
#> 4 4 0.760 0.810 0.894 0.1227 0.845 0.576
#> 5 5 0.635 0.574 0.743 0.0609 0.964 0.860
#> 6 6 0.627 0.430 0.611 0.0400 0.918 0.664
suggest_best_k() suggests the best \(k\) based on these statistics. The rules are as follows:
suggest_best_k(res)
#> [1] 2
Following shows the table of the partitions (You need to click the show/hide
code output link to see it). The membership matrix (columns with name p*)
is inferred by
clue::cl_consensus()
function with the SE method. Basically the value in the membership matrix
represents the probability to belong to a certain group. The finall class
label for an item is determined with the group with highest probability it
belongs to.
In get_classes() function, the entropy is calculated from the membership
matrix and the silhouette score is calculated from the consensus matrix.
cbind(get_classes(res, k = 2), get_membership(res, k = 2))
#> class entropy silhouette p1 p2
#> GSM97138 1 0.0000 0.971 1.000 0.000
#> GSM97145 1 0.0000 0.971 1.000 0.000
#> GSM97147 1 0.0000 0.971 1.000 0.000
#> GSM97125 1 0.0000 0.971 1.000 0.000
#> GSM97127 1 0.0000 0.971 1.000 0.000
#> GSM97130 1 0.0000 0.971 1.000 0.000
#> GSM97133 1 0.0000 0.971 1.000 0.000
#> GSM97134 1 0.0000 0.971 1.000 0.000
#> GSM97120 1 0.0000 0.971 1.000 0.000
#> GSM97126 1 0.0000 0.971 1.000 0.000
#> GSM97112 1 0.0000 0.971 1.000 0.000
#> GSM97115 1 0.0000 0.971 1.000 0.000
#> GSM97116 1 0.0000 0.971 1.000 0.000
#> GSM97117 2 0.0000 0.979 0.000 1.000
#> GSM97119 1 0.0000 0.971 1.000 0.000
#> GSM97122 1 0.0000 0.971 1.000 0.000
#> GSM97135 1 0.0000 0.971 1.000 0.000
#> GSM97136 2 0.1843 0.958 0.028 0.972
#> GSM97139 1 0.0000 0.971 1.000 0.000
#> GSM97146 1 0.0000 0.971 1.000 0.000
#> GSM97123 2 0.0000 0.979 0.000 1.000
#> GSM97129 2 0.2236 0.949 0.036 0.964
#> GSM97143 1 0.0000 0.971 1.000 0.000
#> GSM97113 2 0.5294 0.861 0.120 0.880
#> GSM97056 1 0.0000 0.971 1.000 0.000
#> GSM97124 1 0.0000 0.971 1.000 0.000
#> GSM97132 1 0.0000 0.971 1.000 0.000
#> GSM97144 1 0.0000 0.971 1.000 0.000
#> GSM97149 1 0.0000 0.971 1.000 0.000
#> GSM97068 1 0.9044 0.517 0.680 0.320
#> GSM97071 2 0.6247 0.809 0.156 0.844
#> GSM97086 2 0.0000 0.979 0.000 1.000
#> GSM97103 2 0.0000 0.979 0.000 1.000
#> GSM97057 2 0.7299 0.746 0.204 0.796
#> GSM97060 2 0.0000 0.979 0.000 1.000
#> GSM97075 2 0.0000 0.979 0.000 1.000
#> GSM97098 2 0.0000 0.979 0.000 1.000
#> GSM97099 2 0.0000 0.979 0.000 1.000
#> GSM97101 2 0.0000 0.979 0.000 1.000
#> GSM97105 2 0.0000 0.979 0.000 1.000
#> GSM97106 2 0.0000 0.979 0.000 1.000
#> GSM97121 2 0.0000 0.979 0.000 1.000
#> GSM97128 1 0.2603 0.932 0.956 0.044
#> GSM97131 2 0.0000 0.979 0.000 1.000
#> GSM97137 1 0.0000 0.971 1.000 0.000
#> GSM97118 1 0.0000 0.971 1.000 0.000
#> GSM97114 2 0.9635 0.366 0.388 0.612
#> GSM97142 1 0.0000 0.971 1.000 0.000
#> GSM97140 2 0.0000 0.979 0.000 1.000
#> GSM97141 2 0.0000 0.979 0.000 1.000
#> GSM97055 1 0.0000 0.971 1.000 0.000
#> GSM97090 1 0.0000 0.971 1.000 0.000
#> GSM97091 1 0.0000 0.971 1.000 0.000
#> GSM97148 1 0.0000 0.971 1.000 0.000
#> GSM97063 1 0.0000 0.971 1.000 0.000
#> GSM97053 1 0.0000 0.971 1.000 0.000
#> GSM97066 2 0.0000 0.979 0.000 1.000
#> GSM97079 2 0.0000 0.979 0.000 1.000
#> GSM97083 1 0.0000 0.971 1.000 0.000
#> GSM97084 2 0.4298 0.898 0.088 0.912
#> GSM97094 1 0.0000 0.971 1.000 0.000
#> GSM97096 2 0.0000 0.979 0.000 1.000
#> GSM97097 2 0.0000 0.979 0.000 1.000
#> GSM97107 1 0.0376 0.968 0.996 0.004
#> GSM97054 2 0.0000 0.979 0.000 1.000
#> GSM97062 2 0.0000 0.979 0.000 1.000
#> GSM97069 2 0.0000 0.979 0.000 1.000
#> GSM97070 2 0.0000 0.979 0.000 1.000
#> GSM97073 2 0.0000 0.979 0.000 1.000
#> GSM97076 1 0.0000 0.971 1.000 0.000
#> GSM97077 2 0.0000 0.979 0.000 1.000
#> GSM97095 1 0.0000 0.971 1.000 0.000
#> GSM97102 2 0.0000 0.979 0.000 1.000
#> GSM97109 2 0.2603 0.942 0.044 0.956
#> GSM97110 2 0.0376 0.976 0.004 0.996
#> GSM97074 1 0.0000 0.971 1.000 0.000
#> GSM97085 1 0.9977 0.121 0.528 0.472
#> GSM97059 1 0.0000 0.971 1.000 0.000
#> GSM97072 2 0.0000 0.979 0.000 1.000
#> GSM97078 1 0.2236 0.939 0.964 0.036
#> GSM97067 2 0.0000 0.979 0.000 1.000
#> GSM97087 2 0.0000 0.979 0.000 1.000
#> GSM97111 2 0.0000 0.979 0.000 1.000
#> GSM97064 2 0.0000 0.979 0.000 1.000
#> GSM97065 2 0.0000 0.979 0.000 1.000
#> GSM97081 2 0.0000 0.979 0.000 1.000
#> GSM97082 2 0.0000 0.979 0.000 1.000
#> GSM97088 1 0.9358 0.464 0.648 0.352
#> GSM97100 2 0.0000 0.979 0.000 1.000
#> GSM97104 2 0.0000 0.979 0.000 1.000
#> GSM97108 2 0.0000 0.979 0.000 1.000
#> GSM97050 2 0.0000 0.979 0.000 1.000
#> GSM97080 2 0.0000 0.979 0.000 1.000
#> GSM97089 2 0.0000 0.979 0.000 1.000
#> GSM97092 2 0.0000 0.979 0.000 1.000
#> GSM97093 2 0.0000 0.979 0.000 1.000
#> GSM97058 2 0.0000 0.979 0.000 1.000
#> GSM97051 2 0.0000 0.979 0.000 1.000
#> GSM97052 2 0.0000 0.979 0.000 1.000
#> GSM97061 2 0.0000 0.979 0.000 1.000
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM97138 1 0.0000 0.970 1.000 0.000 0.000
#> GSM97145 1 0.0000 0.970 1.000 0.000 0.000
#> GSM97147 2 0.1031 0.870 0.024 0.976 0.000
#> GSM97125 1 0.0000 0.970 1.000 0.000 0.000
#> GSM97127 1 0.0000 0.970 1.000 0.000 0.000
#> GSM97130 1 0.1031 0.958 0.976 0.024 0.000
#> GSM97133 1 0.0000 0.970 1.000 0.000 0.000
#> GSM97134 1 0.1031 0.958 0.976 0.024 0.000
#> GSM97120 1 0.0000 0.970 1.000 0.000 0.000
#> GSM97126 1 0.0000 0.970 1.000 0.000 0.000
#> GSM97112 1 0.0000 0.970 1.000 0.000 0.000
#> GSM97115 2 0.6008 0.396 0.372 0.628 0.000
#> GSM97116 1 0.0000 0.970 1.000 0.000 0.000
#> GSM97117 3 0.6267 0.170 0.000 0.452 0.548
#> GSM97119 1 0.0000 0.970 1.000 0.000 0.000
#> GSM97122 1 0.0000 0.970 1.000 0.000 0.000
#> GSM97135 1 0.0000 0.970 1.000 0.000 0.000
#> GSM97136 3 0.1289 0.887 0.032 0.000 0.968
#> GSM97139 1 0.0000 0.970 1.000 0.000 0.000
#> GSM97146 1 0.0000 0.970 1.000 0.000 0.000
#> GSM97123 3 0.4291 0.734 0.000 0.180 0.820
#> GSM97129 3 0.7677 0.581 0.120 0.204 0.676
#> GSM97143 1 0.0000 0.970 1.000 0.000 0.000
#> GSM97113 2 0.1315 0.877 0.008 0.972 0.020
#> GSM97056 1 0.0424 0.967 0.992 0.008 0.000
#> GSM97124 1 0.0000 0.970 1.000 0.000 0.000
#> GSM97132 1 0.0000 0.970 1.000 0.000 0.000
#> GSM97144 1 0.1031 0.958 0.976 0.024 0.000
#> GSM97149 1 0.0000 0.970 1.000 0.000 0.000
#> GSM97068 2 0.0000 0.876 0.000 1.000 0.000
#> GSM97071 3 0.1289 0.892 0.000 0.032 0.968
#> GSM97086 2 0.3192 0.841 0.000 0.888 0.112
#> GSM97103 3 0.2066 0.872 0.000 0.060 0.940
#> GSM97057 2 0.0237 0.877 0.000 0.996 0.004
#> GSM97060 3 0.0000 0.908 0.000 0.000 1.000
#> GSM97075 2 0.6274 0.151 0.000 0.544 0.456
#> GSM97098 3 0.1753 0.881 0.000 0.048 0.952
#> GSM97099 2 0.3551 0.816 0.000 0.868 0.132
#> GSM97101 2 0.1031 0.877 0.000 0.976 0.024
#> GSM97105 2 0.0592 0.878 0.000 0.988 0.012
#> GSM97106 3 0.1411 0.890 0.000 0.036 0.964
#> GSM97121 2 0.0592 0.878 0.000 0.988 0.012
#> GSM97128 3 0.5356 0.716 0.196 0.020 0.784
#> GSM97131 2 0.4605 0.779 0.000 0.796 0.204
#> GSM97137 1 0.0237 0.969 0.996 0.004 0.000
#> GSM97118 1 0.0000 0.970 1.000 0.000 0.000
#> GSM97114 2 0.1289 0.869 0.032 0.968 0.000
#> GSM97142 1 0.0000 0.970 1.000 0.000 0.000
#> GSM97140 2 0.0000 0.876 0.000 1.000 0.000
#> GSM97141 2 0.1031 0.877 0.000 0.976 0.024
#> GSM97055 1 0.0892 0.957 0.980 0.000 0.020
#> GSM97090 1 0.3816 0.828 0.852 0.148 0.000
#> GSM97091 1 0.0000 0.970 1.000 0.000 0.000
#> GSM97148 1 0.0000 0.970 1.000 0.000 0.000
#> GSM97063 1 0.0000 0.970 1.000 0.000 0.000
#> GSM97053 1 0.0000 0.970 1.000 0.000 0.000
#> GSM97066 3 0.0000 0.908 0.000 0.000 1.000
#> GSM97079 2 0.4504 0.779 0.000 0.804 0.196
#> GSM97083 1 0.1267 0.956 0.972 0.024 0.004
#> GSM97084 2 0.4346 0.790 0.000 0.816 0.184
#> GSM97094 1 0.1529 0.946 0.960 0.040 0.000
#> GSM97096 3 0.0424 0.905 0.000 0.008 0.992
#> GSM97097 2 0.5529 0.642 0.000 0.704 0.296
#> GSM97107 1 0.4602 0.830 0.852 0.040 0.108
#> GSM97054 2 0.2878 0.849 0.000 0.904 0.096
#> GSM97062 2 0.4399 0.786 0.000 0.812 0.188
#> GSM97069 3 0.0000 0.908 0.000 0.000 1.000
#> GSM97070 3 0.0000 0.908 0.000 0.000 1.000
#> GSM97073 3 0.0000 0.908 0.000 0.000 1.000
#> GSM97076 1 0.3784 0.835 0.864 0.004 0.132
#> GSM97077 2 0.0237 0.877 0.000 0.996 0.004
#> GSM97095 1 0.6180 0.291 0.584 0.416 0.000
#> GSM97102 3 0.0000 0.908 0.000 0.000 1.000
#> GSM97109 2 0.1315 0.874 0.020 0.972 0.008
#> GSM97110 2 0.1643 0.872 0.000 0.956 0.044
#> GSM97074 3 0.5517 0.627 0.268 0.004 0.728
#> GSM97085 3 0.0000 0.908 0.000 0.000 1.000
#> GSM97059 2 0.0592 0.873 0.012 0.988 0.000
#> GSM97072 3 0.0000 0.908 0.000 0.000 1.000
#> GSM97078 3 0.6796 0.459 0.344 0.024 0.632
#> GSM97067 3 0.0000 0.908 0.000 0.000 1.000
#> GSM97087 3 0.0000 0.908 0.000 0.000 1.000
#> GSM97111 2 0.2066 0.865 0.000 0.940 0.060
#> GSM97064 2 0.4555 0.779 0.000 0.800 0.200
#> GSM97065 3 0.5785 0.490 0.000 0.332 0.668
#> GSM97081 3 0.0000 0.908 0.000 0.000 1.000
#> GSM97082 3 0.0000 0.908 0.000 0.000 1.000
#> GSM97088 3 0.0892 0.898 0.000 0.020 0.980
#> GSM97100 2 0.0000 0.876 0.000 1.000 0.000
#> GSM97104 3 0.0000 0.908 0.000 0.000 1.000
#> GSM97108 2 0.0424 0.878 0.000 0.992 0.008
#> GSM97050 2 0.4346 0.800 0.000 0.816 0.184
#> GSM97080 3 0.0000 0.908 0.000 0.000 1.000
#> GSM97089 3 0.0000 0.908 0.000 0.000 1.000
#> GSM97092 3 0.0000 0.908 0.000 0.000 1.000
#> GSM97093 2 0.5760 0.557 0.000 0.672 0.328
#> GSM97058 2 0.1031 0.877 0.000 0.976 0.024
#> GSM97051 2 0.4452 0.786 0.000 0.808 0.192
#> GSM97052 3 0.0237 0.907 0.000 0.004 0.996
#> GSM97061 3 0.2878 0.838 0.000 0.096 0.904
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM97138 1 0.0188 0.9410 0.996 0.000 0.000 0.004
#> GSM97145 1 0.0188 0.9399 0.996 0.004 0.000 0.000
#> GSM97147 2 0.3117 0.7899 0.028 0.880 0.000 0.092
#> GSM97125 1 0.0817 0.9437 0.976 0.000 0.000 0.024
#> GSM97127 1 0.0000 0.9410 1.000 0.000 0.000 0.000
#> GSM97130 4 0.3123 0.7818 0.156 0.000 0.000 0.844
#> GSM97133 1 0.0188 0.9410 0.996 0.000 0.000 0.004
#> GSM97134 4 0.3801 0.6918 0.220 0.000 0.000 0.780
#> GSM97120 1 0.0376 0.9391 0.992 0.004 0.000 0.004
#> GSM97126 1 0.1284 0.9321 0.964 0.024 0.000 0.012
#> GSM97112 1 0.1302 0.9436 0.956 0.000 0.000 0.044
#> GSM97115 4 0.2413 0.8511 0.020 0.064 0.000 0.916
#> GSM97116 1 0.0188 0.9410 0.996 0.000 0.000 0.004
#> GSM97117 2 0.2450 0.8109 0.016 0.912 0.072 0.000
#> GSM97119 1 0.1302 0.9436 0.956 0.000 0.000 0.044
#> GSM97122 1 0.1302 0.9436 0.956 0.000 0.000 0.044
#> GSM97135 1 0.1302 0.9436 0.956 0.000 0.000 0.044
#> GSM97136 3 0.2174 0.8850 0.020 0.052 0.928 0.000
#> GSM97139 1 0.0188 0.9410 0.996 0.000 0.000 0.004
#> GSM97146 1 0.0188 0.9410 0.996 0.000 0.000 0.004
#> GSM97123 3 0.4807 0.6289 0.000 0.248 0.728 0.024
#> GSM97129 2 0.7386 0.2732 0.148 0.484 0.364 0.004
#> GSM97143 1 0.1302 0.9436 0.956 0.000 0.000 0.044
#> GSM97113 2 0.1677 0.8185 0.040 0.948 0.000 0.012
#> GSM97056 1 0.4193 0.6629 0.732 0.000 0.000 0.268
#> GSM97124 1 0.1302 0.9436 0.956 0.000 0.000 0.044
#> GSM97132 1 0.3219 0.8424 0.836 0.000 0.000 0.164
#> GSM97144 4 0.1940 0.8395 0.076 0.000 0.000 0.924
#> GSM97149 1 0.0524 0.9372 0.988 0.008 0.000 0.004
#> GSM97068 4 0.4898 0.2552 0.000 0.416 0.000 0.584
#> GSM97071 4 0.2216 0.8296 0.000 0.000 0.092 0.908
#> GSM97086 4 0.2814 0.7895 0.000 0.132 0.000 0.868
#> GSM97103 3 0.3919 0.8247 0.000 0.104 0.840 0.056
#> GSM97057 2 0.1824 0.8161 0.004 0.936 0.000 0.060
#> GSM97060 3 0.0336 0.9121 0.000 0.000 0.992 0.008
#> GSM97075 2 0.3402 0.7686 0.000 0.832 0.164 0.004
#> GSM97098 3 0.3711 0.8053 0.000 0.140 0.836 0.024
#> GSM97099 2 0.2586 0.8125 0.012 0.912 0.068 0.008
#> GSM97101 2 0.0524 0.8222 0.004 0.988 0.000 0.008
#> GSM97105 2 0.1302 0.8196 0.000 0.956 0.000 0.044
#> GSM97106 3 0.2224 0.8854 0.000 0.032 0.928 0.040
#> GSM97121 2 0.0592 0.8219 0.000 0.984 0.000 0.016
#> GSM97128 4 0.4927 0.6205 0.024 0.000 0.264 0.712
#> GSM97131 2 0.6555 0.6094 0.000 0.632 0.212 0.156
#> GSM97137 1 0.3975 0.6953 0.760 0.000 0.000 0.240
#> GSM97118 1 0.2921 0.8687 0.860 0.000 0.000 0.140
#> GSM97114 2 0.1716 0.8091 0.064 0.936 0.000 0.000
#> GSM97142 1 0.1302 0.9436 0.956 0.000 0.000 0.044
#> GSM97140 2 0.1792 0.8124 0.000 0.932 0.000 0.068
#> GSM97141 2 0.1042 0.8220 0.020 0.972 0.008 0.000
#> GSM97055 1 0.2002 0.9342 0.936 0.000 0.020 0.044
#> GSM97090 4 0.2670 0.8553 0.052 0.040 0.000 0.908
#> GSM97091 1 0.1302 0.9436 0.956 0.000 0.000 0.044
#> GSM97148 1 0.0188 0.9410 0.996 0.000 0.000 0.004
#> GSM97063 1 0.1302 0.9436 0.956 0.000 0.000 0.044
#> GSM97053 1 0.1302 0.9436 0.956 0.000 0.000 0.044
#> GSM97066 3 0.0000 0.9137 0.000 0.000 1.000 0.000
#> GSM97079 4 0.1743 0.8454 0.000 0.056 0.004 0.940
#> GSM97083 4 0.2271 0.8393 0.076 0.000 0.008 0.916
#> GSM97084 4 0.1389 0.8473 0.000 0.048 0.000 0.952
#> GSM97094 4 0.1004 0.8547 0.024 0.004 0.000 0.972
#> GSM97096 3 0.2282 0.8838 0.000 0.052 0.924 0.024
#> GSM97097 4 0.5532 0.6129 0.000 0.068 0.228 0.704
#> GSM97107 4 0.1109 0.8539 0.028 0.004 0.000 0.968
#> GSM97054 4 0.2011 0.8354 0.000 0.080 0.000 0.920
#> GSM97062 4 0.1661 0.8462 0.000 0.052 0.004 0.944
#> GSM97069 3 0.0000 0.9137 0.000 0.000 1.000 0.000
#> GSM97070 3 0.0000 0.9137 0.000 0.000 1.000 0.000
#> GSM97073 3 0.0188 0.9131 0.000 0.004 0.996 0.000
#> GSM97076 1 0.3860 0.8246 0.852 0.012 0.104 0.032
#> GSM97077 2 0.2401 0.8037 0.000 0.904 0.004 0.092
#> GSM97095 4 0.2892 0.8509 0.036 0.068 0.000 0.896
#> GSM97102 3 0.0188 0.9130 0.000 0.004 0.996 0.000
#> GSM97109 2 0.2522 0.8108 0.052 0.920 0.012 0.016
#> GSM97110 2 0.2895 0.8095 0.044 0.908 0.032 0.016
#> GSM97074 3 0.6869 0.3879 0.180 0.000 0.596 0.224
#> GSM97085 3 0.0707 0.9031 0.000 0.000 0.980 0.020
#> GSM97059 2 0.4134 0.6225 0.000 0.740 0.000 0.260
#> GSM97072 3 0.0000 0.9137 0.000 0.000 1.000 0.000
#> GSM97078 4 0.3224 0.8092 0.016 0.000 0.120 0.864
#> GSM97067 3 0.0000 0.9137 0.000 0.000 1.000 0.000
#> GSM97087 3 0.0000 0.9137 0.000 0.000 1.000 0.000
#> GSM97111 2 0.1576 0.8209 0.000 0.948 0.048 0.004
#> GSM97064 2 0.6079 0.4095 0.000 0.568 0.380 0.052
#> GSM97065 2 0.5302 0.6680 0.044 0.720 0.232 0.004
#> GSM97081 3 0.2216 0.8621 0.000 0.092 0.908 0.000
#> GSM97082 3 0.0000 0.9137 0.000 0.000 1.000 0.000
#> GSM97088 3 0.4992 0.0232 0.000 0.000 0.524 0.476
#> GSM97100 2 0.2647 0.7873 0.000 0.880 0.000 0.120
#> GSM97104 3 0.0000 0.9137 0.000 0.000 1.000 0.000
#> GSM97108 2 0.0921 0.8214 0.000 0.972 0.000 0.028
#> GSM97050 2 0.5952 0.6753 0.000 0.692 0.184 0.124
#> GSM97080 3 0.0000 0.9137 0.000 0.000 1.000 0.000
#> GSM97089 3 0.0000 0.9137 0.000 0.000 1.000 0.000
#> GSM97092 3 0.0657 0.9100 0.000 0.012 0.984 0.004
#> GSM97093 2 0.6077 0.1444 0.000 0.496 0.460 0.044
#> GSM97058 2 0.2101 0.8171 0.000 0.928 0.012 0.060
#> GSM97051 2 0.7390 0.4152 0.000 0.512 0.204 0.284
#> GSM97052 3 0.1042 0.9054 0.000 0.020 0.972 0.008
#> GSM97061 3 0.3731 0.8032 0.000 0.120 0.844 0.036
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM97138 1 0.0880 0.7324 0.968 0.000 0.000 0.000 0.032
#> GSM97145 1 0.3530 0.7741 0.784 0.012 0.000 0.000 0.204
#> GSM97147 2 0.4776 0.6171 0.076 0.776 0.000 0.100 0.048
#> GSM97125 1 0.3579 0.7806 0.756 0.000 0.000 0.004 0.240
#> GSM97127 1 0.1851 0.7514 0.912 0.000 0.000 0.000 0.088
#> GSM97130 4 0.4877 0.5872 0.236 0.000 0.000 0.692 0.072
#> GSM97133 1 0.0162 0.7172 0.996 0.000 0.000 0.000 0.004
#> GSM97134 4 0.5952 0.3616 0.136 0.000 0.000 0.560 0.304
#> GSM97120 1 0.0324 0.7174 0.992 0.004 0.000 0.000 0.004
#> GSM97126 1 0.4666 0.7146 0.676 0.040 0.000 0.000 0.284
#> GSM97112 1 0.3838 0.7765 0.716 0.000 0.000 0.004 0.280
#> GSM97115 4 0.4480 0.6687 0.124 0.040 0.000 0.788 0.048
#> GSM97116 1 0.0162 0.7172 0.996 0.000 0.000 0.000 0.004
#> GSM97117 2 0.3996 0.6190 0.012 0.800 0.040 0.000 0.148
#> GSM97119 1 0.3861 0.7747 0.712 0.000 0.000 0.004 0.284
#> GSM97122 1 0.3838 0.7765 0.716 0.000 0.000 0.004 0.280
#> GSM97135 1 0.3838 0.7765 0.716 0.000 0.000 0.004 0.280
#> GSM97136 3 0.6063 0.2558 0.032 0.064 0.564 0.000 0.340
#> GSM97139 1 0.0000 0.7194 1.000 0.000 0.000 0.000 0.000
#> GSM97146 1 0.0162 0.7172 0.996 0.000 0.000 0.000 0.004
#> GSM97123 3 0.5114 0.4967 0.000 0.188 0.720 0.024 0.068
#> GSM97129 5 0.8278 0.1567 0.092 0.308 0.192 0.016 0.392
#> GSM97143 1 0.3814 0.7776 0.720 0.000 0.000 0.004 0.276
#> GSM97113 2 0.5646 0.5895 0.144 0.700 0.020 0.008 0.128
#> GSM97056 1 0.3807 0.4138 0.748 0.000 0.000 0.240 0.012
#> GSM97124 1 0.3838 0.7765 0.716 0.000 0.000 0.004 0.280
#> GSM97132 1 0.6269 0.5173 0.508 0.000 0.000 0.168 0.324
#> GSM97144 4 0.3888 0.6520 0.056 0.000 0.000 0.796 0.148
#> GSM97149 1 0.0451 0.7114 0.988 0.004 0.000 0.000 0.008
#> GSM97068 4 0.6949 0.3277 0.160 0.280 0.000 0.520 0.040
#> GSM97071 4 0.4891 0.5499 0.000 0.004 0.068 0.704 0.224
#> GSM97086 4 0.4314 0.6079 0.000 0.124 0.004 0.780 0.092
#> GSM97103 3 0.6951 0.4151 0.000 0.140 0.580 0.084 0.196
#> GSM97057 2 0.6040 0.5819 0.152 0.688 0.008 0.088 0.064
#> GSM97060 3 0.1571 0.6832 0.000 0.000 0.936 0.004 0.060
#> GSM97075 2 0.4968 0.5576 0.000 0.712 0.136 0.000 0.152
#> GSM97098 3 0.5739 0.4913 0.000 0.148 0.664 0.016 0.172
#> GSM97099 2 0.4708 0.5797 0.012 0.744 0.048 0.004 0.192
#> GSM97101 2 0.1990 0.6598 0.008 0.920 0.000 0.004 0.068
#> GSM97105 2 0.3268 0.6578 0.000 0.856 0.004 0.080 0.060
#> GSM97106 3 0.3327 0.6391 0.000 0.028 0.864 0.036 0.072
#> GSM97121 2 0.2124 0.6702 0.000 0.916 0.000 0.028 0.056
#> GSM97128 4 0.6520 0.0915 0.008 0.000 0.148 0.432 0.412
#> GSM97131 2 0.7767 0.3239 0.000 0.472 0.232 0.184 0.112
#> GSM97137 1 0.4404 0.3451 0.704 0.000 0.000 0.264 0.032
#> GSM97118 1 0.6055 0.4694 0.472 0.000 0.000 0.120 0.408
#> GSM97114 2 0.4357 0.6019 0.104 0.768 0.000 0.000 0.128
#> GSM97142 1 0.3838 0.7765 0.716 0.000 0.000 0.004 0.280
#> GSM97140 2 0.3260 0.6578 0.000 0.856 0.004 0.084 0.056
#> GSM97141 2 0.2733 0.6446 0.012 0.872 0.004 0.000 0.112
#> GSM97055 1 0.4419 0.7204 0.644 0.000 0.008 0.004 0.344
#> GSM97090 4 0.3846 0.6777 0.128 0.004 0.004 0.816 0.048
#> GSM97091 1 0.4047 0.7524 0.676 0.000 0.000 0.004 0.320
#> GSM97148 1 0.0162 0.7172 0.996 0.000 0.000 0.000 0.004
#> GSM97063 1 0.3906 0.7708 0.704 0.000 0.000 0.004 0.292
#> GSM97053 1 0.3607 0.7819 0.752 0.000 0.000 0.004 0.244
#> GSM97066 3 0.3561 0.5913 0.000 0.000 0.740 0.000 0.260
#> GSM97079 4 0.3619 0.6524 0.000 0.040 0.008 0.828 0.124
#> GSM97083 4 0.4398 0.5922 0.040 0.000 0.000 0.720 0.240
#> GSM97084 4 0.1740 0.6943 0.000 0.012 0.000 0.932 0.056
#> GSM97094 4 0.2284 0.6948 0.004 0.004 0.000 0.896 0.096
#> GSM97096 3 0.3831 0.6317 0.000 0.044 0.812 0.008 0.136
#> GSM97097 4 0.6077 0.4669 0.000 0.040 0.152 0.656 0.152
#> GSM97107 4 0.1740 0.6962 0.012 0.000 0.000 0.932 0.056
#> GSM97054 4 0.3323 0.6667 0.000 0.100 0.000 0.844 0.056
#> GSM97062 4 0.2293 0.6826 0.000 0.016 0.000 0.900 0.084
#> GSM97069 3 0.3274 0.6196 0.000 0.000 0.780 0.000 0.220
#> GSM97070 3 0.3612 0.5886 0.000 0.000 0.732 0.000 0.268
#> GSM97073 3 0.3857 0.5636 0.000 0.000 0.688 0.000 0.312
#> GSM97076 5 0.7455 0.4241 0.328 0.040 0.112 0.032 0.488
#> GSM97077 2 0.5314 0.6121 0.000 0.736 0.056 0.120 0.088
#> GSM97095 4 0.5867 0.6503 0.124 0.068 0.008 0.708 0.092
#> GSM97102 3 0.2612 0.6740 0.000 0.008 0.868 0.000 0.124
#> GSM97109 2 0.5504 0.5340 0.092 0.680 0.004 0.012 0.212
#> GSM97110 2 0.6004 0.5063 0.084 0.644 0.016 0.016 0.240
#> GSM97074 5 0.6017 0.3339 0.024 0.000 0.260 0.100 0.616
#> GSM97085 3 0.4009 0.5124 0.000 0.000 0.684 0.004 0.312
#> GSM97059 2 0.6981 0.3216 0.148 0.524 0.000 0.280 0.048
#> GSM97072 3 0.3480 0.6106 0.000 0.000 0.752 0.000 0.248
#> GSM97078 4 0.5464 0.4281 0.004 0.000 0.068 0.596 0.332
#> GSM97067 3 0.3612 0.5896 0.000 0.000 0.732 0.000 0.268
#> GSM97087 3 0.0324 0.6781 0.000 0.004 0.992 0.000 0.004
#> GSM97111 2 0.3461 0.6311 0.004 0.812 0.016 0.000 0.168
#> GSM97064 3 0.6931 -0.0706 0.000 0.396 0.452 0.064 0.088
#> GSM97065 2 0.7744 -0.0838 0.076 0.388 0.160 0.004 0.372
#> GSM97081 3 0.3962 0.6164 0.000 0.112 0.800 0.000 0.088
#> GSM97082 3 0.1410 0.6805 0.000 0.000 0.940 0.000 0.060
#> GSM97088 3 0.6769 -0.2083 0.000 0.000 0.396 0.316 0.288
#> GSM97100 2 0.4854 0.5919 0.000 0.724 0.004 0.184 0.088
#> GSM97104 3 0.1851 0.6756 0.000 0.000 0.912 0.000 0.088
#> GSM97108 2 0.1668 0.6704 0.000 0.940 0.000 0.028 0.032
#> GSM97050 2 0.7703 0.3240 0.000 0.468 0.268 0.152 0.112
#> GSM97080 3 0.2690 0.6543 0.000 0.000 0.844 0.000 0.156
#> GSM97089 3 0.0854 0.6757 0.000 0.008 0.976 0.004 0.012
#> GSM97092 3 0.1560 0.6693 0.000 0.020 0.948 0.004 0.028
#> GSM97093 3 0.7012 0.0607 0.000 0.344 0.488 0.064 0.104
#> GSM97058 2 0.4946 0.6299 0.000 0.768 0.072 0.084 0.076
#> GSM97051 2 0.8147 0.2295 0.000 0.376 0.228 0.280 0.116
#> GSM97052 3 0.2352 0.6545 0.000 0.032 0.912 0.008 0.048
#> GSM97061 3 0.4218 0.5737 0.000 0.112 0.804 0.024 0.060
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM97138 1 0.1918 0.6096 0.904 0.008 0.000 0.000 0.088 0.000
#> GSM97145 1 0.3855 0.6037 0.704 0.016 0.000 0.000 0.276 0.004
#> GSM97147 6 0.6226 0.1447 0.080 0.388 0.000 0.060 0.004 0.468
#> GSM97125 1 0.3742 0.5939 0.648 0.000 0.000 0.000 0.348 0.004
#> GSM97127 1 0.3056 0.6128 0.804 0.008 0.000 0.000 0.184 0.004
#> GSM97130 4 0.5753 0.4275 0.312 0.000 0.000 0.560 0.088 0.040
#> GSM97133 1 0.0508 0.5918 0.984 0.012 0.000 0.004 0.000 0.000
#> GSM97134 4 0.6148 -0.0167 0.088 0.000 0.000 0.436 0.420 0.056
#> GSM97120 1 0.1003 0.6007 0.964 0.016 0.000 0.000 0.020 0.000
#> GSM97126 1 0.6220 0.2911 0.444 0.100 0.000 0.008 0.412 0.036
#> GSM97112 1 0.3966 0.5447 0.552 0.000 0.000 0.000 0.444 0.004
#> GSM97115 4 0.6476 0.5331 0.168 0.016 0.000 0.588 0.076 0.152
#> GSM97116 1 0.0551 0.5956 0.984 0.008 0.000 0.004 0.004 0.000
#> GSM97117 2 0.3003 0.5826 0.000 0.852 0.028 0.000 0.016 0.104
#> GSM97119 1 0.3966 0.5461 0.552 0.000 0.000 0.000 0.444 0.004
#> GSM97122 1 0.3950 0.5542 0.564 0.000 0.000 0.000 0.432 0.004
#> GSM97135 1 0.3937 0.5598 0.572 0.000 0.000 0.000 0.424 0.004
#> GSM97136 3 0.7415 0.2445 0.012 0.192 0.432 0.008 0.276 0.080
#> GSM97139 1 0.0520 0.5987 0.984 0.008 0.000 0.000 0.008 0.000
#> GSM97146 1 0.0653 0.5896 0.980 0.012 0.000 0.004 0.004 0.000
#> GSM97123 3 0.5853 0.5081 0.000 0.068 0.608 0.012 0.056 0.256
#> GSM97129 2 0.8954 0.0612 0.104 0.312 0.112 0.040 0.296 0.136
#> GSM97143 1 0.4111 0.5301 0.536 0.000 0.000 0.004 0.456 0.004
#> GSM97113 2 0.5977 0.3679 0.216 0.596 0.008 0.000 0.032 0.148
#> GSM97056 1 0.4496 0.3228 0.744 0.004 0.000 0.168 0.052 0.032
#> GSM97124 1 0.4289 0.5339 0.540 0.000 0.000 0.012 0.444 0.004
#> GSM97132 5 0.6277 -0.1318 0.364 0.000 0.000 0.144 0.456 0.036
#> GSM97144 4 0.4683 0.5167 0.052 0.000 0.000 0.724 0.176 0.048
#> GSM97149 1 0.0964 0.5803 0.968 0.012 0.000 0.004 0.016 0.000
#> GSM97068 4 0.7717 0.1544 0.184 0.068 0.000 0.348 0.052 0.348
#> GSM97071 4 0.5991 0.3244 0.000 0.004 0.128 0.608 0.204 0.056
#> GSM97086 4 0.3878 0.4249 0.000 0.004 0.000 0.668 0.008 0.320
#> GSM97103 3 0.7335 0.3974 0.000 0.236 0.488 0.160 0.056 0.060
#> GSM97057 6 0.7099 0.0888 0.236 0.328 0.000 0.036 0.020 0.380
#> GSM97060 3 0.2842 0.7106 0.000 0.008 0.868 0.008 0.024 0.092
#> GSM97075 2 0.5470 0.4296 0.004 0.648 0.116 0.004 0.020 0.208
#> GSM97098 3 0.6224 0.4661 0.000 0.320 0.532 0.012 0.052 0.084
#> GSM97099 2 0.3024 0.5660 0.008 0.880 0.028 0.012 0.028 0.044
#> GSM97101 2 0.3104 0.5124 0.000 0.788 0.000 0.004 0.004 0.204
#> GSM97105 6 0.4517 0.1367 0.000 0.444 0.000 0.032 0.000 0.524
#> GSM97106 3 0.5627 0.6196 0.000 0.028 0.676 0.060 0.060 0.176
#> GSM97121 2 0.4325 0.1144 0.000 0.568 0.000 0.016 0.004 0.412
#> GSM97128 5 0.6366 0.1586 0.000 0.000 0.112 0.292 0.520 0.076
#> GSM97131 6 0.6897 0.4273 0.000 0.136 0.112 0.184 0.020 0.548
#> GSM97137 1 0.5146 0.1191 0.644 0.000 0.000 0.260 0.060 0.036
#> GSM97118 5 0.5573 0.2205 0.200 0.000 0.000 0.112 0.640 0.048
#> GSM97114 2 0.3857 0.5549 0.092 0.788 0.000 0.000 0.008 0.112
#> GSM97142 1 0.3961 0.5488 0.556 0.000 0.000 0.000 0.440 0.004
#> GSM97140 6 0.4356 0.1364 0.004 0.432 0.000 0.016 0.000 0.548
#> GSM97141 2 0.2703 0.5422 0.000 0.824 0.000 0.000 0.004 0.172
#> GSM97055 5 0.5364 -0.3603 0.424 0.028 0.028 0.000 0.508 0.012
#> GSM97090 4 0.5701 0.5401 0.188 0.000 0.000 0.640 0.080 0.092
#> GSM97091 5 0.3995 -0.5033 0.480 0.000 0.000 0.000 0.516 0.004
#> GSM97148 1 0.0767 0.5871 0.976 0.012 0.000 0.004 0.008 0.000
#> GSM97063 1 0.3986 0.5131 0.532 0.000 0.000 0.000 0.464 0.004
#> GSM97053 1 0.3672 0.5879 0.632 0.000 0.000 0.000 0.368 0.000
#> GSM97066 3 0.4137 0.6350 0.000 0.012 0.768 0.016 0.168 0.036
#> GSM97079 4 0.4465 0.5009 0.000 0.032 0.004 0.704 0.020 0.240
#> GSM97083 4 0.5278 0.1528 0.004 0.000 0.012 0.504 0.424 0.056
#> GSM97084 4 0.2402 0.5950 0.000 0.000 0.000 0.868 0.012 0.120
#> GSM97094 4 0.2375 0.6021 0.000 0.008 0.000 0.896 0.060 0.036
#> GSM97096 3 0.5840 0.5943 0.000 0.196 0.636 0.012 0.048 0.108
#> GSM97097 4 0.6223 0.4435 0.000 0.080 0.096 0.640 0.036 0.148
#> GSM97107 4 0.1622 0.6082 0.016 0.000 0.000 0.940 0.028 0.016
#> GSM97054 4 0.3758 0.4664 0.000 0.000 0.000 0.668 0.008 0.324
#> GSM97062 4 0.2980 0.5689 0.000 0.000 0.000 0.808 0.012 0.180
#> GSM97069 3 0.3178 0.6799 0.000 0.008 0.848 0.016 0.104 0.024
#> GSM97070 3 0.4002 0.6530 0.000 0.016 0.792 0.016 0.136 0.040
#> GSM97073 3 0.5001 0.6232 0.000 0.060 0.724 0.016 0.156 0.044
#> GSM97076 5 0.9419 0.1306 0.188 0.188 0.192 0.080 0.284 0.068
#> GSM97077 6 0.5450 0.4277 0.004 0.228 0.024 0.048 0.032 0.664
#> GSM97095 4 0.7260 0.4862 0.168 0.032 0.000 0.516 0.116 0.168
#> GSM97102 3 0.3540 0.7020 0.000 0.076 0.840 0.012 0.040 0.032
#> GSM97109 2 0.3512 0.5588 0.036 0.856 0.008 0.024 0.036 0.040
#> GSM97110 2 0.4690 0.5154 0.048 0.792 0.040 0.028 0.036 0.056
#> GSM97074 5 0.6832 0.2135 0.000 0.016 0.284 0.096 0.504 0.100
#> GSM97085 3 0.5459 0.3169 0.000 0.008 0.548 0.016 0.364 0.064
#> GSM97059 6 0.7218 0.2905 0.148 0.172 0.000 0.176 0.012 0.492
#> GSM97072 3 0.3873 0.6722 0.000 0.024 0.812 0.016 0.108 0.040
#> GSM97078 5 0.6415 -0.0231 0.000 0.000 0.096 0.388 0.440 0.076
#> GSM97067 3 0.4280 0.6439 0.000 0.024 0.772 0.016 0.148 0.040
#> GSM97087 3 0.3608 0.6801 0.000 0.000 0.788 0.000 0.064 0.148
#> GSM97111 2 0.2882 0.5823 0.000 0.860 0.020 0.000 0.020 0.100
#> GSM97064 6 0.6161 0.0840 0.000 0.052 0.360 0.020 0.056 0.512
#> GSM97065 2 0.7146 0.2867 0.024 0.540 0.216 0.016 0.124 0.080
#> GSM97081 3 0.4991 0.6409 0.000 0.136 0.712 0.000 0.048 0.104
#> GSM97082 3 0.2401 0.7158 0.000 0.004 0.892 0.000 0.044 0.060
#> GSM97088 5 0.7140 0.1455 0.000 0.000 0.268 0.236 0.404 0.092
#> GSM97100 6 0.5029 0.3767 0.000 0.276 0.000 0.112 0.000 0.612
#> GSM97104 3 0.1390 0.7196 0.000 0.016 0.948 0.000 0.004 0.032
#> GSM97108 2 0.4051 0.0720 0.000 0.560 0.000 0.008 0.000 0.432
#> GSM97050 6 0.6680 0.4206 0.000 0.092 0.156 0.104 0.048 0.600
#> GSM97080 3 0.1901 0.7112 0.000 0.008 0.924 0.004 0.052 0.012
#> GSM97089 3 0.3955 0.6792 0.000 0.012 0.776 0.000 0.064 0.148
#> GSM97092 3 0.3695 0.6626 0.000 0.004 0.776 0.000 0.044 0.176
#> GSM97093 6 0.7338 0.0740 0.000 0.184 0.312 0.020 0.076 0.408
#> GSM97058 6 0.5622 0.3845 0.000 0.296 0.056 0.024 0.024 0.600
#> GSM97051 6 0.5982 0.4381 0.000 0.036 0.104 0.184 0.036 0.640
#> GSM97052 3 0.4038 0.6276 0.000 0.000 0.728 0.000 0.056 0.216
#> GSM97061 3 0.4857 0.5299 0.000 0.008 0.636 0.004 0.056 0.296
Heatmaps for the consensus matrix. It visualizes the probability of two samples to be in a same group.
consensus_heatmap(res, k = 2)
consensus_heatmap(res, k = 3)
consensus_heatmap(res, k = 4)
consensus_heatmap(res, k = 5)
consensus_heatmap(res, k = 6)
Heatmaps for the membership of samples in all partitions to see how consistent they are:
membership_heatmap(res, k = 2)
membership_heatmap(res, k = 3)
membership_heatmap(res, k = 4)
membership_heatmap(res, k = 5)
membership_heatmap(res, k = 6)
As soon as we have had the classes for columns, we can look for signatures which are significantly different between classes which can be candidate marks for certain classes. Following are the heatmaps for signatures.
Signature heatmaps where rows are scaled:
get_signatures(res, k = 2)
get_signatures(res, k = 3)
get_signatures(res, k = 4)
get_signatures(res, k = 5)
get_signatures(res, k = 6)
Signature heatmaps where rows are not scaled:
get_signatures(res, k = 2, scale_rows = FALSE)
get_signatures(res, k = 3, scale_rows = FALSE)
get_signatures(res, k = 4, scale_rows = FALSE)
get_signatures(res, k = 5, scale_rows = FALSE)
get_signatures(res, k = 6, scale_rows = FALSE)
Compare the overlap of signatures from different k:
compare_signatures(res)
get_signature() returns a data frame invisibly. TO get the list of signatures, the function
call should be assigned to a variable explicitly. In following code, if plot argument is set
to FALSE, no heatmap is plotted while only the differential analysis is performed.
# code only for demonstration
tb = get_signature(res, k = ..., plot = FALSE)
An example of the output of tb is:
#> which_row fdr mean_1 mean_2 scaled_mean_1 scaled_mean_2 km
#> 1 38 0.042760348 8.373488 9.131774 -0.5533452 0.5164555 1
#> 2 40 0.018707592 7.106213 8.469186 -0.6173731 0.5762149 1
#> 3 55 0.019134737 10.221463 11.207825 -0.6159697 0.5749050 1
#> 4 59 0.006059896 5.921854 7.869574 -0.6899429 0.6439467 1
#> 5 60 0.018055526 8.928898 10.211722 -0.6204761 0.5791110 1
#> 6 98 0.009384629 15.714769 14.887706 0.6635654 -0.6193277 2
...
The columns in tb are:
which_row: row indices corresponding to the input matrix.fdr: FDR for the differential test. mean_x: The mean value in group x.scaled_mean_x: The mean value in group x after rows are scaled.km: Row groups if k-means clustering is applied to rows.UMAP plot which shows how samples are separated.
dimension_reduction(res, k = 2, method = "UMAP")
dimension_reduction(res, k = 3, method = "UMAP")
dimension_reduction(res, k = 4, method = "UMAP")
dimension_reduction(res, k = 5, method = "UMAP")
dimension_reduction(res, k = 6, method = "UMAP")
Following heatmap shows how subgroups are split when increasing k:
collect_classes(res)
Test correlation between subgroups and known annotations. If the known annotation is numeric, one-way ANOVA test is applied, and if the known annotation is discrete, chi-squared contingency table test is applied.
test_to_known_factors(res)
#> n disease.state(p) specimen(p) cell.type(p) other(p) k
#> SD:skmeans 97 5.64e-04 0.415 3.09e-13 0.1020 2
#> SD:skmeans 94 3.88e-04 0.472 3.48e-13 0.4671 3
#> SD:skmeans 93 6.25e-04 0.424 3.06e-15 0.1922 4
#> SD:skmeans 77 8.93e-06 0.200 2.55e-14 0.0219 5
#> SD:skmeans 56 5.12e-05 0.782 2.75e-09 0.2266 6
If matrix rows can be associated to genes, consider to use functional_enrichment(res,
...) to perform function enrichment for the signature genes. See this vignette for more detailed explanations.
The object with results only for a single top-value method and a single partition method can be extracted as:
res = res_list["SD", "pam"]
# you can also extract it by
# res = res_list["SD:pam"]
A summary of res and all the functions that can be applied to it:
res
#> A 'ConsensusPartition' object with k = 2, 3, 4, 5, 6.
#> On a matrix with 21512 rows and 100 columns.
#> Top rows (1000, 2000, 3000, 4000, 5000) are extracted by 'SD' method.
#> Subgroups are detected by 'pam' method.
#> Performed in total 1250 partitions by row resampling.
#> Best k for subgroups seems to be 3.
#>
#> Following methods can be applied to this 'ConsensusPartition' object:
#> [1] "cola_report" "collect_classes" "collect_plots"
#> [4] "collect_stats" "colnames" "compare_signatures"
#> [7] "consensus_heatmap" "dimension_reduction" "functional_enrichment"
#> [10] "get_anno_col" "get_anno" "get_classes"
#> [13] "get_consensus" "get_matrix" "get_membership"
#> [16] "get_param" "get_signatures" "get_stats"
#> [19] "is_best_k" "is_stable_k" "membership_heatmap"
#> [22] "ncol" "nrow" "plot_ecdf"
#> [25] "rownames" "select_partition_number" "show"
#> [28] "suggest_best_k" "test_to_known_factors"
collect_plots() function collects all the plots made from res for all k (number of partitions)
into one single page to provide an easy and fast comparison between different k.
collect_plots(res)
The plots are:
k and the heatmap of
predicted classes for each k.k.k.k.All the plots in panels can be made by individual functions and they are plotted later in this section.
select_partition_number() produces several plots showing different
statistics for choosing “optimized” k. There are following statistics:
k;k, the area increased is defined as \(A_k - A_{k-1}\).The detailed explanations of these statistics can be found in the cola vignette.
Generally speaking, lower PAC score, higher mean silhouette score or higher
concordance corresponds to better partition. Rand index and Jaccard index
measure how similar the current partition is compared to partition with k-1.
If they are too similar, we won't accept k is better than k-1.
select_partition_number(res)
The numeric values for all these statistics can be obtained by get_stats().
get_stats(res)
#> k 1-PAC mean_silhouette concordance area_increased Rand Jaccard
#> 2 2 0.646 0.818 0.903 0.4445 0.535 0.535
#> 3 3 0.608 0.787 0.879 0.4692 0.687 0.473
#> 4 4 0.626 0.640 0.793 0.0930 0.908 0.748
#> 5 5 0.602 0.589 0.753 0.0738 0.902 0.688
#> 6 6 0.628 0.541 0.736 0.0581 0.866 0.509
suggest_best_k() suggests the best \(k\) based on these statistics. The rules are as follows:
suggest_best_k(res)
#> [1] 3
Following shows the table of the partitions (You need to click the show/hide
code output link to see it). The membership matrix (columns with name p*)
is inferred by
clue::cl_consensus()
function with the SE method. Basically the value in the membership matrix
represents the probability to belong to a certain group. The finall class
label for an item is determined with the group with highest probability it
belongs to.
In get_classes() function, the entropy is calculated from the membership
matrix and the silhouette score is calculated from the consensus matrix.
cbind(get_classes(res, k = 2), get_membership(res, k = 2))
#> class entropy silhouette p1 p2
#> GSM97138 1 0.4298 0.830 0.912 0.088
#> GSM97145 1 0.4298 0.830 0.912 0.088
#> GSM97147 2 0.5408 0.881 0.124 0.876
#> GSM97125 1 0.4298 0.830 0.912 0.088
#> GSM97127 1 0.1633 0.826 0.976 0.024
#> GSM97130 1 0.0000 0.819 1.000 0.000
#> GSM97133 1 0.0000 0.819 1.000 0.000
#> GSM97134 1 0.8443 0.638 0.728 0.272
#> GSM97120 1 0.6048 0.805 0.852 0.148
#> GSM97126 1 0.9686 0.505 0.604 0.396
#> GSM97112 1 0.4298 0.830 0.912 0.088
#> GSM97115 2 0.6247 0.852 0.156 0.844
#> GSM97116 1 0.0672 0.822 0.992 0.008
#> GSM97117 2 0.0000 0.926 0.000 1.000
#> GSM97119 1 0.4298 0.830 0.912 0.088
#> GSM97122 1 0.4298 0.830 0.912 0.088
#> GSM97135 1 0.4298 0.830 0.912 0.088
#> GSM97136 2 0.9170 0.277 0.332 0.668
#> GSM97139 1 0.3274 0.830 0.940 0.060
#> GSM97146 1 0.0000 0.819 1.000 0.000
#> GSM97123 2 0.1633 0.922 0.024 0.976
#> GSM97129 2 0.9954 -0.206 0.460 0.540
#> GSM97143 1 0.9635 0.581 0.612 0.388
#> GSM97113 2 0.0672 0.925 0.008 0.992
#> GSM97056 1 0.0000 0.819 1.000 0.000
#> GSM97124 1 0.4298 0.830 0.912 0.088
#> GSM97132 1 0.5946 0.801 0.856 0.144
#> GSM97144 1 0.5946 0.766 0.856 0.144
#> GSM97149 1 0.3733 0.803 0.928 0.072
#> GSM97068 2 0.4431 0.898 0.092 0.908
#> GSM97071 2 0.6712 0.831 0.176 0.824
#> GSM97086 2 0.4298 0.899 0.088 0.912
#> GSM97103 2 0.0000 0.926 0.000 1.000
#> GSM97057 2 0.4298 0.899 0.088 0.912
#> GSM97060 2 0.0000 0.926 0.000 1.000
#> GSM97075 2 0.0000 0.926 0.000 1.000
#> GSM97098 2 0.0000 0.926 0.000 1.000
#> GSM97099 2 0.0000 0.926 0.000 1.000
#> GSM97101 2 0.0000 0.926 0.000 1.000
#> GSM97105 2 0.4298 0.899 0.088 0.912
#> GSM97106 2 0.0000 0.926 0.000 1.000
#> GSM97121 2 0.4298 0.899 0.088 0.912
#> GSM97128 1 0.9922 0.230 0.552 0.448
#> GSM97131 2 0.1633 0.922 0.024 0.976
#> GSM97137 1 0.1633 0.820 0.976 0.024
#> GSM97118 1 0.9393 0.627 0.644 0.356
#> GSM97114 2 0.0938 0.918 0.012 0.988
#> GSM97142 1 0.4298 0.830 0.912 0.088
#> GSM97140 2 0.4298 0.899 0.088 0.912
#> GSM97141 2 0.0000 0.926 0.000 1.000
#> GSM97055 1 0.9522 0.608 0.628 0.372
#> GSM97090 1 0.9732 0.375 0.596 0.404
#> GSM97091 1 0.4939 0.827 0.892 0.108
#> GSM97148 1 0.0000 0.819 1.000 0.000
#> GSM97063 1 0.4298 0.830 0.912 0.088
#> GSM97053 1 0.0000 0.819 1.000 0.000
#> GSM97066 2 0.0000 0.926 0.000 1.000
#> GSM97079 2 0.4298 0.899 0.088 0.912
#> GSM97083 1 0.8207 0.658 0.744 0.256
#> GSM97084 2 0.5842 0.867 0.140 0.860
#> GSM97094 1 0.9996 0.267 0.512 0.488
#> GSM97096 2 0.0000 0.926 0.000 1.000
#> GSM97097 2 0.0000 0.926 0.000 1.000
#> GSM97107 2 0.6148 0.851 0.152 0.848
#> GSM97054 2 0.6247 0.852 0.156 0.844
#> GSM97062 2 0.6148 0.856 0.152 0.848
#> GSM97069 2 0.0000 0.926 0.000 1.000
#> GSM97070 2 0.0000 0.926 0.000 1.000
#> GSM97073 2 0.0000 0.926 0.000 1.000
#> GSM97076 2 0.9323 0.501 0.348 0.652
#> GSM97077 2 0.4298 0.899 0.088 0.912
#> GSM97095 2 0.6048 0.860 0.148 0.852
#> GSM97102 2 0.0000 0.926 0.000 1.000
#> GSM97109 2 0.0000 0.926 0.000 1.000
#> GSM97110 2 0.0000 0.926 0.000 1.000
#> GSM97074 1 0.9358 0.605 0.648 0.352
#> GSM97085 2 0.3431 0.881 0.064 0.936
#> GSM97059 2 0.4431 0.898 0.092 0.908
#> GSM97072 2 0.0000 0.926 0.000 1.000
#> GSM97078 1 0.9944 0.210 0.544 0.456
#> GSM97067 2 0.0000 0.926 0.000 1.000
#> GSM97087 2 0.0000 0.926 0.000 1.000
#> GSM97111 2 0.0000 0.926 0.000 1.000
#> GSM97064 2 0.4298 0.899 0.088 0.912
#> GSM97065 2 0.2236 0.918 0.036 0.964
#> GSM97081 2 0.0000 0.926 0.000 1.000
#> GSM97082 2 0.0000 0.926 0.000 1.000
#> GSM97088 2 0.4562 0.858 0.096 0.904
#> GSM97100 2 0.4298 0.899 0.088 0.912
#> GSM97104 2 0.0000 0.926 0.000 1.000
#> GSM97108 2 0.0000 0.926 0.000 1.000
#> GSM97050 2 0.4298 0.899 0.088 0.912
#> GSM97080 2 0.0000 0.926 0.000 1.000
#> GSM97089 2 0.0000 0.926 0.000 1.000
#> GSM97092 2 0.0000 0.926 0.000 1.000
#> GSM97093 2 0.4298 0.899 0.088 0.912
#> GSM97058 2 0.4298 0.899 0.088 0.912
#> GSM97051 2 0.4298 0.899 0.088 0.912
#> GSM97052 2 0.0000 0.926 0.000 1.000
#> GSM97061 2 0.0000 0.926 0.000 1.000
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM97138 1 0.0592 0.8613 0.988 0.000 0.012
#> GSM97145 1 0.0000 0.8637 1.000 0.000 0.000
#> GSM97147 2 0.0661 0.8761 0.004 0.988 0.008
#> GSM97125 1 0.0000 0.8637 1.000 0.000 0.000
#> GSM97127 1 0.0000 0.8637 1.000 0.000 0.000
#> GSM97130 2 0.3619 0.7897 0.136 0.864 0.000
#> GSM97133 1 0.5016 0.5972 0.760 0.240 0.000
#> GSM97134 2 0.6215 0.2552 0.428 0.572 0.000
#> GSM97120 1 0.3038 0.8023 0.896 0.000 0.104
#> GSM97126 3 0.9977 -0.1030 0.300 0.348 0.352
#> GSM97112 1 0.0000 0.8637 1.000 0.000 0.000
#> GSM97115 2 0.1643 0.8589 0.044 0.956 0.000
#> GSM97116 1 0.0000 0.8637 1.000 0.000 0.000
#> GSM97117 3 0.3619 0.8823 0.000 0.136 0.864
#> GSM97119 1 0.0747 0.8616 0.984 0.016 0.000
#> GSM97122 1 0.0747 0.8616 0.984 0.016 0.000
#> GSM97135 1 0.0424 0.8634 0.992 0.008 0.000
#> GSM97136 3 0.5919 0.5647 0.276 0.012 0.712
#> GSM97139 1 0.0000 0.8637 1.000 0.000 0.000
#> GSM97146 1 0.2878 0.7937 0.904 0.096 0.000
#> GSM97123 3 0.2711 0.8745 0.000 0.088 0.912
#> GSM97129 3 0.7208 0.4021 0.340 0.040 0.620
#> GSM97143 1 0.6888 0.2304 0.552 0.016 0.432
#> GSM97113 3 0.3989 0.8752 0.012 0.124 0.864
#> GSM97056 2 0.5882 0.5247 0.348 0.652 0.000
#> GSM97124 1 0.0747 0.8616 0.984 0.016 0.000
#> GSM97132 1 0.5744 0.7390 0.800 0.128 0.072
#> GSM97144 2 0.5327 0.6323 0.272 0.728 0.000
#> GSM97149 1 0.6274 0.0108 0.544 0.456 0.000
#> GSM97068 2 0.0424 0.8756 0.000 0.992 0.008
#> GSM97071 2 0.2448 0.8470 0.000 0.924 0.076
#> GSM97086 2 0.1163 0.8756 0.000 0.972 0.028
#> GSM97103 3 0.2959 0.8885 0.000 0.100 0.900
#> GSM97057 2 0.1163 0.8756 0.000 0.972 0.028
#> GSM97060 3 0.0000 0.8812 0.000 0.000 1.000
#> GSM97075 3 0.3686 0.8811 0.000 0.140 0.860
#> GSM97098 3 0.2356 0.8898 0.000 0.072 0.928
#> GSM97099 3 0.3686 0.8811 0.000 0.140 0.860
#> GSM97101 3 0.3686 0.8811 0.000 0.140 0.860
#> GSM97105 2 0.2448 0.8559 0.000 0.924 0.076
#> GSM97106 3 0.2261 0.8905 0.000 0.068 0.932
#> GSM97121 2 0.1163 0.8763 0.000 0.972 0.028
#> GSM97128 2 0.8823 0.3919 0.280 0.564 0.156
#> GSM97131 3 0.4399 0.8457 0.000 0.188 0.812
#> GSM97137 2 0.3686 0.7881 0.140 0.860 0.000
#> GSM97118 1 0.8397 0.4854 0.588 0.116 0.296
#> GSM97114 3 0.4228 0.8768 0.008 0.148 0.844
#> GSM97142 1 0.0237 0.8637 0.996 0.004 0.000
#> GSM97140 2 0.1529 0.8722 0.000 0.960 0.040
#> GSM97141 3 0.3619 0.8823 0.000 0.136 0.864
#> GSM97055 1 0.7458 0.5718 0.676 0.088 0.236
#> GSM97090 2 0.3619 0.7897 0.136 0.864 0.000
#> GSM97091 1 0.1774 0.8526 0.960 0.016 0.024
#> GSM97148 1 0.0237 0.8629 0.996 0.004 0.000
#> GSM97063 1 0.0592 0.8627 0.988 0.012 0.000
#> GSM97053 1 0.0000 0.8637 1.000 0.000 0.000
#> GSM97066 3 0.0892 0.8840 0.000 0.020 0.980
#> GSM97079 2 0.2625 0.8614 0.000 0.916 0.084
#> GSM97083 2 0.4605 0.7292 0.204 0.796 0.000
#> GSM97084 2 0.0592 0.8754 0.000 0.988 0.012
#> GSM97094 3 0.9901 -0.0639 0.336 0.272 0.392
#> GSM97096 3 0.2356 0.8898 0.000 0.072 0.928
#> GSM97097 3 0.3038 0.8881 0.000 0.104 0.896
#> GSM97107 2 0.2711 0.8238 0.000 0.912 0.088
#> GSM97054 2 0.0747 0.8765 0.000 0.984 0.016
#> GSM97062 2 0.0237 0.8746 0.000 0.996 0.004
#> GSM97069 3 0.0000 0.8812 0.000 0.000 1.000
#> GSM97070 3 0.0000 0.8812 0.000 0.000 1.000
#> GSM97073 3 0.2165 0.8908 0.000 0.064 0.936
#> GSM97076 2 0.3910 0.8139 0.104 0.876 0.020
#> GSM97077 2 0.1163 0.8756 0.000 0.972 0.028
#> GSM97095 2 0.0000 0.8733 0.000 1.000 0.000
#> GSM97102 3 0.0000 0.8812 0.000 0.000 1.000
#> GSM97109 3 0.2959 0.8885 0.000 0.100 0.900
#> GSM97110 3 0.3116 0.8876 0.000 0.108 0.892
#> GSM97074 1 0.8140 0.2702 0.524 0.072 0.404
#> GSM97085 3 0.1289 0.8693 0.000 0.032 0.968
#> GSM97059 2 0.1163 0.8756 0.000 0.972 0.028
#> GSM97072 3 0.0000 0.8812 0.000 0.000 1.000
#> GSM97078 2 0.4369 0.8190 0.040 0.864 0.096
#> GSM97067 3 0.0000 0.8812 0.000 0.000 1.000
#> GSM97087 3 0.0000 0.8812 0.000 0.000 1.000
#> GSM97111 3 0.3619 0.8823 0.000 0.136 0.864
#> GSM97064 2 0.2066 0.8640 0.000 0.940 0.060
#> GSM97065 3 0.4654 0.8203 0.000 0.208 0.792
#> GSM97081 3 0.1964 0.8879 0.000 0.056 0.944
#> GSM97082 3 0.1860 0.8858 0.000 0.052 0.948
#> GSM97088 3 0.4217 0.8156 0.032 0.100 0.868
#> GSM97100 2 0.1163 0.8756 0.000 0.972 0.028
#> GSM97104 3 0.0000 0.8812 0.000 0.000 1.000
#> GSM97108 3 0.3686 0.8811 0.000 0.140 0.860
#> GSM97050 2 0.2537 0.8645 0.000 0.920 0.080
#> GSM97080 3 0.0000 0.8812 0.000 0.000 1.000
#> GSM97089 3 0.2356 0.8898 0.000 0.072 0.928
#> GSM97092 3 0.1753 0.8858 0.000 0.048 0.952
#> GSM97093 2 0.6337 0.6416 0.028 0.708 0.264
#> GSM97058 2 0.2711 0.8420 0.000 0.912 0.088
#> GSM97051 2 0.3267 0.8451 0.000 0.884 0.116
#> GSM97052 3 0.1860 0.8858 0.000 0.052 0.948
#> GSM97061 3 0.2165 0.8881 0.000 0.064 0.936
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM97138 1 0.0188 0.7240 0.996 0.000 0.000 0.004
#> GSM97145 1 0.1557 0.7043 0.944 0.000 0.000 0.056
#> GSM97147 2 0.1256 0.8428 0.000 0.964 0.028 0.008
#> GSM97125 1 0.4454 0.4991 0.692 0.000 0.000 0.308
#> GSM97127 1 0.4134 0.5509 0.740 0.000 0.000 0.260
#> GSM97130 2 0.3443 0.7832 0.016 0.848 0.000 0.136
#> GSM97133 1 0.0188 0.7238 0.996 0.004 0.000 0.000
#> GSM97134 2 0.6773 0.3971 0.276 0.588 0.000 0.136
#> GSM97120 1 0.0188 0.7227 0.996 0.000 0.004 0.000
#> GSM97126 2 0.8959 -0.0606 0.104 0.384 0.376 0.136
#> GSM97112 4 0.4948 0.0325 0.440 0.000 0.000 0.560
#> GSM97115 2 0.1953 0.8374 0.004 0.940 0.012 0.044
#> GSM97116 1 0.0000 0.7259 1.000 0.000 0.000 0.000
#> GSM97117 3 0.1743 0.8099 0.000 0.056 0.940 0.004
#> GSM97119 4 0.4585 0.2935 0.332 0.000 0.000 0.668
#> GSM97122 1 0.4933 0.2611 0.568 0.000 0.000 0.432
#> GSM97135 1 0.4746 0.4111 0.632 0.000 0.000 0.368
#> GSM97136 3 0.4571 0.5837 0.252 0.004 0.736 0.008
#> GSM97139 1 0.0000 0.7259 1.000 0.000 0.000 0.000
#> GSM97146 1 0.0000 0.7259 1.000 0.000 0.000 0.000
#> GSM97123 3 0.5564 0.7717 0.000 0.076 0.708 0.216
#> GSM97129 3 0.7213 0.3217 0.260 0.016 0.588 0.136
#> GSM97143 3 0.7704 -0.1942 0.336 0.000 0.432 0.232
#> GSM97113 3 0.2081 0.7936 0.000 0.084 0.916 0.000
#> GSM97056 1 0.4800 0.2509 0.656 0.340 0.000 0.004
#> GSM97124 1 0.4981 0.1641 0.536 0.000 0.000 0.464
#> GSM97132 4 0.8204 0.2131 0.336 0.084 0.088 0.492
#> GSM97144 2 0.6400 0.5411 0.168 0.652 0.000 0.180
#> GSM97149 1 0.1637 0.6582 0.940 0.060 0.000 0.000
#> GSM97068 2 0.1706 0.8406 0.000 0.948 0.016 0.036
#> GSM97071 2 0.3749 0.8013 0.000 0.840 0.032 0.128
#> GSM97086 2 0.0469 0.8449 0.000 0.988 0.012 0.000
#> GSM97103 3 0.0657 0.8148 0.000 0.004 0.984 0.012
#> GSM97057 2 0.1305 0.8442 0.000 0.960 0.036 0.004
#> GSM97060 3 0.4072 0.7831 0.000 0.000 0.748 0.252
#> GSM97075 3 0.2053 0.8066 0.000 0.072 0.924 0.004
#> GSM97098 3 0.1398 0.8159 0.000 0.004 0.956 0.040
#> GSM97099 3 0.1489 0.8109 0.000 0.044 0.952 0.004
#> GSM97101 3 0.1743 0.8101 0.000 0.056 0.940 0.004
#> GSM97105 2 0.3249 0.7861 0.000 0.852 0.140 0.008
#> GSM97106 3 0.2831 0.8153 0.000 0.004 0.876 0.120
#> GSM97121 2 0.1576 0.8401 0.000 0.948 0.048 0.004
#> GSM97128 4 0.5150 0.0140 0.000 0.396 0.008 0.596
#> GSM97131 3 0.3542 0.7798 0.000 0.120 0.852 0.028
#> GSM97137 2 0.2002 0.8342 0.020 0.936 0.000 0.044
#> GSM97118 4 0.9014 0.3057 0.232 0.076 0.264 0.428
#> GSM97114 3 0.2708 0.8041 0.016 0.076 0.904 0.004
#> GSM97142 4 0.4817 0.1959 0.388 0.000 0.000 0.612
#> GSM97140 2 0.1902 0.8340 0.000 0.932 0.064 0.004
#> GSM97141 3 0.1489 0.8109 0.000 0.044 0.952 0.004
#> GSM97055 4 0.6751 0.3333 0.240 0.060 0.048 0.652
#> GSM97090 2 0.1888 0.8340 0.016 0.940 0.000 0.044
#> GSM97091 4 0.4585 0.2935 0.332 0.000 0.000 0.668
#> GSM97148 1 0.0000 0.7259 1.000 0.000 0.000 0.000
#> GSM97063 4 0.4761 0.2337 0.372 0.000 0.000 0.628
#> GSM97053 1 0.4661 0.4432 0.652 0.000 0.000 0.348
#> GSM97066 3 0.4748 0.7738 0.000 0.016 0.716 0.268
#> GSM97079 2 0.2412 0.8287 0.000 0.908 0.084 0.008
#> GSM97083 2 0.5582 0.4080 0.024 0.576 0.000 0.400
#> GSM97084 2 0.1211 0.8405 0.000 0.960 0.000 0.040
#> GSM97094 3 0.9678 -0.2129 0.216 0.212 0.384 0.188
#> GSM97096 3 0.1398 0.8159 0.000 0.004 0.956 0.040
#> GSM97097 3 0.0937 0.8160 0.000 0.012 0.976 0.012
#> GSM97107 2 0.4856 0.7316 0.000 0.780 0.084 0.136
#> GSM97054 2 0.0336 0.8451 0.000 0.992 0.008 0.000
#> GSM97062 2 0.1677 0.8399 0.000 0.948 0.012 0.040
#> GSM97069 3 0.3873 0.7843 0.000 0.000 0.772 0.228
#> GSM97070 3 0.3837 0.7852 0.000 0.000 0.776 0.224
#> GSM97073 3 0.2469 0.8100 0.000 0.000 0.892 0.108
#> GSM97076 2 0.4011 0.8097 0.020 0.844 0.024 0.112
#> GSM97077 2 0.1661 0.8383 0.000 0.944 0.052 0.004
#> GSM97095 2 0.0707 0.8444 0.000 0.980 0.000 0.020
#> GSM97102 3 0.3172 0.8089 0.000 0.000 0.840 0.160
#> GSM97109 3 0.0657 0.8148 0.000 0.004 0.984 0.012
#> GSM97110 3 0.1209 0.8126 0.000 0.032 0.964 0.004
#> GSM97074 4 0.8033 0.3496 0.220 0.036 0.208 0.536
#> GSM97085 4 0.5257 -0.2800 0.000 0.008 0.444 0.548
#> GSM97059 2 0.1398 0.8404 0.000 0.956 0.040 0.004
#> GSM97072 3 0.4072 0.7797 0.000 0.000 0.748 0.252
#> GSM97078 2 0.3074 0.7878 0.000 0.848 0.000 0.152
#> GSM97067 3 0.3907 0.7856 0.000 0.000 0.768 0.232
#> GSM97087 3 0.3907 0.7896 0.000 0.000 0.768 0.232
#> GSM97111 3 0.1824 0.8082 0.000 0.060 0.936 0.004
#> GSM97064 2 0.4499 0.7579 0.000 0.804 0.124 0.072
#> GSM97065 3 0.3400 0.7537 0.004 0.128 0.856 0.012
#> GSM97081 3 0.3877 0.8169 0.000 0.048 0.840 0.112
#> GSM97082 3 0.5267 0.7822 0.000 0.048 0.712 0.240
#> GSM97088 4 0.6340 -0.1893 0.000 0.064 0.408 0.528
#> GSM97100 2 0.1489 0.8396 0.000 0.952 0.044 0.004
#> GSM97104 3 0.4382 0.7589 0.000 0.000 0.704 0.296
#> GSM97108 3 0.2053 0.8066 0.000 0.072 0.924 0.004
#> GSM97050 2 0.2563 0.8287 0.000 0.908 0.072 0.020
#> GSM97080 3 0.4331 0.7596 0.000 0.000 0.712 0.288
#> GSM97089 3 0.1722 0.8181 0.000 0.008 0.944 0.048
#> GSM97092 3 0.5136 0.7891 0.000 0.048 0.728 0.224
#> GSM97093 2 0.4769 0.6010 0.000 0.684 0.308 0.008
#> GSM97058 2 0.3355 0.7709 0.000 0.836 0.160 0.004
#> GSM97051 2 0.4322 0.7462 0.000 0.804 0.044 0.152
#> GSM97052 3 0.5102 0.7890 0.000 0.048 0.732 0.220
#> GSM97061 3 0.4880 0.8000 0.000 0.052 0.760 0.188
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM97138 1 0.3993 0.8833 0.756 0.028 0.000 0.000 0.216
#> GSM97145 1 0.4313 0.7285 0.636 0.008 0.000 0.000 0.356
#> GSM97147 4 0.1915 0.7795 0.040 0.032 0.000 0.928 0.000
#> GSM97125 5 0.4016 0.4033 0.272 0.000 0.012 0.000 0.716
#> GSM97127 5 0.3913 0.2784 0.324 0.000 0.000 0.000 0.676
#> GSM97130 4 0.4524 0.6028 0.000 0.000 0.336 0.644 0.020
#> GSM97133 1 0.3452 0.9074 0.756 0.000 0.000 0.000 0.244
#> GSM97134 4 0.6381 0.4380 0.000 0.000 0.364 0.464 0.172
#> GSM97120 1 0.3819 0.8970 0.756 0.016 0.000 0.000 0.228
#> GSM97126 3 0.8294 0.0223 0.012 0.292 0.348 0.268 0.080
#> GSM97112 5 0.0609 0.6738 0.020 0.000 0.000 0.000 0.980
#> GSM97115 4 0.2351 0.7816 0.000 0.000 0.088 0.896 0.016
#> GSM97116 1 0.3452 0.9074 0.756 0.000 0.000 0.000 0.244
#> GSM97117 2 0.3090 0.6796 0.040 0.856 0.000 0.104 0.000
#> GSM97119 5 0.0162 0.6770 0.000 0.000 0.004 0.000 0.996
#> GSM97122 5 0.2329 0.6369 0.124 0.000 0.000 0.000 0.876
#> GSM97135 5 0.2377 0.6329 0.128 0.000 0.000 0.000 0.872
#> GSM97136 2 0.4960 0.5622 0.024 0.740 0.072 0.000 0.164
#> GSM97139 1 0.3452 0.9074 0.756 0.000 0.000 0.000 0.244
#> GSM97146 1 0.3452 0.9074 0.756 0.000 0.000 0.000 0.244
#> GSM97123 2 0.6408 0.5471 0.156 0.616 0.188 0.040 0.000
#> GSM97129 2 0.6880 0.1600 0.020 0.544 0.264 0.012 0.160
#> GSM97143 5 0.7120 0.1011 0.012 0.320 0.296 0.000 0.372
#> GSM97113 2 0.2329 0.6505 0.000 0.876 0.000 0.124 0.000
#> GSM97056 1 0.5363 0.5453 0.664 0.000 0.004 0.232 0.100
#> GSM97124 5 0.2439 0.6414 0.120 0.000 0.004 0.000 0.876
#> GSM97132 5 0.7099 0.3555 0.036 0.048 0.364 0.056 0.496
#> GSM97144 4 0.6218 0.4508 0.000 0.000 0.364 0.488 0.148
#> GSM97149 1 0.3993 0.8780 0.756 0.000 0.000 0.028 0.216
#> GSM97068 4 0.2474 0.7836 0.000 0.012 0.084 0.896 0.008
#> GSM97071 4 0.5581 0.6348 0.040 0.032 0.256 0.664 0.008
#> GSM97086 4 0.0162 0.7913 0.004 0.000 0.000 0.996 0.000
#> GSM97103 2 0.1041 0.6830 0.032 0.964 0.004 0.000 0.000
#> GSM97057 4 0.2017 0.7800 0.008 0.080 0.000 0.912 0.000
#> GSM97060 2 0.5954 0.5111 0.192 0.592 0.216 0.000 0.000
#> GSM97075 2 0.3649 0.6592 0.040 0.808 0.000 0.152 0.000
#> GSM97098 2 0.2344 0.6684 0.032 0.904 0.064 0.000 0.000
#> GSM97099 2 0.1740 0.6975 0.012 0.932 0.000 0.056 0.000
#> GSM97101 2 0.2470 0.6886 0.012 0.884 0.000 0.104 0.000
#> GSM97105 4 0.2520 0.7689 0.048 0.056 0.000 0.896 0.000
#> GSM97106 2 0.4761 0.6036 0.124 0.732 0.144 0.000 0.000
#> GSM97121 4 0.2438 0.7728 0.040 0.060 0.000 0.900 0.000
#> GSM97128 3 0.6845 -0.2551 0.000 0.004 0.400 0.348 0.248
#> GSM97131 2 0.4514 0.6221 0.072 0.740 0.000 0.188 0.000
#> GSM97137 4 0.2514 0.7818 0.000 0.000 0.060 0.896 0.044
#> GSM97118 5 0.7029 0.1816 0.000 0.216 0.364 0.016 0.404
#> GSM97114 2 0.3409 0.6750 0.052 0.836 0.000 0.112 0.000
#> GSM97142 5 0.0404 0.6777 0.012 0.000 0.000 0.000 0.988
#> GSM97140 4 0.2077 0.7773 0.040 0.040 0.000 0.920 0.000
#> GSM97141 2 0.1740 0.6975 0.012 0.932 0.000 0.056 0.000
#> GSM97055 5 0.4575 0.5150 0.048 0.040 0.012 0.100 0.800
#> GSM97090 4 0.2351 0.7816 0.000 0.000 0.088 0.896 0.016
#> GSM97091 5 0.0162 0.6770 0.000 0.000 0.004 0.000 0.996
#> GSM97148 1 0.3452 0.9074 0.756 0.000 0.000 0.000 0.244
#> GSM97063 5 0.0404 0.6777 0.012 0.000 0.000 0.000 0.988
#> GSM97053 5 0.2813 0.5914 0.168 0.000 0.000 0.000 0.832
#> GSM97066 3 0.5658 0.3611 0.080 0.408 0.512 0.000 0.000
#> GSM97079 4 0.3934 0.7204 0.032 0.168 0.008 0.792 0.000
#> GSM97083 4 0.6596 0.3011 0.000 0.000 0.372 0.416 0.212
#> GSM97084 4 0.2351 0.7816 0.000 0.000 0.088 0.896 0.016
#> GSM97094 2 0.8195 -0.1625 0.000 0.372 0.296 0.136 0.196
#> GSM97096 2 0.2344 0.6684 0.032 0.904 0.064 0.000 0.000
#> GSM97097 2 0.2504 0.6700 0.032 0.900 0.064 0.004 0.000
#> GSM97107 4 0.5848 0.4987 0.000 0.060 0.364 0.556 0.020
#> GSM97054 4 0.0162 0.7919 0.000 0.000 0.004 0.996 0.000
#> GSM97062 4 0.2518 0.7834 0.000 0.008 0.080 0.896 0.016
#> GSM97069 3 0.5689 0.3362 0.080 0.440 0.480 0.000 0.000
#> GSM97070 3 0.5689 0.3362 0.080 0.440 0.480 0.000 0.000
#> GSM97073 2 0.5153 -0.2858 0.040 0.524 0.436 0.000 0.000
#> GSM97076 4 0.5797 0.6758 0.000 0.064 0.228 0.660 0.048
#> GSM97077 4 0.1997 0.7783 0.040 0.036 0.000 0.924 0.000
#> GSM97095 4 0.1628 0.7908 0.000 0.000 0.056 0.936 0.008
#> GSM97102 2 0.3828 0.6448 0.072 0.808 0.120 0.000 0.000
#> GSM97109 2 0.1281 0.6829 0.032 0.956 0.012 0.000 0.000
#> GSM97110 2 0.1041 0.6844 0.004 0.964 0.000 0.032 0.000
#> GSM97074 3 0.3366 0.1695 0.000 0.004 0.784 0.000 0.212
#> GSM97085 3 0.4891 0.3360 0.076 0.036 0.760 0.000 0.128
#> GSM97059 4 0.1915 0.7795 0.040 0.032 0.000 0.928 0.000
#> GSM97072 3 0.5476 0.3333 0.068 0.388 0.544 0.000 0.000
#> GSM97078 4 0.4626 0.5737 0.000 0.000 0.364 0.616 0.020
#> GSM97067 3 0.5680 0.3501 0.080 0.428 0.492 0.000 0.000
#> GSM97087 2 0.5635 0.5403 0.168 0.636 0.196 0.000 0.000
#> GSM97111 2 0.3090 0.6796 0.040 0.856 0.000 0.104 0.000
#> GSM97064 4 0.4883 0.6988 0.100 0.100 0.036 0.764 0.000
#> GSM97065 2 0.3599 0.6457 0.020 0.824 0.016 0.140 0.000
#> GSM97081 2 0.4339 0.6907 0.060 0.808 0.056 0.076 0.000
#> GSM97082 2 0.6041 0.5149 0.168 0.628 0.188 0.016 0.000
#> GSM97088 3 0.6557 0.2940 0.004 0.224 0.608 0.048 0.116
#> GSM97100 4 0.1915 0.7795 0.040 0.032 0.000 0.928 0.000
#> GSM97104 3 0.5815 0.2237 0.104 0.356 0.540 0.000 0.000
#> GSM97108 2 0.3731 0.6537 0.040 0.800 0.000 0.160 0.000
#> GSM97050 4 0.3584 0.7502 0.056 0.108 0.004 0.832 0.000
#> GSM97080 3 0.5533 0.3637 0.084 0.336 0.580 0.000 0.000
#> GSM97089 2 0.1908 0.6881 0.092 0.908 0.000 0.000 0.000
#> GSM97092 2 0.5841 0.5783 0.168 0.664 0.144 0.024 0.000
#> GSM97093 4 0.4315 0.5956 0.024 0.276 0.000 0.700 0.000
#> GSM97058 4 0.2850 0.7583 0.036 0.092 0.000 0.872 0.000
#> GSM97051 4 0.4425 0.7001 0.132 0.036 0.044 0.788 0.000
#> GSM97052 2 0.5803 0.5804 0.168 0.668 0.140 0.024 0.000
#> GSM97061 2 0.5803 0.5998 0.160 0.680 0.124 0.036 0.000
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM97138 1 0.0000 0.8449 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM97145 1 0.4656 0.5908 0.708 0.024 0.000 0.000 0.064 0.204
#> GSM97147 4 0.1923 0.7423 0.000 0.064 0.016 0.916 0.004 0.000
#> GSM97125 1 0.6029 -0.0648 0.396 0.000 0.000 0.000 0.248 0.356
#> GSM97127 1 0.5815 0.0959 0.472 0.000 0.000 0.000 0.200 0.328
#> GSM97130 5 0.3817 0.3131 0.000 0.000 0.000 0.432 0.568 0.000
#> GSM97133 1 0.0000 0.8449 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM97134 5 0.4346 0.5426 0.028 0.000 0.000 0.336 0.632 0.004
#> GSM97120 1 0.0000 0.8449 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM97126 5 0.6168 0.5688 0.036 0.212 0.000 0.184 0.564 0.004
#> GSM97112 6 0.4466 0.5800 0.044 0.000 0.000 0.000 0.336 0.620
#> GSM97115 4 0.2730 0.7003 0.000 0.000 0.000 0.808 0.192 0.000
#> GSM97116 1 0.0000 0.8449 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM97117 2 0.3014 0.6461 0.000 0.804 0.012 0.184 0.000 0.000
#> GSM97119 6 0.4466 0.5800 0.044 0.000 0.000 0.000 0.336 0.620
#> GSM97122 6 0.4766 0.5641 0.072 0.000 0.000 0.000 0.316 0.612
#> GSM97135 6 0.4766 0.5641 0.072 0.000 0.000 0.000 0.316 0.612
#> GSM97136 2 0.5300 0.4919 0.028 0.664 0.224 0.000 0.072 0.012
#> GSM97139 1 0.0000 0.8449 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM97146 1 0.0000 0.8449 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM97123 3 0.3377 0.5015 0.000 0.188 0.784 0.028 0.000 0.000
#> GSM97129 2 0.5559 -0.0591 0.024 0.472 0.000 0.060 0.440 0.004
#> GSM97143 5 0.5072 0.4363 0.048 0.172 0.000 0.000 0.696 0.084
#> GSM97113 2 0.3272 0.6093 0.016 0.820 0.000 0.144 0.020 0.000
#> GSM97056 1 0.2333 0.7314 0.884 0.000 0.000 0.092 0.024 0.000
#> GSM97124 6 0.4847 0.5382 0.064 0.000 0.000 0.000 0.376 0.560
#> GSM97132 5 0.1553 0.4790 0.032 0.004 0.000 0.012 0.944 0.008
#> GSM97144 5 0.3445 0.5996 0.008 0.000 0.000 0.260 0.732 0.000
#> GSM97149 1 0.0000 0.8449 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM97068 4 0.2946 0.7118 0.000 0.012 0.000 0.812 0.176 0.000
#> GSM97071 4 0.4521 0.2197 0.000 0.032 0.012 0.644 0.312 0.000
#> GSM97086 4 0.0937 0.7714 0.000 0.000 0.000 0.960 0.040 0.000
#> GSM97103 2 0.1866 0.6400 0.000 0.908 0.084 0.000 0.000 0.008
#> GSM97057 4 0.2443 0.7514 0.004 0.096 0.000 0.880 0.020 0.000
#> GSM97060 3 0.2854 0.5321 0.000 0.208 0.792 0.000 0.000 0.000
#> GSM97075 2 0.3342 0.6249 0.000 0.760 0.012 0.228 0.000 0.000
#> GSM97098 2 0.3314 0.5454 0.000 0.764 0.224 0.000 0.000 0.012
#> GSM97099 2 0.1714 0.6773 0.000 0.908 0.000 0.092 0.000 0.000
#> GSM97101 2 0.2300 0.6646 0.000 0.856 0.000 0.144 0.000 0.000
#> GSM97105 4 0.1895 0.7391 0.000 0.072 0.016 0.912 0.000 0.000
#> GSM97106 3 0.3955 0.3723 0.000 0.316 0.668 0.004 0.000 0.012
#> GSM97121 4 0.2214 0.7274 0.000 0.096 0.016 0.888 0.000 0.000
#> GSM97128 5 0.2697 0.6243 0.000 0.000 0.000 0.188 0.812 0.000
#> GSM97131 2 0.4146 0.5419 0.000 0.676 0.036 0.288 0.000 0.000
#> GSM97137 4 0.2838 0.7024 0.004 0.000 0.000 0.808 0.188 0.000
#> GSM97118 5 0.3227 0.5191 0.028 0.124 0.000 0.000 0.832 0.016
#> GSM97114 2 0.3386 0.6404 0.008 0.788 0.016 0.188 0.000 0.000
#> GSM97142 6 0.4466 0.5800 0.044 0.000 0.000 0.000 0.336 0.620
#> GSM97140 4 0.1779 0.7418 0.000 0.064 0.016 0.920 0.000 0.000
#> GSM97141 2 0.1714 0.6773 0.000 0.908 0.000 0.092 0.000 0.000
#> GSM97055 6 0.6099 0.4368 0.004 0.008 0.000 0.180 0.344 0.464
#> GSM97090 4 0.2730 0.7003 0.000 0.000 0.000 0.808 0.192 0.000
#> GSM97091 6 0.4466 0.5800 0.044 0.000 0.000 0.000 0.336 0.620
#> GSM97148 1 0.0000 0.8449 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM97063 6 0.4466 0.5800 0.044 0.000 0.000 0.000 0.336 0.620
#> GSM97053 6 0.5406 0.4695 0.160 0.000 0.000 0.000 0.272 0.568
#> GSM97066 3 0.6197 0.3165 0.000 0.228 0.396 0.000 0.008 0.368
#> GSM97079 4 0.4446 0.6450 0.000 0.220 0.032 0.720 0.020 0.008
#> GSM97083 5 0.3198 0.5997 0.000 0.000 0.000 0.260 0.740 0.000
#> GSM97084 4 0.2664 0.7073 0.000 0.000 0.000 0.816 0.184 0.000
#> GSM97094 5 0.5570 0.3802 0.016 0.336 0.008 0.080 0.560 0.000
#> GSM97096 2 0.3314 0.5454 0.000 0.764 0.224 0.000 0.000 0.012
#> GSM97097 2 0.3559 0.5486 0.000 0.744 0.240 0.004 0.000 0.012
#> GSM97107 5 0.4150 0.5388 0.000 0.028 0.000 0.320 0.652 0.000
#> GSM97054 4 0.1204 0.7706 0.000 0.000 0.000 0.944 0.056 0.000
#> GSM97062 4 0.3037 0.7223 0.000 0.016 0.004 0.820 0.160 0.000
#> GSM97069 6 0.6228 -0.3608 0.000 0.292 0.336 0.000 0.004 0.368
#> GSM97070 6 0.6110 -0.3617 0.000 0.296 0.336 0.000 0.000 0.368
#> GSM97073 6 0.6053 -0.3480 0.000 0.372 0.256 0.000 0.000 0.372
#> GSM97076 4 0.5735 0.3175 0.000 0.060 0.004 0.572 0.312 0.052
#> GSM97077 4 0.0260 0.7676 0.000 0.008 0.000 0.992 0.000 0.000
#> GSM97095 4 0.2260 0.7408 0.000 0.000 0.000 0.860 0.140 0.000
#> GSM97102 2 0.4201 0.4588 0.000 0.664 0.300 0.000 0.000 0.036
#> GSM97109 2 0.2056 0.6428 0.012 0.904 0.080 0.000 0.000 0.004
#> GSM97110 2 0.2446 0.6517 0.012 0.904 0.020 0.044 0.020 0.000
#> GSM97074 5 0.4193 0.4661 0.000 0.000 0.024 0.000 0.624 0.352
#> GSM97085 5 0.6082 0.0337 0.000 0.000 0.272 0.000 0.368 0.360
#> GSM97059 4 0.0692 0.7648 0.000 0.020 0.004 0.976 0.000 0.000
#> GSM97072 3 0.5750 0.2860 0.000 0.172 0.448 0.000 0.000 0.380
#> GSM97078 5 0.3717 0.4289 0.000 0.000 0.000 0.384 0.616 0.000
#> GSM97067 6 0.6315 -0.3601 0.000 0.288 0.336 0.000 0.008 0.368
#> GSM97087 3 0.3482 0.5340 0.000 0.316 0.684 0.000 0.000 0.000
#> GSM97111 2 0.3221 0.6431 0.000 0.792 0.020 0.188 0.000 0.000
#> GSM97064 4 0.4443 0.5883 0.000 0.068 0.232 0.696 0.004 0.000
#> GSM97065 2 0.4052 0.6087 0.016 0.772 0.016 0.176 0.016 0.004
#> GSM97081 2 0.4358 0.6514 0.000 0.740 0.108 0.144 0.000 0.008
#> GSM97082 3 0.4302 0.5243 0.000 0.368 0.608 0.020 0.004 0.000
#> GSM97088 5 0.5991 0.5059 0.000 0.024 0.156 0.028 0.624 0.168
#> GSM97100 4 0.1719 0.7445 0.000 0.060 0.016 0.924 0.000 0.000
#> GSM97104 3 0.4011 0.4509 0.000 0.056 0.732 0.000 0.000 0.212
#> GSM97108 2 0.3558 0.6054 0.000 0.736 0.016 0.248 0.000 0.000
#> GSM97050 4 0.3755 0.7121 0.000 0.136 0.048 0.800 0.008 0.008
#> GSM97080 3 0.5765 0.3752 0.000 0.144 0.496 0.000 0.008 0.352
#> GSM97089 2 0.2333 0.6139 0.000 0.872 0.120 0.004 0.000 0.004
#> GSM97092 3 0.4389 0.5065 0.000 0.372 0.596 0.032 0.000 0.000
#> GSM97093 4 0.4165 0.5869 0.000 0.292 0.004 0.676 0.028 0.000
#> GSM97058 4 0.1349 0.7645 0.000 0.056 0.004 0.940 0.000 0.000
#> GSM97051 4 0.4348 0.5584 0.000 0.064 0.248 0.688 0.000 0.000
#> GSM97052 3 0.4312 0.5134 0.000 0.368 0.604 0.028 0.000 0.000
#> GSM97061 3 0.4815 0.4493 0.000 0.384 0.556 0.060 0.000 0.000
Heatmaps for the consensus matrix. It visualizes the probability of two samples to be in a same group.
consensus_heatmap(res, k = 2)
consensus_heatmap(res, k = 3)
consensus_heatmap(res, k = 4)
consensus_heatmap(res, k = 5)
consensus_heatmap(res, k = 6)
Heatmaps for the membership of samples in all partitions to see how consistent they are:
membership_heatmap(res, k = 2)
membership_heatmap(res, k = 3)
membership_heatmap(res, k = 4)
membership_heatmap(res, k = 5)
membership_heatmap(res, k = 6)
As soon as we have had the classes for columns, we can look for signatures which are significantly different between classes which can be candidate marks for certain classes. Following are the heatmaps for signatures.
Signature heatmaps where rows are scaled:
get_signatures(res, k = 2)
get_signatures(res, k = 3)
get_signatures(res, k = 4)
get_signatures(res, k = 5)
get_signatures(res, k = 6)
Signature heatmaps where rows are not scaled:
get_signatures(res, k = 2, scale_rows = FALSE)
get_signatures(res, k = 3, scale_rows = FALSE)
get_signatures(res, k = 4, scale_rows = FALSE)
get_signatures(res, k = 5, scale_rows = FALSE)
get_signatures(res, k = 6, scale_rows = FALSE)
Compare the overlap of signatures from different k:
compare_signatures(res)
get_signature() returns a data frame invisibly. TO get the list of signatures, the function
call should be assigned to a variable explicitly. In following code, if plot argument is set
to FALSE, no heatmap is plotted while only the differential analysis is performed.
# code only for demonstration
tb = get_signature(res, k = ..., plot = FALSE)
An example of the output of tb is:
#> which_row fdr mean_1 mean_2 scaled_mean_1 scaled_mean_2 km
#> 1 38 0.042760348 8.373488 9.131774 -0.5533452 0.5164555 1
#> 2 40 0.018707592 7.106213 8.469186 -0.6173731 0.5762149 1
#> 3 55 0.019134737 10.221463 11.207825 -0.6159697 0.5749050 1
#> 4 59 0.006059896 5.921854 7.869574 -0.6899429 0.6439467 1
#> 5 60 0.018055526 8.928898 10.211722 -0.6204761 0.5791110 1
#> 6 98 0.009384629 15.714769 14.887706 0.6635654 -0.6193277 2
...
The columns in tb are:
which_row: row indices corresponding to the input matrix.fdr: FDR for the differential test. mean_x: The mean value in group x.scaled_mean_x: The mean value in group x after rows are scaled.km: Row groups if k-means clustering is applied to rows.UMAP plot which shows how samples are separated.
dimension_reduction(res, k = 2, method = "UMAP")
dimension_reduction(res, k = 3, method = "UMAP")
dimension_reduction(res, k = 4, method = "UMAP")
dimension_reduction(res, k = 5, method = "UMAP")
dimension_reduction(res, k = 6, method = "UMAP")
Following heatmap shows how subgroups are split when increasing k:
collect_classes(res)
Test correlation between subgroups and known annotations. If the known annotation is numeric, one-way ANOVA test is applied, and if the known annotation is discrete, chi-squared contingency table test is applied.
test_to_known_factors(res)
#> n disease.state(p) specimen(p) cell.type(p) other(p) k
#> SD:pam 94 5.42e-06 0.735 6.12e-16 0.0454 2
#> SD:pam 91 2.19e-04 0.331 4.32e-12 0.0923 3
#> SD:pam 76 1.62e-04 0.602 5.19e-08 0.5594 4
#> SD:pam 76 2.44e-02 0.570 8.38e-10 0.1164 5
#> SD:pam 74 8.01e-02 0.761 1.48e-07 0.2209 6
If matrix rows can be associated to genes, consider to use functional_enrichment(res,
...) to perform function enrichment for the signature genes. See this vignette for more detailed explanations.
The object with results only for a single top-value method and a single partition method can be extracted as:
res = res_list["SD", "mclust"]
# you can also extract it by
# res = res_list["SD:mclust"]
A summary of res and all the functions that can be applied to it:
res
#> A 'ConsensusPartition' object with k = 2, 3, 4, 5, 6.
#> On a matrix with 21512 rows and 100 columns.
#> Top rows (1000, 2000, 3000, 4000, 5000) are extracted by 'SD' method.
#> Subgroups are detected by 'mclust' method.
#> Performed in total 1250 partitions by row resampling.
#> Best k for subgroups seems to be 4.
#>
#> Following methods can be applied to this 'ConsensusPartition' object:
#> [1] "cola_report" "collect_classes" "collect_plots"
#> [4] "collect_stats" "colnames" "compare_signatures"
#> [7] "consensus_heatmap" "dimension_reduction" "functional_enrichment"
#> [10] "get_anno_col" "get_anno" "get_classes"
#> [13] "get_consensus" "get_matrix" "get_membership"
#> [16] "get_param" "get_signatures" "get_stats"
#> [19] "is_best_k" "is_stable_k" "membership_heatmap"
#> [22] "ncol" "nrow" "plot_ecdf"
#> [25] "rownames" "select_partition_number" "show"
#> [28] "suggest_best_k" "test_to_known_factors"
collect_plots() function collects all the plots made from res for all k (number of partitions)
into one single page to provide an easy and fast comparison between different k.
collect_plots(res)
The plots are:
k and the heatmap of
predicted classes for each k.k.k.k.All the plots in panels can be made by individual functions and they are plotted later in this section.
select_partition_number() produces several plots showing different
statistics for choosing “optimized” k. There are following statistics:
k;k, the area increased is defined as \(A_k - A_{k-1}\).The detailed explanations of these statistics can be found in the cola vignette.
Generally speaking, lower PAC score, higher mean silhouette score or higher
concordance corresponds to better partition. Rand index and Jaccard index
measure how similar the current partition is compared to partition with k-1.
If they are too similar, we won't accept k is better than k-1.
select_partition_number(res)
The numeric values for all these statistics can be obtained by get_stats().
get_stats(res)
#> k 1-PAC mean_silhouette concordance area_increased Rand Jaccard
#> 2 2 0.500 0.665 0.861 0.3307 0.677 0.677
#> 3 3 0.520 0.738 0.845 0.8447 0.665 0.521
#> 4 4 0.815 0.874 0.914 0.2114 0.737 0.415
#> 5 5 0.617 0.761 0.853 0.0288 0.949 0.802
#> 6 6 0.721 0.753 0.862 0.0532 0.914 0.646
suggest_best_k() suggests the best \(k\) based on these statistics. The rules are as follows:
suggest_best_k(res)
#> [1] 4
Following shows the table of the partitions (You need to click the show/hide
code output link to see it). The membership matrix (columns with name p*)
is inferred by
clue::cl_consensus()
function with the SE method. Basically the value in the membership matrix
represents the probability to belong to a certain group. The finall class
label for an item is determined with the group with highest probability it
belongs to.
In get_classes() function, the entropy is calculated from the membership
matrix and the silhouette score is calculated from the consensus matrix.
cbind(get_classes(res, k = 2), get_membership(res, k = 2))
#> class entropy silhouette p1 p2
#> GSM97138 1 0.0000 0.813 1.000 0.000
#> GSM97145 1 0.0000 0.813 1.000 0.000
#> GSM97147 1 0.0000 0.813 1.000 0.000
#> GSM97125 1 0.0000 0.813 1.000 0.000
#> GSM97127 1 0.0000 0.813 1.000 0.000
#> GSM97130 1 0.0000 0.813 1.000 0.000
#> GSM97133 1 0.0000 0.813 1.000 0.000
#> GSM97134 1 0.0000 0.813 1.000 0.000
#> GSM97120 1 0.0000 0.813 1.000 0.000
#> GSM97126 1 0.0000 0.813 1.000 0.000
#> GSM97112 1 0.0000 0.813 1.000 0.000
#> GSM97115 1 0.0376 0.813 0.996 0.004
#> GSM97116 1 0.0000 0.813 1.000 0.000
#> GSM97117 1 0.9323 0.474 0.652 0.348
#> GSM97119 1 0.0000 0.813 1.000 0.000
#> GSM97122 1 0.0000 0.813 1.000 0.000
#> GSM97135 1 0.0000 0.813 1.000 0.000
#> GSM97136 1 0.9286 0.479 0.656 0.344
#> GSM97139 1 0.0000 0.813 1.000 0.000
#> GSM97146 1 0.0000 0.813 1.000 0.000
#> GSM97123 2 0.9998 0.127 0.492 0.508
#> GSM97129 1 0.9209 0.490 0.664 0.336
#> GSM97143 1 0.0000 0.813 1.000 0.000
#> GSM97113 1 0.9323 0.474 0.652 0.348
#> GSM97056 1 0.0000 0.813 1.000 0.000
#> GSM97124 1 0.0000 0.813 1.000 0.000
#> GSM97132 1 0.0000 0.813 1.000 0.000
#> GSM97144 1 0.0000 0.813 1.000 0.000
#> GSM97149 1 0.0000 0.813 1.000 0.000
#> GSM97068 1 0.0672 0.812 0.992 0.008
#> GSM97071 1 0.0672 0.812 0.992 0.008
#> GSM97086 1 0.0672 0.812 0.992 0.008
#> GSM97103 1 0.9358 0.463 0.648 0.352
#> GSM97057 1 0.9323 0.474 0.652 0.348
#> GSM97060 2 0.9608 0.466 0.384 0.616
#> GSM97075 1 0.9323 0.474 0.652 0.348
#> GSM97098 2 0.9933 0.283 0.452 0.548
#> GSM97099 1 0.9323 0.474 0.652 0.348
#> GSM97101 1 0.9323 0.474 0.652 0.348
#> GSM97105 1 0.9323 0.474 0.652 0.348
#> GSM97106 2 0.9993 0.163 0.484 0.516
#> GSM97121 1 0.9323 0.474 0.652 0.348
#> GSM97128 1 0.0672 0.812 0.992 0.008
#> GSM97131 1 0.9323 0.474 0.652 0.348
#> GSM97137 1 0.0000 0.813 1.000 0.000
#> GSM97118 1 0.0000 0.813 1.000 0.000
#> GSM97114 1 0.9323 0.474 0.652 0.348
#> GSM97142 1 0.0000 0.813 1.000 0.000
#> GSM97140 1 0.9248 0.485 0.660 0.340
#> GSM97141 1 0.9323 0.474 0.652 0.348
#> GSM97055 1 0.0000 0.813 1.000 0.000
#> GSM97090 1 0.0000 0.813 1.000 0.000
#> GSM97091 1 0.0000 0.813 1.000 0.000
#> GSM97148 1 0.0000 0.813 1.000 0.000
#> GSM97063 1 0.0000 0.813 1.000 0.000
#> GSM97053 1 0.0000 0.813 1.000 0.000
#> GSM97066 2 0.0000 0.716 0.000 1.000
#> GSM97079 1 0.0938 0.810 0.988 0.012
#> GSM97083 1 0.0000 0.813 1.000 0.000
#> GSM97084 1 0.0672 0.812 0.992 0.008
#> GSM97094 1 0.0672 0.812 0.992 0.008
#> GSM97096 2 0.9393 0.515 0.356 0.644
#> GSM97097 1 0.1414 0.806 0.980 0.020
#> GSM97107 1 0.0672 0.812 0.992 0.008
#> GSM97054 1 0.0672 0.812 0.992 0.008
#> GSM97062 1 0.0672 0.812 0.992 0.008
#> GSM97069 2 0.0000 0.716 0.000 1.000
#> GSM97070 2 0.0000 0.716 0.000 1.000
#> GSM97073 2 0.0000 0.716 0.000 1.000
#> GSM97076 1 0.0672 0.812 0.992 0.008
#> GSM97077 1 0.9323 0.474 0.652 0.348
#> GSM97095 1 0.0376 0.813 0.996 0.004
#> GSM97102 2 0.0000 0.716 0.000 1.000
#> GSM97109 1 0.9323 0.474 0.652 0.348
#> GSM97110 1 0.9323 0.474 0.652 0.348
#> GSM97074 1 0.0672 0.812 0.992 0.008
#> GSM97085 1 0.0672 0.812 0.992 0.008
#> GSM97059 1 0.0672 0.812 0.992 0.008
#> GSM97072 2 0.8955 0.558 0.312 0.688
#> GSM97078 1 0.0672 0.812 0.992 0.008
#> GSM97067 2 0.0000 0.716 0.000 1.000
#> GSM97087 2 0.0000 0.716 0.000 1.000
#> GSM97111 1 0.9323 0.474 0.652 0.348
#> GSM97064 1 0.9323 0.474 0.652 0.348
#> GSM97065 1 0.9323 0.474 0.652 0.348
#> GSM97081 2 0.9522 0.491 0.372 0.628
#> GSM97082 2 0.0000 0.716 0.000 1.000
#> GSM97088 1 0.0672 0.812 0.992 0.008
#> GSM97100 1 0.9087 0.506 0.676 0.324
#> GSM97104 2 0.0000 0.716 0.000 1.000
#> GSM97108 1 0.9323 0.474 0.652 0.348
#> GSM97050 1 0.9323 0.474 0.652 0.348
#> GSM97080 2 0.0000 0.716 0.000 1.000
#> GSM97089 1 0.9850 0.222 0.572 0.428
#> GSM97092 2 0.9460 0.505 0.364 0.636
#> GSM97093 1 0.9323 0.474 0.652 0.348
#> GSM97058 1 0.9323 0.474 0.652 0.348
#> GSM97051 1 0.7950 0.608 0.760 0.240
#> GSM97052 2 0.9460 0.505 0.364 0.636
#> GSM97061 2 0.9963 0.242 0.464 0.536
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM97138 1 0.0000 0.8519 1.000 0.000 0.000
#> GSM97145 1 0.0000 0.8519 1.000 0.000 0.000
#> GSM97147 1 0.5639 0.7421 0.752 0.232 0.016
#> GSM97125 1 0.0000 0.8519 1.000 0.000 0.000
#> GSM97127 1 0.0237 0.8523 0.996 0.004 0.000
#> GSM97130 1 0.6476 0.8221 0.748 0.068 0.184
#> GSM97133 1 0.0237 0.8523 0.996 0.004 0.000
#> GSM97134 1 0.7036 0.8152 0.720 0.096 0.184
#> GSM97120 1 0.0000 0.8519 1.000 0.000 0.000
#> GSM97126 1 0.2959 0.8416 0.900 0.100 0.000
#> GSM97112 1 0.0000 0.8519 1.000 0.000 0.000
#> GSM97115 1 0.7108 0.8135 0.716 0.100 0.184
#> GSM97116 1 0.0000 0.8519 1.000 0.000 0.000
#> GSM97117 2 0.0000 0.8459 0.000 1.000 0.000
#> GSM97119 1 0.0000 0.8519 1.000 0.000 0.000
#> GSM97122 1 0.0000 0.8519 1.000 0.000 0.000
#> GSM97135 1 0.0000 0.8519 1.000 0.000 0.000
#> GSM97136 2 0.5061 0.5146 0.208 0.784 0.008
#> GSM97139 1 0.0000 0.8519 1.000 0.000 0.000
#> GSM97146 1 0.0000 0.8519 1.000 0.000 0.000
#> GSM97123 2 0.6307 -0.4608 0.000 0.512 0.488
#> GSM97129 2 0.0000 0.8459 0.000 1.000 0.000
#> GSM97143 1 0.0000 0.8519 1.000 0.000 0.000
#> GSM97113 2 0.0000 0.8459 0.000 1.000 0.000
#> GSM97056 1 0.4409 0.8291 0.824 0.004 0.172
#> GSM97124 1 0.0000 0.8519 1.000 0.000 0.000
#> GSM97132 1 0.0237 0.8521 0.996 0.000 0.004
#> GSM97144 1 0.7036 0.8152 0.720 0.096 0.184
#> GSM97149 1 0.0237 0.8523 0.996 0.004 0.000
#> GSM97068 1 0.9145 0.5864 0.532 0.284 0.184
#> GSM97071 1 0.7108 0.8135 0.716 0.100 0.184
#> GSM97086 1 0.9122 0.5938 0.536 0.280 0.184
#> GSM97103 2 0.0000 0.8459 0.000 1.000 0.000
#> GSM97057 2 0.2261 0.7570 0.068 0.932 0.000
#> GSM97060 3 0.6286 0.5574 0.000 0.464 0.536
#> GSM97075 2 0.0000 0.8459 0.000 1.000 0.000
#> GSM97098 2 0.6308 -0.4712 0.000 0.508 0.492
#> GSM97099 2 0.0000 0.8459 0.000 1.000 0.000
#> GSM97101 2 0.0000 0.8459 0.000 1.000 0.000
#> GSM97105 2 0.0000 0.8459 0.000 1.000 0.000
#> GSM97106 2 0.6295 -0.4139 0.000 0.528 0.472
#> GSM97121 2 0.0000 0.8459 0.000 1.000 0.000
#> GSM97128 1 0.7108 0.8135 0.716 0.100 0.184
#> GSM97131 2 0.0000 0.8459 0.000 1.000 0.000
#> GSM97137 1 0.4099 0.8378 0.852 0.008 0.140
#> GSM97118 1 0.2066 0.8487 0.940 0.060 0.000
#> GSM97114 2 0.2537 0.7321 0.080 0.920 0.000
#> GSM97142 1 0.0000 0.8519 1.000 0.000 0.000
#> GSM97140 2 0.0000 0.8459 0.000 1.000 0.000
#> GSM97141 2 0.0000 0.8459 0.000 1.000 0.000
#> GSM97055 1 0.0424 0.8524 0.992 0.008 0.000
#> GSM97090 1 0.7108 0.8135 0.716 0.100 0.184
#> GSM97091 1 0.0000 0.8519 1.000 0.000 0.000
#> GSM97148 1 0.0000 0.8519 1.000 0.000 0.000
#> GSM97063 1 0.0000 0.8519 1.000 0.000 0.000
#> GSM97053 1 0.0000 0.8519 1.000 0.000 0.000
#> GSM97066 3 0.4555 0.8318 0.000 0.200 0.800
#> GSM97079 1 0.9433 0.3321 0.420 0.404 0.176
#> GSM97083 1 0.7036 0.8152 0.720 0.096 0.184
#> GSM97084 1 0.7862 0.7737 0.668 0.148 0.184
#> GSM97094 1 0.7108 0.8135 0.716 0.100 0.184
#> GSM97096 3 0.6225 0.6315 0.000 0.432 0.568
#> GSM97097 2 0.8839 0.2077 0.256 0.572 0.172
#> GSM97107 1 0.7108 0.8135 0.716 0.100 0.184
#> GSM97054 1 0.9073 0.6079 0.544 0.272 0.184
#> GSM97062 1 0.8845 0.6597 0.576 0.240 0.184
#> GSM97069 3 0.4346 0.8410 0.000 0.184 0.816
#> GSM97070 3 0.4346 0.8410 0.000 0.184 0.816
#> GSM97073 3 0.4346 0.8410 0.000 0.184 0.816
#> GSM97076 1 0.3340 0.8345 0.880 0.120 0.000
#> GSM97077 2 0.0000 0.8459 0.000 1.000 0.000
#> GSM97095 1 0.7108 0.8135 0.716 0.100 0.184
#> GSM97102 3 0.4346 0.8410 0.000 0.184 0.816
#> GSM97109 2 0.0000 0.8459 0.000 1.000 0.000
#> GSM97110 2 0.0000 0.8459 0.000 1.000 0.000
#> GSM97074 1 0.3832 0.8422 0.880 0.100 0.020
#> GSM97085 1 0.4047 0.8200 0.848 0.148 0.004
#> GSM97059 1 0.8350 0.7259 0.628 0.196 0.176
#> GSM97072 3 0.6126 0.6726 0.000 0.400 0.600
#> GSM97078 1 0.7108 0.8135 0.716 0.100 0.184
#> GSM97067 3 0.4346 0.8410 0.000 0.184 0.816
#> GSM97087 3 0.4346 0.8410 0.000 0.184 0.816
#> GSM97111 2 0.0000 0.8459 0.000 1.000 0.000
#> GSM97064 2 0.0000 0.8459 0.000 1.000 0.000
#> GSM97065 2 0.0000 0.8459 0.000 1.000 0.000
#> GSM97081 3 0.6267 0.5929 0.000 0.452 0.548
#> GSM97082 3 0.4346 0.8410 0.000 0.184 0.816
#> GSM97088 1 0.7108 0.8135 0.716 0.100 0.184
#> GSM97100 2 0.0475 0.8382 0.004 0.992 0.004
#> GSM97104 3 0.4346 0.8410 0.000 0.184 0.816
#> GSM97108 2 0.0000 0.8459 0.000 1.000 0.000
#> GSM97050 2 0.0000 0.8459 0.000 1.000 0.000
#> GSM97080 3 0.4346 0.8410 0.000 0.184 0.816
#> GSM97089 2 0.5926 0.0339 0.000 0.644 0.356
#> GSM97092 3 0.6252 0.6096 0.000 0.444 0.556
#> GSM97093 2 0.0000 0.8459 0.000 1.000 0.000
#> GSM97058 2 0.0000 0.8459 0.000 1.000 0.000
#> GSM97051 2 0.0237 0.8416 0.000 0.996 0.004
#> GSM97052 3 0.6267 0.5939 0.000 0.452 0.548
#> GSM97061 2 0.6309 -0.4849 0.000 0.504 0.496
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM97138 1 0.0779 0.9384 0.980 0.004 0.000 0.016
#> GSM97145 1 0.3219 0.8595 0.868 0.020 0.000 0.112
#> GSM97147 2 0.4669 0.7808 0.052 0.780 0.000 0.168
#> GSM97125 1 0.1520 0.9400 0.956 0.020 0.000 0.024
#> GSM97127 1 0.0707 0.9391 0.980 0.020 0.000 0.000
#> GSM97130 1 0.5602 -0.0459 0.508 0.020 0.000 0.472
#> GSM97133 1 0.0469 0.9324 0.988 0.000 0.000 0.012
#> GSM97134 4 0.3853 0.8578 0.160 0.020 0.000 0.820
#> GSM97120 1 0.0469 0.9324 0.988 0.000 0.000 0.012
#> GSM97126 2 0.6560 0.5459 0.248 0.620 0.000 0.132
#> GSM97112 1 0.1389 0.9423 0.952 0.000 0.000 0.048
#> GSM97115 4 0.3090 0.9198 0.056 0.056 0.000 0.888
#> GSM97116 1 0.0469 0.9324 0.988 0.000 0.000 0.012
#> GSM97117 2 0.0188 0.9188 0.000 0.996 0.004 0.000
#> GSM97119 1 0.1389 0.9423 0.952 0.000 0.000 0.048
#> GSM97122 1 0.1389 0.9423 0.952 0.000 0.000 0.048
#> GSM97135 1 0.1389 0.9423 0.952 0.000 0.000 0.048
#> GSM97136 2 0.3746 0.8748 0.040 0.868 0.020 0.072
#> GSM97139 1 0.0469 0.9324 0.988 0.000 0.000 0.012
#> GSM97146 1 0.0469 0.9324 0.988 0.000 0.000 0.012
#> GSM97123 3 0.2101 0.8902 0.000 0.060 0.928 0.012
#> GSM97129 2 0.2255 0.9015 0.012 0.920 0.000 0.068
#> GSM97143 1 0.1520 0.9400 0.956 0.020 0.000 0.024
#> GSM97113 2 0.0188 0.9188 0.000 0.996 0.004 0.000
#> GSM97056 1 0.1624 0.9398 0.952 0.020 0.000 0.028
#> GSM97124 1 0.1624 0.9398 0.952 0.020 0.000 0.028
#> GSM97132 1 0.1624 0.9398 0.952 0.020 0.000 0.028
#> GSM97144 4 0.3900 0.8532 0.164 0.020 0.000 0.816
#> GSM97149 1 0.0592 0.9316 0.984 0.000 0.000 0.016
#> GSM97068 2 0.4764 0.7400 0.032 0.748 0.000 0.220
#> GSM97071 4 0.2670 0.9234 0.052 0.040 0.000 0.908
#> GSM97086 4 0.2048 0.8984 0.008 0.064 0.000 0.928
#> GSM97103 3 0.5256 0.7391 0.000 0.204 0.732 0.064
#> GSM97057 2 0.3508 0.8515 0.012 0.848 0.004 0.136
#> GSM97060 3 0.1305 0.9021 0.000 0.036 0.960 0.004
#> GSM97075 2 0.0188 0.9188 0.000 0.996 0.004 0.000
#> GSM97098 3 0.1792 0.8910 0.000 0.068 0.932 0.000
#> GSM97099 2 0.0188 0.9188 0.000 0.996 0.004 0.000
#> GSM97101 2 0.0188 0.9188 0.000 0.996 0.004 0.000
#> GSM97105 2 0.1716 0.9090 0.000 0.936 0.000 0.064
#> GSM97106 3 0.2256 0.8875 0.000 0.056 0.924 0.020
#> GSM97121 2 0.1398 0.9177 0.000 0.956 0.004 0.040
#> GSM97128 4 0.2413 0.9220 0.064 0.020 0.000 0.916
#> GSM97131 3 0.5750 0.7069 0.000 0.216 0.696 0.088
#> GSM97137 1 0.1624 0.9398 0.952 0.020 0.000 0.028
#> GSM97118 1 0.2843 0.8905 0.892 0.020 0.000 0.088
#> GSM97114 2 0.0188 0.9188 0.000 0.996 0.004 0.000
#> GSM97142 1 0.1389 0.9423 0.952 0.000 0.000 0.048
#> GSM97140 2 0.1389 0.9171 0.000 0.952 0.000 0.048
#> GSM97141 2 0.0188 0.9188 0.000 0.996 0.004 0.000
#> GSM97055 1 0.3447 0.8441 0.852 0.020 0.000 0.128
#> GSM97090 4 0.3862 0.8672 0.152 0.024 0.000 0.824
#> GSM97091 1 0.1389 0.9423 0.952 0.000 0.000 0.048
#> GSM97148 1 0.0469 0.9324 0.988 0.000 0.000 0.012
#> GSM97063 1 0.1389 0.9423 0.952 0.000 0.000 0.048
#> GSM97053 1 0.1624 0.9398 0.952 0.020 0.000 0.028
#> GSM97066 3 0.2011 0.8746 0.000 0.080 0.920 0.000
#> GSM97079 4 0.2814 0.8443 0.000 0.132 0.000 0.868
#> GSM97083 4 0.3853 0.8578 0.160 0.020 0.000 0.820
#> GSM97084 4 0.2021 0.9045 0.012 0.056 0.000 0.932
#> GSM97094 4 0.2443 0.9235 0.060 0.024 0.000 0.916
#> GSM97096 3 0.0817 0.9036 0.000 0.024 0.976 0.000
#> GSM97097 4 0.4719 0.7546 0.000 0.180 0.048 0.772
#> GSM97107 4 0.2521 0.9232 0.064 0.024 0.000 0.912
#> GSM97054 4 0.1970 0.8996 0.008 0.060 0.000 0.932
#> GSM97062 4 0.1970 0.8996 0.008 0.060 0.000 0.932
#> GSM97069 3 0.0000 0.9005 0.000 0.000 1.000 0.000
#> GSM97070 3 0.0000 0.9005 0.000 0.000 1.000 0.000
#> GSM97073 3 0.0000 0.9005 0.000 0.000 1.000 0.000
#> GSM97076 2 0.5267 0.7365 0.076 0.740 0.000 0.184
#> GSM97077 2 0.1211 0.9175 0.000 0.960 0.000 0.040
#> GSM97095 4 0.3009 0.9186 0.052 0.056 0.000 0.892
#> GSM97102 3 0.0000 0.9005 0.000 0.000 1.000 0.000
#> GSM97109 2 0.0188 0.9188 0.000 0.996 0.004 0.000
#> GSM97110 2 0.0188 0.9188 0.000 0.996 0.004 0.000
#> GSM97074 4 0.2782 0.9211 0.068 0.024 0.004 0.904
#> GSM97085 3 0.6287 0.6367 0.068 0.036 0.700 0.196
#> GSM97059 2 0.4719 0.7707 0.048 0.772 0.000 0.180
#> GSM97072 3 0.0592 0.9030 0.000 0.016 0.984 0.000
#> GSM97078 4 0.2413 0.9220 0.064 0.020 0.000 0.916
#> GSM97067 3 0.0000 0.9005 0.000 0.000 1.000 0.000
#> GSM97087 3 0.0000 0.9005 0.000 0.000 1.000 0.000
#> GSM97111 2 0.0188 0.9188 0.000 0.996 0.004 0.000
#> GSM97064 3 0.5913 0.5092 0.000 0.352 0.600 0.048
#> GSM97065 2 0.0188 0.9188 0.000 0.996 0.004 0.000
#> GSM97081 3 0.0921 0.9038 0.000 0.028 0.972 0.000
#> GSM97082 3 0.0000 0.9005 0.000 0.000 1.000 0.000
#> GSM97088 4 0.2413 0.9220 0.064 0.020 0.000 0.916
#> GSM97100 2 0.2281 0.8908 0.000 0.904 0.000 0.096
#> GSM97104 3 0.0000 0.9005 0.000 0.000 1.000 0.000
#> GSM97108 2 0.1211 0.9175 0.000 0.960 0.000 0.040
#> GSM97050 2 0.1389 0.9150 0.000 0.952 0.000 0.048
#> GSM97080 3 0.0000 0.9005 0.000 0.000 1.000 0.000
#> GSM97089 3 0.4386 0.7844 0.004 0.192 0.784 0.020
#> GSM97092 3 0.1022 0.9035 0.000 0.032 0.968 0.000
#> GSM97093 2 0.1305 0.9184 0.000 0.960 0.004 0.036
#> GSM97058 2 0.1798 0.9146 0.000 0.944 0.016 0.040
#> GSM97051 3 0.7234 0.4735 0.000 0.204 0.544 0.252
#> GSM97052 3 0.1022 0.9035 0.000 0.032 0.968 0.000
#> GSM97061 3 0.1661 0.8968 0.000 0.052 0.944 0.004
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM97138 1 0.2806 0.942 0.844 0.000 0.000 0.004 0.152
#> GSM97145 1 0.4996 0.707 0.708 0.000 0.000 0.128 0.164
#> GSM97147 2 0.3883 0.766 0.004 0.744 0.000 0.244 0.008
#> GSM97125 5 0.4038 0.645 0.128 0.000 0.000 0.080 0.792
#> GSM97127 1 0.4075 0.853 0.780 0.000 0.000 0.060 0.160
#> GSM97130 4 0.3969 0.435 0.004 0.000 0.000 0.692 0.304
#> GSM97133 1 0.2516 0.952 0.860 0.000 0.000 0.000 0.140
#> GSM97134 4 0.1357 0.884 0.004 0.000 0.000 0.948 0.048
#> GSM97120 1 0.2516 0.952 0.860 0.000 0.000 0.000 0.140
#> GSM97126 2 0.4180 0.768 0.000 0.744 0.000 0.220 0.036
#> GSM97112 5 0.0703 0.688 0.024 0.000 0.000 0.000 0.976
#> GSM97115 4 0.1116 0.890 0.004 0.004 0.000 0.964 0.028
#> GSM97116 1 0.2561 0.950 0.856 0.000 0.000 0.000 0.144
#> GSM97117 2 0.0000 0.835 0.000 1.000 0.000 0.000 0.000
#> GSM97119 5 0.0865 0.689 0.024 0.000 0.000 0.004 0.972
#> GSM97122 5 0.0703 0.688 0.024 0.000 0.000 0.000 0.976
#> GSM97135 5 0.0703 0.688 0.024 0.000 0.000 0.000 0.976
#> GSM97136 2 0.3937 0.810 0.000 0.784 0.020 0.184 0.012
#> GSM97139 1 0.2516 0.952 0.860 0.000 0.000 0.000 0.140
#> GSM97146 1 0.2516 0.952 0.860 0.000 0.000 0.000 0.140
#> GSM97123 3 0.3745 0.764 0.000 0.196 0.780 0.024 0.000
#> GSM97129 2 0.3048 0.828 0.000 0.820 0.000 0.176 0.004
#> GSM97143 5 0.3527 0.701 0.024 0.000 0.000 0.172 0.804
#> GSM97113 2 0.0000 0.835 0.000 1.000 0.000 0.000 0.000
#> GSM97056 5 0.5003 0.379 0.032 0.000 0.000 0.424 0.544
#> GSM97124 5 0.3635 0.670 0.004 0.000 0.000 0.248 0.748
#> GSM97132 5 0.4430 0.305 0.004 0.000 0.000 0.456 0.540
#> GSM97144 4 0.1205 0.888 0.004 0.000 0.000 0.956 0.040
#> GSM97149 1 0.2516 0.952 0.860 0.000 0.000 0.000 0.140
#> GSM97068 2 0.5036 0.294 0.004 0.520 0.000 0.452 0.024
#> GSM97071 4 0.1830 0.883 0.068 0.000 0.000 0.924 0.008
#> GSM97086 4 0.2583 0.857 0.132 0.004 0.000 0.864 0.000
#> GSM97103 3 0.5756 0.592 0.000 0.204 0.620 0.176 0.000
#> GSM97057 2 0.2929 0.829 0.000 0.820 0.000 0.180 0.000
#> GSM97060 3 0.3132 0.733 0.000 0.008 0.820 0.172 0.000
#> GSM97075 2 0.0162 0.836 0.000 0.996 0.000 0.004 0.000
#> GSM97098 3 0.3659 0.748 0.000 0.220 0.768 0.012 0.000
#> GSM97099 2 0.0000 0.835 0.000 1.000 0.000 0.000 0.000
#> GSM97101 2 0.0000 0.835 0.000 1.000 0.000 0.000 0.000
#> GSM97105 2 0.3586 0.820 0.000 0.792 0.020 0.188 0.000
#> GSM97106 3 0.3954 0.716 0.000 0.036 0.772 0.192 0.000
#> GSM97121 2 0.2179 0.856 0.000 0.888 0.000 0.112 0.000
#> GSM97128 4 0.0798 0.894 0.008 0.000 0.000 0.976 0.016
#> GSM97131 3 0.7462 0.185 0.100 0.108 0.432 0.360 0.000
#> GSM97137 5 0.4696 0.352 0.016 0.000 0.000 0.428 0.556
#> GSM97118 5 0.4350 0.442 0.004 0.000 0.000 0.408 0.588
#> GSM97114 2 0.0000 0.835 0.000 1.000 0.000 0.000 0.000
#> GSM97142 5 0.0703 0.688 0.024 0.000 0.000 0.000 0.976
#> GSM97140 2 0.2329 0.854 0.000 0.876 0.000 0.124 0.000
#> GSM97141 2 0.0000 0.835 0.000 1.000 0.000 0.000 0.000
#> GSM97055 5 0.3901 0.692 0.004 0.024 0.000 0.196 0.776
#> GSM97090 4 0.1205 0.888 0.004 0.000 0.000 0.956 0.040
#> GSM97091 5 0.0703 0.688 0.024 0.000 0.000 0.000 0.976
#> GSM97148 1 0.2516 0.952 0.860 0.000 0.000 0.000 0.140
#> GSM97063 5 0.0703 0.688 0.024 0.000 0.000 0.000 0.976
#> GSM97053 5 0.4029 0.686 0.024 0.000 0.000 0.232 0.744
#> GSM97066 3 0.0000 0.807 0.000 0.000 1.000 0.000 0.000
#> GSM97079 4 0.2707 0.854 0.132 0.008 0.000 0.860 0.000
#> GSM97083 4 0.1331 0.887 0.008 0.000 0.000 0.952 0.040
#> GSM97084 4 0.2583 0.857 0.132 0.004 0.000 0.864 0.000
#> GSM97094 4 0.0794 0.893 0.000 0.000 0.000 0.972 0.028
#> GSM97096 3 0.2732 0.789 0.000 0.160 0.840 0.000 0.000
#> GSM97097 4 0.2741 0.855 0.132 0.004 0.004 0.860 0.000
#> GSM97107 4 0.0693 0.894 0.008 0.000 0.000 0.980 0.012
#> GSM97054 4 0.2583 0.857 0.132 0.004 0.000 0.864 0.000
#> GSM97062 4 0.2583 0.857 0.132 0.004 0.000 0.864 0.000
#> GSM97069 3 0.0000 0.807 0.000 0.000 1.000 0.000 0.000
#> GSM97070 3 0.0000 0.807 0.000 0.000 1.000 0.000 0.000
#> GSM97073 3 0.0000 0.807 0.000 0.000 1.000 0.000 0.000
#> GSM97076 2 0.4104 0.771 0.000 0.748 0.000 0.220 0.032
#> GSM97077 2 0.2516 0.850 0.000 0.860 0.000 0.140 0.000
#> GSM97095 4 0.1285 0.890 0.004 0.004 0.000 0.956 0.036
#> GSM97102 3 0.0000 0.807 0.000 0.000 1.000 0.000 0.000
#> GSM97109 2 0.0000 0.835 0.000 1.000 0.000 0.000 0.000
#> GSM97110 2 0.0000 0.835 0.000 1.000 0.000 0.000 0.000
#> GSM97074 4 0.1978 0.870 0.004 0.024 0.000 0.928 0.044
#> GSM97085 3 0.4881 0.128 0.004 0.000 0.520 0.460 0.016
#> GSM97059 2 0.4908 0.408 0.004 0.560 0.000 0.416 0.020
#> GSM97072 3 0.0579 0.808 0.000 0.008 0.984 0.008 0.000
#> GSM97078 4 0.0798 0.894 0.008 0.000 0.000 0.976 0.016
#> GSM97067 3 0.0000 0.807 0.000 0.000 1.000 0.000 0.000
#> GSM97087 3 0.0000 0.807 0.000 0.000 1.000 0.000 0.000
#> GSM97111 2 0.0000 0.835 0.000 1.000 0.000 0.000 0.000
#> GSM97064 3 0.6247 0.121 0.000 0.424 0.432 0.144 0.000
#> GSM97065 2 0.0000 0.835 0.000 1.000 0.000 0.000 0.000
#> GSM97081 3 0.3109 0.768 0.000 0.200 0.800 0.000 0.000
#> GSM97082 3 0.0000 0.807 0.000 0.000 1.000 0.000 0.000
#> GSM97088 4 0.0912 0.894 0.012 0.000 0.000 0.972 0.016
#> GSM97100 2 0.4716 0.657 0.036 0.656 0.000 0.308 0.000
#> GSM97104 3 0.0000 0.807 0.000 0.000 1.000 0.000 0.000
#> GSM97108 2 0.2127 0.856 0.000 0.892 0.000 0.108 0.000
#> GSM97050 2 0.2732 0.841 0.000 0.840 0.000 0.160 0.000
#> GSM97080 3 0.0000 0.807 0.000 0.000 1.000 0.000 0.000
#> GSM97089 3 0.5025 0.687 0.000 0.172 0.704 0.124 0.000
#> GSM97092 3 0.2629 0.796 0.000 0.136 0.860 0.004 0.000
#> GSM97093 2 0.2179 0.856 0.000 0.888 0.000 0.112 0.000
#> GSM97058 2 0.3002 0.846 0.000 0.856 0.028 0.116 0.000
#> GSM97051 4 0.5894 0.593 0.132 0.020 0.200 0.648 0.000
#> GSM97052 3 0.2719 0.794 0.000 0.144 0.852 0.004 0.000
#> GSM97061 3 0.3760 0.768 0.000 0.188 0.784 0.028 0.000
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM97138 1 0.1780 0.8867 0.924 0.000 0.000 0.000 0.028 0.048
#> GSM97145 1 0.3176 0.7289 0.812 0.000 0.000 0.000 0.032 0.156
#> GSM97147 2 0.3018 0.7542 0.004 0.816 0.000 0.012 0.000 0.168
#> GSM97125 5 0.5737 0.3300 0.368 0.000 0.000 0.000 0.460 0.172
#> GSM97127 1 0.2257 0.8154 0.876 0.000 0.000 0.000 0.008 0.116
#> GSM97130 6 0.2542 0.8023 0.000 0.000 0.000 0.080 0.044 0.876
#> GSM97133 1 0.0146 0.9390 0.996 0.000 0.000 0.000 0.004 0.000
#> GSM97134 6 0.1714 0.8123 0.000 0.000 0.000 0.092 0.000 0.908
#> GSM97120 1 0.0146 0.9390 0.996 0.000 0.000 0.000 0.004 0.000
#> GSM97126 2 0.4560 0.6174 0.004 0.696 0.000 0.000 0.088 0.212
#> GSM97112 5 0.1152 0.7930 0.004 0.000 0.000 0.000 0.952 0.044
#> GSM97115 6 0.3608 0.6107 0.000 0.012 0.000 0.272 0.000 0.716
#> GSM97116 1 0.0146 0.9390 0.996 0.000 0.000 0.000 0.004 0.000
#> GSM97117 2 0.0146 0.8231 0.000 0.996 0.004 0.000 0.000 0.000
#> GSM97119 5 0.1531 0.7900 0.004 0.000 0.000 0.000 0.928 0.068
#> GSM97122 5 0.1152 0.7930 0.004 0.000 0.000 0.000 0.952 0.044
#> GSM97135 5 0.1152 0.7930 0.004 0.000 0.000 0.000 0.952 0.044
#> GSM97136 2 0.6299 0.1551 0.000 0.408 0.364 0.000 0.016 0.212
#> GSM97139 1 0.0146 0.9390 0.996 0.000 0.000 0.000 0.004 0.000
#> GSM97146 1 0.0146 0.9390 0.996 0.000 0.000 0.000 0.004 0.000
#> GSM97123 2 0.5800 0.0558 0.000 0.460 0.424 0.084 0.032 0.000
#> GSM97129 2 0.2491 0.7613 0.000 0.836 0.000 0.000 0.000 0.164
#> GSM97143 5 0.2964 0.7251 0.004 0.000 0.000 0.000 0.792 0.204
#> GSM97113 2 0.0000 0.8244 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97056 6 0.5575 0.5263 0.276 0.000 0.000 0.032 0.096 0.596
#> GSM97124 5 0.3807 0.5105 0.004 0.000 0.000 0.000 0.628 0.368
#> GSM97132 6 0.3620 0.3435 0.000 0.000 0.000 0.000 0.352 0.648
#> GSM97144 6 0.1663 0.8125 0.000 0.000 0.000 0.088 0.000 0.912
#> GSM97149 1 0.0146 0.9390 0.996 0.000 0.000 0.000 0.004 0.000
#> GSM97068 2 0.4859 0.5967 0.004 0.676 0.000 0.140 0.000 0.180
#> GSM97071 6 0.0935 0.8050 0.000 0.000 0.000 0.032 0.004 0.964
#> GSM97086 4 0.2454 0.7918 0.000 0.000 0.000 0.840 0.000 0.160
#> GSM97103 2 0.6331 0.3941 0.000 0.520 0.308 0.124 0.012 0.036
#> GSM97057 2 0.2768 0.7659 0.000 0.832 0.000 0.012 0.000 0.156
#> GSM97060 3 0.2078 0.8888 0.000 0.004 0.912 0.040 0.044 0.000
#> GSM97075 2 0.0000 0.8244 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97098 3 0.4798 0.4052 0.000 0.348 0.600 0.016 0.036 0.000
#> GSM97099 2 0.0000 0.8244 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97101 2 0.0000 0.8244 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97105 2 0.2901 0.7989 0.000 0.840 0.000 0.128 0.000 0.032
#> GSM97106 3 0.4593 0.7599 0.000 0.088 0.748 0.120 0.044 0.000
#> GSM97121 2 0.2266 0.8092 0.000 0.880 0.000 0.108 0.000 0.012
#> GSM97128 6 0.0603 0.7975 0.000 0.000 0.000 0.016 0.004 0.980
#> GSM97131 4 0.4053 0.6868 0.000 0.140 0.036 0.780 0.000 0.044
#> GSM97137 6 0.5655 0.4996 0.296 0.000 0.000 0.032 0.096 0.576
#> GSM97118 5 0.3221 0.6762 0.000 0.000 0.000 0.000 0.736 0.264
#> GSM97114 2 0.0000 0.8244 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97142 5 0.1152 0.7930 0.004 0.000 0.000 0.000 0.952 0.044
#> GSM97140 2 0.2662 0.7915 0.000 0.856 0.000 0.024 0.000 0.120
#> GSM97141 2 0.0000 0.8244 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97055 5 0.2994 0.7217 0.004 0.000 0.000 0.000 0.788 0.208
#> GSM97090 6 0.1765 0.8093 0.000 0.000 0.000 0.096 0.000 0.904
#> GSM97091 5 0.1152 0.7930 0.004 0.000 0.000 0.000 0.952 0.044
#> GSM97148 1 0.0146 0.9390 0.996 0.000 0.000 0.000 0.004 0.000
#> GSM97063 5 0.1152 0.7930 0.004 0.000 0.000 0.000 0.952 0.044
#> GSM97053 5 0.4389 0.6130 0.052 0.000 0.000 0.000 0.660 0.288
#> GSM97066 3 0.0000 0.9044 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM97079 4 0.1610 0.8013 0.000 0.000 0.000 0.916 0.000 0.084
#> GSM97083 6 0.0291 0.8031 0.000 0.000 0.000 0.004 0.004 0.992
#> GSM97084 4 0.2562 0.7770 0.000 0.000 0.000 0.828 0.000 0.172
#> GSM97094 6 0.1663 0.8125 0.000 0.000 0.000 0.088 0.000 0.912
#> GSM97096 3 0.1820 0.8943 0.000 0.012 0.928 0.016 0.044 0.000
#> GSM97097 4 0.1444 0.7954 0.000 0.000 0.000 0.928 0.000 0.072
#> GSM97107 6 0.1663 0.8125 0.000 0.000 0.000 0.088 0.000 0.912
#> GSM97054 4 0.2527 0.7839 0.000 0.000 0.000 0.832 0.000 0.168
#> GSM97062 4 0.2491 0.7889 0.000 0.000 0.000 0.836 0.000 0.164
#> GSM97069 3 0.0000 0.9044 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM97070 3 0.0000 0.9044 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM97073 3 0.0000 0.9044 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM97076 5 0.5803 0.4021 0.004 0.260 0.000 0.000 0.524 0.212
#> GSM97077 2 0.2812 0.8098 0.000 0.856 0.000 0.096 0.000 0.048
#> GSM97095 6 0.2775 0.7821 0.000 0.040 0.000 0.104 0.000 0.856
#> GSM97102 3 0.0000 0.9044 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM97109 2 0.0000 0.8244 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97110 2 0.0000 0.8244 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97074 6 0.4795 -0.0422 0.000 0.000 0.056 0.000 0.400 0.544
#> GSM97085 3 0.3725 0.5223 0.000 0.000 0.676 0.000 0.008 0.316
#> GSM97059 2 0.3885 0.7020 0.000 0.756 0.000 0.064 0.000 0.180
#> GSM97072 3 0.1713 0.8928 0.000 0.000 0.928 0.028 0.044 0.000
#> GSM97078 6 0.0603 0.7975 0.000 0.000 0.000 0.016 0.004 0.980
#> GSM97067 3 0.0000 0.9044 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM97087 3 0.0000 0.9044 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM97111 2 0.0000 0.8244 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97064 2 0.4039 0.7481 0.000 0.776 0.072 0.136 0.000 0.016
#> GSM97065 2 0.0146 0.8231 0.000 0.996 0.004 0.000 0.000 0.000
#> GSM97081 3 0.1793 0.8906 0.000 0.036 0.928 0.004 0.032 0.000
#> GSM97082 3 0.0000 0.9044 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM97088 6 0.0717 0.7982 0.000 0.000 0.016 0.000 0.008 0.976
#> GSM97100 4 0.4858 0.3230 0.004 0.348 0.000 0.588 0.000 0.060
#> GSM97104 3 0.0000 0.9044 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM97108 2 0.2494 0.8059 0.000 0.864 0.000 0.120 0.000 0.016
#> GSM97050 2 0.2706 0.8034 0.000 0.852 0.000 0.124 0.000 0.024
#> GSM97080 3 0.0000 0.9044 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM97089 3 0.3344 0.8149 0.000 0.016 0.844 0.008 0.044 0.088
#> GSM97092 3 0.2007 0.8902 0.000 0.004 0.916 0.036 0.044 0.000
#> GSM97093 2 0.2346 0.7912 0.000 0.868 0.000 0.008 0.000 0.124
#> GSM97058 2 0.2890 0.8013 0.000 0.844 0.004 0.128 0.000 0.024
#> GSM97051 4 0.2294 0.7922 0.000 0.036 0.000 0.892 0.000 0.072
#> GSM97052 3 0.2007 0.8902 0.000 0.004 0.916 0.036 0.044 0.000
#> GSM97061 3 0.4821 0.6897 0.000 0.180 0.712 0.064 0.044 0.000
Heatmaps for the consensus matrix. It visualizes the probability of two samples to be in a same group.
consensus_heatmap(res, k = 2)
consensus_heatmap(res, k = 3)
consensus_heatmap(res, k = 4)
consensus_heatmap(res, k = 5)
consensus_heatmap(res, k = 6)
Heatmaps for the membership of samples in all partitions to see how consistent they are:
membership_heatmap(res, k = 2)
membership_heatmap(res, k = 3)
membership_heatmap(res, k = 4)
membership_heatmap(res, k = 5)
membership_heatmap(res, k = 6)
As soon as we have had the classes for columns, we can look for signatures which are significantly different between classes which can be candidate marks for certain classes. Following are the heatmaps for signatures.
