Date: 2019-12-25 20:17:11 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 8353 rows and 87 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] 8353 87
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 | Optional k | ||
|---|---|---|---|---|---|---|
| ATC:hclust | 2 | 1.000 | 0.984 | 0.994 | ** | |
| ATC:NMF | 2 | 0.999 | 0.957 | 0.980 | ** | |
| CV:skmeans | 2 | 0.976 | 0.917 | 0.970 | ** | |
| ATC:skmeans | 3 | 0.975 | 0.916 | 0.958 | ** | 2 |
| ATC:mclust | 2 | 0.975 | 0.965 | 0.983 | ** | |
| SD:skmeans | 2 | 0.904 | 0.914 | 0.968 | * | |
| SD:NMF | 2 | 0.882 | 0.919 | 0.967 | ||
| ATC:kmeans | 2 | 0.847 | 0.941 | 0.965 | ||
| CV:NMF | 2 | 0.837 | 0.901 | 0.960 | ||
| MAD:skmeans | 2 | 0.820 | 0.893 | 0.954 | ||
| SD:kmeans | 2 | 0.813 | 0.887 | 0.949 | ||
| CV:mclust | 2 | 0.805 | 0.884 | 0.947 | ||
| MAD:kmeans | 2 | 0.800 | 0.885 | 0.950 | ||
| MAD:NMF | 2 | 0.768 | 0.883 | 0.949 | ||
| CV:kmeans | 2 | 0.735 | 0.891 | 0.945 | ||
| SD:hclust | 2 | 0.623 | 0.874 | 0.927 | ||
| ATC:pam | 4 | 0.582 | 0.631 | 0.823 | ||
| MAD:mclust | 2 | 0.578 | 0.718 | 0.882 | ||
| SD:mclust | 2 | 0.573 | 0.737 | 0.884 | ||
| MAD:hclust | 2 | 0.563 | 0.856 | 0.911 | ||
| CV:hclust | 2 | 0.474 | 0.848 | 0.904 | ||
| SD:pam | 3 | 0.305 | 0.624 | 0.819 | ||
| MAD:pam | 3 | 0.226 | 0.570 | 0.780 | ||
| CV:pam | NA | NA | NA | NA |
**: 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 0.882 0.919 0.967 0.4929 0.505 0.505
#> CV:NMF 2 0.837 0.901 0.960 0.4968 0.500 0.500
#> MAD:NMF 2 0.768 0.883 0.949 0.4956 0.502 0.502
#> ATC:NMF 2 0.999 0.957 0.980 0.4297 0.558 0.558
#> SD:skmeans 2 0.904 0.914 0.968 0.5006 0.500 0.500
#> CV:skmeans 2 0.976 0.917 0.970 0.5005 0.497 0.497
#> MAD:skmeans 2 0.820 0.893 0.954 0.4990 0.500 0.500
#> ATC:skmeans 2 1.000 0.985 0.994 0.4639 0.536 0.536
#> SD:mclust 2 0.573 0.737 0.884 0.4647 0.495 0.495
#> CV:mclust 2 0.805 0.884 0.947 0.4928 0.500 0.500
#> MAD:mclust 2 0.578 0.718 0.882 0.4689 0.543 0.543
#> ATC:mclust 2 0.975 0.965 0.983 0.4687 0.524 0.524
#> SD:kmeans 2 0.813 0.887 0.949 0.4460 0.530 0.530
#> CV:kmeans 2 0.735 0.891 0.945 0.4658 0.513 0.513
#> MAD:kmeans 2 0.800 0.885 0.950 0.4413 0.586 0.586
#> ATC:kmeans 2 0.847 0.941 0.965 0.3771 0.607 0.607
#> SD:pam 2 0.666 0.895 0.945 0.2189 0.777 0.777
#> CV:pam 2 0.531 0.731 0.894 0.2928 0.759 0.759
#> MAD:pam 2 0.629 0.788 0.908 0.2684 0.743 0.743
#> ATC:pam 2 0.858 0.896 0.957 0.3858 0.607 0.607
#> SD:hclust 2 0.623 0.874 0.927 0.4100 0.567 0.567
#> CV:hclust 2 0.474 0.848 0.904 0.4070 0.567 0.567
#> MAD:hclust 2 0.563 0.856 0.911 0.4057 0.550 0.550
#> ATC:hclust 2 1.000 0.984 0.994 0.0699 0.933 0.933
get_stats(res_list, k = 3)
#> k 1-PAC mean_silhouette concordance area_increased Rand Jaccard
#> SD:NMF 3 0.451 0.550 0.763 0.318 0.706 0.485
#> CV:NMF 3 0.443 0.462 0.669 0.317 0.680 0.448
#> MAD:NMF 3 0.413 0.417 0.637 0.316 0.628 0.396
#> ATC:NMF 3 0.854 0.877 0.950 0.143 0.906 0.842
#> SD:skmeans 3 0.689 0.826 0.914 0.304 0.786 0.600
#> CV:skmeans 3 0.650 0.667 0.864 0.301 0.795 0.613
#> MAD:skmeans 3 0.609 0.763 0.885 0.321 0.785 0.595
#> ATC:skmeans 3 0.975 0.916 0.958 0.415 0.776 0.595
#> SD:mclust 3 0.524 0.654 0.820 0.356 0.568 0.324
#> CV:mclust 3 0.511 0.794 0.874 0.262 0.558 0.335
#> MAD:mclust 3 0.531 0.747 0.833 0.324 0.759 0.586
#> ATC:mclust 3 0.603 0.748 0.832 0.300 0.732 0.536
#> SD:kmeans 3 0.714 0.857 0.914 0.307 0.871 0.762
#> CV:kmeans 3 0.801 0.868 0.931 0.247 0.866 0.755
#> MAD:kmeans 3 0.739 0.810 0.907 0.345 0.810 0.682
#> ATC:kmeans 3 0.610 0.729 0.843 0.362 0.919 0.870
#> SD:pam 3 0.305 0.624 0.819 1.490 0.592 0.491
#> CV:pam 3 0.335 0.612 0.808 1.044 0.599 0.476
#> MAD:pam 3 0.226 0.570 0.780 1.191 0.599 0.480
#> ATC:pam 3 0.768 0.843 0.937 0.107 0.984 0.974
#> SD:hclust 3 0.508 0.816 0.885 0.124 1.000 1.000
#> CV:hclust 3 0.387 0.699 0.877 0.142 0.954 0.919
#> MAD:hclust 3 0.461 0.778 0.885 0.142 0.942 0.895
#> ATC:hclust 3 0.668 0.743 0.911 3.132 0.875 0.866
get_stats(res_list, k = 4)
#> k 1-PAC mean_silhouette concordance area_increased Rand Jaccard
#> SD:NMF 4 0.441 0.517 0.740 0.0920 0.892 0.707
#> CV:NMF 4 0.427 0.439 0.676 0.0945 0.754 0.451
#> MAD:NMF 4 0.410 0.438 0.694 0.0962 0.761 0.463
#> ATC:NMF 4 0.547 0.747 0.860 0.3346 0.661 0.447
#> SD:skmeans 4 0.581 0.656 0.804 0.1349 0.876 0.663
#> CV:skmeans 4 0.566 0.584 0.776 0.1340 0.826 0.567
#> MAD:skmeans 4 0.539 0.609 0.782 0.1277 0.855 0.617
#> ATC:skmeans 4 0.787 0.871 0.915 0.1317 0.867 0.637
#> SD:mclust 4 0.523 0.606 0.775 0.0854 0.737 0.436
#> CV:mclust 4 0.493 0.641 0.689 0.1173 0.865 0.655
#> MAD:mclust 4 0.503 0.560 0.699 0.1168 0.851 0.621
#> ATC:mclust 4 0.827 0.875 0.927 0.1628 0.875 0.678
#> SD:kmeans 4 0.650 0.650 0.835 0.1205 0.942 0.867
#> CV:kmeans 4 0.668 0.742 0.845 0.1214 0.896 0.773
#> MAD:kmeans 4 0.578 0.631 0.808 0.1263 0.901 0.779
#> ATC:kmeans 4 0.581 0.833 0.884 0.2674 0.764 0.581
#> SD:pam 4 0.298 0.558 0.771 0.1170 0.923 0.827
#> CV:pam 4 0.342 0.622 0.799 0.0273 1.000 1.000
#> MAD:pam 4 0.250 0.564 0.741 0.0620 0.987 0.968
#> ATC:pam 4 0.582 0.631 0.823 0.4854 0.773 0.616
#> SD:hclust 4 0.475 0.709 0.868 0.0608 0.984 0.972
#> CV:hclust 4 0.385 0.692 0.872 0.0542 0.999 0.999
#> MAD:hclust 4 0.452 0.758 0.874 0.0554 0.985 0.969
#> ATC:hclust 4 0.661 -0.508 0.765 0.0922 0.733 0.710
get_stats(res_list, k = 5)
#> k 1-PAC mean_silhouette concordance area_increased Rand Jaccard
#> SD:NMF 5 0.490 0.420 0.651 0.09316 0.823 0.500
#> CV:NMF 5 0.493 0.392 0.636 0.07882 0.864 0.599
#> MAD:NMF 5 0.492 0.424 0.642 0.09767 0.782 0.414
#> ATC:NMF 5 0.587 0.706 0.807 0.15602 0.769 0.428
#> SD:skmeans 5 0.568 0.506 0.715 0.06448 0.910 0.678
#> CV:skmeans 5 0.545 0.511 0.719 0.06870 0.895 0.643
#> MAD:skmeans 5 0.559 0.542 0.712 0.06719 0.898 0.642
#> ATC:skmeans 5 0.773 0.786 0.865 0.06146 0.966 0.866
#> SD:mclust 5 0.600 0.532 0.694 0.07219 0.770 0.433
#> CV:mclust 5 0.611 0.649 0.788 0.07499 0.968 0.886
#> MAD:mclust 5 0.576 0.621 0.777 0.07484 0.910 0.699
#> ATC:mclust 5 0.829 0.848 0.912 -0.00469 0.915 0.748
#> SD:kmeans 5 0.641 0.638 0.801 0.08065 0.893 0.734
#> CV:kmeans 5 0.618 0.623 0.803 0.07989 0.916 0.782
#> MAD:kmeans 5 0.588 0.527 0.760 0.08356 0.871 0.681
#> ATC:kmeans 5 0.778 0.642 0.828 0.10469 0.994 0.983
#> SD:pam 5 0.303 0.518 0.763 0.02865 0.989 0.973
#> CV:pam 5 0.335 0.537 0.793 0.01537 0.997 0.991
#> MAD:pam 5 0.261 0.495 0.720 0.02880 0.977 0.943
#> ATC:pam 5 0.554 0.691 0.794 0.07456 0.893 0.742
#> SD:hclust 5 0.489 0.714 0.858 0.04871 0.941 0.893
#> CV:hclust 5 0.384 0.662 0.861 0.04119 0.985 0.971
#> MAD:hclust 5 0.428 0.749 0.867 0.03248 0.999 0.999
#> ATC:hclust 5 0.599 0.757 0.889 0.17697 0.627 0.567
get_stats(res_list, k = 6)
#> k 1-PAC mean_silhouette concordance area_increased Rand Jaccard
#> SD:NMF 6 0.545 0.495 0.685 0.05110 0.895 0.600
#> CV:NMF 6 0.544 0.444 0.658 0.04792 0.875 0.533
#> MAD:NMF 6 0.606 0.513 0.689 0.04692 0.914 0.651
#> ATC:NMF 6 0.602 0.583 0.772 0.04302 0.963 0.844
#> SD:skmeans 6 0.602 0.501 0.688 0.04317 0.905 0.608
#> CV:skmeans 6 0.542 0.428 0.607 0.04083 0.936 0.721
#> MAD:skmeans 6 0.580 0.490 0.675 0.04081 0.909 0.617
#> ATC:skmeans 6 0.793 0.637 0.807 0.04190 0.942 0.752
#> SD:mclust 6 0.709 0.698 0.811 0.01301 0.836 0.540
#> CV:mclust 6 0.665 0.647 0.734 0.02552 0.905 0.671
#> MAD:mclust 6 0.715 0.677 0.776 0.00145 0.807 0.476
#> ATC:mclust 6 0.722 0.728 0.793 0.08116 0.904 0.679
#> SD:kmeans 6 0.650 0.584 0.757 0.06363 0.952 0.850
#> CV:kmeans 6 0.634 0.594 0.767 0.05623 0.947 0.836
#> MAD:kmeans 6 0.622 0.528 0.711 0.05255 0.895 0.664
#> ATC:kmeans 6 0.727 0.553 0.759 0.05437 0.946 0.842
#> SD:pam 6 0.312 0.553 0.763 0.01817 1.000 1.000
#> CV:pam 6 0.352 0.461 0.792 0.02035 0.991 0.975
#> MAD:pam 6 0.255 0.501 0.704 0.01366 0.965 0.915
#> ATC:pam 6 0.537 0.474 0.716 0.04810 0.856 0.618
#> SD:hclust 6 0.451 0.688 0.850 0.04877 0.975 0.948
#> CV:hclust 6 0.398 0.672 0.845 0.04471 0.985 0.970
#> MAD:hclust 6 0.433 0.727 0.835 0.03631 1.000 1.000
#> ATC:hclust 6 0.565 0.733 0.847 0.16957 0.962 0.937
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 = 835, method = "euler")
top_rows_overlap(res_list, top_n = 1670, method = "euler")
top_rows_overlap(res_list, top_n = 2506, method = "euler")
top_rows_overlap(res_list, top_n = 3341, method = "euler")
top_rows_overlap(res_list, top_n = 4176, method = "euler")
Also visualize the correspondance of rankings between different top-row methods:
top_rows_overlap(res_list, top_n = 835, method = "correspondance")
top_rows_overlap(res_list, top_n = 1670, method = "correspondance")
top_rows_overlap(res_list, top_n = 2506, method = "correspondance")
top_rows_overlap(res_list, top_n = 3341, method = "correspondance")
top_rows_overlap(res_list, top_n = 4176, method = "correspondance")
Heatmaps of the top rows:
top_rows_heatmap(res_list, top_n = 835)
top_rows_heatmap(res_list, top_n = 1670)
top_rows_heatmap(res_list, top_n = 2506)
top_rows_heatmap(res_list, top_n = 3341)
top_rows_heatmap(res_list, top_n = 4176)
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) other(p) protocol(p) k
#> SD:NMF 84 0.0959 5.48e-07 4.41e-05 2
#> CV:NMF 82 0.1504 2.11e-07 1.17e-05 2
#> MAD:NMF 83 0.1379 1.27e-07 2.42e-06 2
#> ATC:NMF 86 0.1284 8.56e-08 4.44e-07 2
#> SD:skmeans 83 0.0736 1.42e-06 8.31e-05 2
#> CV:skmeans 82 0.0814 3.50e-07 1.71e-05 2
#> MAD:skmeans 82 0.0814 1.64e-07 6.97e-06 2
#> ATC:skmeans 87 0.0715 3.03e-07 7.30e-07 2
#> SD:mclust 69 0.1387 1.02e-09 1.23e-08 2
#> CV:mclust 83 1.0000 2.68e-05 2.45e-04 2
#> MAD:mclust 69 0.2090 9.91e-10 1.26e-08 2
#> ATC:mclust 86 0.2058 3.97e-09 1.18e-08 2
#> SD:kmeans 80 0.1285 1.76e-08 2.87e-07 2
#> CV:kmeans 84 0.0731 6.30e-07 8.55e-05 2
#> MAD:kmeans 84 0.1832 7.77e-10 1.25e-09 2
#> ATC:kmeans 85 0.2206 6.68e-07 8.53e-07 2
#> SD:pam 87 0.3413 3.14e-04 2.23e-05 2
#> CV:pam 73 0.2096 2.00e-05 1.10e-04 2
#> MAD:pam 76 0.3153 1.11e-04 3.79e-05 2
#> ATC:pam 82 0.2992 8.44e-07 5.96e-07 2
#> SD:hclust 85 0.1219 5.85e-12 6.19e-13 2
#> CV:hclust 82 0.2713 1.83e-11 1.99e-12 2
#> MAD:hclust 83 0.2514 1.04e-11 1.02e-12 2
#> ATC:hclust 87 1.0000 2.02e-01 1.02e-02 2
test_to_known_factors(res_list, k = 3)
#> n disease.state(p) other(p) protocol(p) k
#> SD:NMF 62 0.04454 1.51e-07 6.20e-07 3
#> CV:NMF 46 0.27464 1.45e-07 1.77e-05 3
#> MAD:NMF 46 0.05942 5.30e-08 1.71e-05 3
#> ATC:NMF 82 0.16365 1.20e-07 1.62e-08 3
#> SD:skmeans 84 0.01736 2.53e-10 1.37e-10 3
#> CV:skmeans 68 0.18544 2.21e-07 1.59e-08 3
#> MAD:skmeans 79 0.27650 5.54e-09 3.42e-09 3
#> ATC:skmeans 84 0.00149 3.65e-09 3.07e-09 3
#> SD:mclust 69 0.00106 2.98e-11 9.01e-09 3
#> CV:mclust 84 0.00684 5.95e-13 2.98e-10 3
#> MAD:mclust 81 0.00209 5.35e-13 6.46e-10 3
#> ATC:mclust 76 0.01616 2.11e-09 8.51e-10 3
#> SD:kmeans 84 0.02516 6.83e-09 2.65e-08 3
#> CV:kmeans 83 0.06557 1.18e-08 3.87e-08 3
#> MAD:kmeans 78 0.03446 3.35e-09 3.56e-09 3
#> ATC:kmeans 82 0.32209 7.22e-06 1.70e-06 3
#> SD:pam 71 0.02270 1.94e-11 1.88e-08 3
#> CV:pam 68 0.00188 4.37e-12 8.74e-09 3
#> MAD:pam 65 0.01086 1.89e-10 3.81e-08 3
#> ATC:pam 80 0.38094 2.38e-05 7.41e-06 3
#> SD:hclust 84 0.11555 2.06e-12 9.05e-13 3
#> CV:hclust 76 0.17615 4.10e-10 3.40e-10 3
#> MAD:hclust 82 0.18675 1.48e-10 2.61e-11 3
#> ATC:hclust 72 0.44420 2.30e-03 2.86e-04 3
test_to_known_factors(res_list, k = 4)
#> n disease.state(p) other(p) protocol(p) k
#> SD:NMF 56 9.28e-03 8.15e-08 8.74e-10 4
#> CV:NMF 50 6.04e-01 2.73e-05 3.51e-08 4
#> MAD:NMF 47 1.09e-02 6.19e-07 3.24e-09 4
#> ATC:NMF 79 5.19e-02 9.69e-06 9.32e-06 4
#> SD:skmeans 73 1.16e-07 3.85e-14 6.37e-15 4
#> CV:skmeans 66 5.34e-03 4.49e-09 2.77e-09 4
#> MAD:skmeans 66 2.60e-11 2.90e-18 4.43e-18 4
#> ATC:skmeans 84 2.71e-11 4.43e-16 4.88e-16 4
#> SD:mclust 66 1.52e-02 8.78e-06 2.01e-08 4
#> CV:mclust 70 8.41e-02 2.03e-10 4.56e-07 4
#> MAD:mclust 63 7.76e-03 4.26e-10 5.61e-09 4
#> ATC:mclust 84 1.39e-02 2.13e-09 2.74e-09 4
#> SD:kmeans 70 5.24e-02 1.50e-06 6.19e-09 4
#> CV:kmeans 77 2.69e-01 3.02e-07 4.17e-08 4
#> MAD:kmeans 65 1.72e-01 2.60e-05 7.02e-07 4
#> ATC:kmeans 83 1.25e-02 3.24e-07 3.76e-08 4
#> SD:pam 58 2.06e-03 1.88e-09 4.74e-10 4
#> CV:pam 69 1.58e-03 1.32e-11 2.29e-08 4
#> MAD:pam 62 7.55e-03 1.18e-10 3.24e-08 4
#> ATC:pam 77 5.51e-02 1.07e-05 2.05e-06 4
#> SD:hclust 77 1.54e-01 8.24e-12 1.35e-11 4
#> CV:hclust 73 1.77e-01 9.63e-12 6.37e-11 4
#> MAD:hclust 79 2.37e-01 8.65e-11 1.16e-10 4
#> ATC:hclust 6 NA NA NA 4
test_to_known_factors(res_list, k = 5)
#> n disease.state(p) other(p) protocol(p) k
#> SD:NMF 38 1.12e-07 6.99e-10 1.45e-10 5
#> CV:NMF 35 5.02e-02 1.84e-04 5.27e-05 5
#> MAD:NMF 36 2.89e-07 7.84e-08 8.44e-11 5
#> ATC:NMF 73 1.03e-07 1.98e-09 8.01e-11 5
#> SD:skmeans 48 NA 1.16e-05 6.35e-07 5
#> CV:skmeans 57 2.46e-04 1.64e-08 2.65e-09 5
#> MAD:skmeans 59 1.29e-05 1.25e-09 2.09e-10 5
#> ATC:skmeans 82 2.54e-11 3.98e-14 6.24e-15 5
#> SD:mclust 59 2.11e-02 1.40e-07 5.06e-08 5
#> CV:mclust 75 6.13e-02 4.22e-11 3.01e-07 5
#> MAD:mclust 61 7.86e-02 1.02e-06 9.16e-08 5
#> ATC:mclust 82 1.79e-02 2.46e-07 3.55e-09 5
#> SD:kmeans 70 1.67e-01 5.61e-06 2.30e-07 5
#> CV:kmeans 73 2.36e-01 2.19e-08 1.21e-09 5
#> MAD:kmeans 59 5.67e-01 1.46e-04 1.67e-06 5
#> ATC:kmeans 62 3.64e-01 1.78e-04 1.18e-05 5
#> SD:pam 57 1.66e-03 6.53e-10 6.63e-09 5
#> CV:pam 59 6.20e-03 1.38e-09 2.67e-07 5
#> MAD:pam 56 2.01e-02 8.84e-10 1.93e-07 5
#> ATC:pam 79 6.55e-02 7.50e-06 2.53e-06 5
#> SD:hclust 75 2.79e-01 1.23e-09 1.74e-09 5
#> CV:hclust 73 1.77e-01 9.63e-12 6.37e-11 5
#> MAD:hclust 78 2.01e-01 4.63e-12 7.17e-12 5
#> ATC:hclust 75 6.34e-01 3.16e-04 5.29e-06 5
test_to_known_factors(res_list, k = 6)
#> n disease.state(p) other(p) protocol(p) k
#> SD:NMF 54 2.10e-10 2.93e-11 4.51e-15 6
#> CV:NMF 47 NA 8.00e-03 1.94e-06 6
#> MAD:NMF 51 8.65e-10 3.73e-10 4.97e-14 6
#> ATC:NMF 60 1.41e-08 2.51e-10 2.40e-11 6
#> SD:skmeans 51 8.65e-10 2.01e-11 1.67e-15 6
#> CV:skmeans 42 1.67e-08 1.25e-08 7.63e-15 6
#> MAD:skmeans 56 2.15e-06 7.30e-09 8.39e-13 6
#> ATC:skmeans 66 3.12e-09 1.04e-10 3.18e-11 6
#> SD:mclust 77 3.91e-01 2.39e-07 1.94e-07 6
#> CV:mclust 64 7.10e-02 7.85e-08 1.91e-05 6
#> MAD:mclust 70 3.82e-01 2.95e-08 7.73e-08 6
#> ATC:mclust 77 3.56e-03 7.56e-09 2.82e-10 6
#> SD:kmeans 68 8.55e-01 2.76e-05 9.53e-07 6
#> CV:kmeans 61 4.58e-01 3.39e-05 1.43e-07 6
#> MAD:kmeans 56 4.43e-01 7.24e-05 8.45e-07 6
#> ATC:kmeans 56 6.37e-01 1.93e-02 3.86e-03 6
#> SD:pam 60 7.47e-03 3.99e-09 2.10e-08 6
#> CV:pam 50 1.41e-02 1.78e-08 5.06e-07 6
#> MAD:pam 56 2.05e-02 7.12e-09 3.69e-07 6
#> ATC:pam 60 5.45e-01 3.78e-03 4.26e-04 6
#> SD:hclust 69 6.25e-01 2.62e-07 2.97e-10 6
#> CV:hclust 70 2.68e-01 2.72e-10 2.61e-12 6
#> MAD:hclust 74 2.81e-01 4.43e-11 3.86e-13 6
#> ATC:hclust 76 4.92e-01 4.42e-05 7.40e-05 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 8353 rows and 87 columns.
#> Top rows (835, 1670, 2506, 3341, 4176) 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 0.623 0.874 0.927 0.4100 0.567 0.567
#> 3 3 0.508 0.816 0.885 0.1241 1.000 1.000
#> 4 4 0.475 0.709 0.868 0.0608 0.984 0.972
#> 5 5 0.489 0.714 0.858 0.0487 0.941 0.893
#> 6 6 0.451 0.688 0.850 0.0488 0.975 0.948
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
#> GSM39104 1 0.0000 0.943 1.000 0.000
#> GSM39105 1 0.0672 0.941 0.992 0.008
#> GSM39106 1 0.5178 0.838 0.884 0.116
#> GSM39107 1 0.8861 0.504 0.696 0.304
#> GSM39108 1 0.1633 0.932 0.976 0.024
#> GSM39109 1 0.3274 0.903 0.940 0.060
#> GSM39110 1 0.3114 0.907 0.944 0.056
#> GSM39111 1 0.1414 0.935 0.980 0.020
#> GSM39112 1 0.8813 0.514 0.700 0.300
#> GSM39113 1 0.8861 0.504 0.696 0.304
#> GSM39114 2 0.8955 0.711 0.312 0.688
#> GSM39115 1 0.0672 0.941 0.992 0.008
#> GSM39148 1 0.0000 0.943 1.000 0.000
#> GSM39149 1 0.0938 0.937 0.988 0.012
#> GSM39150 1 0.0000 0.943 1.000 0.000
#> GSM39151 1 0.0938 0.937 0.988 0.012
#> GSM39152 1 0.0000 0.943 1.000 0.000
#> GSM39153 1 0.0376 0.942 0.996 0.004
#> GSM39154 1 0.0000 0.943 1.000 0.000
#> GSM39155 1 0.0000 0.943 1.000 0.000
#> GSM39156 1 0.2043 0.928 0.968 0.032
#> GSM39157 1 0.0000 0.943 1.000 0.000
#> GSM39158 1 0.0000 0.943 1.000 0.000
#> GSM39159 1 0.0672 0.942 0.992 0.008
#> GSM39160 1 0.0000 0.943 1.000 0.000
#> GSM39161 1 0.1633 0.935 0.976 0.024
#> GSM39162 1 0.0000 0.943 1.000 0.000
#> GSM39163 1 0.0376 0.943 0.996 0.004
#> GSM39164 1 0.0000 0.943 1.000 0.000
#> GSM39165 1 0.0000 0.943 1.000 0.000
#> GSM39166 1 0.0000 0.943 1.000 0.000
#> GSM39167 1 0.0376 0.943 0.996 0.004
#> GSM39168 1 0.0000 0.943 1.000 0.000
#> GSM39169 1 0.0000 0.943 1.000 0.000
#> GSM39170 1 0.0000 0.943 1.000 0.000
#> GSM39171 1 0.0000 0.943 1.000 0.000
#> GSM39172 1 0.0000 0.943 1.000 0.000
#> GSM39173 1 0.3114 0.910 0.944 0.056
#> GSM39174 1 0.0000 0.943 1.000 0.000
#> GSM39175 1 0.0000 0.943 1.000 0.000
#> GSM39176 1 0.0376 0.943 0.996 0.004
#> GSM39177 1 0.0376 0.941 0.996 0.004
#> GSM39178 1 0.0000 0.943 1.000 0.000
#> GSM39179 1 0.0938 0.937 0.988 0.012
#> GSM39180 1 0.6973 0.753 0.812 0.188
#> GSM39181 1 0.0000 0.943 1.000 0.000
#> GSM39182 1 0.1843 0.931 0.972 0.028
#> GSM39183 1 0.0000 0.943 1.000 0.000
#> GSM39184 1 0.0000 0.943 1.000 0.000
#> GSM39185 1 0.1633 0.935 0.976 0.024
#> GSM39186 1 0.0672 0.941 0.992 0.008
#> GSM39187 1 0.1184 0.938 0.984 0.016
#> GSM39116 2 0.6343 0.886 0.160 0.840
#> GSM39117 2 0.1843 0.870 0.028 0.972
#> GSM39118 2 0.4431 0.885 0.092 0.908
#> GSM39119 2 0.1843 0.871 0.028 0.972
#> GSM39120 1 0.9522 0.314 0.628 0.372
#> GSM39121 2 0.8443 0.776 0.272 0.728
#> GSM39122 2 0.8207 0.798 0.256 0.744
#> GSM39123 2 0.1843 0.870 0.028 0.972
#> GSM39124 2 0.7056 0.871 0.192 0.808
#> GSM39125 1 0.9922 0.020 0.552 0.448
#> GSM39126 2 0.8909 0.716 0.308 0.692
#> GSM39127 2 0.6801 0.879 0.180 0.820
#> GSM39128 2 0.7139 0.867 0.196 0.804
#> GSM39129 2 0.1184 0.866 0.016 0.984
#> GSM39130 2 0.1843 0.870 0.028 0.972
#> GSM39131 2 0.6623 0.883 0.172 0.828
#> GSM39132 2 0.6438 0.886 0.164 0.836
#> GSM39133 2 0.1843 0.870 0.028 0.972
#> GSM39134 2 0.1843 0.870 0.028 0.972
#> GSM39135 2 0.6438 0.886 0.164 0.836
#> GSM39136 2 0.6048 0.887 0.148 0.852
#> GSM39137 2 0.7056 0.871 0.192 0.808
#> GSM39138 2 0.1184 0.866 0.016 0.984
#> GSM39139 2 0.1184 0.866 0.016 0.984
#> GSM39140 1 0.7674 0.677 0.776 0.224
#> GSM39141 1 0.2423 0.920 0.960 0.040
#> GSM39142 1 0.2423 0.920 0.960 0.040
#> GSM39143 1 0.2423 0.920 0.960 0.040
#> GSM39144 2 0.1184 0.866 0.016 0.984
#> GSM39145 2 0.4562 0.886 0.096 0.904
#> GSM39146 2 0.6623 0.883 0.172 0.828
#> GSM39147 2 0.7056 0.871 0.192 0.808
#> GSM39188 1 0.1184 0.933 0.984 0.016
#> GSM39189 1 0.0000 0.943 1.000 0.000
#> GSM39190 1 0.0672 0.939 0.992 0.008
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM39104 1 0.0000 0.9032 1.000 0.000 0.000
#> GSM39105 1 0.0424 0.9027 0.992 0.008 0.000
#> GSM39106 1 0.3412 0.8141 0.876 0.124 0.000
#> GSM39107 1 0.5650 0.5087 0.688 0.312 0.000
#> GSM39108 1 0.1163 0.8956 0.972 0.028 0.000
#> GSM39109 1 0.2165 0.8736 0.936 0.064 0.000
#> GSM39110 1 0.2066 0.8766 0.940 0.060 0.000
#> GSM39111 1 0.1031 0.8975 0.976 0.024 0.000
#> GSM39112 1 0.5621 0.5173 0.692 0.308 0.000
#> GSM39113 1 0.5650 0.5087 0.688 0.312 0.000
#> GSM39114 2 0.5591 0.6869 0.304 0.696 0.000
#> GSM39115 1 0.0424 0.9027 0.992 0.008 0.000
#> GSM39148 1 0.0000 0.9032 1.000 0.000 0.000
#> GSM39149 1 0.6155 0.6078 0.664 0.008 0.328
#> GSM39150 1 0.0000 0.9032 1.000 0.000 0.000
#> GSM39151 1 0.5621 0.6470 0.692 0.000 0.308
#> GSM39152 1 0.0000 0.9032 1.000 0.000 0.000
#> GSM39153 1 0.0237 0.9030 0.996 0.004 0.000
#> GSM39154 1 0.0000 0.9032 1.000 0.000 0.000
#> GSM39155 1 0.0000 0.9032 1.000 0.000 0.000
#> GSM39156 1 0.1289 0.8945 0.968 0.032 0.000
#> GSM39157 1 0.0000 0.9032 1.000 0.000 0.000
#> GSM39158 1 0.0892 0.9000 0.980 0.000 0.020
#> GSM39159 1 0.1525 0.8940 0.964 0.004 0.032
#> GSM39160 1 0.0000 0.9032 1.000 0.000 0.000
#> GSM39161 1 0.3207 0.8598 0.904 0.012 0.084
#> GSM39162 1 0.0000 0.9032 1.000 0.000 0.000
#> GSM39163 1 0.0237 0.9033 0.996 0.004 0.000
#> GSM39164 1 0.0000 0.9032 1.000 0.000 0.000
#> GSM39165 1 0.0000 0.9032 1.000 0.000 0.000
#> GSM39166 1 0.1289 0.8950 0.968 0.000 0.032
#> GSM39167 1 0.0237 0.9033 0.996 0.004 0.000
#> GSM39168 1 0.0000 0.9032 1.000 0.000 0.000
#> GSM39169 1 0.0000 0.9032 1.000 0.000 0.000
#> GSM39170 1 0.0747 0.9000 0.984 0.000 0.016
#> GSM39171 1 0.0000 0.9032 1.000 0.000 0.000
#> GSM39172 1 0.0661 0.9027 0.988 0.008 0.004
#> GSM39173 1 0.4097 0.8436 0.880 0.060 0.060
#> GSM39174 1 0.0000 0.9032 1.000 0.000 0.000
#> GSM39175 1 0.0000 0.9032 1.000 0.000 0.000
#> GSM39176 1 0.0237 0.9033 0.996 0.004 0.000
#> GSM39177 1 0.1647 0.8928 0.960 0.004 0.036
#> GSM39178 1 0.0000 0.9032 1.000 0.000 0.000
#> GSM39179 1 0.6104 0.5846 0.648 0.004 0.348
#> GSM39180 1 0.8261 0.5214 0.616 0.124 0.260
#> GSM39181 1 0.1289 0.8950 0.968 0.000 0.032
#> GSM39182 1 0.1585 0.8955 0.964 0.028 0.008
#> GSM39183 1 0.1289 0.8950 0.968 0.000 0.032
#> GSM39184 1 0.0000 0.9032 1.000 0.000 0.000
#> GSM39185 1 0.3207 0.8598 0.904 0.012 0.084
#> GSM39186 1 0.0424 0.9027 0.992 0.008 0.000
#> GSM39187 1 0.0747 0.9007 0.984 0.016 0.000
#> GSM39116 2 0.4228 0.8584 0.148 0.844 0.008
#> GSM39117 2 0.5502 0.7813 0.008 0.744 0.248
#> GSM39118 2 0.3550 0.8549 0.080 0.896 0.024
#> GSM39119 2 0.3454 0.8278 0.008 0.888 0.104
#> GSM39120 1 0.6062 0.3173 0.616 0.384 0.000
#> GSM39121 2 0.5216 0.7533 0.260 0.740 0.000
#> GSM39122 2 0.5058 0.7749 0.244 0.756 0.000
#> GSM39123 2 0.5502 0.7813 0.008 0.744 0.248
#> GSM39124 2 0.4291 0.8438 0.180 0.820 0.000
#> GSM39125 1 0.6280 0.0349 0.540 0.460 0.000
#> GSM39126 2 0.5529 0.6946 0.296 0.704 0.000
#> GSM39127 2 0.4121 0.8515 0.168 0.832 0.000
#> GSM39128 2 0.4346 0.8404 0.184 0.816 0.000
#> GSM39129 2 0.2356 0.8209 0.000 0.928 0.072
#> GSM39130 2 0.5502 0.7813 0.008 0.744 0.248
#> GSM39131 2 0.4233 0.8554 0.160 0.836 0.004
#> GSM39132 2 0.4110 0.8576 0.152 0.844 0.004
#> GSM39133 2 0.5502 0.7813 0.008 0.744 0.248
#> GSM39134 2 0.2860 0.8216 0.004 0.912 0.084
#> GSM39135 2 0.4110 0.8576 0.152 0.844 0.004
#> GSM39136 2 0.4195 0.8592 0.136 0.852 0.012
#> GSM39137 2 0.4291 0.8438 0.180 0.820 0.000
#> GSM39138 2 0.2261 0.8218 0.000 0.932 0.068
#> GSM39139 2 0.2261 0.8218 0.000 0.932 0.068
#> GSM39140 1 0.4931 0.6677 0.768 0.232 0.000
#> GSM39141 1 0.1753 0.8851 0.952 0.048 0.000
#> GSM39142 1 0.1753 0.8851 0.952 0.048 0.000
#> GSM39143 1 0.1753 0.8851 0.952 0.048 0.000
#> GSM39144 2 0.2261 0.8218 0.000 0.932 0.068
#> GSM39145 2 0.4035 0.8521 0.080 0.880 0.040
#> GSM39146 2 0.4575 0.8557 0.160 0.828 0.012
#> GSM39147 2 0.4291 0.8438 0.180 0.820 0.000
#> GSM39188 1 0.6280 0.4378 0.540 0.000 0.460
#> GSM39189 1 0.1129 0.9001 0.976 0.004 0.020
#> GSM39190 1 0.5902 0.6320 0.680 0.004 0.316
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM39104 1 0.0000 0.870 1.000 0.000 0.000 NA
#> GSM39105 1 0.0336 0.870 0.992 0.008 0.000 NA
#> GSM39106 1 0.2704 0.777 0.876 0.124 0.000 NA
#> GSM39107 1 0.4477 0.496 0.688 0.312 0.000 NA
#> GSM39108 1 0.0921 0.863 0.972 0.028 0.000 NA
#> GSM39109 1 0.1902 0.839 0.932 0.064 0.000 NA
#> GSM39110 1 0.1637 0.842 0.940 0.060 0.000 NA
#> GSM39111 1 0.0817 0.865 0.976 0.024 0.000 NA
#> GSM39112 1 0.4454 0.505 0.692 0.308 0.000 NA
#> GSM39113 1 0.4477 0.496 0.688 0.312 0.000 NA
#> GSM39114 2 0.4431 0.622 0.304 0.696 0.000 NA
#> GSM39115 1 0.0336 0.870 0.992 0.008 0.000 NA
#> GSM39148 1 0.0000 0.870 1.000 0.000 0.000 NA
#> GSM39149 1 0.7935 -0.273 0.460 0.008 0.256 NA
#> GSM39150 1 0.0376 0.868 0.992 0.000 0.004 NA
#> GSM39151 1 0.7631 -0.313 0.456 0.000 0.320 NA
#> GSM39152 1 0.0376 0.868 0.992 0.000 0.004 NA
#> GSM39153 1 0.0188 0.870 0.996 0.004 0.000 NA
#> GSM39154 1 0.0000 0.870 1.000 0.000 0.000 NA
#> GSM39155 1 0.0000 0.870 1.000 0.000 0.000 NA
#> GSM39156 1 0.1022 0.862 0.968 0.032 0.000 NA
#> GSM39157 1 0.0000 0.870 1.000 0.000 0.000 NA
#> GSM39158 1 0.1004 0.864 0.972 0.000 0.004 NA
#> GSM39159 1 0.1489 0.852 0.952 0.000 0.004 NA
#> GSM39160 1 0.0376 0.868 0.992 0.000 0.004 NA
#> GSM39161 1 0.2922 0.799 0.884 0.008 0.004 NA
#> GSM39162 1 0.0000 0.870 1.000 0.000 0.000 NA
#> GSM39163 1 0.0188 0.870 0.996 0.004 0.000 NA
#> GSM39164 1 0.0000 0.870 1.000 0.000 0.000 NA
#> GSM39165 1 0.0336 0.870 0.992 0.000 0.008 NA
#> GSM39166 1 0.1305 0.857 0.960 0.000 0.004 NA
#> GSM39167 1 0.0188 0.870 0.996 0.004 0.000 NA
#> GSM39168 1 0.0000 0.870 1.000 0.000 0.000 NA
#> GSM39169 1 0.0000 0.870 1.000 0.000 0.000 NA
#> GSM39170 1 0.0779 0.866 0.980 0.000 0.004 NA
#> GSM39171 1 0.0376 0.868 0.992 0.000 0.004 NA
#> GSM39172 1 0.1247 0.864 0.968 0.004 0.016 NA
#> GSM39173 1 0.4439 0.757 0.840 0.056 0.052 NA
#> GSM39174 1 0.0000 0.870 1.000 0.000 0.000 NA
#> GSM39175 1 0.0188 0.869 0.996 0.000 0.004 NA
#> GSM39176 1 0.0188 0.870 0.996 0.004 0.000 NA
#> GSM39177 1 0.2587 0.817 0.908 0.004 0.076 NA
#> GSM39178 1 0.0376 0.868 0.992 0.000 0.004 NA
#> GSM39179 1 0.7730 -0.218 0.480 0.004 0.236 NA
#> GSM39180 1 0.7617 0.186 0.556 0.088 0.052 NA
#> GSM39181 1 0.1305 0.857 0.960 0.000 0.004 NA
#> GSM39182 1 0.1739 0.860 0.952 0.024 0.008 NA
#> GSM39183 1 0.1305 0.857 0.960 0.000 0.004 NA
#> GSM39184 1 0.0000 0.870 1.000 0.000 0.000 NA
#> GSM39185 1 0.2922 0.799 0.884 0.008 0.004 NA
#> GSM39186 1 0.0336 0.870 0.992 0.008 0.000 NA
#> GSM39187 1 0.0592 0.868 0.984 0.016 0.000 NA
#> GSM39116 2 0.3479 0.792 0.148 0.840 0.000 NA
#> GSM39117 2 0.4406 0.566 0.000 0.700 0.000 NA
#> GSM39118 2 0.3013 0.768 0.080 0.888 0.000 NA
#> GSM39119 2 0.3351 0.690 0.008 0.844 0.000 NA
#> GSM39120 1 0.4804 0.309 0.616 0.384 0.000 NA
#> GSM39121 2 0.4134 0.687 0.260 0.740 0.000 NA
#> GSM39122 2 0.4008 0.708 0.244 0.756 0.000 NA
#> GSM39123 2 0.4406 0.566 0.000 0.700 0.000 NA
#> GSM39124 2 0.3400 0.780 0.180 0.820 0.000 NA
#> GSM39125 1 0.4977 0.033 0.540 0.460 0.000 NA
#> GSM39126 2 0.4382 0.630 0.296 0.704 0.000 NA
#> GSM39127 2 0.3266 0.787 0.168 0.832 0.000 NA
#> GSM39128 2 0.3444 0.776 0.184 0.816 0.000 NA
#> GSM39129 2 0.2654 0.665 0.000 0.888 0.004 NA
#> GSM39130 2 0.4406 0.566 0.000 0.700 0.000 NA
#> GSM39131 2 0.3355 0.791 0.160 0.836 0.000 NA
#> GSM39132 2 0.3257 0.792 0.152 0.844 0.000 NA
#> GSM39133 2 0.4406 0.566 0.000 0.700 0.000 NA
#> GSM39134 2 0.2888 0.668 0.000 0.872 0.004 NA
#> GSM39135 2 0.3401 0.792 0.152 0.840 0.000 NA
#> GSM39136 2 0.3443 0.790 0.136 0.848 0.000 NA
#> GSM39137 2 0.3400 0.780 0.180 0.820 0.000 NA
#> GSM39138 2 0.2654 0.664 0.000 0.888 0.004 NA
#> GSM39139 2 0.2654 0.664 0.000 0.888 0.004 NA
#> GSM39140 1 0.3907 0.639 0.768 0.232 0.000 NA
#> GSM39141 1 0.1389 0.852 0.952 0.048 0.000 NA
#> GSM39142 1 0.1389 0.852 0.952 0.048 0.000 NA
#> GSM39143 1 0.1389 0.852 0.952 0.048 0.000 NA
#> GSM39144 2 0.2654 0.664 0.000 0.888 0.004 NA
#> GSM39145 2 0.3761 0.751 0.080 0.852 0.000 NA
#> GSM39146 2 0.3625 0.791 0.160 0.828 0.000 NA
#> GSM39147 2 0.3400 0.780 0.180 0.820 0.000 NA
#> GSM39188 3 0.3024 0.000 0.148 0.000 0.852 NA
#> GSM39189 1 0.1892 0.851 0.944 0.004 0.016 NA
#> GSM39190 1 0.7745 -0.343 0.436 0.004 0.360 NA
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM39104 1 0.0162 0.87775 0.996 0.000 0.000 0.000 NA
#> GSM39105 1 0.0290 0.87751 0.992 0.008 0.000 0.000 NA
#> GSM39106 1 0.2536 0.74364 0.868 0.128 0.000 0.000 NA
#> GSM39107 1 0.4047 0.38379 0.676 0.320 0.000 0.000 NA
#> GSM39108 1 0.1041 0.86292 0.964 0.032 0.000 0.000 NA
#> GSM39109 1 0.1928 0.82458 0.920 0.072 0.004 0.000 NA
#> GSM39110 1 0.1704 0.83008 0.928 0.068 0.000 0.000 NA
#> GSM39111 1 0.0865 0.86849 0.972 0.024 0.000 0.000 NA
#> GSM39112 1 0.4029 0.39197 0.680 0.316 0.000 0.000 NA
#> GSM39113 1 0.4047 0.38379 0.676 0.320 0.000 0.000 NA
#> GSM39114 2 0.3884 0.58877 0.288 0.708 0.000 0.000 NA
#> GSM39115 1 0.0290 0.87751 0.992 0.008 0.000 0.000 NA
#> GSM39148 1 0.0000 0.87785 1.000 0.000 0.000 0.000 NA
#> GSM39149 3 0.5062 0.60224 0.308 0.004 0.648 0.032 NA
#> GSM39150 1 0.0486 0.87666 0.988 0.000 0.004 0.004 NA
#> GSM39151 3 0.7141 0.61258 0.344 0.000 0.448 0.172 NA
#> GSM39152 1 0.0566 0.87270 0.984 0.000 0.012 0.004 NA
#> GSM39153 1 0.0162 0.87790 0.996 0.004 0.000 0.000 NA
#> GSM39154 1 0.0000 0.87785 1.000 0.000 0.000 0.000 NA
#> GSM39155 1 0.0000 0.87785 1.000 0.000 0.000 0.000 NA
#> GSM39156 1 0.0880 0.86661 0.968 0.032 0.000 0.000 NA
#> GSM39157 1 0.0000 0.87785 1.000 0.000 0.000 0.000 NA
#> GSM39158 1 0.1059 0.86675 0.968 0.000 0.004 0.008 NA
#> GSM39159 1 0.1282 0.85452 0.952 0.000 0.000 0.004 NA
#> GSM39160 1 0.0324 0.87564 0.992 0.000 0.004 0.004 NA
#> GSM39161 1 0.2919 0.77768 0.880 0.008 0.016 0.008 NA
#> GSM39162 1 0.0000 0.87785 1.000 0.000 0.000 0.000 NA
#> GSM39163 1 0.0162 0.87836 0.996 0.004 0.000 0.000 NA
#> GSM39164 1 0.0000 0.87785 1.000 0.000 0.000 0.000 NA
#> GSM39165 1 0.0324 0.87723 0.992 0.000 0.004 0.004 NA
#> GSM39166 1 0.1329 0.85701 0.956 0.000 0.004 0.008 NA
#> GSM39167 1 0.0162 0.87836 0.996 0.004 0.000 0.000 NA
#> GSM39168 1 0.0000 0.87785 1.000 0.000 0.000 0.000 NA
#> GSM39169 1 0.0000 0.87785 1.000 0.000 0.000 0.000 NA
#> GSM39170 1 0.0833 0.86939 0.976 0.000 0.004 0.004 NA
#> GSM39171 1 0.0324 0.87564 0.992 0.000 0.004 0.004 NA
#> GSM39172 1 0.1329 0.85733 0.956 0.000 0.032 0.004 NA
#> GSM39173 1 0.5019 0.59793 0.776 0.036 0.100 0.020 NA
#> GSM39174 1 0.0000 0.87785 1.000 0.000 0.000 0.000 NA
#> GSM39175 1 0.0162 0.87662 0.996 0.000 0.000 0.004 NA
#> GSM39176 1 0.0162 0.87836 0.996 0.004 0.000 0.000 NA
#> GSM39177 1 0.3154 0.68255 0.836 0.004 0.148 0.012 NA
#> GSM39178 1 0.0324 0.87564 0.992 0.000 0.004 0.004 NA
#> GSM39179 3 0.5996 0.62627 0.364 0.004 0.552 0.060 NA
#> GSM39180 1 0.7104 -0.20726 0.524 0.084 0.052 0.020 NA
#> GSM39181 1 0.1329 0.85701 0.956 0.000 0.004 0.008 NA
#> GSM39182 1 0.1757 0.85964 0.944 0.028 0.012 0.004 NA
#> GSM39183 1 0.1329 0.85701 0.956 0.000 0.004 0.008 NA
#> GSM39184 1 0.0000 0.87785 1.000 0.000 0.000 0.000 NA
#> GSM39185 1 0.2919 0.77768 0.880 0.008 0.016 0.008 NA
#> GSM39186 1 0.0290 0.87751 0.992 0.008 0.000 0.000 NA
#> GSM39187 1 0.0510 0.87507 0.984 0.016 0.000 0.000 NA
#> GSM39116 2 0.3106 0.76964 0.140 0.840 0.000 0.000 NA
#> GSM39117 2 0.4201 0.43542 0.000 0.592 0.000 0.000 NA
#> GSM39118 2 0.2726 0.73324 0.064 0.884 0.000 0.000 NA
#> GSM39119 2 0.3783 0.60077 0.008 0.740 0.000 0.000 NA
#> GSM39120 1 0.4299 0.23790 0.608 0.388 0.000 0.000 NA
#> GSM39121 2 0.3607 0.66088 0.244 0.752 0.000 0.000 NA
#> GSM39122 2 0.3491 0.68440 0.228 0.768 0.000 0.000 NA
#> GSM39123 2 0.4201 0.43542 0.000 0.592 0.000 0.000 NA
#> GSM39124 2 0.2813 0.75730 0.168 0.832 0.000 0.000 NA
#> GSM39125 1 0.4440 0.00666 0.528 0.468 0.000 0.000 NA
#> GSM39126 2 0.3838 0.59855 0.280 0.716 0.000 0.000 NA
#> GSM39127 2 0.2690 0.76489 0.156 0.844 0.000 0.000 NA
#> GSM39128 2 0.2852 0.75369 0.172 0.828 0.000 0.000 NA
#> GSM39129 2 0.3863 0.56025 0.000 0.792 0.012 0.020 NA
#> GSM39130 2 0.4201 0.43542 0.000 0.592 0.000 0.000 NA
#> GSM39131 2 0.2763 0.76835 0.148 0.848 0.000 0.000 NA
#> GSM39132 2 0.2674 0.76923 0.140 0.856 0.000 0.000 NA
#> GSM39133 2 0.4201 0.43542 0.000 0.592 0.000 0.000 NA
#> GSM39134 2 0.3333 0.58116 0.000 0.788 0.000 0.004 NA
#> GSM39135 2 0.2909 0.76963 0.140 0.848 0.000 0.000 NA
#> GSM39136 2 0.3229 0.76629 0.128 0.840 0.000 0.000 NA
#> GSM39137 2 0.2813 0.75730 0.168 0.832 0.000 0.000 NA
#> GSM39138 2 0.3209 0.57717 0.000 0.812 0.000 0.008 NA
#> GSM39139 2 0.3209 0.57599 0.000 0.812 0.000 0.008 NA
#> GSM39140 1 0.3521 0.56369 0.764 0.232 0.000 0.000 NA
#> GSM39141 1 0.1357 0.85105 0.948 0.048 0.000 0.000 NA
#> GSM39142 1 0.1357 0.85105 0.948 0.048 0.000 0.000 NA
#> GSM39143 1 0.1357 0.85105 0.948 0.048 0.000 0.000 NA
#> GSM39144 2 0.3209 0.57599 0.000 0.812 0.000 0.008 NA
#> GSM39145 2 0.3586 0.71141 0.076 0.828 0.000 0.000 NA
#> GSM39146 2 0.3326 0.76736 0.152 0.824 0.000 0.000 NA
#> GSM39147 2 0.2813 0.75730 0.168 0.832 0.000 0.000 NA
#> GSM39188 4 0.2304 0.00000 0.048 0.000 0.044 0.908 NA
#> GSM39189 1 0.1978 0.83141 0.928 0.000 0.044 0.004 NA
#> GSM39190 3 0.8270 0.05575 0.228 0.000 0.396 0.172 NA
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM39104 1 0.0260 0.8908 0.992 0.000 0.000 0.008 0.000 0.000
#> GSM39105 1 0.0260 0.8914 0.992 0.008 0.000 0.000 0.000 0.000
#> GSM39106 1 0.2473 0.7787 0.856 0.136 0.000 0.008 0.000 0.000
#> GSM39107 1 0.3789 0.4579 0.660 0.332 0.000 0.008 0.000 0.000
#> GSM39108 1 0.1453 0.8709 0.944 0.040 0.008 0.008 0.000 0.000
#> GSM39109 1 0.2395 0.8354 0.892 0.076 0.020 0.012 0.000 0.000
#> GSM39110 1 0.2058 0.8434 0.908 0.072 0.008 0.012 0.000 0.000
#> GSM39111 1 0.1230 0.8787 0.956 0.028 0.008 0.008 0.000 0.000
#> GSM39112 1 0.3774 0.4674 0.664 0.328 0.000 0.008 0.000 0.000
#> GSM39113 1 0.3789 0.4579 0.660 0.332 0.000 0.008 0.000 0.000
#> GSM39114 2 0.3512 0.5349 0.272 0.720 0.000 0.008 0.000 0.000
#> GSM39115 1 0.0260 0.8914 0.992 0.008 0.000 0.000 0.000 0.000
#> GSM39148 1 0.0000 0.8913 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM39149 3 0.5516 0.4967 0.172 0.000 0.612 0.008 0.004 0.204
#> GSM39150 1 0.0622 0.8891 0.980 0.000 0.012 0.008 0.000 0.000
#> GSM39151 3 0.6967 0.4622 0.156 0.000 0.572 0.052 0.104 0.116
#> GSM39152 1 0.0692 0.8852 0.976 0.000 0.020 0.000 0.000 0.004
#> GSM39153 1 0.0146 0.8915 0.996 0.004 0.000 0.000 0.000 0.000
#> GSM39154 1 0.0000 0.8913 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM39155 1 0.0000 0.8913 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM39156 1 0.0790 0.8835 0.968 0.032 0.000 0.000 0.000 0.000
#> GSM39157 1 0.0000 0.8913 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM39158 1 0.0951 0.8833 0.968 0.000 0.008 0.000 0.004 0.020
#> GSM39159 1 0.1425 0.8755 0.952 0.000 0.008 0.020 0.008 0.012
#> GSM39160 1 0.0508 0.8888 0.984 0.000 0.012 0.004 0.000 0.000
#> GSM39161 1 0.3129 0.8113 0.872 0.008 0.032 0.028 0.008 0.052
#> GSM39162 1 0.0000 0.8913 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM39163 1 0.0146 0.8919 0.996 0.004 0.000 0.000 0.000 0.000
#> GSM39164 1 0.0000 0.8913 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM39165 1 0.0260 0.8911 0.992 0.000 0.008 0.000 0.000 0.000
#> GSM39166 1 0.1294 0.8757 0.956 0.000 0.008 0.008 0.004 0.024
#> GSM39167 1 0.0146 0.8919 0.996 0.004 0.000 0.000 0.000 0.000
#> GSM39168 1 0.0000 0.8913 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM39169 1 0.0000 0.8913 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM39170 1 0.0779 0.8852 0.976 0.000 0.008 0.008 0.000 0.008
#> GSM39171 1 0.0405 0.8902 0.988 0.000 0.008 0.004 0.000 0.000
#> GSM39172 1 0.1478 0.8699 0.944 0.000 0.032 0.004 0.000 0.020
#> GSM39173 1 0.5025 0.6419 0.748 0.028 0.132 0.044 0.020 0.028
#> GSM39174 1 0.0000 0.8913 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM39175 1 0.0146 0.8903 0.996 0.000 0.004 0.000 0.000 0.000
#> GSM39176 1 0.0146 0.8919 0.996 0.004 0.000 0.000 0.000 0.000
#> GSM39177 1 0.3134 0.6670 0.784 0.000 0.208 0.000 0.004 0.004
#> GSM39178 1 0.0405 0.8902 0.988 0.000 0.008 0.004 0.000 0.000
#> GSM39179 3 0.3433 0.6100 0.200 0.000 0.780 0.008 0.004 0.008
#> GSM39180 1 0.7901 0.0208 0.508 0.084 0.096 0.180 0.024 0.108
#> GSM39181 1 0.1294 0.8757 0.956 0.000 0.008 0.008 0.004 0.024
#> GSM39182 1 0.1975 0.8714 0.928 0.028 0.020 0.012 0.000 0.012
#> GSM39183 1 0.1294 0.8757 0.956 0.000 0.008 0.008 0.004 0.024
#> GSM39184 1 0.0000 0.8913 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM39185 1 0.3129 0.8113 0.872 0.008 0.032 0.028 0.008 0.052
#> GSM39186 1 0.0260 0.8914 0.992 0.008 0.000 0.000 0.000 0.000
#> GSM39187 1 0.0458 0.8896 0.984 0.016 0.000 0.000 0.000 0.000
#> GSM39116 2 0.2826 0.6601 0.128 0.844 0.000 0.028 0.000 0.000
#> GSM39117 4 0.3823 1.0000 0.000 0.436 0.000 0.564 0.000 0.000
#> GSM39118 2 0.2672 0.4926 0.052 0.868 0.000 0.080 0.000 0.000
#> GSM39119 2 0.4010 -0.6323 0.008 0.584 0.000 0.408 0.000 0.000
#> GSM39120 1 0.3993 0.2690 0.592 0.400 0.000 0.008 0.000 0.000
#> GSM39121 2 0.3245 0.5972 0.228 0.764 0.000 0.008 0.000 0.000
#> GSM39122 2 0.3133 0.6163 0.212 0.780 0.000 0.008 0.000 0.000
#> GSM39123 4 0.3823 1.0000 0.000 0.436 0.000 0.564 0.000 0.000
#> GSM39124 2 0.2416 0.6767 0.156 0.844 0.000 0.000 0.000 0.000
#> GSM39125 1 0.4095 -0.0446 0.512 0.480 0.000 0.008 0.000 0.000
#> GSM39126 2 0.3468 0.5435 0.264 0.728 0.000 0.008 0.000 0.000
#> GSM39127 2 0.2300 0.6778 0.144 0.856 0.000 0.000 0.000 0.000
#> GSM39128 2 0.2454 0.6743 0.160 0.840 0.000 0.000 0.000 0.000
#> GSM39129 2 0.4567 0.1241 0.000 0.644 0.016 0.316 0.008 0.016
#> GSM39130 4 0.3823 1.0000 0.000 0.436 0.000 0.564 0.000 0.000
#> GSM39131 2 0.2362 0.6747 0.136 0.860 0.000 0.004 0.000 0.000
#> GSM39132 2 0.2278 0.6691 0.128 0.868 0.000 0.004 0.000 0.000
#> GSM39133 4 0.3823 1.0000 0.000 0.436 0.000 0.564 0.000 0.000
#> GSM39134 2 0.4171 -0.2603 0.000 0.656 0.012 0.320 0.000 0.012
#> GSM39135 2 0.2581 0.6656 0.128 0.856 0.000 0.016 0.000 0.000
#> GSM39136 2 0.2979 0.6323 0.116 0.840 0.000 0.044 0.000 0.000
#> GSM39137 2 0.2416 0.6767 0.156 0.844 0.000 0.000 0.000 0.000
#> GSM39138 2 0.4222 0.1645 0.000 0.676 0.016 0.292 0.000 0.016
#> GSM39139 2 0.4184 0.1718 0.000 0.684 0.016 0.284 0.000 0.016
#> GSM39140 1 0.3298 0.6398 0.756 0.236 0.000 0.008 0.000 0.000
#> GSM39141 1 0.1398 0.8682 0.940 0.052 0.000 0.008 0.000 0.000
#> GSM39142 1 0.1398 0.8682 0.940 0.052 0.000 0.008 0.000 0.000
#> GSM39143 1 0.1398 0.8682 0.940 0.052 0.000 0.008 0.000 0.000
#> GSM39144 2 0.4240 0.1642 0.000 0.672 0.016 0.296 0.000 0.016
#> GSM39145 2 0.3384 0.4977 0.068 0.812 0.000 0.120 0.000 0.000
#> GSM39146 2 0.2949 0.6671 0.140 0.832 0.000 0.028 0.000 0.000
#> GSM39147 2 0.2416 0.6767 0.156 0.844 0.000 0.000 0.000 0.000
#> GSM39188 5 0.0777 0.0000 0.004 0.000 0.024 0.000 0.972 0.000
#> GSM39189 1 0.2094 0.8464 0.912 0.000 0.060 0.004 0.004 0.020
#> GSM39190 6 0.3833 0.0000 0.120 0.000 0.004 0.000 0.092 0.784
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) other(p) protocol(p) k
#> SD:hclust 85 0.122 5.85e-12 6.19e-13 2
#> SD:hclust 84 0.116 2.06e-12 9.05e-13 3
#> SD:hclust 77 0.154 8.24e-12 1.35e-11 4
#> SD:hclust 75 0.279 1.23e-09 1.74e-09 5
#> SD:hclust 69 0.625 2.62e-07 2.97e-10 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 8353 rows and 87 columns.