Signature heatmaps where rows are scaled:
get_signatures(res, k = 2)
get_signatures(res, k = 3)
get_signatures(res, k = 4)
get_signatures(res, k = 5)
get_signatures(res, k = 6)
Signature heatmaps where rows are not scaled:
get_signatures(res, k = 2, scale_rows = FALSE)
get_signatures(res, k = 3, scale_rows = FALSE)
get_signatures(res, k = 4, scale_rows = FALSE)
get_signatures(res, k = 5, scale_rows = FALSE)
get_signatures(res, k = 6, scale_rows = FALSE)
Compare the overlap of signatures from different k:
compare_signatures(res)
get_signature() returns a data frame invisibly. TO get the list of signatures, the function
call should be assigned to a variable explicitly. In following code, if plot argument is set
to FALSE, no heatmap is plotted while only the differential analysis is performed.
# code only for demonstration
tb = get_signature(res, k = ..., plot = FALSE)
An example of the output of tb is:
#> which_row fdr mean_1 mean_2 scaled_mean_1 scaled_mean_2 km
#> 1 38 0.042760348 8.373488 9.131774 -0.5533452 0.5164555 1
#> 2 40 0.018707592 7.106213 8.469186 -0.6173731 0.5762149 1
#> 3 55 0.019134737 10.221463 11.207825 -0.6159697 0.5749050 1
#> 4 59 0.006059896 5.921854 7.869574 -0.6899429 0.6439467 1
#> 5 60 0.018055526 8.928898 10.211722 -0.6204761 0.5791110 1
#> 6 98 0.009384629 15.714769 14.887706 0.6635654 -0.6193277 2
...
The columns in tb are:
which_row: row indices corresponding to the input matrix.fdr: FDR for the differential test. mean_x: The mean value in group x.scaled_mean_x: The mean value in group x after rows are scaled.km: Row groups if k-means clustering is applied to rows.UMAP plot which shows how samples are separated.
dimension_reduction(res, k = 2, method = "UMAP")
dimension_reduction(res, k = 3, method = "UMAP")
dimension_reduction(res, k = 4, method = "UMAP")
dimension_reduction(res, k = 5, method = "UMAP")
dimension_reduction(res, k = 6, method = "UMAP")
Following heatmap shows how subgroups are split when increasing k:
collect_classes(res)
Test correlation between subgroups and known annotations. If the known annotation is numeric, one-way ANOVA test is applied, and if the known annotation is discrete, chi-squared contingency table test is applied.
test_to_known_factors(res)
#> n disease.state(p) specimen(p) cell.type(p) other(p) k
#> SD:mclust 68 0.002023 0.1868 1.59e-05 0.247 2
#> SD:mclust 93 0.001622 0.0534 3.99e-10 0.178 3
#> SD:mclust 98 0.000289 0.6454 1.37e-15 0.239 4
#> SD:mclust 90 0.000347 0.6504 9.82e-13 0.346 5
#> SD:mclust 90 0.000104 0.3602 4.01e-11 0.298 6
If matrix rows can be associated to genes, consider to use functional_enrichment(res,
...) to perform function enrichment for the signature genes. See this vignette for more detailed explanations.
The object with results only for a single top-value method and a single partition method can be extracted as:
res = res_list["SD", "NMF"]
# you can also extract it by
# res = res_list["SD:NMF"]
A summary of res and all the functions that can be applied to it:
res
#> A 'ConsensusPartition' object with k = 2, 3, 4, 5, 6.
#> On a matrix with 21512 rows and 100 columns.
#> Top rows (1000, 2000, 3000, 4000, 5000) are extracted by 'SD' method.
#> Subgroups are detected by 'NMF' method.
#> Performed in total 1250 partitions by row resampling.
#> Best k for subgroups seems to be 2.
#>
#> Following methods can be applied to this 'ConsensusPartition' object:
#> [1] "cola_report" "collect_classes" "collect_plots"
#> [4] "collect_stats" "colnames" "compare_signatures"
#> [7] "consensus_heatmap" "dimension_reduction" "functional_enrichment"
#> [10] "get_anno_col" "get_anno" "get_classes"
#> [13] "get_consensus" "get_matrix" "get_membership"
#> [16] "get_param" "get_signatures" "get_stats"
#> [19] "is_best_k" "is_stable_k" "membership_heatmap"
#> [22] "ncol" "nrow" "plot_ecdf"
#> [25] "rownames" "select_partition_number" "show"
#> [28] "suggest_best_k" "test_to_known_factors"
collect_plots() function collects all the plots made from res for all k (number of partitions)
into one single page to provide an easy and fast comparison between different k.
collect_plots(res)
The plots are:
k and the heatmap of
predicted classes for each k.k.k.k.All the plots in panels can be made by individual functions and they are plotted later in this section.
select_partition_number() produces several plots showing different
statistics for choosing “optimized” k. There are following statistics:
k;k, the area increased is defined as \(A_k - A_{k-1}\).The detailed explanations of these statistics can be found in the cola vignette.
Generally speaking, lower PAC score, higher mean silhouette score or higher
concordance corresponds to better partition. Rand index and Jaccard index
measure how similar the current partition is compared to partition with k-1.
If they are too similar, we won't accept k is better than k-1.
select_partition_number(res)
The numeric values for all these statistics can be obtained by get_stats().
get_stats(res)
#> k 1-PAC mean_silhouette concordance area_increased Rand Jaccard
#> 2 2 0.917 0.934 0.973 0.4991 0.500 0.500
#> 3 3 0.432 0.484 0.662 0.3074 0.779 0.589
#> 4 4 0.602 0.586 0.754 0.1313 0.670 0.311
#> 5 5 0.550 0.434 0.694 0.0780 0.819 0.445
#> 6 6 0.643 0.536 0.729 0.0458 0.864 0.452
suggest_best_k() suggests the best \(k\) based on these statistics. The rules are as follows:
suggest_best_k(res)
#> [1] 2
Following shows the table of the partitions (You need to click the show/hide
code output link to see it). The membership matrix (columns with name p*)
is inferred by
clue::cl_consensus()
function with the SE method. Basically the value in the membership matrix
represents the probability to belong to a certain group. The finall class
label for an item is determined with the group with highest probability it
belongs to.
In get_classes() function, the entropy is calculated from the membership
matrix and the silhouette score is calculated from the consensus matrix.
cbind(get_classes(res, k = 2), get_membership(res, k = 2))
#> class entropy silhouette p1 p2
#> GSM97138 1 0.0000 0.968 1.000 0.000
#> GSM97145 1 0.0000 0.968 1.000 0.000
#> GSM97147 1 0.0000 0.968 1.000 0.000
#> GSM97125 1 0.0000 0.968 1.000 0.000
#> GSM97127 1 0.0000 0.968 1.000 0.000
#> GSM97130 1 0.0000 0.968 1.000 0.000
#> GSM97133 1 0.0000 0.968 1.000 0.000
#> GSM97134 1 0.0000 0.968 1.000 0.000
#> GSM97120 1 0.0000 0.968 1.000 0.000
#> GSM97126 1 0.0000 0.968 1.000 0.000
#> GSM97112 1 0.0000 0.968 1.000 0.000
#> GSM97115 1 0.0000 0.968 1.000 0.000
#> GSM97116 1 0.0000 0.968 1.000 0.000
#> GSM97117 2 0.0000 0.973 0.000 1.000
#> GSM97119 1 0.0000 0.968 1.000 0.000
#> GSM97122 1 0.0000 0.968 1.000 0.000
#> GSM97135 1 0.0000 0.968 1.000 0.000
#> GSM97136 2 0.3584 0.913 0.068 0.932
#> GSM97139 1 0.0000 0.968 1.000 0.000
#> GSM97146 1 0.0000 0.968 1.000 0.000
#> GSM97123 2 0.0000 0.973 0.000 1.000
#> GSM97129 2 0.9170 0.505 0.332 0.668
#> GSM97143 1 0.0000 0.968 1.000 0.000
#> GSM97113 2 0.6531 0.795 0.168 0.832
#> GSM97056 1 0.0000 0.968 1.000 0.000
#> GSM97124 1 0.0000 0.968 1.000 0.000
#> GSM97132 1 0.0000 0.968 1.000 0.000
#> GSM97144 1 0.0000 0.968 1.000 0.000
#> GSM97149 1 0.0000 0.968 1.000 0.000
#> GSM97068 1 0.2603 0.929 0.956 0.044
#> GSM97071 2 0.0000 0.973 0.000 1.000
#> GSM97086 2 0.0000 0.973 0.000 1.000
#> GSM97103 2 0.0000 0.973 0.000 1.000
#> GSM97057 1 0.7453 0.725 0.788 0.212
#> GSM97060 2 0.0000 0.973 0.000 1.000
#> GSM97075 2 0.0000 0.973 0.000 1.000
#> GSM97098 2 0.0000 0.973 0.000 1.000
#> GSM97099 2 0.0000 0.973 0.000 1.000
#> GSM97101 2 0.0000 0.973 0.000 1.000
#> GSM97105 2 0.0000 0.973 0.000 1.000
#> GSM97106 2 0.0000 0.973 0.000 1.000
#> GSM97121 2 0.0000 0.973 0.000 1.000
#> GSM97128 1 0.7745 0.699 0.772 0.228
#> GSM97131 2 0.0000 0.973 0.000 1.000
#> GSM97137 1 0.0000 0.968 1.000 0.000
#> GSM97118 1 0.0000 0.968 1.000 0.000
#> GSM97114 1 0.0672 0.962 0.992 0.008
#> GSM97142 1 0.0000 0.968 1.000 0.000
#> GSM97140 2 0.9209 0.500 0.336 0.664
#> GSM97141 2 0.0000 0.973 0.000 1.000
#> GSM97055 1 0.0000 0.968 1.000 0.000
#> GSM97090 1 0.0000 0.968 1.000 0.000
#> GSM97091 1 0.0000 0.968 1.000 0.000
#> GSM97148 1 0.0000 0.968 1.000 0.000
#> GSM97063 1 0.0000 0.968 1.000 0.000
#> GSM97053 1 0.0000 0.968 1.000 0.000
#> GSM97066 2 0.0000 0.973 0.000 1.000
#> GSM97079 2 0.0000 0.973 0.000 1.000
#> GSM97083 1 0.0000 0.968 1.000 0.000
#> GSM97084 2 0.3431 0.917 0.064 0.936
#> GSM97094 1 0.0000 0.968 1.000 0.000
#> GSM97096 2 0.0000 0.973 0.000 1.000
#> GSM97097 2 0.0000 0.973 0.000 1.000
#> GSM97107 1 0.0376 0.965 0.996 0.004
#> GSM97054 2 0.4815 0.875 0.104 0.896
#> GSM97062 2 0.0000 0.973 0.000 1.000
#> GSM97069 2 0.0000 0.973 0.000 1.000
#> GSM97070 2 0.0000 0.973 0.000 1.000
#> GSM97073 2 0.0000 0.973 0.000 1.000
#> GSM97076 1 0.0376 0.965 0.996 0.004
#> GSM97077 2 0.0000 0.973 0.000 1.000
#> GSM97095 1 0.0000 0.968 1.000 0.000
#> GSM97102 2 0.0000 0.973 0.000 1.000
#> GSM97109 2 0.9087 0.524 0.324 0.676
#> GSM97110 2 0.0000 0.973 0.000 1.000
#> GSM97074 1 0.9896 0.211 0.560 0.440
#> GSM97085 2 0.0000 0.973 0.000 1.000
#> GSM97059 1 0.0000 0.968 1.000 0.000
#> GSM97072 2 0.0000 0.973 0.000 1.000
#> GSM97078 1 0.9661 0.354 0.608 0.392
#> GSM97067 2 0.0000 0.973 0.000 1.000
#> GSM97087 2 0.0000 0.973 0.000 1.000
#> GSM97111 2 0.0000 0.973 0.000 1.000
#> GSM97064 2 0.0000 0.973 0.000 1.000
#> GSM97065 2 0.0000 0.973 0.000 1.000
#> GSM97081 2 0.0000 0.973 0.000 1.000
#> GSM97082 2 0.0000 0.973 0.000 1.000
#> GSM97088 2 0.0000 0.973 0.000 1.000
#> GSM97100 2 0.0000 0.973 0.000 1.000
#> GSM97104 2 0.0000 0.973 0.000 1.000
#> GSM97108 2 0.0000 0.973 0.000 1.000
#> GSM97050 2 0.0000 0.973 0.000 1.000
#> GSM97080 2 0.0000 0.973 0.000 1.000
#> GSM97089 2 0.0000 0.973 0.000 1.000
#> GSM97092 2 0.0000 0.973 0.000 1.000
#> GSM97093 2 0.0938 0.963 0.012 0.988
#> GSM97058 2 0.0000 0.973 0.000 1.000
#> GSM97051 2 0.0000 0.973 0.000 1.000
#> GSM97052 2 0.0000 0.973 0.000 1.000
#> GSM97061 2 0.0000 0.973 0.000 1.000
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM97138 1 0.2261 0.7238 0.932 0.000 0.068
#> GSM97145 1 0.1643 0.7331 0.956 0.000 0.044
#> GSM97147 1 0.5948 0.4629 0.640 0.360 0.000
#> GSM97125 1 0.3267 0.7003 0.884 0.000 0.116
#> GSM97127 1 0.1765 0.7396 0.956 0.040 0.004
#> GSM97130 1 0.5223 0.6955 0.800 0.176 0.024
#> GSM97133 1 0.4121 0.6970 0.832 0.168 0.000
#> GSM97134 1 0.3181 0.7375 0.912 0.024 0.064
#> GSM97120 1 0.1182 0.7400 0.976 0.012 0.012
#> GSM97126 1 0.2400 0.7301 0.932 0.004 0.064
#> GSM97112 1 0.6235 0.3396 0.564 0.000 0.436
#> GSM97115 1 0.6647 0.3900 0.592 0.396 0.012
#> GSM97116 1 0.0892 0.7383 0.980 0.000 0.020
#> GSM97117 2 0.5397 0.4711 0.000 0.720 0.280
#> GSM97119 1 0.5529 0.5562 0.704 0.000 0.296
#> GSM97122 1 0.5465 0.5651 0.712 0.000 0.288
#> GSM97135 1 0.5363 0.5773 0.724 0.000 0.276
#> GSM97136 3 0.3369 0.5361 0.052 0.040 0.908
#> GSM97139 1 0.0661 0.7402 0.988 0.008 0.004
#> GSM97146 1 0.1267 0.7381 0.972 0.004 0.024
#> GSM97123 2 0.4931 0.5232 0.000 0.768 0.232
#> GSM97129 2 0.7309 0.5572 0.168 0.708 0.124
#> GSM97143 1 0.5254 0.5865 0.736 0.000 0.264
#> GSM97113 2 0.5254 0.4999 0.264 0.736 0.000
#> GSM97056 1 0.3120 0.7345 0.908 0.080 0.012
#> GSM97124 1 0.3686 0.6925 0.860 0.000 0.140
#> GSM97132 1 0.3038 0.7141 0.896 0.000 0.104
#> GSM97144 1 0.4799 0.7189 0.836 0.132 0.032
#> GSM97149 1 0.4452 0.6763 0.808 0.192 0.000
#> GSM97068 2 0.6521 -0.1259 0.492 0.504 0.004
#> GSM97071 2 0.6204 0.1991 0.000 0.576 0.424
#> GSM97086 2 0.3618 0.6439 0.104 0.884 0.012
#> GSM97103 2 0.5397 0.4736 0.000 0.720 0.280
#> GSM97057 2 0.6432 0.0903 0.428 0.568 0.004
#> GSM97060 2 0.6215 0.1656 0.000 0.572 0.428
#> GSM97075 2 0.5291 0.4861 0.000 0.732 0.268
#> GSM97098 2 0.5560 0.4415 0.000 0.700 0.300
#> GSM97099 2 0.2448 0.6391 0.000 0.924 0.076
#> GSM97101 2 0.4047 0.6282 0.148 0.848 0.004
#> GSM97105 2 0.2066 0.6573 0.060 0.940 0.000
#> GSM97106 2 0.5431 0.4681 0.000 0.716 0.284
#> GSM97121 2 0.3192 0.6439 0.112 0.888 0.000
#> GSM97128 3 0.4682 0.3678 0.192 0.004 0.804
#> GSM97131 2 0.1753 0.6485 0.000 0.952 0.048
#> GSM97137 1 0.4033 0.7169 0.856 0.136 0.008
#> GSM97118 1 0.6235 0.3472 0.564 0.000 0.436
#> GSM97114 1 0.6235 0.2856 0.564 0.436 0.000
#> GSM97142 1 0.6062 0.4323 0.616 0.000 0.384
#> GSM97140 2 0.6079 0.2188 0.388 0.612 0.000
#> GSM97141 2 0.4602 0.6309 0.152 0.832 0.016
#> GSM97055 3 0.6180 -0.0524 0.416 0.000 0.584
#> GSM97090 1 0.6570 0.5510 0.668 0.308 0.024
#> GSM97091 3 0.6215 -0.1002 0.428 0.000 0.572
#> GSM97148 1 0.2066 0.7358 0.940 0.060 0.000
#> GSM97063 3 0.6299 -0.2055 0.476 0.000 0.524
#> GSM97053 1 0.3116 0.7071 0.892 0.000 0.108
#> GSM97066 3 0.5291 0.4588 0.000 0.268 0.732
#> GSM97079 2 0.2050 0.6593 0.020 0.952 0.028
#> GSM97083 3 0.6944 -0.2338 0.468 0.016 0.516
#> GSM97084 2 0.6226 0.4759 0.252 0.720 0.028
#> GSM97094 1 0.5403 0.7231 0.816 0.124 0.060
#> GSM97096 2 0.6180 0.1960 0.000 0.584 0.416
#> GSM97097 2 0.2165 0.6463 0.000 0.936 0.064
#> GSM97107 1 0.6698 0.5849 0.684 0.280 0.036
#> GSM97054 2 0.6090 0.4659 0.264 0.716 0.020
#> GSM97062 2 0.3550 0.6504 0.080 0.896 0.024
#> GSM97069 3 0.5968 0.3855 0.000 0.364 0.636
#> GSM97070 3 0.6154 0.3115 0.000 0.408 0.592
#> GSM97073 3 0.6180 0.2957 0.000 0.416 0.584
#> GSM97076 1 0.6521 0.2145 0.504 0.004 0.492
#> GSM97077 2 0.3030 0.6506 0.092 0.904 0.004
#> GSM97095 1 0.6879 0.4660 0.616 0.360 0.024
#> GSM97102 3 0.5882 0.4055 0.000 0.348 0.652
#> GSM97109 2 0.6738 0.3239 0.356 0.624 0.020
#> GSM97110 2 0.2689 0.6612 0.036 0.932 0.032
#> GSM97074 3 0.3551 0.4488 0.132 0.000 0.868
#> GSM97085 3 0.0848 0.5335 0.008 0.008 0.984
#> GSM97059 1 0.6410 0.3322 0.576 0.420 0.004
#> GSM97072 2 0.6244 0.1298 0.000 0.560 0.440
#> GSM97078 3 0.4749 0.3884 0.172 0.012 0.816
#> GSM97067 3 0.5621 0.4384 0.000 0.308 0.692
#> GSM97087 3 0.6204 0.2762 0.000 0.424 0.576
#> GSM97111 2 0.3272 0.6263 0.004 0.892 0.104
#> GSM97064 2 0.3412 0.6104 0.000 0.876 0.124
#> GSM97065 2 0.6180 0.3917 0.008 0.660 0.332
#> GSM97081 2 0.6274 0.0710 0.000 0.544 0.456
#> GSM97082 3 0.5785 0.4210 0.000 0.332 0.668
#> GSM97088 3 0.2116 0.5335 0.040 0.012 0.948
#> GSM97100 2 0.4883 0.5733 0.208 0.788 0.004
#> GSM97104 3 0.5926 0.3966 0.000 0.356 0.644
#> GSM97108 2 0.3619 0.6330 0.136 0.864 0.000
#> GSM97050 2 0.2063 0.6603 0.044 0.948 0.008
#> GSM97080 3 0.6204 0.2770 0.000 0.424 0.576
#> GSM97089 3 0.6274 0.1877 0.000 0.456 0.544
#> GSM97092 2 0.6045 0.2896 0.000 0.620 0.380
#> GSM97093 2 0.3856 0.6601 0.072 0.888 0.040
#> GSM97058 2 0.1163 0.6547 0.000 0.972 0.028
#> GSM97051 2 0.1453 0.6577 0.008 0.968 0.024
#> GSM97052 2 0.5882 0.3550 0.000 0.652 0.348
#> GSM97061 2 0.5178 0.4994 0.000 0.744 0.256
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM97138 1 0.4998 -0.4564 0.512 0.488 0.000 0.000
#> GSM97145 2 0.5105 0.5670 0.432 0.564 0.004 0.000
#> GSM97147 2 0.6471 0.5806 0.340 0.592 0.052 0.016
#> GSM97125 1 0.4406 0.1738 0.700 0.300 0.000 0.000
#> GSM97127 2 0.4948 0.5585 0.440 0.560 0.000 0.000
#> GSM97130 4 0.3037 0.8544 0.100 0.020 0.000 0.880
#> GSM97133 2 0.4925 0.5693 0.428 0.572 0.000 0.000
#> GSM97134 4 0.4328 0.6978 0.244 0.008 0.000 0.748
#> GSM97120 2 0.5112 0.5633 0.436 0.560 0.004 0.000
#> GSM97126 2 0.5168 0.4485 0.496 0.500 0.004 0.000
#> GSM97112 1 0.0188 0.6127 0.996 0.004 0.000 0.000
#> GSM97115 4 0.2131 0.8846 0.032 0.036 0.000 0.932
#> GSM97116 1 0.4977 -0.3783 0.540 0.460 0.000 0.000
#> GSM97117 3 0.2081 0.7168 0.000 0.084 0.916 0.000
#> GSM97119 1 0.1398 0.5985 0.956 0.040 0.000 0.004
#> GSM97122 1 0.1902 0.5872 0.932 0.064 0.000 0.004
#> GSM97135 1 0.2266 0.5731 0.912 0.084 0.000 0.004
#> GSM97136 3 0.6993 0.5321 0.148 0.296 0.556 0.000
#> GSM97139 2 0.4977 0.5272 0.460 0.540 0.000 0.000
#> GSM97146 2 0.4977 0.5272 0.460 0.540 0.000 0.000
#> GSM97123 3 0.0927 0.7369 0.000 0.008 0.976 0.016
#> GSM97129 3 0.6346 0.4224 0.116 0.244 0.640 0.000
#> GSM97143 1 0.1867 0.5821 0.928 0.072 0.000 0.000
#> GSM97113 2 0.5696 0.1864 0.024 0.496 0.480 0.000
#> GSM97056 1 0.7769 0.0502 0.432 0.272 0.000 0.296
#> GSM97124 1 0.3108 0.5502 0.872 0.112 0.000 0.016
#> GSM97132 1 0.4139 0.5618 0.816 0.040 0.000 0.144
#> GSM97144 4 0.2675 0.8617 0.100 0.008 0.000 0.892
#> GSM97149 2 0.5161 0.5787 0.400 0.592 0.008 0.000
#> GSM97068 4 0.2744 0.8734 0.024 0.052 0.012 0.912
#> GSM97071 4 0.0188 0.8963 0.004 0.000 0.000 0.996
#> GSM97086 4 0.0336 0.8947 0.000 0.000 0.008 0.992
#> GSM97103 3 0.3245 0.7419 0.000 0.056 0.880 0.064
#> GSM97057 2 0.6466 0.5279 0.092 0.588 0.320 0.000
#> GSM97060 3 0.5292 0.7072 0.000 0.208 0.728 0.064
#> GSM97075 3 0.0707 0.7357 0.000 0.020 0.980 0.000
#> GSM97098 3 0.0524 0.7392 0.000 0.008 0.988 0.004
#> GSM97099 3 0.2216 0.7037 0.000 0.092 0.908 0.000
#> GSM97101 3 0.5151 -0.0753 0.004 0.464 0.532 0.000
#> GSM97105 3 0.5815 0.5506 0.000 0.140 0.708 0.152
#> GSM97106 3 0.2882 0.7354 0.000 0.024 0.892 0.084
#> GSM97121 3 0.5371 0.2494 0.000 0.364 0.616 0.020
#> GSM97128 1 0.7439 0.3921 0.532 0.264 0.004 0.200
#> GSM97131 3 0.5097 0.2859 0.000 0.004 0.568 0.428
#> GSM97137 1 0.7569 -0.1797 0.436 0.368 0.000 0.196
#> GSM97118 1 0.4364 0.6081 0.808 0.136 0.000 0.056
#> GSM97114 2 0.6950 0.5590 0.180 0.584 0.236 0.000
#> GSM97142 1 0.0524 0.6132 0.988 0.008 0.000 0.004
#> GSM97140 2 0.5708 0.3571 0.028 0.556 0.416 0.000
#> GSM97141 2 0.5296 0.1448 0.008 0.500 0.492 0.000
#> GSM97055 1 0.4399 0.5845 0.760 0.224 0.016 0.000
#> GSM97090 4 0.1798 0.8901 0.040 0.016 0.000 0.944
#> GSM97091 1 0.3907 0.5885 0.768 0.232 0.000 0.000
#> GSM97148 2 0.4961 0.5473 0.448 0.552 0.000 0.000
#> GSM97063 1 0.2814 0.6158 0.868 0.132 0.000 0.000
#> GSM97053 1 0.3498 0.4866 0.832 0.160 0.000 0.008
#> GSM97066 3 0.6568 0.4865 0.080 0.408 0.512 0.000
#> GSM97079 4 0.0336 0.8947 0.000 0.000 0.008 0.992
#> GSM97083 4 0.4018 0.7203 0.224 0.004 0.000 0.772
#> GSM97084 4 0.0000 0.8959 0.000 0.000 0.000 1.000
#> GSM97094 4 0.1211 0.8912 0.040 0.000 0.000 0.960
#> GSM97096 3 0.2125 0.7433 0.000 0.076 0.920 0.004
#> GSM97097 4 0.0707 0.8887 0.000 0.000 0.020 0.980
#> GSM97107 4 0.0592 0.8956 0.016 0.000 0.000 0.984
#> GSM97054 4 0.0188 0.8959 0.000 0.000 0.004 0.996
#> GSM97062 4 0.0188 0.8959 0.000 0.000 0.004 0.996
#> GSM97069 3 0.5846 0.5912 0.032 0.372 0.592 0.004
#> GSM97070 3 0.4594 0.6845 0.008 0.280 0.712 0.000
#> GSM97073 3 0.4509 0.6822 0.004 0.288 0.708 0.000
#> GSM97076 1 0.3450 0.6101 0.836 0.156 0.008 0.000
#> GSM97077 3 0.5307 0.5623 0.000 0.188 0.736 0.076
#> GSM97095 4 0.2179 0.8814 0.064 0.012 0.000 0.924
#> GSM97102 3 0.4814 0.6643 0.008 0.316 0.676 0.000
#> GSM97109 2 0.6087 0.3620 0.048 0.540 0.412 0.000
#> GSM97110 3 0.3569 0.6094 0.000 0.196 0.804 0.000
#> GSM97074 1 0.5933 0.4598 0.552 0.408 0.040 0.000
#> GSM97085 1 0.6862 0.3786 0.488 0.408 0.104 0.000
#> GSM97059 2 0.8459 0.4223 0.184 0.524 0.072 0.220
#> GSM97072 3 0.4098 0.7158 0.000 0.204 0.784 0.012
#> GSM97078 4 0.5050 0.6188 0.268 0.028 0.000 0.704
#> GSM97067 3 0.5905 0.5586 0.040 0.396 0.564 0.000
#> GSM97087 3 0.4262 0.7041 0.008 0.236 0.756 0.000
#> GSM97111 3 0.2011 0.7133 0.000 0.080 0.920 0.000
#> GSM97064 3 0.1936 0.7275 0.000 0.028 0.940 0.032
#> GSM97065 3 0.1824 0.7239 0.004 0.060 0.936 0.000
#> GSM97081 3 0.1398 0.7443 0.004 0.040 0.956 0.000
#> GSM97082 3 0.5306 0.6295 0.020 0.348 0.632 0.000
#> GSM97088 1 0.7026 0.4632 0.540 0.372 0.036 0.052
#> GSM97100 4 0.5074 0.6068 0.000 0.040 0.236 0.724
#> GSM97104 3 0.5220 0.6302 0.016 0.352 0.632 0.000
#> GSM97108 3 0.5143 0.2733 0.000 0.360 0.628 0.012
#> GSM97050 3 0.4966 0.6056 0.000 0.156 0.768 0.076
#> GSM97080 3 0.4456 0.6859 0.004 0.280 0.716 0.000
#> GSM97089 3 0.3791 0.7165 0.004 0.200 0.796 0.000
#> GSM97092 3 0.2329 0.7443 0.000 0.072 0.916 0.012
#> GSM97093 3 0.4163 0.5666 0.004 0.220 0.772 0.004
#> GSM97058 3 0.3312 0.7011 0.000 0.052 0.876 0.072
#> GSM97051 4 0.2149 0.8330 0.000 0.000 0.088 0.912
#> GSM97052 3 0.1854 0.7447 0.000 0.048 0.940 0.012
#> GSM97061 3 0.1305 0.7381 0.000 0.004 0.960 0.036
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM97138 1 0.388 0.5404 0.776 0.000 0.032 0.000 0.192
#> GSM97145 1 0.461 0.5730 0.760 0.008 0.092 0.000 0.140
#> GSM97147 1 0.678 0.4901 0.612 0.192 0.140 0.036 0.020
#> GSM97125 1 0.464 0.3336 0.660 0.000 0.032 0.000 0.308
#> GSM97127 1 0.257 0.6382 0.892 0.000 0.040 0.000 0.068
#> GSM97130 4 0.239 0.8023 0.044 0.000 0.004 0.908 0.044
#> GSM97133 1 0.131 0.6552 0.956 0.000 0.024 0.000 0.020
#> GSM97134 4 0.517 0.5760 0.040 0.004 0.020 0.688 0.248
#> GSM97120 1 0.136 0.6527 0.952 0.000 0.012 0.000 0.036
#> GSM97126 1 0.451 0.4846 0.728 0.012 0.020 0.004 0.236
#> GSM97112 5 0.447 0.5181 0.292 0.000 0.020 0.004 0.684
#> GSM97115 4 0.386 0.7874 0.076 0.076 0.008 0.832 0.008
#> GSM97116 1 0.214 0.6281 0.904 0.000 0.008 0.000 0.088
#> GSM97117 2 0.593 0.2880 0.100 0.536 0.360 0.000 0.004
#> GSM97119 5 0.477 0.4649 0.340 0.000 0.024 0.004 0.632
#> GSM97122 5 0.495 0.4423 0.356 0.000 0.024 0.008 0.612
#> GSM97135 5 0.506 0.3405 0.412 0.000 0.028 0.004 0.556
#> GSM97136 5 0.743 -0.2044 0.052 0.204 0.292 0.000 0.452
#> GSM97139 1 0.120 0.6496 0.952 0.000 0.000 0.000 0.048
#> GSM97146 1 0.153 0.6483 0.948 0.004 0.012 0.000 0.036
#> GSM97123 2 0.304 0.4961 0.000 0.840 0.148 0.004 0.008
#> GSM97129 1 0.816 0.1773 0.376 0.220 0.284 0.000 0.120
#> GSM97143 5 0.474 0.4160 0.380 0.000 0.016 0.004 0.600
#> GSM97113 1 0.471 0.3571 0.684 0.268 0.048 0.000 0.000
#> GSM97056 1 0.630 0.3146 0.592 0.008 0.016 0.272 0.112
#> GSM97124 5 0.575 0.4098 0.356 0.000 0.040 0.032 0.572
#> GSM97132 5 0.653 0.4248 0.232 0.000 0.016 0.196 0.556
#> GSM97144 4 0.144 0.8128 0.004 0.000 0.004 0.948 0.044
#> GSM97149 1 0.111 0.6460 0.964 0.024 0.012 0.000 0.000
#> GSM97068 4 0.476 0.7414 0.104 0.116 0.012 0.764 0.004
#> GSM97071 4 0.160 0.8164 0.000 0.008 0.024 0.948 0.020
#> GSM97086 4 0.225 0.7921 0.000 0.012 0.088 0.900 0.000
#> GSM97103 3 0.353 0.4872 0.000 0.076 0.832 0.092 0.000
#> GSM97057 1 0.500 -0.0400 0.508 0.468 0.016 0.008 0.000
#> GSM97060 2 0.613 -0.2269 0.000 0.476 0.416 0.008 0.100
#> GSM97075 2 0.350 0.4905 0.016 0.816 0.160 0.000 0.008
#> GSM97098 3 0.316 0.5042 0.000 0.164 0.824 0.012 0.000
#> GSM97099 3 0.505 0.3810 0.068 0.248 0.680 0.004 0.000
#> GSM97101 2 0.629 0.3198 0.332 0.500 0.168 0.000 0.000
#> GSM97105 2 0.593 0.4085 0.048 0.632 0.260 0.060 0.000
#> GSM97106 3 0.498 0.1884 0.000 0.484 0.488 0.028 0.000
#> GSM97121 2 0.683 0.2908 0.156 0.480 0.340 0.024 0.000
#> GSM97128 5 0.488 0.3743 0.000 0.032 0.024 0.236 0.708
#> GSM97131 2 0.679 0.1206 0.000 0.376 0.284 0.340 0.000
#> GSM97137 1 0.495 0.4849 0.732 0.000 0.012 0.164 0.092
#> GSM97118 5 0.409 0.5736 0.096 0.000 0.008 0.092 0.804
#> GSM97114 1 0.545 0.5146 0.668 0.128 0.200 0.000 0.004
#> GSM97142 5 0.451 0.5176 0.284 0.000 0.024 0.004 0.688
#> GSM97140 2 0.561 0.4393 0.244 0.648 0.096 0.012 0.000
#> GSM97141 2 0.651 0.2536 0.348 0.452 0.200 0.000 0.000
#> GSM97055 5 0.302 0.5783 0.088 0.012 0.028 0.000 0.872
#> GSM97090 4 0.478 0.7767 0.072 0.076 0.012 0.792 0.048
#> GSM97091 5 0.239 0.5863 0.104 0.000 0.004 0.004 0.888
#> GSM97148 1 0.110 0.6492 0.968 0.008 0.012 0.000 0.012
#> GSM97063 5 0.289 0.5784 0.160 0.000 0.004 0.000 0.836
#> GSM97053 1 0.529 -0.1124 0.516 0.000 0.008 0.032 0.444
#> GSM97066 5 0.691 -0.3794 0.000 0.336 0.276 0.004 0.384
#> GSM97079 4 0.378 0.6328 0.000 0.008 0.252 0.740 0.000
#> GSM97083 4 0.377 0.7132 0.000 0.008 0.012 0.780 0.200
#> GSM97084 4 0.164 0.8054 0.000 0.004 0.064 0.932 0.000
#> GSM97094 4 0.385 0.7306 0.000 0.000 0.172 0.788 0.040
#> GSM97096 3 0.372 0.5204 0.000 0.208 0.776 0.004 0.012
#> GSM97097 3 0.465 -0.1220 0.000 0.012 0.520 0.468 0.000
#> GSM97107 4 0.213 0.8023 0.000 0.000 0.080 0.908 0.012
#> GSM97054 4 0.217 0.8029 0.000 0.088 0.004 0.904 0.004
#> GSM97062 4 0.157 0.8141 0.000 0.012 0.032 0.948 0.008
#> GSM97069 3 0.671 0.3187 0.000 0.296 0.424 0.000 0.280
#> GSM97070 3 0.656 0.2684 0.000 0.372 0.424 0.000 0.204
#> GSM97073 3 0.494 0.5347 0.000 0.176 0.720 0.004 0.100
#> GSM97076 3 0.592 0.2549 0.072 0.004 0.632 0.028 0.264
#> GSM97077 2 0.434 0.5118 0.080 0.812 0.040 0.064 0.004
#> GSM97095 4 0.409 0.7905 0.040 0.060 0.008 0.832 0.060
#> GSM97102 3 0.517 0.5269 0.000 0.184 0.688 0.000 0.128
#> GSM97109 3 0.496 0.3580 0.176 0.084 0.728 0.012 0.000
#> GSM97110 3 0.408 0.4990 0.076 0.104 0.808 0.012 0.000
#> GSM97074 5 0.359 0.4688 0.004 0.040 0.100 0.012 0.844
#> GSM97085 5 0.476 0.2642 0.000 0.140 0.128 0.000 0.732
#> GSM97059 1 0.720 0.0886 0.424 0.352 0.012 0.200 0.012
#> GSM97072 3 0.474 0.5314 0.000 0.216 0.724 0.012 0.048
#> GSM97078 4 0.488 0.5318 0.000 0.020 0.012 0.636 0.332
#> GSM97067 3 0.666 0.3827 0.000 0.236 0.480 0.004 0.280
#> GSM97087 2 0.457 0.3706 0.000 0.748 0.148 0.000 0.104
#> GSM97111 3 0.517 0.0725 0.048 0.376 0.576 0.000 0.000
#> GSM97064 2 0.177 0.5263 0.016 0.944 0.024 0.012 0.004
#> GSM97065 3 0.589 0.2578 0.096 0.360 0.540 0.000 0.004
#> GSM97081 2 0.485 0.3414 0.000 0.692 0.240 0.000 0.068
#> GSM97082 2 0.645 -0.0203 0.000 0.500 0.228 0.000 0.272
#> GSM97088 5 0.535 0.4295 0.000 0.128 0.036 0.112 0.724
#> GSM97100 4 0.656 0.0029 0.016 0.416 0.128 0.440 0.000
#> GSM97104 3 0.663 0.3366 0.000 0.332 0.436 0.000 0.232
#> GSM97108 2 0.673 0.3291 0.140 0.512 0.320 0.028 0.000
#> GSM97050 2 0.483 0.5102 0.108 0.776 0.072 0.040 0.004
#> GSM97080 2 0.621 -0.0263 0.000 0.532 0.296 0.000 0.172
#> GSM97089 2 0.422 0.3817 0.000 0.772 0.156 0.000 0.072
#> GSM97092 2 0.367 0.4371 0.000 0.812 0.140 0.000 0.048
#> GSM97093 2 0.372 0.5105 0.160 0.808 0.024 0.004 0.004
#> GSM97058 2 0.265 0.5327 0.028 0.900 0.052 0.020 0.000
#> GSM97051 2 0.494 -0.0862 0.004 0.516 0.012 0.464 0.004
#> GSM97052 2 0.306 0.4708 0.000 0.860 0.096 0.000 0.044
#> GSM97061 2 0.255 0.4968 0.000 0.896 0.072 0.004 0.028
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM97138 1 0.5290 -0.0452 0.468 0.048 0.000 0.000 0.460 0.024
#> GSM97145 5 0.6706 0.2356 0.212 0.272 0.000 0.000 0.460 0.056
#> GSM97147 2 0.4612 0.5764 0.080 0.752 0.000 0.012 0.132 0.024
#> GSM97125 5 0.4430 0.5889 0.196 0.052 0.000 0.000 0.728 0.024
#> GSM97127 1 0.5963 0.1653 0.488 0.108 0.000 0.000 0.372 0.032
#> GSM97130 4 0.2132 0.8237 0.020 0.008 0.000 0.920 0.032 0.020
#> GSM97133 1 0.1856 0.8204 0.920 0.032 0.000 0.000 0.048 0.000
#> GSM97134 5 0.5327 0.1155 0.000 0.044 0.012 0.448 0.484 0.012
#> GSM97120 1 0.1477 0.8296 0.940 0.004 0.000 0.000 0.048 0.008
#> GSM97126 5 0.4448 0.6086 0.176 0.080 0.000 0.008 0.732 0.004
#> GSM97112 5 0.1261 0.7297 0.028 0.004 0.008 0.000 0.956 0.004
#> GSM97115 4 0.3258 0.7974 0.104 0.012 0.020 0.848 0.008 0.008
#> GSM97116 1 0.1285 0.8301 0.944 0.004 0.000 0.000 0.052 0.000
#> GSM97117 2 0.3991 0.6487 0.032 0.820 0.036 0.000 0.052 0.060
#> GSM97119 5 0.1737 0.7291 0.040 0.020 0.000 0.008 0.932 0.000
#> GSM97122 5 0.2077 0.7265 0.056 0.012 0.008 0.008 0.916 0.000
#> GSM97135 5 0.2144 0.7149 0.092 0.004 0.000 0.004 0.896 0.004
#> GSM97136 5 0.6074 0.3485 0.008 0.040 0.256 0.000 0.576 0.120
#> GSM97139 1 0.1349 0.8283 0.940 0.004 0.000 0.000 0.056 0.000
#> GSM97146 1 0.0547 0.8327 0.980 0.000 0.000 0.000 0.020 0.000
#> GSM97123 2 0.4951 0.0551 0.008 0.476 0.476 0.004 0.000 0.036
#> GSM97129 2 0.5895 0.2166 0.056 0.548 0.012 0.000 0.336 0.048
#> GSM97143 5 0.1956 0.7203 0.080 0.008 0.000 0.004 0.908 0.000
#> GSM97113 1 0.2713 0.7616 0.884 0.036 0.040 0.000 0.000 0.040
#> GSM97056 1 0.3252 0.7219 0.832 0.004 0.008 0.124 0.032 0.000
#> GSM97124 5 0.3531 0.7123 0.052 0.052 0.004 0.032 0.848 0.012
#> GSM97132 5 0.3783 0.6766 0.024 0.004 0.016 0.160 0.792 0.004
#> GSM97144 4 0.2245 0.8199 0.000 0.008 0.004 0.908 0.052 0.028
#> GSM97149 1 0.0291 0.8280 0.992 0.004 0.000 0.000 0.004 0.000
#> GSM97068 4 0.3733 0.7376 0.180 0.020 0.016 0.780 0.004 0.000
#> GSM97071 4 0.3690 0.7864 0.000 0.012 0.024 0.824 0.040 0.100
#> GSM97086 4 0.2218 0.8077 0.000 0.012 0.000 0.884 0.000 0.104
#> GSM97103 6 0.3699 0.6121 0.000 0.088 0.028 0.068 0.000 0.816
#> GSM97057 1 0.3327 0.7256 0.844 0.076 0.060 0.016 0.000 0.004
#> GSM97060 3 0.4407 0.3698 0.004 0.020 0.660 0.012 0.000 0.304
#> GSM97075 2 0.4749 0.3995 0.008 0.612 0.332 0.000 0.000 0.048
#> GSM97098 6 0.3915 0.6263 0.000 0.092 0.128 0.004 0.000 0.776
#> GSM97099 6 0.6507 0.4748 0.060 0.200 0.180 0.000 0.008 0.552
#> GSM97101 2 0.3851 0.6538 0.096 0.804 0.072 0.000 0.000 0.028
#> GSM97105 2 0.1950 0.6621 0.012 0.928 0.032 0.008 0.000 0.020
#> GSM97106 3 0.5838 0.0256 0.004 0.096 0.468 0.020 0.000 0.412
#> GSM97121 2 0.2987 0.6525 0.024 0.876 0.004 0.008 0.028 0.060
#> GSM97128 5 0.5911 0.2517 0.000 0.000 0.180 0.316 0.496 0.008
#> GSM97131 2 0.3550 0.6432 0.000 0.816 0.044 0.120 0.000 0.020
#> GSM97137 1 0.2406 0.7941 0.896 0.004 0.004 0.060 0.036 0.000
#> GSM97118 5 0.2933 0.7073 0.000 0.000 0.056 0.076 0.860 0.008
#> GSM97114 2 0.5365 0.5103 0.216 0.652 0.000 0.000 0.044 0.088
#> GSM97142 5 0.0806 0.7300 0.020 0.008 0.000 0.000 0.972 0.000
#> GSM97140 2 0.3627 0.6441 0.036 0.820 0.112 0.028 0.004 0.000
#> GSM97141 2 0.3291 0.6594 0.084 0.848 0.028 0.000 0.004 0.036
#> GSM97055 5 0.3533 0.6286 0.008 0.000 0.196 0.000 0.776 0.020
#> GSM97090 4 0.4163 0.7777 0.096 0.016 0.064 0.800 0.020 0.004
#> GSM97091 5 0.1970 0.7066 0.000 0.000 0.092 0.000 0.900 0.008
#> GSM97148 1 0.0363 0.8312 0.988 0.000 0.000 0.000 0.012 0.000
#> GSM97063 5 0.1655 0.7197 0.008 0.000 0.052 0.000 0.932 0.008
#> GSM97053 5 0.4160 0.6088 0.228 0.008 0.012 0.016 0.732 0.004
#> GSM97066 3 0.5213 0.3488 0.000 0.032 0.640 0.000 0.072 0.256
#> GSM97079 4 0.3905 0.5059 0.000 0.004 0.004 0.636 0.000 0.356
#> GSM97083 4 0.3663 0.7046 0.000 0.000 0.040 0.776 0.180 0.004
#> GSM97084 4 0.2400 0.8032 0.000 0.004 0.008 0.872 0.000 0.116
#> GSM97094 4 0.4471 0.5774 0.000 0.020 0.000 0.648 0.020 0.312
#> GSM97096 6 0.4062 0.5793 0.000 0.068 0.196 0.000 0.000 0.736
#> GSM97097 6 0.4731 0.2087 0.000 0.044 0.008 0.344 0.000 0.604
#> GSM97107 4 0.2593 0.7889 0.000 0.008 0.000 0.844 0.000 0.148
#> GSM97054 4 0.1794 0.8095 0.000 0.036 0.040 0.924 0.000 0.000
#> GSM97062 4 0.1769 0.8180 0.000 0.012 0.004 0.924 0.000 0.060
#> GSM97069 3 0.5035 0.3146 0.000 0.028 0.612 0.000 0.044 0.316
#> GSM97070 3 0.5129 0.2687 0.000 0.076 0.576 0.000 0.008 0.340
#> GSM97073 6 0.4200 0.4746 0.000 0.032 0.264 0.000 0.008 0.696
#> GSM97076 6 0.5542 0.5268 0.016 0.040 0.076 0.016 0.152 0.700
#> GSM97077 2 0.4960 0.5143 0.008 0.656 0.264 0.060 0.000 0.012
#> GSM97095 4 0.3282 0.8012 0.020 0.044 0.048 0.864 0.020 0.004
#> GSM97102 6 0.4319 0.4390 0.000 0.024 0.320 0.000 0.008 0.648
#> GSM97109 6 0.4724 0.5374 0.036 0.208 0.000 0.008 0.036 0.712
#> GSM97110 6 0.3748 0.6409 0.056 0.068 0.048 0.004 0.000 0.824
#> GSM97074 5 0.5910 0.2477 0.000 0.000 0.292 0.008 0.508 0.192
#> GSM97085 3 0.5002 0.0835 0.000 0.000 0.516 0.000 0.412 0.072
#> GSM97059 2 0.6706 0.3896 0.216 0.472 0.060 0.252 0.000 0.000
#> GSM97072 6 0.3719 0.4941 0.000 0.024 0.248 0.000 0.000 0.728
#> GSM97078 4 0.5228 0.5886 0.000 0.004 0.128 0.656 0.200 0.012
#> GSM97067 3 0.4656 0.1603 0.000 0.012 0.552 0.004 0.016 0.416
#> GSM97087 3 0.3258 0.5221 0.008 0.092 0.836 0.000 0.000 0.064
#> GSM97111 2 0.4266 0.5942 0.032 0.776 0.020 0.000 0.028 0.144
#> GSM97064 3 0.5350 -0.0890 0.012 0.412 0.516 0.048 0.000 0.012
#> GSM97065 6 0.6939 0.3073 0.120 0.200 0.192 0.000 0.000 0.488
#> GSM97081 2 0.5399 0.1963 0.000 0.528 0.360 0.000 0.004 0.108
#> GSM97082 3 0.3916 0.4994 0.000 0.040 0.804 0.000 0.072 0.084
#> GSM97088 5 0.6351 0.2241 0.000 0.000 0.344 0.212 0.424 0.020
#> GSM97100 2 0.3354 0.6255 0.000 0.796 0.036 0.168 0.000 0.000
#> GSM97104 3 0.4321 0.3205 0.000 0.012 0.652 0.000 0.020 0.316
#> GSM97108 2 0.2676 0.6652 0.020 0.900 0.024 0.008 0.028 0.020
#> GSM97050 2 0.6576 0.2276 0.040 0.456 0.392 0.080 0.004 0.028
#> GSM97080 3 0.3986 0.4706 0.000 0.036 0.756 0.000 0.016 0.192
#> GSM97089 3 0.3564 0.5167 0.012 0.084 0.824 0.004 0.000 0.076
#> GSM97092 3 0.4357 0.4267 0.008 0.244 0.704 0.004 0.000 0.040
#> GSM97093 3 0.6458 0.0516 0.108 0.336 0.504 0.016 0.008 0.028
#> GSM97058 2 0.4962 0.4272 0.008 0.608 0.332 0.040 0.000 0.012
#> GSM97051 2 0.5927 0.3730 0.004 0.484 0.208 0.304 0.000 0.000
#> GSM97052 3 0.4497 0.3204 0.008 0.284 0.672 0.012 0.000 0.024
#> GSM97061 3 0.4978 0.1038 0.008 0.372 0.576 0.020 0.000 0.024
Heatmaps for the consensus matrix. It visualizes the probability of two samples to be in a same group.
consensus_heatmap(res, k = 2)
consensus_heatmap(res, k = 3)
consensus_heatmap(res, k = 4)
consensus_heatmap(res, k = 5)
consensus_heatmap(res, k = 6)
Heatmaps for the membership of samples in all partitions to see how consistent they are:
membership_heatmap(res, k = 2)
membership_heatmap(res, k = 3)
membership_heatmap(res, k = 4)
membership_heatmap(res, k = 5)
membership_heatmap(res, k = 6)
As soon as we have had the classes for columns, we can look for signatures which are significantly different between classes which can be candidate marks for certain classes. Following are the heatmaps for signatures.
Signature heatmaps where rows are scaled:
get_signatures(res, k = 2)
get_signatures(res, k = 3)
get_signatures(res, k = 4)
get_signatures(res, k = 5)
get_signatures(res, k = 6)
Signature heatmaps where rows are not scaled:
get_signatures(res, k = 2, scale_rows = FALSE)
get_signatures(res, k = 3, scale_rows = FALSE)
get_signatures(res, k = 4, scale_rows = FALSE)
get_signatures(res, k = 5, scale_rows = FALSE)
get_signatures(res, k = 6, scale_rows = FALSE)
Compare the overlap of signatures from different k:
compare_signatures(res)
get_signature() returns a data frame invisibly. TO get the list of signatures, the function
call should be assigned to a variable explicitly. In following code, if plot argument is set
to FALSE, no heatmap is plotted while only the differential analysis is performed.
# code only for demonstration
tb = get_signature(res, k = ..., plot = FALSE)
An example of the output of tb is:
#> which_row fdr mean_1 mean_2 scaled_mean_1 scaled_mean_2 km
#> 1 38 0.042760348 8.373488 9.131774 -0.5533452 0.5164555 1
#> 2 40 0.018707592 7.106213 8.469186 -0.6173731 0.5762149 1
#> 3 55 0.019134737 10.221463 11.207825 -0.6159697 0.5749050 1
#> 4 59 0.006059896 5.921854 7.869574 -0.6899429 0.6439467 1
#> 5 60 0.018055526 8.928898 10.211722 -0.6204761 0.5791110 1
#> 6 98 0.009384629 15.714769 14.887706 0.6635654 -0.6193277 2
...
The columns in tb are:
which_row: row indices corresponding to the input matrix.fdr: FDR for the differential test. mean_x: The mean value in group x.scaled_mean_x: The mean value in group x after rows are scaled.km: Row groups if k-means clustering is applied to rows.UMAP plot which shows how samples are separated.
dimension_reduction(res, k = 2, method = "UMAP")
dimension_reduction(res, k = 3, method = "UMAP")
dimension_reduction(res, k = 4, method = "UMAP")
dimension_reduction(res, k = 5, method = "UMAP")
dimension_reduction(res, k = 6, method = "UMAP")
Following heatmap shows how subgroups are split when increasing k:
collect_classes(res)
Test correlation between subgroups and known annotations. If the known annotation is numeric, one-way ANOVA test is applied, and if the known annotation is discrete, chi-squared contingency table test is applied.
test_to_known_factors(res)
#> n disease.state(p) specimen(p) cell.type(p) other(p) k
#> SD:NMF 97 7.18e-05 0.738 3.31e-13 0.149 2
#> SD:NMF 52 1.69e-02 0.437 4.61e-08 0.614 3
#> SD:NMF 78 1.01e-03 0.993 5.54e-12 0.604 4
#> SD:NMF 44 2.78e-03 0.476 8.31e-09 0.410 5
#> SD:NMF 63 5.59e-02 0.869 2.11e-11 0.121 6
If matrix rows can be associated to genes, consider to use functional_enrichment(res,
...) to perform function enrichment for the signature genes. See this vignette for more detailed explanations.
The object with results only for a single top-value method and a single partition method can be extracted as:
res = res_list["CV", "hclust"]
# you can also extract it by
# res = res_list["CV:hclust"]
A summary of res and all the functions that can be applied to it:
res
#> A 'ConsensusPartition' object with k = 2, 3, 4, 5, 6.
#> On a matrix with 21512 rows and 100 columns.
#> Top rows (1000, 2000, 3000, 4000, 5000) are extracted by 'CV' method.
#> Subgroups are detected by 'hclust' method.
#> Performed in total 1250 partitions by row resampling.
#> Best k for subgroups seems to be 2.
#>
#> Following methods can be applied to this 'ConsensusPartition' object:
#> [1] "cola_report" "collect_classes" "collect_plots"
#> [4] "collect_stats" "colnames" "compare_signatures"
#> [7] "consensus_heatmap" "dimension_reduction" "functional_enrichment"
#> [10] "get_anno_col" "get_anno" "get_classes"
#> [13] "get_consensus" "get_matrix" "get_membership"
#> [16] "get_param" "get_signatures" "get_stats"
#> [19] "is_best_k" "is_stable_k" "membership_heatmap"
#> [22] "ncol" "nrow" "plot_ecdf"
#> [25] "rownames" "select_partition_number" "show"
#> [28] "suggest_best_k" "test_to_known_factors"
collect_plots() function collects all the plots made from res for all k (number of partitions)
into one single page to provide an easy and fast comparison between different k.
collect_plots(res)
The plots are:
k and the heatmap of
predicted classes for each k.k.k.k.All the plots in panels can be made by individual functions and they are plotted later in this section.
select_partition_number() produces several plots showing different
statistics for choosing “optimized” k. There are following statistics:
k;k, the area increased is defined as \(A_k - A_{k-1}\).The detailed explanations of these statistics can be found in the cola vignette.
Generally speaking, lower PAC score, higher mean silhouette score or higher
concordance corresponds to better partition. Rand index and Jaccard index
measure how similar the current partition is compared to partition with k-1.
If they are too similar, we won't accept k is better than k-1.
select_partition_number(res)
The numeric values for all these statistics can be obtained by get_stats().
get_stats(res)
#> k 1-PAC mean_silhouette concordance area_increased Rand Jaccard
#> 2 2 0.426 0.786 0.886 0.4182 0.547 0.547
#> 3 3 0.297 0.657 0.801 0.3678 0.890 0.804
#> 4 4 0.345 0.588 0.747 0.1706 0.886 0.759
#> 5 5 0.406 0.453 0.685 0.0911 0.921 0.786
#> 6 6 0.473 0.528 0.698 0.0463 0.929 0.767
suggest_best_k() suggests the best \(k\) based on these statistics. The rules are as follows:
suggest_best_k(res)
#> [1] 2
Following shows the table of the partitions (You need to click the show/hide
code output link to see it). The membership matrix (columns with name p*)
is inferred by
clue::cl_consensus()
function with the SE method. Basically the value in the membership matrix
represents the probability to belong to a certain group. The finall class
label for an item is determined with the group with highest probability it
belongs to.
In get_classes() function, the entropy is calculated from the membership
matrix and the silhouette score is calculated from the consensus matrix.
cbind(get_classes(res, k = 2), get_membership(res, k = 2))
#> class entropy silhouette p1 p2
#> GSM97138 1 0.4161 0.809 0.916 0.084
#> GSM97145 1 0.3733 0.806 0.928 0.072
#> GSM97147 2 0.7453 0.725 0.212 0.788
#> GSM97125 1 0.3733 0.806 0.928 0.072
#> GSM97127 1 0.3879 0.807 0.924 0.076
#> GSM97130 1 0.8144 0.728 0.748 0.252
#> GSM97133 1 0.2043 0.792 0.968 0.032
#> GSM97134 1 0.9988 0.313 0.520 0.480
#> GSM97120 1 0.2236 0.794 0.964 0.036
#> GSM97126 1 0.9954 0.376 0.540 0.460
#> GSM97112 1 0.6343 0.810 0.840 0.160
#> GSM97115 2 1.0000 -0.227 0.496 0.504
#> GSM97116 1 0.2236 0.794 0.964 0.036
#> GSM97117 2 0.6048 0.819 0.148 0.852
#> GSM97119 1 0.6343 0.810 0.840 0.160
#> GSM97122 1 0.6343 0.810 0.840 0.160
#> GSM97135 1 0.6247 0.810 0.844 0.156
#> GSM97136 2 0.9954 -0.169 0.460 0.540
#> GSM97139 1 0.1633 0.788 0.976 0.024
#> GSM97146 1 0.0000 0.774 1.000 0.000
#> GSM97123 2 0.0000 0.913 0.000 1.000
#> GSM97129 1 0.9988 0.313 0.520 0.480
#> GSM97143 1 0.8909 0.692 0.692 0.308
#> GSM97113 1 0.9970 0.244 0.532 0.468
#> GSM97056 1 0.1184 0.785 0.984 0.016
#> GSM97124 1 0.6438 0.808 0.836 0.164
#> GSM97132 1 0.8608 0.721 0.716 0.284
#> GSM97144 1 0.9129 0.656 0.672 0.328
#> GSM97149 1 0.0000 0.774 1.000 0.000
#> GSM97068 1 0.9795 0.460 0.584 0.416
#> GSM97071 2 0.1633 0.914 0.024 0.976
#> GSM97086 2 0.0000 0.913 0.000 1.000
#> GSM97103 2 0.0672 0.914 0.008 0.992
#> GSM97057 1 0.9775 0.408 0.588 0.412
#> GSM97060 2 0.0000 0.913 0.000 1.000
#> GSM97075 2 0.2948 0.905 0.052 0.948
#> GSM97098 2 0.0672 0.914 0.008 0.992
#> GSM97099 2 0.5059 0.859 0.112 0.888
#> GSM97101 2 0.4939 0.862 0.108 0.892
#> GSM97105 2 0.1414 0.914 0.020 0.980
#> GSM97106 2 0.0000 0.913 0.000 1.000
#> GSM97121 2 0.6048 0.812 0.148 0.852
#> GSM97128 2 0.3733 0.888 0.072 0.928
#> GSM97131 2 0.0000 0.913 0.000 1.000
#> GSM97137 1 0.6343 0.785 0.840 0.160
#> GSM97118 1 0.9427 0.619 0.640 0.360
#> GSM97114 2 0.6048 0.819 0.148 0.852
#> GSM97142 1 0.6343 0.810 0.840 0.160
#> GSM97140 2 0.4022 0.888 0.080 0.920
#> GSM97141 2 0.6048 0.819 0.148 0.852
#> GSM97055 2 0.8267 0.611 0.260 0.740
#> GSM97090 2 0.9983 -0.160 0.476 0.524
#> GSM97091 1 0.6531 0.807 0.832 0.168
#> GSM97148 1 0.0000 0.774 1.000 0.000
#> GSM97063 1 0.6531 0.807 0.832 0.168
#> GSM97053 1 0.5408 0.812 0.876 0.124
#> GSM97066 2 0.1414 0.914 0.020 0.980
#> GSM97079 2 0.0376 0.914 0.004 0.996
#> GSM97083 2 0.3733 0.888 0.072 0.928
#> GSM97084 2 0.0000 0.913 0.000 1.000
#> GSM97094 2 0.4298 0.867 0.088 0.912
#> GSM97096 2 0.0672 0.914 0.008 0.992
#> GSM97097 2 0.0000 0.913 0.000 1.000
#> GSM97107 2 0.3431 0.888 0.064 0.936
#> GSM97054 2 0.1633 0.914 0.024 0.976
#> GSM97062 2 0.0000 0.913 0.000 1.000
#> GSM97069 2 0.1414 0.914 0.020 0.980
#> GSM97070 2 0.1414 0.914 0.020 0.980
#> GSM97073 2 0.1414 0.914 0.020 0.980
#> GSM97076 2 0.3114 0.904 0.056 0.944
#> GSM97077 2 0.3114 0.903 0.056 0.944
#> GSM97095 1 0.9970 0.310 0.532 0.468
#> GSM97102 2 0.0672 0.914 0.008 0.992
#> GSM97109 2 0.2603 0.909 0.044 0.956
#> GSM97110 2 0.2603 0.909 0.044 0.956
#> GSM97074 2 0.5842 0.811 0.140 0.860
#> GSM97085 2 0.2778 0.903 0.048 0.952
#> GSM97059 2 0.8144 0.643 0.252 0.748
#> GSM97072 2 0.0000 0.913 0.000 1.000
#> GSM97078 2 0.3733 0.888 0.072 0.928
#> GSM97067 2 0.1414 0.914 0.020 0.980
#> GSM97087 2 0.0376 0.914 0.004 0.996
#> GSM97111 2 0.3114 0.905 0.056 0.944
#> GSM97064 2 0.0672 0.915 0.008 0.992
#> GSM97065 2 0.3584 0.900 0.068 0.932
#> GSM97081 2 0.2236 0.913 0.036 0.964
#> GSM97082 2 0.0000 0.913 0.000 1.000
#> GSM97088 2 0.2948 0.901 0.052 0.948
#> GSM97100 2 0.2778 0.906 0.048 0.952
#> GSM97104 2 0.0000 0.913 0.000 1.000
#> GSM97108 2 0.4690 0.867 0.100 0.900
#> GSM97050 2 0.0376 0.914 0.004 0.996
#> GSM97080 2 0.0672 0.914 0.008 0.992
#> GSM97089 2 0.0376 0.914 0.004 0.996
#> GSM97092 2 0.0000 0.913 0.000 1.000
#> GSM97093 2 0.6973 0.744 0.188 0.812
#> GSM97058 2 0.2236 0.911 0.036 0.964
#> GSM97051 2 0.0000 0.913 0.000 1.000
#> GSM97052 2 0.0000 0.913 0.000 1.000
#> GSM97061 2 0.0000 0.913 0.000 1.000
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM97138 1 0.286 0.7065 0.912 0.004 0.084
#> GSM97145 1 0.241 0.7094 0.940 0.020 0.040
#> GSM97147 2 0.583 0.6445 0.204 0.764 0.032
#> GSM97125 1 0.253 0.7098 0.936 0.020 0.044
#> GSM97127 1 0.255 0.7093 0.936 0.024 0.040
#> GSM97130 1 0.678 0.5768 0.732 0.188 0.080
#> GSM97133 1 0.162 0.6990 0.964 0.012 0.024
#> GSM97134 1 0.846 0.2442 0.528 0.376 0.096
#> GSM97120 1 0.127 0.7000 0.972 0.004 0.024
#> GSM97126 1 0.882 0.3000 0.540 0.324 0.136
#> GSM97112 1 0.445 0.6924 0.836 0.012 0.152
#> GSM97115 1 0.814 0.1611 0.480 0.452 0.068
#> GSM97116 1 0.186 0.6983 0.948 0.000 0.052
#> GSM97117 2 0.547 0.7259 0.140 0.808 0.052
#> GSM97119 1 0.445 0.6924 0.836 0.012 0.152
#> GSM97122 1 0.445 0.6924 0.836 0.012 0.152
#> GSM97135 1 0.439 0.6936 0.840 0.012 0.148
#> GSM97136 1 0.960 0.1291 0.460 0.320 0.220
#> GSM97139 1 0.140 0.6939 0.968 0.004 0.028
#> GSM97146 1 0.175 0.6856 0.952 0.000 0.048
#> GSM97123 2 0.186 0.7978 0.000 0.948 0.052
#> GSM97129 1 0.846 0.2442 0.528 0.376 0.096
#> GSM97143 1 0.730 0.5866 0.688 0.084 0.228
#> GSM97113 1 0.757 0.1202 0.512 0.448 0.040
#> GSM97056 1 0.175 0.6987 0.952 0.000 0.048
#> GSM97124 1 0.462 0.6959 0.836 0.020 0.144
#> GSM97132 1 0.677 0.6184 0.724 0.068 0.208
#> GSM97144 1 0.771 0.5161 0.660 0.240 0.100
#> GSM97149 1 0.175 0.6856 0.952 0.000 0.048
#> GSM97068 1 0.774 0.2942 0.568 0.376 0.056
#> GSM97071 2 0.611 0.6676 0.044 0.756 0.200
#> GSM97086 2 0.236 0.7933 0.000 0.928 0.072
#> GSM97103 2 0.392 0.7838 0.004 0.856 0.140
#> GSM97057 1 0.744 0.2225 0.568 0.392 0.040
#> GSM97060 2 0.355 0.7765 0.000 0.868 0.132
#> GSM97075 2 0.415 0.7995 0.044 0.876 0.080
#> GSM97098 2 0.385 0.7863 0.004 0.860 0.136
#> GSM97099 2 0.482 0.7619 0.108 0.844 0.048
#> GSM97101 2 0.474 0.7642 0.104 0.848 0.048
#> GSM97105 2 0.191 0.7973 0.016 0.956 0.028
#> GSM97106 2 0.254 0.7878 0.000 0.920 0.080
#> GSM97121 2 0.493 0.7276 0.140 0.828 0.032
#> GSM97128 3 0.543 0.8413 0.048 0.144 0.808
#> GSM97131 2 0.175 0.7976 0.000 0.952 0.048
#> GSM97137 1 0.524 0.6488 0.824 0.120 0.056
#> GSM97118 1 0.792 0.5198 0.636 0.100 0.264
#> GSM97114 2 0.547 0.7259 0.140 0.808 0.052
#> GSM97142 1 0.445 0.6924 0.836 0.012 0.152
#> GSM97140 2 0.376 0.7847 0.068 0.892 0.040
#> GSM97141 2 0.527 0.7289 0.140 0.816 0.044
#> GSM97055 3 0.917 0.6068 0.248 0.212 0.540
#> GSM97090 2 0.821 -0.1949 0.460 0.468 0.072
#> GSM97091 1 0.466 0.6893 0.828 0.016 0.156
#> GSM97148 1 0.175 0.6856 0.952 0.000 0.048
#> GSM97063 1 0.466 0.6893 0.828 0.016 0.156
#> GSM97053 1 0.364 0.7011 0.872 0.004 0.124
#> GSM97066 2 0.581 0.5957 0.004 0.692 0.304
#> GSM97079 2 0.250 0.7950 0.004 0.928 0.068
#> GSM97083 3 0.547 0.8364 0.052 0.140 0.808
#> GSM97084 2 0.303 0.7881 0.004 0.904 0.092
#> GSM97094 2 0.631 0.6862 0.100 0.772 0.128
#> GSM97096 2 0.385 0.7849 0.004 0.860 0.136
#> GSM97097 2 0.286 0.7923 0.004 0.912 0.084
#> GSM97107 2 0.530 0.7421 0.068 0.824 0.108
#> GSM97054 2 0.611 0.6676 0.044 0.756 0.200
#> GSM97062 2 0.268 0.7927 0.004 0.920 0.076
#> GSM97069 2 0.537 0.6781 0.004 0.744 0.252
#> GSM97070 2 0.581 0.5957 0.004 0.692 0.304
#> GSM97073 2 0.596 0.5961 0.008 0.692 0.300
#> GSM97076 2 0.716 0.5153 0.044 0.640 0.316
#> GSM97077 2 0.334 0.7924 0.060 0.908 0.032
#> GSM97095 1 0.781 0.1830 0.512 0.436 0.052
#> GSM97102 2 0.392 0.7838 0.004 0.856 0.140
#> GSM97109 2 0.493 0.7814 0.044 0.836 0.120
#> GSM97110 2 0.493 0.7814 0.044 0.836 0.120
#> GSM97074 2 0.900 -0.0733 0.136 0.488 0.376
#> GSM97085 3 0.638 0.7926 0.032 0.256 0.712
#> GSM97059 2 0.638 0.5599 0.244 0.720 0.036
#> GSM97072 2 0.440 0.7504 0.000 0.812 0.188
#> GSM97078 3 0.543 0.8413 0.048 0.144 0.808
#> GSM97067 2 0.578 0.6030 0.004 0.696 0.300
#> GSM97087 2 0.475 0.7537 0.008 0.808 0.184
#> GSM97111 2 0.514 0.7774 0.052 0.828 0.120
#> GSM97064 2 0.217 0.8029 0.008 0.944 0.048
#> GSM97065 2 0.689 0.6372 0.060 0.704 0.236
#> GSM97081 2 0.492 0.7773 0.036 0.832 0.132
#> GSM97082 2 0.501 0.7422 0.008 0.788 0.204
#> GSM97088 3 0.696 0.7305 0.040 0.300 0.660
#> GSM97100 2 0.292 0.7950 0.044 0.924 0.032
#> GSM97104 2 0.394 0.7749 0.000 0.844 0.156
#> GSM97108 2 0.459 0.7694 0.096 0.856 0.048
#> GSM97050 2 0.116 0.7994 0.000 0.972 0.028
#> GSM97080 2 0.455 0.7489 0.000 0.800 0.200
#> GSM97089 2 0.475 0.7537 0.008 0.808 0.184
#> GSM97092 2 0.388 0.7762 0.000 0.848 0.152
#> GSM97093 2 0.782 0.5234 0.176 0.672 0.152
#> GSM97058 2 0.266 0.8012 0.024 0.932 0.044
#> GSM97051 2 0.175 0.7990 0.000 0.952 0.048
#> GSM97052 2 0.296 0.7902 0.000 0.900 0.100
#> GSM97061 2 0.196 0.7974 0.000 0.944 0.056
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM97138 1 0.2401 0.711 0.904 0.000 0.004 0.092
#> GSM97145 1 0.2188 0.715 0.936 0.020 0.012 0.032
#> GSM97147 2 0.5886 0.585 0.200 0.720 0.040 0.040
#> GSM97125 1 0.2284 0.716 0.932 0.020 0.012 0.036
#> GSM97127 1 0.2269 0.716 0.932 0.028 0.008 0.032
#> GSM97130 1 0.6302 0.624 0.720 0.128 0.040 0.112
#> GSM97133 1 0.1271 0.704 0.968 0.012 0.008 0.012
#> GSM97134 1 0.7693 0.386 0.516 0.332 0.028 0.124
#> GSM97120 1 0.0992 0.706 0.976 0.004 0.008 0.012
#> GSM97126 1 0.7988 0.421 0.524 0.268 0.032 0.176
#> GSM97112 1 0.3539 0.691 0.820 0.000 0.004 0.176
#> GSM97115 1 0.7767 0.291 0.472 0.388 0.036 0.104
#> GSM97116 1 0.1767 0.704 0.944 0.000 0.012 0.044
#> GSM97117 2 0.5540 0.631 0.132 0.760 0.088 0.020
#> GSM97119 1 0.3539 0.691 0.820 0.000 0.004 0.176
#> GSM97122 1 0.3539 0.691 0.820 0.000 0.004 0.176
#> GSM97135 1 0.3494 0.692 0.824 0.000 0.004 0.172
#> GSM97136 1 0.8950 0.268 0.444 0.252 0.076 0.228
#> GSM97139 1 0.1124 0.700 0.972 0.004 0.012 0.012
#> GSM97146 1 0.1575 0.693 0.956 0.004 0.012 0.028
#> GSM97123 2 0.3128 0.690 0.000 0.884 0.076 0.040
#> GSM97129 1 0.7693 0.386 0.516 0.332 0.028 0.124
#> GSM97143 1 0.6187 0.601 0.672 0.052 0.024 0.252
#> GSM97113 1 0.6287 0.140 0.508 0.448 0.020 0.024
#> GSM97056 1 0.1822 0.707 0.944 0.004 0.008 0.044
#> GSM97124 1 0.3950 0.698 0.820 0.012 0.008 0.160
#> GSM97132 1 0.5725 0.627 0.708 0.032 0.028 0.232
#> GSM97144 1 0.7126 0.586 0.648 0.172 0.040 0.140
#> GSM97149 1 0.1575 0.693 0.956 0.004 0.012 0.028
#> GSM97068 1 0.7087 0.422 0.564 0.336 0.032 0.068
#> GSM97071 2 0.7929 0.321 0.024 0.516 0.188 0.272
#> GSM97086 2 0.4758 0.633 0.000 0.780 0.156 0.064
#> GSM97103 2 0.6055 0.443 0.004 0.604 0.344 0.048
#> GSM97057 1 0.6099 0.305 0.564 0.396 0.016 0.024
#> GSM97060 2 0.5619 0.566 0.000 0.688 0.248 0.064
#> GSM97075 2 0.4416 0.678 0.040 0.832 0.100 0.028
#> GSM97098 2 0.5733 0.493 0.004 0.648 0.308 0.040
#> GSM97099 2 0.5055 0.653 0.096 0.792 0.096 0.016
#> GSM97101 2 0.4994 0.654 0.092 0.796 0.096 0.016
#> GSM97105 2 0.1975 0.693 0.012 0.944 0.028 0.016
#> GSM97106 2 0.4054 0.633 0.000 0.796 0.188 0.016
#> GSM97121 2 0.4830 0.646 0.136 0.800 0.036 0.028
#> GSM97128 4 0.3048 0.776 0.028 0.016 0.056 0.900
#> GSM97131 2 0.2131 0.695 0.000 0.932 0.032 0.036
#> GSM97137 1 0.4821 0.672 0.812 0.088 0.024 0.076
#> GSM97118 1 0.6338 0.525 0.620 0.024 0.040 0.316
#> GSM97114 2 0.5540 0.631 0.132 0.760 0.088 0.020
#> GSM97142 1 0.3539 0.691 0.820 0.000 0.004 0.176
#> GSM97140 2 0.3793 0.690 0.064 0.868 0.024 0.044
#> GSM97141 2 0.4984 0.642 0.132 0.788 0.068 0.012
#> GSM97055 4 0.8120 0.518 0.232 0.116 0.088 0.564
#> GSM97090 1 0.7857 0.236 0.452 0.404 0.040 0.104
#> GSM97091 1 0.3765 0.688 0.812 0.004 0.004 0.180
#> GSM97148 1 0.1575 0.693 0.956 0.004 0.012 0.028
#> GSM97063 1 0.3765 0.688 0.812 0.004 0.004 0.180
#> GSM97053 1 0.2999 0.705 0.864 0.004 0.000 0.132
#> GSM97066 3 0.4944 0.723 0.000 0.160 0.768 0.072
#> GSM97079 2 0.4711 0.639 0.000 0.784 0.152 0.064
#> GSM97083 4 0.3065 0.774 0.032 0.016 0.052 0.900
#> GSM97084 2 0.5457 0.594 0.000 0.728 0.184 0.088
#> GSM97094 2 0.7688 0.506 0.092 0.624 0.160 0.124
#> GSM97096 2 0.6004 0.461 0.004 0.616 0.332 0.048
#> GSM97097 2 0.5200 0.613 0.000 0.744 0.184 0.072
#> GSM97107 2 0.7126 0.540 0.060 0.660 0.168 0.112
#> GSM97054 2 0.7929 0.321 0.024 0.516 0.188 0.272
#> GSM97062 2 0.4829 0.636 0.000 0.776 0.156 0.068
#> GSM97069 3 0.6182 0.617 0.000 0.308 0.616 0.076
#> GSM97070 3 0.4944 0.723 0.000 0.160 0.768 0.072
#> GSM97073 3 0.5265 0.710 0.000 0.160 0.748 0.092
#> GSM97076 3 0.6400 0.658 0.036 0.148 0.708 0.108
#> GSM97077 2 0.3853 0.693 0.052 0.868 0.040 0.040
#> GSM97095 1 0.7314 0.290 0.508 0.388 0.036 0.068
#> GSM97102 2 0.6055 0.443 0.004 0.604 0.344 0.048
#> GSM97109 2 0.6681 0.486 0.036 0.624 0.288 0.052
#> GSM97110 2 0.6681 0.486 0.036 0.624 0.288 0.052
#> GSM97074 3 0.8399 0.195 0.124 0.092 0.528 0.256
#> GSM97085 4 0.6196 0.674 0.024 0.136 0.124 0.716
#> GSM97059 2 0.5781 0.526 0.240 0.700 0.024 0.036
#> GSM97072 3 0.4482 0.628 0.000 0.264 0.728 0.008
#> GSM97078 4 0.3048 0.776 0.028 0.016 0.056 0.900
#> GSM97067 3 0.4776 0.721 0.000 0.164 0.776 0.060
#> GSM97087 2 0.6355 0.512 0.004 0.656 0.228 0.112
#> GSM97111 2 0.6128 0.589 0.044 0.708 0.200 0.048
#> GSM97064 2 0.2555 0.696 0.008 0.920 0.040 0.032
#> GSM97065 3 0.7494 0.508 0.048 0.332 0.544 0.076
#> GSM97081 2 0.6142 0.562 0.024 0.684 0.236 0.056
#> GSM97082 2 0.6565 0.488 0.004 0.640 0.224 0.132
#> GSM97088 4 0.6770 0.626 0.036 0.168 0.120 0.676
#> GSM97100 2 0.3301 0.694 0.040 0.892 0.024 0.044
#> GSM97104 2 0.5894 0.285 0.000 0.568 0.392 0.040
#> GSM97108 2 0.5126 0.668 0.088 0.800 0.072 0.040
#> GSM97050 2 0.2036 0.693 0.000 0.936 0.032 0.032
#> GSM97080 3 0.6179 0.411 0.000 0.392 0.552 0.056
#> GSM97089 2 0.6355 0.512 0.004 0.656 0.228 0.112
#> GSM97092 2 0.5594 0.578 0.000 0.716 0.192 0.092
#> GSM97093 2 0.8124 0.450 0.172 0.584 0.100 0.144
#> GSM97058 2 0.2719 0.696 0.024 0.916 0.040 0.020
#> GSM97051 2 0.2500 0.693 0.000 0.916 0.044 0.040
#> GSM97052 2 0.4881 0.603 0.000 0.756 0.196 0.048
#> GSM97061 2 0.2845 0.682 0.000 0.896 0.076 0.028
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM97138 1 0.233 0.7254 0.908 0.004 0.000 0.024 0.064
#> GSM97145 1 0.207 0.7289 0.932 0.024 0.008 0.008 0.028
#> GSM97147 2 0.493 0.3819 0.196 0.736 0.008 0.036 0.024
#> GSM97125 1 0.215 0.7293 0.928 0.024 0.008 0.008 0.032
#> GSM97127 1 0.211 0.7296 0.928 0.032 0.004 0.008 0.028
#> GSM97130 1 0.573 0.6440 0.712 0.092 0.004 0.128 0.064
#> GSM97133 1 0.199 0.7189 0.936 0.012 0.008 0.028 0.016
#> GSM97134 1 0.692 0.4346 0.528 0.316 0.004 0.064 0.088
#> GSM97120 1 0.193 0.7189 0.936 0.004 0.008 0.032 0.020
#> GSM97126 1 0.724 0.4690 0.544 0.260 0.016 0.052 0.128
#> GSM97112 1 0.288 0.7063 0.848 0.004 0.004 0.000 0.144
#> GSM97115 1 0.696 0.3554 0.484 0.352 0.000 0.108 0.056
#> GSM97116 1 0.247 0.7127 0.908 0.000 0.012 0.040 0.040
#> GSM97117 2 0.464 0.4656 0.140 0.772 0.068 0.016 0.004
#> GSM97119 1 0.288 0.7063 0.848 0.004 0.004 0.000 0.144
#> GSM97122 1 0.288 0.7063 0.848 0.004 0.004 0.000 0.144
#> GSM97135 1 0.283 0.7076 0.852 0.004 0.004 0.000 0.140
#> GSM97136 1 0.824 0.3155 0.468 0.232 0.072 0.040 0.188
#> GSM97139 1 0.209 0.7112 0.928 0.004 0.008 0.040 0.020
#> GSM97146 1 0.281 0.6970 0.896 0.004 0.016 0.048 0.036
#> GSM97123 2 0.469 0.3874 0.000 0.740 0.056 0.192 0.012
#> GSM97129 1 0.692 0.4346 0.528 0.316 0.004 0.064 0.088
#> GSM97143 1 0.541 0.6250 0.700 0.056 0.020 0.012 0.212
#> GSM97113 1 0.553 0.1583 0.492 0.464 0.012 0.016 0.016
#> GSM97056 1 0.258 0.7125 0.908 0.004 0.016 0.036 0.036
#> GSM97124 1 0.317 0.7124 0.844 0.020 0.004 0.000 0.132
#> GSM97132 1 0.509 0.6491 0.728 0.032 0.020 0.020 0.200
#> GSM97144 1 0.618 0.6169 0.668 0.124 0.000 0.124 0.084
#> GSM97149 1 0.281 0.6970 0.896 0.004 0.016 0.048 0.036
#> GSM97068 1 0.668 0.4551 0.548 0.316 0.008 0.088 0.040
#> GSM97071 4 0.791 0.2662 0.024 0.340 0.040 0.396 0.200
#> GSM97086 2 0.441 0.0149 0.000 0.604 0.000 0.388 0.008
#> GSM97103 2 0.744 0.0906 0.004 0.388 0.300 0.284 0.024
#> GSM97057 1 0.556 0.3122 0.544 0.408 0.012 0.016 0.020
#> GSM97060 4 0.727 0.1289 0.000 0.332 0.144 0.464 0.060
#> GSM97075 2 0.484 0.4973 0.036 0.788 0.096 0.060 0.020
#> GSM97098 2 0.713 0.1936 0.004 0.456 0.276 0.248 0.016
#> GSM97099 2 0.473 0.4917 0.100 0.784 0.072 0.040 0.004
#> GSM97101 2 0.468 0.4926 0.096 0.788 0.072 0.040 0.004
#> GSM97105 2 0.189 0.4824 0.012 0.936 0.008 0.040 0.004
#> GSM97106 2 0.623 -0.0234 0.000 0.480 0.092 0.412 0.016
#> GSM97121 2 0.413 0.4109 0.136 0.800 0.008 0.052 0.004
#> GSM97128 5 0.241 0.7898 0.028 0.016 0.044 0.000 0.912
#> GSM97131 2 0.344 0.4464 0.000 0.824 0.024 0.148 0.004
#> GSM97137 1 0.459 0.6877 0.796 0.060 0.008 0.100 0.036
#> GSM97118 1 0.566 0.5484 0.644 0.024 0.040 0.012 0.280
#> GSM97114 2 0.464 0.4656 0.140 0.772 0.068 0.016 0.004
#> GSM97142 1 0.288 0.7063 0.848 0.004 0.004 0.000 0.144
#> GSM97140 2 0.319 0.4775 0.064 0.876 0.004 0.032 0.024
#> GSM97141 2 0.411 0.4717 0.136 0.800 0.048 0.016 0.000
#> GSM97055 5 0.776 0.5359 0.240 0.100 0.076 0.048 0.536
#> GSM97090 1 0.703 0.3250 0.464 0.368 0.000 0.112 0.056
#> GSM97091 1 0.304 0.7038 0.840 0.008 0.004 0.000 0.148
#> GSM97148 1 0.281 0.6970 0.896 0.004 0.016 0.048 0.036
#> GSM97063 1 0.304 0.7038 0.840 0.008 0.004 0.000 0.148
#> GSM97053 1 0.241 0.7201 0.884 0.008 0.000 0.000 0.108
#> GSM97066 3 0.194 0.7019 0.000 0.068 0.920 0.000 0.012
#> GSM97079 2 0.430 0.0481 0.000 0.608 0.000 0.388 0.004
#> GSM97083 5 0.258 0.7861 0.040 0.016 0.040 0.000 0.904
#> GSM97084 4 0.456 0.1082 0.000 0.484 0.000 0.508 0.008
#> GSM97094 4 0.679 0.1904 0.100 0.404 0.000 0.452 0.044
#> GSM97096 2 0.740 0.1208 0.004 0.408 0.288 0.276 0.024
#> GSM97097 2 0.475 -0.1794 0.000 0.500 0.016 0.484 0.000
#> GSM97107 4 0.621 0.1872 0.068 0.424 0.000 0.480 0.028
#> GSM97054 4 0.791 0.2662 0.024 0.340 0.040 0.396 0.200
#> GSM97062 2 0.443 0.0271 0.000 0.600 0.000 0.392 0.008
#> GSM97069 3 0.554 0.6227 0.000 0.136 0.704 0.128 0.032
#> GSM97070 3 0.194 0.7019 0.000 0.068 0.920 0.000 0.012
#> GSM97073 3 0.286 0.6901 0.000 0.076 0.884 0.016 0.024
#> GSM97076 3 0.402 0.6484 0.040 0.068 0.840 0.024 0.028
#> GSM97077 2 0.366 0.4968 0.048 0.860 0.020 0.048 0.024
#> GSM97095 1 0.675 0.3419 0.500 0.368 0.008 0.088 0.036
#> GSM97102 2 0.744 0.0906 0.004 0.388 0.300 0.284 0.024
#> GSM97109 2 0.765 0.2562 0.040 0.484 0.276 0.176 0.024
#> GSM97110 2 0.765 0.2562 0.040 0.484 0.276 0.176 0.024
#> GSM97074 3 0.678 0.2853 0.140 0.060 0.612 0.008 0.180
#> GSM97085 5 0.591 0.7059 0.016 0.100 0.084 0.084 0.716
#> GSM97059 2 0.489 0.3219 0.236 0.712 0.008 0.028 0.016
#> GSM97072 3 0.515 0.6322 0.000 0.100 0.720 0.164 0.016
#> GSM97078 5 0.241 0.7898 0.028 0.016 0.044 0.000 0.912
#> GSM97067 3 0.258 0.7047 0.000 0.084 0.892 0.016 0.008
#> GSM97087 4 0.786 0.1329 0.000 0.360 0.144 0.380 0.116
#> GSM97111 2 0.612 0.4280 0.048 0.672 0.192 0.072 0.016
#> GSM97064 2 0.373 0.4666 0.008 0.832 0.036 0.116 0.008
#> GSM97065 3 0.635 0.4693 0.052 0.264 0.620 0.028 0.036
#> GSM97081 2 0.656 0.3767 0.028 0.624 0.216 0.108 0.024
#> GSM97082 4 0.813 0.1219 0.004 0.344 0.160 0.368 0.124
#> GSM97088 5 0.637 0.6775 0.028 0.104 0.084 0.096 0.688
#> GSM97100 2 0.276 0.4831 0.040 0.900 0.004 0.036 0.020
#> GSM97104 3 0.767 0.0261 0.000 0.276 0.352 0.324 0.048
#> GSM97108 2 0.505 0.4966 0.088 0.780 0.056 0.052 0.024
#> GSM97050 2 0.281 0.4599 0.000 0.872 0.012 0.108 0.008
#> GSM97080 3 0.646 0.5149 0.000 0.180 0.600 0.188 0.032
#> GSM97089 4 0.786 0.1329 0.000 0.360 0.144 0.380 0.116
#> GSM97092 2 0.732 -0.0817 0.000 0.452 0.116 0.352 0.080
#> GSM97093 2 0.842 0.1102 0.176 0.500 0.060 0.144 0.120
#> GSM97058 2 0.305 0.4956 0.020 0.884 0.024 0.064 0.008
#> GSM97051 2 0.356 0.4301 0.000 0.804 0.012 0.176 0.008
#> GSM97052 2 0.683 -0.0844 0.000 0.460 0.112 0.388 0.040
#> GSM97061 2 0.441 0.3865 0.000 0.764 0.048 0.176 0.012
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM97138 1 0.283 0.7381 0.876 0.004 0.000 0.056 0.052 0.012
#> GSM97145 1 0.160 0.7416 0.944 0.020 0.000 0.020 0.012 0.004
#> GSM97147 2 0.517 0.4844 0.192 0.704 0.012 0.052 0.024 0.016
#> GSM97125 1 0.169 0.7420 0.940 0.020 0.000 0.020 0.016 0.004
#> GSM97127 1 0.162 0.7421 0.940 0.028 0.000 0.020 0.012 0.000
#> GSM97130 1 0.566 0.6514 0.688 0.068 0.012 0.160 0.056 0.016
#> GSM97133 1 0.239 0.7279 0.900 0.012 0.000 0.064 0.008 0.016
#> GSM97134 1 0.641 0.4233 0.540 0.300 0.004 0.068 0.076 0.012
#> GSM97120 1 0.215 0.7290 0.916 0.004 0.004 0.052 0.012 0.012
#> GSM97126 1 0.666 0.4965 0.556 0.244 0.004 0.056 0.116 0.024
#> GSM97112 1 0.242 0.7248 0.864 0.004 0.000 0.000 0.128 0.004
#> GSM97115 1 0.676 0.3105 0.480 0.308 0.008 0.156 0.040 0.008
#> GSM97116 1 0.338 0.7140 0.844 0.000 0.008 0.088 0.036 0.024
#> GSM97117 2 0.444 0.5640 0.140 0.760 0.008 0.024 0.000 0.068
#> GSM97119 1 0.242 0.7248 0.864 0.004 0.000 0.000 0.128 0.004
#> GSM97122 1 0.242 0.7248 0.864 0.004 0.000 0.000 0.128 0.004
#> GSM97135 1 0.238 0.7260 0.868 0.004 0.000 0.000 0.124 0.004
#> GSM97136 1 0.776 0.3582 0.480 0.208 0.036 0.020 0.168 0.088
#> GSM97139 1 0.268 0.7171 0.884 0.004 0.004 0.076 0.012 0.020
#> GSM97146 1 0.375 0.6859 0.824 0.004 0.024 0.104 0.016 0.028
#> GSM97123 2 0.470 0.2268 0.000 0.624 0.316 0.056 0.000 0.004
#> GSM97129 1 0.641 0.4233 0.540 0.300 0.004 0.068 0.076 0.012
#> GSM97143 1 0.483 0.6460 0.712 0.044 0.000 0.008 0.196 0.040
#> GSM97113 2 0.595 -0.0971 0.440 0.456 0.016 0.064 0.008 0.016
#> GSM97056 1 0.316 0.7174 0.868 0.004 0.020 0.064 0.016 0.028
#> GSM97124 1 0.273 0.7299 0.860 0.016 0.000 0.004 0.116 0.004
#> GSM97132 1 0.477 0.6687 0.724 0.020 0.000 0.028 0.188 0.040
#> GSM97144 1 0.549 0.6197 0.680 0.080 0.004 0.168 0.064 0.004
#> GSM97149 1 0.375 0.6859 0.824 0.004 0.024 0.104 0.016 0.028
#> GSM97068 1 0.633 0.4252 0.536 0.312 0.012 0.100 0.028 0.012
#> GSM97071 4 0.757 0.5005 0.024 0.140 0.136 0.520 0.156 0.024
#> GSM97086 2 0.440 -0.2454 0.000 0.508 0.024 0.468 0.000 0.000
#> GSM97103 3 0.646 0.5146 0.004 0.248 0.476 0.016 0.004 0.252
#> GSM97057 1 0.600 0.2187 0.492 0.400 0.020 0.064 0.008 0.016
#> GSM97060 3 0.398 0.6238 0.000 0.128 0.792 0.052 0.004 0.024
#> GSM97075 2 0.482 0.5492 0.036 0.764 0.068 0.016 0.012 0.104
#> GSM97098 2 0.672 -0.3740 0.004 0.364 0.352 0.028 0.000 0.252
#> GSM97099 2 0.468 0.5826 0.096 0.764 0.032 0.024 0.000 0.084
#> GSM97101 2 0.463 0.5829 0.092 0.768 0.032 0.024 0.000 0.084
#> GSM97105 2 0.207 0.5768 0.012 0.920 0.036 0.028 0.000 0.004
#> GSM97106 3 0.439 0.6290 0.000 0.300 0.652 0.048 0.000 0.000
#> GSM97121 2 0.408 0.5094 0.136 0.784 0.008 0.060 0.004 0.008
#> GSM97128 5 0.112 0.7707 0.004 0.008 0.008 0.000 0.964 0.016
#> GSM97131 2 0.423 0.4604 0.000 0.736 0.180 0.080 0.000 0.004
#> GSM97137 1 0.444 0.7000 0.780 0.048 0.012 0.124 0.020 0.016
#> GSM97118 1 0.505 0.5754 0.652 0.016 0.000 0.008 0.264 0.060
#> GSM97114 2 0.444 0.5640 0.140 0.760 0.008 0.024 0.000 0.068
#> GSM97142 1 0.242 0.7248 0.864 0.004 0.000 0.000 0.128 0.004
#> GSM97140 2 0.347 0.5728 0.064 0.852 0.024 0.036 0.020 0.004
#> GSM97141 2 0.403 0.5710 0.136 0.788 0.008 0.020 0.000 0.048
#> GSM97055 5 0.736 0.4868 0.236 0.084 0.064 0.012 0.524 0.080
#> GSM97090 1 0.674 0.2838 0.460 0.332 0.004 0.152 0.044 0.008
#> GSM97091 1 0.250 0.7221 0.856 0.004 0.000 0.000 0.136 0.004
#> GSM97148 1 0.375 0.6859 0.824 0.004 0.024 0.104 0.016 0.028
#> GSM97063 1 0.250 0.7221 0.856 0.004 0.000 0.000 0.136 0.004
#> GSM97053 1 0.216 0.7367 0.892 0.008 0.000 0.004 0.096 0.000
#> GSM97066 6 0.195 0.7525 0.000 0.012 0.072 0.000 0.004 0.912
#> GSM97079 2 0.440 -0.2043 0.000 0.516 0.024 0.460 0.000 0.000
#> GSM97083 5 0.132 0.7675 0.016 0.008 0.004 0.000 0.956 0.016
#> GSM97084 4 0.398 0.6712 0.000 0.284 0.020 0.692 0.000 0.004
#> GSM97094 4 0.617 0.6726 0.104 0.240 0.020 0.596 0.036 0.004
#> GSM97096 3 0.655 0.4837 0.004 0.288 0.444 0.016 0.004 0.244
#> GSM97097 4 0.491 0.6351 0.000 0.304 0.068 0.620 0.000 0.008
#> GSM97107 4 0.555 0.7049 0.068 0.236 0.024 0.648 0.020 0.004
#> GSM97054 4 0.757 0.5005 0.024 0.140 0.136 0.520 0.156 0.024
#> GSM97062 2 0.434 -0.2488 0.000 0.496 0.020 0.484 0.000 0.000
#> GSM97069 6 0.452 0.5395 0.000 0.048 0.300 0.000 0.004 0.648
#> GSM97070 6 0.195 0.7525 0.000 0.012 0.072 0.000 0.004 0.912
#> GSM97073 6 0.186 0.7460 0.000 0.016 0.044 0.004 0.008 0.928
#> GSM97076 6 0.194 0.7165 0.040 0.016 0.004 0.004 0.008 0.928
#> GSM97077 2 0.385 0.5921 0.048 0.840 0.024 0.040 0.024 0.024
#> GSM97095 1 0.644 0.3056 0.484 0.360 0.008 0.108 0.024 0.016
#> GSM97102 3 0.646 0.5146 0.004 0.248 0.476 0.016 0.004 0.252
#> GSM97109 2 0.712 -0.1747 0.040 0.408 0.280 0.012 0.004 0.256
#> GSM97110 2 0.712 -0.1747 0.040 0.408 0.280 0.012 0.004 0.256
#> GSM97074 6 0.497 0.3987 0.160 0.012 0.000 0.000 0.148 0.680
#> GSM97085 5 0.530 0.7031 0.008 0.056 0.144 0.008 0.712 0.072
#> GSM97059 2 0.486 0.4342 0.224 0.700 0.008 0.040 0.016 0.012
#> GSM97072 6 0.457 0.5797 0.000 0.020 0.304 0.020 0.004 0.652
#> GSM97078 5 0.112 0.7707 0.004 0.008 0.008 0.000 0.964 0.016
#> GSM97067 6 0.235 0.7466 0.000 0.020 0.100 0.000 0.000 0.880
#> GSM97087 3 0.490 0.6894 0.000 0.204 0.696 0.004 0.072 0.024
#> GSM97111 2 0.612 0.4016 0.048 0.632 0.084 0.024 0.008 0.204
#> GSM97064 2 0.383 0.5128 0.008 0.808 0.128 0.036 0.008 0.012
#> GSM97065 6 0.519 0.4336 0.048 0.228 0.016 0.016 0.012 0.680
#> GSM97081 2 0.634 0.2517 0.028 0.588 0.140 0.020 0.008 0.216
#> GSM97082 3 0.566 0.6771 0.004 0.184 0.664 0.008 0.080 0.060
#> GSM97088 5 0.529 0.6633 0.012 0.068 0.172 0.004 0.700 0.044
#> GSM97100 2 0.309 0.5764 0.040 0.876 0.024 0.036 0.020 0.004
#> GSM97104 3 0.555 0.4559 0.000 0.136 0.608 0.008 0.008 0.240
#> GSM97108 2 0.530 0.5919 0.088 0.744 0.040 0.036 0.020 0.072
#> GSM97050 2 0.301 0.5379 0.000 0.844 0.068 0.088 0.000 0.000
#> GSM97080 6 0.542 0.3043 0.000 0.084 0.368 0.004 0.008 0.536
#> GSM97089 3 0.490 0.6894 0.000 0.204 0.696 0.004 0.072 0.024
#> GSM97092 3 0.521 0.6211 0.000 0.312 0.612 0.016 0.044 0.016
#> GSM97093 2 0.833 0.1705 0.172 0.428 0.200 0.088 0.096 0.016
#> GSM97058 2 0.316 0.5716 0.020 0.868 0.068 0.020 0.008 0.016
#> GSM97051 2 0.405 0.4558 0.000 0.752 0.152 0.096 0.000 0.000
#> GSM97052 3 0.417 0.6651 0.000 0.288 0.684 0.012 0.004 0.012
#> GSM97061 2 0.429 0.2826 0.000 0.684 0.276 0.032 0.004 0.004
Heatmaps for the consensus matrix. It visualizes the probability of two samples to be in a same group.
consensus_heatmap(res, k = 2)
consensus_heatmap(res, k = 3)
consensus_heatmap(res, k = 4)
consensus_heatmap(res, k = 5)
consensus_heatmap(res, k = 6)
Heatmaps for the membership of samples in all partitions to see how consistent they are:
membership_heatmap(res, k = 2)
membership_heatmap(res, k = 3)
membership_heatmap(res, k = 4)
membership_heatmap(res, k = 5)
membership_heatmap(res, k = 6)
As soon as we have had the classes for columns, we can look for signatures which are significantly different between classes which can be candidate marks for certain classes. Following are the heatmaps for signatures.
Signature heatmaps where rows are scaled:
get_signatures(res, k = 2)
get_signatures(res, k = 3)
get_signatures(res, k = 4)
get_signatures(res, k = 5)
get_signatures(res, k = 6)
Signature heatmaps where rows are not scaled:
get_signatures(res, k = 2, scale_rows = FALSE)
get_signatures(res, k = 3, scale_rows = FALSE)
get_signatures(res, k = 4, scale_rows = FALSE)
get_signatures(res, k = 5, scale_rows = FALSE)
get_signatures(res, k = 6, scale_rows = FALSE)
Compare the overlap of signatures from different k:
compare_signatures(res)
get_signature() returns a data frame invisibly. TO get the list of signatures, the function
call should be assigned to a variable explicitly. In following code, if plot argument is set
to FALSE, no heatmap is plotted while only the differential analysis is performed.
# code only for demonstration
tb = get_signature(res, k = ..., plot = FALSE)
An example of the output of tb is:
#> which_row fdr mean_1 mean_2 scaled_mean_1 scaled_mean_2 km
#> 1 38 0.042760348 8.373488 9.131774 -0.5533452 0.5164555 1
#> 2 40 0.018707592 7.106213 8.469186 -0.6173731 0.5762149 1
#> 3 55 0.019134737 10.221463 11.207825 -0.6159697 0.5749050 1
#> 4 59 0.006059896 5.921854 7.869574 -0.6899429 0.6439467 1
#> 5 60 0.018055526 8.928898 10.211722 -0.6204761 0.5791110 1
#> 6 98 0.009384629 15.714769 14.887706 0.6635654 -0.6193277 2
...
The columns in tb are:
which_row: row indices corresponding to the input matrix.fdr: FDR for the differential test. mean_x: The mean value in group x.scaled_mean_x: The mean value in group x after rows are scaled.km: Row groups if k-means clustering is applied to rows.UMAP plot which shows how samples are separated.
dimension_reduction(res, k = 2, method = "UMAP")
dimension_reduction(res, k = 3, method = "UMAP")
dimension_reduction(res, k = 4, method = "UMAP")
dimension_reduction(res, k = 5, method = "UMAP")
dimension_reduction(res, k = 6, method = "UMAP")
Following heatmap shows how subgroups are split when increasing k:
collect_classes(res)
Test correlation between subgroups and known annotations. If the known annotation is numeric, one-way ANOVA test is applied, and if the known annotation is discrete, chi-squared contingency table test is applied.
test_to_known_factors(res)
#> n disease.state(p) specimen(p) cell.type(p) other(p) k
#> CV:hclust 90 6.48e-08 1.000 1.15e-16 0.0559 2
#> CV:hclust 89 2.43e-06 0.956 3.86e-15 0.2565 3
#> CV:hclust 77 5.86e-05 0.460 7.45e-14 0.3050 4
#> CV:hclust 41 4.10e-04 0.341 1.08e-07 0.3823 5
#> CV:hclust 69 2.07e-04 0.595 1.65e-16 0.1259 6
If matrix rows can be associated to genes, consider to use functional_enrichment(res,
...) to perform function enrichment for the signature genes. See this vignette for more detailed explanations.
The object with results only for a single top-value method and a single partition method can be extracted as:
res = res_list["CV", "kmeans"]
# you can also extract it by
# res = res_list["CV:kmeans"]
A summary of res and all the functions that can be applied to it:
res
#> A 'ConsensusPartition' object with k = 2, 3, 4, 5, 6.
#> On a matrix with 21512 rows and 100 columns.
#> Top rows (1000, 2000, 3000, 4000, 5000) are extracted by 'CV' method.
#> Subgroups are detected by 'kmeans' method.
#> Performed in total 1250 partitions by row resampling.
#> Best k for subgroups seems to be 2.
#>
#> Following methods can be applied to this 'ConsensusPartition' object:
#> [1] "cola_report" "collect_classes" "collect_plots"
#> [4] "collect_stats" "colnames" "compare_signatures"
#> [7] "consensus_heatmap" "dimension_reduction" "functional_enrichment"
#> [10] "get_anno_col" "get_anno" "get_classes"
#> [13] "get_consensus" "get_matrix" "get_membership"
#> [16] "get_param" "get_signatures" "get_stats"
#> [19] "is_best_k" "is_stable_k" "membership_heatmap"
#> [22] "ncol" "nrow" "plot_ecdf"
#> [25] "rownames" "select_partition_number" "show"
#> [28] "suggest_best_k" "test_to_known_factors"
collect_plots() function collects all the plots made from res for all k (number of partitions)
into one single page to provide an easy and fast comparison between different k.
collect_plots(res)
The plots are:
k and the heatmap of
predicted classes for each k.k.k.k.All the plots in panels can be made by individual functions and they are plotted later in this section.
select_partition_number() produces several plots showing different
statistics for choosing “optimized” k. There are following statistics:
k;k, the area increased is defined as \(A_k - A_{k-1}\).The detailed explanations of these statistics can be found in the cola vignette.
Generally speaking, lower PAC score, higher mean silhouette score or higher
concordance corresponds to better partition. Rand index and Jaccard index
measure how similar the current partition is compared to partition with k-1.
If they are too similar, we won't accept k is better than k-1.
select_partition_number(res)
The numeric values for all these statistics can be obtained by get_stats().
get_stats(res)
#> k 1-PAC mean_silhouette concordance area_increased Rand Jaccard
#> 2 2 0.704 0.885 0.937 0.4923 0.505 0.505
#> 3 3 0.656 0.834 0.889 0.3295 0.697 0.472
#> 4 4 0.604 0.577 0.782 0.1225 0.864 0.632
#> 5 5 0.609 0.528 0.716 0.0684 0.896 0.644
#> 6 6 0.667 0.430 0.693 0.0440 0.889 0.563
suggest_best_k() suggests the best \(k\) based on these statistics. The rules are as follows:
suggest_best_k(res)
#> [1] 2
Following shows the table of the partitions (You need to click the show/hide
code output link to see it). The membership matrix (columns with name p*)
is inferred by
clue::cl_consensus()
function with the SE method. Basically the value in the membership matrix
represents the probability to belong to a certain group. The finall class
label for an item is determined with the group with highest probability it
belongs to.
In get_classes() function, the entropy is calculated from the membership
matrix and the silhouette score is calculated from the consensus matrix.
cbind(get_classes(res, k = 2), get_membership(res, k = 2))
#> class entropy silhouette p1 p2
#> GSM97138 1 0.0000 0.946 1.000 0.000
#> GSM97145 1 0.0000 0.946 1.000 0.000
#> GSM97147 1 0.2603 0.921 0.956 0.044
#> GSM97125 1 0.0000 0.946 1.000 0.000
#> GSM97127 1 0.0376 0.945 0.996 0.004
#> GSM97130 1 0.0376 0.945 0.996 0.004
#> GSM97133 1 0.0376 0.945 0.996 0.004
#> GSM97134 1 0.0000 0.946 1.000 0.000
#> GSM97120 1 0.0000 0.946 1.000 0.000
#> GSM97126 1 0.0000 0.946 1.000 0.000
#> GSM97112 1 0.0000 0.946 1.000 0.000
#> GSM97115 1 0.0938 0.942 0.988 0.012
#> GSM97116 1 0.0000 0.946 1.000 0.000
#> GSM97117 2 0.7883 0.777 0.236 0.764
#> GSM97119 1 0.0000 0.946 1.000 0.000
#> GSM97122 1 0.0000 0.946 1.000 0.000
#> GSM97135 1 0.0000 0.946 1.000 0.000
#> GSM97136 2 0.6531 0.822 0.168 0.832
#> GSM97139 1 0.0000 0.946 1.000 0.000
#> GSM97146 1 0.0000 0.946 1.000 0.000
#> GSM97123 2 0.0000 0.918 0.000 1.000
#> GSM97129 2 0.7815 0.781 0.232 0.768
#> GSM97143 1 0.0000 0.946 1.000 0.000
#> GSM97113 2 0.8144 0.753 0.252 0.748
#> GSM97056 1 0.0376 0.945 0.996 0.004
#> GSM97124 1 0.0000 0.946 1.000 0.000
#> GSM97132 1 0.0000 0.946 1.000 0.000
#> GSM97144 1 0.0376 0.945 0.996 0.004
#> GSM97149 1 0.0376 0.945 0.996 0.004
#> GSM97068 1 0.9427 0.361 0.640 0.360
#> GSM97071 2 0.0376 0.918 0.004 0.996
#> GSM97086 2 0.2778 0.915 0.048 0.952
#> GSM97103 2 0.2778 0.915 0.048 0.952
#> GSM97057 2 0.8499 0.718 0.276 0.724
#> GSM97060 2 0.0000 0.918 0.000 1.000
#> GSM97075 2 0.0000 0.918 0.000 1.000
#> GSM97098 2 0.0000 0.918 0.000 1.000
#> GSM97099 2 0.7815 0.778 0.232 0.768
#> GSM97101 2 0.7815 0.778 0.232 0.768
#> GSM97105 2 0.2778 0.915 0.048 0.952
#> GSM97106 2 0.0000 0.918 0.000 1.000
#> GSM97121 2 0.7528 0.794 0.216 0.784
#> GSM97128 1 0.9000 0.598 0.684 0.316
#> GSM97131 2 0.2778 0.915 0.048 0.952
#> GSM97137 1 0.0376 0.945 0.996 0.004
#> GSM97118 1 0.0000 0.946 1.000 0.000
#> GSM97114 2 0.8386 0.730 0.268 0.732
#> GSM97142 1 0.0000 0.946 1.000 0.000
#> GSM97140 2 0.7815 0.778 0.232 0.768
#> GSM97141 2 0.7815 0.778 0.232 0.768
#> GSM97055 1 0.2778 0.914 0.952 0.048
#> GSM97090 1 0.0938 0.942 0.988 0.012
#> GSM97091 1 0.2778 0.914 0.952 0.048
#> GSM97148 1 0.0376 0.945 0.996 0.004
#> GSM97063 1 0.2603 0.917 0.956 0.044
#> GSM97053 1 0.0000 0.946 1.000 0.000
#> GSM97066 2 0.0376 0.918 0.004 0.996
#> GSM97079 2 0.2423 0.917 0.040 0.960
#> GSM97083 1 0.3274 0.907 0.940 0.060
#> GSM97084 2 0.2778 0.915 0.048 0.952
#> GSM97094 1 0.3584 0.902 0.932 0.068
#> GSM97096 2 0.0000 0.918 0.000 1.000
#> GSM97097 2 0.2778 0.915 0.048 0.952
#> GSM97107 1 0.3584 0.902 0.932 0.068
#> GSM97054 2 0.2778 0.915 0.048 0.952
#> GSM97062 2 0.2778 0.915 0.048 0.952
#> GSM97069 2 0.0376 0.918 0.004 0.996
#> GSM97070 2 0.0376 0.918 0.004 0.996
#> GSM97073 2 0.0376 0.918 0.004 0.996
#> GSM97076 1 0.4161 0.883 0.916 0.084
#> GSM97077 2 0.2423 0.917 0.040 0.960
#> GSM97095 1 0.2778 0.918 0.952 0.048
#> GSM97102 2 0.0376 0.918 0.004 0.996
#> GSM97109 2 0.7815 0.778 0.232 0.768
#> GSM97110 2 0.6973 0.821 0.188 0.812
#> GSM97074 1 0.8207 0.684 0.744 0.256
#> GSM97085 2 0.3584 0.880 0.068 0.932
#> GSM97059 1 0.6712 0.762 0.824 0.176
#> GSM97072 2 0.0376 0.918 0.004 0.996
#> GSM97078 1 0.9795 0.387 0.584 0.416
#> GSM97067 2 0.0376 0.918 0.004 0.996
#> GSM97087 2 0.0376 0.918 0.004 0.996
#> GSM97111 2 0.4298 0.895 0.088 0.912
#> GSM97064 2 0.0000 0.918 0.000 1.000
#> GSM97065 2 0.0376 0.918 0.004 0.996
#> GSM97081 2 0.0000 0.918 0.000 1.000
#> GSM97082 2 0.0376 0.918 0.004 0.996
#> GSM97088 2 0.5178 0.835 0.116 0.884
#> GSM97100 2 0.3274 0.910 0.060 0.940
#> GSM97104 2 0.0376 0.918 0.004 0.996
#> GSM97108 2 0.7815 0.778 0.232 0.768
#> GSM97050 2 0.2423 0.917 0.040 0.960
#> GSM97080 2 0.0376 0.918 0.004 0.996
#> GSM97089 2 0.0376 0.918 0.004 0.996
#> GSM97092 2 0.0000 0.918 0.000 1.000
#> GSM97093 2 0.2423 0.917 0.040 0.960
#> GSM97058 2 0.2423 0.917 0.040 0.960
#> GSM97051 2 0.2236 0.917 0.036 0.964
#> GSM97052 2 0.0000 0.918 0.000 1.000
#> GSM97061 2 0.0000 0.918 0.000 1.000
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM97138 1 0.2356 0.9293 0.928 0.072 0.000
#> GSM97145 1 0.2537 0.9275 0.920 0.080 0.000
#> GSM97147 2 0.1267 0.8278 0.024 0.972 0.004
#> GSM97125 1 0.0747 0.9328 0.984 0.016 0.000
#> GSM97127 1 0.2537 0.9275 0.920 0.080 0.000
#> GSM97130 1 0.4002 0.8962 0.840 0.160 0.000
#> GSM97133 1 0.2711 0.9253 0.912 0.088 0.000
#> GSM97134 1 0.2625 0.8999 0.916 0.084 0.000
#> GSM97120 1 0.2711 0.9253 0.912 0.088 0.000
#> GSM97126 1 0.3715 0.8870 0.868 0.128 0.004
#> GSM97112 1 0.0592 0.9304 0.988 0.000 0.012
#> GSM97115 2 0.3267 0.7453 0.116 0.884 0.000
#> GSM97116 1 0.2711 0.9253 0.912 0.088 0.000
#> GSM97117 2 0.3879 0.8794 0.000 0.848 0.152
#> GSM97119 1 0.0592 0.9304 0.988 0.000 0.012
#> GSM97122 1 0.0592 0.9304 0.988 0.000 0.012
#> GSM97135 1 0.0592 0.9304 0.988 0.000 0.012
#> GSM97136 3 0.4891 0.7895 0.040 0.124 0.836
#> GSM97139 1 0.2711 0.9253 0.912 0.088 0.000
#> GSM97146 1 0.2711 0.9253 0.912 0.088 0.000
#> GSM97123 2 0.5465 0.7546 0.000 0.712 0.288
#> GSM97129 2 0.3752 0.8819 0.000 0.856 0.144
#> GSM97143 1 0.0592 0.9304 0.988 0.000 0.012
#> GSM97113 2 0.2625 0.8658 0.000 0.916 0.084
#> GSM97056 1 0.2796 0.9245 0.908 0.092 0.000
#> GSM97124 1 0.0000 0.9317 1.000 0.000 0.000
#> GSM97132 1 0.1753 0.9192 0.952 0.048 0.000
#> GSM97144 1 0.2537 0.9021 0.920 0.080 0.000
#> GSM97149 1 0.2711 0.9253 0.912 0.088 0.000
#> GSM97068 2 0.0747 0.8248 0.016 0.984 0.000
#> GSM97071 3 0.4209 0.8155 0.020 0.120 0.860
#> GSM97086 2 0.2448 0.8693 0.000 0.924 0.076
#> GSM97103 2 0.5431 0.7566 0.000 0.716 0.284
#> GSM97057 2 0.1482 0.8384 0.012 0.968 0.020
#> GSM97060 3 0.0747 0.8794 0.000 0.016 0.984
#> GSM97075 2 0.4062 0.8742 0.000 0.836 0.164
#> GSM97098 2 0.5560 0.7382 0.000 0.700 0.300
#> GSM97099 2 0.3686 0.8828 0.000 0.860 0.140
#> GSM97101 2 0.3619 0.8837 0.000 0.864 0.136
#> GSM97105 2 0.3752 0.8827 0.000 0.856 0.144
#> GSM97106 3 0.6299 -0.1763 0.000 0.476 0.524
#> GSM97121 2 0.3619 0.8837 0.000 0.864 0.136
#> GSM97128 3 0.6109 0.7214 0.140 0.080 0.780
#> GSM97131 2 0.4062 0.8725 0.000 0.836 0.164
#> GSM97137 1 0.3267 0.9194 0.884 0.116 0.000
#> GSM97118 1 0.2446 0.9149 0.936 0.052 0.012
#> GSM97114 2 0.3120 0.8605 0.012 0.908 0.080
#> GSM97142 1 0.0592 0.9304 0.988 0.000 0.012
#> GSM97140 2 0.2261 0.8720 0.000 0.932 0.068
#> GSM97141 2 0.3686 0.8828 0.000 0.860 0.140
#> GSM97055 1 0.1878 0.9154 0.952 0.004 0.044
#> GSM97090 2 0.4887 0.5987 0.228 0.772 0.000
#> GSM97091 1 0.0592 0.9304 0.988 0.000 0.012
#> GSM97148 1 0.2711 0.9253 0.912 0.088 0.000
#> GSM97063 1 0.0592 0.9304 0.988 0.000 0.012
#> GSM97053 1 0.0000 0.9317 1.000 0.000 0.000
#> GSM97066 3 0.0592 0.8810 0.000 0.012 0.988
#> GSM97079 2 0.3116 0.8778 0.000 0.892 0.108
#> GSM97083 1 0.4146 0.8785 0.876 0.080 0.044
#> GSM97084 2 0.3031 0.8649 0.012 0.912 0.076
#> GSM97094 2 0.5873 0.5315 0.312 0.684 0.004
#> GSM97096 3 0.4796 0.6390 0.000 0.220 0.780
#> GSM97097 2 0.4750 0.8348 0.000 0.784 0.216
#> GSM97107 2 0.6359 0.2986 0.404 0.592 0.004
#> GSM97054 2 0.2448 0.8693 0.000 0.924 0.076
#> GSM97062 2 0.2537 0.8691 0.000 0.920 0.080
#> GSM97069 3 0.0592 0.8810 0.000 0.012 0.988
#> GSM97070 3 0.0592 0.8810 0.000 0.012 0.988
#> GSM97073 3 0.0592 0.8810 0.000 0.012 0.988
#> GSM97076 1 0.7816 0.4814 0.628 0.084 0.288
#> GSM97077 2 0.3116 0.8786 0.000 0.892 0.108
#> GSM97095 2 0.2711 0.7735 0.088 0.912 0.000
#> GSM97102 3 0.0592 0.8810 0.000 0.012 0.988
#> GSM97109 2 0.3851 0.8833 0.004 0.860 0.136
#> GSM97110 2 0.3686 0.8828 0.000 0.860 0.140
#> GSM97074 3 0.5875 0.7153 0.160 0.056 0.784
#> GSM97085 3 0.3129 0.8174 0.088 0.008 0.904
#> GSM97059 2 0.0747 0.8248 0.016 0.984 0.000
#> GSM97072 3 0.0747 0.8794 0.000 0.016 0.984
#> GSM97078 3 0.6109 0.7214 0.140 0.080 0.780
#> GSM97067 3 0.0592 0.8810 0.000 0.012 0.988
#> GSM97087 3 0.0592 0.8810 0.000 0.012 0.988
#> GSM97111 2 0.3752 0.8819 0.000 0.856 0.144
#> GSM97064 2 0.4178 0.8693 0.000 0.828 0.172
#> GSM97065 3 0.6267 0.0329 0.000 0.452 0.548
#> GSM97081 3 0.4291 0.7066 0.000 0.180 0.820
#> GSM97082 3 0.0592 0.8810 0.000 0.012 0.988
#> GSM97088 3 0.5093 0.7721 0.088 0.076 0.836
#> GSM97100 2 0.2261 0.8720 0.000 0.932 0.068
#> GSM97104 3 0.0592 0.8810 0.000 0.012 0.988
#> GSM97108 2 0.3551 0.8845 0.000 0.868 0.132
#> GSM97050 2 0.3816 0.8799 0.000 0.852 0.148
#> GSM97080 3 0.0592 0.8810 0.000 0.012 0.988
#> GSM97089 3 0.0592 0.8810 0.000 0.012 0.988
#> GSM97092 3 0.0592 0.8810 0.000 0.012 0.988
#> GSM97093 2 0.3619 0.8825 0.000 0.864 0.136
#> GSM97058 2 0.4062 0.8740 0.000 0.836 0.164
#> GSM97051 2 0.3551 0.8796 0.000 0.868 0.132
#> GSM97052 3 0.0747 0.8787 0.000 0.016 0.984
#> GSM97061 2 0.5621 0.7255 0.000 0.692 0.308
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM97138 1 0.1256 0.7523 0.964 0.008 0.000 0.028
#> GSM97145 1 0.3102 0.7641 0.872 0.008 0.004 0.116
#> GSM97147 2 0.2742 0.7850 0.024 0.900 0.000 0.076
#> GSM97125 1 0.3538 0.7651 0.832 0.004 0.004 0.160
#> GSM97127 1 0.1042 0.7522 0.972 0.008 0.000 0.020
#> GSM97130 1 0.5678 0.0401 0.524 0.024 0.000 0.452
#> GSM97133 1 0.0336 0.7450 0.992 0.008 0.000 0.000
#> GSM97134 4 0.4483 0.0619 0.284 0.004 0.000 0.712
#> GSM97120 1 0.0336 0.7450 0.992 0.008 0.000 0.000
#> GSM97126 1 0.7344 0.4535 0.528 0.224 0.000 0.248
#> GSM97112 1 0.4713 0.7410 0.700 0.004 0.004 0.292
#> GSM97115 4 0.6658 0.0806 0.084 0.444 0.000 0.472
#> GSM97116 1 0.0336 0.7450 0.992 0.008 0.000 0.000
#> GSM97117 2 0.2214 0.7801 0.028 0.928 0.044 0.000
#> GSM97119 1 0.4713 0.7410 0.700 0.004 0.004 0.292
#> GSM97122 1 0.4661 0.7446 0.708 0.004 0.004 0.284
#> GSM97135 1 0.4579 0.7485 0.720 0.004 0.004 0.272
#> GSM97136 3 0.7609 0.1739 0.004 0.396 0.428 0.172
#> GSM97139 1 0.0336 0.7450 0.992 0.008 0.000 0.000
#> GSM97146 1 0.0336 0.7450 0.992 0.008 0.000 0.000
#> GSM97123 2 0.6369 0.3237 0.000 0.572 0.352 0.076
#> GSM97129 2 0.1471 0.7933 0.024 0.960 0.004 0.012
#> GSM97143 1 0.4687 0.7425 0.704 0.004 0.004 0.288
#> GSM97113 2 0.2149 0.7668 0.088 0.912 0.000 0.000
#> GSM97056 1 0.3450 0.6010 0.836 0.008 0.000 0.156
#> GSM97124 1 0.4579 0.7481 0.720 0.004 0.004 0.272
#> GSM97132 1 0.4933 0.5678 0.568 0.000 0.000 0.432
#> GSM97144 4 0.5062 0.0810 0.300 0.020 0.000 0.680
#> GSM97149 1 0.0336 0.7450 0.992 0.008 0.000 0.000
#> GSM97068 2 0.5374 0.5474 0.052 0.704 0.000 0.244
#> GSM97071 4 0.5478 0.3408 0.000 0.056 0.248 0.696
#> GSM97086 2 0.5137 0.1806 0.000 0.544 0.004 0.452
#> GSM97103 2 0.5657 0.4476 0.000 0.644 0.312 0.044
#> GSM97057 2 0.3687 0.7674 0.080 0.856 0.000 0.064
#> GSM97060 3 0.1398 0.8149 0.000 0.004 0.956 0.040
#> GSM97075 2 0.1471 0.7933 0.012 0.960 0.024 0.004
#> GSM97098 2 0.5888 0.1429 0.000 0.540 0.424 0.036
#> GSM97099 2 0.1209 0.7932 0.032 0.964 0.004 0.000
#> GSM97101 2 0.1109 0.7941 0.028 0.968 0.004 0.000
#> GSM97105 2 0.1890 0.7911 0.000 0.936 0.008 0.056
#> GSM97106 3 0.6352 0.5546 0.000 0.188 0.656 0.156
#> GSM97121 2 0.1114 0.7959 0.016 0.972 0.004 0.008
#> GSM97128 4 0.4608 0.2759 0.004 0.000 0.304 0.692
#> GSM97131 2 0.3485 0.7633 0.000 0.856 0.028 0.116
#> GSM97137 1 0.3810 0.5701 0.804 0.008 0.000 0.188
#> GSM97118 4 0.5000 -0.5219 0.496 0.000 0.000 0.504
#> GSM97114 2 0.1557 0.7840 0.056 0.944 0.000 0.000
#> GSM97142 1 0.4713 0.7410 0.700 0.004 0.004 0.292
#> GSM97140 2 0.2450 0.7852 0.016 0.912 0.000 0.072
#> GSM97141 2 0.1209 0.7932 0.032 0.964 0.004 0.000
#> GSM97055 1 0.5545 0.6683 0.612 0.004 0.020 0.364
#> GSM97090 4 0.6894 0.2557 0.112 0.376 0.000 0.512
#> GSM97091 1 0.5013 0.7004 0.644 0.004 0.004 0.348
#> GSM97148 1 0.0336 0.7450 0.992 0.008 0.000 0.000
#> GSM97063 1 0.4995 0.7042 0.648 0.004 0.004 0.344
#> GSM97053 1 0.4198 0.7595 0.768 0.004 0.004 0.224
#> GSM97066 3 0.2342 0.8109 0.000 0.008 0.912 0.080
#> GSM97079 2 0.5151 0.1383 0.000 0.532 0.004 0.464
#> GSM97083 4 0.4467 0.2368 0.172 0.000 0.040 0.788
#> GSM97084 4 0.5151 0.0111 0.000 0.464 0.004 0.532
#> GSM97094 4 0.4997 0.4643 0.036 0.216 0.004 0.744
#> GSM97096 3 0.6078 0.4723 0.000 0.312 0.620 0.068
#> GSM97097 2 0.7193 0.2648 0.000 0.508 0.152 0.340
#> GSM97107 4 0.5215 0.4739 0.052 0.204 0.004 0.740
#> GSM97054 4 0.4999 -0.0420 0.000 0.492 0.000 0.508
#> GSM97062 4 0.5151 0.0111 0.000 0.464 0.004 0.532
#> GSM97069 3 0.2053 0.8111 0.000 0.004 0.924 0.072
#> GSM97070 3 0.2342 0.8109 0.000 0.008 0.912 0.080
#> GSM97073 3 0.2125 0.8103 0.000 0.004 0.920 0.076
#> GSM97076 4 0.9001 -0.0322 0.300 0.124 0.132 0.444
#> GSM97077 2 0.2149 0.7793 0.000 0.912 0.000 0.088
#> GSM97095 4 0.6334 0.0733 0.060 0.456 0.000 0.484
#> GSM97102 3 0.0592 0.8209 0.000 0.016 0.984 0.000
#> GSM97109 2 0.2221 0.7874 0.044 0.932 0.008 0.016
#> GSM97110 2 0.2221 0.7874 0.044 0.932 0.008 0.016
#> GSM97074 4 0.5290 -0.0765 0.008 0.000 0.476 0.516
#> GSM97085 3 0.4335 0.6166 0.004 0.004 0.752 0.240
#> GSM97059 2 0.4720 0.6516 0.044 0.768 0.000 0.188
#> GSM97072 3 0.1978 0.8145 0.000 0.004 0.928 0.068
#> GSM97078 4 0.4372 0.3262 0.004 0.000 0.268 0.728
#> GSM97067 3 0.2125 0.8103 0.000 0.004 0.920 0.076
#> GSM97087 3 0.1624 0.8157 0.000 0.020 0.952 0.028
#> GSM97111 2 0.0895 0.7951 0.020 0.976 0.004 0.000
#> GSM97064 2 0.3697 0.7674 0.000 0.852 0.048 0.100
#> GSM97065 2 0.5664 0.5963 0.032 0.748 0.164 0.056
#> GSM97081 3 0.5576 0.2106 0.000 0.444 0.536 0.020
#> GSM97082 3 0.0524 0.8209 0.000 0.008 0.988 0.004
#> GSM97088 4 0.5165 -0.0526 0.004 0.000 0.484 0.512
#> GSM97100 2 0.2469 0.7681 0.000 0.892 0.000 0.108
#> GSM97104 3 0.0188 0.8204 0.000 0.004 0.996 0.000
#> GSM97108 2 0.0992 0.7959 0.012 0.976 0.004 0.008
#> GSM97050 2 0.3166 0.7719 0.000 0.868 0.016 0.116
#> GSM97080 3 0.1890 0.8157 0.000 0.008 0.936 0.056
#> GSM97089 3 0.1733 0.8148 0.000 0.024 0.948 0.028
#> GSM97092 3 0.2996 0.7837 0.000 0.064 0.892 0.044
#> GSM97093 2 0.3778 0.7637 0.000 0.848 0.052 0.100
#> GSM97058 2 0.2198 0.7856 0.000 0.920 0.008 0.072
#> GSM97051 2 0.4225 0.7220 0.000 0.792 0.024 0.184
#> GSM97052 3 0.3392 0.7722 0.000 0.072 0.872 0.056
#> GSM97061 2 0.6932 0.1885 0.000 0.492 0.396 0.112
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM97138 1 0.2612 0.64006 0.868 0.008 0.000 0.000 0.124
#> GSM97145 1 0.4924 0.56775 0.668 0.060 0.000 0.000 0.272
#> GSM97147 2 0.2914 0.73088 0.016 0.872 0.000 0.100 0.012
#> GSM97125 1 0.4135 0.56342 0.656 0.004 0.000 0.000 0.340
#> GSM97127 1 0.1809 0.65189 0.928 0.012 0.000 0.000 0.060
#> GSM97130 4 0.6548 0.36926 0.292 0.012 0.000 0.524 0.172
#> GSM97133 1 0.0566 0.65055 0.984 0.012 0.000 0.004 0.000
#> GSM97134 4 0.5921 -0.00263 0.088 0.004 0.000 0.460 0.448
#> GSM97120 1 0.0404 0.65195 0.988 0.012 0.000 0.000 0.000
#> GSM97126 2 0.7196 -0.27515 0.244 0.368 0.000 0.020 0.368
#> GSM97112 1 0.4262 0.47152 0.560 0.000 0.000 0.000 0.440
#> GSM97115 4 0.6587 0.65107 0.080 0.164 0.000 0.624 0.132
#> GSM97116 1 0.0290 0.65293 0.992 0.008 0.000 0.000 0.000
#> GSM97117 2 0.1267 0.74239 0.004 0.960 0.012 0.000 0.024
#> GSM97119 1 0.4262 0.47152 0.560 0.000 0.000 0.000 0.440
#> GSM97122 1 0.4262 0.47152 0.560 0.000 0.000 0.000 0.440
#> GSM97135 1 0.4227 0.49710 0.580 0.000 0.000 0.000 0.420
#> GSM97136 2 0.6465 0.31350 0.004 0.548 0.204 0.004 0.240
#> GSM97139 1 0.0290 0.65293 0.992 0.008 0.000 0.000 0.000
#> GSM97146 1 0.0566 0.65055 0.984 0.012 0.000 0.004 0.000
#> GSM97123 2 0.7197 0.18993 0.000 0.460 0.360 0.092 0.088
#> GSM97129 2 0.1847 0.74857 0.004 0.940 0.008 0.020 0.028
#> GSM97143 1 0.4278 0.45170 0.548 0.000 0.000 0.000 0.452
#> GSM97113 2 0.1651 0.74531 0.036 0.944 0.000 0.008 0.012
#> GSM97056 1 0.3576 0.49119 0.840 0.012 0.000 0.100 0.048
#> GSM97124 1 0.4367 0.50073 0.580 0.004 0.000 0.000 0.416
#> GSM97132 5 0.6215 0.09841 0.340 0.004 0.000 0.136 0.520
#> GSM97144 4 0.5763 0.37944 0.108 0.004 0.000 0.600 0.288
#> GSM97149 1 0.0566 0.65055 0.984 0.012 0.000 0.004 0.000
#> GSM97068 2 0.6241 0.29573 0.068 0.556 0.000 0.336 0.040
#> GSM97071 4 0.6141 0.27880 0.004 0.000 0.164 0.572 0.260
#> GSM97086 4 0.3171 0.63462 0.000 0.176 0.000 0.816 0.008
#> GSM97103 2 0.6227 0.31366 0.000 0.584 0.300 0.072 0.044
#> GSM97057 2 0.4943 0.66926 0.088 0.752 0.000 0.132 0.028
#> GSM97060 3 0.1799 0.74200 0.000 0.020 0.940 0.012 0.028
#> GSM97075 2 0.1372 0.74198 0.004 0.956 0.016 0.000 0.024
#> GSM97098 2 0.6233 0.21304 0.000 0.548 0.348 0.060 0.044
#> GSM97099 2 0.1153 0.74236 0.008 0.964 0.004 0.000 0.024
#> GSM97101 2 0.0898 0.74790 0.008 0.972 0.000 0.020 0.000
#> GSM97105 2 0.3340 0.72272 0.000 0.840 0.004 0.124 0.032
#> GSM97106 3 0.6618 0.54339 0.000 0.092 0.616 0.192 0.100
#> GSM97121 2 0.1121 0.74690 0.000 0.956 0.000 0.044 0.000
#> GSM97128 5 0.5435 0.33552 0.004 0.000 0.104 0.236 0.656
#> GSM97131 2 0.5788 0.60922 0.000 0.648 0.064 0.248 0.040
#> GSM97137 1 0.4397 0.40365 0.784 0.012 0.000 0.116 0.088
#> GSM97118 5 0.5543 0.32067 0.224 0.000 0.000 0.136 0.640
#> GSM97114 2 0.1710 0.74418 0.020 0.944 0.000 0.012 0.024
#> GSM97142 1 0.4262 0.47152 0.560 0.000 0.000 0.000 0.440
#> GSM97140 2 0.3059 0.72607 0.000 0.860 0.004 0.108 0.028
#> GSM97141 2 0.1200 0.74643 0.008 0.964 0.000 0.012 0.016
#> GSM97055 5 0.4748 -0.11478 0.384 0.000 0.004 0.016 0.596
#> GSM97090 4 0.6416 0.65147 0.100 0.116 0.000 0.648 0.136
#> GSM97091 5 0.4443 -0.34320 0.472 0.000 0.000 0.004 0.524
#> GSM97148 1 0.0566 0.65055 0.984 0.012 0.000 0.004 0.000
#> GSM97063 5 0.4448 -0.36182 0.480 0.000 0.000 0.004 0.516
#> GSM97053 1 0.4088 0.54392 0.632 0.000 0.000 0.000 0.368
#> GSM97066 3 0.4505 0.69497 0.004 0.016 0.776 0.052 0.152
#> GSM97079 4 0.3622 0.63683 0.000 0.172 0.008 0.804 0.016
#> GSM97083 5 0.5063 0.14325 0.024 0.000 0.012 0.360 0.604
#> GSM97084 4 0.2124 0.68939 0.000 0.096 0.000 0.900 0.004
#> GSM97094 4 0.3953 0.63991 0.000 0.048 0.000 0.784 0.168
#> GSM97096 3 0.6381 0.44666 0.000 0.292 0.580 0.068 0.060
#> GSM97097 4 0.5668 0.53227 0.000 0.168 0.112 0.688 0.032
#> GSM97107 4 0.4191 0.64486 0.012 0.044 0.000 0.784 0.160
#> GSM97054 4 0.3909 0.67669 0.000 0.148 0.004 0.800 0.048
#> GSM97062 4 0.2286 0.68826 0.000 0.108 0.000 0.888 0.004
#> GSM97069 3 0.4211 0.70597 0.004 0.016 0.796 0.040 0.144
#> GSM97070 3 0.4396 0.70111 0.004 0.016 0.784 0.048 0.148
#> GSM97073 3 0.4437 0.70139 0.004 0.016 0.780 0.048 0.152
#> GSM97076 5 0.8483 0.21066 0.056 0.228 0.132 0.112 0.472
#> GSM97077 2 0.3880 0.68879 0.000 0.784 0.004 0.184 0.028
#> GSM97095 4 0.6782 0.62589 0.064 0.204 0.000 0.588 0.144
#> GSM97102 3 0.2474 0.74033 0.000 0.040 0.908 0.012 0.040
#> GSM97109 2 0.2438 0.72514 0.008 0.912 0.004 0.044 0.032
#> GSM97110 2 0.2438 0.72514 0.008 0.912 0.004 0.044 0.032
#> GSM97074 5 0.5963 0.24513 0.004 0.000 0.288 0.128 0.580
#> GSM97085 3 0.5037 0.39406 0.000 0.000 0.584 0.040 0.376
#> GSM97059 2 0.5806 0.52846 0.072 0.648 0.000 0.244 0.036
#> GSM97072 3 0.4133 0.71699 0.004 0.016 0.808 0.048 0.124
#> GSM97078 5 0.5870 0.24667 0.000 0.000 0.140 0.276 0.584
#> GSM97067 3 0.4396 0.70111 0.004 0.016 0.784 0.048 0.148
#> GSM97087 3 0.2742 0.72876 0.000 0.020 0.892 0.020 0.068
#> GSM97111 2 0.1153 0.74249 0.004 0.964 0.008 0.000 0.024
#> GSM97064 2 0.6275 0.61687 0.000 0.644 0.076 0.192 0.088
#> GSM97065 2 0.5242 0.59657 0.008 0.752 0.084 0.044 0.112
#> GSM97081 3 0.5551 0.08127 0.000 0.464 0.484 0.016 0.036
#> GSM97082 3 0.1787 0.74019 0.000 0.016 0.940 0.012 0.032
#> GSM97088 5 0.6410 0.23863 0.000 0.000 0.320 0.192 0.488
#> GSM97100 2 0.4238 0.65622 0.000 0.740 0.004 0.228 0.028
#> GSM97104 3 0.2040 0.74280 0.000 0.032 0.928 0.008 0.032
#> GSM97108 2 0.1357 0.74613 0.000 0.948 0.000 0.048 0.004
#> GSM97050 2 0.5985 0.60050 0.000 0.640 0.040 0.236 0.084
#> GSM97080 3 0.3585 0.72173 0.004 0.016 0.844 0.032 0.104
#> GSM97089 3 0.2742 0.72876 0.000 0.020 0.892 0.020 0.068
#> GSM97092 3 0.3828 0.70355 0.000 0.068 0.832 0.020 0.080
#> GSM97093 2 0.5907 0.65224 0.000 0.688 0.104 0.140 0.068
#> GSM97058 2 0.3951 0.68425 0.000 0.776 0.004 0.192 0.028
#> GSM97051 2 0.6933 0.44584 0.000 0.516 0.076 0.320 0.088
#> GSM97052 3 0.4559 0.68249 0.000 0.072 0.792 0.048 0.088
#> GSM97061 3 0.7592 0.11289 0.000 0.308 0.452 0.152 0.088
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM97138 1 0.2300 0.62369 0.856 0.000 0.000 0.000 0.144 0.000
#> GSM97145 1 0.4897 0.07791 0.556 0.056 0.000 0.004 0.384 0.000
#> GSM97147 2 0.4088 0.65328 0.000 0.780 0.104 0.096 0.020 0.000
#> GSM97125 1 0.3971 -0.04908 0.548 0.000 0.000 0.004 0.448 0.000
#> GSM97127 1 0.1908 0.68821 0.900 0.000 0.000 0.004 0.096 0.000
#> GSM97130 4 0.5948 0.44012 0.272 0.000 0.008 0.508 0.212 0.000
#> GSM97133 1 0.0436 0.74760 0.988 0.000 0.004 0.004 0.004 0.000
#> GSM97134 5 0.5452 -0.03264 0.048 0.012 0.024 0.348 0.568 0.000
#> GSM97120 1 0.0363 0.74766 0.988 0.000 0.000 0.000 0.012 0.000
#> GSM97126 2 0.6233 0.00385 0.124 0.448 0.020 0.012 0.396 0.000
#> GSM97112 5 0.3857 0.22416 0.468 0.000 0.000 0.000 0.532 0.000
#> GSM97115 4 0.6829 0.63411 0.064 0.096 0.088 0.588 0.164 0.000
#> GSM97116 1 0.0260 0.74799 0.992 0.000 0.000 0.000 0.008 0.000
#> GSM97117 2 0.0146 0.70750 0.000 0.996 0.000 0.000 0.004 0.000
#> GSM97119 5 0.3854 0.22502 0.464 0.000 0.000 0.000 0.536 0.000
#> GSM97122 5 0.3860 0.21621 0.472 0.000 0.000 0.000 0.528 0.000
#> GSM97135 5 0.3866 0.18277 0.484 0.000 0.000 0.000 0.516 0.000
#> GSM97136 2 0.5328 0.42891 0.000 0.676 0.136 0.004 0.152 0.032
#> GSM97139 1 0.0260 0.74799 0.992 0.000 0.000 0.000 0.008 0.000
#> GSM97146 1 0.0291 0.74842 0.992 0.000 0.004 0.004 0.000 0.000
#> GSM97123 3 0.4947 0.46196 0.000 0.196 0.692 0.032 0.000 0.080
#> GSM97129 2 0.1341 0.70915 0.000 0.948 0.028 0.000 0.024 0.000
#> GSM97143 5 0.3961 0.24889 0.440 0.000 0.000 0.004 0.556 0.000
#> GSM97113 2 0.0858 0.70596 0.028 0.968 0.004 0.000 0.000 0.000
#> GSM97056 1 0.3419 0.56827 0.824 0.000 0.008 0.096 0.072 0.000
#> GSM97124 5 0.4086 0.20430 0.464 0.000 0.000 0.008 0.528 0.000
#> GSM97132 5 0.5494 0.33783 0.208 0.000 0.016 0.160 0.616 0.000
#> GSM97144 4 0.4880 0.48200 0.048 0.000 0.012 0.596 0.344 0.000
#> GSM97149 1 0.0291 0.74842 0.992 0.000 0.004 0.004 0.000 0.000
#> GSM97068 2 0.7227 0.22450 0.060 0.456 0.124 0.312 0.048 0.000
#> GSM97071 4 0.6777 0.18808 0.000 0.000 0.072 0.420 0.160 0.348
#> GSM97086 4 0.3029 0.65639 0.000 0.052 0.088 0.852 0.008 0.000
#> GSM97103 2 0.6388 0.29609 0.000 0.596 0.224 0.068 0.040 0.072
#> GSM97057 2 0.6140 0.54241 0.084 0.620 0.156 0.132 0.008 0.000
#> GSM97060 3 0.4664 0.09828 0.000 0.000 0.488 0.004 0.032 0.476
#> GSM97075 2 0.0405 0.70805 0.000 0.988 0.008 0.000 0.004 0.000
#> GSM97098 2 0.5902 0.21802 0.000 0.588 0.276 0.012 0.040 0.084
#> GSM97099 2 0.0508 0.70548 0.000 0.984 0.012 0.000 0.004 0.000
#> GSM97101 2 0.0547 0.70950 0.000 0.980 0.020 0.000 0.000 0.000
#> GSM97105 2 0.4062 0.63782 0.000 0.764 0.160 0.064 0.012 0.000
#> GSM97106 3 0.5108 0.39862 0.000 0.004 0.692 0.092 0.032 0.180
#> GSM97121 2 0.1745 0.70012 0.000 0.920 0.068 0.000 0.012 0.000
#> GSM97128 5 0.6166 0.15700 0.000 0.000 0.148 0.168 0.596 0.088
#> GSM97131 2 0.6259 0.29064 0.000 0.464 0.312 0.208 0.012 0.004
#> GSM97137 1 0.3860 0.51827 0.788 0.000 0.008 0.108 0.096 0.000
#> GSM97118 5 0.4344 0.34545 0.036 0.000 0.060 0.120 0.776 0.008
#> GSM97114 2 0.0000 0.70822 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97142 5 0.3857 0.22416 0.468 0.000 0.000 0.000 0.532 0.000
#> GSM97140 2 0.4010 0.63898 0.000 0.772 0.148 0.068 0.012 0.000
#> GSM97141 2 0.0000 0.70822 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97055 5 0.5512 0.36422 0.236 0.012 0.076 0.020 0.648 0.008
#> GSM97090 4 0.6235 0.65096 0.064 0.044 0.084 0.632 0.176 0.000
#> GSM97091 5 0.3938 0.34188 0.324 0.000 0.016 0.000 0.660 0.000
#> GSM97148 1 0.0291 0.74842 0.992 0.000 0.004 0.004 0.000 0.000
#> GSM97063 5 0.3984 0.33482 0.336 0.000 0.016 0.000 0.648 0.000
#> GSM97053 1 0.3866 -0.18390 0.516 0.000 0.000 0.000 0.484 0.000
#> GSM97066 6 0.0291 0.67204 0.000 0.000 0.004 0.000 0.004 0.992
#> GSM97079 4 0.3127 0.65140 0.000 0.040 0.104 0.844 0.012 0.000
#> GSM97083 5 0.5387 0.13003 0.008 0.000 0.136 0.212 0.636 0.008
#> GSM97084 4 0.2224 0.69040 0.000 0.020 0.064 0.904 0.012 0.000
#> GSM97094 4 0.3360 0.66743 0.004 0.016 0.008 0.804 0.168 0.000
#> GSM97096 3 0.6694 0.25533 0.000 0.292 0.480 0.012 0.040 0.176
#> GSM97097 4 0.5191 0.52694 0.000 0.108 0.160 0.692 0.036 0.004
#> GSM97107 4 0.3431 0.66881 0.012 0.012 0.012 0.812 0.152 0.000
#> GSM97054 4 0.4403 0.63055 0.000 0.044 0.172 0.744 0.040 0.000
#> GSM97062 4 0.2290 0.67804 0.000 0.020 0.084 0.892 0.004 0.000
#> GSM97069 6 0.0632 0.66763 0.000 0.000 0.024 0.000 0.000 0.976
#> GSM97070 6 0.0260 0.67241 0.000 0.000 0.008 0.000 0.000 0.992
#> GSM97073 6 0.0458 0.67056 0.000 0.000 0.016 0.000 0.000 0.984
#> GSM97076 6 0.6966 0.29270 0.004 0.272 0.048 0.048 0.108 0.520
#> GSM97077 2 0.5098 0.56421 0.000 0.664 0.176 0.148 0.012 0.000
#> GSM97095 4 0.7243 0.59594 0.056 0.140 0.088 0.532 0.184 0.000
#> GSM97102 6 0.5453 -0.02932 0.000 0.024 0.432 0.008 0.044 0.492
#> GSM97109 2 0.1382 0.69269 0.000 0.948 0.036 0.008 0.008 0.000
#> GSM97110 2 0.1382 0.69269 0.000 0.948 0.036 0.008 0.008 0.000
#> GSM97074 6 0.5896 0.36785 0.000 0.000 0.116 0.052 0.240 0.592
#> GSM97085 6 0.6023 0.39312 0.000 0.000 0.176 0.016 0.304 0.504
#> GSM97059 2 0.6875 0.38630 0.056 0.516 0.132 0.264 0.032 0.000
#> GSM97072 6 0.1257 0.65550 0.000 0.000 0.028 0.000 0.020 0.952
#> GSM97078 5 0.6610 0.09051 0.000 0.000 0.148 0.184 0.544 0.124
#> GSM97067 6 0.0146 0.67250 0.000 0.000 0.004 0.000 0.000 0.996
#> GSM97087 3 0.3797 0.25762 0.000 0.000 0.580 0.000 0.000 0.420
#> GSM97111 2 0.0748 0.70217 0.000 0.976 0.016 0.004 0.004 0.000
#> GSM97064 3 0.5770 0.03585 0.000 0.348 0.512 0.128 0.008 0.004
#> GSM97065 2 0.4223 0.33785 0.000 0.612 0.016 0.000 0.004 0.368
#> GSM97081 2 0.5762 -0.25432 0.000 0.440 0.424 0.004 0.004 0.128
#> GSM97082 3 0.3996 0.12798 0.000 0.000 0.512 0.000 0.004 0.484
#> GSM97088 5 0.6733 0.07604 0.000 0.000 0.192 0.164 0.524 0.120
#> GSM97100 2 0.5552 0.50595 0.000 0.600 0.180 0.208 0.012 0.000
#> GSM97104 6 0.4682 -0.01243 0.000 0.000 0.420 0.004 0.036 0.540
#> GSM97108 2 0.2019 0.69460 0.000 0.900 0.088 0.000 0.012 0.000
#> GSM97050 3 0.6224 -0.10934 0.000 0.364 0.412 0.212 0.012 0.000
#> GSM97080 6 0.2994 0.47644 0.000 0.000 0.208 0.000 0.004 0.788
#> GSM97089 3 0.3782 0.26925 0.000 0.000 0.588 0.000 0.000 0.412
#> GSM97092 3 0.3905 0.34127 0.000 0.004 0.636 0.004 0.000 0.356
#> GSM97093 3 0.5502 -0.08319 0.000 0.408 0.484 0.100 0.008 0.000
#> GSM97058 2 0.5098 0.56406 0.000 0.664 0.176 0.148 0.012 0.000
#> GSM97051 3 0.6400 0.15763 0.000 0.216 0.476 0.284 0.016 0.008
#> GSM97052 3 0.3894 0.36732 0.000 0.004 0.664 0.008 0.000 0.324
#> GSM97061 3 0.4717 0.46478 0.000 0.080 0.744 0.068 0.000 0.108
Heatmaps for the consensus matrix. It visualizes the probability of two samples to be in a same group.
consensus_heatmap(res, k = 2)
consensus_heatmap(res, k = 3)
consensus_heatmap(res, k = 4)
consensus_heatmap(res, k = 5)
consensus_heatmap(res, k = 6)
Heatmaps for the membership of samples in all partitions to see how consistent they are:
membership_heatmap(res, k = 2)
membership_heatmap(res, k = 3)
membership_heatmap(res, k = 4)
membership_heatmap(res, k = 5)
membership_heatmap(res, k = 6)
As soon as we have had the classes for columns, we can look for signatures which are significantly different between classes which can be candidate marks for certain classes. Following are the heatmaps for signatures.
Signature heatmaps where rows are scaled:
get_signatures(res, k = 2)
get_signatures(res, k = 3)
get_signatures(res, k = 4)
get_signatures(res, k = 5)
get_signatures(res, k = 6)
Signature heatmaps where rows are not scaled:
get_signatures(res, k = 2, scale_rows = FALSE)
get_signatures(res, k = 3, scale_rows = FALSE)
get_signatures(res, k = 4, scale_rows = FALSE)
get_signatures(res, k = 5, scale_rows = FALSE)
get_signatures(res, k = 6, scale_rows = FALSE)
Compare the overlap of signatures from different k:
compare_signatures(res)
get_signature() returns a data frame invisibly. TO get the list of signatures, the function
call should be assigned to a variable explicitly. In following code, if plot argument is set
to FALSE, no heatmap is plotted while only the differential analysis is performed.
# code only for demonstration
tb = get_signature(res, k = ..., plot = FALSE)
An example of the output of tb is:
#> which_row fdr mean_1 mean_2 scaled_mean_1 scaled_mean_2 km
#> 1 38 0.042760348 8.373488 9.131774 -0.5533452 0.5164555 1
#> 2 40 0.018707592 7.106213 8.469186 -0.6173731 0.5762149 1
#> 3 55 0.019134737 10.221463 11.207825 -0.6159697 0.5749050 1
#> 4 59 0.006059896 5.921854 7.869574 -0.6899429 0.6439467 1
#> 5 60 0.018055526 8.928898 10.211722 -0.6204761 0.5791110 1
#> 6 98 0.009384629 15.714769 14.887706 0.6635654 -0.6193277 2
...
The columns in tb are:
which_row: row indices corresponding to the input matrix.fdr: FDR for the differential test. mean_x: The mean value in group x.scaled_mean_x: The mean value in group x after rows are scaled.km: Row groups if k-means clustering is applied to rows.UMAP plot which shows how samples are separated.
dimension_reduction(res, k = 2, method = "UMAP")
dimension_reduction(res, k = 3, method = "UMAP")
dimension_reduction(res, k = 4, method = "UMAP")
dimension_reduction(res, k = 5, method = "UMAP")
dimension_reduction(res, k = 6, method = "UMAP")
Following heatmap shows how subgroups are split when increasing k:
collect_classes(res)
Test correlation between subgroups and known annotations. If the known annotation is numeric, one-way ANOVA test is applied, and if the known annotation is discrete, chi-squared contingency table test is applied.
test_to_known_factors(res)
#> n disease.state(p) specimen(p) cell.type(p) other(p) k
#> CV:kmeans 98 2.21e-04 0.403 2.16e-13 0.146 2
#> CV:kmeans 96 1.37e-05 0.175 1.05e-13 0.139 3
#> CV:kmeans 70 8.41e-05 0.158 4.56e-12 0.156 4
#> CV:kmeans 66 1.02e-04 0.191 1.61e-11 0.242 5
#> CV:kmeans 48 2.87e-03 0.093 5.36e-09 0.118 6
If matrix rows can be associated to genes, consider to use functional_enrichment(res,
...) to perform function enrichment for the signature genes. See this vignette for more detailed explanations.
The object with results only for a single top-value method and a single partition method can be extracted as:
res = res_list["CV", "skmeans"]
# you can also extract it by
# res = res_list["CV:skmeans"]
A summary of res and all the functions that can be applied to it:
res
#> A 'ConsensusPartition' object with k = 2, 3, 4, 5, 6.
#> On a matrix with 21512 rows and 100 columns.
#> Top rows (1000, 2000, 3000, 4000, 5000) are extracted by 'CV' method.
#> Subgroups are detected by 'skmeans' method.
#> Performed in total 1250 partitions by row resampling.
#> Best k for subgroups seems to be 2.
#>
#> Following methods can be applied to this 'ConsensusPartition' object:
#> [1] "cola_report" "collect_classes" "collect_plots"
#> [4] "collect_stats" "colnames" "compare_signatures"
#> [7] "consensus_heatmap" "dimension_reduction" "functional_enrichment"
#> [10] "get_anno_col" "get_anno" "get_classes"
#> [13] "get_consensus" "get_matrix" "get_membership"
#> [16] "get_param" "get_signatures" "get_stats"
#> [19] "is_best_k" "is_stable_k" "membership_heatmap"
#> [22] "ncol" "nrow" "plot_ecdf"
#> [25] "rownames" "select_partition_number" "show"
#> [28] "suggest_best_k" "test_to_known_factors"
collect_plots() function collects all the plots made from res for all k (number of partitions)
into one single page to provide an easy and fast comparison between different k.
collect_plots(res)
The plots are:
k and the heatmap of
predicted classes for each k.k.k.k.All the plots in panels can be made by individual functions and they are plotted later in this section.
select_partition_number() produces several plots showing different
statistics for choosing “optimized” k. There are following statistics:
k;k, the area increased is defined as \(A_k - A_{k-1}\).The detailed explanations of these statistics can be found in the cola vignette.
Generally speaking, lower PAC score, higher mean silhouette score or higher
concordance corresponds to better partition. Rand index and Jaccard index
measure how similar the current partition is compared to partition with k-1.
If they are too similar, we won't accept k is better than k-1.
select_partition_number(res)
The numeric values for all these statistics can be obtained by get_stats().
get_stats(res)
#> k 1-PAC mean_silhouette concordance area_increased Rand Jaccard
#> 2 2 0.999 0.965 0.985 0.5000 0.500 0.500
#> 3 3 0.768 0.834 0.924 0.3409 0.731 0.510
#> 4 4 0.564 0.587 0.791 0.1197 0.828 0.543
#> 5 5 0.552 0.381 0.648 0.0657 0.896 0.642
#> 6 6 0.589 0.398 0.638 0.0407 0.867 0.498
suggest_best_k() suggests the best \(k\) based on these statistics. The rules are as follows:
suggest_best_k(res)
#> [1] 2
Following shows the table of the partitions (You need to click the show/hide
code output link to see it). The membership matrix (columns with name p*)
is inferred by
clue::cl_consensus()
function with the SE method. Basically the value in the membership matrix
represents the probability to belong to a certain group. The finall class
label for an item is determined with the group with highest probability it
belongs to.
In get_classes() function, the entropy is calculated from the membership
matrix and the silhouette score is calculated from the consensus matrix.
cbind(get_classes(res, k = 2), get_membership(res, k = 2))
#> class entropy silhouette p1 p2
#> GSM97138 1 0.0000 0.980 1.000 0.000
#> GSM97145 1 0.0000 0.980 1.000 0.000
#> GSM97147 1 0.0000 0.980 1.000 0.000
#> GSM97125 1 0.0000 0.980 1.000 0.000
#> GSM97127 1 0.0000 0.980 1.000 0.000
#> GSM97130 1 0.0000 0.980 1.000 0.000
#> GSM97133 1 0.0000 0.980 1.000 0.000
#> GSM97134 1 0.0000 0.980 1.000 0.000
#> GSM97120 1 0.0000 0.980 1.000 0.000
#> GSM97126 1 0.0000 0.980 1.000 0.000
#> GSM97112 1 0.0000 0.980 1.000 0.000
#> GSM97115 1 0.0000 0.980 1.000 0.000
#> GSM97116 1 0.0000 0.980 1.000 0.000
#> GSM97117 2 0.0376 0.986 0.004 0.996
#> GSM97119 1 0.0000 0.980 1.000 0.000
#> GSM97122 1 0.0000 0.980 1.000 0.000
#> GSM97135 1 0.0000 0.980 1.000 0.000
#> GSM97136 2 0.8608 0.604 0.284 0.716
#> GSM97139 1 0.0000 0.980 1.000 0.000
#> GSM97146 1 0.0000 0.980 1.000 0.000
#> GSM97123 2 0.0000 0.988 0.000 1.000
#> GSM97129 2 0.3114 0.939 0.056 0.944
#> GSM97143 1 0.0000 0.980 1.000 0.000
#> GSM97113 2 0.0672 0.983 0.008 0.992
#> GSM97056 1 0.0000 0.980 1.000 0.000
#> GSM97124 1 0.0000 0.980 1.000 0.000
#> GSM97132 1 0.0000 0.980 1.000 0.000
#> GSM97144 1 0.0000 0.980 1.000 0.000
#> GSM97149 1 0.0000 0.980 1.000 0.000
#> GSM97068 1 0.9170 0.504 0.668 0.332
#> GSM97071 2 0.3431 0.930 0.064 0.936
#> GSM97086 2 0.0000 0.988 0.000 1.000
#> GSM97103 2 0.0000 0.988 0.000 1.000
#> GSM97057 2 0.4022 0.914 0.080 0.920
#> GSM97060 2 0.0000 0.988 0.000 1.000
#> GSM97075 2 0.0000 0.988 0.000 1.000
#> GSM97098 2 0.0000 0.988 0.000 1.000
#> GSM97099 2 0.0376 0.986 0.004 0.996
#> GSM97101 2 0.0376 0.986 0.004 0.996
#> GSM97105 2 0.0000 0.988 0.000 1.000
#> GSM97106 2 0.0000 0.988 0.000 1.000
#> GSM97121 2 0.0376 0.986 0.004 0.996
#> GSM97128 1 0.0376 0.977 0.996 0.004
#> GSM97131 2 0.0000 0.988 0.000 1.000
#> GSM97137 1 0.0000 0.980 1.000 0.000
#> GSM97118 1 0.0000 0.980 1.000 0.000
#> GSM97114 2 0.4298 0.905 0.088 0.912
#> GSM97142 1 0.0000 0.980 1.000 0.000
#> GSM97140 2 0.0938 0.980 0.012 0.988
#> GSM97141 2 0.0376 0.986 0.004 0.996
#> GSM97055 1 0.0000 0.980 1.000 0.000
#> GSM97090 1 0.0000 0.980 1.000 0.000
#> GSM97091 1 0.0000 0.980 1.000 0.000
#> GSM97148 1 0.0000 0.980 1.000 0.000
#> GSM97063 1 0.0000 0.980 1.000 0.000
#> GSM97053 1 0.0000 0.980 1.000 0.000
#> GSM97066 2 0.0000 0.988 0.000 1.000
#> GSM97079 2 0.0000 0.988 0.000 1.000
#> GSM97083 1 0.0376 0.977 0.996 0.004
#> GSM97084 2 0.0376 0.985 0.004 0.996
#> GSM97094 1 0.0000 0.980 1.000 0.000
#> GSM97096 2 0.0000 0.988 0.000 1.000
#> GSM97097 2 0.0000 0.988 0.000 1.000
#> GSM97107 1 0.0000 0.980 1.000 0.000
#> GSM97054 2 0.1843 0.965 0.028 0.972
#> GSM97062 2 0.0000 0.988 0.000 1.000
#> GSM97069 2 0.0000 0.988 0.000 1.000
#> GSM97070 2 0.0000 0.988 0.000 1.000
#> GSM97073 2 0.0000 0.988 0.000 1.000
#> GSM97076 1 0.0000 0.980 1.000 0.000
#> GSM97077 2 0.0000 0.988 0.000 1.000
#> GSM97095 1 0.0000 0.980 1.000 0.000
#> GSM97102 2 0.0000 0.988 0.000 1.000
#> GSM97109 2 0.0672 0.983 0.008 0.992
#> GSM97110 2 0.0000 0.988 0.000 1.000
#> GSM97074 1 0.0376 0.977 0.996 0.004
#> GSM97085 1 0.9608 0.382 0.616 0.384
#> GSM97059 1 0.3274 0.924 0.940 0.060
#> GSM97072 2 0.0000 0.988 0.000 1.000
#> GSM97078 1 0.0376 0.977 0.996 0.004
#> GSM97067 2 0.0000 0.988 0.000 1.000
#> GSM97087 2 0.0000 0.988 0.000 1.000
#> GSM97111 2 0.0000 0.988 0.000 1.000
#> GSM97064 2 0.0000 0.988 0.000 1.000
#> GSM97065 2 0.0000 0.988 0.000 1.000
#> GSM97081 2 0.0000 0.988 0.000 1.000
#> GSM97082 2 0.0000 0.988 0.000 1.000
#> GSM97088 1 0.3114 0.930 0.944 0.056
#> GSM97100 2 0.0000 0.988 0.000 1.000
#> GSM97104 2 0.0000 0.988 0.000 1.000
#> GSM97108 2 0.0376 0.986 0.004 0.996
#> GSM97050 2 0.0000 0.988 0.000 1.000
#> GSM97080 2 0.0000 0.988 0.000 1.000
#> GSM97089 2 0.0000 0.988 0.000 1.000
#> GSM97092 2 0.0000 0.988 0.000 1.000
#> GSM97093 2 0.0000 0.988 0.000 1.000
#> GSM97058 2 0.0000 0.988 0.000 1.000
#> GSM97051 2 0.0000 0.988 0.000 1.000
#> GSM97052 2 0.0000 0.988 0.000 1.000
#> GSM97061 2 0.0000 0.988 0.000 1.000
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM97138 1 0.0000 0.95996 1.000 0.000 0.000
#> GSM97145 1 0.0000 0.95996 1.000 0.000 0.000
#> GSM97147 2 0.2878 0.82308 0.096 0.904 0.000
#> GSM97125 1 0.0000 0.95996 1.000 0.000 0.000
#> GSM97127 1 0.0000 0.95996 1.000 0.000 0.000
#> GSM97130 1 0.0000 0.95996 1.000 0.000 0.000
#> GSM97133 1 0.0000 0.95996 1.000 0.000 0.000
#> GSM97134 1 0.0424 0.95553 0.992 0.008 0.000
#> GSM97120 1 0.0000 0.95996 1.000 0.000 0.000
#> GSM97126 1 0.0424 0.95527 0.992 0.008 0.000
#> GSM97112 1 0.0000 0.95996 1.000 0.000 0.000
#> GSM97115 2 0.6235 0.21408 0.436 0.564 0.000
#> GSM97116 1 0.0000 0.95996 1.000 0.000 0.000
#> GSM97117 2 0.6745 0.24102 0.012 0.560 0.428
#> GSM97119 1 0.0000 0.95996 1.000 0.000 0.000
#> GSM97122 1 0.0000 0.95996 1.000 0.000 0.000
#> GSM97135 1 0.0000 0.95996 1.000 0.000 0.000
#> GSM97136 3 0.2063 0.87693 0.044 0.008 0.948
#> GSM97139 1 0.0000 0.95996 1.000 0.000 0.000
#> GSM97146 1 0.0000 0.95996 1.000 0.000 0.000
#> GSM97123 2 0.6244 0.28439 0.000 0.560 0.440
#> GSM97129 2 0.8898 0.24309 0.128 0.500 0.372
#> GSM97143 1 0.0000 0.95996 1.000 0.000 0.000
#> GSM97113 2 0.0661 0.87824 0.004 0.988 0.008
#> GSM97056 1 0.0000 0.95996 1.000 0.000 0.000
#> GSM97124 1 0.0000 0.95996 1.000 0.000 0.000
#> GSM97132 1 0.0000 0.95996 1.000 0.000 0.000
#> GSM97144 1 0.0747 0.95087 0.984 0.016 0.000
#> GSM97149 1 0.0000 0.95996 1.000 0.000 0.000
#> GSM97068 2 0.0592 0.87553 0.012 0.988 0.000
#> GSM97071 3 0.0892 0.89847 0.000 0.020 0.980
#> GSM97086 2 0.0747 0.87892 0.000 0.984 0.016
#> GSM97103 3 0.5363 0.58847 0.000 0.276 0.724
#> GSM97057 2 0.0000 0.87674 0.000 1.000 0.000
#> GSM97060 3 0.0237 0.90397 0.000 0.004 0.996
#> GSM97075 3 0.6308 -0.00393 0.000 0.492 0.508
#> GSM97098 3 0.5882 0.42562 0.000 0.348 0.652
#> GSM97099 2 0.2878 0.84653 0.000 0.904 0.096
#> GSM97101 2 0.0424 0.87771 0.000 0.992 0.008
#> GSM97105 2 0.0237 0.87750 0.000 0.996 0.004
#> GSM97106 3 0.4887 0.67231 0.000 0.228 0.772
#> GSM97121 2 0.0237 0.87750 0.000 0.996 0.004
#> GSM97128 3 0.4912 0.73294 0.196 0.008 0.796
#> GSM97131 2 0.2796 0.85557 0.000 0.908 0.092
#> GSM97137 1 0.0000 0.95996 1.000 0.000 0.000
#> GSM97118 1 0.0237 0.95789 0.996 0.000 0.004
#> GSM97114 2 0.1989 0.86301 0.048 0.948 0.004
#> GSM97142 1 0.0000 0.95996 1.000 0.000 0.000
#> GSM97140 2 0.0000 0.87674 0.000 1.000 0.000
#> GSM97141 2 0.0592 0.87797 0.000 0.988 0.012
#> GSM97055 1 0.3267 0.85488 0.884 0.000 0.116
#> GSM97090 1 0.4504 0.75062 0.804 0.196 0.000
#> GSM97091 1 0.0424 0.95544 0.992 0.000 0.008
#> GSM97148 1 0.0000 0.95996 1.000 0.000 0.000
#> GSM97063 1 0.0237 0.95789 0.996 0.000 0.004
#> GSM97053 1 0.0000 0.95996 1.000 0.000 0.000
#> GSM97066 3 0.0000 0.90534 0.000 0.000 1.000
#> GSM97079 2 0.3412 0.83528 0.000 0.876 0.124
#> GSM97083 1 0.2486 0.91058 0.932 0.008 0.060
#> GSM97084 2 0.2066 0.86913 0.000 0.940 0.060
#> GSM97094 1 0.2550 0.91363 0.932 0.056 0.012
#> GSM97096 3 0.1860 0.87580 0.000 0.052 0.948
#> GSM97097 2 0.4452 0.76555 0.000 0.808 0.192
#> GSM97107 1 0.3263 0.89864 0.912 0.048 0.040
#> GSM97054 2 0.0892 0.87904 0.000 0.980 0.020
#> GSM97062 2 0.2448 0.86301 0.000 0.924 0.076
#> GSM97069 3 0.0000 0.90534 0.000 0.000 1.000
#> GSM97070 3 0.0000 0.90534 0.000 0.000 1.000
#> GSM97073 3 0.0000 0.90534 0.000 0.000 1.000
#> GSM97076 1 0.6598 0.22215 0.564 0.008 0.428
#> GSM97077 2 0.1289 0.87803 0.000 0.968 0.032
#> GSM97095 1 0.5859 0.47423 0.656 0.344 0.000
#> GSM97102 3 0.0000 0.90534 0.000 0.000 1.000
#> GSM97109 2 0.1525 0.87641 0.004 0.964 0.032
#> GSM97110 2 0.1860 0.87102 0.000 0.948 0.052
#> GSM97074 3 0.4291 0.75374 0.180 0.000 0.820
#> GSM97085 3 0.0000 0.90534 0.000 0.000 1.000
#> GSM97059 2 0.2261 0.84484 0.068 0.932 0.000
#> GSM97072 3 0.0000 0.90534 0.000 0.000 1.000
#> GSM97078 3 0.4228 0.78472 0.148 0.008 0.844
#> GSM97067 3 0.0000 0.90534 0.000 0.000 1.000
#> GSM97087 3 0.0000 0.90534 0.000 0.000 1.000
#> GSM97111 2 0.2356 0.86294 0.000 0.928 0.072
#> GSM97064 2 0.3619 0.82419 0.000 0.864 0.136
#> GSM97065 3 0.4654 0.72601 0.000 0.208 0.792
#> GSM97081 3 0.1753 0.88287 0.000 0.048 0.952
#> GSM97082 3 0.0000 0.90534 0.000 0.000 1.000
#> GSM97088 3 0.0661 0.90083 0.004 0.008 0.988
#> GSM97100 2 0.0000 0.87674 0.000 1.000 0.000
#> GSM97104 3 0.0000 0.90534 0.000 0.000 1.000
#> GSM97108 2 0.0000 0.87674 0.000 1.000 0.000
#> GSM97050 2 0.2356 0.86904 0.000 0.928 0.072
#> GSM97080 3 0.0000 0.90534 0.000 0.000 1.000
#> GSM97089 3 0.0000 0.90534 0.000 0.000 1.000
#> GSM97092 3 0.0592 0.90150 0.000 0.012 0.988
#> GSM97093 2 0.3752 0.81125 0.000 0.856 0.144
#> GSM97058 2 0.0892 0.87924 0.000 0.980 0.020
#> GSM97051 2 0.2878 0.85208 0.000 0.904 0.096
#> GSM97052 3 0.1289 0.89171 0.000 0.032 0.968
#> GSM97061 2 0.6299 0.17238 0.000 0.524 0.476
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM97138 1 0.2589 0.8634 0.912 0.044 0.000 0.044
#> GSM97145 1 0.2500 0.8647 0.916 0.044 0.000 0.040
#> GSM97147 2 0.5792 0.4121 0.056 0.648 0.000 0.296
#> GSM97125 1 0.1042 0.8711 0.972 0.008 0.000 0.020
#> GSM97127 1 0.2751 0.8622 0.904 0.040 0.000 0.056
#> GSM97130 4 0.5724 0.0759 0.424 0.028 0.000 0.548
#> GSM97133 1 0.3071 0.8567 0.888 0.044 0.000 0.068
#> GSM97134 1 0.5163 -0.0404 0.516 0.004 0.000 0.480
#> GSM97120 1 0.2926 0.8596 0.896 0.048 0.000 0.056
#> GSM97126 1 0.2936 0.8453 0.900 0.056 0.004 0.040
#> GSM97112 1 0.0469 0.8685 0.988 0.000 0.000 0.012
#> GSM97115 4 0.4344 0.5957 0.076 0.108 0.000 0.816
#> GSM97116 1 0.2996 0.8582 0.892 0.044 0.000 0.064
#> GSM97117 2 0.3627 0.6263 0.008 0.840 0.144 0.008
#> GSM97119 1 0.0469 0.8685 0.988 0.000 0.000 0.012
#> GSM97122 1 0.0469 0.8685 0.988 0.000 0.000 0.012
#> GSM97135 1 0.0469 0.8685 0.988 0.000 0.000 0.012
#> GSM97136 3 0.6358 0.5190 0.128 0.204 0.664 0.004
#> GSM97139 1 0.2996 0.8582 0.892 0.044 0.000 0.064
#> GSM97146 1 0.3071 0.8567 0.888 0.044 0.000 0.068
#> GSM97123 3 0.7151 -0.1635 0.000 0.420 0.448 0.132
#> GSM97129 2 0.8054 0.4148 0.160 0.580 0.184 0.076
#> GSM97143 1 0.0336 0.8690 0.992 0.000 0.000 0.008
#> GSM97113 2 0.1854 0.6386 0.012 0.940 0.000 0.048
#> GSM97056 1 0.5308 0.6301 0.684 0.036 0.000 0.280
#> GSM97124 1 0.0592 0.8677 0.984 0.000 0.000 0.016
#> GSM97132 1 0.2760 0.7917 0.872 0.000 0.000 0.128
#> GSM97144 4 0.4776 0.3415 0.376 0.000 0.000 0.624
#> GSM97149 1 0.3156 0.8551 0.884 0.048 0.000 0.068
#> GSM97068 4 0.4585 0.3025 0.000 0.332 0.000 0.668
#> GSM97071 4 0.5600 -0.0575 0.020 0.000 0.468 0.512
#> GSM97086 4 0.3908 0.4859 0.000 0.212 0.004 0.784
#> GSM97103 2 0.7034 0.1533 0.000 0.468 0.412 0.120
#> GSM97057 2 0.4356 0.5240 0.000 0.708 0.000 0.292
#> GSM97060 3 0.0707 0.7989 0.000 0.000 0.980 0.020
#> GSM97075 2 0.5384 0.4873 0.000 0.648 0.324 0.028
#> GSM97098 2 0.6442 0.1234 0.000 0.492 0.440 0.068
#> GSM97099 2 0.2965 0.6504 0.000 0.892 0.072 0.036
#> GSM97101 2 0.1022 0.6502 0.000 0.968 0.000 0.032
#> GSM97105 2 0.3266 0.6121 0.000 0.832 0.000 0.168
#> GSM97106 3 0.5839 0.5575 0.000 0.104 0.696 0.200
#> GSM97121 2 0.2345 0.6422 0.000 0.900 0.000 0.100
#> GSM97128 3 0.7214 0.0984 0.144 0.000 0.476 0.380
#> GSM97131 2 0.6836 0.4779 0.000 0.580 0.140 0.280
#> GSM97137 1 0.5512 0.5923 0.660 0.040 0.000 0.300
#> GSM97118 1 0.3196 0.7731 0.856 0.000 0.008 0.136
#> GSM97114 2 0.2494 0.6212 0.036 0.916 0.000 0.048
#> GSM97142 1 0.0469 0.8685 0.988 0.000 0.000 0.012
#> GSM97140 2 0.3801 0.5804 0.000 0.780 0.000 0.220
#> GSM97141 2 0.0707 0.6493 0.000 0.980 0.000 0.020
#> GSM97055 1 0.3668 0.7658 0.852 0.004 0.116 0.028
#> GSM97090 4 0.3948 0.6099 0.136 0.036 0.000 0.828
#> GSM97091 1 0.1042 0.8630 0.972 0.000 0.008 0.020
#> GSM97148 1 0.3071 0.8567 0.888 0.044 0.000 0.068
#> GSM97063 1 0.0657 0.8674 0.984 0.000 0.004 0.012
#> GSM97053 1 0.0921 0.8707 0.972 0.000 0.000 0.028
#> GSM97066 3 0.0336 0.8009 0.000 0.000 0.992 0.008
#> GSM97079 4 0.4586 0.5209 0.000 0.136 0.068 0.796
#> GSM97083 4 0.6553 0.4141 0.316 0.000 0.100 0.584
#> GSM97084 4 0.2546 0.5817 0.000 0.092 0.008 0.900
#> GSM97094 4 0.4960 0.5821 0.212 0.020 0.016 0.752
#> GSM97096 3 0.5421 0.5720 0.000 0.200 0.724 0.076
#> GSM97097 4 0.7558 -0.1332 0.000 0.380 0.192 0.428
#> GSM97107 4 0.4527 0.5939 0.192 0.020 0.008 0.780
#> GSM97054 4 0.3498 0.5524 0.000 0.160 0.008 0.832
#> GSM97062 4 0.2593 0.5753 0.000 0.104 0.004 0.892
#> GSM97069 3 0.0336 0.8009 0.000 0.000 0.992 0.008
#> GSM97070 3 0.0336 0.8009 0.000 0.000 0.992 0.008
#> GSM97073 3 0.0779 0.7997 0.000 0.004 0.980 0.016
#> GSM97076 1 0.8105 0.3498 0.544 0.084 0.272 0.100
#> GSM97077 2 0.6817 0.2796 0.000 0.492 0.100 0.408
#> GSM97095 4 0.5029 0.6009 0.140 0.072 0.008 0.780
#> GSM97102 3 0.1388 0.7892 0.000 0.028 0.960 0.012
#> GSM97109 2 0.3177 0.6312 0.016 0.892 0.024 0.068
#> GSM97110 2 0.3855 0.6277 0.012 0.860 0.060 0.068
#> GSM97074 3 0.6426 0.4406 0.256 0.000 0.628 0.116
#> GSM97085 3 0.2032 0.7754 0.036 0.000 0.936 0.028
#> GSM97059 4 0.5543 0.0574 0.020 0.424 0.000 0.556
#> GSM97072 3 0.0592 0.8009 0.000 0.000 0.984 0.016
#> GSM97078 3 0.6745 0.0824 0.092 0.000 0.480 0.428
#> GSM97067 3 0.0336 0.8009 0.000 0.000 0.992 0.008
#> GSM97087 3 0.0524 0.8004 0.000 0.004 0.988 0.008
#> GSM97111 2 0.2861 0.6503 0.000 0.888 0.096 0.016
#> GSM97064 2 0.7754 0.3291 0.000 0.420 0.336 0.244
#> GSM97065 2 0.5775 0.0395 0.004 0.488 0.488 0.020
#> GSM97081 3 0.3873 0.5976 0.000 0.228 0.772 0.000
#> GSM97082 3 0.0336 0.8012 0.000 0.000 0.992 0.008
#> GSM97088 3 0.5657 0.5205 0.068 0.000 0.688 0.244
#> GSM97100 2 0.4564 0.4679 0.000 0.672 0.000 0.328
#> GSM97104 3 0.0000 0.8008 0.000 0.000 1.000 0.000
#> GSM97108 2 0.2408 0.6389 0.000 0.896 0.000 0.104
#> GSM97050 2 0.7275 0.3111 0.000 0.472 0.152 0.376
#> GSM97080 3 0.0188 0.8009 0.000 0.000 0.996 0.004
#> GSM97089 3 0.0657 0.7997 0.000 0.004 0.984 0.012
#> GSM97092 3 0.2002 0.7795 0.000 0.020 0.936 0.044
#> GSM97093 2 0.7824 0.2580 0.000 0.400 0.264 0.336
#> GSM97058 2 0.6378 0.5120 0.000 0.628 0.108 0.264
#> GSM97051 4 0.7268 0.0295 0.000 0.312 0.172 0.516
#> GSM97052 3 0.3652 0.7222 0.000 0.052 0.856 0.092
#> GSM97061 3 0.7049 0.2603 0.000 0.236 0.572 0.192
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM97138 1 0.4384 0.47558 0.660 0.016 0.000 0.000 0.324
#> GSM97145 1 0.3942 0.53647 0.748 0.020 0.000 0.000 0.232
#> GSM97147 2 0.7419 0.14047 0.084 0.460 0.000 0.328 0.128
#> GSM97125 1 0.3039 0.56352 0.808 0.000 0.000 0.000 0.192
#> GSM97127 1 0.4327 0.44811 0.632 0.008 0.000 0.000 0.360
#> GSM97130 5 0.6166 0.52499 0.180 0.000 0.000 0.272 0.548
#> GSM97133 1 0.4867 0.36701 0.544 0.024 0.000 0.000 0.432
#> GSM97134 1 0.6660 -0.26312 0.468 0.000 0.004 0.216 0.312
#> GSM97120 1 0.4760 0.39292 0.564 0.020 0.000 0.000 0.416
#> GSM97126 1 0.4776 0.49987 0.744 0.084 0.004 0.004 0.164
#> GSM97112 1 0.0290 0.59461 0.992 0.000 0.000 0.000 0.008
#> GSM97115 4 0.5755 -0.11630 0.024 0.040 0.000 0.504 0.432
#> GSM97116 1 0.4682 0.39010 0.564 0.016 0.000 0.000 0.420
#> GSM97117 2 0.2913 0.62529 0.004 0.880 0.080 0.004 0.032
#> GSM97119 1 0.0609 0.59786 0.980 0.000 0.000 0.000 0.020
#> GSM97122 1 0.0794 0.59915 0.972 0.000 0.000 0.000 0.028
#> GSM97135 1 0.1043 0.59965 0.960 0.000 0.000 0.000 0.040
#> GSM97136 3 0.7607 0.30426 0.244 0.212 0.468 0.000 0.076
#> GSM97139 1 0.4689 0.38567 0.560 0.016 0.000 0.000 0.424
#> GSM97146 1 0.4793 0.36467 0.544 0.020 0.000 0.000 0.436
#> GSM97123 3 0.8130 -0.01570 0.000 0.300 0.340 0.260 0.100
#> GSM97129 2 0.8851 0.29140 0.172 0.464 0.116 0.108 0.140
#> GSM97143 1 0.1341 0.58481 0.944 0.000 0.000 0.000 0.056
#> GSM97113 2 0.4333 0.57880 0.000 0.752 0.000 0.060 0.188
#> GSM97056 5 0.6490 0.16652 0.376 0.008 0.000 0.148 0.468
#> GSM97124 1 0.2389 0.58272 0.880 0.000 0.000 0.004 0.116
#> GSM97132 1 0.4527 0.33055 0.700 0.000 0.000 0.040 0.260
#> GSM97144 5 0.6747 0.33121 0.260 0.000 0.000 0.364 0.376
#> GSM97149 1 0.4872 0.36009 0.540 0.024 0.000 0.000 0.436
#> GSM97068 4 0.6193 0.24710 0.000 0.184 0.000 0.544 0.272
#> GSM97071 3 0.7547 0.14284 0.036 0.004 0.416 0.280 0.264
#> GSM97086 4 0.2719 0.48076 0.000 0.068 0.000 0.884 0.048
#> GSM97103 2 0.7540 0.25782 0.000 0.484 0.268 0.156 0.092
#> GSM97057 2 0.6642 0.06338 0.000 0.420 0.000 0.352 0.228
#> GSM97060 3 0.3256 0.64837 0.000 0.012 0.864 0.064 0.060
#> GSM97075 2 0.5896 0.38814 0.000 0.604 0.304 0.052 0.040
#> GSM97098 2 0.6703 0.25069 0.000 0.544 0.308 0.068 0.080
#> GSM97099 2 0.2812 0.63044 0.004 0.896 0.028 0.020 0.052
#> GSM97101 2 0.2189 0.61089 0.000 0.904 0.000 0.084 0.012
#> GSM97105 2 0.4109 0.42479 0.000 0.700 0.000 0.288 0.012
#> GSM97106 3 0.7243 0.26857 0.000 0.072 0.480 0.324 0.124
#> GSM97121 2 0.3318 0.56067 0.000 0.808 0.000 0.180 0.012
#> GSM97128 3 0.7892 0.13225 0.300 0.000 0.340 0.068 0.292
#> GSM97131 4 0.6126 0.13081 0.000 0.380 0.056 0.528 0.036
#> GSM97137 5 0.6170 0.17910 0.364 0.008 0.000 0.112 0.516
#> GSM97118 1 0.4111 0.37454 0.756 0.000 0.016 0.012 0.216
#> GSM97114 2 0.3421 0.60179 0.008 0.824 0.000 0.016 0.152
#> GSM97142 1 0.0404 0.59307 0.988 0.000 0.000 0.000 0.012
#> GSM97140 2 0.5335 0.11725 0.000 0.536 0.004 0.416 0.044
#> GSM97141 2 0.1300 0.63001 0.000 0.956 0.000 0.028 0.016
#> GSM97055 1 0.4171 0.44264 0.784 0.000 0.104 0.000 0.112
#> GSM97090 4 0.5682 -0.19133 0.060 0.008 0.000 0.512 0.420
#> GSM97091 1 0.2110 0.54797 0.912 0.000 0.016 0.000 0.072
#> GSM97148 1 0.4793 0.36467 0.544 0.020 0.000 0.000 0.436
#> GSM97063 1 0.1121 0.57758 0.956 0.000 0.000 0.000 0.044
#> GSM97053 1 0.2773 0.57254 0.836 0.000 0.000 0.000 0.164
#> GSM97066 3 0.1270 0.66290 0.000 0.000 0.948 0.000 0.052
#> GSM97079 4 0.4295 0.45303 0.000 0.052 0.048 0.808 0.092
#> GSM97083 1 0.7832 -0.26032 0.368 0.000 0.084 0.192 0.356
#> GSM97084 4 0.3160 0.33400 0.000 0.004 0.000 0.808 0.188
#> GSM97094 4 0.6802 -0.13190 0.152 0.020 0.004 0.508 0.316
#> GSM97096 3 0.6860 0.38584 0.000 0.256 0.564 0.080 0.100
#> GSM97097 4 0.7343 0.09165 0.000 0.332 0.104 0.468 0.096
#> GSM97107 4 0.6211 -0.22222 0.128 0.000 0.004 0.500 0.368
#> GSM97054 4 0.3497 0.44575 0.000 0.048 0.004 0.836 0.112
#> GSM97062 4 0.2488 0.40771 0.000 0.004 0.000 0.872 0.124
#> GSM97069 3 0.0794 0.66555 0.000 0.000 0.972 0.000 0.028
#> GSM97070 3 0.1121 0.66521 0.000 0.000 0.956 0.000 0.044
#> GSM97073 3 0.1862 0.66306 0.000 0.016 0.932 0.004 0.048
#> GSM97076 3 0.8901 0.02271 0.276 0.132 0.316 0.028 0.248
#> GSM97077 4 0.7016 0.27476 0.008 0.296 0.084 0.540 0.072
#> GSM97095 5 0.6681 0.06080 0.060 0.068 0.000 0.428 0.444
#> GSM97102 3 0.2864 0.65172 0.000 0.064 0.884 0.008 0.044
#> GSM97109 2 0.4181 0.58926 0.000 0.784 0.008 0.052 0.156
#> GSM97110 2 0.4351 0.58686 0.000 0.776 0.020 0.040 0.164
#> GSM97074 3 0.6814 0.35072 0.296 0.000 0.508 0.024 0.172
#> GSM97085 3 0.4766 0.57347 0.136 0.000 0.732 0.000 0.132
#> GSM97059 4 0.6615 0.26266 0.008 0.276 0.000 0.508 0.208
#> GSM97072 3 0.1329 0.66736 0.000 0.008 0.956 0.004 0.032
#> GSM97078 3 0.8109 0.23467 0.216 0.000 0.396 0.120 0.268
#> GSM97067 3 0.1121 0.66403 0.000 0.000 0.956 0.000 0.044
#> GSM97087 3 0.3298 0.64384 0.000 0.012 0.856 0.036 0.096
#> GSM97111 2 0.1978 0.63017 0.000 0.932 0.024 0.032 0.012
#> GSM97064 4 0.7760 0.24025 0.000 0.204 0.240 0.460 0.096
#> GSM97065 3 0.6188 -0.00711 0.000 0.416 0.448 0.000 0.136
#> GSM97081 3 0.6183 0.36524 0.000 0.300 0.588 0.048 0.064
#> GSM97082 3 0.1710 0.66536 0.000 0.004 0.940 0.016 0.040
#> GSM97088 3 0.7476 0.39916 0.180 0.004 0.496 0.064 0.256
#> GSM97100 4 0.4996 0.10805 0.000 0.420 0.000 0.548 0.032
#> GSM97104 3 0.1862 0.66328 0.000 0.016 0.932 0.004 0.048
#> GSM97108 2 0.3318 0.54231 0.000 0.800 0.000 0.192 0.008
#> GSM97050 4 0.6789 0.34469 0.000 0.204 0.092 0.596 0.108
#> GSM97080 3 0.1124 0.66803 0.000 0.004 0.960 0.000 0.036
#> GSM97089 3 0.3699 0.64089 0.000 0.028 0.836 0.032 0.104
#> GSM97092 3 0.5273 0.58010 0.000 0.044 0.736 0.108 0.112
#> GSM97093 4 0.8153 0.20619 0.000 0.260 0.168 0.408 0.164
#> GSM97058 4 0.6506 0.12832 0.000 0.388 0.080 0.492 0.040
#> GSM97051 4 0.6181 0.38413 0.000 0.168 0.096 0.660 0.076
#> GSM97052 3 0.5835 0.52744 0.000 0.048 0.684 0.160 0.108
#> GSM97061 3 0.7751 0.02858 0.000 0.148 0.396 0.356 0.100
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM97138 5 0.4111 -0.33296 0.456 0.004 0.000 0.004 0.536 0.000
#> GSM97145 5 0.4350 0.19229 0.292 0.048 0.000 0.000 0.660 0.000
#> GSM97147 2 0.8273 0.22586 0.180 0.392 0.188 0.164 0.076 0.000
#> GSM97125 5 0.2854 0.44832 0.208 0.000 0.000 0.000 0.792 0.000
#> GSM97127 1 0.4331 0.44111 0.516 0.020 0.000 0.000 0.464 0.000
#> GSM97130 4 0.6008 0.10615 0.424 0.008 0.020 0.444 0.104 0.000
#> GSM97133 1 0.3922 0.66848 0.664 0.016 0.000 0.000 0.320 0.000
#> GSM97134 5 0.6484 0.09278 0.184 0.016 0.012 0.360 0.428 0.000
#> GSM97120 1 0.3955 0.61372 0.608 0.008 0.000 0.000 0.384 0.000
#> GSM97126 5 0.5636 0.39268 0.240 0.092 0.016 0.024 0.628 0.000
#> GSM97112 5 0.0260 0.64435 0.008 0.000 0.000 0.000 0.992 0.000
#> GSM97115 4 0.6300 0.38959 0.364 0.028 0.120 0.476 0.012 0.000
#> GSM97116 1 0.3684 0.63936 0.628 0.000 0.000 0.000 0.372 0.000
#> GSM97117 2 0.2831 0.59529 0.020 0.884 0.028 0.000 0.016 0.052
#> GSM97119 5 0.0951 0.64528 0.020 0.000 0.004 0.008 0.968 0.000
#> GSM97122 5 0.1462 0.62976 0.056 0.000 0.000 0.008 0.936 0.000
#> GSM97135 5 0.1588 0.61947 0.072 0.000 0.000 0.004 0.924 0.000
#> GSM97136 5 0.8505 -0.22926 0.056 0.188 0.136 0.012 0.308 0.300
#> GSM97139 1 0.3659 0.64555 0.636 0.000 0.000 0.000 0.364 0.000
#> GSM97146 1 0.3547 0.66996 0.668 0.000 0.000 0.000 0.332 0.000
#> GSM97123 3 0.6325 0.47033 0.024 0.208 0.560 0.020 0.000 0.188
#> GSM97129 2 0.9241 0.16150 0.140 0.336 0.188 0.052 0.176 0.108
#> GSM97143 5 0.1594 0.64241 0.052 0.000 0.000 0.016 0.932 0.000
#> GSM97113 2 0.4798 0.46207 0.324 0.620 0.044 0.004 0.000 0.008
#> GSM97056 1 0.5475 0.55158 0.620 0.000 0.016 0.164 0.200 0.000
#> GSM97124 5 0.2740 0.59042 0.120 0.000 0.000 0.028 0.852 0.000
#> GSM97132 5 0.5741 0.40430 0.200 0.004 0.012 0.196 0.588 0.000
#> GSM97144 4 0.5528 0.39750 0.172 0.004 0.004 0.596 0.224 0.000
#> GSM97149 1 0.3619 0.67289 0.680 0.000 0.004 0.000 0.316 0.000
#> GSM97068 4 0.7113 0.37306 0.296 0.144 0.136 0.424 0.000 0.000
#> GSM97071 6 0.6375 0.03269 0.072 0.000 0.044 0.428 0.024 0.432
#> GSM97086 4 0.5568 0.21797 0.032 0.064 0.336 0.564 0.000 0.004
#> GSM97103 2 0.7750 0.20380 0.052 0.428 0.120 0.128 0.000 0.272
#> GSM97057 1 0.7115 -0.12616 0.432 0.268 0.188 0.112 0.000 0.000
#> GSM97060 6 0.4347 0.43171 0.012 0.000 0.304 0.024 0.000 0.660
#> GSM97075 2 0.6875 0.21215 0.052 0.496 0.220 0.016 0.000 0.216
#> GSM97098 2 0.7227 0.22288 0.044 0.476 0.156 0.056 0.000 0.268
#> GSM97099 2 0.3804 0.59334 0.052 0.836 0.048 0.028 0.004 0.032
#> GSM97101 2 0.2263 0.58909 0.036 0.900 0.060 0.004 0.000 0.000
#> GSM97105 2 0.5485 0.31771 0.032 0.584 0.308 0.076 0.000 0.000
#> GSM97106 3 0.5774 0.30291 0.032 0.016 0.592 0.072 0.000 0.288
#> GSM97121 2 0.3839 0.55041 0.032 0.796 0.132 0.040 0.000 0.000
#> GSM97128 5 0.8295 0.00524 0.128 0.004 0.052 0.212 0.340 0.264
#> GSM97131 3 0.6961 0.22795 0.040 0.292 0.484 0.144 0.000 0.040
#> GSM97137 1 0.5563 0.51660 0.616 0.000 0.020 0.184 0.180 0.000
#> GSM97118 5 0.5749 0.48845 0.104 0.004 0.020 0.172 0.664 0.036
#> GSM97114 2 0.3500 0.57378 0.168 0.800 0.016 0.004 0.008 0.004
#> GSM97142 5 0.0653 0.64464 0.012 0.000 0.004 0.004 0.980 0.000
#> GSM97140 2 0.6641 0.22988 0.096 0.464 0.344 0.092 0.004 0.000
#> GSM97141 2 0.1856 0.59599 0.032 0.920 0.048 0.000 0.000 0.000
#> GSM97055 5 0.5133 0.55018 0.092 0.008 0.028 0.056 0.748 0.068
#> GSM97090 4 0.5510 0.49851 0.248 0.008 0.100 0.624 0.020 0.000
#> GSM97091 5 0.2113 0.63012 0.032 0.000 0.012 0.028 0.920 0.008
#> GSM97148 1 0.3652 0.67245 0.672 0.000 0.000 0.004 0.324 0.000
#> GSM97063 5 0.1198 0.64162 0.020 0.000 0.004 0.012 0.960 0.004
#> GSM97053 5 0.3364 0.47922 0.196 0.000 0.000 0.024 0.780 0.000
#> GSM97066 6 0.1949 0.60355 0.036 0.000 0.020 0.020 0.000 0.924
#> GSM97079 4 0.6353 0.26712 0.064 0.060 0.256 0.580 0.000 0.040
#> GSM97083 4 0.7082 0.07233 0.124 0.004 0.028 0.448 0.340 0.056
#> GSM97084 4 0.3820 0.47005 0.040 0.016 0.148 0.792 0.000 0.004
#> GSM97094 4 0.4388 0.51591 0.036 0.016 0.052 0.800 0.080 0.016
#> GSM97096 6 0.7386 0.16145 0.044 0.180 0.296 0.052 0.000 0.428
#> GSM97097 4 0.7292 0.13485 0.036 0.216 0.208 0.476 0.000 0.064
#> GSM97107 4 0.3869 0.52905 0.080 0.020 0.016 0.828 0.044 0.012
#> GSM97054 4 0.5603 0.25407 0.072 0.036 0.340 0.552 0.000 0.000
#> GSM97062 4 0.4570 0.37484 0.028 0.032 0.256 0.684 0.000 0.000
#> GSM97069 6 0.1649 0.60841 0.016 0.000 0.040 0.008 0.000 0.936
#> GSM97070 6 0.1405 0.60741 0.024 0.000 0.024 0.004 0.000 0.948
#> GSM97073 6 0.1971 0.59768 0.024 0.024 0.016 0.008 0.000 0.928
#> GSM97076 6 0.8368 0.28379 0.144 0.120 0.040 0.096 0.124 0.476
#> GSM97077 3 0.6968 0.35853 0.080 0.204 0.544 0.136 0.000 0.036
#> GSM97095 4 0.6755 0.47392 0.236 0.052 0.096 0.564 0.048 0.004
#> GSM97102 6 0.5402 0.51710 0.032 0.080 0.144 0.032 0.004 0.708
#> GSM97109 2 0.4489 0.57118 0.128 0.772 0.040 0.044 0.004 0.012
#> GSM97110 2 0.5733 0.53591 0.120 0.692 0.068 0.048 0.000 0.072
#> GSM97074 6 0.6931 0.31308 0.096 0.000 0.028 0.104 0.244 0.528
#> GSM97085 6 0.5780 0.50587 0.072 0.004 0.048 0.048 0.140 0.688
#> GSM97059 1 0.7771 -0.26297 0.368 0.184 0.188 0.248 0.012 0.000
#> GSM97072 6 0.1837 0.60329 0.020 0.012 0.032 0.004 0.000 0.932
#> GSM97078 4 0.8329 -0.01822 0.120 0.000 0.072 0.312 0.208 0.288
#> GSM97067 6 0.0951 0.60711 0.020 0.000 0.008 0.004 0.000 0.968
#> GSM97087 6 0.4788 0.27440 0.028 0.008 0.420 0.004 0.000 0.540
#> GSM97111 2 0.3882 0.58588 0.040 0.812 0.100 0.008 0.000 0.040
#> GSM97064 3 0.5223 0.60309 0.032 0.064 0.728 0.060 0.000 0.116
#> GSM97065 6 0.6702 0.12121 0.132 0.304 0.068 0.008 0.000 0.488
#> GSM97081 6 0.6672 0.24249 0.040 0.188 0.292 0.008 0.000 0.472
#> GSM97082 6 0.3932 0.50560 0.028 0.004 0.248 0.000 0.000 0.720
#> GSM97088 6 0.8641 0.23389 0.104 0.004 0.160 0.204 0.176 0.352
#> GSM97100 2 0.6716 -0.03804 0.052 0.384 0.376 0.188 0.000 0.000
#> GSM97104 6 0.3194 0.56181 0.012 0.004 0.172 0.004 0.000 0.808
#> GSM97108 2 0.4518 0.49161 0.036 0.736 0.172 0.056 0.000 0.000
#> GSM97050 3 0.5664 0.50508 0.044 0.108 0.676 0.148 0.000 0.024
#> GSM97080 6 0.2263 0.59497 0.016 0.000 0.100 0.000 0.000 0.884
#> GSM97089 6 0.5047 0.24891 0.032 0.012 0.424 0.008 0.000 0.524
#> GSM97092 6 0.4805 0.14248 0.016 0.012 0.468 0.008 0.000 0.496
#> GSM97093 3 0.6184 0.55232 0.064 0.100 0.660 0.084 0.000 0.092
#> GSM97058 3 0.6957 0.40288 0.060 0.208 0.556 0.104 0.000 0.072
#> GSM97051 3 0.5310 0.48954 0.024 0.076 0.684 0.192 0.000 0.024
#> GSM97052 3 0.4591 -0.01563 0.008 0.012 0.552 0.008 0.000 0.420
#> GSM97061 3 0.4377 0.50949 0.000 0.040 0.728 0.028 0.000 0.204
Heatmaps for the consensus matrix. It visualizes the probability of two samples to be in a same group.
consensus_heatmap(res, k = 2)
consensus_heatmap(res, k = 3)
consensus_heatmap(res, k = 4)
consensus_heatmap(res, k = 5)
consensus_heatmap(res, k = 6)
Heatmaps for the membership of samples in all partitions to see how consistent they are:
membership_heatmap(res, k = 2)
membership_heatmap(res, k = 3)
membership_heatmap(res, k = 4)
membership_heatmap(res, k = 5)
membership_heatmap(res, k = 6)
As soon as we have had the classes for columns, we can look for signatures which are significantly different between classes which can be candidate marks for certain classes. Following are the heatmaps for signatures.
Signature heatmaps where rows are scaled:
get_signatures(res, k = 2)
get_signatures(res, k = 3)
get_signatures(res, k = 4)
get_signatures(res, k = 5)
get_signatures(res, k = 6)
Signature heatmaps where rows are not scaled:
get_signatures(res, k = 2, scale_rows = FALSE)
get_signatures(res, k = 3, scale_rows = FALSE)
get_signatures(res, k = 4, scale_rows = FALSE)
get_signatures(res, k = 5, scale_rows = FALSE)
get_signatures(res, k = 6, scale_rows = FALSE)
Compare the overlap of signatures from different k:
compare_signatures(res)
get_signature() returns a data frame invisibly. TO get the list of signatures, the function
call should be assigned to a variable explicitly. In following code, if plot argument is set
to FALSE, no heatmap is plotted while only the differential analysis is performed.
# code only for demonstration
tb = get_signature(res, k = ..., plot = FALSE)
An example of the output of tb is:
#> which_row fdr mean_1 mean_2 scaled_mean_1 scaled_mean_2 km
#> 1 38 0.042760348 8.373488 9.131774 -0.5533452 0.5164555 1
#> 2 40 0.018707592 7.106213 8.469186 -0.6173731 0.5762149 1
#> 3 55 0.019134737 10.221463 11.207825 -0.6159697 0.5749050 1
#> 4 59 0.006059896 5.921854 7.869574 -0.6899429 0.6439467 1
#> 5 60 0.018055526 8.928898 10.211722 -0.6204761 0.5791110 1
#> 6 98 0.009384629 15.714769 14.887706 0.6635654 -0.6193277 2
...
The columns in tb are:
which_row: row indices corresponding to the input matrix.fdr: FDR for the differential test. mean_x: The mean value in group x.scaled_mean_x: The mean value in group x after rows are scaled.km: Row groups if k-means clustering is applied to rows.UMAP plot which shows how samples are separated.
dimension_reduction(res, k = 2, method = "UMAP")
dimension_reduction(res, k = 3, method = "UMAP")
dimension_reduction(res, k = 4, method = "UMAP")
dimension_reduction(res, k = 5, method = "UMAP")
dimension_reduction(res, k = 6, method = "UMAP")
Following heatmap shows how subgroups are split when increasing k:
collect_classes(res)
Test correlation between subgroups and known annotations. If the known annotation is numeric, one-way ANOVA test is applied, and if the known annotation is discrete, chi-squared contingency table test is applied.
test_to_known_factors(res)
#> n disease.state(p) specimen(p) cell.type(p) other(p) k
#> CV:skmeans 99 9.78e-04 0.4095 3.76e-12 0.1858 2
#> CV:skmeans 91 2.44e-05 0.4304 4.93e-14 0.3416 3
#> CV:skmeans 72 1.31e-05 0.0481 5.42e-14 0.0260 4
#> CV:skmeans 40 2.93e-03 0.2278 5.02e-06 0.2191 5
#> CV:skmeans 45 7.18e-03 0.3311 6.31e-08 0.0441 6
If matrix rows can be associated to genes, consider to use functional_enrichment(res,
...) to perform function enrichment for the signature genes. See this vignette for more detailed explanations.
The object with results only for a single top-value method and a single partition method can be extracted as:
res = res_list["CV", "pam"]
# you can also extract it by
# res = res_list["CV:pam"]
A summary of res and all the functions that can be applied to it:
res
#> A 'ConsensusPartition' object with k = 2, 3, 4, 5, 6.
#> On a matrix with 21512 rows and 100 columns.
#> Top rows (1000, 2000, 3000, 4000, 5000) are extracted by 'CV' method.
#> Subgroups are detected by 'pam' method.
#> Performed in total 1250 partitions by row resampling.
#> Best k for subgroups seems to be 3.
#>
#> Following methods can be applied to this 'ConsensusPartition' object:
#> [1] "cola_report" "collect_classes" "collect_plots"
#> [4] "collect_stats" "colnames" "compare_signatures"
#> [7] "consensus_heatmap" "dimension_reduction" "functional_enrichment"
#> [10] "get_anno_col" "get_anno" "get_classes"
#> [13] "get_consensus" "get_matrix" "get_membership"
#> [16] "get_param" "get_signatures" "get_stats"
#> [19] "is_best_k" "is_stable_k" "membership_heatmap"
#> [22] "ncol" "nrow" "plot_ecdf"
#> [25] "rownames" "select_partition_number" "show"
#> [28] "suggest_best_k" "test_to_known_factors"
collect_plots() function collects all the plots made from res for all k (number of partitions)
into one single page to provide an easy and fast comparison between different k.
collect_plots(res)
The plots are:
k and the heatmap of
predicted classes for each k.k.k.k.All the plots in panels can be made by individual functions and they are plotted later in this section.
select_partition_number() produces several plots showing different
statistics for choosing “optimized” k. There are following statistics:
k;k, the area increased is defined as \(A_k - A_{k-1}\).The detailed explanations of these statistics can be found in the cola vignette.
Generally speaking, lower PAC score, higher mean silhouette score or higher
concordance corresponds to better partition. Rand index and Jaccard index
measure how similar the current partition is compared to partition with k-1.
If they are too similar, we won't accept k is better than k-1.
select_partition_number(res)
The numeric values for all these statistics can be obtained by get_stats().
get_stats(res)
#> k 1-PAC mean_silhouette concordance area_increased Rand Jaccard
#> 2 2 0.344 0.538 0.787 0.4632 0.576 0.576
#> 3 3 0.494 0.720 0.855 0.4006 0.634 0.431
#> 4 4 0.579 0.747 0.830 0.0954 0.901 0.737
#> 5 5 0.565 0.476 0.739 0.0776 0.943 0.820
#> 6 6 0.594 0.404 0.661 0.0511 0.884 0.613
suggest_best_k() suggests the best \(k\) based on these statistics. The rules are as follows:
suggest_best_k(res)
#> [1] 3
Following shows the table of the partitions (You need to click the show/hide
code output link to see it). The membership matrix (columns with name p*)
is inferred by
clue::cl_consensus()
function with the SE method. Basically the value in the membership matrix
represents the probability to belong to a certain group. The finall class
label for an item is determined with the group with highest probability it
belongs to.
In get_classes() function, the entropy is calculated from the membership
matrix and the silhouette score is calculated from the consensus matrix.
cbind(get_classes(res, k = 2), get_membership(res, k = 2))
#> class entropy silhouette p1 p2
#> GSM97138 1 0.9996 0.4693 0.512 0.488
#> GSM97145 1 0.9909 0.5379 0.556 0.444
#> GSM97147 2 0.9833 0.4931 0.424 0.576
#> GSM97125 1 0.9833 0.5574 0.576 0.424
#> GSM97127 1 0.5408 0.6363 0.876 0.124
#> GSM97130 1 0.0000 0.6241 1.000 0.000
#> GSM97133 1 0.0000 0.6241 1.000 0.000
#> GSM97134 1 0.3114 0.5985 0.944 0.056
#> GSM97120 2 0.9998 -0.4617 0.492 0.508
#> GSM97126 2 0.9954 -0.0792 0.460 0.540
#> GSM97112 1 0.9833 0.5574 0.576 0.424
#> GSM97115 2 0.9983 0.4332 0.476 0.524
#> GSM97116 1 0.7950 0.6265 0.760 0.240
#> GSM97117 2 0.0000 0.6848 0.000 1.000
#> GSM97119 1 0.9833 0.5574 0.576 0.424
#> GSM97122 1 0.9710 0.5719 0.600 0.400
#> GSM97135 1 0.9833 0.5574 0.576 0.424
#> GSM97136 2 0.4690 0.5755 0.100 0.900
#> GSM97139 1 0.9209 0.5994 0.664 0.336
#> GSM97146 1 0.0938 0.6220 0.988 0.012
#> GSM97123 2 0.9087 0.5454 0.324 0.676
#> GSM97129 2 0.7056 0.4206 0.192 0.808
#> GSM97143 2 0.9850 -0.3166 0.428 0.572
#> GSM97113 2 0.1843 0.6791 0.028 0.972
#> GSM97056 1 0.0000 0.6241 1.000 0.000
#> GSM97124 1 0.9954 0.5184 0.540 0.460
#> GSM97132 1 0.8386 0.5961 0.732 0.268
#> GSM97144 1 0.2043 0.6143 0.968 0.032
#> GSM97149 1 0.2948 0.6010 0.948 0.052
#> GSM97068 2 0.9833 0.4931 0.424 0.576
#> GSM97071 2 0.9963 0.4275 0.464 0.536
#> GSM97086 2 0.9833 0.4931 0.424 0.576
#> GSM97103 2 0.0000 0.6848 0.000 1.000
#> GSM97057 2 0.9833 0.4931 0.424 0.576
#> GSM97060 2 0.0000 0.6848 0.000 1.000
#> GSM97075 2 0.0000 0.6848 0.000 1.000
#> GSM97098 2 0.0000 0.6848 0.000 1.000
#> GSM97099 2 0.0000 0.6848 0.000 1.000
#> GSM97101 2 0.0000 0.6848 0.000 1.000
#> GSM97105 2 0.9833 0.4931 0.424 0.576
#> GSM97106 2 0.0000 0.6848 0.000 1.000
#> GSM97121 2 0.9833 0.4931 0.424 0.576
#> GSM97128 1 0.9635 0.0586 0.612 0.388
#> GSM97131 2 0.1843 0.6794 0.028 0.972
#> GSM97137 1 0.3114 0.5950 0.944 0.056
#> GSM97118 2 0.9833 -0.2917 0.424 0.576
#> GSM97114 2 0.0938 0.6753 0.012 0.988
#> GSM97142 1 0.9833 0.5574 0.576 0.424
#> GSM97140 2 0.9833 0.4931 0.424 0.576
#> GSM97141 2 0.0000 0.6848 0.000 1.000
#> GSM97055 2 0.8661 0.1445 0.288 0.712
#> GSM97090 1 0.6973 0.4149 0.812 0.188
#> GSM97091 1 0.9881 0.5471 0.564 0.436
#> GSM97148 1 0.0376 0.6238 0.996 0.004
#> GSM97063 1 0.9833 0.5574 0.576 0.424
#> GSM97053 1 0.0000 0.6241 1.000 0.000
#> GSM97066 2 0.0672 0.6797 0.008 0.992
#> GSM97079 2 0.9833 0.4931 0.424 0.576
#> GSM97083 1 0.2423 0.6101 0.960 0.040
#> GSM97084 2 0.9922 0.4693 0.448 0.552
#> GSM97094 2 0.9087 0.4177 0.324 0.676
#> GSM97096 2 0.0000 0.6848 0.000 1.000
#> GSM97097 2 0.0000 0.6848 0.000 1.000
#> GSM97107 2 0.7219 0.5970 0.200 0.800
#> GSM97054 2 0.9896 0.4788 0.440 0.560
#> GSM97062 2 0.9993 0.4207 0.484 0.516
#> GSM97069 2 0.0000 0.6848 0.000 1.000
#> GSM97070 2 0.0376 0.6823 0.004 0.996
#> GSM97073 2 0.0000 0.6848 0.000 1.000
#> GSM97076 2 0.9963 0.4445 0.464 0.536
#> GSM97077 2 0.9833 0.4931 0.424 0.576
#> GSM97095 2 0.9970 0.4445 0.468 0.532
#> GSM97102 2 0.0000 0.6848 0.000 1.000
#> GSM97109 2 0.0376 0.6818 0.004 0.996
#> GSM97110 2 0.0000 0.6848 0.000 1.000
#> GSM97074 1 0.9358 0.5776 0.648 0.352
#> GSM97085 2 0.4022 0.6105 0.080 0.920
#> GSM97059 2 0.9833 0.4931 0.424 0.576
#> GSM97072 2 0.0000 0.6848 0.000 1.000
#> GSM97078 1 0.9087 0.0565 0.676 0.324
#> GSM97067 2 0.0938 0.6768 0.012 0.988
#> GSM97087 2 0.0000 0.6848 0.000 1.000
#> GSM97111 2 0.0000 0.6848 0.000 1.000
#> GSM97064 2 0.9833 0.4931 0.424 0.576
#> GSM97065 2 0.8909 0.5526 0.308 0.692
#> GSM97081 2 0.0000 0.6848 0.000 1.000
#> GSM97082 2 0.0000 0.6848 0.000 1.000
#> GSM97088 2 0.5946 0.5963 0.144 0.856
#> GSM97100 2 0.9833 0.4931 0.424 0.576
#> GSM97104 2 0.0000 0.6848 0.000 1.000
#> GSM97108 2 0.0000 0.6848 0.000 1.000
#> GSM97050 2 0.9850 0.4891 0.428 0.572
#> GSM97080 2 0.0000 0.6848 0.000 1.000
#> GSM97089 2 0.1843 0.6608 0.028 0.972
#> GSM97092 2 0.0000 0.6848 0.000 1.000
#> GSM97093 2 0.9909 0.4715 0.444 0.556
#> GSM97058 2 0.9833 0.4931 0.424 0.576
#> GSM97051 2 0.9833 0.4931 0.424 0.576
#> GSM97052 2 0.0000 0.6848 0.000 1.000
#> GSM97061 2 0.5408 0.6424 0.124 0.876
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM97138 1 0.5785 0.49417 0.668 0.000 0.332
#> GSM97145 1 0.1585 0.79583 0.964 0.008 0.028
#> GSM97147 2 0.3921 0.74401 0.112 0.872 0.016
#> GSM97125 1 0.0592 0.80012 0.988 0.000 0.012
#> GSM97127 1 0.0747 0.79577 0.984 0.016 0.000
#> GSM97130 2 0.4291 0.70302 0.180 0.820 0.000
#> GSM97133 1 0.5591 0.45711 0.696 0.304 0.000
#> GSM97134 2 0.6154 0.29846 0.408 0.592 0.000
#> GSM97120 1 0.5650 0.50114 0.688 0.000 0.312
#> GSM97126 1 0.9464 0.16284 0.412 0.180 0.408
#> GSM97112 1 0.0424 0.79989 0.992 0.008 0.000
#> GSM97115 2 0.4068 0.77801 0.016 0.864 0.120
#> GSM97116 1 0.2569 0.79029 0.936 0.032 0.032
#> GSM97117 3 0.4178 0.83119 0.000 0.172 0.828
#> GSM97119 1 0.0747 0.79964 0.984 0.016 0.000
#> GSM97122 1 0.0747 0.79964 0.984 0.016 0.000
#> GSM97135 1 0.0424 0.79989 0.992 0.008 0.000
#> GSM97136 3 0.5633 0.68104 0.208 0.024 0.768
#> GSM97139 1 0.1015 0.79911 0.980 0.008 0.012
#> GSM97146 1 0.6079 0.27021 0.612 0.388 0.000
#> GSM97123 2 0.6079 0.27612 0.000 0.612 0.388
#> GSM97129 3 0.8623 0.57857 0.224 0.176 0.600
#> GSM97143 3 0.6228 0.46777 0.316 0.012 0.672
#> GSM97113 3 0.3038 0.85387 0.000 0.104 0.896
#> GSM97056 2 0.4702 0.68301 0.212 0.788 0.000
#> GSM97124 1 0.2998 0.78219 0.916 0.016 0.068
#> GSM97132 1 0.8483 0.47821 0.600 0.260 0.140
#> GSM97144 2 0.5070 0.66412 0.224 0.772 0.004
#> GSM97149 2 0.6143 0.55586 0.304 0.684 0.012
#> GSM97068 2 0.2066 0.80644 0.000 0.940 0.060
#> GSM97071 2 0.3030 0.78495 0.004 0.904 0.092
#> GSM97086 2 0.0747 0.80348 0.000 0.984 0.016
#> GSM97103 3 0.0237 0.87879 0.000 0.004 0.996
#> GSM97057 2 0.3482 0.78221 0.000 0.872 0.128
#> GSM97060 3 0.1031 0.87829 0.000 0.024 0.976
#> GSM97075 3 0.4002 0.84011 0.000 0.160 0.840
#> GSM97098 3 0.0000 0.87850 0.000 0.000 1.000
#> GSM97099 3 0.2165 0.87250 0.000 0.064 0.936
#> GSM97101 3 0.2165 0.87250 0.000 0.064 0.936
#> GSM97105 2 0.3340 0.75580 0.000 0.880 0.120
#> GSM97106 3 0.0000 0.87850 0.000 0.000 1.000
#> GSM97121 2 0.2066 0.79523 0.000 0.940 0.060
#> GSM97128 2 0.8109 0.46822 0.272 0.620 0.108
#> GSM97131 3 0.4654 0.80703 0.000 0.208 0.792
#> GSM97137 2 0.5174 0.73456 0.128 0.824 0.048
#> GSM97118 3 0.8840 -0.14645 0.428 0.116 0.456
#> GSM97114 3 0.7348 0.73663 0.120 0.176 0.704
#> GSM97142 1 0.0592 0.80013 0.988 0.012 0.000
#> GSM97140 2 0.0747 0.80348 0.000 0.984 0.016
#> GSM97141 3 0.2165 0.87250 0.000 0.064 0.936
#> GSM97055 1 0.9395 0.00244 0.432 0.172 0.396
#> GSM97090 2 0.4575 0.71215 0.160 0.828 0.012
#> GSM97091 1 0.2383 0.79050 0.940 0.016 0.044
#> GSM97148 1 0.6081 0.38453 0.652 0.344 0.004
#> GSM97063 1 0.0592 0.80013 0.988 0.012 0.000
#> GSM97053 1 0.0000 0.79900 1.000 0.000 0.000
#> GSM97066 3 0.0237 0.87926 0.000 0.004 0.996
#> GSM97079 2 0.3340 0.79296 0.000 0.880 0.120
#> GSM97083 2 0.5216 0.62087 0.260 0.740 0.000
#> GSM97084 2 0.0592 0.80337 0.000 0.988 0.012
#> GSM97094 2 0.7741 0.29012 0.056 0.568 0.376
#> GSM97096 3 0.0000 0.87850 0.000 0.000 1.000
#> GSM97097 3 0.2959 0.85824 0.000 0.100 0.900
#> GSM97107 2 0.6566 0.29577 0.012 0.612 0.376
#> GSM97054 2 0.2796 0.79773 0.000 0.908 0.092
#> GSM97062 2 0.1491 0.80320 0.016 0.968 0.016
#> GSM97069 3 0.0000 0.87850 0.000 0.000 1.000
#> GSM97070 3 0.0000 0.87850 0.000 0.000 1.000
#> GSM97073 3 0.0000 0.87850 0.000 0.000 1.000
#> GSM97076 2 0.7065 0.60464 0.040 0.644 0.316
#> GSM97077 2 0.0747 0.80348 0.000 0.984 0.016
#> GSM97095 2 0.0829 0.80354 0.004 0.984 0.012
#> GSM97102 3 0.0000 0.87850 0.000 0.000 1.000
#> GSM97109 3 0.0829 0.87911 0.004 0.012 0.984
#> GSM97110 3 0.0592 0.87933 0.000 0.012 0.988
#> GSM97074 1 0.7741 0.61313 0.660 0.104 0.236
#> GSM97085 3 0.2269 0.85212 0.040 0.016 0.944
#> GSM97059 2 0.0747 0.80348 0.000 0.984 0.016
#> GSM97072 3 0.0000 0.87850 0.000 0.000 1.000
#> GSM97078 2 0.5883 0.73498 0.112 0.796 0.092
#> GSM97067 3 0.0000 0.87850 0.000 0.000 1.000
#> GSM97087 3 0.1289 0.87797 0.000 0.032 0.968
#> GSM97111 3 0.4178 0.83119 0.000 0.172 0.828
#> GSM97064 2 0.1031 0.80563 0.000 0.976 0.024
#> GSM97065 2 0.6683 0.18991 0.008 0.500 0.492
#> GSM97081 3 0.4062 0.83560 0.000 0.164 0.836
#> GSM97082 3 0.3941 0.83900 0.000 0.156 0.844
#> GSM97088 3 0.6372 0.70102 0.084 0.152 0.764
#> GSM97100 2 0.0747 0.80348 0.000 0.984 0.016
#> GSM97104 3 0.0000 0.87850 0.000 0.000 1.000
#> GSM97108 3 0.4235 0.82957 0.000 0.176 0.824
#> GSM97050 2 0.3412 0.79384 0.000 0.876 0.124
#> GSM97080 3 0.0237 0.87926 0.000 0.004 0.996
#> GSM97089 3 0.0000 0.87850 0.000 0.000 1.000
#> GSM97092 3 0.4121 0.83334 0.000 0.168 0.832
#> GSM97093 2 0.4700 0.75132 0.008 0.812 0.180
#> GSM97058 2 0.1411 0.80473 0.000 0.964 0.036
#> GSM97051 2 0.1289 0.80296 0.000 0.968 0.032
#> GSM97052 3 0.4121 0.83334 0.000 0.168 0.832
#> GSM97061 3 0.5529 0.68934 0.000 0.296 0.704
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM97138 1 0.0000 0.921 1.000 0.000 0.000 0.000
#> GSM97145 1 0.0992 0.913 0.976 0.004 0.008 0.012
#> GSM97147 2 0.2300 0.782 0.000 0.920 0.016 0.064
#> GSM97125 1 0.0336 0.920 0.992 0.000 0.000 0.008
#> GSM97127 1 0.0336 0.919 0.992 0.000 0.000 0.008
#> GSM97130 2 0.3754 0.740 0.064 0.852 0.000 0.084
#> GSM97133 1 0.0469 0.914 0.988 0.012 0.000 0.000
#> GSM97134 2 0.5952 0.589 0.184 0.692 0.000 0.124
#> GSM97120 1 0.0469 0.913 0.988 0.000 0.012 0.000
#> GSM97126 3 0.8030 0.337 0.164 0.300 0.504 0.032
#> GSM97112 4 0.3649 0.809 0.204 0.000 0.000 0.796
#> GSM97115 2 0.3659 0.759 0.016 0.868 0.032 0.084
#> GSM97116 1 0.0000 0.921 1.000 0.000 0.000 0.000
#> GSM97117 3 0.4872 0.804 0.000 0.148 0.776 0.076
#> GSM97119 4 0.3610 0.809 0.200 0.000 0.000 0.800
#> GSM97122 1 0.4605 0.386 0.664 0.000 0.000 0.336
#> GSM97135 1 0.2921 0.776 0.860 0.000 0.000 0.140
#> GSM97136 3 0.4007 0.815 0.068 0.020 0.856 0.056
#> GSM97139 1 0.0000 0.921 1.000 0.000 0.000 0.000
#> GSM97146 1 0.0000 0.921 1.000 0.000 0.000 0.000
#> GSM97123 2 0.6684 0.260 0.000 0.560 0.336 0.104
#> GSM97129 3 0.6137 0.797 0.044 0.152 0.728 0.076
#> GSM97143 3 0.5911 0.649 0.196 0.000 0.692 0.112
#> GSM97113 3 0.5339 0.758 0.088 0.144 0.760 0.008
#> GSM97056 2 0.4990 0.484 0.352 0.640 0.000 0.008
#> GSM97124 3 0.7506 0.250 0.308 0.000 0.484 0.208
#> GSM97132 2 0.8758 0.387 0.164 0.500 0.236 0.100
#> GSM97144 2 0.4581 0.708 0.080 0.800 0.000 0.120
#> GSM97149 1 0.2480 0.811 0.904 0.088 0.008 0.000
#> GSM97068 2 0.0804 0.797 0.000 0.980 0.012 0.008
#> GSM97071 2 0.3833 0.773 0.000 0.848 0.080 0.072
#> GSM97086 2 0.0188 0.796 0.000 0.996 0.000 0.004
#> GSM97103 3 0.1022 0.833 0.000 0.000 0.968 0.032
#> GSM97057 2 0.3450 0.759 0.084 0.872 0.040 0.004
#> GSM97060 3 0.2179 0.826 0.000 0.012 0.924 0.064
#> GSM97075 3 0.4775 0.809 0.000 0.140 0.784 0.076
#> GSM97098 3 0.0469 0.829 0.000 0.000 0.988 0.012
#> GSM97099 3 0.4318 0.822 0.000 0.116 0.816 0.068
#> GSM97101 3 0.4344 0.822 0.000 0.108 0.816 0.076
#> GSM97105 2 0.4318 0.729 0.000 0.816 0.116 0.068
#> GSM97106 3 0.1637 0.824 0.000 0.000 0.940 0.060
#> GSM97121 2 0.3323 0.768 0.000 0.876 0.060 0.064
#> GSM97128 4 0.4576 0.596 0.012 0.260 0.000 0.728
#> GSM97131 3 0.5021 0.784 0.000 0.180 0.756 0.064
#> GSM97137 2 0.3741 0.739 0.036 0.852 0.004 0.108
#> GSM97118 3 0.8463 0.309 0.064 0.156 0.492 0.288
#> GSM97114 3 0.5390 0.814 0.028 0.120 0.776 0.076
#> GSM97142 4 0.3649 0.808 0.204 0.000 0.000 0.796
#> GSM97140 2 0.2489 0.780 0.000 0.912 0.020 0.068
#> GSM97141 3 0.4344 0.822 0.000 0.108 0.816 0.076
#> GSM97055 4 0.4274 0.740 0.120 0.040 0.012 0.828
#> GSM97090 2 0.3470 0.732 0.008 0.852 0.008 0.132
#> GSM97091 4 0.3688 0.806 0.208 0.000 0.000 0.792
#> GSM97148 1 0.0707 0.908 0.980 0.020 0.000 0.000
#> GSM97063 4 0.4222 0.733 0.272 0.000 0.000 0.728
#> GSM97053 1 0.2530 0.830 0.888 0.000 0.000 0.112
#> GSM97066 3 0.3088 0.809 0.000 0.008 0.864 0.128
#> GSM97079 2 0.3088 0.771 0.000 0.864 0.128 0.008
#> GSM97083 4 0.3542 0.766 0.028 0.120 0.000 0.852
#> GSM97084 2 0.0657 0.796 0.000 0.984 0.004 0.012
#> GSM97094 2 0.6815 0.230 0.016 0.552 0.364 0.068
#> GSM97096 3 0.0817 0.828 0.000 0.000 0.976 0.024
#> GSM97097 3 0.3088 0.834 0.000 0.060 0.888 0.052
#> GSM97107 2 0.6626 0.248 0.000 0.544 0.364 0.092
#> GSM97054 2 0.1584 0.795 0.000 0.952 0.036 0.012
#> GSM97062 2 0.2053 0.777 0.000 0.924 0.004 0.072
#> GSM97069 3 0.2530 0.811 0.000 0.000 0.888 0.112
#> GSM97070 3 0.2255 0.835 0.000 0.012 0.920 0.068
#> GSM97073 3 0.1118 0.830 0.000 0.000 0.964 0.036
#> GSM97076 2 0.6791 0.667 0.036 0.668 0.192 0.104
#> GSM97077 2 0.0000 0.796 0.000 1.000 0.000 0.000
#> GSM97095 2 0.0336 0.796 0.000 0.992 0.000 0.008
#> GSM97102 3 0.2081 0.812 0.000 0.000 0.916 0.084
#> GSM97109 3 0.3349 0.826 0.064 0.004 0.880 0.052
#> GSM97110 3 0.3241 0.816 0.040 0.072 0.884 0.004
#> GSM97074 4 0.2973 0.815 0.144 0.000 0.000 0.856
#> GSM97085 4 0.2530 0.735 0.000 0.000 0.112 0.888
#> GSM97059 2 0.1635 0.791 0.000 0.948 0.008 0.044
#> GSM97072 3 0.1022 0.826 0.000 0.000 0.968 0.032
#> GSM97078 2 0.5300 0.371 0.000 0.580 0.012 0.408
#> GSM97067 3 0.2530 0.804 0.000 0.000 0.888 0.112
#> GSM97087 3 0.3697 0.834 0.000 0.048 0.852 0.100
#> GSM97111 3 0.4872 0.804 0.000 0.148 0.776 0.076
#> GSM97064 2 0.1545 0.797 0.000 0.952 0.008 0.040
#> GSM97065 2 0.7404 0.336 0.096 0.508 0.372 0.024
#> GSM97081 3 0.4605 0.816 0.000 0.132 0.796 0.072
#> GSM97082 3 0.5528 0.812 0.000 0.124 0.732 0.144
#> GSM97088 4 0.2945 0.772 0.012 0.052 0.032 0.904
#> GSM97100 2 0.2413 0.781 0.000 0.916 0.020 0.064
#> GSM97104 3 0.2530 0.803 0.000 0.000 0.888 0.112
#> GSM97108 3 0.4829 0.803 0.000 0.156 0.776 0.068
#> GSM97050 2 0.3464 0.779 0.000 0.860 0.108 0.032
#> GSM97080 3 0.2944 0.807 0.000 0.004 0.868 0.128
#> GSM97089 3 0.1888 0.830 0.016 0.000 0.940 0.044
#> GSM97092 3 0.5314 0.811 0.000 0.144 0.748 0.108
#> GSM97093 2 0.4185 0.751 0.012 0.832 0.120 0.036
#> GSM97058 2 0.2797 0.782 0.000 0.900 0.032 0.068
#> GSM97051 2 0.2706 0.777 0.000 0.900 0.020 0.080
#> GSM97052 3 0.5110 0.824 0.000 0.132 0.764 0.104
#> GSM97061 3 0.6323 0.655 0.000 0.272 0.628 0.100
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM97138 1 0.0290 0.84673 0.992 0.000 0.000 0.000 0.008
#> GSM97145 1 0.3878 0.80127 0.808 0.144 0.012 0.000 0.036
#> GSM97147 4 0.5414 0.66665 0.000 0.140 0.200 0.660 0.000
#> GSM97125 1 0.4887 0.74170 0.720 0.148 0.000 0.000 0.132
#> GSM97127 1 0.4437 0.77404 0.760 0.140 0.000 0.000 0.100
#> GSM97130 4 0.2523 0.75020 0.028 0.024 0.000 0.908 0.040
#> GSM97133 1 0.2329 0.82391 0.876 0.124 0.000 0.000 0.000
#> GSM97134 4 0.4017 0.71173 0.060 0.032 0.000 0.824 0.084
#> GSM97120 1 0.2068 0.83451 0.904 0.092 0.000 0.000 0.004
#> GSM97126 3 0.8545 0.09581 0.152 0.156 0.436 0.228 0.028
#> GSM97112 5 0.1544 0.71553 0.068 0.000 0.000 0.000 0.932
#> GSM97115 4 0.1967 0.75981 0.000 0.012 0.020 0.932 0.036
#> GSM97116 1 0.0000 0.84787 1.000 0.000 0.000 0.000 0.000
#> GSM97117 3 0.2079 0.45252 0.000 0.020 0.916 0.064 0.000
#> GSM97119 5 0.3868 0.63113 0.060 0.140 0.000 0.000 0.800
#> GSM97122 5 0.5744 0.18789 0.332 0.104 0.000 0.000 0.564
#> GSM97135 1 0.4114 0.42749 0.624 0.000 0.000 0.000 0.376
#> GSM97136 3 0.5985 0.10151 0.004 0.372 0.544 0.016 0.064
#> GSM97139 1 0.0000 0.84787 1.000 0.000 0.000 0.000 0.000
#> GSM97146 1 0.0000 0.84787 1.000 0.000 0.000 0.000 0.000
#> GSM97123 3 0.6822 -0.00214 0.000 0.332 0.348 0.320 0.000
#> GSM97129 3 0.4017 0.38972 0.000 0.148 0.788 0.064 0.000
#> GSM97143 3 0.5933 0.20668 0.080 0.028 0.676 0.016 0.200
#> GSM97113 3 0.4795 0.29682 0.032 0.020 0.740 0.200 0.008
#> GSM97056 4 0.4299 0.38490 0.388 0.000 0.000 0.608 0.004
#> GSM97124 5 0.7871 0.27542 0.128 0.188 0.220 0.000 0.464
#> GSM97132 4 0.8131 0.22440 0.080 0.028 0.204 0.476 0.212
#> GSM97144 4 0.3152 0.73312 0.016 0.032 0.000 0.868 0.084
#> GSM97149 1 0.0703 0.83118 0.976 0.000 0.000 0.024 0.000
#> GSM97068 4 0.1197 0.78222 0.000 0.000 0.048 0.952 0.000
#> GSM97071 4 0.4475 0.73913 0.000 0.056 0.180 0.756 0.008
#> GSM97086 4 0.1124 0.78074 0.000 0.004 0.036 0.960 0.000
#> GSM97103 3 0.3612 0.27222 0.000 0.228 0.764 0.000 0.008
#> GSM97057 4 0.2308 0.76809 0.036 0.004 0.048 0.912 0.000
#> GSM97060 3 0.4747 -0.41435 0.000 0.484 0.500 0.000 0.016
#> GSM97075 3 0.1983 0.45592 0.000 0.008 0.924 0.060 0.008
#> GSM97098 3 0.4288 0.13340 0.000 0.324 0.664 0.000 0.012
#> GSM97099 3 0.1356 0.45601 0.000 0.004 0.956 0.028 0.012
#> GSM97101 3 0.1356 0.45510 0.000 0.004 0.956 0.028 0.012
#> GSM97105 4 0.4687 0.65016 0.000 0.040 0.288 0.672 0.000
#> GSM97106 2 0.4150 0.42072 0.000 0.612 0.388 0.000 0.000
#> GSM97121 4 0.4384 0.70369 0.000 0.044 0.228 0.728 0.000
#> GSM97128 5 0.5411 0.64957 0.000 0.176 0.000 0.160 0.664
#> GSM97131 3 0.3003 0.43863 0.000 0.044 0.864 0.092 0.000
#> GSM97137 4 0.1901 0.75682 0.024 0.004 0.000 0.932 0.040
#> GSM97118 3 0.7134 -0.04190 0.008 0.028 0.428 0.144 0.392
#> GSM97114 3 0.4725 0.40391 0.040 0.128 0.772 0.060 0.000
#> GSM97142 5 0.1478 0.71668 0.064 0.000 0.000 0.000 0.936
#> GSM97140 4 0.4129 0.72147 0.000 0.040 0.204 0.756 0.000
#> GSM97141 3 0.1243 0.45589 0.000 0.004 0.960 0.028 0.008
#> GSM97055 5 0.5859 0.67425 0.064 0.152 0.040 0.032 0.712
#> GSM97090 4 0.1626 0.75727 0.000 0.016 0.000 0.940 0.044
#> GSM97091 5 0.1410 0.71706 0.060 0.000 0.000 0.000 0.940
#> GSM97148 1 0.0162 0.84638 0.996 0.000 0.000 0.004 0.000
#> GSM97063 5 0.1851 0.70467 0.088 0.000 0.000 0.000 0.912
#> GSM97053 1 0.5260 0.44135 0.592 0.060 0.000 0.000 0.348
#> GSM97066 2 0.5289 0.51135 0.000 0.500 0.452 0.000 0.048
#> GSM97079 4 0.4205 0.71393 0.000 0.164 0.056 0.776 0.004
#> GSM97083 5 0.4972 0.69923 0.008 0.172 0.000 0.096 0.724
#> GSM97084 4 0.1267 0.78066 0.000 0.012 0.024 0.960 0.004
#> GSM97094 4 0.6361 0.14190 0.000 0.048 0.412 0.484 0.056
#> GSM97096 3 0.4341 -0.00957 0.000 0.404 0.592 0.000 0.004
#> GSM97097 3 0.4735 0.26337 0.000 0.304 0.664 0.024 0.008
#> GSM97107 4 0.5993 0.09429 0.000 0.056 0.440 0.480 0.024
#> GSM97054 4 0.1662 0.77893 0.000 0.004 0.056 0.936 0.004
#> GSM97062 4 0.1168 0.76514 0.000 0.008 0.000 0.960 0.032
#> GSM97069 3 0.5086 -0.39196 0.000 0.396 0.564 0.000 0.040
#> GSM97070 3 0.4595 0.13761 0.000 0.236 0.716 0.004 0.044
#> GSM97073 3 0.3958 0.24523 0.000 0.184 0.776 0.000 0.040
#> GSM97076 4 0.6098 0.65532 0.080 0.016 0.168 0.684 0.052
#> GSM97077 4 0.1768 0.78011 0.000 0.004 0.072 0.924 0.000
#> GSM97095 4 0.1116 0.78078 0.000 0.004 0.028 0.964 0.004
#> GSM97102 3 0.4446 -0.22938 0.000 0.476 0.520 0.000 0.004
#> GSM97109 3 0.5210 0.25127 0.088 0.200 0.700 0.000 0.012
#> GSM97110 3 0.4794 0.32404 0.052 0.112 0.784 0.040 0.012
#> GSM97074 5 0.4319 0.71648 0.028 0.176 0.000 0.024 0.772
#> GSM97085 5 0.4268 0.65580 0.000 0.268 0.000 0.024 0.708
#> GSM97059 4 0.3370 0.75575 0.000 0.028 0.148 0.824 0.000
#> GSM97072 3 0.4227 -0.04408 0.000 0.420 0.580 0.000 0.000
#> GSM97078 4 0.6544 0.11871 0.000 0.196 0.004 0.492 0.308
#> GSM97067 3 0.5344 -0.51177 0.000 0.448 0.500 0.000 0.052
#> GSM97087 3 0.5679 -0.27590 0.000 0.364 0.560 0.008 0.068
#> GSM97111 3 0.3055 0.44508 0.000 0.072 0.864 0.064 0.000
#> GSM97064 4 0.3233 0.77733 0.000 0.028 0.112 0.852 0.008
#> GSM97065 4 0.6911 0.29269 0.108 0.020 0.356 0.496 0.020
#> GSM97081 3 0.3260 0.44281 0.000 0.084 0.856 0.056 0.004
#> GSM97082 3 0.5670 -0.40932 0.000 0.452 0.488 0.016 0.044
#> GSM97088 5 0.5155 0.61679 0.000 0.284 0.020 0.036 0.660
#> GSM97100 4 0.4096 0.72357 0.000 0.040 0.200 0.760 0.000
#> GSM97104 2 0.3752 0.57388 0.000 0.708 0.292 0.000 0.000
#> GSM97108 3 0.2504 0.44666 0.000 0.040 0.896 0.064 0.000
#> GSM97050 4 0.3443 0.76499 0.000 0.076 0.060 0.852 0.012
#> GSM97080 2 0.5204 0.58878 0.000 0.560 0.392 0.000 0.048
#> GSM97089 3 0.4387 0.21529 0.000 0.232 0.732 0.008 0.028
#> GSM97092 3 0.4941 0.24437 0.000 0.244 0.696 0.048 0.012
#> GSM97093 4 0.3089 0.75590 0.000 0.076 0.040 0.872 0.012
#> GSM97058 4 0.3977 0.72908 0.000 0.032 0.204 0.764 0.000
#> GSM97051 4 0.4369 0.71500 0.000 0.052 0.208 0.740 0.000
#> GSM97052 3 0.5007 0.05043 0.000 0.320 0.640 0.024 0.016
#> GSM97061 3 0.5638 0.23258 0.000 0.192 0.636 0.172 0.000
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM97138 1 0.0146 0.79622 0.996 0.000 0.000 0.000 0.004 0.000
#> GSM97145 1 0.3969 0.63128 0.644 0.000 0.008 0.000 0.344 0.004
#> GSM97147 2 0.5887 -0.29796 0.000 0.408 0.000 0.392 0.200 0.000
#> GSM97125 1 0.4172 0.47184 0.528 0.000 0.000 0.000 0.460 0.012
#> GSM97127 1 0.3706 0.59873 0.620 0.000 0.000 0.000 0.380 0.000
#> GSM97130 4 0.3142 0.68249 0.016 0.000 0.000 0.848 0.044 0.092
#> GSM97133 1 0.2730 0.73418 0.808 0.000 0.000 0.000 0.192 0.000
#> GSM97134 4 0.3703 0.64464 0.000 0.000 0.000 0.788 0.108 0.104
#> GSM97120 1 0.2178 0.76330 0.868 0.000 0.000 0.000 0.132 0.000
#> GSM97126 5 0.8931 -0.10182 0.124 0.212 0.024 0.256 0.296 0.088
#> GSM97112 5 0.3857 0.48319 0.000 0.000 0.000 0.000 0.532 0.468
#> GSM97115 4 0.1148 0.72511 0.000 0.016 0.000 0.960 0.004 0.020
#> GSM97116 1 0.0000 0.79817 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM97117 2 0.1854 0.43603 0.000 0.932 0.016 0.028 0.020 0.004
#> GSM97119 5 0.2912 0.50403 0.000 0.000 0.000 0.000 0.784 0.216
#> GSM97122 5 0.3825 0.48287 0.160 0.000 0.000 0.000 0.768 0.072
#> GSM97135 1 0.4097 -0.07167 0.504 0.000 0.000 0.000 0.488 0.008
#> GSM97136 3 0.7441 0.35331 0.004 0.288 0.376 0.008 0.244 0.080
#> GSM97139 1 0.0000 0.79817 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM97146 1 0.0000 0.79817 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM97123 2 0.7487 0.15099 0.000 0.400 0.316 0.184 0.048 0.052
#> GSM97129 2 0.5237 0.29600 0.000 0.652 0.024 0.036 0.260 0.028
#> GSM97143 2 0.6392 0.08929 0.012 0.456 0.012 0.024 0.400 0.096
#> GSM97113 2 0.5983 0.18845 0.012 0.588 0.036 0.292 0.016 0.056
#> GSM97056 4 0.3955 0.30015 0.436 0.000 0.000 0.560 0.004 0.000
#> GSM97124 5 0.3401 0.43082 0.036 0.072 0.000 0.000 0.840 0.052
#> GSM97132 4 0.7850 0.08025 0.048 0.188 0.000 0.412 0.248 0.104
#> GSM97144 4 0.3566 0.64839 0.000 0.000 0.000 0.800 0.096 0.104
#> GSM97149 1 0.0000 0.79817 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM97068 4 0.1686 0.73473 0.000 0.064 0.000 0.924 0.000 0.012
#> GSM97071 4 0.5187 0.63741 0.000 0.196 0.012 0.688 0.032 0.072
#> GSM97086 4 0.1663 0.73030 0.000 0.088 0.000 0.912 0.000 0.000
#> GSM97103 2 0.5290 -0.33089 0.000 0.496 0.440 0.012 0.036 0.016
#> GSM97057 4 0.2466 0.72408 0.028 0.052 0.000 0.896 0.000 0.024
#> GSM97060 3 0.5165 0.38865 0.000 0.244 0.636 0.000 0.012 0.108
#> GSM97075 2 0.1971 0.43587 0.000 0.928 0.016 0.024 0.008 0.024
#> GSM97098 3 0.5285 0.53489 0.000 0.376 0.556 0.012 0.032 0.024
#> GSM97099 2 0.2601 0.40021 0.000 0.896 0.040 0.012 0.016 0.036
#> GSM97101 2 0.2627 0.40494 0.000 0.892 0.032 0.008 0.016 0.052
#> GSM97105 4 0.4644 0.45342 0.000 0.440 0.004 0.524 0.032 0.000
#> GSM97106 3 0.2201 0.54455 0.000 0.056 0.904 0.000 0.004 0.036
#> GSM97121 4 0.4494 0.47772 0.000 0.424 0.000 0.544 0.032 0.000
#> GSM97128 6 0.5631 0.15575 0.000 0.000 0.016 0.140 0.264 0.580
#> GSM97131 2 0.1856 0.43302 0.000 0.920 0.000 0.048 0.032 0.000
#> GSM97137 4 0.1237 0.72428 0.020 0.000 0.000 0.956 0.004 0.020
#> GSM97118 2 0.7221 -0.01801 0.000 0.384 0.004 0.120 0.344 0.148
#> GSM97114 2 0.3880 0.36686 0.028 0.772 0.000 0.024 0.176 0.000
#> GSM97142 5 0.3857 0.48319 0.000 0.000 0.000 0.000 0.532 0.468
#> GSM97140 4 0.4461 0.50199 0.000 0.404 0.000 0.564 0.032 0.000
#> GSM97141 2 0.2533 0.40369 0.000 0.900 0.032 0.008 0.028 0.032
#> GSM97055 6 0.5247 0.14038 0.056 0.080 0.000 0.012 0.144 0.708
#> GSM97090 4 0.1010 0.72105 0.000 0.000 0.000 0.960 0.004 0.036
#> GSM97091 5 0.3857 0.48319 0.000 0.000 0.000 0.000 0.532 0.468
#> GSM97148 1 0.0000 0.79817 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM97063 5 0.3857 0.48319 0.000 0.000 0.000 0.000 0.532 0.468
#> GSM97053 5 0.4742 -0.00636 0.440 0.000 0.000 0.000 0.512 0.048
#> GSM97066 6 0.6352 -0.12436 0.000 0.364 0.228 0.000 0.016 0.392
#> GSM97079 4 0.4842 0.51040 0.000 0.032 0.292 0.648 0.020 0.008
#> GSM97083 6 0.4316 0.21810 0.000 0.000 0.000 0.128 0.144 0.728
#> GSM97084 4 0.1508 0.73501 0.000 0.048 0.004 0.940 0.004 0.004
#> GSM97094 2 0.6816 0.07792 0.000 0.408 0.092 0.392 0.100 0.008
#> GSM97096 3 0.4201 0.61396 0.000 0.280 0.688 0.004 0.020 0.008
#> GSM97097 3 0.5085 0.52961 0.000 0.356 0.584 0.016 0.032 0.012
#> GSM97107 2 0.7226 0.02417 0.000 0.400 0.068 0.388 0.064 0.080
#> GSM97054 4 0.1829 0.73202 0.000 0.056 0.000 0.920 0.000 0.024
#> GSM97062 4 0.0777 0.72670 0.000 0.000 0.000 0.972 0.004 0.024
#> GSM97069 2 0.6292 -0.07868 0.000 0.432 0.284 0.000 0.012 0.272
#> GSM97070 2 0.5814 0.10464 0.000 0.568 0.164 0.000 0.020 0.248
#> GSM97073 2 0.5722 -0.08264 0.000 0.576 0.296 0.004 0.028 0.096
#> GSM97076 4 0.5634 0.65745 0.040 0.060 0.016 0.716 0.056 0.112
#> GSM97077 4 0.2340 0.71367 0.000 0.148 0.000 0.852 0.000 0.000
#> GSM97095 4 0.1531 0.73294 0.000 0.068 0.000 0.928 0.000 0.004
#> GSM97102 3 0.3612 0.62214 0.000 0.200 0.764 0.000 0.000 0.036
#> GSM97109 3 0.6510 0.47111 0.056 0.396 0.468 0.012 0.032 0.036
#> GSM97110 2 0.6131 -0.11208 0.020 0.588 0.288 0.032 0.036 0.036
#> GSM97074 6 0.3714 0.14595 0.000 0.000 0.000 0.044 0.196 0.760
#> GSM97085 6 0.3469 0.27051 0.000 0.000 0.064 0.004 0.120 0.812
#> GSM97059 4 0.3879 0.61694 0.000 0.292 0.000 0.688 0.020 0.000
#> GSM97072 3 0.3445 0.62590 0.000 0.260 0.732 0.000 0.000 0.008
#> GSM97078 4 0.5867 0.10504 0.000 0.000 0.016 0.472 0.128 0.384
#> GSM97067 2 0.6375 -0.07897 0.000 0.404 0.216 0.008 0.008 0.364
#> GSM97087 3 0.6862 0.08895 0.000 0.284 0.396 0.008 0.032 0.280
#> GSM97111 2 0.3083 0.40276 0.000 0.860 0.060 0.028 0.052 0.000
#> GSM97064 4 0.3895 0.71328 0.000 0.108 0.032 0.800 0.000 0.060
#> GSM97065 4 0.7224 0.27961 0.084 0.288 0.032 0.504 0.024 0.068
#> GSM97081 2 0.4940 0.20886 0.000 0.696 0.220 0.028 0.032 0.024
#> GSM97082 6 0.7005 -0.08091 0.000 0.324 0.296 0.012 0.032 0.336
#> GSM97088 6 0.3428 0.30708 0.000 0.004 0.044 0.028 0.084 0.840
#> GSM97100 4 0.4403 0.50162 0.000 0.408 0.000 0.564 0.028 0.000
#> GSM97104 3 0.2962 0.57005 0.000 0.084 0.848 0.000 0.000 0.068
#> GSM97108 2 0.1498 0.43368 0.000 0.940 0.000 0.028 0.032 0.000
#> GSM97050 4 0.3827 0.70724 0.000 0.048 0.060 0.812 0.000 0.080
#> GSM97080 6 0.6702 -0.07831 0.000 0.248 0.352 0.000 0.036 0.364
#> GSM97089 2 0.6089 0.08287 0.000 0.588 0.180 0.012 0.028 0.192
#> GSM97092 2 0.6734 0.28345 0.000 0.560 0.196 0.028 0.064 0.152
#> GSM97093 4 0.3485 0.70438 0.000 0.052 0.024 0.828 0.000 0.096
#> GSM97058 4 0.4570 0.54755 0.000 0.352 0.000 0.608 0.032 0.008
#> GSM97051 4 0.5196 0.48184 0.000 0.396 0.036 0.536 0.032 0.000
#> GSM97052 2 0.6412 0.17980 0.000 0.500 0.308 0.012 0.028 0.152
#> GSM97061 2 0.6639 0.31607 0.000 0.596 0.168 0.124 0.052 0.060
Heatmaps for the consensus matrix. It visualizes the probability of two samples to be in a same group.
consensus_heatmap(res, k = 2)
consensus_heatmap(res, k = 3)
consensus_heatmap(res, k = 4)
consensus_heatmap(res, k = 5)
consensus_heatmap(res, k = 6)
Heatmaps for the membership of samples in all partitions to see how consistent they are:
membership_heatmap(res, k = 2)
membership_heatmap(res, k = 3)
membership_heatmap(res, k = 4)
membership_heatmap(res, k = 5)
membership_heatmap(res, k = 6)
As soon as we have had the classes for columns, we can look for signatures which are significantly different between classes which can be candidate marks for certain classes. Following are the heatmaps for signatures.
Signature heatmaps where rows are scaled:
get_signatures(res, k = 2)
get_signatures(res, k = 3)
get_signatures(res, k = 4)
get_signatures(res, k = 5)
get_signatures(res, k = 6)
Signature heatmaps where rows are not scaled:
get_signatures(res, k = 2, scale_rows = FALSE)
get_signatures(res, k = 3, scale_rows = FALSE)
get_signatures(res, k = 4, scale_rows = FALSE)
get_signatures(res, k = 5, scale_rows = FALSE)
get_signatures(res, k = 6, scale_rows = FALSE)
Compare the overlap of signatures from different k:
compare_signatures(res)
get_signature() returns a data frame invisibly. TO get the list of signatures, the function
call should be assigned to a variable explicitly. In following code, if plot argument is set
to FALSE, no heatmap is plotted while only the differential analysis is performed.
# code only for demonstration
tb = get_signature(res, k = ..., plot = FALSE)
An example of the output of tb is:
#> which_row fdr mean_1 mean_2 scaled_mean_1 scaled_mean_2 km
#> 1 38 0.042760348 8.373488 9.131774 -0.5533452 0.5164555 1
#> 2 40 0.018707592 7.106213 8.469186 -0.6173731 0.5762149 1
#> 3 55 0.019134737 10.221463 11.207825 -0.6159697 0.5749050 1
#> 4 59 0.006059896 5.921854 7.869574 -0.6899429 0.6439467 1
#> 5 60 0.018055526 8.928898 10.211722 -0.6204761 0.5791110 1
#> 6 98 0.009384629 15.714769 14.887706 0.6635654 -0.6193277 2
...
The columns in tb are:
which_row: row indices corresponding to the input matrix.fdr: FDR for the differential test. mean_x: The mean value in group x.scaled_mean_x: The mean value in group x after rows are scaled.km: Row groups if k-means clustering is applied to rows.UMAP plot which shows how samples are separated.
dimension_reduction(res, k = 2, method = "UMAP")
dimension_reduction(res, k = 3, method = "UMAP")
dimension_reduction(res, k = 4, method = "UMAP")
dimension_reduction(res, k = 5, method = "UMAP")
dimension_reduction(res, k = 6, method = "UMAP")
Following heatmap shows how subgroups are split when increasing k:
collect_classes(res)
Test correlation between subgroups and known annotations. If the known annotation is numeric, one-way ANOVA test is applied, and if the known annotation is discrete, chi-squared contingency table test is applied.
test_to_known_factors(res)
#> n disease.state(p) specimen(p) cell.type(p) other(p) k
#> CV:pam 66 0.001310 0.993 4.68e-11 0.1244 2
#> CV:pam 85 0.000627 0.188 4.51e-08 0.0449 3
#> CV:pam 89 0.002305 0.263 1.49e-06 0.3288 4
#> CV:pam 53 0.004999 0.413 1.39e-03 0.6961 5
#> CV:pam 42 0.007248 0.746 6.80e-04 0.4629 6
If matrix rows can be associated to genes, consider to use functional_enrichment(res,
...) to perform function enrichment for the signature genes. See this vignette for more detailed explanations.
The object with results only for a single top-value method and a single partition method can be extracted as:
res = res_list["CV", "mclust"]
# you can also extract it by
# res = res_list["CV:mclust"]
A summary of res and all the functions that can be applied to it:
res
#> A 'ConsensusPartition' object with k = 2, 3, 4, 5, 6.
#> On a matrix with 21512 rows and 100 columns.
#> Top rows (1000, 2000, 3000, 4000, 5000) are extracted by 'CV' method.
#> Subgroups are detected by 'mclust' method.
#> Performed in total 1250 partitions by row resampling.
#> Best k for subgroups seems to be 3.
#>
#> Following methods can be applied to this 'ConsensusPartition' object:
#> [1] "cola_report" "collect_classes" "collect_plots"
#> [4] "collect_stats" "colnames" "compare_signatures"
#> [7] "consensus_heatmap" "dimension_reduction" "functional_enrichment"
#> [10] "get_anno_col" "get_anno" "get_classes"
#> [13] "get_consensus" "get_matrix" "get_membership"
#> [16] "get_param" "get_signatures" "get_stats"
#> [19] "is_best_k" "is_stable_k" "membership_heatmap"
#> [22] "ncol" "nrow" "plot_ecdf"
#> [25] "rownames" "select_partition_number" "show"
#> [28] "suggest_best_k" "test_to_known_factors"
collect_plots() function collects all the plots made from res for all k (number of partitions)
into one single page to provide an easy and fast comparison between different k.
collect_plots(res)
The plots are:
k and the heatmap of
predicted classes for each k.k.k.k.All the plots in panels can be made by individual functions and they are plotted later in this section.
select_partition_number() produces several plots showing different
statistics for choosing “optimized” k. There are following statistics:
k;k, the area increased is defined as \(A_k - A_{k-1}\).The detailed explanations of these statistics can be found in the cola vignette.
Generally speaking, lower PAC score, higher mean silhouette score or higher
concordance corresponds to better partition. Rand index and Jaccard index
measure how similar the current partition is compared to partition with k-1.
If they are too similar, we won't accept k is better than k-1.
select_partition_number(res)
The numeric values for all these statistics can be obtained by get_stats().
get_stats(res)
#> k 1-PAC mean_silhouette concordance area_increased Rand Jaccard
#> 2 2 0.395 0.736 0.768 0.3575 0.495 0.495
#> 3 3 0.865 0.901 0.950 0.6521 0.795 0.628
#> 4 4 0.529 0.282 0.691 0.1710 0.851 0.675
#> 5 5 0.579 0.499 0.711 0.0934 0.698 0.318
#> 6 6 0.679 0.661 0.809 0.0703 0.845 0.426
suggest_best_k() suggests the best \(k\) based on these statistics. The rules are as follows:
suggest_best_k(res)
#> [1] 3
Following shows the table of the partitions (You need to click the show/hide
code output link to see it). The membership matrix (columns with name p*)
is inferred by
clue::cl_consensus()
function with the SE method. Basically the value in the membership matrix
represents the probability to belong to a certain group. The finall class
label for an item is determined with the group with highest probability it
belongs to.