#> Top rows (835, 1670, 2506, 3341, 4176) 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 0.813 0.887 0.949 0.4460 0.530 0.530
#> 3 3 0.714 0.857 0.914 0.3067 0.871 0.762
#> 4 4 0.650 0.650 0.835 0.1205 0.942 0.867
#> 5 5 0.641 0.638 0.801 0.0806 0.893 0.734
#> 6 6 0.650 0.584 0.757 0.0636 0.952 0.850
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
#> GSM39104 1 0.0000 0.986 1.000 0.000
#> GSM39105 1 0.0000 0.986 1.000 0.000
#> GSM39106 1 0.0000 0.986 1.000 0.000
#> GSM39107 1 0.0000 0.986 1.000 0.000
#> GSM39108 1 0.0000 0.986 1.000 0.000
#> GSM39109 1 0.4161 0.894 0.916 0.084
#> GSM39110 1 0.0000 0.986 1.000 0.000
#> GSM39111 1 0.0000 0.986 1.000 0.000
#> GSM39112 1 0.0000 0.986 1.000 0.000
#> GSM39113 1 0.0000 0.986 1.000 0.000
#> GSM39114 2 0.8081 0.664 0.248 0.752
#> GSM39115 1 0.0000 0.986 1.000 0.000
#> GSM39148 1 0.0000 0.986 1.000 0.000
#> GSM39149 2 0.9933 0.325 0.452 0.548
#> GSM39150 1 0.0000 0.986 1.000 0.000
#> GSM39151 2 0.9896 0.356 0.440 0.560
#> GSM39152 1 0.0000 0.986 1.000 0.000
#> GSM39153 1 0.0000 0.986 1.000 0.000
#> GSM39154 1 0.0000 0.986 1.000 0.000
#> GSM39155 1 0.0000 0.986 1.000 0.000
#> GSM39156 1 0.0000 0.986 1.000 0.000
#> GSM39157 1 0.0000 0.986 1.000 0.000
#> GSM39158 1 0.0000 0.986 1.000 0.000
#> GSM39159 1 0.0000 0.986 1.000 0.000
#> GSM39160 1 0.0000 0.986 1.000 0.000
#> GSM39161 1 0.0000 0.986 1.000 0.000
#> GSM39162 1 0.0000 0.986 1.000 0.000
#> GSM39163 1 0.0000 0.986 1.000 0.000
#> GSM39164 1 0.0000 0.986 1.000 0.000
#> GSM39165 1 0.0000 0.986 1.000 0.000
#> GSM39166 1 0.0000 0.986 1.000 0.000
#> GSM39167 1 0.0000 0.986 1.000 0.000
#> GSM39168 1 0.0000 0.986 1.000 0.000
#> GSM39169 1 0.0000 0.986 1.000 0.000
#> GSM39170 1 0.0000 0.986 1.000 0.000
#> GSM39171 1 0.0000 0.986 1.000 0.000
#> GSM39172 2 0.9933 0.325 0.452 0.548
#> GSM39173 2 0.9608 0.468 0.384 0.616
#> GSM39174 1 0.0000 0.986 1.000 0.000
#> GSM39175 1 0.0000 0.986 1.000 0.000
#> GSM39176 1 0.0000 0.986 1.000 0.000
#> GSM39177 1 0.4815 0.867 0.896 0.104
#> GSM39178 1 0.0000 0.986 1.000 0.000
#> GSM39179 2 0.9954 0.302 0.460 0.540
#> GSM39180 2 0.0376 0.868 0.004 0.996
#> GSM39181 1 0.0000 0.986 1.000 0.000
#> GSM39182 1 0.5294 0.845 0.880 0.120
#> GSM39183 1 0.0000 0.986 1.000 0.000
#> GSM39184 1 0.0000 0.986 1.000 0.000
#> GSM39185 1 0.2948 0.933 0.948 0.052
#> GSM39186 1 0.0000 0.986 1.000 0.000
#> GSM39187 1 0.0000 0.986 1.000 0.000
#> GSM39116 2 0.0000 0.870 0.000 1.000
#> GSM39117 2 0.0000 0.870 0.000 1.000
#> GSM39118 2 0.0000 0.870 0.000 1.000
#> GSM39119 2 0.0000 0.870 0.000 1.000
#> GSM39120 1 0.0000 0.986 1.000 0.000
#> GSM39121 1 0.1843 0.959 0.972 0.028
#> GSM39122 1 0.2423 0.948 0.960 0.040
#> GSM39123 2 0.0000 0.870 0.000 1.000
#> GSM39124 2 0.5737 0.778 0.136 0.864
#> GSM39125 1 0.0000 0.986 1.000 0.000
#> GSM39126 1 0.2603 0.943 0.956 0.044
#> GSM39127 2 0.0000 0.870 0.000 1.000
#> GSM39128 2 0.3114 0.838 0.056 0.944
#> GSM39129 2 0.0000 0.870 0.000 1.000
#> GSM39130 2 0.0000 0.870 0.000 1.000
#> GSM39131 2 0.0000 0.870 0.000 1.000
#> GSM39132 2 0.0000 0.870 0.000 1.000
#> GSM39133 2 0.0000 0.870 0.000 1.000
#> GSM39134 2 0.0000 0.870 0.000 1.000
#> GSM39135 2 0.0000 0.870 0.000 1.000
#> GSM39136 2 0.0000 0.870 0.000 1.000
#> GSM39137 2 0.8499 0.630 0.276 0.724
#> GSM39138 2 0.0000 0.870 0.000 1.000
#> GSM39139 2 0.0000 0.870 0.000 1.000
#> GSM39140 1 0.0000 0.986 1.000 0.000
#> GSM39141 1 0.0000 0.986 1.000 0.000
#> GSM39142 1 0.0000 0.986 1.000 0.000
#> GSM39143 1 0.0000 0.986 1.000 0.000
#> GSM39144 2 0.0000 0.870 0.000 1.000
#> GSM39145 2 0.0000 0.870 0.000 1.000
#> GSM39146 2 0.0000 0.870 0.000 1.000
#> GSM39147 2 0.0000 0.870 0.000 1.000
#> GSM39188 2 0.9850 0.383 0.428 0.572
#> GSM39189 1 0.6973 0.735 0.812 0.188
#> GSM39190 2 0.9896 0.356 0.440 0.560
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM39104 1 0.0000 0.929 1.000 0.000 0.000
#> GSM39105 1 0.0000 0.929 1.000 0.000 0.000
#> GSM39106 1 0.1129 0.920 0.976 0.004 0.020
#> GSM39107 1 0.3805 0.844 0.884 0.092 0.024
#> GSM39108 1 0.0829 0.924 0.984 0.004 0.012
#> GSM39109 1 0.4964 0.802 0.836 0.116 0.048
#> GSM39110 1 0.0829 0.924 0.984 0.004 0.012
#> GSM39111 1 0.0000 0.929 1.000 0.000 0.000
#> GSM39112 1 0.3370 0.863 0.904 0.072 0.024
#> GSM39113 1 0.3805 0.844 0.884 0.092 0.024
#> GSM39114 2 0.2689 0.872 0.032 0.932 0.036
#> GSM39115 1 0.0000 0.929 1.000 0.000 0.000
#> GSM39148 1 0.0237 0.928 0.996 0.000 0.004
#> GSM39149 3 0.3989 0.905 0.124 0.012 0.864
#> GSM39150 1 0.0237 0.927 0.996 0.000 0.004
#> GSM39151 3 0.3921 0.900 0.112 0.016 0.872
#> GSM39152 3 0.5098 0.824 0.248 0.000 0.752
#> GSM39153 1 0.0000 0.929 1.000 0.000 0.000
#> GSM39154 1 0.0000 0.929 1.000 0.000 0.000
#> GSM39155 1 0.0000 0.929 1.000 0.000 0.000
#> GSM39156 1 0.1129 0.920 0.976 0.004 0.020
#> GSM39157 1 0.0000 0.929 1.000 0.000 0.000
#> GSM39158 1 0.0000 0.929 1.000 0.000 0.000
#> GSM39159 1 0.6286 -0.168 0.536 0.000 0.464
#> GSM39160 1 0.0892 0.915 0.980 0.000 0.020
#> GSM39161 3 0.5785 0.704 0.332 0.000 0.668
#> GSM39162 1 0.0424 0.927 0.992 0.000 0.008
#> GSM39163 1 0.0000 0.929 1.000 0.000 0.000
#> GSM39164 1 0.0000 0.929 1.000 0.000 0.000
#> GSM39165 1 0.4002 0.747 0.840 0.000 0.160
#> GSM39166 1 0.0237 0.927 0.996 0.000 0.004
#> GSM39167 1 0.0000 0.929 1.000 0.000 0.000
#> GSM39168 1 0.0237 0.928 0.996 0.000 0.004
#> GSM39169 1 0.0000 0.929 1.000 0.000 0.000
#> GSM39170 1 0.0000 0.929 1.000 0.000 0.000
#> GSM39171 1 0.0000 0.929 1.000 0.000 0.000
#> GSM39172 3 0.3896 0.905 0.128 0.008 0.864
#> GSM39173 3 0.4479 0.878 0.096 0.044 0.860
#> GSM39174 1 0.0000 0.929 1.000 0.000 0.000
#> GSM39175 1 0.0000 0.929 1.000 0.000 0.000
#> GSM39176 1 0.0000 0.929 1.000 0.000 0.000
#> GSM39177 3 0.4465 0.880 0.176 0.004 0.820
#> GSM39178 1 0.5882 0.307 0.652 0.000 0.348
#> GSM39179 3 0.3896 0.905 0.128 0.008 0.864
#> GSM39180 3 0.3141 0.770 0.020 0.068 0.912
#> GSM39181 1 0.0237 0.927 0.996 0.000 0.004
#> GSM39182 1 0.6140 0.119 0.596 0.000 0.404
#> GSM39183 1 0.0237 0.927 0.996 0.000 0.004
#> GSM39184 1 0.0000 0.929 1.000 0.000 0.000
#> GSM39185 3 0.5706 0.724 0.320 0.000 0.680
#> GSM39186 1 0.0000 0.929 1.000 0.000 0.000
#> GSM39187 1 0.0000 0.929 1.000 0.000 0.000
#> GSM39116 2 0.0592 0.902 0.000 0.988 0.012
#> GSM39117 2 0.5254 0.788 0.000 0.736 0.264
#> GSM39118 2 0.2959 0.889 0.000 0.900 0.100
#> GSM39119 2 0.3551 0.877 0.000 0.868 0.132
#> GSM39120 1 0.1620 0.913 0.964 0.012 0.024
#> GSM39121 1 0.5402 0.737 0.792 0.180 0.028
#> GSM39122 1 0.5610 0.716 0.776 0.196 0.028
#> GSM39123 2 0.5254 0.788 0.000 0.736 0.264
#> GSM39124 2 0.1411 0.898 0.000 0.964 0.036
#> GSM39125 1 0.1774 0.910 0.960 0.016 0.024
#> GSM39126 1 0.5728 0.712 0.772 0.196 0.032
#> GSM39127 2 0.1411 0.898 0.000 0.964 0.036
#> GSM39128 2 0.1411 0.898 0.000 0.964 0.036
#> GSM39129 2 0.3482 0.881 0.000 0.872 0.128
#> GSM39130 2 0.5254 0.788 0.000 0.736 0.264
#> GSM39131 2 0.1411 0.898 0.000 0.964 0.036
#> GSM39132 2 0.1411 0.898 0.000 0.964 0.036
#> GSM39133 2 0.4702 0.835 0.000 0.788 0.212
#> GSM39134 2 0.3192 0.885 0.000 0.888 0.112
#> GSM39135 2 0.0592 0.902 0.000 0.988 0.012
#> GSM39136 2 0.0424 0.901 0.000 0.992 0.008
#> GSM39137 2 0.5526 0.667 0.172 0.792 0.036
#> GSM39138 2 0.3482 0.881 0.000 0.872 0.128
#> GSM39139 2 0.2165 0.897 0.000 0.936 0.064
#> GSM39140 1 0.1453 0.916 0.968 0.008 0.024
#> GSM39141 1 0.1453 0.916 0.968 0.008 0.024
#> GSM39142 1 0.1267 0.918 0.972 0.004 0.024
#> GSM39143 1 0.1453 0.916 0.968 0.008 0.024
#> GSM39144 2 0.3482 0.881 0.000 0.872 0.128
#> GSM39145 2 0.0892 0.901 0.000 0.980 0.020
#> GSM39146 2 0.1411 0.898 0.000 0.964 0.036
#> GSM39147 2 0.1411 0.898 0.000 0.964 0.036
#> GSM39188 3 0.3690 0.888 0.100 0.016 0.884
#> GSM39189 3 0.3965 0.904 0.132 0.008 0.860
#> GSM39190 3 0.3921 0.900 0.112 0.016 0.872
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM39104 1 0.2546 0.8644 0.900 0.000 0.008 0.092
#> GSM39105 1 0.1356 0.8773 0.960 0.000 0.008 0.032
#> GSM39106 1 0.2852 0.8585 0.904 0.024 0.008 0.064
#> GSM39107 1 0.5503 0.6736 0.728 0.204 0.008 0.060
#> GSM39108 1 0.2380 0.8657 0.920 0.008 0.008 0.064
#> GSM39109 1 0.7013 0.5567 0.616 0.256 0.024 0.104
#> GSM39110 1 0.2510 0.8639 0.916 0.012 0.008 0.064
#> GSM39111 1 0.2271 0.8684 0.916 0.000 0.008 0.076
#> GSM39112 1 0.4712 0.7645 0.800 0.132 0.008 0.060
#> GSM39113 1 0.5539 0.6672 0.724 0.208 0.008 0.060
#> GSM39114 2 0.3453 0.4768 0.080 0.868 0.000 0.052
#> GSM39115 1 0.1256 0.8785 0.964 0.000 0.008 0.028
#> GSM39148 1 0.0000 0.8833 1.000 0.000 0.000 0.000
#> GSM39149 3 0.1722 0.8621 0.008 0.000 0.944 0.048
#> GSM39150 1 0.3978 0.7886 0.796 0.000 0.012 0.192
#> GSM39151 3 0.1890 0.8595 0.008 0.000 0.936 0.056
#> GSM39152 3 0.2805 0.8328 0.012 0.000 0.888 0.100
#> GSM39153 1 0.0188 0.8826 0.996 0.000 0.000 0.004
#> GSM39154 1 0.0000 0.8833 1.000 0.000 0.000 0.000
#> GSM39155 1 0.0000 0.8833 1.000 0.000 0.000 0.000
#> GSM39156 1 0.0804 0.8810 0.980 0.012 0.000 0.008
#> GSM39157 1 0.0000 0.8833 1.000 0.000 0.000 0.000
#> GSM39158 1 0.3172 0.8017 0.840 0.000 0.000 0.160
#> GSM39159 1 0.7602 -0.1028 0.420 0.000 0.380 0.200
#> GSM39160 1 0.4801 0.7595 0.764 0.000 0.048 0.188
#> GSM39161 3 0.7289 0.4856 0.268 0.000 0.532 0.200
#> GSM39162 1 0.0000 0.8833 1.000 0.000 0.000 0.000
#> GSM39163 1 0.0000 0.8833 1.000 0.000 0.000 0.000
#> GSM39164 1 0.0000 0.8833 1.000 0.000 0.000 0.000
#> GSM39165 1 0.4700 0.7577 0.792 0.000 0.124 0.084
#> GSM39166 1 0.3751 0.7766 0.800 0.000 0.004 0.196
#> GSM39167 1 0.0000 0.8833 1.000 0.000 0.000 0.000
#> GSM39168 1 0.0000 0.8833 1.000 0.000 0.000 0.000
#> GSM39169 1 0.0000 0.8833 1.000 0.000 0.000 0.000
#> GSM39170 1 0.3074 0.8075 0.848 0.000 0.000 0.152
#> GSM39171 1 0.1867 0.8647 0.928 0.000 0.000 0.072
#> GSM39172 3 0.1356 0.8677 0.008 0.000 0.960 0.032
#> GSM39173 3 0.1339 0.8689 0.008 0.004 0.964 0.024
#> GSM39174 1 0.0000 0.8833 1.000 0.000 0.000 0.000
#> GSM39175 1 0.0188 0.8826 0.996 0.000 0.000 0.004
#> GSM39176 1 0.0000 0.8833 1.000 0.000 0.000 0.000
#> GSM39177 3 0.1284 0.8706 0.012 0.000 0.964 0.024
#> GSM39178 1 0.7304 0.3519 0.532 0.000 0.260 0.208
#> GSM39179 3 0.1807 0.8609 0.008 0.000 0.940 0.052
#> GSM39180 3 0.1792 0.8620 0.000 0.000 0.932 0.068
#> GSM39181 1 0.3626 0.7795 0.812 0.000 0.004 0.184
#> GSM39182 1 0.7496 0.2302 0.512 0.008 0.320 0.160
#> GSM39183 1 0.3668 0.7785 0.808 0.000 0.004 0.188
#> GSM39184 1 0.0000 0.8833 1.000 0.000 0.000 0.000
#> GSM39185 3 0.7269 0.4907 0.264 0.000 0.536 0.200
#> GSM39186 1 0.1209 0.8791 0.964 0.000 0.004 0.032
#> GSM39187 1 0.0000 0.8833 1.000 0.000 0.000 0.000
#> GSM39116 2 0.0921 0.5495 0.000 0.972 0.000 0.028
#> GSM39117 4 0.5440 0.9949 0.000 0.384 0.020 0.596
#> GSM39118 2 0.4222 0.0548 0.000 0.728 0.000 0.272
#> GSM39119 2 0.5165 -0.6441 0.000 0.512 0.004 0.484
#> GSM39120 1 0.2413 0.8609 0.924 0.036 0.004 0.036
#> GSM39121 1 0.5861 0.0356 0.488 0.480 0.000 0.032
#> GSM39122 2 0.5938 -0.0649 0.476 0.488 0.000 0.036
#> GSM39123 4 0.5440 0.9949 0.000 0.384 0.020 0.596
#> GSM39124 2 0.1677 0.5440 0.040 0.948 0.000 0.012
#> GSM39125 1 0.1833 0.8683 0.944 0.032 0.000 0.024
#> GSM39126 2 0.5861 -0.0716 0.480 0.488 0.000 0.032
#> GSM39127 2 0.0000 0.5642 0.000 1.000 0.000 0.000
#> GSM39128 2 0.0895 0.5595 0.020 0.976 0.000 0.004
#> GSM39129 2 0.5163 -0.4628 0.000 0.516 0.004 0.480
#> GSM39130 4 0.5440 0.9949 0.000 0.384 0.020 0.596
#> GSM39131 2 0.0188 0.5638 0.000 0.996 0.000 0.004
#> GSM39132 2 0.0000 0.5642 0.000 1.000 0.000 0.000
#> GSM39133 4 0.5364 0.9846 0.000 0.392 0.016 0.592
#> GSM39134 2 0.4948 -0.4471 0.000 0.560 0.000 0.440
#> GSM39135 2 0.0921 0.5495 0.000 0.972 0.000 0.028
#> GSM39136 2 0.1118 0.5427 0.000 0.964 0.000 0.036
#> GSM39137 2 0.3647 0.4250 0.152 0.832 0.000 0.016
#> GSM39138 2 0.5163 -0.4628 0.000 0.516 0.004 0.480
#> GSM39139 2 0.4193 0.2262 0.000 0.732 0.000 0.268
#> GSM39140 1 0.1151 0.8774 0.968 0.008 0.000 0.024
#> GSM39141 1 0.0672 0.8809 0.984 0.008 0.000 0.008
#> GSM39142 1 0.0524 0.8818 0.988 0.004 0.000 0.008
#> GSM39143 1 0.0672 0.8809 0.984 0.008 0.000 0.008
#> GSM39144 2 0.5163 -0.4628 0.000 0.516 0.004 0.480
#> GSM39145 2 0.2469 0.5155 0.000 0.892 0.000 0.108
#> GSM39146 2 0.0000 0.5642 0.000 1.000 0.000 0.000
#> GSM39147 2 0.0707 0.5631 0.000 0.980 0.000 0.020
#> GSM39188 3 0.1970 0.8586 0.008 0.000 0.932 0.060
#> GSM39189 3 0.1824 0.8616 0.004 0.000 0.936 0.060
#> GSM39190 3 0.0672 0.8695 0.008 0.000 0.984 0.008
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM39104 1 0.4210 0.6319 0.756 0.000 0.004 0.036 0.204
#> GSM39105 1 0.2470 0.7307 0.884 0.000 0.000 0.012 0.104
#> GSM39106 1 0.4530 0.6389 0.768 0.036 0.000 0.032 0.164
#> GSM39107 1 0.6954 0.1593 0.468 0.348 0.000 0.032 0.152
#> GSM39108 1 0.4004 0.6635 0.792 0.012 0.000 0.032 0.164
#> GSM39109 2 0.7761 0.0832 0.292 0.404 0.008 0.044 0.252
#> GSM39110 1 0.4430 0.6466 0.772 0.020 0.004 0.032 0.172
#> GSM39111 1 0.4202 0.6525 0.776 0.004 0.008 0.032 0.180
#> GSM39112 1 0.6691 0.2718 0.544 0.280 0.000 0.032 0.144
#> GSM39113 1 0.6974 0.1220 0.456 0.360 0.000 0.032 0.152
#> GSM39114 2 0.3531 0.6395 0.036 0.852 0.000 0.032 0.080
#> GSM39115 1 0.1670 0.7605 0.936 0.000 0.000 0.012 0.052
#> GSM39148 1 0.0000 0.7826 1.000 0.000 0.000 0.000 0.000
#> GSM39149 3 0.1018 0.8372 0.000 0.000 0.968 0.016 0.016
#> GSM39150 1 0.4491 0.2733 0.648 0.000 0.004 0.012 0.336
#> GSM39151 3 0.1750 0.8270 0.000 0.000 0.936 0.036 0.028
#> GSM39152 3 0.4156 0.7321 0.008 0.000 0.700 0.004 0.288
#> GSM39153 1 0.0162 0.7808 0.996 0.000 0.000 0.000 0.004
#> GSM39154 1 0.0162 0.7808 0.996 0.000 0.000 0.000 0.004
#> GSM39155 1 0.0000 0.7826 1.000 0.000 0.000 0.000 0.000
#> GSM39156 1 0.1205 0.7705 0.956 0.004 0.000 0.000 0.040
#> GSM39157 1 0.0000 0.7826 1.000 0.000 0.000 0.000 0.000
#> GSM39158 1 0.3715 0.3803 0.736 0.000 0.000 0.004 0.260
#> GSM39159 5 0.5991 0.8405 0.352 0.000 0.108 0.004 0.536
#> GSM39160 1 0.4620 0.0931 0.612 0.000 0.004 0.012 0.372
#> GSM39161 5 0.6383 0.7990 0.264 0.000 0.196 0.004 0.536
#> GSM39162 1 0.0000 0.7826 1.000 0.000 0.000 0.000 0.000
#> GSM39163 1 0.0000 0.7826 1.000 0.000 0.000 0.000 0.000
#> GSM39164 1 0.0000 0.7826 1.000 0.000 0.000 0.000 0.000
#> GSM39165 1 0.3936 0.5488 0.800 0.000 0.052 0.004 0.144
#> GSM39166 1 0.4251 0.0622 0.624 0.000 0.000 0.004 0.372
#> GSM39167 1 0.0000 0.7826 1.000 0.000 0.000 0.000 0.000
#> GSM39168 1 0.0000 0.7826 1.000 0.000 0.000 0.000 0.000
#> GSM39169 1 0.0000 0.7826 1.000 0.000 0.000 0.000 0.000
#> GSM39170 1 0.3814 0.3570 0.720 0.000 0.000 0.004 0.276
#> GSM39171 1 0.2127 0.7165 0.892 0.000 0.000 0.000 0.108
#> GSM39172 3 0.4119 0.8320 0.000 0.000 0.752 0.036 0.212
#> GSM39173 3 0.3988 0.8422 0.000 0.000 0.768 0.036 0.196
#> GSM39174 1 0.0000 0.7826 1.000 0.000 0.000 0.000 0.000
#> GSM39175 1 0.0162 0.7808 0.996 0.000 0.000 0.000 0.004
#> GSM39176 1 0.0000 0.7826 1.000 0.000 0.000 0.000 0.000
#> GSM39177 3 0.1704 0.8541 0.000 0.000 0.928 0.004 0.068
#> GSM39178 5 0.5606 0.7867 0.360 0.000 0.072 0.004 0.564
#> GSM39179 3 0.0912 0.8398 0.000 0.000 0.972 0.016 0.012
#> GSM39180 3 0.4495 0.8258 0.000 0.000 0.712 0.044 0.244
#> GSM39181 1 0.4182 0.0981 0.644 0.000 0.000 0.004 0.352
#> GSM39182 5 0.6944 0.7743 0.376 0.004 0.168 0.016 0.436
#> GSM39183 1 0.4264 0.0424 0.620 0.000 0.000 0.004 0.376
#> GSM39184 1 0.0162 0.7808 0.996 0.000 0.000 0.000 0.004
#> GSM39185 5 0.6313 0.8170 0.272 0.000 0.180 0.004 0.544
#> GSM39186 1 0.0794 0.7748 0.972 0.000 0.000 0.000 0.028
#> GSM39187 1 0.0000 0.7826 1.000 0.000 0.000 0.000 0.000
#> GSM39116 2 0.1041 0.6829 0.000 0.964 0.000 0.032 0.004
#> GSM39117 4 0.6263 0.6654 0.000 0.192 0.000 0.532 0.276
#> GSM39118 2 0.4302 -0.4574 0.000 0.520 0.000 0.480 0.000
#> GSM39119 4 0.4485 0.7069 0.000 0.292 0.000 0.680 0.028
#> GSM39120 1 0.4737 0.6155 0.768 0.096 0.000 0.024 0.112
#> GSM39121 2 0.5331 0.4928 0.256 0.668 0.000 0.020 0.056
#> GSM39122 2 0.5402 0.4954 0.248 0.668 0.000 0.020 0.064
#> GSM39123 4 0.6263 0.6654 0.000 0.192 0.000 0.532 0.276
#> GSM39124 2 0.0609 0.7084 0.020 0.980 0.000 0.000 0.000
#> GSM39125 1 0.3546 0.6931 0.848 0.060 0.000 0.016 0.076
#> GSM39126 2 0.5466 0.4976 0.240 0.668 0.000 0.020 0.072
#> GSM39127 2 0.0324 0.7064 0.000 0.992 0.000 0.004 0.004
#> GSM39128 2 0.0671 0.7098 0.016 0.980 0.000 0.004 0.000
#> GSM39129 4 0.3906 0.6999 0.000 0.292 0.004 0.704 0.000
#> GSM39130 4 0.6263 0.6654 0.000 0.192 0.000 0.532 0.276
#> GSM39131 2 0.0324 0.7079 0.000 0.992 0.000 0.004 0.004
#> GSM39132 2 0.0324 0.7064 0.000 0.992 0.000 0.004 0.004
#> GSM39133 4 0.6263 0.6654 0.000 0.192 0.000 0.532 0.276
#> GSM39134 4 0.3932 0.6889 0.000 0.328 0.000 0.672 0.000
#> GSM39135 2 0.1041 0.6829 0.000 0.964 0.000 0.032 0.004
#> GSM39136 2 0.1205 0.6733 0.000 0.956 0.000 0.040 0.004
#> GSM39137 2 0.1768 0.6792 0.072 0.924 0.000 0.000 0.004
#> GSM39138 4 0.3906 0.6999 0.000 0.292 0.004 0.704 0.000
#> GSM39139 4 0.4297 0.3878 0.000 0.472 0.000 0.528 0.000
#> GSM39140 1 0.1280 0.7720 0.960 0.008 0.000 0.008 0.024
#> GSM39141 1 0.0771 0.7771 0.976 0.004 0.000 0.000 0.020
#> GSM39142 1 0.0404 0.7803 0.988 0.000 0.000 0.000 0.012
#> GSM39143 1 0.0771 0.7771 0.976 0.004 0.000 0.000 0.020
#> GSM39144 4 0.3906 0.6999 0.000 0.292 0.004 0.704 0.000
#> GSM39145 2 0.3949 0.1476 0.000 0.668 0.000 0.332 0.000
#> GSM39146 2 0.0324 0.7064 0.000 0.992 0.000 0.004 0.004
#> GSM39147 2 0.0510 0.7055 0.000 0.984 0.000 0.016 0.000
#> GSM39188 3 0.2304 0.8236 0.000 0.000 0.908 0.048 0.044
#> GSM39189 3 0.4276 0.8104 0.000 0.000 0.724 0.032 0.244
#> GSM39190 3 0.3649 0.8535 0.000 0.000 0.808 0.040 0.152
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM39104 1 0.5905 0.3179 0.492 0.004 0.000 0.296 0.208 0.000
#> GSM39105 1 0.4640 0.5398 0.680 0.000 0.000 0.212 0.108 0.000
#> GSM39106 1 0.6096 0.3936 0.528 0.028 0.000 0.276 0.168 0.000
#> GSM39107 2 0.7287 0.1388 0.248 0.388 0.000 0.252 0.112 0.000
#> GSM39108 1 0.5877 0.4168 0.548 0.020 0.000 0.276 0.156 0.000
#> GSM39109 4 0.7518 -0.2197 0.128 0.324 0.004 0.328 0.216 0.000
#> GSM39110 1 0.5998 0.3928 0.528 0.020 0.000 0.280 0.172 0.000
#> GSM39111 1 0.5780 0.3870 0.532 0.008 0.000 0.288 0.172 0.000
#> GSM39112 1 0.7324 0.1315 0.380 0.252 0.000 0.252 0.116 0.000
#> GSM39113 2 0.7305 0.1415 0.228 0.392 0.000 0.260 0.120 0.000
#> GSM39114 2 0.3191 0.6649 0.016 0.832 0.000 0.128 0.024 0.000
#> GSM39115 1 0.2812 0.6800 0.856 0.000 0.000 0.096 0.048 0.000
#> GSM39148 1 0.0146 0.7398 0.996 0.000 0.000 0.000 0.004 0.000
#> GSM39149 3 0.0717 0.7071 0.000 0.000 0.976 0.000 0.016 0.008
#> GSM39150 1 0.6053 -0.1085 0.440 0.000 0.008 0.192 0.360 0.000
#> GSM39151 3 0.1710 0.6894 0.000 0.000 0.936 0.016 0.020 0.028
#> GSM39152 3 0.5154 0.4823 0.000 0.000 0.524 0.076 0.396 0.004
#> GSM39153 1 0.0000 0.7402 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM39154 1 0.0146 0.7399 0.996 0.000 0.000 0.000 0.004 0.000
#> GSM39155 1 0.0260 0.7393 0.992 0.000 0.000 0.000 0.008 0.000
#> GSM39156 1 0.2444 0.7028 0.896 0.016 0.000 0.052 0.036 0.000
#> GSM39157 1 0.0146 0.7399 0.996 0.000 0.000 0.000 0.004 0.000
#> GSM39158 1 0.3482 0.3526 0.684 0.000 0.000 0.000 0.316 0.000
#> GSM39159 5 0.3719 0.7808 0.248 0.000 0.024 0.000 0.728 0.000
#> GSM39160 1 0.6038 -0.2007 0.420 0.000 0.008 0.184 0.388 0.000
#> GSM39161 5 0.3861 0.7908 0.184 0.000 0.060 0.000 0.756 0.000
#> GSM39162 1 0.0260 0.7395 0.992 0.000 0.000 0.000 0.008 0.000
#> GSM39163 1 0.0146 0.7399 0.996 0.000 0.000 0.000 0.004 0.000
#> GSM39164 1 0.0000 0.7402 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM39165 1 0.3767 0.5588 0.780 0.000 0.028 0.020 0.172 0.000
#> GSM39166 1 0.4305 -0.0259 0.544 0.000 0.000 0.020 0.436 0.000
#> GSM39167 1 0.0146 0.7399 0.996 0.000 0.000 0.000 0.004 0.000
#> GSM39168 1 0.0260 0.7395 0.992 0.000 0.000 0.000 0.008 0.000
#> GSM39169 1 0.0260 0.7393 0.992 0.000 0.000 0.000 0.008 0.000
#> GSM39170 1 0.3619 0.3559 0.680 0.000 0.000 0.004 0.316 0.000
#> GSM39171 1 0.3566 0.6034 0.788 0.000 0.000 0.056 0.156 0.000
#> GSM39172 3 0.5553 0.6512 0.000 0.000 0.524 0.104 0.360 0.012
#> GSM39173 3 0.5536 0.6812 0.000 0.000 0.540 0.080 0.356 0.024
#> GSM39174 1 0.0146 0.7399 0.996 0.000 0.000 0.000 0.004 0.000
#> GSM39175 1 0.0260 0.7393 0.992 0.000 0.000 0.000 0.008 0.000
#> GSM39176 1 0.0146 0.7399 0.996 0.000 0.000 0.000 0.004 0.000
#> GSM39177 3 0.2872 0.7267 0.000 0.000 0.832 0.012 0.152 0.004
#> GSM39178 5 0.5369 0.7363 0.196 0.000 0.024 0.120 0.656 0.004
#> GSM39179 3 0.1152 0.7139 0.000 0.000 0.952 0.000 0.044 0.004
#> GSM39180 3 0.6053 0.6185 0.000 0.000 0.440 0.100 0.420 0.040
#> GSM39181 1 0.3950 0.0209 0.564 0.000 0.000 0.004 0.432 0.000
#> GSM39182 5 0.6417 0.6833 0.220 0.000 0.072 0.132 0.568 0.008
#> GSM39183 1 0.4238 -0.0451 0.540 0.000 0.000 0.016 0.444 0.000
#> GSM39184 1 0.0260 0.7393 0.992 0.000 0.000 0.000 0.008 0.000
#> GSM39185 5 0.4000 0.7897 0.184 0.000 0.060 0.004 0.752 0.000
#> GSM39186 1 0.1088 0.7286 0.960 0.000 0.000 0.016 0.024 0.000
#> GSM39187 1 0.0146 0.7399 0.996 0.000 0.000 0.000 0.004 0.000
#> GSM39116 2 0.1080 0.7141 0.000 0.960 0.000 0.004 0.004 0.032
#> GSM39117 4 0.5104 0.6043 0.000 0.088 0.000 0.540 0.000 0.372
#> GSM39118 6 0.4415 0.6467 0.000 0.420 0.000 0.004 0.020 0.556
#> GSM39119 6 0.4541 0.6099 0.000 0.160 0.000 0.088 0.020 0.732
#> GSM39120 1 0.6379 0.4142 0.568 0.116 0.000 0.200 0.116 0.000
#> GSM39121 2 0.4467 0.6230 0.132 0.756 0.000 0.064 0.048 0.000
#> GSM39122 2 0.4455 0.6290 0.120 0.760 0.000 0.072 0.048 0.000
#> GSM39123 4 0.5104 0.6043 0.000 0.088 0.000 0.540 0.000 0.372
#> GSM39124 2 0.0976 0.7324 0.016 0.968 0.000 0.008 0.008 0.000
#> GSM39125 1 0.4169 0.6223 0.788 0.080 0.000 0.076 0.056 0.000
#> GSM39126 2 0.4574 0.6244 0.120 0.752 0.000 0.072 0.056 0.000
#> GSM39127 2 0.0547 0.7269 0.000 0.980 0.000 0.000 0.000 0.020
#> GSM39128 2 0.1223 0.7324 0.016 0.960 0.000 0.004 0.008 0.012
#> GSM39129 6 0.3198 0.8124 0.000 0.188 0.000 0.008 0.008 0.796
#> GSM39130 4 0.5104 0.6043 0.000 0.088 0.000 0.540 0.000 0.372
#> GSM39131 2 0.0291 0.7321 0.000 0.992 0.000 0.004 0.000 0.004
#> GSM39132 2 0.0632 0.7244 0.000 0.976 0.000 0.000 0.000 0.024
#> GSM39133 4 0.5127 0.5987 0.000 0.092 0.000 0.544 0.000 0.364
#> GSM39134 6 0.3371 0.8085 0.000 0.200 0.000 0.004 0.016 0.780
#> GSM39135 2 0.0935 0.7152 0.000 0.964 0.000 0.000 0.004 0.032
#> GSM39136 2 0.1080 0.7141 0.000 0.960 0.000 0.004 0.004 0.032
#> GSM39137 2 0.1692 0.7197 0.048 0.932 0.000 0.012 0.008 0.000
#> GSM39138 6 0.2838 0.8142 0.000 0.188 0.000 0.000 0.004 0.808
#> GSM39139 6 0.4079 0.6563 0.000 0.380 0.000 0.004 0.008 0.608
#> GSM39140 1 0.1798 0.7216 0.932 0.020 0.000 0.020 0.028 0.000
#> GSM39141 1 0.1346 0.7288 0.952 0.016 0.000 0.008 0.024 0.000
#> GSM39142 1 0.1251 0.7305 0.956 0.012 0.000 0.008 0.024 0.000
#> GSM39143 1 0.1346 0.7288 0.952 0.016 0.000 0.008 0.024 0.000
#> GSM39144 6 0.3012 0.8158 0.000 0.196 0.000 0.000 0.008 0.796
#> GSM39145 2 0.4211 -0.3955 0.000 0.532 0.000 0.004 0.008 0.456
#> GSM39146 2 0.0603 0.7277 0.000 0.980 0.000 0.004 0.000 0.016
#> GSM39147 2 0.1080 0.7242 0.000 0.960 0.000 0.004 0.004 0.032
#> GSM39188 3 0.3481 0.6715 0.000 0.000 0.836 0.068 0.044 0.052
#> GSM39189 3 0.5495 0.6405 0.000 0.000 0.524 0.096 0.368 0.012
#> GSM39190 3 0.5628 0.7068 0.000 0.000 0.600 0.080 0.272 0.048
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) other(p) protocol(p) k
#> SD:kmeans 80 0.1285 1.76e-08 2.87e-07 2
#> SD:kmeans 84 0.0252 6.83e-09 2.65e-08 3
#> SD:kmeans 70 0.0524 1.50e-06 6.19e-09 4
#> SD:kmeans 70 0.1673 5.61e-06 2.30e-07 5
#> SD:kmeans 68 0.8553 2.76e-05 9.53e-07 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 8353 rows and 87 columns.