In get_classes() function, the entropy is calculated from the membership
matrix and the silhouette score is calculated from the consensus matrix.
cbind(get_classes(res, k = 2), get_membership(res, k = 2))
#> class entropy silhouette p1 p2
#> GSM97138 1 0.0000 0.8743 1.000 0.000
#> GSM97145 1 0.0376 0.8712 0.996 0.004
#> GSM97147 1 0.9209 0.0651 0.664 0.336
#> GSM97125 1 0.0000 0.8743 1.000 0.000
#> GSM97127 1 0.0000 0.8743 1.000 0.000
#> GSM97130 1 0.0000 0.8743 1.000 0.000
#> GSM97133 1 0.0000 0.8743 1.000 0.000
#> GSM97134 1 0.0000 0.8743 1.000 0.000
#> GSM97120 1 0.0000 0.8743 1.000 0.000
#> GSM97126 1 0.0000 0.8743 1.000 0.000
#> GSM97112 1 0.0000 0.8743 1.000 0.000
#> GSM97115 1 0.1414 0.8592 0.980 0.020
#> GSM97116 1 0.0000 0.8743 1.000 0.000
#> GSM97117 2 0.9881 0.8008 0.436 0.564
#> GSM97119 1 0.0000 0.8743 1.000 0.000
#> GSM97122 1 0.0000 0.8743 1.000 0.000
#> GSM97135 1 0.0000 0.8743 1.000 0.000
#> GSM97136 2 0.9881 0.8008 0.436 0.564
#> GSM97139 1 0.0000 0.8743 1.000 0.000
#> GSM97146 1 0.0000 0.8743 1.000 0.000
#> GSM97123 2 0.9866 0.7970 0.432 0.568
#> GSM97129 2 0.9881 0.8008 0.436 0.564
#> GSM97143 1 0.0000 0.8743 1.000 0.000
#> GSM97113 2 0.9881 0.8008 0.436 0.564
#> GSM97056 1 0.0000 0.8743 1.000 0.000
#> GSM97124 1 0.0000 0.8743 1.000 0.000
#> GSM97132 1 0.0000 0.8743 1.000 0.000
#> GSM97144 1 0.0000 0.8743 1.000 0.000
#> GSM97149 1 0.0000 0.8743 1.000 0.000
#> GSM97068 1 0.8861 0.1951 0.696 0.304
#> GSM97071 1 0.6438 0.6848 0.836 0.164
#> GSM97086 2 0.9988 0.7005 0.480 0.520
#> GSM97103 2 0.9881 0.8008 0.436 0.564
#> GSM97057 2 0.9881 0.8008 0.436 0.564
#> GSM97060 2 0.9866 0.7973 0.432 0.568
#> GSM97075 2 0.9881 0.8008 0.436 0.564
#> GSM97098 2 0.9881 0.8008 0.436 0.564
#> GSM97099 2 0.9881 0.8008 0.436 0.564
#> GSM97101 2 0.9881 0.8008 0.436 0.564
#> GSM97105 2 0.9881 0.8008 0.436 0.564
#> GSM97106 2 0.9866 0.7970 0.432 0.568
#> GSM97121 2 0.9881 0.8008 0.436 0.564
#> GSM97128 1 0.6247 0.6980 0.844 0.156
#> GSM97131 2 0.9881 0.8008 0.436 0.564
#> GSM97137 1 0.0000 0.8743 1.000 0.000
#> GSM97118 1 0.0000 0.8743 1.000 0.000
#> GSM97114 2 0.9881 0.8008 0.436 0.564
#> GSM97142 1 0.0000 0.8743 1.000 0.000
#> GSM97140 2 0.9881 0.8008 0.436 0.564
#> GSM97141 2 0.9881 0.8008 0.436 0.564
#> GSM97055 1 0.0000 0.8743 1.000 0.000
#> GSM97090 1 0.0672 0.8688 0.992 0.008
#> GSM97091 1 0.0000 0.8743 1.000 0.000
#> GSM97148 1 0.0000 0.8743 1.000 0.000
#> GSM97063 1 0.0000 0.8743 1.000 0.000
#> GSM97053 1 0.0000 0.8743 1.000 0.000
#> GSM97066 2 0.0672 0.4750 0.008 0.992
#> GSM97079 2 0.9998 0.6683 0.492 0.508
#> GSM97083 1 0.0000 0.8743 1.000 0.000
#> GSM97084 1 0.9286 0.0626 0.656 0.344
#> GSM97094 1 0.6438 0.6848 0.836 0.164
#> GSM97096 2 0.9881 0.8008 0.436 0.564
#> GSM97097 2 0.9866 0.7970 0.432 0.568
#> GSM97107 1 0.6438 0.6848 0.836 0.164
#> GSM97054 1 0.9209 0.1074 0.664 0.336
#> GSM97062 1 0.9393 -0.0117 0.644 0.356
#> GSM97069 2 0.0672 0.4750 0.008 0.992
#> GSM97070 2 0.0672 0.4750 0.008 0.992
#> GSM97073 2 0.0672 0.4750 0.008 0.992
#> GSM97076 1 0.7815 0.5169 0.768 0.232
#> GSM97077 2 0.9881 0.8008 0.436 0.564
#> GSM97095 1 0.3114 0.8260 0.944 0.056
#> GSM97102 2 0.0672 0.4750 0.008 0.992
#> GSM97109 2 0.9881 0.8008 0.436 0.564
#> GSM97110 2 0.9881 0.8008 0.436 0.564
#> GSM97074 1 0.4690 0.7768 0.900 0.100
#> GSM97085 1 0.6623 0.6685 0.828 0.172
#> GSM97059 1 0.8661 0.2951 0.712 0.288
#> GSM97072 2 0.7950 0.6085 0.240 0.760
#> GSM97078 1 0.6438 0.6848 0.836 0.164
#> GSM97067 2 0.0672 0.4750 0.008 0.992
#> GSM97087 2 0.0672 0.4750 0.008 0.992
#> GSM97111 2 0.9881 0.8008 0.436 0.564
#> GSM97064 2 0.9881 0.8008 0.436 0.564
#> GSM97065 2 0.9881 0.8008 0.436 0.564
#> GSM97081 2 0.9881 0.8008 0.436 0.564
#> GSM97082 2 0.0672 0.4750 0.008 0.992
#> GSM97088 1 0.6438 0.6848 0.836 0.164
#> GSM97100 2 0.9881 0.8008 0.436 0.564
#> GSM97104 2 0.0672 0.4750 0.008 0.992
#> GSM97108 2 0.9881 0.8008 0.436 0.564
#> GSM97050 2 0.9881 0.8008 0.436 0.564
#> GSM97080 2 0.0672 0.4750 0.008 0.992
#> GSM97089 2 0.9881 0.8008 0.436 0.564
#> GSM97092 2 0.9881 0.8008 0.436 0.564
#> GSM97093 2 0.9881 0.8008 0.436 0.564
#> GSM97058 2 0.9881 0.8008 0.436 0.564
#> GSM97051 2 0.9881 0.8008 0.436 0.564
#> GSM97052 2 0.9881 0.8008 0.436 0.564
#> GSM97061 2 0.9866 0.7970 0.432 0.568
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM97138 1 0.0000 0.956 1.000 0.000 0.000
#> GSM97145 1 0.0000 0.956 1.000 0.000 0.000
#> GSM97147 2 0.6809 0.151 0.464 0.524 0.012
#> GSM97125 1 0.0000 0.956 1.000 0.000 0.000
#> GSM97127 1 0.0424 0.956 0.992 0.008 0.000
#> GSM97130 1 0.1482 0.951 0.968 0.020 0.012
#> GSM97133 1 0.0424 0.956 0.992 0.008 0.000
#> GSM97134 1 0.1877 0.945 0.956 0.032 0.012
#> GSM97120 1 0.0000 0.956 1.000 0.000 0.000
#> GSM97126 1 0.1031 0.952 0.976 0.024 0.000
#> GSM97112 1 0.0000 0.956 1.000 0.000 0.000
#> GSM97115 1 0.5020 0.757 0.796 0.192 0.012
#> GSM97116 1 0.0000 0.956 1.000 0.000 0.000
#> GSM97117 2 0.2537 0.877 0.080 0.920 0.000
#> GSM97119 1 0.0000 0.956 1.000 0.000 0.000
#> GSM97122 1 0.0000 0.956 1.000 0.000 0.000
#> GSM97135 1 0.0000 0.956 1.000 0.000 0.000
#> GSM97136 1 0.5692 0.642 0.724 0.268 0.008
#> GSM97139 1 0.0000 0.956 1.000 0.000 0.000
#> GSM97146 1 0.0000 0.956 1.000 0.000 0.000
#> GSM97123 2 0.0237 0.928 0.004 0.996 0.000
#> GSM97129 2 0.0592 0.929 0.012 0.988 0.000
#> GSM97143 1 0.0000 0.956 1.000 0.000 0.000
#> GSM97113 2 0.1753 0.907 0.048 0.952 0.000
#> GSM97056 1 0.1015 0.954 0.980 0.012 0.008
#> GSM97124 1 0.0424 0.955 0.992 0.000 0.008
#> GSM97132 1 0.0000 0.956 1.000 0.000 0.000
#> GSM97144 1 0.1620 0.949 0.964 0.024 0.012
#> GSM97149 1 0.0592 0.955 0.988 0.012 0.000
#> GSM97068 2 0.5536 0.689 0.236 0.752 0.012
#> GSM97071 1 0.6113 0.568 0.688 0.300 0.012
#> GSM97086 2 0.2031 0.914 0.032 0.952 0.016
#> GSM97103 2 0.0661 0.928 0.008 0.988 0.004
#> GSM97057 2 0.2280 0.904 0.052 0.940 0.008
#> GSM97060 2 0.2173 0.908 0.008 0.944 0.048
#> GSM97075 2 0.0424 0.929 0.008 0.992 0.000
#> GSM97098 2 0.1015 0.926 0.008 0.980 0.012
#> GSM97099 2 0.0424 0.929 0.008 0.992 0.000
#> GSM97101 2 0.0424 0.929 0.008 0.992 0.000
#> GSM97105 2 0.0424 0.929 0.008 0.992 0.000
#> GSM97106 2 0.0237 0.928 0.004 0.996 0.000
#> GSM97121 2 0.0424 0.929 0.008 0.992 0.000
#> GSM97128 1 0.2152 0.941 0.948 0.036 0.016
#> GSM97131 2 0.0237 0.928 0.004 0.996 0.000
#> GSM97137 1 0.0848 0.955 0.984 0.008 0.008
#> GSM97118 1 0.0237 0.956 0.996 0.000 0.004
#> GSM97114 2 0.4974 0.676 0.236 0.764 0.000
#> GSM97142 1 0.0000 0.956 1.000 0.000 0.000
#> GSM97140 2 0.0848 0.927 0.008 0.984 0.008
#> GSM97141 2 0.0237 0.927 0.004 0.996 0.000
#> GSM97055 1 0.0000 0.956 1.000 0.000 0.000
#> GSM97090 1 0.2446 0.931 0.936 0.052 0.012
#> GSM97091 1 0.0000 0.956 1.000 0.000 0.000
#> GSM97148 1 0.0000 0.956 1.000 0.000 0.000
#> GSM97063 1 0.0000 0.956 1.000 0.000 0.000
#> GSM97053 1 0.0000 0.956 1.000 0.000 0.000
#> GSM97066 3 0.0747 0.957 0.000 0.016 0.984
#> GSM97079 2 0.2651 0.893 0.060 0.928 0.012
#> GSM97083 1 0.1620 0.949 0.964 0.024 0.012
#> GSM97084 2 0.4663 0.788 0.156 0.828 0.016
#> GSM97094 1 0.2550 0.928 0.932 0.056 0.012
#> GSM97096 2 0.1878 0.909 0.004 0.952 0.044
#> GSM97097 2 0.0237 0.928 0.004 0.996 0.000
#> GSM97107 1 0.2446 0.932 0.936 0.052 0.012
#> GSM97054 2 0.4602 0.795 0.152 0.832 0.016
#> GSM97062 2 0.4277 0.815 0.132 0.852 0.016
#> GSM97069 3 0.0747 0.957 0.000 0.016 0.984
#> GSM97070 3 0.0747 0.957 0.000 0.016 0.984
#> GSM97073 3 0.0747 0.957 0.000 0.016 0.984
#> GSM97076 1 0.1765 0.944 0.956 0.040 0.004
#> GSM97077 2 0.0848 0.927 0.008 0.984 0.008
#> GSM97095 1 0.3695 0.869 0.880 0.108 0.012
#> GSM97102 3 0.0747 0.957 0.000 0.016 0.984
#> GSM97109 2 0.1860 0.904 0.052 0.948 0.000
#> GSM97110 2 0.0592 0.929 0.012 0.988 0.000
#> GSM97074 1 0.1525 0.948 0.964 0.032 0.004
#> GSM97085 1 0.2280 0.933 0.940 0.052 0.008
#> GSM97059 2 0.6529 0.447 0.368 0.620 0.012
#> GSM97072 3 0.6095 0.348 0.000 0.392 0.608
#> GSM97078 1 0.2269 0.939 0.944 0.040 0.016
#> GSM97067 3 0.0747 0.957 0.000 0.016 0.984
#> GSM97087 3 0.1289 0.945 0.000 0.032 0.968
#> GSM97111 2 0.0424 0.929 0.008 0.992 0.000
#> GSM97064 2 0.0424 0.929 0.008 0.992 0.000
#> GSM97065 2 0.1964 0.904 0.056 0.944 0.000
#> GSM97081 2 0.1765 0.912 0.004 0.956 0.040
#> GSM97082 3 0.0983 0.954 0.004 0.016 0.980
#> GSM97088 1 0.2414 0.937 0.940 0.040 0.020
#> GSM97100 2 0.0848 0.927 0.008 0.984 0.008
#> GSM97104 3 0.0747 0.957 0.000 0.016 0.984
#> GSM97108 2 0.0424 0.929 0.008 0.992 0.000
#> GSM97050 2 0.0424 0.929 0.008 0.992 0.000
#> GSM97080 3 0.0747 0.957 0.000 0.016 0.984
#> GSM97089 2 0.1585 0.919 0.008 0.964 0.028
#> GSM97092 2 0.1399 0.917 0.004 0.968 0.028
#> GSM97093 2 0.0424 0.929 0.008 0.992 0.000
#> GSM97058 2 0.0424 0.929 0.008 0.992 0.000
#> GSM97051 2 0.0848 0.927 0.008 0.984 0.008
#> GSM97052 2 0.1647 0.913 0.004 0.960 0.036
#> GSM97061 2 0.0237 0.928 0.004 0.996 0.000
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM97138 1 0.3649 0.7235 0.796 0.000 0.000 0.204
#> GSM97145 1 0.3853 0.7427 0.820 0.020 0.000 0.160
#> GSM97147 2 0.7338 -0.3006 0.156 0.440 0.000 0.404
#> GSM97125 1 0.1059 0.7832 0.972 0.012 0.000 0.016
#> GSM97127 1 0.3881 0.7382 0.812 0.016 0.000 0.172
#> GSM97130 1 0.2676 0.7639 0.896 0.092 0.000 0.012
#> GSM97133 1 0.4431 0.6669 0.696 0.000 0.000 0.304
#> GSM97134 1 0.5388 0.4834 0.532 0.456 0.000 0.012
#> GSM97120 1 0.4608 0.6667 0.692 0.004 0.000 0.304
#> GSM97126 1 0.5250 0.5931 0.736 0.068 0.000 0.196
#> GSM97112 1 0.0188 0.7815 0.996 0.000 0.000 0.004
#> GSM97115 2 0.5110 -0.1357 0.328 0.656 0.000 0.016
#> GSM97116 1 0.4431 0.6669 0.696 0.000 0.000 0.304
#> GSM97117 2 0.6393 -0.9442 0.000 0.480 0.064 0.456
#> GSM97119 1 0.0524 0.7831 0.988 0.008 0.000 0.004
#> GSM97122 1 0.0000 0.7813 1.000 0.000 0.000 0.000
#> GSM97135 1 0.0000 0.7813 1.000 0.000 0.000 0.000
#> GSM97136 2 0.8410 -0.6078 0.112 0.420 0.072 0.396
#> GSM97139 1 0.4431 0.6669 0.696 0.000 0.000 0.304
#> GSM97146 1 0.4431 0.6669 0.696 0.000 0.000 0.304
#> GSM97123 2 0.6024 0.1574 0.000 0.540 0.416 0.044
#> GSM97129 2 0.5833 -0.7437 0.000 0.528 0.032 0.440
#> GSM97143 1 0.0469 0.7829 0.988 0.012 0.000 0.000
#> GSM97113 4 0.6395 0.9929 0.000 0.460 0.064 0.476
#> GSM97056 1 0.2342 0.7703 0.912 0.080 0.000 0.008
#> GSM97124 1 0.1854 0.7790 0.940 0.048 0.000 0.012
#> GSM97132 1 0.1109 0.7825 0.968 0.028 0.000 0.004
#> GSM97144 1 0.5310 0.5275 0.576 0.412 0.000 0.012
#> GSM97149 1 0.4608 0.6667 0.692 0.004 0.000 0.304
#> GSM97068 2 0.5228 -0.2558 0.024 0.664 0.000 0.312
#> GSM97071 2 0.4453 0.1043 0.244 0.744 0.000 0.012
#> GSM97086 2 0.2704 0.1878 0.000 0.876 0.000 0.124
#> GSM97103 2 0.7277 -0.3610 0.000 0.540 0.228 0.232
#> GSM97057 2 0.4972 -0.6095 0.000 0.544 0.000 0.456
#> GSM97060 2 0.6242 0.1633 0.000 0.520 0.424 0.056
#> GSM97075 2 0.6340 -0.8099 0.000 0.528 0.064 0.408
#> GSM97098 2 0.5158 0.1289 0.000 0.524 0.472 0.004
#> GSM97099 2 0.6395 -0.9542 0.000 0.476 0.064 0.460
#> GSM97101 2 0.5933 -0.8255 0.000 0.500 0.036 0.464
#> GSM97105 2 0.4955 -0.6151 0.000 0.556 0.000 0.444
#> GSM97106 2 0.6206 0.1602 0.000 0.540 0.404 0.056
#> GSM97121 2 0.4977 -0.6521 0.000 0.540 0.000 0.460
#> GSM97128 1 0.5512 0.4444 0.492 0.492 0.000 0.016
#> GSM97131 2 0.6732 0.0984 0.000 0.556 0.336 0.108
#> GSM97137 1 0.2342 0.7703 0.912 0.080 0.000 0.008
#> GSM97118 1 0.1284 0.7831 0.964 0.024 0.000 0.012
#> GSM97114 4 0.6560 0.9814 0.004 0.456 0.064 0.476
#> GSM97142 1 0.0188 0.7815 0.996 0.000 0.000 0.004
#> GSM97140 2 0.4981 -0.6172 0.000 0.536 0.000 0.464
#> GSM97141 4 0.6395 0.9850 0.000 0.464 0.064 0.472
#> GSM97055 1 0.1305 0.7831 0.960 0.036 0.000 0.004
#> GSM97090 2 0.5268 -0.2913 0.396 0.592 0.000 0.012
#> GSM97091 1 0.0188 0.7815 0.996 0.000 0.000 0.004
#> GSM97148 1 0.4431 0.6669 0.696 0.000 0.000 0.304
#> GSM97063 1 0.0188 0.7815 0.996 0.000 0.000 0.004
#> GSM97053 1 0.1118 0.7826 0.964 0.036 0.000 0.000
#> GSM97066 3 0.0188 0.9731 0.000 0.000 0.996 0.004
#> GSM97079 2 0.3074 0.1786 0.000 0.848 0.000 0.152
#> GSM97083 1 0.5508 0.4609 0.508 0.476 0.000 0.016
#> GSM97084 2 0.3208 0.2010 0.004 0.848 0.000 0.148
#> GSM97094 2 0.5279 -0.2993 0.400 0.588 0.000 0.012
#> GSM97096 2 0.4996 0.1311 0.000 0.516 0.484 0.000
#> GSM97097 2 0.5392 0.1288 0.000 0.724 0.204 0.072
#> GSM97107 2 0.5268 -0.2867 0.396 0.592 0.000 0.012
#> GSM97054 2 0.3208 0.2010 0.004 0.848 0.000 0.148
#> GSM97062 2 0.3208 0.2010 0.004 0.848 0.000 0.148
#> GSM97069 3 0.0000 0.9761 0.000 0.000 1.000 0.000
#> GSM97070 3 0.0000 0.9761 0.000 0.000 1.000 0.000
#> GSM97073 3 0.0000 0.9761 0.000 0.000 1.000 0.000
#> GSM97076 1 0.7921 0.0812 0.464 0.120 0.036 0.380
#> GSM97077 2 0.4941 -0.6062 0.000 0.564 0.000 0.436
#> GSM97095 2 0.4936 -0.1559 0.340 0.652 0.000 0.008
#> GSM97102 3 0.0000 0.9761 0.000 0.000 1.000 0.000
#> GSM97109 4 0.6395 0.9929 0.000 0.460 0.064 0.476
#> GSM97110 4 0.6395 0.9929 0.000 0.460 0.064 0.476
#> GSM97074 1 0.5236 0.5035 0.560 0.432 0.000 0.008
#> GSM97085 1 0.6462 0.1674 0.520 0.060 0.416 0.004
#> GSM97059 2 0.6096 0.1053 0.136 0.680 0.000 0.184
#> GSM97072 3 0.2973 0.7654 0.000 0.144 0.856 0.000
#> GSM97078 1 0.5512 0.4444 0.492 0.492 0.000 0.016
#> GSM97067 3 0.0000 0.9761 0.000 0.000 1.000 0.000
#> GSM97087 3 0.0592 0.9598 0.000 0.016 0.984 0.000
#> GSM97111 2 0.6395 -0.9542 0.000 0.476 0.064 0.460
#> GSM97064 2 0.6337 0.1485 0.000 0.552 0.380 0.068
#> GSM97065 2 0.6395 -0.9542 0.000 0.476 0.064 0.460
#> GSM97081 2 0.5161 0.1300 0.000 0.520 0.476 0.004
#> GSM97082 3 0.0188 0.9733 0.000 0.004 0.996 0.000
#> GSM97088 1 0.5512 0.4444 0.492 0.492 0.000 0.016
#> GSM97100 2 0.4916 -0.5245 0.000 0.576 0.000 0.424
#> GSM97104 3 0.0000 0.9761 0.000 0.000 1.000 0.000
#> GSM97108 2 0.4972 -0.6422 0.000 0.544 0.000 0.456
#> GSM97050 2 0.5543 -0.5929 0.000 0.556 0.020 0.424
#> GSM97080 3 0.0000 0.9761 0.000 0.000 1.000 0.000
#> GSM97089 2 0.5392 0.1225 0.000 0.528 0.460 0.012
#> GSM97092 2 0.6031 0.1587 0.000 0.536 0.420 0.044
#> GSM97093 2 0.4955 -0.6151 0.000 0.556 0.000 0.444
#> GSM97058 2 0.6928 -0.3561 0.000 0.556 0.136 0.308
#> GSM97051 2 0.3758 0.1619 0.000 0.848 0.048 0.104
#> GSM97052 2 0.6222 0.1629 0.000 0.532 0.412 0.056
#> GSM97061 2 0.6206 0.1602 0.000 0.540 0.404 0.056
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM97138 1 0.2424 0.7657 0.868 0.000 0.000 0.000 0.132
#> GSM97145 1 0.2865 0.7570 0.856 0.004 0.000 0.008 0.132
#> GSM97147 2 0.5778 0.3684 0.000 0.528 0.000 0.376 0.096
#> GSM97125 1 0.4549 -0.0954 0.528 0.000 0.000 0.008 0.464
#> GSM97127 1 0.2843 0.7503 0.848 0.000 0.000 0.008 0.144
#> GSM97130 5 0.4757 0.6013 0.120 0.000 0.000 0.148 0.732
#> GSM97133 1 0.0000 0.8655 1.000 0.000 0.000 0.000 0.000
#> GSM97134 5 0.3845 0.5990 0.024 0.000 0.000 0.208 0.768
#> GSM97120 1 0.0000 0.8655 1.000 0.000 0.000 0.000 0.000
#> GSM97126 5 0.6814 0.2850 0.052 0.392 0.000 0.092 0.464
#> GSM97112 5 0.5534 0.1272 0.424 0.000 0.000 0.068 0.508
#> GSM97115 5 0.4235 0.2460 0.000 0.000 0.000 0.424 0.576
#> GSM97116 1 0.0510 0.8596 0.984 0.000 0.000 0.000 0.016
#> GSM97117 2 0.1768 0.6601 0.000 0.924 0.004 0.072 0.000
#> GSM97119 5 0.3884 0.4385 0.288 0.000 0.000 0.004 0.708
#> GSM97122 5 0.5431 0.1400 0.424 0.000 0.000 0.060 0.516
#> GSM97135 5 0.5431 0.1400 0.424 0.000 0.000 0.060 0.516
#> GSM97136 2 0.4547 0.4140 0.000 0.712 0.012 0.024 0.252
#> GSM97139 1 0.0162 0.8650 0.996 0.000 0.000 0.000 0.004
#> GSM97146 1 0.0000 0.8655 1.000 0.000 0.000 0.000 0.000
#> GSM97123 3 0.6690 0.2984 0.000 0.232 0.468 0.296 0.004
#> GSM97129 2 0.4268 0.6485 0.000 0.708 0.000 0.268 0.024
#> GSM97143 5 0.4211 0.3424 0.360 0.000 0.000 0.004 0.636
#> GSM97113 2 0.0162 0.6202 0.000 0.996 0.004 0.000 0.000
#> GSM97056 5 0.5093 0.5848 0.180 0.000 0.000 0.124 0.696
#> GSM97124 5 0.4647 0.5835 0.184 0.000 0.000 0.084 0.732
#> GSM97132 5 0.4541 0.5874 0.172 0.000 0.000 0.084 0.744
#> GSM97144 5 0.3690 0.5886 0.012 0.000 0.000 0.224 0.764
#> GSM97149 1 0.0000 0.8655 1.000 0.000 0.000 0.000 0.000
#> GSM97068 4 0.6036 -0.1529 0.000 0.432 0.000 0.452 0.116
#> GSM97071 4 0.3730 0.1113 0.000 0.000 0.000 0.712 0.288
#> GSM97086 4 0.2408 0.6333 0.000 0.092 0.000 0.892 0.016
#> GSM97103 2 0.6616 0.1583 0.000 0.456 0.292 0.252 0.000
#> GSM97057 2 0.3796 0.6439 0.000 0.700 0.000 0.300 0.000
#> GSM97060 3 0.5905 0.4123 0.000 0.136 0.572 0.292 0.000
#> GSM97075 2 0.3607 0.6631 0.000 0.752 0.004 0.244 0.000
#> GSM97098 3 0.6326 0.3925 0.000 0.248 0.528 0.224 0.000
#> GSM97099 2 0.1571 0.6556 0.000 0.936 0.004 0.060 0.000
#> GSM97101 2 0.3635 0.6641 0.000 0.748 0.004 0.248 0.000
#> GSM97105 2 0.4066 0.6199 0.000 0.672 0.004 0.324 0.000
#> GSM97106 3 0.6666 0.3041 0.000 0.224 0.472 0.300 0.004
#> GSM97121 2 0.3796 0.6437 0.000 0.700 0.000 0.300 0.000
#> GSM97128 5 0.4201 0.4917 0.000 0.000 0.000 0.408 0.592
#> GSM97131 4 0.6820 -0.0981 0.000 0.240 0.348 0.408 0.004
#> GSM97137 5 0.5201 0.5809 0.188 0.000 0.000 0.128 0.684
#> GSM97118 5 0.2423 0.5947 0.080 0.000 0.000 0.024 0.896
#> GSM97114 2 0.0162 0.6202 0.000 0.996 0.004 0.000 0.000
#> GSM97142 5 0.5534 0.1272 0.424 0.000 0.000 0.068 0.508
#> GSM97140 2 0.4067 0.6374 0.000 0.692 0.000 0.300 0.008
#> GSM97141 2 0.0324 0.6234 0.000 0.992 0.004 0.004 0.000
#> GSM97055 5 0.3646 0.5596 0.120 0.044 0.000 0.008 0.828
#> GSM97090 5 0.3774 0.5376 0.000 0.000 0.000 0.296 0.704
#> GSM97091 5 0.4679 0.4471 0.216 0.000 0.000 0.068 0.716
#> GSM97148 1 0.0000 0.8655 1.000 0.000 0.000 0.000 0.000
#> GSM97063 5 0.5182 0.3427 0.300 0.000 0.000 0.068 0.632
#> GSM97053 5 0.4793 0.5593 0.216 0.000 0.000 0.076 0.708
#> GSM97066 3 0.0000 0.6321 0.000 0.000 1.000 0.000 0.000
#> GSM97079 4 0.2179 0.6201 0.000 0.112 0.000 0.888 0.000
#> GSM97083 5 0.4380 0.5183 0.008 0.000 0.000 0.376 0.616
#> GSM97084 4 0.2193 0.6353 0.000 0.060 0.000 0.912 0.028
#> GSM97094 5 0.4256 0.4480 0.000 0.000 0.000 0.436 0.564
#> GSM97096 3 0.5946 0.4570 0.000 0.184 0.592 0.224 0.000
#> GSM97097 4 0.5629 0.3775 0.000 0.220 0.132 0.644 0.004
#> GSM97107 5 0.4420 0.3547 0.004 0.000 0.000 0.448 0.548
#> GSM97054 4 0.2221 0.6300 0.000 0.052 0.000 0.912 0.036
#> GSM97062 4 0.2104 0.6356 0.000 0.060 0.000 0.916 0.024
#> GSM97069 3 0.0000 0.6321 0.000 0.000 1.000 0.000 0.000
#> GSM97070 3 0.0000 0.6321 0.000 0.000 1.000 0.000 0.000
#> GSM97073 3 0.0000 0.6321 0.000 0.000 1.000 0.000 0.000
#> GSM97076 2 0.6363 -0.0162 0.024 0.512 0.004 0.080 0.380
#> GSM97077 2 0.3913 0.6243 0.000 0.676 0.000 0.324 0.000
#> GSM97095 5 0.4201 0.3615 0.000 0.000 0.000 0.408 0.592
#> GSM97102 3 0.0000 0.6321 0.000 0.000 1.000 0.000 0.000
#> GSM97109 2 0.0162 0.6202 0.000 0.996 0.004 0.000 0.000
#> GSM97110 2 0.0451 0.6261 0.000 0.988 0.004 0.008 0.000
#> GSM97074 5 0.4470 0.5229 0.012 0.000 0.000 0.372 0.616
#> GSM97085 3 0.5713 -0.0293 0.000 0.000 0.500 0.084 0.416
#> GSM97059 4 0.6458 0.3328 0.000 0.216 0.000 0.492 0.292
#> GSM97072 3 0.1557 0.6184 0.000 0.052 0.940 0.008 0.000
#> GSM97078 5 0.4201 0.4917 0.000 0.000 0.000 0.408 0.592
#> GSM97067 3 0.0000 0.6321 0.000 0.000 1.000 0.000 0.000
#> GSM97087 3 0.0451 0.6305 0.000 0.004 0.988 0.008 0.000
#> GSM97111 2 0.2068 0.6637 0.000 0.904 0.004 0.092 0.000
#> GSM97064 3 0.6653 0.2110 0.000 0.240 0.432 0.328 0.000
#> GSM97065 2 0.1704 0.6589 0.000 0.928 0.004 0.068 0.000
#> GSM97081 3 0.6010 0.4550 0.000 0.204 0.584 0.212 0.000
#> GSM97082 3 0.0000 0.6321 0.000 0.000 1.000 0.000 0.000
#> GSM97088 5 0.4235 0.4820 0.000 0.000 0.000 0.424 0.576
#> GSM97100 2 0.4448 0.3379 0.000 0.516 0.000 0.480 0.004
#> GSM97104 3 0.0000 0.6321 0.000 0.000 1.000 0.000 0.000
#> GSM97108 2 0.3816 0.6406 0.000 0.696 0.000 0.304 0.000
#> GSM97050 2 0.5551 0.2797 0.000 0.488 0.068 0.444 0.000
#> GSM97080 3 0.0000 0.6321 0.000 0.000 1.000 0.000 0.000
#> GSM97089 3 0.6441 0.4086 0.000 0.196 0.544 0.252 0.008
#> GSM97092 3 0.6392 0.3992 0.000 0.192 0.532 0.272 0.004
#> GSM97093 2 0.3816 0.6406 0.000 0.696 0.000 0.304 0.000
#> GSM97058 4 0.6784 -0.0165 0.000 0.352 0.280 0.368 0.000
#> GSM97051 4 0.3155 0.6061 0.000 0.128 0.016 0.848 0.008
#> GSM97052 3 0.6295 0.4247 0.000 0.188 0.552 0.256 0.004
#> GSM97061 3 0.6701 0.2908 0.000 0.232 0.464 0.300 0.004
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM97138 1 0.1498 0.9185 0.940 0.000 0.000 0.028 0.032 0.000
#> GSM97145 1 0.2112 0.8547 0.896 0.000 0.000 0.088 0.016 0.000
#> GSM97147 3 0.6028 0.4105 0.004 0.308 0.484 0.200 0.004 0.000
#> GSM97125 5 0.5585 0.4840 0.364 0.000 0.000 0.148 0.488 0.000
#> GSM97127 1 0.2237 0.8695 0.896 0.000 0.000 0.068 0.036 0.000
#> GSM97130 4 0.0865 0.7884 0.000 0.000 0.000 0.964 0.036 0.000
#> GSM97133 1 0.0000 0.9589 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM97134 4 0.0363 0.7977 0.000 0.000 0.000 0.988 0.012 0.000
#> GSM97120 1 0.0000 0.9589 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM97126 5 0.6990 0.4900 0.068 0.248 0.008 0.212 0.464 0.000
#> GSM97112 5 0.1633 0.7516 0.044 0.000 0.000 0.024 0.932 0.000
#> GSM97115 4 0.1531 0.7809 0.000 0.000 0.068 0.928 0.004 0.000
#> GSM97116 1 0.0146 0.9569 0.996 0.000 0.000 0.000 0.004 0.000
#> GSM97117 2 0.1858 0.7361 0.000 0.904 0.092 0.000 0.000 0.004
#> GSM97119 5 0.4024 0.7635 0.072 0.000 0.000 0.184 0.744 0.000
#> GSM97122 5 0.1789 0.7573 0.044 0.000 0.000 0.032 0.924 0.000
#> GSM97135 5 0.1789 0.7573 0.044 0.000 0.000 0.032 0.924 0.000
#> GSM97136 2 0.5001 0.5514 0.000 0.700 0.024 0.200 0.016 0.060
#> GSM97139 1 0.0000 0.9589 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM97146 1 0.0000 0.9589 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM97123 3 0.4911 0.5959 0.000 0.196 0.688 0.000 0.020 0.096
#> GSM97129 2 0.5200 0.3538 0.000 0.620 0.276 0.088 0.000 0.016
#> GSM97143 5 0.3927 0.7664 0.072 0.000 0.000 0.172 0.756 0.000
#> GSM97113 2 0.0291 0.7341 0.000 0.992 0.004 0.004 0.000 0.000
#> GSM97056 4 0.2009 0.7507 0.024 0.000 0.000 0.908 0.068 0.000
#> GSM97124 5 0.5398 0.6300 0.136 0.000 0.000 0.320 0.544 0.000
#> GSM97132 5 0.4774 0.5277 0.052 0.000 0.000 0.420 0.528 0.000
#> GSM97144 4 0.0363 0.7977 0.000 0.000 0.000 0.988 0.012 0.000
#> GSM97149 1 0.0000 0.9589 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM97068 3 0.4764 0.5796 0.000 0.108 0.660 0.232 0.000 0.000
#> GSM97071 4 0.2837 0.7759 0.000 0.000 0.088 0.856 0.056 0.000
#> GSM97086 3 0.3986 -0.0908 0.000 0.004 0.532 0.464 0.000 0.000
#> GSM97103 3 0.5867 0.3578 0.000 0.344 0.516 0.004 0.016 0.120
#> GSM97057 3 0.4781 0.5623 0.000 0.320 0.608 0.072 0.000 0.000
#> GSM97060 6 0.4962 0.6193 0.000 0.032 0.236 0.028 0.020 0.684
#> GSM97075 2 0.3093 0.6698 0.000 0.816 0.164 0.012 0.008 0.000
#> GSM97098 2 0.4967 0.5641 0.000 0.688 0.196 0.004 0.016 0.096
#> GSM97099 2 0.1501 0.7393 0.000 0.924 0.076 0.000 0.000 0.000
#> GSM97101 2 0.2805 0.6784 0.000 0.828 0.160 0.012 0.000 0.000
#> GSM97105 3 0.3409 0.6816 0.000 0.192 0.780 0.028 0.000 0.000
#> GSM97106 3 0.3956 0.6390 0.000 0.072 0.792 0.000 0.024 0.112
#> GSM97121 2 0.4469 -0.1641 0.000 0.504 0.468 0.028 0.000 0.000
#> GSM97128 4 0.2728 0.7558 0.000 0.000 0.040 0.860 0.100 0.000
#> GSM97131 3 0.2153 0.6932 0.000 0.084 0.900 0.008 0.004 0.004
#> GSM97137 4 0.4393 0.0902 0.044 0.000 0.000 0.640 0.316 0.000
#> GSM97118 5 0.3817 0.7283 0.028 0.000 0.000 0.252 0.720 0.000
#> GSM97114 2 0.0146 0.7326 0.000 0.996 0.004 0.000 0.000 0.000
#> GSM97142 5 0.1633 0.7516 0.044 0.000 0.000 0.024 0.932 0.000
#> GSM97140 3 0.4569 0.5851 0.000 0.304 0.636 0.060 0.000 0.000
#> GSM97141 2 0.0260 0.7349 0.000 0.992 0.008 0.000 0.000 0.000
#> GSM97055 5 0.3900 0.7588 0.044 0.008 0.000 0.188 0.760 0.000
#> GSM97090 4 0.0405 0.7991 0.000 0.000 0.004 0.988 0.008 0.000
#> GSM97091 5 0.1418 0.7541 0.024 0.000 0.000 0.032 0.944 0.000
#> GSM97148 1 0.0000 0.9589 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM97063 5 0.1418 0.7520 0.032 0.000 0.000 0.024 0.944 0.000
#> GSM97053 5 0.4756 0.5612 0.052 0.000 0.000 0.408 0.540 0.000
#> GSM97066 6 0.0000 0.8537 0.000 0.000 0.000 0.000 0.000 1.000
#> GSM97079 3 0.3872 0.1549 0.000 0.004 0.604 0.392 0.000 0.000
#> GSM97083 4 0.2822 0.7543 0.000 0.000 0.040 0.852 0.108 0.000
#> GSM97084 4 0.4080 0.2496 0.000 0.008 0.456 0.536 0.000 0.000
#> GSM97094 4 0.0909 0.8018 0.000 0.000 0.020 0.968 0.012 0.000
#> GSM97096 6 0.5467 0.5656 0.000 0.156 0.196 0.000 0.020 0.628
#> GSM97097 3 0.2094 0.6865 0.000 0.064 0.908 0.024 0.004 0.000
#> GSM97107 4 0.0692 0.7991 0.000 0.000 0.020 0.976 0.004 0.000
#> GSM97054 4 0.4018 0.3743 0.000 0.008 0.412 0.580 0.000 0.000
#> GSM97062 4 0.4083 0.2369 0.000 0.008 0.460 0.532 0.000 0.000
#> GSM97069 6 0.0000 0.8537 0.000 0.000 0.000 0.000 0.000 1.000
#> GSM97070 6 0.0000 0.8537 0.000 0.000 0.000 0.000 0.000 1.000
#> GSM97073 6 0.0000 0.8537 0.000 0.000 0.000 0.000 0.000 1.000
#> GSM97076 2 0.7161 -0.2634 0.040 0.376 0.004 0.216 0.348 0.016
#> GSM97077 3 0.3745 0.6484 0.000 0.240 0.732 0.028 0.000 0.000
#> GSM97095 4 0.1138 0.7964 0.000 0.012 0.024 0.960 0.004 0.000
#> GSM97102 6 0.0146 0.8534 0.000 0.000 0.004 0.000 0.000 0.996
#> GSM97109 2 0.0146 0.7326 0.000 0.996 0.004 0.000 0.000 0.000
#> GSM97110 2 0.0363 0.7361 0.000 0.988 0.012 0.000 0.000 0.000
#> GSM97074 5 0.4172 0.6425 0.000 0.000 0.040 0.280 0.680 0.000
#> GSM97085 6 0.5868 0.4143 0.000 0.000 0.040 0.228 0.140 0.592
#> GSM97059 3 0.4863 0.5319 0.000 0.092 0.624 0.284 0.000 0.000
#> GSM97072 6 0.2020 0.8258 0.000 0.020 0.040 0.000 0.020 0.920
#> GSM97078 4 0.2679 0.7580 0.000 0.000 0.040 0.864 0.096 0.000
#> GSM97067 6 0.0000 0.8537 0.000 0.000 0.000 0.000 0.000 1.000
#> GSM97087 6 0.0146 0.8531 0.000 0.004 0.000 0.000 0.000 0.996
#> GSM97111 2 0.2257 0.7203 0.000 0.876 0.116 0.008 0.000 0.000
#> GSM97064 3 0.3406 0.6985 0.000 0.100 0.832 0.024 0.000 0.044
#> GSM97065 2 0.2169 0.7407 0.000 0.900 0.080 0.008 0.000 0.012
#> GSM97081 2 0.6028 0.1918 0.000 0.464 0.156 0.000 0.016 0.364
#> GSM97082 6 0.0146 0.8530 0.000 0.000 0.004 0.000 0.000 0.996
#> GSM97088 4 0.3491 0.7389 0.000 0.000 0.040 0.828 0.100 0.032
#> GSM97100 3 0.2384 0.6876 0.000 0.048 0.888 0.064 0.000 0.000
#> GSM97104 6 0.0000 0.8537 0.000 0.000 0.000 0.000 0.000 1.000
#> GSM97108 3 0.4371 0.4246 0.000 0.392 0.580 0.028 0.000 0.000
#> GSM97050 3 0.2554 0.7004 0.000 0.092 0.876 0.028 0.000 0.004
#> GSM97080 6 0.0000 0.8537 0.000 0.000 0.000 0.000 0.000 1.000
#> GSM97089 6 0.4322 0.7067 0.000 0.020 0.152 0.040 0.020 0.768
#> GSM97092 6 0.4849 0.5747 0.000 0.044 0.288 0.000 0.024 0.644
#> GSM97093 3 0.4203 0.5620 0.000 0.316 0.652 0.032 0.000 0.000
#> GSM97058 3 0.3353 0.6945 0.000 0.156 0.808 0.028 0.000 0.008
#> GSM97051 3 0.3217 0.5424 0.000 0.008 0.768 0.224 0.000 0.000
#> GSM97052 6 0.4849 0.5754 0.000 0.044 0.288 0.000 0.024 0.644
#> GSM97061 3 0.3977 0.6476 0.000 0.084 0.792 0.000 0.024 0.100
Heatmaps for the consensus matrix. It visualizes the probability of two samples to be in a same group.
consensus_heatmap(res, k = 2)
consensus_heatmap(res, k = 3)
consensus_heatmap(res, k = 4)
consensus_heatmap(res, k = 5)
consensus_heatmap(res, k = 6)
Heatmaps for the membership of samples in all partitions to see how consistent they are:
membership_heatmap(res, k = 2)
membership_heatmap(res, k = 3)
membership_heatmap(res, k = 4)
membership_heatmap(res, k = 5)
membership_heatmap(res, k = 6)
As soon as we have had the classes for columns, we can look for signatures which are significantly different between classes which can be candidate marks for certain classes. Following are the heatmaps for signatures.
Signature heatmaps where rows are scaled:
get_signatures(res, k = 2)
get_signatures(res, k = 3)
get_signatures(res, k = 4)
get_signatures(res, k = 5)
get_signatures(res, k = 6)
Signature heatmaps where rows are not scaled:
get_signatures(res, k = 2, scale_rows = FALSE)
get_signatures(res, k = 3, scale_rows = FALSE)
get_signatures(res, k = 4, scale_rows = FALSE)
get_signatures(res, k = 5, scale_rows = FALSE)
get_signatures(res, k = 6, scale_rows = FALSE)
Compare the overlap of signatures from different k:
compare_signatures(res)
get_signature() returns a data frame invisibly. TO get the list of signatures, the function
call should be assigned to a variable explicitly. In following code, if plot argument is set
to FALSE, no heatmap is plotted while only the differential analysis is performed.
# code only for demonstration
tb = get_signature(res, k = ..., plot = FALSE)
An example of the output of tb is:
#> which_row fdr mean_1 mean_2 scaled_mean_1 scaled_mean_2 km
#> 1 38 0.042760348 8.373488 9.131774 -0.5533452 0.5164555 1
#> 2 40 0.018707592 7.106213 8.469186 -0.6173731 0.5762149 1
#> 3 55 0.019134737 10.221463 11.207825 -0.6159697 0.5749050 1
#> 4 59 0.006059896 5.921854 7.869574 -0.6899429 0.6439467 1
#> 5 60 0.018055526 8.928898 10.211722 -0.6204761 0.5791110 1
#> 6 98 0.009384629 15.714769 14.887706 0.6635654 -0.6193277 2
...
The columns in tb are:
which_row: row indices corresponding to the input matrix.fdr: FDR for the differential test. mean_x: The mean value in group x.scaled_mean_x: The mean value in group x after rows are scaled.km: Row groups if k-means clustering is applied to rows.UMAP plot which shows how samples are separated.
dimension_reduction(res, k = 2, method = "UMAP")
dimension_reduction(res, k = 3, method = "UMAP")
dimension_reduction(res, k = 4, method = "UMAP")
dimension_reduction(res, k = 5, method = "UMAP")
dimension_reduction(res, k = 6, method = "UMAP")
Following heatmap shows how subgroups are split when increasing k:
collect_classes(res)
Test correlation between subgroups and known annotations. If the known annotation is numeric, one-way ANOVA test is applied, and if the known annotation is discrete, chi-squared contingency table test is applied.
test_to_known_factors(res)
#> n disease.state(p) specimen(p) cell.type(p) other(p) k
#> CV:mclust 84 2.21e-02 0.1703 2.41e-10 0.4651 2
#> CV:mclust 97 2.50e-04 0.0331 1.46e-11 0.0353 3
#> CV:mclust 46 1.52e-04 0.2732 8.96e-08 0.1054 4
#> CV:mclust 59 8.03e-05 0.0740 4.15e-09 0.2242 5
#> CV:mclust 84 2.66e-04 0.3095 4.48e-10 0.4903 6
If matrix rows can be associated to genes, consider to use functional_enrichment(res,
...) to perform function enrichment for the signature genes. See this vignette for more detailed explanations.
The object with results only for a single top-value method and a single partition method can be extracted as:
res = res_list["CV", "NMF"]
# you can also extract it by
# res = res_list["CV:NMF"]
A summary of res and all the functions that can be applied to it:
res
#> A 'ConsensusPartition' object with k = 2, 3, 4, 5, 6.
#> On a matrix with 21512 rows and 100 columns.
#> Top rows (1000, 2000, 3000, 4000, 5000) are extracted by 'CV' method.
#> Subgroups are detected by 'NMF' method.
#> Performed in total 1250 partitions by row resampling.
#> Best k for subgroups seems to be 2.
#>
#> Following methods can be applied to this 'ConsensusPartition' object:
#> [1] "cola_report" "collect_classes" "collect_plots"
#> [4] "collect_stats" "colnames" "compare_signatures"
#> [7] "consensus_heatmap" "dimension_reduction" "functional_enrichment"
#> [10] "get_anno_col" "get_anno" "get_classes"
#> [13] "get_consensus" "get_matrix" "get_membership"
#> [16] "get_param" "get_signatures" "get_stats"
#> [19] "is_best_k" "is_stable_k" "membership_heatmap"
#> [22] "ncol" "nrow" "plot_ecdf"
#> [25] "rownames" "select_partition_number" "show"
#> [28] "suggest_best_k" "test_to_known_factors"
collect_plots() function collects all the plots made from res for all k (number of partitions)
into one single page to provide an easy and fast comparison between different k.
collect_plots(res)
The plots are:
k and the heatmap of
predicted classes for each k.k.k.k.All the plots in panels can be made by individual functions and they are plotted later in this section.
select_partition_number() produces several plots showing different
statistics for choosing “optimized” k. There are following statistics:
k;k, the area increased is defined as \(A_k - A_{k-1}\).The detailed explanations of these statistics can be found in the cola vignette.
Generally speaking, lower PAC score, higher mean silhouette score or higher
concordance corresponds to better partition. Rand index and Jaccard index
measure how similar the current partition is compared to partition with k-1.
If they are too similar, we won't accept k is better than k-1.
select_partition_number(res)
The numeric values for all these statistics can be obtained by get_stats().
get_stats(res)
#> k 1-PAC mean_silhouette concordance area_increased Rand Jaccard
#> 2 2 0.935 0.944 0.975 0.4958 0.508 0.508
#> 3 3 0.470 0.630 0.806 0.3180 0.784 0.595
#> 4 4 0.531 0.507 0.763 0.1344 0.766 0.446
#> 5 5 0.538 0.464 0.701 0.0765 0.826 0.450
#> 6 6 0.618 0.481 0.690 0.0446 0.892 0.535
suggest_best_k() suggests the best \(k\) based on these statistics. The rules are as follows:
suggest_best_k(res)
#> [1] 2
Following shows the table of the partitions (You need to click the show/hide
code output link to see it). The membership matrix (columns with name p*)
is inferred by
clue::cl_consensus()
function with the SE method. Basically the value in the membership matrix
represents the probability to belong to a certain group. The finall class
label for an item is determined with the group with highest probability it
belongs to.
In get_classes() function, the entropy is calculated from the membership
matrix and the silhouette score is calculated from the consensus matrix.
cbind(get_classes(res, k = 2), get_membership(res, k = 2))
#> class entropy silhouette p1 p2
#> GSM97138 1 0.0000 0.987 1.000 0.000
#> GSM97145 1 0.0000 0.987 1.000 0.000
#> GSM97147 1 0.0000 0.987 1.000 0.000
#> GSM97125 1 0.0000 0.987 1.000 0.000
#> GSM97127 1 0.0000 0.987 1.000 0.000
#> GSM97130 1 0.0000 0.987 1.000 0.000
#> GSM97133 1 0.0000 0.987 1.000 0.000
#> GSM97134 1 0.0000 0.987 1.000 0.000
#> GSM97120 1 0.0000 0.987 1.000 0.000
#> GSM97126 1 0.0000 0.987 1.000 0.000
#> GSM97112 1 0.0000 0.987 1.000 0.000
#> GSM97115 1 0.0000 0.987 1.000 0.000
#> GSM97116 1 0.0000 0.987 1.000 0.000
#> GSM97117 2 0.0672 0.959 0.008 0.992
#> GSM97119 1 0.0000 0.987 1.000 0.000
#> GSM97122 1 0.0000 0.987 1.000 0.000
#> GSM97135 1 0.0000 0.987 1.000 0.000
#> GSM97136 2 0.2948 0.924 0.052 0.948
#> GSM97139 1 0.0000 0.987 1.000 0.000
#> GSM97146 1 0.0000 0.987 1.000 0.000
#> GSM97123 2 0.0000 0.964 0.000 1.000
#> GSM97129 2 0.6887 0.780 0.184 0.816
#> GSM97143 1 0.0000 0.987 1.000 0.000
#> GSM97113 2 0.7950 0.703 0.240 0.760
#> GSM97056 1 0.0000 0.987 1.000 0.000
#> GSM97124 1 0.0000 0.987 1.000 0.000
#> GSM97132 1 0.0000 0.987 1.000 0.000
#> GSM97144 1 0.0000 0.987 1.000 0.000
#> GSM97149 1 0.0000 0.987 1.000 0.000
#> GSM97068 1 0.3114 0.937 0.944 0.056
#> GSM97071 2 0.0000 0.964 0.000 1.000
#> GSM97086 2 0.0000 0.964 0.000 1.000
#> GSM97103 2 0.0000 0.964 0.000 1.000
#> GSM97057 1 0.5519 0.852 0.872 0.128
#> GSM97060 2 0.0000 0.964 0.000 1.000
#> GSM97075 2 0.0000 0.964 0.000 1.000
#> GSM97098 2 0.0000 0.964 0.000 1.000
#> GSM97099 2 0.0000 0.964 0.000 1.000
#> GSM97101 2 0.0672 0.959 0.008 0.992
#> GSM97105 2 0.0000 0.964 0.000 1.000
#> GSM97106 2 0.0000 0.964 0.000 1.000
#> GSM97121 2 0.0000 0.964 0.000 1.000
#> GSM97128 2 0.9460 0.468 0.364 0.636
#> GSM97131 2 0.0000 0.964 0.000 1.000
#> GSM97137 1 0.0000 0.987 1.000 0.000
#> GSM97118 1 0.0000 0.987 1.000 0.000
#> GSM97114 1 0.3879 0.916 0.924 0.076
#> GSM97142 1 0.0000 0.987 1.000 0.000
#> GSM97140 2 0.8909 0.586 0.308 0.692
#> GSM97141 2 0.1414 0.950 0.020 0.980
#> GSM97055 1 0.0000 0.987 1.000 0.000
#> GSM97090 1 0.0000 0.987 1.000 0.000
#> GSM97091 1 0.0000 0.987 1.000 0.000
#> GSM97148 1 0.0000 0.987 1.000 0.000
#> GSM97063 1 0.0000 0.987 1.000 0.000
#> GSM97053 1 0.0000 0.987 1.000 0.000
#> GSM97066 2 0.0000 0.964 0.000 1.000
#> GSM97079 2 0.0000 0.964 0.000 1.000
#> GSM97083 1 0.0000 0.987 1.000 0.000
#> GSM97084 2 0.0000 0.964 0.000 1.000
#> GSM97094 1 0.1414 0.972 0.980 0.020
#> GSM97096 2 0.0000 0.964 0.000 1.000
#> GSM97097 2 0.0000 0.964 0.000 1.000
#> GSM97107 1 0.1184 0.975 0.984 0.016
#> GSM97054 2 0.2778 0.927 0.048 0.952
#> GSM97062 2 0.0000 0.964 0.000 1.000
#> GSM97069 2 0.0000 0.964 0.000 1.000
#> GSM97070 2 0.0000 0.964 0.000 1.000
#> GSM97073 2 0.0000 0.964 0.000 1.000
#> GSM97076 1 0.6623 0.790 0.828 0.172
#> GSM97077 2 0.0000 0.964 0.000 1.000
#> GSM97095 1 0.1633 0.968 0.976 0.024
#> GSM97102 2 0.0000 0.964 0.000 1.000
#> GSM97109 2 0.9552 0.428 0.376 0.624
#> GSM97110 2 0.0376 0.961 0.004 0.996
#> GSM97074 2 0.9087 0.557 0.324 0.676
#> GSM97085 2 0.0000 0.964 0.000 1.000
#> GSM97059 1 0.0000 0.987 1.000 0.000
#> GSM97072 2 0.0000 0.964 0.000 1.000
#> GSM97078 2 0.4562 0.882 0.096 0.904
#> GSM97067 2 0.0000 0.964 0.000 1.000
#> GSM97087 2 0.0000 0.964 0.000 1.000
#> GSM97111 2 0.0000 0.964 0.000 1.000
#> GSM97064 2 0.0000 0.964 0.000 1.000
#> GSM97065 2 0.0000 0.964 0.000 1.000
#> GSM97081 2 0.0000 0.964 0.000 1.000
#> GSM97082 2 0.0000 0.964 0.000 1.000
#> GSM97088 2 0.0000 0.964 0.000 1.000
#> GSM97100 2 0.0000 0.964 0.000 1.000
#> GSM97104 2 0.0000 0.964 0.000 1.000
#> GSM97108 2 0.0376 0.961 0.004 0.996
#> GSM97050 2 0.0000 0.964 0.000 1.000
#> GSM97080 2 0.0000 0.964 0.000 1.000
#> GSM97089 2 0.0000 0.964 0.000 1.000
#> GSM97092 2 0.0000 0.964 0.000 1.000
#> GSM97093 2 0.0672 0.959 0.008 0.992
#> GSM97058 2 0.0000 0.964 0.000 1.000
#> GSM97051 2 0.0000 0.964 0.000 1.000
#> GSM97052 2 0.0000 0.964 0.000 1.000
#> GSM97061 2 0.0000 0.964 0.000 1.000
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM97138 1 0.2878 0.7582 0.904 0.000 0.096
#> GSM97145 1 0.1182 0.7789 0.976 0.012 0.012
#> GSM97147 1 0.5363 0.5867 0.724 0.276 0.000
#> GSM97125 1 0.2959 0.7571 0.900 0.000 0.100
#> GSM97127 1 0.2280 0.7755 0.940 0.052 0.008
#> GSM97130 1 0.2939 0.7700 0.916 0.072 0.012
#> GSM97133 1 0.2878 0.7559 0.904 0.096 0.000
#> GSM97134 1 0.3670 0.7710 0.888 0.020 0.092
#> GSM97120 1 0.1337 0.7787 0.972 0.016 0.012
#> GSM97126 1 0.2384 0.7730 0.936 0.008 0.056
#> GSM97112 1 0.6280 0.3384 0.540 0.000 0.460
#> GSM97115 1 0.5431 0.5745 0.716 0.284 0.000
#> GSM97116 1 0.2261 0.7686 0.932 0.000 0.068
#> GSM97117 2 0.3752 0.7098 0.000 0.856 0.144
#> GSM97119 1 0.5591 0.6010 0.696 0.000 0.304
#> GSM97122 1 0.5560 0.6057 0.700 0.000 0.300
#> GSM97135 1 0.5291 0.6388 0.732 0.000 0.268
#> GSM97136 3 0.1905 0.6576 0.016 0.028 0.956
#> GSM97139 1 0.0829 0.7784 0.984 0.004 0.012
#> GSM97146 1 0.1411 0.7749 0.964 0.000 0.036
#> GSM97123 2 0.2959 0.7408 0.000 0.900 0.100
#> GSM97129 2 0.4891 0.7396 0.124 0.836 0.040
#> GSM97143 1 0.5497 0.6136 0.708 0.000 0.292
#> GSM97113 2 0.5497 0.5586 0.292 0.708 0.000
#> GSM97056 1 0.1860 0.7733 0.948 0.052 0.000
#> GSM97124 1 0.3752 0.7377 0.856 0.000 0.144
#> GSM97132 1 0.3752 0.7372 0.856 0.000 0.144
#> GSM97144 1 0.3791 0.7780 0.892 0.060 0.048
#> GSM97149 1 0.3412 0.7410 0.876 0.124 0.000
#> GSM97068 1 0.6299 0.0895 0.524 0.476 0.000
#> GSM97071 3 0.6291 0.2263 0.000 0.468 0.532
#> GSM97086 2 0.3038 0.7628 0.104 0.896 0.000
#> GSM97103 2 0.3816 0.7062 0.000 0.852 0.148
#> GSM97057 2 0.6095 0.3294 0.392 0.608 0.000
#> GSM97060 2 0.5650 0.4457 0.000 0.688 0.312
#> GSM97075 2 0.4062 0.6898 0.000 0.836 0.164
#> GSM97098 2 0.3941 0.6983 0.000 0.844 0.156
#> GSM97099 2 0.1529 0.7824 0.040 0.960 0.000
#> GSM97101 2 0.3879 0.7363 0.152 0.848 0.000
#> GSM97105 2 0.2625 0.7713 0.084 0.916 0.000
#> GSM97106 2 0.4002 0.6958 0.000 0.840 0.160
#> GSM97121 2 0.3752 0.7418 0.144 0.856 0.000
#> GSM97128 3 0.2537 0.5962 0.080 0.000 0.920
#> GSM97131 2 0.1453 0.7800 0.008 0.968 0.024
#> GSM97137 1 0.2448 0.7646 0.924 0.076 0.000
#> GSM97118 1 0.6305 0.2830 0.516 0.000 0.484
#> GSM97114 1 0.6095 0.3562 0.608 0.392 0.000
#> GSM97142 1 0.6140 0.4507 0.596 0.000 0.404
#> GSM97140 2 0.5733 0.4949 0.324 0.676 0.000
#> GSM97141 2 0.4399 0.7048 0.188 0.812 0.000
#> GSM97055 3 0.4887 0.4210 0.228 0.000 0.772
#> GSM97090 1 0.5220 0.6785 0.780 0.208 0.012
#> GSM97091 3 0.5431 0.3205 0.284 0.000 0.716
#> GSM97148 1 0.1411 0.7762 0.964 0.036 0.000
#> GSM97063 3 0.5859 0.1779 0.344 0.000 0.656
#> GSM97053 1 0.3686 0.7392 0.860 0.000 0.140
#> GSM97066 3 0.4842 0.6379 0.000 0.224 0.776
#> GSM97079 2 0.1399 0.7787 0.004 0.968 0.028
#> GSM97083 3 0.5254 0.3602 0.264 0.000 0.736
#> GSM97084 2 0.3267 0.7576 0.116 0.884 0.000
#> GSM97094 1 0.5355 0.7397 0.800 0.032 0.168
#> GSM97096 2 0.5178 0.5597 0.000 0.744 0.256
#> GSM97097 2 0.1289 0.7759 0.000 0.968 0.032
#> GSM97107 1 0.6719 0.7232 0.744 0.160 0.096
#> GSM97054 2 0.4062 0.7292 0.164 0.836 0.000
#> GSM97062 2 0.2173 0.7823 0.048 0.944 0.008
#> GSM97069 3 0.5465 0.6073 0.000 0.288 0.712
#> GSM97070 3 0.5810 0.5518 0.000 0.336 0.664
#> GSM97073 3 0.5760 0.5643 0.000 0.328 0.672
#> GSM97076 3 0.5404 0.3846 0.256 0.004 0.740
#> GSM97077 2 0.2749 0.7813 0.064 0.924 0.012
#> GSM97095 1 0.5325 0.6280 0.748 0.248 0.004
#> GSM97102 3 0.5465 0.6077 0.000 0.288 0.712
#> GSM97109 2 0.5859 0.4550 0.344 0.656 0.000
#> GSM97110 2 0.1860 0.7808 0.052 0.948 0.000
#> GSM97074 3 0.2448 0.5997 0.076 0.000 0.924
#> GSM97085 3 0.0747 0.6574 0.000 0.016 0.984
#> GSM97059 1 0.5926 0.4392 0.644 0.356 0.000
#> GSM97072 2 0.5882 0.3542 0.000 0.652 0.348
#> GSM97078 3 0.1711 0.6370 0.032 0.008 0.960
#> GSM97067 3 0.5216 0.6244 0.000 0.260 0.740
#> GSM97087 3 0.6235 0.3567 0.000 0.436 0.564
#> GSM97111 2 0.1647 0.7760 0.004 0.960 0.036
#> GSM97064 2 0.2261 0.7589 0.000 0.932 0.068
#> GSM97065 2 0.4555 0.6402 0.000 0.800 0.200
#> GSM97081 2 0.5835 0.3748 0.000 0.660 0.340
#> GSM97082 3 0.5431 0.6109 0.000 0.284 0.716
#> GSM97088 3 0.1337 0.6484 0.016 0.012 0.972
#> GSM97100 2 0.3879 0.7359 0.152 0.848 0.000
#> GSM97104 3 0.5650 0.5831 0.000 0.312 0.688
#> GSM97108 2 0.3879 0.7362 0.152 0.848 0.000
#> GSM97050 2 0.1031 0.7823 0.024 0.976 0.000
#> GSM97080 3 0.6126 0.4397 0.000 0.400 0.600
#> GSM97089 3 0.6302 0.2259 0.000 0.480 0.520
#> GSM97092 2 0.5327 0.5320 0.000 0.728 0.272
#> GSM97093 2 0.2772 0.7754 0.080 0.916 0.004
#> GSM97058 2 0.1031 0.7782 0.000 0.976 0.024
#> GSM97051 2 0.1031 0.7783 0.000 0.976 0.024
#> GSM97052 2 0.5016 0.5878 0.000 0.760 0.240
#> GSM97061 2 0.3267 0.7308 0.000 0.884 0.116
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM97138 1 0.1302 0.7094 0.956 0.000 0.000 0.044
#> GSM97145 1 0.0921 0.6973 0.972 0.000 0.028 0.000
#> GSM97147 1 0.5208 0.5496 0.748 0.172 0.080 0.000
#> GSM97125 1 0.3157 0.6784 0.852 0.004 0.000 0.144
#> GSM97127 1 0.0564 0.7101 0.988 0.004 0.004 0.004
#> GSM97130 2 0.5387 0.5789 0.256 0.696 0.000 0.048
#> GSM97133 1 0.0895 0.7023 0.976 0.004 0.020 0.000
#> GSM97134 2 0.5889 0.5932 0.188 0.696 0.000 0.116
#> GSM97120 1 0.0707 0.7010 0.980 0.000 0.020 0.000
#> GSM97126 1 0.1824 0.7085 0.936 0.000 0.004 0.060
#> GSM97112 4 0.5263 -0.2646 0.448 0.008 0.000 0.544
#> GSM97115 2 0.2401 0.7793 0.092 0.904 0.000 0.004
#> GSM97116 1 0.2197 0.7021 0.916 0.004 0.000 0.080
#> GSM97117 3 0.1637 0.6803 0.060 0.000 0.940 0.000
#> GSM97119 1 0.5459 0.4234 0.552 0.016 0.000 0.432
#> GSM97122 1 0.5398 0.4644 0.580 0.016 0.000 0.404
#> GSM97135 1 0.5024 0.5210 0.632 0.008 0.000 0.360
#> GSM97136 4 0.5781 0.0198 0.028 0.000 0.484 0.488
#> GSM97139 1 0.0188 0.7098 0.996 0.004 0.000 0.000
#> GSM97146 1 0.0524 0.7110 0.988 0.004 0.000 0.008
#> GSM97123 3 0.1733 0.6860 0.000 0.028 0.948 0.024
#> GSM97129 3 0.3841 0.6505 0.144 0.004 0.832 0.020
#> GSM97143 1 0.4978 0.4951 0.612 0.004 0.000 0.384
#> GSM97113 3 0.4746 0.4432 0.368 0.000 0.632 0.000
#> GSM97056 1 0.6483 0.1893 0.532 0.392 0.000 0.076
#> GSM97124 1 0.5839 0.5597 0.648 0.060 0.000 0.292
#> GSM97132 1 0.6674 0.5051 0.584 0.116 0.000 0.300
#> GSM97144 2 0.4364 0.7174 0.136 0.808 0.000 0.056
#> GSM97149 1 0.1305 0.6931 0.960 0.004 0.036 0.000
#> GSM97068 2 0.4070 0.7472 0.132 0.824 0.044 0.000
#> GSM97071 2 0.1209 0.7846 0.000 0.964 0.004 0.032
#> GSM97086 2 0.0921 0.7851 0.000 0.972 0.028 0.000
#> GSM97103 3 0.1584 0.6813 0.000 0.012 0.952 0.036
#> GSM97057 1 0.6733 -0.0603 0.492 0.092 0.416 0.000
#> GSM97060 3 0.7093 0.3390 0.000 0.172 0.556 0.272
#> GSM97075 3 0.0817 0.6830 0.000 0.000 0.976 0.024
#> GSM97098 3 0.0895 0.6838 0.004 0.000 0.976 0.020
#> GSM97099 3 0.1637 0.6836 0.060 0.000 0.940 0.000
#> GSM97101 3 0.4608 0.5382 0.304 0.004 0.692 0.000
#> GSM97105 3 0.4964 0.6208 0.068 0.168 0.764 0.000
#> GSM97106 3 0.4840 0.6003 0.000 0.240 0.732 0.028
#> GSM97121 3 0.4831 0.6020 0.208 0.040 0.752 0.000
#> GSM97128 4 0.4824 0.3856 0.024 0.228 0.004 0.744
#> GSM97131 3 0.4643 0.4649 0.000 0.344 0.656 0.000
#> GSM97137 1 0.5035 0.6086 0.748 0.196 0.000 0.056
#> GSM97118 4 0.6123 -0.0439 0.336 0.064 0.000 0.600
#> GSM97114 1 0.4992 -0.1168 0.524 0.000 0.476 0.000
#> GSM97142 1 0.5294 0.3306 0.508 0.008 0.000 0.484
#> GSM97140 3 0.7577 0.2514 0.316 0.216 0.468 0.000
#> GSM97141 3 0.4605 0.4955 0.336 0.000 0.664 0.000
#> GSM97055 4 0.3032 0.4666 0.124 0.000 0.008 0.868
#> GSM97090 2 0.2179 0.7866 0.064 0.924 0.000 0.012
#> GSM97091 4 0.3636 0.3901 0.172 0.008 0.000 0.820
#> GSM97148 1 0.0336 0.7108 0.992 0.008 0.000 0.000
#> GSM97063 4 0.4792 0.1104 0.312 0.008 0.000 0.680
#> GSM97053 1 0.5925 0.5608 0.648 0.068 0.000 0.284
#> GSM97066 4 0.4973 0.3085 0.000 0.008 0.348 0.644
#> GSM97079 2 0.1867 0.7628 0.000 0.928 0.072 0.000
#> GSM97083 2 0.5599 0.5088 0.040 0.644 0.000 0.316
#> GSM97084 2 0.0336 0.7901 0.000 0.992 0.008 0.000
#> GSM97094 2 0.2675 0.7810 0.044 0.908 0.000 0.048
#> GSM97096 3 0.2081 0.6617 0.000 0.000 0.916 0.084
#> GSM97097 2 0.5016 0.2140 0.000 0.600 0.396 0.004
#> GSM97107 2 0.1610 0.7920 0.032 0.952 0.000 0.016
#> GSM97054 2 0.0336 0.7901 0.000 0.992 0.008 0.000
#> GSM97062 2 0.0469 0.7889 0.000 0.988 0.012 0.000
#> GSM97069 4 0.5290 0.2126 0.000 0.012 0.404 0.584
#> GSM97070 3 0.5273 0.0904 0.000 0.008 0.536 0.456
#> GSM97073 3 0.5155 0.0632 0.000 0.004 0.528 0.468
#> GSM97076 4 0.5690 0.4790 0.168 0.000 0.116 0.716
#> GSM97077 3 0.5691 0.1744 0.024 0.468 0.508 0.000
#> GSM97095 2 0.3224 0.7563 0.120 0.864 0.000 0.016
#> GSM97102 3 0.4941 0.1536 0.000 0.000 0.564 0.436
#> GSM97109 3 0.4898 0.3484 0.416 0.000 0.584 0.000
#> GSM97110 3 0.1867 0.6812 0.072 0.000 0.928 0.000
#> GSM97074 4 0.0469 0.5408 0.012 0.000 0.000 0.988
#> GSM97085 4 0.1118 0.5557 0.000 0.000 0.036 0.964
#> GSM97059 2 0.5660 0.3901 0.396 0.576 0.028 0.000
#> GSM97072 3 0.4690 0.4639 0.000 0.012 0.712 0.276
#> GSM97078 2 0.5064 0.4707 0.004 0.632 0.004 0.360
#> GSM97067 4 0.5172 0.2151 0.000 0.008 0.404 0.588
#> GSM97087 3 0.5256 0.2518 0.000 0.012 0.596 0.392
#> GSM97111 3 0.0817 0.6858 0.024 0.000 0.976 0.000
#> GSM97064 3 0.3311 0.6541 0.000 0.172 0.828 0.000
#> GSM97065 3 0.1767 0.6806 0.012 0.000 0.944 0.044
#> GSM97081 3 0.2011 0.6641 0.000 0.000 0.920 0.080
#> GSM97082 4 0.5212 0.1763 0.000 0.008 0.420 0.572
#> GSM97088 4 0.2861 0.5201 0.012 0.092 0.004 0.892
#> GSM97100 2 0.3900 0.6971 0.020 0.816 0.164 0.000
#> GSM97104 4 0.5167 -0.0109 0.000 0.004 0.488 0.508
#> GSM97108 3 0.4574 0.6016 0.220 0.024 0.756 0.000
#> GSM97050 2 0.5158 -0.0545 0.004 0.524 0.472 0.000
#> GSM97080 3 0.5399 0.0507 0.000 0.012 0.520 0.468
#> GSM97089 3 0.5110 0.3352 0.000 0.012 0.636 0.352
#> GSM97092 3 0.4206 0.6189 0.000 0.048 0.816 0.136
#> GSM97093 3 0.3858 0.6686 0.100 0.056 0.844 0.000
#> GSM97058 3 0.3626 0.6482 0.004 0.184 0.812 0.000
#> GSM97051 2 0.1637 0.7720 0.000 0.940 0.060 0.000
#> GSM97052 3 0.4784 0.6259 0.000 0.100 0.788 0.112
#> GSM97061 3 0.3280 0.6707 0.000 0.124 0.860 0.016
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM97138 1 0.4040 0.4453 0.712 0.012 0.000 0.000 0.276
#> GSM97145 1 0.4390 0.5632 0.760 0.084 0.000 0.000 0.156
#> GSM97147 1 0.7139 0.4923 0.616 0.128 0.060 0.156 0.040
#> GSM97125 1 0.4701 0.2311 0.612 0.016 0.000 0.004 0.368
#> GSM97127 1 0.2707 0.6089 0.860 0.008 0.000 0.000 0.132
#> GSM97130 4 0.4079 0.7256 0.092 0.012 0.004 0.816 0.076
#> GSM97133 1 0.0798 0.6462 0.976 0.008 0.000 0.000 0.016
#> GSM97134 4 0.5042 0.6437 0.044 0.040 0.000 0.728 0.188
#> GSM97120 1 0.1331 0.6449 0.952 0.008 0.000 0.000 0.040
#> GSM97126 1 0.4631 0.4486 0.696 0.020 0.008 0.004 0.272
#> GSM97112 5 0.3895 0.5352 0.264 0.004 0.000 0.004 0.728
#> GSM97115 4 0.2443 0.7863 0.036 0.004 0.040 0.912 0.008
#> GSM97116 1 0.2970 0.5702 0.828 0.000 0.000 0.004 0.168
#> GSM97117 2 0.4751 0.5205 0.036 0.692 0.264 0.000 0.008
#> GSM97119 5 0.4875 0.4545 0.336 0.008 0.000 0.024 0.632
#> GSM97122 5 0.4835 0.3877 0.384 0.004 0.000 0.020 0.592
#> GSM97135 5 0.4837 0.3078 0.424 0.004 0.000 0.016 0.556
#> GSM97136 5 0.6727 -0.1657 0.040 0.400 0.100 0.000 0.460
#> GSM97139 1 0.1502 0.6419 0.940 0.004 0.000 0.000 0.056
#> GSM97146 1 0.1041 0.6427 0.964 0.004 0.000 0.000 0.032
#> GSM97123 3 0.4088 0.2632 0.000 0.304 0.688 0.008 0.000
#> GSM97129 2 0.7074 0.4698 0.156 0.580 0.180 0.004 0.080
#> GSM97143 5 0.4505 0.4239 0.368 0.008 0.000 0.004 0.620
#> GSM97113 1 0.6010 0.1382 0.584 0.208 0.208 0.000 0.000
#> GSM97056 1 0.5707 0.3916 0.644 0.008 0.004 0.244 0.100
#> GSM97124 5 0.6356 0.3558 0.344 0.020 0.000 0.108 0.528
#> GSM97132 5 0.6473 0.3663 0.280 0.004 0.000 0.200 0.516
#> GSM97144 4 0.3243 0.7584 0.012 0.036 0.000 0.860 0.092
#> GSM97149 1 0.1430 0.6235 0.944 0.004 0.052 0.000 0.000
#> GSM97068 4 0.4531 0.7212 0.128 0.024 0.068 0.780 0.000
#> GSM97071 4 0.1498 0.7899 0.000 0.008 0.024 0.952 0.016
#> GSM97086 4 0.1764 0.7727 0.000 0.064 0.008 0.928 0.000
#> GSM97103 2 0.2635 0.5708 0.000 0.888 0.016 0.088 0.008
#> GSM97057 1 0.5122 0.2443 0.584 0.012 0.380 0.024 0.000
#> GSM97060 3 0.5703 0.5245 0.000 0.148 0.684 0.028 0.140
#> GSM97075 3 0.4009 0.2790 0.004 0.312 0.684 0.000 0.000
#> GSM97098 2 0.2077 0.5865 0.000 0.908 0.084 0.008 0.000
#> GSM97099 2 0.3274 0.6026 0.064 0.856 0.076 0.004 0.000
#> GSM97101 2 0.6780 0.3722 0.280 0.448 0.268 0.004 0.000
#> GSM97105 2 0.6291 0.3355 0.040 0.520 0.376 0.064 0.000
#> GSM97106 3 0.5688 0.3321 0.000 0.296 0.608 0.088 0.008
#> GSM97121 2 0.5814 0.5502 0.116 0.688 0.148 0.048 0.000
#> GSM97128 5 0.5491 0.3417 0.000 0.004 0.116 0.224 0.656
#> GSM97131 2 0.6729 0.2947 0.004 0.460 0.244 0.292 0.000
#> GSM97137 1 0.4878 0.4801 0.728 0.004 0.000 0.164 0.104
#> GSM97118 5 0.4436 0.5664 0.140 0.000 0.004 0.088 0.768
#> GSM97114 1 0.5924 -0.0601 0.504 0.400 0.092 0.004 0.000
#> GSM97142 5 0.4253 0.5177 0.284 0.008 0.000 0.008 0.700
#> GSM97140 3 0.7524 0.0723 0.296 0.136 0.472 0.096 0.000
#> GSM97141 2 0.6735 0.3509 0.340 0.436 0.220 0.004 0.000
#> GSM97055 5 0.3572 0.5333 0.040 0.008 0.120 0.000 0.832
#> GSM97090 4 0.4586 0.7466 0.052 0.008 0.104 0.796 0.040
#> GSM97091 5 0.2983 0.5776 0.096 0.000 0.032 0.004 0.868
#> GSM97148 1 0.0671 0.6433 0.980 0.004 0.000 0.000 0.016
#> GSM97063 5 0.3726 0.5784 0.152 0.004 0.028 0.004 0.812
#> GSM97053 1 0.5390 -0.0438 0.536 0.004 0.000 0.048 0.412
#> GSM97066 3 0.6114 0.3931 0.000 0.128 0.472 0.000 0.400
#> GSM97079 4 0.3527 0.6873 0.000 0.192 0.016 0.792 0.000
#> GSM97083 4 0.4669 0.5332 0.004 0.004 0.016 0.656 0.320
#> GSM97084 4 0.1041 0.7839 0.000 0.032 0.004 0.964 0.000
#> GSM97094 4 0.4469 0.7020 0.000 0.148 0.000 0.756 0.096
#> GSM97096 2 0.3419 0.5257 0.000 0.804 0.180 0.000 0.016
#> GSM97097 2 0.4150 0.1329 0.000 0.612 0.000 0.388 0.000
#> GSM97107 4 0.2879 0.7647 0.000 0.100 0.000 0.868 0.032
#> GSM97054 4 0.3243 0.7242 0.000 0.004 0.180 0.812 0.004
#> GSM97062 4 0.0898 0.7856 0.000 0.008 0.020 0.972 0.000
#> GSM97069 3 0.6347 0.3556 0.000 0.160 0.432 0.000 0.408
#> GSM97070 3 0.6303 0.4174 0.000 0.196 0.524 0.000 0.280
#> GSM97073 2 0.6326 0.1385 0.000 0.524 0.268 0.000 0.208
#> GSM97076 2 0.5902 0.1630 0.016 0.556 0.028 0.024 0.376
#> GSM97077 3 0.5380 0.4158 0.100 0.032 0.716 0.152 0.000
#> GSM97095 4 0.3412 0.7663 0.060 0.004 0.020 0.864 0.052
#> GSM97102 2 0.5703 0.3085 0.000 0.628 0.184 0.000 0.188
#> GSM97109 2 0.2973 0.5905 0.084 0.880 0.012 0.016 0.008
#> GSM97110 2 0.2569 0.5993 0.040 0.892 0.068 0.000 0.000
#> GSM97074 5 0.2616 0.4937 0.000 0.020 0.100 0.000 0.880
#> GSM97085 5 0.4398 0.2068 0.000 0.040 0.240 0.000 0.720
#> GSM97059 1 0.6162 0.3346 0.572 0.004 0.252 0.172 0.000
#> GSM97072 2 0.5440 0.2625 0.000 0.620 0.300 0.004 0.076
#> GSM97078 4 0.6464 0.3917 0.000 0.008 0.168 0.520 0.304
#> GSM97067 3 0.6519 0.3267 0.000 0.192 0.408 0.000 0.400
#> GSM97087 3 0.2722 0.5732 0.000 0.020 0.872 0.000 0.108
#> GSM97111 2 0.3760 0.5707 0.028 0.784 0.188 0.000 0.000
#> GSM97064 3 0.3279 0.5131 0.016 0.048 0.864 0.072 0.000
#> GSM97065 2 0.5908 0.3000 0.080 0.552 0.356 0.000 0.012
#> GSM97081 3 0.5843 0.1710 0.000 0.388 0.512 0.000 0.100
#> GSM97082 3 0.5040 0.5107 0.000 0.068 0.660 0.000 0.272
#> GSM97088 5 0.5375 0.2996 0.000 0.004 0.220 0.108 0.668
#> GSM97100 4 0.6649 0.3860 0.024 0.164 0.268 0.544 0.000
#> GSM97104 3 0.6659 0.3337 0.000 0.248 0.436 0.000 0.316
#> GSM97108 2 0.6417 0.4918 0.160 0.596 0.216 0.028 0.000
#> GSM97050 3 0.5186 0.4404 0.068 0.052 0.740 0.140 0.000
#> GSM97080 3 0.5592 0.4995 0.000 0.144 0.636 0.000 0.220
#> GSM97089 3 0.2959 0.5736 0.000 0.036 0.864 0.000 0.100
#> GSM97092 3 0.3291 0.5583 0.000 0.088 0.848 0.000 0.064
#> GSM97093 3 0.5469 0.3828 0.164 0.092 0.708 0.036 0.000
#> GSM97058 3 0.4849 0.4548 0.040 0.128 0.764 0.068 0.000
#> GSM97051 4 0.4735 0.2320 0.004 0.004 0.472 0.516 0.004
#> GSM97052 3 0.3135 0.5659 0.000 0.036 0.876 0.028 0.060
#> GSM97061 3 0.3243 0.5185 0.000 0.092 0.860 0.036 0.012
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM97138 1 0.4821 0.24996 0.584 0.040 0.000 0.000 0.364 0.012
#> GSM97145 5 0.6707 0.10079 0.304 0.268 0.000 0.000 0.392 0.036
#> GSM97147 2 0.5307 0.51987 0.128 0.700 0.000 0.080 0.088 0.004
#> GSM97125 5 0.4136 0.54542 0.248 0.040 0.000 0.000 0.708 0.004
#> GSM97127 1 0.4901 0.43780 0.640 0.060 0.000 0.000 0.284 0.016
#> GSM97130 4 0.3211 0.77121 0.028 0.016 0.000 0.852 0.092 0.012
#> GSM97133 1 0.1367 0.80079 0.944 0.012 0.000 0.000 0.044 0.000
#> GSM97134 5 0.5218 0.20626 0.000 0.076 0.000 0.376 0.540 0.008
#> GSM97120 1 0.1245 0.80018 0.952 0.016 0.000 0.000 0.032 0.000
#> GSM97126 5 0.4998 0.53428 0.164 0.172 0.000 0.000 0.660 0.004
#> GSM97112 5 0.2247 0.69235 0.060 0.000 0.024 0.000 0.904 0.012
#> GSM97115 4 0.3827 0.77235 0.036 0.068 0.024 0.828 0.044 0.000
#> GSM97116 1 0.1471 0.79337 0.932 0.004 0.000 0.000 0.064 0.000
#> GSM97117 2 0.4370 0.56177 0.020 0.772 0.036 0.000 0.032 0.140
#> GSM97119 5 0.2252 0.69085 0.056 0.016 0.000 0.016 0.908 0.004
#> GSM97122 5 0.2323 0.68307 0.084 0.012 0.000 0.012 0.892 0.000
#> GSM97135 5 0.2951 0.64668 0.168 0.004 0.000 0.004 0.820 0.004
#> GSM97136 5 0.6890 0.26312 0.004 0.204 0.104 0.000 0.508 0.180
#> GSM97139 1 0.1462 0.79683 0.936 0.008 0.000 0.000 0.056 0.000
#> GSM97146 1 0.0713 0.80170 0.972 0.000 0.000 0.000 0.028 0.000
#> GSM97123 2 0.4498 0.31739 0.000 0.632 0.324 0.004 0.000 0.040
#> GSM97129 2 0.5044 0.49312 0.044 0.712 0.008 0.000 0.164 0.072
#> GSM97143 5 0.2580 0.69127 0.068 0.008 0.008 0.004 0.892 0.020
#> GSM97113 1 0.4192 0.59154 0.748 0.188 0.028 0.000 0.000 0.036
#> GSM97056 1 0.3514 0.72587 0.828 0.008 0.008 0.112 0.040 0.004
#> GSM97124 5 0.4290 0.65826 0.080 0.048 0.000 0.084 0.784 0.004
#> GSM97132 5 0.4552 0.60208 0.052 0.012 0.000 0.188 0.732 0.016
#> GSM97144 4 0.3384 0.72161 0.000 0.024 0.000 0.800 0.168 0.008
#> GSM97149 1 0.0508 0.79684 0.984 0.012 0.004 0.000 0.000 0.000
#> GSM97068 4 0.4768 0.61847 0.240 0.040 0.012 0.692 0.004 0.012
#> GSM97071 4 0.3614 0.74212 0.004 0.008 0.052 0.832 0.016 0.088
#> GSM97086 4 0.2203 0.76355 0.000 0.016 0.004 0.896 0.000 0.084
#> GSM97103 6 0.4703 0.56001 0.000 0.204 0.008 0.084 0.004 0.700
#> GSM97057 1 0.4036 0.64476 0.772 0.100 0.120 0.008 0.000 0.000
#> GSM97060 3 0.4877 0.38976 0.000 0.040 0.688 0.032 0.008 0.232
#> GSM97075 2 0.6213 -0.08083 0.016 0.412 0.384 0.000 0.000 0.188
#> GSM97098 6 0.4127 0.58216 0.000 0.236 0.044 0.004 0.000 0.716
#> GSM97099 6 0.5531 0.26704 0.040 0.416 0.040 0.000 0.004 0.500
#> GSM97101 2 0.2781 0.62814 0.036 0.880 0.044 0.000 0.000 0.040
#> GSM97105 2 0.2635 0.61112 0.004 0.880 0.076 0.004 0.000 0.036
#> GSM97106 3 0.6634 0.28069 0.000 0.192 0.500 0.068 0.000 0.240
#> GSM97121 2 0.2879 0.59209 0.012 0.864 0.000 0.012 0.012 0.100
#> GSM97128 5 0.6039 0.42496 0.008 0.004 0.184 0.196 0.588 0.020
#> GSM97131 2 0.4704 0.55317 0.000 0.732 0.092 0.140 0.000 0.036
#> GSM97137 1 0.2653 0.76041 0.876 0.004 0.000 0.064 0.056 0.000
#> GSM97118 5 0.2978 0.68235 0.012 0.000 0.020 0.068 0.872 0.028
#> GSM97114 2 0.5821 0.21886 0.372 0.488 0.000 0.000 0.016 0.124
#> GSM97142 5 0.1862 0.69372 0.044 0.000 0.008 0.004 0.928 0.016
#> GSM97140 2 0.4522 0.48370 0.040 0.736 0.192 0.020 0.008 0.004
#> GSM97141 2 0.2943 0.63241 0.052 0.872 0.020 0.000 0.004 0.052
#> GSM97055 5 0.4813 0.49597 0.016 0.000 0.272 0.004 0.660 0.048
#> GSM97090 4 0.5151 0.72039 0.080 0.024 0.120 0.728 0.048 0.000
#> GSM97091 5 0.2592 0.67433 0.012 0.000 0.080 0.004 0.884 0.020
#> GSM97148 1 0.0692 0.80160 0.976 0.000 0.004 0.000 0.020 0.000
#> GSM97063 5 0.2650 0.68485 0.036 0.000 0.056 0.004 0.888 0.016
#> GSM97053 5 0.4395 0.51759 0.276 0.004 0.008 0.024 0.684 0.004
#> GSM97066 3 0.5650 0.22898 0.000 0.016 0.548 0.016 0.068 0.352
#> GSM97079 4 0.3955 0.62043 0.000 0.004 0.032 0.724 0.000 0.240
#> GSM97083 4 0.5176 0.42034 0.008 0.004 0.056 0.580 0.348 0.004
#> GSM97084 4 0.2489 0.76389 0.004 0.004 0.016 0.888 0.004 0.084
#> GSM97094 4 0.4799 0.66746 0.000 0.028 0.000 0.704 0.076 0.192
#> GSM97096 6 0.4723 0.57741 0.000 0.124 0.156 0.012 0.000 0.708
#> GSM97097 6 0.5562 0.11076 0.000 0.124 0.000 0.376 0.004 0.496
#> GSM97107 4 0.3676 0.73960 0.000 0.020 0.000 0.808 0.052 0.120
#> GSM97054 4 0.3579 0.71872 0.008 0.064 0.120 0.808 0.000 0.000
#> GSM97062 4 0.2012 0.77569 0.000 0.008 0.028 0.924 0.008 0.032
#> GSM97069 3 0.5228 0.17743 0.000 0.004 0.532 0.004 0.072 0.388
#> GSM97070 3 0.5677 0.16235 0.000 0.040 0.500 0.008 0.044 0.408
#> GSM97073 6 0.4791 0.37418 0.000 0.032 0.252 0.008 0.028 0.680
#> GSM97076 6 0.4421 0.57029 0.012 0.040 0.052 0.036 0.056 0.804
#> GSM97077 3 0.6496 -0.02390 0.016 0.420 0.420 0.116 0.012 0.016
#> GSM97095 4 0.3392 0.77503 0.024 0.024 0.016 0.844 0.092 0.000
#> GSM97102 6 0.4806 0.51729 0.000 0.072 0.200 0.000 0.028 0.700
#> GSM97109 6 0.4898 0.49146 0.032 0.292 0.004 0.016 0.008 0.648
#> GSM97110 6 0.4103 0.63284 0.052 0.136 0.016 0.012 0.000 0.784
#> GSM97074 5 0.6565 0.23561 0.000 0.000 0.240 0.040 0.464 0.256
#> GSM97085 5 0.5834 -0.00175 0.000 0.000 0.424 0.008 0.424 0.144
#> GSM97059 1 0.7417 0.11035 0.404 0.204 0.188 0.204 0.000 0.000
#> GSM97072 6 0.4058 0.48077 0.000 0.044 0.196 0.012 0.000 0.748
#> GSM97078 4 0.6490 0.39742 0.008 0.008 0.284 0.496 0.188 0.016
#> GSM97067 3 0.5297 0.04978 0.000 0.008 0.472 0.012 0.048 0.460
#> GSM97087 3 0.3544 0.46728 0.000 0.140 0.804 0.000 0.008 0.048
#> GSM97111 2 0.4446 0.43006 0.012 0.708 0.036 0.000 0.008 0.236
#> GSM97064 3 0.4944 0.25479 0.012 0.340 0.600 0.044 0.000 0.004
#> GSM97065 6 0.6515 0.30636 0.156 0.084 0.224 0.000 0.000 0.536
#> GSM97081 2 0.6292 -0.12352 0.000 0.388 0.388 0.000 0.016 0.208
#> GSM97082 3 0.4394 0.41278 0.000 0.020 0.760 0.004 0.092 0.124
#> GSM97088 5 0.6081 0.25170 0.004 0.004 0.384 0.084 0.488 0.036
#> GSM97100 2 0.4873 0.46925 0.000 0.676 0.104 0.212 0.004 0.004
#> GSM97104 3 0.5543 0.14449 0.000 0.028 0.532 0.000 0.072 0.368
#> GSM97108 2 0.3300 0.61200 0.036 0.856 0.000 0.024 0.016 0.068
#> GSM97050 3 0.5949 0.21973 0.016 0.316 0.540 0.116 0.000 0.012
#> GSM97080 3 0.5006 0.34029 0.004 0.024 0.660 0.008 0.036 0.268
#> GSM97089 3 0.4007 0.46910 0.000 0.144 0.776 0.000 0.016 0.064
#> GSM97092 3 0.4165 0.35232 0.000 0.292 0.676 0.000 0.004 0.028
#> GSM97093 2 0.5938 0.03826 0.032 0.484 0.420 0.032 0.008 0.024
#> GSM97058 3 0.6619 0.14927 0.024 0.372 0.464 0.072 0.000 0.068
#> GSM97051 3 0.6620 0.01786 0.012 0.308 0.356 0.316 0.000 0.008
#> GSM97052 3 0.3897 0.37008 0.000 0.264 0.712 0.016 0.000 0.008
#> GSM97061 3 0.4650 0.13518 0.000 0.416 0.548 0.028 0.000 0.008
Heatmaps for the consensus matrix. It visualizes the probability of two samples to be in a same group.
consensus_heatmap(res, k = 2)
consensus_heatmap(res, k = 3)
consensus_heatmap(res, k = 4)
consensus_heatmap(res, k = 5)
consensus_heatmap(res, k = 6)
Heatmaps for the membership of samples in all partitions to see how consistent they are:
membership_heatmap(res, k = 2)
membership_heatmap(res, k = 3)
membership_heatmap(res, k = 4)
membership_heatmap(res, k = 5)
membership_heatmap(res, k = 6)
As soon as we have had the classes for columns, we can look for signatures which are significantly different between classes which can be candidate marks for certain classes. Following are the heatmaps for signatures.
Signature heatmaps where rows are scaled:
get_signatures(res, k = 2)
get_signatures(res, k = 3)
get_signatures(res, k = 4)
get_signatures(res, k = 5)
get_signatures(res, k = 6)
Signature heatmaps where rows are not scaled:
get_signatures(res, k = 2, scale_rows = FALSE)
get_signatures(res, k = 3, scale_rows = FALSE)
get_signatures(res, k = 4, scale_rows = FALSE)
get_signatures(res, k = 5, scale_rows = FALSE)
get_signatures(res, k = 6, scale_rows = FALSE)
Compare the overlap of signatures from different k:
compare_signatures(res)
get_signature() returns a data frame invisibly. TO get the list of signatures, the function
call should be assigned to a variable explicitly. In following code, if plot argument is set
to FALSE, no heatmap is plotted while only the differential analysis is performed.
# code only for demonstration
tb = get_signature(res, k = ..., plot = FALSE)
An example of the output of tb is:
#> which_row fdr mean_1 mean_2 scaled_mean_1 scaled_mean_2 km
#> 1 38 0.042760348 8.373488 9.131774 -0.5533452 0.5164555 1
#> 2 40 0.018707592 7.106213 8.469186 -0.6173731 0.5762149 1
#> 3 55 0.019134737 10.221463 11.207825 -0.6159697 0.5749050 1
#> 4 59 0.006059896 5.921854 7.869574 -0.6899429 0.6439467 1
#> 5 60 0.018055526 8.928898 10.211722 -0.6204761 0.5791110 1
#> 6 98 0.009384629 15.714769 14.887706 0.6635654 -0.6193277 2
...
The columns in tb are:
which_row: row indices corresponding to the input matrix.fdr: FDR for the differential test. mean_x: The mean value in group x.scaled_mean_x: The mean value in group x after rows are scaled.km: Row groups if k-means clustering is applied to rows.UMAP plot which shows how samples are separated.
dimension_reduction(res, k = 2, method = "UMAP")
dimension_reduction(res, k = 3, method = "UMAP")
dimension_reduction(res, k = 4, method = "UMAP")
dimension_reduction(res, k = 5, method = "UMAP")
dimension_reduction(res, k = 6, method = "UMAP")
Following heatmap shows how subgroups are split when increasing k:
collect_classes(res)
Test correlation between subgroups and known annotations. If the known annotation is numeric, one-way ANOVA test is applied, and if the known annotation is discrete, chi-squared contingency table test is applied.
test_to_known_factors(res)
#> n disease.state(p) specimen(p) cell.type(p) other(p) k
#> CV:NMF 98 8.95e-05 0.650 4.19e-14 0.184 2
#> CV:NMF 79 7.34e-05 0.107 9.83e-13 0.402 3
#> CV:NMF 62 2.76e-03 0.253 1.41e-11 0.180 4
#> CV:NMF 48 2.44e-04 0.228 6.96e-10 0.384 5
#> CV:NMF 53 7.07e-02 0.720 8.65e-10 0.108 6
If matrix rows can be associated to genes, consider to use functional_enrichment(res,
...) to perform function enrichment for the signature genes. See this vignette for more detailed explanations.
The object with results only for a single top-value method and a single partition method can be extracted as:
res = res_list["MAD", "hclust"]
# you can also extract it by
# res = res_list["MAD:hclust"]
A summary of res and all the functions that can be applied to it:
res
#> A 'ConsensusPartition' object with k = 2, 3, 4, 5, 6.
#> On a matrix with 21512 rows and 100 columns.
#> Top rows (1000, 2000, 3000, 4000, 5000) are extracted by 'MAD' method.
#> Subgroups are detected by 'hclust' method.
#> Performed in total 1250 partitions by row resampling.
#> Best k for subgroups seems to be 2.
#>
#> Following methods can be applied to this 'ConsensusPartition' object:
#> [1] "cola_report" "collect_classes" "collect_plots"
#> [4] "collect_stats" "colnames" "compare_signatures"
#> [7] "consensus_heatmap" "dimension_reduction" "functional_enrichment"
#> [10] "get_anno_col" "get_anno" "get_classes"
#> [13] "get_consensus" "get_matrix" "get_membership"
#> [16] "get_param" "get_signatures" "get_stats"
#> [19] "is_best_k" "is_stable_k" "membership_heatmap"
#> [22] "ncol" "nrow" "plot_ecdf"
#> [25] "rownames" "select_partition_number" "show"
#> [28] "suggest_best_k" "test_to_known_factors"
collect_plots() function collects all the plots made from res for all k (number of partitions)
into one single page to provide an easy and fast comparison between different k.
collect_plots(res)
The plots are:
k and the heatmap of
predicted classes for each k.k.k.k.All the plots in panels can be made by individual functions and they are plotted later in this section.
select_partition_number() produces several plots showing different
statistics for choosing “optimized” k. There are following statistics:
k;k, the area increased is defined as \(A_k - A_{k-1}\).The detailed explanations of these statistics can be found in the cola vignette.
Generally speaking, lower PAC score, higher mean silhouette score or higher
concordance corresponds to better partition. Rand index and Jaccard index
measure how similar the current partition is compared to partition with k-1.
If they are too similar, we won't accept k is better than k-1.
select_partition_number(res)
The numeric values for all these statistics can be obtained by get_stats().
get_stats(res)
#> k 1-PAC mean_silhouette concordance area_increased Rand Jaccard
#> 2 2 0.219 0.661 0.817 0.4233 0.540 0.540
#> 3 3 0.248 0.351 0.653 0.4089 0.835 0.738
#> 4 4 0.337 0.527 0.696 0.1696 0.761 0.568
#> 5 5 0.434 0.568 0.697 0.0667 0.890 0.672
#> 6 6 0.576 0.538 0.702 0.0665 0.977 0.909
suggest_best_k() suggests the best \(k\) based on these statistics. The rules are as follows:
suggest_best_k(res)
#> [1] 2
Following shows the table of the partitions (You need to click the show/hide
code output link to see it). The membership matrix (columns with name p*)
is inferred by
clue::cl_consensus()
function with the SE method. Basically the value in the membership matrix
represents the probability to belong to a certain group. The finall class
label for an item is determined with the group with highest probability it
belongs to.
In get_classes() function, the entropy is calculated from the membership
matrix and the silhouette score is calculated from the consensus matrix.
cbind(get_classes(res, k = 2), get_membership(res, k = 2))
#> class entropy silhouette p1 p2
#> GSM97138 1 0.1633 0.77905 0.976 0.024
#> GSM97145 1 0.0938 0.78277 0.988 0.012
#> GSM97147 2 0.8861 0.71951 0.304 0.696
#> GSM97125 1 0.0938 0.78277 0.988 0.012
#> GSM97127 1 0.0938 0.78277 0.988 0.012
#> GSM97130 1 0.8144 0.57446 0.748 0.252
#> GSM97133 1 0.0376 0.78306 0.996 0.004
#> GSM97134 1 0.9983 -0.14646 0.524 0.476
#> GSM97120 1 0.0376 0.78150 0.996 0.004
#> GSM97126 1 0.9881 0.02908 0.564 0.436
#> GSM97112 1 0.0000 0.78251 1.000 0.000
#> GSM97115 1 0.9866 0.08085 0.568 0.432
#> GSM97116 1 0.0000 0.78251 1.000 0.000
#> GSM97117 2 0.8661 0.73553 0.288 0.712
#> GSM97119 1 0.0000 0.78251 1.000 0.000
#> GSM97122 1 0.0000 0.78251 1.000 0.000
#> GSM97135 1 0.0000 0.78251 1.000 0.000
#> GSM97136 2 0.9977 0.28813 0.472 0.528
#> GSM97139 1 0.0000 0.78251 1.000 0.000
#> GSM97146 1 0.0000 0.78251 1.000 0.000
#> GSM97123 2 0.2236 0.77300 0.036 0.964
#> GSM97129 1 0.9988 -0.16330 0.520 0.480
#> GSM97143 1 0.7453 0.61867 0.788 0.212
#> GSM97113 2 0.9358 0.65026 0.352 0.648
#> GSM97056 1 0.3274 0.76052 0.940 0.060
#> GSM97124 1 0.1184 0.78204 0.984 0.016
#> GSM97132 1 0.8207 0.57089 0.744 0.256
#> GSM97144 1 0.8713 0.50735 0.708 0.292
#> GSM97149 1 0.0000 0.78251 1.000 0.000
#> GSM97068 1 0.9922 0.00378 0.552 0.448
#> GSM97071 2 0.8207 0.74548 0.256 0.744
#> GSM97086 2 0.7376 0.77761 0.208 0.792
#> GSM97103 2 0.3584 0.78227 0.068 0.932
#> GSM97057 2 0.9358 0.65143 0.352 0.648
#> GSM97060 2 0.0000 0.74787 0.000 1.000
#> GSM97075 2 0.8144 0.76220 0.252 0.748
#> GSM97098 2 0.3584 0.78227 0.068 0.932
#> GSM97099 2 0.8608 0.73843 0.284 0.716
#> GSM97101 2 0.8499 0.74541 0.276 0.724
#> GSM97105 2 0.5629 0.79337 0.132 0.868
#> GSM97106 2 0.0000 0.74787 0.000 1.000
#> GSM97121 2 0.8555 0.73399 0.280 0.720
#> GSM97128 2 0.9963 0.36549 0.464 0.536
#> GSM97131 2 0.5178 0.79295 0.116 0.884
#> GSM97137 1 0.8016 0.58613 0.756 0.244
#> GSM97118 1 0.7602 0.60782 0.780 0.220
#> GSM97114 2 0.8661 0.73553 0.288 0.712
#> GSM97142 1 0.0000 0.78251 1.000 0.000
#> GSM97140 2 0.8081 0.76790 0.248 0.752
#> GSM97141 2 0.8499 0.74420 0.276 0.724
#> GSM97055 1 0.9248 0.36432 0.660 0.340
#> GSM97090 1 0.9909 0.03821 0.556 0.444
#> GSM97091 1 0.2603 0.77011 0.956 0.044
#> GSM97148 1 0.0000 0.78251 1.000 0.000
#> GSM97063 1 0.2603 0.77011 0.956 0.044
#> GSM97053 1 0.0000 0.78251 1.000 0.000
#> GSM97066 2 0.2603 0.77621 0.044 0.956
#> GSM97079 2 0.7453 0.77533 0.212 0.788
#> GSM97083 2 0.9970 0.35371 0.468 0.532
#> GSM97084 2 0.7745 0.76494 0.228 0.772
#> GSM97094 2 0.7745 0.76494 0.228 0.772
#> GSM97096 2 0.3584 0.78227 0.068 0.932
#> GSM97097 2 0.7745 0.76494 0.228 0.772
#> GSM97107 2 0.8016 0.75300 0.244 0.756
#> GSM97054 2 0.8207 0.74548 0.256 0.744
#> GSM97062 2 0.7453 0.77533 0.212 0.788
#> GSM97069 2 0.1184 0.76026 0.016 0.984
#> GSM97070 2 0.2603 0.77621 0.044 0.956
#> GSM97073 2 0.3584 0.78451 0.068 0.932
#> GSM97076 2 0.9710 0.53441 0.400 0.600
#> GSM97077 2 0.7219 0.78772 0.200 0.800
#> GSM97095 1 0.9988 -0.15209 0.520 0.480
#> GSM97102 2 0.3584 0.78227 0.068 0.932
#> GSM97109 2 0.7453 0.77457 0.212 0.788
#> GSM97110 2 0.7453 0.77457 0.212 0.788
#> GSM97074 2 0.9866 0.44192 0.432 0.568
#> GSM97085 2 0.9775 0.48842 0.412 0.588
#> GSM97059 2 0.9209 0.67693 0.336 0.664
#> GSM97072 2 0.0000 0.74787 0.000 1.000
#> GSM97078 2 0.9963 0.36549 0.464 0.536
#> GSM97067 2 0.1414 0.76330 0.020 0.980
#> GSM97087 2 0.0000 0.74787 0.000 1.000
#> GSM97111 2 0.8386 0.75213 0.268 0.732
#> GSM97064 2 0.6623 0.79326 0.172 0.828
#> GSM97065 2 0.9427 0.61986 0.360 0.640
#> GSM97081 2 0.8016 0.76743 0.244 0.756
#> GSM97082 2 0.3274 0.77808 0.060 0.940
#> GSM97088 2 0.9775 0.48842 0.412 0.588
#> GSM97100 2 0.7883 0.77464 0.236 0.764
#> GSM97104 2 0.0000 0.74787 0.000 1.000
#> GSM97108 2 0.8267 0.75939 0.260 0.740
#> GSM97050 2 0.6148 0.79673 0.152 0.848
#> GSM97080 2 0.1414 0.76312 0.020 0.980
#> GSM97089 2 0.0000 0.74787 0.000 1.000
#> GSM97092 2 0.1184 0.76040 0.016 0.984
#> GSM97093 2 0.9815 0.45269 0.420 0.580
#> GSM97058 2 0.6623 0.79326 0.172 0.828
#> GSM97051 2 0.4939 0.79182 0.108 0.892
#> GSM97052 2 0.0000 0.74787 0.000 1.000
#> GSM97061 2 0.2236 0.77320 0.036 0.964
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM97138 1 0.409 0.77129 0.880 0.052 0.068
#> GSM97145 1 0.191 0.77398 0.956 0.028 0.016
#> GSM97147 2 0.517 0.33763 0.148 0.816 0.036
#> GSM97125 1 0.191 0.77398 0.956 0.028 0.016
#> GSM97127 1 0.191 0.77398 0.956 0.028 0.016
#> GSM97130 1 0.898 0.20433 0.548 0.284 0.168
#> GSM97133 1 0.192 0.76695 0.956 0.020 0.024
#> GSM97134 2 0.836 -0.11222 0.384 0.528 0.088
#> GSM97120 1 0.192 0.76614 0.956 0.020 0.024
#> GSM97126 2 0.866 -0.16524 0.408 0.488 0.104
#> GSM97112 1 0.492 0.76690 0.832 0.036 0.132
#> GSM97115 2 0.922 -0.29130 0.376 0.468 0.156
#> GSM97116 1 0.301 0.77399 0.920 0.028 0.052
#> GSM97117 2 0.518 0.39437 0.156 0.812 0.032
#> GSM97119 1 0.492 0.76690 0.832 0.036 0.132
#> GSM97122 1 0.492 0.76690 0.832 0.036 0.132
#> GSM97135 1 0.492 0.76690 0.832 0.036 0.132
#> GSM97136 2 0.900 -0.08467 0.312 0.532 0.156
#> GSM97139 1 0.164 0.76808 0.964 0.016 0.020
#> GSM97146 1 0.145 0.76260 0.968 0.008 0.024
#> GSM97123 2 0.553 0.45633 0.000 0.704 0.296
#> GSM97129 2 0.835 -0.10669 0.380 0.532 0.088
#> GSM97143 1 0.884 0.41167 0.580 0.208 0.212
#> GSM97113 2 0.634 0.32151 0.252 0.716 0.032
#> GSM97056 1 0.606 0.67356 0.784 0.084 0.132
#> GSM97124 1 0.550 0.76254 0.804 0.048 0.148
#> GSM97132 1 0.907 0.25190 0.544 0.272 0.184
#> GSM97144 1 0.944 -0.00238 0.480 0.324 0.196
#> GSM97149 1 0.145 0.76260 0.968 0.008 0.024
#> GSM97068 2 0.894 -0.27052 0.368 0.500 0.132
#> GSM97071 2 0.576 -0.14019 0.008 0.716 0.276
#> GSM97086 2 0.493 0.12012 0.004 0.784 0.212
#> GSM97103 2 0.626 0.43558 0.004 0.616 0.380
#> GSM97057 2 0.634 0.32028 0.252 0.716 0.032
#> GSM97060 2 0.628 0.38697 0.000 0.540 0.460
#> GSM97075 2 0.649 0.36854 0.064 0.744 0.192
#> GSM97098 2 0.623 0.43849 0.004 0.624 0.372
#> GSM97099 2 0.556 0.39939 0.152 0.800 0.048
#> GSM97101 2 0.543 0.40441 0.144 0.808 0.048
#> GSM97105 2 0.238 0.44314 0.008 0.936 0.056
#> GSM97106 2 0.627 0.39318 0.000 0.548 0.452
#> GSM97121 2 0.518 0.37407 0.164 0.808 0.028
#> GSM97128 3 0.784 0.99129 0.052 0.460 0.488
#> GSM97131 2 0.296 0.45179 0.000 0.900 0.100
#> GSM97137 1 0.884 0.25205 0.568 0.268 0.164
#> GSM97118 1 0.924 0.31149 0.532 0.220 0.248
#> GSM97114 2 0.518 0.39437 0.156 0.812 0.032
#> GSM97142 1 0.492 0.76690 0.832 0.036 0.132
#> GSM97140 2 0.369 0.38397 0.100 0.884 0.016
#> GSM97141 2 0.529 0.40672 0.148 0.812 0.040
#> GSM97055 1 0.996 -0.28877 0.376 0.324 0.300
#> GSM97090 2 0.906 -0.27023 0.364 0.492 0.144
#> GSM97091 1 0.595 0.73669 0.776 0.052 0.172
#> GSM97148 1 0.145 0.76260 0.968 0.008 0.024
#> GSM97063 1 0.595 0.73669 0.776 0.052 0.172
#> GSM97053 1 0.480 0.76885 0.836 0.032 0.132
#> GSM97066 2 0.615 0.43144 0.000 0.592 0.408
#> GSM97079 2 0.498 0.09545 0.004 0.780 0.216
#> GSM97083 3 0.792 0.98266 0.056 0.456 0.488
#> GSM97084 2 0.540 -0.02875 0.004 0.740 0.256
#> GSM97094 2 0.544 -0.03876 0.004 0.736 0.260
#> GSM97096 2 0.623 0.43849 0.004 0.624 0.372
#> GSM97097 2 0.544 -0.03876 0.004 0.736 0.260
#> GSM97107 2 0.573 -0.09367 0.008 0.720 0.272
#> GSM97054 2 0.576 -0.14019 0.008 0.716 0.276
#> GSM97062 2 0.498 0.09545 0.004 0.780 0.216
#> GSM97069 2 0.615 0.43817 0.000 0.592 0.408
#> GSM97070 2 0.614 0.43378 0.000 0.596 0.404
#> GSM97073 2 0.623 0.43975 0.004 0.624 0.372
#> GSM97076 2 0.902 0.00457 0.212 0.560 0.228
#> GSM97077 2 0.285 0.42134 0.056 0.924 0.020
#> GSM97095 2 0.873 -0.24456 0.316 0.552 0.132
#> GSM97102 2 0.626 0.43558 0.004 0.616 0.380
#> GSM97109 2 0.777 0.39229 0.088 0.640 0.272
#> GSM97110 2 0.777 0.39229 0.088 0.640 0.272
#> GSM97074 2 0.843 -0.64225 0.088 0.500 0.412
#> GSM97085 2 0.740 -0.77695 0.032 0.488 0.480
#> GSM97059 2 0.538 0.28765 0.188 0.788 0.024
#> GSM97072 2 0.628 0.40389 0.000 0.540 0.460
#> GSM97078 3 0.784 0.99129 0.052 0.460 0.488
#> GSM97067 2 0.615 0.43563 0.000 0.592 0.408
#> GSM97087 2 0.626 0.39373 0.000 0.552 0.448
#> GSM97111 2 0.657 0.38067 0.088 0.752 0.160
#> GSM97064 2 0.325 0.44672 0.036 0.912 0.052
#> GSM97065 2 0.860 0.11258 0.184 0.604 0.212
#> GSM97081 2 0.645 0.37434 0.056 0.740 0.204
#> GSM97082 2 0.621 0.40812 0.000 0.572 0.428
#> GSM97088 2 0.740 -0.77695 0.032 0.488 0.480
#> GSM97100 2 0.361 0.39372 0.096 0.888 0.016
#> GSM97104 2 0.629 0.39829 0.000 0.536 0.464
#> GSM97108 2 0.500 0.41287 0.124 0.832 0.044
#> GSM97050 2 0.337 0.44777 0.040 0.908 0.052
#> GSM97080 2 0.610 0.44223 0.000 0.608 0.392
#> GSM97089 2 0.626 0.39373 0.000 0.552 0.448
#> GSM97092 2 0.604 0.42901 0.000 0.620 0.380
#> GSM97093 2 0.799 0.12974 0.292 0.616 0.092
#> GSM97058 2 0.315 0.44448 0.036 0.916 0.048
#> GSM97051 2 0.236 0.44326 0.000 0.928 0.072
#> GSM97052 2 0.618 0.41321 0.000 0.584 0.416
#> GSM97061 2 0.576 0.44294 0.000 0.672 0.328
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM97138 1 0.298 0.75628 0.888 0.016 0.004 0.092
#> GSM97145 1 0.151 0.76252 0.956 0.028 0.000 0.016
#> GSM97147 2 0.441 0.57761 0.152 0.808 0.012 0.028
#> GSM97125 1 0.151 0.76252 0.956 0.028 0.000 0.016
#> GSM97127 1 0.151 0.76252 0.956 0.028 0.000 0.016
#> GSM97130 1 0.761 0.29315 0.536 0.196 0.012 0.256
#> GSM97133 1 0.148 0.75457 0.960 0.020 0.004 0.016
#> GSM97134 2 0.777 0.27409 0.360 0.460 0.012 0.168
#> GSM97120 1 0.112 0.75680 0.972 0.012 0.004 0.012
#> GSM97126 2 0.795 0.15601 0.384 0.412 0.012 0.192
#> GSM97112 1 0.359 0.74416 0.824 0.008 0.000 0.168
#> GSM97115 2 0.827 0.13526 0.368 0.372 0.016 0.244
#> GSM97116 1 0.216 0.76131 0.928 0.008 0.004 0.060
#> GSM97117 2 0.431 0.56097 0.152 0.812 0.024 0.012
#> GSM97119 1 0.359 0.74416 0.824 0.008 0.000 0.168
#> GSM97122 1 0.359 0.74416 0.824 0.008 0.000 0.168
#> GSM97135 1 0.359 0.74416 0.824 0.008 0.000 0.168
#> GSM97136 2 0.900 0.18394 0.300 0.404 0.068 0.228
#> GSM97139 1 0.111 0.75851 0.972 0.008 0.004 0.016
#> GSM97146 1 0.123 0.75481 0.968 0.008 0.004 0.020
#> GSM97123 3 0.569 0.52137 0.000 0.464 0.512 0.024
#> GSM97129 2 0.777 0.27811 0.356 0.464 0.012 0.168
#> GSM97143 1 0.663 0.36889 0.564 0.064 0.012 0.360
#> GSM97113 2 0.525 0.53881 0.252 0.712 0.028 0.008
#> GSM97056 1 0.510 0.64154 0.772 0.056 0.012 0.160
#> GSM97124 1 0.404 0.74349 0.804 0.020 0.000 0.176
#> GSM97132 1 0.734 0.31201 0.532 0.168 0.004 0.296
#> GSM97144 1 0.796 0.16436 0.472 0.228 0.012 0.288
#> GSM97149 1 0.123 0.75481 0.968 0.008 0.004 0.020
#> GSM97068 2 0.815 0.19181 0.352 0.416 0.016 0.216
#> GSM97071 2 0.637 0.26368 0.008 0.564 0.052 0.376
#> GSM97086 2 0.570 0.41539 0.004 0.680 0.052 0.264
#> GSM97103 3 0.628 0.66144 0.008 0.304 0.624 0.064
#> GSM97057 2 0.519 0.53868 0.256 0.712 0.024 0.008
#> GSM97060 3 0.443 0.70531 0.000 0.204 0.772 0.024
#> GSM97075 2 0.782 0.27358 0.072 0.604 0.168 0.156
#> GSM97098 3 0.634 0.64932 0.008 0.316 0.612 0.064
#> GSM97099 2 0.467 0.55552 0.148 0.800 0.036 0.016
#> GSM97101 2 0.482 0.55003 0.140 0.796 0.048 0.016
#> GSM97105 2 0.317 0.45028 0.004 0.872 0.112 0.012
#> GSM97106 3 0.457 0.70528 0.000 0.220 0.756 0.024
#> GSM97121 2 0.538 0.56524 0.156 0.764 0.056 0.024
#> GSM97128 4 0.318 0.80151 0.032 0.052 0.020 0.896
#> GSM97131 2 0.409 0.38348 0.000 0.804 0.172 0.024
#> GSM97137 1 0.761 0.31957 0.548 0.184 0.016 0.252
#> GSM97118 1 0.676 0.22299 0.512 0.064 0.012 0.412
#> GSM97114 2 0.431 0.56097 0.152 0.812 0.024 0.012
#> GSM97142 1 0.359 0.74416 0.824 0.008 0.000 0.168
#> GSM97140 2 0.390 0.56371 0.092 0.856 0.032 0.020
#> GSM97141 2 0.458 0.55698 0.148 0.804 0.032 0.016
#> GSM97055 4 0.754 0.19184 0.372 0.108 0.024 0.496
#> GSM97090 2 0.822 0.16721 0.356 0.396 0.016 0.232
#> GSM97091 1 0.439 0.70107 0.768 0.012 0.004 0.216
#> GSM97148 1 0.123 0.75481 0.968 0.008 0.004 0.020
#> GSM97063 1 0.439 0.70107 0.768 0.012 0.004 0.216
#> GSM97053 1 0.354 0.74762 0.828 0.008 0.000 0.164
#> GSM97066 3 0.569 0.67758 0.000 0.240 0.688 0.072
#> GSM97079 2 0.570 0.40754 0.004 0.680 0.052 0.264
#> GSM97083 4 0.316 0.79238 0.036 0.052 0.016 0.896
#> GSM97084 2 0.604 0.33684 0.004 0.620 0.052 0.324
#> GSM97094 2 0.606 0.33047 0.004 0.616 0.052 0.328
#> GSM97096 3 0.634 0.64932 0.008 0.316 0.612 0.064
#> GSM97097 2 0.606 0.33047 0.004 0.616 0.052 0.328
#> GSM97107 2 0.632 0.28490 0.008 0.580 0.052 0.360
#> GSM97054 2 0.637 0.26368 0.008 0.564 0.052 0.376
#> GSM97062 2 0.570 0.40754 0.004 0.680 0.052 0.264
#> GSM97069 3 0.508 0.73098 0.000 0.248 0.716 0.036
#> GSM97070 3 0.572 0.68030 0.000 0.244 0.684 0.072
#> GSM97073 3 0.639 0.61143 0.004 0.292 0.620 0.084
#> GSM97076 2 0.971 -0.00147 0.216 0.376 0.224 0.184
#> GSM97077 2 0.347 0.53198 0.052 0.880 0.056 0.012
#> GSM97095 2 0.787 0.29011 0.312 0.488 0.016 0.184
#> GSM97102 3 0.628 0.66144 0.008 0.304 0.624 0.064
#> GSM97109 2 0.837 -0.08601 0.092 0.472 0.344 0.092
#> GSM97110 2 0.837 -0.08601 0.092 0.472 0.344 0.092
#> GSM97074 4 0.761 0.67882 0.080 0.156 0.136 0.628
#> GSM97085 4 0.570 0.78460 0.024 0.092 0.132 0.752
#> GSM97059 2 0.479 0.56630 0.184 0.776 0.016 0.024
#> GSM97072 3 0.440 0.71808 0.000 0.212 0.768 0.020
#> GSM97078 4 0.318 0.80151 0.032 0.052 0.020 0.896
#> GSM97067 3 0.522 0.70059 0.000 0.244 0.712 0.044
#> GSM97087 3 0.460 0.70852 0.000 0.212 0.760 0.028
#> GSM97111 2 0.759 0.34764 0.100 0.636 0.140 0.124
#> GSM97064 2 0.415 0.48275 0.032 0.836 0.116 0.016
#> GSM97065 2 0.955 0.08525 0.192 0.404 0.236 0.168
#> GSM97081 2 0.791 0.22034 0.064 0.588 0.180 0.168
#> GSM97082 3 0.650 0.70026 0.000 0.216 0.636 0.148
#> GSM97088 4 0.570 0.78460 0.024 0.092 0.132 0.752
#> GSM97100 2 0.382 0.56080 0.088 0.860 0.036 0.016
#> GSM97104 3 0.363 0.73078 0.000 0.160 0.828 0.012
#> GSM97108 2 0.464 0.54788 0.120 0.812 0.052 0.016
#> GSM97050 2 0.419 0.48384 0.032 0.840 0.104 0.024
#> GSM97080 3 0.537 0.73647 0.000 0.276 0.684 0.040
#> GSM97089 3 0.460 0.70852 0.000 0.212 0.760 0.028
#> GSM97092 3 0.527 0.68689 0.000 0.320 0.656 0.024
#> GSM97093 2 0.778 0.45193 0.288 0.556 0.060 0.096
#> GSM97058 2 0.404 0.48962 0.032 0.844 0.108 0.016
#> GSM97051 2 0.401 0.40498 0.000 0.816 0.156 0.028
#> GSM97052 3 0.493 0.70845 0.000 0.264 0.712 0.024
#> GSM97061 3 0.561 0.58234 0.000 0.412 0.564 0.024
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM97138 1 0.324 0.7190 0.856 0.028 0.000 0.012 0.104
#> GSM97145 1 0.117 0.7356 0.960 0.032 0.000 0.000 0.008
#> GSM97147 2 0.331 0.6126 0.124 0.844 0.000 0.020 0.012
#> GSM97125 1 0.117 0.7356 0.960 0.032 0.000 0.000 0.008
#> GSM97127 1 0.117 0.7356 0.960 0.032 0.000 0.000 0.008
#> GSM97130 1 0.722 0.3689 0.524 0.136 0.000 0.260 0.080
#> GSM97133 1 0.206 0.7259 0.928 0.036 0.000 0.012 0.024
#> GSM97134 2 0.715 0.3014 0.356 0.456 0.004 0.036 0.148
#> GSM97120 1 0.169 0.7291 0.944 0.028 0.000 0.008 0.020
#> GSM97126 2 0.725 0.1915 0.368 0.412 0.008 0.020 0.192
#> GSM97112 1 0.317 0.7088 0.828 0.008 0.000 0.004 0.160
#> GSM97115 1 0.799 -0.1277 0.340 0.328 0.000 0.248 0.084
#> GSM97116 1 0.272 0.7283 0.892 0.028 0.000 0.012 0.068
#> GSM97117 2 0.336 0.6352 0.120 0.844 0.024 0.000 0.012
#> GSM97119 1 0.317 0.7088 0.828 0.008 0.000 0.004 0.160
#> GSM97122 1 0.317 0.7088 0.828 0.008 0.000 0.004 0.160
#> GSM97135 1 0.317 0.7088 0.828 0.008 0.000 0.004 0.160
#> GSM97136 2 0.838 0.2508 0.292 0.396 0.084 0.024 0.204
#> GSM97139 1 0.189 0.7289 0.936 0.028 0.000 0.012 0.024
#> GSM97146 1 0.218 0.7230 0.924 0.028 0.000 0.020 0.028
#> GSM97123 3 0.565 0.3992 0.000 0.460 0.472 0.064 0.004
#> GSM97129 2 0.714 0.3101 0.352 0.460 0.004 0.036 0.148
#> GSM97143 1 0.605 0.3473 0.560 0.056 0.016 0.012 0.356
#> GSM97113 2 0.467 0.5816 0.200 0.744 0.020 0.032 0.004
#> GSM97056 1 0.505 0.6316 0.748 0.044 0.000 0.140 0.068
#> GSM97124 1 0.382 0.7102 0.808 0.020 0.000 0.020 0.152
#> GSM97132 1 0.754 0.4019 0.516 0.116 0.000 0.192 0.176
#> GSM97144 1 0.760 0.2416 0.464 0.148 0.000 0.292 0.096
#> GSM97149 1 0.218 0.7230 0.924 0.028 0.000 0.020 0.028
#> GSM97068 2 0.767 0.0324 0.340 0.384 0.000 0.216 0.060
#> GSM97071 4 0.516 0.8520 0.000 0.248 0.004 0.672 0.076
#> GSM97086 4 0.448 0.7734 0.000 0.416 0.000 0.576 0.008
#> GSM97103 3 0.509 0.6581 0.004 0.212 0.716 0.036 0.032
#> GSM97057 2 0.452 0.5797 0.204 0.748 0.012 0.032 0.004
#> GSM97060 3 0.514 0.6441 0.000 0.200 0.696 0.100 0.004
#> GSM97075 2 0.750 0.2127 0.048 0.576 0.200 0.064 0.112
#> GSM97098 3 0.511 0.6516 0.004 0.224 0.708 0.032 0.032
#> GSM97099 2 0.405 0.6355 0.108 0.824 0.036 0.016 0.016
#> GSM97101 2 0.415 0.6366 0.104 0.820 0.044 0.016 0.016
#> GSM97105 2 0.320 0.5434 0.000 0.852 0.096 0.052 0.000
#> GSM97106 3 0.523 0.6404 0.000 0.212 0.684 0.100 0.004
#> GSM97121 2 0.499 0.6005 0.148 0.760 0.048 0.028 0.016
#> GSM97128 5 0.248 0.7609 0.028 0.004 0.000 0.068 0.900
#> GSM97131 2 0.423 0.4827 0.000 0.776 0.140 0.084 0.000
#> GSM97137 1 0.713 0.3950 0.536 0.128 0.000 0.256 0.080
#> GSM97118 1 0.618 0.2277 0.512 0.052 0.016 0.016 0.404
#> GSM97114 2 0.336 0.6352 0.120 0.844 0.024 0.000 0.012
#> GSM97142 1 0.317 0.7088 0.828 0.008 0.000 0.004 0.160
#> GSM97140 2 0.318 0.6152 0.072 0.876 0.020 0.024 0.008
#> GSM97141 2 0.332 0.6355 0.116 0.844 0.036 0.000 0.004
#> GSM97055 5 0.774 0.2234 0.336 0.104 0.020 0.084 0.456
#> GSM97090 2 0.792 -0.0298 0.328 0.360 0.000 0.232 0.080
#> GSM97091 1 0.395 0.6639 0.768 0.012 0.000 0.012 0.208
#> GSM97148 1 0.218 0.7230 0.924 0.028 0.000 0.020 0.028
#> GSM97063 1 0.395 0.6639 0.768 0.012 0.000 0.012 0.208
#> GSM97053 1 0.301 0.7121 0.832 0.008 0.000 0.000 0.160
#> GSM97066 3 0.420 0.6424 0.000 0.156 0.788 0.036 0.020
#> GSM97079 4 0.446 0.7932 0.000 0.408 0.000 0.584 0.008
#> GSM97083 5 0.279 0.7496 0.028 0.004 0.000 0.088 0.880
#> GSM97084 4 0.394 0.8786 0.000 0.272 0.004 0.720 0.004
#> GSM97094 4 0.412 0.8813 0.000 0.264 0.004 0.720 0.012
#> GSM97096 3 0.511 0.6516 0.004 0.224 0.708 0.032 0.032
#> GSM97097 4 0.399 0.8796 0.000 0.260 0.004 0.728 0.008
#> GSM97107 4 0.471 0.8666 0.004 0.240 0.004 0.712 0.040
#> GSM97054 4 0.516 0.8520 0.000 0.248 0.004 0.672 0.076
#> GSM97062 4 0.446 0.7932 0.000 0.408 0.000 0.584 0.008
#> GSM97069 3 0.363 0.6883 0.000 0.176 0.800 0.020 0.004
#> GSM97070 3 0.424 0.6465 0.000 0.160 0.784 0.036 0.020
#> GSM97073 3 0.520 0.6118 0.000 0.200 0.712 0.052 0.036
#> GSM97076 3 0.943 -0.0345 0.176 0.300 0.300 0.108 0.116
#> GSM97077 2 0.304 0.6001 0.040 0.888 0.044 0.020 0.008
#> GSM97095 2 0.731 0.0879 0.288 0.464 0.000 0.204 0.044
#> GSM97102 3 0.509 0.6581 0.004 0.212 0.716 0.036 0.032
#> GSM97109 3 0.751 0.2730 0.068 0.388 0.444 0.040 0.060
#> GSM97110 3 0.751 0.2730 0.068 0.388 0.444 0.040 0.060
#> GSM97074 5 0.663 0.6635 0.080 0.104 0.160 0.012 0.644
#> GSM97085 5 0.484 0.7545 0.020 0.044 0.148 0.020 0.768
#> GSM97059 2 0.377 0.5900 0.156 0.808 0.000 0.020 0.016
#> GSM97072 3 0.353 0.6676 0.000 0.128 0.824 0.048 0.000
#> GSM97078 5 0.242 0.7616 0.028 0.004 0.000 0.064 0.904
#> GSM97067 3 0.369 0.6639 0.000 0.156 0.808 0.032 0.004
#> GSM97087 3 0.510 0.6434 0.000 0.208 0.696 0.092 0.004
#> GSM97111 2 0.731 0.3454 0.076 0.612 0.164 0.064 0.084
#> GSM97064 2 0.342 0.5740 0.020 0.852 0.104 0.020 0.004
#> GSM97065 2 0.919 -0.1051 0.144 0.336 0.312 0.108 0.100
#> GSM97081 2 0.760 0.1478 0.040 0.560 0.212 0.076 0.112
#> GSM97082 3 0.639 0.6480 0.000 0.208 0.628 0.080 0.084
#> GSM97088 5 0.484 0.7545 0.020 0.044 0.148 0.020 0.768
#> GSM97100 2 0.330 0.6114 0.068 0.872 0.024 0.028 0.008
#> GSM97104 3 0.411 0.6769 0.000 0.140 0.784 0.076 0.000
#> GSM97108 2 0.406 0.6345 0.092 0.828 0.048 0.020 0.012
#> GSM97050 2 0.362 0.5795 0.024 0.852 0.084 0.032 0.008
#> GSM97080 3 0.432 0.6887 0.000 0.212 0.748 0.032 0.008
#> GSM97089 3 0.510 0.6434 0.000 0.208 0.696 0.092 0.004
#> GSM97092 3 0.547 0.5897 0.000 0.320 0.604 0.072 0.004
#> GSM97093 2 0.739 0.4679 0.252 0.560 0.072 0.048 0.068
#> GSM97058 2 0.331 0.5767 0.020 0.860 0.096 0.020 0.004
#> GSM97051 2 0.419 0.4940 0.000 0.796 0.128 0.064 0.012
#> GSM97052 3 0.530 0.6214 0.000 0.272 0.648 0.076 0.004
#> GSM97061 3 0.566 0.4754 0.000 0.412 0.516 0.068 0.004
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM97138 1 0.4071 0.70473 0.796 0.024 0.000 0.008 0.072 0.100
#> GSM97145 1 0.1624 0.72057 0.936 0.040 0.000 0.000 0.004 0.020
#> GSM97147 2 0.3573 0.65114 0.076 0.836 0.000 0.032 0.008 0.048
#> GSM97125 1 0.1552 0.72160 0.940 0.036 0.000 0.000 0.004 0.020
#> GSM97127 1 0.1624 0.72057 0.936 0.040 0.000 0.000 0.004 0.020
#> GSM97130 1 0.7545 0.30891 0.460 0.096 0.000 0.276 0.072 0.096
#> GSM97133 1 0.2867 0.70476 0.848 0.040 0.000 0.000 0.000 0.112
#> GSM97134 2 0.7452 0.34739 0.312 0.448 0.004 0.052 0.108 0.076
#> GSM97120 1 0.2848 0.70741 0.856 0.036 0.000 0.004 0.000 0.104
#> GSM97126 2 0.7538 0.23952 0.332 0.404 0.008 0.020 0.136 0.100
#> GSM97112 1 0.2821 0.69662 0.832 0.000 0.000 0.000 0.152 0.016
#> GSM97115 2 0.8236 0.07916 0.272 0.292 0.000 0.268 0.060 0.108
#> GSM97116 1 0.3705 0.71051 0.812 0.024 0.000 0.004 0.040 0.120
#> GSM97117 2 0.2903 0.64462 0.076 0.864 0.004 0.000 0.004 0.052
#> GSM97119 1 0.2821 0.69662 0.832 0.000 0.000 0.000 0.152 0.016
#> GSM97122 1 0.2821 0.69662 0.832 0.000 0.000 0.000 0.152 0.016
#> GSM97135 1 0.2821 0.69662 0.832 0.000 0.000 0.000 0.152 0.016
#> GSM97136 2 0.8303 0.11423 0.260 0.368 0.068 0.004 0.124 0.176
#> GSM97139 1 0.2776 0.70791 0.860 0.032 0.000 0.004 0.000 0.104
#> GSM97146 1 0.3164 0.69425 0.824 0.032 0.000 0.004 0.000 0.140
#> GSM97123 3 0.5862 0.28421 0.000 0.388 0.476 0.012 0.004 0.120
#> GSM97129 2 0.7411 0.35285 0.312 0.452 0.004 0.052 0.108 0.072
#> GSM97143 1 0.6179 0.31974 0.544 0.020 0.012 0.004 0.292 0.128
#> GSM97113 2 0.4358 0.61178 0.144 0.756 0.016 0.004 0.000 0.080
#> GSM97056 1 0.5512 0.60148 0.684 0.016 0.000 0.144 0.044 0.112
#> GSM97124 1 0.3708 0.69834 0.808 0.012 0.000 0.024 0.136 0.020
#> GSM97132 1 0.7730 0.36662 0.476 0.080 0.000 0.200 0.148 0.096
#> GSM97144 1 0.7462 0.20772 0.432 0.092 0.000 0.324 0.088 0.064
#> GSM97149 1 0.3164 0.69425 0.824 0.032 0.000 0.004 0.000 0.140
#> GSM97068 2 0.8099 0.15626 0.268 0.344 0.000 0.228 0.052 0.108
#> GSM97071 4 0.3217 0.79323 0.000 0.044 0.000 0.852 0.068 0.036
#> GSM97086 4 0.3819 0.70061 0.000 0.280 0.000 0.700 0.000 0.020
#> GSM97103 3 0.3548 0.47914 0.000 0.048 0.796 0.000 0.004 0.152
#> GSM97057 2 0.4209 0.61285 0.148 0.760 0.008 0.004 0.000 0.080
#> GSM97060 3 0.3659 0.58433 0.000 0.012 0.752 0.012 0.000 0.224
#> GSM97075 2 0.6867 -0.00532 0.020 0.472 0.184 0.000 0.040 0.284
#> GSM97098 3 0.3728 0.47243 0.000 0.060 0.784 0.000 0.004 0.152
#> GSM97099 2 0.3862 0.62884 0.064 0.804 0.020 0.000 0.004 0.108
#> GSM97101 2 0.4019 0.62950 0.064 0.796 0.028 0.000 0.004 0.108
#> GSM97105 2 0.3162 0.60877 0.000 0.860 0.064 0.040 0.004 0.032
#> GSM97106 3 0.4312 0.58391 0.000 0.032 0.728 0.020 0.004 0.216
#> GSM97121 2 0.4536 0.63951 0.116 0.780 0.032 0.032 0.008 0.032
#> GSM97128 5 0.0260 0.69517 0.008 0.000 0.000 0.000 0.992 0.000
#> GSM97131 2 0.4346 0.56550 0.000 0.776 0.112 0.060 0.004 0.048
#> GSM97137 1 0.7520 0.32931 0.468 0.092 0.000 0.268 0.072 0.100
#> GSM97118 1 0.6216 0.21149 0.500 0.016 0.012 0.004 0.344 0.124
#> GSM97114 2 0.2903 0.64462 0.076 0.864 0.004 0.000 0.004 0.052
#> GSM97142 1 0.2821 0.69662 0.832 0.000 0.000 0.000 0.152 0.016
#> GSM97140 2 0.3153 0.65106 0.036 0.872 0.016 0.028 0.004 0.044
#> GSM97141 2 0.2797 0.64770 0.076 0.872 0.016 0.000 0.000 0.036
#> GSM97055 5 0.7090 0.20008 0.296 0.032 0.024 0.000 0.404 0.244
#> GSM97090 2 0.8153 0.12067 0.264 0.324 0.000 0.252 0.056 0.104
#> GSM97091 1 0.3618 0.65506 0.768 0.000 0.000 0.000 0.192 0.040
#> GSM97148 1 0.3164 0.69425 0.824 0.032 0.000 0.004 0.000 0.140
#> GSM97063 1 0.3618 0.65506 0.768 0.000 0.000 0.000 0.192 0.040
#> GSM97053 1 0.2869 0.70111 0.832 0.000 0.000 0.000 0.148 0.020
#> GSM97066 3 0.3650 0.44126 0.000 0.024 0.756 0.000 0.004 0.216
#> GSM97079 4 0.3799 0.71102 0.000 0.276 0.000 0.704 0.000 0.020
#> GSM97083 5 0.0862 0.68804 0.008 0.000 0.000 0.016 0.972 0.004
#> GSM97084 4 0.1265 0.82673 0.000 0.044 0.000 0.948 0.000 0.008
#> GSM97094 4 0.1082 0.82853 0.000 0.040 0.000 0.956 0.004 0.000
#> GSM97096 3 0.3728 0.47243 0.000 0.060 0.784 0.000 0.004 0.152
#> GSM97097 4 0.0865 0.82622 0.000 0.036 0.000 0.964 0.000 0.000
#> GSM97107 4 0.1760 0.81027 0.004 0.020 0.000 0.936 0.028 0.012
#> GSM97054 4 0.3217 0.79323 0.000 0.044 0.000 0.852 0.068 0.036
#> GSM97062 4 0.3778 0.71541 0.000 0.272 0.000 0.708 0.000 0.020
#> GSM97069 3 0.2250 0.55294 0.000 0.020 0.888 0.000 0.000 0.092
#> GSM97070 3 0.3622 0.44586 0.000 0.024 0.760 0.000 0.004 0.212
#> GSM97073 3 0.4495 0.31444 0.000 0.044 0.664 0.000 0.008 0.284
#> GSM97076 6 0.6904 0.86097 0.136 0.088 0.244 0.000 0.016 0.516
#> GSM97077 2 0.2556 0.64348 0.016 0.904 0.028 0.020 0.004 0.028
#> GSM97095 2 0.7649 0.23509 0.236 0.416 0.000 0.228 0.036 0.084
#> GSM97102 3 0.3548 0.47914 0.000 0.048 0.796 0.000 0.004 0.152
#> GSM97109 3 0.6789 -0.25389 0.044 0.264 0.472 0.004 0.004 0.212
#> GSM97110 3 0.6789 -0.25389 0.044 0.264 0.472 0.004 0.004 0.212
#> GSM97074 5 0.6750 0.44248 0.076 0.020 0.116 0.004 0.556 0.228
#> GSM97085 5 0.5080 0.62204 0.016 0.008 0.116 0.004 0.704 0.152
#> GSM97059 2 0.4227 0.63073 0.104 0.788 0.000 0.036 0.008 0.064
#> GSM97072 3 0.3168 0.52847 0.000 0.016 0.792 0.000 0.000 0.192
#> GSM97078 5 0.0622 0.69649 0.012 0.000 0.000 0.000 0.980 0.008
#> GSM97067 3 0.3263 0.48519 0.000 0.020 0.800 0.000 0.004 0.176
#> GSM97087 3 0.3534 0.58435 0.000 0.016 0.740 0.000 0.000 0.244
#> GSM97111 2 0.6389 0.24765 0.040 0.560 0.132 0.000 0.020 0.248
#> GSM97064 2 0.3558 0.60344 0.008 0.832 0.088 0.020 0.000 0.052
#> GSM97065 6 0.7037 0.86589 0.100 0.148 0.248 0.000 0.012 0.492
#> GSM97081 2 0.6921 -0.07408 0.016 0.444 0.204 0.000 0.040 0.296
#> GSM97082 3 0.4496 0.55675 0.000 0.008 0.708 0.000 0.076 0.208
#> GSM97088 5 0.5045 0.62246 0.016 0.008 0.116 0.004 0.708 0.148
#> GSM97100 2 0.3088 0.65076 0.032 0.876 0.016 0.032 0.004 0.040
#> GSM97104 3 0.2917 0.60133 0.000 0.016 0.840 0.008 0.000 0.136
#> GSM97108 2 0.4050 0.63022 0.056 0.804 0.036 0.004 0.004 0.096
#> GSM97050 2 0.3060 0.62108 0.004 0.868 0.056 0.016 0.004 0.052
#> GSM97080 3 0.2398 0.57726 0.000 0.020 0.876 0.000 0.000 0.104
#> GSM97089 3 0.3534 0.58435 0.000 0.016 0.740 0.000 0.000 0.244
#> GSM97092 3 0.5414 0.51199 0.000 0.184 0.628 0.008 0.004 0.176
#> GSM97093 2 0.7288 0.47043 0.180 0.548 0.056 0.036 0.028 0.152
#> GSM97058 2 0.3402 0.61140 0.008 0.844 0.076 0.020 0.000 0.052
#> GSM97051 2 0.4246 0.56448 0.000 0.784 0.108 0.052 0.004 0.052
#> GSM97052 3 0.4803 0.56443 0.000 0.108 0.684 0.008 0.000 0.200
#> GSM97061 3 0.5986 0.38338 0.000 0.308 0.516 0.012 0.004 0.160
Heatmaps for the consensus matrix. It visualizes the probability of two samples to be in a same group.
consensus_heatmap(res, k = 2)
consensus_heatmap(res, k = 3)
consensus_heatmap(res, k = 4)
consensus_heatmap(res, k = 5)
consensus_heatmap(res, k = 6)
Heatmaps for the membership of samples in all partitions to see how consistent they are:
membership_heatmap(res, k = 2)
membership_heatmap(res, k = 3)
membership_heatmap(res, k = 4)
membership_heatmap(res, k = 5)
membership_heatmap(res, k = 6)
As soon as we have had the classes for columns, we can look for signatures which are significantly different between classes which can be candidate marks for certain classes. Following are the heatmaps for signatures.
Signature heatmaps where rows are scaled:
get_signatures(res, k = 2)
get_signatures(res, k = 3)
get_signatures(res, k = 4)
get_signatures(res, k = 5)
get_signatures(res, k = 6)
Signature heatmaps where rows are not scaled:
get_signatures(res, k = 2, scale_rows = FALSE)
get_signatures(res, k = 3, scale_rows = FALSE)
get_signatures(res, k = 4, scale_rows = FALSE)
get_signatures(res, k = 5, scale_rows = FALSE)
get_signatures(res, k = 6, scale_rows = FALSE)
Compare the overlap of signatures from different k:
compare_signatures(res)
get_signature() returns a data frame invisibly. TO get the list of signatures, the function
call should be assigned to a variable explicitly. In following code, if plot argument is set
to FALSE, no heatmap is plotted while only the differential analysis is performed.
# code only for demonstration
tb = get_signature(res, k = ..., plot = FALSE)
An example of the output of tb is:
#> which_row fdr mean_1 mean_2 scaled_mean_1 scaled_mean_2 km
#> 1 38 0.042760348 8.373488 9.131774 -0.5533452 0.5164555 1
#> 2 40 0.018707592 7.106213 8.469186 -0.6173731 0.5762149 1
#> 3 55 0.019134737 10.221463 11.207825 -0.6159697 0.5749050 1
#> 4 59 0.006059896 5.921854 7.869574 -0.6899429 0.6439467 1
#> 5 60 0.018055526 8.928898 10.211722 -0.6204761 0.5791110 1
#> 6 98 0.009384629 15.714769 14.887706 0.6635654 -0.6193277 2
...
The columns in tb are:
which_row: row indices corresponding to the input matrix.fdr: FDR for the differential test. mean_x: The mean value in group x.scaled_mean_x: The mean value in group x after rows are scaled.km: Row groups if k-means clustering is applied to rows.UMAP plot which shows how samples are separated.
dimension_reduction(res, k = 2, method = "UMAP")
dimension_reduction(res, k = 3, method = "UMAP")
dimension_reduction(res, k = 4, method = "UMAP")
dimension_reduction(res, k = 5, method = "UMAP")
dimension_reduction(res, k = 6, method = "UMAP")
Following heatmap shows how subgroups are split when increasing k:
collect_classes(res)
Test correlation between subgroups and known annotations. If the known annotation is numeric, one-way ANOVA test is applied, and if the known annotation is discrete, chi-squared contingency table test is applied.
test_to_known_factors(res)
#> n disease.state(p) specimen(p) cell.type(p) other(p) k
#> MAD:hclust 84 1.73e-06 1.000 2.80e-16 0.0431 2
#> MAD:hclust 24 1.75e-01 1.000 6.14e-06 0.7279 3
#> MAD:hclust 62 1.61e-04 0.472 1.18e-10 0.2285 4
#> MAD:hclust 73 7.71e-05 0.701 1.95e-15 0.0744 5
#> MAD:hclust 68 1.63e-03 0.645 4.50e-15 0.0769 6
If matrix rows can be associated to genes, consider to use functional_enrichment(res,
...) to perform function enrichment for the signature genes. See this vignette for more detailed explanations.
The object with results only for a single top-value method and a single partition method can be extracted as:
res = res_list["MAD", "kmeans"]
# you can also extract it by
# res = res_list["MAD:kmeans"]
A summary of res and all the functions that can be applied to it:
res
#> A 'ConsensusPartition' object with k = 2, 3, 4, 5, 6.
#> On a matrix with 21512 rows and 100 columns.
#> Top rows (1000, 2000, 3000, 4000, 5000) are extracted by 'MAD' method.
#> Subgroups are detected by 'kmeans' method.
#> Performed in total 1250 partitions by row resampling.
#> Best k for subgroups seems to be 2.
#>
#> Following methods can be applied to this 'ConsensusPartition' object:
#> [1] "cola_report" "collect_classes" "collect_plots"
#> [4] "collect_stats" "colnames" "compare_signatures"
#> [7] "consensus_heatmap" "dimension_reduction" "functional_enrichment"
#> [10] "get_anno_col" "get_anno" "get_classes"
#> [13] "get_consensus" "get_matrix" "get_membership"
#> [16] "get_param" "get_signatures" "get_stats"
#> [19] "is_best_k" "is_stable_k" "membership_heatmap"
#> [22] "ncol" "nrow" "plot_ecdf"
#> [25] "rownames" "select_partition_number" "show"
#> [28] "suggest_best_k" "test_to_known_factors"
collect_plots() function collects all the plots made from res for all k (number of partitions)
into one single page to provide an easy and fast comparison between different k.
collect_plots(res)
The plots are:
k and the heatmap of
predicted classes for each k.k.k.k.All the plots in panels can be made by individual functions and they are plotted later in this section.
select_partition_number() produces several plots showing different
statistics for choosing “optimized” k. There are following statistics:
k;k, the area increased is defined as \(A_k - A_{k-1}\).The detailed explanations of these statistics can be found in the cola vignette.
Generally speaking, lower PAC score, higher mean silhouette score or higher
concordance corresponds to better partition. Rand index and Jaccard index
measure how similar the current partition is compared to partition with k-1.
If they are too similar, we won't accept k is better than k-1.
select_partition_number(res)
The numeric values for all these statistics can be obtained by get_stats().
get_stats(res)
#> k 1-PAC mean_silhouette concordance area_increased Rand Jaccard
#> 2 2 1.000 0.977 0.988 0.4919 0.508 0.508
#> 3 3 0.545 0.755 0.846 0.3360 0.725 0.509
#> 4 4 0.733 0.778 0.865 0.1317 0.836 0.560
#> 5 5 0.696 0.665 0.791 0.0609 0.960 0.843
#> 6 6 0.699 0.462 0.639 0.0390 0.941 0.755
suggest_best_k() suggests the best \(k\) based on these statistics. The rules are as follows:
suggest_best_k(res)
#> [1] 2
Following shows the table of the partitions (You need to click the show/hide
code output link to see it). The membership matrix (columns with name p*)
is inferred by
clue::cl_consensus()
function with the SE method. Basically the value in the membership matrix
represents the probability to belong to a certain group. The finall class
label for an item is determined with the group with highest probability it
belongs to.
In get_classes() function, the entropy is calculated from the membership
matrix and the silhouette score is calculated from the consensus matrix.
cbind(get_classes(res, k = 2), get_membership(res, k = 2))
#> class entropy silhouette p1 p2
#> GSM97138 1 0.0000 0.986 1.000 0.000
#> GSM97145 1 0.0000 0.986 1.000 0.000
#> GSM97147 1 0.0672 0.981 0.992 0.008
#> GSM97125 1 0.0000 0.986 1.000 0.000
#> GSM97127 1 0.0000 0.986 1.000 0.000
#> GSM97130 1 0.0000 0.986 1.000 0.000
#> GSM97133 1 0.0000 0.986 1.000 0.000
#> GSM97134 1 0.0000 0.986 1.000 0.000
#> GSM97120 1 0.0000 0.986 1.000 0.000
#> GSM97126 1 0.0000 0.986 1.000 0.000
#> GSM97112 1 0.0000 0.986 1.000 0.000
#> GSM97115 1 0.0376 0.984 0.996 0.004
#> GSM97116 1 0.0000 0.986 1.000 0.000
#> GSM97117 2 0.0938 0.984 0.012 0.988
#> GSM97119 1 0.0000 0.986 1.000 0.000
#> GSM97122 1 0.0000 0.986 1.000 0.000
#> GSM97135 1 0.0000 0.986 1.000 0.000
#> GSM97136 2 0.0938 0.984 0.012 0.988
#> GSM97139 1 0.0000 0.986 1.000 0.000
#> GSM97146 1 0.0000 0.986 1.000 0.000
#> GSM97123 2 0.0000 0.989 0.000 1.000
#> GSM97129 2 0.0938 0.984 0.012 0.988
#> GSM97143 1 0.0000 0.986 1.000 0.000
#> GSM97113 2 0.0938 0.984 0.012 0.988
#> GSM97056 1 0.0000 0.986 1.000 0.000
#> GSM97124 1 0.0000 0.986 1.000 0.000
#> GSM97132 1 0.0000 0.986 1.000 0.000
#> GSM97144 1 0.0000 0.986 1.000 0.000
#> GSM97149 1 0.0000 0.986 1.000 0.000
#> GSM97068 2 0.7299 0.749 0.204 0.796
#> GSM97071 2 0.0000 0.989 0.000 1.000
#> GSM97086 2 0.0000 0.989 0.000 1.000
#> GSM97103 2 0.0000 0.989 0.000 1.000
#> GSM97057 2 0.0938 0.984 0.012 0.988
#> GSM97060 2 0.0000 0.989 0.000 1.000
#> GSM97075 2 0.0000 0.989 0.000 1.000
#> GSM97098 2 0.0000 0.989 0.000 1.000
#> GSM97099 2 0.0938 0.984 0.012 0.988
#> GSM97101 2 0.0938 0.984 0.012 0.988
#> GSM97105 2 0.0000 0.989 0.000 1.000
#> GSM97106 2 0.0000 0.989 0.000 1.000
#> GSM97121 2 0.0938 0.984 0.012 0.988
#> GSM97128 1 0.4161 0.916 0.916 0.084
#> GSM97131 2 0.0000 0.989 0.000 1.000
#> GSM97137 1 0.0000 0.986 1.000 0.000
#> GSM97118 1 0.0000 0.986 1.000 0.000
#> GSM97114 2 0.6623 0.798 0.172 0.828
#> GSM97142 1 0.0000 0.986 1.000 0.000
#> GSM97140 2 0.0938 0.984 0.012 0.988
#> GSM97141 2 0.0938 0.984 0.012 0.988
#> GSM97055 1 0.0000 0.986 1.000 0.000
#> GSM97090 1 0.0376 0.984 0.996 0.004
#> GSM97091 1 0.0000 0.986 1.000 0.000
#> GSM97148 1 0.0000 0.986 1.000 0.000
#> GSM97063 1 0.0000 0.986 1.000 0.000
#> GSM97053 1 0.0000 0.986 1.000 0.000
#> GSM97066 2 0.0000 0.989 0.000 1.000
#> GSM97079 2 0.0000 0.989 0.000 1.000
#> GSM97083 1 0.0000 0.986 1.000 0.000
#> GSM97084 2 0.0000 0.989 0.000 1.000
#> GSM97094 1 0.0938 0.978 0.988 0.012
#> GSM97096 2 0.0000 0.989 0.000 1.000
#> GSM97097 2 0.0000 0.989 0.000 1.000
#> GSM97107 1 0.1414 0.973 0.980 0.020
#> GSM97054 2 0.0000 0.989 0.000 1.000
#> GSM97062 2 0.0000 0.989 0.000 1.000
#> GSM97069 2 0.0000 0.989 0.000 1.000
#> GSM97070 2 0.0000 0.989 0.000 1.000
#> GSM97073 2 0.0000 0.989 0.000 1.000
#> GSM97076 1 0.1633 0.970 0.976 0.024
#> GSM97077 2 0.0000 0.989 0.000 1.000
#> GSM97095 1 0.5059 0.878 0.888 0.112
#> GSM97102 2 0.0000 0.989 0.000 1.000
#> GSM97109 2 0.0938 0.984 0.012 0.988
#> GSM97110 2 0.0938 0.984 0.012 0.988
#> GSM97074 1 0.1414 0.973 0.980 0.020
#> GSM97085 2 0.1414 0.974 0.020 0.980
#> GSM97059 1 0.7453 0.739 0.788 0.212
#> GSM97072 2 0.0000 0.989 0.000 1.000
#> GSM97078 1 0.4161 0.916 0.916 0.084
#> GSM97067 2 0.0000 0.989 0.000 1.000
#> GSM97087 2 0.0000 0.989 0.000 1.000
#> GSM97111 2 0.0938 0.984 0.012 0.988
#> GSM97064 2 0.0000 0.989 0.000 1.000
#> GSM97065 2 0.0000 0.989 0.000 1.000
#> GSM97081 2 0.0000 0.989 0.000 1.000
#> GSM97082 2 0.0000 0.989 0.000 1.000
#> GSM97088 2 0.2603 0.951 0.044 0.956
#> GSM97100 2 0.0000 0.989 0.000 1.000
#> GSM97104 2 0.0000 0.989 0.000 1.000
#> GSM97108 2 0.0938 0.984 0.012 0.988
#> GSM97050 2 0.0000 0.989 0.000 1.000
#> GSM97080 2 0.0000 0.989 0.000 1.000
#> GSM97089 2 0.0000 0.989 0.000 1.000
#> GSM97092 2 0.0000 0.989 0.000 1.000
#> GSM97093 2 0.0000 0.989 0.000 1.000
#> GSM97058 2 0.0000 0.989 0.000 1.000
#> GSM97051 2 0.0000 0.989 0.000 1.000
#> GSM97052 2 0.0000 0.989 0.000 1.000
#> GSM97061 2 0.0000 0.989 0.000 1.000
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM97138 1 0.3192 0.8715 0.888 0.112 0.000
#> GSM97145 1 0.2959 0.8740 0.900 0.100 0.000
#> GSM97147 2 0.0237 0.7230 0.004 0.996 0.000
#> GSM97125 1 0.1529 0.8796 0.960 0.040 0.000
#> GSM97127 1 0.3192 0.8715 0.888 0.112 0.000
#> GSM97130 1 0.5785 0.7181 0.668 0.332 0.000
#> GSM97133 1 0.3192 0.8715 0.888 0.112 0.000
#> GSM97134 1 0.5397 0.6890 0.720 0.280 0.000
#> GSM97120 1 0.3192 0.8715 0.888 0.112 0.000
#> GSM97126 1 0.2796 0.8757 0.908 0.092 0.000
#> GSM97112 1 0.0000 0.8773 1.000 0.000 0.000
#> GSM97115 2 0.1289 0.7046 0.032 0.968 0.000
#> GSM97116 1 0.3192 0.8715 0.888 0.112 0.000
#> GSM97117 2 0.6104 0.6982 0.004 0.648 0.348
#> GSM97119 1 0.0000 0.8773 1.000 0.000 0.000
#> GSM97122 1 0.0000 0.8773 1.000 0.000 0.000
#> GSM97135 1 0.0000 0.8773 1.000 0.000 0.000
#> GSM97136 3 0.1832 0.8668 0.008 0.036 0.956
#> GSM97139 1 0.3192 0.8715 0.888 0.112 0.000
#> GSM97146 1 0.3192 0.8715 0.888 0.112 0.000
#> GSM97123 3 0.4796 0.5608 0.000 0.220 0.780
#> GSM97129 2 0.6102 0.7247 0.008 0.672 0.320
#> GSM97143 1 0.0000 0.8773 1.000 0.000 0.000
#> GSM97113 2 0.5722 0.7063 0.004 0.704 0.292
#> GSM97056 1 0.3116 0.8740 0.892 0.108 0.000
#> GSM97124 1 0.0424 0.8781 0.992 0.008 0.000
#> GSM97132 1 0.2356 0.8601 0.928 0.072 0.000
#> GSM97144 1 0.5327 0.6984 0.728 0.272 0.000
#> GSM97149 1 0.3192 0.8715 0.888 0.112 0.000
#> GSM97068 2 0.0000 0.7222 0.000 1.000 0.000
#> GSM97071 2 0.7159 0.1382 0.024 0.528 0.448
#> GSM97086 2 0.3192 0.7642 0.000 0.888 0.112
#> GSM97103 3 0.6192 -0.2191 0.000 0.420 0.580
#> GSM97057 2 0.3030 0.7595 0.004 0.904 0.092
#> GSM97060 3 0.0000 0.8968 0.000 0.000 1.000
#> GSM97075 2 0.6008 0.6798 0.000 0.628 0.372
#> GSM97098 3 0.3879 0.7058 0.000 0.152 0.848
#> GSM97099 2 0.6126 0.6961 0.004 0.644 0.352
#> GSM97101 2 0.6008 0.7141 0.004 0.664 0.332
#> GSM97105 2 0.5178 0.7680 0.000 0.744 0.256
#> GSM97106 3 0.1411 0.8688 0.000 0.036 0.964
#> GSM97121 2 0.4978 0.7798 0.004 0.780 0.216
#> GSM97128 1 0.9896 0.0876 0.376 0.264 0.360
#> GSM97131 2 0.5058 0.7721 0.000 0.756 0.244
#> GSM97137 1 0.4062 0.8577 0.836 0.164 0.000
#> GSM97118 1 0.2261 0.8569 0.932 0.068 0.000
#> GSM97114 2 0.6063 0.7161 0.084 0.784 0.132
#> GSM97142 1 0.0000 0.8773 1.000 0.000 0.000
#> GSM97140 2 0.3983 0.7828 0.004 0.852 0.144
#> GSM97141 2 0.6081 0.7004 0.004 0.652 0.344
#> GSM97055 1 0.0000 0.8773 1.000 0.000 0.000
#> GSM97090 2 0.4178 0.5306 0.172 0.828 0.000
#> GSM97091 1 0.0000 0.8773 1.000 0.000 0.000
#> GSM97148 1 0.3192 0.8715 0.888 0.112 0.000
#> GSM97063 1 0.0000 0.8773 1.000 0.000 0.000
#> GSM97053 1 0.0424 0.8781 0.992 0.008 0.000
#> GSM97066 3 0.0000 0.8968 0.000 0.000 1.000
#> GSM97079 2 0.3412 0.7610 0.000 0.876 0.124
#> GSM97083 1 0.5812 0.6876 0.724 0.264 0.012
#> GSM97084 2 0.3267 0.7591 0.000 0.884 0.116
#> GSM97094 2 0.5953 0.4564 0.280 0.708 0.012
#> GSM97096 3 0.0592 0.8904 0.000 0.012 0.988
#> GSM97097 2 0.4504 0.7693 0.000 0.804 0.196
#> GSM97107 2 0.5919 0.4650 0.276 0.712 0.012
#> GSM97054 2 0.3192 0.7642 0.000 0.888 0.112
#> GSM97062 2 0.3412 0.7610 0.000 0.876 0.124
#> GSM97069 3 0.0000 0.8968 0.000 0.000 1.000
#> GSM97070 3 0.0000 0.8968 0.000 0.000 1.000
#> GSM97073 3 0.0000 0.8968 0.000 0.000 1.000
#> GSM97076 1 0.7504 0.5426 0.628 0.312 0.060
#> GSM97077 2 0.4555 0.7802 0.000 0.800 0.200
#> GSM97095 2 0.1529 0.6995 0.040 0.960 0.000
#> GSM97102 3 0.0000 0.8968 0.000 0.000 1.000
#> GSM97109 2 0.5982 0.7088 0.004 0.668 0.328
#> GSM97110 2 0.6057 0.7048 0.004 0.656 0.340
#> GSM97074 3 0.8295 0.2333 0.364 0.088 0.548
#> GSM97085 3 0.3192 0.7784 0.112 0.000 0.888
#> GSM97059 2 0.0237 0.7230 0.004 0.996 0.000
#> GSM97072 3 0.0000 0.8968 0.000 0.000 1.000
#> GSM97078 1 0.9956 0.1466 0.376 0.328 0.296
#> GSM97067 3 0.0000 0.8968 0.000 0.000 1.000
#> GSM97087 3 0.0237 0.8958 0.000 0.004 0.996
#> GSM97111 2 0.6126 0.6961 0.004 0.644 0.352
#> GSM97064 2 0.6026 0.6760 0.000 0.624 0.376
#> GSM97065 2 0.6192 0.5957 0.000 0.580 0.420
#> GSM97081 3 0.0892 0.8853 0.000 0.020 0.980
#> GSM97082 3 0.0000 0.8968 0.000 0.000 1.000
#> GSM97088 3 0.7165 0.5905 0.112 0.172 0.716
#> GSM97100 2 0.2959 0.7668 0.000 0.900 0.100
#> GSM97104 3 0.0000 0.8968 0.000 0.000 1.000
#> GSM97108 2 0.4978 0.7802 0.004 0.780 0.216
#> GSM97050 2 0.5291 0.7635 0.000 0.732 0.268
#> GSM97080 3 0.0000 0.8968 0.000 0.000 1.000
#> GSM97089 3 0.0237 0.8958 0.000 0.004 0.996
#> GSM97092 3 0.0237 0.8958 0.000 0.004 0.996
#> GSM97093 2 0.5706 0.7291 0.000 0.680 0.320
#> GSM97058 2 0.5497 0.7517 0.000 0.708 0.292
#> GSM97051 2 0.4504 0.7789 0.000 0.804 0.196
#> GSM97052 3 0.0237 0.8958 0.000 0.004 0.996
#> GSM97061 3 0.2711 0.8097 0.000 0.088 0.912
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM97138 1 0.1109 0.8659 0.968 0.028 0.000 0.004
#> GSM97145 1 0.1406 0.8675 0.960 0.024 0.000 0.016
#> GSM97147 2 0.1624 0.9053 0.020 0.952 0.000 0.028
#> GSM97125 1 0.1209 0.8678 0.964 0.004 0.000 0.032
#> GSM97127 1 0.0921 0.8643 0.972 0.028 0.000 0.000
#> GSM97130 4 0.4574 0.6264 0.220 0.024 0.000 0.756
#> GSM97133 1 0.1022 0.8635 0.968 0.032 0.000 0.000
#> GSM97134 4 0.2654 0.6583 0.108 0.004 0.000 0.888
#> GSM97120 1 0.1022 0.8633 0.968 0.032 0.000 0.000
#> GSM97126 2 0.7333 0.0628 0.332 0.496 0.000 0.172
#> GSM97112 1 0.3356 0.8489 0.824 0.000 0.000 0.176
#> GSM97115 4 0.4934 0.6720 0.028 0.252 0.000 0.720
#> GSM97116 1 0.0921 0.8647 0.972 0.028 0.000 0.000
#> GSM97117 2 0.1389 0.9249 0.000 0.952 0.048 0.000
#> GSM97119 1 0.3356 0.8489 0.824 0.000 0.000 0.176
#> GSM97122 1 0.3311 0.8508 0.828 0.000 0.000 0.172
#> GSM97135 1 0.3123 0.8556 0.844 0.000 0.000 0.156
#> GSM97136 3 0.6061 0.3451 0.000 0.400 0.552 0.048
#> GSM97139 1 0.0921 0.8647 0.972 0.028 0.000 0.000
#> GSM97146 1 0.1022 0.8635 0.968 0.032 0.000 0.000
#> GSM97123 3 0.4936 0.5915 0.000 0.316 0.672 0.012
#> GSM97129 2 0.1677 0.9259 0.000 0.948 0.040 0.012
#> GSM97143 1 0.3400 0.8467 0.820 0.000 0.000 0.180
#> GSM97113 2 0.1575 0.9232 0.012 0.956 0.028 0.004
#> GSM97056 1 0.4149 0.6891 0.804 0.028 0.000 0.168
#> GSM97124 1 0.3123 0.8556 0.844 0.000 0.000 0.156
#> GSM97132 4 0.5000 -0.2157 0.496 0.000 0.000 0.504
#> GSM97144 4 0.2859 0.6605 0.112 0.008 0.000 0.880
#> GSM97149 1 0.1118 0.8615 0.964 0.036 0.000 0.000
#> GSM97068 2 0.4464 0.6523 0.024 0.768 0.000 0.208
#> GSM97071 4 0.3320 0.7159 0.000 0.068 0.056 0.876
#> GSM97086 4 0.4746 0.6236 0.000 0.304 0.008 0.688
#> GSM97103 3 0.4820 0.6270 0.000 0.296 0.692 0.012
#> GSM97057 2 0.1733 0.9013 0.024 0.948 0.000 0.028
#> GSM97060 3 0.0927 0.8902 0.000 0.008 0.976 0.016
#> GSM97075 2 0.1389 0.9249 0.000 0.952 0.048 0.000
#> GSM97098 3 0.3972 0.7551 0.000 0.204 0.788 0.008
#> GSM97099 2 0.1389 0.9249 0.000 0.952 0.048 0.000
#> GSM97101 2 0.1489 0.9259 0.000 0.952 0.044 0.004
#> GSM97105 2 0.1929 0.9191 0.000 0.940 0.024 0.036
#> GSM97106 3 0.1388 0.8879 0.000 0.028 0.960 0.012
#> GSM97121 2 0.1042 0.9256 0.000 0.972 0.020 0.008
#> GSM97128 4 0.2317 0.6747 0.032 0.004 0.036 0.928
#> GSM97131 2 0.2313 0.9097 0.000 0.924 0.032 0.044
#> GSM97137 1 0.4934 0.5391 0.720 0.028 0.000 0.252
#> GSM97118 4 0.4967 -0.0970 0.452 0.000 0.000 0.548
#> GSM97114 2 0.1575 0.9093 0.028 0.956 0.004 0.012
#> GSM97142 1 0.3356 0.8489 0.824 0.000 0.000 0.176
#> GSM97140 2 0.1509 0.9217 0.008 0.960 0.012 0.020
#> GSM97141 2 0.1489 0.9259 0.000 0.952 0.044 0.004
#> GSM97055 1 0.3528 0.8387 0.808 0.000 0.000 0.192
#> GSM97090 4 0.5031 0.7007 0.048 0.212 0.000 0.740
#> GSM97091 1 0.3486 0.8419 0.812 0.000 0.000 0.188
#> GSM97148 1 0.1022 0.8635 0.968 0.032 0.000 0.000
#> GSM97063 1 0.3444 0.8447 0.816 0.000 0.000 0.184
#> GSM97053 1 0.2589 0.8629 0.884 0.000 0.000 0.116
#> GSM97066 3 0.1807 0.8838 0.000 0.008 0.940 0.052
#> GSM97079 4 0.4722 0.6288 0.000 0.300 0.008 0.692
#> GSM97083 4 0.2216 0.6615 0.092 0.000 0.000 0.908
#> GSM97084 4 0.4391 0.6773 0.000 0.252 0.008 0.740
#> GSM97094 4 0.3172 0.7280 0.012 0.112 0.004 0.872
#> GSM97096 3 0.1256 0.8879 0.000 0.028 0.964 0.008
#> GSM97097 4 0.6252 0.5895 0.000 0.288 0.088 0.624
#> GSM97107 4 0.3436 0.7286 0.016 0.112 0.008 0.864
#> GSM97054 4 0.4422 0.6753 0.000 0.256 0.008 0.736
#> GSM97062 4 0.4391 0.6773 0.000 0.252 0.008 0.740
#> GSM97069 3 0.1576 0.8838 0.000 0.004 0.948 0.048
#> GSM97070 3 0.1807 0.8838 0.000 0.008 0.940 0.052
#> GSM97073 3 0.1807 0.8838 0.000 0.008 0.940 0.052
#> GSM97076 4 0.8511 0.0160 0.316 0.288 0.024 0.372
#> GSM97077 2 0.2214 0.9210 0.000 0.928 0.028 0.044
#> GSM97095 4 0.5137 0.6282 0.024 0.296 0.000 0.680
#> GSM97102 3 0.0937 0.8918 0.000 0.012 0.976 0.012
#> GSM97109 2 0.1639 0.9261 0.008 0.952 0.036 0.004
#> GSM97110 2 0.1576 0.9255 0.000 0.948 0.048 0.004
#> GSM97074 4 0.5311 0.3573 0.024 0.000 0.328 0.648
#> GSM97085 3 0.3625 0.7709 0.012 0.000 0.828 0.160
#> GSM97059 2 0.2521 0.8777 0.024 0.912 0.000 0.064
#> GSM97072 3 0.1389 0.8841 0.000 0.000 0.952 0.048
#> GSM97078 4 0.1697 0.6834 0.028 0.004 0.016 0.952
#> GSM97067 3 0.1807 0.8838 0.000 0.008 0.940 0.052
#> GSM97087 3 0.1004 0.8921 0.000 0.024 0.972 0.004
#> GSM97111 2 0.1389 0.9249 0.000 0.952 0.048 0.000
#> GSM97064 2 0.2742 0.9039 0.000 0.900 0.076 0.024
#> GSM97065 2 0.3081 0.8787 0.000 0.888 0.064 0.048
#> GSM97081 3 0.3074 0.8122 0.000 0.152 0.848 0.000
#> GSM97082 3 0.0927 0.8924 0.000 0.016 0.976 0.008
#> GSM97088 4 0.5140 0.4474 0.020 0.004 0.284 0.692
#> GSM97100 2 0.1722 0.9068 0.000 0.944 0.008 0.048
#> GSM97104 3 0.0657 0.8907 0.000 0.004 0.984 0.012
#> GSM97108 2 0.1624 0.9203 0.000 0.952 0.020 0.028
#> GSM97050 2 0.2494 0.9204 0.000 0.916 0.048 0.036
#> GSM97080 3 0.1452 0.8878 0.000 0.008 0.956 0.036
#> GSM97089 3 0.1004 0.8921 0.000 0.024 0.972 0.004
#> GSM97092 3 0.1151 0.8917 0.000 0.024 0.968 0.008
#> GSM97093 2 0.2197 0.8976 0.000 0.916 0.080 0.004
#> GSM97058 2 0.2224 0.9225 0.000 0.928 0.040 0.032
#> GSM97051 2 0.2111 0.9118 0.000 0.932 0.024 0.044
#> GSM97052 3 0.1151 0.8917 0.000 0.024 0.968 0.008
#> GSM97061 3 0.3271 0.8287 0.000 0.132 0.856 0.012
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM97138 1 0.2017 0.7590 0.912 0.008 0.000 0.000 0.080
#> GSM97145 1 0.1356 0.7612 0.956 0.028 0.000 0.012 0.004
#> GSM97147 2 0.1522 0.8383 0.000 0.944 0.000 0.044 0.012
#> GSM97125 1 0.1211 0.7640 0.960 0.000 0.000 0.016 0.024
#> GSM97127 1 0.2361 0.7553 0.892 0.012 0.000 0.000 0.096
#> GSM97130 4 0.5644 0.3833 0.144 0.000 0.000 0.628 0.228
#> GSM97133 1 0.2997 0.7428 0.840 0.012 0.000 0.000 0.148
#> GSM97134 4 0.5065 -0.1121 0.036 0.000 0.000 0.544 0.420
#> GSM97120 1 0.2997 0.7428 0.840 0.012 0.000 0.000 0.148
#> GSM97126 2 0.6319 0.3936 0.200 0.620 0.000 0.036 0.144
#> GSM97112 1 0.3445 0.7341 0.824 0.000 0.000 0.036 0.140
#> GSM97115 4 0.4832 0.6113 0.008 0.064 0.000 0.720 0.208
#> GSM97116 1 0.2886 0.7445 0.844 0.008 0.000 0.000 0.148
#> GSM97117 2 0.0451 0.8412 0.000 0.988 0.004 0.000 0.008
#> GSM97119 1 0.3445 0.7341 0.824 0.000 0.000 0.036 0.140
#> GSM97122 1 0.3445 0.7341 0.824 0.000 0.000 0.036 0.140
#> GSM97135 1 0.3355 0.7385 0.832 0.000 0.000 0.036 0.132
#> GSM97136 2 0.6803 0.3384 0.052 0.556 0.264 0.000 0.128
#> GSM97139 1 0.2886 0.7445 0.844 0.008 0.000 0.000 0.148
#> GSM97146 1 0.2886 0.7445 0.844 0.008 0.000 0.000 0.148
#> GSM97123 3 0.4528 0.6787 0.000 0.172 0.756 0.008 0.064
#> GSM97129 2 0.0865 0.8379 0.000 0.972 0.004 0.000 0.024
#> GSM97143 1 0.3919 0.6878 0.776 0.000 0.000 0.036 0.188
#> GSM97113 2 0.0162 0.8425 0.000 0.996 0.004 0.000 0.000
#> GSM97056 1 0.5531 0.4593 0.632 0.000 0.000 0.120 0.248
#> GSM97124 1 0.3355 0.7385 0.832 0.000 0.000 0.036 0.132
#> GSM97132 5 0.6515 0.3529 0.388 0.000 0.000 0.192 0.420
#> GSM97144 4 0.4301 0.4349 0.028 0.000 0.000 0.712 0.260
#> GSM97149 1 0.2997 0.7428 0.840 0.012 0.000 0.000 0.148
#> GSM97068 2 0.4860 0.6020 0.004 0.664 0.000 0.292 0.040
#> GSM97071 4 0.4017 0.5312 0.000 0.004 0.012 0.736 0.248
#> GSM97086 4 0.1981 0.6554 0.000 0.064 0.000 0.920 0.016
#> GSM97103 3 0.6199 0.4993 0.000 0.328 0.564 0.036 0.072
#> GSM97057 2 0.2984 0.8115 0.000 0.860 0.000 0.108 0.032
#> GSM97060 3 0.2037 0.8060 0.000 0.004 0.920 0.012 0.064
#> GSM97075 2 0.0798 0.8410 0.000 0.976 0.016 0.000 0.008
#> GSM97098 3 0.4961 0.5962 0.000 0.276 0.672 0.008 0.044
#> GSM97099 2 0.0566 0.8408 0.000 0.984 0.004 0.000 0.012
#> GSM97101 2 0.0162 0.8425 0.000 0.996 0.004 0.000 0.000
#> GSM97105 2 0.3425 0.8112 0.000 0.840 0.004 0.112 0.044
#> GSM97106 3 0.2760 0.7837 0.000 0.016 0.892 0.028 0.064
#> GSM97121 2 0.0510 0.8422 0.000 0.984 0.000 0.016 0.000
#> GSM97128 5 0.4726 0.3247 0.020 0.000 0.000 0.400 0.580
#> GSM97131 2 0.5765 0.6493 0.000 0.644 0.036 0.256 0.064
#> GSM97137 1 0.6233 0.1956 0.520 0.000 0.000 0.168 0.312
#> GSM97118 5 0.6442 0.4779 0.324 0.000 0.000 0.196 0.480
#> GSM97114 2 0.0613 0.8421 0.004 0.984 0.000 0.004 0.008
#> GSM97142 1 0.3445 0.7341 0.824 0.000 0.000 0.036 0.140
#> GSM97140 2 0.1408 0.8391 0.000 0.948 0.000 0.044 0.008
#> GSM97141 2 0.0162 0.8425 0.000 0.996 0.004 0.000 0.000
#> GSM97055 1 0.4451 0.6018 0.712 0.000 0.000 0.040 0.248
#> GSM97090 4 0.4832 0.6035 0.008 0.060 0.000 0.716 0.216
#> GSM97091 1 0.4193 0.6574 0.748 0.000 0.000 0.040 0.212
#> GSM97148 1 0.2886 0.7445 0.844 0.008 0.000 0.000 0.148
#> GSM97063 1 0.3691 0.7196 0.804 0.000 0.000 0.040 0.156
#> GSM97053 1 0.2616 0.7545 0.888 0.000 0.000 0.036 0.076
#> GSM97066 3 0.3661 0.7483 0.000 0.000 0.724 0.000 0.276
#> GSM97079 4 0.1571 0.6672 0.000 0.060 0.000 0.936 0.004
#> GSM97083 4 0.4829 -0.2709 0.020 0.000 0.000 0.500 0.480
#> GSM97084 4 0.1270 0.6721 0.000 0.052 0.000 0.948 0.000
#> GSM97094 4 0.2813 0.6495 0.000 0.024 0.000 0.868 0.108
#> GSM97096 3 0.3113 0.7751 0.000 0.080 0.868 0.008 0.044
#> GSM97097 4 0.3389 0.5893 0.000 0.048 0.052 0.864 0.036
#> GSM97107 4 0.2813 0.6495 0.000 0.024 0.000 0.868 0.108
#> GSM97054 4 0.2124 0.6690 0.000 0.056 0.000 0.916 0.028
#> GSM97062 4 0.1341 0.6708 0.000 0.056 0.000 0.944 0.000
#> GSM97069 3 0.3741 0.7536 0.000 0.000 0.732 0.004 0.264
#> GSM97070 3 0.3636 0.7504 0.000 0.000 0.728 0.000 0.272
#> GSM97073 3 0.3661 0.7483 0.000 0.000 0.724 0.000 0.276
#> GSM97076 2 0.6893 0.0478 0.108 0.432 0.024 0.012 0.424
#> GSM97077 2 0.3495 0.8035 0.000 0.836 0.008 0.120 0.036
#> GSM97095 4 0.6610 0.3246 0.004 0.288 0.000 0.488 0.220
#> GSM97102 3 0.2942 0.8025 0.000 0.008 0.856 0.008 0.128
#> GSM97109 2 0.0671 0.8404 0.000 0.980 0.004 0.000 0.016
#> GSM97110 2 0.0671 0.8404 0.000 0.980 0.004 0.000 0.016
#> GSM97074 5 0.5540 0.4344 0.100 0.000 0.084 0.092 0.724
#> GSM97085 3 0.4895 0.4674 0.008 0.000 0.528 0.012 0.452
#> GSM97059 2 0.3722 0.7697 0.004 0.796 0.000 0.176 0.024
#> GSM97072 3 0.3809 0.7610 0.000 0.000 0.736 0.008 0.256
#> GSM97078 5 0.4746 0.1106 0.016 0.000 0.000 0.480 0.504
#> GSM97067 3 0.3661 0.7483 0.000 0.000 0.724 0.000 0.276
#> GSM97087 3 0.1168 0.8011 0.000 0.008 0.960 0.000 0.032
#> GSM97111 2 0.0566 0.8408 0.000 0.984 0.004 0.000 0.012
#> GSM97064 2 0.5590 0.7406 0.000 0.712 0.124 0.112 0.052
#> GSM97065 2 0.3283 0.7567 0.000 0.832 0.028 0.000 0.140
#> GSM97081 3 0.3639 0.7210 0.000 0.184 0.792 0.000 0.024
#> GSM97082 3 0.1638 0.8062 0.000 0.004 0.932 0.000 0.064
#> GSM97088 5 0.6126 0.3761 0.008 0.000 0.128 0.300 0.564
#> GSM97100 2 0.3841 0.7650 0.000 0.780 0.000 0.188 0.032
#> GSM97104 3 0.2570 0.8005 0.000 0.004 0.880 0.008 0.108
#> GSM97108 2 0.1469 0.8395 0.000 0.948 0.000 0.036 0.016
#> GSM97050 2 0.5557 0.7404 0.000 0.712 0.104 0.136 0.048
#> GSM97080 3 0.3242 0.7730 0.000 0.000 0.784 0.000 0.216
#> GSM97089 3 0.1168 0.8011 0.000 0.008 0.960 0.000 0.032
#> GSM97092 3 0.1862 0.7913 0.000 0.016 0.932 0.004 0.048
#> GSM97093 2 0.3922 0.7524 0.000 0.796 0.164 0.012 0.028
#> GSM97058 2 0.3579 0.8052 0.000 0.836 0.016 0.116 0.032
#> GSM97051 2 0.6776 0.6008 0.000 0.576 0.120 0.240 0.064
#> GSM97052 3 0.1862 0.7913 0.000 0.016 0.932 0.004 0.048
#> GSM97061 3 0.3164 0.7676 0.000 0.040 0.872 0.020 0.068
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM97138 1 0.1910 0.549897 0.892 0.000 0.000 0.000 0.108 0.000
#> GSM97145 1 0.4212 0.365896 0.560 0.016 0.000 0.000 0.424 0.000
#> GSM97147 2 0.3286 0.791489 0.000 0.844 0.000 0.044 0.084 0.028
#> GSM97125 1 0.3833 0.362153 0.556 0.000 0.000 0.000 0.444 0.000
#> GSM97127 1 0.2300 0.543294 0.856 0.000 0.000 0.000 0.144 0.000
#> GSM97130 4 0.7356 0.439345 0.152 0.000 0.000 0.408 0.228 0.212
#> GSM97133 1 0.0146 0.571019 0.996 0.000 0.000 0.000 0.004 0.000
#> GSM97134 5 0.6415 -0.375556 0.012 0.000 0.000 0.312 0.348 0.328
#> GSM97120 1 0.0363 0.573493 0.988 0.000 0.000 0.000 0.012 0.000
#> GSM97126 2 0.5102 0.552320 0.028 0.680 0.000 0.000 0.184 0.108
#> GSM97112 1 0.3868 0.299464 0.504 0.000 0.000 0.000 0.496 0.000
#> GSM97115 4 0.6827 0.523214 0.020 0.032 0.000 0.476 0.236 0.236
#> GSM97116 1 0.0363 0.574262 0.988 0.000 0.000 0.000 0.012 0.000
#> GSM97117 2 0.0810 0.800767 0.004 0.976 0.008 0.000 0.004 0.008
#> GSM97119 1 0.3868 0.299464 0.504 0.000 0.000 0.000 0.496 0.000
#> GSM97122 1 0.3868 0.306654 0.508 0.000 0.000 0.000 0.492 0.000
#> GSM97135 1 0.3868 0.306654 0.508 0.000 0.000 0.000 0.492 0.000
#> GSM97136 2 0.6239 0.302244 0.004 0.560 0.228 0.000 0.044 0.164
#> GSM97139 1 0.0363 0.574262 0.988 0.000 0.000 0.000 0.012 0.000
#> GSM97146 1 0.0000 0.572253 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM97123 3 0.3773 0.600189 0.000 0.136 0.804 0.016 0.032 0.012
#> GSM97129 2 0.1498 0.797068 0.004 0.948 0.012 0.000 0.012 0.024
#> GSM97143 5 0.4587 -0.281035 0.456 0.000 0.000 0.000 0.508 0.036
#> GSM97113 2 0.0436 0.803398 0.004 0.988 0.004 0.000 0.004 0.000
#> GSM97056 1 0.4481 0.253073 0.736 0.000 0.000 0.056 0.176 0.032
#> GSM97124 1 0.3868 0.304807 0.504 0.000 0.000 0.000 0.496 0.000
#> GSM97132 5 0.6588 0.136437 0.124 0.000 0.000 0.080 0.472 0.324
#> GSM97144 4 0.6177 0.488835 0.016 0.000 0.000 0.488 0.264 0.232
#> GSM97149 1 0.0000 0.572253 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM97068 2 0.5836 0.620889 0.016 0.620 0.000 0.220 0.116 0.028
#> GSM97071 4 0.4755 0.586618 0.000 0.000 0.004 0.664 0.088 0.244
#> GSM97086 4 0.1448 0.656821 0.000 0.016 0.000 0.948 0.024 0.012
#> GSM97103 3 0.7500 0.396770 0.000 0.300 0.448 0.072 0.092 0.088
#> GSM97057 2 0.4393 0.774740 0.016 0.792 0.008 0.084 0.072 0.028
#> GSM97060 3 0.3142 0.671864 0.000 0.000 0.840 0.008 0.044 0.108
#> GSM97075 2 0.0881 0.805066 0.000 0.972 0.008 0.000 0.008 0.012
#> GSM97098 3 0.6475 0.469024 0.000 0.284 0.536 0.012 0.092 0.076
#> GSM97099 2 0.1377 0.793891 0.004 0.952 0.004 0.000 0.016 0.024
#> GSM97101 2 0.0653 0.805516 0.004 0.980 0.004 0.000 0.012 0.000
#> GSM97105 2 0.4272 0.775150 0.000 0.788 0.012 0.096 0.072 0.032
#> GSM97106 3 0.3467 0.668023 0.000 0.012 0.836 0.048 0.092 0.012
#> GSM97121 2 0.1434 0.805947 0.000 0.948 0.000 0.012 0.028 0.012
#> GSM97128 6 0.6018 -0.053215 0.000 0.000 0.008 0.192 0.336 0.464
#> GSM97131 2 0.6713 0.551279 0.000 0.528 0.076 0.284 0.080 0.032
#> GSM97137 1 0.5151 0.150391 0.668 0.000 0.000 0.068 0.220 0.044
#> GSM97118 5 0.5924 0.032592 0.048 0.000 0.000 0.076 0.476 0.400
#> GSM97114 2 0.0551 0.802626 0.004 0.984 0.000 0.000 0.004 0.008
#> GSM97142 1 0.3868 0.299464 0.504 0.000 0.000 0.000 0.496 0.000
#> GSM97140 2 0.2803 0.797396 0.000 0.876 0.000 0.032 0.064 0.028
#> GSM97141 2 0.0436 0.803398 0.004 0.988 0.004 0.000 0.004 0.000
#> GSM97055 5 0.5510 -0.000949 0.340 0.000 0.000 0.000 0.516 0.144
#> GSM97090 4 0.6827 0.522415 0.020 0.032 0.000 0.476 0.236 0.236
#> GSM97091 5 0.4468 -0.196827 0.408 0.000 0.000 0.000 0.560 0.032
#> GSM97148 1 0.0000 0.572253 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM97063 5 0.4067 -0.277092 0.444 0.000 0.000 0.000 0.548 0.008
#> GSM97053 1 0.3867 0.312662 0.512 0.000 0.000 0.000 0.488 0.000
#> GSM97066 6 0.3867 -0.263564 0.000 0.000 0.488 0.000 0.000 0.512
#> GSM97079 4 0.0520 0.678203 0.000 0.008 0.000 0.984 0.000 0.008
#> GSM97083 6 0.6130 -0.151259 0.004 0.000 0.000 0.240 0.352 0.404
#> GSM97084 4 0.0551 0.680267 0.000 0.008 0.004 0.984 0.000 0.004
#> GSM97094 4 0.3418 0.674147 0.000 0.004 0.000 0.820 0.084 0.092
#> GSM97096 3 0.5263 0.638948 0.000 0.104 0.716 0.012 0.092 0.076
#> GSM97097 4 0.2973 0.593100 0.000 0.004 0.032 0.864 0.084 0.016
#> GSM97107 4 0.3658 0.671615 0.000 0.004 0.004 0.808 0.092 0.092
#> GSM97054 4 0.2460 0.661252 0.000 0.016 0.004 0.896 0.064 0.020
#> GSM97062 4 0.0520 0.680810 0.000 0.008 0.000 0.984 0.000 0.008
#> GSM97069 3 0.3866 0.206563 0.000 0.000 0.516 0.000 0.000 0.484
#> GSM97070 6 0.3869 -0.282764 0.000 0.000 0.500 0.000 0.000 0.500
#> GSM97073 6 0.3864 -0.264336 0.000 0.000 0.480 0.000 0.000 0.520
#> GSM97076 6 0.4607 0.082445 0.000 0.356 0.012 0.000 0.028 0.604
#> GSM97077 2 0.4788 0.760443 0.000 0.756 0.028 0.108 0.076 0.032
#> GSM97095 4 0.7939 0.336610 0.016 0.160 0.000 0.300 0.264 0.260
#> GSM97102 3 0.4891 0.608747 0.000 0.004 0.664 0.004 0.092 0.236
#> GSM97109 2 0.1924 0.785441 0.004 0.928 0.004 0.004 0.024 0.036
#> GSM97110 2 0.1924 0.785441 0.004 0.928 0.004 0.004 0.024 0.036
#> GSM97074 6 0.3811 0.250731 0.004 0.000 0.028 0.024 0.152 0.792
#> GSM97085 6 0.5121 0.113746 0.000 0.000 0.272 0.000 0.124 0.604
#> GSM97059 2 0.5378 0.720800 0.016 0.704 0.008 0.136 0.108 0.028
#> GSM97072 3 0.4468 0.223301 0.000 0.000 0.492 0.004 0.020 0.484
#> GSM97078 6 0.5932 -0.103640 0.000 0.000 0.000 0.224 0.336 0.440
#> GSM97067 6 0.3866 -0.265688 0.000 0.000 0.484 0.000 0.000 0.516
#> GSM97087 3 0.1657 0.691647 0.000 0.016 0.928 0.000 0.000 0.056
#> GSM97111 2 0.1096 0.797180 0.004 0.964 0.004 0.000 0.008 0.020
#> GSM97064 2 0.6644 0.615613 0.000 0.560 0.240 0.100 0.068 0.032
#> GSM97065 2 0.3670 0.575326 0.000 0.704 0.012 0.000 0.000 0.284
#> GSM97081 3 0.4051 0.635035 0.000 0.164 0.756 0.000 0.004 0.076
#> GSM97082 3 0.2178 0.664381 0.000 0.000 0.868 0.000 0.000 0.132
#> GSM97088 6 0.6433 -0.017422 0.000 0.000 0.044 0.172 0.300 0.484
#> GSM97100 2 0.5055 0.719231 0.000 0.700 0.008 0.184 0.076 0.032
#> GSM97104 3 0.4121 0.610340 0.000 0.000 0.732 0.004 0.056 0.208
#> GSM97108 2 0.2589 0.799373 0.000 0.888 0.000 0.024 0.060 0.028
#> GSM97050 2 0.6683 0.629308 0.000 0.572 0.176 0.160 0.060 0.032
#> GSM97080 3 0.3727 0.379624 0.000 0.000 0.612 0.000 0.000 0.388
#> GSM97089 3 0.1700 0.692863 0.000 0.024 0.928 0.000 0.000 0.048
#> GSM97092 3 0.1109 0.692315 0.000 0.016 0.964 0.004 0.012 0.004
#> GSM97093 2 0.4823 0.572540 0.000 0.648 0.296 0.016 0.020 0.020
#> GSM97058 2 0.4788 0.759311 0.000 0.756 0.028 0.108 0.076 0.032
#> GSM97051 2 0.7584 0.432181 0.000 0.412 0.216 0.260 0.080 0.032
#> GSM97052 3 0.1109 0.692315 0.000 0.016 0.964 0.004 0.012 0.004
#> GSM97061 3 0.2145 0.674666 0.000 0.020 0.920 0.032 0.016 0.012
Heatmaps for the consensus matrix. It visualizes the probability of two samples to be in a same group.
consensus_heatmap(res, k = 2)
consensus_heatmap(res, k = 3)
consensus_heatmap(res, k = 4)
consensus_heatmap(res, k = 5)
consensus_heatmap(res, k = 6)
Heatmaps for the membership of samples in all partitions to see how consistent they are:
membership_heatmap(res, k = 2)
membership_heatmap(res, k = 3)
membership_heatmap(res, k = 4)
membership_heatmap(res, k = 5)
membership_heatmap(res, k = 6)
As soon as we have had the classes for columns, we can look for signatures which are significantly different between classes which can be candidate marks for certain classes. Following are the heatmaps for signatures.
Signature heatmaps where rows are scaled:
get_signatures(res, k = 2)
get_signatures(res, k = 3)
get_signatures(res, k = 4)
get_signatures(res, k = 5)
get_signatures(res, k = 6)
Signature heatmaps where rows are not scaled:
get_signatures(res, k = 2, scale_rows = FALSE)
get_signatures(res, k = 3, scale_rows = FALSE)
get_signatures(res, k = 4, scale_rows = FALSE)
get_signatures(res, k = 5, scale_rows = FALSE)
get_signatures(res, k = 6, scale_rows = FALSE)
Compare the overlap of signatures from different k:
compare_signatures(res)
get_signature() returns a data frame invisibly. TO get the list of signatures, the function
call should be assigned to a variable explicitly. In following code, if plot argument is set
to FALSE, no heatmap is plotted while only the differential analysis is performed.
# code only for demonstration
tb = get_signature(res, k = ..., plot = FALSE)
An example of the output of tb is:
#> which_row fdr mean_1 mean_2 scaled_mean_1 scaled_mean_2 km
#> 1 38 0.042760348 8.373488 9.131774 -0.5533452 0.5164555 1
#> 2 40 0.018707592 7.106213 8.469186 -0.6173731 0.5762149 1
#> 3 55 0.019134737 10.221463 11.207825 -0.6159697 0.5749050 1
#> 4 59 0.006059896 5.921854 7.869574 -0.6899429 0.6439467 1
#> 5 60 0.018055526 8.928898 10.211722 -0.6204761 0.5791110 1
#> 6 98 0.009384629 15.714769 14.887706 0.6635654 -0.6193277 2
...
The columns in tb are:
which_row: row indices corresponding to the input matrix.fdr: FDR for the differential test. mean_x: The mean value in group x.scaled_mean_x: The mean value in group x after rows are scaled.km: Row groups if k-means clustering is applied to rows.UMAP plot which shows how samples are separated.
dimension_reduction(res, k = 2, method = "UMAP")
dimension_reduction(res, k = 3, method = "UMAP")
dimension_reduction(res, k = 4, method = "UMAP")
dimension_reduction(res, k = 5, method = "UMAP")
dimension_reduction(res, k = 6, method = "UMAP")
Following heatmap shows how subgroups are split when increasing k:
collect_classes(res)
Test correlation between subgroups and known annotations. If the known annotation is numeric, one-way ANOVA test is applied, and if the known annotation is discrete, chi-squared contingency table test is applied.
test_to_known_factors(res)
#> n disease.state(p) specimen(p) cell.type(p) other(p) k
#> MAD:kmeans 100 0.000277 0.298 2.66e-13 0.0975 2
#> MAD:kmeans 93 0.000141 0.249 1.94e-12 0.2283 3
#> MAD:kmeans 93 0.000266 0.331 7.45e-16 0.1535 4
#> MAD:kmeans 82 0.000097 0.187 1.56e-15 0.0497 5
#> MAD:kmeans 61 0.002758 0.310 5.40e-10 0.2320 6
If matrix rows can be associated to genes, consider to use functional_enrichment(res,
...) to perform function enrichment for the signature genes. See this vignette for more detailed explanations.
The object with results only for a single top-value method and a single partition method can be extracted as:
res = res_list["MAD", "skmeans"]
# you can also extract it by
# res = res_list["MAD:skmeans"]
A summary of res and all the functions that can be applied to it:
res
#> A 'ConsensusPartition' object with k = 2, 3, 4, 5, 6.
#> On a matrix with 21512 rows and 100 columns.
#> Top rows (1000, 2000, 3000, 4000, 5000) are extracted by 'MAD' method.
#> Subgroups are detected by 'skmeans' method.
#> Performed in total 1250 partitions by row resampling.
#> Best k for subgroups seems to be 2.
#>
#> Following methods can be applied to this 'ConsensusPartition' object:
#> [1] "cola_report" "collect_classes" "collect_plots"
#> [4] "collect_stats" "colnames" "compare_signatures"
#> [7] "consensus_heatmap" "dimension_reduction" "functional_enrichment"
#> [10] "get_anno_col" "get_anno" "get_classes"
#> [13] "get_consensus" "get_matrix" "get_membership"
#> [16] "get_param" "get_signatures" "get_stats"
#> [19] "is_best_k" "is_stable_k" "membership_heatmap"
#> [22] "ncol" "nrow" "plot_ecdf"
#> [25] "rownames" "select_partition_number" "show"
#> [28] "suggest_best_k" "test_to_known_factors"
collect_plots() function collects all the plots made from res for all k (number of partitions)
into one single page to provide an easy and fast comparison between different k.
collect_plots(res)
The plots are:
k and the heatmap of
predicted classes for each k.k.k.k.All the plots in panels can be made by individual functions and they are plotted later in this section.
select_partition_number() produces several plots showing different
statistics for choosing “optimized” k. There are following statistics:
k;k, the area increased is defined as \(A_k - A_{k-1}\).The detailed explanations of these statistics can be found in the cola vignette.
Generally speaking, lower PAC score, higher mean silhouette score or higher
concordance corresponds to better partition. Rand index and Jaccard index
measure how similar the current partition is compared to partition with k-1.
If they are too similar, we won't accept k is better than k-1.
select_partition_number(res)
The numeric values for all these statistics can be obtained by get_stats().
get_stats(res)
#> k 1-PAC mean_silhouette concordance area_increased Rand Jaccard
#> 2 2 0.979 0.953 0.982 0.4995 0.500 0.500
#> 3 3 0.820 0.849 0.931 0.3415 0.753 0.541
#> 4 4 0.719 0.793 0.887 0.1215 0.829 0.544
#> 5 5 0.616 0.547 0.719 0.0615 0.976 0.904
#> 6 6 0.614 0.429 0.637 0.0393 0.951 0.789
suggest_best_k() suggests the best \(k\) based on these statistics. The rules are as follows:
suggest_best_k(res)
#> [1] 2
Following shows the table of the partitions (You need to click the show/hide
code output link to see it). The membership matrix (columns with name p*)
is inferred by
clue::cl_consensus()
function with the SE method. Basically the value in the membership matrix
represents the probability to belong to a certain group. The finall class
label for an item is determined with the group with highest probability it
belongs to.
In get_classes() function, the entropy is calculated from the membership
matrix and the silhouette score is calculated from the consensus matrix.
cbind(get_classes(res, k = 2), get_membership(res, k = 2))
#> class entropy silhouette p1 p2
#> GSM97138 1 0.0000 0.974 1.000 0.000
#> GSM97145 1 0.0000 0.974 1.000 0.000
#> GSM97147 1 0.0000 0.974 1.000 0.000
#> GSM97125 1 0.0000 0.974 1.000 0.000
#> GSM97127 1 0.0000 0.974 1.000 0.000
#> GSM97130 1 0.0000 0.974 1.000 0.000
#> GSM97133 1 0.0000 0.974 1.000 0.000
#> GSM97134 1 0.0000 0.974 1.000 0.000
#> GSM97120 1 0.0000 0.974 1.000 0.000
#> GSM97126 1 0.0000 0.974 1.000 0.000
#> GSM97112 1 0.0000 0.974 1.000 0.000
#> GSM97115 1 0.0000 0.974 1.000 0.000
#> GSM97116 1 0.0000 0.974 1.000 0.000
#> GSM97117 2 0.0000 0.986 0.000 1.000
#> GSM97119 1 0.0000 0.974 1.000 0.000
#> GSM97122 1 0.0000 0.974 1.000 0.000
#> GSM97135 1 0.0000 0.974 1.000 0.000
#> GSM97136 2 0.4690 0.885 0.100 0.900
#> GSM97139 1 0.0000 0.974 1.000 0.000
#> GSM97146 1 0.0000 0.974 1.000 0.000
#> GSM97123 2 0.0000 0.986 0.000 1.000
#> GSM97129 2 0.6343 0.805 0.160 0.840
#> GSM97143 1 0.0000 0.974 1.000 0.000
#> GSM97113 2 0.0376 0.983 0.004 0.996
#> GSM97056 1 0.0000 0.974 1.000 0.000
#> GSM97124 1 0.0000 0.974 1.000 0.000
#> GSM97132 1 0.0000 0.974 1.000 0.000
#> GSM97144 1 0.0000 0.974 1.000 0.000
#> GSM97149 1 0.0000 0.974 1.000 0.000
#> GSM97068 1 0.8386 0.631 0.732 0.268
#> GSM97071 2 0.2603 0.946 0.044 0.956
#> GSM97086 2 0.0000 0.986 0.000 1.000
#> GSM97103 2 0.0000 0.986 0.000 1.000
#> GSM97057 2 0.0938 0.977 0.012 0.988
#> GSM97060 2 0.0000 0.986 0.000 1.000
#> GSM97075 2 0.0000 0.986 0.000 1.000
#> GSM97098 2 0.0000 0.986 0.000 1.000
#> GSM97099 2 0.0000 0.986 0.000 1.000
#> GSM97101 2 0.0000 0.986 0.000 1.000
#> GSM97105 2 0.0000 0.986 0.000 1.000
#> GSM97106 2 0.0000 0.986 0.000 1.000
#> GSM97121 2 0.0000 0.986 0.000 1.000
#> GSM97128 1 0.0000 0.974 1.000 0.000
#> GSM97131 2 0.0000 0.986 0.000 1.000
#> GSM97137 1 0.0000 0.974 1.000 0.000
#> GSM97118 1 0.0000 0.974 1.000 0.000
#> GSM97114 1 0.9944 0.166 0.544 0.456
#> GSM97142 1 0.0000 0.974 1.000 0.000
#> GSM97140 2 0.0376 0.983 0.004 0.996
#> GSM97141 2 0.0000 0.986 0.000 1.000
#> GSM97055 1 0.0000 0.974 1.000 0.000
#> GSM97090 1 0.0000 0.974 1.000 0.000
#> GSM97091 1 0.0000 0.974 1.000 0.000
#> GSM97148 1 0.0000 0.974 1.000 0.000
#> GSM97063 1 0.0000 0.974 1.000 0.000
#> GSM97053 1 0.0000 0.974 1.000 0.000
#> GSM97066 2 0.0000 0.986 0.000 1.000
#> GSM97079 2 0.0000 0.986 0.000 1.000
#> GSM97083 1 0.0000 0.974 1.000 0.000
#> GSM97084 2 0.0938 0.977 0.012 0.988
#> GSM97094 1 0.0000 0.974 1.000 0.000
#> GSM97096 2 0.0000 0.986 0.000 1.000
#> GSM97097 2 0.0000 0.986 0.000 1.000
#> GSM97107 1 0.0000 0.974 1.000 0.000
#> GSM97054 2 0.0000 0.986 0.000 1.000
#> GSM97062 2 0.0000 0.986 0.000 1.000
#> GSM97069 2 0.0000 0.986 0.000 1.000
#> GSM97070 2 0.0000 0.986 0.000 1.000
#> GSM97073 2 0.0000 0.986 0.000 1.000
#> GSM97076 1 0.0000 0.974 1.000 0.000
#> GSM97077 2 0.0000 0.986 0.000 1.000
#> GSM97095 1 0.0000 0.974 1.000 0.000
#> GSM97102 2 0.0000 0.986 0.000 1.000
#> GSM97109 2 0.0376 0.983 0.004 0.996
#> GSM97110 2 0.0000 0.986 0.000 1.000
#> GSM97074 1 0.0000 0.974 1.000 0.000
#> GSM97085 2 0.9491 0.403 0.368 0.632
#> GSM97059 1 0.0376 0.970 0.996 0.004
#> GSM97072 2 0.0000 0.986 0.000 1.000
#> GSM97078 1 0.0000 0.974 1.000 0.000
#> GSM97067 2 0.0000 0.986 0.000 1.000
#> GSM97087 2 0.0000 0.986 0.000 1.000
#> GSM97111 2 0.0000 0.986 0.000 1.000
#> GSM97064 2 0.0000 0.986 0.000 1.000
#> GSM97065 2 0.0000 0.986 0.000 1.000
#> GSM97081 2 0.0000 0.986 0.000 1.000
#> GSM97082 2 0.0000 0.986 0.000 1.000
#> GSM97088 1 0.9686 0.343 0.604 0.396
#> GSM97100 2 0.0000 0.986 0.000 1.000
#> GSM97104 2 0.0000 0.986 0.000 1.000
#> GSM97108 2 0.0000 0.986 0.000 1.000
#> GSM97050 2 0.0000 0.986 0.000 1.000
#> GSM97080 2 0.0000 0.986 0.000 1.000
#> GSM97089 2 0.0000 0.986 0.000 1.000
#> GSM97092 2 0.0000 0.986 0.000 1.000
#> GSM97093 2 0.0376 0.983 0.004 0.996
#> GSM97058 2 0.0000 0.986 0.000 1.000
#> GSM97051 2 0.0000 0.986 0.000 1.000
#> GSM97052 2 0.0000 0.986 0.000 1.000
#> GSM97061 2 0.0000 0.986 0.000 1.000
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM97138 1 0.0000 0.9485 1.000 0.000 0.000
#> GSM97145 1 0.0000 0.9485 1.000 0.000 0.000
#> GSM97147 2 0.1529 0.9099 0.040 0.960 0.000
#> GSM97125 1 0.0000 0.9485 1.000 0.000 0.000
#> GSM97127 1 0.0000 0.9485 1.000 0.000 0.000
#> GSM97130 1 0.1031 0.9374 0.976 0.024 0.000
#> GSM97133 1 0.0000 0.9485 1.000 0.000 0.000
#> GSM97134 1 0.1031 0.9369 0.976 0.024 0.000
#> GSM97120 1 0.0000 0.9485 1.000 0.000 0.000
#> GSM97126 1 0.0000 0.9485 1.000 0.000 0.000
#> GSM97112 1 0.0000 0.9485 1.000 0.000 0.000
#> GSM97115 1 0.6286 0.2064 0.536 0.464 0.000
#> GSM97116 1 0.0000 0.9485 1.000 0.000 0.000
#> GSM97117 3 0.6295 0.0797 0.000 0.472 0.528
#> GSM97119 1 0.0000 0.9485 1.000 0.000 0.000
#> GSM97122 1 0.0000 0.9485 1.000 0.000 0.000
#> GSM97135 1 0.0000 0.9485 1.000 0.000 0.000
#> GSM97136 3 0.0892 0.8839 0.020 0.000 0.980
#> GSM97139 1 0.0000 0.9485 1.000 0.000 0.000
#> GSM97146 1 0.0000 0.9485 1.000 0.000 0.000
#> GSM97123 3 0.4291 0.7414 0.000 0.180 0.820
#> GSM97129 3 0.9171 0.1612 0.152 0.372 0.476
#> GSM97143 1 0.0000 0.9485 1.000 0.000 0.000
#> GSM97113 2 0.1643 0.9235 0.000 0.956 0.044
#> GSM97056 1 0.0237 0.9469 0.996 0.004 0.000
#> GSM97124 1 0.0000 0.9485 1.000 0.000 0.000
#> GSM97132 1 0.0237 0.9470 0.996 0.004 0.000
#> GSM97144 1 0.1411 0.9296 0.964 0.036 0.000
#> GSM97149 1 0.0000 0.9485 1.000 0.000 0.000
#> GSM97068 2 0.0000 0.9252 0.000 1.000 0.000
#> GSM97071 3 0.2959 0.8320 0.000 0.100 0.900
#> GSM97086 2 0.0747 0.9266 0.000 0.984 0.016
#> GSM97103 3 0.2537 0.8469 0.000 0.080 0.920
#> GSM97057 2 0.0424 0.9280 0.000 0.992 0.008
#> GSM97060 3 0.0000 0.8955 0.000 0.000 1.000
#> GSM97075 3 0.5678 0.5246 0.000 0.316 0.684
#> GSM97098 3 0.1753 0.8697 0.000 0.048 0.952
#> GSM97099 2 0.5138 0.6956 0.000 0.748 0.252
#> GSM97101 2 0.1289 0.9272 0.000 0.968 0.032
#> GSM97105 2 0.0592 0.9287 0.000 0.988 0.012
#> GSM97106 3 0.0592 0.8911 0.000 0.012 0.988
#> GSM97121 2 0.0592 0.9287 0.000 0.988 0.012
#> GSM97128 3 0.7353 0.1596 0.436 0.032 0.532
#> GSM97131 2 0.1860 0.9249 0.000 0.948 0.052
#> GSM97137 1 0.0000 0.9485 1.000 0.000 0.000
#> GSM97118 1 0.0237 0.9470 0.996 0.004 0.000
#> GSM97114 2 0.1753 0.9077 0.048 0.952 0.000
#> GSM97142 1 0.0000 0.9485 1.000 0.000 0.000
#> GSM97140 2 0.0237 0.9268 0.000 0.996 0.004
#> GSM97141 2 0.1643 0.9235 0.000 0.956 0.044
#> GSM97055 1 0.0424 0.9438 0.992 0.000 0.008
#> GSM97090 1 0.4178 0.8004 0.828 0.172 0.000
#> GSM97091 1 0.0000 0.9485 1.000 0.000 0.000
#> GSM97148 1 0.0000 0.9485 1.000 0.000 0.000
#> GSM97063 1 0.0000 0.9485 1.000 0.000 0.000
#> GSM97053 1 0.0000 0.9485 1.000 0.000 0.000
#> GSM97066 3 0.0000 0.8955 0.000 0.000 1.000
#> GSM97079 2 0.1964 0.9146 0.000 0.944 0.056
#> GSM97083 1 0.1289 0.9319 0.968 0.032 0.000
#> GSM97084 2 0.1289 0.9236 0.000 0.968 0.032
#> GSM97094 1 0.2711 0.8913 0.912 0.088 0.000
#> GSM97096 3 0.0237 0.8941 0.000 0.004 0.996
#> GSM97097 2 0.4452 0.7729 0.000 0.808 0.192
#> GSM97107 1 0.3532 0.8671 0.884 0.108 0.008
#> GSM97054 2 0.0424 0.9266 0.000 0.992 0.008
#> GSM97062 2 0.1964 0.9135 0.000 0.944 0.056
#> GSM97069 3 0.0000 0.8955 0.000 0.000 1.000
#> GSM97070 3 0.0000 0.8955 0.000 0.000 1.000
#> GSM97073 3 0.0000 0.8955 0.000 0.000 1.000
#> GSM97076 1 0.1860 0.9070 0.948 0.000 0.052
#> GSM97077 2 0.1163 0.9304 0.000 0.972 0.028
#> GSM97095 1 0.6192 0.3447 0.580 0.420 0.000
#> GSM97102 3 0.0000 0.8955 0.000 0.000 1.000
#> GSM97109 2 0.2550 0.9144 0.012 0.932 0.056
#> GSM97110 2 0.2959 0.8881 0.000 0.900 0.100
#> GSM97074 3 0.6104 0.4512 0.348 0.004 0.648
#> GSM97085 3 0.0237 0.8934 0.000 0.004 0.996
#> GSM97059 2 0.0000 0.9252 0.000 1.000 0.000
#> GSM97072 3 0.0000 0.8955 0.000 0.000 1.000
#> GSM97078 1 0.7411 0.2078 0.548 0.036 0.416
#> GSM97067 3 0.0000 0.8955 0.000 0.000 1.000
#> GSM97087 3 0.0000 0.8955 0.000 0.000 1.000
#> GSM97111 2 0.4121 0.8182 0.000 0.832 0.168
#> GSM97064 2 0.4235 0.8150 0.000 0.824 0.176
#> GSM97065 3 0.5138 0.6414 0.000 0.252 0.748
#> GSM97081 3 0.0237 0.8938 0.000 0.004 0.996
#> GSM97082 3 0.0000 0.8955 0.000 0.000 1.000
#> GSM97088 3 0.0892 0.8851 0.000 0.020 0.980
#> GSM97100 2 0.0237 0.9268 0.000 0.996 0.004
#> GSM97104 3 0.0000 0.8955 0.000 0.000 1.000
#> GSM97108 2 0.0592 0.9287 0.000 0.988 0.012
#> GSM97050 2 0.2537 0.9122 0.000 0.920 0.080
#> GSM97080 3 0.0000 0.8955 0.000 0.000 1.000
#> GSM97089 3 0.0000 0.8955 0.000 0.000 1.000
#> GSM97092 3 0.0000 0.8955 0.000 0.000 1.000
#> GSM97093 2 0.6140 0.3474 0.000 0.596 0.404
#> GSM97058 2 0.1411 0.9288 0.000 0.964 0.036
#> GSM97051 2 0.2356 0.9144 0.000 0.928 0.072
#> GSM97052 3 0.0000 0.8955 0.000 0.000 1.000
#> GSM97061 3 0.2959 0.8289 0.000 0.100 0.900
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM97138 1 0.0779 0.952 0.980 0.016 0.000 0.004
#> GSM97145 1 0.0895 0.950 0.976 0.020 0.000 0.004
#> GSM97147 2 0.4638 0.729 0.060 0.788 0.000 0.152
#> GSM97125 1 0.0376 0.953 0.992 0.004 0.000 0.004
#> GSM97127 1 0.0779 0.952 0.980 0.016 0.000 0.004
#> GSM97130 4 0.4761 0.526 0.332 0.004 0.000 0.664
#> GSM97133 1 0.0779 0.952 0.980 0.016 0.000 0.004
#> GSM97134 4 0.4040 0.697 0.248 0.000 0.000 0.752
#> GSM97120 1 0.0895 0.950 0.976 0.020 0.000 0.004
#> GSM97126 1 0.0921 0.939 0.972 0.028 0.000 0.000
#> GSM97112 1 0.0188 0.954 0.996 0.000 0.000 0.004
#> GSM97115 4 0.1610 0.846 0.032 0.016 0.000 0.952
#> GSM97116 1 0.0779 0.952 0.980 0.016 0.000 0.004
#> GSM97117 2 0.1824 0.807 0.004 0.936 0.060 0.000
#> GSM97119 1 0.0188 0.954 0.996 0.000 0.000 0.004
#> GSM97122 1 0.0188 0.954 0.996 0.000 0.000 0.004
#> GSM97135 1 0.0188 0.954 0.996 0.000 0.000 0.004
#> GSM97136 3 0.4359 0.770 0.084 0.100 0.816 0.000
#> GSM97139 1 0.0779 0.952 0.980 0.016 0.000 0.004
#> GSM97146 1 0.0779 0.952 0.980 0.016 0.000 0.004
#> GSM97123 3 0.4675 0.651 0.000 0.244 0.736 0.020
#> GSM97129 2 0.8125 0.451 0.144 0.544 0.252 0.060
#> GSM97143 1 0.0188 0.954 0.996 0.000 0.000 0.004
#> GSM97113 2 0.0779 0.811 0.004 0.980 0.000 0.016
#> GSM97056 1 0.4228 0.689 0.760 0.008 0.000 0.232
#> GSM97124 1 0.0188 0.954 0.996 0.000 0.000 0.004
#> GSM97132 1 0.2589 0.859 0.884 0.000 0.000 0.116
#> GSM97144 4 0.2921 0.801 0.140 0.000 0.000 0.860
#> GSM97149 1 0.0779 0.952 0.980 0.016 0.000 0.004
#> GSM97068 4 0.4877 0.148 0.000 0.408 0.000 0.592
#> GSM97071 4 0.2983 0.807 0.008 0.004 0.108 0.880
#> GSM97086 4 0.2011 0.798 0.000 0.080 0.000 0.920
#> GSM97103 3 0.4719 0.726 0.000 0.180 0.772 0.048
#> GSM97057 2 0.1716 0.806 0.000 0.936 0.000 0.064
#> GSM97060 3 0.0188 0.884 0.000 0.004 0.996 0.000
#> GSM97075 2 0.4072 0.672 0.000 0.748 0.252 0.000
#> GSM97098 3 0.3893 0.738 0.000 0.196 0.796 0.008
#> GSM97099 2 0.2281 0.795 0.000 0.904 0.096 0.000
#> GSM97101 2 0.0336 0.811 0.000 0.992 0.000 0.008
#> GSM97105 2 0.2011 0.802 0.000 0.920 0.000 0.080
#> GSM97106 3 0.2197 0.863 0.000 0.048 0.928 0.024
#> GSM97121 2 0.0921 0.811 0.000 0.972 0.000 0.028
#> GSM97128 4 0.6074 0.634 0.104 0.000 0.228 0.668
#> GSM97131 2 0.6531 0.640 0.000 0.636 0.160 0.204
#> GSM97137 1 0.4284 0.708 0.764 0.012 0.000 0.224
#> GSM97118 1 0.2589 0.857 0.884 0.000 0.000 0.116
#> GSM97114 2 0.1109 0.804 0.028 0.968 0.000 0.004
#> GSM97142 1 0.0188 0.954 0.996 0.000 0.000 0.004
#> GSM97140 2 0.1557 0.808 0.000 0.944 0.000 0.056
#> GSM97141 2 0.0376 0.809 0.004 0.992 0.004 0.000
#> GSM97055 1 0.0376 0.952 0.992 0.000 0.004 0.004
#> GSM97090 4 0.1118 0.846 0.036 0.000 0.000 0.964
#> GSM97091 1 0.0188 0.954 0.996 0.000 0.000 0.004
#> GSM97148 1 0.0779 0.952 0.980 0.016 0.000 0.004
#> GSM97063 1 0.0188 0.954 0.996 0.000 0.000 0.004
#> GSM97053 1 0.0188 0.954 0.996 0.000 0.000 0.004
#> GSM97066 3 0.0000 0.885 0.000 0.000 1.000 0.000
#> GSM97079 4 0.1109 0.834 0.000 0.028 0.004 0.968
#> GSM97083 4 0.3052 0.806 0.136 0.000 0.004 0.860
#> GSM97084 4 0.0469 0.838 0.000 0.012 0.000 0.988
#> GSM97094 4 0.1022 0.846 0.032 0.000 0.000 0.968
#> GSM97096 3 0.1661 0.869 0.000 0.052 0.944 0.004
#> GSM97097 4 0.4257 0.721 0.000 0.048 0.140 0.812
#> GSM97107 4 0.0921 0.846 0.028 0.000 0.000 0.972
#> GSM97054 4 0.1022 0.833 0.000 0.032 0.000 0.968
#> GSM97062 4 0.0336 0.839 0.000 0.008 0.000 0.992
#> GSM97069 3 0.0000 0.885 0.000 0.000 1.000 0.000
#> GSM97070 3 0.0000 0.885 0.000 0.000 1.000 0.000
#> GSM97073 3 0.0000 0.885 0.000 0.000 1.000 0.000
#> GSM97076 1 0.5173 0.783 0.800 0.068 0.080 0.052
#> GSM97077 2 0.5988 0.672 0.000 0.676 0.100 0.224
#> GSM97095 4 0.3617 0.803 0.064 0.076 0.000 0.860
#> GSM97102 3 0.0336 0.884 0.000 0.008 0.992 0.000
#> GSM97109 2 0.1697 0.808 0.016 0.952 0.028 0.004
#> GSM97110 2 0.1732 0.809 0.004 0.948 0.040 0.008
#> GSM97074 3 0.7469 0.187 0.312 0.000 0.488 0.200
#> GSM97085 3 0.0672 0.879 0.008 0.000 0.984 0.008
#> GSM97059 2 0.4679 0.516 0.000 0.648 0.000 0.352
#> GSM97072 3 0.0000 0.885 0.000 0.000 1.000 0.000
#> GSM97078 4 0.3325 0.802 0.024 0.000 0.112 0.864
#> GSM97067 3 0.0000 0.885 0.000 0.000 1.000 0.000
#> GSM97087 3 0.0336 0.884 0.000 0.008 0.992 0.000
#> GSM97111 2 0.1296 0.812 0.004 0.964 0.028 0.004
#> GSM97064 2 0.6111 0.354 0.000 0.556 0.392 0.052
#> GSM97065 2 0.4741 0.546 0.004 0.668 0.328 0.000
#> GSM97081 3 0.2469 0.828 0.000 0.108 0.892 0.000
#> GSM97082 3 0.0000 0.885 0.000 0.000 1.000 0.000
#> GSM97088 3 0.5376 0.259 0.016 0.000 0.588 0.396
#> GSM97100 2 0.3942 0.696 0.000 0.764 0.000 0.236
#> GSM97104 3 0.0188 0.885 0.000 0.004 0.996 0.000
#> GSM97108 2 0.1302 0.810 0.000 0.956 0.000 0.044
#> GSM97050 2 0.7211 0.540 0.000 0.548 0.248 0.204
#> GSM97080 3 0.0000 0.885 0.000 0.000 1.000 0.000
#> GSM97089 3 0.0336 0.884 0.000 0.008 0.992 0.000
#> GSM97092 3 0.1305 0.875 0.000 0.036 0.960 0.004
#> GSM97093 3 0.6808 0.223 0.004 0.368 0.536 0.092
#> GSM97058 2 0.4171 0.785 0.000 0.828 0.084 0.088
#> GSM97051 2 0.7875 0.278 0.000 0.384 0.288 0.328
#> GSM97052 3 0.1576 0.870 0.000 0.048 0.948 0.004
#> GSM97061 3 0.3278 0.810 0.000 0.116 0.864 0.020
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM97138 1 0.3143 0.79222 0.796 0.000 0.000 0.000 0.204
#> GSM97145 1 0.2929 0.80206 0.856 0.012 0.000 0.004 0.128
#> GSM97147 2 0.6421 0.37402 0.060 0.608 0.000 0.092 0.240
#> GSM97125 1 0.1851 0.80636 0.912 0.000 0.000 0.000 0.088
#> GSM97127 1 0.3123 0.79388 0.812 0.000 0.000 0.004 0.184
#> GSM97130 4 0.5510 0.49389 0.184 0.000 0.000 0.652 0.164
#> GSM97133 1 0.3689 0.77428 0.740 0.000 0.000 0.004 0.256
#> GSM97134 4 0.4960 0.53657 0.268 0.000 0.000 0.668 0.064
#> GSM97120 1 0.3579 0.78036 0.756 0.000 0.000 0.004 0.240
#> GSM97126 1 0.5536 0.69131 0.712 0.108 0.000 0.044 0.136
#> GSM97112 1 0.0693 0.79746 0.980 0.000 0.000 0.008 0.012
#> GSM97115 4 0.4096 0.62234 0.024 0.012 0.000 0.772 0.192
#> GSM97116 1 0.3607 0.77882 0.752 0.000 0.000 0.004 0.244
#> GSM97117 2 0.1329 0.60377 0.004 0.956 0.032 0.000 0.008
#> GSM97119 1 0.0451 0.79946 0.988 0.000 0.000 0.008 0.004
#> GSM97122 1 0.0451 0.79946 0.988 0.000 0.000 0.008 0.004
#> GSM97135 1 0.0290 0.80013 0.992 0.000 0.000 0.008 0.000
#> GSM97136 3 0.7991 0.21539 0.160 0.192 0.496 0.012 0.140
#> GSM97139 1 0.3635 0.77751 0.748 0.000 0.000 0.004 0.248
#> GSM97146 1 0.3689 0.77428 0.740 0.000 0.000 0.004 0.256
#> GSM97123 3 0.6194 0.21454 0.000 0.208 0.588 0.008 0.196
#> GSM97129 2 0.8437 0.10453 0.180 0.476 0.152 0.036 0.156
#> GSM97143 1 0.1403 0.78844 0.952 0.000 0.000 0.024 0.024
#> GSM97113 2 0.2970 0.58866 0.000 0.828 0.000 0.004 0.168
#> GSM97056 1 0.6493 0.47084 0.492 0.000 0.000 0.260 0.248
#> GSM97124 1 0.0912 0.80367 0.972 0.000 0.000 0.012 0.016
#> GSM97132 1 0.5304 0.44611 0.640 0.000 0.000 0.272 0.088
#> GSM97144 4 0.3736 0.63043 0.140 0.000 0.000 0.808 0.052
#> GSM97149 1 0.3989 0.76923 0.728 0.004 0.000 0.008 0.260
#> GSM97068 4 0.6418 0.22027 0.000 0.184 0.000 0.472 0.344
#> GSM97071 4 0.5184 0.50314 0.000 0.000 0.176 0.688 0.136
#> GSM97086 4 0.4851 0.42469 0.000 0.036 0.000 0.624 0.340
#> GSM97103 3 0.6703 0.34253 0.000 0.208 0.600 0.076 0.116
#> GSM97057 2 0.4540 0.40119 0.000 0.640 0.000 0.020 0.340
#> GSM97060 3 0.1608 0.71125 0.000 0.000 0.928 0.000 0.072
#> GSM97075 2 0.5955 0.16347 0.000 0.596 0.256 0.004 0.144
#> GSM97098 3 0.5441 0.35200 0.000 0.280 0.624 0.000 0.096
#> GSM97099 2 0.3441 0.54888 0.000 0.848 0.088 0.008 0.056
#> GSM97101 2 0.1121 0.60498 0.000 0.956 0.000 0.000 0.044
#> GSM97105 2 0.3993 0.45754 0.000 0.756 0.000 0.028 0.216
#> GSM97106 3 0.4295 0.56558 0.000 0.020 0.740 0.012 0.228
#> GSM97121 2 0.1410 0.60253 0.000 0.940 0.000 0.000 0.060
#> GSM97128 4 0.7779 0.37497 0.188 0.000 0.180 0.488 0.144
#> GSM97131 2 0.7934 -0.43905 0.000 0.376 0.124 0.148 0.352
#> GSM97137 1 0.6687 0.31453 0.420 0.000 0.000 0.332 0.248
#> GSM97118 1 0.5976 0.40180 0.616 0.000 0.016 0.252 0.116
#> GSM97114 2 0.2408 0.59338 0.008 0.892 0.000 0.004 0.096
#> GSM97142 1 0.0693 0.79713 0.980 0.000 0.000 0.008 0.012
#> GSM97140 2 0.3805 0.51365 0.000 0.784 0.000 0.032 0.184
#> GSM97141 2 0.0794 0.60743 0.000 0.972 0.000 0.000 0.028
#> GSM97055 1 0.2954 0.76527 0.888 0.004 0.024 0.024 0.060
#> GSM97090 4 0.3608 0.63918 0.040 0.000 0.000 0.812 0.148
#> GSM97091 1 0.1522 0.78207 0.944 0.000 0.000 0.012 0.044
#> GSM97148 1 0.3809 0.77217 0.736 0.000 0.000 0.008 0.256
#> GSM97063 1 0.0898 0.79427 0.972 0.000 0.000 0.008 0.020
#> GSM97053 1 0.1430 0.80714 0.944 0.000 0.000 0.004 0.052
#> GSM97066 3 0.1892 0.70901 0.000 0.000 0.916 0.004 0.080
#> GSM97079 4 0.4698 0.48640 0.000 0.028 0.004 0.664 0.304
#> GSM97083 4 0.4819 0.58917 0.148 0.000 0.004 0.736 0.112
#> GSM97084 4 0.3274 0.58355 0.000 0.000 0.000 0.780 0.220
#> GSM97094 4 0.2813 0.64655 0.032 0.004 0.000 0.880 0.084
#> GSM97096 3 0.3937 0.65718 0.000 0.072 0.808 0.004 0.116
#> GSM97097 4 0.6840 0.29109 0.000 0.060 0.116 0.552 0.272
#> GSM97107 4 0.1965 0.65026 0.024 0.000 0.000 0.924 0.052
#> GSM97054 4 0.4445 0.50702 0.000 0.024 0.000 0.676 0.300
#> GSM97062 4 0.3790 0.54557 0.000 0.004 0.000 0.724 0.272
#> GSM97069 3 0.1121 0.72217 0.000 0.000 0.956 0.000 0.044
#> GSM97070 3 0.1638 0.71816 0.000 0.000 0.932 0.004 0.064
#> GSM97073 3 0.1892 0.71295 0.000 0.000 0.916 0.004 0.080
#> GSM97076 1 0.8563 0.36553 0.464 0.084 0.144 0.076 0.232
#> GSM97077 5 0.7331 0.22034 0.000 0.388 0.064 0.136 0.412
#> GSM97095 4 0.5355 0.57286 0.024 0.076 0.000 0.696 0.204
#> GSM97102 3 0.1484 0.72309 0.000 0.008 0.944 0.000 0.048
#> GSM97109 2 0.3476 0.57498 0.004 0.844 0.016 0.020 0.116
#> GSM97110 2 0.4361 0.54898 0.000 0.780 0.040 0.024 0.156
#> GSM97074 3 0.8368 0.00668 0.220 0.000 0.372 0.228 0.180
#> GSM97085 3 0.5081 0.54547 0.064 0.000 0.736 0.036 0.164
#> GSM97059 2 0.6773 0.04705 0.004 0.424 0.000 0.228 0.344
#> GSM97072 3 0.1430 0.72092 0.000 0.000 0.944 0.004 0.052
#> GSM97078 4 0.6240 0.54583 0.104 0.000 0.076 0.656 0.164
#> GSM97067 3 0.1732 0.71313 0.000 0.000 0.920 0.000 0.080
#> GSM97087 3 0.2074 0.70003 0.000 0.000 0.896 0.000 0.104
#> GSM97111 2 0.2012 0.60338 0.000 0.920 0.020 0.000 0.060
#> GSM97064 5 0.7350 0.54102 0.000 0.228 0.368 0.032 0.372
#> GSM97065 2 0.6362 0.03610 0.000 0.484 0.364 0.004 0.148
#> GSM97081 3 0.4238 0.60668 0.000 0.136 0.776 0.000 0.088
#> GSM97082 3 0.1197 0.72076 0.000 0.000 0.952 0.000 0.048
#> GSM97088 3 0.7640 -0.10204 0.084 0.000 0.388 0.376 0.152
#> GSM97100 2 0.6134 -0.01013 0.000 0.516 0.000 0.144 0.340
#> GSM97104 3 0.0703 0.72459 0.000 0.000 0.976 0.000 0.024
#> GSM97108 2 0.2011 0.58849 0.000 0.908 0.000 0.004 0.088
#> GSM97050 5 0.7880 0.61436 0.000 0.252 0.204 0.104 0.440
#> GSM97080 3 0.1197 0.72406 0.000 0.000 0.952 0.000 0.048
#> GSM97089 3 0.2127 0.69956 0.000 0.000 0.892 0.000 0.108
#> GSM97092 3 0.3013 0.64883 0.000 0.008 0.832 0.000 0.160
#> GSM97093 5 0.8122 0.47303 0.008 0.292 0.292 0.068 0.340
#> GSM97058 2 0.6622 -0.19259 0.000 0.500 0.088 0.044 0.368
#> GSM97051 5 0.7968 0.58896 0.000 0.180 0.192 0.168 0.460
#> GSM97052 3 0.3475 0.61714 0.000 0.012 0.804 0.004 0.180
#> GSM97061 3 0.4870 0.41427 0.000 0.040 0.680 0.008 0.272
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM97138 1 0.2445 0.6196 0.868 0.008 0.000 0.004 0.120 0.000
#> GSM97145 1 0.4127 0.6203 0.716 0.016 0.000 0.016 0.248 0.004
#> GSM97147 2 0.7724 0.2418 0.116 0.476 0.000 0.116 0.080 0.212
#> GSM97125 1 0.3972 0.6265 0.664 0.000 0.000 0.012 0.320 0.004
#> GSM97127 1 0.3051 0.6241 0.824 0.004 0.000 0.012 0.156 0.004
#> GSM97130 4 0.6018 0.2646 0.344 0.000 0.000 0.488 0.148 0.020
#> GSM97133 1 0.0405 0.5956 0.988 0.000 0.000 0.004 0.008 0.000
#> GSM97134 4 0.6324 0.0383 0.084 0.004 0.000 0.440 0.408 0.064
#> GSM97120 1 0.0632 0.5972 0.976 0.000 0.000 0.000 0.024 0.000
#> GSM97126 1 0.6200 0.2451 0.440 0.092 0.000 0.016 0.424 0.028
#> GSM97112 1 0.3950 0.5855 0.564 0.000 0.000 0.004 0.432 0.000
#> GSM97115 4 0.6012 0.5150 0.136 0.012 0.000 0.644 0.116 0.092
#> GSM97116 1 0.0632 0.6036 0.976 0.000 0.000 0.000 0.024 0.000
#> GSM97117 2 0.1707 0.5838 0.000 0.928 0.004 0.000 0.012 0.056
#> GSM97119 1 0.4088 0.5867 0.556 0.000 0.000 0.004 0.436 0.004
#> GSM97122 1 0.4056 0.6010 0.576 0.000 0.000 0.004 0.416 0.004
#> GSM97135 1 0.4002 0.6059 0.588 0.000 0.000 0.008 0.404 0.000
#> GSM97136 3 0.8387 0.1330 0.060 0.212 0.372 0.012 0.232 0.112
#> GSM97139 1 0.0547 0.6035 0.980 0.000 0.000 0.000 0.020 0.000
#> GSM97146 1 0.0291 0.5911 0.992 0.000 0.000 0.004 0.004 0.000
#> GSM97123 3 0.6347 0.2759 0.000 0.144 0.452 0.012 0.020 0.372
#> GSM97129 2 0.8697 0.1658 0.100 0.412 0.092 0.044 0.208 0.144
#> GSM97143 1 0.4304 0.5566 0.536 0.000 0.000 0.008 0.448 0.008
#> GSM97113 2 0.5116 0.4940 0.128 0.696 0.004 0.000 0.028 0.144
#> GSM97056 1 0.4792 0.1977 0.672 0.000 0.000 0.232 0.088 0.008
#> GSM97124 1 0.4325 0.5966 0.568 0.000 0.000 0.016 0.412 0.004
#> GSM97132 5 0.6528 -0.0544 0.360 0.000 0.000 0.192 0.412 0.036
#> GSM97144 4 0.5204 0.3937 0.068 0.000 0.000 0.636 0.264 0.032
#> GSM97149 1 0.0551 0.5852 0.984 0.008 0.000 0.004 0.004 0.000
#> GSM97068 4 0.8124 0.1896 0.140 0.124 0.000 0.416 0.092 0.228
#> GSM97071 4 0.6212 0.2416 0.000 0.000 0.240 0.556 0.148 0.056
#> GSM97086 4 0.4552 0.3855 0.000 0.024 0.000 0.668 0.028 0.280
#> GSM97103 3 0.8021 0.2348 0.000 0.208 0.432 0.124 0.076 0.160
#> GSM97057 2 0.7050 0.1758 0.172 0.468 0.000 0.044 0.032 0.284
#> GSM97060 3 0.3730 0.6497 0.000 0.004 0.796 0.020 0.028 0.152
#> GSM97075 2 0.6651 0.0934 0.000 0.480 0.256 0.004 0.044 0.216
#> GSM97098 3 0.7096 0.2755 0.000 0.296 0.432 0.012 0.064 0.196
#> GSM97099 2 0.4298 0.5283 0.004 0.796 0.048 0.012 0.052 0.088
#> GSM97101 2 0.2349 0.5809 0.008 0.892 0.000 0.000 0.020 0.080
#> GSM97105 2 0.5115 0.3607 0.000 0.624 0.004 0.040 0.032 0.300
#> GSM97106 3 0.6157 0.4244 0.000 0.044 0.528 0.056 0.028 0.344
#> GSM97121 2 0.3652 0.5515 0.004 0.796 0.000 0.004 0.048 0.148
#> GSM97128 5 0.6865 0.2449 0.004 0.000 0.212 0.220 0.488 0.076
#> GSM97131 6 0.7229 0.2803 0.000 0.320 0.076 0.148 0.024 0.432
#> GSM97137 1 0.5090 0.1133 0.624 0.000 0.000 0.272 0.096 0.008
#> GSM97118 5 0.6598 0.1789 0.260 0.000 0.012 0.132 0.532 0.064
#> GSM97114 2 0.3340 0.5723 0.100 0.840 0.000 0.004 0.024 0.032
#> GSM97142 1 0.3955 0.5846 0.560 0.000 0.000 0.004 0.436 0.000
#> GSM97140 2 0.5654 0.4124 0.012 0.620 0.000 0.044 0.064 0.260
#> GSM97141 2 0.2094 0.5857 0.016 0.908 0.000 0.000 0.008 0.068
#> GSM97055 5 0.5315 -0.4128 0.448 0.008 0.040 0.004 0.488 0.012
#> GSM97090 4 0.5457 0.5413 0.116 0.000 0.000 0.676 0.132 0.076
#> GSM97091 1 0.3996 0.5208 0.512 0.000 0.000 0.004 0.484 0.000
#> GSM97148 1 0.0291 0.5911 0.992 0.000 0.000 0.004 0.004 0.000
#> GSM97063 1 0.3847 0.5631 0.544 0.000 0.000 0.000 0.456 0.000
#> GSM97053 1 0.4090 0.6137 0.604 0.000 0.000 0.008 0.384 0.004
#> GSM97066 3 0.2398 0.6333 0.000 0.000 0.876 0.000 0.104 0.020
#> GSM97079 4 0.4648 0.3998 0.000 0.004 0.016 0.668 0.036 0.276
#> GSM97083 4 0.5300 0.1277 0.008 0.000 0.008 0.492 0.436 0.056
#> GSM97084 4 0.2882 0.5140 0.000 0.000 0.000 0.812 0.008 0.180
#> GSM97094 4 0.3417 0.5591 0.004 0.000 0.000 0.812 0.132 0.052
#> GSM97096 3 0.6261 0.5218 0.000 0.100 0.584 0.016 0.060 0.240
#> GSM97097 4 0.6206 0.3188 0.000 0.068 0.036 0.612 0.064 0.220
#> GSM97107 4 0.2619 0.5742 0.008 0.000 0.000 0.880 0.072 0.040
#> GSM97054 4 0.4222 0.4302 0.000 0.016 0.000 0.708 0.028 0.248
#> GSM97062 4 0.3424 0.4934 0.000 0.004 0.000 0.780 0.020 0.196
#> GSM97069 3 0.1584 0.6563 0.000 0.000 0.928 0.000 0.064 0.008
#> GSM97070 3 0.1858 0.6494 0.000 0.000 0.912 0.000 0.076 0.012
#> GSM97073 3 0.2701 0.6361 0.000 0.004 0.864 0.000 0.104 0.028
#> GSM97076 5 0.8952 0.2796 0.244 0.080 0.244 0.068 0.304 0.060
#> GSM97077 6 0.6562 0.4452 0.000 0.232 0.048 0.092 0.052 0.576
#> GSM97095 4 0.7744 0.4012 0.164 0.064 0.000 0.468 0.196 0.108
#> GSM97102 3 0.3971 0.6445 0.000 0.036 0.808 0.008 0.056 0.092
#> GSM97109 2 0.5154 0.5018 0.040 0.748 0.016 0.024 0.088 0.084
#> GSM97110 2 0.6139 0.4554 0.040 0.676 0.048 0.024 0.084 0.128
#> GSM97074 5 0.6284 0.1117 0.004 0.000 0.424 0.084 0.428 0.060
#> GSM97085 3 0.5050 0.3201 0.000 0.000 0.640 0.032 0.276 0.052
#> GSM97059 6 0.8554 0.0537 0.168 0.252 0.000 0.252 0.072 0.256
#> GSM97072 3 0.2442 0.6609 0.000 0.000 0.884 0.000 0.068 0.048
#> GSM97078 5 0.6871 -0.0549 0.000 0.000 0.136 0.376 0.392 0.096
#> GSM97067 3 0.2006 0.6457 0.000 0.000 0.904 0.000 0.080 0.016
#> GSM97087 3 0.3828 0.5887 0.000 0.000 0.696 0.004 0.012 0.288
#> GSM97111 2 0.3380 0.5692 0.000 0.832 0.016 0.000 0.056 0.096
#> GSM97064 6 0.5771 0.5182 0.000 0.116 0.188 0.048 0.008 0.640
#> GSM97065 2 0.7658 0.1402 0.036 0.412 0.308 0.004 0.148 0.092
#> GSM97081 3 0.5276 0.5816 0.000 0.108 0.676 0.000 0.044 0.172
#> GSM97082 3 0.2445 0.6646 0.000 0.000 0.872 0.000 0.020 0.108
#> GSM97088 3 0.6769 -0.0600 0.000 0.000 0.460 0.144 0.308 0.088
#> GSM97100 2 0.6543 -0.0581 0.000 0.420 0.000 0.172 0.044 0.364
#> GSM97104 3 0.1728 0.6706 0.000 0.000 0.924 0.004 0.008 0.064
#> GSM97108 2 0.4238 0.4770 0.000 0.720 0.000 0.016 0.036 0.228
#> GSM97050 6 0.6010 0.5600 0.000 0.132 0.068 0.132 0.020 0.648
#> GSM97080 3 0.1856 0.6662 0.000 0.000 0.920 0.000 0.032 0.048
#> GSM97089 3 0.4283 0.5853 0.000 0.004 0.676 0.004 0.028 0.288
#> GSM97092 3 0.4197 0.5563 0.000 0.012 0.660 0.004 0.008 0.316
#> GSM97093 6 0.7211 0.3886 0.012 0.120 0.152 0.068 0.076 0.572
#> GSM97058 6 0.6379 0.4507 0.000 0.252 0.100 0.048 0.028 0.572
#> GSM97051 6 0.5517 0.5657 0.000 0.076 0.068 0.152 0.016 0.688
#> GSM97052 3 0.4289 0.5294 0.000 0.012 0.636 0.004 0.008 0.340
#> GSM97061 3 0.5210 0.3055 0.000 0.036 0.488 0.016 0.008 0.452
Heatmaps for the consensus matrix. It visualizes the probability of two samples to be in a same group.
consensus_heatmap(res, k = 2)
consensus_heatmap(res, k = 3)
consensus_heatmap(res, k = 4)
consensus_heatmap(res, k = 5)
consensus_heatmap(res, k = 6)
Heatmaps for the membership of samples in all partitions to see how consistent they are:
membership_heatmap(res, k = 2)
membership_heatmap(res, k = 3)
membership_heatmap(res, k = 4)
membership_heatmap(res, k = 5)
membership_heatmap(res, k = 6)
As soon as we have had the classes for columns, we can look for signatures which are significantly different between classes which can be candidate marks for certain classes. Following are the heatmaps for signatures.
Signature heatmaps where rows are scaled:
get_signatures(res, k = 2)
get_signatures(res, k = 3)
get_signatures(res, k = 4)
get_signatures(res, k = 5)
get_signatures(res, k = 6)
Signature heatmaps where rows are not scaled:
get_signatures(res, k = 2, scale_rows = FALSE)
get_signatures(res, k = 3, scale_rows = FALSE)
get_signatures(res, k = 4, scale_rows = FALSE)
get_signatures(res, k = 5, scale_rows = FALSE)
get_signatures(res, k = 6, scale_rows = FALSE)
Compare the overlap of signatures from different k:
compare_signatures(res)
get_signature() returns a data frame invisibly. TO get the list of signatures, the function
call should be assigned to a variable explicitly. In following code, if plot argument is set
to FALSE, no heatmap is plotted while only the differential analysis is performed.
# code only for demonstration
tb = get_signature(res, k = ..., plot = FALSE)
An example of the output of tb is:
#> which_row fdr mean_1 mean_2 scaled_mean_1 scaled_mean_2 km
#> 1 38 0.042760348 8.373488 9.131774 -0.5533452 0.5164555 1
#> 2 40 0.018707592 7.106213 8.469186 -0.6173731 0.5762149 1
#> 3 55 0.019134737 10.221463 11.207825 -0.6159697 0.5749050 1
#> 4 59 0.006059896 5.921854 7.869574 -0.6899429 0.6439467 1
#> 5 60 0.018055526 8.928898 10.211722 -0.6204761 0.5791110 1
#> 6 98 0.009384629 15.714769 14.887706 0.6635654 -0.6193277 2
...
The columns in tb are:
which_row: row indices corresponding to the input matrix.fdr: FDR for the differential test. mean_x: The mean value in group x.scaled_mean_x: The mean value in group x after rows are scaled.km: Row groups if k-means clustering is applied to rows.UMAP plot which shows how samples are separated.
dimension_reduction(res, k = 2, method = "UMAP")
dimension_reduction(res, k = 3, method = "UMAP")
dimension_reduction(res, k = 4, method = "UMAP")
dimension_reduction(res, k = 5, method = "UMAP")
dimension_reduction(res, k = 6, method = "UMAP")
Following heatmap shows how subgroups are split when increasing k:
collect_classes(res)
Test correlation between subgroups and known annotations. If the known annotation is numeric, one-way ANOVA test is applied, and if the known annotation is discrete, chi-squared contingency table test is applied.
test_to_known_factors(res)
#> n disease.state(p) specimen(p) cell.type(p) other(p) k
#> MAD:skmeans 97 5.64e-04 0.415 3.09e-13 0.102 2
#> MAD:skmeans 92 1.86e-04 0.410 1.35e-13 0.379 3
#> MAD:skmeans 93 6.25e-04 0.424 3.06e-15 0.192 4
#> MAD:skmeans 70 2.27e-05 0.422 1.45e-12 0.193 5
#> MAD:skmeans 54 2.48e-05 0.181 7.58e-09 0.181 6
If matrix rows can be associated to genes, consider to use functional_enrichment(res,
...) to perform function enrichment for the signature genes. See this vignette for more detailed explanations.
The object with results only for a single top-value method and a single partition method can be extracted as:
res = res_list["MAD", "pam"]
# you can also extract it by
# res = res_list["MAD:pam"]
A summary of res and all the functions that can be applied to it:
res
#> A 'ConsensusPartition' object with k = 2, 3, 4, 5, 6.
#> On a matrix with 21512 rows and 100 columns.
#> Top rows (1000, 2000, 3000, 4000, 5000) are extracted by 'MAD' method.
#> Subgroups are detected by 'pam' method.
#> Performed in total 1250 partitions by row resampling.
#> Best k for subgroups seems to be 3.
#>
#> Following methods can be applied to this 'ConsensusPartition' object:
#> [1] "cola_report" "collect_classes" "collect_plots"
#> [4] "collect_stats" "colnames" "compare_signatures"
#> [7] "consensus_heatmap" "dimension_reduction" "functional_enrichment"
#> [10] "get_anno_col" "get_anno" "get_classes"
#> [13] "get_consensus" "get_matrix" "get_membership"
#> [16] "get_param" "get_signatures" "get_stats"
#> [19] "is_best_k" "is_stable_k" "membership_heatmap"
#> [22] "ncol" "nrow" "plot_ecdf"
#> [25] "rownames" "select_partition_number" "show"
#> [28] "suggest_best_k" "test_to_known_factors"
collect_plots() function collects all the plots made from res for all k (number of partitions)
into one single page to provide an easy and fast comparison between different k.
collect_plots(res)
The plots are:
k and the heatmap of
predicted classes for each k.k.k.k.All the plots in panels can be made by individual functions and they are plotted later in this section.
select_partition_number() produces several plots showing different
statistics for choosing “optimized” k. There are following statistics:
k;k, the area increased is defined as \(A_k - A_{k-1}\).The detailed explanations of these statistics can be found in the cola vignette.
Generally speaking, lower PAC score, higher mean silhouette score or higher
concordance corresponds to better partition. Rand index and Jaccard index
measure how similar the current partition is compared to partition with k-1.
If they are too similar, we won't accept k is better than k-1.
select_partition_number(res)
The numeric values for all these statistics can be obtained by get_stats().
get_stats(res)
#> k 1-PAC mean_silhouette concordance area_increased Rand Jaccard
#> 2 2 0.597 0.733 0.899 0.4544 0.529 0.529
#> 3 3 0.727 0.839 0.924 0.4358 0.666 0.447
#> 4 4 0.662 0.741 0.839 0.1101 0.892 0.704
#> 5 5 0.629 0.684 0.767 0.0615 0.959 0.855
#> 6 6 0.684 0.515 0.720 0.0584 0.919 0.683
suggest_best_k() suggests the best \(k\) based on these statistics. The rules are as follows:
suggest_best_k(res)
#> [1] 3
Following shows the table of the partitions (You need to click the show/hide
code output link to see it). The membership matrix (columns with name p*)
is inferred by
clue::cl_consensus()
function with the SE method. Basically the value in the membership matrix
represents the probability to belong to a certain group. The finall class
label for an item is determined with the group with highest probability it
belongs to.
In get_classes() function, the entropy is calculated from the membership
matrix and the silhouette score is calculated from the consensus matrix.
cbind(get_classes(res, k = 2), get_membership(res, k = 2))
#> class entropy silhouette p1 p2
#> GSM97138 1 0.0376 0.82719 0.996 0.004
#> GSM97145 1 0.2423 0.80884 0.960 0.040
#> GSM97147 2 0.7219 0.66637 0.200 0.800
#> GSM97125 1 0.0000 0.82823 1.000 0.000
#> GSM97127 1 0.1633 0.81770 0.976 0.024
#> GSM97130 1 0.0000 0.82823 1.000 0.000
#> GSM97133 1 0.0000 0.82823 1.000 0.000
#> GSM97134 1 0.8713 0.62042 0.708 0.292
#> GSM97120 1 0.9044 0.48274 0.680 0.320
#> GSM97126 1 0.9850 0.34449 0.572 0.428
#> GSM97112 1 0.0000 0.82823 1.000 0.000
#> GSM97115 1 0.9998 0.13424 0.508 0.492
#> GSM97116 1 0.0000 0.82823 1.000 0.000
#> GSM97117 2 0.0000 0.89973 0.000 1.000
#> GSM97119 1 0.0000 0.82823 1.000 0.000
#> GSM97122 1 0.0000 0.82823 1.000 0.000
#> GSM97135 1 0.0000 0.82823 1.000 0.000
#> GSM97136 2 0.7299 0.65587 0.204 0.796
#> GSM97139 1 0.0000 0.82823 1.000 0.000
#> GSM97146 1 0.0672 0.82535 0.992 0.008
#> GSM97123 2 0.0000 0.89973 0.000 1.000
#> GSM97129 2 0.9635 0.25777 0.388 0.612
#> GSM97143 1 0.8861 0.60526 0.696 0.304
#> GSM97113 2 0.0000 0.89973 0.000 1.000
#> GSM97056 1 0.0000 0.82823 1.000 0.000
#> GSM97124 1 0.0000 0.82823 1.000 0.000
#> GSM97132 1 0.5946 0.75304 0.856 0.144
#> GSM97144 1 0.0000 0.82823 1.000 0.000
#> GSM97149 1 0.8713 0.53002 0.708 0.292
#> GSM97068 2 0.1414 0.88352 0.020 0.980
#> GSM97071 2 0.9954 0.00951 0.460 0.540
#> GSM97086 2 0.0000 0.89973 0.000 1.000
#> GSM97103 2 0.0000 0.89973 0.000 1.000
#> GSM97057 2 0.0376 0.89706 0.004 0.996
#> GSM97060 2 0.0000 0.89973 0.000 1.000
#> GSM97075 2 0.0000 0.89973 0.000 1.000
#> GSM97098 2 0.0000 0.89973 0.000 1.000
#> GSM97099 2 0.0000 0.89973 0.000 1.000
#> GSM97101 2 0.0000 0.89973 0.000 1.000
#> GSM97105 2 0.0000 0.89973 0.000 1.000
#> GSM97106 2 0.0000 0.89973 0.000 1.000
#> GSM97121 2 0.0000 0.89973 0.000 1.000
#> GSM97128 1 0.9491 0.49497 0.632 0.368
#> GSM97131 2 0.0000 0.89973 0.000 1.000
#> GSM97137 1 0.0376 0.82729 0.996 0.004
#> GSM97118 1 0.8763 0.61533 0.704 0.296
#> GSM97114 2 0.0376 0.89706 0.004 0.996
#> GSM97142 1 0.0000 0.82823 1.000 0.000
#> GSM97140 2 0.0672 0.89413 0.008 0.992
#> GSM97141 2 0.0000 0.89973 0.000 1.000
#> GSM97055 1 1.0000 0.15082 0.504 0.496
#> GSM97090 1 0.9552 0.47464 0.624 0.376
#> GSM97091 1 0.0000 0.82823 1.000 0.000
#> GSM97148 1 0.0000 0.82823 1.000 0.000
#> GSM97063 1 0.0000 0.82823 1.000 0.000
#> GSM97053 1 0.0000 0.82823 1.000 0.000
#> GSM97066 2 0.0376 0.89713 0.004 0.996
#> GSM97079 2 0.0376 0.89705 0.004 0.996
#> GSM97083 1 0.8713 0.62042 0.708 0.292
#> GSM97084 2 0.9815 0.15139 0.420 0.580
#> GSM97094 1 0.9522 0.49801 0.628 0.372
#> GSM97096 2 0.0000 0.89973 0.000 1.000
#> GSM97097 2 0.0000 0.89973 0.000 1.000
#> GSM97107 2 1.0000 -0.13281 0.496 0.504
#> GSM97054 2 0.9427 0.33172 0.360 0.640
#> GSM97062 2 0.9996 -0.09865 0.488 0.512
#> GSM97069 2 0.0000 0.89973 0.000 1.000
#> GSM97070 2 0.0000 0.89973 0.000 1.000
#> GSM97073 2 0.0000 0.89973 0.000 1.000
#> GSM97076 2 0.9963 -0.00893 0.464 0.536
#> GSM97077 2 0.1184 0.88709 0.016 0.984
#> GSM97095 2 0.9833 0.13998 0.424 0.576
#> GSM97102 2 0.0000 0.89973 0.000 1.000
#> GSM97109 2 0.0000 0.89973 0.000 1.000
#> GSM97110 2 0.0000 0.89973 0.000 1.000
#> GSM97074 1 0.8955 0.59354 0.688 0.312
#> GSM97085 2 0.9996 -0.09865 0.488 0.512
#> GSM97059 2 0.0672 0.89431 0.008 0.992
#> GSM97072 2 0.0000 0.89973 0.000 1.000
#> GSM97078 1 0.9000 0.58765 0.684 0.316
#> GSM97067 2 0.0000 0.89973 0.000 1.000
#> GSM97087 2 0.0000 0.89973 0.000 1.000
#> GSM97111 2 0.0000 0.89973 0.000 1.000
#> GSM97064 2 0.0000 0.89973 0.000 1.000
#> GSM97065 2 0.0000 0.89973 0.000 1.000
#> GSM97081 2 0.0000 0.89973 0.000 1.000
#> GSM97082 2 0.0000 0.89973 0.000 1.000
#> GSM97088 2 0.9998 -0.11388 0.492 0.508
#> GSM97100 2 0.0000 0.89973 0.000 1.000
#> GSM97104 2 0.0000 0.89973 0.000 1.000
#> GSM97108 2 0.0000 0.89973 0.000 1.000
#> GSM97050 2 0.0376 0.89711 0.004 0.996
#> GSM97080 2 0.0000 0.89973 0.000 1.000
#> GSM97089 2 0.0000 0.89973 0.000 1.000
#> GSM97092 2 0.0000 0.89973 0.000 1.000
#> GSM97093 2 0.0000 0.89973 0.000 1.000
#> GSM97058 2 0.0000 0.89973 0.000 1.000
#> GSM97051 2 0.0000 0.89973 0.000 1.000
#> GSM97052 2 0.0000 0.89973 0.000 1.000
#> GSM97061 2 0.0000 0.89973 0.000 1.000
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM97138 1 0.0000 0.914 1.000 0.000 0.000
#> GSM97145 1 0.0000 0.914 1.000 0.000 0.000
#> GSM97147 2 0.0424 0.886 0.008 0.992 0.000
#> GSM97125 1 0.0000 0.914 1.000 0.000 0.000
#> GSM97127 1 0.0000 0.914 1.000 0.000 0.000
#> GSM97130 2 0.0424 0.885 0.008 0.992 0.000
#> GSM97133 1 0.1753 0.880 0.952 0.048 0.000
#> GSM97134 2 0.3482 0.804 0.128 0.872 0.000
#> GSM97120 1 0.4702 0.704 0.788 0.000 0.212
#> GSM97126 2 0.7104 0.378 0.032 0.608 0.360
#> GSM97112 1 0.0000 0.914 1.000 0.000 0.000
#> GSM97115 2 0.0237 0.886 0.004 0.996 0.000
#> GSM97116 1 0.0000 0.914 1.000 0.000 0.000
#> GSM97117 3 0.0424 0.940 0.000 0.008 0.992
#> GSM97119 1 0.0237 0.913 0.996 0.004 0.000
#> GSM97122 1 0.0424 0.912 0.992 0.008 0.000
#> GSM97135 1 0.0424 0.912 0.992 0.008 0.000
#> GSM97136 3 0.0983 0.930 0.016 0.004 0.980
#> GSM97139 1 0.0000 0.914 1.000 0.000 0.000
#> GSM97146 1 0.2878 0.831 0.904 0.096 0.000
#> GSM97123 3 0.0237 0.941 0.000 0.004 0.996
#> GSM97129 3 0.5486 0.751 0.024 0.196 0.780
#> GSM97143 3 0.6228 0.505 0.316 0.012 0.672
#> GSM97113 3 0.4235 0.800 0.000 0.176 0.824
#> GSM97056 2 0.4974 0.672 0.236 0.764 0.000
#> GSM97124 1 0.0592 0.911 0.988 0.012 0.000
#> GSM97132 1 0.7027 0.665 0.724 0.172 0.104
#> GSM97144 2 0.5760 0.511 0.328 0.672 0.000
#> GSM97149 2 0.6291 0.163 0.468 0.532 0.000
#> GSM97068 2 0.0661 0.888 0.004 0.988 0.008
#> GSM97071 2 0.0237 0.887 0.000 0.996 0.004
#> GSM97086 2 0.0592 0.888 0.000 0.988 0.012
#> GSM97103 3 0.0000 0.942 0.000 0.000 1.000
#> GSM97057 2 0.0747 0.887 0.000 0.984 0.016
#> GSM97060 3 0.0000 0.942 0.000 0.000 1.000
#> GSM97075 3 0.4796 0.749 0.000 0.220 0.780
#> GSM97098 3 0.0000 0.942 0.000 0.000 1.000
#> GSM97099 3 0.0237 0.941 0.000 0.004 0.996
#> GSM97101 3 0.1860 0.913 0.000 0.052 0.948
#> GSM97105 2 0.4605 0.740 0.000 0.796 0.204
#> GSM97106 3 0.0000 0.942 0.000 0.000 1.000
#> GSM97121 2 0.1529 0.879 0.000 0.960 0.040
#> GSM97128 2 0.4750 0.707 0.216 0.784 0.000
#> GSM97131 3 0.4842 0.740 0.000 0.224 0.776
#> GSM97137 2 0.0892 0.884 0.020 0.980 0.000
#> GSM97118 1 0.9311 0.232 0.468 0.168 0.364
#> GSM97114 3 0.4953 0.796 0.016 0.176 0.808
#> GSM97142 1 0.0424 0.912 0.992 0.008 0.000
#> GSM97140 2 0.0592 0.888 0.000 0.988 0.012
#> GSM97141 3 0.0424 0.940 0.000 0.008 0.992
#> GSM97055 1 0.7013 0.195 0.548 0.020 0.432
#> GSM97090 2 0.0424 0.885 0.008 0.992 0.000
#> GSM97091 1 0.0592 0.911 0.988 0.012 0.000
#> GSM97148 1 0.0424 0.911 0.992 0.008 0.000
#> GSM97063 1 0.0000 0.914 1.000 0.000 0.000
#> GSM97053 1 0.0000 0.914 1.000 0.000 0.000
#> GSM97066 3 0.0000 0.942 0.000 0.000 1.000
#> GSM97079 2 0.0892 0.887 0.000 0.980 0.020
#> GSM97083 2 0.3192 0.822 0.112 0.888 0.000
#> GSM97084 2 0.0000 0.886 0.000 1.000 0.000
#> GSM97094 2 0.6737 0.704 0.156 0.744 0.100
#> GSM97096 3 0.0000 0.942 0.000 0.000 1.000
#> GSM97097 3 0.0000 0.942 0.000 0.000 1.000
#> GSM97107 2 0.2860 0.842 0.004 0.912 0.084
#> GSM97054 2 0.0424 0.888 0.000 0.992 0.008
#> GSM97062 2 0.0237 0.886 0.004 0.996 0.000
#> GSM97069 3 0.0000 0.942 0.000 0.000 1.000
#> GSM97070 3 0.0237 0.941 0.000 0.004 0.996
#> GSM97073 3 0.0000 0.942 0.000 0.000 1.000
#> GSM97076 2 0.1015 0.885 0.008 0.980 0.012
#> GSM97077 2 0.0592 0.888 0.000 0.988 0.012
#> GSM97095 2 0.0000 0.886 0.000 1.000 0.000
#> GSM97102 3 0.0000 0.942 0.000 0.000 1.000
#> GSM97109 3 0.0237 0.940 0.004 0.000 0.996
#> GSM97110 3 0.2165 0.901 0.000 0.064 0.936
#> GSM97074 2 0.9284 0.360 0.192 0.512 0.296
#> GSM97085 3 0.0829 0.933 0.004 0.012 0.984
#> GSM97059 2 0.0592 0.888 0.000 0.988 0.012
#> GSM97072 3 0.0000 0.942 0.000 0.000 1.000
#> GSM97078 2 0.0424 0.885 0.008 0.992 0.000
#> GSM97067 3 0.0000 0.942 0.000 0.000 1.000
#> GSM97087 3 0.0000 0.942 0.000 0.000 1.000
#> GSM97111 3 0.0424 0.940 0.000 0.008 0.992
#> GSM97064 2 0.1031 0.886 0.000 0.976 0.024
#> GSM97065 2 0.6330 0.319 0.004 0.600 0.396
#> GSM97081 3 0.0000 0.942 0.000 0.000 1.000
#> GSM97082 3 0.0424 0.940 0.000 0.008 0.992
#> GSM97088 3 0.5722 0.631 0.004 0.292 0.704
#> GSM97100 2 0.0592 0.888 0.000 0.988 0.012
#> GSM97104 3 0.0000 0.942 0.000 0.000 1.000
#> GSM97108 3 0.4291 0.799 0.000 0.180 0.820
#> GSM97050 2 0.3941 0.788 0.000 0.844 0.156
#> GSM97080 3 0.0000 0.942 0.000 0.000 1.000
#> GSM97089 3 0.0000 0.942 0.000 0.000 1.000
#> GSM97092 3 0.0424 0.940 0.000 0.008 0.992
#> GSM97093 2 0.2165 0.866 0.000 0.936 0.064
#> GSM97058 2 0.1031 0.885 0.000 0.976 0.024
#> GSM97051 2 0.0892 0.887 0.000 0.980 0.020
#> GSM97052 3 0.0237 0.941 0.000 0.004 0.996
#> GSM97061 3 0.0000 0.942 0.000 0.000 1.000
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM97138 1 0.0188 0.8720 0.996 0.000 0.000 0.004
#> GSM97145 1 0.0188 0.8721 0.996 0.000 0.000 0.004
#> GSM97147 2 0.0921 0.8339 0.000 0.972 0.000 0.028
#> GSM97125 1 0.3266 0.7285 0.832 0.000 0.000 0.168
#> GSM97127 1 0.0707 0.8680 0.980 0.000 0.000 0.020
#> GSM97130 2 0.4277 0.5356 0.000 0.720 0.000 0.280
#> GSM97133 1 0.0469 0.8686 0.988 0.000 0.000 0.012
#> GSM97134 4 0.6615 0.3261 0.084 0.404 0.000 0.512
#> GSM97120 1 0.0376 0.8700 0.992 0.000 0.004 0.004
#> GSM97126 4 0.8774 0.4546 0.084 0.300 0.156 0.460
#> GSM97112 1 0.4888 0.3980 0.588 0.000 0.000 0.412
#> GSM97115 2 0.1389 0.8220 0.000 0.952 0.000 0.048
#> GSM97116 1 0.0336 0.8709 0.992 0.000 0.000 0.008
#> GSM97117 3 0.3877 0.8749 0.000 0.048 0.840 0.112
#> GSM97119 4 0.4661 0.3387 0.348 0.000 0.000 0.652
#> GSM97122 1 0.3688 0.7153 0.792 0.000 0.000 0.208
#> GSM97135 1 0.2704 0.8121 0.876 0.000 0.000 0.124
#> GSM97136 3 0.2408 0.8687 0.044 0.000 0.920 0.036
#> GSM97139 1 0.0000 0.8722 1.000 0.000 0.000 0.000
#> GSM97146 1 0.0000 0.8722 1.000 0.000 0.000 0.000
#> GSM97123 3 0.1635 0.9021 0.000 0.008 0.948 0.044
#> GSM97129 3 0.7081 0.2499 0.028 0.060 0.496 0.416
#> GSM97143 4 0.7307 0.4829 0.192 0.000 0.284 0.524
#> GSM97113 3 0.3335 0.8351 0.016 0.128 0.856 0.000
#> GSM97056 2 0.5387 0.2998 0.400 0.584 0.000 0.016
#> GSM97124 4 0.4916 0.2936 0.424 0.000 0.000 0.576
#> GSM97132 4 0.7514 0.5366 0.276 0.068 0.072 0.584
#> GSM97144 4 0.6851 0.5186 0.132 0.300 0.000 0.568
#> GSM97149 1 0.3219 0.6503 0.836 0.164 0.000 0.000
#> GSM97068 2 0.1302 0.8241 0.000 0.956 0.000 0.044
#> GSM97071 2 0.2053 0.8241 0.000 0.924 0.004 0.072
#> GSM97086 2 0.0000 0.8359 0.000 1.000 0.000 0.000
#> GSM97103 3 0.0376 0.9022 0.000 0.004 0.992 0.004
#> GSM97057 2 0.0188 0.8368 0.000 0.996 0.004 0.000
#> GSM97060 3 0.1118 0.8991 0.000 0.000 0.964 0.036
#> GSM97075 3 0.4773 0.8429 0.000 0.092 0.788 0.120
#> GSM97098 3 0.0000 0.9010 0.000 0.000 1.000 0.000
#> GSM97099 3 0.1256 0.9030 0.000 0.008 0.964 0.028
#> GSM97101 3 0.3934 0.8728 0.000 0.048 0.836 0.116
#> GSM97105 2 0.4055 0.7381 0.000 0.832 0.108 0.060
#> GSM97106 3 0.1022 0.8993 0.000 0.000 0.968 0.032
#> GSM97121 2 0.1936 0.8235 0.000 0.940 0.028 0.032
#> GSM97128 4 0.6584 0.4487 0.080 0.348 0.004 0.568
#> GSM97131 3 0.5250 0.7405 0.000 0.196 0.736 0.068
#> GSM97137 2 0.1489 0.8192 0.044 0.952 0.000 0.004
#> GSM97118 4 0.7574 0.5965 0.144 0.060 0.176 0.620
#> GSM97114 3 0.5850 0.8199 0.076 0.052 0.756 0.116
#> GSM97142 4 0.4605 0.3410 0.336 0.000 0.000 0.664
#> GSM97140 2 0.1118 0.8323 0.000 0.964 0.000 0.036
#> GSM97141 3 0.2036 0.8981 0.000 0.032 0.936 0.032
#> GSM97055 4 0.6099 0.4692 0.076 0.040 0.156 0.728
#> GSM97090 2 0.1389 0.8220 0.000 0.952 0.000 0.048
#> GSM97091 4 0.3688 0.5300 0.208 0.000 0.000 0.792
#> GSM97148 1 0.0000 0.8722 1.000 0.000 0.000 0.000
#> GSM97063 1 0.4585 0.5523 0.668 0.000 0.000 0.332
#> GSM97053 1 0.2216 0.8321 0.908 0.000 0.000 0.092
#> GSM97066 3 0.2973 0.8883 0.000 0.000 0.856 0.144
#> GSM97079 2 0.1302 0.8235 0.000 0.956 0.044 0.000
#> GSM97083 4 0.4595 0.6249 0.044 0.176 0.000 0.780
#> GSM97084 2 0.0000 0.8359 0.000 1.000 0.000 0.000
#> GSM97094 2 0.7631 0.2073 0.084 0.556 0.056 0.304
#> GSM97096 3 0.0000 0.9010 0.000 0.000 1.000 0.000
#> GSM97097 3 0.0779 0.9019 0.000 0.004 0.980 0.016
#> GSM97107 2 0.5866 0.3802 0.000 0.624 0.052 0.324
#> GSM97054 2 0.0000 0.8359 0.000 1.000 0.000 0.000
#> GSM97062 2 0.1389 0.8220 0.000 0.952 0.000 0.048
#> GSM97069 3 0.2589 0.8910 0.000 0.000 0.884 0.116
#> GSM97070 3 0.2676 0.8947 0.000 0.012 0.896 0.092
#> GSM97073 3 0.1792 0.9010 0.000 0.000 0.932 0.068
#> GSM97076 2 0.3455 0.7571 0.004 0.852 0.012 0.132
#> GSM97077 2 0.0817 0.8351 0.000 0.976 0.000 0.024
#> GSM97095 2 0.0592 0.8342 0.000 0.984 0.000 0.016
#> GSM97102 3 0.0817 0.8999 0.000 0.000 0.976 0.024
#> GSM97109 3 0.1975 0.8893 0.048 0.000 0.936 0.016
#> GSM97110 3 0.1743 0.8908 0.004 0.056 0.940 0.000
#> GSM97074 4 0.5400 0.6281 0.124 0.068 0.032 0.776
#> GSM97085 4 0.3024 0.5832 0.000 0.000 0.148 0.852
#> GSM97059 2 0.0707 0.8356 0.000 0.980 0.000 0.020
#> GSM97072 3 0.1022 0.8993 0.000 0.000 0.968 0.032
#> GSM97078 2 0.4999 -0.0966 0.000 0.508 0.000 0.492
#> GSM97067 3 0.0921 0.8988 0.000 0.000 0.972 0.028
#> GSM97087 3 0.0817 0.9010 0.000 0.000 0.976 0.024
#> GSM97111 3 0.3934 0.8728 0.000 0.048 0.836 0.116
#> GSM97064 2 0.1677 0.8292 0.000 0.948 0.012 0.040
#> GSM97065 2 0.7630 0.0475 0.036 0.484 0.388 0.092
#> GSM97081 3 0.2654 0.8978 0.000 0.004 0.888 0.108
#> GSM97082 3 0.4370 0.8680 0.000 0.044 0.800 0.156
#> GSM97088 4 0.5051 0.6302 0.000 0.132 0.100 0.768
#> GSM97100 2 0.1022 0.8327 0.000 0.968 0.000 0.032
#> GSM97104 3 0.1474 0.8943 0.000 0.000 0.948 0.052
#> GSM97108 3 0.4094 0.8698 0.000 0.056 0.828 0.116
#> GSM97050 2 0.3552 0.7547 0.000 0.848 0.128 0.024
#> GSM97080 3 0.2921 0.8894 0.000 0.000 0.860 0.140
#> GSM97089 3 0.0188 0.9018 0.000 0.000 0.996 0.004
#> GSM97092 3 0.4307 0.8708 0.000 0.048 0.808 0.144
#> GSM97093 2 0.2197 0.8076 0.000 0.916 0.080 0.004
#> GSM97058 2 0.2987 0.7733 0.000 0.880 0.016 0.104
#> GSM97051 2 0.3157 0.7530 0.000 0.852 0.004 0.144
#> GSM97052 3 0.3606 0.8880 0.000 0.024 0.844 0.132
#> GSM97061 3 0.1576 0.9026 0.000 0.004 0.948 0.048
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM97138 1 0.0290 0.8066 0.992 0.000 0.000 0.000 0.008
#> GSM97145 1 0.1341 0.7701 0.944 0.000 0.000 0.000 0.056
#> GSM97147 2 0.2803 0.8018 0.004 0.892 0.008 0.036 0.060
#> GSM97125 1 0.5355 0.3882 0.660 0.000 0.000 0.220 0.120
#> GSM97127 1 0.3366 0.5364 0.768 0.000 0.000 0.000 0.232
#> GSM97130 2 0.3796 0.4684 0.000 0.700 0.000 0.300 0.000
#> GSM97133 1 0.0000 0.8110 1.000 0.000 0.000 0.000 0.000
#> GSM97134 4 0.5198 0.6582 0.004 0.284 0.000 0.648 0.064
#> GSM97120 1 0.0000 0.8110 1.000 0.000 0.000 0.000 0.000
#> GSM97126 4 0.7057 0.5864 0.156 0.224 0.068 0.552 0.000
#> GSM97112 5 0.3495 0.7213 0.160 0.000 0.000 0.028 0.812
#> GSM97115 2 0.1341 0.8014 0.000 0.944 0.000 0.056 0.000
#> GSM97116 1 0.0000 0.8110 1.000 0.000 0.000 0.000 0.000
#> GSM97117 3 0.4664 0.7677 0.000 0.052 0.784 0.100 0.064
#> GSM97119 5 0.3734 0.6936 0.060 0.000 0.000 0.128 0.812
#> GSM97122 5 0.5229 0.5659 0.324 0.000 0.000 0.064 0.612
#> GSM97135 5 0.4576 0.4713 0.376 0.000 0.000 0.016 0.608
#> GSM97136 3 0.2515 0.7837 0.020 0.000 0.908 0.032 0.040
#> GSM97139 1 0.0000 0.8110 1.000 0.000 0.000 0.000 0.000
#> GSM97146 1 0.0000 0.8110 1.000 0.000 0.000 0.000 0.000
#> GSM97123 3 0.4462 0.7862 0.000 0.004 0.768 0.128 0.100
#> GSM97129 3 0.6761 0.2920 0.028 0.060 0.476 0.408 0.028
#> GSM97143 4 0.6379 0.5054 0.048 0.000 0.132 0.624 0.196
#> GSM97113 3 0.3848 0.7303 0.040 0.172 0.788 0.000 0.000
#> GSM97056 1 0.4977 -0.0381 0.500 0.472 0.000 0.028 0.000
#> GSM97124 5 0.5568 0.5237 0.096 0.000 0.000 0.308 0.596
#> GSM97132 4 0.6420 0.6149 0.052 0.076 0.020 0.648 0.204
#> GSM97144 4 0.5820 0.6780 0.012 0.168 0.000 0.648 0.172
#> GSM97149 1 0.0162 0.8073 0.996 0.004 0.000 0.000 0.000
#> GSM97068 2 0.1270 0.8035 0.000 0.948 0.000 0.052 0.000
#> GSM97071 2 0.3265 0.7959 0.000 0.860 0.012 0.088 0.040
#> GSM97086 2 0.0000 0.8210 0.000 1.000 0.000 0.000 0.000
#> GSM97103 3 0.0162 0.7989 0.000 0.000 0.996 0.000 0.004
#> GSM97057 2 0.0162 0.8219 0.000 0.996 0.004 0.000 0.000
#> GSM97060 3 0.5016 0.7358 0.000 0.000 0.704 0.176 0.120
#> GSM97075 3 0.5454 0.7450 0.000 0.104 0.728 0.104 0.064
#> GSM97098 3 0.0579 0.7975 0.000 0.000 0.984 0.008 0.008
#> GSM97099 3 0.2696 0.7948 0.000 0.012 0.896 0.040 0.052
#> GSM97101 3 0.4679 0.7682 0.000 0.056 0.784 0.096 0.064
#> GSM97105 2 0.4990 0.7160 0.000 0.764 0.092 0.080 0.064
#> GSM97106 3 0.3427 0.7879 0.000 0.000 0.836 0.108 0.056
#> GSM97121 2 0.3670 0.7760 0.000 0.848 0.044 0.044 0.064
#> GSM97128 4 0.5349 0.6993 0.004 0.204 0.000 0.676 0.116
#> GSM97131 3 0.6047 0.6780 0.000 0.184 0.664 0.088 0.064
#> GSM97137 2 0.1430 0.8016 0.052 0.944 0.000 0.004 0.000
#> GSM97118 4 0.5743 0.5896 0.004 0.040 0.048 0.652 0.256
#> GSM97114 3 0.6569 0.7251 0.108 0.052 0.680 0.096 0.064
#> GSM97142 5 0.4627 0.6741 0.080 0.000 0.000 0.188 0.732
#> GSM97140 2 0.2983 0.7939 0.000 0.880 0.012 0.048 0.060
#> GSM97141 3 0.3310 0.7877 0.000 0.036 0.868 0.040 0.056
#> GSM97055 5 0.6569 0.3240 0.024 0.048 0.056 0.272 0.600
#> GSM97090 2 0.1341 0.8014 0.000 0.944 0.000 0.056 0.000
#> GSM97091 5 0.3280 0.6425 0.012 0.000 0.000 0.176 0.812
#> GSM97148 1 0.0000 0.8110 1.000 0.000 0.000 0.000 0.000
#> GSM97063 5 0.3003 0.6964 0.188 0.000 0.000 0.000 0.812
#> GSM97053 1 0.4287 -0.1461 0.540 0.000 0.000 0.000 0.460
#> GSM97066 3 0.5828 0.7109 0.000 0.004 0.596 0.284 0.116
#> GSM97079 2 0.1341 0.8024 0.000 0.944 0.056 0.000 0.000
#> GSM97083 4 0.5329 0.7002 0.000 0.184 0.000 0.672 0.144
#> GSM97084 2 0.0404 0.8200 0.000 0.988 0.000 0.012 0.000
#> GSM97094 2 0.7575 -0.3046 0.004 0.384 0.044 0.364 0.204
#> GSM97096 3 0.0579 0.7975 0.000 0.000 0.984 0.008 0.008
#> GSM97097 3 0.0865 0.7993 0.000 0.000 0.972 0.024 0.004
#> GSM97107 2 0.5014 0.2288 0.000 0.592 0.040 0.368 0.000
#> GSM97054 2 0.0000 0.8210 0.000 1.000 0.000 0.000 0.000
#> GSM97062 2 0.1341 0.8014 0.000 0.944 0.000 0.056 0.000
#> GSM97069 3 0.5663 0.7169 0.000 0.004 0.628 0.252 0.116
#> GSM97070 3 0.5060 0.7712 0.000 0.020 0.720 0.192 0.068
#> GSM97073 3 0.3779 0.7790 0.000 0.000 0.804 0.144 0.052
#> GSM97076 2 0.3403 0.7087 0.008 0.820 0.012 0.160 0.000
#> GSM97077 2 0.1393 0.8200 0.000 0.956 0.008 0.024 0.012
#> GSM97095 2 0.0703 0.8171 0.000 0.976 0.000 0.024 0.000
#> GSM97102 3 0.1774 0.7995 0.000 0.000 0.932 0.052 0.016
#> GSM97109 3 0.2020 0.7828 0.100 0.000 0.900 0.000 0.000
#> GSM97110 3 0.2471 0.7683 0.000 0.136 0.864 0.000 0.000
#> GSM97074 4 0.4804 0.6242 0.000 0.044 0.016 0.720 0.220
#> GSM97085 4 0.4847 0.4221 0.000 0.000 0.080 0.704 0.216
#> GSM97059 2 0.2158 0.8106 0.000 0.920 0.008 0.020 0.052
#> GSM97072 3 0.3953 0.7645 0.000 0.000 0.792 0.148 0.060
#> GSM97078 4 0.4114 0.5515 0.000 0.376 0.000 0.624 0.000
#> GSM97067 3 0.4219 0.7510 0.000 0.000 0.772 0.156 0.072
#> GSM97087 3 0.3682 0.7810 0.000 0.000 0.820 0.108 0.072
#> GSM97111 3 0.4664 0.7677 0.000 0.052 0.784 0.100 0.064
#> GSM97064 2 0.1597 0.8179 0.000 0.940 0.012 0.048 0.000
#> GSM97065 2 0.7056 0.0504 0.084 0.480 0.368 0.060 0.008
#> GSM97081 3 0.3184 0.8019 0.000 0.000 0.852 0.100 0.048
#> GSM97082 3 0.7368 0.6805 0.000 0.056 0.472 0.288 0.184
#> GSM97088 4 0.4857 0.6474 0.000 0.100 0.068 0.772 0.060
#> GSM97100 2 0.3079 0.7899 0.000 0.876 0.016 0.044 0.064
#> GSM97104 3 0.5434 0.6941 0.000 0.000 0.648 0.232 0.120
#> GSM97108 3 0.4919 0.7630 0.000 0.068 0.768 0.100 0.064
#> GSM97050 2 0.2921 0.7563 0.000 0.856 0.124 0.020 0.000
#> GSM97080 3 0.6043 0.6802 0.000 0.004 0.560 0.308 0.128
#> GSM97089 3 0.2139 0.8032 0.000 0.000 0.916 0.032 0.052
#> GSM97092 3 0.6520 0.7398 0.000 0.056 0.612 0.204 0.128
#> GSM97093 2 0.1908 0.7890 0.000 0.908 0.092 0.000 0.000
#> GSM97058 2 0.4116 0.7490 0.000 0.816 0.032 0.096 0.056
#> GSM97051 2 0.4512 0.7180 0.000 0.776 0.020 0.140 0.064
#> GSM97052 3 0.6626 0.7401 0.000 0.032 0.572 0.228 0.168
#> GSM97061 3 0.4785 0.7824 0.000 0.004 0.740 0.140 0.116
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM97138 1 0.0146 0.8455 0.996 0.000 0.000 0.000 0.004 0.000
#> GSM97145 1 0.1267 0.8009 0.940 0.000 0.000 0.000 0.060 0.000
#> GSM97147 4 0.3126 0.7305 0.000 0.248 0.000 0.752 0.000 0.000
#> GSM97125 1 0.5510 0.3136 0.560 0.000 0.000 0.000 0.192 0.248
#> GSM97127 1 0.3464 0.4109 0.688 0.000 0.000 0.000 0.312 0.000
#> GSM97130 4 0.3728 0.3371 0.004 0.000 0.000 0.652 0.000 0.344
#> GSM97133 1 0.0000 0.8479 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM97134 6 0.3307 0.7782 0.000 0.000 0.000 0.148 0.044 0.808
#> GSM97120 1 0.0000 0.8479 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM97126 6 0.5402 0.6660 0.152 0.012 0.008 0.180 0.000 0.648
#> GSM97112 5 0.1245 0.8369 0.032 0.000 0.000 0.000 0.952 0.016
#> GSM97115 4 0.0458 0.7963 0.000 0.000 0.000 0.984 0.000 0.016
#> GSM97116 1 0.0000 0.8479 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM97117 2 0.3938 0.4500 0.000 0.660 0.324 0.016 0.000 0.000
#> GSM97119 5 0.1225 0.8306 0.012 0.000 0.000 0.000 0.952 0.036
#> GSM97122 5 0.2730 0.7953 0.152 0.000 0.000 0.000 0.836 0.012
#> GSM97135 5 0.2491 0.7864 0.164 0.000 0.000 0.000 0.836 0.000
#> GSM97136 3 0.5347 0.0755 0.020 0.320 0.596 0.000 0.012 0.052
#> GSM97139 1 0.0000 0.8479 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM97146 1 0.0000 0.8479 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM97123 2 0.3543 0.1640 0.000 0.756 0.224 0.000 0.004 0.016
#> GSM97129 2 0.6603 0.1309 0.012 0.400 0.164 0.028 0.000 0.396
#> GSM97143 6 0.3593 0.7180 0.016 0.000 0.020 0.000 0.180 0.784
#> GSM97113 3 0.6671 -0.0851 0.028 0.328 0.332 0.312 0.000 0.000
#> GSM97056 1 0.4648 -0.0198 0.496 0.000 0.000 0.464 0.000 0.040
#> GSM97124 5 0.2848 0.7358 0.008 0.000 0.000 0.000 0.816 0.176
#> GSM97132 6 0.3621 0.7590 0.024 0.000 0.000 0.036 0.132 0.808
#> GSM97144 6 0.3472 0.7839 0.000 0.000 0.000 0.092 0.100 0.808
#> GSM97149 1 0.0000 0.8479 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM97068 4 0.0458 0.7963 0.000 0.000 0.000 0.984 0.000 0.016
#> GSM97071 4 0.3824 0.7499 0.000 0.164 0.016 0.780 0.000 0.040
#> GSM97086 4 0.0000 0.8007 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM97103 3 0.3620 0.0647 0.000 0.352 0.648 0.000 0.000 0.000
#> GSM97057 4 0.0000 0.8007 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM97060 3 0.4494 0.1407 0.000 0.400 0.572 0.000 0.016 0.012
#> GSM97075 2 0.4403 0.4444 0.000 0.648 0.304 0.048 0.000 0.000
#> GSM97098 3 0.3547 0.1066 0.000 0.332 0.668 0.000 0.000 0.000
#> GSM97099 2 0.3989 0.2819 0.000 0.528 0.468 0.004 0.000 0.000
#> GSM97101 2 0.3922 0.4506 0.000 0.664 0.320 0.016 0.000 0.000
#> GSM97105 4 0.4285 0.6430 0.000 0.320 0.036 0.644 0.000 0.000
#> GSM97106 2 0.3854 -0.0470 0.000 0.536 0.464 0.000 0.000 0.000
#> GSM97121 4 0.3802 0.6658 0.000 0.312 0.012 0.676 0.000 0.000
#> GSM97128 6 0.3424 0.7872 0.000 0.008 0.000 0.076 0.092 0.824
#> GSM97131 2 0.5156 0.3862 0.000 0.600 0.272 0.128 0.000 0.000
#> GSM97137 4 0.0458 0.7965 0.016 0.000 0.000 0.984 0.000 0.000
#> GSM97118 6 0.3158 0.7460 0.000 0.000 0.004 0.020 0.164 0.812
#> GSM97114 2 0.5218 0.4163 0.088 0.616 0.280 0.016 0.000 0.000
#> GSM97142 5 0.2554 0.8238 0.048 0.000 0.000 0.000 0.876 0.076
#> GSM97140 4 0.3266 0.7145 0.000 0.272 0.000 0.728 0.000 0.000
#> GSM97141 2 0.3756 0.3864 0.000 0.600 0.400 0.000 0.000 0.000
#> GSM97055 5 0.5795 0.4885 0.016 0.232 0.012 0.008 0.620 0.112
#> GSM97090 4 0.0458 0.7963 0.000 0.000 0.000 0.984 0.000 0.016
#> GSM97091 5 0.1075 0.8223 0.000 0.000 0.000 0.000 0.952 0.048
#> GSM97148 1 0.0000 0.8479 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM97063 5 0.1075 0.8351 0.048 0.000 0.000 0.000 0.952 0.000
#> GSM97053 5 0.3684 0.4785 0.372 0.000 0.000 0.000 0.628 0.000
#> GSM97066 3 0.5961 0.2745 0.000 0.168 0.596 0.000 0.048 0.188
#> GSM97079 4 0.0508 0.7969 0.000 0.004 0.012 0.984 0.000 0.000
#> GSM97083 6 0.3094 0.7817 0.000 0.000 0.000 0.140 0.036 0.824
#> GSM97084 4 0.0000 0.8007 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM97094 6 0.7762 0.2610 0.000 0.160 0.032 0.284 0.128 0.396
#> GSM97096 3 0.3515 0.1125 0.000 0.324 0.676 0.000 0.000 0.000
#> GSM97097 3 0.3659 0.0362 0.000 0.364 0.636 0.000 0.000 0.000
#> GSM97107 4 0.5092 -0.0956 0.000 0.044 0.016 0.492 0.000 0.448
#> GSM97054 4 0.0000 0.8007 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM97062 4 0.0458 0.7963 0.000 0.000 0.000 0.984 0.000 0.016
#> GSM97069 3 0.6278 0.2575 0.000 0.216 0.544 0.000 0.048 0.192
#> GSM97070 3 0.5724 0.2759 0.000 0.144 0.628 0.000 0.048 0.180
#> GSM97073 3 0.5301 0.2709 0.000 0.120 0.676 0.000 0.044 0.160
#> GSM97076 4 0.2853 0.7149 0.004 0.012 0.012 0.856 0.000 0.116
#> GSM97077 4 0.2048 0.7841 0.000 0.120 0.000 0.880 0.000 0.000
#> GSM97095 4 0.0260 0.7990 0.000 0.000 0.000 0.992 0.000 0.008
#> GSM97102 3 0.3490 0.1372 0.000 0.268 0.724 0.000 0.000 0.008
#> GSM97109 3 0.5016 0.0504 0.092 0.324 0.584 0.000 0.000 0.000
#> GSM97110 3 0.5940 -0.0617 0.000 0.332 0.440 0.228 0.000 0.000
#> GSM97074 6 0.1148 0.7309 0.000 0.000 0.020 0.004 0.016 0.960
#> GSM97085 6 0.3962 0.5703 0.000 0.000 0.116 0.000 0.120 0.764
#> GSM97059 4 0.2854 0.7509 0.000 0.208 0.000 0.792 0.000 0.000
#> GSM97072 3 0.2669 0.3103 0.000 0.000 0.836 0.000 0.008 0.156
#> GSM97078 6 0.3198 0.7013 0.000 0.000 0.000 0.260 0.000 0.740
#> GSM97067 3 0.4354 0.3071 0.000 0.028 0.740 0.000 0.048 0.184
#> GSM97087 2 0.4388 0.0095 0.000 0.648 0.312 0.000 0.004 0.036
#> GSM97111 2 0.3852 0.4489 0.000 0.664 0.324 0.012 0.000 0.000
#> GSM97064 4 0.1531 0.7922 0.000 0.068 0.004 0.928 0.000 0.000
#> GSM97065 4 0.6675 0.2096 0.080 0.236 0.132 0.540 0.000 0.012
#> GSM97081 2 0.3996 0.2059 0.000 0.512 0.484 0.000 0.000 0.004
#> GSM97082 2 0.5162 -0.0591 0.000 0.612 0.312 0.004 0.040 0.032
#> GSM97088 6 0.1508 0.7419 0.000 0.020 0.004 0.016 0.012 0.948
#> GSM97100 4 0.3409 0.6872 0.000 0.300 0.000 0.700 0.000 0.000
#> GSM97104 3 0.3898 0.1949 0.000 0.296 0.684 0.000 0.000 0.020
#> GSM97108 2 0.4062 0.4513 0.000 0.660 0.316 0.024 0.000 0.000
#> GSM97050 4 0.2869 0.7303 0.000 0.020 0.148 0.832 0.000 0.000
#> GSM97080 3 0.6680 0.1835 0.000 0.360 0.400 0.000 0.048 0.192
#> GSM97089 3 0.4381 0.0291 0.000 0.456 0.524 0.000 0.004 0.016
#> GSM97092 2 0.1109 0.3123 0.000 0.964 0.012 0.004 0.004 0.016
#> GSM97093 4 0.1471 0.7835 0.000 0.004 0.064 0.932 0.000 0.000
#> GSM97058 4 0.3464 0.6813 0.000 0.312 0.000 0.688 0.000 0.000
#> GSM97051 4 0.3695 0.6239 0.000 0.376 0.000 0.624 0.000 0.000
#> GSM97052 2 0.3805 0.1073 0.000 0.728 0.248 0.000 0.008 0.016
#> GSM97061 2 0.3000 0.2412 0.000 0.824 0.156 0.000 0.004 0.016
Heatmaps for the consensus matrix. It visualizes the probability of two samples to be in a same group.
consensus_heatmap(res, k = 2)
consensus_heatmap(res, k = 3)
consensus_heatmap(res, k = 4)
consensus_heatmap(res, k = 5)
consensus_heatmap(res, k = 6)
Heatmaps for the membership of samples in all partitions to see how consistent they are:
membership_heatmap(res, k = 2)
membership_heatmap(res, k = 3)
membership_heatmap(res, k = 4)
membership_heatmap(res, k = 5)
membership_heatmap(res, k = 6)
As soon as we have had the classes for columns, we can look for signatures which are significantly different between classes which can be candidate marks for certain classes. Following are the heatmaps for signatures.