#> Top rows (835, 1670, 2506, 3341, 4176) 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 0.904 0.914 0.968 0.5006 0.500 0.500
#> 3 3 0.689 0.826 0.914 0.3039 0.786 0.600
#> 4 4 0.581 0.656 0.804 0.1349 0.876 0.663
#> 5 5 0.568 0.506 0.715 0.0645 0.910 0.678
#> 6 6 0.602 0.501 0.688 0.0432 0.905 0.608
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
#> GSM39104 1 0.0000 0.96929 1.000 0.000
#> GSM39105 1 0.0000 0.96929 1.000 0.000
#> GSM39106 1 0.0000 0.96929 1.000 0.000
#> GSM39107 1 0.0000 0.96929 1.000 0.000
#> GSM39108 1 0.0000 0.96929 1.000 0.000
#> GSM39109 2 0.0672 0.95494 0.008 0.992
#> GSM39110 1 0.0000 0.96929 1.000 0.000
#> GSM39111 1 0.0000 0.96929 1.000 0.000
#> GSM39112 1 0.0000 0.96929 1.000 0.000
#> GSM39113 1 0.0000 0.96929 1.000 0.000
#> GSM39114 2 0.0000 0.96092 0.000 1.000
#> GSM39115 1 0.0000 0.96929 1.000 0.000
#> GSM39148 1 0.0000 0.96929 1.000 0.000
#> GSM39149 2 0.0000 0.96092 0.000 1.000
#> GSM39150 1 0.0000 0.96929 1.000 0.000
#> GSM39151 2 0.0000 0.96092 0.000 1.000
#> GSM39152 2 0.9983 0.09761 0.476 0.524
#> GSM39153 1 0.0000 0.96929 1.000 0.000
#> GSM39154 1 0.0000 0.96929 1.000 0.000
#> GSM39155 1 0.0000 0.96929 1.000 0.000
#> GSM39156 1 0.0000 0.96929 1.000 0.000
#> GSM39157 1 0.0000 0.96929 1.000 0.000
#> GSM39158 1 0.0000 0.96929 1.000 0.000
#> GSM39159 1 0.7139 0.73569 0.804 0.196
#> GSM39160 1 0.0000 0.96929 1.000 0.000
#> GSM39161 1 1.0000 -0.05080 0.500 0.500
#> GSM39162 1 0.0000 0.96929 1.000 0.000
#> GSM39163 1 0.0000 0.96929 1.000 0.000
#> GSM39164 1 0.0000 0.96929 1.000 0.000
#> GSM39165 1 0.0000 0.96929 1.000 0.000
#> GSM39166 1 0.0000 0.96929 1.000 0.000
#> GSM39167 1 0.0000 0.96929 1.000 0.000
#> GSM39168 1 0.0000 0.96929 1.000 0.000
#> GSM39169 1 0.0000 0.96929 1.000 0.000
#> GSM39170 1 0.0000 0.96929 1.000 0.000
#> GSM39171 1 0.0000 0.96929 1.000 0.000
#> GSM39172 2 0.0000 0.96092 0.000 1.000
#> GSM39173 2 0.0000 0.96092 0.000 1.000
#> GSM39174 1 0.0000 0.96929 1.000 0.000
#> GSM39175 1 0.0000 0.96929 1.000 0.000
#> GSM39176 1 0.0000 0.96929 1.000 0.000
#> GSM39177 2 0.6801 0.76990 0.180 0.820
#> GSM39178 1 0.0000 0.96929 1.000 0.000
#> GSM39179 2 0.0000 0.96092 0.000 1.000
#> GSM39180 2 0.0000 0.96092 0.000 1.000
#> GSM39181 1 0.0000 0.96929 1.000 0.000
#> GSM39182 2 0.5408 0.84018 0.124 0.876
#> GSM39183 1 0.0000 0.96929 1.000 0.000
#> GSM39184 1 0.0000 0.96929 1.000 0.000
#> GSM39185 2 0.4815 0.86469 0.104 0.896
#> GSM39186 1 0.0000 0.96929 1.000 0.000
#> GSM39187 1 0.0000 0.96929 1.000 0.000
#> GSM39116 2 0.0000 0.96092 0.000 1.000
#> GSM39117 2 0.0000 0.96092 0.000 1.000
#> GSM39118 2 0.0000 0.96092 0.000 1.000
#> GSM39119 2 0.0000 0.96092 0.000 1.000
#> GSM39120 1 0.0000 0.96929 1.000 0.000
#> GSM39121 1 0.8144 0.64837 0.748 0.252
#> GSM39122 1 0.9775 0.28717 0.588 0.412
#> GSM39123 2 0.0000 0.96092 0.000 1.000
#> GSM39124 2 0.0000 0.96092 0.000 1.000
#> GSM39125 1 0.0000 0.96929 1.000 0.000
#> GSM39126 2 0.9998 0.00163 0.492 0.508
#> GSM39127 2 0.0000 0.96092 0.000 1.000
#> GSM39128 2 0.0000 0.96092 0.000 1.000
#> GSM39129 2 0.0000 0.96092 0.000 1.000
#> GSM39130 2 0.0000 0.96092 0.000 1.000
#> GSM39131 2 0.0000 0.96092 0.000 1.000
#> GSM39132 2 0.0000 0.96092 0.000 1.000
#> GSM39133 2 0.0000 0.96092 0.000 1.000
#> GSM39134 2 0.0000 0.96092 0.000 1.000
#> GSM39135 2 0.0000 0.96092 0.000 1.000
#> GSM39136 2 0.0000 0.96092 0.000 1.000
#> GSM39137 2 0.0376 0.95804 0.004 0.996
#> GSM39138 2 0.0000 0.96092 0.000 1.000
#> GSM39139 2 0.0000 0.96092 0.000 1.000
#> GSM39140 1 0.0000 0.96929 1.000 0.000
#> GSM39141 1 0.0000 0.96929 1.000 0.000
#> GSM39142 1 0.0000 0.96929 1.000 0.000
#> GSM39143 1 0.0000 0.96929 1.000 0.000
#> GSM39144 2 0.0000 0.96092 0.000 1.000
#> GSM39145 2 0.0000 0.96092 0.000 1.000
#> GSM39146 2 0.0000 0.96092 0.000 1.000
#> GSM39147 2 0.0000 0.96092 0.000 1.000
#> GSM39188 2 0.0000 0.96092 0.000 1.000
#> GSM39189 2 0.1843 0.93844 0.028 0.972
#> GSM39190 2 0.0000 0.96092 0.000 1.000
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM39104 1 0.3816 0.797 0.852 0.000 0.148
#> GSM39105 1 0.0000 0.911 1.000 0.000 0.000
#> GSM39106 1 0.4058 0.852 0.880 0.076 0.044
#> GSM39107 1 0.4178 0.783 0.828 0.172 0.000
#> GSM39108 1 0.0424 0.909 0.992 0.008 0.000
#> GSM39109 2 0.6427 0.510 0.012 0.640 0.348
#> GSM39110 1 0.5598 0.772 0.800 0.052 0.148
#> GSM39111 1 0.5363 0.617 0.724 0.000 0.276
#> GSM39112 1 0.3686 0.815 0.860 0.140 0.000
#> GSM39113 1 0.5560 0.614 0.700 0.300 0.000
#> GSM39114 2 0.0000 0.886 0.000 1.000 0.000
#> GSM39115 1 0.0000 0.911 1.000 0.000 0.000
#> GSM39148 1 0.0000 0.911 1.000 0.000 0.000
#> GSM39149 3 0.0000 0.899 0.000 0.000 1.000
#> GSM39150 3 0.6307 0.026 0.488 0.000 0.512
#> GSM39151 3 0.0000 0.899 0.000 0.000 1.000
#> GSM39152 3 0.0592 0.897 0.012 0.000 0.988
#> GSM39153 1 0.0000 0.911 1.000 0.000 0.000
#> GSM39154 1 0.0000 0.911 1.000 0.000 0.000
#> GSM39155 1 0.0000 0.911 1.000 0.000 0.000
#> GSM39156 1 0.0424 0.909 0.992 0.008 0.000
#> GSM39157 1 0.0000 0.911 1.000 0.000 0.000
#> GSM39158 1 0.1163 0.898 0.972 0.000 0.028
#> GSM39159 3 0.2356 0.860 0.072 0.000 0.928
#> GSM39160 3 0.5882 0.465 0.348 0.000 0.652
#> GSM39161 3 0.0747 0.895 0.016 0.000 0.984
#> GSM39162 1 0.0000 0.911 1.000 0.000 0.000
#> GSM39163 1 0.0000 0.911 1.000 0.000 0.000
#> GSM39164 1 0.0000 0.911 1.000 0.000 0.000
#> GSM39165 3 0.5254 0.637 0.264 0.000 0.736
#> GSM39166 1 0.5706 0.521 0.680 0.000 0.320
#> GSM39167 1 0.0000 0.911 1.000 0.000 0.000
#> GSM39168 1 0.0000 0.911 1.000 0.000 0.000
#> GSM39169 1 0.0237 0.910 0.996 0.000 0.004
#> GSM39170 1 0.1031 0.901 0.976 0.000 0.024
#> GSM39171 1 0.5650 0.542 0.688 0.000 0.312
#> GSM39172 3 0.0000 0.899 0.000 0.000 1.000
#> GSM39173 3 0.0592 0.892 0.000 0.012 0.988
#> GSM39174 1 0.0000 0.911 1.000 0.000 0.000
#> GSM39175 1 0.1964 0.881 0.944 0.000 0.056
#> GSM39176 1 0.0000 0.911 1.000 0.000 0.000
#> GSM39177 3 0.0237 0.899 0.004 0.000 0.996
#> GSM39178 3 0.3482 0.813 0.128 0.000 0.872
#> GSM39179 3 0.0000 0.899 0.000 0.000 1.000
#> GSM39180 3 0.1163 0.879 0.000 0.028 0.972
#> GSM39181 1 0.5497 0.577 0.708 0.000 0.292
#> GSM39182 3 0.1878 0.862 0.004 0.044 0.952
#> GSM39183 1 0.6267 0.147 0.548 0.000 0.452
#> GSM39184 1 0.0000 0.911 1.000 0.000 0.000
#> GSM39185 3 0.0237 0.898 0.004 0.000 0.996
#> GSM39186 1 0.0237 0.910 0.996 0.000 0.004
#> GSM39187 1 0.0000 0.911 1.000 0.000 0.000
#> GSM39116 2 0.0424 0.886 0.000 0.992 0.008
#> GSM39117 2 0.5178 0.766 0.000 0.744 0.256
#> GSM39118 2 0.3482 0.857 0.000 0.872 0.128
#> GSM39119 2 0.4399 0.827 0.000 0.812 0.188
#> GSM39120 1 0.3896 0.823 0.864 0.128 0.008
#> GSM39121 2 0.4887 0.658 0.228 0.772 0.000
#> GSM39122 2 0.3482 0.783 0.128 0.872 0.000
#> GSM39123 2 0.5178 0.766 0.000 0.744 0.256
#> GSM39124 2 0.0000 0.886 0.000 1.000 0.000
#> GSM39125 1 0.3272 0.846 0.892 0.104 0.004
#> GSM39126 2 0.2625 0.827 0.084 0.916 0.000
#> GSM39127 2 0.0000 0.886 0.000 1.000 0.000
#> GSM39128 2 0.0000 0.886 0.000 1.000 0.000
#> GSM39129 2 0.4235 0.834 0.000 0.824 0.176
#> GSM39130 2 0.5138 0.770 0.000 0.748 0.252
#> GSM39131 2 0.0000 0.886 0.000 1.000 0.000
#> GSM39132 2 0.0237 0.886 0.000 0.996 0.004
#> GSM39133 2 0.4796 0.801 0.000 0.780 0.220
#> GSM39134 2 0.4002 0.842 0.000 0.840 0.160
#> GSM39135 2 0.0424 0.886 0.000 0.992 0.008
#> GSM39136 2 0.0424 0.887 0.000 0.992 0.008
#> GSM39137 2 0.0000 0.886 0.000 1.000 0.000
#> GSM39138 2 0.4291 0.832 0.000 0.820 0.180
#> GSM39139 2 0.2066 0.878 0.000 0.940 0.060
#> GSM39140 1 0.0747 0.905 0.984 0.016 0.000
#> GSM39141 1 0.0237 0.910 0.996 0.004 0.000
#> GSM39142 1 0.0237 0.910 0.996 0.004 0.000
#> GSM39143 1 0.0237 0.910 0.996 0.004 0.000
#> GSM39144 2 0.4062 0.840 0.000 0.836 0.164
#> GSM39145 2 0.0892 0.886 0.000 0.980 0.020
#> GSM39146 2 0.0237 0.886 0.000 0.996 0.004
#> GSM39147 2 0.0000 0.886 0.000 1.000 0.000
#> GSM39188 3 0.0000 0.899 0.000 0.000 1.000
#> GSM39189 3 0.0000 0.899 0.000 0.000 1.000
#> GSM39190 3 0.0000 0.899 0.000 0.000 1.000
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM39104 4 0.7285 0.2693 0.300 0.000 0.180 0.520
#> GSM39105 1 0.4891 0.5706 0.680 0.000 0.012 0.308
#> GSM39106 4 0.4405 0.6126 0.152 0.000 0.048 0.800
#> GSM39107 4 0.3383 0.6523 0.076 0.052 0.000 0.872
#> GSM39108 4 0.5855 0.3557 0.356 0.000 0.044 0.600
#> GSM39109 4 0.7122 0.3605 0.000 0.248 0.192 0.560
#> GSM39110 4 0.6954 0.5115 0.224 0.008 0.156 0.612
#> GSM39111 4 0.7700 0.2376 0.304 0.000 0.248 0.448
#> GSM39112 4 0.3946 0.6453 0.168 0.020 0.000 0.812
#> GSM39113 4 0.3398 0.6450 0.068 0.060 0.000 0.872
#> GSM39114 4 0.4585 0.3247 0.000 0.332 0.000 0.668
#> GSM39115 1 0.3982 0.7046 0.776 0.000 0.004 0.220
#> GSM39148 1 0.0921 0.8264 0.972 0.000 0.000 0.028
#> GSM39149 3 0.2101 0.7556 0.000 0.060 0.928 0.012
#> GSM39150 3 0.7904 0.0432 0.324 0.000 0.368 0.308
#> GSM39151 3 0.2255 0.7561 0.000 0.068 0.920 0.012
#> GSM39152 3 0.2707 0.7323 0.008 0.016 0.908 0.068
#> GSM39153 1 0.1022 0.8314 0.968 0.000 0.000 0.032
#> GSM39154 1 0.0657 0.8318 0.984 0.000 0.004 0.012
#> GSM39155 1 0.1211 0.8295 0.960 0.000 0.000 0.040
#> GSM39156 1 0.3801 0.6598 0.780 0.000 0.000 0.220
#> GSM39157 1 0.0188 0.8314 0.996 0.000 0.000 0.004
#> GSM39158 1 0.4245 0.7467 0.820 0.000 0.064 0.116
#> GSM39159 3 0.5771 0.5711 0.212 0.004 0.704 0.080
#> GSM39160 3 0.7594 0.2523 0.264 0.000 0.480 0.256
#> GSM39161 3 0.4805 0.6867 0.088 0.016 0.808 0.088
#> GSM39162 1 0.1118 0.8232 0.964 0.000 0.000 0.036
#> GSM39163 1 0.0188 0.8312 0.996 0.000 0.000 0.004
#> GSM39164 1 0.1557 0.8284 0.944 0.000 0.000 0.056
#> GSM39165 3 0.6077 0.0523 0.460 0.000 0.496 0.044
#> GSM39166 1 0.7004 0.4674 0.580 0.000 0.200 0.220
#> GSM39167 1 0.0592 0.8308 0.984 0.000 0.000 0.016
#> GSM39168 1 0.1022 0.8249 0.968 0.000 0.000 0.032
#> GSM39169 1 0.1854 0.8255 0.940 0.000 0.012 0.048
#> GSM39170 1 0.4920 0.7111 0.776 0.000 0.088 0.136
#> GSM39171 1 0.6634 0.5034 0.624 0.000 0.212 0.164
#> GSM39172 3 0.2408 0.7439 0.000 0.104 0.896 0.000
#> GSM39173 3 0.2944 0.7322 0.000 0.128 0.868 0.004
#> GSM39174 1 0.0188 0.8317 0.996 0.000 0.000 0.004
#> GSM39175 1 0.2124 0.8200 0.932 0.000 0.040 0.028
#> GSM39176 1 0.0707 0.8317 0.980 0.000 0.000 0.020
#> GSM39177 3 0.2245 0.7551 0.008 0.040 0.932 0.020
#> GSM39178 3 0.5184 0.5998 0.060 0.000 0.736 0.204
#> GSM39179 3 0.2542 0.7530 0.000 0.084 0.904 0.012
#> GSM39180 3 0.4428 0.5687 0.000 0.276 0.720 0.004
#> GSM39181 1 0.6133 0.5899 0.676 0.000 0.188 0.136
#> GSM39182 3 0.6199 0.5113 0.028 0.288 0.648 0.036
#> GSM39183 1 0.7345 0.2941 0.508 0.000 0.308 0.184
#> GSM39184 1 0.1576 0.8225 0.948 0.000 0.004 0.048
#> GSM39185 3 0.5194 0.6890 0.056 0.040 0.792 0.112
#> GSM39186 1 0.2859 0.7946 0.880 0.000 0.008 0.112
#> GSM39187 1 0.0817 0.8320 0.976 0.000 0.000 0.024
#> GSM39116 2 0.1637 0.7939 0.000 0.940 0.000 0.060
#> GSM39117 2 0.3528 0.7204 0.000 0.808 0.192 0.000
#> GSM39118 2 0.2125 0.7990 0.000 0.920 0.076 0.004
#> GSM39119 2 0.2868 0.7722 0.000 0.864 0.136 0.000
#> GSM39120 4 0.4955 0.5960 0.244 0.024 0.004 0.728
#> GSM39121 4 0.6478 0.5001 0.132 0.236 0.000 0.632
#> GSM39122 4 0.5827 0.3788 0.052 0.316 0.000 0.632
#> GSM39123 2 0.3444 0.7291 0.000 0.816 0.184 0.000
#> GSM39124 2 0.4500 0.5779 0.000 0.684 0.000 0.316
#> GSM39125 4 0.6012 0.3548 0.404 0.024 0.012 0.560
#> GSM39126 4 0.5646 0.4115 0.048 0.296 0.000 0.656
#> GSM39127 2 0.4072 0.6647 0.000 0.748 0.000 0.252
#> GSM39128 2 0.4477 0.5864 0.000 0.688 0.000 0.312
#> GSM39129 2 0.2408 0.7884 0.000 0.896 0.104 0.000
#> GSM39130 2 0.3486 0.7252 0.000 0.812 0.188 0.000
#> GSM39131 2 0.4431 0.5975 0.000 0.696 0.000 0.304
#> GSM39132 2 0.3266 0.7395 0.000 0.832 0.000 0.168
#> GSM39133 2 0.2868 0.7718 0.000 0.864 0.136 0.000
#> GSM39134 2 0.1940 0.7980 0.000 0.924 0.076 0.000
#> GSM39135 2 0.2081 0.7850 0.000 0.916 0.000 0.084
#> GSM39136 2 0.1743 0.7960 0.000 0.940 0.004 0.056
#> GSM39137 2 0.5112 0.4370 0.008 0.608 0.000 0.384
#> GSM39138 2 0.2530 0.7848 0.000 0.888 0.112 0.000
#> GSM39139 2 0.1733 0.8026 0.000 0.948 0.028 0.024
#> GSM39140 1 0.4304 0.5253 0.716 0.000 0.000 0.284
#> GSM39141 1 0.3172 0.7236 0.840 0.000 0.000 0.160
#> GSM39142 1 0.3074 0.7371 0.848 0.000 0.000 0.152
#> GSM39143 1 0.3266 0.7172 0.832 0.000 0.000 0.168
#> GSM39144 2 0.2081 0.7957 0.000 0.916 0.084 0.000
#> GSM39145 2 0.2255 0.7980 0.000 0.920 0.012 0.068
#> GSM39146 2 0.2345 0.7795 0.000 0.900 0.000 0.100
#> GSM39147 2 0.3311 0.7383 0.000 0.828 0.000 0.172
#> GSM39188 3 0.2011 0.7525 0.000 0.080 0.920 0.000
#> GSM39189 3 0.2443 0.7408 0.000 0.024 0.916 0.060
#> GSM39190 3 0.1978 0.7549 0.000 0.068 0.928 0.004
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM39104 4 0.7685 0.2094 0.136 0.000 0.116 0.460 0.288
#> GSM39105 1 0.6418 0.2533 0.508 0.000 0.004 0.316 0.172
#> GSM39106 5 0.6675 0.1737 0.108 0.000 0.036 0.356 0.500
#> GSM39107 5 0.4841 0.4438 0.052 0.012 0.000 0.220 0.716
#> GSM39108 4 0.7426 -0.0631 0.216 0.000 0.040 0.384 0.360
#> GSM39109 5 0.8508 0.1006 0.004 0.192 0.172 0.304 0.328
#> GSM39110 5 0.7652 0.0391 0.100 0.000 0.132 0.364 0.404
#> GSM39111 4 0.8299 0.2429 0.152 0.000 0.232 0.376 0.240
#> GSM39112 5 0.5792 0.3530 0.128 0.004 0.004 0.228 0.636
#> GSM39113 5 0.4165 0.4413 0.032 0.004 0.000 0.208 0.756
#> GSM39114 5 0.4400 0.4487 0.000 0.196 0.000 0.060 0.744
#> GSM39115 1 0.5942 0.4124 0.580 0.000 0.004 0.292 0.124
#> GSM39148 1 0.1216 0.7792 0.960 0.000 0.000 0.020 0.020
#> GSM39149 3 0.1300 0.7170 0.000 0.016 0.956 0.028 0.000
#> GSM39150 4 0.7583 0.4588 0.164 0.000 0.236 0.496 0.104
#> GSM39151 3 0.1403 0.7223 0.000 0.024 0.952 0.024 0.000
#> GSM39152 3 0.2873 0.6320 0.000 0.000 0.856 0.128 0.016
#> GSM39153 1 0.1682 0.7864 0.940 0.000 0.004 0.044 0.012
#> GSM39154 1 0.1557 0.7844 0.940 0.000 0.008 0.052 0.000
#> GSM39155 1 0.2110 0.7774 0.912 0.000 0.000 0.072 0.016
#> GSM39156 1 0.4258 0.6723 0.768 0.000 0.000 0.072 0.160
#> GSM39157 1 0.1444 0.7857 0.948 0.000 0.000 0.040 0.012
#> GSM39158 1 0.4928 0.1709 0.548 0.000 0.020 0.428 0.004
#> GSM39159 4 0.6657 0.2525 0.112 0.008 0.392 0.472 0.016
#> GSM39160 4 0.7496 0.3267 0.164 0.000 0.368 0.404 0.064
#> GSM39161 4 0.6144 0.1213 0.060 0.016 0.456 0.460 0.008
#> GSM39162 1 0.1403 0.7771 0.952 0.000 0.000 0.024 0.024
#> GSM39163 1 0.1205 0.7843 0.956 0.000 0.000 0.040 0.004
#> GSM39164 1 0.2632 0.7776 0.888 0.000 0.000 0.072 0.040
#> GSM39165 3 0.7093 -0.2192 0.340 0.000 0.444 0.188 0.028
#> GSM39166 4 0.5403 0.4207 0.292 0.000 0.076 0.628 0.004
#> GSM39167 1 0.0609 0.7827 0.980 0.000 0.000 0.020 0.000
#> GSM39168 1 0.1211 0.7794 0.960 0.000 0.000 0.024 0.016
#> GSM39169 1 0.3387 0.7445 0.836 0.000 0.004 0.128 0.032
#> GSM39170 1 0.4888 0.0818 0.508 0.000 0.016 0.472 0.004
#> GSM39171 1 0.7378 -0.0734 0.464 0.000 0.168 0.308 0.060
#> GSM39172 3 0.4045 0.6465 0.000 0.136 0.796 0.064 0.004
#> GSM39173 3 0.4460 0.6300 0.000 0.136 0.772 0.084 0.008
#> GSM39174 1 0.1484 0.7855 0.944 0.000 0.000 0.048 0.008
#> GSM39175 1 0.3449 0.7269 0.844 0.000 0.064 0.088 0.004
#> GSM39176 1 0.1041 0.7845 0.964 0.000 0.000 0.032 0.004
#> GSM39177 3 0.1830 0.6928 0.012 0.004 0.932 0.052 0.000
#> GSM39178 4 0.6005 0.2300 0.044 0.000 0.412 0.508 0.036
#> GSM39179 3 0.1399 0.7216 0.000 0.028 0.952 0.020 0.000
#> GSM39180 3 0.5882 0.3469 0.000 0.376 0.528 0.092 0.004
#> GSM39181 4 0.5535 0.2309 0.392 0.000 0.072 0.536 0.000
#> GSM39182 3 0.7524 0.2977 0.024 0.364 0.420 0.168 0.024
#> GSM39183 4 0.5681 0.4624 0.268 0.000 0.124 0.608 0.000
#> GSM39184 1 0.2907 0.7483 0.864 0.000 0.008 0.116 0.012
#> GSM39185 4 0.6578 0.1597 0.032 0.080 0.360 0.520 0.008
#> GSM39186 1 0.4967 0.6072 0.716 0.000 0.012 0.204 0.068
#> GSM39187 1 0.1717 0.7855 0.936 0.000 0.004 0.052 0.008
#> GSM39116 2 0.2784 0.7232 0.000 0.872 0.004 0.016 0.108
#> GSM39117 2 0.3867 0.6729 0.000 0.804 0.144 0.048 0.004
#> GSM39118 2 0.2990 0.7488 0.000 0.876 0.080 0.012 0.032
#> GSM39119 2 0.2990 0.7331 0.000 0.868 0.100 0.024 0.008
#> GSM39120 5 0.6646 0.3423 0.196 0.012 0.016 0.192 0.584
#> GSM39121 5 0.5277 0.4757 0.092 0.144 0.000 0.036 0.728
#> GSM39122 5 0.4536 0.4630 0.044 0.176 0.000 0.020 0.760
#> GSM39123 2 0.3779 0.6830 0.000 0.812 0.136 0.048 0.004
#> GSM39124 5 0.4965 -0.1828 0.000 0.452 0.000 0.028 0.520
#> GSM39125 5 0.6773 0.1397 0.336 0.012 0.000 0.188 0.464
#> GSM39126 5 0.4406 0.4613 0.016 0.172 0.000 0.044 0.768
#> GSM39127 2 0.4510 0.3280 0.000 0.560 0.000 0.008 0.432
#> GSM39128 5 0.4743 -0.1819 0.000 0.472 0.000 0.016 0.512
#> GSM39129 2 0.3530 0.7451 0.000 0.844 0.104 0.024 0.028
#> GSM39130 2 0.3795 0.6743 0.000 0.808 0.144 0.044 0.004
#> GSM39131 2 0.4767 0.3236 0.000 0.560 0.000 0.020 0.420
#> GSM39132 2 0.4339 0.4997 0.000 0.652 0.000 0.012 0.336
#> GSM39133 2 0.3107 0.7210 0.000 0.864 0.096 0.032 0.008
#> GSM39134 2 0.2882 0.7509 0.000 0.888 0.060 0.028 0.024
#> GSM39135 2 0.3011 0.7054 0.000 0.844 0.000 0.016 0.140
#> GSM39136 2 0.2352 0.7304 0.000 0.896 0.004 0.008 0.092
#> GSM39137 5 0.5127 -0.0214 0.016 0.416 0.000 0.016 0.552
#> GSM39138 2 0.3361 0.7449 0.000 0.856 0.092 0.032 0.020
#> GSM39139 2 0.4041 0.7054 0.000 0.804 0.024 0.032 0.140
#> GSM39140 1 0.4824 0.6139 0.720 0.004 0.000 0.076 0.200
#> GSM39141 1 0.3714 0.7054 0.812 0.000 0.000 0.056 0.132
#> GSM39142 1 0.3946 0.7157 0.800 0.000 0.000 0.080 0.120
#> GSM39143 1 0.4069 0.6973 0.788 0.000 0.000 0.076 0.136
#> GSM39144 2 0.3811 0.7469 0.000 0.836 0.080 0.028 0.056
#> GSM39145 2 0.4755 0.6569 0.000 0.732 0.028 0.032 0.208
#> GSM39146 2 0.3914 0.6427 0.000 0.760 0.004 0.016 0.220
#> GSM39147 2 0.4920 0.4274 0.000 0.584 0.000 0.032 0.384
#> GSM39188 3 0.2153 0.7200 0.000 0.040 0.916 0.044 0.000
#> GSM39189 3 0.3300 0.6634 0.000 0.020 0.856 0.100 0.024
#> GSM39190 3 0.2221 0.7162 0.000 0.036 0.912 0.052 0.000
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM39104 6 0.636 0.439 0.080 0.040 0.068 0.000 0.204 0.608
#> GSM39105 6 0.641 0.156 0.352 0.016 0.012 0.000 0.172 0.448
#> GSM39106 6 0.590 0.588 0.068 0.168 0.024 0.000 0.084 0.656
#> GSM39107 6 0.538 0.313 0.036 0.440 0.000 0.004 0.032 0.488
#> GSM39108 6 0.631 0.532 0.160 0.084 0.028 0.000 0.104 0.624
#> GSM39109 6 0.789 0.391 0.008 0.136 0.112 0.132 0.112 0.500
#> GSM39110 6 0.699 0.522 0.120 0.104 0.128 0.000 0.068 0.580
#> GSM39111 6 0.705 0.377 0.100 0.028 0.152 0.000 0.184 0.536
#> GSM39112 6 0.578 0.500 0.124 0.296 0.000 0.000 0.024 0.556
#> GSM39113 6 0.464 0.376 0.020 0.416 0.000 0.004 0.008 0.552
#> GSM39114 2 0.406 0.418 0.000 0.744 0.000 0.060 0.004 0.192
#> GSM39115 1 0.663 0.203 0.456 0.028 0.008 0.000 0.220 0.288
#> GSM39148 1 0.130 0.759 0.952 0.012 0.000 0.000 0.004 0.032
#> GSM39149 3 0.267 0.771 0.000 0.008 0.892 0.028 0.028 0.044
#> GSM39150 5 0.720 0.214 0.096 0.012 0.152 0.000 0.432 0.308
#> GSM39151 3 0.262 0.778 0.000 0.008 0.892 0.056 0.024 0.020
#> GSM39152 3 0.410 0.674 0.000 0.004 0.776 0.012 0.132 0.076
#> GSM39153 1 0.374 0.736 0.820 0.028 0.004 0.000 0.080 0.068
#> GSM39154 1 0.301 0.748 0.864 0.004 0.012 0.000 0.064 0.056
#> GSM39155 1 0.415 0.698 0.772 0.012 0.004 0.000 0.132 0.080
#> GSM39156 1 0.524 0.542 0.672 0.100 0.004 0.000 0.028 0.196
#> GSM39157 1 0.298 0.750 0.856 0.012 0.000 0.000 0.092 0.040
#> GSM39158 5 0.465 0.354 0.348 0.012 0.000 0.000 0.608 0.032
#> GSM39159 5 0.614 0.392 0.080 0.008 0.304 0.012 0.560 0.036
#> GSM39160 5 0.783 0.137 0.120 0.020 0.256 0.000 0.324 0.280
#> GSM39161 5 0.495 0.500 0.032 0.004 0.224 0.024 0.696 0.020
#> GSM39162 1 0.160 0.757 0.940 0.024 0.000 0.000 0.008 0.028
#> GSM39163 1 0.321 0.738 0.832 0.012 0.000 0.000 0.124 0.032
#> GSM39164 1 0.277 0.755 0.864 0.004 0.000 0.000 0.040 0.092
#> GSM39165 3 0.794 -0.130 0.276 0.020 0.384 0.012 0.188 0.120
#> GSM39166 5 0.381 0.591 0.112 0.000 0.028 0.000 0.804 0.056
#> GSM39167 1 0.126 0.761 0.952 0.000 0.000 0.000 0.028 0.020
#> GSM39168 1 0.131 0.760 0.952 0.008 0.000 0.000 0.008 0.032
#> GSM39169 1 0.510 0.637 0.688 0.016 0.008 0.000 0.176 0.112
#> GSM39170 5 0.522 0.286 0.380 0.008 0.004 0.000 0.544 0.064
#> GSM39171 1 0.768 -0.188 0.332 0.016 0.108 0.000 0.308 0.236
#> GSM39172 3 0.561 0.590 0.000 0.008 0.644 0.220 0.076 0.052
#> GSM39173 3 0.507 0.699 0.000 0.028 0.732 0.124 0.080 0.036
#> GSM39174 1 0.285 0.762 0.872 0.008 0.004 0.000 0.060 0.056
#> GSM39175 1 0.549 0.608 0.684 0.012 0.064 0.000 0.160 0.080
#> GSM39176 1 0.204 0.761 0.912 0.008 0.000 0.000 0.064 0.016
#> GSM39177 3 0.336 0.752 0.008 0.008 0.856 0.020 0.044 0.064
#> GSM39178 5 0.567 0.368 0.012 0.004 0.276 0.000 0.576 0.132
#> GSM39179 3 0.149 0.775 0.000 0.000 0.944 0.024 0.004 0.028
#> GSM39180 4 0.635 -0.223 0.000 0.016 0.412 0.428 0.120 0.024
#> GSM39181 5 0.386 0.583 0.208 0.004 0.020 0.000 0.756 0.012
#> GSM39182 4 0.822 -0.117 0.032 0.028 0.248 0.404 0.176 0.112
#> GSM39183 5 0.377 0.600 0.112 0.004 0.036 0.000 0.812 0.036
#> GSM39184 1 0.484 0.613 0.688 0.008 0.008 0.000 0.216 0.080
#> GSM39185 5 0.540 0.457 0.012 0.004 0.240 0.052 0.656 0.036
#> GSM39186 1 0.670 0.334 0.504 0.020 0.032 0.000 0.236 0.208
#> GSM39187 1 0.268 0.767 0.884 0.024 0.000 0.000 0.056 0.036
#> GSM39116 4 0.377 0.507 0.000 0.296 0.004 0.692 0.000 0.008
#> GSM39117 4 0.259 0.647 0.000 0.004 0.052 0.892 0.036 0.016
#> GSM39118 4 0.333 0.680 0.000 0.096 0.032 0.844 0.012 0.016
#> GSM39119 4 0.182 0.677 0.000 0.028 0.028 0.932 0.008 0.004
#> GSM39120 2 0.735 -0.343 0.236 0.340 0.012 0.000 0.072 0.340
#> GSM39121 2 0.438 0.389 0.084 0.772 0.000 0.012 0.020 0.112
#> GSM39122 2 0.371 0.456 0.016 0.800 0.000 0.036 0.004 0.144
#> GSM39123 4 0.260 0.647 0.000 0.004 0.052 0.892 0.032 0.020
#> GSM39124 2 0.343 0.520 0.000 0.764 0.000 0.216 0.000 0.020
#> GSM39125 2 0.746 -0.264 0.308 0.324 0.000 0.000 0.132 0.236
#> GSM39126 2 0.405 0.467 0.020 0.792 0.000 0.048 0.012 0.128
#> GSM39127 2 0.390 0.374 0.000 0.652 0.000 0.336 0.000 0.012
#> GSM39128 2 0.403 0.480 0.000 0.708 0.000 0.260 0.008 0.024
#> GSM39129 4 0.358 0.670 0.000 0.116 0.040 0.820 0.004 0.020
#> GSM39130 4 0.251 0.648 0.000 0.004 0.052 0.896 0.032 0.016
#> GSM39131 2 0.444 0.410 0.000 0.660 0.004 0.300 0.008 0.028
#> GSM39132 2 0.450 0.149 0.000 0.564 0.000 0.408 0.008 0.020
#> GSM39133 4 0.180 0.670 0.000 0.020 0.016 0.936 0.020 0.008
#> GSM39134 4 0.276 0.681 0.000 0.084 0.012 0.876 0.008 0.020
#> GSM39135 4 0.427 0.391 0.000 0.356 0.004 0.620 0.000 0.020
#> GSM39136 4 0.383 0.525 0.000 0.280 0.004 0.704 0.004 0.008
#> GSM39137 2 0.389 0.546 0.016 0.776 0.000 0.176 0.008 0.024
#> GSM39138 4 0.327 0.679 0.000 0.096 0.024 0.848 0.012 0.020
#> GSM39139 4 0.461 0.482 0.000 0.312 0.020 0.644 0.004 0.020
#> GSM39140 1 0.575 0.522 0.640 0.140 0.008 0.000 0.040 0.172
#> GSM39141 1 0.394 0.702 0.796 0.072 0.000 0.000 0.028 0.104
#> GSM39142 1 0.399 0.694 0.780 0.040 0.000 0.000 0.032 0.148
#> GSM39143 1 0.396 0.698 0.792 0.072 0.000 0.000 0.024 0.112
#> GSM39144 4 0.409 0.645 0.000 0.160 0.040 0.772 0.004 0.024
#> GSM39145 4 0.498 0.410 0.000 0.336 0.012 0.604 0.008 0.040
#> GSM39146 4 0.442 0.279 0.000 0.384 0.000 0.588 0.004 0.024
#> GSM39147 2 0.484 0.203 0.000 0.560 0.000 0.384 0.004 0.052
#> GSM39188 3 0.234 0.773 0.000 0.004 0.900 0.068 0.016 0.012
#> GSM39189 3 0.472 0.667 0.000 0.008 0.740 0.024 0.120 0.108
#> GSM39190 3 0.339 0.761 0.000 0.008 0.844 0.084 0.040 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) other(p) protocol(p) k
#> SD:skmeans 83 7.36e-02 1.42e-06 8.31e-05 2
#> SD:skmeans 84 1.74e-02 2.53e-10 1.37e-10 3
#> SD:skmeans 73 1.16e-07 3.85e-14 6.37e-15 4
#> SD:skmeans 48 NA 1.16e-05 6.35e-07 5
#> SD:skmeans 51 8.65e-10 2.01e-11 1.67e-15 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 8353 rows and 87 columns.
#> Top rows (835, 1670, 2506, 3341, 4176) 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 0.666 0.895 0.945 0.2189 0.777 0.777
#> 3 3 0.305 0.624 0.819 1.4903 0.592 0.491
#> 4 4 0.298 0.558 0.771 0.1170 0.923 0.827
#> 5 5 0.303 0.518 0.763 0.0286 0.989 0.973
#> 6 6 0.312 0.553 0.763 0.0182 1.000 1.000
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
#> GSM39104 1 0.0000 0.956 1.000 0.000
#> GSM39105 1 0.0000 0.956 1.000 0.000
#> GSM39106 1 0.0000 0.956 1.000 0.000
#> GSM39107 1 0.1414 0.943 0.980 0.020
#> GSM39108 1 0.0000 0.956 1.000 0.000
#> GSM39109 1 0.0376 0.954 0.996 0.004
#> GSM39110 1 0.0000 0.956 1.000 0.000
#> GSM39111 1 0.0000 0.956 1.000 0.000
#> GSM39112 1 0.0000 0.956 1.000 0.000
#> GSM39113 1 0.0000 0.956 1.000 0.000
#> GSM39114 1 0.4690 0.870 0.900 0.100
#> GSM39115 1 0.0000 0.956 1.000 0.000
#> GSM39148 1 0.0000 0.956 1.000 0.000
#> GSM39149 1 0.2043 0.933 0.968 0.032
#> GSM39150 1 0.0000 0.956 1.000 0.000
#> GSM39151 1 0.0938 0.949 0.988 0.012
#> GSM39152 1 0.0000 0.956 1.000 0.000
#> GSM39153 1 0.0000 0.956 1.000 0.000
#> GSM39154 1 0.0000 0.956 1.000 0.000
#> GSM39155 1 0.0000 0.956 1.000 0.000
#> GSM39156 1 0.0000 0.956 1.000 0.000
#> GSM39157 1 0.0000 0.956 1.000 0.000
#> GSM39158 1 0.0000 0.956 1.000 0.000
#> GSM39159 1 0.0000 0.956 1.000 0.000
#> GSM39160 1 0.0000 0.956 1.000 0.000
#> GSM39161 1 0.0000 0.956 1.000 0.000
#> GSM39162 1 0.0000 0.956 1.000 0.000
#> GSM39163 1 0.0000 0.956 1.000 0.000
#> GSM39164 1 0.0000 0.956 1.000 0.000
#> GSM39165 1 0.0000 0.956 1.000 0.000
#> GSM39166 1 0.0000 0.956 1.000 0.000
#> GSM39167 1 0.0000 0.956 1.000 0.000
#> GSM39168 1 0.0000 0.956 1.000 0.000
#> GSM39169 1 0.0000 0.956 1.000 0.000
#> GSM39170 1 0.0000 0.956 1.000 0.000
#> GSM39171 1 0.0000 0.956 1.000 0.000
#> GSM39172 1 0.4562 0.866 0.904 0.096
#> GSM39173 1 0.0376 0.954 0.996 0.004
#> GSM39174 1 0.0000 0.956 1.000 0.000
#> GSM39175 1 0.0000 0.956 1.000 0.000
#> GSM39176 1 0.0000 0.956 1.000 0.000
#> GSM39177 1 0.0000 0.956 1.000 0.000
#> GSM39178 1 0.0000 0.956 1.000 0.000
#> GSM39179 1 0.2236 0.931 0.964 0.036
#> GSM39180 1 0.7745 0.668 0.772 0.228
#> GSM39181 1 0.0000 0.956 1.000 0.000
#> GSM39182 1 0.0000 0.956 1.000 0.000
#> GSM39183 1 0.0000 0.956 1.000 0.000
#> GSM39184 1 0.0000 0.956 1.000 0.000
#> GSM39185 1 0.0000 0.956 1.000 0.000
#> GSM39186 1 0.0000 0.956 1.000 0.000
#> GSM39187 1 0.0000 0.956 1.000 0.000
#> GSM39116 1 0.7950 0.653 0.760 0.240
#> GSM39117 2 0.0000 0.757 0.000 1.000
#> GSM39118 2 0.9815 0.532 0.420 0.580
#> GSM39119 2 0.7219 0.786 0.200 0.800
#> GSM39120 1 0.0000 0.956 1.000 0.000
#> GSM39121 1 0.0000 0.956 1.000 0.000
#> GSM39122 1 0.2236 0.931 0.964 0.036
#> GSM39123 2 0.0000 0.757 0.000 1.000
#> GSM39124 1 0.4431 0.879 0.908 0.092
#> GSM39125 1 0.0000 0.956 1.000 0.000
#> GSM39126 1 0.0672 0.951 0.992 0.008
#> GSM39127 1 0.6531 0.785 0.832 0.168
#> GSM39128 1 0.5178 0.854 0.884 0.116
#> GSM39129 2 0.7299 0.785 0.204 0.796
#> GSM39130 2 0.0000 0.757 0.000 1.000
#> GSM39131 1 0.4690 0.870 0.900 0.100
#> GSM39132 1 0.5737 0.830 0.864 0.136
#> GSM39133 2 0.2043 0.767 0.032 0.968
#> GSM39134 2 0.8443 0.754 0.272 0.728
#> GSM39135 1 0.7299 0.725 0.796 0.204
#> GSM39136 2 0.9686 0.589 0.396 0.604
#> GSM39137 1 0.4562 0.875 0.904 0.096
#> GSM39138 2 0.8443 0.754 0.272 0.728
#> GSM39139 1 0.7219 0.733 0.800 0.200
#> GSM39140 1 0.0000 0.956 1.000 0.000
#> GSM39141 1 0.0000 0.956 1.000 0.000
#> GSM39142 1 0.0000 0.956 1.000 0.000
#> GSM39143 1 0.0000 0.956 1.000 0.000
#> GSM39144 2 0.9710 0.582 0.400 0.600
#> GSM39145 1 0.4815 0.867 0.896 0.104
#> GSM39146 1 0.4939 0.863 0.892 0.108
#> GSM39147 1 0.4690 0.870 0.900 0.100
#> GSM39188 1 0.6623 0.752 0.828 0.172
#> GSM39189 1 0.2043 0.933 0.968 0.032
#> GSM39190 1 0.5842 0.820 0.860 0.140
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM39104 1 0.6280 -0.09826 0.540 0.460 0.000
#> GSM39105 2 0.6026 0.56104 0.376 0.624 0.000
#> GSM39106 1 0.3412 0.76422 0.876 0.124 0.000
#> GSM39107 2 0.5178 0.68780 0.256 0.744 0.000
#> GSM39108 2 0.5733 0.64917 0.324 0.676 0.000
#> GSM39109 2 0.5363 0.68076 0.276 0.724 0.000
#> GSM39110 1 0.4842 0.58499 0.776 0.224 0.000
#> GSM39111 2 0.5926 0.59633 0.356 0.644 0.000
#> GSM39112 2 0.5591 0.65905 0.304 0.696 0.000
#> GSM39113 2 0.5810 0.61973 0.336 0.664 0.000
#> GSM39114 2 0.0592 0.65363 0.012 0.988 0.000
#> GSM39115 2 0.6309 0.23102 0.496 0.504 0.000
#> GSM39148 1 0.0000 0.79329 1.000 0.000 0.000
#> GSM39149 1 0.6159 0.66728 0.756 0.196 0.048
#> GSM39150 1 0.1411 0.79406 0.964 0.036 0.000
#> GSM39151 2 0.6924 0.47005 0.400 0.580 0.020
#> GSM39152 1 0.0892 0.79661 0.980 0.020 0.000
#> GSM39153 1 0.0000 0.79329 1.000 0.000 0.000
#> GSM39154 1 0.0892 0.79491 0.980 0.020 0.000
#> GSM39155 1 0.6079 0.14258 0.612 0.388 0.000
#> GSM39156 1 0.1163 0.79711 0.972 0.028 0.000
#> GSM39157 2 0.5785 0.64714 0.332 0.668 0.000
#> GSM39158 1 0.1411 0.79465 0.964 0.036 0.000
#> GSM39159 1 0.4750 0.66195 0.784 0.216 0.000
#> GSM39160 1 0.1860 0.79297 0.948 0.052 0.000
#> GSM39161 1 0.1289 0.79508 0.968 0.032 0.000
#> GSM39162 1 0.0237 0.79416 0.996 0.004 0.000
#> GSM39163 1 0.3482 0.74876 0.872 0.128 0.000
#> GSM39164 1 0.0000 0.79329 1.000 0.000 0.000
#> GSM39165 1 0.0592 0.79532 0.988 0.012 0.000
#> GSM39166 1 0.3116 0.76711 0.892 0.108 0.000
#> GSM39167 1 0.0000 0.79329 1.000 0.000 0.000
#> GSM39168 1 0.0237 0.79416 0.996 0.004 0.000
#> GSM39169 1 0.0000 0.79329 1.000 0.000 0.000
#> GSM39170 1 0.0237 0.79461 0.996 0.004 0.000
#> GSM39171 1 0.6267 -0.13069 0.548 0.452 0.000
#> GSM39172 1 0.6892 0.61540 0.736 0.112 0.152
#> GSM39173 1 0.0592 0.79146 0.988 0.012 0.000
#> GSM39174 1 0.1289 0.79280 0.968 0.032 0.000
#> GSM39175 1 0.0424 0.79543 0.992 0.008 0.000
#> GSM39176 1 0.0000 0.79329 1.000 0.000 0.000
#> GSM39177 1 0.5785 0.44120 0.668 0.332 0.000
#> GSM39178 1 0.2878 0.77546 0.904 0.096 0.000
#> GSM39179 1 0.6662 0.56184 0.704 0.252 0.044
#> GSM39180 1 0.9730 0.08220 0.428 0.340 0.232
#> GSM39181 1 0.4062 0.71143 0.836 0.164 0.000
#> GSM39182 1 0.4452 0.69861 0.808 0.192 0.000
#> GSM39183 1 0.3340 0.76403 0.880 0.120 0.000
#> GSM39184 1 0.5810 0.41298 0.664 0.336 0.000
#> GSM39185 1 0.5926 0.39085 0.644 0.356 0.000
#> GSM39186 1 0.6244 -0.00249 0.560 0.440 0.000
#> GSM39187 1 0.2261 0.78509 0.932 0.068 0.000
#> GSM39116 2 0.0000 0.64101 0.000 1.000 0.000
#> GSM39117 3 0.0000 0.84367 0.000 0.000 1.000
#> GSM39118 2 0.4796 0.53103 0.000 0.780 0.220
#> GSM39119 3 0.4842 0.74484 0.000 0.224 0.776
#> GSM39120 2 0.6286 0.32288 0.464 0.536 0.000
#> GSM39121 2 0.5650 0.66423 0.312 0.688 0.000
#> GSM39122 2 0.5016 0.69961 0.240 0.760 0.000
#> GSM39123 3 0.0000 0.84367 0.000 0.000 1.000
#> GSM39124 2 0.3941 0.71379 0.156 0.844 0.000
#> GSM39125 2 0.5905 0.59738 0.352 0.648 0.000
#> GSM39126 1 0.6274 -0.06853 0.544 0.456 0.000
#> GSM39127 2 0.0000 0.64101 0.000 1.000 0.000
#> GSM39128 2 0.5810 0.26751 0.336 0.664 0.000
#> GSM39129 3 0.5402 0.77686 0.028 0.180 0.792
#> GSM39130 3 0.0000 0.84367 0.000 0.000 1.000
#> GSM39131 2 0.0000 0.64101 0.000 1.000 0.000
#> GSM39132 2 0.2959 0.61550 0.100 0.900 0.000
#> GSM39133 3 0.1163 0.84496 0.000 0.028 0.972
#> GSM39134 3 0.7101 0.72040 0.080 0.216 0.704
#> GSM39135 2 0.1647 0.64041 0.036 0.960 0.004
#> GSM39136 2 0.4887 0.32910 0.000 0.772 0.228
#> GSM39137 2 0.2959 0.70479 0.100 0.900 0.000
#> GSM39138 3 0.7605 0.68929 0.192 0.124 0.684
#> GSM39139 2 0.0829 0.64592 0.012 0.984 0.004
#> GSM39140 2 0.6140 0.51927 0.404 0.596 0.000
#> GSM39141 2 0.5529 0.66760 0.296 0.704 0.000
#> GSM39142 2 0.5591 0.65823 0.304 0.696 0.000
#> GSM39143 2 0.5465 0.66829 0.288 0.712 0.000
#> GSM39144 2 0.4887 0.36271 0.000 0.772 0.228
#> GSM39145 2 0.1031 0.64999 0.024 0.976 0.000
#> GSM39146 2 0.0000 0.64101 0.000 1.000 0.000
#> GSM39147 2 0.1031 0.66031 0.024 0.976 0.000
#> GSM39188 1 0.8109 0.44465 0.628 0.116 0.256
#> GSM39189 1 0.3472 0.77621 0.904 0.056 0.040
#> GSM39190 2 0.7245 0.67617 0.168 0.712 0.120
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM39104 2 0.6090 0.2621 0.384 0.564 0.052 0.000
#> GSM39105 2 0.3486 0.5923 0.188 0.812 0.000 0.000
#> GSM39106 1 0.5463 0.6053 0.692 0.256 0.052 0.000
#> GSM39107 2 0.2928 0.6006 0.052 0.896 0.052 0.000
#> GSM39108 2 0.3569 0.6084 0.196 0.804 0.000 0.000
#> GSM39109 2 0.3056 0.6086 0.072 0.888 0.040 0.000
#> GSM39110 1 0.3907 0.5669 0.768 0.232 0.000 0.000
#> GSM39111 2 0.3837 0.5631 0.224 0.776 0.000 0.000
#> GSM39112 2 0.3453 0.6044 0.080 0.868 0.052 0.000
#> GSM39113 2 0.3778 0.6060 0.100 0.848 0.052 0.000
#> GSM39114 2 0.3306 0.5300 0.004 0.840 0.156 0.000
#> GSM39115 2 0.4605 0.4136 0.336 0.664 0.000 0.000
#> GSM39148 1 0.0000 0.7756 1.000 0.000 0.000 0.000
#> GSM39149 1 0.4939 0.6423 0.740 0.220 0.000 0.040
#> GSM39150 1 0.3278 0.7423 0.864 0.116 0.020 0.000
#> GSM39151 2 0.5530 0.4042 0.360 0.616 0.004 0.020
#> GSM39152 1 0.0707 0.7791 0.980 0.020 0.000 0.000
#> GSM39153 1 0.0000 0.7756 1.000 0.000 0.000 0.000
#> GSM39154 1 0.0707 0.7757 0.980 0.020 0.000 0.000
#> GSM39155 1 0.4961 0.0202 0.552 0.448 0.000 0.000
#> GSM39156 1 0.0921 0.7787 0.972 0.028 0.000 0.000
#> GSM39157 2 0.3907 0.5997 0.232 0.768 0.000 0.000
#> GSM39158 1 0.1398 0.7775 0.956 0.040 0.004 0.000
#> GSM39159 1 0.5244 0.4818 0.600 0.388 0.012 0.000
#> GSM39160 1 0.3597 0.7403 0.836 0.148 0.016 0.000
#> GSM39161 1 0.2596 0.7648 0.908 0.068 0.024 0.000
#> GSM39162 1 0.0000 0.7756 1.000 0.000 0.000 0.000
#> GSM39163 1 0.2760 0.7316 0.872 0.128 0.000 0.000
#> GSM39164 1 0.0000 0.7756 1.000 0.000 0.000 0.000
#> GSM39165 1 0.2011 0.7553 0.920 0.080 0.000 0.000
#> GSM39166 1 0.4225 0.7202 0.792 0.184 0.024 0.000
#> GSM39167 1 0.0000 0.7756 1.000 0.000 0.000 0.000
#> GSM39168 1 0.0000 0.7756 1.000 0.000 0.000 0.000
#> GSM39169 1 0.0000 0.7756 1.000 0.000 0.000 0.000
#> GSM39170 1 0.1902 0.7653 0.932 0.064 0.004 0.000
#> GSM39171 2 0.4981 0.3070 0.464 0.536 0.000 0.000
#> GSM39172 1 0.5416 0.6228 0.740 0.112 0.000 0.148
#> GSM39173 1 0.0657 0.7755 0.984 0.004 0.012 0.000
#> GSM39174 1 0.0921 0.7748 0.972 0.028 0.000 0.000
#> GSM39175 1 0.0188 0.7767 0.996 0.004 0.000 0.000
#> GSM39176 1 0.0336 0.7752 0.992 0.008 0.000 0.000
#> GSM39177 1 0.4661 0.4561 0.652 0.348 0.000 0.000
#> GSM39178 1 0.3946 0.7344 0.812 0.168 0.020 0.000
#> GSM39179 1 0.5442 0.5397 0.672 0.288 0.000 0.040
#> GSM39180 1 0.7843 -0.0393 0.420 0.356 0.004 0.220
#> GSM39181 1 0.4898 0.6398 0.716 0.260 0.024 0.000
#> GSM39182 1 0.3688 0.6858 0.792 0.208 0.000 0.000
#> GSM39183 1 0.4464 0.7075 0.768 0.208 0.024 0.000
#> GSM39184 1 0.4776 0.3545 0.624 0.376 0.000 0.000
#> GSM39185 1 0.5716 0.3602 0.552 0.420 0.028 0.000
#> GSM39186 1 0.4992 -0.0846 0.524 0.476 0.000 0.000
#> GSM39187 1 0.1867 0.7653 0.928 0.072 0.000 0.000
#> GSM39116 2 0.4730 0.3182 0.000 0.636 0.364 0.000
#> GSM39117 4 0.0000 0.7790 0.000 0.000 0.000 1.000
#> GSM39118 2 0.5109 0.4081 0.000 0.736 0.052 0.212
#> GSM39119 4 0.4914 0.5443 0.000 0.208 0.044 0.748
#> GSM39120 2 0.5636 0.4291 0.308 0.648 0.044 0.000
#> GSM39121 2 0.3649 0.6109 0.204 0.796 0.000 0.000
#> GSM39122 2 0.3172 0.6229 0.160 0.840 0.000 0.000
#> GSM39123 4 0.0000 0.7790 0.000 0.000 0.000 1.000
#> GSM39124 2 0.4236 0.6051 0.088 0.824 0.088 0.000
#> GSM39125 2 0.5624 0.5322 0.280 0.668 0.052 0.000
#> GSM39126 2 0.5372 0.2558 0.444 0.544 0.012 0.000
#> GSM39127 2 0.4730 0.3182 0.000 0.636 0.364 0.000
#> GSM39128 3 0.7912 0.1672 0.328 0.312 0.360 0.000
#> GSM39129 4 0.5078 0.7303 0.008 0.072 0.144 0.776
#> GSM39130 4 0.0000 0.7790 0.000 0.000 0.000 1.000
#> GSM39131 2 0.4643 0.3447 0.000 0.656 0.344 0.000
#> GSM39132 2 0.6665 0.1128 0.096 0.544 0.360 0.000
#> GSM39133 4 0.0921 0.7788 0.000 0.028 0.000 0.972
#> GSM39134 4 0.6608 0.4970 0.080 0.192 0.044 0.684
#> GSM39135 2 0.5389 0.3247 0.032 0.660 0.308 0.000
#> GSM39136 2 0.6773 0.0715 0.000 0.532 0.364 0.104
#> GSM39137 2 0.3991 0.5867 0.048 0.832 0.120 0.000
#> GSM39138 4 0.6772 0.4657 0.192 0.076 0.056 0.676
#> GSM39139 2 0.4673 0.3838 0.008 0.700 0.292 0.000
#> GSM39140 2 0.4585 0.4865 0.332 0.668 0.000 0.000
#> GSM39141 2 0.3444 0.6213 0.184 0.816 0.000 0.000
#> GSM39142 2 0.3528 0.6232 0.192 0.808 0.000 0.000
#> GSM39143 2 0.3311 0.6235 0.172 0.828 0.000 0.000
#> GSM39144 3 0.6377 0.1461 0.000 0.256 0.632 0.112
#> GSM39145 2 0.4546 0.4313 0.012 0.732 0.256 0.000
#> GSM39146 2 0.3873 0.4610 0.000 0.772 0.228 0.000
#> GSM39147 2 0.2859 0.5549 0.008 0.880 0.112 0.000
#> GSM39188 1 0.8499 0.2635 0.528 0.096 0.140 0.236
#> GSM39189 1 0.3940 0.7333 0.824 0.152 0.020 0.004
#> GSM39190 2 0.5755 0.5557 0.136 0.752 0.032 0.080
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM39104 2 0.5880 0.3570 0.360 0.560 0.028 0.000 0.052
#> GSM39105 2 0.2997 0.6262 0.148 0.840 0.012 0.000 0.000
#> GSM39106 1 0.5567 0.3625 0.640 0.280 0.028 0.000 0.052
#> GSM39107 2 0.2745 0.6257 0.024 0.896 0.028 0.000 0.052
#> GSM39108 2 0.3430 0.6470 0.220 0.776 0.004 0.000 0.000
#> GSM39109 2 0.2698 0.6324 0.036 0.900 0.028 0.000 0.036
#> GSM39110 1 0.3366 0.4949 0.768 0.232 0.000 0.000 0.000
#> GSM39111 2 0.3010 0.6417 0.172 0.824 0.004 0.000 0.000
#> GSM39112 2 0.3152 0.6262 0.044 0.876 0.028 0.000 0.052
#> GSM39113 2 0.3432 0.6251 0.060 0.860 0.028 0.000 0.052
#> GSM39114 2 0.2848 0.6007 0.000 0.868 0.028 0.000 0.104
#> GSM39115 2 0.3949 0.4677 0.332 0.668 0.000 0.000 0.000
#> GSM39148 1 0.0000 0.7008 1.000 0.000 0.000 0.000 0.000
#> GSM39149 1 0.4629 0.5361 0.724 0.224 0.008 0.044 0.000
#> GSM39150 1 0.3471 0.6288 0.836 0.072 0.092 0.000 0.000
#> GSM39151 2 0.5371 0.4955 0.312 0.624 0.052 0.012 0.000
#> GSM39152 1 0.1836 0.6954 0.932 0.036 0.032 0.000 0.000
#> GSM39153 1 0.0000 0.7008 1.000 0.000 0.000 0.000 0.000
#> GSM39154 1 0.0963 0.6933 0.964 0.036 0.000 0.000 0.000
#> GSM39155 1 0.4235 0.0240 0.576 0.424 0.000 0.000 0.000
#> GSM39156 1 0.0794 0.7039 0.972 0.028 0.000 0.000 0.000
#> GSM39157 2 0.3561 0.6221 0.260 0.740 0.000 0.000 0.000
#> GSM39158 1 0.1469 0.7037 0.948 0.036 0.016 0.000 0.000
#> GSM39159 1 0.5246 0.3544 0.596 0.344 0.060 0.000 0.000
#> GSM39160 1 0.3906 0.6271 0.804 0.112 0.084 0.000 0.000
#> GSM39161 1 0.3647 0.6035 0.816 0.052 0.132 0.000 0.000
#> GSM39162 1 0.0000 0.7008 1.000 0.000 0.000 0.000 0.000
#> GSM39163 1 0.2471 0.6467 0.864 0.136 0.000 0.000 0.000
#> GSM39164 1 0.0000 0.7008 1.000 0.000 0.000 0.000 0.000
#> GSM39165 1 0.0955 0.6998 0.968 0.028 0.004 0.000 0.000
#> GSM39166 1 0.4926 0.5587 0.716 0.152 0.132 0.000 0.000
#> GSM39167 1 0.0000 0.7008 1.000 0.000 0.000 0.000 0.000
#> GSM39168 1 0.0000 0.7008 1.000 0.000 0.000 0.000 0.000
#> GSM39169 1 0.0000 0.7008 1.000 0.000 0.000 0.000 0.000
#> GSM39170 1 0.1915 0.6851 0.928 0.040 0.032 0.000 0.000
#> GSM39171 2 0.4449 0.2600 0.484 0.512 0.004 0.000 0.000
#> GSM39172 1 0.5199 0.4443 0.720 0.124 0.016 0.140 0.000
#> GSM39173 1 0.0912 0.6977 0.972 0.000 0.016 0.000 0.012
#> GSM39174 1 0.0794 0.6996 0.972 0.028 0.000 0.000 0.000
#> GSM39175 1 0.0162 0.7023 0.996 0.004 0.000 0.000 0.000
#> GSM39176 1 0.0162 0.7005 0.996 0.000 0.004 0.000 0.000
#> GSM39177 1 0.4511 0.3530 0.628 0.356 0.016 0.000 0.000
#> GSM39178 1 0.4766 0.5723 0.732 0.136 0.132 0.000 0.000
#> GSM39179 1 0.4815 0.4081 0.660 0.304 0.008 0.028 0.000
#> GSM39180 1 0.6751 -0.1870 0.424 0.352 0.004 0.220 0.000
#> GSM39181 1 0.5365 0.4953 0.664 0.204 0.132 0.000 0.000
#> GSM39182 1 0.3366 0.5899 0.784 0.212 0.004 0.000 0.000
#> GSM39183 1 0.5109 0.5436 0.696 0.172 0.132 0.000 0.000
#> GSM39184 1 0.4114 0.3288 0.624 0.376 0.000 0.000 0.000
#> GSM39185 1 0.6339 0.1947 0.484 0.368 0.144 0.000 0.004
#> GSM39186 1 0.4443 -0.0935 0.524 0.472 0.004 0.000 0.000
#> GSM39187 1 0.1671 0.6873 0.924 0.076 0.000 0.000 0.000
#> GSM39116 2 0.4299 0.4566 0.000 0.608 0.004 0.000 0.388
#> GSM39117 4 0.0000 0.7111 0.000 0.000 0.000 1.000 0.000
#> GSM39118 2 0.4400 0.5231 0.000 0.736 0.000 0.212 0.052
#> GSM39119 4 0.4270 0.4997 0.000 0.204 0.000 0.748 0.048
#> GSM39120 2 0.5340 0.4691 0.336 0.608 0.012 0.000 0.044
#> GSM39121 2 0.3366 0.6359 0.232 0.768 0.000 0.000 0.000
#> GSM39122 2 0.2890 0.6600 0.160 0.836 0.000 0.000 0.004
#> GSM39123 4 0.0000 0.7111 0.000 0.000 0.000 1.000 0.000
#> GSM39124 2 0.4166 0.6567 0.088 0.792 0.004 0.000 0.116
#> GSM39125 2 0.5218 0.5956 0.280 0.656 0.012 0.000 0.052
#> GSM39126 2 0.5094 0.2270 0.468 0.504 0.016 0.000 0.012
#> GSM39127 2 0.4161 0.4540 0.000 0.608 0.000 0.000 0.392
#> GSM39128 5 0.6775 -0.1799 0.328 0.284 0.000 0.000 0.388
#> GSM39129 4 0.4536 0.5493 0.000 0.008 0.344 0.640 0.008
#> GSM39130 4 0.0000 0.7111 0.000 0.000 0.000 1.000 0.000
#> GSM39131 2 0.4101 0.4748 0.000 0.628 0.000 0.000 0.372
#> GSM39132 2 0.5843 0.3059 0.100 0.512 0.000 0.000 0.388
#> GSM39133 4 0.0794 0.7086 0.000 0.028 0.000 0.972 0.000
#> GSM39134 4 0.5739 0.3905 0.076 0.188 0.000 0.684 0.052
#> GSM39135 2 0.4763 0.4638 0.032 0.632 0.000 0.000 0.336
#> GSM39136 2 0.5889 0.2835 0.000 0.504 0.000 0.104 0.392
#> GSM39137 2 0.3639 0.6435 0.044 0.812 0.000 0.000 0.144
#> GSM39138 4 0.5916 0.1253 0.188 0.080 0.000 0.672 0.060
#> GSM39139 2 0.4165 0.5039 0.008 0.672 0.000 0.000 0.320
#> GSM39140 2 0.4015 0.5193 0.348 0.652 0.000 0.000 0.000
#> GSM39141 2 0.3003 0.6541 0.188 0.812 0.000 0.000 0.000
#> GSM39142 2 0.3109 0.6549 0.200 0.800 0.000 0.000 0.000
#> GSM39143 2 0.2929 0.6560 0.180 0.820 0.000 0.000 0.000
#> GSM39144 5 0.2663 -0.1104 0.000 0.048 0.008 0.048 0.896
#> GSM39145 2 0.4086 0.5393 0.012 0.704 0.000 0.000 0.284
#> GSM39146 2 0.3534 0.5595 0.000 0.744 0.000 0.000 0.256
#> GSM39147 2 0.2997 0.6218 0.012 0.840 0.000 0.000 0.148
#> GSM39188 3 0.7518 0.0000 0.288 0.072 0.464 0.176 0.000
#> GSM39189 1 0.4487 0.5722 0.756 0.104 0.140 0.000 0.000
#> GSM39190 2 0.5779 0.5733 0.072 0.708 0.164 0.024 0.032
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM39104 2 0.5540 0.3663 0.356 0.552 0.024 0.000 NA 0.060
#> GSM39105 2 0.2695 0.6301 0.144 0.844 0.008 0.000 NA 0.000
#> GSM39106 1 0.5146 0.5289 0.644 0.268 0.020 0.000 NA 0.060
#> GSM39107 2 0.2683 0.6220 0.020 0.888 0.024 0.000 NA 0.060
#> GSM39108 2 0.3081 0.6424 0.220 0.776 0.000 0.000 NA 0.000
#> GSM39109 2 0.2666 0.6288 0.032 0.892 0.024 0.000 NA 0.044
#> GSM39110 1 0.3050 0.5669 0.764 0.236 0.000 0.000 NA 0.000
#> GSM39111 2 0.2703 0.6416 0.172 0.824 0.004 0.000 NA 0.000
#> GSM39112 2 0.2992 0.6224 0.036 0.872 0.024 0.000 NA 0.060
#> GSM39113 2 0.3320 0.6214 0.056 0.852 0.024 0.000 NA 0.060
#> GSM39114 2 0.2781 0.6021 0.000 0.860 0.024 0.000 NA 0.108
#> GSM39115 2 0.3499 0.4919 0.320 0.680 0.000 0.000 NA 0.000
#> GSM39148 1 0.0000 0.7696 1.000 0.000 0.000 0.000 NA 0.000
#> GSM39149 1 0.4356 0.6426 0.728 0.212 0.008 0.040 NA 0.000
#> GSM39150 1 0.3324 0.7245 0.840 0.060 0.080 0.000 NA 0.000
#> GSM39151 2 0.6390 0.3831 0.284 0.544 0.032 0.012 NA 0.008
#> GSM39152 1 0.1829 0.7674 0.928 0.036 0.028 0.000 NA 0.000
#> GSM39153 1 0.0000 0.7696 1.000 0.000 0.000 0.000 NA 0.000
#> GSM39154 1 0.0865 0.7661 0.964 0.036 0.000 0.000 NA 0.000
#> GSM39155 1 0.3804 0.0281 0.576 0.424 0.000 0.000 NA 0.000
#> GSM39156 1 0.0632 0.7719 0.976 0.024 0.000 0.000 NA 0.000
#> GSM39157 2 0.3151 0.6233 0.252 0.748 0.000 0.000 NA 0.000
#> GSM39158 1 0.1245 0.7724 0.952 0.032 0.016 0.000 NA 0.000
#> GSM39159 1 0.4805 0.5236 0.608 0.332 0.052 0.000 NA 0.000
#> GSM39160 1 0.3817 0.7182 0.796 0.104 0.088 0.000 NA 0.000
#> GSM39161 1 0.3720 0.7039 0.812 0.044 0.108 0.000 NA 0.000
#> GSM39162 1 0.0000 0.7696 1.000 0.000 0.000 0.000 NA 0.000
#> GSM39163 1 0.2260 0.7155 0.860 0.140 0.000 0.000 NA 0.000
#> GSM39164 1 0.0000 0.7696 1.000 0.000 0.000 0.000 NA 0.000
#> GSM39165 1 0.0858 0.7700 0.968 0.028 0.004 0.000 NA 0.000
#> GSM39166 1 0.4881 0.6711 0.716 0.140 0.108 0.000 NA 0.000
#> GSM39167 1 0.0000 0.7696 1.000 0.000 0.000 0.000 NA 0.000
#> GSM39168 1 0.0000 0.7696 1.000 0.000 0.000 0.000 NA 0.000
#> GSM39169 1 0.0000 0.7696 1.000 0.000 0.000 0.000 NA 0.000
#> GSM39170 1 0.1636 0.7655 0.936 0.036 0.024 0.000 NA 0.000
#> GSM39171 2 0.3986 0.3104 0.464 0.532 0.004 0.000 NA 0.000
#> GSM39172 1 0.4640 0.6341 0.728 0.116 0.012 0.140 NA 0.000
#> GSM39173 1 0.0881 0.7690 0.972 0.000 0.008 0.000 NA 0.012
#> GSM39174 1 0.0713 0.7676 0.972 0.028 0.000 0.000 NA 0.000
#> GSM39175 1 0.0146 0.7707 0.996 0.004 0.000 0.000 NA 0.000
#> GSM39176 1 0.0146 0.7696 0.996 0.000 0.004 0.000 NA 0.000
#> GSM39177 1 0.4266 0.4554 0.628 0.348 0.016 0.000 NA 0.000
#> GSM39178 1 0.4729 0.6830 0.732 0.124 0.108 0.000 NA 0.000
#> GSM39179 1 0.5342 0.5103 0.628 0.288 0.016 0.024 NA 0.004
#> GSM39180 1 0.6182 0.0944 0.428 0.344 0.004 0.220 NA 0.000
#> GSM39181 1 0.5349 0.6100 0.656 0.200 0.108 0.000 NA 0.000
#> GSM39182 1 0.2871 0.6934 0.804 0.192 0.004 0.000 NA 0.000
#> GSM39183 1 0.5054 0.6610 0.696 0.160 0.108 0.000 NA 0.000
#> GSM39184 1 0.3647 0.3778 0.640 0.360 0.000 0.000 NA 0.000
#> GSM39185 1 0.6255 0.3267 0.480 0.360 0.116 0.000 NA 0.004
#> GSM39186 1 0.3993 -0.1158 0.520 0.476 0.004 0.000 NA 0.000
#> GSM39187 1 0.1501 0.7557 0.924 0.076 0.000 0.000 NA 0.000
#> GSM39116 2 0.3915 0.4510 0.000 0.584 0.004 0.000 NA 0.412
#> GSM39117 4 0.0000 0.6330 0.000 0.000 0.000 1.000 NA 0.000
#> GSM39118 2 0.4176 0.5222 0.000 0.720 0.000 0.212 NA 0.068
#> GSM39119 4 0.3896 0.4584 0.000 0.204 0.000 0.744 NA 0.052
#> GSM39120 2 0.4855 0.4799 0.320 0.616 0.012 0.000 NA 0.052
#> GSM39121 2 0.2996 0.6353 0.228 0.772 0.000 0.000 NA 0.000
#> GSM39122 2 0.2520 0.6583 0.152 0.844 0.000 0.000 NA 0.004
#> GSM39123 4 0.0000 0.6330 0.000 0.000 0.000 1.000 NA 0.000
#> GSM39124 2 0.3598 0.6558 0.080 0.804 0.004 0.000 NA 0.112
#> GSM39125 2 0.4744 0.6011 0.264 0.668 0.012 0.000 NA 0.052
#> GSM39126 2 0.4713 0.2737 0.448 0.520 0.012 0.000 NA 0.012
#> GSM39127 2 0.3789 0.4484 0.000 0.584 0.000 0.000 NA 0.416
#> GSM39128 6 0.6047 -0.0881 0.316 0.272 0.000 0.000 NA 0.412
#> GSM39129 4 0.3765 0.2927 0.000 0.000 0.000 0.596 NA 0.000
#> GSM39130 4 0.0000 0.6330 0.000 0.000 0.000 1.000 NA 0.000
#> GSM39131 2 0.3747 0.4685 0.000 0.604 0.000 0.000 NA 0.396
#> GSM39132 2 0.5278 0.2960 0.100 0.488 0.000 0.000 NA 0.412
#> GSM39133 4 0.0713 0.6348 0.000 0.028 0.000 0.972 NA 0.000
#> GSM39134 4 0.5155 0.4462 0.076 0.188 0.000 0.684 NA 0.052
#> GSM39135 2 0.4332 0.4628 0.032 0.616 0.000 0.000 NA 0.352
#> GSM39136 2 0.5171 0.3190 0.000 0.496 0.000 0.088 NA 0.416
#> GSM39137 2 0.3163 0.6428 0.040 0.820 0.000 0.000 NA 0.140
#> GSM39138 4 0.5562 0.3721 0.188 0.076 0.008 0.664 NA 0.064
#> GSM39139 2 0.3819 0.4982 0.008 0.652 0.000 0.000 NA 0.340
#> GSM39140 2 0.3607 0.5131 0.348 0.652 0.000 0.000 NA 0.000
#> GSM39141 2 0.2631 0.6527 0.180 0.820 0.000 0.000 NA 0.000
#> GSM39142 2 0.2697 0.6541 0.188 0.812 0.000 0.000 NA 0.000
#> GSM39143 2 0.2562 0.6544 0.172 0.828 0.000 0.000 NA 0.000
#> GSM39144 6 0.2728 -0.3997 0.000 0.012 0.004 0.024 NA 0.876
#> GSM39145 2 0.3729 0.5409 0.012 0.692 0.000 0.000 NA 0.296
#> GSM39146 2 0.3244 0.5605 0.000 0.732 0.000 0.000 NA 0.268
#> GSM39147 2 0.2593 0.6231 0.008 0.844 0.000 0.000 NA 0.148
#> GSM39188 3 0.2664 0.0000 0.016 0.000 0.848 0.136 NA 0.000
#> GSM39189 1 0.4439 0.6873 0.760 0.084 0.116 0.000 NA 0.000
#> GSM39190 2 0.5437 0.3842 0.036 0.564 0.036 0.004 NA 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) other(p) protocol(p) k
#> SD:pam 87 0.34127 3.14e-04 2.23e-05 2
#> SD:pam 71 0.02270 1.94e-11 1.88e-08 3
#> SD:pam 58 0.00206 1.88e-09 4.74e-10 4
#> SD:pam 57 0.00166 6.53e-10 6.63e-09 5
#> SD:pam 60 0.00747 3.99e-09 2.10e-08 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 8353 rows and 87 columns.