Signature heatmaps where rows are scaled:
get_signatures(res, k = 2)
get_signatures(res, k = 3)
get_signatures(res, k = 4)
get_signatures(res, k = 5)
get_signatures(res, k = 6)
Signature heatmaps where rows are not scaled:
get_signatures(res, k = 2, scale_rows = FALSE)
get_signatures(res, k = 3, scale_rows = FALSE)
get_signatures(res, k = 4, scale_rows = FALSE)
get_signatures(res, k = 5, scale_rows = FALSE)
get_signatures(res, k = 6, scale_rows = FALSE)
Compare the overlap of signatures from different k:
compare_signatures(res)
get_signature() returns a data frame invisibly. TO get the list of signatures, the function
call should be assigned to a variable explicitly. In following code, if plot argument is set
to FALSE, no heatmap is plotted while only the differential analysis is performed.
# code only for demonstration
tb = get_signature(res, k = ..., plot = FALSE)
An example of the output of tb is:
#> which_row fdr mean_1 mean_2 scaled_mean_1 scaled_mean_2 km
#> 1 38 0.042760348 8.373488 9.131774 -0.5533452 0.5164555 1
#> 2 40 0.018707592 7.106213 8.469186 -0.6173731 0.5762149 1
#> 3 55 0.019134737 10.221463 11.207825 -0.6159697 0.5749050 1
#> 4 59 0.006059896 5.921854 7.869574 -0.6899429 0.6439467 1
#> 5 60 0.018055526 8.928898 10.211722 -0.6204761 0.5791110 1
#> 6 98 0.009384629 15.714769 14.887706 0.6635654 -0.6193277 2
...
The columns in tb are:
which_row: row indices corresponding to the input matrix.fdr: FDR for the differential test. mean_x: The mean value in group x.scaled_mean_x: The mean value in group x after rows are scaled.km: Row groups if k-means clustering is applied to rows.UMAP plot which shows how samples are separated.
dimension_reduction(res, k = 2, method = "UMAP")
dimension_reduction(res, k = 3, method = "UMAP")
dimension_reduction(res, k = 4, method = "UMAP")
dimension_reduction(res, k = 5, method = "UMAP")
dimension_reduction(res, k = 6, method = "UMAP")
Following heatmap shows how subgroups are split when increasing k:
collect_classes(res)
Test correlation between subgroups and known annotations. If the known annotation is numeric, one-way ANOVA test is applied, and if the known annotation is discrete, chi-squared contingency table test is applied.
test_to_known_factors(res)
#> n disease.state(p) specimen(p) cell.type(p) other(p) k
#> MAD:pam 83 0.000194 0.697 4.28e-13 0.1590 2
#> MAD:pam 94 0.000647 0.278 1.32e-10 0.0607 3
#> MAD:pam 85 0.000311 0.506 1.05e-08 0.0238 4
#> MAD:pam 89 0.004272 0.663 1.34e-08 0.1969 5
#> MAD:pam 54 0.029447 0.576 6.07e-05 0.0453 6
If matrix rows can be associated to genes, consider to use functional_enrichment(res,
...) to perform function enrichment for the signature genes. See this vignette for more detailed explanations.
The object with results only for a single top-value method and a single partition method can be extracted as:
res = res_list["MAD", "mclust"]
# you can also extract it by
# res = res_list["MAD:mclust"]
A summary of res and all the functions that can be applied to it:
res
#> A 'ConsensusPartition' object with k = 2, 3, 4, 5, 6.
#> On a matrix with 21512 rows and 100 columns.
#> Top rows (1000, 2000, 3000, 4000, 5000) are extracted by 'MAD' method.
#> Subgroups are detected by 'mclust' method.
#> Performed in total 1250 partitions by row resampling.
#> Best k for subgroups seems to be 4.
#>
#> Following methods can be applied to this 'ConsensusPartition' object:
#> [1] "cola_report" "collect_classes" "collect_plots"
#> [4] "collect_stats" "colnames" "compare_signatures"
#> [7] "consensus_heatmap" "dimension_reduction" "functional_enrichment"
#> [10] "get_anno_col" "get_anno" "get_classes"
#> [13] "get_consensus" "get_matrix" "get_membership"
#> [16] "get_param" "get_signatures" "get_stats"
#> [19] "is_best_k" "is_stable_k" "membership_heatmap"
#> [22] "ncol" "nrow" "plot_ecdf"
#> [25] "rownames" "select_partition_number" "show"
#> [28] "suggest_best_k" "test_to_known_factors"
collect_plots() function collects all the plots made from res for all k (number of partitions)
into one single page to provide an easy and fast comparison between different k.
collect_plots(res)
The plots are:
k and the heatmap of
predicted classes for each k.k.k.k.All the plots in panels can be made by individual functions and they are plotted later in this section.
select_partition_number() produces several plots showing different
statistics for choosing “optimized” k. There are following statistics:
k;k, the area increased is defined as \(A_k - A_{k-1}\).The detailed explanations of these statistics can be found in the cola vignette.
Generally speaking, lower PAC score, higher mean silhouette score or higher
concordance corresponds to better partition. Rand index and Jaccard index
measure how similar the current partition is compared to partition with k-1.
If they are too similar, we won't accept k is better than k-1.
select_partition_number(res)
The numeric values for all these statistics can be obtained by get_stats().
get_stats(res)
#> k 1-PAC mean_silhouette concordance area_increased Rand Jaccard
#> 2 2 0.538 0.906 0.905 0.4401 0.496 0.496
#> 3 3 0.669 0.846 0.890 0.3900 0.894 0.785
#> 4 4 0.908 0.905 0.945 0.2215 0.754 0.442
#> 5 5 0.692 0.678 0.849 0.0152 0.864 0.550
#> 6 6 0.745 0.752 0.840 0.0664 0.925 0.690
suggest_best_k() suggests the best \(k\) based on these statistics. The rules are as follows:
suggest_best_k(res)
#> [1] 4
Following shows the table of the partitions (You need to click the show/hide
code output link to see it). The membership matrix (columns with name p*)
is inferred by
clue::cl_consensus()
function with the SE method. Basically the value in the membership matrix
represents the probability to belong to a certain group. The finall class
label for an item is determined with the group with highest probability it
belongs to.
In get_classes() function, the entropy is calculated from the membership
matrix and the silhouette score is calculated from the consensus matrix.
cbind(get_classes(res, k = 2), get_membership(res, k = 2))
#> class entropy silhouette p1 p2
#> GSM97138 1 0.000 0.976 1.000 0.000
#> GSM97145 1 0.000 0.976 1.000 0.000
#> GSM97147 1 0.204 0.966 0.968 0.032
#> GSM97125 1 0.000 0.976 1.000 0.000
#> GSM97127 1 0.000 0.976 1.000 0.000
#> GSM97130 1 0.000 0.976 1.000 0.000
#> GSM97133 1 0.000 0.976 1.000 0.000
#> GSM97134 1 0.000 0.976 1.000 0.000
#> GSM97120 1 0.000 0.976 1.000 0.000
#> GSM97126 1 0.000 0.976 1.000 0.000
#> GSM97112 1 0.000 0.976 1.000 0.000
#> GSM97115 1 0.163 0.969 0.976 0.024
#> GSM97116 1 0.000 0.976 1.000 0.000
#> GSM97117 2 0.767 0.892 0.224 0.776
#> GSM97119 1 0.000 0.976 1.000 0.000
#> GSM97122 1 0.000 0.976 1.000 0.000
#> GSM97135 1 0.000 0.976 1.000 0.000
#> GSM97136 2 0.767 0.892 0.224 0.776
#> GSM97139 1 0.000 0.976 1.000 0.000
#> GSM97146 1 0.000 0.976 1.000 0.000
#> GSM97123 2 0.714 0.890 0.196 0.804
#> GSM97129 2 0.952 0.689 0.372 0.628
#> GSM97143 1 0.000 0.976 1.000 0.000
#> GSM97113 2 0.767 0.892 0.224 0.776
#> GSM97056 1 0.000 0.976 1.000 0.000
#> GSM97124 1 0.000 0.976 1.000 0.000
#> GSM97132 1 0.000 0.976 1.000 0.000
#> GSM97144 1 0.000 0.976 1.000 0.000
#> GSM97149 1 0.000 0.976 1.000 0.000
#> GSM97068 1 0.242 0.962 0.960 0.040
#> GSM97071 1 0.242 0.962 0.960 0.040
#> GSM97086 1 0.278 0.954 0.952 0.048
#> GSM97103 2 0.753 0.892 0.216 0.784
#> GSM97057 2 0.767 0.892 0.224 0.776
#> GSM97060 2 0.615 0.877 0.152 0.848
#> GSM97075 2 0.767 0.892 0.224 0.776
#> GSM97098 2 0.706 0.889 0.192 0.808
#> GSM97099 2 0.767 0.892 0.224 0.776
#> GSM97101 2 0.767 0.892 0.224 0.776
#> GSM97105 2 0.767 0.892 0.224 0.776
#> GSM97106 2 0.689 0.887 0.184 0.816
#> GSM97121 2 0.767 0.892 0.224 0.776
#> GSM97128 1 0.242 0.962 0.960 0.040
#> GSM97131 2 0.767 0.892 0.224 0.776
#> GSM97137 1 0.000 0.976 1.000 0.000
#> GSM97118 1 0.000 0.976 1.000 0.000
#> GSM97114 2 0.767 0.892 0.224 0.776
#> GSM97142 1 0.000 0.976 1.000 0.000
#> GSM97140 2 0.814 0.864 0.252 0.748
#> GSM97141 2 0.767 0.892 0.224 0.776
#> GSM97055 1 0.000 0.976 1.000 0.000
#> GSM97090 1 0.141 0.970 0.980 0.020
#> GSM97091 1 0.000 0.976 1.000 0.000
#> GSM97148 1 0.000 0.976 1.000 0.000
#> GSM97063 1 0.000 0.976 1.000 0.000
#> GSM97053 1 0.000 0.976 1.000 0.000
#> GSM97066 2 0.000 0.795 0.000 1.000
#> GSM97079 1 0.295 0.949 0.948 0.052
#> GSM97083 1 0.000 0.976 1.000 0.000
#> GSM97084 1 0.242 0.962 0.960 0.040
#> GSM97094 1 0.242 0.962 0.960 0.040
#> GSM97096 2 0.563 0.869 0.132 0.868
#> GSM97097 1 0.662 0.754 0.828 0.172
#> GSM97107 1 0.242 0.962 0.960 0.040
#> GSM97054 1 0.242 0.962 0.960 0.040
#> GSM97062 1 0.242 0.962 0.960 0.040
#> GSM97069 2 0.000 0.795 0.000 1.000
#> GSM97070 2 0.000 0.795 0.000 1.000
#> GSM97073 2 0.000 0.795 0.000 1.000
#> GSM97076 1 0.242 0.962 0.960 0.040
#> GSM97077 2 0.992 0.521 0.448 0.552
#> GSM97095 1 0.204 0.966 0.968 0.032
#> GSM97102 2 0.000 0.795 0.000 1.000
#> GSM97109 2 0.767 0.892 0.224 0.776
#> GSM97110 2 0.767 0.892 0.224 0.776
#> GSM97074 1 0.224 0.964 0.964 0.036
#> GSM97085 1 0.260 0.958 0.956 0.044
#> GSM97059 1 0.242 0.962 0.960 0.040
#> GSM97072 2 0.343 0.831 0.064 0.936
#> GSM97078 1 0.242 0.962 0.960 0.040
#> GSM97067 2 0.000 0.795 0.000 1.000
#> GSM97087 2 0.000 0.795 0.000 1.000
#> GSM97111 2 0.767 0.892 0.224 0.776
#> GSM97064 2 0.767 0.892 0.224 0.776
#> GSM97065 2 0.767 0.892 0.224 0.776
#> GSM97081 2 0.605 0.876 0.148 0.852
#> GSM97082 2 0.000 0.795 0.000 1.000
#> GSM97088 1 0.242 0.962 0.960 0.040
#> GSM97100 2 0.998 0.443 0.476 0.524
#> GSM97104 2 0.000 0.795 0.000 1.000
#> GSM97108 2 0.767 0.892 0.224 0.776
#> GSM97050 2 0.767 0.892 0.224 0.776
#> GSM97080 2 0.000 0.795 0.000 1.000
#> GSM97089 2 0.689 0.887 0.184 0.816
#> GSM97092 2 0.574 0.871 0.136 0.864
#> GSM97093 2 0.775 0.888 0.228 0.772
#> GSM97058 2 0.767 0.892 0.224 0.776
#> GSM97051 2 0.983 0.580 0.424 0.576
#> GSM97052 2 0.574 0.871 0.136 0.864
#> GSM97061 2 0.697 0.888 0.188 0.812
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM97138 1 0.0000 0.907 1.000 0.000 0.000
#> GSM97145 1 0.0237 0.908 0.996 0.004 0.000
#> GSM97147 1 0.4324 0.872 0.860 0.112 0.028
#> GSM97125 1 0.0000 0.907 1.000 0.000 0.000
#> GSM97127 1 0.0237 0.908 0.996 0.004 0.000
#> GSM97130 1 0.4748 0.889 0.832 0.024 0.144
#> GSM97133 1 0.0237 0.908 0.996 0.004 0.000
#> GSM97134 1 0.5222 0.886 0.816 0.040 0.144
#> GSM97120 1 0.0000 0.907 1.000 0.000 0.000
#> GSM97126 1 0.1031 0.908 0.976 0.024 0.000
#> GSM97112 1 0.0237 0.907 0.996 0.000 0.004
#> GSM97115 1 0.5222 0.886 0.816 0.040 0.144
#> GSM97116 1 0.0000 0.907 1.000 0.000 0.000
#> GSM97117 2 0.0000 0.923 0.000 1.000 0.000
#> GSM97119 1 0.0237 0.907 0.996 0.000 0.004
#> GSM97122 1 0.0237 0.907 0.996 0.000 0.004
#> GSM97135 1 0.0237 0.907 0.996 0.000 0.004
#> GSM97136 2 0.2261 0.818 0.068 0.932 0.000
#> GSM97139 1 0.0000 0.907 1.000 0.000 0.000
#> GSM97146 1 0.0000 0.907 1.000 0.000 0.000
#> GSM97123 2 0.5363 0.436 0.000 0.724 0.276
#> GSM97129 2 0.0237 0.918 0.004 0.996 0.000
#> GSM97143 1 0.0237 0.908 0.996 0.004 0.000
#> GSM97113 2 0.0000 0.923 0.000 1.000 0.000
#> GSM97056 1 0.3213 0.901 0.900 0.008 0.092
#> GSM97124 1 0.0237 0.908 0.996 0.004 0.000
#> GSM97132 1 0.0661 0.909 0.988 0.004 0.008
#> GSM97144 1 0.4874 0.888 0.828 0.028 0.144
#> GSM97149 1 0.0237 0.908 0.996 0.004 0.000
#> GSM97068 1 0.6721 0.837 0.748 0.116 0.136
#> GSM97071 1 0.5730 0.876 0.796 0.060 0.144
#> GSM97086 1 0.6705 0.840 0.748 0.108 0.144
#> GSM97103 2 0.0000 0.923 0.000 1.000 0.000
#> GSM97057 2 0.0000 0.923 0.000 1.000 0.000
#> GSM97060 3 0.5835 0.758 0.000 0.340 0.660
#> GSM97075 2 0.0000 0.923 0.000 1.000 0.000
#> GSM97098 2 0.6260 -0.233 0.000 0.552 0.448
#> GSM97099 2 0.0000 0.923 0.000 1.000 0.000
#> GSM97101 2 0.0000 0.923 0.000 1.000 0.000
#> GSM97105 2 0.0000 0.923 0.000 1.000 0.000
#> GSM97106 2 0.5948 0.154 0.000 0.640 0.360
#> GSM97121 2 0.0000 0.923 0.000 1.000 0.000
#> GSM97128 1 0.5222 0.886 0.816 0.040 0.144
#> GSM97131 2 0.0000 0.923 0.000 1.000 0.000
#> GSM97137 1 0.2384 0.907 0.936 0.008 0.056
#> GSM97118 1 0.1482 0.909 0.968 0.020 0.012
#> GSM97114 2 0.0747 0.902 0.016 0.984 0.000
#> GSM97142 1 0.0237 0.907 0.996 0.000 0.004
#> GSM97140 2 0.0000 0.923 0.000 1.000 0.000
#> GSM97141 2 0.0000 0.923 0.000 1.000 0.000
#> GSM97055 1 0.0592 0.908 0.988 0.012 0.000
#> GSM97090 1 0.5222 0.886 0.816 0.040 0.144
#> GSM97091 1 0.0237 0.907 0.996 0.000 0.004
#> GSM97148 1 0.0000 0.907 1.000 0.000 0.000
#> GSM97063 1 0.0237 0.907 0.996 0.000 0.004
#> GSM97053 1 0.0237 0.908 0.996 0.004 0.000
#> GSM97066 3 0.3816 0.885 0.000 0.148 0.852
#> GSM97079 1 0.7661 0.772 0.684 0.172 0.144
#> GSM97083 1 0.5222 0.886 0.816 0.040 0.144
#> GSM97084 1 0.6087 0.865 0.780 0.076 0.144
#> GSM97094 1 0.5222 0.886 0.816 0.040 0.144
#> GSM97096 3 0.5882 0.749 0.000 0.348 0.652
#> GSM97097 1 0.9024 0.285 0.448 0.420 0.132
#> GSM97107 1 0.5222 0.886 0.816 0.040 0.144
#> GSM97054 1 0.6634 0.843 0.752 0.104 0.144
#> GSM97062 1 0.6486 0.850 0.760 0.096 0.144
#> GSM97069 3 0.3816 0.885 0.000 0.148 0.852
#> GSM97070 3 0.3816 0.885 0.000 0.148 0.852
#> GSM97073 3 0.3816 0.885 0.000 0.148 0.852
#> GSM97076 1 0.2959 0.880 0.900 0.100 0.000
#> GSM97077 2 0.0000 0.923 0.000 1.000 0.000
#> GSM97095 1 0.5435 0.882 0.808 0.048 0.144
#> GSM97102 3 0.3816 0.885 0.000 0.148 0.852
#> GSM97109 2 0.0000 0.923 0.000 1.000 0.000
#> GSM97110 2 0.0000 0.923 0.000 1.000 0.000
#> GSM97074 1 0.2743 0.903 0.928 0.052 0.020
#> GSM97085 1 0.3116 0.875 0.892 0.108 0.000
#> GSM97059 1 0.6462 0.843 0.764 0.120 0.116
#> GSM97072 3 0.4555 0.862 0.000 0.200 0.800
#> GSM97078 1 0.5222 0.886 0.816 0.040 0.144
#> GSM97067 3 0.3816 0.885 0.000 0.148 0.852
#> GSM97087 3 0.3816 0.885 0.000 0.148 0.852
#> GSM97111 2 0.0000 0.923 0.000 1.000 0.000
#> GSM97064 2 0.0000 0.923 0.000 1.000 0.000
#> GSM97065 2 0.0000 0.923 0.000 1.000 0.000
#> GSM97081 3 0.6026 0.702 0.000 0.376 0.624
#> GSM97082 3 0.3816 0.885 0.000 0.148 0.852
#> GSM97088 1 0.5222 0.886 0.816 0.040 0.144
#> GSM97100 2 0.1129 0.895 0.004 0.976 0.020
#> GSM97104 3 0.3816 0.885 0.000 0.148 0.852
#> GSM97108 2 0.0000 0.923 0.000 1.000 0.000
#> GSM97050 2 0.0000 0.923 0.000 1.000 0.000
#> GSM97080 3 0.3816 0.885 0.000 0.148 0.852
#> GSM97089 3 0.6252 0.530 0.000 0.444 0.556
#> GSM97092 3 0.5905 0.744 0.000 0.352 0.648
#> GSM97093 2 0.0000 0.923 0.000 1.000 0.000
#> GSM97058 2 0.0000 0.923 0.000 1.000 0.000
#> GSM97051 2 0.0892 0.899 0.000 0.980 0.020
#> GSM97052 3 0.5905 0.744 0.000 0.352 0.648
#> GSM97061 2 0.6192 -0.116 0.000 0.580 0.420
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM97138 1 0.0336 0.9675 0.992 0.000 0.000 0.008
#> GSM97145 1 0.0707 0.9654 0.980 0.000 0.000 0.020
#> GSM97147 2 0.3486 0.8518 0.044 0.864 0.000 0.092
#> GSM97125 1 0.0707 0.9692 0.980 0.000 0.000 0.020
#> GSM97127 1 0.0000 0.9667 1.000 0.000 0.000 0.000
#> GSM97130 1 0.4509 0.6099 0.708 0.004 0.000 0.288
#> GSM97133 1 0.0000 0.9667 1.000 0.000 0.000 0.000
#> GSM97134 4 0.1661 0.9302 0.052 0.004 0.000 0.944
#> GSM97120 1 0.0188 0.9661 0.996 0.000 0.000 0.004
#> GSM97126 2 0.5923 0.3562 0.376 0.580 0.000 0.044
#> GSM97112 1 0.1022 0.9682 0.968 0.000 0.000 0.032
#> GSM97115 4 0.1913 0.9345 0.040 0.020 0.000 0.940
#> GSM97116 1 0.0000 0.9667 1.000 0.000 0.000 0.000
#> GSM97117 2 0.0000 0.9299 0.000 1.000 0.000 0.000
#> GSM97119 1 0.0895 0.9688 0.976 0.004 0.000 0.020
#> GSM97122 1 0.0921 0.9690 0.972 0.000 0.000 0.028
#> GSM97135 1 0.0921 0.9690 0.972 0.000 0.000 0.028
#> GSM97136 2 0.1811 0.9018 0.028 0.948 0.004 0.020
#> GSM97139 1 0.0000 0.9667 1.000 0.000 0.000 0.000
#> GSM97146 1 0.0000 0.9667 1.000 0.000 0.000 0.000
#> GSM97123 3 0.1716 0.9359 0.000 0.064 0.936 0.000
#> GSM97129 2 0.0707 0.9256 0.000 0.980 0.000 0.020
#> GSM97143 1 0.0895 0.9688 0.976 0.004 0.000 0.020
#> GSM97113 2 0.0000 0.9299 0.000 1.000 0.000 0.000
#> GSM97056 1 0.0895 0.9688 0.976 0.004 0.000 0.020
#> GSM97124 1 0.0895 0.9688 0.976 0.004 0.000 0.020
#> GSM97132 1 0.1209 0.9627 0.964 0.004 0.000 0.032
#> GSM97144 4 0.1661 0.9302 0.052 0.004 0.000 0.944
#> GSM97149 1 0.0188 0.9661 0.996 0.000 0.000 0.004
#> GSM97068 4 0.5168 -0.0323 0.004 0.496 0.000 0.500
#> GSM97071 4 0.1724 0.9344 0.032 0.020 0.000 0.948
#> GSM97086 4 0.1022 0.9183 0.000 0.032 0.000 0.968
#> GSM97103 3 0.2342 0.9222 0.000 0.080 0.912 0.008
#> GSM97057 2 0.0336 0.9295 0.000 0.992 0.000 0.008
#> GSM97060 3 0.0817 0.9508 0.000 0.024 0.976 0.000
#> GSM97075 2 0.0000 0.9299 0.000 1.000 0.000 0.000
#> GSM97098 3 0.1637 0.9383 0.000 0.060 0.940 0.000
#> GSM97099 2 0.0000 0.9299 0.000 1.000 0.000 0.000
#> GSM97101 2 0.0000 0.9299 0.000 1.000 0.000 0.000
#> GSM97105 2 0.2739 0.8822 0.000 0.904 0.060 0.036
#> GSM97106 3 0.1557 0.9403 0.000 0.056 0.944 0.000
#> GSM97121 2 0.0336 0.9295 0.000 0.992 0.000 0.008
#> GSM97128 4 0.1305 0.9360 0.036 0.004 0.000 0.960
#> GSM97131 3 0.3176 0.9016 0.000 0.084 0.880 0.036
#> GSM97137 1 0.0895 0.9688 0.976 0.004 0.000 0.020
#> GSM97118 1 0.3751 0.7745 0.800 0.004 0.000 0.196
#> GSM97114 2 0.0188 0.9289 0.004 0.996 0.000 0.000
#> GSM97142 1 0.1022 0.9682 0.968 0.000 0.000 0.032
#> GSM97140 2 0.0336 0.9295 0.000 0.992 0.000 0.008
#> GSM97141 2 0.0000 0.9299 0.000 1.000 0.000 0.000
#> GSM97055 1 0.1398 0.9584 0.956 0.004 0.000 0.040
#> GSM97090 4 0.1489 0.9338 0.044 0.004 0.000 0.952
#> GSM97091 1 0.1022 0.9682 0.968 0.000 0.000 0.032
#> GSM97148 1 0.0000 0.9667 1.000 0.000 0.000 0.000
#> GSM97063 1 0.1022 0.9682 0.968 0.000 0.000 0.032
#> GSM97053 1 0.0895 0.9688 0.976 0.004 0.000 0.020
#> GSM97066 3 0.0000 0.9489 0.000 0.000 1.000 0.000
#> GSM97079 4 0.1557 0.9060 0.000 0.056 0.000 0.944
#> GSM97083 4 0.1661 0.9302 0.052 0.004 0.000 0.944
#> GSM97084 4 0.1022 0.9183 0.000 0.032 0.000 0.968
#> GSM97094 4 0.1305 0.9360 0.036 0.004 0.000 0.960
#> GSM97096 3 0.0707 0.9511 0.000 0.020 0.980 0.000
#> GSM97097 4 0.4875 0.7307 0.000 0.068 0.160 0.772
#> GSM97107 4 0.1398 0.9352 0.040 0.004 0.000 0.956
#> GSM97054 4 0.1022 0.9183 0.000 0.032 0.000 0.968
#> GSM97062 4 0.1022 0.9183 0.000 0.032 0.000 0.968
#> GSM97069 3 0.0000 0.9489 0.000 0.000 1.000 0.000
#> GSM97070 3 0.0000 0.9489 0.000 0.000 1.000 0.000
#> GSM97073 3 0.0000 0.9489 0.000 0.000 1.000 0.000
#> GSM97076 2 0.3764 0.8290 0.040 0.844 0.000 0.116
#> GSM97077 2 0.2011 0.8886 0.000 0.920 0.000 0.080
#> GSM97095 4 0.1929 0.9331 0.036 0.024 0.000 0.940
#> GSM97102 3 0.0000 0.9489 0.000 0.000 1.000 0.000
#> GSM97109 2 0.0000 0.9299 0.000 1.000 0.000 0.000
#> GSM97110 2 0.0000 0.9299 0.000 1.000 0.000 0.000
#> GSM97074 4 0.1492 0.9355 0.036 0.004 0.004 0.956
#> GSM97085 3 0.4232 0.7974 0.036 0.004 0.816 0.144
#> GSM97059 2 0.3659 0.8206 0.024 0.840 0.000 0.136
#> GSM97072 3 0.0592 0.9511 0.000 0.016 0.984 0.000
#> GSM97078 4 0.1305 0.9360 0.036 0.004 0.000 0.960
#> GSM97067 3 0.0000 0.9489 0.000 0.000 1.000 0.000
#> GSM97087 3 0.0000 0.9489 0.000 0.000 1.000 0.000
#> GSM97111 2 0.0000 0.9299 0.000 1.000 0.000 0.000
#> GSM97064 3 0.3450 0.8507 0.000 0.156 0.836 0.008
#> GSM97065 2 0.0000 0.9299 0.000 1.000 0.000 0.000
#> GSM97081 3 0.0817 0.9506 0.000 0.024 0.976 0.000
#> GSM97082 3 0.0000 0.9489 0.000 0.000 1.000 0.000
#> GSM97088 4 0.1305 0.9360 0.036 0.004 0.000 0.960
#> GSM97100 2 0.2469 0.8663 0.000 0.892 0.000 0.108
#> GSM97104 3 0.0000 0.9489 0.000 0.000 1.000 0.000
#> GSM97108 2 0.0336 0.9295 0.000 0.992 0.000 0.008
#> GSM97050 2 0.1557 0.9056 0.000 0.944 0.000 0.056
#> GSM97080 3 0.0188 0.9497 0.000 0.004 0.996 0.000
#> GSM97089 3 0.3401 0.8495 0.000 0.152 0.840 0.008
#> GSM97092 3 0.0707 0.9511 0.000 0.020 0.980 0.000
#> GSM97093 2 0.0336 0.9295 0.000 0.992 0.000 0.008
#> GSM97058 2 0.5085 0.3477 0.000 0.616 0.376 0.008
#> GSM97051 3 0.4700 0.8134 0.000 0.084 0.792 0.124
#> GSM97052 3 0.0817 0.9508 0.000 0.024 0.976 0.000
#> GSM97061 3 0.1557 0.9403 0.000 0.056 0.944 0.000
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM97138 1 0.2984 0.9038 0.860 0.000 0.000 0.032 0.108
#> GSM97145 1 0.1768 0.9836 0.924 0.000 0.000 0.004 0.072
#> GSM97147 2 0.4009 0.4605 0.004 0.684 0.000 0.312 0.000
#> GSM97125 5 0.5670 0.2396 0.388 0.000 0.000 0.084 0.528
#> GSM97127 1 0.1608 0.9868 0.928 0.000 0.000 0.000 0.072
#> GSM97130 4 0.3395 0.5250 0.000 0.000 0.000 0.764 0.236
#> GSM97133 1 0.1608 0.9868 0.928 0.000 0.000 0.000 0.072
#> GSM97134 4 0.0290 0.7890 0.000 0.000 0.000 0.992 0.008
#> GSM97120 1 0.1608 0.9868 0.928 0.000 0.000 0.000 0.072
#> GSM97126 2 0.5447 0.2267 0.008 0.560 0.000 0.384 0.048
#> GSM97112 5 0.1410 0.7274 0.000 0.000 0.000 0.060 0.940
#> GSM97115 4 0.0566 0.7922 0.004 0.012 0.000 0.984 0.000
#> GSM97116 1 0.2069 0.9730 0.912 0.000 0.000 0.012 0.076
#> GSM97117 2 0.0290 0.8185 0.000 0.992 0.000 0.000 0.008
#> GSM97119 5 0.3395 0.7102 0.000 0.000 0.000 0.236 0.764
#> GSM97122 5 0.1410 0.7274 0.000 0.000 0.000 0.060 0.940
#> GSM97135 5 0.1410 0.7274 0.000 0.000 0.000 0.060 0.940
#> GSM97136 2 0.4141 0.6043 0.000 0.736 0.028 0.236 0.000
#> GSM97139 1 0.1608 0.9868 0.928 0.000 0.000 0.000 0.072
#> GSM97146 1 0.1608 0.9868 0.928 0.000 0.000 0.000 0.072
#> GSM97123 2 0.4227 0.0801 0.000 0.580 0.420 0.000 0.000
#> GSM97129 2 0.1908 0.7561 0.000 0.908 0.000 0.092 0.000
#> GSM97143 5 0.3949 0.6382 0.000 0.000 0.000 0.332 0.668
#> GSM97113 2 0.0290 0.8185 0.000 0.992 0.000 0.000 0.008
#> GSM97056 4 0.4583 0.3657 0.032 0.000 0.000 0.672 0.296
#> GSM97124 5 0.4182 0.5404 0.000 0.000 0.000 0.400 0.600
#> GSM97132 4 0.4045 0.2505 0.000 0.000 0.000 0.644 0.356
#> GSM97144 4 0.0000 0.7921 0.000 0.000 0.000 1.000 0.000
#> GSM97149 1 0.1608 0.9868 0.928 0.000 0.000 0.000 0.072
#> GSM97068 4 0.4298 0.4884 0.008 0.352 0.000 0.640 0.000
#> GSM97071 4 0.1082 0.7893 0.028 0.008 0.000 0.964 0.000
#> GSM97086 4 0.3561 0.7531 0.068 0.028 0.000 0.852 0.052
#> GSM97103 2 0.4937 0.0154 0.000 0.544 0.428 0.028 0.000
#> GSM97057 2 0.0162 0.8176 0.004 0.996 0.000 0.000 0.000
#> GSM97060 3 0.5535 0.5114 0.000 0.108 0.620 0.272 0.000
#> GSM97075 2 0.0290 0.8185 0.000 0.992 0.000 0.000 0.008
#> GSM97098 2 0.4242 0.0588 0.000 0.572 0.428 0.000 0.000
#> GSM97099 2 0.0290 0.8185 0.000 0.992 0.000 0.000 0.008
#> GSM97101 2 0.0290 0.8185 0.000 0.992 0.000 0.000 0.008
#> GSM97105 2 0.0451 0.8149 0.004 0.988 0.000 0.008 0.000
#> GSM97106 3 0.6417 0.4037 0.000 0.280 0.504 0.216 0.000
#> GSM97121 2 0.0162 0.8176 0.004 0.996 0.000 0.000 0.000
#> GSM97128 4 0.0000 0.7921 0.000 0.000 0.000 1.000 0.000
#> GSM97131 2 0.7079 0.0415 0.016 0.432 0.276 0.276 0.000
#> GSM97137 4 0.4583 0.3657 0.032 0.000 0.000 0.672 0.296
#> GSM97118 5 0.4268 0.4767 0.000 0.000 0.000 0.444 0.556
#> GSM97114 2 0.0290 0.8185 0.000 0.992 0.000 0.000 0.008
#> GSM97142 5 0.1410 0.7274 0.000 0.000 0.000 0.060 0.940
#> GSM97140 2 0.0162 0.8176 0.004 0.996 0.000 0.000 0.000
#> GSM97141 2 0.0290 0.8185 0.000 0.992 0.000 0.000 0.008
#> GSM97055 5 0.4074 0.6027 0.000 0.000 0.000 0.364 0.636
#> GSM97090 4 0.0162 0.7920 0.004 0.000 0.000 0.996 0.000
#> GSM97091 5 0.1410 0.7274 0.000 0.000 0.000 0.060 0.940
#> GSM97148 1 0.1608 0.9868 0.928 0.000 0.000 0.000 0.072
#> GSM97063 5 0.1410 0.7274 0.000 0.000 0.000 0.060 0.940
#> GSM97053 5 0.4300 0.3431 0.000 0.000 0.000 0.476 0.524
#> GSM97066 3 0.0000 0.8007 0.000 0.000 1.000 0.000 0.000
#> GSM97079 4 0.3918 0.7308 0.068 0.080 0.000 0.828 0.024
#> GSM97083 4 0.0000 0.7921 0.000 0.000 0.000 1.000 0.000
#> GSM97084 4 0.3383 0.7563 0.068 0.020 0.000 0.860 0.052
#> GSM97094 4 0.0000 0.7921 0.000 0.000 0.000 1.000 0.000
#> GSM97096 3 0.3932 0.5537 0.000 0.328 0.672 0.000 0.000
#> GSM97097 4 0.4955 0.6641 0.072 0.136 0.008 0.760 0.024
#> GSM97107 4 0.0000 0.7921 0.000 0.000 0.000 1.000 0.000
#> GSM97054 4 0.3474 0.7552 0.068 0.024 0.000 0.856 0.052
#> GSM97062 4 0.3474 0.7552 0.068 0.024 0.000 0.856 0.052
#> GSM97069 3 0.0000 0.8007 0.000 0.000 1.000 0.000 0.000
#> GSM97070 3 0.0000 0.8007 0.000 0.000 1.000 0.000 0.000
#> GSM97073 3 0.0000 0.8007 0.000 0.000 1.000 0.000 0.000
#> GSM97076 2 0.4482 0.3139 0.000 0.612 0.000 0.376 0.012
#> GSM97077 2 0.0162 0.8176 0.004 0.996 0.000 0.000 0.000
#> GSM97095 4 0.0703 0.7891 0.000 0.024 0.000 0.976 0.000
#> GSM97102 3 0.0000 0.8007 0.000 0.000 1.000 0.000 0.000
#> GSM97109 2 0.0290 0.8185 0.000 0.992 0.000 0.000 0.008
#> GSM97110 2 0.0290 0.8185 0.000 0.992 0.000 0.000 0.008
#> GSM97074 4 0.1043 0.7700 0.000 0.000 0.000 0.960 0.040
#> GSM97085 4 0.3816 0.4866 0.000 0.000 0.304 0.696 0.000
#> GSM97059 4 0.4420 0.2711 0.004 0.448 0.000 0.548 0.000
#> GSM97072 3 0.0880 0.7962 0.000 0.032 0.968 0.000 0.000
#> GSM97078 4 0.0000 0.7921 0.000 0.000 0.000 1.000 0.000
#> GSM97067 3 0.0000 0.8007 0.000 0.000 1.000 0.000 0.000
#> GSM97087 3 0.0000 0.8007 0.000 0.000 1.000 0.000 0.000
#> GSM97111 2 0.0290 0.8185 0.000 0.992 0.000 0.000 0.008
#> GSM97064 2 0.3928 0.4035 0.004 0.700 0.296 0.000 0.000
#> GSM97065 2 0.0290 0.8185 0.000 0.992 0.000 0.000 0.008
#> GSM97081 3 0.4242 0.3400 0.000 0.428 0.572 0.000 0.000
#> GSM97082 3 0.0000 0.8007 0.000 0.000 1.000 0.000 0.000
#> GSM97088 4 0.0000 0.7921 0.000 0.000 0.000 1.000 0.000
#> GSM97100 2 0.4206 0.5149 0.020 0.708 0.000 0.272 0.000
#> GSM97104 3 0.0000 0.8007 0.000 0.000 1.000 0.000 0.000
#> GSM97108 2 0.0162 0.8176 0.004 0.996 0.000 0.000 0.000
#> GSM97050 2 0.0162 0.8176 0.004 0.996 0.000 0.000 0.000
#> GSM97080 3 0.0290 0.8002 0.000 0.008 0.992 0.000 0.000
#> GSM97089 3 0.3421 0.7043 0.000 0.204 0.788 0.008 0.000
#> GSM97092 3 0.3534 0.6544 0.000 0.256 0.744 0.000 0.000
#> GSM97093 2 0.0324 0.8166 0.004 0.992 0.000 0.004 0.000
#> GSM97058 2 0.2583 0.6902 0.004 0.864 0.132 0.000 0.000
#> GSM97051 4 0.7683 0.2701 0.072 0.260 0.196 0.468 0.004
#> GSM97052 3 0.3730 0.6155 0.000 0.288 0.712 0.000 0.000
#> GSM97061 3 0.4306 0.1611 0.000 0.492 0.508 0.000 0.000
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM97138 1 0.1492 0.897 0.940 0.000 0.000 0.000 0.036 0.024
#> GSM97145 1 0.0146 0.947 0.996 0.000 0.000 0.000 0.000 0.004
#> GSM97147 2 0.4513 0.709 0.000 0.704 0.000 0.172 0.000 0.124
#> GSM97125 1 0.4687 0.386 0.632 0.000 0.000 0.000 0.296 0.072
#> GSM97127 1 0.0000 0.950 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM97130 6 0.2129 0.807 0.000 0.000 0.000 0.056 0.040 0.904
#> GSM97133 1 0.0000 0.950 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM97134 6 0.1204 0.817 0.000 0.000 0.000 0.056 0.000 0.944
#> GSM97120 1 0.0000 0.950 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM97126 2 0.5617 0.465 0.036 0.628 0.000 0.000 0.140 0.196
#> GSM97112 5 0.1007 0.827 0.000 0.000 0.000 0.000 0.956 0.044
#> GSM97115 6 0.2340 0.741 0.000 0.000 0.000 0.148 0.000 0.852
#> GSM97116 1 0.0000 0.950 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM97117 2 0.0363 0.784 0.000 0.988 0.000 0.012 0.000 0.000
#> GSM97119 5 0.2491 0.805 0.000 0.000 0.000 0.000 0.836 0.164
#> GSM97122 5 0.1007 0.827 0.000 0.000 0.000 0.000 0.956 0.044
#> GSM97135 5 0.1007 0.827 0.000 0.000 0.000 0.000 0.956 0.044
#> GSM97136 2 0.3620 0.637 0.000 0.772 0.044 0.000 0.000 0.184
#> GSM97139 1 0.0000 0.950 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM97146 1 0.0000 0.950 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM97123 2 0.6053 0.425 0.000 0.488 0.304 0.196 0.012 0.000
#> GSM97129 2 0.3552 0.744 0.000 0.800 0.000 0.084 0.000 0.116
#> GSM97143 5 0.2664 0.794 0.000 0.000 0.000 0.000 0.816 0.184
#> GSM97113 2 0.0000 0.787 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97056 6 0.4858 0.566 0.180 0.000 0.000 0.000 0.156 0.664
#> GSM97124 5 0.3952 0.629 0.020 0.000 0.000 0.000 0.672 0.308
#> GSM97132 6 0.3221 0.554 0.000 0.000 0.000 0.000 0.264 0.736
#> GSM97144 6 0.1204 0.817 0.000 0.000 0.000 0.056 0.000 0.944
#> GSM97149 1 0.0000 0.950 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM97068 4 0.5160 0.251 0.000 0.320 0.000 0.572 0.000 0.108
#> GSM97071 6 0.0865 0.812 0.000 0.000 0.000 0.036 0.000 0.964
#> GSM97086 4 0.3076 0.720 0.000 0.000 0.000 0.760 0.000 0.240
#> GSM97103 2 0.5908 0.506 0.000 0.520 0.256 0.216 0.000 0.008
#> GSM97057 2 0.3320 0.769 0.000 0.772 0.000 0.212 0.000 0.016
#> GSM97060 3 0.3414 0.841 0.000 0.004 0.844 0.036 0.044 0.072
#> GSM97075 2 0.0458 0.786 0.000 0.984 0.000 0.016 0.000 0.000
#> GSM97098 2 0.4546 0.505 0.000 0.660 0.288 0.040 0.012 0.000
#> GSM97099 2 0.0260 0.789 0.000 0.992 0.000 0.008 0.000 0.000
#> GSM97101 2 0.0458 0.789 0.000 0.984 0.000 0.016 0.000 0.000
#> GSM97105 2 0.3758 0.732 0.000 0.700 0.000 0.284 0.000 0.016
#> GSM97106 3 0.4367 0.771 0.000 0.044 0.752 0.160 0.044 0.000
#> GSM97121 2 0.2562 0.780 0.000 0.828 0.000 0.172 0.000 0.000
#> GSM97128 6 0.0146 0.813 0.000 0.000 0.000 0.004 0.000 0.996
#> GSM97131 4 0.3090 0.603 0.000 0.140 0.004 0.828 0.000 0.028
#> GSM97137 6 0.5224 0.488 0.228 0.000 0.000 0.000 0.164 0.608
#> GSM97118 5 0.3409 0.668 0.000 0.000 0.000 0.000 0.700 0.300
#> GSM97114 2 0.0000 0.787 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97142 5 0.1007 0.827 0.000 0.000 0.000 0.000 0.956 0.044
#> GSM97140 2 0.3323 0.760 0.000 0.752 0.000 0.240 0.000 0.008
#> GSM97141 2 0.0000 0.787 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97055 5 0.2664 0.794 0.000 0.000 0.000 0.000 0.816 0.184
#> GSM97090 6 0.1267 0.816 0.000 0.000 0.000 0.060 0.000 0.940
#> GSM97091 5 0.1007 0.827 0.000 0.000 0.000 0.000 0.956 0.044
#> GSM97148 1 0.0000 0.950 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM97063 5 0.1007 0.827 0.000 0.000 0.000 0.000 0.956 0.044
#> GSM97053 5 0.3614 0.752 0.028 0.000 0.000 0.000 0.752 0.220
#> GSM97066 3 0.0000 0.903 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM97079 4 0.3342 0.729 0.000 0.012 0.000 0.760 0.000 0.228
#> GSM97083 6 0.0146 0.813 0.000 0.000 0.000 0.004 0.000 0.996
#> GSM97084 4 0.3351 0.657 0.000 0.000 0.000 0.712 0.000 0.288
#> GSM97094 6 0.1204 0.817 0.000 0.000 0.000 0.056 0.000 0.944
#> GSM97096 3 0.2252 0.884 0.000 0.020 0.908 0.044 0.028 0.000
#> GSM97097 4 0.3287 0.729 0.000 0.012 0.000 0.768 0.000 0.220
#> GSM97107 6 0.1204 0.817 0.000 0.000 0.000 0.056 0.000 0.944
#> GSM97054 4 0.3221 0.700 0.000 0.000 0.000 0.736 0.000 0.264
#> GSM97062 4 0.3266 0.689 0.000 0.000 0.000 0.728 0.000 0.272
#> GSM97069 3 0.0000 0.903 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM97070 3 0.0000 0.903 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM97073 3 0.0000 0.903 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM97076 5 0.5965 0.338 0.000 0.352 0.000 0.004 0.448 0.196
#> GSM97077 2 0.3734 0.743 0.000 0.716 0.000 0.264 0.000 0.020
#> GSM97095 6 0.3247 0.704 0.000 0.036 0.000 0.156 0.000 0.808
#> GSM97102 3 0.0000 0.903 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM97109 2 0.0000 0.787 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97110 2 0.0146 0.786 0.000 0.996 0.000 0.004 0.000 0.000
#> GSM97074 6 0.4879 -0.192 0.000 0.000 0.048 0.004 0.448 0.500
#> GSM97085 3 0.3636 0.522 0.000 0.000 0.676 0.004 0.000 0.320
#> GSM97059 2 0.4871 0.653 0.000 0.652 0.000 0.224 0.000 0.124
#> GSM97072 3 0.1480 0.892 0.000 0.000 0.940 0.020 0.040 0.000
#> GSM97078 6 0.0146 0.813 0.000 0.000 0.000 0.004 0.000 0.996
#> GSM97067 3 0.0000 0.903 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM97087 3 0.0000 0.903 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM97111 2 0.0363 0.784 0.000 0.988 0.000 0.012 0.000 0.000
#> GSM97064 2 0.4389 0.706 0.000 0.660 0.052 0.288 0.000 0.000
#> GSM97065 2 0.0363 0.784 0.000 0.988 0.000 0.012 0.000 0.000
#> GSM97081 3 0.3748 0.737 0.000 0.212 0.756 0.020 0.012 0.000
#> GSM97082 3 0.0000 0.903 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM97088 6 0.2482 0.662 0.000 0.000 0.148 0.004 0.000 0.848
#> GSM97100 4 0.3744 0.577 0.000 0.184 0.000 0.764 0.000 0.052
#> GSM97104 3 0.0000 0.903 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM97108 2 0.3101 0.761 0.000 0.756 0.000 0.244 0.000 0.000
#> GSM97050 2 0.3608 0.743 0.000 0.716 0.000 0.272 0.000 0.012
#> GSM97080 3 0.0146 0.903 0.000 0.000 0.996 0.004 0.000 0.000
#> GSM97089 3 0.2030 0.882 0.000 0.036 0.924 0.016 0.008 0.016
#> GSM97092 3 0.2889 0.862 0.000 0.004 0.856 0.096 0.044 0.000
#> GSM97093 2 0.3348 0.769 0.000 0.768 0.000 0.216 0.000 0.016
#> GSM97058 2 0.3767 0.740 0.000 0.708 0.004 0.276 0.000 0.012
#> GSM97051 4 0.3006 0.687 0.000 0.092 0.000 0.844 0.000 0.064
#> GSM97052 3 0.2984 0.856 0.000 0.004 0.848 0.104 0.044 0.000
#> GSM97061 3 0.5815 0.538 0.000 0.208 0.608 0.140 0.044 0.000
Heatmaps for the consensus matrix. It visualizes the probability of two samples to be in a same group.
consensus_heatmap(res, k = 2)
consensus_heatmap(res, k = 3)
consensus_heatmap(res, k = 4)
consensus_heatmap(res, k = 5)
consensus_heatmap(res, k = 6)
Heatmaps for the membership of samples in all partitions to see how consistent they are:
membership_heatmap(res, k = 2)
membership_heatmap(res, k = 3)
membership_heatmap(res, k = 4)
membership_heatmap(res, k = 5)
membership_heatmap(res, k = 6)
As soon as we have had the classes for columns, we can look for signatures which are significantly different between classes which can be candidate marks for certain classes. Following are the heatmaps for signatures.
Signature heatmaps where rows are scaled:
get_signatures(res, k = 2)
get_signatures(res, k = 3)
get_signatures(res, k = 4)
get_signatures(res, k = 5)
get_signatures(res, k = 6)
Signature heatmaps where rows are not scaled:
get_signatures(res, k = 2, scale_rows = FALSE)
get_signatures(res, k = 3, scale_rows = FALSE)
get_signatures(res, k = 4, scale_rows = FALSE)
get_signatures(res, k = 5, scale_rows = FALSE)
get_signatures(res, k = 6, scale_rows = FALSE)
Compare the overlap of signatures from different k:
compare_signatures(res)
get_signature() returns a data frame invisibly. TO get the list of signatures, the function
call should be assigned to a variable explicitly. In following code, if plot argument is set
to FALSE, no heatmap is plotted while only the differential analysis is performed.
# code only for demonstration
tb = get_signature(res, k = ..., plot = FALSE)
An example of the output of tb is:
#> which_row fdr mean_1 mean_2 scaled_mean_1 scaled_mean_2 km
#> 1 38 0.042760348 8.373488 9.131774 -0.5533452 0.5164555 1
#> 2 40 0.018707592 7.106213 8.469186 -0.6173731 0.5762149 1
#> 3 55 0.019134737 10.221463 11.207825 -0.6159697 0.5749050 1
#> 4 59 0.006059896 5.921854 7.869574 -0.6899429 0.6439467 1
#> 5 60 0.018055526 8.928898 10.211722 -0.6204761 0.5791110 1
#> 6 98 0.009384629 15.714769 14.887706 0.6635654 -0.6193277 2
...
The columns in tb are:
which_row: row indices corresponding to the input matrix.fdr: FDR for the differential test. mean_x: The mean value in group x.scaled_mean_x: The mean value in group x after rows are scaled.km: Row groups if k-means clustering is applied to rows.UMAP plot which shows how samples are separated.
dimension_reduction(res, k = 2, method = "UMAP")
dimension_reduction(res, k = 3, method = "UMAP")
dimension_reduction(res, k = 4, method = "UMAP")
dimension_reduction(res, k = 5, method = "UMAP")
dimension_reduction(res, k = 6, method = "UMAP")
Following heatmap shows how subgroups are split when increasing k:
collect_classes(res)
Test correlation between subgroups and known annotations. If the known annotation is numeric, one-way ANOVA test is applied, and if the known annotation is discrete, chi-squared contingency table test is applied.
test_to_known_factors(res)
#> n disease.state(p) specimen(p) cell.type(p) other(p) k
#> MAD:mclust 99 2.75e-02 0.4512 1.16e-11 0.441 2
#> MAD:mclust 95 1.29e-03 0.0433 2.87e-10 0.161 3
#> MAD:mclust 97 1.01e-04 0.6347 2.67e-16 0.275 4
#> MAD:mclust 79 1.15e-04 0.2310 5.81e-12 0.150 5
#> MAD:mclust 93 3.34e-05 0.2292 3.99e-11 0.266 6
If matrix rows can be associated to genes, consider to use functional_enrichment(res,
...) to perform function enrichment for the signature genes. See this vignette for more detailed explanations.
The object with results only for a single top-value method and a single partition method can be extracted as:
res = res_list["MAD", "NMF"]
# you can also extract it by
# res = res_list["MAD:NMF"]
A summary of res and all the functions that can be applied to it:
res
#> A 'ConsensusPartition' object with k = 2, 3, 4, 5, 6.
#> On a matrix with 21512 rows and 100 columns.
#> Top rows (1000, 2000, 3000, 4000, 5000) are extracted by 'MAD' method.
#> Subgroups are detected by 'NMF' method.
#> Performed in total 1250 partitions by row resampling.
#> Best k for subgroups seems to be 2.
#>
#> Following methods can be applied to this 'ConsensusPartition' object:
#> [1] "cola_report" "collect_classes" "collect_plots"
#> [4] "collect_stats" "colnames" "compare_signatures"
#> [7] "consensus_heatmap" "dimension_reduction" "functional_enrichment"
#> [10] "get_anno_col" "get_anno" "get_classes"
#> [13] "get_consensus" "get_matrix" "get_membership"
#> [16] "get_param" "get_signatures" "get_stats"
#> [19] "is_best_k" "is_stable_k" "membership_heatmap"
#> [22] "ncol" "nrow" "plot_ecdf"
#> [25] "rownames" "select_partition_number" "show"
#> [28] "suggest_best_k" "test_to_known_factors"
collect_plots() function collects all the plots made from res for all k (number of partitions)
into one single page to provide an easy and fast comparison between different k.
collect_plots(res)
The plots are:
k and the heatmap of
predicted classes for each k.k.k.k.All the plots in panels can be made by individual functions and they are plotted later in this section.
select_partition_number() produces several plots showing different
statistics for choosing “optimized” k. There are following statistics:
k;k, the area increased is defined as \(A_k - A_{k-1}\).The detailed explanations of these statistics can be found in the cola vignette.
Generally speaking, lower PAC score, higher mean silhouette score or higher
concordance corresponds to better partition. Rand index and Jaccard index
measure how similar the current partition is compared to partition with k-1.
If they are too similar, we won't accept k is better than k-1.
select_partition_number(res)
The numeric values for all these statistics can be obtained by get_stats().
get_stats(res)
#> k 1-PAC mean_silhouette concordance area_increased Rand Jaccard
#> 2 2 0.917 0.944 0.975 0.4983 0.502 0.502
#> 3 3 0.449 0.504 0.714 0.3127 0.727 0.510
#> 4 4 0.564 0.536 0.742 0.1290 0.671 0.300
#> 5 5 0.538 0.399 0.675 0.0745 0.836 0.484
#> 6 6 0.595 0.458 0.646 0.0467 0.844 0.401
suggest_best_k() suggests the best \(k\) based on these statistics. The rules are as follows:
suggest_best_k(res)
#> [1] 2
Following shows the table of the partitions (You need to click the show/hide
code output link to see it). The membership matrix (columns with name p*)
is inferred by
clue::cl_consensus()
function with the SE method. Basically the value in the membership matrix
represents the probability to belong to a certain group. The finall class
label for an item is determined with the group with highest probability it
belongs to.
In get_classes() function, the entropy is calculated from the membership
matrix and the silhouette score is calculated from the consensus matrix.
cbind(get_classes(res, k = 2), get_membership(res, k = 2))
#> class entropy silhouette p1 p2
#> GSM97138 1 0.0000 0.978 1.000 0.000
#> GSM97145 1 0.0000 0.978 1.000 0.000
#> GSM97147 1 0.0000 0.978 1.000 0.000
#> GSM97125 1 0.0000 0.978 1.000 0.000
#> GSM97127 1 0.0000 0.978 1.000 0.000
#> GSM97130 1 0.0000 0.978 1.000 0.000
#> GSM97133 1 0.0000 0.978 1.000 0.000
#> GSM97134 1 0.0000 0.978 1.000 0.000
#> GSM97120 1 0.0000 0.978 1.000 0.000
#> GSM97126 1 0.0000 0.978 1.000 0.000
#> GSM97112 1 0.0000 0.978 1.000 0.000
#> GSM97115 1 0.0000 0.978 1.000 0.000
#> GSM97116 1 0.0000 0.978 1.000 0.000
#> GSM97117 2 0.0000 0.971 0.000 1.000
#> GSM97119 1 0.0000 0.978 1.000 0.000
#> GSM97122 1 0.0000 0.978 1.000 0.000
#> GSM97135 1 0.0000 0.978 1.000 0.000
#> GSM97136 2 0.5842 0.830 0.140 0.860
#> GSM97139 1 0.0000 0.978 1.000 0.000
#> GSM97146 1 0.0000 0.978 1.000 0.000
#> GSM97123 2 0.0000 0.971 0.000 1.000
#> GSM97129 2 0.9393 0.457 0.356 0.644
#> GSM97143 1 0.0000 0.978 1.000 0.000
#> GSM97113 2 0.2948 0.926 0.052 0.948
#> GSM97056 1 0.0000 0.978 1.000 0.000
#> GSM97124 1 0.0000 0.978 1.000 0.000
#> GSM97132 1 0.0000 0.978 1.000 0.000
#> GSM97144 1 0.0000 0.978 1.000 0.000
#> GSM97149 1 0.0000 0.978 1.000 0.000
#> GSM97068 1 0.7376 0.740 0.792 0.208
#> GSM97071 2 0.0000 0.971 0.000 1.000
#> GSM97086 2 0.0000 0.971 0.000 1.000
#> GSM97103 2 0.0000 0.971 0.000 1.000
#> GSM97057 2 0.8499 0.631 0.276 0.724
#> GSM97060 2 0.0000 0.971 0.000 1.000
#> GSM97075 2 0.0000 0.971 0.000 1.000
#> GSM97098 2 0.0000 0.971 0.000 1.000
#> GSM97099 2 0.0000 0.971 0.000 1.000
#> GSM97101 2 0.0000 0.971 0.000 1.000
#> GSM97105 2 0.0000 0.971 0.000 1.000
#> GSM97106 2 0.0000 0.971 0.000 1.000
#> GSM97121 2 0.0000 0.971 0.000 1.000
#> GSM97128 1 0.6801 0.783 0.820 0.180
#> GSM97131 2 0.0000 0.971 0.000 1.000
#> GSM97137 1 0.0000 0.978 1.000 0.000
#> GSM97118 1 0.0000 0.978 1.000 0.000
#> GSM97114 1 0.2043 0.951 0.968 0.032
#> GSM97142 1 0.0000 0.978 1.000 0.000
#> GSM97140 2 0.8713 0.600 0.292 0.708
#> GSM97141 2 0.0000 0.971 0.000 1.000
#> GSM97055 1 0.0000 0.978 1.000 0.000
#> GSM97090 1 0.0000 0.978 1.000 0.000
#> GSM97091 1 0.0000 0.978 1.000 0.000
#> GSM97148 1 0.0000 0.978 1.000 0.000
#> GSM97063 1 0.0000 0.978 1.000 0.000
#> GSM97053 1 0.0000 0.978 1.000 0.000
#> GSM97066 2 0.0000 0.971 0.000 1.000
#> GSM97079 2 0.0000 0.971 0.000 1.000
#> GSM97083 1 0.0000 0.978 1.000 0.000
#> GSM97084 2 0.0672 0.965 0.008 0.992
#> GSM97094 1 0.0000 0.978 1.000 0.000
#> GSM97096 2 0.0000 0.971 0.000 1.000
#> GSM97097 2 0.0000 0.971 0.000 1.000
#> GSM97107 1 0.0000 0.978 1.000 0.000
#> GSM97054 2 0.0376 0.968 0.004 0.996
#> GSM97062 2 0.0000 0.971 0.000 1.000
#> GSM97069 2 0.0000 0.971 0.000 1.000
#> GSM97070 2 0.0000 0.971 0.000 1.000
#> GSM97073 2 0.0000 0.971 0.000 1.000
#> GSM97076 1 0.0938 0.969 0.988 0.012
#> GSM97077 2 0.0000 0.971 0.000 1.000
#> GSM97095 1 0.0672 0.972 0.992 0.008
#> GSM97102 2 0.0000 0.971 0.000 1.000
#> GSM97109 2 0.9393 0.463 0.356 0.644
#> GSM97110 2 0.0000 0.971 0.000 1.000
#> GSM97074 1 0.7376 0.741 0.792 0.208
#> GSM97085 2 0.0000 0.971 0.000 1.000
#> GSM97059 1 0.0000 0.978 1.000 0.000
#> GSM97072 2 0.0000 0.971 0.000 1.000
#> GSM97078 1 0.8081 0.675 0.752 0.248
#> GSM97067 2 0.0000 0.971 0.000 1.000
#> GSM97087 2 0.0000 0.971 0.000 1.000
#> GSM97111 2 0.0000 0.971 0.000 1.000
#> GSM97064 2 0.0000 0.971 0.000 1.000
#> GSM97065 2 0.0000 0.971 0.000 1.000
#> GSM97081 2 0.0000 0.971 0.000 1.000
#> GSM97082 2 0.0000 0.971 0.000 1.000
#> GSM97088 2 0.0000 0.971 0.000 1.000
#> GSM97100 2 0.0000 0.971 0.000 1.000
#> GSM97104 2 0.0000 0.971 0.000 1.000
#> GSM97108 2 0.0000 0.971 0.000 1.000
#> GSM97050 2 0.0000 0.971 0.000 1.000
#> GSM97080 2 0.0000 0.971 0.000 1.000
#> GSM97089 2 0.0000 0.971 0.000 1.000
#> GSM97092 2 0.0000 0.971 0.000 1.000
#> GSM97093 2 0.3879 0.903 0.076 0.924
#> GSM97058 2 0.0000 0.971 0.000 1.000
#> GSM97051 2 0.0000 0.971 0.000 1.000
#> GSM97052 2 0.0000 0.971 0.000 1.000
#> GSM97061 2 0.0000 0.971 0.000 1.000
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM97138 1 0.2711 0.7651 0.912 0.088 0.000
#> GSM97145 1 0.4504 0.7433 0.804 0.196 0.000
#> GSM97147 2 0.6235 -0.1587 0.436 0.564 0.000
#> GSM97125 1 0.2537 0.7649 0.920 0.080 0.000
#> GSM97127 1 0.4974 0.7202 0.764 0.236 0.000
#> GSM97130 1 0.5678 0.6518 0.684 0.316 0.000
#> GSM97133 1 0.5733 0.6424 0.676 0.324 0.000
#> GSM97134 1 0.4413 0.7568 0.832 0.160 0.008
#> GSM97120 1 0.4605 0.7394 0.796 0.204 0.000
#> GSM97126 1 0.3116 0.7641 0.892 0.108 0.000
#> GSM97112 1 0.4974 0.6057 0.764 0.000 0.236
#> GSM97115 2 0.6215 -0.1321 0.428 0.572 0.000
#> GSM97116 1 0.3619 0.7608 0.864 0.136 0.000
#> GSM97117 2 0.6095 0.2212 0.000 0.608 0.392
#> GSM97119 1 0.2448 0.7213 0.924 0.000 0.076
#> GSM97122 1 0.2711 0.7154 0.912 0.000 0.088
#> GSM97135 1 0.2261 0.7247 0.932 0.000 0.068
#> GSM97136 3 0.5016 0.3989 0.240 0.000 0.760
#> GSM97139 1 0.4346 0.7478 0.816 0.184 0.000
#> GSM97146 1 0.4002 0.7552 0.840 0.160 0.000
#> GSM97123 2 0.5905 0.3234 0.000 0.648 0.352
#> GSM97129 2 0.8396 0.4837 0.196 0.624 0.180
#> GSM97143 1 0.3116 0.7037 0.892 0.000 0.108
#> GSM97113 2 0.0747 0.6233 0.016 0.984 0.000
#> GSM97056 1 0.5465 0.6804 0.712 0.288 0.000
#> GSM97124 1 0.1453 0.7578 0.968 0.024 0.008
#> GSM97132 1 0.1711 0.7594 0.960 0.032 0.008
#> GSM97144 1 0.5254 0.6991 0.736 0.264 0.000
#> GSM97149 1 0.5905 0.6047 0.648 0.352 0.000
#> GSM97068 2 0.5327 0.3161 0.272 0.728 0.000
#> GSM97071 3 0.6045 0.4353 0.000 0.380 0.620
#> GSM97086 2 0.1753 0.6218 0.000 0.952 0.048
#> GSM97103 2 0.6267 0.0303 0.000 0.548 0.452
#> GSM97057 2 0.4062 0.5158 0.164 0.836 0.000
#> GSM97060 3 0.6140 0.3969 0.000 0.404 0.596
#> GSM97075 2 0.6291 -0.0402 0.000 0.532 0.468
#> GSM97098 2 0.6299 -0.0711 0.000 0.524 0.476
#> GSM97099 2 0.5431 0.4536 0.000 0.716 0.284
#> GSM97101 2 0.1182 0.6251 0.012 0.976 0.012
#> GSM97105 2 0.3340 0.6005 0.000 0.880 0.120
#> GSM97106 2 0.6215 0.1168 0.000 0.572 0.428
#> GSM97121 2 0.1267 0.6227 0.024 0.972 0.004
#> GSM97128 3 0.5810 0.2262 0.336 0.000 0.664
#> GSM97131 2 0.4887 0.5247 0.000 0.772 0.228
#> GSM97137 1 0.5621 0.6605 0.692 0.308 0.000
#> GSM97118 1 0.4931 0.6087 0.768 0.000 0.232
#> GSM97114 2 0.6008 0.0513 0.372 0.628 0.000
#> GSM97142 1 0.4346 0.6509 0.816 0.000 0.184
#> GSM97140 2 0.4555 0.4601 0.200 0.800 0.000
#> GSM97141 2 0.1620 0.6243 0.024 0.964 0.012
#> GSM97055 1 0.6235 0.3140 0.564 0.000 0.436
#> GSM97090 1 0.6309 0.3030 0.500 0.500 0.000
#> GSM97091 1 0.6062 0.4105 0.616 0.000 0.384
#> GSM97148 1 0.5291 0.6975 0.732 0.268 0.000
#> GSM97063 1 0.5810 0.4867 0.664 0.000 0.336
#> GSM97053 1 0.1950 0.7606 0.952 0.040 0.008
#> GSM97066 3 0.2066 0.6290 0.000 0.060 0.940
#> GSM97079 2 0.4235 0.5682 0.000 0.824 0.176
#> GSM97083 1 0.5621 0.5239 0.692 0.000 0.308
#> GSM97084 2 0.1964 0.6081 0.056 0.944 0.000
#> GSM97094 1 0.4589 0.7429 0.820 0.172 0.008
#> GSM97096 3 0.6168 0.3808 0.000 0.412 0.588
#> GSM97097 2 0.5254 0.4837 0.000 0.736 0.264
#> GSM97107 1 0.6345 0.5049 0.596 0.400 0.004
#> GSM97054 2 0.1643 0.6136 0.044 0.956 0.000
#> GSM97062 2 0.2356 0.6172 0.000 0.928 0.072
#> GSM97069 3 0.3412 0.6453 0.000 0.124 0.876
#> GSM97070 3 0.4605 0.6276 0.000 0.204 0.796
#> GSM97073 3 0.4291 0.6360 0.000 0.180 0.820
#> GSM97076 1 0.5956 0.4995 0.672 0.004 0.324
#> GSM97077 2 0.3267 0.6035 0.000 0.884 0.116
#> GSM97095 2 0.6286 -0.2380 0.464 0.536 0.000
#> GSM97102 3 0.2959 0.6423 0.000 0.100 0.900
#> GSM97109 2 0.3816 0.5412 0.148 0.852 0.000
#> GSM97110 2 0.4555 0.5510 0.000 0.800 0.200
#> GSM97074 3 0.5560 0.2993 0.300 0.000 0.700
#> GSM97085 3 0.4750 0.4342 0.216 0.000 0.784
#> GSM97059 2 0.6062 0.0113 0.384 0.616 0.000
#> GSM97072 3 0.5926 0.4804 0.000 0.356 0.644
#> GSM97078 3 0.5905 0.1892 0.352 0.000 0.648
#> GSM97067 3 0.2356 0.6340 0.000 0.072 0.928
#> GSM97087 3 0.4931 0.6128 0.000 0.232 0.768
#> GSM97111 2 0.5529 0.4355 0.000 0.704 0.296
#> GSM97064 2 0.5591 0.4203 0.000 0.696 0.304
#> GSM97065 3 0.6307 0.1603 0.000 0.488 0.512
#> GSM97081 3 0.5882 0.4918 0.000 0.348 0.652
#> GSM97082 3 0.3619 0.6448 0.000 0.136 0.864
#> GSM97088 3 0.4796 0.4288 0.220 0.000 0.780
#> GSM97100 2 0.0892 0.6225 0.020 0.980 0.000
#> GSM97104 3 0.3192 0.6444 0.000 0.112 0.888
#> GSM97108 2 0.0983 0.6251 0.004 0.980 0.016
#> GSM97050 2 0.4002 0.5799 0.000 0.840 0.160
#> GSM97080 3 0.5098 0.6019 0.000 0.248 0.752
#> GSM97089 3 0.5098 0.6022 0.000 0.248 0.752
#> GSM97092 3 0.6225 0.3319 0.000 0.432 0.568
#> GSM97093 2 0.3644 0.5971 0.004 0.872 0.124
#> GSM97058 2 0.5178 0.4926 0.000 0.744 0.256
#> GSM97051 2 0.5058 0.5084 0.000 0.756 0.244
#> GSM97052 3 0.6267 0.2755 0.000 0.452 0.548
#> GSM97061 2 0.6168 0.1674 0.000 0.588 0.412
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM97138 1 0.3172 0.5204 0.840 0.000 0.000 0.160
#> GSM97145 1 0.0804 0.6021 0.980 0.000 0.012 0.008
#> GSM97147 1 0.3108 0.5436 0.872 0.016 0.112 0.000
#> GSM97125 1 0.3942 0.4203 0.764 0.000 0.000 0.236
#> GSM97127 1 0.0707 0.6033 0.980 0.000 0.000 0.020
#> GSM97130 2 0.5077 0.7146 0.160 0.760 0.000 0.080
#> GSM97133 1 0.0188 0.6024 0.996 0.000 0.004 0.000
#> GSM97134 2 0.5167 0.7117 0.108 0.760 0.000 0.132
#> GSM97120 1 0.0469 0.6000 0.988 0.000 0.012 0.000
#> GSM97126 1 0.2334 0.5797 0.908 0.000 0.004 0.088
#> GSM97112 4 0.4907 0.2843 0.420 0.000 0.000 0.580
#> GSM97115 2 0.1716 0.8478 0.064 0.936 0.000 0.000
#> GSM97116 1 0.3219 0.5148 0.836 0.000 0.000 0.164
#> GSM97117 3 0.1637 0.7336 0.060 0.000 0.940 0.000
#> GSM97119 4 0.5168 0.1093 0.496 0.004 0.000 0.500
#> GSM97122 4 0.5168 0.1068 0.496 0.004 0.000 0.500
#> GSM97135 1 0.5132 -0.0523 0.548 0.004 0.000 0.448
#> GSM97136 3 0.5755 0.5431 0.044 0.000 0.624 0.332
#> GSM97139 1 0.1637 0.5913 0.940 0.000 0.000 0.060
#> GSM97146 1 0.1389 0.5962 0.952 0.000 0.000 0.048
#> GSM97123 3 0.1892 0.7459 0.004 0.036 0.944 0.016
#> GSM97129 3 0.5423 0.4656 0.332 0.000 0.640 0.028
#> GSM97143 1 0.4985 -0.1063 0.532 0.000 0.000 0.468
#> GSM97113 3 0.4855 0.3355 0.400 0.000 0.600 0.000
#> GSM97056 1 0.7384 0.1623 0.476 0.352 0.000 0.172
#> GSM97124 1 0.5004 0.1037 0.604 0.004 0.000 0.392
#> GSM97132 1 0.7283 -0.1436 0.432 0.148 0.000 0.420
#> GSM97144 2 0.3691 0.8042 0.068 0.856 0.000 0.076
#> GSM97149 1 0.1637 0.5788 0.940 0.000 0.060 0.000
#> GSM97068 2 0.1388 0.8561 0.028 0.960 0.012 0.000
#> GSM97071 2 0.0592 0.8580 0.000 0.984 0.000 0.016
#> GSM97086 2 0.0188 0.8565 0.000 0.996 0.004 0.000
#> GSM97103 3 0.3810 0.7391 0.000 0.060 0.848 0.092
#> GSM97057 1 0.5581 0.0262 0.532 0.020 0.448 0.000
#> GSM97060 3 0.6102 0.6563 0.000 0.116 0.672 0.212
#> GSM97075 3 0.0779 0.7444 0.004 0.000 0.980 0.016
#> GSM97098 3 0.0817 0.7452 0.000 0.000 0.976 0.024
#> GSM97099 3 0.1211 0.7370 0.040 0.000 0.960 0.000
#> GSM97101 3 0.4776 0.3846 0.376 0.000 0.624 0.000
#> GSM97105 3 0.5384 0.6054 0.076 0.196 0.728 0.000
#> GSM97106 3 0.4782 0.7114 0.000 0.152 0.780 0.068
#> GSM97121 3 0.5110 0.4155 0.352 0.012 0.636 0.000
#> GSM97128 4 0.4182 0.4582 0.024 0.180 0.000 0.796
#> GSM97131 2 0.4955 0.0702 0.000 0.556 0.444 0.000
#> GSM97137 1 0.6313 0.3645 0.652 0.220 0.000 0.128
#> GSM97118 4 0.5200 0.4613 0.264 0.036 0.000 0.700
#> GSM97114 1 0.4382 0.3951 0.704 0.000 0.296 0.000
#> GSM97142 4 0.4933 0.2636 0.432 0.000 0.000 0.568
#> GSM97140 1 0.5594 -0.0141 0.520 0.020 0.460 0.000
#> GSM97141 3 0.4916 0.2790 0.424 0.000 0.576 0.000
#> GSM97055 4 0.3450 0.5274 0.156 0.000 0.008 0.836
#> GSM97090 2 0.1706 0.8525 0.036 0.948 0.000 0.016
#> GSM97091 4 0.3688 0.5139 0.208 0.000 0.000 0.792
#> GSM97148 1 0.0592 0.6032 0.984 0.000 0.000 0.016
#> GSM97063 4 0.4304 0.4656 0.284 0.000 0.000 0.716
#> GSM97053 1 0.5883 0.0698 0.572 0.040 0.000 0.388
#> GSM97066 4 0.4994 -0.3647 0.000 0.000 0.480 0.520
#> GSM97079 2 0.0000 0.8577 0.000 1.000 0.000 0.000
#> GSM97083 2 0.4678 0.6704 0.024 0.744 0.000 0.232
#> GSM97084 2 0.0000 0.8577 0.000 1.000 0.000 0.000
#> GSM97094 2 0.2060 0.8448 0.016 0.932 0.000 0.052
#> GSM97096 3 0.2921 0.7296 0.000 0.000 0.860 0.140
#> GSM97097 2 0.0524 0.8544 0.000 0.988 0.008 0.004
#> GSM97107 2 0.0707 0.8568 0.000 0.980 0.000 0.020
#> GSM97054 2 0.0000 0.8577 0.000 1.000 0.000 0.000
#> GSM97062 2 0.0000 0.8577 0.000 1.000 0.000 0.000
#> GSM97069 3 0.4972 0.4349 0.000 0.000 0.544 0.456
#> GSM97070 3 0.4331 0.6516 0.000 0.000 0.712 0.288
#> GSM97073 3 0.4564 0.6156 0.000 0.000 0.672 0.328
#> GSM97076 4 0.5344 0.4571 0.300 0.000 0.032 0.668
#> GSM97077 3 0.6248 0.5423 0.104 0.252 0.644 0.000
#> GSM97095 2 0.3205 0.8141 0.104 0.872 0.000 0.024
#> GSM97102 3 0.4697 0.5861 0.000 0.000 0.644 0.356
#> GSM97109 1 0.5000 -0.1192 0.500 0.000 0.500 0.000
#> GSM97110 3 0.2704 0.7010 0.124 0.000 0.876 0.000
#> GSM97074 4 0.0804 0.5185 0.012 0.000 0.008 0.980
#> GSM97085 4 0.2408 0.4645 0.000 0.000 0.104 0.896
#> GSM97059 2 0.6412 0.3940 0.348 0.572 0.080 0.000
#> GSM97072 3 0.4542 0.6858 0.000 0.020 0.752 0.228
#> GSM97078 2 0.4053 0.6971 0.004 0.768 0.000 0.228
#> GSM97067 4 0.5000 -0.3969 0.000 0.000 0.496 0.504
#> GSM97087 3 0.4193 0.6669 0.000 0.000 0.732 0.268
#> GSM97111 3 0.1474 0.7346 0.052 0.000 0.948 0.000
#> GSM97064 3 0.1557 0.7405 0.000 0.056 0.944 0.000
#> GSM97065 3 0.1004 0.7419 0.024 0.000 0.972 0.004
#> GSM97081 3 0.2647 0.7350 0.000 0.000 0.880 0.120
#> GSM97082 3 0.4855 0.5280 0.000 0.000 0.600 0.400
#> GSM97088 4 0.1406 0.5126 0.000 0.024 0.016 0.960
#> GSM97100 2 0.3196 0.7629 0.008 0.856 0.136 0.000
#> GSM97104 3 0.4843 0.5335 0.000 0.000 0.604 0.396
#> GSM97108 3 0.4983 0.5328 0.272 0.024 0.704 0.000
#> GSM97050 3 0.5111 0.6291 0.056 0.204 0.740 0.000
#> GSM97080 3 0.4697 0.6424 0.000 0.008 0.696 0.296
#> GSM97089 3 0.4103 0.6757 0.000 0.000 0.744 0.256
#> GSM97092 3 0.4050 0.7273 0.000 0.036 0.820 0.144
#> GSM97093 3 0.3808 0.6547 0.176 0.012 0.812 0.000
#> GSM97058 3 0.3612 0.7142 0.044 0.100 0.856 0.000
#> GSM97051 2 0.2081 0.8029 0.000 0.916 0.084 0.000
#> GSM97052 3 0.3638 0.7352 0.000 0.032 0.848 0.120
#> GSM97061 3 0.2996 0.7434 0.000 0.064 0.892 0.044
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM97138 1 0.5014 -0.00901 0.536 0.032 0.000 0.000 0.432
#> GSM97145 1 0.5840 0.33310 0.604 0.164 0.000 0.000 0.232
#> GSM97147 1 0.5589 0.41138 0.688 0.224 0.020 0.028 0.040
#> GSM97125 1 0.5379 -0.10666 0.492 0.044 0.000 0.004 0.460
#> GSM97127 1 0.4481 0.38914 0.720 0.048 0.000 0.000 0.232
#> GSM97130 4 0.3273 0.74231 0.036 0.004 0.000 0.848 0.112
#> GSM97133 1 0.2676 0.50427 0.884 0.036 0.000 0.000 0.080
#> GSM97134 4 0.5023 0.55768 0.028 0.024 0.000 0.676 0.272
#> GSM97120 1 0.2795 0.49993 0.872 0.028 0.000 0.000 0.100
#> GSM97126 1 0.4934 0.18979 0.616 0.024 0.000 0.008 0.352
#> GSM97112 5 0.3849 0.56192 0.232 0.016 0.000 0.000 0.752
#> GSM97115 4 0.2624 0.77092 0.032 0.012 0.004 0.904 0.048
#> GSM97116 1 0.3607 0.32736 0.752 0.004 0.000 0.000 0.244
#> GSM97117 2 0.6825 0.35412 0.208 0.468 0.312 0.000 0.012
#> GSM97119 5 0.4805 0.54625 0.252 0.032 0.000 0.016 0.700
#> GSM97122 5 0.4491 0.53178 0.280 0.024 0.000 0.004 0.692
#> GSM97135 5 0.4636 0.49864 0.308 0.024 0.000 0.004 0.664
#> GSM97136 5 0.7863 -0.23803 0.064 0.296 0.316 0.000 0.324
#> GSM97139 1 0.3462 0.41934 0.792 0.012 0.000 0.000 0.196
#> GSM97146 1 0.2011 0.48853 0.908 0.004 0.000 0.000 0.088
#> GSM97123 3 0.4133 0.40716 0.012 0.232 0.744 0.012 0.000
#> GSM97129 2 0.7767 0.03083 0.344 0.408 0.132 0.000 0.116
#> GSM97143 5 0.4607 0.49196 0.320 0.020 0.000 0.004 0.656
#> GSM97113 1 0.5957 0.03223 0.572 0.148 0.280 0.000 0.000
#> GSM97056 1 0.7147 0.07952 0.492 0.020 0.008 0.252 0.228
#> GSM97124 5 0.5396 0.50040 0.284 0.036 0.000 0.032 0.648
#> GSM97132 5 0.6325 0.47808 0.204 0.024 0.000 0.168 0.604
#> GSM97144 4 0.3070 0.75153 0.012 0.016 0.000 0.860 0.112
#> GSM97149 1 0.0807 0.51224 0.976 0.012 0.000 0.000 0.012
#> GSM97068 4 0.3990 0.74343 0.096 0.032 0.040 0.828 0.004
#> GSM97071 4 0.1992 0.77763 0.000 0.044 0.000 0.924 0.032
#> GSM97086 4 0.1671 0.76699 0.000 0.076 0.000 0.924 0.000
#> GSM97103 2 0.4348 0.45567 0.000 0.788 0.068 0.128 0.016
#> GSM97057 1 0.5839 0.15103 0.568 0.056 0.352 0.024 0.000
#> GSM97060 3 0.4989 0.53377 0.000 0.176 0.736 0.032 0.056
#> GSM97075 3 0.4010 0.42287 0.032 0.208 0.760 0.000 0.000
#> GSM97098 2 0.4103 0.41103 0.012 0.748 0.228 0.012 0.000
#> GSM97099 2 0.5004 0.45458 0.076 0.696 0.224 0.004 0.000
#> GSM97101 1 0.6775 -0.27691 0.388 0.328 0.284 0.000 0.000
#> GSM97105 2 0.7293 0.28151 0.088 0.460 0.348 0.104 0.000
#> GSM97106 3 0.5448 0.26457 0.000 0.340 0.584 0.076 0.000
#> GSM97121 2 0.6605 0.45307 0.212 0.588 0.160 0.040 0.000
#> GSM97128 5 0.6603 0.25235 0.004 0.024 0.140 0.268 0.564
#> GSM97131 4 0.5996 0.13968 0.004 0.432 0.096 0.468 0.000
#> GSM97137 1 0.5969 0.25826 0.644 0.020 0.000 0.160 0.176
#> GSM97118 5 0.4392 0.56739 0.088 0.012 0.020 0.072 0.808
#> GSM97114 1 0.5257 0.22777 0.640 0.296 0.056 0.000 0.008
#> GSM97142 5 0.4286 0.54691 0.260 0.020 0.000 0.004 0.716
#> GSM97140 1 0.7145 -0.12176 0.380 0.256 0.348 0.016 0.000
#> GSM97141 1 0.6646 -0.24510 0.416 0.356 0.228 0.000 0.000
#> GSM97055 5 0.3677 0.53623 0.032 0.032 0.096 0.000 0.840
#> GSM97090 4 0.5021 0.71568 0.068 0.016 0.044 0.776 0.096
#> GSM97091 5 0.2581 0.56523 0.048 0.020 0.028 0.000 0.904
#> GSM97148 1 0.2166 0.49523 0.912 0.012 0.000 0.004 0.072
#> GSM97063 5 0.3173 0.57931 0.112 0.016 0.016 0.000 0.856
#> GSM97053 5 0.5403 0.42288 0.368 0.016 0.000 0.036 0.580
#> GSM97066 3 0.6009 0.41932 0.000 0.180 0.580 0.000 0.240
#> GSM97079 4 0.2597 0.75075 0.000 0.120 0.004 0.872 0.004
#> GSM97083 4 0.4756 0.57203 0.004 0.016 0.012 0.676 0.292
#> GSM97084 4 0.1341 0.77209 0.000 0.056 0.000 0.944 0.000
#> GSM97094 4 0.4649 0.66885 0.000 0.212 0.000 0.720 0.068
#> GSM97096 2 0.4478 0.09836 0.000 0.628 0.360 0.004 0.008
#> GSM97097 4 0.4843 0.39894 0.000 0.428 0.004 0.552 0.016
#> GSM97107 4 0.2900 0.75676 0.000 0.108 0.000 0.864 0.028
#> GSM97054 4 0.1988 0.76962 0.000 0.016 0.048 0.928 0.008
#> GSM97062 4 0.0880 0.77401 0.000 0.032 0.000 0.968 0.000
#> GSM97069 3 0.6106 0.41455 0.000 0.228 0.568 0.000 0.204
#> GSM97070 3 0.5490 0.45805 0.000 0.248 0.636 0.000 0.116
#> GSM97073 3 0.6316 0.25018 0.000 0.396 0.464 0.004 0.136
#> GSM97076 2 0.6524 -0.11059 0.056 0.488 0.036 0.012 0.408
#> GSM97077 3 0.7218 0.21136 0.124 0.096 0.560 0.216 0.004
#> GSM97095 4 0.5228 0.70096 0.100 0.020 0.020 0.752 0.108
#> GSM97102 2 0.6213 -0.20836 0.000 0.452 0.408 0.000 0.140
#> GSM97109 2 0.4753 0.49276 0.136 0.780 0.024 0.028 0.032
#> GSM97110 2 0.4752 0.50322 0.080 0.760 0.144 0.012 0.004
#> GSM97074 5 0.4911 0.44488 0.000 0.124 0.132 0.008 0.736
#> GSM97085 5 0.5932 0.03555 0.000 0.096 0.336 0.008 0.560
#> GSM97059 1 0.7507 0.02162 0.460 0.040 0.176 0.312 0.012
#> GSM97072 3 0.5439 0.20528 0.000 0.464 0.484 0.004 0.048
#> GSM97078 4 0.5581 0.55229 0.004 0.016 0.064 0.648 0.268
#> GSM97067 3 0.6171 0.40522 0.000 0.240 0.556 0.000 0.204
#> GSM97087 3 0.1357 0.56136 0.000 0.004 0.948 0.000 0.048
#> GSM97111 2 0.5789 0.44920 0.124 0.612 0.260 0.000 0.004
#> GSM97064 3 0.4346 0.47207 0.044 0.080 0.812 0.060 0.004
#> GSM97065 3 0.6243 -0.00465 0.124 0.432 0.440 0.000 0.004
#> GSM97081 3 0.4535 0.42284 0.004 0.288 0.684 0.000 0.024
#> GSM97082 3 0.4496 0.51791 0.000 0.092 0.752 0.000 0.156
#> GSM97088 5 0.6602 0.30406 0.000 0.024 0.240 0.176 0.560
#> GSM97100 4 0.6109 0.56574 0.040 0.164 0.144 0.652 0.000
#> GSM97104 3 0.6128 0.40964 0.000 0.252 0.560 0.000 0.188
#> GSM97108 2 0.6847 0.41576 0.208 0.520 0.248 0.024 0.000
#> GSM97050 3 0.6969 0.27200 0.124 0.116 0.604 0.152 0.004
#> GSM97080 3 0.4300 0.53486 0.000 0.132 0.772 0.000 0.096
#> GSM97089 3 0.1579 0.56374 0.000 0.032 0.944 0.000 0.024
#> GSM97092 3 0.2199 0.55051 0.000 0.060 0.916 0.016 0.008
#> GSM97093 3 0.5746 0.27663 0.228 0.108 0.648 0.016 0.000
#> GSM97058 3 0.5726 0.37875 0.064 0.128 0.704 0.104 0.000
#> GSM97051 4 0.5315 0.35591 0.004 0.036 0.396 0.560 0.004
#> GSM97052 3 0.2053 0.54618 0.000 0.048 0.924 0.024 0.004
#> GSM97061 3 0.3163 0.51047 0.012 0.092 0.864 0.032 0.000
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM97138 5 0.542 0.1477 0.428 0.068 0.012 0.000 0.488 0.004
#> GSM97145 5 0.686 0.2236 0.252 0.248 0.052 0.000 0.444 0.004
#> GSM97147 2 0.545 0.4790 0.160 0.672 0.012 0.028 0.128 0.000
#> GSM97125 5 0.487 0.5362 0.228 0.064 0.020 0.004 0.684 0.000
#> GSM97127 5 0.578 0.0907 0.420 0.096 0.024 0.000 0.460 0.000
#> GSM97130 4 0.321 0.7645 0.036 0.008 0.016 0.852 0.088 0.000
#> GSM97133 1 0.396 0.7006 0.748 0.040 0.008 0.000 0.204 0.000
#> GSM97134 4 0.523 0.1563 0.000 0.040 0.020 0.500 0.436 0.004
#> GSM97120 1 0.370 0.7544 0.784 0.044 0.008 0.000 0.164 0.000
#> GSM97126 5 0.488 0.5445 0.216 0.092 0.008 0.000 0.680 0.004
#> GSM97112 5 0.186 0.6759 0.044 0.004 0.008 0.000 0.928 0.016
#> GSM97115 4 0.384 0.7565 0.084 0.004 0.048 0.816 0.048 0.000
#> GSM97116 1 0.230 0.7766 0.856 0.000 0.000 0.000 0.144 0.000
#> GSM97117 2 0.324 0.5335 0.024 0.864 0.040 0.000 0.048 0.024
#> GSM97119 5 0.239 0.6782 0.064 0.028 0.000 0.012 0.896 0.000
#> GSM97122 5 0.234 0.6732 0.088 0.020 0.000 0.004 0.888 0.000
#> GSM97135 5 0.272 0.6598 0.128 0.016 0.000 0.004 0.852 0.000
#> GSM97136 5 0.768 0.0690 0.020 0.192 0.140 0.000 0.412 0.236
#> GSM97139 1 0.290 0.7373 0.800 0.004 0.000 0.000 0.196 0.000
#> GSM97146 1 0.207 0.7966 0.896 0.012 0.000 0.000 0.092 0.000
#> GSM97123 3 0.581 0.3799 0.004 0.384 0.452 0.000 0.000 0.160
#> GSM97129 2 0.635 0.2736 0.092 0.552 0.052 0.000 0.284 0.020
#> GSM97143 5 0.216 0.6790 0.076 0.008 0.008 0.000 0.904 0.004
#> GSM97113 1 0.537 0.4550 0.660 0.176 0.128 0.000 0.000 0.036
#> GSM97056 1 0.432 0.6952 0.772 0.008 0.020 0.116 0.084 0.000
#> GSM97124 5 0.419 0.6516 0.084 0.068 0.004 0.052 0.792 0.000
#> GSM97132 5 0.456 0.5962 0.072 0.008 0.008 0.172 0.736 0.004
#> GSM97144 4 0.267 0.7610 0.008 0.008 0.008 0.868 0.108 0.000
#> GSM97149 1 0.219 0.7925 0.904 0.032 0.004 0.000 0.060 0.000
#> GSM97068 4 0.527 0.6288 0.216 0.036 0.068 0.672 0.004 0.004
#> GSM97071 4 0.303 0.7641 0.000 0.016 0.032 0.872 0.020 0.060
#> GSM97086 4 0.222 0.7639 0.012 0.020 0.036 0.916 0.000 0.016
#> GSM97103 3 0.845 -0.2874 0.044 0.212 0.296 0.160 0.008 0.280
#> GSM97057 1 0.528 0.3863 0.620 0.192 0.184 0.004 0.000 0.000
#> GSM97060 6 0.491 0.1615 0.000 0.024 0.356 0.032 0.000 0.588
#> GSM97075 2 0.641 -0.0740 0.032 0.476 0.280 0.000 0.000 0.212
#> GSM97098 6 0.688 0.1963 0.036 0.296 0.304 0.004 0.000 0.360
#> GSM97099 2 0.717 -0.1838 0.056 0.360 0.248 0.004 0.004 0.328
#> GSM97101 2 0.428 0.4648 0.128 0.732 0.140 0.000 0.000 0.000
#> GSM97105 2 0.368 0.4196 0.012 0.796 0.160 0.020 0.000 0.012
#> GSM97106 3 0.627 0.2501 0.004 0.160 0.544 0.040 0.000 0.252
#> GSM97121 2 0.256 0.5365 0.036 0.904 0.024 0.008 0.016 0.012
#> GSM97128 5 0.716 0.2232 0.016 0.000 0.100 0.228 0.488 0.168
#> GSM97131 2 0.569 0.3416 0.020 0.600 0.092 0.276 0.004 0.008
#> GSM97137 1 0.388 0.7297 0.784 0.000 0.008 0.080 0.128 0.000
#> GSM97118 5 0.382 0.6229 0.016 0.000 0.028 0.060 0.824 0.072
#> GSM97114 2 0.438 0.4960 0.220 0.720 0.008 0.000 0.044 0.008
#> GSM97142 5 0.166 0.6786 0.052 0.008 0.000 0.000 0.932 0.008
#> GSM97140 2 0.477 0.3730 0.064 0.704 0.208 0.008 0.016 0.000
#> GSM97141 2 0.375 0.5144 0.104 0.804 0.076 0.000 0.016 0.000
#> GSM97055 5 0.502 0.3965 0.012 0.008 0.060 0.000 0.644 0.276
#> GSM97090 4 0.524 0.6848 0.100 0.012 0.152 0.700 0.036 0.000
#> GSM97091 5 0.340 0.6184 0.016 0.000 0.040 0.000 0.824 0.120
#> GSM97148 1 0.207 0.7963 0.904 0.024 0.000 0.000 0.072 0.000
#> GSM97063 5 0.288 0.6424 0.012 0.000 0.040 0.000 0.864 0.084
#> GSM97053 5 0.388 0.6163 0.200 0.012 0.004 0.024 0.760 0.000
#> GSM97066 6 0.388 0.4779 0.004 0.012 0.116 0.000 0.072 0.796
#> GSM97079 4 0.350 0.7198 0.004 0.008 0.100 0.824 0.000 0.064
#> GSM97083 4 0.550 0.5977 0.028 0.000 0.056 0.648 0.240 0.028
#> GSM97084 4 0.105 0.7671 0.000 0.004 0.020 0.964 0.000 0.012
#> GSM97094 4 0.550 0.6475 0.024 0.040 0.160 0.708 0.028 0.040
#> GSM97096 6 0.655 0.3548 0.028 0.188 0.280 0.012 0.000 0.492
#> GSM97097 4 0.736 0.3081 0.036 0.116 0.252 0.480 0.000 0.116
#> GSM97107 4 0.357 0.7459 0.016 0.020 0.072 0.848 0.028 0.016
#> GSM97054 4 0.316 0.7398 0.012 0.028 0.104 0.848 0.008 0.000
#> GSM97062 4 0.158 0.7683 0.000 0.008 0.036 0.940 0.000 0.016
#> GSM97069 6 0.326 0.5083 0.004 0.004 0.100 0.008 0.040 0.844
#> GSM97070 6 0.387 0.4893 0.008 0.068 0.116 0.004 0.004 0.800
#> GSM97073 6 0.400 0.5268 0.012 0.060 0.100 0.012 0.008 0.808
#> GSM97076 6 0.803 0.3550 0.036 0.104 0.192 0.052 0.128 0.488
#> GSM97077 3 0.666 0.1675 0.040 0.392 0.432 0.112 0.000 0.024
#> GSM97095 4 0.570 0.6922 0.088 0.056 0.112 0.700 0.040 0.004
#> GSM97102 6 0.545 0.4798 0.020 0.092 0.196 0.000 0.024 0.668
#> GSM97109 2 0.799 -0.0210 0.064 0.372 0.296 0.020 0.036 0.212
#> GSM97110 6 0.780 0.2385 0.080 0.224 0.272 0.028 0.008 0.388
#> GSM97074 6 0.512 0.1573 0.004 0.000 0.036 0.020 0.392 0.548
#> GSM97085 6 0.535 0.2395 0.004 0.000 0.100 0.000 0.372 0.524
#> GSM97059 2 0.785 0.0779 0.244 0.324 0.212 0.212 0.008 0.000
#> GSM97072 6 0.446 0.5067 0.012 0.052 0.148 0.028 0.000 0.760
#> GSM97078 4 0.628 0.5965 0.016 0.000 0.160 0.604 0.160 0.060
#> GSM97067 6 0.241 0.5286 0.000 0.004 0.068 0.008 0.024 0.896
#> GSM97087 3 0.536 0.3394 0.004 0.076 0.544 0.004 0.004 0.368
#> GSM97111 2 0.394 0.5127 0.020 0.816 0.064 0.000 0.024 0.076
#> GSM97064 3 0.627 0.5140 0.024 0.256 0.568 0.032 0.000 0.120
#> GSM97065 6 0.643 0.3279 0.140 0.232 0.060 0.004 0.004 0.560
#> GSM97081 2 0.616 -0.1190 0.008 0.452 0.252 0.000 0.000 0.288
#> GSM97082 6 0.569 0.1135 0.004 0.052 0.352 0.000 0.048 0.544
#> GSM97088 5 0.740 -0.0642 0.012 0.000 0.148 0.120 0.372 0.348
#> GSM97100 2 0.581 0.2559 0.016 0.552 0.128 0.300 0.004 0.000
#> GSM97104 6 0.418 0.4508 0.000 0.028 0.228 0.000 0.020 0.724
#> GSM97108 2 0.355 0.5279 0.040 0.852 0.056 0.016 0.024 0.012
#> GSM97050 3 0.699 0.4961 0.064 0.200 0.556 0.068 0.000 0.112
#> GSM97080 6 0.430 0.3044 0.000 0.032 0.284 0.000 0.008 0.676
#> GSM97089 3 0.572 0.2905 0.012 0.080 0.516 0.004 0.008 0.380
#> GSM97092 3 0.599 0.4876 0.004 0.180 0.512 0.008 0.000 0.296
#> GSM97093 3 0.695 0.4683 0.128 0.228 0.520 0.008 0.004 0.112
#> GSM97058 3 0.666 0.1840 0.028 0.412 0.424 0.080 0.000 0.056
#> GSM97051 3 0.683 0.1518 0.024 0.316 0.388 0.260 0.000 0.012
#> GSM97052 3 0.587 0.5173 0.004 0.192 0.528 0.004 0.000 0.272
#> GSM97061 3 0.580 0.5509 0.004 0.244 0.548 0.004 0.000 0.200
Heatmaps for the consensus matrix. It visualizes the probability of two samples to be in a same group.
consensus_heatmap(res, k = 2)
consensus_heatmap(res, k = 3)
consensus_heatmap(res, k = 4)
consensus_heatmap(res, k = 5)
consensus_heatmap(res, k = 6)
Heatmaps for the membership of samples in all partitions to see how consistent they are:
membership_heatmap(res, k = 2)
membership_heatmap(res, k = 3)
membership_heatmap(res, k = 4)
membership_heatmap(res, k = 5)
membership_heatmap(res, k = 6)
As soon as we have had the classes for columns, we can look for signatures which are significantly different between classes which can be candidate marks for certain classes. Following are the heatmaps for signatures.
Signature heatmaps where rows are scaled:
get_signatures(res, k = 2)
get_signatures(res, k = 3)
get_signatures(res, k = 4)
get_signatures(res, k = 5)
get_signatures(res, k = 6)
Signature heatmaps where rows are not scaled:
get_signatures(res, k = 2, scale_rows = FALSE)
get_signatures(res, k = 3, scale_rows = FALSE)
get_signatures(res, k = 4, scale_rows = FALSE)
get_signatures(res, k = 5, scale_rows = FALSE)
get_signatures(res, k = 6, scale_rows = FALSE)
Compare the overlap of signatures from different k:
compare_signatures(res)
get_signature() returns a data frame invisibly. TO get the list of signatures, the function
call should be assigned to a variable explicitly. In following code, if plot argument is set
to FALSE, no heatmap is plotted while only the differential analysis is performed.
# code only for demonstration
tb = get_signature(res, k = ..., plot = FALSE)
An example of the output of tb is:
#> which_row fdr mean_1 mean_2 scaled_mean_1 scaled_mean_2 km
#> 1 38 0.042760348 8.373488 9.131774 -0.5533452 0.5164555 1
#> 2 40 0.018707592 7.106213 8.469186 -0.6173731 0.5762149 1
#> 3 55 0.019134737 10.221463 11.207825 -0.6159697 0.5749050 1
#> 4 59 0.006059896 5.921854 7.869574 -0.6899429 0.6439467 1
#> 5 60 0.018055526 8.928898 10.211722 -0.6204761 0.5791110 1
#> 6 98 0.009384629 15.714769 14.887706 0.6635654 -0.6193277 2
...
The columns in tb are:
which_row: row indices corresponding to the input matrix.fdr: FDR for the differential test. mean_x: The mean value in group x.scaled_mean_x: The mean value in group x after rows are scaled.km: Row groups if k-means clustering is applied to rows.UMAP plot which shows how samples are separated.
dimension_reduction(res, k = 2, method = "UMAP")
dimension_reduction(res, k = 3, method = "UMAP")
dimension_reduction(res, k = 4, method = "UMAP")
dimension_reduction(res, k = 5, method = "UMAP")
dimension_reduction(res, k = 6, method = "UMAP")
Following heatmap shows how subgroups are split when increasing k:
collect_classes(res)
Test correlation between subgroups and known annotations. If the known annotation is numeric, one-way ANOVA test is applied, and if the known annotation is discrete, chi-squared contingency table test is applied.
test_to_known_factors(res)
#> n disease.state(p) specimen(p) cell.type(p) other(p) k
#> MAD:NMF 98 1.37e-04 0.5025 3.14e-14 0.0947 2
#> MAD:NMF 61 9.50e-06 0.0726 1.84e-10 0.0259 3
#> MAD:NMF 69 1.83e-05 0.5813 3.99e-10 0.3021 4
#> MAD:NMF 38 2.02e-01 0.6930 1.89e-08 0.3341 5
#> MAD:NMF 51 7.90e-02 0.8392 1.49e-11 0.3598 6
If matrix rows can be associated to genes, consider to use functional_enrichment(res,
...) to perform function enrichment for the signature genes. See this vignette for more detailed explanations.
The object with results only for a single top-value method and a single partition method can be extracted as:
res = res_list["ATC", "hclust"]
# you can also extract it by
# res = res_list["ATC:hclust"]
A summary of res and all the functions that can be applied to it:
res
#> A 'ConsensusPartition' object with k = 2, 3, 4, 5, 6.
#> On a matrix with 21512 rows and 100 columns.
#> Top rows (1000, 2000, 3000, 4000, 5000) are extracted by 'ATC' method.
#> Subgroups are detected by 'hclust' method.
#> Performed in total 1250 partitions by row resampling.
#> Best k for subgroups seems to be 2.
#>
#> Following methods can be applied to this 'ConsensusPartition' object:
#> [1] "cola_report" "collect_classes" "collect_plots"
#> [4] "collect_stats" "colnames" "compare_signatures"
#> [7] "consensus_heatmap" "dimension_reduction" "functional_enrichment"
#> [10] "get_anno_col" "get_anno" "get_classes"
#> [13] "get_consensus" "get_matrix" "get_membership"
#> [16] "get_param" "get_signatures" "get_stats"
#> [19] "is_best_k" "is_stable_k" "membership_heatmap"
#> [22] "ncol" "nrow" "plot_ecdf"
#> [25] "rownames" "select_partition_number" "show"
#> [28] "suggest_best_k" "test_to_known_factors"
collect_plots() function collects all the plots made from res for all k (number of partitions)
into one single page to provide an easy and fast comparison between different k.
collect_plots(res)
The plots are:
k and the heatmap of
predicted classes for each k.k.k.k.All the plots in panels can be made by individual functions and they are plotted later in this section.
select_partition_number() produces several plots showing different
statistics for choosing “optimized” k. There are following statistics:
k;k, the area increased is defined as \(A_k - A_{k-1}\).The detailed explanations of these statistics can be found in the cola vignette.
Generally speaking, lower PAC score, higher mean silhouette score or higher
concordance corresponds to better partition. Rand index and Jaccard index
measure how similar the current partition is compared to partition with k-1.
If they are too similar, we won't accept k is better than k-1.
select_partition_number(res)
The numeric values for all these statistics can be obtained by get_stats().
get_stats(res)
#> k 1-PAC mean_silhouette concordance area_increased Rand Jaccard
#> 2 2 0.344 0.688 0.802 0.4431 0.553 0.553
#> 3 3 0.369 0.598 0.723 0.2693 0.906 0.834
#> 4 4 0.530 0.668 0.803 0.2476 0.787 0.572
#> 5 5 0.593 0.682 0.815 0.0633 0.946 0.819
#> 6 6 0.620 0.517 0.729 0.0606 0.973 0.890
suggest_best_k() suggests the best \(k\) based on these statistics. The rules are as follows:
suggest_best_k(res)
#> [1] 2
Following shows the table of the partitions (You need to click the show/hide
code output link to see it). The membership matrix (columns with name p*)
is inferred by
clue::cl_consensus()
function with the SE method. Basically the value in the membership matrix
represents the probability to belong to a certain group. The finall class
label for an item is determined with the group with highest probability it
belongs to.
In get_classes() function, the entropy is calculated from the membership
matrix and the silhouette score is calculated from the consensus matrix.
cbind(get_classes(res, k = 2), get_membership(res, k = 2))
#> class entropy silhouette p1 p2
#> GSM97138 1 0.981 0.7956 0.580 0.420
#> GSM97145 2 0.388 0.6701 0.076 0.924
#> GSM97147 2 0.000 0.7081 0.000 1.000
#> GSM97125 2 0.929 -0.0501 0.344 0.656
#> GSM97127 2 0.388 0.6701 0.076 0.924
#> GSM97130 2 0.358 0.6795 0.068 0.932
#> GSM97133 2 0.388 0.6701 0.076 0.924
#> GSM97134 2 0.402 0.6693 0.080 0.920
#> GSM97120 2 0.388 0.6701 0.076 0.924
#> GSM97126 1 0.886 0.9574 0.696 0.304
#> GSM97112 1 0.886 0.9574 0.696 0.304
#> GSM97115 2 0.278 0.7033 0.048 0.952
#> GSM97116 2 0.939 -0.0939 0.356 0.644
#> GSM97117 2 0.242 0.6940 0.040 0.960
#> GSM97119 1 0.904 0.9565 0.680 0.320
#> GSM97122 1 0.909 0.9548 0.676 0.324
#> GSM97135 1 0.886 0.9574 0.696 0.304
#> GSM97136 1 0.981 0.8049 0.580 0.420
#> GSM97139 2 0.469 0.6483 0.100 0.900
#> GSM97146 2 0.605 0.5853 0.148 0.852
#> GSM97123 2 0.917 0.6700 0.332 0.668
#> GSM97129 2 0.311 0.6856 0.056 0.944
#> GSM97143 1 0.886 0.9574 0.696 0.304
#> GSM97113 2 0.775 0.6944 0.228 0.772
#> GSM97056 2 0.358 0.6783 0.068 0.932
#> GSM97124 2 0.443 0.6562 0.092 0.908
#> GSM97132 2 0.988 -0.4203 0.436 0.564
#> GSM97144 2 0.358 0.6795 0.068 0.932
#> GSM97149 2 0.469 0.6483 0.100 0.900
#> GSM97068 2 0.839 0.6835 0.268 0.732
#> GSM97071 2 0.775 0.3909 0.228 0.772
#> GSM97086 2 0.886 0.6696 0.304 0.696
#> GSM97103 2 0.844 0.6780 0.272 0.728
#> GSM97057 2 0.844 0.6822 0.272 0.728
#> GSM97060 2 0.939 0.6299 0.356 0.644
#> GSM97075 2 1.000 -0.6067 0.492 0.508
#> GSM97098 2 0.844 0.6780 0.272 0.728
#> GSM97099 2 0.722 0.6811 0.200 0.800
#> GSM97101 2 0.295 0.7161 0.052 0.948
#> GSM97105 2 0.886 0.6696 0.304 0.696
#> GSM97106 2 0.891 0.6688 0.308 0.692
#> GSM97121 2 0.295 0.7161 0.052 0.948
#> GSM97128 1 0.886 0.9574 0.696 0.304
#> GSM97131 2 0.886 0.6696 0.304 0.696
#> GSM97137 2 0.278 0.6905 0.048 0.952
#> GSM97118 1 0.886 0.9574 0.696 0.304
#> GSM97114 2 0.242 0.6940 0.040 0.960
#> GSM97142 1 0.886 0.9574 0.696 0.304
#> GSM97140 2 0.000 0.7081 0.000 1.000
#> GSM97141 2 0.373 0.7135 0.072 0.928
#> GSM97055 1 0.886 0.9574 0.696 0.304
#> GSM97090 2 0.260 0.6931 0.044 0.956
#> GSM97091 1 0.886 0.9574 0.696 0.304
#> GSM97148 2 0.469 0.6483 0.100 0.900
#> GSM97063 1 0.886 0.9574 0.696 0.304
#> GSM97053 2 0.541 0.6196 0.124 0.876
#> GSM97066 1 0.909 0.9563 0.676 0.324
#> GSM97079 2 0.886 0.6696 0.304 0.696
#> GSM97083 1 0.886 0.9574 0.696 0.304
#> GSM97084 2 0.886 0.6696 0.304 0.696
#> GSM97094 2 0.745 0.4429 0.212 0.788
#> GSM97096 2 1.000 -0.6161 0.496 0.504
#> GSM97097 2 0.886 0.6696 0.304 0.696
#> GSM97107 2 0.443 0.6559 0.092 0.908
#> GSM97054 2 0.886 0.6696 0.304 0.696
#> GSM97062 2 0.886 0.6696 0.304 0.696
#> GSM97069 1 0.909 0.9547 0.676 0.324
#> GSM97070 1 0.909 0.9563 0.676 0.324
#> GSM97073 1 0.909 0.9563 0.676 0.324
#> GSM97076 1 0.909 0.9563 0.676 0.324
#> GSM97077 2 0.430 0.7157 0.088 0.912
#> GSM97095 2 0.706 0.4878 0.192 0.808
#> GSM97102 1 0.913 0.9501 0.672 0.328
#> GSM97109 2 0.242 0.6940 0.040 0.960
#> GSM97110 2 0.775 0.6956 0.228 0.772
#> GSM97074 1 0.909 0.9563 0.676 0.324
#> GSM97085 1 0.886 0.9574 0.696 0.304
#> GSM97059 2 0.000 0.7081 0.000 1.000
#> GSM97072 1 0.925 0.9422 0.660 0.340
#> GSM97078 1 0.988 0.7799 0.564 0.436
#> GSM97067 1 0.909 0.9563 0.676 0.324
#> GSM97087 1 0.946 0.9077 0.636 0.364
#> GSM97111 2 0.242 0.6940 0.040 0.960
#> GSM97064 2 0.895 0.6744 0.312 0.688
#> GSM97065 1 0.909 0.9563 0.676 0.324
#> GSM97081 2 1.000 -0.6067 0.492 0.508
#> GSM97082 1 0.946 0.9077 0.636 0.364
#> GSM97088 1 0.886 0.9574 0.696 0.304
#> GSM97100 2 0.886 0.6696 0.304 0.696
#> GSM97104 1 0.913 0.9501 0.672 0.328
#> GSM97108 2 0.295 0.7161 0.052 0.948
#> GSM97050 2 0.886 0.6696 0.304 0.696
#> GSM97080 1 0.917 0.9482 0.668 0.332
#> GSM97089 1 0.949 0.9031 0.632 0.368
#> GSM97092 2 0.939 0.6611 0.356 0.644
#> GSM97093 2 0.958 0.3923 0.380 0.620
#> GSM97058 2 0.844 0.6822 0.272 0.728
#> GSM97051 2 0.886 0.6696 0.304 0.696
#> GSM97052 2 0.943 0.6592 0.360 0.640
#> GSM97061 2 0.943 0.6592 0.360 0.640
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM97138 1 0.7395 0.1268 0.580 0.380 0.040
#> GSM97145 2 0.1832 0.6703 0.008 0.956 0.036
#> GSM97147 2 0.1643 0.7083 0.000 0.956 0.044
#> GSM97125 2 0.6998 0.0787 0.292 0.664 0.044
#> GSM97127 2 0.1832 0.6703 0.008 0.956 0.036
#> GSM97130 2 0.2280 0.6780 0.008 0.940 0.052
#> GSM97133 2 0.1832 0.6703 0.008 0.956 0.036
#> GSM97134 2 0.2056 0.6705 0.024 0.952 0.024
#> GSM97120 2 0.1832 0.6703 0.008 0.956 0.036
#> GSM97126 1 0.5660 0.5068 0.772 0.200 0.028
#> GSM97112 1 0.2492 0.6008 0.936 0.048 0.016
#> GSM97115 2 0.2173 0.7031 0.008 0.944 0.048
#> GSM97116 2 0.7084 0.0440 0.304 0.652 0.044
#> GSM97117 2 0.0424 0.6921 0.000 0.992 0.008
#> GSM97119 1 0.7027 0.3339 0.660 0.296 0.044
#> GSM97122 1 0.7056 0.3261 0.656 0.300 0.044
#> GSM97135 1 0.6839 0.3625 0.684 0.272 0.044
#> GSM97136 1 0.8882 0.3874 0.540 0.316 0.144
#> GSM97139 2 0.2663 0.6515 0.024 0.932 0.044
#> GSM97146 2 0.4058 0.6016 0.076 0.880 0.044
#> GSM97123 2 0.6617 0.6459 0.012 0.600 0.388
#> GSM97129 2 0.1129 0.6839 0.004 0.976 0.020
#> GSM97143 1 0.5660 0.5068 0.772 0.200 0.028
#> GSM97113 2 0.5465 0.6906 0.000 0.712 0.288
#> GSM97056 2 0.1636 0.6783 0.020 0.964 0.016
#> GSM97124 2 0.2434 0.6578 0.024 0.940 0.036
#> GSM97132 2 0.7508 -0.2374 0.416 0.544 0.040
#> GSM97144 2 0.2280 0.6780 0.008 0.940 0.052
#> GSM97149 2 0.2663 0.6515 0.024 0.932 0.044
#> GSM97068 2 0.5785 0.6779 0.000 0.668 0.332
#> GSM97071 2 0.5455 0.3959 0.020 0.776 0.204
#> GSM97086 2 0.5988 0.6641 0.000 0.632 0.368
#> GSM97103 2 0.6313 0.6585 0.016 0.676 0.308
#> GSM97057 2 0.5810 0.6766 0.000 0.664 0.336
#> GSM97060 2 0.9405 0.4911 0.204 0.496 0.300
#> GSM97075 1 0.9648 0.2577 0.408 0.384 0.208
#> GSM97098 2 0.6313 0.6585 0.016 0.676 0.308
#> GSM97099 2 0.5595 0.6719 0.016 0.756 0.228
#> GSM97101 2 0.2796 0.7159 0.000 0.908 0.092
#> GSM97105 2 0.5988 0.6641 0.000 0.632 0.368
#> GSM97106 2 0.6026 0.6599 0.000 0.624 0.376
#> GSM97121 2 0.2796 0.7159 0.000 0.908 0.092
#> GSM97128 1 0.1711 0.6054 0.960 0.032 0.008
#> GSM97131 2 0.5988 0.6641 0.000 0.632 0.368
#> GSM97137 2 0.1170 0.6882 0.008 0.976 0.016
#> GSM97118 1 0.5610 0.5111 0.776 0.196 0.028
#> GSM97114 2 0.0424 0.6921 0.000 0.992 0.008
#> GSM97142 1 0.6839 0.3625 0.684 0.272 0.044
#> GSM97140 2 0.1643 0.7083 0.000 0.956 0.044
#> GSM97141 2 0.2537 0.7135 0.000 0.920 0.080
#> GSM97055 1 0.1315 0.6047 0.972 0.020 0.008
#> GSM97090 2 0.1453 0.6928 0.008 0.968 0.024
#> GSM97091 1 0.1315 0.6047 0.972 0.020 0.008
#> GSM97148 2 0.2663 0.6515 0.024 0.932 0.044
#> GSM97063 1 0.1315 0.6047 0.972 0.020 0.008
#> GSM97053 2 0.3369 0.6270 0.052 0.908 0.040
#> GSM97066 3 0.9575 0.9295 0.216 0.320 0.464
#> GSM97079 2 0.5988 0.6641 0.000 0.632 0.368
#> GSM97083 1 0.1585 0.6050 0.964 0.028 0.008
#> GSM97084 2 0.5988 0.6641 0.000 0.632 0.368
#> GSM97094 2 0.6211 0.4196 0.228 0.736 0.036
#> GSM97096 1 0.9645 0.2654 0.412 0.380 0.208
#> GSM97097 2 0.5988 0.6641 0.000 0.632 0.368
#> GSM97107 2 0.3043 0.6589 0.008 0.908 0.084
#> GSM97054 2 0.5988 0.6641 0.000 0.632 0.368
#> GSM97062 2 0.5988 0.6641 0.000 0.632 0.368
#> GSM97069 1 0.8924 0.4071 0.524 0.140 0.336
#> GSM97070 3 0.9575 0.9295 0.216 0.320 0.464
#> GSM97073 3 0.9575 0.9295 0.216 0.320 0.464
#> GSM97076 3 0.9899 0.8671 0.280 0.320 0.400
#> GSM97077 2 0.3482 0.7136 0.000 0.872 0.128
#> GSM97095 2 0.5987 0.4595 0.208 0.756 0.036
#> GSM97102 1 0.6624 0.5532 0.708 0.044 0.248
#> GSM97109 2 0.0424 0.6921 0.000 0.992 0.008
#> GSM97110 2 0.5919 0.6840 0.016 0.724 0.260
#> GSM97074 3 0.9899 0.8671 0.280 0.320 0.400
#> GSM97085 1 0.1315 0.6047 0.972 0.020 0.008
#> GSM97059 2 0.1643 0.7083 0.000 0.956 0.044
#> GSM97072 3 0.7378 0.7084 0.052 0.320 0.628
#> GSM97078 1 0.8841 0.3573 0.528 0.340 0.132
#> GSM97067 3 0.9575 0.9295 0.216 0.320 0.464
#> GSM97087 1 0.7331 0.5471 0.672 0.072 0.256
#> GSM97111 2 0.0424 0.6921 0.000 0.992 0.008
#> GSM97064 2 0.6209 0.6632 0.004 0.628 0.368
#> GSM97065 3 0.9575 0.9295 0.216 0.320 0.464
#> GSM97081 1 0.9648 0.2577 0.408 0.384 0.208
#> GSM97082 1 0.7331 0.5471 0.672 0.072 0.256
#> GSM97088 1 0.1585 0.6050 0.964 0.028 0.008
#> GSM97100 2 0.5988 0.6641 0.000 0.632 0.368
#> GSM97104 1 0.6624 0.5532 0.708 0.044 0.248
#> GSM97108 2 0.2796 0.7159 0.000 0.908 0.092
#> GSM97050 2 0.5988 0.6641 0.000 0.632 0.368
#> GSM97080 1 0.7565 0.5464 0.660 0.084 0.256
#> GSM97089 1 0.7489 0.5453 0.664 0.080 0.256
#> GSM97092 2 0.7410 0.6285 0.040 0.576 0.384
#> GSM97093 2 0.9223 0.2850 0.272 0.528 0.200
#> GSM97058 2 0.5810 0.6766 0.000 0.664 0.336
#> GSM97051 2 0.5988 0.6641 0.000 0.632 0.368
#> GSM97052 2 0.7424 0.6256 0.040 0.572 0.388
#> GSM97061 2 0.7424 0.6256 0.040 0.572 0.388
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM97138 1 0.7416 -0.367 0.440 0.000 0.392 0.168
#> GSM97145 1 0.0707 0.790 0.980 0.020 0.000 0.000
#> GSM97147 1 0.4874 0.733 0.764 0.180 0.000 0.056
#> GSM97125 1 0.6162 0.466 0.704 0.012 0.148 0.136
#> GSM97127 1 0.0707 0.790 0.980 0.020 0.000 0.000
#> GSM97130 1 0.3972 0.787 0.840 0.080 0.000 0.080
#> GSM97133 1 0.0707 0.790 0.980 0.020 0.000 0.000
#> GSM97134 1 0.1917 0.794 0.944 0.036 0.008 0.012
#> GSM97120 1 0.0707 0.790 0.980 0.020 0.000 0.000
#> GSM97126 3 0.6780 0.604 0.232 0.000 0.604 0.164
#> GSM97112 3 0.4864 0.627 0.060 0.000 0.768 0.172
#> GSM97115 1 0.2944 0.785 0.868 0.128 0.000 0.004
#> GSM97116 1 0.5724 0.455 0.716 0.000 0.144 0.140
#> GSM97117 1 0.3323 0.793 0.876 0.064 0.000 0.060
#> GSM97119 3 0.7277 0.523 0.360 0.000 0.484 0.156
#> GSM97122 3 0.7286 0.519 0.364 0.000 0.480 0.156
#> GSM97135 3 0.7314 0.533 0.336 0.000 0.496 0.168
#> GSM97136 3 0.6619 0.545 0.328 0.004 0.580 0.088
#> GSM97139 1 0.0188 0.777 0.996 0.000 0.000 0.004
#> GSM97146 1 0.1833 0.754 0.944 0.000 0.032 0.024
#> GSM97123 2 0.1724 0.827 0.000 0.948 0.032 0.020
#> GSM97129 1 0.2300 0.798 0.924 0.048 0.000 0.028
#> GSM97143 3 0.6780 0.604 0.232 0.000 0.604 0.164
#> GSM97113 2 0.4415 0.737 0.140 0.804 0.000 0.056
#> GSM97056 1 0.1022 0.794 0.968 0.032 0.000 0.000
#> GSM97124 1 0.1377 0.787 0.964 0.020 0.008 0.008
#> GSM97132 1 0.6731 0.175 0.604 0.000 0.248 0.148
#> GSM97144 1 0.3972 0.787 0.840 0.080 0.000 0.080
#> GSM97149 1 0.0188 0.777 0.996 0.000 0.000 0.004
#> GSM97068 2 0.2469 0.816 0.108 0.892 0.000 0.000
#> GSM97071 1 0.5986 0.538 0.620 0.060 0.000 0.320
#> GSM97086 2 0.0817 0.858 0.024 0.976 0.000 0.000
#> GSM97103 2 0.7708 0.442 0.268 0.576 0.076 0.080
#> GSM97057 2 0.2408 0.819 0.104 0.896 0.000 0.000
#> GSM97060 2 0.4574 0.641 0.000 0.756 0.220 0.024
#> GSM97075 3 0.8166 0.445 0.324 0.092 0.504 0.080
#> GSM97098 2 0.7708 0.442 0.268 0.576 0.076 0.080
#> GSM97099 1 0.8148 0.174 0.456 0.384 0.076 0.084
#> GSM97101 1 0.5769 0.592 0.652 0.292 0.000 0.056
#> GSM97105 2 0.0817 0.858 0.024 0.976 0.000 0.000
#> GSM97106 2 0.0188 0.840 0.000 0.996 0.004 0.000
#> GSM97121 1 0.5769 0.592 0.652 0.292 0.000 0.056
#> GSM97128 3 0.3925 0.625 0.016 0.000 0.808 0.176
#> GSM97131 2 0.0817 0.858 0.024 0.976 0.000 0.000
#> GSM97137 1 0.1661 0.797 0.944 0.052 0.000 0.004
#> GSM97118 3 0.6714 0.608 0.228 0.000 0.612 0.160
#> GSM97114 1 0.3323 0.793 0.876 0.064 0.000 0.060
#> GSM97142 3 0.7314 0.533 0.336 0.000 0.496 0.168
#> GSM97140 1 0.4874 0.733 0.764 0.180 0.000 0.056
#> GSM97141 1 0.5550 0.653 0.692 0.248 0.000 0.060
#> GSM97055 3 0.3311 0.622 0.000 0.000 0.828 0.172
#> GSM97090 1 0.2542 0.798 0.904 0.084 0.000 0.012
#> GSM97091 3 0.3311 0.622 0.000 0.000 0.828 0.172
#> GSM97148 1 0.0188 0.777 0.996 0.000 0.000 0.004
#> GSM97063 3 0.3311 0.622 0.000 0.000 0.828 0.172
#> GSM97053 1 0.1362 0.768 0.964 0.004 0.020 0.012
#> GSM97066 4 0.0336 0.934 0.008 0.000 0.000 0.992
#> GSM97079 2 0.0817 0.858 0.024 0.976 0.000 0.000
#> GSM97083 3 0.3681 0.624 0.008 0.000 0.816 0.176
#> GSM97084 2 0.0817 0.858 0.024 0.976 0.000 0.000
#> GSM97094 1 0.6040 0.543 0.708 0.044 0.208 0.040
#> GSM97096 3 0.8153 0.452 0.320 0.092 0.508 0.080
#> GSM97097 2 0.0817 0.858 0.024 0.976 0.000 0.000
#> GSM97107 1 0.5080 0.751 0.764 0.092 0.000 0.144
#> GSM97054 2 0.0921 0.857 0.028 0.972 0.000 0.000
#> GSM97062 2 0.0817 0.858 0.024 0.976 0.000 0.000
#> GSM97069 3 0.4883 0.329 0.000 0.016 0.696 0.288
#> GSM97070 4 0.0336 0.934 0.008 0.000 0.000 0.992
#> GSM97073 4 0.0336 0.934 0.008 0.000 0.000 0.992
#> GSM97076 4 0.2124 0.876 0.008 0.000 0.068 0.924
#> GSM97077 1 0.6421 0.228 0.508 0.432 0.004 0.056
#> GSM97095 1 0.5972 0.574 0.720 0.048 0.192 0.040
#> GSM97102 3 0.3080 0.550 0.000 0.024 0.880 0.096
#> GSM97109 1 0.3323 0.793 0.876 0.064 0.000 0.060
#> GSM97110 2 0.7505 0.301 0.332 0.544 0.052 0.072
#> GSM97074 4 0.2124 0.876 0.008 0.000 0.068 0.924
#> GSM97085 3 0.3311 0.622 0.000 0.000 0.828 0.172
#> GSM97059 1 0.4874 0.733 0.764 0.180 0.000 0.056
#> GSM97072 4 0.4088 0.724 0.008 0.012 0.172 0.808
#> GSM97078 3 0.7680 0.527 0.328 0.028 0.520 0.124
#> GSM97067 4 0.0336 0.934 0.008 0.000 0.000 0.992
#> GSM97087 3 0.4022 0.548 0.000 0.068 0.836 0.096
#> GSM97111 1 0.3323 0.793 0.876 0.064 0.000 0.060
#> GSM97064 2 0.3215 0.828 0.064 0.892 0.024 0.020
#> GSM97065 4 0.0336 0.934 0.008 0.000 0.000 0.992
#> GSM97081 3 0.8166 0.445 0.324 0.092 0.504 0.080
#> GSM97082 3 0.4022 0.548 0.000 0.068 0.836 0.096
#> GSM97088 3 0.3681 0.624 0.008 0.000 0.816 0.176
#> GSM97100 2 0.0817 0.858 0.024 0.976 0.000 0.000
#> GSM97104 3 0.3080 0.550 0.000 0.024 0.880 0.096
#> GSM97108 1 0.5769 0.592 0.652 0.292 0.000 0.056
#> GSM97050 2 0.0817 0.858 0.024 0.976 0.000 0.000
#> GSM97080 3 0.3763 0.521 0.000 0.024 0.832 0.144
#> GSM97089 3 0.4402 0.549 0.012 0.064 0.828 0.096
#> GSM97092 2 0.2667 0.813 0.008 0.912 0.060 0.020
#> GSM97093 2 0.8382 -0.133 0.324 0.348 0.312 0.016
#> GSM97058 2 0.2408 0.819 0.104 0.896 0.000 0.000
#> GSM97051 2 0.0817 0.858 0.024 0.976 0.000 0.000
#> GSM97052 2 0.2335 0.812 0.000 0.920 0.060 0.020
#> GSM97061 2 0.2335 0.812 0.000 0.920 0.060 0.020
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM97138 5 0.5811 0.5049 0.340 0.000 0.000 0.108 0.552
#> GSM97145 1 0.1787 0.7702 0.936 0.016 0.004 0.000 0.044
#> GSM97147 1 0.4352 0.7179 0.772 0.172 0.020 0.036 0.000
#> GSM97125 1 0.5811 0.4021 0.640 0.012 0.004 0.100 0.244
#> GSM97127 1 0.1787 0.7702 0.936 0.016 0.004 0.000 0.044
#> GSM97130 1 0.3646 0.7630 0.848 0.064 0.016 0.068 0.004
#> GSM97133 1 0.1787 0.7702 0.936 0.016 0.004 0.000 0.044
#> GSM97134 1 0.1975 0.7750 0.936 0.020 0.004 0.016 0.024
#> GSM97120 1 0.1787 0.7702 0.936 0.016 0.004 0.000 0.044
#> GSM97126 5 0.5276 0.6819 0.148 0.000 0.024 0.108 0.720
#> GSM97112 5 0.2228 0.6805 0.020 0.000 0.056 0.008 0.916
#> GSM97115 1 0.2780 0.7669 0.872 0.112 0.008 0.004 0.004
#> GSM97116 1 0.5470 0.3386 0.628 0.000 0.000 0.104 0.268
#> GSM97117 1 0.2901 0.7701 0.888 0.048 0.020 0.044 0.000
#> GSM97119 5 0.5449 0.6258 0.256 0.000 0.000 0.108 0.636
#> GSM97122 5 0.5472 0.6216 0.260 0.000 0.000 0.108 0.632
#> GSM97135 5 0.5303 0.6439 0.232 0.000 0.000 0.108 0.660
#> GSM97136 3 0.7511 0.2548 0.296 0.000 0.412 0.044 0.248
#> GSM97139 1 0.1732 0.7485 0.920 0.000 0.000 0.000 0.080
#> GSM97146 1 0.2921 0.7134 0.856 0.000 0.000 0.020 0.124
#> GSM97123 2 0.2280 0.7998 0.000 0.880 0.120 0.000 0.000
#> GSM97129 1 0.2227 0.7795 0.924 0.032 0.004 0.028 0.012
#> GSM97143 5 0.5276 0.6819 0.148 0.000 0.024 0.108 0.720
#> GSM97113 2 0.3935 0.7182 0.140 0.808 0.016 0.036 0.000
#> GSM97056 1 0.1525 0.7720 0.948 0.012 0.004 0.000 0.036
#> GSM97124 1 0.2082 0.7679 0.928 0.012 0.004 0.012 0.044
#> GSM97132 1 0.5876 -0.0385 0.512 0.000 0.000 0.104 0.384
#> GSM97144 1 0.3646 0.7630 0.848 0.064 0.016 0.068 0.004
#> GSM97149 1 0.1792 0.7465 0.916 0.000 0.000 0.000 0.084
#> GSM97068 2 0.2074 0.8045 0.104 0.896 0.000 0.000 0.000
#> GSM97071 1 0.5751 0.5314 0.620 0.048 0.028 0.300 0.004
#> GSM97086 2 0.0162 0.8588 0.004 0.996 0.000 0.000 0.000
#> GSM97103 2 0.6520 0.4191 0.268 0.576 0.116 0.040 0.000
#> GSM97057 2 0.2020 0.8078 0.100 0.900 0.000 0.000 0.000
#> GSM97060 2 0.3857 0.5934 0.000 0.688 0.312 0.000 0.000
#> GSM97075 3 0.7411 0.4523 0.308 0.032 0.500 0.032 0.128
#> GSM97098 2 0.6520 0.4191 0.268 0.576 0.116 0.040 0.000
#> GSM97099 1 0.6906 0.1986 0.464 0.376 0.116 0.044 0.000
#> GSM97101 1 0.5134 0.6110 0.664 0.280 0.020 0.036 0.000
#> GSM97105 2 0.0162 0.8588 0.004 0.996 0.000 0.000 0.000
#> GSM97106 2 0.1608 0.8210 0.000 0.928 0.072 0.000 0.000
#> GSM97121 1 0.5134 0.6110 0.664 0.280 0.020 0.036 0.000
#> GSM97128 5 0.2408 0.6706 0.004 0.000 0.096 0.008 0.892
#> GSM97131 2 0.0162 0.8588 0.004 0.996 0.000 0.000 0.000
#> GSM97137 1 0.1573 0.7779 0.948 0.036 0.008 0.004 0.004
#> GSM97118 5 0.5223 0.6829 0.144 0.000 0.028 0.100 0.728
#> GSM97114 1 0.2901 0.7701 0.888 0.048 0.020 0.044 0.000
#> GSM97142 5 0.5303 0.6439 0.232 0.000 0.000 0.108 0.660
#> GSM97140 1 0.4352 0.7179 0.772 0.172 0.020 0.036 0.000
#> GSM97141 1 0.4911 0.6658 0.708 0.232 0.020 0.040 0.000
#> GSM97055 5 0.2249 0.6681 0.000 0.000 0.096 0.008 0.896
#> GSM97090 1 0.2512 0.7799 0.904 0.068 0.008 0.012 0.008
#> GSM97091 5 0.2249 0.6681 0.000 0.000 0.096 0.008 0.896
#> GSM97148 1 0.1792 0.7465 0.916 0.000 0.000 0.000 0.084
#> GSM97063 5 0.2249 0.6681 0.000 0.000 0.096 0.008 0.896
#> GSM97053 1 0.2305 0.7378 0.896 0.000 0.000 0.012 0.092
#> GSM97066 4 0.1410 0.9438 0.000 0.000 0.060 0.940 0.000
#> GSM97079 2 0.0162 0.8588 0.004 0.996 0.000 0.000 0.000
#> GSM97083 5 0.2304 0.6664 0.000 0.000 0.100 0.008 0.892
#> GSM97084 2 0.0162 0.8588 0.004 0.996 0.000 0.000 0.000
#> GSM97094 1 0.5978 0.5489 0.672 0.040 0.024 0.048 0.216
#> GSM97096 3 0.7398 0.4546 0.304 0.032 0.504 0.032 0.128
#> GSM97097 2 0.0162 0.8588 0.004 0.996 0.000 0.000 0.000
#> GSM97107 1 0.4965 0.7218 0.764 0.076 0.028 0.124 0.008
#> GSM97054 2 0.0451 0.8571 0.008 0.988 0.004 0.000 0.000
#> GSM97062 2 0.0162 0.8588 0.004 0.996 0.000 0.000 0.000
#> GSM97069 3 0.4367 0.4592 0.000 0.000 0.748 0.192 0.060
#> GSM97070 4 0.1410 0.9438 0.000 0.000 0.060 0.940 0.000
#> GSM97073 4 0.1410 0.9438 0.000 0.000 0.060 0.940 0.000
#> GSM97076 4 0.0566 0.8972 0.000 0.000 0.004 0.984 0.012
#> GSM97077 1 0.5690 0.2518 0.508 0.432 0.024 0.036 0.000
#> GSM97095 1 0.5930 0.5803 0.684 0.044 0.024 0.048 0.200
#> GSM97102 3 0.1608 0.6219 0.000 0.000 0.928 0.000 0.072
#> GSM97109 1 0.2901 0.7701 0.888 0.048 0.020 0.044 0.000
#> GSM97110 2 0.6442 0.2766 0.332 0.544 0.080 0.044 0.000
#> GSM97074 4 0.0566 0.8972 0.000 0.000 0.004 0.984 0.012
#> GSM97085 5 0.2249 0.6681 0.000 0.000 0.096 0.008 0.896
#> GSM97059 1 0.4352 0.7179 0.772 0.172 0.020 0.036 0.000
#> GSM97072 4 0.3750 0.7435 0.000 0.012 0.232 0.756 0.000
#> GSM97078 5 0.7703 0.3289 0.284 0.028 0.152 0.048 0.488
#> GSM97067 4 0.1410 0.9438 0.000 0.000 0.060 0.940 0.000
#> GSM97087 3 0.1041 0.6254 0.000 0.004 0.964 0.000 0.032
#> GSM97111 1 0.2901 0.7701 0.888 0.048 0.020 0.044 0.000
#> GSM97064 2 0.3102 0.8148 0.056 0.860 0.084 0.000 0.000
#> GSM97065 4 0.1410 0.9438 0.000 0.000 0.060 0.940 0.000
#> GSM97081 3 0.7411 0.4523 0.308 0.032 0.500 0.032 0.128
#> GSM97082 3 0.1041 0.6254 0.000 0.004 0.964 0.000 0.032
#> GSM97088 5 0.2304 0.6664 0.000 0.000 0.100 0.008 0.892
#> GSM97100 2 0.0162 0.8588 0.004 0.996 0.000 0.000 0.000
#> GSM97104 3 0.1608 0.6219 0.000 0.000 0.928 0.000 0.072
#> GSM97108 1 0.5134 0.6110 0.664 0.280 0.020 0.036 0.000
#> GSM97050 2 0.0162 0.8588 0.004 0.996 0.000 0.000 0.000
#> GSM97080 3 0.2782 0.6018 0.000 0.000 0.880 0.048 0.072
#> GSM97089 3 0.1757 0.6318 0.012 0.004 0.936 0.000 0.048
#> GSM97092 2 0.2886 0.7807 0.008 0.844 0.148 0.000 0.000
#> GSM97093 3 0.7801 0.2756 0.308 0.280 0.352 0.000 0.060
#> GSM97058 2 0.2020 0.8078 0.100 0.900 0.000 0.000 0.000
#> GSM97051 2 0.0162 0.8588 0.004 0.996 0.000 0.000 0.000
#> GSM97052 2 0.2605 0.7809 0.000 0.852 0.148 0.000 0.000
#> GSM97061 2 0.2605 0.7809 0.000 0.852 0.148 0.000 0.000
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM97138 5 0.6198 0.0963 0.408 0.156 0.000 0.000 0.412 0.024
#> GSM97145 2 0.3765 0.0263 0.404 0.596 0.000 0.000 0.000 0.000
#> GSM97147 2 0.2092 0.4687 0.000 0.876 0.000 0.124 0.000 0.000
#> GSM97125 1 0.6122 0.6401 0.440 0.384 0.000 0.000 0.156 0.020
#> GSM97127 2 0.3765 0.0263 0.404 0.596 0.000 0.000 0.000 0.000
#> GSM97130 2 0.3967 0.3543 0.356 0.632 0.000 0.012 0.000 0.000
#> GSM97133 2 0.3756 0.0318 0.400 0.600 0.000 0.000 0.000 0.000
#> GSM97134 2 0.3636 0.2018 0.320 0.676 0.000 0.000 0.000 0.004
#> GSM97120 2 0.3765 0.0263 0.404 0.596 0.000 0.000 0.000 0.000
#> GSM97126 5 0.4668 0.5972 0.324 0.024 0.000 0.000 0.628 0.024
#> GSM97112 5 0.1124 0.6842 0.036 0.008 0.000 0.000 0.956 0.000
#> GSM97115 2 0.4619 0.3442 0.244 0.668 0.000 0.088 0.000 0.000
#> GSM97116 1 0.6000 0.6781 0.504 0.324 0.000 0.000 0.152 0.020
#> GSM97117 2 0.0653 0.4634 0.012 0.980 0.000 0.004 0.000 0.004
#> GSM97119 5 0.5500 0.4445 0.412 0.068 0.000 0.000 0.496 0.024
#> GSM97122 5 0.5504 0.4386 0.416 0.068 0.000 0.000 0.492 0.024
#> GSM97135 5 0.5299 0.4833 0.404 0.052 0.000 0.000 0.520 0.024
#> GSM97136 3 0.8146 0.3322 0.192 0.180 0.396 0.000 0.176 0.056
#> GSM97139 2 0.4095 -0.2570 0.480 0.512 0.000 0.000 0.008 0.000
#> GSM97146 1 0.4500 0.1799 0.496 0.480 0.000 0.000 0.012 0.012
#> GSM97123 4 0.2964 0.7949 0.032 0.004 0.108 0.852 0.004 0.000
#> GSM97129 2 0.3109 0.3317 0.224 0.772 0.000 0.000 0.000 0.004
#> GSM97143 5 0.4668 0.5972 0.324 0.024 0.000 0.000 0.628 0.024
#> GSM97113 4 0.3109 0.7038 0.004 0.224 0.000 0.772 0.000 0.000
#> GSM97056 2 0.3892 0.1567 0.352 0.640 0.000 0.004 0.004 0.000
#> GSM97124 2 0.3819 0.0681 0.372 0.624 0.000 0.000 0.004 0.000
#> GSM97132 1 0.6376 0.5074 0.448 0.272 0.000 0.000 0.260 0.020
#> GSM97144 2 0.3967 0.3543 0.356 0.632 0.000 0.012 0.000 0.000
#> GSM97149 2 0.4080 -0.2005 0.456 0.536 0.000 0.000 0.008 0.000
#> GSM97068 4 0.2048 0.7949 0.000 0.120 0.000 0.880 0.000 0.000
#> GSM97071 2 0.5925 0.2373 0.244 0.552 0.004 0.012 0.000 0.188
#> GSM97086 4 0.0146 0.8519 0.000 0.004 0.000 0.996 0.000 0.000
#> GSM97103 4 0.5854 0.4278 0.032 0.344 0.088 0.532 0.000 0.004
#> GSM97057 4 0.2003 0.7984 0.000 0.116 0.000 0.884 0.000 0.000
#> GSM97060 4 0.4408 0.5903 0.032 0.004 0.300 0.660 0.004 0.000
#> GSM97075 3 0.7776 0.4865 0.168 0.212 0.468 0.008 0.096 0.048
#> GSM97098 4 0.5854 0.4278 0.032 0.344 0.088 0.532 0.000 0.004
#> GSM97099 2 0.5788 0.1419 0.032 0.556 0.088 0.320 0.000 0.004
#> GSM97101 2 0.3023 0.4186 0.000 0.768 0.000 0.232 0.000 0.000
#> GSM97105 4 0.0146 0.8519 0.000 0.004 0.000 0.996 0.000 0.000
#> GSM97106 4 0.2069 0.8198 0.020 0.000 0.068 0.908 0.004 0.000
#> GSM97121 2 0.3023 0.4186 0.000 0.768 0.000 0.232 0.000 0.000
#> GSM97128 5 0.1390 0.6849 0.016 0.004 0.032 0.000 0.948 0.000
#> GSM97131 4 0.0146 0.8519 0.000 0.004 0.000 0.996 0.000 0.000
#> GSM97137 2 0.3695 0.3204 0.272 0.712 0.000 0.016 0.000 0.000
#> GSM97118 5 0.4622 0.6045 0.316 0.020 0.004 0.000 0.640 0.020
#> GSM97114 2 0.0653 0.4634 0.012 0.980 0.000 0.004 0.000 0.004
#> GSM97142 5 0.5299 0.4833 0.404 0.052 0.000 0.000 0.520 0.024
#> GSM97140 2 0.2234 0.4699 0.004 0.872 0.000 0.124 0.000 0.000
#> GSM97141 2 0.2805 0.4360 0.000 0.812 0.000 0.184 0.000 0.004
#> GSM97055 5 0.0547 0.6808 0.000 0.000 0.020 0.000 0.980 0.000
#> GSM97090 2 0.4193 0.3333 0.272 0.684 0.000 0.044 0.000 0.000
#> GSM97091 5 0.0547 0.6808 0.000 0.000 0.020 0.000 0.980 0.000
#> GSM97148 2 0.4080 -0.2005 0.456 0.536 0.000 0.000 0.008 0.000
#> GSM97063 5 0.0547 0.6808 0.000 0.000 0.020 0.000 0.980 0.000
#> GSM97053 2 0.4333 -0.3005 0.468 0.512 0.000 0.000 0.020 0.000
#> GSM97066 6 0.0000 0.9472 0.000 0.000 0.000 0.000 0.000 1.000
#> GSM97079 4 0.0146 0.8519 0.000 0.004 0.000 0.996 0.000 0.000
#> GSM97083 5 0.1151 0.6813 0.012 0.000 0.032 0.000 0.956 0.000
#> GSM97084 4 0.0000 0.8502 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM97094 2 0.6200 -0.0887 0.324 0.472 0.000 0.008 0.188 0.008
#> GSM97096 3 0.7758 0.4883 0.168 0.208 0.472 0.008 0.096 0.048
#> GSM97097 4 0.0291 0.8503 0.004 0.004 0.000 0.992 0.000 0.000
#> GSM97107 2 0.4254 0.3098 0.352 0.624 0.004 0.020 0.000 0.000
#> GSM97054 4 0.0508 0.8493 0.004 0.012 0.000 0.984 0.000 0.000
#> GSM97062 4 0.0146 0.8519 0.000 0.004 0.000 0.996 0.000 0.000
#> GSM97069 3 0.3854 0.5223 0.016 0.000 0.768 0.000 0.032 0.184
#> GSM97070 6 0.0000 0.9472 0.000 0.000 0.000 0.000 0.000 1.000
#> GSM97073 6 0.0000 0.9472 0.000 0.000 0.000 0.000 0.000 1.000
#> GSM97076 6 0.1471 0.9085 0.064 0.000 0.000 0.000 0.004 0.932
#> GSM97077 2 0.4066 0.1476 0.012 0.596 0.000 0.392 0.000 0.000
#> GSM97095 2 0.6073 -0.0128 0.292 0.512 0.000 0.008 0.180 0.008
#> GSM97102 3 0.1461 0.6623 0.016 0.000 0.940 0.000 0.044 0.000
#> GSM97109 2 0.0653 0.4634 0.012 0.980 0.000 0.004 0.000 0.004
#> GSM97110 4 0.5556 0.3040 0.028 0.412 0.056 0.500 0.000 0.004
#> GSM97074 6 0.1471 0.9085 0.064 0.000 0.000 0.000 0.004 0.932
#> GSM97085 5 0.0547 0.6808 0.000 0.000 0.020 0.000 0.980 0.000
#> GSM97059 2 0.2234 0.4699 0.004 0.872 0.000 0.124 0.000 0.000
#> GSM97072 6 0.3121 0.7523 0.012 0.000 0.180 0.004 0.000 0.804
#> GSM97078 5 0.7783 0.3199 0.204 0.180 0.088 0.008 0.468 0.052
#> GSM97067 6 0.0000 0.9472 0.000 0.000 0.000 0.000 0.000 1.000
#> GSM97087 3 0.0363 0.6673 0.000 0.000 0.988 0.000 0.012 0.000
#> GSM97111 2 0.0653 0.4634 0.012 0.980 0.000 0.004 0.000 0.004
#> GSM97064 4 0.3594 0.8033 0.032 0.068 0.064 0.832 0.004 0.000
#> GSM97065 6 0.0000 0.9472 0.000 0.000 0.000 0.000 0.000 1.000
#> GSM97081 3 0.7776 0.4865 0.168 0.212 0.468 0.008 0.096 0.048
#> GSM97082 3 0.0363 0.6673 0.000 0.000 0.988 0.000 0.012 0.000
#> GSM97088 5 0.1151 0.6813 0.012 0.000 0.032 0.000 0.956 0.000
#> GSM97100 4 0.0146 0.8519 0.000 0.004 0.000 0.996 0.000 0.000
#> GSM97104 3 0.1461 0.6623 0.016 0.000 0.940 0.000 0.044 0.000
#> GSM97108 2 0.3023 0.4186 0.000 0.768 0.000 0.232 0.000 0.000
#> GSM97050 4 0.0146 0.8519 0.000 0.004 0.000 0.996 0.000 0.000
#> GSM97080 3 0.2171 0.6470 0.016 0.000 0.912 0.000 0.032 0.040
#> GSM97089 3 0.1074 0.6723 0.000 0.012 0.960 0.000 0.028 0.000
#> GSM97092 4 0.3475 0.7754 0.032 0.012 0.136 0.816 0.004 0.000
#> GSM97093 3 0.7731 0.2862 0.072 0.272 0.360 0.264 0.028 0.004
#> GSM97058 4 0.2003 0.7984 0.000 0.116 0.000 0.884 0.000 0.000
#> GSM97051 4 0.0146 0.8519 0.000 0.004 0.000 0.996 0.000 0.000
#> GSM97052 4 0.3266 0.7770 0.032 0.004 0.136 0.824 0.004 0.000
#> GSM97061 4 0.3266 0.7770 0.032 0.004 0.136 0.824 0.004 0.000
Heatmaps for the consensus matrix. It visualizes the probability of two samples to be in a same group.
consensus_heatmap(res, k = 2)
consensus_heatmap(res, k = 3)
consensus_heatmap(res, k = 4)
consensus_heatmap(res, k = 5)
consensus_heatmap(res, k = 6)
Heatmaps for the membership of samples in all partitions to see how consistent they are:
membership_heatmap(res, k = 2)
membership_heatmap(res, k = 3)
membership_heatmap(res, k = 4)
membership_heatmap(res, k = 5)
membership_heatmap(res, k = 6)
As soon as we have had the classes for columns, we can look for signatures which are significantly different between classes which can be candidate marks for certain classes. Following are the heatmaps for signatures.
Signature heatmaps where rows are scaled:
get_signatures(res, k = 2)
get_signatures(res, k = 3)
get_signatures(res, k = 4)
get_signatures(res, k = 5)
get_signatures(res, k = 6)
Signature heatmaps where rows are not scaled:
get_signatures(res, k = 2, scale_rows = FALSE)
get_signatures(res, k = 3, scale_rows = FALSE)
get_signatures(res, k = 4, scale_rows = FALSE)
get_signatures(res, k = 5, scale_rows = FALSE)
get_signatures(res, k = 6, scale_rows = FALSE)
Compare the overlap of signatures from different k:
compare_signatures(res)
get_signature() returns a data frame invisibly. TO get the list of signatures, the function
call should be assigned to a variable explicitly. In following code, if plot argument is set
to FALSE, no heatmap is plotted while only the differential analysis is performed.
# code only for demonstration
tb = get_signature(res, k = ..., plot = FALSE)
An example of the output of tb is:
#> which_row fdr mean_1 mean_2 scaled_mean_1 scaled_mean_2 km
#> 1 38 0.042760348 8.373488 9.131774 -0.5533452 0.5164555 1
#> 2 40 0.018707592 7.106213 8.469186 -0.6173731 0.5762149 1
#> 3 55 0.019134737 10.221463 11.207825 -0.6159697 0.5749050 1
#> 4 59 0.006059896 5.921854 7.869574 -0.6899429 0.6439467 1
#> 5 60 0.018055526 8.928898 10.211722 -0.6204761 0.5791110 1
#> 6 98 0.009384629 15.714769 14.887706 0.6635654 -0.6193277 2
...
The columns in tb are:
which_row: row indices corresponding to the input matrix.fdr: FDR for the differential test. mean_x: The mean value in group x.scaled_mean_x: The mean value in group x after rows are scaled.km: Row groups if k-means clustering is applied to rows.UMAP plot which shows how samples are separated.
dimension_reduction(res, k = 2, method = "UMAP")
dimension_reduction(res, k = 3, method = "UMAP")
dimension_reduction(res, k = 4, method = "UMAP")
dimension_reduction(res, k = 5, method = "UMAP")
dimension_reduction(res, k = 6, method = "UMAP")
Following heatmap shows how subgroups are split when increasing k:
collect_classes(res)
Test correlation between subgroups and known annotations. If the known annotation is numeric, one-way ANOVA test is applied, and if the known annotation is discrete, chi-squared contingency table test is applied.
test_to_known_factors(res)
#> n disease.state(p) specimen(p) cell.type(p) other(p) k
#> ATC:hclust 90 0.01978 0.0110 3.85e-01 0.121 2
#> ATC:hclust 81 0.03939 0.0731 2.47e-02 0.182 3
#> ATC:hclust 86 0.00566 0.0555 1.79e-06 0.460 4
#> ATC:hclust 85 0.00205 0.1029 3.40e-08 0.358 5
#> ATC:hclust 51 0.01687 0.1213 1.94e-06 0.382 6
If matrix rows can be associated to genes, consider to use functional_enrichment(res,
...) to perform function enrichment for the signature genes. See this vignette for more detailed explanations.
The object with results only for a single top-value method and a single partition method can be extracted as:
res = res_list["ATC", "kmeans"]
# you can also extract it by
# res = res_list["ATC:kmeans"]
A summary of res and all the functions that can be applied to it:
res
#> A 'ConsensusPartition' object with k = 2, 3, 4, 5, 6.
#> On a matrix with 21512 rows and 100 columns.
#> Top rows (1000, 2000, 3000, 4000, 5000) are extracted by 'ATC' method.
#> Subgroups are detected by 'kmeans' method.
#> Performed in total 1250 partitions by row resampling.
#> Best k for subgroups seems to be 2.
#>
#> Following methods can be applied to this 'ConsensusPartition' object:
#> [1] "cola_report" "collect_classes" "collect_plots"
#> [4] "collect_stats" "colnames" "compare_signatures"
#> [7] "consensus_heatmap" "dimension_reduction" "functional_enrichment"
#> [10] "get_anno_col" "get_anno" "get_classes"
#> [13] "get_consensus" "get_matrix" "get_membership"
#> [16] "get_param" "get_signatures" "get_stats"
#> [19] "is_best_k" "is_stable_k" "membership_heatmap"
#> [22] "ncol" "nrow" "plot_ecdf"
#> [25] "rownames" "select_partition_number" "show"
#> [28] "suggest_best_k" "test_to_known_factors"
collect_plots() function collects all the plots made from res for all k (number of partitions)
into one single page to provide an easy and fast comparison between different k.
collect_plots(res)
The plots are:
k and the heatmap of
predicted classes for each k.k.k.k.All the plots in panels can be made by individual functions and they are plotted later in this section.
select_partition_number() produces several plots showing different
statistics for choosing “optimized” k. There are following statistics:
k;k, the area increased is defined as \(A_k - A_{k-1}\).The detailed explanations of these statistics can be found in the cola vignette.
Generally speaking, lower PAC score, higher mean silhouette score or higher
concordance corresponds to better partition. Rand index and Jaccard index
measure how similar the current partition is compared to partition with k-1.
If they are too similar, we won't accept k is better than k-1.
select_partition_number(res)
The numeric values for all these statistics can be obtained by get_stats().
get_stats(res)
#> k 1-PAC mean_silhouette concordance area_increased Rand Jaccard
#> 2 2 0.625 0.870 0.928 0.4997 0.495 0.495
#> 3 3 0.732 0.856 0.929 0.3259 0.699 0.466
#> 4 4 0.534 0.536 0.741 0.0990 0.897 0.709
#> 5 5 0.611 0.555 0.767 0.0692 0.778 0.377
#> 6 6 0.773 0.830 0.868 0.0557 0.872 0.511
suggest_best_k() suggests the best \(k\) based on these statistics. The rules are as follows:
suggest_best_k(res)
#> [1] 2
Following shows the table of the partitions (You need to click the show/hide
code output link to see it). The membership matrix (columns with name p*)
is inferred by
clue::cl_consensus()
function with the SE method. Basically the value in the membership matrix
represents the probability to belong to a certain group. The finall class
label for an item is determined with the group with highest probability it
belongs to.
In get_classes() function, the entropy is calculated from the membership
matrix and the silhouette score is calculated from the consensus matrix.
cbind(get_classes(res, k = 2), get_membership(res, k = 2))
#> class entropy silhouette p1 p2
#> GSM97138 1 0.0376 0.892 0.996 0.004
#> GSM97145 1 0.3431 0.865 0.936 0.064
#> GSM97147 2 0.0000 0.942 0.000 1.000
#> GSM97125 1 0.0376 0.892 0.996 0.004
#> GSM97127 2 0.8207 0.700 0.256 0.744
#> GSM97130 2 0.8016 0.714 0.244 0.756
#> GSM97133 2 0.8016 0.714 0.244 0.756
#> GSM97134 2 0.8813 0.637 0.300 0.700
#> GSM97120 1 0.3431 0.865 0.936 0.064
#> GSM97126 1 0.0376 0.892 0.996 0.004
#> GSM97112 1 0.0000 0.892 1.000 0.000
#> GSM97115 2 0.0000 0.942 0.000 1.000
#> GSM97116 1 0.0376 0.892 0.996 0.004
#> GSM97117 1 0.1184 0.892 0.984 0.016
#> GSM97119 1 0.0376 0.892 0.996 0.004
#> GSM97122 1 0.0376 0.892 0.996 0.004
#> GSM97135 1 0.0376 0.892 0.996 0.004
#> GSM97136 1 0.0000 0.892 1.000 0.000
#> GSM97139 1 0.3431 0.865 0.936 0.064
#> GSM97146 1 0.0376 0.892 0.996 0.004
#> GSM97123 2 0.0376 0.939 0.004 0.996
#> GSM97129 2 0.4431 0.872 0.092 0.908
#> GSM97143 1 0.0000 0.892 1.000 0.000
#> GSM97113 2 0.0000 0.942 0.000 1.000
#> GSM97056 2 0.8016 0.714 0.244 0.756
#> GSM97124 1 0.1843 0.883 0.972 0.028
#> GSM97132 1 0.0376 0.892 0.996 0.004
#> GSM97144 2 0.8016 0.714 0.244 0.756
#> GSM97149 2 0.8016 0.714 0.244 0.756
#> GSM97068 2 0.0000 0.942 0.000 1.000
#> GSM97071 1 0.8443 0.743 0.728 0.272
#> GSM97086 2 0.0000 0.942 0.000 1.000
#> GSM97103 2 0.0000 0.942 0.000 1.000
#> GSM97057 2 0.0000 0.942 0.000 1.000
#> GSM97060 2 0.0376 0.939 0.004 0.996
#> GSM97075 1 0.8661 0.725 0.712 0.288
#> GSM97098 2 0.0376 0.939 0.004 0.996
#> GSM97099 2 0.0000 0.942 0.000 1.000
#> GSM97101 2 0.0000 0.942 0.000 1.000
#> GSM97105 2 0.0000 0.942 0.000 1.000
#> GSM97106 2 0.0376 0.939 0.004 0.996
#> GSM97121 2 0.0000 0.942 0.000 1.000
#> GSM97128 1 0.0000 0.892 1.000 0.000
#> GSM97131 2 0.0000 0.942 0.000 1.000
#> GSM97137 2 0.8016 0.714 0.244 0.756
#> GSM97118 1 0.0000 0.892 1.000 0.000
#> GSM97114 2 0.2948 0.902 0.052 0.948
#> GSM97142 1 0.0000 0.892 1.000 0.000
#> GSM97140 2 0.0000 0.942 0.000 1.000
#> GSM97141 2 0.0000 0.942 0.000 1.000
#> GSM97055 1 0.0000 0.892 1.000 0.000
#> GSM97090 2 0.0000 0.942 0.000 1.000
#> GSM97091 1 0.0000 0.892 1.000 0.000
#> GSM97148 1 0.3431 0.865 0.936 0.064
#> GSM97063 1 0.0000 0.892 1.000 0.000
#> GSM97053 1 0.0672 0.891 0.992 0.008
#> GSM97066 1 0.7056 0.816 0.808 0.192
#> GSM97079 2 0.0000 0.942 0.000 1.000
#> GSM97083 1 0.0000 0.892 1.000 0.000
#> GSM97084 2 0.0000 0.942 0.000 1.000
#> GSM97094 1 0.3431 0.865 0.936 0.064
#> GSM97096 1 0.8016 0.775 0.756 0.244
#> GSM97097 2 0.0000 0.942 0.000 1.000
#> GSM97107 2 0.1414 0.928 0.020 0.980
#> GSM97054 2 0.0000 0.942 0.000 1.000
#> GSM97062 2 0.0000 0.942 0.000 1.000
#> GSM97069 1 0.7602 0.796 0.780 0.220
#> GSM97070 1 0.7056 0.816 0.808 0.192
#> GSM97073 1 0.7056 0.816 0.808 0.192
#> GSM97076 1 0.0376 0.892 0.996 0.004
#> GSM97077 2 0.0000 0.942 0.000 1.000
#> GSM97095 2 0.8909 0.486 0.308 0.692
#> GSM97102 1 0.7602 0.796 0.780 0.220
#> GSM97109 2 0.0000 0.942 0.000 1.000
#> GSM97110 2 0.0000 0.942 0.000 1.000
#> GSM97074 1 0.0000 0.892 1.000 0.000
#> GSM97085 1 0.0000 0.892 1.000 0.000
#> GSM97059 2 0.0000 0.942 0.000 1.000
#> GSM97072 2 0.1843 0.919 0.028 0.972
#> GSM97078 1 0.6973 0.818 0.812 0.188
#> GSM97067 1 0.7056 0.816 0.808 0.192
#> GSM97087 1 0.8016 0.775 0.756 0.244
#> GSM97111 1 0.9460 0.600 0.636 0.364
#> GSM97064 2 0.0000 0.942 0.000 1.000
#> GSM97065 1 0.7139 0.815 0.804 0.196
#> GSM97081 1 0.8016 0.775 0.756 0.244
#> GSM97082 1 0.8016 0.775 0.756 0.244
#> GSM97088 1 0.0000 0.892 1.000 0.000
#> GSM97100 2 0.0000 0.942 0.000 1.000
#> GSM97104 1 0.8016 0.775 0.756 0.244
#> GSM97108 2 0.0000 0.942 0.000 1.000
#> GSM97050 2 0.0000 0.942 0.000 1.000
#> GSM97080 1 0.8016 0.775 0.756 0.244
#> GSM97089 1 0.8016 0.775 0.756 0.244
#> GSM97092 2 0.0376 0.939 0.004 0.996
#> GSM97093 2 0.0376 0.939 0.004 0.996
#> GSM97058 2 0.0000 0.942 0.000 1.000
#> GSM97051 2 0.0000 0.942 0.000 1.000
#> GSM97052 2 0.0376 0.939 0.004 0.996
#> GSM97061 2 0.0376 0.939 0.004 0.996
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM97138 1 0.0237 0.873 0.996 0.000 0.004
#> GSM97145 1 0.0000 0.874 1.000 0.000 0.000
#> GSM97147 2 0.0237 0.976 0.004 0.996 0.000
#> GSM97125 1 0.0000 0.874 1.000 0.000 0.000
#> GSM97127 1 0.2066 0.857 0.940 0.060 0.000
#> GSM97130 1 0.5327 0.695 0.728 0.272 0.000
#> GSM97133 1 0.4235 0.795 0.824 0.176 0.000
#> GSM97134 1 0.3816 0.813 0.852 0.148 0.000
#> GSM97120 1 0.0000 0.874 1.000 0.000 0.000
#> GSM97126 1 0.0000 0.874 1.000 0.000 0.000
#> GSM97112 1 0.3412 0.778 0.876 0.000 0.124
#> GSM97115 2 0.0237 0.976 0.004 0.996 0.000
#> GSM97116 1 0.0000 0.874 1.000 0.000 0.000
#> GSM97117 1 0.0424 0.872 0.992 0.000 0.008
#> GSM97119 1 0.0237 0.873 0.996 0.000 0.004
#> GSM97122 1 0.0237 0.873 0.996 0.000 0.004
#> GSM97135 1 0.0237 0.873 0.996 0.000 0.004
#> GSM97136 3 0.1031 0.886 0.024 0.000 0.976
#> GSM97139 1 0.0000 0.874 1.000 0.000 0.000
#> GSM97146 1 0.0000 0.874 1.000 0.000 0.000
#> GSM97123 2 0.0424 0.973 0.000 0.992 0.008
#> GSM97129 1 0.4346 0.788 0.816 0.184 0.000
#> GSM97143 1 0.2878 0.807 0.904 0.000 0.096
#> GSM97113 2 0.0000 0.977 0.000 1.000 0.000
#> GSM97056 1 0.2878 0.842 0.904 0.096 0.000
#> GSM97124 1 0.0000 0.874 1.000 0.000 0.000
#> GSM97132 1 0.0000 0.874 1.000 0.000 0.000
#> GSM97144 1 0.5327 0.695 0.728 0.272 0.000
#> GSM97149 1 0.3192 0.834 0.888 0.112 0.000
#> GSM97068 2 0.0237 0.976 0.004 0.996 0.000
#> GSM97071 3 0.6180 0.622 0.260 0.024 0.716
#> GSM97086 2 0.0000 0.977 0.000 1.000 0.000
#> GSM97103 2 0.0747 0.966 0.000 0.984 0.016
#> GSM97057 2 0.0237 0.976 0.004 0.996 0.000
#> GSM97060 3 0.5882 0.482 0.000 0.348 0.652
#> GSM97075 3 0.6264 0.630 0.256 0.028 0.716
#> GSM97098 2 0.1031 0.960 0.000 0.976 0.024
#> GSM97099 2 0.0747 0.966 0.000 0.984 0.016
#> GSM97101 2 0.0237 0.976 0.004 0.996 0.000
#> GSM97105 2 0.0000 0.977 0.000 1.000 0.000
#> GSM97106 2 0.0424 0.973 0.000 0.992 0.008
#> GSM97121 2 0.0237 0.976 0.004 0.996 0.000
#> GSM97128 3 0.4062 0.784 0.164 0.000 0.836
#> GSM97131 2 0.0000 0.977 0.000 1.000 0.000
#> GSM97137 1 0.5327 0.695 0.728 0.272 0.000
#> GSM97118 1 0.6215 0.164 0.572 0.000 0.428
#> GSM97114 1 0.5216 0.712 0.740 0.260 0.000
#> GSM97142 1 0.1860 0.844 0.948 0.000 0.052
#> GSM97140 2 0.0237 0.976 0.004 0.996 0.000
#> GSM97141 2 0.0237 0.976 0.004 0.996 0.000
#> GSM97055 3 0.5733 0.565 0.324 0.000 0.676
#> GSM97090 2 0.0237 0.976 0.004 0.996 0.000
#> GSM97091 3 0.5760 0.558 0.328 0.000 0.672
#> GSM97148 1 0.0000 0.874 1.000 0.000 0.000
#> GSM97063 1 0.5327 0.558 0.728 0.000 0.272
#> GSM97053 1 0.0000 0.874 1.000 0.000 0.000
#> GSM97066 3 0.0475 0.889 0.004 0.004 0.992
#> GSM97079 2 0.0000 0.977 0.000 1.000 0.000
#> GSM97083 3 0.4291 0.768 0.180 0.000 0.820
#> GSM97084 2 0.0000 0.977 0.000 1.000 0.000
#> GSM97094 1 0.0000 0.874 1.000 0.000 0.000
#> GSM97096 3 0.0829 0.889 0.012 0.004 0.984
#> GSM97097 2 0.0000 0.977 0.000 1.000 0.000
#> GSM97107 2 0.3551 0.824 0.132 0.868 0.000
#> GSM97054 2 0.0000 0.977 0.000 1.000 0.000
#> GSM97062 2 0.0000 0.977 0.000 1.000 0.000
#> GSM97069 3 0.0237 0.888 0.000 0.004 0.996
#> GSM97070 3 0.0475 0.889 0.004 0.004 0.992
#> GSM97073 3 0.0475 0.889 0.004 0.004 0.992
#> GSM97076 1 0.3816 0.763 0.852 0.000 0.148
#> GSM97077 2 0.0000 0.977 0.000 1.000 0.000
#> GSM97095 1 0.6019 0.661 0.700 0.288 0.012
#> GSM97102 3 0.0237 0.888 0.000 0.004 0.996
#> GSM97109 2 0.5363 0.561 0.276 0.724 0.000
#> GSM97110 2 0.0000 0.977 0.000 1.000 0.000
#> GSM97074 3 0.1031 0.883 0.024 0.000 0.976
#> GSM97085 3 0.1289 0.882 0.032 0.000 0.968
#> GSM97059 2 0.0237 0.976 0.004 0.996 0.000
#> GSM97072 3 0.1964 0.857 0.000 0.056 0.944
#> GSM97078 3 0.1399 0.886 0.028 0.004 0.968
#> GSM97067 3 0.0475 0.889 0.004 0.004 0.992
#> GSM97087 3 0.0829 0.889 0.012 0.004 0.984
#> GSM97111 1 0.5268 0.760 0.776 0.212 0.012
#> GSM97064 2 0.0237 0.975 0.000 0.996 0.004
#> GSM97065 3 0.5443 0.640 0.260 0.004 0.736
#> GSM97081 3 0.0829 0.889 0.012 0.004 0.984
#> GSM97082 3 0.0829 0.889 0.012 0.004 0.984
#> GSM97088 3 0.0747 0.888 0.016 0.000 0.984
#> GSM97100 2 0.0000 0.977 0.000 1.000 0.000
#> GSM97104 3 0.0237 0.888 0.000 0.004 0.996
#> GSM97108 2 0.0237 0.976 0.004 0.996 0.000
#> GSM97050 2 0.0000 0.977 0.000 1.000 0.000
#> GSM97080 3 0.0237 0.888 0.000 0.004 0.996
#> GSM97089 3 0.0829 0.889 0.012 0.004 0.984
#> GSM97092 3 0.5882 0.482 0.000 0.348 0.652
#> GSM97093 2 0.0747 0.968 0.000 0.984 0.016
#> GSM97058 2 0.0000 0.977 0.000 1.000 0.000
#> GSM97051 2 0.0000 0.977 0.000 1.000 0.000
#> GSM97052 2 0.4346 0.753 0.000 0.816 0.184
#> GSM97061 2 0.0424 0.973 0.000 0.992 0.008
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM97138 1 0.4907 0.0550 0.580 0.000 0.000 0.420
#> GSM97145 1 0.0817 0.7171 0.976 0.000 0.000 0.024
#> GSM97147 2 0.5936 0.5572 0.324 0.620 0.056 0.000
#> GSM97125 1 0.2921 0.6400 0.860 0.000 0.000 0.140
#> GSM97127 1 0.1022 0.7243 0.968 0.032 0.000 0.000
#> GSM97130 1 0.3743 0.6707 0.824 0.160 0.000 0.016
#> GSM97133 1 0.2814 0.6933 0.868 0.132 0.000 0.000
#> GSM97134 1 0.3190 0.7054 0.880 0.096 0.008 0.016
#> GSM97120 1 0.0817 0.7171 0.976 0.000 0.000 0.024
#> GSM97126 1 0.3311 0.6006 0.828 0.000 0.000 0.172
#> GSM97112 4 0.4955 0.2893 0.444 0.000 0.000 0.556
#> GSM97115 2 0.5997 0.4830 0.368 0.592 0.028 0.012
#> GSM97116 1 0.3123 0.6183 0.844 0.000 0.000 0.156
#> GSM97117 1 0.2401 0.6931 0.904 0.000 0.092 0.004
#> GSM97119 1 0.4907 0.0550 0.580 0.000 0.000 0.420
#> GSM97122 1 0.4907 0.0550 0.580 0.000 0.000 0.420
#> GSM97135 1 0.4907 0.0550 0.580 0.000 0.000 0.420
#> GSM97136 4 0.5755 -0.1691 0.028 0.000 0.444 0.528
#> GSM97139 1 0.1474 0.7072 0.948 0.000 0.000 0.052
#> GSM97146 1 0.3123 0.6183 0.844 0.000 0.000 0.156
#> GSM97123 2 0.4277 0.5673 0.000 0.720 0.280 0.000
#> GSM97129 1 0.4617 0.6676 0.812 0.100 0.080 0.008
#> GSM97143 1 0.4999 -0.1940 0.508 0.000 0.000 0.492
#> GSM97113 2 0.0000 0.7861 0.000 1.000 0.000 0.000
#> GSM97056 1 0.1576 0.7237 0.948 0.048 0.000 0.004
#> GSM97124 1 0.1389 0.7108 0.952 0.000 0.000 0.048
#> GSM97132 1 0.1557 0.7120 0.944 0.000 0.000 0.056
#> GSM97144 1 0.3743 0.6707 0.824 0.160 0.000 0.016
#> GSM97149 1 0.1557 0.7229 0.944 0.056 0.000 0.000
#> GSM97068 2 0.1042 0.7839 0.020 0.972 0.000 0.008
#> GSM97071 3 0.7795 0.2626 0.264 0.004 0.468 0.264
#> GSM97086 2 0.0336 0.7849 0.000 0.992 0.000 0.008
#> GSM97103 2 0.5926 0.5512 0.060 0.632 0.308 0.000
#> GSM97057 2 0.0000 0.7861 0.000 1.000 0.000 0.000
#> GSM97060 3 0.4901 0.5207 0.000 0.112 0.780 0.108
#> GSM97075 3 0.8004 0.2174 0.368 0.044 0.472 0.116
#> GSM97098 2 0.4996 0.2335 0.000 0.516 0.484 0.000
#> GSM97099 2 0.7453 0.4658 0.300 0.496 0.204 0.000
#> GSM97101 2 0.2816 0.7687 0.036 0.900 0.064 0.000
#> GSM97105 2 0.0000 0.7861 0.000 1.000 0.000 0.000
#> GSM97106 2 0.4103 0.5780 0.000 0.744 0.256 0.000
#> GSM97121 2 0.5491 0.6373 0.260 0.688 0.052 0.000
#> GSM97128 4 0.4938 0.5995 0.148 0.000 0.080 0.772
#> GSM97131 2 0.0000 0.7861 0.000 1.000 0.000 0.000
#> GSM97137 1 0.3743 0.6707 0.824 0.160 0.000 0.016
#> GSM97118 4 0.4746 0.5514 0.304 0.000 0.008 0.688
#> GSM97114 1 0.4979 0.6030 0.760 0.176 0.064 0.000
#> GSM97142 4 0.4994 0.1852 0.480 0.000 0.000 0.520
#> GSM97140 2 0.6033 0.5641 0.316 0.620 0.064 0.000
#> GSM97141 2 0.6182 0.5714 0.308 0.616 0.076 0.000
#> GSM97055 4 0.4098 0.6414 0.204 0.000 0.012 0.784
#> GSM97090 2 0.6616 0.2694 0.456 0.480 0.052 0.012
#> GSM97091 4 0.4098 0.6414 0.204 0.000 0.012 0.784
#> GSM97148 1 0.1389 0.7095 0.952 0.000 0.000 0.048
#> GSM97063 4 0.4428 0.6041 0.276 0.000 0.004 0.720
#> GSM97053 1 0.2469 0.6675 0.892 0.000 0.000 0.108
#> GSM97066 3 0.4564 0.4858 0.000 0.000 0.672 0.328
#> GSM97079 2 0.0336 0.7849 0.000 0.992 0.000 0.008
#> GSM97083 4 0.5032 0.6059 0.156 0.000 0.080 0.764
#> GSM97084 2 0.0336 0.7849 0.000 0.992 0.000 0.008
#> GSM97094 1 0.2188 0.7175 0.936 0.012 0.032 0.020
#> GSM97096 3 0.3668 0.5424 0.000 0.004 0.808 0.188
#> GSM97097 2 0.0336 0.7849 0.000 0.992 0.000 0.008
#> GSM97107 2 0.6611 0.2335 0.460 0.480 0.040 0.020
#> GSM97054 2 0.0336 0.7849 0.000 0.992 0.000 0.008
#> GSM97062 2 0.0336 0.7849 0.000 0.992 0.000 0.008
#> GSM97069 3 0.4500 0.4925 0.000 0.000 0.684 0.316
#> GSM97070 3 0.4564 0.4858 0.000 0.000 0.672 0.328
#> GSM97073 3 0.4477 0.4881 0.000 0.000 0.688 0.312
#> GSM97076 4 0.7550 -0.0905 0.192 0.000 0.372 0.436
#> GSM97077 2 0.2282 0.7742 0.024 0.924 0.052 0.000
#> GSM97095 1 0.5711 0.6116 0.744 0.140 0.100 0.016
#> GSM97102 3 0.4356 0.5369 0.000 0.000 0.708 0.292
#> GSM97109 1 0.6835 0.0272 0.540 0.360 0.096 0.004
#> GSM97110 2 0.5792 0.6829 0.168 0.708 0.124 0.000
#> GSM97074 3 0.4998 0.2595 0.000 0.000 0.512 0.488
#> GSM97085 4 0.4857 0.2530 0.016 0.000 0.284 0.700
#> GSM97059 2 0.5973 0.5463 0.332 0.612 0.056 0.000
#> GSM97072 3 0.4122 0.5123 0.000 0.004 0.760 0.236
#> GSM97078 3 0.8017 0.3778 0.124 0.048 0.512 0.316
#> GSM97067 3 0.4564 0.4858 0.000 0.000 0.672 0.328
#> GSM97087 3 0.4277 0.5427 0.000 0.000 0.720 0.280
#> GSM97111 1 0.5106 0.6426 0.780 0.112 0.100 0.008
#> GSM97064 2 0.1474 0.7716 0.000 0.948 0.052 0.000
#> GSM97065 3 0.7458 0.2969 0.240 0.000 0.508 0.252
#> GSM97081 3 0.3810 0.5415 0.000 0.008 0.804 0.188
#> GSM97082 3 0.4277 0.5427 0.000 0.000 0.720 0.280
#> GSM97088 4 0.4857 0.1628 0.008 0.000 0.324 0.668
#> GSM97100 2 0.0000 0.7861 0.000 1.000 0.000 0.000
#> GSM97104 3 0.4304 0.5419 0.000 0.000 0.716 0.284
#> GSM97108 2 0.4514 0.7270 0.148 0.796 0.056 0.000
#> GSM97050 2 0.0000 0.7861 0.000 1.000 0.000 0.000
#> GSM97080 3 0.4304 0.5419 0.000 0.000 0.716 0.284
#> GSM97089 3 0.4250 0.5441 0.000 0.000 0.724 0.276
#> GSM97092 3 0.4728 0.5228 0.000 0.104 0.792 0.104
#> GSM97093 3 0.9359 0.0544 0.300 0.220 0.376 0.104
#> GSM97058 2 0.0188 0.7855 0.000 0.996 0.004 0.000
#> GSM97051 2 0.0000 0.7861 0.000 1.000 0.000 0.000
#> GSM97052 3 0.6712 0.2302 0.000 0.344 0.552 0.104
#> GSM97061 2 0.4304 0.5615 0.000 0.716 0.284 0.000
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM97138 5 0.2719 0.5081 0.144 0.000 0.000 0.004 0.852
#> GSM97145 1 0.4211 0.4733 0.636 0.000 0.000 0.004 0.360
#> GSM97147 1 0.4325 0.3958 0.684 0.300 0.012 0.004 0.000
#> GSM97125 5 0.4449 -0.2256 0.484 0.000 0.000 0.004 0.512
#> GSM97127 1 0.3884 0.5531 0.708 0.000 0.000 0.004 0.288
#> GSM97130 1 0.3099 0.6320 0.848 0.008 0.000 0.012 0.132
#> GSM97133 1 0.3790 0.5655 0.724 0.000 0.000 0.004 0.272
#> GSM97134 1 0.2818 0.6321 0.856 0.000 0.000 0.012 0.132
#> GSM97120 1 0.4251 0.4563 0.624 0.000 0.000 0.004 0.372
#> GSM97126 5 0.4437 -0.1788 0.464 0.000 0.004 0.000 0.532
#> GSM97112 5 0.3379 0.5834 0.016 0.000 0.008 0.148 0.828
#> GSM97115 1 0.4536 0.4110 0.656 0.324 0.004 0.016 0.000
#> GSM97116 5 0.4443 -0.1924 0.472 0.000 0.000 0.004 0.524
#> GSM97117 1 0.1956 0.6415 0.928 0.008 0.012 0.000 0.052
#> GSM97119 5 0.2719 0.5081 0.144 0.000 0.000 0.004 0.852
#> GSM97122 5 0.2561 0.5044 0.144 0.000 0.000 0.000 0.856
#> GSM97135 5 0.2719 0.5081 0.144 0.000 0.000 0.004 0.852
#> GSM97136 3 0.5992 -0.1171 0.000 0.000 0.472 0.112 0.416
#> GSM97139 1 0.4375 0.3740 0.576 0.000 0.000 0.004 0.420
#> GSM97146 5 0.4443 -0.1955 0.472 0.000 0.000 0.004 0.524
#> GSM97123 3 0.5974 0.3641 0.100 0.380 0.516 0.004 0.000
#> GSM97129 1 0.2302 0.6453 0.904 0.008 0.008 0.000 0.080
#> GSM97143 5 0.2871 0.5883 0.040 0.000 0.000 0.088 0.872
#> GSM97113 2 0.0000 0.8446 0.000 1.000 0.000 0.000 0.000
#> GSM97056 1 0.3849 0.5761 0.752 0.000 0.000 0.016 0.232
#> GSM97124 1 0.4658 0.3981 0.576 0.000 0.000 0.016 0.408
#> GSM97132 1 0.4430 0.4382 0.628 0.000 0.000 0.012 0.360
#> GSM97144 1 0.3099 0.6320 0.848 0.008 0.000 0.012 0.132
#> GSM97149 1 0.3838 0.5640 0.716 0.000 0.000 0.004 0.280
#> GSM97068 2 0.1270 0.8201 0.052 0.948 0.000 0.000 0.000
#> GSM97071 4 0.4281 0.7034 0.204 0.000 0.028 0.756 0.012
#> GSM97086 2 0.0451 0.8426 0.008 0.988 0.000 0.004 0.000
#> GSM97103 2 0.6842 0.1941 0.360 0.392 0.244 0.004 0.000
#> GSM97057 2 0.0000 0.8446 0.000 1.000 0.000 0.000 0.000
#> GSM97060 3 0.3396 0.6816 0.112 0.036 0.844 0.008 0.000
#> GSM97075 1 0.4633 0.2514 0.632 0.016 0.348 0.004 0.000
#> GSM97098 3 0.5335 0.6045 0.112 0.208 0.676 0.004 0.000
#> GSM97099 1 0.5759 0.3811 0.636 0.188 0.172 0.004 0.000
#> GSM97101 2 0.4500 0.5670 0.316 0.664 0.016 0.004 0.000
#> GSM97105 2 0.0000 0.8446 0.000 1.000 0.000 0.000 0.000
#> GSM97106 3 0.5814 0.2533 0.092 0.436 0.472 0.000 0.000
#> GSM97121 1 0.4661 0.2608 0.624 0.356 0.016 0.004 0.000
#> GSM97128 5 0.6606 0.4512 0.028 0.000 0.232 0.172 0.568
#> GSM97131 2 0.0162 0.8439 0.000 0.996 0.000 0.004 0.000
#> GSM97137 1 0.3099 0.6320 0.848 0.008 0.000 0.012 0.132
#> GSM97118 5 0.5430 0.5465 0.008 0.000 0.144 0.164 0.684
#> GSM97114 1 0.2569 0.6509 0.896 0.032 0.004 0.000 0.068
#> GSM97142 5 0.2625 0.5875 0.016 0.000 0.000 0.108 0.876
#> GSM97140 1 0.4359 0.4011 0.692 0.288 0.016 0.004 0.000
#> GSM97141 1 0.4359 0.4011 0.692 0.288 0.016 0.004 0.000
#> GSM97055 5 0.5530 0.5319 0.004 0.000 0.160 0.172 0.664
#> GSM97090 1 0.3504 0.5770 0.816 0.160 0.008 0.016 0.000
#> GSM97091 5 0.5494 0.5353 0.004 0.000 0.156 0.172 0.668
#> GSM97148 1 0.4367 0.3894 0.580 0.000 0.000 0.004 0.416
#> GSM97063 5 0.5434 0.5461 0.008 0.000 0.152 0.156 0.684
#> GSM97053 1 0.4437 0.2791 0.532 0.000 0.000 0.004 0.464
#> GSM97066 4 0.2690 0.8867 0.000 0.000 0.156 0.844 0.000
#> GSM97079 2 0.0451 0.8426 0.008 0.988 0.000 0.004 0.000
#> GSM97083 5 0.6606 0.4512 0.028 0.000 0.232 0.172 0.568
#> GSM97084 2 0.0451 0.8426 0.008 0.988 0.000 0.004 0.000
#> GSM97094 1 0.2407 0.6411 0.896 0.000 0.004 0.012 0.088
#> GSM97096 3 0.2624 0.6790 0.116 0.000 0.872 0.012 0.000
#> GSM97097 2 0.0451 0.8426 0.008 0.988 0.000 0.004 0.000
#> GSM97107 1 0.3988 0.5185 0.732 0.252 0.000 0.016 0.000
#> GSM97054 2 0.0451 0.8426 0.008 0.988 0.000 0.004 0.000
#> GSM97062 2 0.0451 0.8426 0.008 0.988 0.000 0.004 0.000
#> GSM97069 4 0.2852 0.8777 0.000 0.000 0.172 0.828 0.000
#> GSM97070 4 0.2690 0.8867 0.000 0.000 0.156 0.844 0.000
#> GSM97073 4 0.3362 0.8858 0.008 0.000 0.156 0.824 0.012
#> GSM97076 4 0.4274 0.7388 0.032 0.000 0.020 0.776 0.172
#> GSM97077 2 0.4843 0.5949 0.276 0.676 0.044 0.004 0.000
#> GSM97095 1 0.1439 0.6348 0.956 0.004 0.020 0.016 0.004
#> GSM97102 3 0.2124 0.6214 0.000 0.000 0.900 0.096 0.004
#> GSM97109 1 0.2984 0.6147 0.856 0.124 0.016 0.004 0.000
#> GSM97110 2 0.6388 0.2156 0.424 0.428 0.144 0.004 0.000
#> GSM97074 4 0.2824 0.8626 0.000 0.000 0.116 0.864 0.020
#> GSM97085 5 0.6497 0.3027 0.000 0.000 0.312 0.212 0.476
#> GSM97059 1 0.4283 0.4064 0.692 0.292 0.012 0.004 0.000
#> GSM97072 4 0.3318 0.8668 0.008 0.000 0.192 0.800 0.000
#> GSM97078 1 0.7613 0.0199 0.464 0.004 0.300 0.076 0.156
#> GSM97067 4 0.2690 0.8867 0.000 0.000 0.156 0.844 0.000
#> GSM97087 3 0.1544 0.6464 0.000 0.000 0.932 0.068 0.000
#> GSM97111 1 0.2376 0.6356 0.916 0.012 0.024 0.004 0.044
#> GSM97064 2 0.3567 0.7164 0.092 0.836 0.068 0.004 0.000
#> GSM97065 4 0.4870 0.7858 0.140 0.000 0.104 0.744 0.012
#> GSM97081 3 0.2624 0.6790 0.116 0.000 0.872 0.012 0.000
#> GSM97082 3 0.1544 0.6464 0.000 0.000 0.932 0.068 0.000
#> GSM97088 5 0.6545 0.2797 0.000 0.000 0.324 0.216 0.460
#> GSM97100 2 0.0000 0.8446 0.000 1.000 0.000 0.000 0.000
#> GSM97104 3 0.1792 0.6339 0.000 0.000 0.916 0.084 0.000
#> GSM97108 2 0.4877 0.2586 0.456 0.524 0.016 0.004 0.000
#> GSM97050 2 0.0000 0.8446 0.000 1.000 0.000 0.000 0.000
#> GSM97080 3 0.1965 0.6305 0.000 0.000 0.904 0.096 0.000
#> GSM97089 3 0.1544 0.6464 0.000 0.000 0.932 0.068 0.000
#> GSM97092 3 0.2915 0.6814 0.116 0.024 0.860 0.000 0.000
#> GSM97093 3 0.4812 0.5275 0.312 0.032 0.652 0.004 0.000
#> GSM97058 2 0.0963 0.8230 0.036 0.964 0.000 0.000 0.000
#> GSM97051 2 0.0000 0.8446 0.000 1.000 0.000 0.000 0.000
#> GSM97052 3 0.4277 0.6684 0.112 0.100 0.784 0.004 0.000
#> GSM97061 3 0.5974 0.3641 0.100 0.380 0.516 0.004 0.000
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM97138 1 0.3934 0.635 0.728 0.012 0.000 0.000 0.240 0.020
#> GSM97145 1 0.2006 0.839 0.892 0.104 0.000 0.000 0.004 0.000
#> GSM97147 2 0.3104 0.827 0.028 0.852 0.028 0.092 0.000 0.000
#> GSM97125 1 0.2100 0.837 0.916 0.036 0.000 0.000 0.032 0.016
#> GSM97127 1 0.2730 0.771 0.808 0.192 0.000 0.000 0.000 0.000
#> GSM97130 2 0.5057 0.625 0.280 0.648 0.000 0.024 0.024 0.024
#> GSM97133 1 0.2823 0.757 0.796 0.204 0.000 0.000 0.000 0.000
#> GSM97134 2 0.4505 0.683 0.240 0.704 0.000 0.008 0.024 0.024
#> GSM97120 1 0.2006 0.839 0.892 0.104 0.000 0.000 0.004 0.000
#> GSM97126 1 0.2772 0.818 0.876 0.036 0.000 0.000 0.068 0.020
#> GSM97112 5 0.2070 0.866 0.092 0.000 0.000 0.000 0.896 0.012
#> GSM97115 2 0.4319 0.806 0.080 0.796 0.008 0.076 0.024 0.016
#> GSM97116 1 0.1851 0.838 0.928 0.024 0.000 0.000 0.036 0.012
#> GSM97117 2 0.2822 0.811 0.108 0.852 0.040 0.000 0.000 0.000
#> GSM97119 1 0.3876 0.635 0.728 0.012 0.000 0.000 0.244 0.016
#> GSM97122 1 0.3636 0.682 0.764 0.012 0.000 0.000 0.208 0.016
#> GSM97135 1 0.3876 0.635 0.728 0.012 0.000 0.000 0.244 0.016
#> GSM97136 5 0.4346 0.662 0.008 0.016 0.240 0.000 0.712 0.024
#> GSM97139 1 0.1644 0.846 0.920 0.076 0.000 0.000 0.004 0.000
#> GSM97146 1 0.1498 0.841 0.940 0.028 0.000 0.000 0.032 0.000
#> GSM97123 3 0.4341 0.739 0.008 0.080 0.732 0.180 0.000 0.000
#> GSM97129 2 0.2869 0.795 0.148 0.832 0.020 0.000 0.000 0.000
#> GSM97143 5 0.3831 0.716 0.224 0.012 0.000 0.000 0.744 0.020
#> GSM97113 4 0.0632 0.968 0.000 0.024 0.000 0.976 0.000 0.000
#> GSM97056 1 0.3971 0.679 0.748 0.208 0.000 0.000 0.024 0.020
#> GSM97124 1 0.1779 0.842 0.920 0.064 0.000 0.000 0.000 0.016
#> GSM97132 1 0.3425 0.789 0.824 0.120 0.000 0.000 0.024 0.032
#> GSM97144 2 0.5038 0.631 0.276 0.652 0.000 0.024 0.024 0.024
#> GSM97149 1 0.2793 0.762 0.800 0.200 0.000 0.000 0.000 0.000
#> GSM97068 4 0.0146 0.970 0.000 0.004 0.000 0.996 0.000 0.000
#> GSM97071 6 0.2086 0.888 0.008 0.064 0.012 0.000 0.004 0.912
#> GSM97086 4 0.0000 0.972 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM97103 2 0.3838 0.753 0.004 0.784 0.096 0.116 0.000 0.000
#> GSM97057 4 0.0632 0.968 0.000 0.024 0.000 0.976 0.000 0.000
#> GSM97060 3 0.1265 0.847 0.008 0.044 0.948 0.000 0.000 0.000
#> GSM97075 2 0.2257 0.783 0.008 0.876 0.116 0.000 0.000 0.000
#> GSM97098 3 0.2911 0.819 0.008 0.100 0.856 0.036 0.000 0.000
#> GSM97099 2 0.2649 0.807 0.004 0.876 0.068 0.052 0.000 0.000
#> GSM97101 2 0.4338 0.683 0.012 0.700 0.040 0.248 0.000 0.000
#> GSM97105 4 0.0547 0.970 0.000 0.020 0.000 0.980 0.000 0.000
#> GSM97106 3 0.4340 0.708 0.008 0.060 0.716 0.216 0.000 0.000
#> GSM97121 2 0.3123 0.818 0.012 0.840 0.032 0.116 0.000 0.000
#> GSM97128 5 0.1337 0.877 0.012 0.008 0.016 0.000 0.956 0.008
#> GSM97131 4 0.0363 0.973 0.000 0.012 0.000 0.988 0.000 0.000
#> GSM97137 2 0.5076 0.629 0.284 0.644 0.000 0.024 0.024 0.024
#> GSM97118 5 0.1951 0.885 0.060 0.004 0.000 0.000 0.916 0.020
#> GSM97114 2 0.2859 0.793 0.156 0.828 0.016 0.000 0.000 0.000
#> GSM97142 5 0.2877 0.801 0.168 0.000 0.000 0.000 0.820 0.012
#> GSM97140 2 0.3179 0.826 0.028 0.848 0.032 0.092 0.000 0.000
#> GSM97141 2 0.3025 0.828 0.028 0.860 0.032 0.080 0.000 0.000
#> GSM97055 5 0.1333 0.889 0.048 0.000 0.008 0.000 0.944 0.000
#> GSM97090 2 0.4127 0.796 0.120 0.800 0.008 0.028 0.024 0.020
#> GSM97091 5 0.1333 0.889 0.048 0.000 0.008 0.000 0.944 0.000
#> GSM97148 1 0.1644 0.846 0.920 0.076 0.000 0.000 0.004 0.000
#> GSM97063 5 0.1411 0.887 0.060 0.000 0.004 0.000 0.936 0.000
#> GSM97053 1 0.1341 0.845 0.948 0.028 0.000 0.000 0.024 0.000
#> GSM97066 6 0.1082 0.958 0.000 0.000 0.040 0.000 0.004 0.956
#> GSM97079 4 0.0000 0.972 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM97083 5 0.1337 0.877 0.012 0.008 0.016 0.000 0.956 0.008
#> GSM97084 4 0.0000 0.972 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM97094 2 0.3994 0.782 0.124 0.804 0.008 0.012 0.028 0.024
#> GSM97096 3 0.1900 0.846 0.000 0.068 0.916 0.000 0.008 0.008
#> GSM97097 4 0.0000 0.972 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM97107 2 0.3925 0.794 0.104 0.812 0.000 0.036 0.024 0.024
#> GSM97054 4 0.0000 0.972 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM97062 4 0.0000 0.972 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM97069 6 0.2203 0.916 0.004 0.000 0.084 0.000 0.016 0.896
#> GSM97070 6 0.1082 0.958 0.000 0.000 0.040 0.000 0.004 0.956
#> GSM97073 6 0.1484 0.958 0.004 0.004 0.040 0.000 0.008 0.944
#> GSM97076 6 0.1850 0.902 0.052 0.016 0.000 0.000 0.008 0.924
#> GSM97077 2 0.4282 0.621 0.000 0.656 0.040 0.304 0.000 0.000
#> GSM97095 2 0.3480 0.807 0.072 0.852 0.016 0.012 0.028 0.020
#> GSM97102 3 0.2907 0.811 0.008 0.004 0.868 0.000 0.064 0.056
#> GSM97109 2 0.2541 0.829 0.052 0.892 0.024 0.032 0.000 0.000
#> GSM97110 2 0.3297 0.782 0.000 0.820 0.068 0.112 0.000 0.000
#> GSM97074 6 0.1370 0.956 0.004 0.000 0.036 0.000 0.012 0.948
#> GSM97085 5 0.2563 0.824 0.008 0.004 0.108 0.000 0.872 0.008
#> GSM97059 2 0.2589 0.833 0.028 0.888 0.024 0.060 0.000 0.000
#> GSM97072 6 0.1152 0.957 0.000 0.004 0.044 0.000 0.000 0.952
#> GSM97078 2 0.4642 0.776 0.052 0.792 0.052 0.016 0.052 0.036
#> GSM97067 6 0.1082 0.958 0.000 0.000 0.040 0.000 0.004 0.956
#> GSM97087 3 0.2656 0.819 0.008 0.004 0.884 0.000 0.060 0.044
#> GSM97111 2 0.2527 0.817 0.084 0.876 0.040 0.000 0.000 0.000
#> GSM97064 4 0.4024 0.741 0.008 0.092 0.128 0.772 0.000 0.000
#> GSM97065 6 0.2217 0.927 0.004 0.048 0.036 0.000 0.004 0.908
#> GSM97081 3 0.1957 0.846 0.000 0.072 0.912 0.000 0.008 0.008
#> GSM97082 3 0.2722 0.818 0.008 0.004 0.880 0.000 0.060 0.048
#> GSM97088 5 0.2912 0.815 0.008 0.008 0.112 0.000 0.856 0.016
#> GSM97100 4 0.0363 0.973 0.000 0.012 0.000 0.988 0.000 0.000
#> GSM97104 3 0.2848 0.814 0.008 0.004 0.872 0.000 0.060 0.056
#> GSM97108 2 0.3194 0.811 0.008 0.828 0.032 0.132 0.000 0.000
#> GSM97050 4 0.0547 0.970 0.000 0.020 0.000 0.980 0.000 0.000
#> GSM97080 3 0.2848 0.814 0.008 0.004 0.872 0.000 0.060 0.056
#> GSM97089 3 0.2394 0.824 0.008 0.004 0.900 0.000 0.052 0.036
#> GSM97092 3 0.1524 0.844 0.008 0.060 0.932 0.000 0.000 0.000
#> GSM97093 3 0.3470 0.688 0.012 0.248 0.740 0.000 0.000 0.000
#> GSM97058 4 0.0993 0.957 0.000 0.024 0.012 0.964 0.000 0.000
#> GSM97051 4 0.0363 0.973 0.000 0.012 0.000 0.988 0.000 0.000
#> GSM97052 3 0.2001 0.841 0.008 0.068 0.912 0.012 0.000 0.000
#> GSM97061 3 0.4341 0.739 0.008 0.080 0.732 0.180 0.000 0.000
Heatmaps for the consensus matrix. It visualizes the probability of two samples to be in a same group.
consensus_heatmap(res, k = 2)
consensus_heatmap(res, k = 3)
consensus_heatmap(res, k = 4)
consensus_heatmap(res, k = 5)
consensus_heatmap(res, k = 6)
Heatmaps for the membership of samples in all partitions to see how consistent they are:
membership_heatmap(res, k = 2)
membership_heatmap(res, k = 3)
membership_heatmap(res, k = 4)
membership_heatmap(res, k = 5)
membership_heatmap(res, k = 6)
As soon as we have had the classes for columns, we can look for signatures which are significantly different between classes which can be candidate marks for certain classes. Following are the heatmaps for signatures.
Signature heatmaps where rows are scaled:
get_signatures(res, k = 2)
get_signatures(res, k = 3)
get_signatures(res, k = 4)
get_signatures(res, k = 5)
get_signatures(res, k = 6)
Signature heatmaps where rows are not scaled:
get_signatures(res, k = 2, scale_rows = FALSE)
get_signatures(res, k = 3, scale_rows = FALSE)
get_signatures(res, k = 4, scale_rows = FALSE)
get_signatures(res, k = 5, scale_rows = FALSE)
get_signatures(res, k = 6, scale_rows = FALSE)
Compare the overlap of signatures from different k:
compare_signatures(res)
get_signature() returns a data frame invisibly. TO get the list of signatures, the function
call should be assigned to a variable explicitly. In following code, if plot argument is set
to FALSE, no heatmap is plotted while only the differential analysis is performed.
# code only for demonstration
tb = get_signature(res, k = ..., plot = FALSE)
An example of the output of tb is:
#> which_row fdr mean_1 mean_2 scaled_mean_1 scaled_mean_2 km
#> 1 38 0.042760348 8.373488 9.131774 -0.5533452 0.5164555 1
#> 2 40 0.018707592 7.106213 8.469186 -0.6173731 0.5762149 1
#> 3 55 0.019134737 10.221463 11.207825 -0.6159697 0.5749050 1
#> 4 59 0.006059896 5.921854 7.869574 -0.6899429 0.6439467 1
#> 5 60 0.018055526 8.928898 10.211722 -0.6204761 0.5791110 1
#> 6 98 0.009384629 15.714769 14.887706 0.6635654 -0.6193277 2
...
The columns in tb are:
which_row: row indices corresponding to the input matrix.fdr: FDR for the differential test. mean_x: The mean value in group x.scaled_mean_x: The mean value in group x after rows are scaled.km: Row groups if k-means clustering is applied to rows.UMAP plot which shows how samples are separated.
dimension_reduction(res, k = 2, method = "UMAP")
dimension_reduction(res, k = 3, method = "UMAP")
dimension_reduction(res, k = 4, method = "UMAP")
dimension_reduction(res, k = 5, method = "UMAP")
dimension_reduction(res, k = 6, method = "UMAP")
Following heatmap shows how subgroups are split when increasing k:
collect_classes(res)
Test correlation between subgroups and known annotations. If the known annotation is numeric, one-way ANOVA test is applied, and if the known annotation is discrete, chi-squared contingency table test is applied.
test_to_known_factors(res)
#> n disease.state(p) specimen(p) cell.type(p) other(p) k
#> ATC:kmeans 99 5.34e-02 0.0201 3.41e-02 0.0991 2
#> ATC:kmeans 97 9.51e-07 0.1400 2.13e-12 0.0897 3
#> ATC:kmeans 71 1.03e-03 0.3798 2.62e-09 0.2825 4
#> ATC:kmeans 69 7.22e-04 0.2582 9.31e-12 0.2071 5
#> ATC:kmeans 100 4.95e-05 0.6295 3.66e-13 0.3836 6
If matrix rows can be associated to genes, consider to use functional_enrichment(res,
...) to perform function enrichment for the signature genes. See this vignette for more detailed explanations.
The object with results only for a single top-value method and a single partition method can be extracted as:
res = res_list["ATC", "skmeans"]
# you can also extract it by
# res = res_list["ATC:skmeans"]
A summary of res and all the functions that can be applied to it:
res
#> A 'ConsensusPartition' object with k = 2, 3, 4, 5, 6.
#> On a matrix with 21512 rows and 100 columns.
#> Top rows (1000, 2000, 3000, 4000, 5000) are extracted by 'ATC' method.
#> Subgroups are detected by 'skmeans' method.
#> Performed in total 1250 partitions by row resampling.
#> Best k for subgroups seems to be 3.
#>
#> Following methods can be applied to this 'ConsensusPartition' object:
#> [1] "cola_report" "collect_classes" "collect_plots"
#> [4] "collect_stats" "colnames" "compare_signatures"
#> [7] "consensus_heatmap" "dimension_reduction" "functional_enrichment"
#> [10] "get_anno_col" "get_anno" "get_classes"
#> [13] "get_consensus" "get_matrix" "get_membership"
#> [16] "get_param" "get_signatures" "get_stats"
#> [19] "is_best_k" "is_stable_k" "membership_heatmap"
#> [22] "ncol" "nrow" "plot_ecdf"
#> [25] "rownames" "select_partition_number" "show"
#> [28] "suggest_best_k" "test_to_known_factors"
collect_plots() function collects all the plots made from res for all k (number of partitions)
into one single page to provide an easy and fast comparison between different k.
collect_plots(res)
The plots are:
k and the heatmap of
predicted classes for each k.k.k.k.All the plots in panels can be made by individual functions and they are plotted later in this section.
select_partition_number() produces several plots showing different
statistics for choosing “optimized” k. There are following statistics:
k;k, the area increased is defined as \(A_k - A_{k-1}\).The detailed explanations of these statistics can be found in the cola vignette.
Generally speaking, lower PAC score, higher mean silhouette score or higher
concordance corresponds to better partition. Rand index and Jaccard index
measure how similar the current partition is compared to partition with k-1.
If they are too similar, we won't accept k is better than k-1.
select_partition_number(res)
The numeric values for all these statistics can be obtained by get_stats().
get_stats(res)
#> k 1-PAC mean_silhouette concordance area_increased Rand Jaccard
#> 2 2 0.696 0.948 0.973 0.5055 0.495 0.495
#> 3 3 0.936 0.945 0.976 0.3264 0.706 0.475
#> 4 4 0.720 0.699 0.843 0.0956 0.908 0.735
#> 5 5 0.772 0.804 0.888 0.0726 0.875 0.591
#> 6 6 0.774 0.640 0.794 0.0462 0.952 0.783
suggest_best_k() suggests the best \(k\) based on these statistics. The rules are as follows:
suggest_best_k(res)
#> [1] 3
Following shows the table of the partitions (You need to click the show/hide
code output link to see it). The membership matrix (columns with name p*)
is inferred by
clue::cl_consensus()
function with the SE method. Basically the value in the membership matrix
represents the probability to belong to a certain group. The finall class
label for an item is determined with the group with highest probability it
belongs to.
In get_classes() function, the entropy is calculated from the membership
matrix and the silhouette score is calculated from the consensus matrix.
cbind(get_classes(res, k = 2), get_membership(res, k = 2))
#> class entropy silhouette p1 p2
#> GSM97138 1 0.000 0.967 1.000 0.000
#> GSM97145 1 0.000 0.967 1.000 0.000
#> GSM97147 2 0.000 0.974 0.000 1.000
#> GSM97125 1 0.000 0.967 1.000 0.000
#> GSM97127 2 0.469 0.906 0.100 0.900
#> GSM97130 2 0.469 0.906 0.100 0.900
#> GSM97133 2 0.469 0.906 0.100 0.900
#> GSM97134 2 0.662 0.832 0.172 0.828
#> GSM97120 1 0.000 0.967 1.000 0.000
#> GSM97126 1 0.000 0.967 1.000 0.000
#> GSM97112 1 0.000 0.967 1.000 0.000
#> GSM97115 2 0.000 0.974 0.000 1.000
#> GSM97116 1 0.000 0.967 1.000 0.000
#> GSM97117 1 0.000 0.967 1.000 0.000
#> GSM97119 1 0.000 0.967 1.000 0.000
#> GSM97122 1 0.000 0.967 1.000 0.000
#> GSM97135 1 0.000 0.967 1.000 0.000
#> GSM97136 1 0.000 0.967 1.000 0.000
#> GSM97139 1 0.000 0.967 1.000 0.000
#> GSM97146 1 0.000 0.967 1.000 0.000
#> GSM97123 2 0.000 0.974 0.000 1.000
#> GSM97129 2 0.605 0.861 0.148 0.852
#> GSM97143 1 0.000 0.967 1.000 0.000
#> GSM97113 2 0.000 0.974 0.000 1.000
#> GSM97056 2 0.469 0.906 0.100 0.900
#> GSM97124 1 0.000 0.967 1.000 0.000
#> GSM97132 1 0.000 0.967 1.000 0.000
#> GSM97144 2 0.469 0.906 0.100 0.900
#> GSM97149 2 0.469 0.906 0.100 0.900
#> GSM97068 2 0.000 0.974 0.000 1.000
#> GSM97071 1 0.000 0.967 1.000 0.000
#> GSM97086 2 0.000 0.974 0.000 1.000
#> GSM97103 2 0.000 0.974 0.000 1.000
#> GSM97057 2 0.000 0.974 0.000 1.000
#> GSM97060 2 0.000 0.974 0.000 1.000
#> GSM97075 1 0.469 0.903 0.900 0.100
#> GSM97098 2 0.000 0.974 0.000 1.000
#> GSM97099 2 0.000 0.974 0.000 1.000
#> GSM97101 2 0.000 0.974 0.000 1.000
#> GSM97105 2 0.000 0.974 0.000 1.000
#> GSM97106 2 0.000 0.974 0.000 1.000
#> GSM97121 2 0.000 0.974 0.000 1.000
#> GSM97128 1 0.000 0.967 1.000 0.000
#> GSM97131 2 0.000 0.974 0.000 1.000
#> GSM97137 2 0.469 0.906 0.100 0.900
#> GSM97118 1 0.000 0.967 1.000 0.000
#> GSM97114 2 0.469 0.906 0.100 0.900
#> GSM97142 1 0.000 0.967 1.000 0.000
#> GSM97140 2 0.000 0.974 0.000 1.000
#> GSM97141 2 0.000 0.974 0.000 1.000
#> GSM97055 1 0.000 0.967 1.000 0.000
#> GSM97090 2 0.000 0.974 0.000 1.000
#> GSM97091 1 0.000 0.967 1.000 0.000
#> GSM97148 1 0.000 0.967 1.000 0.000
#> GSM97063 1 0.000 0.967 1.000 0.000
#> GSM97053 1 0.000 0.967 1.000 0.000
#> GSM97066 1 0.000 0.967 1.000 0.000
#> GSM97079 2 0.000 0.974 0.000 1.000
#> GSM97083 1 0.000 0.967 1.000 0.000
#> GSM97084 2 0.000 0.974 0.000 1.000
#> GSM97094 1 0.000 0.967 1.000 0.000
#> GSM97096 1 0.469 0.903 0.900 0.100
#> GSM97097 2 0.000 0.974 0.000 1.000
#> GSM97107 2 0.456 0.909 0.096 0.904
#> GSM97054 2 0.000 0.974 0.000 1.000
#> GSM97062 2 0.000 0.974 0.000 1.000
#> GSM97069 1 0.469 0.903 0.900 0.100
#> GSM97070 1 0.000 0.967 1.000 0.000
#> GSM97073 1 0.000 0.967 1.000 0.000
#> GSM97076 1 0.000 0.967 1.000 0.000
#> GSM97077 2 0.000 0.974 0.000 1.000
#> GSM97095 1 0.963 0.326 0.612 0.388
#> GSM97102 1 0.469 0.903 0.900 0.100
#> GSM97109 2 0.000 0.974 0.000 1.000
#> GSM97110 2 0.000 0.974 0.000 1.000
#> GSM97074 1 0.000 0.967 1.000 0.000
#> GSM97085 1 0.000 0.967 1.000 0.000
#> GSM97059 2 0.000 0.974 0.000 1.000
#> GSM97072 2 0.000 0.974 0.000 1.000
#> GSM97078 1 0.000 0.967 1.000 0.000
#> GSM97067 1 0.000 0.967 1.000 0.000
#> GSM97087 1 0.469 0.903 0.900 0.100
#> GSM97111 1 0.552 0.878 0.872 0.128
#> GSM97064 2 0.000 0.974 0.000 1.000
#> GSM97065 1 0.000 0.967 1.000 0.000
#> GSM97081 1 0.469 0.903 0.900 0.100
#> GSM97082 1 0.469 0.903 0.900 0.100
#> GSM97088 1 0.000 0.967 1.000 0.000
#> GSM97100 2 0.000 0.974 0.000 1.000
#> GSM97104 1 0.469 0.903 0.900 0.100
#> GSM97108 2 0.000 0.974 0.000 1.000
#> GSM97050 2 0.000 0.974 0.000 1.000
#> GSM97080 1 0.469 0.903 0.900 0.100
#> GSM97089 1 0.469 0.903 0.900 0.100
#> GSM97092 2 0.000 0.974 0.000 1.000
#> GSM97093 2 0.000 0.974 0.000 1.000
#> GSM97058 2 0.000 0.974 0.000 1.000
#> GSM97051 2 0.000 0.974 0.000 1.000
#> GSM97052 2 0.000 0.974 0.000 1.000
#> GSM97061 2 0.000 0.974 0.000 1.000
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM97138 1 0.0000 0.961 1.000 0.000 0.000
#> GSM97145 1 0.0000 0.961 1.000 0.000 0.000
#> GSM97147 2 0.0000 0.985 0.000 1.000 0.000
#> GSM97125 1 0.0000 0.961 1.000 0.000 0.000
#> GSM97127 1 0.0000 0.961 1.000 0.000 0.000
#> GSM97130 1 0.0000 0.961 1.000 0.000 0.000
#> GSM97133 1 0.0000 0.961 1.000 0.000 0.000
#> GSM97134 1 0.0000 0.961 1.000 0.000 0.000
#> GSM97120 1 0.0000 0.961 1.000 0.000 0.000
#> GSM97126 1 0.0000 0.961 1.000 0.000 0.000
#> GSM97112 1 0.4346 0.783 0.816 0.000 0.184
#> GSM97115 2 0.0000 0.985 0.000 1.000 0.000
#> GSM97116 1 0.0000 0.961 1.000 0.000 0.000
#> GSM97117 1 0.0000 0.961 1.000 0.000 0.000
#> GSM97119 1 0.0000 0.961 1.000 0.000 0.000
#> GSM97122 1 0.0000 0.961 1.000 0.000 0.000
#> GSM97135 1 0.0000 0.961 1.000 0.000 0.000
#> GSM97136 3 0.0000 0.974 0.000 0.000 1.000
#> GSM97139 1 0.0000 0.961 1.000 0.000 0.000
#> GSM97146 1 0.0000 0.961 1.000 0.000 0.000
#> GSM97123 2 0.0000 0.985 0.000 1.000 0.000
#> GSM97129 1 0.0000 0.961 1.000 0.000 0.000
#> GSM97143 1 0.4346 0.783 0.816 0.000 0.184
#> GSM97113 2 0.0000 0.985 0.000 1.000 0.000
#> GSM97056 1 0.0000 0.961 1.000 0.000 0.000
#> GSM97124 1 0.0000 0.961 1.000 0.000 0.000
#> GSM97132 1 0.0000 0.961 1.000 0.000 0.000
#> GSM97144 1 0.0000 0.961 1.000 0.000 0.000
#> GSM97149 1 0.0000 0.961 1.000 0.000 0.000
#> GSM97068 2 0.0000 0.985 0.000 1.000 0.000
#> GSM97071 3 0.0000 0.974 0.000 0.000 1.000
#> GSM97086 2 0.0000 0.985 0.000 1.000 0.000
#> GSM97103 2 0.0000 0.985 0.000 1.000 0.000
#> GSM97057 2 0.0000 0.985 0.000 1.000 0.000
#> GSM97060 3 0.4555 0.740 0.000 0.200 0.800
#> GSM97075 3 0.0000 0.974 0.000 0.000 1.000
#> GSM97098 2 0.0000 0.985 0.000 1.000 0.000
#> GSM97099 2 0.0000 0.985 0.000 1.000 0.000
#> GSM97101 2 0.0000 0.985 0.000 1.000 0.000
#> GSM97105 2 0.0000 0.985 0.000 1.000 0.000
#> GSM97106 2 0.0000 0.985 0.000 1.000 0.000
#> GSM97121 2 0.0000 0.985 0.000 1.000 0.000
#> GSM97128 3 0.0000 0.974 0.000 0.000 1.000
#> GSM97131 2 0.0000 0.985 0.000 1.000 0.000
#> GSM97137 1 0.0000 0.961 1.000 0.000 0.000
#> GSM97118 3 0.5138 0.630 0.252 0.000 0.748
#> GSM97114 1 0.0237 0.958 0.996 0.004 0.000
#> GSM97142 1 0.0000 0.961 1.000 0.000 0.000
#> GSM97140 2 0.0000 0.985 0.000 1.000 0.000
#> GSM97141 2 0.0000 0.985 0.000 1.000 0.000
#> GSM97055 3 0.0000 0.974 0.000 0.000 1.000
#> GSM97090 2 0.0000 0.985 0.000 1.000 0.000
#> GSM97091 3 0.0424 0.967 0.008 0.000 0.992
#> GSM97148 1 0.0000 0.961 1.000 0.000 0.000
#> GSM97063 1 0.4605 0.757 0.796 0.000 0.204
#> GSM97053 1 0.0000 0.961 1.000 0.000 0.000
#> GSM97066 3 0.0000 0.974 0.000 0.000 1.000
#> GSM97079 2 0.0000 0.985 0.000 1.000 0.000
#> GSM97083 3 0.0000 0.974 0.000 0.000 1.000
#> GSM97084 2 0.0000 0.985 0.000 1.000 0.000
#> GSM97094 1 0.0000 0.961 1.000 0.000 0.000
#> GSM97096 3 0.0000 0.974 0.000 0.000 1.000
#> GSM97097 2 0.0000 0.985 0.000 1.000 0.000
#> GSM97107 2 0.0237 0.981 0.004 0.996 0.000
#> GSM97054 2 0.0000 0.985 0.000 1.000 0.000
#> GSM97062 2 0.0000 0.985 0.000 1.000 0.000
#> GSM97069 3 0.0000 0.974 0.000 0.000 1.000
#> GSM97070 3 0.0000 0.974 0.000 0.000 1.000
#> GSM97073 3 0.0000 0.974 0.000 0.000 1.000
#> GSM97076 1 0.5988 0.455 0.632 0.000 0.368
#> GSM97077 2 0.0000 0.985 0.000 1.000 0.000
#> GSM97095 1 0.6723 0.629 0.704 0.248 0.048
#> GSM97102 3 0.0000 0.974 0.000 0.000 1.000
#> GSM97109 2 0.0000 0.985 0.000 1.000 0.000
#> GSM97110 2 0.0000 0.985 0.000 1.000 0.000
#> GSM97074 3 0.0000 0.974 0.000 0.000 1.000
#> GSM97085 3 0.0000 0.974 0.000 0.000 1.000
#> GSM97059 2 0.0000 0.985 0.000 1.000 0.000
#> GSM97072 3 0.0000 0.974 0.000 0.000 1.000
#> GSM97078 3 0.0000 0.974 0.000 0.000 1.000
#> GSM97067 3 0.0000 0.974 0.000 0.000 1.000
#> GSM97087 3 0.0000 0.974 0.000 0.000 1.000
#> GSM97111 1 0.0892 0.945 0.980 0.020 0.000
#> GSM97064 2 0.0000 0.985 0.000 1.000 0.000
#> GSM97065 3 0.0000 0.974 0.000 0.000 1.000
#> GSM97081 3 0.0000 0.974 0.000 0.000 1.000
#> GSM97082 3 0.0000 0.974 0.000 0.000 1.000
#> GSM97088 3 0.0000 0.974 0.000 0.000 1.000
#> GSM97100 2 0.0000 0.985 0.000 1.000 0.000
#> GSM97104 3 0.0000 0.974 0.000 0.000 1.000
#> GSM97108 2 0.0000 0.985 0.000 1.000 0.000
#> GSM97050 2 0.0000 0.985 0.000 1.000 0.000
#> GSM97080 3 0.0000 0.974 0.000 0.000 1.000
#> GSM97089 3 0.0000 0.974 0.000 0.000 1.000
#> GSM97092 3 0.4504 0.746 0.000 0.196 0.804
#> GSM97093 2 0.5138 0.664 0.000 0.748 0.252
#> GSM97058 2 0.0000 0.985 0.000 1.000 0.000
#> GSM97051 2 0.0000 0.985 0.000 1.000 0.000
#> GSM97052 2 0.5291 0.636 0.000 0.732 0.268
#> GSM97061 2 0.0000 0.985 0.000 1.000 0.000
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM97138 1 0.4605 0.705 0.664 0.000 0.000 0.336
#> GSM97145 1 0.0000 0.810 1.000 0.000 0.000 0.000
#> GSM97147 2 0.1004 0.919 0.024 0.972 0.000 0.004
#> GSM97125 1 0.3266 0.795 0.832 0.000 0.000 0.168
#> GSM97127 1 0.0000 0.810 1.000 0.000 0.000 0.000
#> GSM97130 1 0.2773 0.723 0.880 0.116 0.000 0.004
#> GSM97133 1 0.1302 0.784 0.956 0.044 0.000 0.000
#> GSM97134 1 0.0188 0.809 0.996 0.000 0.000 0.004
#> GSM97120 1 0.0000 0.810 1.000 0.000 0.000 0.000
#> GSM97126 1 0.4605 0.705 0.664 0.000 0.000 0.336
#> GSM97112 1 0.4661 0.693 0.652 0.000 0.000 0.348
#> GSM97115 2 0.1489 0.907 0.044 0.952 0.000 0.004
#> GSM97116 1 0.3528 0.788 0.808 0.000 0.000 0.192
#> GSM97117 1 0.3400 0.792 0.820 0.000 0.000 0.180
#> GSM97119 1 0.4382 0.738 0.704 0.000 0.000 0.296
#> GSM97122 1 0.4331 0.743 0.712 0.000 0.000 0.288
#> GSM97135 1 0.4382 0.738 0.704 0.000 0.000 0.296
#> GSM97136 3 0.4877 0.390 0.000 0.000 0.592 0.408
#> GSM97139 1 0.0000 0.810 1.000 0.000 0.000 0.000
#> GSM97146 1 0.2281 0.807 0.904 0.000 0.000 0.096
#> GSM97123 2 0.4331 0.615 0.000 0.712 0.288 0.000
#> GSM97129 1 0.0188 0.808 0.996 0.000 0.000 0.004
#> GSM97143 1 0.4661 0.693 0.652 0.000 0.000 0.348
#> GSM97113 2 0.0000 0.928 0.000 1.000 0.000 0.000
#> GSM97056 1 0.0188 0.809 0.996 0.000 0.000 0.004
#> GSM97124 1 0.0707 0.810 0.980 0.000 0.000 0.020
#> GSM97132 1 0.3873 0.774 0.772 0.000 0.000 0.228
#> GSM97144 1 0.2888 0.715 0.872 0.124 0.000 0.004
#> GSM97149 1 0.0000 0.810 1.000 0.000 0.000 0.000
#> GSM97068 2 0.0188 0.928 0.000 0.996 0.000 0.004
#> GSM97071 4 0.4431 0.682 0.000 0.000 0.304 0.696
#> GSM97086 2 0.0000 0.928 0.000 1.000 0.000 0.000
#> GSM97103 2 0.6742 0.456 0.000 0.608 0.232 0.160
#> GSM97057 2 0.0000 0.928 0.000 1.000 0.000 0.000
#> GSM97060 3 0.0000 0.664 0.000 0.000 1.000 0.000
#> GSM97075 3 0.0188 0.662 0.000 0.000 0.996 0.004
#> GSM97098 3 0.5000 -0.185 0.000 0.496 0.504 0.000
#> GSM97099 2 0.6482 0.527 0.000 0.640 0.208 0.152
#> GSM97101 2 0.0188 0.927 0.000 0.996 0.000 0.004
#> GSM97105 2 0.0000 0.928 0.000 1.000 0.000 0.000
#> GSM97106 2 0.4356 0.608 0.000 0.708 0.292 0.000
#> GSM97121 2 0.1004 0.919 0.024 0.972 0.000 0.004
#> GSM97128 3 0.6552 0.274 0.076 0.000 0.484 0.440
#> GSM97131 2 0.0000 0.928 0.000 1.000 0.000 0.000
#> GSM97137 1 0.2888 0.715 0.872 0.124 0.000 0.004
#> GSM97118 4 0.7641 -0.179 0.344 0.000 0.216 0.440
#> GSM97114 1 0.2714 0.727 0.884 0.112 0.000 0.004
#> GSM97142 1 0.4661 0.693 0.652 0.000 0.000 0.348
#> GSM97140 2 0.1004 0.919 0.024 0.972 0.000 0.004
#> GSM97141 2 0.1004 0.919 0.024 0.972 0.000 0.004
#> GSM97055 3 0.7047 0.202 0.120 0.000 0.440 0.440
#> GSM97090 2 0.1398 0.910 0.040 0.956 0.000 0.004
#> GSM97091 4 0.7391 -0.231 0.164 0.000 0.396 0.440
#> GSM97148 1 0.0000 0.810 1.000 0.000 0.000 0.000
#> GSM97063 1 0.7617 0.303 0.424 0.000 0.204 0.372
#> GSM97053 1 0.0921 0.810 0.972 0.000 0.000 0.028
#> GSM97066 4 0.4454 0.683 0.000 0.000 0.308 0.692
#> GSM97079 2 0.0188 0.928 0.000 0.996 0.000 0.004
#> GSM97083 3 0.6130 0.312 0.048 0.000 0.512 0.440
#> GSM97084 2 0.0188 0.928 0.000 0.996 0.000 0.004
#> GSM97094 1 0.4164 0.760 0.736 0.000 0.000 0.264
#> GSM97096 3 0.0000 0.664 0.000 0.000 1.000 0.000
#> GSM97097 2 0.0188 0.928 0.000 0.996 0.000 0.004
#> GSM97107 2 0.4300 0.791 0.092 0.820 0.000 0.088
#> GSM97054 2 0.0188 0.928 0.000 0.996 0.000 0.004
#> GSM97062 2 0.0188 0.928 0.000 0.996 0.000 0.004
#> GSM97069 4 0.4522 0.671 0.000 0.000 0.320 0.680
#> GSM97070 4 0.4454 0.683 0.000 0.000 0.308 0.692
#> GSM97073 4 0.4454 0.683 0.000 0.000 0.308 0.692
#> GSM97076 4 0.0804 0.503 0.008 0.000 0.012 0.980
#> GSM97077 2 0.0000 0.928 0.000 1.000 0.000 0.000
#> GSM97095 3 0.9365 0.150 0.200 0.112 0.376 0.312
#> GSM97102 3 0.0188 0.661 0.000 0.000 0.996 0.004
#> GSM97109 2 0.2831 0.840 0.120 0.876 0.000 0.004
#> GSM97110 2 0.0000 0.928 0.000 1.000 0.000 0.000
#> GSM97074 4 0.1867 0.543 0.000 0.000 0.072 0.928
#> GSM97085 3 0.4790 0.406 0.000 0.000 0.620 0.380
#> GSM97059 2 0.1004 0.919 0.024 0.972 0.000 0.004
#> GSM97072 4 0.4713 0.615 0.000 0.000 0.360 0.640
#> GSM97078 4 0.4008 0.371 0.000 0.000 0.244 0.756
#> GSM97067 4 0.4454 0.683 0.000 0.000 0.308 0.692
#> GSM97087 3 0.0000 0.664 0.000 0.000 1.000 0.000
#> GSM97111 1 0.6080 0.496 0.660 0.012 0.272 0.056
#> GSM97064 2 0.1867 0.879 0.000 0.928 0.072 0.000
#> GSM97065 4 0.4431 0.682 0.000 0.000 0.304 0.696
#> GSM97081 3 0.0000 0.664 0.000 0.000 1.000 0.000
#> GSM97082 3 0.0000 0.664 0.000 0.000 1.000 0.000
#> GSM97088 3 0.4776 0.408 0.000 0.000 0.624 0.376
#> GSM97100 2 0.0000 0.928 0.000 1.000 0.000 0.000
#> GSM97104 3 0.0000 0.664 0.000 0.000 1.000 0.000
#> GSM97108 2 0.0188 0.927 0.000 0.996 0.000 0.004
#> GSM97050 2 0.0000 0.928 0.000 1.000 0.000 0.000
#> GSM97080 3 0.0000 0.664 0.000 0.000 1.000 0.000
#> GSM97089 3 0.0000 0.664 0.000 0.000 1.000 0.000
#> GSM97092 3 0.0000 0.664 0.000 0.000 1.000 0.000
#> GSM97093 3 0.4222 0.435 0.000 0.272 0.728 0.000
#> GSM97058 2 0.0000 0.928 0.000 1.000 0.000 0.000
#> GSM97051 2 0.0000 0.928 0.000 1.000 0.000 0.000
#> GSM97052 3 0.1637 0.621 0.000 0.060 0.940 0.000
#> GSM97061 2 0.4331 0.615 0.000 0.712 0.288 0.000
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM97138 5 0.0880 0.8460 0.032 0.000 0.000 0.000 0.968
#> GSM97145 1 0.2179 0.8418 0.888 0.000 0.000 0.000 0.112
#> GSM97147 2 0.2694 0.8571 0.128 0.864 0.004 0.004 0.000
#> GSM97125 1 0.4114 0.6334 0.624 0.000 0.000 0.000 0.376
#> GSM97127 1 0.1608 0.8399 0.928 0.000 0.000 0.000 0.072
#> GSM97130 1 0.2676 0.7932 0.884 0.080 0.000 0.000 0.036
#> GSM97133 1 0.1205 0.8299 0.956 0.004 0.000 0.000 0.040
#> GSM97134 1 0.2193 0.8268 0.900 0.008 0.000 0.000 0.092
#> GSM97120 1 0.2329 0.8418 0.876 0.000 0.000 0.000 0.124
#> GSM97126 5 0.0703 0.8508 0.024 0.000 0.000 0.000 0.976
#> GSM97112 5 0.0510 0.8533 0.016 0.000 0.000 0.000 0.984
#> GSM97115 2 0.3003 0.8094 0.188 0.812 0.000 0.000 0.000
#> GSM97116 1 0.4114 0.6336 0.624 0.000 0.000 0.000 0.376
#> GSM97117 1 0.4370 0.6019 0.656 0.000 0.004 0.008 0.332
#> GSM97119 5 0.2377 0.7499 0.128 0.000 0.000 0.000 0.872
#> GSM97122 5 0.3480 0.5282 0.248 0.000 0.000 0.000 0.752
#> GSM97135 5 0.3274 0.5927 0.220 0.000 0.000 0.000 0.780
#> GSM97136 5 0.3942 0.7008 0.000 0.000 0.232 0.020 0.748
#> GSM97139 1 0.2424 0.8416 0.868 0.000 0.000 0.000 0.132
#> GSM97146 1 0.3752 0.7445 0.708 0.000 0.000 0.000 0.292
#> GSM97123 3 0.3395 0.7117 0.000 0.236 0.764 0.000 0.000
#> GSM97129 1 0.1924 0.8317 0.924 0.000 0.004 0.008 0.064
#> GSM97143 5 0.0510 0.8533 0.016 0.000 0.000 0.000 0.984
#> GSM97113 2 0.0000 0.9052 0.000 1.000 0.000 0.000 0.000
#> GSM97056 1 0.1952 0.8269 0.912 0.004 0.000 0.000 0.084
#> GSM97124 1 0.3305 0.8013 0.776 0.000 0.000 0.000 0.224
#> GSM97132 1 0.4304 0.2989 0.516 0.000 0.000 0.000 0.484
#> GSM97144 1 0.2676 0.7932 0.884 0.080 0.000 0.000 0.036
#> GSM97149 1 0.1341 0.8349 0.944 0.000 0.000 0.000 0.056
#> GSM97068 2 0.0162 0.9043 0.004 0.996 0.000 0.000 0.000
#> GSM97071 4 0.0324 0.9395 0.004 0.000 0.000 0.992 0.004
#> GSM97086 2 0.0000 0.9052 0.000 1.000 0.000 0.000 0.000
#> GSM97103 2 0.6568 0.3183 0.000 0.528 0.276 0.184 0.012
#> GSM97057 2 0.0000 0.9052 0.000 1.000 0.000 0.000 0.000
#> GSM97060 3 0.0290 0.8805 0.000 0.008 0.992 0.000 0.000
#> GSM97075 3 0.0613 0.8804 0.004 0.000 0.984 0.008 0.004
#> GSM97098 3 0.2753 0.7916 0.000 0.136 0.856 0.000 0.008
#> GSM97099 2 0.6894 0.3986 0.020 0.548 0.248 0.172 0.012
#> GSM97101 2 0.1990 0.8798 0.068 0.920 0.008 0.004 0.000
#> GSM97105 2 0.0000 0.9052 0.000 1.000 0.000 0.000 0.000
#> GSM97106 3 0.3612 0.6615 0.000 0.268 0.732 0.000 0.000
#> GSM97121 2 0.2964 0.8426 0.152 0.840 0.004 0.004 0.000
#> GSM97128 5 0.3059 0.8155 0.028 0.000 0.108 0.004 0.860
#> GSM97131 2 0.0000 0.9052 0.000 1.000 0.000 0.000 0.000
#> GSM97137 1 0.2597 0.7861 0.884 0.092 0.000 0.000 0.024
#> GSM97118 5 0.0613 0.8550 0.008 0.000 0.004 0.004 0.984
#> GSM97114 1 0.1996 0.8212 0.932 0.016 0.004 0.008 0.040
#> GSM97142 5 0.0794 0.8481 0.028 0.000 0.000 0.000 0.972
#> GSM97140 2 0.2741 0.8550 0.132 0.860 0.004 0.004 0.000
#> GSM97141 2 0.3402 0.8173 0.184 0.804 0.008 0.004 0.000
#> GSM97055 5 0.1638 0.8434 0.000 0.000 0.064 0.004 0.932
#> GSM97090 2 0.2732 0.8277 0.160 0.840 0.000 0.000 0.000
#> GSM97091 5 0.0955 0.8531 0.000 0.000 0.028 0.004 0.968
#> GSM97148 1 0.2280 0.8443 0.880 0.000 0.000 0.000 0.120
#> GSM97063 5 0.0451 0.8549 0.008 0.000 0.004 0.000 0.988
#> GSM97053 1 0.3586 0.7714 0.736 0.000 0.000 0.000 0.264
#> GSM97066 4 0.0324 0.9412 0.000 0.000 0.004 0.992 0.004
#> GSM97079 2 0.0000 0.9052 0.000 1.000 0.000 0.000 0.000
#> GSM97083 5 0.3059 0.8155 0.028 0.000 0.108 0.004 0.860
#> GSM97084 2 0.0000 0.9052 0.000 1.000 0.000 0.000 0.000
#> GSM97094 5 0.2471 0.7999 0.136 0.000 0.000 0.000 0.864
#> GSM97096 3 0.0671 0.8821 0.000 0.000 0.980 0.016 0.004
#> GSM97097 2 0.0404 0.9017 0.000 0.988 0.000 0.000 0.012
#> GSM97107 2 0.3858 0.7951 0.156 0.804 0.000 0.024 0.016
#> GSM97054 2 0.0000 0.9052 0.000 1.000 0.000 0.000 0.000
#> GSM97062 2 0.0000 0.9052 0.000 1.000 0.000 0.000 0.000
#> GSM97069 4 0.0566 0.9365 0.000 0.000 0.012 0.984 0.004
#> GSM97070 4 0.0324 0.9412 0.000 0.000 0.004 0.992 0.004
#> GSM97073 4 0.0324 0.9412 0.000 0.000 0.004 0.992 0.004
#> GSM97076 4 0.0703 0.9259 0.000 0.000 0.000 0.976 0.024
#> GSM97077 2 0.0000 0.9052 0.000 1.000 0.000 0.000 0.000
#> GSM97095 5 0.3773 0.7983 0.100 0.020 0.048 0.000 0.832
#> GSM97102 3 0.1018 0.8801 0.000 0.000 0.968 0.016 0.016
#> GSM97109 2 0.4438 0.7307 0.252 0.720 0.008 0.008 0.012
#> GSM97110 2 0.0912 0.8962 0.000 0.972 0.016 0.000 0.012
#> GSM97074 4 0.0290 0.9398 0.000 0.000 0.000 0.992 0.008
#> GSM97085 5 0.3852 0.7168 0.000 0.000 0.220 0.020 0.760
#> GSM97059 2 0.2488 0.8615 0.124 0.872 0.004 0.000 0.000
#> GSM97072 4 0.0290 0.9382 0.000 0.000 0.008 0.992 0.000
#> GSM97078 4 0.6067 -0.0041 0.028 0.000 0.056 0.472 0.444
#> GSM97067 4 0.0324 0.9412 0.000 0.000 0.004 0.992 0.004
#> GSM97087 3 0.0912 0.8814 0.000 0.000 0.972 0.016 0.012
#> GSM97111 3 0.6782 -0.0588 0.408 0.004 0.412 0.008 0.168
#> GSM97064 2 0.3561 0.6142 0.000 0.740 0.260 0.000 0.000
#> GSM97065 4 0.0162 0.9400 0.000 0.000 0.000 0.996 0.004
#> GSM97081 3 0.0798 0.8820 0.000 0.000 0.976 0.016 0.008
#> GSM97082 3 0.1018 0.8801 0.000 0.000 0.968 0.016 0.016
#> GSM97088 5 0.3970 0.6968 0.000 0.000 0.236 0.020 0.744
#> GSM97100 2 0.0000 0.9052 0.000 1.000 0.000 0.000 0.000
#> GSM97104 3 0.1018 0.8801 0.000 0.000 0.968 0.016 0.016
#> GSM97108 2 0.1518 0.8899 0.048 0.944 0.004 0.004 0.000
#> GSM97050 2 0.0000 0.9052 0.000 1.000 0.000 0.000 0.000
#> GSM97080 3 0.0955 0.8779 0.000 0.000 0.968 0.028 0.004
#> GSM97089 3 0.0912 0.8814 0.000 0.000 0.972 0.016 0.012
#> GSM97092 3 0.0162 0.8809 0.000 0.004 0.996 0.000 0.000
#> GSM97093 3 0.1124 0.8705 0.000 0.036 0.960 0.000 0.004
#> GSM97058 2 0.0000 0.9052 0.000 1.000 0.000 0.000 0.000
#> GSM97051 2 0.0000 0.9052 0.000 1.000 0.000 0.000 0.000
#> GSM97052 3 0.0609 0.8769 0.000 0.020 0.980 0.000 0.000
#> GSM97061 3 0.3366 0.7164 0.000 0.232 0.768 0.000 0.000
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM97138 5 0.1643 0.78016 0.068 0.000 0.000 0.008 0.924 0.000
#> GSM97145 1 0.1320 0.62776 0.948 0.000 0.000 0.016 0.036 0.000
#> GSM97147 2 0.4517 0.47379 0.060 0.648 0.000 0.292 0.000 0.000
#> GSM97125 1 0.4124 0.42816 0.644 0.000 0.000 0.024 0.332 0.000
#> GSM97127 1 0.1700 0.61883 0.928 0.000 0.000 0.048 0.024 0.000
#> GSM97130 1 0.4184 0.29751 0.500 0.012 0.000 0.488 0.000 0.000
#> GSM97133 1 0.1204 0.60473 0.944 0.000 0.000 0.056 0.000 0.000
#> GSM97134 1 0.3854 0.33377 0.536 0.000 0.000 0.464 0.000 0.000
#> GSM97120 1 0.1434 0.63050 0.940 0.000 0.000 0.012 0.048 0.000
#> GSM97126 5 0.1285 0.79198 0.052 0.000 0.000 0.004 0.944 0.000
#> GSM97112 5 0.0777 0.79980 0.024 0.000 0.000 0.004 0.972 0.000
#> GSM97115 4 0.5114 0.34049 0.080 0.452 0.000 0.468 0.000 0.000
#> GSM97116 1 0.4139 0.41913 0.640 0.000 0.000 0.024 0.336 0.000
#> GSM97117 1 0.5650 0.28130 0.508 0.000 0.004 0.344 0.144 0.000
#> GSM97119 5 0.3592 0.56445 0.240 0.000 0.000 0.020 0.740 0.000
#> GSM97122 5 0.4193 0.32934 0.352 0.000 0.000 0.024 0.624 0.000
#> GSM97135 5 0.4139 0.36806 0.336 0.000 0.000 0.024 0.640 0.000
#> GSM97136 5 0.3248 0.66145 0.000 0.000 0.224 0.004 0.768 0.004
#> GSM97139 1 0.1333 0.63080 0.944 0.000 0.000 0.008 0.048 0.000
#> GSM97146 1 0.3629 0.53106 0.724 0.000 0.000 0.016 0.260 0.000
#> GSM97123 3 0.4431 0.70067 0.000 0.200 0.704 0.096 0.000 0.000
#> GSM97129 1 0.3499 0.39471 0.680 0.000 0.000 0.320 0.000 0.000
#> GSM97143 5 0.0777 0.79980 0.024 0.000 0.000 0.004 0.972 0.000
#> GSM97113 2 0.0000 0.80227 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97056 1 0.3866 0.31538 0.516 0.000 0.000 0.484 0.000 0.000
#> GSM97124 1 0.3276 0.60662 0.816 0.000 0.000 0.052 0.132 0.000
#> GSM97132 1 0.5917 0.16025 0.400 0.000 0.000 0.208 0.392 0.000
#> GSM97144 1 0.4184 0.29751 0.500 0.012 0.000 0.488 0.000 0.000
#> GSM97149 1 0.1398 0.61024 0.940 0.000 0.000 0.052 0.008 0.000
#> GSM97068 2 0.0146 0.80053 0.000 0.996 0.000 0.004 0.000 0.000
#> GSM97071 6 0.0000 0.98971 0.000 0.000 0.000 0.000 0.000 1.000
#> GSM97086 2 0.0000 0.80227 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97103 2 0.6238 0.21979 0.000 0.584 0.156 0.176 0.000 0.084
#> GSM97057 2 0.0146 0.80055 0.000 0.996 0.000 0.004 0.000 0.000
#> GSM97060 3 0.1714 0.86754 0.000 0.000 0.908 0.092 0.000 0.000
#> GSM97075 3 0.2278 0.83593 0.000 0.000 0.868 0.128 0.004 0.000
#> GSM97098 3 0.4121 0.75467 0.000 0.136 0.748 0.116 0.000 0.000
#> GSM97099 4 0.6829 -0.11569 0.024 0.376 0.116 0.432 0.000 0.052
#> GSM97101 2 0.4219 0.49273 0.036 0.660 0.000 0.304 0.000 0.000
#> GSM97105 2 0.0000 0.80227 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97106 3 0.4513 0.68357 0.000 0.212 0.692 0.096 0.000 0.000
#> GSM97121 2 0.5070 0.36248 0.096 0.576 0.000 0.328 0.000 0.000
#> GSM97128 5 0.2118 0.75510 0.000 0.000 0.104 0.008 0.888 0.000
#> GSM97131 2 0.0000 0.80227 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97137 1 0.4264 0.29057 0.496 0.016 0.000 0.488 0.000 0.000
#> GSM97118 5 0.0260 0.80012 0.008 0.000 0.000 0.000 0.992 0.000
#> GSM97114 1 0.3672 0.34800 0.632 0.000 0.000 0.368 0.000 0.000
#> GSM97142 5 0.1265 0.79343 0.044 0.000 0.000 0.008 0.948 0.000
#> GSM97140 2 0.4736 0.43351 0.072 0.620 0.000 0.308 0.000 0.000
#> GSM97141 2 0.5574 0.21699 0.152 0.504 0.000 0.344 0.000 0.000
#> GSM97055 5 0.0547 0.79364 0.000 0.000 0.020 0.000 0.980 0.000
#> GSM97090 4 0.4948 0.33339 0.064 0.460 0.000 0.476 0.000 0.000
#> GSM97091 5 0.0000 0.79902 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM97148 1 0.1333 0.63129 0.944 0.000 0.000 0.008 0.048 0.000
#> GSM97063 5 0.0458 0.80058 0.016 0.000 0.000 0.000 0.984 0.000
#> GSM97053 1 0.3333 0.58971 0.784 0.000 0.000 0.024 0.192 0.000
#> GSM97066 6 0.0000 0.98971 0.000 0.000 0.000 0.000 0.000 1.000
#> GSM97079 2 0.0146 0.80053 0.000 0.996 0.000 0.004 0.000 0.000
#> GSM97083 5 0.2118 0.75504 0.000 0.000 0.104 0.008 0.888 0.000
#> GSM97084 2 0.0363 0.79503 0.000 0.988 0.000 0.012 0.000 0.000
#> GSM97094 4 0.5745 -0.05626 0.212 0.000 0.000 0.508 0.280 0.000
#> GSM97096 3 0.0862 0.87822 0.000 0.000 0.972 0.016 0.008 0.004
#> GSM97097 2 0.1007 0.77106 0.000 0.956 0.000 0.044 0.000 0.000
#> GSM97107 4 0.4868 0.36896 0.060 0.416 0.000 0.524 0.000 0.000
#> GSM97054 2 0.0363 0.79503 0.000 0.988 0.000 0.012 0.000 0.000
#> GSM97062 2 0.0146 0.80053 0.000 0.996 0.000 0.004 0.000 0.000
#> GSM97069 6 0.1588 0.91031 0.000 0.000 0.072 0.004 0.000 0.924
#> GSM97070 6 0.0000 0.98971 0.000 0.000 0.000 0.000 0.000 1.000
#> GSM97073 6 0.0000 0.98971 0.000 0.000 0.000 0.000 0.000 1.000
#> GSM97076 6 0.0146 0.98485 0.000 0.000 0.000 0.000 0.004 0.996
#> GSM97077 2 0.0000 0.80227 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97095 4 0.5864 -0.02095 0.052 0.012 0.040 0.492 0.404 0.000
#> GSM97102 3 0.1003 0.86788 0.000 0.000 0.964 0.004 0.028 0.004
#> GSM97109 4 0.5999 0.00961 0.256 0.312 0.000 0.432 0.000 0.000
#> GSM97110 2 0.1556 0.74120 0.000 0.920 0.000 0.080 0.000 0.000
#> GSM97074 6 0.0000 0.98971 0.000 0.000 0.000 0.000 0.000 1.000
#> GSM97085 5 0.3192 0.66694 0.000 0.000 0.216 0.004 0.776 0.004
#> GSM97059 2 0.4110 0.53165 0.052 0.712 0.000 0.236 0.000 0.000
#> GSM97072 6 0.0000 0.98971 0.000 0.000 0.000 0.000 0.000 1.000
#> GSM97078 5 0.5856 0.07297 0.000 0.000 0.084 0.036 0.460 0.420
#> GSM97067 6 0.0000 0.98971 0.000 0.000 0.000 0.000 0.000 1.000
#> GSM97087 3 0.0748 0.87565 0.000 0.000 0.976 0.004 0.016 0.004
#> GSM97111 1 0.5753 0.19283 0.468 0.008 0.044 0.436 0.044 0.000
#> GSM97064 2 0.3426 0.57936 0.000 0.808 0.124 0.068 0.000 0.000
#> GSM97065 6 0.0000 0.98971 0.000 0.000 0.000 0.000 0.000 1.000
#> GSM97081 3 0.0291 0.87791 0.000 0.000 0.992 0.000 0.004 0.004
#> GSM97082 3 0.0748 0.87565 0.000 0.000 0.976 0.004 0.016 0.004
#> GSM97088 5 0.3354 0.64153 0.000 0.000 0.240 0.004 0.752 0.004
#> GSM97100 2 0.0000 0.80227 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97104 3 0.0748 0.87565 0.000 0.000 0.976 0.004 0.016 0.004
#> GSM97108 2 0.4165 0.50724 0.036 0.672 0.000 0.292 0.000 0.000
#> GSM97050 2 0.0000 0.80227 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97080 3 0.0748 0.87565 0.000 0.000 0.976 0.004 0.016 0.004
#> GSM97089 3 0.0748 0.87565 0.000 0.000 0.976 0.004 0.016 0.004
#> GSM97092 3 0.1714 0.86754 0.000 0.000 0.908 0.092 0.000 0.000
#> GSM97093 3 0.2070 0.86392 0.000 0.008 0.892 0.100 0.000 0.000
#> GSM97058 2 0.0000 0.80227 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97051 2 0.0000 0.80227 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97052 3 0.1714 0.86754 0.000 0.000 0.908 0.092 0.000 0.000
#> GSM97061 3 0.4431 0.70067 0.000 0.200 0.704 0.096 0.000 0.000
Heatmaps for the consensus matrix. It visualizes the probability of two samples to be in a same group.
consensus_heatmap(res, k = 2)
consensus_heatmap(res, k = 3)
consensus_heatmap(res, k = 4)
consensus_heatmap(res, k = 5)
consensus_heatmap(res, k = 6)
Heatmaps for the membership of samples in all partitions to see how consistent they are:
membership_heatmap(res, k = 2)
membership_heatmap(res, k = 3)
membership_heatmap(res, k = 4)
membership_heatmap(res, k = 5)
membership_heatmap(res, k = 6)
As soon as we have had the classes for columns, we can look for signatures which are significantly different between classes which can be candidate marks for certain classes. Following are the heatmaps for signatures.
Signature heatmaps where rows are scaled:
get_signatures(res, k = 2)
get_signatures(res, k = 3)
get_signatures(res, k = 4)
get_signatures(res, k = 5)
get_signatures(res, k = 6)
Signature heatmaps where rows are not scaled:
get_signatures(res, k = 2, scale_rows = FALSE)
get_signatures(res, k = 3, scale_rows = FALSE)
get_signatures(res, k = 4, scale_rows = FALSE)
get_signatures(res, k = 5, scale_rows = FALSE)
get_signatures(res, k = 6, scale_rows = FALSE)
Compare the overlap of signatures from different k:
compare_signatures(res)
get_signature() returns a data frame invisibly. TO get the list of signatures, the function
call should be assigned to a variable explicitly. In following code, if plot argument is set
to FALSE, no heatmap is plotted while only the differential analysis is performed.
# code only for demonstration
tb = get_signature(res, k = ..., plot = FALSE)
An example of the output of tb is:
#> which_row fdr mean_1 mean_2 scaled_mean_1 scaled_mean_2 km
#> 1 38 0.042760348 8.373488 9.131774 -0.5533452 0.5164555 1
#> 2 40 0.018707592 7.106213 8.469186 -0.6173731 0.5762149 1
#> 3 55 0.019134737 10.221463 11.207825 -0.6159697 0.5749050 1
#> 4 59 0.006059896 5.921854 7.869574 -0.6899429 0.6439467 1
#> 5 60 0.018055526 8.928898 10.211722 -0.6204761 0.5791110 1
#> 6 98 0.009384629 15.714769 14.887706 0.6635654 -0.6193277 2
...
The columns in tb are:
which_row: row indices corresponding to the input matrix.fdr: FDR for the differential test. mean_x: The mean value in group x.scaled_mean_x: The mean value in group x after rows are scaled.km: Row groups if k-means clustering is applied to rows.UMAP plot which shows how samples are separated.
dimension_reduction(res, k = 2, method = "UMAP")
dimension_reduction(res, k = 3, method = "UMAP")
dimension_reduction(res, k = 4, method = "UMAP")
dimension_reduction(res, k = 5, method = "UMAP")
dimension_reduction(res, k = 6, method = "UMAP")
Following heatmap shows how subgroups are split when increasing k:
collect_classes(res)
Test correlation between subgroups and known annotations. If the known annotation is numeric, one-way ANOVA test is applied, and if the known annotation is discrete, chi-squared contingency table test is applied.
test_to_known_factors(res)
#> n disease.state(p) specimen(p) cell.type(p) other(p) k
#> ATC:skmeans 99 5.34e-02 0.0201 3.41e-02 0.0991 2
#> ATC:skmeans 99 9.07e-07 0.1940 1.22e-12 0.0406 3
#> ATC:skmeans 85 1.78e-06 0.5372 5.62e-16 0.1843 4
#> ATC:skmeans 95 2.61e-04 0.5584 2.02e-15 0.3617 5
#> ATC:skmeans 72 9.22e-04 0.6769 3.37e-13 0.6042 6
If matrix rows can be associated to genes, consider to use functional_enrichment(res,
...) to perform function enrichment for the signature genes. See this vignette for more detailed explanations.
The object with results only for a single top-value method and a single partition method can be extracted as:
res = res_list["ATC", "pam"]
# you can also extract it by
# res = res_list["ATC:pam"]
A summary of res and all the functions that can be applied to it:
res
#> A 'ConsensusPartition' object with k = 2, 3, 4, 5, 6.
#> On a matrix with 21512 rows and 100 columns.
#> Top rows (1000, 2000, 3000, 4000, 5000) are extracted by 'ATC' method.
#> Subgroups are detected by 'pam' method.
#> Performed in total 1250 partitions by row resampling.
#> Best k for subgroups seems to be 2.
#>
#> Following methods can be applied to this 'ConsensusPartition' object:
#> [1] "cola_report" "collect_classes" "collect_plots"
#> [4] "collect_stats" "colnames" "compare_signatures"
#> [7] "consensus_heatmap" "dimension_reduction" "functional_enrichment"
#> [10] "get_anno_col" "get_anno" "get_classes"
#> [13] "get_consensus" "get_matrix" "get_membership"
#> [16] "get_param" "get_signatures" "get_stats"
#> [19] "is_best_k" "is_stable_k" "membership_heatmap"
#> [22] "ncol" "nrow" "plot_ecdf"
#> [25] "rownames" "select_partition_number" "show"
#> [28] "suggest_best_k" "test_to_known_factors"
collect_plots() function collects all the plots made from res for all k (number of partitions)
into one single page to provide an easy and fast comparison between different k.
collect_plots(res)
The plots are:
k and the heatmap of
predicted classes for each k.k.k.k.All the plots in panels can be made by individual functions and they are plotted later in this section.
select_partition_number() produces several plots showing different
statistics for choosing “optimized” k. There are following statistics:
k;k, the area increased is defined as \(A_k - A_{k-1}\).The detailed explanations of these statistics can be found in the cola vignette.
Generally speaking, lower PAC score, higher mean silhouette score or higher
concordance corresponds to better partition. Rand index and Jaccard index
measure how similar the current partition is compared to partition with k-1.
If they are too similar, we won't accept k is better than k-1.
select_partition_number(res)
The numeric values for all these statistics can be obtained by get_stats().
get_stats(res)
#> k 1-PAC mean_silhouette concordance area_increased Rand Jaccard
#> 2 2 0.624 0.902 0.944 0.4649 0.547 0.547
#> 3 3 0.864 0.900 0.949 0.4122 0.772 0.593
#> 4 4 0.807 0.822 0.921 0.0943 0.754 0.439
#> 5 5 0.761 0.562 0.772 0.0750 0.891 0.643
#> 6 6 0.890 0.871 0.933 0.0531 0.899 0.599
suggest_best_k() suggests the best \(k\) based on these statistics. The rules are as follows:
suggest_best_k(res)
#> [1] 2
Following shows the table of the partitions (You need to click the show/hide
code output link to see it). The membership matrix (columns with name p*)
is inferred by
clue::cl_consensus()
function with the SE method. Basically the value in the membership matrix
represents the probability to belong to a certain group. The finall class
label for an item is determined with the group with highest probability it
belongs to.