#> Top rows (835, 1670, 2506, 3341, 4176) 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 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 0.573 0.737 0.884 0.4647 0.495 0.495
#> 3 3 0.524 0.654 0.820 0.3556 0.568 0.324
#> 4 4 0.523 0.606 0.775 0.0854 0.737 0.436
#> 5 5 0.600 0.532 0.694 0.0722 0.770 0.433
#> 6 6 0.709 0.698 0.811 0.0130 0.836 0.540
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
#> GSM39104 1 0.0000 0.934 1.000 0.000
#> GSM39105 1 0.0000 0.934 1.000 0.000
#> GSM39106 1 0.0000 0.934 1.000 0.000
#> GSM39107 1 0.9944 -0.110 0.544 0.456
#> GSM39108 1 0.0000 0.934 1.000 0.000
#> GSM39109 2 0.9896 0.414 0.440 0.560
#> GSM39110 1 0.7056 0.682 0.808 0.192
#> GSM39111 1 0.3114 0.883 0.944 0.056
#> GSM39112 1 0.4939 0.831 0.892 0.108
#> GSM39113 1 0.9427 0.291 0.640 0.360
#> GSM39114 2 0.0000 0.771 0.000 1.000
#> GSM39115 1 0.0000 0.934 1.000 0.000
#> GSM39148 1 0.0000 0.934 1.000 0.000
#> GSM39149 2 0.9988 0.358 0.480 0.520
#> GSM39150 1 0.0000 0.934 1.000 0.000
#> GSM39151 2 0.9988 0.358 0.480 0.520
#> GSM39152 2 0.9988 0.358 0.480 0.520
#> GSM39153 1 0.0000 0.934 1.000 0.000
#> GSM39154 1 0.0000 0.934 1.000 0.000
#> GSM39155 1 0.0000 0.934 1.000 0.000
#> GSM39156 1 0.0000 0.934 1.000 0.000
#> GSM39157 1 0.0000 0.934 1.000 0.000
#> GSM39158 1 0.0000 0.934 1.000 0.000
#> GSM39159 1 0.9580 0.158 0.620 0.380
#> GSM39160 1 0.2043 0.907 0.968 0.032
#> GSM39161 2 0.9988 0.358 0.480 0.520
#> GSM39162 1 0.0000 0.934 1.000 0.000
#> GSM39163 1 0.0000 0.934 1.000 0.000
#> GSM39164 1 0.0000 0.934 1.000 0.000
#> GSM39165 1 0.7745 0.614 0.772 0.228
#> GSM39166 1 0.0000 0.934 1.000 0.000
#> GSM39167 1 0.0000 0.934 1.000 0.000
#> GSM39168 1 0.0000 0.934 1.000 0.000
#> GSM39169 1 0.0000 0.934 1.000 0.000
#> GSM39170 1 0.0000 0.934 1.000 0.000
#> GSM39171 1 0.0000 0.934 1.000 0.000
#> GSM39172 2 0.9988 0.358 0.480 0.520
#> GSM39173 2 0.9988 0.358 0.480 0.520
#> GSM39174 1 0.0000 0.934 1.000 0.000
#> GSM39175 1 0.0000 0.934 1.000 0.000
#> GSM39176 1 0.0000 0.934 1.000 0.000
#> GSM39177 2 0.9988 0.358 0.480 0.520
#> GSM39178 1 0.8386 0.522 0.732 0.268
#> GSM39179 2 0.9988 0.358 0.480 0.520
#> GSM39180 2 0.9988 0.358 0.480 0.520
#> GSM39181 1 0.4298 0.846 0.912 0.088
#> GSM39182 2 0.9988 0.358 0.480 0.520
#> GSM39183 1 0.2236 0.903 0.964 0.036
#> GSM39184 1 0.0000 0.934 1.000 0.000
#> GSM39185 2 0.9988 0.358 0.480 0.520
#> GSM39186 1 0.0000 0.934 1.000 0.000
#> GSM39187 1 0.0000 0.934 1.000 0.000
#> GSM39116 2 0.0000 0.771 0.000 1.000
#> GSM39117 2 0.0000 0.771 0.000 1.000
#> GSM39118 2 0.0000 0.771 0.000 1.000
#> GSM39119 2 0.0000 0.771 0.000 1.000
#> GSM39120 1 0.0000 0.934 1.000 0.000
#> GSM39121 2 0.7602 0.655 0.220 0.780
#> GSM39122 2 0.7056 0.675 0.192 0.808
#> GSM39123 2 0.0000 0.771 0.000 1.000
#> GSM39124 2 0.0000 0.771 0.000 1.000
#> GSM39125 1 0.0000 0.934 1.000 0.000
#> GSM39126 2 0.8207 0.625 0.256 0.744
#> GSM39127 2 0.0000 0.771 0.000 1.000
#> GSM39128 2 0.0376 0.770 0.004 0.996
#> GSM39129 2 0.0000 0.771 0.000 1.000
#> GSM39130 2 0.0000 0.771 0.000 1.000
#> GSM39131 2 0.0376 0.770 0.004 0.996
#> GSM39132 2 0.0000 0.771 0.000 1.000
#> GSM39133 2 0.0000 0.771 0.000 1.000
#> GSM39134 2 0.0000 0.771 0.000 1.000
#> GSM39135 2 0.0000 0.771 0.000 1.000
#> GSM39136 2 0.0000 0.771 0.000 1.000
#> GSM39137 2 0.2236 0.758 0.036 0.964
#> GSM39138 2 0.0000 0.771 0.000 1.000
#> GSM39139 2 0.0000 0.771 0.000 1.000
#> GSM39140 1 0.0672 0.927 0.992 0.008
#> GSM39141 1 0.0376 0.931 0.996 0.004
#> GSM39142 1 0.0000 0.934 1.000 0.000
#> GSM39143 1 0.0376 0.931 0.996 0.004
#> GSM39144 2 0.0000 0.771 0.000 1.000
#> GSM39145 2 0.0000 0.771 0.000 1.000
#> GSM39146 2 0.0000 0.771 0.000 1.000
#> GSM39147 2 0.0000 0.771 0.000 1.000
#> GSM39188 2 0.9988 0.358 0.480 0.520
#> GSM39189 2 0.9988 0.358 0.480 0.520
#> GSM39190 2 0.9988 0.358 0.480 0.520
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM39104 3 0.4974 0.7687 0.236 0.000 0.764
#> GSM39105 1 0.6126 0.1269 0.600 0.000 0.400
#> GSM39106 1 0.2066 0.6935 0.940 0.000 0.060
#> GSM39107 1 0.3461 0.6783 0.900 0.076 0.024
#> GSM39108 1 0.2625 0.6828 0.916 0.000 0.084
#> GSM39109 1 0.4209 0.6620 0.856 0.016 0.128
#> GSM39110 1 0.3038 0.6722 0.896 0.000 0.104
#> GSM39111 3 0.4654 0.7821 0.208 0.000 0.792
#> GSM39112 1 0.1585 0.7011 0.964 0.008 0.028
#> GSM39113 1 0.3590 0.6823 0.896 0.076 0.028
#> GSM39114 1 0.5722 0.4834 0.704 0.292 0.004
#> GSM39115 3 0.5465 0.7468 0.288 0.000 0.712
#> GSM39148 1 0.6062 0.1091 0.616 0.000 0.384
#> GSM39149 3 0.0592 0.7715 0.000 0.012 0.988
#> GSM39150 3 0.4654 0.7821 0.208 0.000 0.792
#> GSM39151 3 0.0592 0.7715 0.000 0.012 0.988
#> GSM39152 3 0.1015 0.7748 0.008 0.012 0.980
#> GSM39153 3 0.6235 0.4826 0.436 0.000 0.564
#> GSM39154 3 0.6126 0.5671 0.400 0.000 0.600
#> GSM39155 1 0.6244 -0.1041 0.560 0.000 0.440
#> GSM39156 1 0.1643 0.6913 0.956 0.000 0.044
#> GSM39157 1 0.6295 -0.2360 0.528 0.000 0.472
#> GSM39158 3 0.5291 0.7618 0.268 0.000 0.732
#> GSM39159 3 0.4172 0.7964 0.156 0.004 0.840
#> GSM39160 3 0.4555 0.7847 0.200 0.000 0.800
#> GSM39161 3 0.2599 0.7748 0.052 0.016 0.932
#> GSM39162 1 0.5465 0.3644 0.712 0.000 0.288
#> GSM39163 3 0.6111 0.5755 0.396 0.000 0.604
#> GSM39164 1 0.6299 -0.2457 0.524 0.000 0.476
#> GSM39165 3 0.4178 0.7955 0.172 0.000 0.828
#> GSM39166 3 0.5216 0.7630 0.260 0.000 0.740
#> GSM39167 3 0.5988 0.6276 0.368 0.000 0.632
#> GSM39168 1 0.5905 0.2161 0.648 0.000 0.352
#> GSM39169 1 0.6309 -0.3306 0.500 0.000 0.500
#> GSM39170 3 0.5363 0.7562 0.276 0.000 0.724
#> GSM39171 3 0.4842 0.7775 0.224 0.000 0.776
#> GSM39172 3 0.0983 0.7672 0.004 0.016 0.980
#> GSM39173 3 0.0592 0.7715 0.000 0.012 0.988
#> GSM39174 1 0.6126 0.0579 0.600 0.000 0.400
#> GSM39175 3 0.5327 0.7594 0.272 0.000 0.728
#> GSM39176 3 0.5810 0.6833 0.336 0.000 0.664
#> GSM39177 3 0.0592 0.7715 0.000 0.012 0.988
#> GSM39178 3 0.2711 0.7936 0.088 0.000 0.912
#> GSM39179 3 0.0592 0.7715 0.000 0.012 0.988
#> GSM39180 3 0.1636 0.7702 0.020 0.016 0.964
#> GSM39181 3 0.4062 0.7960 0.164 0.000 0.836
#> GSM39182 3 0.2703 0.7752 0.056 0.016 0.928
#> GSM39183 3 0.4291 0.7943 0.180 0.000 0.820
#> GSM39184 3 0.5465 0.7440 0.288 0.000 0.712
#> GSM39185 3 0.2599 0.7748 0.052 0.016 0.932
#> GSM39186 1 0.5810 0.2932 0.664 0.000 0.336
#> GSM39187 3 0.6308 0.3165 0.492 0.000 0.508
#> GSM39116 2 0.3009 0.9155 0.028 0.920 0.052
#> GSM39117 2 0.3715 0.9010 0.004 0.868 0.128
#> GSM39118 2 0.2261 0.9318 0.000 0.932 0.068
#> GSM39119 2 0.2537 0.9279 0.000 0.920 0.080
#> GSM39120 1 0.1267 0.7007 0.972 0.004 0.024
#> GSM39121 1 0.4605 0.5882 0.796 0.204 0.000
#> GSM39122 1 0.5016 0.5508 0.760 0.240 0.000
#> GSM39123 2 0.3715 0.9010 0.004 0.868 0.128
#> GSM39124 1 0.5529 0.4775 0.704 0.296 0.000
#> GSM39125 1 0.1129 0.7005 0.976 0.004 0.020
#> GSM39126 1 0.4452 0.6002 0.808 0.192 0.000
#> GSM39127 1 0.5733 0.4372 0.676 0.324 0.000
#> GSM39128 1 0.5560 0.4732 0.700 0.300 0.000
#> GSM39129 2 0.2356 0.9307 0.000 0.928 0.072
#> GSM39130 2 0.3715 0.9010 0.004 0.868 0.128
#> GSM39131 1 0.5560 0.4732 0.700 0.300 0.000
#> GSM39132 2 0.2959 0.8186 0.100 0.900 0.000
#> GSM39133 2 0.3644 0.9036 0.004 0.872 0.124
#> GSM39134 2 0.2261 0.9318 0.000 0.932 0.068
#> GSM39135 2 0.2443 0.8970 0.032 0.940 0.028
#> GSM39136 2 0.3406 0.9229 0.028 0.904 0.068
#> GSM39137 1 0.5529 0.4779 0.704 0.296 0.000
#> GSM39138 2 0.2261 0.9318 0.000 0.932 0.068
#> GSM39139 2 0.2261 0.9318 0.000 0.932 0.068
#> GSM39140 1 0.0892 0.6997 0.980 0.000 0.020
#> GSM39141 1 0.0892 0.6997 0.980 0.000 0.020
#> GSM39142 1 0.0892 0.6997 0.980 0.000 0.020
#> GSM39143 1 0.0892 0.6997 0.980 0.000 0.020
#> GSM39144 2 0.2261 0.9318 0.000 0.932 0.068
#> GSM39145 2 0.2261 0.9318 0.000 0.932 0.068
#> GSM39146 2 0.5953 0.5304 0.280 0.708 0.012
#> GSM39147 2 0.3551 0.7863 0.132 0.868 0.000
#> GSM39188 3 0.0592 0.7715 0.000 0.012 0.988
#> GSM39189 3 0.0592 0.7715 0.000 0.012 0.988
#> GSM39190 3 0.0592 0.7715 0.000 0.012 0.988
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM39104 1 0.3172 0.68973 0.840 0.000 0.160 0.000
#> GSM39105 1 0.1994 0.76793 0.936 0.004 0.052 0.008
#> GSM39106 1 0.5229 0.72165 0.772 0.156 0.048 0.024
#> GSM39107 1 0.6107 0.58582 0.652 0.288 0.024 0.036
#> GSM39108 1 0.4742 0.73440 0.800 0.140 0.044 0.016
#> GSM39109 1 0.6923 0.60543 0.636 0.244 0.084 0.036
#> GSM39110 1 0.5146 0.74020 0.784 0.120 0.080 0.016
#> GSM39111 1 0.4857 0.47644 0.668 0.008 0.324 0.000
#> GSM39112 1 0.5909 0.61232 0.672 0.272 0.020 0.036
#> GSM39113 1 0.5996 0.56789 0.644 0.304 0.016 0.036
#> GSM39114 2 0.3875 0.48205 0.068 0.852 0.004 0.076
#> GSM39115 1 0.2530 0.72238 0.888 0.000 0.112 0.000
#> GSM39148 1 0.1284 0.76875 0.964 0.012 0.024 0.000
#> GSM39149 3 0.1302 0.82987 0.044 0.000 0.956 0.000
#> GSM39150 3 0.4866 0.56601 0.404 0.000 0.596 0.000
#> GSM39151 3 0.1302 0.82987 0.044 0.000 0.956 0.000
#> GSM39152 3 0.1867 0.83129 0.072 0.000 0.928 0.000
#> GSM39153 1 0.1867 0.75367 0.928 0.000 0.072 0.000
#> GSM39154 1 0.2281 0.73504 0.904 0.000 0.096 0.000
#> GSM39155 1 0.1557 0.75999 0.944 0.000 0.056 0.000
#> GSM39156 1 0.4781 0.72592 0.788 0.160 0.040 0.012
#> GSM39157 1 0.1022 0.76669 0.968 0.000 0.032 0.000
#> GSM39158 1 0.4564 0.29992 0.672 0.000 0.328 0.000
#> GSM39159 3 0.4193 0.72537 0.268 0.000 0.732 0.000
#> GSM39160 3 0.4730 0.63761 0.364 0.000 0.636 0.000
#> GSM39161 3 0.2610 0.82220 0.088 0.000 0.900 0.012
#> GSM39162 1 0.1722 0.76800 0.944 0.048 0.008 0.000
#> GSM39163 1 0.1716 0.75632 0.936 0.000 0.064 0.000
#> GSM39164 1 0.1211 0.76552 0.960 0.000 0.040 0.000
#> GSM39165 3 0.4543 0.67982 0.324 0.000 0.676 0.000
#> GSM39166 3 0.4855 0.58265 0.400 0.000 0.600 0.000
#> GSM39167 1 0.1940 0.74911 0.924 0.000 0.076 0.000
#> GSM39168 1 0.1284 0.76910 0.964 0.024 0.012 0.000
#> GSM39169 1 0.1637 0.75827 0.940 0.000 0.060 0.000
#> GSM39170 1 0.2921 0.68888 0.860 0.000 0.140 0.000
#> GSM39171 1 0.4888 -0.00631 0.588 0.000 0.412 0.000
#> GSM39172 3 0.0657 0.78902 0.000 0.004 0.984 0.012
#> GSM39173 3 0.1022 0.82354 0.032 0.000 0.968 0.000
#> GSM39174 1 0.1118 0.76608 0.964 0.000 0.036 0.000
#> GSM39175 1 0.3356 0.64007 0.824 0.000 0.176 0.000
#> GSM39176 1 0.1940 0.74911 0.924 0.000 0.076 0.000
#> GSM39177 3 0.1389 0.83058 0.048 0.000 0.952 0.000
#> GSM39178 3 0.4331 0.72499 0.288 0.000 0.712 0.000
#> GSM39179 3 0.1302 0.82987 0.044 0.000 0.956 0.000
#> GSM39180 3 0.0844 0.79139 0.004 0.004 0.980 0.012
#> GSM39181 3 0.5110 0.66345 0.352 0.000 0.636 0.012
#> GSM39182 3 0.3617 0.80307 0.108 0.020 0.860 0.012
#> GSM39183 3 0.4679 0.65729 0.352 0.000 0.648 0.000
#> GSM39184 1 0.2469 0.72727 0.892 0.000 0.108 0.000
#> GSM39185 3 0.1767 0.81215 0.044 0.000 0.944 0.012
#> GSM39186 1 0.1824 0.76507 0.936 0.004 0.060 0.000
#> GSM39187 1 0.1389 0.76306 0.952 0.000 0.048 0.000
#> GSM39116 2 0.3726 0.38350 0.000 0.788 0.000 0.212
#> GSM39117 4 0.1474 0.62547 0.000 0.052 0.000 0.948
#> GSM39118 2 0.5167 -0.51488 0.000 0.508 0.004 0.488
#> GSM39119 4 0.4730 0.58324 0.000 0.364 0.000 0.636
#> GSM39120 1 0.5430 0.70376 0.752 0.180 0.032 0.036
#> GSM39121 1 0.5678 0.26662 0.500 0.480 0.004 0.016
#> GSM39122 1 0.5510 0.26959 0.504 0.480 0.000 0.016
#> GSM39123 4 0.1474 0.62547 0.000 0.052 0.000 0.948
#> GSM39124 2 0.4706 0.46124 0.140 0.788 0.000 0.072
#> GSM39125 1 0.5478 0.70383 0.752 0.176 0.036 0.036
#> GSM39126 1 0.5859 0.28507 0.504 0.468 0.004 0.024
#> GSM39127 2 0.2081 0.50456 0.000 0.916 0.000 0.084
#> GSM39128 2 0.5267 0.43750 0.184 0.740 0.000 0.076
#> GSM39129 4 0.5161 0.51748 0.000 0.476 0.004 0.520
#> GSM39130 4 0.1474 0.62547 0.000 0.052 0.000 0.948
#> GSM39131 2 0.5307 0.43605 0.188 0.736 0.000 0.076
#> GSM39132 2 0.2868 0.48461 0.000 0.864 0.000 0.136
#> GSM39133 4 0.1637 0.62644 0.000 0.060 0.000 0.940
#> GSM39134 4 0.4998 0.49918 0.000 0.488 0.000 0.512
#> GSM39135 2 0.3528 0.41213 0.000 0.808 0.000 0.192
#> GSM39136 2 0.4134 0.25694 0.000 0.740 0.000 0.260
#> GSM39137 2 0.5924 -0.03470 0.404 0.556 0.000 0.040
#> GSM39138 4 0.4992 0.51949 0.000 0.476 0.000 0.524
#> GSM39139 2 0.4925 -0.31156 0.000 0.572 0.000 0.428
#> GSM39140 1 0.5716 0.68461 0.728 0.200 0.036 0.036
#> GSM39141 1 0.5400 0.69396 0.744 0.196 0.024 0.036
#> GSM39142 1 0.5065 0.71407 0.772 0.172 0.024 0.032
#> GSM39143 1 0.5548 0.67864 0.728 0.212 0.024 0.036
#> GSM39144 4 0.5161 0.51748 0.000 0.476 0.004 0.520
#> GSM39145 2 0.4304 0.22824 0.000 0.716 0.000 0.284
#> GSM39146 2 0.3569 0.47713 0.000 0.804 0.000 0.196
#> GSM39147 2 0.2973 0.48633 0.000 0.856 0.000 0.144
#> GSM39188 3 0.1118 0.82724 0.036 0.000 0.964 0.000
#> GSM39189 3 0.1211 0.82888 0.040 0.000 0.960 0.000
#> GSM39190 3 0.1118 0.82724 0.036 0.000 0.964 0.000
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM39104 1 0.2325 0.7009 0.904 0.000 0.028 0.000 0.068
#> GSM39105 1 0.1251 0.6983 0.956 0.000 0.008 0.000 0.036
#> GSM39106 5 0.4811 0.6752 0.452 0.000 0.020 0.000 0.528
#> GSM39107 5 0.5836 0.7349 0.348 0.076 0.012 0.000 0.564
#> GSM39108 1 0.5014 -0.4995 0.536 0.000 0.032 0.000 0.432
#> GSM39109 5 0.6631 0.6663 0.340 0.048 0.060 0.012 0.540
#> GSM39110 1 0.5393 -0.5268 0.504 0.000 0.056 0.000 0.440
#> GSM39111 1 0.5435 0.5359 0.668 0.000 0.124 0.004 0.204
#> GSM39112 5 0.5210 0.7486 0.384 0.028 0.012 0.000 0.576
#> GSM39113 5 0.5778 0.7169 0.324 0.096 0.004 0.000 0.576
#> GSM39114 2 0.4691 0.4554 0.004 0.636 0.000 0.020 0.340
#> GSM39115 1 0.1026 0.7157 0.968 0.000 0.004 0.004 0.024
#> GSM39148 1 0.1502 0.6690 0.940 0.000 0.004 0.000 0.056
#> GSM39149 3 0.0963 0.7852 0.036 0.000 0.964 0.000 0.000
#> GSM39150 1 0.5796 0.5403 0.624 0.000 0.092 0.016 0.268
#> GSM39151 3 0.1124 0.7848 0.036 0.000 0.960 0.000 0.004
#> GSM39152 3 0.4901 0.7101 0.084 0.000 0.716 0.004 0.196
#> GSM39153 1 0.0324 0.7145 0.992 0.000 0.004 0.000 0.004
#> GSM39154 1 0.0162 0.7136 0.996 0.000 0.000 0.000 0.004
#> GSM39155 1 0.0162 0.7098 0.996 0.000 0.000 0.000 0.004
#> GSM39156 1 0.4708 -0.5250 0.548 0.000 0.016 0.000 0.436
#> GSM39157 1 0.0404 0.7041 0.988 0.000 0.000 0.000 0.012
#> GSM39158 1 0.3073 0.6779 0.856 0.000 0.024 0.004 0.116
#> GSM39159 1 0.7952 0.1145 0.400 0.000 0.200 0.100 0.300
#> GSM39160 1 0.5977 0.5250 0.612 0.000 0.100 0.020 0.268
#> GSM39161 3 0.8141 0.3933 0.220 0.000 0.372 0.116 0.292
#> GSM39162 1 0.2563 0.5764 0.872 0.000 0.008 0.000 0.120
#> GSM39163 1 0.0000 0.7120 1.000 0.000 0.000 0.000 0.000
#> GSM39164 1 0.0290 0.7073 0.992 0.000 0.000 0.000 0.008
#> GSM39165 1 0.6235 0.4548 0.572 0.000 0.164 0.008 0.256
#> GSM39166 1 0.6200 0.5188 0.612 0.000 0.100 0.036 0.252
#> GSM39167 1 0.0000 0.7120 1.000 0.000 0.000 0.000 0.000
#> GSM39168 1 0.1894 0.6552 0.920 0.000 0.008 0.000 0.072
#> GSM39169 1 0.0451 0.7092 0.988 0.000 0.004 0.000 0.008
#> GSM39170 1 0.1638 0.7061 0.932 0.000 0.004 0.000 0.064
#> GSM39171 1 0.4701 0.5942 0.704 0.000 0.060 0.000 0.236
#> GSM39172 3 0.6760 0.6096 0.064 0.000 0.572 0.112 0.252
#> GSM39173 3 0.1834 0.7776 0.032 0.004 0.940 0.008 0.016
#> GSM39174 1 0.0290 0.7073 0.992 0.000 0.000 0.000 0.008
#> GSM39175 1 0.2304 0.6917 0.892 0.000 0.008 0.000 0.100
#> GSM39176 1 0.0162 0.7131 0.996 0.000 0.004 0.000 0.000
#> GSM39177 3 0.3622 0.7580 0.048 0.000 0.816 0.000 0.136
#> GSM39178 1 0.7071 0.3236 0.504 0.000 0.180 0.040 0.276
#> GSM39179 3 0.1124 0.7864 0.036 0.000 0.960 0.000 0.004
#> GSM39180 3 0.4898 0.6894 0.020 0.004 0.760 0.120 0.096
#> GSM39181 1 0.6558 0.4626 0.568 0.000 0.108 0.044 0.280
#> GSM39182 5 0.8243 -0.4175 0.260 0.000 0.304 0.116 0.320
#> GSM39183 1 0.6300 0.4895 0.592 0.000 0.092 0.040 0.276
#> GSM39184 1 0.1121 0.7141 0.956 0.000 0.000 0.000 0.044
#> GSM39185 3 0.7969 0.4643 0.172 0.000 0.400 0.116 0.312
#> GSM39186 1 0.1012 0.7121 0.968 0.000 0.020 0.000 0.012
#> GSM39187 1 0.0290 0.7091 0.992 0.000 0.000 0.000 0.008
#> GSM39116 2 0.0963 0.5296 0.000 0.964 0.000 0.036 0.000
#> GSM39117 4 0.2439 0.9894 0.000 0.120 0.000 0.876 0.004
#> GSM39118 2 0.5513 0.2744 0.000 0.632 0.000 0.252 0.116
#> GSM39119 2 0.5844 -0.0515 0.000 0.484 0.000 0.420 0.096
#> GSM39120 5 0.4769 0.7187 0.440 0.004 0.012 0.000 0.544
#> GSM39121 5 0.6613 0.2406 0.184 0.384 0.000 0.004 0.428
#> GSM39122 2 0.6620 -0.2809 0.184 0.408 0.000 0.004 0.404
#> GSM39123 4 0.2230 0.9879 0.000 0.116 0.000 0.884 0.000
#> GSM39124 2 0.4462 0.4667 0.004 0.672 0.000 0.016 0.308
#> GSM39125 5 0.5067 0.7272 0.436 0.012 0.016 0.000 0.536
#> GSM39126 5 0.6656 0.2902 0.196 0.364 0.000 0.004 0.436
#> GSM39127 2 0.2864 0.5566 0.000 0.864 0.000 0.024 0.112
#> GSM39128 2 0.4920 0.4484 0.020 0.660 0.000 0.020 0.300
#> GSM39129 2 0.5811 0.1992 0.000 0.568 0.000 0.316 0.116
#> GSM39130 4 0.2439 0.9894 0.000 0.120 0.000 0.876 0.004
#> GSM39131 2 0.5005 0.4465 0.020 0.656 0.000 0.024 0.300
#> GSM39132 2 0.2012 0.5559 0.000 0.920 0.000 0.020 0.060
#> GSM39133 4 0.2536 0.9757 0.000 0.128 0.000 0.868 0.004
#> GSM39134 2 0.5838 0.1686 0.000 0.552 0.000 0.336 0.112
#> GSM39135 2 0.1168 0.5355 0.000 0.960 0.000 0.032 0.008
#> GSM39136 2 0.2798 0.4846 0.000 0.852 0.000 0.140 0.008
#> GSM39137 2 0.6606 -0.1176 0.176 0.480 0.000 0.008 0.336
#> GSM39138 2 0.5862 0.1512 0.000 0.544 0.000 0.344 0.112
#> GSM39139 2 0.5790 0.2702 0.000 0.608 0.004 0.268 0.120
#> GSM39140 5 0.5012 0.7488 0.404 0.016 0.012 0.000 0.568
#> GSM39141 5 0.4936 0.7455 0.412 0.012 0.012 0.000 0.564
#> GSM39142 5 0.4779 0.7116 0.448 0.004 0.012 0.000 0.536
#> GSM39143 5 0.4928 0.7473 0.408 0.012 0.012 0.000 0.568
#> GSM39144 2 0.5811 0.1992 0.000 0.568 0.000 0.316 0.116
#> GSM39145 2 0.3741 0.4569 0.000 0.816 0.000 0.076 0.108
#> GSM39146 2 0.2813 0.5468 0.000 0.876 0.000 0.040 0.084
#> GSM39147 2 0.2416 0.5613 0.000 0.888 0.000 0.012 0.100
#> GSM39188 3 0.1124 0.7848 0.036 0.000 0.960 0.000 0.004
#> GSM39189 3 0.1741 0.7874 0.040 0.000 0.936 0.000 0.024
#> GSM39190 3 0.0963 0.7852 0.036 0.000 0.964 0.000 0.000
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM39104 1 0.1970 0.746 0.920 0.000 0.044 0.008 0.028 0.000
#> GSM39105 1 0.1155 0.764 0.956 0.000 0.036 0.004 0.004 0.000
#> GSM39106 1 0.4361 0.498 0.544 0.016 0.436 0.004 0.000 0.000
#> GSM39107 1 0.5579 0.340 0.444 0.120 0.432 0.000 0.004 0.000
#> GSM39108 1 0.4402 0.573 0.632 0.016 0.336 0.016 0.000 0.000
#> GSM39109 3 0.6694 -0.356 0.372 0.156 0.408 0.000 0.064 0.000
#> GSM39110 1 0.4957 0.567 0.628 0.016 0.304 0.048 0.004 0.000
#> GSM39111 1 0.4120 0.662 0.788 0.000 0.056 0.104 0.052 0.000
#> GSM39112 1 0.4336 0.455 0.504 0.020 0.476 0.000 0.000 0.000
#> GSM39113 3 0.5879 -0.184 0.284 0.240 0.476 0.000 0.000 0.000
#> GSM39114 2 0.2260 0.861 0.000 0.860 0.140 0.000 0.000 0.000
#> GSM39115 1 0.0146 0.772 0.996 0.000 0.000 0.000 0.004 0.000
#> GSM39148 1 0.0547 0.771 0.980 0.000 0.020 0.000 0.000 0.000
#> GSM39149 3 0.4783 0.710 0.000 0.000 0.520 0.428 0.052 0.000
#> GSM39150 1 0.3111 0.674 0.836 0.000 0.032 0.008 0.124 0.000
#> GSM39151 3 0.4783 0.710 0.000 0.000 0.520 0.428 0.052 0.000
#> GSM39152 3 0.5811 0.589 0.008 0.000 0.492 0.348 0.152 0.000
#> GSM39153 1 0.0508 0.769 0.984 0.000 0.000 0.004 0.012 0.000
#> GSM39154 1 0.0632 0.766 0.976 0.000 0.000 0.000 0.024 0.000
#> GSM39155 1 0.0146 0.772 0.996 0.000 0.000 0.000 0.004 0.000
#> GSM39156 1 0.4074 0.579 0.656 0.016 0.324 0.000 0.004 0.000
#> GSM39157 1 0.0000 0.772 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM39158 1 0.1196 0.756 0.952 0.000 0.008 0.000 0.040 0.000
#> GSM39159 5 0.4118 0.535 0.396 0.000 0.008 0.004 0.592 0.000
#> GSM39160 1 0.3491 0.631 0.804 0.000 0.040 0.008 0.148 0.000
#> GSM39161 5 0.2994 0.707 0.208 0.000 0.004 0.000 0.788 0.000
#> GSM39162 1 0.0713 0.769 0.972 0.000 0.028 0.000 0.000 0.000
#> GSM39163 1 0.0000 0.772 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM39164 1 0.0146 0.772 0.996 0.000 0.000 0.000 0.004 0.000
#> GSM39165 1 0.4502 0.583 0.732 0.000 0.012 0.116 0.140 0.000
#> GSM39166 1 0.2020 0.716 0.896 0.000 0.008 0.000 0.096 0.000
#> GSM39167 1 0.0260 0.772 0.992 0.000 0.000 0.000 0.008 0.000
#> GSM39168 1 0.0692 0.771 0.976 0.000 0.020 0.000 0.004 0.000
#> GSM39169 1 0.0405 0.772 0.988 0.000 0.004 0.000 0.008 0.000
#> GSM39170 1 0.0806 0.763 0.972 0.000 0.008 0.000 0.020 0.000
#> GSM39171 1 0.2215 0.724 0.900 0.000 0.012 0.012 0.076 0.000
#> GSM39172 5 0.5420 0.568 0.144 0.000 0.044 0.148 0.664 0.000
#> GSM39173 3 0.4873 0.708 0.000 0.000 0.520 0.420 0.060 0.000
#> GSM39174 1 0.0000 0.772 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM39175 1 0.0891 0.761 0.968 0.000 0.008 0.000 0.024 0.000
#> GSM39176 1 0.0000 0.772 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM39177 3 0.4962 0.703 0.000 0.000 0.516 0.416 0.068 0.000
#> GSM39178 5 0.4702 0.480 0.440 0.000 0.012 0.024 0.524 0.000
#> GSM39179 3 0.4783 0.710 0.000 0.000 0.520 0.428 0.052 0.000
#> GSM39180 5 0.6049 -0.154 0.004 0.000 0.216 0.288 0.488 0.004
#> GSM39181 1 0.3298 0.525 0.756 0.000 0.008 0.000 0.236 0.000
#> GSM39182 5 0.2883 0.706 0.212 0.000 0.000 0.000 0.788 0.000
#> GSM39183 1 0.3437 0.507 0.752 0.000 0.008 0.004 0.236 0.000
#> GSM39184 1 0.0458 0.768 0.984 0.000 0.000 0.000 0.016 0.000
#> GSM39185 5 0.2933 0.703 0.200 0.000 0.004 0.000 0.796 0.000
#> GSM39186 1 0.0976 0.769 0.968 0.000 0.016 0.008 0.008 0.000
#> GSM39187 1 0.0146 0.772 0.996 0.000 0.004 0.000 0.000 0.000
#> GSM39116 2 0.1524 0.878 0.000 0.932 0.000 0.008 0.000 0.060
#> GSM39117 4 0.5696 0.997 0.000 0.008 0.000 0.564 0.200 0.228
#> GSM39118 6 0.2340 0.822 0.000 0.148 0.000 0.000 0.000 0.852
#> GSM39119 6 0.4252 0.695 0.000 0.128 0.000 0.076 0.028 0.768
#> GSM39120 1 0.4176 0.516 0.580 0.016 0.404 0.000 0.000 0.000
#> GSM39121 2 0.2595 0.842 0.004 0.836 0.160 0.000 0.000 0.000
#> GSM39122 2 0.2482 0.852 0.004 0.848 0.148 0.000 0.000 0.000
#> GSM39123 4 0.5696 0.997 0.000 0.008 0.000 0.564 0.200 0.228
#> GSM39124 2 0.1075 0.901 0.000 0.952 0.048 0.000 0.000 0.000
#> GSM39125 1 0.4404 0.514 0.576 0.016 0.400 0.000 0.008 0.000
#> GSM39126 2 0.2668 0.836 0.004 0.828 0.168 0.000 0.000 0.000
#> GSM39127 2 0.1453 0.893 0.000 0.944 0.008 0.008 0.000 0.040
#> GSM39128 2 0.1007 0.901 0.000 0.956 0.044 0.000 0.000 0.000
#> GSM39129 6 0.0865 0.850 0.000 0.036 0.000 0.000 0.000 0.964
#> GSM39130 4 0.5696 0.997 0.000 0.008 0.000 0.564 0.200 0.228
#> GSM39131 2 0.1007 0.901 0.000 0.956 0.044 0.000 0.000 0.000
#> GSM39132 2 0.1219 0.887 0.000 0.948 0.000 0.004 0.000 0.048
#> GSM39133 4 0.5766 0.992 0.000 0.012 0.000 0.564 0.200 0.224
#> GSM39134 6 0.2510 0.819 0.000 0.080 0.000 0.028 0.008 0.884
#> GSM39135 2 0.1411 0.880 0.000 0.936 0.000 0.004 0.000 0.060
#> GSM39136 2 0.2480 0.836 0.000 0.872 0.000 0.024 0.000 0.104
#> GSM39137 2 0.1610 0.889 0.000 0.916 0.084 0.000 0.000 0.000
#> GSM39138 6 0.0767 0.824 0.000 0.012 0.000 0.008 0.004 0.976
#> GSM39139 6 0.1700 0.843 0.000 0.080 0.004 0.000 0.000 0.916
#> GSM39140 1 0.4439 0.480 0.540 0.028 0.432 0.000 0.000 0.000
#> GSM39141 1 0.4294 0.492 0.552 0.020 0.428 0.000 0.000 0.000
#> GSM39142 1 0.4121 0.537 0.604 0.016 0.380 0.000 0.000 0.000
#> GSM39143 1 0.4366 0.488 0.548 0.024 0.428 0.000 0.000 0.000
#> GSM39144 6 0.0865 0.850 0.000 0.036 0.000 0.000 0.000 0.964
#> GSM39145 6 0.3189 0.705 0.000 0.236 0.004 0.000 0.000 0.760
#> GSM39146 2 0.0632 0.893 0.000 0.976 0.000 0.000 0.000 0.024
#> GSM39147 2 0.1588 0.875 0.000 0.924 0.004 0.000 0.000 0.072
#> GSM39188 3 0.4921 0.705 0.000 0.000 0.516 0.420 0.064 0.000
#> GSM39189 3 0.5275 0.689 0.008 0.000 0.504 0.412 0.076 0.000
#> GSM39190 3 0.4783 0.710 0.000 0.000 0.520 0.428 0.052 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) other(p) protocol(p) k
#> SD:mclust 69 0.13866 1.02e-09 1.23e-08 2
#> SD:mclust 69 0.00106 2.98e-11 9.01e-09 3
#> SD:mclust 66 0.01518 8.78e-06 2.01e-08 4
#> SD:mclust 59 0.02108 1.40e-07 5.06e-08 5
#> SD:mclust 77 0.39118 2.39e-07 1.94e-07 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 8353 rows and 87 columns.
#> Top rows (835, 1670, 2506, 3341, 4176) 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 0.882 0.919 0.967 0.4929 0.505 0.505
#> 3 3 0.451 0.550 0.763 0.3178 0.706 0.485
#> 4 4 0.441 0.517 0.740 0.0920 0.892 0.707
#> 5 5 0.490 0.420 0.651 0.0932 0.823 0.500
#> 6 6 0.545 0.495 0.685 0.0511 0.895 0.600
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
#> GSM39104 1 0.0000 0.9698 1.000 0.000
#> GSM39105 1 0.0000 0.9698 1.000 0.000
#> GSM39106 1 0.0000 0.9698 1.000 0.000
#> GSM39107 1 0.0000 0.9698 1.000 0.000
#> GSM39108 1 0.0000 0.9698 1.000 0.000
#> GSM39109 2 0.8608 0.6146 0.284 0.716
#> GSM39110 1 0.0000 0.9698 1.000 0.000
#> GSM39111 1 0.0000 0.9698 1.000 0.000
#> GSM39112 1 0.0000 0.9698 1.000 0.000
#> GSM39113 1 0.0376 0.9669 0.996 0.004
#> GSM39114 2 0.2236 0.9323 0.036 0.964
#> GSM39115 1 0.0000 0.9698 1.000 0.000
#> GSM39148 1 0.0000 0.9698 1.000 0.000
#> GSM39149 2 0.0000 0.9586 0.000 1.000
#> GSM39150 1 0.0000 0.9698 1.000 0.000
#> GSM39151 2 0.0938 0.9506 0.012 0.988
#> GSM39152 1 0.7056 0.7520 0.808 0.192
#> GSM39153 1 0.0000 0.9698 1.000 0.000
#> GSM39154 1 0.0000 0.9698 1.000 0.000
#> GSM39155 1 0.0000 0.9698 1.000 0.000
#> GSM39156 1 0.0000 0.9698 1.000 0.000
#> GSM39157 1 0.0000 0.9698 1.000 0.000
#> GSM39158 1 0.0000 0.9698 1.000 0.000
#> GSM39159 1 0.2043 0.9422 0.968 0.032
#> GSM39160 1 0.0000 0.9698 1.000 0.000
#> GSM39161 1 0.4939 0.8624 0.892 0.108
#> GSM39162 1 0.0000 0.9698 1.000 0.000
#> GSM39163 1 0.0000 0.9698 1.000 0.000
#> GSM39164 1 0.0000 0.9698 1.000 0.000
#> GSM39165 1 0.0376 0.9668 0.996 0.004
#> GSM39166 1 0.0000 0.9698 1.000 0.000
#> GSM39167 1 0.0000 0.9698 1.000 0.000
#> GSM39168 1 0.0000 0.9698 1.000 0.000
#> GSM39169 1 0.0000 0.9698 1.000 0.000
#> GSM39170 1 0.0000 0.9698 1.000 0.000
#> GSM39171 1 0.0000 0.9698 1.000 0.000
#> GSM39172 2 0.0000 0.9586 0.000 1.000
#> GSM39173 2 0.0000 0.9586 0.000 1.000
#> GSM39174 1 0.0000 0.9698 1.000 0.000
#> GSM39175 1 0.0000 0.9698 1.000 0.000
#> GSM39176 1 0.0000 0.9698 1.000 0.000
#> GSM39177 2 0.9996 0.0498 0.488 0.512
#> GSM39178 1 0.0000 0.9698 1.000 0.000
#> GSM39179 2 0.0000 0.9586 0.000 1.000
#> GSM39180 2 0.0000 0.9586 0.000 1.000
#> GSM39181 1 0.0000 0.9698 1.000 0.000
#> GSM39182 2 0.5294 0.8542 0.120 0.880
#> GSM39183 1 0.0000 0.9698 1.000 0.000
#> GSM39184 1 0.0000 0.9698 1.000 0.000
#> GSM39185 2 0.8661 0.6034 0.288 0.712
#> GSM39186 1 0.0000 0.9698 1.000 0.000
#> GSM39187 1 0.0000 0.9698 1.000 0.000
#> GSM39116 2 0.0000 0.9586 0.000 1.000
#> GSM39117 2 0.0000 0.9586 0.000 1.000
#> GSM39118 2 0.0000 0.9586 0.000 1.000
#> GSM39119 2 0.0000 0.9586 0.000 1.000
#> GSM39120 1 0.0376 0.9669 0.996 0.004
#> GSM39121 1 0.7453 0.7185 0.788 0.212
#> GSM39122 1 0.9608 0.3607 0.616 0.384
#> GSM39123 2 0.0000 0.9586 0.000 1.000
#> GSM39124 2 0.0376 0.9561 0.004 0.996
#> GSM39125 1 0.0376 0.9669 0.996 0.004
#> GSM39126 1 0.9933 0.1504 0.548 0.452
#> GSM39127 2 0.0000 0.9586 0.000 1.000
#> GSM39128 2 0.0000 0.9586 0.000 1.000
#> GSM39129 2 0.0000 0.9586 0.000 1.000
#> GSM39130 2 0.0000 0.9586 0.000 1.000
#> GSM39131 2 0.0000 0.9586 0.000 1.000
#> GSM39132 2 0.0000 0.9586 0.000 1.000
#> GSM39133 2 0.0000 0.9586 0.000 1.000
#> GSM39134 2 0.0000 0.9586 0.000 1.000
#> GSM39135 2 0.0000 0.9586 0.000 1.000
#> GSM39136 2 0.0000 0.9586 0.000 1.000
#> GSM39137 2 0.5737 0.8329 0.136 0.864
#> GSM39138 2 0.0000 0.9586 0.000 1.000
#> GSM39139 2 0.0000 0.9586 0.000 1.000
#> GSM39140 1 0.0000 0.9698 1.000 0.000
#> GSM39141 1 0.0000 0.9698 1.000 0.000
#> GSM39142 1 0.0000 0.9698 1.000 0.000
#> GSM39143 1 0.0000 0.9698 1.000 0.000
#> GSM39144 2 0.0000 0.9586 0.000 1.000
#> GSM39145 2 0.0000 0.9586 0.000 1.000
#> GSM39146 2 0.0000 0.9586 0.000 1.000
#> GSM39147 2 0.0000 0.9586 0.000 1.000
#> GSM39188 2 0.0000 0.9586 0.000 1.000
#> GSM39189 2 0.3431 0.9097 0.064 0.936
#> GSM39190 2 0.0000 0.9586 0.000 1.000
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM39104 1 0.271 0.7419 0.912 0.088 0.000
#> GSM39105 1 0.533 0.6640 0.728 0.272 0.000
#> GSM39106 2 0.613 0.1774 0.400 0.600 0.000
#> GSM39107 2 0.388 0.6345 0.152 0.848 0.000
#> GSM39108 1 0.619 0.4094 0.580 0.420 0.000
#> GSM39109 2 0.818 0.5532 0.208 0.640 0.152
#> GSM39110 1 0.610 0.4728 0.608 0.392 0.000
#> GSM39111 1 0.206 0.7308 0.948 0.044 0.008
#> GSM39112 2 0.440 0.6025 0.188 0.812 0.000
#> GSM39113 2 0.355 0.6493 0.132 0.868 0.000
#> GSM39114 2 0.153 0.6805 0.040 0.960 0.000
#> GSM39115 1 0.536 0.6591 0.724 0.276 0.000
#> GSM39148 1 0.595 0.5350 0.640 0.360 0.000
#> GSM39149 3 0.588 0.5603 0.348 0.000 0.652
#> GSM39150 1 0.254 0.6551 0.920 0.000 0.080
#> GSM39151 3 0.550 0.6194 0.292 0.000 0.708
#> GSM39152 1 0.588 0.1589 0.652 0.000 0.348
#> GSM39153 1 0.327 0.7448 0.884 0.116 0.000
#> GSM39154 1 0.312 0.7448 0.892 0.108 0.000
#> GSM39155 1 0.394 0.7382 0.844 0.156 0.000
#> GSM39156 2 0.617 0.1377 0.412 0.588 0.000
#> GSM39157 1 0.543 0.6490 0.716 0.284 0.000
#> GSM39158 1 0.312 0.7446 0.892 0.108 0.000
#> GSM39159 1 0.445 0.5065 0.808 0.000 0.192
#> GSM39160 1 0.288 0.6380 0.904 0.000 0.096
#> GSM39161 1 0.619 -0.0736 0.580 0.000 0.420
#> GSM39162 1 0.625 0.3443 0.556 0.444 0.000
#> GSM39163 1 0.445 0.7239 0.808 0.192 0.000
#> GSM39164 1 0.514 0.6818 0.748 0.252 0.000
#> GSM39165 1 0.319 0.6204 0.888 0.000 0.112
#> GSM39166 1 0.216 0.6699 0.936 0.000 0.064
#> GSM39167 1 0.510 0.6855 0.752 0.248 0.000
#> GSM39168 1 0.601 0.5122 0.628 0.372 0.000
#> GSM39169 1 0.455 0.7199 0.800 0.200 0.000
#> GSM39170 1 0.382 0.7405 0.852 0.148 0.000
#> GSM39171 1 0.175 0.6837 0.952 0.000 0.048
#> GSM39172 3 0.559 0.6095 0.304 0.000 0.696
#> GSM39173 3 0.369 0.6694 0.140 0.000 0.860
#> GSM39174 1 0.460 0.7174 0.796 0.204 0.000
#> GSM39175 1 0.103 0.7005 0.976 0.000 0.024
#> GSM39176 1 0.489 0.7017 0.772 0.228 0.000
#> GSM39177 1 0.620 -0.0864 0.576 0.000 0.424
#> GSM39178 1 0.424 0.5321 0.824 0.000 0.176
#> GSM39179 3 0.550 0.6199 0.292 0.000 0.708
#> GSM39180 3 0.259 0.6621 0.072 0.004 0.924
#> GSM39181 1 0.164 0.6866 0.956 0.000 0.044
#> GSM39182 3 0.595 0.5431 0.360 0.000 0.640
#> GSM39183 1 0.236 0.6634 0.928 0.000 0.072
#> GSM39184 1 0.348 0.7435 0.872 0.128 0.000
#> GSM39185 3 0.629 0.3498 0.464 0.000 0.536
#> GSM39186 1 0.382 0.7404 0.852 0.148 0.000
#> GSM39187 1 0.565 0.6126 0.688 0.312 0.000
#> GSM39116 2 0.593 0.3302 0.000 0.644 0.356
#> GSM39117 3 0.334 0.6234 0.000 0.120 0.880
#> GSM39118 3 0.620 0.2045 0.000 0.424 0.576
#> GSM39119 3 0.429 0.5923 0.000 0.180 0.820
#> GSM39120 2 0.455 0.5901 0.200 0.800 0.000
#> GSM39121 2 0.288 0.6701 0.096 0.904 0.000
#> GSM39122 2 0.271 0.6734 0.088 0.912 0.000
#> GSM39123 3 0.355 0.6196 0.000 0.132 0.868
#> GSM39124 2 0.240 0.6433 0.004 0.932 0.064
#> GSM39125 2 0.502 0.5391 0.240 0.760 0.000
#> GSM39126 2 0.280 0.6720 0.092 0.908 0.000
#> GSM39127 2 0.382 0.5853 0.000 0.852 0.148
#> GSM39128 2 0.295 0.6306 0.004 0.908 0.088
#> GSM39129 3 0.506 0.5392 0.000 0.244 0.756
#> GSM39130 3 0.355 0.6196 0.000 0.132 0.868
#> GSM39131 2 0.341 0.6043 0.000 0.876 0.124
#> GSM39132 2 0.497 0.4979 0.000 0.764 0.236
#> GSM39133 3 0.540 0.4919 0.000 0.280 0.720
#> GSM39134 3 0.556 0.4692 0.000 0.300 0.700
#> GSM39135 2 0.581 0.3628 0.000 0.664 0.336
#> GSM39136 2 0.597 0.3146 0.000 0.636 0.364
#> GSM39137 2 0.257 0.6691 0.032 0.936 0.032
#> GSM39138 3 0.470 0.5686 0.000 0.212 0.788
#> GSM39139 2 0.628 0.0793 0.000 0.540 0.460
#> GSM39140 2 0.533 0.4895 0.272 0.728 0.000
#> GSM39141 2 0.536 0.4827 0.276 0.724 0.000
#> GSM39142 2 0.603 0.2536 0.376 0.624 0.000
#> GSM39143 2 0.581 0.3578 0.336 0.664 0.000
#> GSM39144 3 0.559 0.4616 0.000 0.304 0.696
#> GSM39145 2 0.604 0.2838 0.000 0.620 0.380
#> GSM39146 2 0.502 0.4944 0.000 0.760 0.240
#> GSM39147 2 0.455 0.5363 0.000 0.800 0.200
#> GSM39188 3 0.533 0.6325 0.272 0.000 0.728
#> GSM39189 3 0.618 0.4526 0.416 0.000 0.584
#> GSM39190 3 0.525 0.6367 0.264 0.000 0.736
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM39104 1 0.404 0.7508 0.852 0.036 0.088 0.024
#> GSM39105 1 0.527 0.6488 0.740 0.212 0.028 0.020
#> GSM39106 2 0.609 0.3549 0.348 0.604 0.012 0.036
#> GSM39107 2 0.673 0.5531 0.216 0.636 0.008 0.140
#> GSM39108 1 0.685 0.1869 0.500 0.428 0.036 0.036
#> GSM39109 4 0.814 0.0205 0.212 0.284 0.024 0.480
#> GSM39110 1 0.686 0.2137 0.496 0.428 0.056 0.020
#> GSM39111 1 0.615 0.6799 0.700 0.072 0.204 0.024
#> GSM39112 2 0.523 0.5864 0.244 0.716 0.004 0.036
#> GSM39113 2 0.476 0.5869 0.160 0.784 0.004 0.052
#> GSM39114 2 0.249 0.5018 0.028 0.920 0.004 0.048
#> GSM39115 1 0.447 0.7084 0.820 0.116 0.012 0.052
#> GSM39148 1 0.476 0.5502 0.724 0.260 0.004 0.012
#> GSM39149 3 0.251 0.7197 0.064 0.008 0.916 0.012
#> GSM39150 1 0.476 0.6504 0.764 0.000 0.192 0.044
#> GSM39151 3 0.273 0.7214 0.088 0.000 0.896 0.016
#> GSM39152 3 0.488 0.5699 0.272 0.000 0.708 0.020
#> GSM39153 1 0.212 0.7649 0.936 0.032 0.028 0.004
#> GSM39154 1 0.232 0.7648 0.928 0.032 0.036 0.004
#> GSM39155 1 0.248 0.7623 0.916 0.052 0.032 0.000
#> GSM39156 2 0.558 0.1461 0.468 0.516 0.008 0.008
#> GSM39157 1 0.421 0.6673 0.804 0.172 0.008 0.016
#> GSM39158 1 0.252 0.7520 0.920 0.012 0.016 0.052
#> GSM39159 1 0.499 0.5925 0.744 0.000 0.208 0.048
#> GSM39160 1 0.499 0.6222 0.740 0.000 0.216 0.044
#> GSM39161 1 0.635 0.4135 0.636 0.000 0.252 0.112
#> GSM39162 1 0.516 0.3421 0.624 0.364 0.000 0.012
#> GSM39163 1 0.230 0.7522 0.924 0.060 0.008 0.008
#> GSM39164 1 0.352 0.7165 0.848 0.136 0.008 0.008
#> GSM39165 1 0.511 0.4841 0.648 0.004 0.340 0.008
#> GSM39166 1 0.360 0.7177 0.860 0.000 0.084 0.056
#> GSM39167 1 0.362 0.7139 0.856 0.116 0.012 0.016
#> GSM39168 1 0.509 0.4733 0.672 0.312 0.004 0.012
#> GSM39169 1 0.330 0.7511 0.876 0.092 0.028 0.004
#> GSM39170 1 0.184 0.7593 0.948 0.016 0.008 0.028
#> GSM39171 1 0.420 0.6704 0.788 0.004 0.196 0.012
#> GSM39172 3 0.763 0.4023 0.320 0.000 0.456 0.224
#> GSM39173 3 0.246 0.6883 0.028 0.040 0.924 0.008
#> GSM39174 1 0.339 0.7452 0.868 0.104 0.024 0.004
#> GSM39175 1 0.327 0.7241 0.860 0.004 0.128 0.008
#> GSM39176 1 0.343 0.7254 0.868 0.104 0.008 0.020
#> GSM39177 3 0.371 0.6664 0.192 0.000 0.804 0.004
#> GSM39178 1 0.568 0.5232 0.680 0.000 0.256 0.064
#> GSM39179 3 0.185 0.7193 0.048 0.000 0.940 0.012
#> GSM39180 3 0.423 0.6320 0.036 0.004 0.816 0.144
#> GSM39181 1 0.395 0.7081 0.828 0.000 0.036 0.136
#> GSM39182 4 0.513 0.3939 0.184 0.000 0.068 0.748
#> GSM39183 1 0.402 0.7096 0.836 0.000 0.068 0.096
#> GSM39184 1 0.184 0.7646 0.948 0.028 0.016 0.008
#> GSM39185 1 0.706 0.1777 0.540 0.000 0.312 0.148
#> GSM39186 1 0.334 0.7616 0.884 0.052 0.056 0.008
#> GSM39187 1 0.403 0.6835 0.824 0.148 0.008 0.020
#> GSM39116 4 0.532 0.6144 0.000 0.312 0.028 0.660
#> GSM39117 4 0.343 0.6426 0.000 0.028 0.112 0.860
#> GSM39118 4 0.780 0.4800 0.000 0.340 0.256 0.404
#> GSM39119 4 0.677 0.5119 0.000 0.128 0.292 0.580
#> GSM39120 2 0.487 0.5874 0.244 0.728 0.000 0.028
#> GSM39121 2 0.327 0.5842 0.132 0.856 0.000 0.012
#> GSM39122 2 0.322 0.5768 0.112 0.868 0.000 0.020
#> GSM39123 4 0.277 0.6532 0.008 0.024 0.060 0.908
#> GSM39124 2 0.375 0.3757 0.012 0.836 0.008 0.144
#> GSM39125 2 0.787 0.3511 0.364 0.428 0.008 0.200
#> GSM39126 2 0.359 0.5714 0.104 0.860 0.004 0.032
#> GSM39127 2 0.532 -0.1914 0.012 0.572 0.000 0.416
#> GSM39128 2 0.468 0.2937 0.020 0.764 0.008 0.208
#> GSM39129 3 0.659 0.2612 0.000 0.212 0.628 0.160
#> GSM39130 4 0.294 0.6579 0.000 0.032 0.076 0.892
#> GSM39131 2 0.484 0.2648 0.016 0.748 0.012 0.224
#> GSM39132 2 0.590 -0.2670 0.000 0.564 0.040 0.396
#> GSM39133 4 0.302 0.6759 0.004 0.060 0.040 0.896
#> GSM39134 4 0.756 0.4733 0.000 0.220 0.304 0.476
#> GSM39135 4 0.575 0.5195 0.000 0.396 0.032 0.572
#> GSM39136 4 0.422 0.6787 0.000 0.184 0.024 0.792
#> GSM39137 2 0.264 0.4953 0.032 0.908 0.000 0.060
#> GSM39138 3 0.668 0.2176 0.000 0.156 0.616 0.228
#> GSM39139 2 0.712 -0.2539 0.000 0.440 0.432 0.128
#> GSM39140 2 0.494 0.4934 0.316 0.672 0.000 0.012
#> GSM39141 2 0.539 0.4476 0.344 0.632 0.000 0.024
#> GSM39142 2 0.578 0.1483 0.468 0.504 0.000 0.028
#> GSM39143 2 0.570 0.3209 0.412 0.560 0.000 0.028
#> GSM39144 3 0.654 0.2484 0.000 0.252 0.620 0.128
#> GSM39145 2 0.698 -0.1579 0.000 0.536 0.332 0.132
#> GSM39146 4 0.521 0.5768 0.004 0.336 0.012 0.648
#> GSM39147 2 0.471 0.2591 0.000 0.792 0.088 0.120
#> GSM39188 3 0.318 0.7178 0.084 0.000 0.880 0.036
#> GSM39189 3 0.531 0.5672 0.280 0.000 0.684 0.036
#> GSM39190 3 0.234 0.7216 0.060 0.000 0.920 0.020
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM39104 5 0.546 0.4944 0.232 0.004 0.108 0.000 0.656
#> GSM39105 5 0.497 0.4548 0.300 0.044 0.004 0.000 0.652
#> GSM39106 5 0.542 0.5348 0.136 0.160 0.012 0.000 0.692
#> GSM39107 5 0.739 0.0638 0.088 0.344 0.000 0.116 0.452
#> GSM39108 5 0.553 0.5601 0.216 0.076 0.028 0.000 0.680
#> GSM39109 5 0.745 0.3810 0.068 0.076 0.064 0.212 0.580
#> GSM39110 5 0.649 0.5650 0.200 0.092 0.084 0.000 0.624
#> GSM39111 5 0.613 0.4899 0.196 0.012 0.184 0.000 0.608
#> GSM39112 5 0.638 0.2606 0.104 0.316 0.000 0.028 0.552
#> GSM39113 5 0.657 0.1363 0.056 0.348 0.000 0.072 0.524
#> GSM39114 2 0.614 0.3396 0.028 0.568 0.000 0.080 0.324
#> GSM39115 1 0.525 0.2666 0.560 0.028 0.000 0.012 0.400
#> GSM39148 1 0.346 0.5646 0.836 0.096 0.000 0.000 0.068
#> GSM39149 3 0.199 0.7052 0.008 0.016 0.928 0.000 0.048
#> GSM39150 5 0.716 0.1704 0.308 0.012 0.208 0.012 0.460
#> GSM39151 3 0.361 0.6660 0.036 0.004 0.840 0.012 0.108
#> GSM39152 3 0.559 0.4490 0.096 0.008 0.660 0.004 0.232
#> GSM39153 1 0.338 0.5913 0.844 0.004 0.012 0.016 0.124
#> GSM39154 1 0.305 0.6084 0.888 0.028 0.020 0.012 0.052
#> GSM39155 1 0.305 0.5903 0.856 0.008 0.008 0.004 0.124
#> GSM39156 1 0.633 0.0203 0.484 0.168 0.000 0.000 0.348
#> GSM39157 1 0.268 0.5836 0.880 0.092 0.000 0.000 0.028
#> GSM39158 1 0.505 0.5367 0.728 0.012 0.020 0.040 0.200
#> GSM39159 1 0.721 0.4262 0.580 0.024 0.144 0.052 0.200
#> GSM39160 5 0.727 0.1333 0.328 0.012 0.224 0.012 0.424
#> GSM39161 1 0.823 0.3314 0.476 0.024 0.140 0.136 0.224
#> GSM39162 1 0.496 0.4609 0.708 0.180 0.000 0.000 0.112
#> GSM39163 1 0.176 0.6122 0.940 0.016 0.000 0.008 0.036
#> GSM39164 1 0.381 0.5474 0.800 0.036 0.004 0.000 0.160
#> GSM39165 1 0.610 0.3457 0.588 0.020 0.312 0.008 0.072
#> GSM39166 1 0.661 0.4338 0.592 0.024 0.072 0.036 0.276
#> GSM39167 1 0.231 0.6076 0.916 0.044 0.000 0.012 0.028
#> GSM39168 1 0.475 0.4764 0.728 0.100 0.000 0.000 0.172
#> GSM39169 1 0.272 0.6072 0.880 0.020 0.000 0.004 0.096
#> GSM39170 1 0.538 0.5088 0.688 0.016 0.028 0.028 0.240
#> GSM39171 1 0.669 0.2059 0.516 0.008 0.164 0.008 0.304
#> GSM39172 4 0.810 -0.0925 0.112 0.016 0.324 0.420 0.128
#> GSM39173 3 0.393 0.6444 0.008 0.168 0.792 0.000 0.032
#> GSM39174 1 0.306 0.5844 0.868 0.028 0.004 0.004 0.096
#> GSM39175 1 0.358 0.5863 0.848 0.012 0.036 0.008 0.096
#> GSM39176 1 0.191 0.6112 0.936 0.024 0.000 0.016 0.024
#> GSM39177 3 0.350 0.6761 0.080 0.020 0.852 0.000 0.048
#> GSM39178 1 0.780 0.0276 0.364 0.020 0.308 0.024 0.284
#> GSM39179 3 0.314 0.6899 0.008 0.016 0.876 0.024 0.076
#> GSM39180 3 0.694 0.5719 0.048 0.076 0.648 0.104 0.124
#> GSM39181 1 0.604 0.4855 0.648 0.020 0.016 0.084 0.232
#> GSM39182 4 0.371 0.5952 0.052 0.012 0.040 0.856 0.040
#> GSM39183 1 0.723 0.3995 0.544 0.024 0.072 0.076 0.284
#> GSM39184 1 0.299 0.5978 0.884 0.016 0.008 0.020 0.072
#> GSM39185 1 0.836 0.2987 0.452 0.024 0.148 0.136 0.240
#> GSM39186 1 0.473 0.4789 0.704 0.012 0.024 0.004 0.256
#> GSM39187 1 0.276 0.5972 0.892 0.060 0.000 0.012 0.036
#> GSM39116 4 0.529 0.4651 0.000 0.280 0.004 0.644 0.072
#> GSM39117 4 0.216 0.6736 0.000 0.024 0.036 0.924 0.016
#> GSM39118 2 0.721 -0.0814 0.000 0.412 0.192 0.364 0.032
#> GSM39119 4 0.595 0.4906 0.000 0.204 0.144 0.636 0.016
#> GSM39120 2 0.713 -0.1098 0.204 0.400 0.000 0.024 0.372
#> GSM39121 2 0.549 0.3875 0.212 0.660 0.000 0.004 0.124
#> GSM39122 2 0.545 0.4611 0.116 0.692 0.000 0.016 0.176
#> GSM39123 4 0.143 0.6789 0.004 0.024 0.012 0.956 0.004
#> GSM39124 2 0.458 0.5555 0.068 0.796 0.004 0.088 0.044
#> GSM39125 1 0.771 0.1325 0.464 0.256 0.000 0.096 0.184
#> GSM39126 2 0.472 0.4846 0.144 0.736 0.000 0.000 0.120
#> GSM39127 2 0.582 0.3003 0.012 0.588 0.000 0.316 0.084
#> GSM39128 2 0.563 0.5256 0.076 0.716 0.008 0.152 0.048
#> GSM39129 3 0.611 0.2912 0.000 0.388 0.516 0.076 0.020
#> GSM39130 4 0.150 0.6795 0.000 0.024 0.016 0.952 0.008
#> GSM39131 2 0.648 0.4133 0.016 0.592 0.008 0.224 0.160
#> GSM39132 2 0.513 0.3178 0.000 0.656 0.008 0.284 0.052
#> GSM39133 4 0.163 0.6738 0.000 0.044 0.000 0.940 0.016
#> GSM39134 2 0.710 -0.1677 0.000 0.392 0.196 0.388 0.024
#> GSM39135 4 0.515 0.0736 0.004 0.480 0.008 0.492 0.016
#> GSM39136 4 0.479 0.5504 0.000 0.224 0.000 0.704 0.072
#> GSM39137 2 0.434 0.5417 0.132 0.792 0.000 0.028 0.048
#> GSM39138 3 0.659 0.2855 0.000 0.364 0.500 0.104 0.032
#> GSM39139 2 0.531 0.1239 0.000 0.612 0.336 0.032 0.020
#> GSM39140 1 0.620 0.1785 0.492 0.380 0.000 0.004 0.124
#> GSM39141 1 0.641 0.2298 0.528 0.304 0.000 0.008 0.160
#> GSM39142 1 0.608 0.3021 0.592 0.200 0.000 0.004 0.204
#> GSM39143 1 0.665 0.1808 0.512 0.284 0.000 0.012 0.192
#> GSM39144 3 0.560 0.2976 0.000 0.416 0.528 0.032 0.024
#> GSM39145 2 0.504 0.2644 0.000 0.664 0.284 0.040 0.012
#> GSM39146 4 0.528 0.5100 0.008 0.224 0.004 0.688 0.076
#> GSM39147 2 0.325 0.5100 0.012 0.872 0.068 0.040 0.008
#> GSM39188 3 0.287 0.6969 0.024 0.008 0.896 0.024 0.048
#> GSM39189 3 0.595 0.4713 0.068 0.008 0.648 0.032 0.244
#> GSM39190 3 0.207 0.7062 0.004 0.016 0.920 0.000 0.060
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM39104 6 0.441 0.55327 0.068 0.000 0.060 0.000 0.104 0.768
#> GSM39105 6 0.493 0.52814 0.168 0.008 0.024 0.000 0.088 0.712
#> GSM39106 6 0.338 0.59601 0.028 0.088 0.000 0.008 0.032 0.844
#> GSM39107 6 0.586 0.40918 0.020 0.228 0.000 0.052 0.072 0.628
#> GSM39108 6 0.349 0.61843 0.100 0.020 0.016 0.000 0.028 0.836
#> GSM39109 6 0.463 0.56660 0.004 0.060 0.056 0.096 0.012 0.772
#> GSM39110 6 0.544 0.56494 0.132 0.052 0.112 0.000 0.012 0.692
#> GSM39111 6 0.553 0.44787 0.100 0.004 0.192 0.000 0.048 0.656
#> GSM39112 6 0.470 0.50103 0.040 0.200 0.000 0.016 0.024 0.720
#> GSM39113 6 0.434 0.48760 0.012 0.212 0.000 0.020 0.024 0.732
#> GSM39114 6 0.529 0.15628 0.008 0.388 0.000 0.032 0.028 0.544
#> GSM39115 5 0.679 0.23683 0.240 0.020 0.004 0.008 0.376 0.352
#> GSM39148 1 0.170 0.71633 0.936 0.028 0.000 0.000 0.024 0.012
#> GSM39149 3 0.314 0.71825 0.016 0.024 0.872 0.004 0.040 0.044
#> GSM39150 6 0.676 0.22562 0.120 0.004 0.108 0.000 0.260 0.508
#> GSM39151 3 0.418 0.69823 0.004 0.012 0.772 0.000 0.120 0.092
#> GSM39152 3 0.594 0.44338 0.040 0.004 0.568 0.000 0.104 0.284
#> GSM39153 1 0.368 0.69337 0.840 0.012 0.048 0.004 0.044 0.052
#> GSM39154 1 0.177 0.71987 0.940 0.012 0.012 0.004 0.020 0.012
#> GSM39155 1 0.509 0.53539 0.688 0.012 0.020 0.000 0.200 0.080
#> GSM39156 1 0.526 0.58723 0.692 0.076 0.036 0.004 0.008 0.184
#> GSM39157 1 0.293 0.70515 0.856 0.024 0.000 0.000 0.104 0.016
#> GSM39158 5 0.472 0.51245 0.352 0.012 0.004 0.004 0.608 0.020
#> GSM39159 5 0.442 0.71690 0.188 0.004 0.068 0.004 0.732 0.004
#> GSM39160 6 0.741 0.12832 0.136 0.012 0.164 0.000 0.248 0.440
#> GSM39161 5 0.474 0.70359 0.144 0.020 0.044 0.044 0.748 0.000
#> GSM39162 1 0.209 0.70813 0.908 0.068 0.000 0.000 0.008 0.016
#> GSM39163 1 0.360 0.61622 0.764 0.004 0.000 0.004 0.212 0.016
#> GSM39164 1 0.330 0.70010 0.844 0.008 0.020 0.000 0.028 0.100
#> GSM39165 1 0.558 0.38486 0.588 0.016 0.312 0.000 0.060 0.024
#> GSM39166 5 0.472 0.71968 0.196 0.004 0.028 0.004 0.720 0.048
#> GSM39167 1 0.245 0.70127 0.876 0.016 0.000 0.004 0.104 0.000
#> GSM39168 1 0.219 0.71581 0.912 0.036 0.000 0.004 0.008 0.040
#> GSM39169 1 0.383 0.62423 0.764 0.004 0.004 0.000 0.192 0.036
#> GSM39170 5 0.418 0.64012 0.280 0.004 0.000 0.000 0.684 0.032
#> GSM39171 1 0.770 0.08311 0.424 0.016 0.152 0.004 0.168 0.236
#> GSM39172 4 0.649 0.25175 0.020 0.020 0.304 0.548 0.056 0.052
#> GSM39173 3 0.493 0.53574 0.000 0.196 0.672 0.000 0.124 0.008
#> GSM39174 1 0.155 0.72010 0.944 0.004 0.012 0.000 0.032 0.008
#> GSM39175 1 0.338 0.68745 0.848 0.004 0.052 0.000 0.056 0.040
#> GSM39176 1 0.350 0.63749 0.780 0.008 0.000 0.000 0.192 0.020
#> GSM39177 3 0.437 0.67485 0.100 0.020 0.788 0.004 0.064 0.024
#> GSM39178 5 0.671 0.44555 0.116 0.000 0.168 0.004 0.544 0.168
#> GSM39179 3 0.441 0.69579 0.044 0.028 0.808 0.028 0.036 0.056
#> GSM39180 5 0.627 -0.13710 0.008 0.084 0.324 0.044 0.532 0.008
#> GSM39181 5 0.470 0.67478 0.248 0.012 0.004 0.028 0.692 0.016
#> GSM39182 4 0.298 0.66796 0.016 0.016 0.024 0.884 0.044 0.016
#> GSM39183 5 0.479 0.72933 0.168 0.004 0.020 0.020 0.736 0.052
#> GSM39184 1 0.418 0.60618 0.752 0.008 0.008 0.008 0.196 0.028
#> GSM39185 5 0.441 0.66118 0.096 0.020 0.052 0.032 0.792 0.008
#> GSM39186 1 0.615 0.43971 0.612 0.016 0.048 0.000 0.168 0.156
#> GSM39187 1 0.310 0.69076 0.836 0.008 0.000 0.004 0.132 0.020
#> GSM39116 4 0.577 0.44874 0.000 0.248 0.004 0.588 0.020 0.140
#> GSM39117 4 0.207 0.69979 0.000 0.012 0.036 0.920 0.028 0.004
#> GSM39118 4 0.691 0.13110 0.000 0.384 0.172 0.388 0.028 0.028
#> GSM39119 4 0.502 0.62363 0.000 0.148 0.096 0.716 0.028 0.012
#> GSM39120 6 0.700 0.22294 0.116 0.276 0.000 0.028 0.080 0.500
#> GSM39121 2 0.569 0.31712 0.356 0.516 0.004 0.000 0.008 0.116
#> GSM39122 2 0.591 0.36890 0.188 0.572 0.000 0.008 0.012 0.220
#> GSM39123 4 0.108 0.70401 0.000 0.012 0.008 0.964 0.016 0.000
#> GSM39124 2 0.540 0.52615 0.132 0.712 0.016 0.048 0.008 0.084
#> GSM39125 1 0.838 0.05821 0.372 0.160 0.000 0.100 0.232 0.136
#> GSM39126 2 0.559 0.45353 0.180 0.628 0.000 0.004 0.020 0.168
#> GSM39127 2 0.677 0.22796 0.020 0.500 0.000 0.272 0.044 0.164
#> GSM39128 2 0.570 0.49714 0.100 0.696 0.004 0.092 0.024 0.084
#> GSM39129 2 0.659 -0.06966 0.000 0.440 0.404 0.044 0.076 0.036
#> GSM39130 4 0.133 0.70370 0.000 0.012 0.012 0.956 0.016 0.004
#> GSM39131 2 0.654 0.20318 0.008 0.480 0.004 0.120 0.044 0.344
#> GSM39132 2 0.622 0.31969 0.004 0.592 0.008 0.208 0.044 0.144
#> GSM39133 4 0.184 0.70191 0.000 0.040 0.000 0.928 0.012 0.020
#> GSM39134 2 0.704 -0.06023 0.000 0.436 0.168 0.316 0.068 0.012
#> GSM39135 4 0.519 0.32534 0.004 0.404 0.004 0.536 0.016 0.036
#> GSM39136 4 0.506 0.58495 0.000 0.212 0.004 0.684 0.036 0.064
#> GSM39137 2 0.516 0.50133 0.184 0.692 0.000 0.020 0.016 0.088
#> GSM39138 3 0.653 0.00822 0.000 0.404 0.432 0.060 0.092 0.012
#> GSM39139 2 0.524 0.24174 0.004 0.624 0.296 0.020 0.048 0.008
#> GSM39140 1 0.477 0.55104 0.688 0.236 0.000 0.004 0.024 0.048
#> GSM39141 1 0.421 0.61147 0.744 0.196 0.000 0.008 0.008 0.044
#> GSM39142 1 0.445 0.65059 0.752 0.112 0.000 0.004 0.016 0.116
#> GSM39143 1 0.491 0.61830 0.708 0.136 0.000 0.000 0.028 0.128
#> GSM39144 2 0.561 -0.11234 0.000 0.456 0.452 0.024 0.064 0.004
#> GSM39145 2 0.479 0.36261 0.000 0.692 0.232 0.012 0.048 0.016
#> GSM39146 4 0.403 0.63726 0.004 0.172 0.000 0.764 0.008 0.052
#> GSM39147 2 0.471 0.50418 0.080 0.780 0.064 0.012 0.036 0.028
#> GSM39188 3 0.420 0.71284 0.004 0.024 0.792 0.036 0.124 0.020
#> GSM39189 3 0.639 0.49908 0.012 0.012 0.532 0.008 0.208 0.228
#> GSM39190 3 0.440 0.68672 0.004 0.056 0.756 0.004 0.160 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) other(p) protocol(p) k
#> SD:NMF 84 9.59e-02 5.48e-07 4.41e-05 2
#> SD:NMF 62 4.45e-02 1.51e-07 6.20e-07 3
#> SD:NMF 56 9.28e-03 8.15e-08 8.74e-10 4
#> SD:NMF 38 1.12e-07 6.99e-10 1.45e-10 5
#> SD:NMF 54 2.10e-10 2.93e-11 4.51e-15 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 8353 rows and 87 columns.
#> Top rows (835, 1670, 2506, 3341, 4176) 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 0.474 0.848 0.904 0.4070 0.567 0.567
#> 3 3 0.387 0.699 0.877 0.1416 0.954 0.919
#> 4 4 0.385 0.692 0.872 0.0542 0.999 0.999
#> 5 5 0.384 0.662 0.861 0.0412 0.985 0.971
#> 6 6 0.398 0.672 0.845 0.0447 0.985 0.970
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
#> GSM39104 1 0.0376 0.9182 0.996 0.004
#> GSM39105 1 0.0672 0.9188 0.992 0.008
#> GSM39106 1 0.5294 0.8256 0.880 0.120
#> GSM39107 1 0.9460 0.3261 0.636 0.364
#> GSM39108 1 0.0938 0.9179 0.988 0.012
#> GSM39109 1 0.3879 0.8758 0.924 0.076
#> GSM39110 1 0.2236 0.9067 0.964 0.036
#> GSM39111 1 0.0938 0.9179 0.988 0.012
#> GSM39112 1 0.9427 0.3393 0.640 0.360
#> GSM39113 1 0.9522 0.2983 0.628 0.372
#> GSM39114 2 0.8207 0.8354 0.256 0.744
#> GSM39115 1 0.0672 0.9188 0.992 0.008
#> GSM39148 1 0.0672 0.9188 0.992 0.008
#> GSM39149 1 0.2603 0.8899 0.956 0.044
#> GSM39150 1 0.0376 0.9182 0.996 0.004
#> GSM39151 1 0.2236 0.8929 0.964 0.036
#> GSM39152 1 0.0376 0.9154 0.996 0.004
#> GSM39153 1 0.0938 0.9184 0.988 0.012
#> GSM39154 1 0.0938 0.9184 0.988 0.012
#> GSM39155 1 0.0672 0.9188 0.992 0.008
#> GSM39156 1 0.2948 0.8951 0.948 0.052
#> GSM39157 1 0.0672 0.9188 0.992 0.008
#> GSM39158 1 0.0672 0.9171 0.992 0.008
#> GSM39159 1 0.1843 0.9132 0.972 0.028
#> GSM39160 1 0.0672 0.9173 0.992 0.008
#> GSM39161 1 0.1843 0.9105 0.972 0.028
#> GSM39162 1 0.0672 0.9188 0.992 0.008
#> GSM39163 1 0.0672 0.9188 0.992 0.008
#> GSM39164 1 0.0672 0.9188 0.992 0.008
#> GSM39165 1 0.0376 0.9182 0.996 0.004
#> GSM39166 1 0.0672 0.9171 0.992 0.008
#> GSM39167 1 0.0938 0.9183 0.988 0.012
#> GSM39168 1 0.0672 0.9188 0.992 0.008
#> GSM39169 1 0.0672 0.9188 0.992 0.008
#> GSM39170 1 0.0000 0.9171 1.000 0.000
#> GSM39171 1 0.0672 0.9173 0.992 0.008
#> GSM39172 1 0.4815 0.8410 0.896 0.104
#> GSM39173 1 0.1633 0.9115 0.976 0.024
#> GSM39174 1 0.0672 0.9188 0.992 0.008
#> GSM39175 1 0.0672 0.9173 0.992 0.008
#> GSM39176 1 0.0938 0.9183 0.988 0.012
#> GSM39177 1 0.1184 0.9084 0.984 0.016
#> GSM39178 1 0.0672 0.9173 0.992 0.008
#> GSM39179 1 0.2236 0.8929 0.964 0.036
#> GSM39180 1 0.5629 0.8115 0.868 0.132
#> GSM39181 1 0.0672 0.9171 0.992 0.008
#> GSM39182 1 0.8555 0.5632 0.720 0.280
#> GSM39183 1 0.0672 0.9171 0.992 0.008
#> GSM39184 1 0.0672 0.9188 0.992 0.008
#> GSM39185 1 0.1843 0.9105 0.972 0.028
#> GSM39186 1 0.0672 0.9188 0.992 0.008
#> GSM39187 1 0.1633 0.9137 0.976 0.024
#> GSM39116 2 0.5842 0.9032 0.140 0.860
#> GSM39117 2 0.7883 0.8161 0.236 0.764
#> GSM39118 2 0.3584 0.8782 0.068 0.932
#> GSM39119 2 0.3274 0.8728 0.060 0.940
#> GSM39120 1 0.9491 0.3260 0.632 0.368
#> GSM39121 2 0.7950 0.8563 0.240 0.760
#> GSM39122 2 0.7883 0.8608 0.236 0.764
#> GSM39123 2 0.7883 0.8161 0.236 0.764
#> GSM39124 2 0.7602 0.8753 0.220 0.780
#> GSM39125 1 0.9866 0.0775 0.568 0.432
#> GSM39126 2 0.8081 0.8462 0.248 0.752
#> GSM39127 2 0.6712 0.8988 0.176 0.824
#> GSM39128 2 0.7219 0.8880 0.200 0.800
#> GSM39129 2 0.2778 0.8655 0.048 0.952
#> GSM39130 2 0.7883 0.8161 0.236 0.764
#> GSM39131 2 0.6438 0.9024 0.164 0.836
#> GSM39132 2 0.6343 0.9028 0.160 0.840
#> GSM39133 2 0.5294 0.8832 0.120 0.880
#> GSM39134 2 0.3584 0.8774 0.068 0.932
#> GSM39135 2 0.5842 0.9032 0.140 0.860
#> GSM39136 2 0.5737 0.9027 0.136 0.864
#> GSM39137 2 0.7602 0.8753 0.220 0.780
#> GSM39138 2 0.2778 0.8655 0.048 0.952
#> GSM39139 2 0.2778 0.8655 0.048 0.952
#> GSM39140 1 0.8207 0.6176 0.744 0.256
#> GSM39141 1 0.7056 0.7298 0.808 0.192
#> GSM39142 1 0.6343 0.7759 0.840 0.160
#> GSM39143 1 0.6343 0.7759 0.840 0.160
#> GSM39144 2 0.2778 0.8655 0.048 0.952
#> GSM39145 2 0.5059 0.8960 0.112 0.888
#> GSM39146 2 0.6623 0.9013 0.172 0.828
#> GSM39147 2 0.7602 0.8753 0.220 0.780
#> GSM39188 1 0.2778 0.8817 0.952 0.048
#> GSM39189 1 0.0000 0.9171 1.000 0.000
#> GSM39190 1 0.2423 0.8965 0.960 0.040
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM39104 1 0.0000 0.8092 1.000 0.000 0.000
#> GSM39105 1 0.0237 0.8113 0.996 0.004 0.000
#> GSM39106 1 0.3267 0.6738 0.884 0.116 0.000
#> GSM39107 1 0.5948 0.1674 0.640 0.360 0.000
#> GSM39108 1 0.0661 0.8095 0.988 0.008 0.004
#> GSM39109 1 0.2845 0.7418 0.920 0.068 0.012
#> GSM39110 1 0.1525 0.7908 0.964 0.032 0.004
#> GSM39111 1 0.0661 0.8095 0.988 0.008 0.004
#> GSM39112 1 0.5926 0.1754 0.644 0.356 0.000
#> GSM39113 1 0.5988 0.1502 0.632 0.368 0.000
#> GSM39114 2 0.5016 0.7768 0.240 0.760 0.000
#> GSM39115 1 0.0237 0.8113 0.996 0.004 0.000
#> GSM39148 1 0.0237 0.8113 0.996 0.004 0.000
#> GSM39149 3 0.6489 0.7266 0.456 0.004 0.540
#> GSM39150 1 0.0000 0.8092 1.000 0.000 0.000
#> GSM39151 1 0.6169 -0.3730 0.636 0.004 0.360
#> GSM39152 1 0.0661 0.8023 0.988 0.004 0.008
#> GSM39153 1 0.0475 0.8104 0.992 0.004 0.004
#> GSM39154 1 0.0475 0.8104 0.992 0.004 0.004
#> GSM39155 1 0.0237 0.8113 0.996 0.004 0.000
#> GSM39156 1 0.1860 0.7717 0.948 0.052 0.000
#> GSM39157 1 0.0237 0.8113 0.996 0.004 0.000
#> GSM39158 1 0.0592 0.8065 0.988 0.012 0.000
#> GSM39159 1 0.1129 0.8004 0.976 0.020 0.004
#> GSM39160 1 0.0237 0.8082 0.996 0.000 0.004
#> GSM39161 1 0.1774 0.7878 0.960 0.024 0.016
#> GSM39162 1 0.0237 0.8113 0.996 0.004 0.000
#> GSM39163 1 0.0237 0.8113 0.996 0.004 0.000
#> GSM39164 1 0.0237 0.8113 0.996 0.004 0.000
#> GSM39165 1 0.0237 0.8087 0.996 0.000 0.004
#> GSM39166 1 0.0592 0.8065 0.988 0.012 0.000
#> GSM39167 1 0.0424 0.8107 0.992 0.008 0.000
#> GSM39168 1 0.0237 0.8113 0.996 0.004 0.000
#> GSM39169 1 0.0237 0.8113 0.996 0.004 0.000
#> GSM39170 1 0.0237 0.8059 0.996 0.004 0.000
#> GSM39171 1 0.0237 0.8082 0.996 0.000 0.004
#> GSM39172 1 0.4982 0.5889 0.840 0.096 0.064
#> GSM39173 1 0.5036 0.4953 0.808 0.020 0.172
#> GSM39174 1 0.0237 0.8113 0.996 0.004 0.000
#> GSM39175 1 0.0237 0.8082 0.996 0.000 0.004
#> GSM39176 1 0.0424 0.8107 0.992 0.008 0.000
#> GSM39177 1 0.3030 0.6883 0.904 0.004 0.092
#> GSM39178 1 0.0237 0.8082 0.996 0.000 0.004
#> GSM39179 1 0.6434 -0.4395 0.612 0.008 0.380
#> GSM39180 1 0.7762 0.0472 0.668 0.120 0.212
#> GSM39181 1 0.0592 0.8065 0.988 0.012 0.000
#> GSM39182 1 0.7213 0.1546 0.668 0.272 0.060
#> GSM39183 1 0.0592 0.8065 0.988 0.012 0.000
#> GSM39184 1 0.0237 0.8113 0.996 0.004 0.000
#> GSM39185 1 0.1774 0.7878 0.960 0.024 0.016
#> GSM39186 1 0.0237 0.8113 0.996 0.004 0.000
#> GSM39187 1 0.0892 0.8041 0.980 0.020 0.000
#> GSM39116 2 0.3644 0.8589 0.124 0.872 0.004
#> GSM39117 2 0.6402 0.7252 0.200 0.744 0.056
#> GSM39118 2 0.1765 0.8176 0.040 0.956 0.004
#> GSM39119 2 0.2879 0.8180 0.052 0.924 0.024
#> GSM39120 1 0.6008 0.1608 0.628 0.372 0.000
#> GSM39121 2 0.4842 0.8020 0.224 0.776 0.000
#> GSM39122 2 0.4796 0.8074 0.220 0.780 0.000
#> GSM39123 2 0.6402 0.7252 0.200 0.744 0.056
#> GSM39124 2 0.4605 0.8247 0.204 0.796 0.000
#> GSM39125 1 0.6235 0.0413 0.564 0.436 0.000
#> GSM39126 2 0.4931 0.7900 0.232 0.768 0.000
#> GSM39127 2 0.4002 0.8523 0.160 0.840 0.000
#> GSM39128 2 0.4346 0.8395 0.184 0.816 0.000
#> GSM39129 2 0.2063 0.7746 0.008 0.948 0.044
#> GSM39130 2 0.6402 0.7252 0.200 0.744 0.056
#> GSM39131 2 0.3983 0.8581 0.144 0.852 0.004
#> GSM39132 2 0.3918 0.8585 0.140 0.856 0.004
#> GSM39133 2 0.4217 0.8175 0.100 0.868 0.032
#> GSM39134 2 0.1950 0.8166 0.040 0.952 0.008
#> GSM39135 2 0.3644 0.8589 0.124 0.872 0.004
#> GSM39136 2 0.3500 0.8572 0.116 0.880 0.004
#> GSM39137 2 0.4605 0.8247 0.204 0.796 0.000
#> GSM39138 2 0.2063 0.7745 0.008 0.948 0.044
#> GSM39139 2 0.2173 0.7725 0.008 0.944 0.048
#> GSM39140 1 0.5178 0.4040 0.744 0.256 0.000
#> GSM39141 1 0.4399 0.5456 0.812 0.188 0.000
#> GSM39142 1 0.3941 0.6072 0.844 0.156 0.000
#> GSM39143 1 0.3941 0.6072 0.844 0.156 0.000
#> GSM39144 2 0.2173 0.7725 0.008 0.944 0.048
#> GSM39145 2 0.2955 0.8417 0.080 0.912 0.008
#> GSM39146 2 0.4228 0.8577 0.148 0.844 0.008
#> GSM39147 2 0.4605 0.8247 0.204 0.796 0.000
#> GSM39188 3 0.6460 0.7425 0.440 0.004 0.556
#> GSM39189 1 0.3425 0.6587 0.884 0.004 0.112
#> GSM39190 3 0.6672 0.7139 0.472 0.008 0.520
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM39104 1 0.0000 0.8278 1.000 0.000 0.000 0.000
#> GSM39105 1 0.0188 0.8294 0.996 0.004 0.000 0.000
#> GSM39106 1 0.2831 0.7157 0.876 0.120 0.004 0.000
#> GSM39107 1 0.4936 0.2661 0.624 0.372 0.004 0.000
#> GSM39108 1 0.0804 0.8253 0.980 0.008 0.012 0.000
#> GSM39109 1 0.2489 0.7704 0.912 0.068 0.020 0.000
#> GSM39110 1 0.1488 0.8100 0.956 0.032 0.012 0.000
#> GSM39111 1 0.0804 0.8253 0.980 0.008 0.012 0.000
#> GSM39112 1 0.4920 0.2741 0.628 0.368 0.004 0.000
#> GSM39113 1 0.4964 0.2489 0.616 0.380 0.004 0.000
#> GSM39114 2 0.3873 0.7664 0.228 0.772 0.000 0.000
#> GSM39115 1 0.0188 0.8294 0.996 0.004 0.000 0.000
#> GSM39148 1 0.0188 0.8294 0.996 0.004 0.000 0.000
#> GSM39149 3 0.6919 0.4128 0.352 0.000 0.528 0.120
#> GSM39150 1 0.0000 0.8278 1.000 0.000 0.000 0.000
#> GSM39151 1 0.7270 -0.3673 0.548 0.004 0.280 0.168
#> GSM39152 1 0.0524 0.8230 0.988 0.000 0.008 0.004
#> GSM39153 1 0.0376 0.8288 0.992 0.004 0.004 0.000
#> GSM39154 1 0.0376 0.8288 0.992 0.004 0.004 0.000
#> GSM39155 1 0.0188 0.8294 0.996 0.004 0.000 0.000
#> GSM39156 1 0.1474 0.7991 0.948 0.052 0.000 0.000
#> GSM39157 1 0.0188 0.8294 0.996 0.004 0.000 0.000
#> GSM39158 1 0.0524 0.8260 0.988 0.008 0.000 0.004
#> GSM39159 1 0.1059 0.8189 0.972 0.016 0.012 0.000
#> GSM39160 1 0.0188 0.8272 0.996 0.000 0.004 0.000
#> GSM39161 1 0.1593 0.8078 0.956 0.016 0.024 0.004
#> GSM39162 1 0.0188 0.8294 0.996 0.004 0.000 0.000
#> GSM39163 1 0.0188 0.8294 0.996 0.004 0.000 0.000
#> GSM39164 1 0.0188 0.8294 0.996 0.004 0.000 0.000
#> GSM39165 1 0.0188 0.8277 0.996 0.000 0.004 0.000
#> GSM39166 1 0.0524 0.8260 0.988 0.008 0.000 0.004
#> GSM39167 1 0.0336 0.8289 0.992 0.008 0.000 0.000
#> GSM39168 1 0.0188 0.8294 0.996 0.004 0.000 0.000
#> GSM39169 1 0.0188 0.8294 0.996 0.004 0.000 0.000
#> GSM39170 1 0.0188 0.8253 0.996 0.000 0.000 0.004
#> GSM39171 1 0.0188 0.8272 0.996 0.000 0.004 0.000
#> GSM39172 1 0.4977 0.6034 0.804 0.096 0.072 0.028
#> GSM39173 1 0.5403 0.4362 0.732 0.024 0.216 0.028
#> GSM39174 1 0.0188 0.8294 0.996 0.004 0.000 0.000
#> GSM39175 1 0.0188 0.8272 0.996 0.000 0.004 0.000
#> GSM39176 1 0.0336 0.8289 0.992 0.008 0.000 0.000
#> GSM39177 1 0.3239 0.7075 0.880 0.000 0.052 0.068
#> GSM39178 1 0.0188 0.8272 0.996 0.000 0.004 0.000
#> GSM39179 1 0.7001 -0.3705 0.544 0.000 0.316 0.140
#> GSM39180 1 0.7718 -0.0724 0.572 0.120 0.260 0.048
#> GSM39181 1 0.0524 0.8260 0.988 0.008 0.000 0.004
#> GSM39182 1 0.6557 0.2432 0.636 0.268 0.080 0.016
#> GSM39183 1 0.0524 0.8260 0.988 0.008 0.000 0.004
#> GSM39184 1 0.0188 0.8294 0.996 0.004 0.000 0.000
#> GSM39185 1 0.1593 0.8078 0.956 0.016 0.024 0.004
#> GSM39186 1 0.0188 0.8294 0.996 0.004 0.000 0.000
#> GSM39187 1 0.0707 0.8240 0.980 0.020 0.000 0.000
#> GSM39116 2 0.2839 0.8408 0.108 0.884 0.004 0.004
#> GSM39117 2 0.5679 0.7084 0.160 0.744 0.076 0.020
#> GSM39118 2 0.1590 0.7862 0.028 0.956 0.008 0.008
#> GSM39119 2 0.2669 0.7729 0.032 0.912 0.052 0.004
#> GSM39120 1 0.4978 0.2599 0.612 0.384 0.004 0.000
#> GSM39121 2 0.3726 0.7908 0.212 0.788 0.000 0.000
#> GSM39122 2 0.3688 0.7959 0.208 0.792 0.000 0.000
#> GSM39123 2 0.5679 0.7084 0.160 0.744 0.076 0.020
#> GSM39124 2 0.3528 0.8125 0.192 0.808 0.000 0.000
#> GSM39125 1 0.5132 0.0932 0.548 0.448 0.004 0.000
#> GSM39126 2 0.3801 0.7794 0.220 0.780 0.000 0.000
#> GSM39127 2 0.3024 0.8384 0.148 0.852 0.000 0.000
#> GSM39128 2 0.3311 0.8270 0.172 0.828 0.000 0.000
#> GSM39129 2 0.3577 0.6963 0.004 0.868 0.056 0.072
#> GSM39130 2 0.5679 0.7084 0.160 0.744 0.076 0.020
#> GSM39131 2 0.2999 0.8424 0.132 0.864 0.000 0.004
#> GSM39132 2 0.2944 0.8425 0.128 0.868 0.000 0.004
#> GSM39133 2 0.3705 0.7830 0.076 0.868 0.040 0.016
#> GSM39134 2 0.2210 0.7801 0.028 0.936 0.016 0.020
#> GSM39135 2 0.2839 0.8408 0.108 0.884 0.004 0.004
#> GSM39136 2 0.2715 0.8380 0.100 0.892 0.004 0.004
#> GSM39137 2 0.3528 0.8125 0.192 0.808 0.000 0.000
#> GSM39138 2 0.3241 0.7062 0.004 0.884 0.040 0.072
#> GSM39139 2 0.3400 0.7014 0.004 0.876 0.044 0.076
#> GSM39140 1 0.4193 0.4891 0.732 0.268 0.000 0.000
#> GSM39141 1 0.3610 0.6064 0.800 0.200 0.000 0.000
#> GSM39142 1 0.3219 0.6605 0.836 0.164 0.000 0.000
#> GSM39143 1 0.3219 0.6605 0.836 0.164 0.000 0.000
#> GSM39144 2 0.3648 0.6909 0.004 0.864 0.056 0.076
#> GSM39145 2 0.3328 0.8185 0.076 0.884 0.020 0.020
#> GSM39146 2 0.3326 0.8431 0.132 0.856 0.008 0.004
#> GSM39147 2 0.3528 0.8125 0.192 0.808 0.000 0.000
#> GSM39188 4 0.2999 0.0000 0.132 0.000 0.004 0.864
#> GSM39189 1 0.3612 0.6523 0.840 0.004 0.144 0.012
#> GSM39190 3 0.6808 0.3856 0.320 0.000 0.560 0.120
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM39104 1 0.0000 0.8162 1.000 0.000 0.000 0.000 0.000
#> GSM39105 1 0.0162 0.8179 0.996 0.004 0.000 0.000 0.000
#> GSM39106 1 0.2488 0.6920 0.872 0.124 0.004 0.000 0.000
#> GSM39107 1 0.4276 0.2177 0.616 0.380 0.004 0.000 0.000
#> GSM39108 1 0.0740 0.8133 0.980 0.008 0.008 0.000 0.004
#> GSM39109 1 0.2585 0.7392 0.896 0.072 0.008 0.000 0.024
#> GSM39110 1 0.1682 0.7886 0.944 0.032 0.012 0.000 0.012
#> GSM39111 1 0.0740 0.8133 0.980 0.008 0.008 0.000 0.004
#> GSM39112 1 0.4264 0.2260 0.620 0.376 0.004 0.000 0.000
#> GSM39113 1 0.4299 0.2003 0.608 0.388 0.004 0.000 0.000
#> GSM39114 2 0.3274 0.7520 0.220 0.780 0.000 0.000 0.000
#> GSM39115 1 0.0162 0.8179 0.996 0.004 0.000 0.000 0.000
#> GSM39148 1 0.0162 0.8179 0.996 0.004 0.000 0.000 0.000
#> GSM39149 3 0.7694 0.2501 0.244 0.000 0.460 0.084 0.212
#> GSM39150 1 0.0000 0.8162 1.000 0.000 0.000 0.000 0.000
#> GSM39151 5 0.7603 0.0000 0.380 0.000 0.124 0.100 0.396
#> GSM39152 1 0.0566 0.8091 0.984 0.004 0.000 0.000 0.012
#> GSM39153 1 0.0324 0.8172 0.992 0.004 0.000 0.000 0.004
#> GSM39154 1 0.0324 0.8172 0.992 0.004 0.000 0.000 0.004
#> GSM39155 1 0.0162 0.8179 0.996 0.004 0.000 0.000 0.000
#> GSM39156 1 0.1270 0.7855 0.948 0.052 0.000 0.000 0.000
#> GSM39157 1 0.0162 0.8179 0.996 0.004 0.000 0.000 0.000
#> GSM39158 1 0.0404 0.8142 0.988 0.012 0.000 0.000 0.000
#> GSM39159 1 0.0960 0.8065 0.972 0.016 0.004 0.000 0.008
#> GSM39160 1 0.0162 0.8154 0.996 0.000 0.000 0.000 0.004
#> GSM39161 1 0.1471 0.7910 0.952 0.020 0.004 0.000 0.024
#> GSM39162 1 0.0162 0.8179 0.996 0.004 0.000 0.000 0.000
#> GSM39163 1 0.0162 0.8179 0.996 0.004 0.000 0.000 0.000
#> GSM39164 1 0.0162 0.8179 0.996 0.004 0.000 0.000 0.000
#> GSM39165 1 0.0451 0.8133 0.988 0.000 0.004 0.000 0.008
#> GSM39166 1 0.0404 0.8142 0.988 0.012 0.000 0.000 0.000
#> GSM39167 1 0.0290 0.8174 0.992 0.008 0.000 0.000 0.000
#> GSM39168 1 0.0162 0.8179 0.996 0.004 0.000 0.000 0.000
#> GSM39169 1 0.0162 0.8179 0.996 0.004 0.000 0.000 0.000
#> GSM39170 1 0.0162 0.8135 0.996 0.004 0.000 0.000 0.000
#> GSM39171 1 0.0162 0.8154 0.996 0.000 0.000 0.000 0.004
#> GSM39172 1 0.4812 0.5388 0.784 0.108 0.040 0.012 0.056
#> GSM39173 1 0.6004 0.1911 0.664 0.028 0.160 0.004 0.144
#> GSM39174 1 0.0162 0.8179 0.996 0.004 0.000 0.000 0.000
#> GSM39175 1 0.0162 0.8154 0.996 0.000 0.000 0.000 0.004
#> GSM39176 1 0.0290 0.8174 0.992 0.008 0.000 0.000 0.000
#> GSM39177 1 0.3581 0.6328 0.852 0.004 0.020 0.044 0.080
#> GSM39178 1 0.0162 0.8154 0.996 0.000 0.000 0.000 0.004
#> GSM39179 1 0.7832 -0.7448 0.436 0.004 0.204 0.076 0.280
#> GSM39180 1 0.7933 -0.2609 0.512 0.132 0.188 0.016 0.152
#> GSM39181 1 0.0404 0.8142 0.988 0.012 0.000 0.000 0.000
#> GSM39182 1 0.6023 0.1637 0.620 0.276 0.036 0.004 0.064
#> GSM39183 1 0.0404 0.8142 0.988 0.012 0.000 0.000 0.000
#> GSM39184 1 0.0162 0.8179 0.996 0.004 0.000 0.000 0.000
#> GSM39185 1 0.1471 0.7910 0.952 0.020 0.004 0.000 0.024
#> GSM39186 1 0.0162 0.8179 0.996 0.004 0.000 0.000 0.000
#> GSM39187 1 0.0609 0.8121 0.980 0.020 0.000 0.000 0.000
#> GSM39116 2 0.2519 0.8220 0.100 0.884 0.000 0.000 0.016
#> GSM39117 2 0.5590 0.6652 0.136 0.716 0.032 0.008 0.108
#> GSM39118 2 0.1493 0.7699 0.024 0.948 0.000 0.000 0.028
#> GSM39119 2 0.2482 0.7446 0.024 0.892 0.000 0.000 0.084
#> GSM39120 1 0.4299 0.2202 0.608 0.388 0.004 0.000 0.000
#> GSM39121 2 0.3143 0.7754 0.204 0.796 0.000 0.000 0.000
#> GSM39122 2 0.3109 0.7801 0.200 0.800 0.000 0.000 0.000
#> GSM39123 2 0.5590 0.6652 0.136 0.716 0.032 0.008 0.108
#> GSM39124 2 0.2966 0.7953 0.184 0.816 0.000 0.000 0.000
#> GSM39125 1 0.4425 0.0986 0.544 0.452 0.004 0.000 0.000
#> GSM39126 2 0.3210 0.7646 0.212 0.788 0.000 0.000 0.000
#> GSM39127 2 0.2516 0.8192 0.140 0.860 0.000 0.000 0.000
#> GSM39128 2 0.2773 0.8083 0.164 0.836 0.000 0.000 0.000
#> GSM39129 2 0.3491 0.6127 0.004 0.768 0.000 0.000 0.228
#> GSM39130 2 0.5590 0.6652 0.136 0.716 0.032 0.008 0.108
#> GSM39131 2 0.2488 0.8236 0.124 0.872 0.000 0.000 0.004
#> GSM39132 2 0.2563 0.8237 0.120 0.872 0.000 0.000 0.008
#> GSM39133 2 0.3777 0.7410 0.056 0.836 0.008 0.008 0.092
#> GSM39134 2 0.2208 0.7534 0.020 0.908 0.000 0.000 0.072
#> GSM39135 2 0.2416 0.8220 0.100 0.888 0.000 0.000 0.012
#> GSM39136 2 0.2408 0.8195 0.092 0.892 0.000 0.000 0.016
#> GSM39137 2 0.2966 0.7953 0.184 0.816 0.000 0.000 0.000
#> GSM39138 2 0.3333 0.6305 0.004 0.788 0.000 0.000 0.208
#> GSM39139 2 0.3333 0.6296 0.004 0.788 0.000 0.000 0.208
#> GSM39140 1 0.3612 0.4606 0.732 0.268 0.000 0.000 0.000
#> GSM39141 1 0.3109 0.5836 0.800 0.200 0.000 0.000 0.000
#> GSM39142 1 0.2773 0.6399 0.836 0.164 0.000 0.000 0.000
#> GSM39143 1 0.2773 0.6399 0.836 0.164 0.000 0.000 0.000
#> GSM39144 2 0.3491 0.6095 0.004 0.768 0.000 0.000 0.228
#> GSM39145 2 0.3354 0.7848 0.068 0.844 0.000 0.000 0.088
#> GSM39146 2 0.2825 0.8243 0.124 0.860 0.000 0.000 0.016
#> GSM39147 2 0.2966 0.7953 0.184 0.816 0.000 0.000 0.000
#> GSM39188 4 0.0290 0.0000 0.008 0.000 0.000 0.992 0.000
#> GSM39189 1 0.4253 0.5511 0.796 0.012 0.100 0.000 0.092
#> GSM39190 3 0.3942 0.3724 0.232 0.000 0.748 0.020 0.000
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM39104 1 0.0000 0.8435 1.000 0.000 0.000 0.000 NA 0.000
#> GSM39105 1 0.0146 0.8449 0.996 0.004 0.000 0.000 NA 0.000
#> GSM39106 1 0.2489 0.7371 0.860 0.128 0.012 0.000 NA 0.000
#> GSM39107 1 0.4131 0.3097 0.600 0.384 0.016 0.000 NA 0.000
#> GSM39108 1 0.1086 0.8349 0.964 0.012 0.012 0.000 NA 0.012
#> GSM39109 1 0.3051 0.7599 0.864 0.076 0.036 0.000 NA 0.008
#> GSM39110 1 0.2451 0.7968 0.904 0.036 0.024 0.000 NA 0.028
#> GSM39111 1 0.1086 0.8349 0.964 0.012 0.012 0.000 NA 0.012
#> GSM39112 1 0.4121 0.3178 0.604 0.380 0.016 0.000 NA 0.000
#> GSM39113 1 0.4150 0.2926 0.592 0.392 0.016 0.000 NA 0.000
#> GSM39114 2 0.3103 0.7210 0.208 0.784 0.008 0.000 NA 0.000
#> GSM39115 1 0.0146 0.8449 0.996 0.004 0.000 0.000 NA 0.000
#> GSM39148 1 0.0146 0.8449 0.996 0.004 0.000 0.000 NA 0.000
#> GSM39149 3 0.6801 0.2138 0.172 0.000 0.516 0.052 NA 0.240
#> GSM39150 1 0.0000 0.8435 1.000 0.000 0.000 0.000 NA 0.000
#> GSM39151 6 0.3974 0.0536 0.224 0.000 0.000 0.048 NA 0.728
#> GSM39152 1 0.0622 0.8366 0.980 0.000 0.012 0.000 NA 0.008
#> GSM39153 1 0.0291 0.8446 0.992 0.004 0.000 0.000 NA 0.004
#> GSM39154 1 0.0291 0.8446 0.992 0.004 0.000 0.000 NA 0.004
#> GSM39155 1 0.0146 0.8449 0.996 0.004 0.000 0.000 NA 0.000
#> GSM39156 1 0.1349 0.8166 0.940 0.056 0.004 0.000 NA 0.000
#> GSM39157 1 0.0146 0.8449 0.996 0.004 0.000 0.000 NA 0.000
#> GSM39158 1 0.0405 0.8423 0.988 0.008 0.004 0.000 NA 0.000
#> GSM39159 1 0.1173 0.8298 0.960 0.016 0.000 0.000 NA 0.016
#> GSM39160 1 0.0146 0.8432 0.996 0.000 0.000 0.000 NA 0.004
#> GSM39161 1 0.1627 0.8185 0.944 0.016 0.016 0.000 NA 0.008
#> GSM39162 1 0.0146 0.8449 0.996 0.004 0.000 0.000 NA 0.000
#> GSM39163 1 0.0146 0.8449 0.996 0.004 0.000 0.000 NA 0.000
#> GSM39164 1 0.0146 0.8449 0.996 0.004 0.000 0.000 NA 0.000
#> GSM39165 1 0.0837 0.8368 0.972 0.000 0.020 0.000 NA 0.004
#> GSM39166 1 0.0405 0.8423 0.988 0.008 0.004 0.000 NA 0.000
#> GSM39167 1 0.0260 0.8446 0.992 0.008 0.000 0.000 NA 0.000
#> GSM39168 1 0.0146 0.8449 0.996 0.004 0.000 0.000 NA 0.000
#> GSM39169 1 0.0146 0.8449 0.996 0.004 0.000 0.000 NA 0.000
#> GSM39170 1 0.0146 0.8415 0.996 0.000 0.004 0.000 NA 0.000
#> GSM39171 1 0.0291 0.8440 0.992 0.000 0.004 0.000 NA 0.004
#> GSM39172 1 0.4754 0.5631 0.748 0.108 0.068 0.000 NA 0.004
#> GSM39173 1 0.5391 0.2190 0.604 0.024 0.316 0.004 NA 0.036
#> GSM39174 1 0.0146 0.8449 0.996 0.004 0.000 0.000 NA 0.000
#> GSM39175 1 0.0146 0.8432 0.996 0.000 0.000 0.000 NA 0.004
#> GSM39176 1 0.0260 0.8446 0.992 0.008 0.000 0.000 NA 0.000
#> GSM39177 1 0.3463 0.6230 0.800 0.000 0.032 0.008 NA 0.160
#> GSM39178 1 0.0146 0.8432 0.996 0.000 0.000 0.000 NA 0.004
#> GSM39179 6 0.7407 0.0740 0.312 0.000 0.236 0.012 NA 0.360
#> GSM39180 1 0.7212 -0.2369 0.448 0.132 0.320 0.008 NA 0.016
#> GSM39181 1 0.0405 0.8423 0.988 0.008 0.004 0.000 NA 0.000
#> GSM39182 1 0.5781 0.2392 0.588 0.272 0.056 0.000 NA 0.000
#> GSM39183 1 0.0405 0.8423 0.988 0.008 0.004 0.000 NA 0.000
#> GSM39184 1 0.0146 0.8449 0.996 0.004 0.000 0.000 NA 0.000
#> GSM39185 1 0.1729 0.8155 0.940 0.016 0.016 0.000 NA 0.012
#> GSM39186 1 0.0146 0.8449 0.996 0.004 0.000 0.000 NA 0.000
#> GSM39187 1 0.0777 0.8381 0.972 0.024 0.004 0.000 NA 0.000
#> GSM39116 2 0.2163 0.7844 0.092 0.892 0.000 0.000 NA 0.000
#> GSM39117 2 0.5399 0.5677 0.104 0.652 0.040 0.000 NA 0.000
#> GSM39118 2 0.1829 0.7344 0.024 0.920 0.000 0.000 NA 0.000
#> GSM39119 2 0.2886 0.6830 0.016 0.836 0.000 0.000 NA 0.004
#> GSM39120 1 0.4150 0.3091 0.592 0.392 0.016 0.000 NA 0.000
#> GSM39121 2 0.2980 0.7432 0.192 0.800 0.008 0.000 NA 0.000
#> GSM39122 2 0.2948 0.7478 0.188 0.804 0.008 0.000 NA 0.000
#> GSM39123 2 0.5399 0.5677 0.104 0.652 0.040 0.000 NA 0.000
#> GSM39124 2 0.2814 0.7627 0.172 0.820 0.008 0.000 NA 0.000
#> GSM39125 1 0.4250 0.0898 0.528 0.456 0.016 0.000 NA 0.000
#> GSM39126 2 0.3043 0.7325 0.200 0.792 0.008 0.000 NA 0.000
#> GSM39127 2 0.2219 0.7827 0.136 0.864 0.000 0.000 NA 0.000
#> GSM39128 2 0.2631 0.7748 0.152 0.840 0.008 0.000 NA 0.000
#> GSM39129 2 0.3717 0.4291 0.000 0.616 0.000 0.000 NA 0.000
#> GSM39130 2 0.5399 0.5677 0.104 0.652 0.040 0.000 NA 0.000
#> GSM39131 2 0.2146 0.7872 0.116 0.880 0.000 0.000 NA 0.000
#> GSM39132 2 0.2212 0.7871 0.112 0.880 0.000 0.000 NA 0.000
#> GSM39133 2 0.3620 0.6530 0.044 0.772 0.000 0.000 NA 0.000
#> GSM39134 2 0.2704 0.6879 0.016 0.844 0.000 0.000 NA 0.000
#> GSM39135 2 0.2070 0.7844 0.092 0.896 0.000 0.000 NA 0.000
#> GSM39136 2 0.2147 0.7813 0.084 0.896 0.000 0.000 NA 0.000
#> GSM39137 2 0.2814 0.7627 0.172 0.820 0.008 0.000 NA 0.000
#> GSM39138 2 0.3659 0.4602 0.000 0.636 0.000 0.000 NA 0.000
#> GSM39139 2 0.3620 0.4770 0.000 0.648 0.000 0.000 NA 0.000
#> GSM39140 1 0.3512 0.5393 0.720 0.272 0.008 0.000 NA 0.000
#> GSM39141 1 0.3073 0.6449 0.788 0.204 0.008 0.000 NA 0.000
#> GSM39142 1 0.2778 0.6921 0.824 0.168 0.008 0.000 NA 0.000
#> GSM39143 1 0.2778 0.6921 0.824 0.168 0.008 0.000 NA 0.000
#> GSM39144 2 0.3747 0.4245 0.000 0.604 0.000 0.000 NA 0.000
#> GSM39145 2 0.3570 0.7232 0.064 0.792 0.000 0.000 NA 0.000
#> GSM39146 2 0.2699 0.7878 0.108 0.864 0.008 0.000 NA 0.000
#> GSM39147 2 0.2814 0.7627 0.172 0.820 0.008 0.000 NA 0.000
#> GSM39188 4 0.0146 0.0000 0.000 0.000 0.000 0.996 NA 0.004
#> GSM39189 1 0.3703 0.6009 0.768 0.008 0.200 0.000 NA 0.004
#> GSM39190 3 0.6719 0.3414 0.176 0.000 0.464 0.020 NA 0.028
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) other(p) protocol(p) k
#> CV:hclust 82 0.271 1.83e-11 1.99e-12 2
#> CV:hclust 76 0.176 4.10e-10 3.40e-10 3
#> CV:hclust 73 0.177 9.63e-12 6.37e-11 4
#> CV:hclust 73 0.177 9.63e-12 6.37e-11 5
#> CV:hclust 70 0.268 2.72e-10 2.61e-12 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 8353 rows and 87 columns.