In get_classes() function, the entropy is calculated from the membership
matrix and the silhouette score is calculated from the consensus matrix.
cbind(get_classes(res, k = 2), get_membership(res, k = 2))
#> class entropy silhouette p1 p2
#> GSM97138 1 0.0000 0.961 1.000 0.000
#> GSM97145 1 0.0000 0.961 1.000 0.000
#> GSM97147 2 0.0938 0.920 0.012 0.988
#> GSM97125 1 0.0000 0.961 1.000 0.000
#> GSM97127 1 0.0000 0.961 1.000 0.000
#> GSM97130 1 0.6887 0.799 0.816 0.184
#> GSM97133 1 0.6801 0.801 0.820 0.180
#> GSM97134 1 0.0376 0.958 0.996 0.004
#> GSM97120 1 0.0000 0.961 1.000 0.000
#> GSM97126 1 0.0000 0.961 1.000 0.000
#> GSM97112 1 0.0000 0.961 1.000 0.000
#> GSM97115 2 0.0000 0.927 0.000 1.000
#> GSM97116 1 0.0000 0.961 1.000 0.000
#> GSM97117 2 0.7139 0.826 0.196 0.804
#> GSM97119 1 0.0000 0.961 1.000 0.000
#> GSM97122 1 0.0000 0.961 1.000 0.000
#> GSM97135 1 0.0000 0.961 1.000 0.000
#> GSM97136 2 0.9000 0.660 0.316 0.684
#> GSM97139 1 0.0000 0.961 1.000 0.000
#> GSM97146 1 0.0000 0.961 1.000 0.000
#> GSM97123 2 0.0000 0.927 0.000 1.000
#> GSM97129 2 0.6343 0.853 0.160 0.840
#> GSM97143 1 0.0000 0.961 1.000 0.000
#> GSM97113 2 0.0000 0.927 0.000 1.000
#> GSM97056 1 0.2948 0.922 0.948 0.052
#> GSM97124 1 0.0000 0.961 1.000 0.000
#> GSM97132 1 0.0000 0.961 1.000 0.000
#> GSM97144 1 0.7299 0.780 0.796 0.204
#> GSM97149 1 0.6801 0.801 0.820 0.180
#> GSM97068 2 0.0000 0.927 0.000 1.000
#> GSM97071 2 0.0000 0.927 0.000 1.000
#> GSM97086 2 0.0000 0.927 0.000 1.000
#> GSM97103 2 0.0000 0.927 0.000 1.000
#> GSM97057 2 0.0000 0.927 0.000 1.000
#> GSM97060 2 0.0000 0.927 0.000 1.000
#> GSM97075 2 0.0672 0.924 0.008 0.992
#> GSM97098 2 0.0000 0.927 0.000 1.000
#> GSM97099 2 0.0000 0.927 0.000 1.000
#> GSM97101 2 0.0000 0.927 0.000 1.000
#> GSM97105 2 0.0000 0.927 0.000 1.000
#> GSM97106 2 0.0000 0.927 0.000 1.000
#> GSM97121 2 0.0000 0.927 0.000 1.000
#> GSM97128 1 0.0938 0.953 0.988 0.012
#> GSM97131 2 0.0000 0.927 0.000 1.000
#> GSM97137 1 0.6887 0.797 0.816 0.184
#> GSM97118 1 0.0000 0.961 1.000 0.000
#> GSM97114 2 0.0938 0.920 0.012 0.988
#> GSM97142 1 0.0000 0.961 1.000 0.000
#> GSM97140 2 0.0000 0.927 0.000 1.000
#> GSM97141 2 0.0000 0.927 0.000 1.000
#> GSM97055 1 0.0000 0.961 1.000 0.000
#> GSM97090 2 0.0672 0.923 0.008 0.992
#> GSM97091 1 0.0000 0.961 1.000 0.000
#> GSM97148 1 0.0000 0.961 1.000 0.000
#> GSM97063 1 0.0000 0.961 1.000 0.000
#> GSM97053 1 0.0000 0.961 1.000 0.000
#> GSM97066 2 0.6887 0.835 0.184 0.816
#> GSM97079 2 0.0000 0.927 0.000 1.000
#> GSM97083 1 0.0672 0.956 0.992 0.008
#> GSM97084 2 0.0000 0.927 0.000 1.000
#> GSM97094 2 0.6343 0.852 0.160 0.840
#> GSM97096 2 0.6801 0.839 0.180 0.820
#> GSM97097 2 0.0000 0.927 0.000 1.000
#> GSM97107 2 0.0000 0.927 0.000 1.000
#> GSM97054 2 0.0000 0.927 0.000 1.000
#> GSM97062 2 0.0000 0.927 0.000 1.000
#> GSM97069 2 0.6801 0.839 0.180 0.820
#> GSM97070 2 0.6887 0.835 0.184 0.816
#> GSM97073 2 0.6801 0.839 0.180 0.820
#> GSM97076 1 0.0938 0.953 0.988 0.012
#> GSM97077 2 0.0000 0.927 0.000 1.000
#> GSM97095 2 0.6623 0.844 0.172 0.828
#> GSM97102 2 0.6887 0.835 0.184 0.816
#> GSM97109 2 0.0000 0.927 0.000 1.000
#> GSM97110 2 0.0000 0.927 0.000 1.000
#> GSM97074 2 0.9922 0.354 0.448 0.552
#> GSM97085 1 0.5519 0.825 0.872 0.128
#> GSM97059 2 0.0000 0.927 0.000 1.000
#> GSM97072 2 0.0000 0.927 0.000 1.000
#> GSM97078 2 0.6801 0.839 0.180 0.820
#> GSM97067 2 0.6887 0.835 0.184 0.816
#> GSM97087 2 0.6801 0.839 0.180 0.820
#> GSM97111 2 0.0672 0.924 0.008 0.992
#> GSM97064 2 0.0000 0.927 0.000 1.000
#> GSM97065 2 0.6801 0.839 0.180 0.820
#> GSM97081 2 0.6801 0.839 0.180 0.820
#> GSM97082 2 0.6801 0.839 0.180 0.820
#> GSM97088 2 0.7056 0.828 0.192 0.808
#> GSM97100 2 0.0000 0.927 0.000 1.000
#> GSM97104 2 0.6801 0.839 0.180 0.820
#> GSM97108 2 0.0000 0.927 0.000 1.000
#> GSM97050 2 0.0000 0.927 0.000 1.000
#> GSM97080 2 0.6801 0.839 0.180 0.820
#> GSM97089 2 0.6801 0.839 0.180 0.820
#> GSM97092 2 0.0000 0.927 0.000 1.000
#> GSM97093 2 0.0000 0.927 0.000 1.000
#> GSM97058 2 0.0000 0.927 0.000 1.000
#> GSM97051 2 0.0000 0.927 0.000 1.000
#> GSM97052 2 0.0000 0.927 0.000 1.000
#> GSM97061 2 0.0000 0.927 0.000 1.000
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM97138 1 0.0000 0.9753 1.000 0.000 0.000
#> GSM97145 1 0.0000 0.9753 1.000 0.000 0.000
#> GSM97147 2 0.0747 0.9490 0.000 0.984 0.016
#> GSM97125 1 0.0000 0.9753 1.000 0.000 0.000
#> GSM97127 1 0.0000 0.9753 1.000 0.000 0.000
#> GSM97130 2 0.3686 0.8102 0.140 0.860 0.000
#> GSM97133 1 0.6026 0.3798 0.624 0.376 0.000
#> GSM97134 1 0.0424 0.9698 0.992 0.000 0.008
#> GSM97120 1 0.0000 0.9753 1.000 0.000 0.000
#> GSM97126 1 0.0000 0.9753 1.000 0.000 0.000
#> GSM97112 1 0.0000 0.9753 1.000 0.000 0.000
#> GSM97115 2 0.0747 0.9490 0.000 0.984 0.016
#> GSM97116 1 0.0000 0.9753 1.000 0.000 0.000
#> GSM97117 3 0.1647 0.9188 0.036 0.004 0.960
#> GSM97119 1 0.0000 0.9753 1.000 0.000 0.000
#> GSM97122 1 0.0000 0.9753 1.000 0.000 0.000
#> GSM97135 1 0.0000 0.9753 1.000 0.000 0.000
#> GSM97136 3 0.4842 0.7296 0.224 0.000 0.776
#> GSM97139 1 0.0000 0.9753 1.000 0.000 0.000
#> GSM97146 1 0.0000 0.9753 1.000 0.000 0.000
#> GSM97123 3 0.1643 0.9259 0.000 0.044 0.956
#> GSM97129 3 0.1751 0.9262 0.012 0.028 0.960
#> GSM97143 1 0.0000 0.9753 1.000 0.000 0.000
#> GSM97113 2 0.0592 0.9507 0.000 0.988 0.012
#> GSM97056 1 0.0000 0.9753 1.000 0.000 0.000
#> GSM97124 1 0.0000 0.9753 1.000 0.000 0.000
#> GSM97132 1 0.0000 0.9753 1.000 0.000 0.000
#> GSM97144 2 0.0747 0.9454 0.016 0.984 0.000
#> GSM97149 1 0.1411 0.9430 0.964 0.036 0.000
#> GSM97068 2 0.0747 0.9490 0.000 0.984 0.016
#> GSM97071 3 0.1643 0.9233 0.000 0.044 0.956
#> GSM97086 2 0.0000 0.9544 0.000 1.000 0.000
#> GSM97103 3 0.1643 0.9233 0.000 0.044 0.956
#> GSM97057 2 0.0000 0.9544 0.000 1.000 0.000
#> GSM97060 3 0.1643 0.9259 0.000 0.044 0.956
#> GSM97075 3 0.1751 0.9262 0.012 0.028 0.960
#> GSM97098 3 0.1163 0.9263 0.000 0.028 0.972
#> GSM97099 3 0.1529 0.9242 0.000 0.040 0.960
#> GSM97101 3 0.4931 0.7430 0.000 0.232 0.768
#> GSM97105 2 0.0000 0.9544 0.000 1.000 0.000
#> GSM97106 3 0.3816 0.8495 0.000 0.148 0.852
#> GSM97121 2 0.4654 0.7134 0.000 0.792 0.208
#> GSM97128 1 0.1163 0.9549 0.972 0.000 0.028
#> GSM97131 2 0.0747 0.9490 0.000 0.984 0.016
#> GSM97137 2 0.1643 0.9182 0.044 0.956 0.000
#> GSM97118 1 0.0000 0.9753 1.000 0.000 0.000
#> GSM97114 3 0.1751 0.9262 0.012 0.028 0.960
#> GSM97142 1 0.0000 0.9753 1.000 0.000 0.000
#> GSM97140 2 0.6302 -0.0342 0.000 0.520 0.480
#> GSM97141 3 0.4390 0.8362 0.012 0.148 0.840
#> GSM97055 1 0.0892 0.9598 0.980 0.000 0.020
#> GSM97090 2 0.0000 0.9544 0.000 1.000 0.000
#> GSM97091 1 0.0000 0.9753 1.000 0.000 0.000
#> GSM97148 1 0.0000 0.9753 1.000 0.000 0.000
#> GSM97063 1 0.0000 0.9753 1.000 0.000 0.000
#> GSM97053 1 0.0000 0.9753 1.000 0.000 0.000
#> GSM97066 3 0.0000 0.9239 0.000 0.000 1.000
#> GSM97079 2 0.0000 0.9544 0.000 1.000 0.000
#> GSM97083 1 0.1289 0.9519 0.968 0.000 0.032
#> GSM97084 2 0.0000 0.9544 0.000 1.000 0.000
#> GSM97094 3 0.3192 0.8630 0.112 0.000 0.888
#> GSM97096 3 0.0237 0.9251 0.000 0.004 0.996
#> GSM97097 3 0.5016 0.7323 0.000 0.240 0.760
#> GSM97107 3 0.6244 0.2770 0.000 0.440 0.560
#> GSM97054 2 0.0000 0.9544 0.000 1.000 0.000
#> GSM97062 2 0.0000 0.9544 0.000 1.000 0.000
#> GSM97069 3 0.0000 0.9239 0.000 0.000 1.000
#> GSM97070 3 0.0000 0.9239 0.000 0.000 1.000
#> GSM97073 3 0.0592 0.9231 0.012 0.000 0.988
#> GSM97076 1 0.0592 0.9676 0.988 0.000 0.012
#> GSM97077 2 0.0000 0.9544 0.000 1.000 0.000
#> GSM97095 3 0.3482 0.8595 0.000 0.128 0.872
#> GSM97102 3 0.0000 0.9239 0.000 0.000 1.000
#> GSM97109 3 0.1751 0.9262 0.012 0.028 0.960
#> GSM97110 3 0.1964 0.9178 0.000 0.056 0.944
#> GSM97074 3 0.6244 0.2456 0.440 0.000 0.560
#> GSM97085 1 0.4399 0.7789 0.812 0.000 0.188
#> GSM97059 2 0.0747 0.9490 0.000 0.984 0.016
#> GSM97072 3 0.0237 0.9247 0.000 0.004 0.996
#> GSM97078 3 0.2793 0.9133 0.028 0.044 0.928
#> GSM97067 3 0.0000 0.9239 0.000 0.000 1.000
#> GSM97087 3 0.0747 0.9224 0.000 0.016 0.984
#> GSM97111 3 0.1751 0.9262 0.012 0.028 0.960
#> GSM97064 2 0.0892 0.9417 0.000 0.980 0.020
#> GSM97065 3 0.1636 0.9255 0.020 0.016 0.964
#> GSM97081 3 0.1643 0.9259 0.000 0.044 0.956
#> GSM97082 3 0.0747 0.9224 0.000 0.016 0.984
#> GSM97088 3 0.1529 0.9076 0.040 0.000 0.960
#> GSM97100 2 0.0000 0.9544 0.000 1.000 0.000
#> GSM97104 3 0.0000 0.9239 0.000 0.000 1.000
#> GSM97108 3 0.5058 0.7262 0.000 0.244 0.756
#> GSM97050 2 0.0000 0.9544 0.000 1.000 0.000
#> GSM97080 3 0.0747 0.9224 0.000 0.016 0.984
#> GSM97089 3 0.0747 0.9224 0.000 0.016 0.984
#> GSM97092 3 0.1643 0.9259 0.000 0.044 0.956
#> GSM97093 3 0.1643 0.9259 0.000 0.044 0.956
#> GSM97058 2 0.0000 0.9544 0.000 1.000 0.000
#> GSM97051 2 0.0000 0.9544 0.000 1.000 0.000
#> GSM97052 3 0.1643 0.9259 0.000 0.044 0.956
#> GSM97061 3 0.1643 0.9259 0.000 0.044 0.956
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM97138 1 0.0000 0.906 1.000 0.000 0.000 0.000
#> GSM97145 1 0.0000 0.906 1.000 0.000 0.000 0.000
#> GSM97147 2 0.0336 0.920 0.008 0.992 0.000 0.000
#> GSM97125 1 0.0000 0.906 1.000 0.000 0.000 0.000
#> GSM97127 1 0.0000 0.906 1.000 0.000 0.000 0.000
#> GSM97130 2 0.2921 0.797 0.140 0.860 0.000 0.000
#> GSM97133 1 0.4624 0.538 0.660 0.340 0.000 0.000
#> GSM97134 1 0.0336 0.902 0.992 0.000 0.008 0.000
#> GSM97120 1 0.0000 0.906 1.000 0.000 0.000 0.000
#> GSM97126 1 0.0000 0.906 1.000 0.000 0.000 0.000
#> GSM97112 1 0.0000 0.906 1.000 0.000 0.000 0.000
#> GSM97115 2 0.0000 0.925 0.000 1.000 0.000 0.000
#> GSM97116 1 0.0000 0.906 1.000 0.000 0.000 0.000
#> GSM97117 1 0.5627 0.616 0.692 0.068 0.240 0.000
#> GSM97119 1 0.0000 0.906 1.000 0.000 0.000 0.000
#> GSM97122 1 0.0000 0.906 1.000 0.000 0.000 0.000
#> GSM97135 1 0.0000 0.906 1.000 0.000 0.000 0.000
#> GSM97136 1 0.2149 0.846 0.912 0.000 0.088 0.000
#> GSM97139 1 0.0000 0.906 1.000 0.000 0.000 0.000
#> GSM97146 1 0.0000 0.906 1.000 0.000 0.000 0.000
#> GSM97123 3 0.0336 0.826 0.000 0.008 0.992 0.000
#> GSM97129 1 0.5757 0.607 0.684 0.076 0.240 0.000
#> GSM97143 1 0.0000 0.906 1.000 0.000 0.000 0.000
#> GSM97113 2 0.0000 0.925 0.000 1.000 0.000 0.000
#> GSM97056 1 0.0000 0.906 1.000 0.000 0.000 0.000
#> GSM97124 1 0.0000 0.906 1.000 0.000 0.000 0.000
#> GSM97132 1 0.0000 0.906 1.000 0.000 0.000 0.000
#> GSM97144 2 0.0000 0.925 0.000 1.000 0.000 0.000
#> GSM97149 1 0.0921 0.889 0.972 0.028 0.000 0.000
#> GSM97068 2 0.0000 0.925 0.000 1.000 0.000 0.000
#> GSM97071 2 0.7768 0.065 0.000 0.400 0.240 0.360
#> GSM97086 2 0.0000 0.925 0.000 1.000 0.000 0.000
#> GSM97103 3 0.4746 0.326 0.000 0.368 0.632 0.000
#> GSM97057 2 0.0000 0.925 0.000 1.000 0.000 0.000
#> GSM97060 3 0.0000 0.830 0.000 0.000 1.000 0.000
#> GSM97075 3 0.6179 0.346 0.320 0.072 0.608 0.000
#> GSM97098 3 0.0000 0.830 0.000 0.000 1.000 0.000
#> GSM97099 2 0.4193 0.658 0.000 0.732 0.268 0.000
#> GSM97101 2 0.1474 0.896 0.000 0.948 0.052 0.000
#> GSM97105 2 0.0000 0.925 0.000 1.000 0.000 0.000
#> GSM97106 3 0.2011 0.777 0.000 0.080 0.920 0.000
#> GSM97121 2 0.0000 0.925 0.000 1.000 0.000 0.000
#> GSM97128 1 0.2011 0.839 0.920 0.000 0.080 0.000
#> GSM97131 2 0.0000 0.925 0.000 1.000 0.000 0.000
#> GSM97137 2 0.1940 0.854 0.076 0.924 0.000 0.000
#> GSM97118 1 0.0000 0.906 1.000 0.000 0.000 0.000
#> GSM97114 1 0.7010 0.466 0.576 0.184 0.240 0.000
#> GSM97142 1 0.0000 0.906 1.000 0.000 0.000 0.000
#> GSM97140 2 0.1474 0.897 0.000 0.948 0.052 0.000
#> GSM97141 1 0.6954 0.460 0.568 0.280 0.152 0.000
#> GSM97055 1 0.0188 0.904 0.996 0.000 0.004 0.000
#> GSM97090 2 0.0000 0.925 0.000 1.000 0.000 0.000
#> GSM97091 1 0.0188 0.904 0.996 0.000 0.004 0.000
#> GSM97148 1 0.0000 0.906 1.000 0.000 0.000 0.000
#> GSM97063 1 0.0000 0.906 1.000 0.000 0.000 0.000
#> GSM97053 1 0.0000 0.906 1.000 0.000 0.000 0.000
#> GSM97066 4 0.0000 0.989 0.000 0.000 0.000 1.000
#> GSM97079 2 0.0000 0.925 0.000 1.000 0.000 0.000
#> GSM97083 3 0.4933 0.273 0.432 0.000 0.568 0.000
#> GSM97084 2 0.0000 0.925 0.000 1.000 0.000 0.000
#> GSM97094 1 0.4491 0.749 0.800 0.060 0.140 0.000
#> GSM97096 3 0.0000 0.830 0.000 0.000 1.000 0.000
#> GSM97097 2 0.2011 0.875 0.000 0.920 0.080 0.000
#> GSM97107 2 0.3569 0.755 0.000 0.804 0.196 0.000
#> GSM97054 2 0.0000 0.925 0.000 1.000 0.000 0.000
#> GSM97062 2 0.0000 0.925 0.000 1.000 0.000 0.000
#> GSM97069 4 0.0000 0.989 0.000 0.000 0.000 1.000
#> GSM97070 4 0.0000 0.989 0.000 0.000 0.000 1.000
#> GSM97073 4 0.0000 0.989 0.000 0.000 0.000 1.000
#> GSM97076 4 0.0188 0.985 0.004 0.000 0.000 0.996
#> GSM97077 2 0.0000 0.925 0.000 1.000 0.000 0.000
#> GSM97095 2 0.3975 0.698 0.000 0.760 0.240 0.000
#> GSM97102 3 0.3907 0.622 0.000 0.000 0.768 0.232
#> GSM97109 1 0.7074 0.453 0.568 0.192 0.240 0.000
#> GSM97110 2 0.3975 0.698 0.000 0.760 0.240 0.000
#> GSM97074 4 0.0000 0.989 0.000 0.000 0.000 1.000
#> GSM97085 4 0.0188 0.985 0.000 0.000 0.004 0.996
#> GSM97059 2 0.0000 0.925 0.000 1.000 0.000 0.000
#> GSM97072 4 0.0000 0.989 0.000 0.000 0.000 1.000
#> GSM97078 3 0.6362 0.251 0.072 0.368 0.560 0.000
#> GSM97067 4 0.0000 0.989 0.000 0.000 0.000 1.000
#> GSM97087 3 0.0000 0.830 0.000 0.000 1.000 0.000
#> GSM97111 1 0.5817 0.596 0.676 0.076 0.248 0.000
#> GSM97064 2 0.3528 0.726 0.000 0.808 0.192 0.000
#> GSM97065 4 0.1867 0.902 0.000 0.000 0.072 0.928
#> GSM97081 3 0.0000 0.830 0.000 0.000 1.000 0.000
#> GSM97082 3 0.0000 0.830 0.000 0.000 1.000 0.000
#> GSM97088 3 0.3123 0.709 0.000 0.000 0.844 0.156
#> GSM97100 2 0.0000 0.925 0.000 1.000 0.000 0.000
#> GSM97104 3 0.3975 0.607 0.000 0.000 0.760 0.240
#> GSM97108 2 0.1474 0.896 0.000 0.948 0.052 0.000
#> GSM97050 2 0.0000 0.925 0.000 1.000 0.000 0.000
#> GSM97080 3 0.4072 0.594 0.000 0.000 0.748 0.252
#> GSM97089 3 0.0000 0.830 0.000 0.000 1.000 0.000
#> GSM97092 3 0.0000 0.830 0.000 0.000 1.000 0.000
#> GSM97093 3 0.0000 0.830 0.000 0.000 1.000 0.000
#> GSM97058 2 0.0000 0.925 0.000 1.000 0.000 0.000
#> GSM97051 2 0.0000 0.925 0.000 1.000 0.000 0.000
#> GSM97052 3 0.0000 0.830 0.000 0.000 1.000 0.000
#> GSM97061 3 0.0336 0.826 0.000 0.008 0.992 0.000
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM97138 1 0.4294 0.7066 0.532 0.000 0.000 0.000 0.468
#> GSM97145 1 0.4294 0.7066 0.532 0.000 0.000 0.000 0.468
#> GSM97147 2 0.0404 0.8530 0.000 0.988 0.000 0.012 0.000
#> GSM97125 1 0.4294 0.7066 0.532 0.000 0.000 0.000 0.468
#> GSM97127 1 0.4294 0.7066 0.532 0.000 0.000 0.000 0.468
#> GSM97130 2 0.2966 0.6835 0.184 0.816 0.000 0.000 0.000
#> GSM97133 1 0.5504 0.5848 0.488 0.064 0.000 0.000 0.448
#> GSM97134 1 0.4294 0.7066 0.532 0.000 0.000 0.000 0.468
#> GSM97120 1 0.4294 0.7066 0.532 0.000 0.000 0.000 0.468
#> GSM97126 1 0.4294 0.7066 0.532 0.000 0.000 0.000 0.468
#> GSM97112 5 0.0000 0.7013 0.000 0.000 0.000 0.000 1.000
#> GSM97115 2 0.0000 0.8552 0.000 1.000 0.000 0.000 0.000
#> GSM97116 1 0.4294 0.7066 0.532 0.000 0.000 0.000 0.468
#> GSM97117 1 0.5459 0.1210 0.472 0.000 0.060 0.468 0.000
#> GSM97119 1 0.4294 0.7066 0.532 0.000 0.000 0.000 0.468
#> GSM97122 1 0.4294 0.7066 0.532 0.000 0.000 0.000 0.468
#> GSM97135 5 0.4304 -0.6360 0.484 0.000 0.000 0.000 0.516
#> GSM97136 1 0.6036 0.1040 0.472 0.000 0.436 0.080 0.012
#> GSM97139 1 0.4294 0.7066 0.532 0.000 0.000 0.000 0.468
#> GSM97146 1 0.4294 0.7066 0.532 0.000 0.000 0.000 0.468
#> GSM97123 3 0.3835 0.6118 0.000 0.012 0.744 0.244 0.000
#> GSM97129 1 0.5320 0.1500 0.488 0.004 0.040 0.468 0.000
#> GSM97143 5 0.4278 -0.5600 0.452 0.000 0.000 0.000 0.548
#> GSM97113 2 0.0162 0.8555 0.000 0.996 0.004 0.000 0.000
#> GSM97056 1 0.4294 0.7066 0.532 0.000 0.000 0.000 0.468
#> GSM97124 1 0.4294 0.7066 0.532 0.000 0.000 0.000 0.468
#> GSM97132 1 0.4294 0.7066 0.532 0.000 0.000 0.000 0.468
#> GSM97144 2 0.1121 0.8290 0.044 0.956 0.000 0.000 0.000
#> GSM97149 1 0.4440 0.6989 0.528 0.004 0.000 0.000 0.468
#> GSM97068 2 0.0000 0.8552 0.000 1.000 0.000 0.000 0.000
#> GSM97071 4 0.5310 0.1536 0.072 0.108 0.080 0.740 0.000
#> GSM97086 2 0.0162 0.8555 0.000 0.996 0.004 0.000 0.000
#> GSM97103 4 0.5557 -0.3807 0.000 0.068 0.464 0.468 0.000
#> GSM97057 2 0.0000 0.8552 0.000 1.000 0.000 0.000 0.000
#> GSM97060 3 0.0000 0.7561 0.000 0.000 1.000 0.000 0.000
#> GSM97075 4 0.6684 -0.2209 0.152 0.016 0.364 0.468 0.000
#> GSM97098 3 0.4283 0.4132 0.000 0.000 0.544 0.456 0.000
#> GSM97099 4 0.6578 -0.0921 0.000 0.284 0.248 0.468 0.000
#> GSM97101 2 0.4294 0.3875 0.000 0.532 0.000 0.468 0.000
#> GSM97105 2 0.0404 0.8530 0.000 0.988 0.000 0.012 0.000
#> GSM97106 3 0.1893 0.7297 0.000 0.048 0.928 0.024 0.000
#> GSM97121 2 0.2929 0.7350 0.000 0.820 0.000 0.180 0.000
#> GSM97128 5 0.1205 0.6639 0.040 0.000 0.004 0.000 0.956
#> GSM97131 2 0.0404 0.8530 0.000 0.988 0.000 0.012 0.000
#> GSM97137 2 0.2074 0.7728 0.104 0.896 0.000 0.000 0.000
#> GSM97118 5 0.3177 0.3378 0.208 0.000 0.000 0.000 0.792
#> GSM97114 1 0.5320 0.1500 0.488 0.004 0.040 0.468 0.000
#> GSM97142 5 0.0000 0.7013 0.000 0.000 0.000 0.000 1.000
#> GSM97140 2 0.3730 0.6275 0.000 0.712 0.000 0.288 0.000
#> GSM97141 4 0.7258 0.0904 0.284 0.208 0.040 0.468 0.000
#> GSM97055 5 0.0000 0.7013 0.000 0.000 0.000 0.000 1.000
#> GSM97090 2 0.0162 0.8555 0.000 0.996 0.004 0.000 0.000
#> GSM97091 5 0.0000 0.7013 0.000 0.000 0.000 0.000 1.000
#> GSM97148 1 0.4294 0.7066 0.532 0.000 0.000 0.000 0.468
#> GSM97063 5 0.0000 0.7013 0.000 0.000 0.000 0.000 1.000
#> GSM97053 1 0.4294 0.7066 0.532 0.000 0.000 0.000 0.468
#> GSM97066 4 0.4294 0.5257 0.468 0.000 0.000 0.532 0.000
#> GSM97079 2 0.0162 0.8555 0.000 0.996 0.004 0.000 0.000
#> GSM97083 5 0.0000 0.7013 0.000 0.000 0.000 0.000 1.000
#> GSM97084 2 0.0162 0.8555 0.000 0.996 0.004 0.000 0.000
#> GSM97094 1 0.5496 0.1296 0.476 0.004 0.052 0.468 0.000
#> GSM97096 3 0.4283 0.4132 0.000 0.000 0.544 0.456 0.000
#> GSM97097 2 0.4549 0.3817 0.000 0.528 0.008 0.464 0.000
#> GSM97107 2 0.4555 0.6442 0.000 0.732 0.068 0.200 0.000
#> GSM97054 2 0.0162 0.8555 0.000 0.996 0.004 0.000 0.000
#> GSM97062 2 0.0162 0.8555 0.000 0.996 0.004 0.000 0.000
#> GSM97069 4 0.4294 0.5257 0.468 0.000 0.000 0.532 0.000
#> GSM97070 4 0.4294 0.5257 0.468 0.000 0.000 0.532 0.000
#> GSM97073 4 0.4294 0.5257 0.468 0.000 0.000 0.532 0.000
#> GSM97076 4 0.4294 0.5257 0.468 0.000 0.000 0.532 0.000
#> GSM97077 2 0.0162 0.8555 0.000 0.996 0.004 0.000 0.000
#> GSM97095 2 0.7058 0.2564 0.084 0.456 0.080 0.380 0.000
#> GSM97102 3 0.2773 0.6487 0.000 0.000 0.836 0.164 0.000
#> GSM97109 4 0.6202 -0.1377 0.432 0.020 0.080 0.468 0.000
#> GSM97110 2 0.5737 0.2606 0.000 0.460 0.084 0.456 0.000
#> GSM97074 4 0.4294 0.5257 0.468 0.000 0.000 0.532 0.000
#> GSM97085 5 0.5735 -0.0947 0.092 0.000 0.376 0.000 0.532
#> GSM97059 2 0.0404 0.8530 0.000 0.988 0.000 0.012 0.000
#> GSM97072 4 0.4294 0.5257 0.468 0.000 0.000 0.532 0.000
#> GSM97078 3 0.6069 0.3005 0.008 0.092 0.456 0.444 0.000
#> GSM97067 4 0.4294 0.5257 0.468 0.000 0.000 0.532 0.000
#> GSM97087 3 0.0000 0.7561 0.000 0.000 1.000 0.000 0.000
#> GSM97111 4 0.6038 -0.1522 0.440 0.012 0.080 0.468 0.000
#> GSM97064 2 0.1908 0.7947 0.000 0.908 0.092 0.000 0.000
#> GSM97065 4 0.5046 0.5066 0.468 0.000 0.032 0.500 0.000
#> GSM97081 3 0.4060 0.5195 0.000 0.000 0.640 0.360 0.000
#> GSM97082 3 0.0000 0.7561 0.000 0.000 1.000 0.000 0.000
#> GSM97088 3 0.3039 0.6228 0.000 0.000 0.808 0.000 0.192
#> GSM97100 2 0.0000 0.8552 0.000 1.000 0.000 0.000 0.000
#> GSM97104 3 0.2648 0.6590 0.000 0.000 0.848 0.152 0.000
#> GSM97108 2 0.4294 0.3875 0.000 0.532 0.000 0.468 0.000
#> GSM97050 2 0.0162 0.8555 0.000 0.996 0.004 0.000 0.000
#> GSM97080 3 0.2970 0.6423 0.004 0.000 0.828 0.168 0.000
#> GSM97089 3 0.0000 0.7561 0.000 0.000 1.000 0.000 0.000
#> GSM97092 3 0.0000 0.7561 0.000 0.000 1.000 0.000 0.000
#> GSM97093 3 0.0000 0.7561 0.000 0.000 1.000 0.000 0.000
#> GSM97058 2 0.0290 0.8540 0.000 0.992 0.000 0.008 0.000
#> GSM97051 2 0.0162 0.8555 0.000 0.996 0.004 0.000 0.000
#> GSM97052 3 0.0000 0.7561 0.000 0.000 1.000 0.000 0.000
#> GSM97061 3 0.4482 0.4944 0.000 0.012 0.612 0.376 0.000
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM97138 1 0.0291 0.943 0.992 0.004 0.000 0.000 0.004 0.000
#> GSM97145 1 0.0000 0.945 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM97147 4 0.1858 0.895 0.004 0.092 0.000 0.904 0.000 0.000
#> GSM97125 1 0.0146 0.944 0.996 0.004 0.000 0.000 0.000 0.000
#> GSM97127 1 0.0000 0.945 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM97130 4 0.3071 0.744 0.180 0.016 0.000 0.804 0.000 0.000
#> GSM97133 1 0.1152 0.906 0.952 0.004 0.000 0.044 0.000 0.000
#> GSM97134 1 0.0146 0.944 0.996 0.004 0.000 0.000 0.000 0.000
#> GSM97120 1 0.0000 0.945 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM97126 1 0.0000 0.945 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM97112 5 0.0000 0.989 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM97115 4 0.0363 0.939 0.000 0.012 0.000 0.988 0.000 0.000
#> GSM97116 1 0.0146 0.944 0.996 0.004 0.000 0.000 0.000 0.000
#> GSM97117 1 0.1863 0.865 0.896 0.104 0.000 0.000 0.000 0.000
#> GSM97119 1 0.0291 0.943 0.992 0.004 0.000 0.000 0.004 0.000
#> GSM97122 1 0.0291 0.943 0.992 0.004 0.000 0.000 0.004 0.000
#> GSM97135 1 0.1285 0.912 0.944 0.004 0.000 0.000 0.052 0.000
#> GSM97136 1 0.5788 0.238 0.488 0.204 0.308 0.000 0.000 0.000
#> GSM97139 1 0.0000 0.945 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM97146 1 0.0000 0.945 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM97123 3 0.3330 0.671 0.000 0.284 0.716 0.000 0.000 0.000
#> GSM97129 1 0.1075 0.916 0.952 0.048 0.000 0.000 0.000 0.000
#> GSM97143 1 0.1958 0.867 0.896 0.004 0.000 0.000 0.100 0.000
#> GSM97113 4 0.0363 0.939 0.000 0.012 0.000 0.988 0.000 0.000
#> GSM97056 1 0.0146 0.943 0.996 0.000 0.000 0.004 0.000 0.000
#> GSM97124 1 0.0146 0.944 0.996 0.004 0.000 0.000 0.000 0.000
#> GSM97132 1 0.0146 0.944 0.996 0.004 0.000 0.000 0.000 0.000
#> GSM97144 4 0.1594 0.900 0.052 0.016 0.000 0.932 0.000 0.000
#> GSM97149 1 0.0260 0.941 0.992 0.008 0.000 0.000 0.000 0.000
#> GSM97068 4 0.0363 0.939 0.000 0.012 0.000 0.988 0.000 0.000
#> GSM97071 2 0.0146 0.835 0.000 0.996 0.000 0.000 0.000 0.004
#> GSM97086 4 0.0000 0.939 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM97103 2 0.0363 0.835 0.000 0.988 0.012 0.000 0.000 0.000
#> GSM97057 4 0.0363 0.939 0.000 0.012 0.000 0.988 0.000 0.000
#> GSM97060 3 0.1267 0.867 0.000 0.060 0.940 0.000 0.000 0.000
#> GSM97075 2 0.0363 0.832 0.000 0.988 0.012 0.000 0.000 0.000
#> GSM97098 2 0.2260 0.761 0.000 0.860 0.140 0.000 0.000 0.000
#> GSM97099 2 0.0146 0.834 0.000 0.996 0.004 0.000 0.000 0.000
#> GSM97101 2 0.2730 0.758 0.000 0.808 0.000 0.192 0.000 0.000
#> GSM97105 4 0.1501 0.902 0.000 0.076 0.000 0.924 0.000 0.000
#> GSM97106 3 0.3699 0.731 0.000 0.212 0.752 0.036 0.000 0.000
#> GSM97121 4 0.3126 0.684 0.000 0.248 0.000 0.752 0.000 0.000
#> GSM97128 5 0.1718 0.927 0.044 0.008 0.016 0.000 0.932 0.000
#> GSM97131 4 0.1501 0.902 0.000 0.076 0.000 0.924 0.000 0.000
#> GSM97137 4 0.1556 0.880 0.080 0.000 0.000 0.920 0.000 0.000
#> GSM97118 1 0.4367 0.398 0.604 0.032 0.000 0.000 0.364 0.000
#> GSM97114 1 0.1075 0.916 0.952 0.048 0.000 0.000 0.000 0.000
#> GSM97142 5 0.0000 0.989 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM97140 2 0.3862 0.130 0.000 0.524 0.000 0.476 0.000 0.000
#> GSM97141 2 0.1663 0.822 0.000 0.912 0.000 0.088 0.000 0.000
#> GSM97055 5 0.0000 0.989 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM97090 4 0.0363 0.939 0.000 0.012 0.000 0.988 0.000 0.000
#> GSM97091 5 0.0000 0.989 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM97148 1 0.0000 0.945 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM97063 5 0.0000 0.989 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM97053 1 0.0000 0.945 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM97066 6 0.0000 0.989 0.000 0.000 0.000 0.000 0.000 1.000
#> GSM97079 4 0.0000 0.939 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM97083 5 0.0000 0.989 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM97084 4 0.0000 0.939 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM97094 2 0.0937 0.825 0.040 0.960 0.000 0.000 0.000 0.000
#> GSM97096 2 0.1714 0.777 0.000 0.908 0.092 0.000 0.000 0.000
#> GSM97097 2 0.2697 0.767 0.000 0.812 0.000 0.188 0.000 0.000
#> GSM97107 2 0.3727 0.473 0.000 0.612 0.000 0.388 0.000 0.000
#> GSM97054 4 0.0000 0.939 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM97062 4 0.0000 0.939 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM97069 6 0.0000 0.989 0.000 0.000 0.000 0.000 0.000 1.000
#> GSM97070 6 0.0000 0.989 0.000 0.000 0.000 0.000 0.000 1.000
#> GSM97073 6 0.0000 0.989 0.000 0.000 0.000 0.000 0.000 1.000
#> GSM97076 6 0.0000 0.989 0.000 0.000 0.000 0.000 0.000 1.000
#> GSM97077 4 0.0363 0.939 0.000 0.012 0.000 0.988 0.000 0.000
#> GSM97095 2 0.2006 0.814 0.004 0.892 0.000 0.104 0.000 0.000
#> GSM97102 3 0.1387 0.858 0.000 0.000 0.932 0.000 0.000 0.068
#> GSM97109 2 0.0458 0.835 0.016 0.984 0.000 0.000 0.000 0.000
#> GSM97110 2 0.1556 0.821 0.000 0.920 0.000 0.080 0.000 0.000
#> GSM97074 6 0.0000 0.989 0.000 0.000 0.000 0.000 0.000 1.000
#> GSM97085 5 0.0146 0.986 0.000 0.000 0.000 0.000 0.996 0.004
#> GSM97059 4 0.1663 0.898 0.000 0.088 0.000 0.912 0.000 0.000
#> GSM97072 6 0.0000 0.989 0.000 0.000 0.000 0.000 0.000 1.000
#> GSM97078 2 0.2315 0.813 0.008 0.892 0.016 0.084 0.000 0.000
#> GSM97067 6 0.0000 0.989 0.000 0.000 0.000 0.000 0.000 1.000
#> GSM97087 3 0.0937 0.876 0.000 0.040 0.960 0.000 0.000 0.000
#> GSM97111 2 0.1444 0.803 0.072 0.928 0.000 0.000 0.000 0.000
#> GSM97064 4 0.2848 0.779 0.000 0.008 0.176 0.816 0.000 0.000
#> GSM97065 6 0.1501 0.906 0.000 0.076 0.000 0.000 0.000 0.924
#> GSM97081 3 0.2762 0.793 0.000 0.196 0.804 0.000 0.000 0.000
#> GSM97082 3 0.0937 0.876 0.000 0.040 0.960 0.000 0.000 0.000
#> GSM97088 3 0.1387 0.849 0.000 0.000 0.932 0.000 0.068 0.000
#> GSM97100 4 0.0000 0.939 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM97104 3 0.1267 0.862 0.000 0.000 0.940 0.000 0.000 0.060
#> GSM97108 2 0.2793 0.752 0.000 0.800 0.000 0.200 0.000 0.000
#> GSM97050 4 0.0000 0.939 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM97080 3 0.1444 0.855 0.000 0.000 0.928 0.000 0.000 0.072
#> GSM97089 3 0.0937 0.876 0.000 0.040 0.960 0.000 0.000 0.000
#> GSM97092 3 0.0000 0.871 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM97093 3 0.2597 0.823 0.000 0.176 0.824 0.000 0.000 0.000
#> GSM97058 4 0.1387 0.913 0.000 0.068 0.000 0.932 0.000 0.000
#> GSM97051 4 0.0000 0.939 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM97052 3 0.0146 0.872 0.000 0.004 0.996 0.000 0.000 0.000
#> GSM97061 3 0.3446 0.634 0.000 0.308 0.692 0.000 0.000 0.000
Heatmaps for the consensus matrix. It visualizes the probability of two samples to be in a same group.
consensus_heatmap(res, k = 2)
consensus_heatmap(res, k = 3)
consensus_heatmap(res, k = 4)
consensus_heatmap(res, k = 5)
consensus_heatmap(res, k = 6)
Heatmaps for the membership of samples in all partitions to see how consistent they are:
membership_heatmap(res, k = 2)
membership_heatmap(res, k = 3)
membership_heatmap(res, k = 4)
membership_heatmap(res, k = 5)
membership_heatmap(res, k = 6)
As soon as we have had the classes for columns, we can look for signatures which are significantly different between classes which can be candidate marks for certain classes. Following are the heatmaps for signatures.
Signature heatmaps where rows are scaled:
get_signatures(res, k = 2)
get_signatures(res, k = 3)
get_signatures(res, k = 4)
get_signatures(res, k = 5)
get_signatures(res, k = 6)
Signature heatmaps where rows are not scaled:
get_signatures(res, k = 2, scale_rows = FALSE)
get_signatures(res, k = 3, scale_rows = FALSE)
get_signatures(res, k = 4, scale_rows = FALSE)
get_signatures(res, k = 5, scale_rows = FALSE)
get_signatures(res, k = 6, scale_rows = FALSE)
Compare the overlap of signatures from different k:
compare_signatures(res)
get_signature() returns a data frame invisibly. TO get the list of signatures, the function
call should be assigned to a variable explicitly. In following code, if plot argument is set
to FALSE, no heatmap is plotted while only the differential analysis is performed.
# code only for demonstration
tb = get_signature(res, k = ..., plot = FALSE)
An example of the output of tb is:
#> which_row fdr mean_1 mean_2 scaled_mean_1 scaled_mean_2 km
#> 1 38 0.042760348 8.373488 9.131774 -0.5533452 0.5164555 1
#> 2 40 0.018707592 7.106213 8.469186 -0.6173731 0.5762149 1
#> 3 55 0.019134737 10.221463 11.207825 -0.6159697 0.5749050 1
#> 4 59 0.006059896 5.921854 7.869574 -0.6899429 0.6439467 1
#> 5 60 0.018055526 8.928898 10.211722 -0.6204761 0.5791110 1
#> 6 98 0.009384629 15.714769 14.887706 0.6635654 -0.6193277 2
...
The columns in tb are:
which_row: row indices corresponding to the input matrix.fdr: FDR for the differential test. mean_x: The mean value in group x.scaled_mean_x: The mean value in group x after rows are scaled.km: Row groups if k-means clustering is applied to rows.UMAP plot which shows how samples are separated.
dimension_reduction(res, k = 2, method = "UMAP")
dimension_reduction(res, k = 3, method = "UMAP")
dimension_reduction(res, k = 4, method = "UMAP")
dimension_reduction(res, k = 5, method = "UMAP")
dimension_reduction(res, k = 6, method = "UMAP")
Following heatmap shows how subgroups are split when increasing k:
collect_classes(res)
Test correlation between subgroups and known annotations. If the known annotation is numeric, one-way ANOVA test is applied, and if the known annotation is discrete, chi-squared contingency table test is applied.
test_to_known_factors(res)
#> n disease.state(p) specimen(p) cell.type(p) other(p) k
#> ATC:pam 99 6.38e-05 0.842 2.96e-13 0.0823 2
#> ATC:pam 96 3.29e-03 0.458 1.35e-09 0.1412 3
#> ATC:pam 92 1.73e-05 0.199 2.61e-15 0.0358 4
#> ATC:pam 75 1.08e-04 0.334 6.07e-12 0.2391 5
#> ATC:pam 96 2.10e-06 0.212 4.71e-15 0.0833 6
If matrix rows can be associated to genes, consider to use functional_enrichment(res,
...) to perform function enrichment for the signature genes. See this vignette for more detailed explanations.
The object with results only for a single top-value method and a single partition method can be extracted as:
res = res_list["ATC", "mclust"]
# you can also extract it by
# res = res_list["ATC:mclust"]
A summary of res and all the functions that can be applied to it:
res
#> A 'ConsensusPartition' object with k = 2, 3, 4, 5, 6.
#> On a matrix with 21512 rows and 100 columns.
#> Top rows (1000, 2000, 3000, 4000, 5000) are extracted by 'ATC' method.
#> Subgroups are detected by 'mclust' method.
#> Performed in total 1250 partitions by row resampling.
#> Best k for subgroups seems to be 4.
#>
#> Following methods can be applied to this 'ConsensusPartition' object:
#> [1] "cola_report" "collect_classes" "collect_plots"
#> [4] "collect_stats" "colnames" "compare_signatures"
#> [7] "consensus_heatmap" "dimension_reduction" "functional_enrichment"
#> [10] "get_anno_col" "get_anno" "get_classes"
#> [13] "get_consensus" "get_matrix" "get_membership"
#> [16] "get_param" "get_signatures" "get_stats"
#> [19] "is_best_k" "is_stable_k" "membership_heatmap"
#> [22] "ncol" "nrow" "plot_ecdf"
#> [25] "rownames" "select_partition_number" "show"
#> [28] "suggest_best_k" "test_to_known_factors"
collect_plots() function collects all the plots made from res for all k (number of partitions)
into one single page to provide an easy and fast comparison between different k.
collect_plots(res)
The plots are:
k and the heatmap of
predicted classes for each k.k.k.k.All the plots in panels can be made by individual functions and they are plotted later in this section.
select_partition_number() produces several plots showing different
statistics for choosing “optimized” k. There are following statistics:
k;k, the area increased is defined as \(A_k - A_{k-1}\).The detailed explanations of these statistics can be found in the cola vignette.
Generally speaking, lower PAC score, higher mean silhouette score or higher
concordance corresponds to better partition. Rand index and Jaccard index
measure how similar the current partition is compared to partition with k-1.
If they are too similar, we won't accept k is better than k-1.
select_partition_number(res)
The numeric values for all these statistics can be obtained by get_stats().
get_stats(res)
#> k 1-PAC mean_silhouette concordance area_increased Rand Jaccard
#> 2 2 0.599 0.822 0.911 0.2817 0.802 0.802
#> 3 3 0.581 0.885 0.917 1.0116 0.602 0.510
#> 4 4 0.936 0.937 0.974 0.2631 0.806 0.570
#> 5 5 0.785 0.763 0.872 0.0631 0.964 0.877
#> 6 6 0.887 0.833 0.924 0.0671 0.871 0.553
suggest_best_k() suggests the best \(k\) based on these statistics. The rules are as follows:
suggest_best_k(res)
#> [1] 4
Following shows the table of the partitions (You need to click the show/hide
code output link to see it). The membership matrix (columns with name p*)
is inferred by
clue::cl_consensus()
function with the SE method. Basically the value in the membership matrix
represents the probability to belong to a certain group. The finall class
label for an item is determined with the group with highest probability it
belongs to.
In get_classes() function, the entropy is calculated from the membership
matrix and the silhouette score is calculated from the consensus matrix.
cbind(get_classes(res, k = 2), get_membership(res, k = 2))
#> class entropy silhouette p1 p2
#> GSM97138 2 0.0000 0.885 0.000 1.000
#> GSM97145 2 0.0000 0.885 0.000 1.000
#> GSM97147 2 0.0000 0.885 0.000 1.000
#> GSM97125 2 0.0000 0.885 0.000 1.000
#> GSM97127 2 0.0000 0.885 0.000 1.000
#> GSM97130 2 0.0000 0.885 0.000 1.000
#> GSM97133 2 0.0000 0.885 0.000 1.000
#> GSM97134 2 0.0000 0.885 0.000 1.000
#> GSM97120 2 0.0000 0.885 0.000 1.000
#> GSM97126 2 0.0000 0.885 0.000 1.000
#> GSM97112 2 0.0000 0.885 0.000 1.000
#> GSM97115 2 0.0000 0.885 0.000 1.000
#> GSM97116 2 0.0000 0.885 0.000 1.000
#> GSM97117 2 0.0000 0.885 0.000 1.000
#> GSM97119 2 0.0000 0.885 0.000 1.000
#> GSM97122 2 0.0000 0.885 0.000 1.000
#> GSM97135 2 0.0000 0.885 0.000 1.000
#> GSM97136 2 0.8909 0.663 0.308 0.692
#> GSM97139 2 0.0000 0.885 0.000 1.000
#> GSM97146 2 0.0000 0.885 0.000 1.000
#> GSM97123 2 0.9000 0.653 0.316 0.684
#> GSM97129 2 0.0000 0.885 0.000 1.000
#> GSM97143 2 0.0000 0.885 0.000 1.000
#> GSM97113 2 0.0000 0.885 0.000 1.000
#> GSM97056 2 0.0000 0.885 0.000 1.000
#> GSM97124 2 0.0000 0.885 0.000 1.000
#> GSM97132 2 0.0000 0.885 0.000 1.000
#> GSM97144 2 0.0000 0.885 0.000 1.000
#> GSM97149 2 0.0000 0.885 0.000 1.000
#> GSM97068 2 0.0000 0.885 0.000 1.000
#> GSM97071 1 0.0000 0.956 1.000 0.000
#> GSM97086 2 0.0000 0.885 0.000 1.000
#> GSM97103 2 0.9170 0.632 0.332 0.668
#> GSM97057 2 0.0000 0.885 0.000 1.000
#> GSM97060 2 0.9170 0.632 0.332 0.668
#> GSM97075 2 0.0672 0.881 0.008 0.992
#> GSM97098 2 0.9087 0.643 0.324 0.676
#> GSM97099 2 0.0000 0.885 0.000 1.000
#> GSM97101 2 0.0000 0.885 0.000 1.000
#> GSM97105 2 0.0000 0.885 0.000 1.000
#> GSM97106 2 0.9170 0.632 0.332 0.668
#> GSM97121 2 0.0000 0.885 0.000 1.000
#> GSM97128 2 0.8909 0.663 0.308 0.692
#> GSM97131 2 0.0000 0.885 0.000 1.000
#> GSM97137 2 0.0000 0.885 0.000 1.000
#> GSM97118 2 0.8909 0.663 0.308 0.692
#> GSM97114 2 0.0000 0.885 0.000 1.000
#> GSM97142 2 0.0000 0.885 0.000 1.000
#> GSM97140 2 0.0000 0.885 0.000 1.000
#> GSM97141 2 0.0000 0.885 0.000 1.000
#> GSM97055 2 0.8909 0.663 0.308 0.692
#> GSM97090 2 0.0000 0.885 0.000 1.000
#> GSM97091 2 0.8909 0.663 0.308 0.692
#> GSM97148 2 0.0000 0.885 0.000 1.000
#> GSM97063 2 0.8909 0.663 0.308 0.692
#> GSM97053 2 0.0000 0.885 0.000 1.000
#> GSM97066 1 0.0000 0.956 1.000 0.000
#> GSM97079 2 0.0000 0.885 0.000 1.000
#> GSM97083 2 0.8909 0.663 0.308 0.692
#> GSM97084 2 0.0000 0.885 0.000 1.000
#> GSM97094 2 0.0000 0.885 0.000 1.000
#> GSM97096 2 0.8909 0.663 0.308 0.692
#> GSM97097 2 0.0000 0.885 0.000 1.000
#> GSM97107 2 0.0000 0.885 0.000 1.000
#> GSM97054 2 0.0000 0.885 0.000 1.000
#> GSM97062 2 0.0000 0.885 0.000 1.000
#> GSM97069 1 0.0000 0.956 1.000 0.000
#> GSM97070 1 0.0000 0.956 1.000 0.000
#> GSM97073 1 0.0000 0.956 1.000 0.000
#> GSM97076 1 0.0000 0.956 1.000 0.000
#> GSM97077 2 0.0000 0.885 0.000 1.000
#> GSM97095 2 0.0000 0.885 0.000 1.000
#> GSM97102 2 0.9170 0.632 0.332 0.668
#> GSM97109 2 0.0000 0.885 0.000 1.000
#> GSM97110 2 0.0000 0.885 0.000 1.000
#> GSM97074 1 0.0000 0.956 1.000 0.000
#> GSM97085 2 0.8909 0.663 0.308 0.692
#> GSM97059 2 0.0000 0.885 0.000 1.000
#> GSM97072 1 0.0000 0.956 1.000 0.000
#> GSM97078 2 0.8909 0.663 0.308 0.692
#> GSM97067 1 0.0000 0.956 1.000 0.000
#> GSM97087 2 0.9170 0.632 0.332 0.668
#> GSM97111 2 0.0000 0.885 0.000 1.000
#> GSM97064 2 0.8909 0.663 0.308 0.692
#> GSM97065 1 0.0000 0.956 1.000 0.000
#> GSM97081 2 0.8909 0.663 0.308 0.692
#> GSM97082 2 0.9170 0.632 0.332 0.668
#> GSM97088 2 0.8909 0.663 0.308 0.692
#> GSM97100 2 0.0000 0.885 0.000 1.000
#> GSM97104 2 0.9170 0.632 0.332 0.668
#> GSM97108 2 0.0000 0.885 0.000 1.000
#> GSM97050 2 0.0000 0.885 0.000 1.000
#> GSM97080 1 0.9427 0.247 0.640 0.360
#> GSM97089 2 0.9044 0.648 0.320 0.680
#> GSM97092 2 0.9170 0.632 0.332 0.668
#> GSM97093 2 0.8909 0.663 0.308 0.692
#> GSM97058 2 0.0000 0.885 0.000 1.000
#> GSM97051 2 0.0000 0.885 0.000 1.000
#> GSM97052 2 0.9170 0.632 0.332 0.668
#> GSM97061 2 0.8909 0.663 0.308 0.692
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM97138 1 0.3686 0.940 0.860 0.140 0.000
#> GSM97145 1 0.3686 0.940 0.860 0.140 0.000
#> GSM97147 2 0.2066 0.873 0.060 0.940 0.000
#> GSM97125 1 0.3686 0.940 0.860 0.140 0.000
#> GSM97127 1 0.3686 0.940 0.860 0.140 0.000
#> GSM97130 1 0.3686 0.940 0.860 0.140 0.000
#> GSM97133 1 0.3686 0.940 0.860 0.140 0.000
#> GSM97134 1 0.3686 0.940 0.860 0.140 0.000
#> GSM97120 1 0.3686 0.940 0.860 0.140 0.000
#> GSM97126 1 0.3752 0.937 0.856 0.144 0.000
#> GSM97112 1 0.3686 0.940 0.860 0.140 0.000
#> GSM97115 2 0.3482 0.792 0.128 0.872 0.000
#> GSM97116 1 0.3686 0.940 0.860 0.140 0.000
#> GSM97117 1 0.4291 0.902 0.820 0.180 0.000
#> GSM97119 1 0.3686 0.940 0.860 0.140 0.000
#> GSM97122 1 0.3686 0.940 0.860 0.140 0.000
#> GSM97135 1 0.3686 0.940 0.860 0.140 0.000
#> GSM97136 1 0.7004 -0.108 0.552 0.428 0.020
#> GSM97139 1 0.3686 0.940 0.860 0.140 0.000
#> GSM97146 1 0.3686 0.940 0.860 0.140 0.000
#> GSM97123 2 0.4551 0.848 0.140 0.840 0.020
#> GSM97129 2 0.5678 0.401 0.316 0.684 0.000
#> GSM97143 1 0.3686 0.940 0.860 0.140 0.000
#> GSM97113 2 0.0000 0.913 0.000 1.000 0.000
#> GSM97056 1 0.3686 0.940 0.860 0.140 0.000
#> GSM97124 1 0.3686 0.940 0.860 0.140 0.000
#> GSM97132 1 0.3686 0.940 0.860 0.140 0.000
#> GSM97144 1 0.3686 0.940 0.860 0.140 0.000
#> GSM97149 1 0.3686 0.940 0.860 0.140 0.000
#> GSM97068 2 0.0000 0.913 0.000 1.000 0.000
#> GSM97071 3 0.0000 1.000 0.000 0.000 1.000
#> GSM97086 2 0.0000 0.913 0.000 1.000 0.000
#> GSM97103 2 0.0892 0.909 0.000 0.980 0.020
#> GSM97057 2 0.0000 0.913 0.000 1.000 0.000
#> GSM97060 2 0.4551 0.848 0.140 0.840 0.020
#> GSM97075 2 0.0661 0.912 0.004 0.988 0.008
#> GSM97098 2 0.1129 0.909 0.004 0.976 0.020
#> GSM97099 2 0.0000 0.913 0.000 1.000 0.000
#> GSM97101 2 0.0000 0.913 0.000 1.000 0.000
#> GSM97105 2 0.0000 0.913 0.000 1.000 0.000
#> GSM97106 2 0.4551 0.848 0.140 0.840 0.020
#> GSM97121 2 0.0424 0.911 0.008 0.992 0.000
#> GSM97128 1 0.4418 0.662 0.848 0.132 0.020
#> GSM97131 2 0.0000 0.913 0.000 1.000 0.000
#> GSM97137 1 0.3686 0.940 0.860 0.140 0.000
#> GSM97118 1 0.4099 0.934 0.852 0.140 0.008
#> GSM97114 1 0.5397 0.774 0.720 0.280 0.000
#> GSM97142 1 0.3686 0.940 0.860 0.140 0.000
#> GSM97140 2 0.0424 0.911 0.008 0.992 0.000
#> GSM97141 2 0.0237 0.912 0.004 0.996 0.000
#> GSM97055 1 0.0892 0.768 0.980 0.000 0.020
#> GSM97090 2 0.4887 0.620 0.228 0.772 0.000
#> GSM97091 1 0.0892 0.768 0.980 0.000 0.020
#> GSM97148 1 0.3686 0.940 0.860 0.140 0.000
#> GSM97063 1 0.0892 0.768 0.980 0.000 0.020
#> GSM97053 1 0.3686 0.940 0.860 0.140 0.000
#> GSM97066 3 0.0000 1.000 0.000 0.000 1.000
#> GSM97079 2 0.0000 0.913 0.000 1.000 0.000
#> GSM97083 1 0.0892 0.768 0.980 0.000 0.020
#> GSM97084 2 0.0000 0.913 0.000 1.000 0.000
#> GSM97094 1 0.3816 0.934 0.852 0.148 0.000
#> GSM97096 2 0.4551 0.848 0.140 0.840 0.020
#> GSM97097 2 0.0000 0.913 0.000 1.000 0.000
#> GSM97107 2 0.0000 0.913 0.000 1.000 0.000
#> GSM97054 2 0.0000 0.913 0.000 1.000 0.000
#> GSM97062 2 0.0000 0.913 0.000 1.000 0.000
#> GSM97069 3 0.0000 1.000 0.000 0.000 1.000
#> GSM97070 3 0.0000 1.000 0.000 0.000 1.000
#> GSM97073 3 0.0000 1.000 0.000 0.000 1.000
#> GSM97076 3 0.0000 1.000 0.000 0.000 1.000
#> GSM97077 2 0.0000 0.913 0.000 1.000 0.000
#> GSM97095 2 0.0592 0.909 0.012 0.988 0.000
#> GSM97102 2 0.4731 0.848 0.128 0.840 0.032
#> GSM97109 2 0.0424 0.911 0.008 0.992 0.000
#> GSM97110 2 0.0000 0.913 0.000 1.000 0.000
#> GSM97074 3 0.0000 1.000 0.000 0.000 1.000
#> GSM97085 2 0.6832 0.589 0.376 0.604 0.020
#> GSM97059 2 0.1163 0.899 0.028 0.972 0.000
#> GSM97072 3 0.0000 1.000 0.000 0.000 1.000
#> GSM97078 2 0.0892 0.909 0.000 0.980 0.020
#> GSM97067 3 0.0000 1.000 0.000 0.000 1.000
#> GSM97087 2 0.4731 0.848 0.128 0.840 0.032
#> GSM97111 2 0.1163 0.899 0.028 0.972 0.000
#> GSM97064 2 0.4485 0.851 0.136 0.844 0.020
#> GSM97065 3 0.0000 1.000 0.000 0.000 1.000
#> GSM97081 2 0.4551 0.848 0.140 0.840 0.020
#> GSM97082 2 0.4551 0.848 0.140 0.840 0.020
#> GSM97088 2 0.4862 0.841 0.160 0.820 0.020
#> GSM97100 2 0.0000 0.913 0.000 1.000 0.000
#> GSM97104 2 0.4731 0.848 0.128 0.840 0.032
#> GSM97108 2 0.0000 0.913 0.000 1.000 0.000
#> GSM97050 2 0.0000 0.913 0.000 1.000 0.000
#> GSM97080 2 0.4731 0.848 0.128 0.840 0.032
#> GSM97089 2 0.4551 0.848 0.140 0.840 0.020
#> GSM97092 2 0.4551 0.848 0.140 0.840 0.020
#> GSM97093 2 0.4551 0.848 0.140 0.840 0.020
#> GSM97058 2 0.0000 0.913 0.000 1.000 0.000
#> GSM97051 2 0.0000 0.913 0.000 1.000 0.000
#> GSM97052 2 0.4551 0.848 0.140 0.840 0.020
#> GSM97061 2 0.4551 0.848 0.140 0.840 0.020
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM97138 1 0.0000 0.957 1.000 0.000 0.000 0
#> GSM97145 1 0.0000 0.957 1.000 0.000 0.000 0
#> GSM97147 2 0.2589 0.868 0.116 0.884 0.000 0
#> GSM97125 1 0.0000 0.957 1.000 0.000 0.000 0
#> GSM97127 1 0.0000 0.957 1.000 0.000 0.000 0
#> GSM97130 1 0.0000 0.957 1.000 0.000 0.000 0
#> GSM97133 1 0.0000 0.957 1.000 0.000 0.000 0
#> GSM97134 1 0.4790 0.365 0.620 0.380 0.000 0
#> GSM97120 1 0.0000 0.957 1.000 0.000 0.000 0
#> GSM97126 1 0.0000 0.957 1.000 0.000 0.000 0
#> GSM97112 1 0.0000 0.957 1.000 0.000 0.000 0
#> GSM97115 2 0.2216 0.890 0.092 0.908 0.000 0
#> GSM97116 1 0.0000 0.957 1.000 0.000 0.000 0
#> GSM97117 1 0.4382 0.569 0.704 0.296 0.000 0
#> GSM97119 1 0.0000 0.957 1.000 0.000 0.000 0
#> GSM97122 1 0.0000 0.957 1.000 0.000 0.000 0
#> GSM97135 1 0.0000 0.957 1.000 0.000 0.000 0
#> GSM97136 3 0.0000 0.982 0.000 0.000 1.000 0
#> GSM97139 1 0.0000 0.957 1.000 0.000 0.000 0
#> GSM97146 1 0.0000 0.957 1.000 0.000 0.000 0
#> GSM97123 3 0.0000 0.982 0.000 0.000 1.000 0
#> GSM97129 2 0.3024 0.835 0.148 0.852 0.000 0
#> GSM97143 1 0.0000 0.957 1.000 0.000 0.000 0
#> GSM97113 2 0.0000 0.958 0.000 1.000 0.000 0
#> GSM97056 1 0.0000 0.957 1.000 0.000 0.000 0
#> GSM97124 1 0.0000 0.957 1.000 0.000 0.000 0
#> GSM97132 1 0.0000 0.957 1.000 0.000 0.000 0
#> GSM97144 1 0.0000 0.957 1.000 0.000 0.000 0
#> GSM97149 1 0.0000 0.957 1.000 0.000 0.000 0
#> GSM97068 2 0.0000 0.958 0.000 1.000 0.000 0
#> GSM97071 4 0.0000 1.000 0.000 0.000 0.000 1
#> GSM97086 2 0.0000 0.958 0.000 1.000 0.000 0
#> GSM97103 2 0.0000 0.958 0.000 1.000 0.000 0
#> GSM97057 2 0.0000 0.958 0.000 1.000 0.000 0
#> GSM97060 3 0.0000 0.982 0.000 0.000 1.000 0
#> GSM97075 2 0.0000 0.958 0.000 1.000 0.000 0
#> GSM97098 2 0.0000 0.958 0.000 1.000 0.000 0
#> GSM97099 2 0.0000 0.958 0.000 1.000 0.000 0
#> GSM97101 2 0.0000 0.958 0.000 1.000 0.000 0
#> GSM97105 2 0.0000 0.958 0.000 1.000 0.000 0
#> GSM97106 3 0.0000 0.982 0.000 0.000 1.000 0
#> GSM97121 2 0.0000 0.958 0.000 1.000 0.000 0
#> GSM97128 3 0.0469 0.967 0.012 0.000 0.988 0
#> GSM97131 2 0.0000 0.958 0.000 1.000 0.000 0
#> GSM97137 1 0.0000 0.957 1.000 0.000 0.000 0
#> GSM97118 1 0.0000 0.957 1.000 0.000 0.000 0
#> GSM97114 2 0.3074 0.831 0.152 0.848 0.000 0
#> GSM97142 1 0.0000 0.957 1.000 0.000 0.000 0
#> GSM97140 2 0.1022 0.936 0.032 0.968 0.000 0
#> GSM97141 2 0.0000 0.958 0.000 1.000 0.000 0
#> GSM97055 3 0.0000 0.982 0.000 0.000 1.000 0
#> GSM97090 2 0.4746 0.444 0.368 0.632 0.000 0
#> GSM97091 3 0.4406 0.559 0.300 0.000 0.700 0
#> GSM97148 1 0.0000 0.957 1.000 0.000 0.000 0
#> GSM97063 1 0.1557 0.899 0.944 0.000 0.056 0
#> GSM97053 1 0.0000 0.957 1.000 0.000 0.000 0
#> GSM97066 4 0.0000 1.000 0.000 0.000 0.000 1
#> GSM97079 2 0.0000 0.958 0.000 1.000 0.000 0
#> GSM97083 3 0.0000 0.982 0.000 0.000 1.000 0
#> GSM97084 2 0.0000 0.958 0.000 1.000 0.000 0
#> GSM97094 1 0.3942 0.671 0.764 0.236 0.000 0
#> GSM97096 3 0.0000 0.982 0.000 0.000 1.000 0
#> GSM97097 2 0.0000 0.958 0.000 1.000 0.000 0
#> GSM97107 2 0.0000 0.958 0.000 1.000 0.000 0
#> GSM97054 2 0.0000 0.958 0.000 1.000 0.000 0
#> GSM97062 2 0.0000 0.958 0.000 1.000 0.000 0
#> GSM97069 4 0.0000 1.000 0.000 0.000 0.000 1
#> GSM97070 4 0.0000 1.000 0.000 0.000 0.000 1
#> GSM97073 4 0.0000 1.000 0.000 0.000 0.000 1
#> GSM97076 4 0.0000 1.000 0.000 0.000 0.000 1
#> GSM97077 2 0.0000 0.958 0.000 1.000 0.000 0
#> GSM97095 2 0.1867 0.907 0.072 0.928 0.000 0
#> GSM97102 3 0.0000 0.982 0.000 0.000 1.000 0
#> GSM97109 2 0.0000 0.958 0.000 1.000 0.000 0
#> GSM97110 2 0.0000 0.958 0.000 1.000 0.000 0
#> GSM97074 4 0.0000 1.000 0.000 0.000 0.000 1
#> GSM97085 3 0.0000 0.982 0.000 0.000 1.000 0
#> GSM97059 2 0.3311 0.807 0.172 0.828 0.000 0
#> GSM97072 4 0.0000 1.000 0.000 0.000 0.000 1
#> GSM97078 2 0.0000 0.958 0.000 1.000 0.000 0
#> GSM97067 4 0.0000 1.000 0.000 0.000 0.000 1
#> GSM97087 3 0.0000 0.982 0.000 0.000 1.000 0
#> GSM97111 2 0.2921 0.844 0.140 0.860 0.000 0
#> GSM97064 3 0.0000 0.982 0.000 0.000 1.000 0
#> GSM97065 4 0.0000 1.000 0.000 0.000 0.000 1
#> GSM97081 3 0.0000 0.982 0.000 0.000 1.000 0
#> GSM97082 3 0.0000 0.982 0.000 0.000 1.000 0
#> GSM97088 3 0.0000 0.982 0.000 0.000 1.000 0
#> GSM97100 2 0.0000 0.958 0.000 1.000 0.000 0
#> GSM97104 3 0.0000 0.982 0.000 0.000 1.000 0
#> GSM97108 2 0.0000 0.958 0.000 1.000 0.000 0
#> GSM97050 2 0.0000 0.958 0.000 1.000 0.000 0
#> GSM97080 3 0.0000 0.982 0.000 0.000 1.000 0
#> GSM97089 3 0.0000 0.982 0.000 0.000 1.000 0
#> GSM97092 3 0.0000 0.982 0.000 0.000 1.000 0
#> GSM97093 3 0.0000 0.982 0.000 0.000 1.000 0
#> GSM97058 2 0.0000 0.958 0.000 1.000 0.000 0
#> GSM97051 2 0.0000 0.958 0.000 1.000 0.000 0
#> GSM97052 3 0.0000 0.982 0.000 0.000 1.000 0
#> GSM97061 3 0.0000 0.982 0.000 0.000 1.000 0
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM97138 1 0.0000 0.8570 1.000 0.000 0.000 0.000 0.000
#> GSM97145 1 0.0000 0.8570 1.000 0.000 0.000 0.000 0.000
#> GSM97147 2 0.5671 0.6292 0.096 0.568 0.000 0.000 0.336
#> GSM97125 1 0.0000 0.8570 1.000 0.000 0.000 0.000 0.000
#> GSM97127 1 0.0162 0.8560 0.996 0.000 0.000 0.000 0.004
#> GSM97130 1 0.2561 0.7423 0.856 0.000 0.000 0.000 0.144
#> GSM97133 1 0.0404 0.8517 0.988 0.000 0.000 0.000 0.012
#> GSM97134 1 0.4313 0.6054 0.732 0.040 0.000 0.000 0.228
#> GSM97120 1 0.0000 0.8570 1.000 0.000 0.000 0.000 0.000
#> GSM97126 1 0.0000 0.8570 1.000 0.000 0.000 0.000 0.000
#> GSM97112 1 0.3957 0.2838 0.712 0.000 0.008 0.000 0.280
#> GSM97115 2 0.6466 0.4886 0.204 0.480 0.000 0.000 0.316
#> GSM97116 1 0.0000 0.8570 1.000 0.000 0.000 0.000 0.000
#> GSM97117 1 0.6054 0.3508 0.560 0.160 0.000 0.000 0.280
#> GSM97119 1 0.0000 0.8570 1.000 0.000 0.000 0.000 0.000
#> GSM97122 1 0.0000 0.8570 1.000 0.000 0.000 0.000 0.000
#> GSM97135 1 0.0000 0.8570 1.000 0.000 0.000 0.000 0.000
#> GSM97136 3 0.3741 0.5726 0.004 0.000 0.732 0.000 0.264
#> GSM97139 1 0.0000 0.8570 1.000 0.000 0.000 0.000 0.000
#> GSM97146 1 0.0000 0.8570 1.000 0.000 0.000 0.000 0.000
#> GSM97123 3 0.0000 0.8920 0.000 0.000 1.000 0.000 0.000
#> GSM97129 1 0.6739 0.0428 0.400 0.264 0.000 0.000 0.336
#> GSM97143 1 0.0579 0.8437 0.984 0.000 0.008 0.000 0.008
#> GSM97113 2 0.2179 0.7676 0.000 0.888 0.000 0.000 0.112
#> GSM97056 1 0.0162 0.8560 0.996 0.000 0.000 0.000 0.004
#> GSM97124 1 0.0000 0.8570 1.000 0.000 0.000 0.000 0.000
#> GSM97132 1 0.0290 0.8542 0.992 0.000 0.000 0.000 0.008
#> GSM97144 1 0.3048 0.7079 0.820 0.004 0.000 0.000 0.176
#> GSM97149 1 0.0162 0.8560 0.996 0.000 0.000 0.000 0.004
#> GSM97068 2 0.0963 0.8096 0.000 0.964 0.000 0.000 0.036
#> GSM97071 4 0.0000 0.9981 0.000 0.000 0.000 1.000 0.000
#> GSM97086 2 0.1965 0.7769 0.000 0.904 0.000 0.000 0.096
#> GSM97103 2 0.2136 0.7771 0.000 0.904 0.008 0.000 0.088
#> GSM97057 2 0.3177 0.7819 0.000 0.792 0.000 0.000 0.208
#> GSM97060 3 0.0000 0.8920 0.000 0.000 1.000 0.000 0.000
#> GSM97075 2 0.4088 0.7320 0.008 0.688 0.000 0.000 0.304
#> GSM97098 2 0.1648 0.7870 0.000 0.940 0.040 0.000 0.020
#> GSM97099 2 0.0703 0.8087 0.000 0.976 0.000 0.000 0.024
#> GSM97101 2 0.3109 0.7851 0.000 0.800 0.000 0.000 0.200
#> GSM97105 2 0.1410 0.7911 0.000 0.940 0.000 0.000 0.060
#> GSM97106 3 0.0000 0.8920 0.000 0.000 1.000 0.000 0.000
#> GSM97121 2 0.3521 0.7701 0.004 0.764 0.000 0.000 0.232
#> GSM97128 3 0.5131 0.3844 0.048 0.008 0.648 0.000 0.296
#> GSM97131 2 0.2280 0.7635 0.000 0.880 0.000 0.000 0.120
#> GSM97137 1 0.2970 0.7158 0.828 0.004 0.000 0.000 0.168
#> GSM97118 1 0.1082 0.8211 0.964 0.000 0.028 0.000 0.008
#> GSM97114 1 0.5678 0.4171 0.600 0.116 0.000 0.000 0.284
#> GSM97142 1 0.0000 0.8570 1.000 0.000 0.000 0.000 0.000
#> GSM97140 2 0.4045 0.7048 0.000 0.644 0.000 0.000 0.356
#> GSM97141 2 0.4201 0.7158 0.008 0.664 0.000 0.000 0.328
#> GSM97055 3 0.4307 -0.0121 0.000 0.000 0.504 0.000 0.496
#> GSM97090 2 0.6779 0.3152 0.288 0.388 0.000 0.000 0.324
#> GSM97091 5 0.6107 0.1483 0.132 0.000 0.372 0.000 0.496
#> GSM97148 1 0.0000 0.8570 1.000 0.000 0.000 0.000 0.000
#> GSM97063 5 0.5498 0.3133 0.440 0.000 0.064 0.000 0.496
#> GSM97053 1 0.0000 0.8570 1.000 0.000 0.000 0.000 0.000
#> GSM97066 4 0.0000 0.9981 0.000 0.000 0.000 1.000 0.000
#> GSM97079 2 0.2424 0.7544 0.000 0.868 0.000 0.000 0.132
#> GSM97083 3 0.4262 0.1767 0.000 0.000 0.560 0.000 0.440
#> GSM97084 2 0.1544 0.7886 0.000 0.932 0.000 0.000 0.068
#> GSM97094 1 0.4588 0.5897 0.720 0.060 0.000 0.000 0.220
#> GSM97096 3 0.0000 0.8920 0.000 0.000 1.000 0.000 0.000
#> GSM97097 2 0.2516 0.7492 0.000 0.860 0.000 0.000 0.140
#> GSM97107 2 0.2249 0.7982 0.000 0.896 0.000 0.008 0.096
#> GSM97054 2 0.2020 0.8069 0.000 0.900 0.000 0.000 0.100
#> GSM97062 2 0.2280 0.7627 0.000 0.880 0.000 0.000 0.120
#> GSM97069 4 0.0404 0.9823 0.000 0.000 0.012 0.988 0.000
#> GSM97070 4 0.0000 0.9981 0.000 0.000 0.000 1.000 0.000
#> GSM97073 4 0.0000 0.9981 0.000 0.000 0.000 1.000 0.000
#> GSM97076 4 0.0000 0.9981 0.000 0.000 0.000 1.000 0.000
#> GSM97077 2 0.1608 0.8093 0.000 0.928 0.000 0.000 0.072
#> GSM97095 2 0.5822 0.6076 0.108 0.548 0.000 0.000 0.344
#> GSM97102 3 0.0000 0.8920 0.000 0.000 1.000 0.000 0.000
#> GSM97109 2 0.3913 0.7231 0.000 0.676 0.000 0.000 0.324
#> GSM97110 2 0.0290 0.8067 0.000 0.992 0.000 0.000 0.008
#> GSM97074 4 0.0000 0.9981 0.000 0.000 0.000 1.000 0.000
#> GSM97085 3 0.3424 0.6204 0.000 0.000 0.760 0.000 0.240
#> GSM97059 2 0.5989 0.5925 0.128 0.536 0.000 0.000 0.336
#> GSM97072 4 0.0000 0.9981 0.000 0.000 0.000 1.000 0.000
#> GSM97078 2 0.2732 0.7968 0.000 0.840 0.000 0.000 0.160
#> GSM97067 4 0.0000 0.9981 0.000 0.000 0.000 1.000 0.000
#> GSM97087 3 0.0000 0.8920 0.000 0.000 1.000 0.000 0.000
#> GSM97111 2 0.6394 0.5042 0.180 0.476 0.000 0.000 0.344
#> GSM97064 3 0.0290 0.8818 0.000 0.008 0.992 0.000 0.000
#> GSM97065 4 0.0000 0.9981 0.000 0.000 0.000 1.000 0.000
#> GSM97081 3 0.0000 0.8920 0.000 0.000 1.000 0.000 0.000
#> GSM97082 3 0.0000 0.8920 0.000 0.000 1.000 0.000 0.000
#> GSM97088 3 0.0000 0.8920 0.000 0.000 1.000 0.000 0.000
#> GSM97100 2 0.0510 0.8031 0.000 0.984 0.000 0.000 0.016
#> GSM97104 3 0.0000 0.8920 0.000 0.000 1.000 0.000 0.000
#> GSM97108 2 0.2648 0.7981 0.000 0.848 0.000 0.000 0.152
#> GSM97050 2 0.1792 0.8088 0.000 0.916 0.000 0.000 0.084
#> GSM97080 3 0.0000 0.8920 0.000 0.000 1.000 0.000 0.000
#> GSM97089 3 0.0000 0.8920 0.000 0.000 1.000 0.000 0.000
#> GSM97092 3 0.0000 0.8920 0.000 0.000 1.000 0.000 0.000
#> GSM97093 3 0.0000 0.8920 0.000 0.000 1.000 0.000 0.000
#> GSM97058 2 0.1121 0.8008 0.000 0.956 0.000 0.000 0.044
#> GSM97051 2 0.1270 0.8101 0.000 0.948 0.000 0.000 0.052
#> GSM97052 3 0.0000 0.8920 0.000 0.000 1.000 0.000 0.000
#> GSM97061 3 0.0000 0.8920 0.000 0.000 1.000 0.000 0.000
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM97138 1 0.0000 0.9084 1.000 0.000 0.000 0.000 0.000 0
#> GSM97145 1 0.0260 0.9090 0.992 0.008 0.000 0.000 0.000 0
#> GSM97147 2 0.1225 0.8419 0.012 0.952 0.000 0.036 0.000 0
#> GSM97125 1 0.0146 0.9091 0.996 0.004 0.000 0.000 0.000 0
#> GSM97127 1 0.0458 0.9065 0.984 0.016 0.000 0.000 0.000 0
#> GSM97130 1 0.2219 0.7866 0.864 0.136 0.000 0.000 0.000 0
#> GSM97133 1 0.0632 0.9014 0.976 0.024 0.000 0.000 0.000 0
#> GSM97134 2 0.2823 0.6529 0.204 0.796 0.000 0.000 0.000 0
#> GSM97120 1 0.0260 0.9090 0.992 0.008 0.000 0.000 0.000 0
#> GSM97126 1 0.3409 0.5607 0.700 0.300 0.000 0.000 0.000 0
#> GSM97112 5 0.3620 0.3723 0.352 0.000 0.000 0.000 0.648 0
#> GSM97115 2 0.1261 0.8418 0.024 0.952 0.000 0.024 0.000 0
#> GSM97116 1 0.0000 0.9084 1.000 0.000 0.000 0.000 0.000 0
#> GSM97117 2 0.0363 0.8336 0.012 0.988 0.000 0.000 0.000 0
#> GSM97119 1 0.0000 0.9084 1.000 0.000 0.000 0.000 0.000 0
#> GSM97122 1 0.0000 0.9084 1.000 0.000 0.000 0.000 0.000 0
#> GSM97135 1 0.0000 0.9084 1.000 0.000 0.000 0.000 0.000 0
#> GSM97136 5 0.5928 0.5273 0.004 0.264 0.236 0.000 0.496 0
#> GSM97139 1 0.0146 0.9091 0.996 0.004 0.000 0.000 0.000 0
#> GSM97146 1 0.0000 0.9084 1.000 0.000 0.000 0.000 0.000 0
#> GSM97123 3 0.0146 0.9938 0.000 0.004 0.996 0.000 0.000 0
#> GSM97129 2 0.1196 0.8308 0.040 0.952 0.000 0.008 0.000 0
#> GSM97143 1 0.1007 0.8742 0.956 0.000 0.000 0.000 0.044 0
#> GSM97113 4 0.0146 0.9133 0.000 0.004 0.000 0.996 0.000 0
#> GSM97056 1 0.0458 0.9065 0.984 0.016 0.000 0.000 0.000 0
#> GSM97124 1 0.0260 0.9090 0.992 0.008 0.000 0.000 0.000 0
#> GSM97132 1 0.0458 0.9065 0.984 0.016 0.000 0.000 0.000 0
#> GSM97144 1 0.3390 0.5738 0.704 0.296 0.000 0.000 0.000 0
#> GSM97149 1 0.0458 0.9065 0.984 0.016 0.000 0.000 0.000 0
#> GSM97068 4 0.1204 0.9010 0.000 0.056 0.000 0.944 0.000 0
#> GSM97071 6 0.0000 1.0000 0.000 0.000 0.000 0.000 0.000 1
#> GSM97086 4 0.0146 0.9116 0.000 0.000 0.000 0.996 0.004 0
#> GSM97103 4 0.1663 0.8618 0.000 0.088 0.000 0.912 0.000 0
#> GSM97057 4 0.2146 0.8543 0.000 0.116 0.000 0.880 0.004 0
#> GSM97060 3 0.0146 0.9938 0.000 0.004 0.996 0.000 0.000 0
#> GSM97075 2 0.0717 0.8398 0.008 0.976 0.000 0.016 0.000 0
#> GSM97098 4 0.1010 0.8923 0.000 0.004 0.036 0.960 0.000 0
#> GSM97099 2 0.3547 0.5536 0.000 0.668 0.000 0.332 0.000 0
#> GSM97101 4 0.3854 0.0506 0.000 0.464 0.000 0.536 0.000 0
#> GSM97105 4 0.0291 0.9133 0.000 0.004 0.000 0.992 0.004 0
#> GSM97106 3 0.0146 0.9938 0.000 0.004 0.996 0.000 0.000 0
#> GSM97121 2 0.3360 0.6456 0.004 0.732 0.000 0.264 0.000 0
#> GSM97128 5 0.4703 0.1559 0.000 0.464 0.044 0.000 0.492 0
#> GSM97131 4 0.0000 0.9124 0.000 0.000 0.000 1.000 0.000 0
#> GSM97137 1 0.3797 0.2832 0.580 0.420 0.000 0.000 0.000 0
#> GSM97118 1 0.4835 0.3622 0.580 0.352 0.000 0.000 0.068 0
#> GSM97114 2 0.2092 0.7533 0.124 0.876 0.000 0.000 0.000 0
#> GSM97142 1 0.0146 0.9058 0.996 0.000 0.000 0.000 0.004 0
#> GSM97140 2 0.0790 0.8396 0.000 0.968 0.000 0.032 0.000 0
#> GSM97141 2 0.1219 0.8360 0.004 0.948 0.000 0.048 0.000 0
#> GSM97055 5 0.0260 0.7179 0.000 0.000 0.008 0.000 0.992 0
#> GSM97090 2 0.0891 0.8379 0.024 0.968 0.000 0.008 0.000 0
#> GSM97091 5 0.0260 0.7179 0.000 0.000 0.008 0.000 0.992 0
#> GSM97148 1 0.0146 0.9091 0.996 0.004 0.000 0.000 0.000 0
#> GSM97063 5 0.0260 0.7179 0.000 0.000 0.008 0.000 0.992 0
#> GSM97053 1 0.0000 0.9084 1.000 0.000 0.000 0.000 0.000 0
#> GSM97066 6 0.0000 1.0000 0.000 0.000 0.000 0.000 0.000 1
#> GSM97079 4 0.0000 0.9124 0.000 0.000 0.000 1.000 0.000 0
#> GSM97083 5 0.1765 0.7091 0.000 0.000 0.096 0.000 0.904 0
#> GSM97084 4 0.0291 0.9133 0.000 0.004 0.000 0.992 0.004 0
#> GSM97094 2 0.1327 0.8103 0.064 0.936 0.000 0.000 0.000 0
#> GSM97096 3 0.0000 0.9941 0.000 0.000 1.000 0.000 0.000 0
#> GSM97097 4 0.0000 0.9124 0.000 0.000 0.000 1.000 0.000 0
#> GSM97107 2 0.3647 0.5238 0.000 0.640 0.000 0.360 0.000 0
#> GSM97054 4 0.1411 0.9005 0.000 0.060 0.000 0.936 0.004 0
#> GSM97062 4 0.0000 0.9124 0.000 0.000 0.000 1.000 0.000 0
#> GSM97069 6 0.0000 1.0000 0.000 0.000 0.000 0.000 0.000 1
#> GSM97070 6 0.0000 1.0000 0.000 0.000 0.000 0.000 0.000 1
#> GSM97073 6 0.0000 1.0000 0.000 0.000 0.000 0.000 0.000 1
#> GSM97076 6 0.0000 1.0000 0.000 0.000 0.000 0.000 0.000 1
#> GSM97077 4 0.3499 0.4963 0.000 0.320 0.000 0.680 0.000 0
#> GSM97095 2 0.0405 0.8356 0.008 0.988 0.000 0.004 0.000 0
#> GSM97102 3 0.0000 0.9941 0.000 0.000 1.000 0.000 0.000 0
#> GSM97109 2 0.1124 0.8417 0.008 0.956 0.000 0.036 0.000 0
#> GSM97110 2 0.3851 0.2332 0.000 0.540 0.000 0.460 0.000 0
#> GSM97074 6 0.0000 1.0000 0.000 0.000 0.000 0.000 0.000 1
#> GSM97085 5 0.3446 0.5153 0.000 0.000 0.308 0.000 0.692 0
#> GSM97059 2 0.0891 0.8423 0.008 0.968 0.000 0.024 0.000 0
#> GSM97072 6 0.0000 1.0000 0.000 0.000 0.000 0.000 0.000 1
#> GSM97078 2 0.0603 0.8375 0.004 0.980 0.000 0.016 0.000 0
#> GSM97067 6 0.0000 1.0000 0.000 0.000 0.000 0.000 0.000 1
#> GSM97087 3 0.0000 0.9941 0.000 0.000 1.000 0.000 0.000 0
#> GSM97111 2 0.0405 0.8356 0.008 0.988 0.000 0.004 0.000 0
#> GSM97064 3 0.0146 0.9938 0.000 0.004 0.996 0.000 0.000 0
#> GSM97065 6 0.0000 1.0000 0.000 0.000 0.000 0.000 0.000 1
#> GSM97081 3 0.0000 0.9941 0.000 0.000 1.000 0.000 0.000 0
#> GSM97082 3 0.0000 0.9941 0.000 0.000 1.000 0.000 0.000 0
#> GSM97088 3 0.0632 0.9709 0.000 0.000 0.976 0.000 0.024 0
#> GSM97100 4 0.0405 0.9132 0.000 0.008 0.000 0.988 0.004 0
#> GSM97104 3 0.0000 0.9941 0.000 0.000 1.000 0.000 0.000 0
#> GSM97108 2 0.3804 0.3324 0.000 0.576 0.000 0.424 0.000 0
#> GSM97050 4 0.1411 0.9004 0.000 0.060 0.000 0.936 0.004 0
#> GSM97080 3 0.0000 0.9941 0.000 0.000 1.000 0.000 0.000 0
#> GSM97089 3 0.0000 0.9941 0.000 0.000 1.000 0.000 0.000 0
#> GSM97092 3 0.0146 0.9938 0.000 0.004 0.996 0.000 0.000 0
#> GSM97093 3 0.0790 0.9627 0.000 0.032 0.968 0.000 0.000 0
#> GSM97058 4 0.0937 0.9082 0.000 0.040 0.000 0.960 0.000 0
#> GSM97051 4 0.1010 0.9093 0.000 0.036 0.000 0.960 0.004 0
#> GSM97052 3 0.0146 0.9938 0.000 0.004 0.996 0.000 0.000 0
#> GSM97061 3 0.0146 0.9938 0.000 0.004 0.996 0.000 0.000 0
Heatmaps for the consensus matrix. It visualizes the probability of two samples to be in a same group.
consensus_heatmap(res, k = 2)
consensus_heatmap(res, k = 3)
consensus_heatmap(res, k = 4)
consensus_heatmap(res, k = 5)
consensus_heatmap(res, k = 6)
Heatmaps for the membership of samples in all partitions to see how consistent they are:
membership_heatmap(res, k = 2)
membership_heatmap(res, k = 3)
membership_heatmap(res, k = 4)
membership_heatmap(res, k = 5)
membership_heatmap(res, k = 6)
As soon as we have had the classes for columns, we can look for signatures which are significantly different between classes which can be candidate marks for certain classes. Following are the heatmaps for signatures.
Signature heatmaps where rows are scaled:
get_signatures(res, k = 2)
get_signatures(res, k = 3)
get_signatures(res, k = 4)
get_signatures(res, k = 5)
get_signatures(res, k = 6)
Signature heatmaps where rows are not scaled:
get_signatures(res, k = 2, scale_rows = FALSE)
get_signatures(res, k = 3, scale_rows = FALSE)
get_signatures(res, k = 4, scale_rows = FALSE)
get_signatures(res, k = 5, scale_rows = FALSE)
get_signatures(res, k = 6, scale_rows = FALSE)
Compare the overlap of signatures from different k:
compare_signatures(res)
get_signature() returns a data frame invisibly. TO get the list of signatures, the function
call should be assigned to a variable explicitly. In following code, if plot argument is set
to FALSE, no heatmap is plotted while only the differential analysis is performed.
# code only for demonstration
tb = get_signature(res, k = ..., plot = FALSE)
An example of the output of tb is:
#> which_row fdr mean_1 mean_2 scaled_mean_1 scaled_mean_2 km
#> 1 38 0.042760348 8.373488 9.131774 -0.5533452 0.5164555 1
#> 2 40 0.018707592 7.106213 8.469186 -0.6173731 0.5762149 1
#> 3 55 0.019134737 10.221463 11.207825 -0.6159697 0.5749050 1
#> 4 59 0.006059896 5.921854 7.869574 -0.6899429 0.6439467 1
#> 5 60 0.018055526 8.928898 10.211722 -0.6204761 0.5791110 1
#> 6 98 0.009384629 15.714769 14.887706 0.6635654 -0.6193277 2
...
The columns in tb are:
which_row: row indices corresponding to the input matrix.fdr: FDR for the differential test. mean_x: The mean value in group x.scaled_mean_x: The mean value in group x after rows are scaled.km: Row groups if k-means clustering is applied to rows.UMAP plot which shows how samples are separated.
dimension_reduction(res, k = 2, method = "UMAP")
dimension_reduction(res, k = 3, method = "UMAP")
dimension_reduction(res, k = 4, method = "UMAP")
dimension_reduction(res, k = 5, method = "UMAP")
dimension_reduction(res, k = 6, method = "UMAP")
Following heatmap shows how subgroups are split when increasing k:
collect_classes(res)
Test correlation between subgroups and known annotations. If the known annotation is numeric, one-way ANOVA test is applied, and if the known annotation is discrete, chi-squared contingency table test is applied.
test_to_known_factors(res)
#> n disease.state(p) specimen(p) cell.type(p) other(p) k
#> ATC:mclust 99 5.43e-02 0.5156 4.66e-04 0.5101 2
#> ATC:mclust 98 3.01e-06 0.4748 2.61e-17 0.1297 3
#> ATC:mclust 98 2.32e-06 0.1255 1.64e-15 0.0431 4
#> ATC:mclust 89 8.31e-07 0.1198 1.82e-17 0.1541 5
#> ATC:mclust 92 4.34e-06 0.0833 9.96e-15 0.1335 6
If matrix rows can be associated to genes, consider to use functional_enrichment(res,
...) to perform function enrichment for the signature genes. See this vignette for more detailed explanations.
The object with results only for a single top-value method and a single partition method can be extracted as:
res = res_list["ATC", "NMF"]
# you can also extract it by
# res = res_list["ATC:NMF"]
A summary of res and all the functions that can be applied to it:
res
#> A 'ConsensusPartition' object with k = 2, 3, 4, 5, 6.
#> On a matrix with 21512 rows and 100 columns.
#> Top rows (1000, 2000, 3000, 4000, 5000) are extracted by 'ATC' method.
#> Subgroups are detected by 'NMF' method.
#> Performed in total 1250 partitions by row resampling.
#> Best k for subgroups seems to be 2.
#>
#> Following methods can be applied to this 'ConsensusPartition' object:
#> [1] "cola_report" "collect_classes" "collect_plots"
#> [4] "collect_stats" "colnames" "compare_signatures"
#> [7] "consensus_heatmap" "dimension_reduction" "functional_enrichment"
#> [10] "get_anno_col" "get_anno" "get_classes"
#> [13] "get_consensus" "get_matrix" "get_membership"
#> [16] "get_param" "get_signatures" "get_stats"
#> [19] "is_best_k" "is_stable_k" "membership_heatmap"
#> [22] "ncol" "nrow" "plot_ecdf"
#> [25] "rownames" "select_partition_number" "show"
#> [28] "suggest_best_k" "test_to_known_factors"
collect_plots() function collects all the plots made from res for all k (number of partitions)
into one single page to provide an easy and fast comparison between different k.
collect_plots(res)
The plots are:
k and the heatmap of
predicted classes for each k.k.k.k.All the plots in panels can be made by individual functions and they are plotted later in this section.
select_partition_number() produces several plots showing different
statistics for choosing “optimized” k. There are following statistics:
k;k, the area increased is defined as \(A_k - A_{k-1}\).The detailed explanations of these statistics can be found in the cola vignette.
Generally speaking, lower PAC score, higher mean silhouette score or higher
concordance corresponds to better partition. Rand index and Jaccard index
measure how similar the current partition is compared to partition with k-1.
If they are too similar, we won't accept k is better than k-1.
select_partition_number(res)
The numeric values for all these statistics can be obtained by get_stats().
get_stats(res)
#> k 1-PAC mean_silhouette concordance area_increased Rand Jaccard
#> 2 2 0.918 0.918 0.968 0.5025 0.497 0.497
#> 3 3 0.618 0.710 0.875 0.3265 0.661 0.418
#> 4 4 0.654 0.703 0.842 0.1026 0.905 0.727
#> 5 5 0.639 0.592 0.793 0.0831 0.780 0.362
#> 6 6 0.637 0.474 0.694 0.0402 0.908 0.610
suggest_best_k() suggests the best \(k\) based on these statistics. The rules are as follows:
suggest_best_k(res)
#> [1] 2
Following shows the table of the partitions (You need to click the show/hide
code output link to see it). The membership matrix (columns with name p*)
is inferred by
clue::cl_consensus()
function with the SE method. Basically the value in the membership matrix
represents the probability to belong to a certain group. The finall class
label for an item is determined with the group with highest probability it
belongs to.
In get_classes() function, the entropy is calculated from the membership
matrix and the silhouette score is calculated from the consensus matrix.
cbind(get_classes(res, k = 2), get_membership(res, k = 2))
#> class entropy silhouette p1 p2
#> GSM97138 1 0.0000 0.95809 1.000 0.000
#> GSM97145 1 0.0000 0.95809 1.000 0.000
#> GSM97147 2 0.0000 0.97421 0.000 1.000
#> GSM97125 1 0.0000 0.95809 1.000 0.000
#> GSM97127 1 0.2043 0.93722 0.968 0.032
#> GSM97130 2 0.1184 0.96094 0.016 0.984
#> GSM97133 2 0.1184 0.96094 0.016 0.984
#> GSM97134 1 0.9922 0.22790 0.552 0.448
#> GSM97120 1 0.0000 0.95809 1.000 0.000
#> GSM97126 1 0.0000 0.95809 1.000 0.000
#> GSM97112 1 0.0000 0.95809 1.000 0.000
#> GSM97115 2 0.0000 0.97421 0.000 1.000
#> GSM97116 1 0.0000 0.95809 1.000 0.000
#> GSM97117 1 0.0000 0.95809 1.000 0.000
#> GSM97119 1 0.0000 0.95809 1.000 0.000
#> GSM97122 1 0.0000 0.95809 1.000 0.000
#> GSM97135 1 0.0000 0.95809 1.000 0.000
#> GSM97136 1 0.0000 0.95809 1.000 0.000
#> GSM97139 1 0.0000 0.95809 1.000 0.000
#> GSM97146 1 0.0000 0.95809 1.000 0.000
#> GSM97123 2 0.0000 0.97421 0.000 1.000
#> GSM97129 1 0.9983 0.13308 0.524 0.476
#> GSM97143 1 0.0000 0.95809 1.000 0.000
#> GSM97113 2 0.0000 0.97421 0.000 1.000
#> GSM97056 1 0.4939 0.86802 0.892 0.108
#> GSM97124 1 0.0000 0.95809 1.000 0.000
#> GSM97132 1 0.0000 0.95809 1.000 0.000
#> GSM97144 2 0.0376 0.97112 0.004 0.996
#> GSM97149 2 0.9963 0.08774 0.464 0.536
#> GSM97068 2 0.0000 0.97421 0.000 1.000
#> GSM97071 2 0.9993 -0.00864 0.484 0.516
#> GSM97086 2 0.0000 0.97421 0.000 1.000
#> GSM97103 2 0.0000 0.97421 0.000 1.000
#> GSM97057 2 0.0000 0.97421 0.000 1.000
#> GSM97060 2 0.0000 0.97421 0.000 1.000
#> GSM97075 1 0.4939 0.87054 0.892 0.108
#> GSM97098 2 0.0000 0.97421 0.000 1.000
#> GSM97099 2 0.0000 0.97421 0.000 1.000
#> GSM97101 2 0.0000 0.97421 0.000 1.000
#> GSM97105 2 0.0000 0.97421 0.000 1.000
#> GSM97106 2 0.0000 0.97421 0.000 1.000
#> GSM97121 2 0.0000 0.97421 0.000 1.000
#> GSM97128 1 0.0000 0.95809 1.000 0.000
#> GSM97131 2 0.0000 0.97421 0.000 1.000
#> GSM97137 2 0.0938 0.96446 0.012 0.988
#> GSM97118 1 0.0000 0.95809 1.000 0.000
#> GSM97114 2 0.0000 0.97421 0.000 1.000
#> GSM97142 1 0.0000 0.95809 1.000 0.000
#> GSM97140 2 0.0000 0.97421 0.000 1.000
#> GSM97141 2 0.0000 0.97421 0.000 1.000
#> GSM97055 1 0.0000 0.95809 1.000 0.000
#> GSM97090 2 0.0000 0.97421 0.000 1.000
#> GSM97091 1 0.0000 0.95809 1.000 0.000
#> GSM97148 1 0.0000 0.95809 1.000 0.000
#> GSM97063 1 0.0000 0.95809 1.000 0.000
#> GSM97053 1 0.0000 0.95809 1.000 0.000
#> GSM97066 1 0.0000 0.95809 1.000 0.000
#> GSM97079 2 0.0000 0.97421 0.000 1.000
#> GSM97083 1 0.0000 0.95809 1.000 0.000
#> GSM97084 2 0.0000 0.97421 0.000 1.000
#> GSM97094 1 0.0376 0.95592 0.996 0.004
#> GSM97096 1 0.2043 0.93786 0.968 0.032
#> GSM97097 2 0.0000 0.97421 0.000 1.000
#> GSM97107 2 0.0000 0.97421 0.000 1.000
#> GSM97054 2 0.0000 0.97421 0.000 1.000
#> GSM97062 2 0.0000 0.97421 0.000 1.000
#> GSM97069 1 0.0376 0.95592 0.996 0.004
#> GSM97070 1 0.0000 0.95809 1.000 0.000
#> GSM97073 1 0.3431 0.91281 0.936 0.064
#> GSM97076 1 0.0000 0.95809 1.000 0.000
#> GSM97077 2 0.0000 0.97421 0.000 1.000
#> GSM97095 1 0.9209 0.51922 0.664 0.336
#> GSM97102 1 0.0000 0.95809 1.000 0.000
#> GSM97109 2 0.0672 0.96787 0.008 0.992
#> GSM97110 2 0.0000 0.97421 0.000 1.000
#> GSM97074 1 0.0000 0.95809 1.000 0.000
#> GSM97085 1 0.0000 0.95809 1.000 0.000
#> GSM97059 2 0.0000 0.97421 0.000 1.000
#> GSM97072 2 0.0000 0.97421 0.000 1.000
#> GSM97078 1 0.1184 0.94853 0.984 0.016
#> GSM97067 1 0.0000 0.95809 1.000 0.000
#> GSM97087 1 0.0000 0.95809 1.000 0.000
#> GSM97111 1 0.3733 0.90632 0.928 0.072
#> GSM97064 2 0.0000 0.97421 0.000 1.000
#> GSM97065 1 0.2778 0.92591 0.952 0.048
#> GSM97081 1 0.4298 0.89049 0.912 0.088
#> GSM97082 1 0.0000 0.95809 1.000 0.000
#> GSM97088 1 0.0000 0.95809 1.000 0.000
#> GSM97100 2 0.0000 0.97421 0.000 1.000
#> GSM97104 1 0.0000 0.95809 1.000 0.000
#> GSM97108 2 0.0000 0.97421 0.000 1.000
#> GSM97050 2 0.0000 0.97421 0.000 1.000
#> GSM97080 1 0.8555 0.62791 0.720 0.280
#> GSM97089 1 0.0000 0.95809 1.000 0.000
#> GSM97092 2 0.0000 0.97421 0.000 1.000
#> GSM97093 2 0.5178 0.84730 0.116 0.884
#> GSM97058 2 0.0000 0.97421 0.000 1.000
#> GSM97051 2 0.0000 0.97421 0.000 1.000
#> GSM97052 2 0.0000 0.97421 0.000 1.000
#> GSM97061 2 0.0000 0.97421 0.000 1.000
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM97138 1 0.1643 0.84002 0.956 0.000 0.044
#> GSM97145 1 0.0237 0.86359 0.996 0.004 0.000
#> GSM97147 1 0.6154 0.37844 0.592 0.408 0.000
#> GSM97125 1 0.0000 0.86282 1.000 0.000 0.000
#> GSM97127 1 0.1031 0.86229 0.976 0.024 0.000
#> GSM97130 1 0.3038 0.83012 0.896 0.104 0.000
#> GSM97133 1 0.3412 0.81570 0.876 0.124 0.000
#> GSM97134 1 0.1964 0.85357 0.944 0.056 0.000
#> GSM97120 1 0.0237 0.86359 0.996 0.004 0.000
#> GSM97126 1 0.2878 0.79957 0.904 0.000 0.096
#> GSM97112 1 0.5785 0.40660 0.668 0.000 0.332
#> GSM97115 1 0.6111 0.41041 0.604 0.396 0.000
#> GSM97116 1 0.0000 0.86282 1.000 0.000 0.000
#> GSM97117 1 0.3038 0.79158 0.896 0.000 0.104
#> GSM97119 1 0.0237 0.86146 0.996 0.000 0.004
#> GSM97122 1 0.0000 0.86282 1.000 0.000 0.000
#> GSM97135 1 0.0237 0.86146 0.996 0.000 0.004
#> GSM97136 3 0.2878 0.78853 0.096 0.000 0.904
#> GSM97139 1 0.0237 0.86359 0.996 0.004 0.000
#> GSM97146 1 0.0000 0.86282 1.000 0.000 0.000
#> GSM97123 2 0.4002 0.73496 0.000 0.840 0.160
#> GSM97129 1 0.2711 0.84045 0.912 0.088 0.000
#> GSM97143 1 0.5016 0.60202 0.760 0.000 0.240
#> GSM97113 2 0.0000 0.86097 0.000 1.000 0.000
#> GSM97056 1 0.1031 0.86229 0.976 0.024 0.000
#> GSM97124 1 0.0237 0.86359 0.996 0.004 0.000
#> GSM97132 1 0.0000 0.86282 1.000 0.000 0.000
#> GSM97144 1 0.3038 0.83012 0.896 0.104 0.000
#> GSM97149 1 0.2537 0.84420 0.920 0.080 0.000
#> GSM97068 2 0.1289 0.84499 0.032 0.968 0.000
#> GSM97071 2 0.7658 0.29664 0.056 0.588 0.356
#> GSM97086 2 0.0000 0.86097 0.000 1.000 0.000
#> GSM97103 2 0.5058 0.62556 0.000 0.756 0.244
#> GSM97057 2 0.1031 0.84960 0.024 0.976 0.000
#> GSM97060 3 0.6204 0.24844 0.000 0.424 0.576
#> GSM97075 3 0.1529 0.80833 0.000 0.040 0.960
#> GSM97098 2 0.5706 0.47842 0.000 0.680 0.320
#> GSM97099 2 0.2625 0.81298 0.000 0.916 0.084
#> GSM97101 2 0.0237 0.85942 0.004 0.996 0.000
#> GSM97105 2 0.0237 0.85942 0.004 0.996 0.000
#> GSM97106 2 0.3752 0.75146 0.000 0.856 0.144
#> GSM97121 2 0.6095 0.25181 0.392 0.608 0.000
#> GSM97128 3 0.3686 0.75609 0.140 0.000 0.860
#> GSM97131 2 0.0237 0.85934 0.000 0.996 0.004
#> GSM97137 1 0.3192 0.82483 0.888 0.112 0.000
#> GSM97118 3 0.6225 0.29784 0.432 0.000 0.568
#> GSM97114 1 0.4002 0.78386 0.840 0.160 0.000
#> GSM97142 1 0.2959 0.79582 0.900 0.000 0.100
#> GSM97140 2 0.6180 0.17595 0.416 0.584 0.000
#> GSM97141 2 0.6026 0.29774 0.376 0.624 0.000
#> GSM97055 3 0.4702 0.68434 0.212 0.000 0.788
#> GSM97090 1 0.5905 0.50662 0.648 0.352 0.000
#> GSM97091 3 0.5529 0.57112 0.296 0.000 0.704
#> GSM97148 1 0.0237 0.86359 0.996 0.004 0.000
#> GSM97063 3 0.6309 0.09168 0.500 0.000 0.500
#> GSM97053 1 0.0000 0.86282 1.000 0.000 0.000
#> GSM97066 3 0.0000 0.82525 0.000 0.000 1.000
#> GSM97079 2 0.0000 0.86097 0.000 1.000 0.000
#> GSM97083 3 0.3816 0.74876 0.148 0.000 0.852
#> GSM97084 2 0.0000 0.86097 0.000 1.000 0.000
#> GSM97094 1 0.0475 0.86315 0.992 0.004 0.004
#> GSM97096 3 0.0424 0.82469 0.000 0.008 0.992
#> GSM97097 2 0.0424 0.85785 0.000 0.992 0.008
#> GSM97107 1 0.6308 0.10536 0.508 0.492 0.000
#> GSM97054 2 0.0000 0.86097 0.000 1.000 0.000
#> GSM97062 2 0.0000 0.86097 0.000 1.000 0.000
#> GSM97069 3 0.0424 0.82469 0.000 0.008 0.992
#> GSM97070 3 0.0237 0.82567 0.000 0.004 0.996
#> GSM97073 3 0.0592 0.82324 0.000 0.012 0.988
#> GSM97076 1 0.4605 0.66153 0.796 0.000 0.204
#> GSM97077 2 0.0000 0.86097 0.000 1.000 0.000
#> GSM97095 1 0.6796 0.66993 0.708 0.236 0.056
#> GSM97102 3 0.0237 0.82567 0.000 0.004 0.996
#> GSM97109 1 0.5882 0.51290 0.652 0.348 0.000
#> GSM97110 2 0.1031 0.85016 0.000 0.976 0.024
#> GSM97074 3 0.2261 0.80262 0.068 0.000 0.932
#> GSM97085 3 0.0747 0.82234 0.016 0.000 0.984
#> GSM97059 2 0.6309 -0.13933 0.500 0.500 0.000
#> GSM97072 3 0.6111 0.31670 0.000 0.396 0.604
#> GSM97078 3 0.0829 0.82466 0.012 0.004 0.984
#> GSM97067 3 0.0000 0.82525 0.000 0.000 1.000
#> GSM97087 3 0.0237 0.82567 0.000 0.004 0.996
#> GSM97111 1 0.4357 0.83202 0.868 0.080 0.052
#> GSM97064 2 0.2625 0.80798 0.000 0.916 0.084
#> GSM97065 3 0.8169 0.28790 0.388 0.076 0.536
#> GSM97081 3 0.0592 0.82324 0.000 0.012 0.988
#> GSM97082 3 0.0237 0.82567 0.000 0.004 0.996
#> GSM97088 3 0.0424 0.82432 0.008 0.000 0.992
#> GSM97100 2 0.0000 0.86097 0.000 1.000 0.000
#> GSM97104 3 0.0237 0.82567 0.000 0.004 0.996
#> GSM97108 2 0.1411 0.84254 0.036 0.964 0.000
#> GSM97050 2 0.0000 0.86097 0.000 1.000 0.000
#> GSM97080 3 0.0747 0.82140 0.000 0.016 0.984
#> GSM97089 3 0.0237 0.82567 0.000 0.004 0.996
#> GSM97092 3 0.5882 0.41803 0.000 0.348 0.652
#> GSM97093 3 0.6309 0.00352 0.000 0.496 0.504
#> GSM97058 2 0.0000 0.86097 0.000 1.000 0.000
#> GSM97051 2 0.0000 0.86097 0.000 1.000 0.000
#> GSM97052 3 0.6295 0.10946 0.000 0.472 0.528
#> GSM97061 2 0.3752 0.75154 0.000 0.856 0.144
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM97138 1 0.2706 0.840 0.900 0.000 0.080 0.020
#> GSM97145 1 0.0592 0.872 0.984 0.016 0.000 0.000
#> GSM97147 1 0.5256 0.345 0.596 0.392 0.000 0.012
#> GSM97125 1 0.1411 0.867 0.960 0.000 0.020 0.020
#> GSM97127 1 0.0921 0.870 0.972 0.028 0.000 0.000
#> GSM97130 1 0.1888 0.862 0.940 0.044 0.000 0.016
#> GSM97133 1 0.2089 0.859 0.932 0.048 0.000 0.020
#> GSM97134 1 0.1284 0.871 0.964 0.024 0.000 0.012
#> GSM97120 1 0.0779 0.873 0.980 0.016 0.004 0.000
#> GSM97126 1 0.3743 0.775 0.824 0.000 0.160 0.016
#> GSM97112 1 0.4910 0.603 0.704 0.000 0.276 0.020
#> GSM97115 1 0.5364 0.333 0.592 0.392 0.000 0.016
#> GSM97116 1 0.1510 0.867 0.956 0.000 0.028 0.016
#> GSM97117 1 0.1489 0.867 0.952 0.000 0.044 0.004
#> GSM97119 1 0.2775 0.839 0.896 0.000 0.084 0.020
#> GSM97122 1 0.2089 0.857 0.932 0.000 0.048 0.020
#> GSM97135 1 0.2256 0.853 0.924 0.000 0.056 0.020
#> GSM97136 3 0.1174 0.753 0.020 0.000 0.968 0.012
#> GSM97139 1 0.0592 0.872 0.984 0.016 0.000 0.000
#> GSM97146 1 0.1677 0.865 0.948 0.000 0.040 0.012
#> GSM97123 2 0.2814 0.720 0.000 0.868 0.132 0.000
#> GSM97129 1 0.1833 0.863 0.944 0.024 0.000 0.032
#> GSM97143 1 0.4797 0.631 0.720 0.000 0.260 0.020
#> GSM97113 2 0.4730 0.555 0.000 0.636 0.000 0.364
#> GSM97056 1 0.1256 0.872 0.964 0.028 0.008 0.000
#> GSM97124 1 0.0779 0.873 0.980 0.016 0.004 0.000
#> GSM97132 1 0.1411 0.866 0.960 0.000 0.020 0.020
#> GSM97144 1 0.2111 0.859 0.932 0.044 0.000 0.024
#> GSM97149 1 0.1256 0.869 0.964 0.028 0.000 0.008
#> GSM97068 2 0.5130 0.613 0.020 0.668 0.000 0.312
#> GSM97071 4 0.0188 0.795 0.000 0.004 0.000 0.996
#> GSM97086 2 0.2530 0.784 0.000 0.888 0.000 0.112
#> GSM97103 4 0.1820 0.800 0.000 0.020 0.036 0.944
#> GSM97057 2 0.1302 0.794 0.044 0.956 0.000 0.000
#> GSM97060 3 0.5933 0.351 0.000 0.408 0.552 0.040
#> GSM97075 3 0.3385 0.752 0.008 0.072 0.880 0.040
#> GSM97098 2 0.5938 0.548 0.000 0.696 0.136 0.168
#> GSM97099 4 0.1929 0.799 0.000 0.024 0.036 0.940
#> GSM97101 2 0.1209 0.794 0.032 0.964 0.004 0.000
#> GSM97105 2 0.2480 0.794 0.008 0.904 0.000 0.088
#> GSM97106 2 0.2814 0.720 0.000 0.868 0.132 0.000
#> GSM97121 2 0.6730 0.521 0.276 0.592 0.000 0.132
#> GSM97128 3 0.2089 0.741 0.048 0.000 0.932 0.020
#> GSM97131 2 0.1867 0.794 0.000 0.928 0.000 0.072
#> GSM97137 1 0.1637 0.860 0.940 0.060 0.000 0.000
#> GSM97118 3 0.5558 0.165 0.432 0.000 0.548 0.020
#> GSM97114 1 0.3958 0.788 0.836 0.052 0.000 0.112
#> GSM97142 1 0.3708 0.782 0.832 0.000 0.148 0.020
#> GSM97140 2 0.3024 0.737 0.148 0.852 0.000 0.000
#> GSM97141 2 0.3837 0.672 0.224 0.776 0.000 0.000
#> GSM97055 3 0.2843 0.719 0.088 0.000 0.892 0.020
#> GSM97090 1 0.4955 0.221 0.556 0.444 0.000 0.000
#> GSM97091 3 0.3806 0.653 0.156 0.000 0.824 0.020
#> GSM97148 1 0.1174 0.873 0.968 0.020 0.012 0.000
#> GSM97063 3 0.5550 0.171 0.428 0.000 0.552 0.020
#> GSM97053 1 0.1042 0.872 0.972 0.008 0.020 0.000
#> GSM97066 4 0.5143 0.557 0.012 0.000 0.360 0.628
#> GSM97079 2 0.4898 0.474 0.000 0.584 0.000 0.416
#> GSM97083 3 0.2174 0.743 0.052 0.000 0.928 0.020
#> GSM97084 2 0.2345 0.789 0.000 0.900 0.000 0.100
#> GSM97094 1 0.2002 0.861 0.936 0.000 0.044 0.020
#> GSM97096 3 0.2500 0.756 0.000 0.044 0.916 0.040
#> GSM97097 2 0.4999 0.302 0.000 0.508 0.000 0.492
#> GSM97107 4 0.2984 0.729 0.084 0.028 0.000 0.888
#> GSM97054 2 0.1151 0.798 0.024 0.968 0.000 0.008
#> GSM97062 2 0.3649 0.732 0.000 0.796 0.000 0.204
#> GSM97069 4 0.5004 0.501 0.004 0.000 0.392 0.604
#> GSM97070 4 0.4452 0.687 0.008 0.000 0.260 0.732
#> GSM97073 4 0.1398 0.803 0.004 0.000 0.040 0.956
#> GSM97076 4 0.2443 0.782 0.024 0.000 0.060 0.916
#> GSM97077 2 0.1938 0.801 0.012 0.936 0.000 0.052
#> GSM97095 1 0.6243 0.665 0.668 0.160 0.172 0.000
#> GSM97102 3 0.1833 0.759 0.000 0.024 0.944 0.032
#> GSM97109 1 0.6522 0.527 0.632 0.144 0.000 0.224
#> GSM97110 4 0.4428 0.388 0.000 0.276 0.004 0.720
#> GSM97074 4 0.4748 0.676 0.016 0.000 0.268 0.716
#> GSM97085 3 0.1042 0.753 0.008 0.000 0.972 0.020
#> GSM97059 2 0.4500 0.511 0.316 0.684 0.000 0.000
#> GSM97072 4 0.2089 0.800 0.000 0.020 0.048 0.932
#> GSM97078 3 0.4964 0.416 0.028 0.000 0.716 0.256
#> GSM97067 4 0.4857 0.612 0.008 0.000 0.324 0.668
#> GSM97087 3 0.2742 0.754 0.000 0.076 0.900 0.024
#> GSM97111 1 0.3616 0.835 0.852 0.036 0.112 0.000
#> GSM97064 2 0.1792 0.768 0.000 0.932 0.068 0.000
#> GSM97065 4 0.0524 0.799 0.004 0.000 0.008 0.988
#> GSM97081 3 0.2443 0.760 0.000 0.060 0.916 0.024
#> GSM97082 3 0.2021 0.763 0.000 0.040 0.936 0.024
#> GSM97088 3 0.0469 0.764 0.000 0.012 0.988 0.000
#> GSM97100 2 0.1938 0.800 0.012 0.936 0.000 0.052
#> GSM97104 3 0.2385 0.749 0.000 0.028 0.920 0.052
#> GSM97108 2 0.4776 0.732 0.060 0.776 0.000 0.164
#> GSM97050 2 0.0524 0.796 0.008 0.988 0.004 0.000
#> GSM97080 3 0.3813 0.652 0.000 0.024 0.828 0.148
#> GSM97089 3 0.1938 0.765 0.000 0.052 0.936 0.012
#> GSM97092 3 0.5465 0.398 0.000 0.392 0.588 0.020
#> GSM97093 3 0.4877 0.369 0.000 0.408 0.592 0.000
#> GSM97058 2 0.1807 0.801 0.008 0.940 0.000 0.052
#> GSM97051 2 0.0524 0.797 0.008 0.988 0.000 0.004
#> GSM97052 2 0.4998 -0.149 0.000 0.512 0.488 0.000
#> GSM97061 2 0.2469 0.742 0.000 0.892 0.108 0.000
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM97138 1 0.4126 0.4271 0.620 0.000 0.000 0.000 0.380
#> GSM97145 1 0.0324 0.7825 0.992 0.004 0.000 0.000 0.004
#> GSM97147 1 0.3339 0.6983 0.836 0.000 0.040 0.124 0.000
#> GSM97125 1 0.3661 0.5988 0.724 0.000 0.000 0.000 0.276
#> GSM97127 1 0.0404 0.7830 0.988 0.000 0.000 0.000 0.012
#> GSM97130 4 0.5368 0.5673 0.136 0.028 0.012 0.736 0.088
#> GSM97133 1 0.0000 0.7823 1.000 0.000 0.000 0.000 0.000
#> GSM97134 4 0.5890 0.3325 0.048 0.024 0.012 0.616 0.300
#> GSM97120 1 0.0510 0.7829 0.984 0.000 0.000 0.000 0.016
#> GSM97126 1 0.3242 0.6787 0.784 0.000 0.000 0.000 0.216
#> GSM97112 5 0.2798 0.6265 0.140 0.000 0.008 0.000 0.852
#> GSM97115 4 0.1780 0.7375 0.028 0.000 0.024 0.940 0.008
#> GSM97116 1 0.3109 0.6855 0.800 0.000 0.000 0.000 0.200
#> GSM97117 1 0.1568 0.7738 0.944 0.020 0.036 0.000 0.000
#> GSM97119 5 0.4249 0.0517 0.432 0.000 0.000 0.000 0.568
#> GSM97122 1 0.4300 0.1678 0.524 0.000 0.000 0.000 0.476
#> GSM97135 1 0.4227 0.3277 0.580 0.000 0.000 0.000 0.420
#> GSM97136 5 0.6303 -0.2234 0.000 0.160 0.364 0.000 0.476
#> GSM97139 1 0.0703 0.7822 0.976 0.000 0.000 0.000 0.024
#> GSM97146 1 0.1792 0.7607 0.916 0.000 0.000 0.000 0.084
#> GSM97123 3 0.1704 0.6946 0.000 0.004 0.928 0.068 0.000
#> GSM97129 1 0.1281 0.7715 0.956 0.032 0.012 0.000 0.000
#> GSM97143 5 0.2707 0.6339 0.132 0.000 0.008 0.000 0.860
#> GSM97113 4 0.7520 0.2402 0.056 0.188 0.360 0.396 0.000
#> GSM97056 4 0.6576 0.0115 0.216 0.000 0.000 0.444 0.340
#> GSM97124 1 0.3508 0.6280 0.748 0.000 0.000 0.000 0.252
#> GSM97132 5 0.5401 -0.1394 0.480 0.004 0.012 0.024 0.480
#> GSM97144 4 0.4313 0.6352 0.020 0.060 0.012 0.812 0.096
#> GSM97149 1 0.0000 0.7823 1.000 0.000 0.000 0.000 0.000
#> GSM97068 4 0.0693 0.7353 0.000 0.008 0.012 0.980 0.000
#> GSM97071 2 0.3653 0.7737 0.000 0.808 0.012 0.164 0.016
#> GSM97086 4 0.1357 0.7351 0.000 0.004 0.048 0.948 0.000
#> GSM97103 2 0.2464 0.8517 0.000 0.888 0.016 0.096 0.000
#> GSM97057 4 0.5344 0.3331 0.052 0.000 0.448 0.500 0.000
#> GSM97060 3 0.3766 0.7242 0.000 0.104 0.828 0.012 0.056
#> GSM97075 3 0.2818 0.7106 0.004 0.128 0.860 0.000 0.008
#> GSM97098 3 0.3635 0.6060 0.000 0.248 0.748 0.004 0.000
#> GSM97099 2 0.2331 0.8665 0.020 0.900 0.080 0.000 0.000
#> GSM97101 3 0.5463 0.4114 0.256 0.004 0.644 0.096 0.000
#> GSM97105 4 0.4491 0.5086 0.008 0.004 0.364 0.624 0.000
#> GSM97106 3 0.2179 0.6710 0.000 0.000 0.896 0.100 0.004
#> GSM97121 1 0.3527 0.6855 0.820 0.028 0.148 0.004 0.000
#> GSM97128 5 0.0486 0.6866 0.004 0.000 0.004 0.004 0.988
#> GSM97131 4 0.3177 0.6716 0.000 0.000 0.208 0.792 0.000
#> GSM97137 4 0.4522 0.6077 0.072 0.008 0.008 0.780 0.132
#> GSM97118 5 0.0703 0.6908 0.024 0.000 0.000 0.000 0.976
#> GSM97114 1 0.1364 0.7703 0.952 0.036 0.012 0.000 0.000
#> GSM97142 5 0.3752 0.4062 0.292 0.000 0.000 0.000 0.708
#> GSM97140 1 0.5092 0.1753 0.524 0.000 0.440 0.036 0.000
#> GSM97141 1 0.4329 0.4858 0.672 0.016 0.312 0.000 0.000
#> GSM97055 5 0.1628 0.6587 0.008 0.000 0.056 0.000 0.936
#> GSM97090 4 0.3047 0.7114 0.004 0.000 0.044 0.868 0.084
#> GSM97091 5 0.0693 0.6849 0.008 0.000 0.012 0.000 0.980
#> GSM97148 1 0.0703 0.7820 0.976 0.000 0.000 0.000 0.024
#> GSM97063 5 0.1364 0.6898 0.036 0.000 0.012 0.000 0.952
#> GSM97053 1 0.3579 0.6407 0.756 0.000 0.000 0.004 0.240
#> GSM97066 2 0.1836 0.8760 0.000 0.932 0.032 0.000 0.036
#> GSM97079 4 0.0324 0.7342 0.000 0.004 0.004 0.992 0.000
#> GSM97083 5 0.1153 0.6876 0.004 0.000 0.008 0.024 0.964
#> GSM97084 4 0.0609 0.7357 0.000 0.000 0.020 0.980 0.000
#> GSM97094 5 0.5315 0.3173 0.028 0.004 0.012 0.368 0.588
#> GSM97096 3 0.5190 0.6320 0.000 0.172 0.688 0.000 0.140
#> GSM97097 4 0.0566 0.7322 0.000 0.012 0.004 0.984 0.000
#> GSM97107 4 0.3049 0.6726 0.000 0.084 0.012 0.872 0.032
#> GSM97054 4 0.1478 0.7331 0.000 0.000 0.064 0.936 0.000
#> GSM97062 4 0.0324 0.7342 0.000 0.004 0.004 0.992 0.000
#> GSM97069 2 0.2795 0.8516 0.000 0.880 0.056 0.000 0.064
#> GSM97070 2 0.1408 0.8776 0.000 0.948 0.044 0.000 0.008
#> GSM97073 2 0.0740 0.8831 0.008 0.980 0.008 0.004 0.000
#> GSM97076 2 0.2821 0.8462 0.052 0.896 0.012 0.032 0.008
#> GSM97077 4 0.3884 0.6104 0.000 0.004 0.288 0.708 0.000
#> GSM97095 5 0.4473 0.2485 0.008 0.000 0.000 0.412 0.580
#> GSM97102 5 0.6480 -0.3356 0.000 0.184 0.404 0.000 0.412
#> GSM97109 1 0.1914 0.7578 0.924 0.060 0.016 0.000 0.000
#> GSM97110 2 0.3934 0.7887 0.016 0.820 0.060 0.104 0.000
#> GSM97074 2 0.2606 0.8617 0.000 0.900 0.012 0.032 0.056
#> GSM97085 5 0.3336 0.5464 0.000 0.060 0.096 0.000 0.844
#> GSM97059 4 0.6709 0.2537 0.384 0.000 0.152 0.448 0.016
#> GSM97072 2 0.1661 0.8836 0.000 0.940 0.036 0.024 0.000
#> GSM97078 5 0.4810 0.2817 0.000 0.008 0.012 0.400 0.580
#> GSM97067 2 0.2074 0.8718 0.000 0.920 0.044 0.000 0.036
#> GSM97087 3 0.5487 0.5509 0.000 0.100 0.620 0.000 0.280
#> GSM97111 1 0.4397 0.5789 0.708 0.024 0.264 0.000 0.004
#> GSM97064 3 0.2127 0.6534 0.000 0.000 0.892 0.108 0.000
#> GSM97065 2 0.1547 0.8764 0.032 0.948 0.016 0.004 0.000
#> GSM97081 3 0.3911 0.6948 0.000 0.144 0.796 0.000 0.060
#> GSM97082 3 0.6269 0.3018 0.000 0.148 0.444 0.000 0.408
#> GSM97088 5 0.2110 0.6254 0.000 0.016 0.072 0.000 0.912
#> GSM97100 4 0.2377 0.7158 0.000 0.000 0.128 0.872 0.000
#> GSM97104 3 0.6368 0.3800 0.000 0.172 0.472 0.000 0.356
#> GSM97108 4 0.7385 0.2836 0.232 0.036 0.320 0.412 0.000
#> GSM97050 3 0.4425 -0.0185 0.008 0.000 0.600 0.392 0.000
#> GSM97080 2 0.5657 0.4278 0.000 0.616 0.256 0.000 0.128
#> GSM97089 3 0.5580 0.5832 0.000 0.132 0.632 0.000 0.236
#> GSM97092 3 0.2843 0.7310 0.000 0.076 0.876 0.000 0.048
#> GSM97093 3 0.1918 0.7301 0.004 0.004 0.932 0.012 0.048
#> GSM97058 4 0.4542 0.3481 0.000 0.008 0.456 0.536 0.000
#> GSM97051 4 0.3707 0.6153 0.000 0.000 0.284 0.716 0.000
#> GSM97052 3 0.1646 0.7288 0.000 0.004 0.944 0.020 0.032
#> GSM97061 3 0.1697 0.7014 0.000 0.008 0.932 0.060 0.000
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM97138 1 0.5466 0.2262 0.512 0.096 0.004 0.000 0.384 0.004
#> GSM97145 1 0.1398 0.6837 0.940 0.052 0.000 0.000 0.008 0.000
#> GSM97147 1 0.6505 0.2286 0.460 0.192 0.040 0.308 0.000 0.000
#> GSM97125 1 0.4716 0.5603 0.680 0.136 0.000 0.000 0.184 0.000
#> GSM97127 1 0.0520 0.6853 0.984 0.008 0.000 0.000 0.008 0.000
#> GSM97130 2 0.5762 0.3958 0.096 0.476 0.000 0.408 0.016 0.004
#> GSM97133 1 0.0363 0.6852 0.988 0.012 0.000 0.000 0.000 0.000
#> GSM97134 4 0.6799 -0.3342 0.048 0.252 0.000 0.452 0.244 0.004
#> GSM97120 1 0.0820 0.6858 0.972 0.016 0.000 0.000 0.012 0.000
#> GSM97126 5 0.3869 0.0366 0.500 0.000 0.000 0.000 0.500 0.000
#> GSM97112 5 0.2033 0.7252 0.056 0.020 0.004 0.000 0.916 0.004
#> GSM97115 4 0.5455 -0.0489 0.080 0.308 0.020 0.588 0.004 0.000
#> GSM97116 1 0.2608 0.6628 0.872 0.048 0.000 0.000 0.080 0.000
#> GSM97117 1 0.6287 0.3852 0.516 0.216 0.240 0.000 0.004 0.024
#> GSM97119 5 0.4585 0.4174 0.308 0.060 0.000 0.000 0.632 0.000
#> GSM97122 5 0.5149 0.0139 0.440 0.084 0.000 0.000 0.476 0.000
#> GSM97135 1 0.5516 0.2315 0.504 0.140 0.000 0.000 0.356 0.000
#> GSM97136 5 0.6478 -0.1799 0.004 0.036 0.364 0.000 0.440 0.156
#> GSM97139 1 0.0891 0.6855 0.968 0.024 0.000 0.000 0.008 0.000
#> GSM97146 1 0.2618 0.6587 0.872 0.052 0.000 0.000 0.076 0.000
#> GSM97123 3 0.2113 0.6797 0.000 0.048 0.912 0.032 0.000 0.008
#> GSM97129 1 0.3970 0.6206 0.756 0.196 0.008 0.000 0.004 0.036
#> GSM97143 5 0.2492 0.7226 0.068 0.036 0.008 0.000 0.888 0.000
#> GSM97113 4 0.7093 0.3324 0.024 0.172 0.092 0.536 0.004 0.172
#> GSM97056 1 0.7681 -0.2092 0.356 0.300 0.012 0.184 0.148 0.000
#> GSM97124 1 0.5019 0.5434 0.652 0.236 0.000 0.004 0.104 0.004
#> GSM97132 1 0.6660 0.1116 0.388 0.388 0.000 0.020 0.188 0.016
#> GSM97144 2 0.5114 0.3457 0.016 0.484 0.000 0.464 0.024 0.012
#> GSM97149 1 0.1036 0.6847 0.964 0.024 0.000 0.004 0.008 0.000
#> GSM97068 4 0.1843 0.4803 0.000 0.080 0.004 0.912 0.000 0.004
#> GSM97071 6 0.4973 0.6490 0.000 0.276 0.004 0.072 0.008 0.640
#> GSM97086 4 0.0291 0.5162 0.000 0.004 0.004 0.992 0.000 0.000
#> GSM97103 6 0.6123 0.5663 0.004 0.312 0.032 0.112 0.004 0.536
#> GSM97057 4 0.5639 0.4366 0.036 0.104 0.232 0.624 0.004 0.000
#> GSM97060 3 0.3700 0.6503 0.000 0.016 0.808 0.012 0.028 0.136
#> GSM97075 3 0.5120 0.6410 0.032 0.116 0.740 0.008 0.028 0.076
#> GSM97098 3 0.4754 0.5802 0.000 0.076 0.704 0.024 0.000 0.196
#> GSM97099 6 0.6102 0.6054 0.036 0.228 0.136 0.012 0.000 0.588
#> GSM97101 3 0.6391 0.4204 0.168 0.144 0.584 0.100 0.000 0.004
#> GSM97105 4 0.4981 0.4682 0.008 0.160 0.132 0.692 0.000 0.008
#> GSM97106 3 0.1969 0.6746 0.000 0.020 0.920 0.052 0.004 0.004
#> GSM97121 1 0.5983 0.4750 0.564 0.264 0.140 0.028 0.000 0.004
#> GSM97128 5 0.1503 0.7270 0.000 0.032 0.016 0.000 0.944 0.008
#> GSM97131 4 0.2799 0.5273 0.000 0.064 0.076 0.860 0.000 0.000
#> GSM97137 4 0.6176 -0.0966 0.080 0.288 0.000 0.552 0.076 0.004
#> GSM97118 5 0.3282 0.6948 0.016 0.116 0.004 0.000 0.836 0.028
#> GSM97114 1 0.3823 0.6164 0.772 0.188 0.008 0.016 0.000 0.016
#> GSM97142 5 0.3455 0.6553 0.144 0.056 0.000 0.000 0.800 0.000
#> GSM97140 3 0.6480 0.0170 0.372 0.128 0.440 0.060 0.000 0.000
#> GSM97141 1 0.5911 0.2731 0.496 0.180 0.316 0.008 0.000 0.000
#> GSM97055 5 0.1995 0.7182 0.004 0.024 0.036 0.000 0.924 0.012
#> GSM97090 4 0.5546 0.2798 0.020 0.204 0.032 0.668 0.072 0.004
#> GSM97091 5 0.1129 0.7338 0.008 0.012 0.012 0.000 0.964 0.004
#> GSM97148 1 0.1461 0.6805 0.940 0.044 0.000 0.000 0.016 0.000
#> GSM97063 5 0.1149 0.7348 0.024 0.008 0.008 0.000 0.960 0.000
#> GSM97053 1 0.3979 0.5805 0.752 0.076 0.000 0.000 0.172 0.000
#> GSM97066 6 0.1542 0.8149 0.000 0.024 0.016 0.000 0.016 0.944
#> GSM97079 4 0.1863 0.5087 0.000 0.060 0.000 0.920 0.004 0.016
#> GSM97083 5 0.2094 0.7175 0.000 0.064 0.024 0.000 0.908 0.004
#> GSM97084 4 0.3023 0.3209 0.000 0.212 0.004 0.784 0.000 0.000
#> GSM97094 2 0.6570 0.4059 0.024 0.504 0.004 0.188 0.268 0.012
#> GSM97096 3 0.4601 0.6548 0.000 0.080 0.760 0.004 0.056 0.100
#> GSM97097 4 0.4235 0.1235 0.000 0.300 0.008 0.668 0.000 0.024
#> GSM97107 2 0.5264 0.3623 0.004 0.500 0.004 0.436 0.012 0.044
#> GSM97054 4 0.3534 0.3272 0.000 0.200 0.024 0.772 0.000 0.004
#> GSM97062 4 0.2362 0.4340 0.000 0.136 0.000 0.860 0.000 0.004
#> GSM97069 6 0.2345 0.7947 0.000 0.028 0.028 0.000 0.040 0.904
#> GSM97070 6 0.1053 0.8147 0.000 0.012 0.020 0.000 0.004 0.964
#> GSM97073 6 0.2663 0.8125 0.000 0.084 0.012 0.028 0.000 0.876
#> GSM97076 6 0.4672 0.7368 0.044 0.216 0.004 0.016 0.008 0.712
#> GSM97077 4 0.4170 0.5188 0.004 0.116 0.064 0.788 0.004 0.024
#> GSM97095 2 0.7238 0.3344 0.008 0.324 0.048 0.316 0.300 0.004
#> GSM97102 3 0.6601 0.3388 0.000 0.044 0.440 0.000 0.320 0.196
#> GSM97109 1 0.6524 0.4076 0.472 0.372 0.056 0.024 0.000 0.076
#> GSM97110 4 0.6564 0.0145 0.004 0.156 0.032 0.420 0.004 0.384
#> GSM97074 6 0.3130 0.7921 0.000 0.144 0.000 0.004 0.028 0.824
#> GSM97085 5 0.3320 0.6678 0.000 0.032 0.076 0.000 0.844 0.048
#> GSM97059 4 0.6446 0.0235 0.396 0.060 0.104 0.436 0.004 0.000
#> GSM97072 6 0.2313 0.8085 0.000 0.044 0.016 0.036 0.000 0.904
#> GSM97078 5 0.4429 0.4144 0.000 0.040 0.004 0.240 0.704 0.012
#> GSM97067 6 0.1232 0.8142 0.000 0.024 0.016 0.000 0.004 0.956
#> GSM97087 3 0.4668 0.5962 0.000 0.020 0.712 0.000 0.188 0.080
#> GSM97111 3 0.6598 -0.0918 0.372 0.220 0.384 0.012 0.004 0.008
#> GSM97064 3 0.4828 -0.0700 0.000 0.044 0.500 0.452 0.000 0.004
#> GSM97065 6 0.3434 0.7898 0.008 0.136 0.012 0.024 0.000 0.820
#> GSM97081 3 0.5052 0.6671 0.000 0.092 0.732 0.012 0.104 0.060
#> GSM97082 3 0.6166 0.2474 0.000 0.036 0.452 0.000 0.388 0.124
#> GSM97088 5 0.2926 0.6800 0.000 0.024 0.112 0.000 0.852 0.012
#> GSM97100 4 0.2003 0.5307 0.000 0.044 0.044 0.912 0.000 0.000
#> GSM97104 3 0.6405 0.4386 0.000 0.040 0.504 0.000 0.240 0.216
#> GSM97108 2 0.7705 -0.1933 0.276 0.340 0.148 0.228 0.000 0.008
#> GSM97050 4 0.5421 0.4143 0.000 0.120 0.264 0.604 0.008 0.004
#> GSM97080 6 0.5566 0.5423 0.000 0.044 0.156 0.012 0.116 0.672
#> GSM97089 3 0.5202 0.5816 0.000 0.024 0.668 0.000 0.176 0.132
#> GSM97092 3 0.1760 0.6854 0.000 0.000 0.928 0.004 0.020 0.048
#> GSM97093 3 0.1065 0.6893 0.000 0.008 0.964 0.008 0.020 0.000
#> GSM97058 4 0.5570 0.4544 0.000 0.156 0.180 0.636 0.004 0.024
#> GSM97051 4 0.3062 0.5289 0.000 0.052 0.112 0.836 0.000 0.000
#> GSM97052 3 0.1242 0.6874 0.000 0.008 0.960 0.012 0.012 0.008
#> GSM97061 3 0.2103 0.6766 0.000 0.012 0.912 0.056 0.000 0.020
Heatmaps for the consensus matrix. It visualizes the probability of two samples to be in a same group.
consensus_heatmap(res, k = 2)
consensus_heatmap(res, k = 3)
consensus_heatmap(res, k = 4)
consensus_heatmap(res, k = 5)
consensus_heatmap(res, k = 6)
Heatmaps for the membership of samples in all partitions to see how consistent they are:
membership_heatmap(res, k = 2)
membership_heatmap(res, k = 3)
membership_heatmap(res, k = 4)
membership_heatmap(res, k = 5)
membership_heatmap(res, k = 6)
As soon as we have had the classes for columns, we can look for signatures which are significantly different between classes which can be candidate marks for certain classes. Following are the heatmaps for signatures.
Signature heatmaps where rows are scaled:
get_signatures(res, k = 2)
get_signatures(res, k = 3)
get_signatures(res, k = 4)
get_signatures(res, k = 5)
get_signatures(res, k = 6)
Signature heatmaps where rows are not scaled:
get_signatures(res, k = 2, scale_rows = FALSE)
get_signatures(res, k = 3, scale_rows = FALSE)
get_signatures(res, k = 4, scale_rows = FALSE)
get_signatures(res, k = 5, scale_rows = FALSE)
get_signatures(res, k = 6, scale_rows = FALSE)
Compare the overlap of signatures from different k:
compare_signatures(res)
get_signature() returns a data frame invisibly. TO get the list of signatures, the function
call should be assigned to a variable explicitly. In following code, if plot argument is set
to FALSE, no heatmap is plotted while only the differential analysis is performed.
# code only for demonstration
tb = get_signature(res, k = ..., plot = FALSE)
An example of the output of tb is:
#> which_row fdr mean_1 mean_2 scaled_mean_1 scaled_mean_2 km
#> 1 38 0.042760348 8.373488 9.131774 -0.5533452 0.5164555 1
#> 2 40 0.018707592 7.106213 8.469186 -0.6173731 0.5762149 1
#> 3 55 0.019134737 10.221463 11.207825 -0.6159697 0.5749050 1
#> 4 59 0.006059896 5.921854 7.869574 -0.6899429 0.6439467 1
#> 5 60 0.018055526 8.928898 10.211722 -0.6204761 0.5791110 1
#> 6 98 0.009384629 15.714769 14.887706 0.6635654 -0.6193277 2
...
The columns in tb are:
which_row: row indices corresponding to the input matrix.fdr: FDR for the differential test. mean_x: The mean value in group x.scaled_mean_x: The mean value in group x after rows are scaled.km: Row groups if k-means clustering is applied to rows.UMAP plot which shows how samples are separated.
dimension_reduction(res, k = 2, method = "UMAP")
dimension_reduction(res, k = 3, method = "UMAP")
dimension_reduction(res, k = 4, method = "UMAP")
dimension_reduction(res, k = 5, method = "UMAP")
dimension_reduction(res, k = 6, method = "UMAP")
Following heatmap shows how subgroups are split when increasing k:
collect_classes(res)
Test correlation between subgroups and known annotations. If the known annotation is numeric, one-way ANOVA test is applied, and if the known annotation is discrete, chi-squared contingency table test is applied.
test_to_known_factors(res)
#> n disease.state(p) specimen(p) cell.type(p) other(p) k
#> ATC:NMF 96 2.74e-02 0.0378 1.75e-03 0.2217 2
#> ATC:NMF 82 1.44e-06 0.0332 3.74e-11 0.0603 3
#> ATC:NMF 87 6.01e-06 0.2059 2.36e-14 0.1471 4
#> ATC:NMF 75 6.08e-03 0.9663 1.16e-10 0.3964 5
#> ATC:NMF 57 2.17e-03 0.9315 1.43e-10 0.8218 6
If matrix rows can be associated to genes, consider to use functional_enrichment(res,
...) to perform function enrichment for the signature genes. See this vignette for more detailed explanations.
sessionInfo()
#> R version 3.6.0 (2019-04-26)
#> Platform: x86_64-pc-linux-gnu (64-bit)
#> Running under: CentOS Linux 7 (Core)
#>
#> Matrix products: default
#> BLAS: /usr/lib64/libblas.so.3.4.2
#> LAPACK: /usr/lib64/liblapack.so.3.4.2
#>
#> locale:
#> [1] LC_CTYPE=en_GB.UTF-8 LC_NUMERIC=C LC_TIME=en_GB.UTF-8
#> [4] LC_COLLATE=en_GB.UTF-8 LC_MONETARY=en_GB.UTF-8 LC_MESSAGES=en_GB.UTF-8
#> [7] LC_PAPER=en_GB.UTF-8 LC_NAME=C LC_ADDRESS=C
#> [10] LC_TELEPHONE=C LC_MEASUREMENT=en_GB.UTF-8 LC_IDENTIFICATION=C
#>
#> attached base packages:
#> [1] grid stats graphics grDevices utils datasets methods base
#>
#> other attached packages:
#> [1] genefilter_1.66.0 ComplexHeatmap_2.3.1 markdown_1.1 knitr_1.26
#> [5] GetoptLong_0.1.7 cola_1.3.2
#>
#> loaded via a namespace (and not attached):
#> [1] circlize_0.4.8 shape_1.4.4 xfun_0.11 slam_0.1-46
#> [5] lattice_0.20-38 splines_3.6.0 colorspace_1.4-1 vctrs_0.2.0
#> [9] stats4_3.6.0 blob_1.2.0 XML_3.98-1.20 survival_2.44-1.1
#> [13] rlang_0.4.2 pillar_1.4.2 DBI_1.0.0 BiocGenerics_0.30.0
#> [17] bit64_0.9-7 RColorBrewer_1.1-2 matrixStats_0.55.0 stringr_1.4.0
#> [21] GlobalOptions_0.1.1 evaluate_0.14 memoise_1.1.0 Biobase_2.44.0
#> [25] IRanges_2.18.3 parallel_3.6.0 AnnotationDbi_1.46.1 highr_0.8
#> [29] Rcpp_1.0.3 xtable_1.8-4 backports_1.1.5 S4Vectors_0.22.1
#> [33] annotate_1.62.0 skmeans_0.2-11 bit_1.1-14 microbenchmark_1.4-7
#> [37] brew_1.0-6 impute_1.58.0 rjson_0.2.20 png_0.1-7
#> [41] digest_0.6.23 stringi_1.4.3 polyclip_1.10-0 clue_0.3-57
#> [45] tools_3.6.0 bitops_1.0-6 magrittr_1.5 eulerr_6.0.0
#> [49] RCurl_1.95-4.12 RSQLite_2.1.4 tibble_2.1.3 cluster_2.1.0
#> [53] crayon_1.3.4 pkgconfig_2.0.3 zeallot_0.1.0 Matrix_1.2-17
#> [57] xml2_1.2.2 httr_1.4.1 R6_2.4.1 mclust_5.4.5
#> [61] compiler_3.6.0