#> Top rows (835, 1670, 2506, 3341, 4176) 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 0.735 0.891 0.945 0.4658 0.513 0.513
#> 3 3 0.801 0.868 0.931 0.2469 0.866 0.755
#> 4 4 0.668 0.742 0.845 0.1214 0.896 0.773
#> 5 5 0.618 0.623 0.803 0.0799 0.916 0.782
#> 6 6 0.634 0.594 0.767 0.0562 0.947 0.836
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
#> GSM39104 1 0.0000 0.978 1.000 0.000
#> GSM39105 1 0.0000 0.978 1.000 0.000
#> GSM39106 1 0.0000 0.978 1.000 0.000
#> GSM39107 1 0.0000 0.978 1.000 0.000
#> GSM39108 1 0.0000 0.978 1.000 0.000
#> GSM39109 2 0.9754 0.473 0.408 0.592
#> GSM39110 1 0.0000 0.978 1.000 0.000
#> GSM39111 1 0.0000 0.978 1.000 0.000
#> GSM39112 1 0.0000 0.978 1.000 0.000
#> GSM39113 1 0.0000 0.978 1.000 0.000
#> GSM39114 2 0.6712 0.769 0.176 0.824
#> GSM39115 1 0.0000 0.978 1.000 0.000
#> GSM39148 1 0.0000 0.978 1.000 0.000
#> GSM39149 2 0.8386 0.719 0.268 0.732
#> GSM39150 1 0.0000 0.978 1.000 0.000
#> GSM39151 2 0.8443 0.715 0.272 0.728
#> GSM39152 1 0.0000 0.978 1.000 0.000
#> GSM39153 1 0.0000 0.978 1.000 0.000
#> GSM39154 1 0.0000 0.978 1.000 0.000
#> GSM39155 1 0.0000 0.978 1.000 0.000
#> GSM39156 1 0.0000 0.978 1.000 0.000
#> GSM39157 1 0.0000 0.978 1.000 0.000
#> GSM39158 1 0.0000 0.978 1.000 0.000
#> GSM39159 1 0.0000 0.978 1.000 0.000
#> GSM39160 1 0.0000 0.978 1.000 0.000
#> GSM39161 1 0.0000 0.978 1.000 0.000
#> GSM39162 1 0.0000 0.978 1.000 0.000
#> GSM39163 1 0.0000 0.978 1.000 0.000
#> GSM39164 1 0.0000 0.978 1.000 0.000
#> GSM39165 1 0.0000 0.978 1.000 0.000
#> GSM39166 1 0.0000 0.978 1.000 0.000
#> GSM39167 1 0.0000 0.978 1.000 0.000
#> GSM39168 1 0.0000 0.978 1.000 0.000
#> GSM39169 1 0.0000 0.978 1.000 0.000
#> GSM39170 1 0.0000 0.978 1.000 0.000
#> GSM39171 1 0.0000 0.978 1.000 0.000
#> GSM39172 2 0.8386 0.719 0.268 0.732
#> GSM39173 2 0.8327 0.723 0.264 0.736
#> GSM39174 1 0.0000 0.978 1.000 0.000
#> GSM39175 1 0.0000 0.978 1.000 0.000
#> GSM39176 1 0.0000 0.978 1.000 0.000
#> GSM39177 1 0.8608 0.535 0.716 0.284
#> GSM39178 1 0.0000 0.978 1.000 0.000
#> GSM39179 2 0.8386 0.719 0.268 0.732
#> GSM39180 2 0.4562 0.843 0.096 0.904
#> GSM39181 1 0.0000 0.978 1.000 0.000
#> GSM39182 2 0.9686 0.498 0.396 0.604
#> GSM39183 1 0.0000 0.978 1.000 0.000
#> GSM39184 1 0.0000 0.978 1.000 0.000
#> GSM39185 1 0.0000 0.978 1.000 0.000
#> GSM39186 1 0.0000 0.978 1.000 0.000
#> GSM39187 1 0.0000 0.978 1.000 0.000
#> GSM39116 2 0.0000 0.880 0.000 1.000
#> GSM39117 2 0.0000 0.880 0.000 1.000
#> GSM39118 2 0.0000 0.880 0.000 1.000
#> GSM39119 2 0.0000 0.880 0.000 1.000
#> GSM39120 1 0.0000 0.978 1.000 0.000
#> GSM39121 1 0.6801 0.754 0.820 0.180
#> GSM39122 1 0.8661 0.547 0.712 0.288
#> GSM39123 2 0.0000 0.880 0.000 1.000
#> GSM39124 2 0.6247 0.788 0.156 0.844
#> GSM39125 1 0.0000 0.978 1.000 0.000
#> GSM39126 1 0.7528 0.692 0.784 0.216
#> GSM39127 2 0.0938 0.877 0.012 0.988
#> GSM39128 2 0.2236 0.867 0.036 0.964
#> GSM39129 2 0.0000 0.880 0.000 1.000
#> GSM39130 2 0.0000 0.880 0.000 1.000
#> GSM39131 2 0.1184 0.876 0.016 0.984
#> GSM39132 2 0.0000 0.880 0.000 1.000
#> GSM39133 2 0.0000 0.880 0.000 1.000
#> GSM39134 2 0.0000 0.880 0.000 1.000
#> GSM39135 2 0.0000 0.880 0.000 1.000
#> GSM39136 2 0.0000 0.880 0.000 1.000
#> GSM39137 2 0.9522 0.477 0.372 0.628
#> GSM39138 2 0.0000 0.880 0.000 1.000
#> GSM39139 2 0.0000 0.880 0.000 1.000
#> GSM39140 1 0.0000 0.978 1.000 0.000
#> GSM39141 1 0.0000 0.978 1.000 0.000
#> GSM39142 1 0.0000 0.978 1.000 0.000
#> GSM39143 1 0.0000 0.978 1.000 0.000
#> GSM39144 2 0.0000 0.880 0.000 1.000
#> GSM39145 2 0.0000 0.880 0.000 1.000
#> GSM39146 2 0.0000 0.880 0.000 1.000
#> GSM39147 2 0.0000 0.880 0.000 1.000
#> GSM39188 2 0.8016 0.742 0.244 0.756
#> GSM39189 2 0.8813 0.675 0.300 0.700
#> GSM39190 2 0.8386 0.719 0.268 0.732
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM39104 1 0.0000 0.9269 1.000 0.000 0.000
#> GSM39105 1 0.0000 0.9269 1.000 0.000 0.000
#> GSM39106 1 0.0592 0.9227 0.988 0.000 0.012
#> GSM39107 1 0.3771 0.8355 0.876 0.112 0.012
#> GSM39108 1 0.0237 0.9258 0.996 0.000 0.004
#> GSM39109 1 0.9587 0.0906 0.440 0.356 0.204
#> GSM39110 1 0.0747 0.9229 0.984 0.000 0.016
#> GSM39111 1 0.0237 0.9250 0.996 0.000 0.004
#> GSM39112 1 0.2651 0.8817 0.928 0.060 0.012
#> GSM39113 1 0.3845 0.8317 0.872 0.116 0.012
#> GSM39114 2 0.1129 0.9057 0.004 0.976 0.020
#> GSM39115 1 0.0000 0.9269 1.000 0.000 0.000
#> GSM39148 1 0.0237 0.9258 0.996 0.000 0.004
#> GSM39149 3 0.1585 0.9682 0.028 0.008 0.964
#> GSM39150 1 0.0000 0.9269 1.000 0.000 0.000
#> GSM39151 3 0.1585 0.9682 0.028 0.008 0.964
#> GSM39152 3 0.4504 0.7608 0.196 0.000 0.804
#> GSM39153 1 0.0000 0.9269 1.000 0.000 0.000
#> GSM39154 1 0.0000 0.9269 1.000 0.000 0.000
#> GSM39155 1 0.0000 0.9269 1.000 0.000 0.000
#> GSM39156 1 0.0829 0.9208 0.984 0.004 0.012
#> GSM39157 1 0.0000 0.9269 1.000 0.000 0.000
#> GSM39158 1 0.0000 0.9269 1.000 0.000 0.000
#> GSM39159 1 0.3192 0.8335 0.888 0.000 0.112
#> GSM39160 1 0.0000 0.9269 1.000 0.000 0.000
#> GSM39161 1 0.5327 0.6085 0.728 0.000 0.272
#> GSM39162 1 0.0237 0.9258 0.996 0.000 0.004
#> GSM39163 1 0.0000 0.9269 1.000 0.000 0.000
#> GSM39164 1 0.0000 0.9269 1.000 0.000 0.000
#> GSM39165 1 0.2356 0.8769 0.928 0.000 0.072
#> GSM39166 1 0.0000 0.9269 1.000 0.000 0.000
#> GSM39167 1 0.0000 0.9269 1.000 0.000 0.000
#> GSM39168 1 0.0237 0.9258 0.996 0.000 0.004
#> GSM39169 1 0.0000 0.9269 1.000 0.000 0.000
#> GSM39170 1 0.0000 0.9269 1.000 0.000 0.000
#> GSM39171 1 0.0000 0.9269 1.000 0.000 0.000
#> GSM39172 3 0.1585 0.9682 0.028 0.008 0.964
#> GSM39173 3 0.1585 0.9682 0.028 0.008 0.964
#> GSM39174 1 0.0000 0.9269 1.000 0.000 0.000
#> GSM39175 1 0.0000 0.9269 1.000 0.000 0.000
#> GSM39176 1 0.0000 0.9269 1.000 0.000 0.000
#> GSM39177 3 0.2584 0.9293 0.064 0.008 0.928
#> GSM39178 1 0.2959 0.8461 0.900 0.000 0.100
#> GSM39179 3 0.1585 0.9682 0.028 0.008 0.964
#> GSM39180 3 0.1482 0.9579 0.020 0.012 0.968
#> GSM39181 1 0.0000 0.9269 1.000 0.000 0.000
#> GSM39182 1 0.6798 0.3058 0.584 0.016 0.400
#> GSM39183 1 0.0000 0.9269 1.000 0.000 0.000
#> GSM39184 1 0.0000 0.9269 1.000 0.000 0.000
#> GSM39185 1 0.5327 0.6085 0.728 0.000 0.272
#> GSM39186 1 0.0000 0.9269 1.000 0.000 0.000
#> GSM39187 1 0.0000 0.9269 1.000 0.000 0.000
#> GSM39116 2 0.0424 0.9091 0.000 0.992 0.008
#> GSM39117 2 0.4452 0.8330 0.000 0.808 0.192
#> GSM39118 2 0.3038 0.8900 0.000 0.896 0.104
#> GSM39119 2 0.3482 0.8779 0.000 0.872 0.128
#> GSM39120 1 0.1337 0.9141 0.972 0.016 0.012
#> GSM39121 1 0.6448 0.5360 0.656 0.328 0.016
#> GSM39122 1 0.6994 0.2966 0.556 0.424 0.020
#> GSM39123 2 0.4452 0.8330 0.000 0.808 0.192
#> GSM39124 2 0.1129 0.9057 0.004 0.976 0.020
#> GSM39125 1 0.1337 0.9141 0.972 0.016 0.012
#> GSM39126 1 0.6497 0.5210 0.648 0.336 0.016
#> GSM39127 2 0.0892 0.9080 0.000 0.980 0.020
#> GSM39128 2 0.0892 0.9080 0.000 0.980 0.020
#> GSM39129 2 0.3482 0.8802 0.000 0.872 0.128
#> GSM39130 2 0.4452 0.8330 0.000 0.808 0.192
#> GSM39131 2 0.0892 0.9080 0.000 0.980 0.020
#> GSM39132 2 0.0892 0.9080 0.000 0.980 0.020
#> GSM39133 2 0.3192 0.8909 0.000 0.888 0.112
#> GSM39134 2 0.3192 0.8886 0.000 0.888 0.112
#> GSM39135 2 0.0424 0.9091 0.000 0.992 0.008
#> GSM39136 2 0.0424 0.9091 0.000 0.992 0.008
#> GSM39137 2 0.6229 0.4925 0.280 0.700 0.020
#> GSM39138 2 0.3482 0.8802 0.000 0.872 0.128
#> GSM39139 2 0.1643 0.9041 0.000 0.956 0.044
#> GSM39140 1 0.0829 0.9208 0.984 0.004 0.012
#> GSM39141 1 0.0829 0.9208 0.984 0.004 0.012
#> GSM39142 1 0.0829 0.9208 0.984 0.004 0.012
#> GSM39143 1 0.0829 0.9208 0.984 0.004 0.012
#> GSM39144 2 0.3482 0.8802 0.000 0.872 0.128
#> GSM39145 2 0.0424 0.9074 0.000 0.992 0.008
#> GSM39146 2 0.0892 0.9080 0.000 0.980 0.020
#> GSM39147 2 0.0892 0.9080 0.000 0.980 0.020
#> GSM39188 3 0.1585 0.9682 0.028 0.008 0.964
#> GSM39189 3 0.1585 0.9682 0.028 0.008 0.964
#> GSM39190 3 0.1585 0.9682 0.028 0.008 0.964
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM39104 1 0.2125 0.8726 0.920 0.000 0.004 0.076
#> GSM39105 1 0.0336 0.8862 0.992 0.000 0.000 0.008
#> GSM39106 1 0.3090 0.8395 0.888 0.056 0.000 0.056
#> GSM39107 1 0.5964 0.1104 0.536 0.424 0.000 0.040
#> GSM39108 1 0.2644 0.8575 0.908 0.032 0.000 0.060
#> GSM39109 2 0.7987 0.3653 0.256 0.544 0.044 0.156
#> GSM39110 1 0.3689 0.8405 0.860 0.048 0.004 0.088
#> GSM39111 1 0.2860 0.8630 0.888 0.008 0.004 0.100
#> GSM39112 1 0.5835 0.2787 0.588 0.372 0.000 0.040
#> GSM39113 1 0.6074 -0.0204 0.500 0.456 0.000 0.044
#> GSM39114 2 0.2565 0.6071 0.056 0.912 0.000 0.032
#> GSM39115 1 0.0000 0.8869 1.000 0.000 0.000 0.000
#> GSM39148 1 0.0000 0.8869 1.000 0.000 0.000 0.000
#> GSM39149 3 0.1743 0.9290 0.004 0.000 0.940 0.056
#> GSM39150 1 0.3249 0.8355 0.852 0.000 0.008 0.140
#> GSM39151 3 0.1489 0.9304 0.004 0.000 0.952 0.044
#> GSM39152 3 0.4546 0.8180 0.056 0.004 0.804 0.136
#> GSM39153 1 0.0000 0.8869 1.000 0.000 0.000 0.000
#> GSM39154 1 0.0188 0.8867 0.996 0.000 0.000 0.004
#> GSM39155 1 0.0000 0.8869 1.000 0.000 0.000 0.000
#> GSM39156 1 0.1584 0.8718 0.952 0.012 0.000 0.036
#> GSM39157 1 0.0000 0.8869 1.000 0.000 0.000 0.000
#> GSM39158 1 0.2197 0.8625 0.916 0.000 0.004 0.080
#> GSM39159 1 0.5077 0.7531 0.760 0.000 0.080 0.160
#> GSM39160 1 0.3401 0.8299 0.840 0.000 0.008 0.152
#> GSM39161 1 0.6323 0.6166 0.660 0.000 0.176 0.164
#> GSM39162 1 0.0188 0.8860 0.996 0.004 0.000 0.000
#> GSM39163 1 0.0000 0.8869 1.000 0.000 0.000 0.000
#> GSM39164 1 0.0000 0.8869 1.000 0.000 0.000 0.000
#> GSM39165 1 0.3818 0.8341 0.844 0.000 0.048 0.108
#> GSM39166 1 0.2888 0.8424 0.872 0.000 0.004 0.124
#> GSM39167 1 0.0000 0.8869 1.000 0.000 0.000 0.000
#> GSM39168 1 0.0000 0.8869 1.000 0.000 0.000 0.000
#> GSM39169 1 0.0336 0.8865 0.992 0.000 0.000 0.008
#> GSM39170 1 0.2334 0.8589 0.908 0.000 0.004 0.088
#> GSM39171 1 0.2675 0.8578 0.892 0.000 0.008 0.100
#> GSM39172 3 0.2530 0.9172 0.004 0.000 0.896 0.100
#> GSM39173 3 0.1902 0.9322 0.004 0.000 0.932 0.064
#> GSM39174 1 0.0000 0.8869 1.000 0.000 0.000 0.000
#> GSM39175 1 0.0188 0.8867 0.996 0.000 0.000 0.004
#> GSM39176 1 0.0000 0.8869 1.000 0.000 0.000 0.000
#> GSM39177 3 0.0895 0.9370 0.004 0.000 0.976 0.020
#> GSM39178 1 0.5102 0.7488 0.748 0.000 0.064 0.188
#> GSM39179 3 0.1576 0.9300 0.004 0.000 0.948 0.048
#> GSM39180 3 0.2401 0.9240 0.004 0.000 0.904 0.092
#> GSM39181 1 0.2773 0.8457 0.880 0.000 0.004 0.116
#> GSM39182 1 0.8970 0.2886 0.480 0.108 0.208 0.204
#> GSM39183 1 0.2888 0.8424 0.872 0.000 0.004 0.124
#> GSM39184 1 0.0336 0.8865 0.992 0.000 0.000 0.008
#> GSM39185 1 0.6439 0.6044 0.648 0.000 0.176 0.176
#> GSM39186 1 0.0336 0.8865 0.992 0.000 0.000 0.008
#> GSM39187 1 0.0000 0.8869 1.000 0.000 0.000 0.000
#> GSM39116 2 0.1716 0.5341 0.000 0.936 0.000 0.064
#> GSM39117 4 0.5800 0.8293 0.000 0.420 0.032 0.548
#> GSM39118 2 0.4989 -0.7791 0.000 0.528 0.000 0.472
#> GSM39119 4 0.4916 0.8436 0.000 0.424 0.000 0.576
#> GSM39120 1 0.4423 0.7123 0.792 0.168 0.000 0.040
#> GSM39121 2 0.5599 0.4197 0.352 0.616 0.000 0.032
#> GSM39122 2 0.5453 0.4428 0.320 0.648 0.000 0.032
#> GSM39123 4 0.5800 0.8293 0.000 0.420 0.032 0.548
#> GSM39124 2 0.1302 0.6210 0.044 0.956 0.000 0.000
#> GSM39125 1 0.2892 0.8329 0.896 0.068 0.000 0.036
#> GSM39126 2 0.5599 0.4197 0.352 0.616 0.000 0.032
#> GSM39127 2 0.0524 0.6104 0.004 0.988 0.000 0.008
#> GSM39128 2 0.0927 0.6180 0.016 0.976 0.000 0.008
#> GSM39129 4 0.4872 0.8213 0.000 0.356 0.004 0.640
#> GSM39130 4 0.5800 0.8293 0.000 0.420 0.032 0.548
#> GSM39131 2 0.1297 0.6203 0.016 0.964 0.000 0.020
#> GSM39132 2 0.0592 0.6032 0.000 0.984 0.000 0.016
#> GSM39133 4 0.5466 0.8274 0.000 0.436 0.016 0.548
#> GSM39134 4 0.4999 0.8009 0.000 0.492 0.000 0.508
#> GSM39135 2 0.1716 0.5341 0.000 0.936 0.000 0.064
#> GSM39136 2 0.1716 0.5341 0.000 0.936 0.000 0.064
#> GSM39137 2 0.4152 0.5446 0.160 0.808 0.000 0.032
#> GSM39138 4 0.4872 0.8213 0.000 0.356 0.004 0.640
#> GSM39139 4 0.5126 0.7395 0.000 0.444 0.004 0.552
#> GSM39140 1 0.1610 0.8698 0.952 0.016 0.000 0.032
#> GSM39141 1 0.0804 0.8816 0.980 0.008 0.000 0.012
#> GSM39142 1 0.0804 0.8816 0.980 0.008 0.000 0.012
#> GSM39143 1 0.0804 0.8816 0.980 0.008 0.000 0.012
#> GSM39144 4 0.4872 0.8213 0.000 0.356 0.004 0.640
#> GSM39145 2 0.3982 0.3464 0.000 0.776 0.004 0.220
#> GSM39146 2 0.0469 0.6031 0.000 0.988 0.000 0.012
#> GSM39147 2 0.0895 0.6094 0.004 0.976 0.000 0.020
#> GSM39188 3 0.1661 0.9284 0.004 0.000 0.944 0.052
#> GSM39189 3 0.2401 0.9204 0.004 0.000 0.904 0.092
#> GSM39190 3 0.1209 0.9355 0.004 0.000 0.964 0.032
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM39104 1 0.3737 0.6361 0.764 0.004 0.000 0.008 0.224
#> GSM39105 1 0.2471 0.7289 0.864 0.000 0.000 0.000 0.136
#> GSM39106 1 0.4975 0.5478 0.712 0.076 0.000 0.008 0.204
#> GSM39107 2 0.5804 0.1822 0.352 0.544 0.000 0.000 0.104
#> GSM39108 1 0.4436 0.6026 0.744 0.040 0.000 0.008 0.208
#> GSM39109 5 0.7081 -0.0530 0.132 0.416 0.012 0.024 0.416
#> GSM39110 1 0.5321 0.4877 0.672 0.068 0.004 0.008 0.248
#> GSM39111 1 0.4568 0.5734 0.716 0.024 0.004 0.008 0.248
#> GSM39112 2 0.5933 -0.0160 0.444 0.452 0.000 0.000 0.104
#> GSM39113 2 0.5902 0.2101 0.320 0.556 0.000 0.000 0.124
#> GSM39114 2 0.1386 0.6809 0.032 0.952 0.000 0.000 0.016
#> GSM39115 1 0.0404 0.8021 0.988 0.000 0.000 0.000 0.012
#> GSM39148 1 0.0000 0.8029 1.000 0.000 0.000 0.000 0.000
#> GSM39149 3 0.0912 0.7812 0.000 0.000 0.972 0.012 0.016
#> GSM39150 1 0.3741 0.5940 0.732 0.000 0.000 0.004 0.264
#> GSM39151 3 0.1493 0.7715 0.000 0.000 0.948 0.024 0.028
#> GSM39152 3 0.5670 0.4725 0.040 0.004 0.476 0.012 0.468
#> GSM39153 1 0.0000 0.8029 1.000 0.000 0.000 0.000 0.000
#> GSM39154 1 0.0290 0.8027 0.992 0.000 0.000 0.000 0.008
#> GSM39155 1 0.0162 0.8031 0.996 0.000 0.000 0.000 0.004
#> GSM39156 1 0.2411 0.7351 0.884 0.008 0.000 0.000 0.108
#> GSM39157 1 0.0162 0.8031 0.996 0.000 0.000 0.000 0.004
#> GSM39158 1 0.3039 0.6818 0.836 0.000 0.000 0.012 0.152
#> GSM39159 1 0.4745 -0.0378 0.560 0.000 0.004 0.012 0.424
#> GSM39160 1 0.3906 0.5505 0.704 0.000 0.000 0.004 0.292
#> GSM39161 1 0.5305 -0.1675 0.524 0.000 0.028 0.012 0.436
#> GSM39162 1 0.0000 0.8029 1.000 0.000 0.000 0.000 0.000
#> GSM39163 1 0.0162 0.8031 0.996 0.000 0.000 0.000 0.004
#> GSM39164 1 0.0162 0.8031 0.996 0.000 0.000 0.000 0.004
#> GSM39165 1 0.4065 0.6106 0.760 0.000 0.020 0.008 0.212
#> GSM39166 1 0.3462 0.6353 0.792 0.000 0.000 0.012 0.196
#> GSM39167 1 0.0162 0.8031 0.996 0.000 0.000 0.000 0.004
#> GSM39168 1 0.0000 0.8029 1.000 0.000 0.000 0.000 0.000
#> GSM39169 1 0.0162 0.8031 0.996 0.000 0.000 0.000 0.004
#> GSM39170 1 0.2997 0.6863 0.840 0.000 0.000 0.012 0.148
#> GSM39171 1 0.2971 0.7125 0.836 0.000 0.000 0.008 0.156
#> GSM39172 3 0.4930 0.7276 0.000 0.000 0.580 0.032 0.388
#> GSM39173 3 0.4520 0.7880 0.000 0.000 0.684 0.032 0.284
#> GSM39174 1 0.0162 0.8031 0.996 0.000 0.000 0.000 0.004
#> GSM39175 1 0.0404 0.8016 0.988 0.000 0.000 0.000 0.012
#> GSM39176 1 0.0000 0.8029 1.000 0.000 0.000 0.000 0.000
#> GSM39177 3 0.3039 0.8088 0.000 0.000 0.836 0.012 0.152
#> GSM39178 5 0.4283 -0.0071 0.456 0.000 0.000 0.000 0.544
#> GSM39179 3 0.1106 0.7887 0.000 0.000 0.964 0.012 0.024
#> GSM39180 3 0.4661 0.7821 0.000 0.000 0.656 0.032 0.312
#> GSM39181 1 0.3355 0.6457 0.804 0.000 0.000 0.012 0.184
#> GSM39182 5 0.6267 0.2247 0.200 0.032 0.064 0.040 0.664
#> GSM39183 1 0.3530 0.6270 0.784 0.000 0.000 0.012 0.204
#> GSM39184 1 0.0290 0.8027 0.992 0.000 0.000 0.000 0.008
#> GSM39185 1 0.5329 -0.2598 0.488 0.000 0.028 0.012 0.472
#> GSM39186 1 0.0404 0.8014 0.988 0.000 0.000 0.000 0.012
#> GSM39187 1 0.0000 0.8029 1.000 0.000 0.000 0.000 0.000
#> GSM39116 2 0.2580 0.5948 0.000 0.892 0.000 0.064 0.044
#> GSM39117 4 0.6547 0.7510 0.000 0.216 0.004 0.504 0.276
#> GSM39118 4 0.4820 0.7231 0.000 0.332 0.000 0.632 0.036
#> GSM39119 4 0.5223 0.7921 0.000 0.220 0.000 0.672 0.108
#> GSM39120 1 0.5649 0.1931 0.596 0.296 0.000 0.000 0.108
#> GSM39121 2 0.3659 0.5441 0.220 0.768 0.000 0.000 0.012
#> GSM39122 2 0.3659 0.5441 0.220 0.768 0.000 0.000 0.012
#> GSM39123 4 0.6547 0.7510 0.000 0.216 0.004 0.504 0.276
#> GSM39124 2 0.1211 0.6852 0.024 0.960 0.000 0.016 0.000
#> GSM39125 1 0.4370 0.4974 0.744 0.200 0.000 0.000 0.056
#> GSM39126 2 0.4024 0.5297 0.220 0.752 0.000 0.000 0.028
#> GSM39127 2 0.0794 0.6714 0.000 0.972 0.000 0.028 0.000
#> GSM39128 2 0.0992 0.6789 0.008 0.968 0.000 0.024 0.000
#> GSM39129 4 0.3093 0.7700 0.000 0.168 0.000 0.824 0.008
#> GSM39130 4 0.6547 0.7510 0.000 0.216 0.004 0.504 0.276
#> GSM39131 2 0.0162 0.6806 0.004 0.996 0.000 0.000 0.000
#> GSM39132 2 0.0794 0.6714 0.000 0.972 0.000 0.028 0.000
#> GSM39133 4 0.6423 0.7495 0.000 0.220 0.000 0.504 0.276
#> GSM39134 4 0.5404 0.7695 0.000 0.292 0.000 0.620 0.088
#> GSM39135 2 0.2645 0.5887 0.000 0.888 0.000 0.068 0.044
#> GSM39136 2 0.2580 0.5948 0.000 0.892 0.000 0.064 0.044
#> GSM39137 2 0.2020 0.6653 0.100 0.900 0.000 0.000 0.000
#> GSM39138 4 0.2813 0.7697 0.000 0.168 0.000 0.832 0.000
#> GSM39139 4 0.3983 0.6083 0.000 0.340 0.000 0.660 0.000
#> GSM39140 1 0.1648 0.7722 0.940 0.040 0.000 0.000 0.020
#> GSM39141 1 0.0912 0.7925 0.972 0.012 0.000 0.000 0.016
#> GSM39142 1 0.0912 0.7925 0.972 0.012 0.000 0.000 0.016
#> GSM39143 1 0.0912 0.7925 0.972 0.012 0.000 0.000 0.016
#> GSM39144 4 0.2813 0.7697 0.000 0.168 0.000 0.832 0.000
#> GSM39145 2 0.4278 -0.1513 0.000 0.548 0.000 0.452 0.000
#> GSM39146 2 0.1493 0.6556 0.000 0.948 0.000 0.028 0.024
#> GSM39147 2 0.1124 0.6740 0.004 0.960 0.000 0.036 0.000
#> GSM39188 3 0.2520 0.7694 0.000 0.000 0.896 0.048 0.056
#> GSM39189 3 0.4763 0.7559 0.000 0.000 0.632 0.032 0.336
#> GSM39190 3 0.3848 0.8067 0.000 0.000 0.788 0.040 0.172
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM39104 1 0.4527 0.460309 0.624 0.004 0.000 0.000 0.332 0.040
#> GSM39105 1 0.2790 0.692665 0.840 0.000 0.000 0.000 0.140 0.020
#> GSM39106 1 0.5451 0.389350 0.572 0.052 0.000 0.000 0.332 0.044
#> GSM39107 2 0.5685 0.558330 0.184 0.632 0.000 0.000 0.136 0.048
#> GSM39108 1 0.4974 0.427419 0.600 0.024 0.000 0.000 0.336 0.040
#> GSM39109 5 0.6598 -0.089051 0.052 0.340 0.032 0.008 0.508 0.060
#> GSM39110 1 0.5488 0.346343 0.552 0.036 0.004 0.000 0.360 0.048
#> GSM39111 1 0.5038 0.385605 0.576 0.016 0.004 0.000 0.364 0.040
#> GSM39112 2 0.6339 0.263994 0.336 0.480 0.000 0.000 0.136 0.048
#> GSM39113 2 0.5731 0.565716 0.160 0.632 0.000 0.000 0.156 0.052
#> GSM39114 2 0.1767 0.782401 0.012 0.932 0.000 0.000 0.036 0.020
#> GSM39115 1 0.0603 0.765271 0.980 0.000 0.000 0.000 0.016 0.004
#> GSM39148 1 0.0146 0.769975 0.996 0.000 0.000 0.000 0.004 0.000
#> GSM39149 3 0.2763 0.716766 0.000 0.000 0.868 0.036 0.008 0.088
#> GSM39150 1 0.4109 0.220445 0.576 0.000 0.000 0.000 0.412 0.012
#> GSM39151 3 0.3993 0.683483 0.000 0.000 0.788 0.028 0.060 0.124
#> GSM39152 5 0.4784 -0.408761 0.012 0.000 0.404 0.004 0.556 0.024
#> GSM39153 1 0.0146 0.770165 0.996 0.000 0.000 0.000 0.000 0.004
#> GSM39154 1 0.0000 0.770579 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM39155 1 0.0000 0.770579 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM39156 1 0.2851 0.687480 0.844 0.004 0.000 0.000 0.132 0.020
#> GSM39157 1 0.0000 0.770579 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM39158 1 0.3536 0.481792 0.736 0.000 0.000 0.008 0.252 0.004
#> GSM39159 5 0.4589 0.379449 0.432 0.000 0.004 0.008 0.540 0.016
#> GSM39160 1 0.4116 0.280262 0.572 0.000 0.000 0.000 0.416 0.012
#> GSM39161 5 0.4902 0.454198 0.400 0.000 0.016 0.012 0.556 0.016
#> GSM39162 1 0.0260 0.769170 0.992 0.000 0.000 0.000 0.008 0.000
#> GSM39163 1 0.0000 0.770579 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM39164 1 0.0000 0.770579 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM39165 1 0.4110 0.514398 0.712 0.000 0.032 0.000 0.248 0.008
#> GSM39166 1 0.4031 0.331178 0.652 0.000 0.000 0.008 0.332 0.008
#> GSM39167 1 0.0000 0.770579 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM39168 1 0.0260 0.769170 0.992 0.000 0.000 0.000 0.008 0.000
#> GSM39169 1 0.0146 0.769922 0.996 0.000 0.000 0.000 0.004 0.000
#> GSM39170 1 0.3714 0.461481 0.720 0.000 0.000 0.008 0.264 0.008
#> GSM39171 1 0.3081 0.613994 0.776 0.000 0.000 0.000 0.220 0.004
#> GSM39172 3 0.5702 0.624669 0.000 0.000 0.524 0.056 0.368 0.052
#> GSM39173 3 0.5082 0.709305 0.000 0.000 0.628 0.016 0.280 0.076
#> GSM39174 1 0.0000 0.770579 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM39175 1 0.0000 0.770579 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM39176 1 0.0000 0.770579 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM39177 3 0.3917 0.731468 0.000 0.000 0.780 0.024 0.156 0.040
#> GSM39178 5 0.3636 0.470961 0.320 0.000 0.000 0.000 0.676 0.004
#> GSM39179 3 0.3063 0.720847 0.000 0.000 0.860 0.024 0.052 0.064
#> GSM39180 3 0.5550 0.686850 0.000 0.000 0.576 0.028 0.308 0.088
#> GSM39181 1 0.3950 0.364928 0.672 0.000 0.000 0.008 0.312 0.008
#> GSM39182 5 0.7406 -0.000111 0.064 0.028 0.112 0.260 0.504 0.032
#> GSM39183 1 0.4031 0.331178 0.652 0.000 0.000 0.008 0.332 0.008
#> GSM39184 1 0.0000 0.770579 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM39185 5 0.4666 0.484855 0.372 0.000 0.012 0.008 0.592 0.016
#> GSM39186 1 0.0363 0.765894 0.988 0.000 0.000 0.000 0.012 0.000
#> GSM39187 1 0.0000 0.770579 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM39116 2 0.2113 0.718860 0.000 0.896 0.000 0.092 0.008 0.004
#> GSM39117 4 0.2135 0.750275 0.000 0.128 0.000 0.872 0.000 0.000
#> GSM39118 6 0.6250 0.449282 0.000 0.212 0.000 0.292 0.020 0.476
#> GSM39119 4 0.5659 0.078352 0.000 0.096 0.000 0.540 0.024 0.340
#> GSM39120 1 0.6333 0.120247 0.472 0.348 0.000 0.000 0.132 0.048
#> GSM39121 2 0.3949 0.711832 0.124 0.788 0.000 0.000 0.068 0.020
#> GSM39122 2 0.3908 0.714933 0.120 0.792 0.000 0.000 0.068 0.020
#> GSM39123 4 0.2135 0.750275 0.000 0.128 0.000 0.872 0.000 0.000
#> GSM39124 2 0.0767 0.791788 0.012 0.976 0.000 0.004 0.008 0.000
#> GSM39125 1 0.5567 0.373980 0.620 0.244 0.000 0.000 0.092 0.044
#> GSM39126 2 0.4312 0.697921 0.124 0.764 0.000 0.000 0.084 0.028
#> GSM39127 2 0.0692 0.783546 0.000 0.976 0.000 0.020 0.000 0.004
#> GSM39128 2 0.0767 0.788704 0.000 0.976 0.000 0.012 0.008 0.004
#> GSM39129 6 0.4753 0.674428 0.000 0.048 0.000 0.308 0.012 0.632
#> GSM39130 4 0.2135 0.750275 0.000 0.128 0.000 0.872 0.000 0.000
#> GSM39131 2 0.0291 0.790127 0.000 0.992 0.000 0.004 0.004 0.000
#> GSM39132 2 0.0692 0.783546 0.000 0.976 0.000 0.020 0.000 0.004
#> GSM39133 4 0.2278 0.748153 0.000 0.128 0.000 0.868 0.004 0.000
#> GSM39134 4 0.6205 -0.133075 0.000 0.208 0.000 0.420 0.012 0.360
#> GSM39135 2 0.2163 0.714317 0.000 0.892 0.000 0.096 0.008 0.004
#> GSM39136 2 0.2113 0.718860 0.000 0.896 0.000 0.092 0.008 0.004
#> GSM39137 2 0.1511 0.783227 0.044 0.940 0.000 0.000 0.012 0.004
#> GSM39138 6 0.4386 0.679151 0.000 0.048 0.000 0.300 0.000 0.652
#> GSM39139 6 0.4904 0.660167 0.000 0.148 0.000 0.196 0.000 0.656
#> GSM39140 1 0.2307 0.726558 0.904 0.032 0.000 0.000 0.048 0.016
#> GSM39141 1 0.1922 0.739044 0.924 0.024 0.000 0.000 0.040 0.012
#> GSM39142 1 0.1838 0.741508 0.928 0.020 0.000 0.000 0.040 0.012
#> GSM39143 1 0.1922 0.739044 0.924 0.024 0.000 0.000 0.040 0.012
#> GSM39144 6 0.4666 0.682194 0.000 0.052 0.000 0.296 0.008 0.644
#> GSM39145 6 0.4432 0.393636 0.000 0.364 0.000 0.036 0.000 0.600
#> GSM39146 2 0.1080 0.775668 0.000 0.960 0.000 0.032 0.004 0.004
#> GSM39147 2 0.0508 0.786983 0.000 0.984 0.000 0.004 0.000 0.012
#> GSM39188 3 0.4735 0.678390 0.000 0.000 0.732 0.056 0.064 0.148
#> GSM39189 3 0.5005 0.632626 0.000 0.000 0.544 0.012 0.396 0.048
#> GSM39190 3 0.5473 0.724416 0.000 0.000 0.652 0.040 0.176 0.132
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) other(p) protocol(p) k
#> CV:kmeans 84 0.0731 6.30e-07 8.55e-05 2
#> CV:kmeans 83 0.0656 1.18e-08 3.87e-08 3
#> CV:kmeans 77 0.2692 3.02e-07 4.17e-08 4
#> CV:kmeans 73 0.2358 2.19e-08 1.21e-09 5
#> CV:kmeans 61 0.4579 3.39e-05 1.43e-07 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 8353 rows and 87 columns.
#> Top rows (835, 1670, 2506, 3341, 4176) 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 0.976 0.917 0.970 0.5005 0.497 0.497
#> 3 3 0.650 0.667 0.864 0.3007 0.795 0.613
#> 4 4 0.566 0.584 0.776 0.1340 0.826 0.567
#> 5 5 0.545 0.511 0.719 0.0687 0.895 0.643
#> 6 6 0.542 0.428 0.607 0.0408 0.936 0.721
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
#> GSM39104 1 0.0000 0.9818 1.000 0.000
#> GSM39105 1 0.0000 0.9818 1.000 0.000
#> GSM39106 1 0.0000 0.9818 1.000 0.000
#> GSM39107 1 0.0000 0.9818 1.000 0.000
#> GSM39108 1 0.0000 0.9818 1.000 0.000
#> GSM39109 2 0.0000 0.9499 0.000 1.000
#> GSM39110 1 0.0376 0.9782 0.996 0.004
#> GSM39111 1 0.0000 0.9818 1.000 0.000
#> GSM39112 1 0.0000 0.9818 1.000 0.000
#> GSM39113 1 0.1633 0.9583 0.976 0.024
#> GSM39114 2 0.0000 0.9499 0.000 1.000
#> GSM39115 1 0.0000 0.9818 1.000 0.000
#> GSM39148 1 0.0000 0.9818 1.000 0.000
#> GSM39149 2 0.0000 0.9499 0.000 1.000
#> GSM39150 1 0.0000 0.9818 1.000 0.000
#> GSM39151 2 0.0000 0.9499 0.000 1.000
#> GSM39152 2 0.9635 0.3759 0.388 0.612
#> GSM39153 1 0.0000 0.9818 1.000 0.000
#> GSM39154 1 0.0000 0.9818 1.000 0.000
#> GSM39155 1 0.0000 0.9818 1.000 0.000
#> GSM39156 1 0.0000 0.9818 1.000 0.000
#> GSM39157 1 0.0000 0.9818 1.000 0.000
#> GSM39158 1 0.0000 0.9818 1.000 0.000
#> GSM39159 1 0.0938 0.9705 0.988 0.012
#> GSM39160 1 0.0000 0.9818 1.000 0.000
#> GSM39161 1 0.8555 0.5774 0.720 0.280
#> GSM39162 1 0.0000 0.9818 1.000 0.000
#> GSM39163 1 0.0000 0.9818 1.000 0.000
#> GSM39164 1 0.0000 0.9818 1.000 0.000
#> GSM39165 1 0.0000 0.9818 1.000 0.000
#> GSM39166 1 0.0000 0.9818 1.000 0.000
#> GSM39167 1 0.0000 0.9818 1.000 0.000
#> GSM39168 1 0.0000 0.9818 1.000 0.000
#> GSM39169 1 0.0000 0.9818 1.000 0.000
#> GSM39170 1 0.0000 0.9818 1.000 0.000
#> GSM39171 1 0.0000 0.9818 1.000 0.000
#> GSM39172 2 0.0000 0.9499 0.000 1.000
#> GSM39173 2 0.0000 0.9499 0.000 1.000
#> GSM39174 1 0.0000 0.9818 1.000 0.000
#> GSM39175 1 0.0000 0.9818 1.000 0.000
#> GSM39176 1 0.0000 0.9818 1.000 0.000
#> GSM39177 2 0.3733 0.8848 0.072 0.928
#> GSM39178 1 0.0000 0.9818 1.000 0.000
#> GSM39179 2 0.0000 0.9499 0.000 1.000
#> GSM39180 2 0.0000 0.9499 0.000 1.000
#> GSM39181 1 0.0000 0.9818 1.000 0.000
#> GSM39182 2 0.1184 0.9371 0.016 0.984
#> GSM39183 1 0.0000 0.9818 1.000 0.000
#> GSM39184 1 0.0000 0.9818 1.000 0.000
#> GSM39185 2 0.9983 0.1152 0.476 0.524
#> GSM39186 1 0.0000 0.9818 1.000 0.000
#> GSM39187 1 0.0000 0.9818 1.000 0.000
#> GSM39116 2 0.0000 0.9499 0.000 1.000
#> GSM39117 2 0.0000 0.9499 0.000 1.000
#> GSM39118 2 0.0000 0.9499 0.000 1.000
#> GSM39119 2 0.0000 0.9499 0.000 1.000
#> GSM39120 1 0.0000 0.9818 1.000 0.000
#> GSM39121 1 0.9944 0.0991 0.544 0.456
#> GSM39122 2 0.9944 0.1820 0.456 0.544
#> GSM39123 2 0.0000 0.9499 0.000 1.000
#> GSM39124 2 0.0000 0.9499 0.000 1.000
#> GSM39125 1 0.0000 0.9818 1.000 0.000
#> GSM39126 2 0.9944 0.1815 0.456 0.544
#> GSM39127 2 0.0000 0.9499 0.000 1.000
#> GSM39128 2 0.0000 0.9499 0.000 1.000
#> GSM39129 2 0.0000 0.9499 0.000 1.000
#> GSM39130 2 0.0000 0.9499 0.000 1.000
#> GSM39131 2 0.0000 0.9499 0.000 1.000
#> GSM39132 2 0.0000 0.9499 0.000 1.000
#> GSM39133 2 0.0000 0.9499 0.000 1.000
#> GSM39134 2 0.0000 0.9499 0.000 1.000
#> GSM39135 2 0.0000 0.9499 0.000 1.000
#> GSM39136 2 0.0000 0.9499 0.000 1.000
#> GSM39137 2 0.0672 0.9437 0.008 0.992
#> GSM39138 2 0.0000 0.9499 0.000 1.000
#> GSM39139 2 0.0000 0.9499 0.000 1.000
#> GSM39140 1 0.0000 0.9818 1.000 0.000
#> GSM39141 1 0.0000 0.9818 1.000 0.000
#> GSM39142 1 0.0000 0.9818 1.000 0.000
#> GSM39143 1 0.0000 0.9818 1.000 0.000
#> GSM39144 2 0.0000 0.9499 0.000 1.000
#> GSM39145 2 0.0000 0.9499 0.000 1.000
#> GSM39146 2 0.0000 0.9499 0.000 1.000
#> GSM39147 2 0.0000 0.9499 0.000 1.000
#> GSM39188 2 0.0000 0.9499 0.000 1.000
#> GSM39189 2 0.0000 0.9499 0.000 1.000
#> GSM39190 2 0.0000 0.9499 0.000 1.000
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM39104 1 0.3030 0.845 0.904 0.004 0.092
#> GSM39105 1 0.0000 0.889 1.000 0.000 0.000
#> GSM39106 1 0.5728 0.724 0.772 0.196 0.032
#> GSM39107 2 0.6308 -0.152 0.492 0.508 0.000
#> GSM39108 1 0.4165 0.833 0.876 0.048 0.076
#> GSM39109 3 0.5810 0.365 0.000 0.336 0.664
#> GSM39110 1 0.9144 0.082 0.448 0.144 0.408
#> GSM39111 1 0.5864 0.610 0.704 0.008 0.288
#> GSM39112 1 0.6062 0.433 0.616 0.384 0.000
#> GSM39113 2 0.5431 0.456 0.284 0.716 0.000
#> GSM39114 2 0.0000 0.758 0.000 1.000 0.000
#> GSM39115 1 0.0000 0.889 1.000 0.000 0.000
#> GSM39148 1 0.0000 0.889 1.000 0.000 0.000
#> GSM39149 3 0.0000 0.773 0.000 0.000 1.000
#> GSM39150 1 0.4178 0.772 0.828 0.000 0.172
#> GSM39151 3 0.0000 0.773 0.000 0.000 1.000
#> GSM39152 3 0.0892 0.764 0.020 0.000 0.980
#> GSM39153 1 0.0000 0.889 1.000 0.000 0.000
#> GSM39154 1 0.0000 0.889 1.000 0.000 0.000
#> GSM39155 1 0.0000 0.889 1.000 0.000 0.000
#> GSM39156 1 0.1289 0.877 0.968 0.032 0.000
#> GSM39157 1 0.0000 0.889 1.000 0.000 0.000
#> GSM39158 1 0.0892 0.883 0.980 0.000 0.020
#> GSM39159 3 0.6126 0.228 0.400 0.000 0.600
#> GSM39160 1 0.5497 0.614 0.708 0.000 0.292
#> GSM39161 3 0.5138 0.578 0.252 0.000 0.748
#> GSM39162 1 0.0000 0.889 1.000 0.000 0.000
#> GSM39163 1 0.0000 0.889 1.000 0.000 0.000
#> GSM39164 1 0.0000 0.889 1.000 0.000 0.000
#> GSM39165 1 0.6309 0.123 0.504 0.000 0.496
#> GSM39166 1 0.2711 0.845 0.912 0.000 0.088
#> GSM39167 1 0.0000 0.889 1.000 0.000 0.000
#> GSM39168 1 0.0000 0.889 1.000 0.000 0.000
#> GSM39169 1 0.0000 0.889 1.000 0.000 0.000
#> GSM39170 1 0.0747 0.885 0.984 0.000 0.016
#> GSM39171 1 0.4750 0.723 0.784 0.000 0.216
#> GSM39172 3 0.0000 0.773 0.000 0.000 1.000
#> GSM39173 3 0.0424 0.770 0.000 0.008 0.992
#> GSM39174 1 0.0000 0.889 1.000 0.000 0.000
#> GSM39175 1 0.1860 0.870 0.948 0.000 0.052
#> GSM39176 1 0.0000 0.889 1.000 0.000 0.000
#> GSM39177 3 0.0424 0.770 0.008 0.000 0.992
#> GSM39178 3 0.6215 0.104 0.428 0.000 0.572
#> GSM39179 3 0.0000 0.773 0.000 0.000 1.000
#> GSM39180 3 0.0892 0.762 0.000 0.020 0.980
#> GSM39181 1 0.2356 0.856 0.928 0.000 0.072
#> GSM39182 3 0.3765 0.702 0.028 0.084 0.888
#> GSM39183 1 0.3619 0.807 0.864 0.000 0.136
#> GSM39184 1 0.0424 0.887 0.992 0.000 0.008
#> GSM39185 3 0.3619 0.685 0.136 0.000 0.864
#> GSM39186 1 0.0237 0.888 0.996 0.000 0.004
#> GSM39187 1 0.0000 0.889 1.000 0.000 0.000
#> GSM39116 2 0.2066 0.744 0.000 0.940 0.060
#> GSM39117 3 0.6286 -0.104 0.000 0.464 0.536
#> GSM39118 2 0.6126 0.392 0.000 0.600 0.400
#> GSM39119 2 0.6302 0.212 0.000 0.520 0.480
#> GSM39120 1 0.6154 0.381 0.592 0.408 0.000
#> GSM39121 2 0.2711 0.696 0.088 0.912 0.000
#> GSM39122 2 0.1860 0.729 0.052 0.948 0.000
#> GSM39123 3 0.6291 -0.117 0.000 0.468 0.532
#> GSM39124 2 0.0000 0.758 0.000 1.000 0.000
#> GSM39125 1 0.5650 0.573 0.688 0.312 0.000
#> GSM39126 2 0.1643 0.735 0.044 0.956 0.000
#> GSM39127 2 0.0000 0.758 0.000 1.000 0.000
#> GSM39128 2 0.0237 0.759 0.000 0.996 0.004
#> GSM39129 2 0.6225 0.334 0.000 0.568 0.432
#> GSM39130 3 0.6291 -0.117 0.000 0.468 0.532
#> GSM39131 2 0.0000 0.758 0.000 1.000 0.000
#> GSM39132 2 0.0237 0.759 0.000 0.996 0.004
#> GSM39133 2 0.6280 0.267 0.000 0.540 0.460
#> GSM39134 2 0.6140 0.384 0.000 0.596 0.404
#> GSM39135 2 0.1289 0.755 0.000 0.968 0.032
#> GSM39136 2 0.1411 0.754 0.000 0.964 0.036
#> GSM39137 2 0.0424 0.756 0.008 0.992 0.000
#> GSM39138 2 0.6267 0.290 0.000 0.548 0.452
#> GSM39139 2 0.5397 0.559 0.000 0.720 0.280
#> GSM39140 1 0.2537 0.845 0.920 0.080 0.000
#> GSM39141 1 0.0892 0.883 0.980 0.020 0.000
#> GSM39142 1 0.0592 0.886 0.988 0.012 0.000
#> GSM39143 1 0.0892 0.883 0.980 0.020 0.000
#> GSM39144 2 0.6235 0.326 0.000 0.564 0.436
#> GSM39145 2 0.3340 0.707 0.000 0.880 0.120
#> GSM39146 2 0.1031 0.757 0.000 0.976 0.024
#> GSM39147 2 0.0000 0.758 0.000 1.000 0.000
#> GSM39188 3 0.0000 0.773 0.000 0.000 1.000
#> GSM39189 3 0.0000 0.773 0.000 0.000 1.000
#> GSM39190 3 0.0000 0.773 0.000 0.000 1.000
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM39104 1 0.7831 0.2116 0.408 0.312 0.280 0.000
#> GSM39105 1 0.4532 0.7453 0.792 0.156 0.052 0.000
#> GSM39106 2 0.6112 0.4892 0.196 0.676 0.128 0.000
#> GSM39107 2 0.3229 0.6700 0.072 0.880 0.000 0.048
#> GSM39108 2 0.7146 -0.0724 0.412 0.456 0.132 0.000
#> GSM39109 4 0.7601 0.2720 0.004 0.216 0.276 0.504
#> GSM39110 2 0.7859 0.0433 0.160 0.428 0.396 0.016
#> GSM39111 3 0.7817 0.1252 0.296 0.288 0.416 0.000
#> GSM39112 2 0.3757 0.6667 0.152 0.828 0.000 0.020
#> GSM39113 2 0.2797 0.6549 0.032 0.900 0.000 0.068
#> GSM39114 2 0.4222 0.4586 0.000 0.728 0.000 0.272
#> GSM39115 1 0.3497 0.7829 0.860 0.104 0.036 0.000
#> GSM39148 1 0.1211 0.8020 0.960 0.040 0.000 0.000
#> GSM39149 3 0.2647 0.7180 0.000 0.000 0.880 0.120
#> GSM39150 1 0.7250 0.3491 0.516 0.168 0.316 0.000
#> GSM39151 3 0.3157 0.7125 0.000 0.004 0.852 0.144
#> GSM39152 3 0.3363 0.6800 0.020 0.052 0.888 0.040
#> GSM39153 1 0.1209 0.8087 0.964 0.032 0.004 0.000
#> GSM39154 1 0.1256 0.8107 0.964 0.028 0.008 0.000
#> GSM39155 1 0.1022 0.8101 0.968 0.032 0.000 0.000
#> GSM39156 1 0.5127 0.4232 0.632 0.356 0.012 0.000
#> GSM39157 1 0.0895 0.8082 0.976 0.020 0.004 0.000
#> GSM39158 1 0.3970 0.7596 0.840 0.084 0.076 0.000
#> GSM39159 3 0.7221 0.3436 0.328 0.092 0.556 0.024
#> GSM39160 3 0.7037 -0.0535 0.416 0.120 0.464 0.000
#> GSM39161 3 0.6834 0.3388 0.340 0.068 0.572 0.020
#> GSM39162 1 0.1637 0.7931 0.940 0.060 0.000 0.000
#> GSM39163 1 0.1174 0.8098 0.968 0.020 0.012 0.000
#> GSM39164 1 0.0921 0.8115 0.972 0.028 0.000 0.000
#> GSM39165 1 0.6267 -0.0370 0.484 0.032 0.472 0.012
#> GSM39166 1 0.6011 0.6293 0.688 0.132 0.180 0.000
#> GSM39167 1 0.0817 0.8058 0.976 0.024 0.000 0.000
#> GSM39168 1 0.1118 0.8035 0.964 0.036 0.000 0.000
#> GSM39169 1 0.2124 0.8071 0.932 0.028 0.040 0.000
#> GSM39170 1 0.4483 0.7437 0.808 0.088 0.104 0.000
#> GSM39171 1 0.6192 0.5687 0.652 0.104 0.244 0.000
#> GSM39172 3 0.4252 0.6141 0.000 0.004 0.744 0.252
#> GSM39173 3 0.3402 0.6989 0.000 0.004 0.832 0.164
#> GSM39174 1 0.1151 0.8102 0.968 0.024 0.008 0.000
#> GSM39175 1 0.2706 0.7888 0.900 0.020 0.080 0.000
#> GSM39176 1 0.0817 0.8064 0.976 0.024 0.000 0.000
#> GSM39177 3 0.3006 0.7214 0.008 0.012 0.888 0.092
#> GSM39178 3 0.6578 0.4424 0.244 0.136 0.620 0.000
#> GSM39179 3 0.2888 0.7194 0.000 0.004 0.872 0.124
#> GSM39180 3 0.4699 0.5131 0.004 0.000 0.676 0.320
#> GSM39181 1 0.4920 0.7094 0.776 0.088 0.136 0.000
#> GSM39182 4 0.7250 -0.1324 0.036 0.060 0.428 0.476
#> GSM39183 1 0.6134 0.5908 0.668 0.116 0.216 0.000
#> GSM39184 1 0.1510 0.8098 0.956 0.016 0.028 0.000
#> GSM39185 3 0.7047 0.5570 0.196 0.092 0.656 0.056
#> GSM39186 1 0.2830 0.7956 0.900 0.060 0.040 0.000
#> GSM39187 1 0.1302 0.8095 0.956 0.044 0.000 0.000
#> GSM39116 4 0.3172 0.6671 0.000 0.160 0.000 0.840
#> GSM39117 4 0.3123 0.6476 0.000 0.000 0.156 0.844
#> GSM39118 4 0.2222 0.7173 0.000 0.016 0.060 0.924
#> GSM39119 4 0.2345 0.7011 0.000 0.000 0.100 0.900
#> GSM39120 2 0.4541 0.6661 0.152 0.804 0.024 0.020
#> GSM39121 2 0.4636 0.5929 0.040 0.772 0.000 0.188
#> GSM39122 2 0.4642 0.5258 0.020 0.740 0.000 0.240
#> GSM39123 4 0.2921 0.6647 0.000 0.000 0.140 0.860
#> GSM39124 4 0.4985 0.2162 0.000 0.468 0.000 0.532
#> GSM39125 2 0.5134 0.5924 0.232 0.732 0.020 0.016
#> GSM39126 2 0.4050 0.6050 0.024 0.808 0.000 0.168
#> GSM39127 4 0.4888 0.3705 0.000 0.412 0.000 0.588
#> GSM39128 4 0.4972 0.2655 0.000 0.456 0.000 0.544
#> GSM39129 4 0.2081 0.7094 0.000 0.000 0.084 0.916
#> GSM39130 4 0.3123 0.6476 0.000 0.000 0.156 0.844
#> GSM39131 4 0.4955 0.3026 0.000 0.444 0.000 0.556
#> GSM39132 4 0.4543 0.5124 0.000 0.324 0.000 0.676
#> GSM39133 4 0.2125 0.7134 0.000 0.004 0.076 0.920
#> GSM39134 4 0.1474 0.7163 0.000 0.000 0.052 0.948
#> GSM39135 4 0.2973 0.6753 0.000 0.144 0.000 0.856
#> GSM39136 4 0.3356 0.6571 0.000 0.176 0.000 0.824
#> GSM39137 2 0.5039 0.1282 0.004 0.592 0.000 0.404
#> GSM39138 4 0.2281 0.7029 0.000 0.000 0.096 0.904
#> GSM39139 4 0.2450 0.7056 0.000 0.072 0.016 0.912
#> GSM39140 1 0.4843 0.3269 0.604 0.396 0.000 0.000
#> GSM39141 1 0.4164 0.5926 0.736 0.264 0.000 0.000
#> GSM39142 1 0.4040 0.6219 0.752 0.248 0.000 0.000
#> GSM39143 1 0.4304 0.5664 0.716 0.284 0.000 0.000
#> GSM39144 4 0.2222 0.7171 0.000 0.016 0.060 0.924
#> GSM39145 4 0.3208 0.6766 0.000 0.148 0.004 0.848
#> GSM39146 4 0.3764 0.6279 0.000 0.216 0.000 0.784
#> GSM39147 4 0.4585 0.5022 0.000 0.332 0.000 0.668
#> GSM39188 3 0.3157 0.7119 0.000 0.004 0.852 0.144
#> GSM39189 3 0.2918 0.7207 0.000 0.008 0.876 0.116
#> GSM39190 3 0.2973 0.7122 0.000 0.000 0.856 0.144
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM39104 4 0.7978 0.3384 0.224 0.000 0.124 0.440 0.212
#> GSM39105 1 0.6586 0.2701 0.564 0.000 0.036 0.272 0.128
#> GSM39106 5 0.7183 0.2544 0.156 0.000 0.064 0.260 0.520
#> GSM39107 5 0.2713 0.5908 0.036 0.004 0.000 0.072 0.888
#> GSM39108 5 0.8178 -0.0898 0.284 0.000 0.112 0.248 0.356
#> GSM39109 3 0.8716 0.0953 0.008 0.280 0.304 0.180 0.228
#> GSM39110 5 0.8591 -0.0420 0.128 0.012 0.312 0.232 0.316
#> GSM39111 4 0.8260 0.2926 0.212 0.000 0.284 0.364 0.140
#> GSM39112 5 0.5175 0.4959 0.172 0.004 0.008 0.100 0.716
#> GSM39113 5 0.2197 0.6028 0.004 0.012 0.004 0.064 0.916
#> GSM39114 5 0.3851 0.4885 0.000 0.212 0.004 0.016 0.768
#> GSM39115 1 0.5590 0.3529 0.628 0.000 0.016 0.288 0.068
#> GSM39148 1 0.0865 0.7054 0.972 0.000 0.000 0.004 0.024
#> GSM39149 3 0.2659 0.7984 0.000 0.060 0.888 0.052 0.000
#> GSM39150 4 0.6857 0.5388 0.268 0.000 0.132 0.548 0.052
#> GSM39151 3 0.2228 0.7951 0.000 0.048 0.912 0.040 0.000
#> GSM39152 3 0.4594 0.6598 0.012 0.028 0.760 0.184 0.016
#> GSM39153 1 0.2364 0.7031 0.908 0.000 0.008 0.064 0.020
#> GSM39154 1 0.3064 0.6755 0.856 0.000 0.036 0.108 0.000
#> GSM39155 1 0.3080 0.6574 0.844 0.000 0.008 0.140 0.008
#> GSM39156 1 0.5548 0.4969 0.668 0.000 0.020 0.084 0.228
#> GSM39157 1 0.1983 0.7053 0.924 0.000 0.008 0.060 0.008
#> GSM39158 4 0.4664 0.3699 0.436 0.000 0.008 0.552 0.004
#> GSM39159 4 0.6794 0.4973 0.200 0.008 0.268 0.516 0.008
#> GSM39160 4 0.7279 0.4219 0.316 0.000 0.208 0.440 0.036
#> GSM39161 4 0.6348 0.5409 0.140 0.020 0.216 0.616 0.008
#> GSM39162 1 0.1168 0.7039 0.960 0.000 0.000 0.008 0.032
#> GSM39163 1 0.2392 0.6843 0.888 0.000 0.004 0.104 0.004
#> GSM39164 1 0.2915 0.6914 0.860 0.000 0.000 0.116 0.024
#> GSM39165 1 0.7478 -0.2034 0.400 0.008 0.348 0.216 0.028
#> GSM39166 4 0.4956 0.5591 0.312 0.000 0.040 0.644 0.004
#> GSM39167 1 0.0992 0.7055 0.968 0.000 0.000 0.024 0.008
#> GSM39168 1 0.0854 0.7062 0.976 0.000 0.004 0.008 0.012
#> GSM39169 1 0.4173 0.5945 0.756 0.000 0.020 0.212 0.012
#> GSM39170 4 0.4886 0.3205 0.468 0.000 0.016 0.512 0.004
#> GSM39171 1 0.7183 -0.3024 0.404 0.000 0.200 0.368 0.028
#> GSM39172 3 0.4647 0.7246 0.000 0.184 0.732 0.084 0.000
#> GSM39173 3 0.3597 0.7926 0.000 0.116 0.832 0.044 0.008
#> GSM39174 1 0.2577 0.6929 0.892 0.000 0.008 0.084 0.016
#> GSM39175 1 0.4367 0.5547 0.748 0.000 0.060 0.192 0.000
#> GSM39176 1 0.1281 0.7084 0.956 0.000 0.000 0.032 0.012
#> GSM39177 3 0.3426 0.7669 0.012 0.052 0.852 0.084 0.000
#> GSM39178 4 0.6021 0.5179 0.124 0.004 0.236 0.624 0.012
#> GSM39179 3 0.2144 0.8042 0.000 0.068 0.912 0.020 0.000
#> GSM39180 3 0.5163 0.5733 0.000 0.296 0.636 0.068 0.000
#> GSM39181 4 0.4626 0.5011 0.364 0.000 0.020 0.616 0.000
#> GSM39182 2 0.7639 -0.2273 0.028 0.400 0.360 0.192 0.020
#> GSM39183 4 0.5098 0.5672 0.300 0.000 0.052 0.644 0.004
#> GSM39184 1 0.4296 0.5212 0.720 0.000 0.012 0.256 0.012
#> GSM39185 4 0.6214 0.4222 0.076 0.048 0.244 0.628 0.004
#> GSM39186 1 0.5148 0.3959 0.664 0.000 0.028 0.280 0.028
#> GSM39187 1 0.1830 0.7112 0.932 0.000 0.000 0.040 0.028
#> GSM39116 2 0.3559 0.6328 0.000 0.804 0.012 0.008 0.176
#> GSM39117 2 0.4101 0.5978 0.000 0.768 0.184 0.048 0.000
#> GSM39118 2 0.3068 0.7108 0.000 0.872 0.084 0.028 0.016
#> GSM39119 2 0.3340 0.6746 0.000 0.840 0.124 0.032 0.004
#> GSM39120 5 0.5496 0.4942 0.196 0.016 0.008 0.084 0.696
#> GSM39121 5 0.3740 0.5707 0.044 0.128 0.000 0.008 0.820
#> GSM39122 5 0.3381 0.5504 0.016 0.160 0.000 0.004 0.820
#> GSM39123 2 0.4065 0.6030 0.000 0.772 0.180 0.048 0.000
#> GSM39124 5 0.4967 -0.0444 0.000 0.464 0.004 0.020 0.512
#> GSM39125 5 0.5860 0.4090 0.240 0.008 0.004 0.116 0.632
#> GSM39126 5 0.3509 0.5748 0.016 0.132 0.000 0.020 0.832
#> GSM39127 2 0.4967 0.0937 0.000 0.512 0.004 0.020 0.464
#> GSM39128 5 0.4973 -0.0382 0.000 0.480 0.004 0.020 0.496
#> GSM39129 2 0.3478 0.6834 0.000 0.828 0.136 0.032 0.004
#> GSM39130 2 0.4101 0.5990 0.000 0.768 0.184 0.048 0.000
#> GSM39131 2 0.5192 0.0152 0.000 0.488 0.004 0.032 0.476
#> GSM39132 2 0.4691 0.3955 0.000 0.636 0.004 0.020 0.340
#> GSM39133 2 0.3412 0.6844 0.000 0.848 0.096 0.048 0.008
#> GSM39134 2 0.2736 0.7158 0.000 0.892 0.068 0.016 0.024
#> GSM39135 2 0.2966 0.6549 0.000 0.848 0.000 0.016 0.136
#> GSM39136 2 0.2723 0.6660 0.000 0.864 0.000 0.012 0.124
#> GSM39137 5 0.5210 0.2784 0.008 0.344 0.004 0.032 0.612
#> GSM39138 2 0.3115 0.6915 0.000 0.852 0.112 0.036 0.000
#> GSM39139 2 0.3600 0.6911 0.000 0.848 0.044 0.028 0.080
#> GSM39140 1 0.5086 0.4452 0.636 0.000 0.000 0.060 0.304
#> GSM39141 1 0.3910 0.5923 0.772 0.000 0.000 0.032 0.196
#> GSM39142 1 0.3885 0.6098 0.784 0.000 0.000 0.040 0.176
#> GSM39143 1 0.4129 0.5807 0.756 0.000 0.000 0.040 0.204
#> GSM39144 2 0.3106 0.6886 0.000 0.844 0.132 0.024 0.000
#> GSM39145 2 0.4126 0.6436 0.000 0.796 0.032 0.024 0.148
#> GSM39146 2 0.4143 0.6084 0.000 0.764 0.004 0.036 0.196
#> GSM39147 2 0.5176 0.3598 0.000 0.608 0.012 0.032 0.348
#> GSM39188 3 0.2628 0.8013 0.000 0.088 0.884 0.028 0.000
#> GSM39189 3 0.4504 0.7481 0.000 0.068 0.768 0.152 0.012
#> GSM39190 3 0.3354 0.7982 0.000 0.088 0.844 0.068 0.000
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM39104 6 0.834 -0.1145 0.108 0.000 0.096 0.172 0.300 0.324
#> GSM39105 1 0.784 0.1847 0.432 0.000 0.056 0.100 0.216 0.196
#> GSM39106 6 0.809 0.2960 0.084 0.044 0.068 0.132 0.176 0.496
#> GSM39107 6 0.394 0.5568 0.044 0.120 0.004 0.008 0.020 0.804
#> GSM39108 6 0.866 0.1308 0.204 0.012 0.084 0.176 0.160 0.364
#> GSM39109 4 0.848 -0.0701 0.004 0.144 0.200 0.356 0.076 0.220
#> GSM39110 6 0.897 0.0858 0.108 0.020 0.216 0.148 0.180 0.328
#> GSM39111 5 0.896 0.0125 0.104 0.008 0.232 0.188 0.256 0.212
#> GSM39112 6 0.517 0.5066 0.204 0.072 0.000 0.028 0.012 0.684
#> GSM39113 6 0.436 0.5275 0.008 0.168 0.008 0.040 0.016 0.760
#> GSM39114 2 0.438 -0.1220 0.000 0.520 0.004 0.016 0.000 0.460
#> GSM39115 1 0.668 0.2804 0.500 0.000 0.008 0.064 0.284 0.144
#> GSM39148 1 0.155 0.6946 0.944 0.000 0.000 0.020 0.020 0.016
#> GSM39149 3 0.324 0.7763 0.000 0.016 0.856 0.080 0.028 0.020
#> GSM39150 5 0.743 0.5396 0.152 0.000 0.120 0.092 0.532 0.104
#> GSM39151 3 0.385 0.7858 0.000 0.020 0.804 0.128 0.036 0.012
#> GSM39152 3 0.502 0.6105 0.012 0.000 0.736 0.080 0.104 0.068
#> GSM39153 1 0.431 0.6847 0.792 0.000 0.028 0.040 0.100 0.040
#> GSM39154 1 0.484 0.6674 0.748 0.000 0.040 0.052 0.132 0.028
#> GSM39155 1 0.455 0.6263 0.720 0.000 0.004 0.032 0.208 0.036
#> GSM39156 1 0.644 0.4607 0.576 0.020 0.004 0.084 0.068 0.248
#> GSM39157 1 0.311 0.6936 0.852 0.000 0.004 0.024 0.100 0.020
#> GSM39158 5 0.442 0.4124 0.332 0.000 0.004 0.020 0.636 0.008
#> GSM39159 5 0.740 0.4354 0.140 0.012 0.224 0.084 0.508 0.032
#> GSM39160 5 0.820 0.4050 0.176 0.000 0.200 0.108 0.412 0.104
#> GSM39161 5 0.513 0.6011 0.084 0.000 0.136 0.044 0.720 0.016
#> GSM39162 1 0.159 0.6934 0.940 0.000 0.000 0.020 0.008 0.032
#> GSM39163 1 0.388 0.6532 0.760 0.000 0.000 0.024 0.196 0.020
#> GSM39164 1 0.451 0.6859 0.772 0.000 0.012 0.068 0.104 0.044
#> GSM39165 1 0.804 0.0132 0.380 0.004 0.264 0.092 0.208 0.052
#> GSM39166 5 0.384 0.6383 0.132 0.000 0.020 0.024 0.804 0.020
#> GSM39167 1 0.170 0.6979 0.928 0.000 0.000 0.024 0.048 0.000
#> GSM39168 1 0.136 0.6979 0.952 0.000 0.000 0.016 0.012 0.020
#> GSM39169 1 0.454 0.6255 0.732 0.000 0.012 0.044 0.192 0.020
#> GSM39170 5 0.539 0.4334 0.320 0.000 0.024 0.016 0.596 0.044
#> GSM39171 1 0.803 -0.0814 0.376 0.000 0.136 0.084 0.304 0.100
#> GSM39172 3 0.561 0.5247 0.000 0.028 0.544 0.364 0.052 0.012
#> GSM39173 3 0.541 0.7511 0.004 0.056 0.708 0.152 0.048 0.032
#> GSM39174 1 0.392 0.6809 0.804 0.000 0.008 0.052 0.112 0.024
#> GSM39175 1 0.675 0.4585 0.584 0.000 0.108 0.068 0.184 0.056
#> GSM39176 1 0.248 0.6978 0.884 0.000 0.000 0.024 0.084 0.008
#> GSM39177 3 0.319 0.7449 0.012 0.008 0.864 0.060 0.048 0.008
#> GSM39178 5 0.630 0.5339 0.048 0.000 0.168 0.100 0.628 0.056
#> GSM39179 3 0.309 0.7798 0.000 0.012 0.860 0.084 0.032 0.012
#> GSM39180 3 0.674 0.4915 0.000 0.080 0.500 0.296 0.112 0.012
#> GSM39181 5 0.380 0.5963 0.196 0.000 0.008 0.012 0.768 0.016
#> GSM39182 4 0.732 0.0706 0.020 0.092 0.236 0.524 0.096 0.032
#> GSM39183 5 0.358 0.6473 0.116 0.000 0.040 0.004 0.820 0.020
#> GSM39184 1 0.545 0.5489 0.644 0.000 0.024 0.044 0.252 0.036
#> GSM39185 5 0.519 0.5452 0.040 0.012 0.156 0.060 0.720 0.012
#> GSM39186 1 0.614 0.4859 0.608 0.000 0.024 0.060 0.228 0.080
#> GSM39187 1 0.336 0.6989 0.836 0.000 0.000 0.024 0.096 0.044
#> GSM39116 2 0.415 0.2547 0.000 0.716 0.008 0.244 0.004 0.028
#> GSM39117 4 0.477 0.5997 0.000 0.364 0.060 0.576 0.000 0.000
#> GSM39118 2 0.488 -0.4661 0.000 0.492 0.048 0.456 0.000 0.004
#> GSM39119 4 0.486 0.4911 0.000 0.436 0.040 0.516 0.000 0.008
#> GSM39120 6 0.570 0.4955 0.220 0.076 0.004 0.024 0.028 0.648
#> GSM39121 6 0.568 0.3439 0.068 0.360 0.004 0.032 0.000 0.536
#> GSM39122 6 0.553 0.1528 0.028 0.444 0.004 0.052 0.000 0.472
#> GSM39123 4 0.483 0.5957 0.000 0.376 0.052 0.568 0.004 0.000
#> GSM39124 2 0.403 0.4190 0.004 0.740 0.004 0.028 0.004 0.220
#> GSM39125 6 0.647 0.4293 0.236 0.052 0.000 0.020 0.128 0.564
#> GSM39126 6 0.551 0.3650 0.032 0.332 0.004 0.060 0.000 0.572
#> GSM39127 2 0.342 0.4757 0.000 0.792 0.000 0.040 0.000 0.168
#> GSM39128 2 0.410 0.3098 0.000 0.676 0.000 0.032 0.000 0.292
#> GSM39129 2 0.514 -0.3842 0.000 0.504 0.072 0.420 0.000 0.004
#> GSM39130 4 0.477 0.5998 0.000 0.364 0.060 0.576 0.000 0.000
#> GSM39131 2 0.408 0.4509 0.000 0.736 0.000 0.072 0.000 0.192
#> GSM39132 2 0.296 0.4529 0.000 0.848 0.000 0.084 0.000 0.068
#> GSM39133 4 0.450 0.5357 0.000 0.432 0.032 0.536 0.000 0.000
#> GSM39134 4 0.465 0.4413 0.000 0.476 0.040 0.484 0.000 0.000
#> GSM39135 2 0.358 0.2599 0.000 0.748 0.008 0.236 0.004 0.004
#> GSM39136 2 0.363 0.2927 0.000 0.760 0.004 0.216 0.004 0.016
#> GSM39137 2 0.486 0.1590 0.008 0.616 0.008 0.040 0.000 0.328
#> GSM39138 2 0.503 -0.4304 0.000 0.484 0.060 0.452 0.000 0.004
#> GSM39139 2 0.461 0.2020 0.000 0.684 0.040 0.252 0.000 0.024
#> GSM39140 1 0.530 0.4635 0.640 0.032 0.000 0.028 0.028 0.272
#> GSM39141 1 0.338 0.6444 0.800 0.000 0.000 0.024 0.008 0.168
#> GSM39142 1 0.443 0.5982 0.708 0.000 0.000 0.032 0.028 0.232
#> GSM39143 1 0.415 0.5878 0.712 0.000 0.000 0.024 0.016 0.248
#> GSM39144 2 0.549 -0.3077 0.000 0.524 0.072 0.384 0.004 0.016
#> GSM39145 2 0.460 0.2559 0.000 0.704 0.052 0.224 0.004 0.016
#> GSM39146 2 0.458 0.1928 0.000 0.652 0.004 0.300 0.008 0.036
#> GSM39147 2 0.388 0.4660 0.000 0.792 0.008 0.076 0.004 0.120
#> GSM39188 3 0.415 0.7760 0.000 0.012 0.776 0.144 0.056 0.012
#> GSM39189 3 0.479 0.7152 0.000 0.000 0.704 0.168 0.112 0.016
#> GSM39190 3 0.423 0.7763 0.000 0.028 0.772 0.140 0.056 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) other(p) protocol(p) k
#> CV:skmeans 82 8.14e-02 3.50e-07 1.71e-05 2
#> CV:skmeans 68 1.85e-01 2.21e-07 1.59e-08 3
#> CV:skmeans 66 5.34e-03 4.49e-09 2.77e-09 4
#> CV:skmeans 57 2.46e-04 1.64e-08 2.65e-09 5
#> CV:skmeans 42 1.67e-08 1.25e-08 7.63e-15 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 8353 rows and 87 columns.
#> Top rows (835, 1670, 2506, 3341, 4176) are extracted by 'CV' method.
#> Subgroups are detected by 'pam' method.
#> Performed in total 1250 partitions by row resampling.
#> There is no best k.
#>
#> 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 0.531 0.731 0.894 0.2928 0.759 0.759
#> 3 3 0.335 0.612 0.808 1.0438 0.599 0.476
#> 4 4 0.342 0.622 0.799 0.0273 1.000 1.000
#> 5 5 0.335 0.537 0.793 0.0154 0.997 0.991
#> 6 6 0.352 0.461 0.792 0.0204 0.991 0.975
suggest_best_k() suggests the best \(k\) based on these statistics. The rules are as follows:
suggest_best_k(res)
#> [1] NA
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
#> GSM39104 1 0.0000 0.8725 1.000 0.000
#> GSM39105 1 0.0000 0.8725 1.000 0.000
#> GSM39106 1 0.0000 0.8725 1.000 0.000
#> GSM39107 1 0.0000 0.8725 1.000 0.000
#> GSM39108 1 0.0000 0.8725 1.000 0.000
#> GSM39109 1 0.8327 0.5760 0.736 0.264
#> GSM39110 1 0.0000 0.8725 1.000 0.000
#> GSM39111 1 0.0000 0.8725 1.000 0.000
#> GSM39112 1 0.0000 0.8725 1.000 0.000
#> GSM39113 1 0.0000 0.8725 1.000 0.000
#> GSM39114 1 0.0376 0.8697 0.996 0.004
#> GSM39115 1 0.0000 0.8725 1.000 0.000
#> GSM39148 1 0.0000 0.8725 1.000 0.000
#> GSM39149 1 0.9710 0.2665 0.600 0.400
#> GSM39150 1 0.0000 0.8725 1.000 0.000
#> GSM39151 1 0.9993 -0.0750 0.516 0.484
#> GSM39152 1 0.0000 0.8725 1.000 0.000
#> GSM39153 1 0.0000 0.8725 1.000 0.000
#> GSM39154 1 0.0000 0.8725 1.000 0.000
#> GSM39155 1 0.0000 0.8725 1.000 0.000
#> GSM39156 1 0.0000 0.8725 1.000 0.000
#> GSM39157 1 0.0000 0.8725 1.000 0.000
#> GSM39158 1 0.0000 0.8725 1.000 0.000
#> GSM39159 1 0.0000 0.8725 1.000 0.000
#> GSM39160 1 0.0000 0.8725 1.000 0.000
#> GSM39161 1 0.0000 0.8725 1.000 0.000
#> GSM39162 1 0.0000 0.8725 1.000 0.000
#> GSM39163 1 0.0000 0.8725 1.000 0.000
#> GSM39164 1 0.0000 0.8725 1.000 0.000
#> GSM39165 1 0.0000 0.8725 1.000 0.000
#> GSM39166 1 0.0000 0.8725 1.000 0.000
#> GSM39167 1 0.0000 0.8725 1.000 0.000
#> GSM39168 1 0.0000 0.8725 1.000 0.000
#> GSM39169 1 0.0000 0.8725 1.000 0.000
#> GSM39170 1 0.0000 0.8725 1.000 0.000
#> GSM39171 1 0.0000 0.8725 1.000 0.000
#> GSM39172 1 0.9993 -0.0779 0.516 0.484
#> GSM39173 1 0.6148 0.7348 0.848 0.152
#> GSM39174 1 0.0000 0.8725 1.000 0.000
#> GSM39175 1 0.0000 0.8725 1.000 0.000
#> GSM39176 1 0.0000 0.8725 1.000 0.000
#> GSM39177 1 0.6438 0.7212 0.836 0.164
#> GSM39178 1 0.0000 0.8725 1.000 0.000
#> GSM39179 1 0.9775 0.2285 0.588 0.412
#> GSM39180 1 0.9993 -0.0730 0.516 0.484
#> GSM39181 1 0.0000 0.8725 1.000 0.000
#> GSM39182 1 0.9580 0.3198 0.620 0.380
#> GSM39183 1 0.0000 0.8725 1.000 0.000
#> GSM39184 1 0.0000 0.8725 1.000 0.000
#> GSM39185 1 0.0000 0.8725 1.000 0.000
#> GSM39186 1 0.0000 0.8725 1.000 0.000
#> GSM39187 1 0.0000 0.8725 1.000 0.000
#> GSM39116 1 0.9983 -0.0372 0.524 0.476
#> GSM39117 2 0.0000 0.7702 0.000 1.000
#> GSM39118 2 0.9044 0.6726 0.320 0.680
#> GSM39119 2 0.7219 0.8213 0.200 0.800
#> GSM39120 1 0.0000 0.8725 1.000 0.000
#> GSM39121 1 0.0000 0.8725 1.000 0.000
#> GSM39122 1 0.0938 0.8641 0.988 0.012
#> GSM39123 2 0.0000 0.7702 0.000 1.000
#> GSM39124 1 0.3274 0.8264 0.940 0.060
#> GSM39125 1 0.0000 0.8725 1.000 0.000
#> GSM39126 1 0.0000 0.8725 1.000 0.000
#> GSM39127 1 0.9686 0.2784 0.604 0.396
#> GSM39128 1 0.7219 0.6746 0.800 0.200
#> GSM39129 2 0.7139 0.8222 0.196 0.804
#> GSM39130 2 0.0000 0.7702 0.000 1.000
#> GSM39131 1 0.4022 0.8087 0.920 0.080
#> GSM39132 1 0.8555 0.5472 0.720 0.280
#> GSM39133 2 0.0000 0.7702 0.000 1.000
#> GSM39134 2 0.7219 0.8213 0.200 0.800
#> GSM39135 1 0.9732 0.2542 0.596 0.404
#> GSM39136 2 0.8713 0.7226 0.292 0.708
#> GSM39137 1 0.0000 0.8725 1.000 0.000
#> GSM39138 2 0.7139 0.8222 0.196 0.804
#> GSM39139 1 0.9754 0.2414 0.592 0.408
#> GSM39140 1 0.0000 0.8725 1.000 0.000
#> GSM39141 1 0.0000 0.8725 1.000 0.000
#> GSM39142 1 0.0000 0.8725 1.000 0.000
#> GSM39143 1 0.0000 0.8725 1.000 0.000
#> GSM39144 2 0.8016 0.7845 0.244 0.756
#> GSM39145 1 0.9522 0.3442 0.628 0.372
#> GSM39146 1 0.9710 0.2665 0.600 0.400
#> GSM39147 1 0.6887 0.6962 0.816 0.184
#> GSM39188 2 0.9460 0.5663 0.364 0.636
#> GSM39189 1 0.8909 0.4933 0.692 0.308
#> GSM39190 1 0.9795 0.2146 0.584 0.416
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM39104 2 0.5733 0.6184 0.324 0.676 0.000
#> GSM39105 2 0.5016 0.7158 0.240 0.760 0.000
#> GSM39106 1 0.5785 0.5531 0.668 0.332 0.000
#> GSM39107 2 0.0892 0.7269 0.020 0.980 0.000
#> GSM39108 2 0.4796 0.7314 0.220 0.780 0.000
#> GSM39109 2 0.2056 0.7232 0.024 0.952 0.024
#> GSM39110 1 0.5988 0.2680 0.632 0.368 0.000
#> GSM39111 2 0.5882 0.5998 0.348 0.652 0.000
#> GSM39112 2 0.0892 0.7269 0.020 0.980 0.000
#> GSM39113 2 0.0892 0.7269 0.020 0.980 0.000
#> GSM39114 2 0.0592 0.7214 0.012 0.988 0.000
#> GSM39115 2 0.5968 0.5774 0.364 0.636 0.000
#> GSM39148 1 0.0000 0.7987 1.000 0.000 0.000
#> GSM39149 1 0.7796 0.1196 0.552 0.056 0.392
#> GSM39150 1 0.1753 0.7980 0.952 0.048 0.000
#> GSM39151 3 0.9520 0.3207 0.200 0.340 0.460
#> GSM39152 1 0.1163 0.8013 0.972 0.028 0.000
#> GSM39153 1 0.0000 0.7987 1.000 0.000 0.000
#> GSM39154 1 0.0892 0.8005 0.980 0.020 0.000
#> GSM39155 1 0.6274 -0.1966 0.544 0.456 0.000
#> GSM39156 1 0.0592 0.7995 0.988 0.012 0.000
#> GSM39157 2 0.5859 0.6325 0.344 0.656 0.000
#> GSM39158 1 0.1031 0.8006 0.976 0.024 0.000
#> GSM39159 1 0.5178 0.6271 0.744 0.256 0.000
#> GSM39160 1 0.2959 0.7839 0.900 0.100 0.000
#> GSM39161 1 0.1529 0.7996 0.960 0.040 0.000
#> GSM39162 1 0.0000 0.7987 1.000 0.000 0.000
#> GSM39163 1 0.1964 0.7939 0.944 0.056 0.000
#> GSM39164 1 0.0000 0.7987 1.000 0.000 0.000
#> GSM39165 1 0.4702 0.6603 0.788 0.212 0.000
#> GSM39166 1 0.1643 0.7993 0.956 0.044 0.000
#> GSM39167 1 0.0000 0.7987 1.000 0.000 0.000
#> GSM39168 1 0.0000 0.7987 1.000 0.000 0.000
#> GSM39169 1 0.0000 0.7987 1.000 0.000 0.000
#> GSM39170 1 0.0747 0.7997 0.984 0.016 0.000
#> GSM39171 2 0.5431 0.6922 0.284 0.716 0.000
#> GSM39172 1 0.7030 0.1664 0.580 0.024 0.396
#> GSM39173 1 0.4551 0.6820 0.840 0.020 0.140
#> GSM39174 1 0.1529 0.7956 0.960 0.040 0.000
#> GSM39175 1 0.0592 0.8012 0.988 0.012 0.000
#> GSM39176 1 0.0000 0.7987 1.000 0.000 0.000
#> GSM39177 1 0.7724 0.5519 0.680 0.156 0.164
#> GSM39178 1 0.2066 0.7968 0.940 0.060 0.000
#> GSM39179 1 0.9147 -0.1577 0.444 0.144 0.412
#> GSM39180 3 0.9224 0.1830 0.408 0.152 0.440
#> GSM39181 1 0.5785 0.4034 0.668 0.332 0.000
#> GSM39182 1 0.7571 0.2519 0.592 0.052 0.356
#> GSM39183 1 0.5178 0.5903 0.744 0.256 0.000
#> GSM39184 1 0.5465 0.5126 0.712 0.288 0.000
#> GSM39185 1 0.5363 0.5905 0.724 0.276 0.000
#> GSM39186 2 0.6274 0.3748 0.456 0.544 0.000
#> GSM39187 1 0.2066 0.7956 0.940 0.060 0.000
#> GSM39116 2 0.6255 0.3856 0.012 0.668 0.320
#> GSM39117 3 0.0000 0.6918 0.000 0.000 1.000
#> GSM39118 3 0.6617 0.3847 0.012 0.388 0.600
#> GSM39119 3 0.4968 0.6818 0.012 0.188 0.800
#> GSM39120 2 0.3412 0.7227 0.124 0.876 0.000
#> GSM39121 2 0.2878 0.7353 0.096 0.904 0.000
#> GSM39122 2 0.1753 0.7353 0.048 0.952 0.000
#> GSM39123 3 0.0000 0.6918 0.000 0.000 1.000
#> GSM39124 2 0.3618 0.7503 0.104 0.884 0.012
#> GSM39125 2 0.3192 0.7494 0.112 0.888 0.000
#> GSM39126 2 0.3267 0.7358 0.116 0.884 0.000
#> GSM39127 2 0.2749 0.6966 0.012 0.924 0.064
#> GSM39128 2 0.6488 0.5584 0.192 0.744 0.064
#> GSM39129 3 0.4733 0.6867 0.004 0.196 0.800
#> GSM39130 3 0.0000 0.6918 0.000 0.000 1.000
#> GSM39131 2 0.0983 0.7213 0.016 0.980 0.004
#> GSM39132 2 0.4475 0.7006 0.072 0.864 0.064
#> GSM39133 3 0.0000 0.6918 0.000 0.000 1.000
#> GSM39134 3 0.5355 0.6939 0.032 0.168 0.800
#> GSM39135 2 0.6357 0.3616 0.012 0.652 0.336
#> GSM39136 3 0.6584 0.4146 0.012 0.380 0.608
#> GSM39137 2 0.4291 0.7431 0.180 0.820 0.000
#> GSM39138 3 0.5292 0.6528 0.172 0.028 0.800
#> GSM39139 2 0.6632 0.2446 0.012 0.596 0.392
#> GSM39140 2 0.6045 0.5561 0.380 0.620 0.000
#> GSM39141 2 0.5178 0.7136 0.256 0.744 0.000
#> GSM39142 2 0.5058 0.7168 0.244 0.756 0.000
#> GSM39143 2 0.5016 0.7190 0.240 0.760 0.000
#> GSM39144 3 0.5775 0.6127 0.012 0.260 0.728
#> GSM39145 2 0.7442 0.3426 0.048 0.604 0.348
#> GSM39146 2 0.5268 0.5927 0.012 0.776 0.212
#> GSM39147 2 0.3112 0.7223 0.028 0.916 0.056
#> GSM39188 3 0.7158 0.3607 0.372 0.032 0.596
#> GSM39189 1 0.6798 0.4802 0.696 0.048 0.256
#> GSM39190 2 0.8370 0.0149 0.084 0.500 0.416
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM39104 2 0.5994 0.6087 0.296 0.636 NA 0.000
#> GSM39105 2 0.5050 0.7253 0.176 0.756 NA 0.000
#> GSM39106 1 0.5901 0.5899 0.652 0.280 NA 0.000
#> GSM39107 2 0.0592 0.7359 0.016 0.984 NA 0.000
#> GSM39108 2 0.5212 0.7252 0.192 0.740 NA 0.000
#> GSM39109 2 0.1411 0.7349 0.020 0.960 NA 0.020
#> GSM39110 1 0.4746 0.2804 0.632 0.368 NA 0.000
#> GSM39111 2 0.5925 0.6362 0.284 0.648 NA 0.000
#> GSM39112 2 0.0592 0.7359 0.016 0.984 NA 0.000
#> GSM39113 2 0.0592 0.7359 0.016 0.984 NA 0.000
#> GSM39114 2 0.0336 0.7306 0.008 0.992 NA 0.000
#> GSM39115 2 0.6016 0.6147 0.300 0.632 NA 0.000
#> GSM39148 1 0.0000 0.7933 1.000 0.000 NA 0.000
#> GSM39149 1 0.7403 0.2186 0.548 0.060 NA 0.336
#> GSM39150 1 0.1888 0.7914 0.940 0.044 NA 0.000
#> GSM39151 4 0.8604 0.3405 0.176 0.324 NA 0.444
#> GSM39152 1 0.1022 0.7958 0.968 0.032 NA 0.000
#> GSM39153 1 0.0000 0.7933 1.000 0.000 NA 0.000
#> GSM39154 1 0.0707 0.7953 0.980 0.020 NA 0.000
#> GSM39155 1 0.6209 -0.2461 0.492 0.456 NA 0.000
#> GSM39156 1 0.0469 0.7944 0.988 0.012 NA 0.000
#> GSM39157 2 0.5697 0.6618 0.292 0.656 NA 0.000
#> GSM39158 1 0.0817 0.7957 0.976 0.024 NA 0.000
#> GSM39159 1 0.5687 0.6046 0.684 0.248 NA 0.000
#> GSM39160 1 0.3948 0.7623 0.840 0.096 NA 0.000
#> GSM39161 1 0.1706 0.7935 0.948 0.036 NA 0.000
#> GSM39162 1 0.0000 0.7933 1.000 0.000 NA 0.000
#> GSM39163 1 0.1557 0.7897 0.944 0.056 NA 0.000
#> GSM39164 1 0.0000 0.7933 1.000 0.000 NA 0.000
#> GSM39165 1 0.4290 0.6550 0.772 0.212 NA 0.000
#> GSM39166 1 0.2319 0.7887 0.924 0.040 NA 0.000
#> GSM39167 1 0.0000 0.7933 1.000 0.000 NA 0.000
#> GSM39168 1 0.0000 0.7933 1.000 0.000 NA 0.000
#> GSM39169 1 0.0000 0.7933 1.000 0.000 NA 0.000
#> GSM39170 1 0.0927 0.7938 0.976 0.008 NA 0.000
#> GSM39171 2 0.4250 0.7023 0.276 0.724 NA 0.000
#> GSM39172 1 0.5660 0.2067 0.576 0.028 NA 0.396
#> GSM39173 1 0.3847 0.6989 0.844 0.012 NA 0.124
#> GSM39174 1 0.1211 0.7910 0.960 0.040 NA 0.000
#> GSM39175 1 0.0469 0.7958 0.988 0.012 NA 0.000
#> GSM39176 1 0.0000 0.7933 1.000 0.000 NA 0.000
#> GSM39177 1 0.6163 0.5641 0.676 0.160 NA 0.164
#> GSM39178 1 0.2363 0.7891 0.920 0.056 NA 0.000
#> GSM39179 1 0.8264 -0.0795 0.440 0.144 NA 0.372
#> GSM39180 4 0.7338 0.1459 0.404 0.156 NA 0.440
#> GSM39181 1 0.6153 0.3566 0.604 0.328 NA 0.000
#> GSM39182 1 0.6069 0.2882 0.588 0.056 NA 0.356
#> GSM39183 1 0.5716 0.5576 0.680 0.252 NA 0.000
#> GSM39184 1 0.5815 0.4758 0.652 0.288 NA 0.000
#> GSM39185 1 0.5772 0.5673 0.672 0.260 NA 0.000
#> GSM39186 2 0.5888 0.4001 0.424 0.540 NA 0.000
#> GSM39187 1 0.1902 0.7906 0.932 0.064 NA 0.000
#> GSM39116 2 0.5205 0.4233 0.008 0.672 NA 0.308
#> GSM39117 4 0.0000 0.7133 0.000 0.000 NA 1.000
#> GSM39118 4 0.5138 0.3614 0.008 0.392 NA 0.600
#> GSM39119 4 0.3852 0.6827 0.008 0.192 NA 0.800
#> GSM39120 2 0.2469 0.7320 0.108 0.892 NA 0.000
#> GSM39121 2 0.2345 0.7431 0.100 0.900 NA 0.000
#> GSM39122 2 0.1302 0.7455 0.044 0.956 NA 0.000
#> GSM39123 4 0.0000 0.7133 0.000 0.000 NA 1.000
#> GSM39124 2 0.2741 0.7587 0.096 0.892 NA 0.012
#> GSM39125 2 0.2345 0.7583 0.100 0.900 NA 0.000
#> GSM39126 2 0.2408 0.7445 0.104 0.896 NA 0.000
#> GSM39127 2 0.2587 0.7047 0.008 0.916 NA 0.056
#> GSM39128 2 0.5616 0.5637 0.180 0.740 NA 0.060
#> GSM39129 4 0.4491 0.7111 0.000 0.140 NA 0.800
#> GSM39130 4 0.0000 0.7133 0.000 0.000 NA 1.000
#> GSM39131 2 0.1042 0.7275 0.008 0.972 NA 0.000
#> GSM39132 2 0.3933 0.7079 0.064 0.860 NA 0.056
#> GSM39133 4 0.0000 0.7133 0.000 0.000 NA 1.000
#> GSM39134 4 0.4194 0.6988 0.028 0.172 NA 0.800
#> GSM39135 2 0.5485 0.4020 0.008 0.652 NA 0.320
#> GSM39136 4 0.5747 0.3747 0.008 0.384 NA 0.588
#> GSM39137 2 0.3219 0.7525 0.164 0.836 NA 0.000
#> GSM39138 4 0.4244 0.6624 0.168 0.032 NA 0.800
#> GSM39139 2 0.5525 0.2884 0.008 0.600 NA 0.380
#> GSM39140 2 0.4761 0.5654 0.372 0.628 NA 0.000
#> GSM39141 2 0.4040 0.7203 0.248 0.752 NA 0.000
#> GSM39142 2 0.3942 0.7242 0.236 0.764 NA 0.000
#> GSM39143 2 0.3907 0.7265 0.232 0.768 NA 0.000
#> GSM39144 4 0.6577 0.6692 0.004 0.176 NA 0.648
#> GSM39145 2 0.6232 0.3806 0.040 0.608 NA 0.336
#> GSM39146 2 0.4612 0.6018 0.008 0.764 NA 0.212
#> GSM39147 2 0.2363 0.7343 0.024 0.920 NA 0.056
#> GSM39188 4 0.8009 0.5480 0.204 0.028 NA 0.520
#> GSM39189 1 0.5463 0.5028 0.692 0.052 NA 0.256
#> GSM39190 2 0.8889 0.1549 0.076 0.456 NA 0.216
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM39104 2 0.5285 0.6057 0.288 0.632 0.080 0.000 NA
#> GSM39105 2 0.4449 0.7245 0.168 0.752 0.080 0.000 NA
#> GSM39106 1 0.5245 0.5868 0.640 0.280 0.080 0.000 NA
#> GSM39107 2 0.0510 0.7359 0.016 0.984 0.000 0.000 NA
#> GSM39108 2 0.4593 0.7264 0.184 0.736 0.080 0.000 NA
#> GSM39109 2 0.1117 0.7334 0.016 0.964 0.000 0.020 NA
#> GSM39110 1 0.4074 0.2869 0.636 0.364 0.000 0.000 NA
#> GSM39111 2 0.5203 0.6386 0.272 0.648 0.080 0.000 NA
#> GSM39112 2 0.0510 0.7359 0.016 0.984 0.000 0.000 NA
#> GSM39113 2 0.0609 0.7367 0.020 0.980 0.000 0.000 NA
#> GSM39114 2 0.0290 0.7308 0.008 0.992 0.000 0.000 NA
#> GSM39115 2 0.5285 0.6175 0.288 0.632 0.080 0.000 NA
#> GSM39148 1 0.0000 0.7889 1.000 0.000 0.000 0.000 NA
#> GSM39149 1 0.7132 0.2638 0.548 0.056 0.104 0.276 NA
#> GSM39150 1 0.1800 0.7862 0.932 0.048 0.020 0.000 NA
#> GSM39151 4 0.8476 0.0725 0.148 0.300 0.080 0.420 NA
#> GSM39152 1 0.0880 0.7926 0.968 0.032 0.000 0.000 NA
#> GSM39153 1 0.0000 0.7889 1.000 0.000 0.000 0.000 NA
#> GSM39154 1 0.0609 0.7914 0.980 0.020 0.000 0.000 NA
#> GSM39155 1 0.5456 -0.2600 0.484 0.456 0.060 0.000 NA
#> GSM39156 1 0.0404 0.7900 0.988 0.012 0.000 0.000 NA
#> GSM39157 2 0.5029 0.6585 0.292 0.648 0.060 0.000 NA
#> GSM39158 1 0.0794 0.7922 0.972 0.028 0.000 0.000 NA
#> GSM39159 1 0.5064 0.6017 0.672 0.248 0.080 0.000 NA
#> GSM39160 1 0.3586 0.7558 0.828 0.096 0.076 0.000 NA
#> GSM39161 1 0.1725 0.7882 0.936 0.044 0.020 0.000 NA
#> GSM39162 1 0.0000 0.7889 1.000 0.000 0.000 0.000 NA
#> GSM39163 1 0.1341 0.7868 0.944 0.056 0.000 0.000 NA
#> GSM39164 1 0.0000 0.7889 1.000 0.000 0.000 0.000 NA
#> GSM39165 1 0.3663 0.6581 0.776 0.208 0.016 0.000 NA
#> GSM39166 1 0.2228 0.7828 0.912 0.048 0.040 0.000 NA
#> GSM39167 1 0.0000 0.7889 1.000 0.000 0.000 0.000 NA
#> GSM39168 1 0.0000 0.7889 1.000 0.000 0.000 0.000 NA
#> GSM39169 1 0.0000 0.7889 1.000 0.000 0.000 0.000 NA
#> GSM39170 1 0.1012 0.7878 0.968 0.012 0.020 0.000 NA
#> GSM39171 2 0.3707 0.6954 0.284 0.716 0.000 0.000 NA
#> GSM39172 1 0.4876 0.2172 0.576 0.028 0.000 0.396 NA
#> GSM39173 1 0.3929 0.6796 0.816 0.000 0.120 0.048 NA
#> GSM39174 1 0.1043 0.7879 0.960 0.040 0.000 0.000 NA
#> GSM39175 1 0.0404 0.7917 0.988 0.012 0.000 0.000 NA
#> GSM39176 1 0.0000 0.7889 1.000 0.000 0.000 0.000 NA
#> GSM39177 1 0.5273 0.5759 0.680 0.156 0.000 0.164 NA
#> GSM39178 1 0.2260 0.7833 0.908 0.064 0.028 0.000 NA
#> GSM39179 1 0.8289 -0.0471 0.428 0.132 0.044 0.316 NA
#> GSM39180 4 0.6320 -0.0236 0.404 0.156 0.000 0.440 NA
#> GSM39181 1 0.5470 0.3431 0.588 0.332 0.080 0.000 NA
#> GSM39182 1 0.5215 0.3179 0.592 0.056 0.000 0.352 NA
#> GSM39183 1 0.5088 0.5533 0.668 0.252 0.080 0.000 NA
#> GSM39184 1 0.5124 0.4676 0.644 0.288 0.068 0.000 NA
#> GSM39185 1 0.5158 0.5581 0.656 0.264 0.080 0.000 NA
#> GSM39186 2 0.5243 0.4046 0.412 0.540 0.048 0.000 NA
#> GSM39187 1 0.1638 0.7880 0.932 0.064 0.004 0.000 NA
#> GSM39116 2 0.4540 0.4428 0.008 0.676 0.016 0.300 NA
#> GSM39117 4 0.0000 0.0232 0.000 0.000 0.000 1.000 NA
#> GSM39118 4 0.4425 0.2061 0.008 0.392 0.000 0.600 NA
#> GSM39119 4 0.3318 0.2250 0.008 0.192 0.000 0.800 NA
#> GSM39120 2 0.2179 0.7326 0.112 0.888 0.000 0.000 NA
#> GSM39121 2 0.2127 0.7422 0.108 0.892 0.000 0.000 NA
#> GSM39122 2 0.1197 0.7458 0.048 0.952 0.000 0.000 NA
#> GSM39123 4 0.0000 0.0232 0.000 0.000 0.000 1.000 NA
#> GSM39124 2 0.2470 0.7590 0.104 0.884 0.000 0.012 NA
#> GSM39125 2 0.2074 0.7589 0.104 0.896 0.000 0.000 NA
#> GSM39126 2 0.2179 0.7459 0.112 0.888 0.000 0.000 NA
#> GSM39127 2 0.2409 0.7016 0.008 0.908 0.028 0.056 NA
#> GSM39128 2 0.4972 0.5688 0.176 0.736 0.028 0.060 NA
#> GSM39129 4 0.4383 -0.0769 0.000 0.048 0.048 0.800 NA
#> GSM39130 4 0.0000 0.0232 0.000 0.000 0.000 1.000 NA
#> GSM39131 2 0.0992 0.7266 0.008 0.968 0.024 0.000 NA
#> GSM39132 2 0.3497 0.7089 0.064 0.856 0.028 0.052 NA
#> GSM39133 4 0.0000 0.0232 0.000 0.000 0.000 1.000 NA
#> GSM39134 4 0.3612 0.2163 0.028 0.172 0.000 0.800 NA
#> GSM39135 2 0.4853 0.4198 0.008 0.652 0.028 0.312 NA
#> GSM39136 4 0.5109 0.1983 0.008 0.384 0.028 0.580 NA
#> GSM39137 2 0.2852 0.7523 0.172 0.828 0.000 0.000 NA
#> GSM39138 4 0.3656 -0.0443 0.168 0.032 0.000 0.800 NA
#> GSM39139 2 0.5395 0.3318 0.008 0.600 0.024 0.352 NA
#> GSM39140 2 0.4126 0.5534 0.380 0.620 0.000 0.000 NA
#> GSM39141 2 0.3534 0.7187 0.256 0.744 0.000 0.000 NA
#> GSM39142 2 0.3452 0.7229 0.244 0.756 0.000 0.000 NA
#> GSM39143 2 0.3424 0.7252 0.240 0.760 0.000 0.000 NA
#> GSM39144 4 0.5656 -0.2502 0.000 0.104 0.000 0.588 NA
#> GSM39145 2 0.5508 0.4007 0.040 0.608 0.024 0.328 NA
#> GSM39146 2 0.4097 0.5987 0.008 0.756 0.020 0.216 NA
#> GSM39147 2 0.2196 0.7336 0.024 0.916 0.004 0.056 NA
#> GSM39188 3 0.5503 0.0000 0.024 0.024 0.484 0.468 NA
#> GSM39189 1 0.4922 0.5036 0.684 0.056 0.004 0.256 NA
#> GSM39190 2 0.7234 0.1267 0.068 0.408 0.000 0.116 NA
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM39104 2 0.4812 0.3620 0.264 0.640 0.000 0.000 NA 0.000
#> GSM39105 2 0.4014 0.5448 0.148 0.756 0.000 0.000 NA 0.000
#> GSM39106 1 0.4869 0.5154 0.628 0.276 0.000 0.000 NA 0.000
#> GSM39107 2 0.0146 0.5725 0.004 0.996 0.000 0.000 NA 0.000
#> GSM39108 2 0.4220 0.5385 0.172 0.732 0.000 0.000 NA 0.000
#> GSM39109 2 0.0717 0.5718 0.008 0.976 0.000 0.016 NA 0.000
#> GSM39110 1 0.3659 0.2767 0.636 0.364 0.000 0.000 NA 0.000
#> GSM39111 2 0.4729 0.4247 0.248 0.656 0.000 0.000 NA 0.000
#> GSM39112 2 0.0146 0.5725 0.004 0.996 0.000 0.000 NA 0.000
#> GSM39113 2 0.0363 0.5754 0.012 0.988 0.000 0.000 NA 0.000
#> GSM39114 2 0.0551 0.5717 0.004 0.984 0.008 0.000 NA 0.000
#> GSM39115 2 0.4812 0.4125 0.264 0.640 0.000 0.000 NA 0.000
#> GSM39148 1 0.0000 0.7726 1.000 0.000 0.000 0.000 NA 0.000
#> GSM39149 1 0.7081 0.2971 0.536 0.060 0.076 0.264 NA 0.024
#> GSM39150 1 0.1682 0.7662 0.928 0.052 0.000 0.000 NA 0.000
#> GSM39151 4 0.8612 -0.1008 0.060 0.172 0.052 0.392 NA 0.240
#> GSM39152 1 0.0790 0.7746 0.968 0.032 0.000 0.000 NA 0.000
#> GSM39153 1 0.0000 0.7726 1.000 0.000 0.000 0.000 NA 0.000
#> GSM39154 1 0.0547 0.7730 0.980 0.020 0.000 0.000 NA 0.000
#> GSM39155 1 0.5075 -0.2844 0.464 0.460 0.000 0.000 NA 0.000
#> GSM39156 1 0.0458 0.7745 0.984 0.016 0.000 0.000 NA 0.000
#> GSM39157 2 0.4616 0.4422 0.280 0.648 0.000 0.000 NA 0.000
#> GSM39158 1 0.0713 0.7745 0.972 0.028 0.000 0.000 NA 0.000
#> GSM39159 1 0.4750 0.5460 0.652 0.252 0.000 0.000 NA 0.000
#> GSM39160 1 0.3423 0.7188 0.812 0.100 0.000 0.000 NA 0.000
#> GSM39161 1 0.1700 0.7682 0.928 0.048 0.000 0.000 NA 0.000
#> GSM39162 1 0.0000 0.7726 1.000 0.000 0.000 0.000 NA 0.000
#> GSM39163 1 0.1267 0.7632 0.940 0.060 0.000 0.000 NA 0.000
#> GSM39164 1 0.0000 0.7726 1.000 0.000 0.000 0.000 NA 0.000
#> GSM39165 1 0.3320 0.6170 0.772 0.212 0.000 0.000 NA 0.000
#> GSM39166 1 0.2197 0.7599 0.900 0.056 0.000 0.000 NA 0.000
#> GSM39167 1 0.0000 0.7726 1.000 0.000 0.000 0.000 NA 0.000
#> GSM39168 1 0.0000 0.7726 1.000 0.000 0.000 0.000 NA 0.000
#> GSM39169 1 0.0000 0.7726 1.000 0.000 0.000 0.000 NA 0.000
#> GSM39170 1 0.1003 0.7704 0.964 0.016 0.000 0.000 NA 0.000
#> GSM39171 2 0.3309 0.5077 0.280 0.720 0.000 0.000 NA 0.000
#> GSM39172 1 0.4379 0.2425 0.576 0.028 0.000 0.396 NA 0.000
#> GSM39173 1 0.3367 0.6653 0.804 0.000 0.164 0.012 NA 0.000
#> GSM39174 1 0.0937 0.7665 0.960 0.040 0.000 0.000 NA 0.000
#> GSM39175 1 0.0363 0.7748 0.988 0.012 0.000 0.000 NA 0.000
#> GSM39176 1 0.0000 0.7726 1.000 0.000 0.000 0.000 NA 0.000
#> GSM39177 1 0.5142 0.5262 0.672 0.160 0.000 0.148 NA 0.000
#> GSM39178 1 0.2088 0.7620 0.904 0.068 0.000 0.000 NA 0.000
#> GSM39179 1 0.7969 -0.0625 0.376 0.120 0.024 0.172 NA 0.012
#> GSM39180 4 0.5825 0.0589 0.400 0.160 0.000 0.436 NA 0.000
#> GSM39181 1 0.4982 0.3119 0.576 0.340 0.000 0.000 NA 0.000
#> GSM39182 1 0.4684 0.3344 0.592 0.056 0.000 0.352 NA 0.000
#> GSM39183 1 0.4792 0.4979 0.644 0.260 0.000 0.000 NA 0.000
#> GSM39184 1 0.4793 0.4220 0.628 0.288 0.000 0.000 NA 0.000
#> GSM39185 1 0.4831 0.5021 0.636 0.268 0.000 0.000 NA 0.000
#> GSM39186 2 0.4750 0.2386 0.404 0.544 0.000 0.000 NA 0.000
#> GSM39187 1 0.1531 0.7646 0.928 0.068 0.000 0.000 NA 0.000
#> GSM39116 2 0.4165 0.1993 0.004 0.676 0.028 0.292 NA 0.000
#> GSM39117 4 0.0000 0.2155 0.000 0.000 0.000 1.000 NA 0.000
#> GSM39118 4 0.3881 0.0536 0.004 0.396 0.000 0.600 NA 0.000
#> GSM39119 4 0.2902 0.3243 0.004 0.196 0.000 0.800 NA 0.000
#> GSM39120 2 0.1910 0.5404 0.108 0.892 0.000 0.000 NA 0.000
#> GSM39121 2 0.1814 0.5667 0.100 0.900 0.000 0.000 NA 0.000
#> GSM39122 2 0.1265 0.5908 0.044 0.948 0.008 0.000 NA 0.000
#> GSM39123 4 0.0000 0.2155 0.000 0.000 0.000 1.000 NA 0.000
#> GSM39124 2 0.2313 0.6112 0.100 0.884 0.004 0.012 NA 0.000
#> GSM39125 2 0.1814 0.6004 0.100 0.900 0.000 0.000 NA 0.000
#> GSM39126 2 0.1910 0.5744 0.108 0.892 0.000 0.000 NA 0.000
#> GSM39127 2 0.2421 0.5141 0.004 0.896 0.044 0.052 NA 0.004
#> GSM39128 2 0.4786 0.2157 0.172 0.724 0.044 0.056 NA 0.004
#> GSM39129 4 0.2793 0.0719 0.000 0.000 0.000 0.800 NA 0.200
#> GSM39130 4 0.0000 0.2155 0.000 0.000 0.000 1.000 NA 0.000
#> GSM39131 2 0.1155 0.5579 0.004 0.956 0.036 0.000 NA 0.004
#> GSM39132 2 0.3508 0.4989 0.060 0.840 0.044 0.052 NA 0.004
#> GSM39133 4 0.0000 0.2155 0.000 0.000 0.000 1.000 NA 0.000
#> GSM39134 4 0.3202 0.3299 0.024 0.176 0.000 0.800 NA 0.000
#> GSM39135 2 0.4508 0.1803 0.004 0.648 0.036 0.308 NA 0.004
#> GSM39136 4 0.4859 0.0144 0.004 0.380 0.044 0.568 NA 0.004
#> GSM39137 2 0.2491 0.5957 0.164 0.836 0.000 0.000 NA 0.000
#> GSM39138 4 0.3351 0.1821 0.160 0.040 0.000 0.800 NA 0.000
#> GSM39139 2 0.5190 0.0960 0.004 0.600 0.064 0.320 NA 0.004
#> GSM39140 2 0.3695 0.3584 0.376 0.624 0.000 0.000 NA 0.000
#> GSM39141 2 0.3126 0.5445 0.248 0.752 0.000 0.000 NA 0.000
#> GSM39142 2 0.3050 0.5543 0.236 0.764 0.000 0.000 NA 0.000
#> GSM39143 2 0.3023 0.5575 0.232 0.768 0.000 0.000 NA 0.000
#> GSM39144 4 0.3817 -0.3453 0.000 0.000 0.000 0.568 NA 0.000
#> GSM39145 2 0.5179 0.1662 0.036 0.608 0.036 0.316 NA 0.004
#> GSM39146 2 0.3733 0.3965 0.004 0.760 0.024 0.208 NA 0.004
#> GSM39147 2 0.2220 0.5765 0.020 0.908 0.020 0.052 NA 0.000
#> GSM39188 3 0.3747 0.0000 0.000 0.000 0.604 0.396 NA 0.000
#> GSM39189 1 0.4476 0.5171 0.680 0.060 0.000 0.256 NA 0.000
#> GSM39190 6 0.5717 0.0000 0.056 0.376 0.000 0.052 NA 0.516
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) other(p) protocol(p) k
#> CV:pam 73 0.20957 2.00e-05 1.10e-04 2
#> CV:pam 68 0.00188 4.37e-12 8.74e-09 3
#> CV:pam 69 0.00158 1.32e-11 2.29e-08 4
#> CV:pam 59 0.00620 1.38e-09 2.67e-07 5
#> CV:pam 50 0.01412 1.78e-08 5.06e-07 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 8353 rows and 87 columns.
#> Top rows (835, 1670, 2506, 3341, 4176) 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 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 0.805 0.884 0.947 0.4928 0.500 0.500
#> 3 3 0.511 0.794 0.874 0.2619 0.558 0.335
#> 4 4 0.493 0.641 0.689 0.1173 0.865 0.655
#> 5 5 0.611 0.649 0.788 0.0750 0.968 0.886
#> 6 6 0.665 0.647 0.734 0.0255 0.905 0.671
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
#> GSM39104 1 0.3733 0.9042 0.928 0.072
#> GSM39105 1 0.1414 0.9243 0.980 0.020
#> GSM39106 1 0.7950 0.7346 0.760 0.240
#> GSM39107 2 0.3114 0.9147 0.056 0.944
#> GSM39108 1 0.7453 0.7728 0.788 0.212
#> GSM39109 2 0.0672 0.9540 0.008 0.992
#> GSM39110 2 0.5519 0.8338 0.128 0.872
#> GSM39111 2 0.9993 -0.0516 0.484 0.516
#> GSM39112 2 0.9393 0.3958 0.356 0.644
#> GSM39113 2 0.3274 0.9107 0.060 0.940
#> GSM39114 2 0.0000 0.9548 0.000 1.000
#> GSM39115 1 0.0376 0.9275 0.996 0.004
#> GSM39148 1 0.0000 0.9278 1.000 0.000
#> GSM39149 2 0.0938 0.9531 0.012 0.988
#> GSM39150 1 0.0672 0.9270 0.992 0.008
#> GSM39151 2 0.0938 0.9531 0.012 0.988
#> GSM39152 2 0.0938 0.9531 0.012 0.988
#> GSM39153 1 0.0000 0.9278 1.000 0.000
#> GSM39154 1 0.0000 0.9278 1.000 0.000
#> GSM39155 1 0.0000 0.9278 1.000 0.000
#> GSM39156 1 0.3879 0.9015 0.924 0.076
#> GSM39157 1 0.0000 0.9278 1.000 0.000
#> GSM39158 1 0.0000 0.9278 1.000 0.000
#> GSM39159 1 0.9635 0.3948 0.612 0.388
#> GSM39160 1 0.2423 0.9185 0.960 0.040
#> GSM39161 2 0.5842 0.8262 0.140 0.860
#> GSM39162 1 0.0000 0.9278 1.000 0.000
#> GSM39163 1 0.0000 0.9278 1.000 0.000
#> GSM39164 1 0.0000 0.9278 1.000 0.000
#> GSM39165 1 0.9460 0.5047 0.636 0.364
#> GSM39166 1 0.0000 0.9278 1.000 0.000
#> GSM39167 1 0.0000 0.9278 1.000 0.000
#> GSM39168 1 0.0000 0.9278 1.000 0.000
#> GSM39169 1 0.0000 0.9278 1.000 0.000
#> GSM39170 1 0.0000 0.9278 1.000 0.000
#> GSM39171 1 0.3584 0.9063 0.932 0.068
#> GSM39172 2 0.0938 0.9531 0.012 0.988
#> GSM39173 2 0.0938 0.9531 0.012 0.988
#> GSM39174 1 0.0000 0.9278 1.000 0.000
#> GSM39175 1 0.0000 0.9278 1.000 0.000
#> GSM39176 1 0.0000 0.9278 1.000 0.000
#> GSM39177 2 0.0938 0.9531 0.012 0.988
#> GSM39178 2 0.9988 0.0612 0.480 0.520
#> GSM39179 2 0.0938 0.9531 0.012 0.988
#> GSM39180 2 0.0938 0.9531 0.012 0.988
#> GSM39181 1 0.5737 0.8490 0.864 0.136
#> GSM39182 2 0.0938 0.9531 0.012 0.988
#> GSM39183 1 0.1414 0.9215 0.980 0.020
#> GSM39184 1 0.0376 0.9275 0.996 0.004
#> GSM39185 2 0.4022 0.8944 0.080 0.920
#> GSM39186 1 0.2236 0.9195 0.964 0.036
#> GSM39187 1 0.0000 0.9278 1.000 0.000
#> GSM39116 2 0.0000 0.9548 0.000 1.000
#> GSM39117 2 0.0000 0.9548 0.000 1.000
#> GSM39118 2 0.0000 0.9548 0.000 1.000
#> GSM39119 2 0.0000 0.9548 0.000 1.000
#> GSM39120 1 0.9286 0.5515 0.656 0.344
#> GSM39121 2 0.0672 0.9540 0.008 0.992
#> GSM39122 2 0.0672 0.9540 0.008 0.992
#> GSM39123 2 0.0000 0.9548 0.000 1.000
#> GSM39124 2 0.0000 0.9548 0.000 1.000
#> GSM39125 1 0.7056 0.7981 0.808 0.192
#> GSM39126 2 0.0672 0.9540 0.008 0.992
#> GSM39127 2 0.0000 0.9548 0.000 1.000
#> GSM39128 2 0.0000 0.9548 0.000 1.000
#> GSM39129 2 0.0000 0.9548 0.000 1.000
#> GSM39130 2 0.0000 0.9548 0.000 1.000
#> GSM39131 2 0.0000 0.9548 0.000 1.000
#> GSM39132 2 0.0000 0.9548 0.000 1.000
#> GSM39133 2 0.0000 0.9548 0.000 1.000
#> GSM39134 2 0.0000 0.9548 0.000 1.000
#> GSM39135 2 0.0000 0.9548 0.000 1.000
#> GSM39136 2 0.0000 0.9548 0.000 1.000
#> GSM39137 2 0.0672 0.9540 0.008 0.992
#> GSM39138 2 0.0000 0.9548 0.000 1.000
#> GSM39139 2 0.0000 0.9548 0.000 1.000
#> GSM39140 1 0.6712 0.8157 0.824 0.176
#> GSM39141 1 0.4022 0.8988 0.920 0.080
#> GSM39142 1 0.3584 0.9057 0.932 0.068
#> GSM39143 1 0.4298 0.8933 0.912 0.088
#> GSM39144 2 0.0000 0.9548 0.000 1.000
#> GSM39145 2 0.0000 0.9548 0.000 1.000
#> GSM39146 2 0.0000 0.9548 0.000 1.000
#> GSM39147 2 0.0000 0.9548 0.000 1.000
#> GSM39188 2 0.0938 0.9531 0.012 0.988
#> GSM39189 2 0.0938 0.9531 0.012 0.988
#> GSM39190 2 0.0938 0.9531 0.012 0.988
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM39104 1 0.1877 0.869 0.956 0.012 0.032
#> GSM39105 1 0.3826 0.811 0.868 0.008 0.124
#> GSM39106 3 0.5315 0.732 0.216 0.012 0.772
#> GSM39107 3 0.3276 0.792 0.024 0.068 0.908
#> GSM39108 3 0.6284 0.612 0.304 0.016 0.680
#> GSM39109 3 0.5842 0.678 0.036 0.196 0.768
#> GSM39110 3 0.8144 0.307 0.380 0.076 0.544
#> GSM39111 1 0.4087 0.851 0.880 0.068 0.052
#> GSM39112 3 0.3618 0.803 0.104 0.012 0.884
#> GSM39113 3 0.1399 0.793 0.004 0.028 0.968
#> GSM39114 3 0.2261 0.780 0.000 0.068 0.932
#> GSM39115 1 0.0000 0.872 1.000 0.000 0.000
#> GSM39148 1 0.4796 0.655 0.780 0.000 0.220
#> GSM39149 1 0.6208 0.786 0.752 0.200 0.048
#> GSM39150 1 0.0424 0.873 0.992 0.008 0.000
#> GSM39151 1 0.6208 0.786 0.752 0.200 0.048
#> GSM39152 1 0.5791 0.807 0.784 0.168 0.048
#> GSM39153 1 0.1753 0.855 0.952 0.000 0.048
#> GSM39154 1 0.0000 0.872 1.000 0.000 0.000
#> GSM39155 1 0.3816 0.763 0.852 0.000 0.148
#> GSM39156 3 0.4702 0.747 0.212 0.000 0.788
#> GSM39157 1 0.1411 0.862 0.964 0.000 0.036
#> GSM39158 1 0.0000 0.872 1.000 0.000 0.000
#> GSM39159 1 0.1289 0.871 0.968 0.000 0.032
#> GSM39160 1 0.1170 0.872 0.976 0.008 0.016
#> GSM39161 1 0.4677 0.829 0.840 0.132 0.028
#> GSM39162 1 0.5327 0.552 0.728 0.000 0.272
#> GSM39163 1 0.0000 0.872 1.000 0.000 0.000
#> GSM39164 1 0.1529 0.861 0.960 0.000 0.040
#> GSM39165 1 0.2806 0.864 0.928 0.032 0.040
#> GSM39166 1 0.0000 0.872 1.000 0.000 0.000
#> GSM39167 1 0.0424 0.871 0.992 0.000 0.008
#> GSM39168 1 0.4504 0.695 0.804 0.000 0.196
#> GSM39169 1 0.1031 0.867 0.976 0.000 0.024
#> GSM39170 1 0.0000 0.872 1.000 0.000 0.000
#> GSM39171 1 0.0848 0.873 0.984 0.008 0.008
#> GSM39172 1 0.6302 0.779 0.744 0.208 0.048
#> GSM39173 1 0.6208 0.786 0.752 0.200 0.048
#> GSM39174 1 0.3879 0.757 0.848 0.000 0.152
#> GSM39175 1 0.0000 0.872 1.000 0.000 0.000
#> GSM39176 1 0.0592 0.870 0.988 0.000 0.012
#> GSM39177 1 0.6208 0.786 0.752 0.200 0.048
#> GSM39178 1 0.1411 0.872 0.964 0.036 0.000
#> GSM39179 1 0.6208 0.786 0.752 0.200 0.048
#> GSM39180 1 0.6255 0.783 0.748 0.204 0.048
#> GSM39181 1 0.0237 0.873 0.996 0.000 0.004
#> GSM39182 1 0.7610 0.706 0.676 0.216 0.108
#> GSM39183 1 0.0000 0.872 1.000 0.000 0.000
#> GSM39184 1 0.0000 0.872 1.000 0.000 0.000
#> GSM39185 1 0.5094 0.824 0.824 0.136 0.040
#> GSM39186 1 0.3116 0.810 0.892 0.000 0.108
#> GSM39187 1 0.3267 0.801 0.884 0.000 0.116
#> GSM39116 2 0.4178 0.849 0.000 0.828 0.172
#> GSM39117 2 0.0747 0.887 0.000 0.984 0.016
#> GSM39118 2 0.2066 0.910 0.000 0.940 0.060
#> GSM39119 2 0.1860 0.906 0.000 0.948 0.052
#> GSM39120 3 0.3551 0.799 0.132 0.000 0.868
#> GSM39121 3 0.1753 0.788 0.000 0.048 0.952
#> GSM39122 3 0.1753 0.788 0.000 0.048 0.952
#> GSM39123 2 0.0747 0.887 0.000 0.984 0.016
#> GSM39124 3 0.2066 0.785 0.000 0.060 0.940
#> GSM39125 3 0.3686 0.797 0.140 0.000 0.860
#> GSM39126 3 0.1753 0.788 0.000 0.048 0.952
#> GSM39127 3 0.4452 0.660 0.000 0.192 0.808
#> GSM39128 3 0.2711 0.769 0.000 0.088 0.912
#> GSM39129 2 0.2066 0.910 0.000 0.940 0.060
#> GSM39130 2 0.0747 0.887 0.000 0.984 0.016
#> GSM39131 3 0.2066 0.785 0.000 0.060 0.940
#> GSM39132 2 0.6215 0.423 0.000 0.572 0.428
#> GSM39133 2 0.1529 0.898 0.000 0.960 0.040
#> GSM39134 2 0.2066 0.909 0.000 0.940 0.060
#> GSM39135 2 0.4235 0.846 0.000 0.824 0.176
#> GSM39136 2 0.4235 0.846 0.000 0.824 0.176
#> GSM39137 3 0.1753 0.788 0.000 0.048 0.952
#> GSM39138 2 0.2066 0.910 0.000 0.940 0.060
#> GSM39139 2 0.2711 0.904 0.000 0.912 0.088
#> GSM39140 3 0.3619 0.798 0.136 0.000 0.864
#> GSM39141 3 0.3941 0.791 0.156 0.000 0.844
#> GSM39142 3 0.4062 0.787 0.164 0.000 0.836
#> GSM39143 3 0.3941 0.791 0.156 0.000 0.844
#> GSM39144 2 0.2066 0.910 0.000 0.940 0.060
#> GSM39145 2 0.2711 0.904 0.000 0.912 0.088
#> GSM39146 2 0.5591 0.670 0.000 0.696 0.304
#> GSM39147 3 0.6308 -0.267 0.000 0.492 0.508
#> GSM39188 1 0.6208 0.786 0.752 0.200 0.048
#> GSM39189 1 0.6208 0.786 0.752 0.200 0.048
#> GSM39190 1 0.6208 0.786 0.752 0.200 0.048
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM39104 1 0.580 0.73005 0.696 0.208 0.096 0.000
#> GSM39105 1 0.531 0.72613 0.700 0.256 0.044 0.000
#> GSM39106 2 0.658 0.57044 0.220 0.640 0.136 0.004
#> GSM39107 2 0.357 0.76016 0.084 0.872 0.024 0.020
#> GSM39108 2 0.759 0.22571 0.320 0.500 0.172 0.008
#> GSM39109 2 0.780 0.41805 0.092 0.516 0.340 0.052
#> GSM39110 2 0.805 0.00251 0.316 0.420 0.256 0.008
#> GSM39111 1 0.750 0.20717 0.472 0.144 0.376 0.008
#> GSM39112 2 0.240 0.76154 0.092 0.904 0.004 0.000
#> GSM39113 2 0.241 0.75680 0.084 0.908 0.000 0.008
#> GSM39114 2 0.401 0.45146 0.000 0.784 0.008 0.208
#> GSM39115 1 0.508 0.74820 0.760 0.160 0.080 0.000
#> GSM39148 1 0.416 0.73239 0.736 0.264 0.000 0.000
#> GSM39149 3 0.387 0.83828 0.208 0.000 0.788 0.004
#> GSM39150 1 0.453 0.42193 0.704 0.004 0.292 0.000
#> GSM39151 3 0.376 0.83879 0.216 0.000 0.784 0.000
#> GSM39152 3 0.481 0.78949 0.268 0.004 0.716 0.012
#> GSM39153 1 0.416 0.77240 0.768 0.224 0.008 0.000
#> GSM39154 1 0.491 0.75981 0.764 0.176 0.060 0.000
#> GSM39155 1 0.419 0.75319 0.752 0.244 0.004 0.000
#> GSM39156 2 0.555 0.67847 0.188 0.728 0.080 0.004
#> GSM39157 1 0.391 0.76540 0.768 0.232 0.000 0.000
#> GSM39158 1 0.514 0.54545 0.716 0.040 0.244 0.000
#> GSM39159 3 0.516 0.29436 0.472 0.000 0.524 0.004
#> GSM39160 1 0.460 0.34304 0.664 0.000 0.336 0.000
#> GSM39161 3 0.523 0.48895 0.384 0.000 0.604 0.012
#> GSM39162 1 0.472 0.63948 0.672 0.324 0.004 0.000
#> GSM39163 1 0.433 0.77221 0.768 0.216 0.016 0.000
#> GSM39164 1 0.430 0.76359 0.752 0.240 0.008 0.000
#> GSM39165 1 0.517 -0.28773 0.500 0.000 0.496 0.004
#> GSM39166 1 0.461 0.40413 0.692 0.004 0.304 0.000
#> GSM39167 1 0.432 0.77123 0.776 0.204 0.020 0.000
#> GSM39168 1 0.425 0.75252 0.744 0.252 0.004 0.000
#> GSM39169 1 0.412 0.77197 0.772 0.220 0.008 0.000
#> GSM39170 1 0.527 0.73082 0.752 0.140 0.108 0.000
#> GSM39171 1 0.465 0.39500 0.684 0.004 0.312 0.000
#> GSM39172 3 0.431 0.74632 0.108 0.004 0.824 0.064
#> GSM39173 3 0.361 0.83814 0.200 0.000 0.800 0.000
#> GSM39174 1 0.426 0.76733 0.756 0.236 0.008 0.000
#> GSM39175 1 0.509 0.73273 0.764 0.140 0.096 0.000
#> GSM39176 1 0.402 0.77068 0.772 0.224 0.004 0.000
#> GSM39177 3 0.391 0.82916 0.232 0.000 0.768 0.000
#> GSM39178 1 0.488 0.09486 0.592 0.000 0.408 0.000
#> GSM39179 3 0.376 0.83879 0.216 0.000 0.784 0.000
#> GSM39180 3 0.279 0.75613 0.088 0.004 0.896 0.012
#> GSM39181 1 0.478 0.28354 0.624 0.000 0.376 0.000
#> GSM39182 3 0.528 0.72685 0.132 0.020 0.776 0.072
#> GSM39183 1 0.472 0.38289 0.672 0.004 0.324 0.000
#> GSM39184 1 0.512 0.75015 0.756 0.164 0.080 0.000
#> GSM39185 3 0.498 0.69965 0.260 0.004 0.716 0.020
#> GSM39186 1 0.430 0.76259 0.752 0.240 0.008 0.000
#> GSM39187 1 0.430 0.76076 0.752 0.240 0.008 0.000
#> GSM39116 4 0.474 0.69524 0.000 0.240 0.024 0.736
#> GSM39117 4 0.642 0.66949 0.216 0.056 0.044 0.684
#> GSM39118 4 0.292 0.78102 0.000 0.044 0.060 0.896
#> GSM39119 4 0.153 0.77205 0.012 0.016 0.012 0.960
#> GSM39120 2 0.355 0.75295 0.136 0.844 0.020 0.000
#> GSM39121 2 0.126 0.72369 0.008 0.964 0.000 0.028
#> GSM39122 2 0.128 0.72752 0.012 0.964 0.000 0.024
#> GSM39123 4 0.634 0.67052 0.216 0.056 0.040 0.688
#> GSM39124 2 0.309 0.62084 0.000 0.864 0.008 0.128
#> GSM39125 2 0.386 0.74835 0.144 0.828 0.028 0.000
#> GSM39126 2 0.162 0.73727 0.028 0.952 0.000 0.020
#> GSM39127 4 0.556 0.42170 0.000 0.432 0.020 0.548
#> GSM39128 2 0.345 0.58411 0.000 0.836 0.008 0.156
#> GSM39129 4 0.283 0.78060 0.000 0.040 0.060 0.900
#> GSM39130 4 0.642 0.66949 0.216 0.056 0.044 0.684
#> GSM39131 2 0.265 0.64427 0.000 0.888 0.004 0.108
#> GSM39132 4 0.544 0.51581 0.000 0.384 0.020 0.596
#> GSM39133 4 0.638 0.67830 0.212 0.076 0.028 0.684
#> GSM39134 4 0.100 0.77923 0.000 0.024 0.004 0.972
#> GSM39135 4 0.478 0.69219 0.000 0.244 0.024 0.732
#> GSM39136 4 0.502 0.68746 0.000 0.264 0.028 0.708
#> GSM39137 2 0.131 0.71814 0.004 0.960 0.000 0.036
#> GSM39138 4 0.274 0.78082 0.000 0.036 0.060 0.904
#> GSM39139 4 0.301 0.78081 0.000 0.056 0.052 0.892
#> GSM39140 2 0.426 0.74619 0.140 0.812 0.048 0.000
#> GSM39141 2 0.424 0.72809 0.168 0.800 0.032 0.000
#> GSM39142 2 0.450 0.70159 0.192 0.776 0.032 0.000
#> GSM39143 2 0.420 0.73300 0.164 0.804 0.032 0.000
#> GSM39144 4 0.283 0.78060 0.000 0.040 0.060 0.900
#> GSM39145 4 0.432 0.76509 0.000 0.116 0.068 0.816
#> GSM39146 4 0.544 0.56275 0.000 0.384 0.020 0.596
#> GSM39147 2 0.569 -0.26111 0.000 0.516 0.024 0.460
#> GSM39188 3 0.385 0.83427 0.192 0.000 0.800 0.008
#> GSM39189 3 0.412 0.83853 0.220 0.000 0.772 0.008
#> GSM39190 3 0.383 0.83880 0.204 0.000 0.792 0.004
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM39104 1 0.3569 0.7863 0.852 0.000 0.036 0.040 0.072
#> GSM39105 1 0.3601 0.7863 0.832 0.000 0.020 0.024 0.124
#> GSM39106 5 0.5052 0.6453 0.156 0.000 0.084 0.024 0.736
#> GSM39107 5 0.2273 0.7372 0.048 0.008 0.016 0.008 0.920
#> GSM39108 5 0.6760 0.4702 0.184 0.000 0.172 0.056 0.588
#> GSM39109 5 0.8394 -0.2611 0.160 0.008 0.152 0.324 0.356
#> GSM39110 5 0.7819 0.2038 0.172 0.000 0.264 0.112 0.452
#> GSM39111 1 0.7097 0.0741 0.544 0.000 0.252 0.116 0.088
#> GSM39112 5 0.2179 0.7356 0.072 0.000 0.008 0.008 0.912
#> GSM39113 5 0.1560 0.7283 0.020 0.000 0.004 0.028 0.948
#> GSM39114 5 0.4270 0.5818 0.000 0.124 0.008 0.080 0.788
#> GSM39115 1 0.2248 0.8232 0.900 0.000 0.000 0.012 0.088
#> GSM39148 1 0.2563 0.8047 0.872 0.000 0.000 0.008 0.120
#> GSM39149 3 0.1557 0.8598 0.052 0.008 0.940 0.000 0.000
#> GSM39150 1 0.1485 0.7682 0.948 0.000 0.020 0.032 0.000
#> GSM39151 3 0.1341 0.8616 0.056 0.000 0.944 0.000 0.000
#> GSM39152 3 0.4088 0.6393 0.176 0.000 0.780 0.036 0.008
#> GSM39153 1 0.2011 0.8254 0.908 0.000 0.004 0.000 0.088
#> GSM39154 1 0.2074 0.8233 0.920 0.000 0.004 0.016 0.060
#> GSM39155 1 0.1892 0.8255 0.916 0.000 0.000 0.004 0.080
#> GSM39156 5 0.4182 0.6800 0.164 0.000 0.036 0.016 0.784
#> GSM39157 1 0.2570 0.8135 0.880 0.000 0.004 0.008 0.108
#> GSM39158 1 0.0898 0.7851 0.972 0.000 0.008 0.020 0.000
#> GSM39159 1 0.6192 -0.3619 0.520 0.000 0.132 0.344 0.004
#> GSM39160 1 0.2438 0.7390 0.900 0.000 0.040 0.060 0.000
#> GSM39161 1 0.6208 -0.5403 0.468 0.000 0.108 0.416 0.008
#> GSM39162 1 0.3044 0.7799 0.840 0.000 0.004 0.008 0.148
#> GSM39163 1 0.1831 0.8247 0.920 0.000 0.000 0.004 0.076
#> GSM39164 1 0.2642 0.8158 0.880 0.000 0.008 0.008 0.104
#> GSM39165 1 0.5572 0.2659 0.668 0.000 0.204 0.116 0.012
#> GSM39166 1 0.2293 0.7171 0.900 0.000 0.016 0.084 0.000
#> GSM39167 1 0.1831 0.8244 0.920 0.000 0.000 0.004 0.076
#> GSM39168 1 0.2358 0.8168 0.888 0.000 0.000 0.008 0.104
#> GSM39169 1 0.1892 0.8238 0.916 0.000 0.000 0.004 0.080
#> GSM39170 1 0.1121 0.8208 0.956 0.000 0.000 0.000 0.044
#> GSM39171 1 0.2074 0.7512 0.920 0.000 0.036 0.044 0.000
#> GSM39172 4 0.7298 0.5468 0.188 0.008 0.360 0.420 0.024
#> GSM39173 3 0.1770 0.8584 0.048 0.008 0.936 0.008 0.000
#> GSM39174 1 0.2177 0.8260 0.908 0.000 0.004 0.008 0.080
#> GSM39175 1 0.1280 0.8050 0.960 0.000 0.008 0.008 0.024
#> GSM39176 1 0.1831 0.8247 0.920 0.000 0.000 0.004 0.076
#> GSM39177 3 0.2127 0.8203 0.108 0.000 0.892 0.000 0.000
#> GSM39178 1 0.4272 0.4942 0.752 0.000 0.052 0.196 0.000
#> GSM39179 3 0.1341 0.8616 0.056 0.000 0.944 0.000 0.000
#> GSM39180 3 0.6465 -0.3151 0.104 0.000 0.496 0.376 0.024
#> GSM39181 1 0.3106 0.6553 0.844 0.000 0.024 0.132 0.000
#> GSM39182 4 0.7455 0.7523 0.264 0.012 0.204 0.484 0.036
#> GSM39183 1 0.2653 0.6917 0.880 0.000 0.024 0.096 0.000
#> GSM39184 1 0.2562 0.8214 0.900 0.000 0.008 0.032 0.060
#> GSM39185 4 0.6721 0.6939 0.360 0.000 0.172 0.456 0.012
#> GSM39186 1 0.2787 0.8178 0.880 0.000 0.004 0.028 0.088
#> GSM39187 1 0.2612 0.8026 0.868 0.000 0.000 0.008 0.124
#> GSM39116 2 0.5085 0.6986 0.000 0.720 0.020 0.072 0.188
#> GSM39117 2 0.4171 0.6122 0.000 0.604 0.000 0.396 0.000
#> GSM39118 2 0.2444 0.7438 0.000 0.904 0.012 0.068 0.016
#> GSM39119 2 0.2286 0.7439 0.000 0.888 0.000 0.108 0.004
#> GSM39120 5 0.3320 0.7216 0.124 0.000 0.016 0.016 0.844
#> GSM39121 5 0.3005 0.6872 0.004 0.048 0.012 0.052 0.884
#> GSM39122 5 0.2710 0.6959 0.004 0.044 0.012 0.040 0.900
#> GSM39123 2 0.4182 0.6146 0.000 0.600 0.000 0.400 0.000
#> GSM39124 5 0.4693 0.5469 0.000 0.148 0.012 0.084 0.756
#> GSM39125 5 0.3601 0.7217 0.124 0.000 0.020 0.024 0.832
#> GSM39126 5 0.2276 0.7030 0.004 0.040 0.008 0.028 0.920
#> GSM39127 2 0.5913 0.4460 0.000 0.548 0.020 0.064 0.368
#> GSM39128 5 0.5232 0.4985 0.000 0.168 0.020 0.096 0.716
#> GSM39129 2 0.2166 0.7384 0.000 0.912 0.012 0.072 0.004
#> GSM39130 2 0.4171 0.6122 0.000 0.604 0.000 0.396 0.000
#> GSM39131 5 0.4761 0.5586 0.000 0.136 0.020 0.084 0.760
#> GSM39132 2 0.5838 0.5540 0.000 0.596 0.020 0.072 0.312
#> GSM39133 2 0.4697 0.6266 0.000 0.592 0.000 0.388 0.020
#> GSM39134 2 0.2102 0.7507 0.000 0.916 0.004 0.068 0.012
#> GSM39135 2 0.5216 0.7003 0.000 0.716 0.024 0.080 0.180
#> GSM39136 2 0.5127 0.7156 0.000 0.728 0.024 0.084 0.164
#> GSM39137 5 0.2934 0.6902 0.004 0.048 0.012 0.048 0.888
#> GSM39138 2 0.2166 0.7384 0.000 0.912 0.012 0.072 0.004
#> GSM39139 2 0.2868 0.7416 0.000 0.884 0.012 0.072 0.032
#> GSM39140 5 0.3416 0.7206 0.124 0.000 0.020 0.016 0.840
#> GSM39141 5 0.3689 0.7105 0.140 0.000 0.024 0.016 0.820
#> GSM39142 5 0.3817 0.7011 0.152 0.000 0.024 0.016 0.808
#> GSM39143 5 0.3775 0.7044 0.148 0.000 0.024 0.016 0.812
#> GSM39144 2 0.2166 0.7384 0.000 0.912 0.012 0.072 0.004
#> GSM39145 2 0.4091 0.7374 0.000 0.808 0.012 0.084 0.096
#> GSM39146 2 0.6301 0.5715 0.000 0.572 0.020 0.124 0.284
#> GSM39147 5 0.5435 -0.0295 0.000 0.404 0.004 0.052 0.540
#> GSM39188 3 0.1651 0.8363 0.036 0.008 0.944 0.012 0.000
#> GSM39189 3 0.2344 0.8364 0.064 0.000 0.904 0.032 0.000
#> GSM39190 3 0.1644 0.8565 0.048 0.004 0.940 0.008 0.000
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM39104 1 0.3036 0.838 0.876 0.004 0.020 0.016 0.056 0.028
#> GSM39105 1 0.2165 0.882 0.912 0.000 0.004 0.008 0.024 0.052
#> GSM39106 6 0.4875 0.771 0.200 0.004 0.040 0.000 0.052 0.704
#> GSM39107 6 0.4089 0.517 0.052 0.184 0.000 0.000 0.012 0.752
#> GSM39108 6 0.6669 0.616 0.240 0.004 0.104 0.012 0.088 0.552
#> GSM39109 5 0.7704 0.129 0.104 0.116 0.040 0.008 0.420 0.312
#> GSM39110 6 0.7584 0.436 0.216 0.004 0.188 0.020 0.120 0.452
#> GSM39111 1 0.6701 0.301 0.580 0.004 0.188 0.020 0.112 0.096
#> GSM39112 6 0.3930 0.764 0.156 0.072 0.004 0.000 0.000 0.768
#> GSM39113 6 0.3982 0.254 0.008 0.280 0.000 0.000 0.016 0.696
#> GSM39114 2 0.4127 0.482 0.000 0.620 0.000 0.012 0.004 0.364
#> GSM39115 1 0.1036 0.896 0.964 0.000 0.000 0.004 0.008 0.024
#> GSM39148 1 0.1668 0.881 0.928 0.000 0.000 0.004 0.008 0.060
#> GSM39149 3 0.0405 0.956 0.008 0.000 0.988 0.000 0.000 0.004
#> GSM39150 1 0.1265 0.886 0.948 0.000 0.008 0.000 0.044 0.000
#> GSM39151 3 0.0146 0.956 0.004 0.000 0.996 0.000 0.000 0.000
#> GSM39152 3 0.3634 0.797 0.048 0.000 0.836 0.012 0.068 0.036
#> GSM39153 1 0.0260 0.899 0.992 0.000 0.000 0.000 0.000 0.008
#> GSM39154 1 0.0692 0.893 0.976 0.000 0.004 0.000 0.020 0.000
#> GSM39155 1 0.0837 0.898 0.972 0.000 0.000 0.004 0.004 0.020
#> GSM39156 6 0.3909 0.797 0.236 0.000 0.012 0.000 0.020 0.732
#> GSM39157 1 0.1370 0.893 0.948 0.000 0.000 0.004 0.012 0.036
#> GSM39158 1 0.0767 0.895 0.976 0.000 0.008 0.004 0.012 0.000
#> GSM39159 1 0.5530 0.149 0.568 0.000 0.056 0.004 0.336 0.036
#> GSM39160 1 0.2407 0.856 0.896 0.000 0.016 0.008 0.072 0.008
#> GSM39161 5 0.5044 0.410 0.404 0.000 0.020 0.008 0.544 0.024
#> GSM39162 1 0.1787 0.876 0.920 0.000 0.000 0.004 0.008 0.068
#> GSM39163 1 0.0922 0.899 0.968 0.000 0.004 0.004 0.000 0.024
#> GSM39164 1 0.1500 0.890 0.936 0.000 0.000 0.000 0.012 0.052
#> GSM39165 1 0.5822 0.501 0.664 0.004 0.148 0.016 0.120 0.048
#> GSM39166 1 0.1382 0.888 0.948 0.000 0.008 0.000 0.036 0.008
#> GSM39167 1 0.1080 0.894 0.960 0.000 0.000 0.004 0.004 0.032
#> GSM39168 1 0.1477 0.889 0.940 0.000 0.000 0.004 0.008 0.048
#> GSM39169 1 0.0551 0.899 0.984 0.000 0.000 0.004 0.004 0.008
#> GSM39170 1 0.0551 0.895 0.984 0.000 0.004 0.004 0.008 0.000
#> GSM39171 1 0.2433 0.858 0.900 0.000 0.016 0.012 0.060 0.012
#> GSM39172 5 0.6425 0.614 0.124 0.012 0.168 0.024 0.620 0.052
#> GSM39173 3 0.0665 0.953 0.004 0.000 0.980 0.000 0.008 0.008
#> GSM39174 1 0.1003 0.899 0.964 0.000 0.000 0.000 0.016 0.020
#> GSM39175 1 0.1155 0.888 0.956 0.000 0.004 0.000 0.036 0.004
#> GSM39176 1 0.0922 0.896 0.968 0.000 0.000 0.004 0.004 0.024
#> GSM39177 3 0.0622 0.953 0.012 0.000 0.980 0.000 0.000 0.008
#> GSM39178 1 0.3847 0.632 0.748 0.000 0.020 0.004 0.220 0.008
#> GSM39179 3 0.0291 0.956 0.004 0.000 0.992 0.000 0.000 0.004
#> GSM39180 5 0.5281 0.334 0.028 0.000 0.336 0.008 0.588 0.040
#> GSM39181 1 0.1598 0.880 0.940 0.000 0.008 0.004 0.040 0.008
#> GSM39182 5 0.6125 0.660 0.148 0.016 0.044 0.036 0.664 0.092
#> GSM39183 1 0.1462 0.875 0.936 0.000 0.008 0.000 0.056 0.000
#> GSM39184 1 0.1226 0.887 0.952 0.000 0.004 0.000 0.040 0.004
#> GSM39185 5 0.5095 0.649 0.256 0.000 0.048 0.004 0.656 0.036
#> GSM39186 1 0.0951 0.900 0.968 0.000 0.004 0.000 0.008 0.020
#> GSM39187 1 0.1477 0.888 0.940 0.000 0.000 0.004 0.008 0.048
#> GSM39116 2 0.2361 0.357 0.000 0.880 0.000 0.104 0.012 0.004
#> GSM39117 4 0.2278 0.548 0.000 0.128 0.000 0.868 0.004 0.000
#> GSM39118 2 0.7598 -0.537 0.000 0.364 0.004 0.272 0.176 0.184
#> GSM39119 4 0.6411 0.489 0.000 0.348 0.000 0.464 0.052 0.136
#> GSM39120 6 0.3549 0.811 0.192 0.028 0.000 0.000 0.004 0.776
#> GSM39121 2 0.4456 0.360 0.000 0.524 0.000 0.000 0.028 0.448
#> GSM39122 2 0.4449 0.373 0.000 0.532 0.000 0.000 0.028 0.440
#> GSM39123 4 0.2320 0.547 0.000 0.132 0.000 0.864 0.004 0.000
#> GSM39124 2 0.4167 0.495 0.000 0.636 0.000 0.012 0.008 0.344
#> GSM39125 6 0.3996 0.816 0.212 0.020 0.012 0.000 0.008 0.748
#> GSM39126 2 0.4463 0.345 0.000 0.516 0.000 0.000 0.028 0.456
#> GSM39127 2 0.3724 0.466 0.000 0.804 0.000 0.088 0.012 0.096
#> GSM39128 2 0.4109 0.506 0.000 0.652 0.000 0.012 0.008 0.328
#> GSM39129 4 0.7743 0.555 0.000 0.260 0.004 0.320 0.232 0.184
#> GSM39130 4 0.2278 0.548 0.000 0.128 0.000 0.868 0.004 0.000
#> GSM39131 2 0.4124 0.504 0.000 0.648 0.000 0.012 0.008 0.332
#> GSM39132 2 0.2088 0.430 0.000 0.904 0.000 0.068 0.000 0.028
#> GSM39133 4 0.2631 0.523 0.000 0.180 0.000 0.820 0.000 0.000
#> GSM39134 2 0.6592 -0.521 0.000 0.400 0.000 0.396 0.060 0.144
#> GSM39135 2 0.2264 0.368 0.000 0.888 0.000 0.096 0.012 0.004
#> GSM39136 2 0.2872 0.348 0.000 0.832 0.000 0.152 0.012 0.004
#> GSM39137 2 0.4423 0.402 0.000 0.552 0.000 0.000 0.028 0.420
#> GSM39138 4 0.7736 0.556 0.000 0.260 0.004 0.324 0.228 0.184
#> GSM39139 4 0.7751 0.548 0.000 0.268 0.004 0.312 0.232 0.184
#> GSM39140 6 0.3449 0.816 0.196 0.016 0.000 0.000 0.008 0.780
#> GSM39141 6 0.3497 0.814 0.224 0.004 0.000 0.004 0.008 0.760
#> GSM39142 6 0.3463 0.802 0.240 0.000 0.000 0.004 0.008 0.748
#> GSM39143 6 0.3384 0.811 0.228 0.000 0.000 0.004 0.008 0.760
#> GSM39144 4 0.7743 0.555 0.000 0.260 0.004 0.320 0.232 0.184
#> GSM39145 2 0.7766 -0.515 0.000 0.356 0.008 0.220 0.224 0.192
#> GSM39146 2 0.2006 0.420 0.000 0.904 0.000 0.080 0.000 0.016
#> GSM39147 2 0.4555 0.544 0.000 0.680 0.000 0.016 0.044 0.260
#> GSM39188 3 0.0458 0.948 0.000 0.000 0.984 0.000 0.016 0.000
#> GSM39189 3 0.1500 0.921 0.012 0.000 0.936 0.000 0.052 0.000
#> GSM39190 3 0.0260 0.953 0.000 0.000 0.992 0.000 0.008 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) other(p) protocol(p) k
#> CV:mclust 83 1.00000 2.68e-05 2.45e-04 2
#> CV:mclust 84 0.00684 5.95e-13 2.98e-10 3
#> CV:mclust 70 0.08415 2.03e-10 4.56e-07 4
#> CV:mclust 75 0.06126 4.22e-11 3.01e-07 5
#> CV:mclust 64 0.07104 7.85e-08 1.91e-05 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 8353 rows and 87 columns.
#> Top rows (835, 1670, 2506, 3341, 4176) 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 0.837 0.901 0.960 0.4968 0.500 0.500
#> 3 3 0.443 0.462 0.669 0.3170 0.680 0.448
#> 4 4 0.427 0.439 0.676 0.0945 0.754 0.451
#> 5 5 0.493 0.392 0.636 0.0788 0.864 0.599
#> 6 6 0.544 0.444 0.658 0.0479 0.875 0.533
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
#> GSM39104 1 0.0000 0.9677 1.000 0.000
#> GSM39105 1 0.0000 0.9677 1.000 0.000
#> GSM39106 1 0.0000 0.9677 1.000 0.000
#> GSM39107 1 0.6623 0.7878 0.828 0.172
#> GSM39108 1 0.0000 0.9677 1.000 0.000
#> GSM39109 2 0.2948 0.9018 0.052 0.948
#> GSM39110 1 0.0376 0.9651 0.996 0.004
#> GSM39111 1 0.0000 0.9677 1.000 0.000
#> GSM39112 1 0.2236 0.9393 0.964 0.036
#> GSM39113 1 0.9686 0.3205 0.604 0.396
#> GSM39114 2 0.0672 0.9363 0.008 0.992
#> GSM39115 1 0.0000 0.9677 1.000 0.000
#> GSM39148 1 0.0000 0.9677 1.000 0.000
#> GSM39149 2 0.0938 0.9336 0.012 0.988
#> GSM39150 1 0.0000 0.9677 1.000 0.000
#> GSM39151 2 0.6247 0.7914 0.156 0.844
#> GSM39152 1 0.7453 0.7229 0.788 0.212
#> GSM39153 1 0.0000 0.9677 1.000 0.000
#> GSM39154 1 0.0000 0.9677 1.000 0.000
#> GSM39155 1 0.0000 0.9677 1.000 0.000
#> GSM39156 1 0.0000 0.9677 1.000 0.000
#> GSM39157 1 0.0000 0.9677 1.000 0.000
#> GSM39158 1 0.0000 0.9677 1.000 0.000
#> GSM39159 1 0.0672 0.9624 0.992 0.008
#> GSM39160 1 0.0000 0.9677 1.000 0.000
#> GSM39161 1 0.1414 0.9528 0.980 0.020
#> GSM39162 1 0.0000 0.9677 1.000 0.000
#> GSM39163 1 0.0000 0.9677 1.000 0.000
#> GSM39164 1 0.0000 0.9677 1.000 0.000
#> GSM39165 1 0.0376 0.9651 0.996 0.004
#> GSM39166 1 0.0000 0.9677 1.000 0.000
#> GSM39167 1 0.0000 0.9677 1.000 0.000
#> GSM39168 1 0.0000 0.9677 1.000 0.000
#> GSM39169 1 0.0000 0.9677 1.000 0.000
#> GSM39170 1 0.0000 0.9677 1.000 0.000
#> GSM39171 1 0.0000 0.9677 1.000 0.000
#> GSM39172 2 0.0000 0.9414 0.000 1.000
#> GSM39173 2 0.0000 0.9414 0.000 1.000
#> GSM39174 1 0.0000 0.9677 1.000 0.000
#> GSM39175 1 0.0000 0.9677 1.000 0.000
#> GSM39176 1 0.0000 0.9677 1.000 0.000
#> GSM39177 2 0.9795 0.2957 0.416 0.584
#> GSM39178 1 0.0000 0.9677 1.000 0.000
#> GSM39179 2 0.0000 0.9414 0.000 1.000
#> GSM39180 2 0.0000 0.9414 0.000 1.000
#> GSM39181 1 0.0000 0.9677 1.000 0.000
#> GSM39182 2 0.0000 0.9414 0.000 1.000
#> GSM39183 1 0.0000 0.9677 1.000 0.000
#> GSM39184 1 0.0000 0.9677 1.000 0.000
#> GSM39185 1 0.7950 0.6768 0.760 0.240
#> GSM39186 1 0.0000 0.9677 1.000 0.000
#> GSM39187 1 0.0000 0.9677 1.000 0.000
#> GSM39116 2 0.0000 0.9414 0.000 1.000
#> GSM39117 2 0.0000 0.9414 0.000 1.000
#> GSM39118 2 0.0000 0.9414 0.000 1.000
#> GSM39119 2 0.0000 0.9414 0.000 1.000
#> GSM39120 1 0.5737 0.8349 0.864 0.136
#> GSM39121 2 1.0000 0.0146 0.500 0.500
#> GSM39122 2 0.9608 0.3911 0.384 0.616
#> GSM39123 2 0.0000 0.9414 0.000 1.000
#> GSM39124 2 0.0000 0.9414 0.000 1.000
#> GSM39125 1 0.5842 0.8296 0.860 0.140
#> GSM39126 2 0.9686 0.3602 0.396 0.604
#> GSM39127 2 0.0000 0.9414 0.000 1.000
#> GSM39128 2 0.0000 0.9414 0.000 1.000
#> GSM39129 2 0.0000 0.9414 0.000 1.000
#> GSM39130 2 0.0000 0.9414 0.000 1.000
#> GSM39131 2 0.0000 0.9414 0.000 1.000
#> GSM39132 2 0.0000 0.9414 0.000 1.000
#> GSM39133 2 0.0000 0.9414 0.000 1.000
#> GSM39134 2 0.0000 0.9414 0.000 1.000
#> GSM39135 2 0.0000 0.9414 0.000 1.000
#> GSM39136 2 0.0000 0.9414 0.000 1.000
#> GSM39137 2 0.5408 0.8304 0.124 0.876
#> GSM39138 2 0.0000 0.9414 0.000 1.000
#> GSM39139 2 0.0000 0.9414 0.000 1.000
#> GSM39140 1 0.2236 0.9392 0.964 0.036
#> GSM39141 1 0.0000 0.9677 1.000 0.000
#> GSM39142 1 0.0000 0.9677 1.000 0.000
#> GSM39143 1 0.0000 0.9677 1.000 0.000
#> GSM39144 2 0.0000 0.9414 0.000 1.000
#> GSM39145 2 0.0000 0.9414 0.000 1.000
#> GSM39146 2 0.0000 0.9414 0.000 1.000
#> GSM39147 2 0.0000 0.9414 0.000 1.000
#> GSM39188 2 0.0000 0.9414 0.000 1.000
#> GSM39189 2 0.2948 0.9032 0.052 0.948
#> GSM39190 2 0.0000 0.9414 0.000 1.000
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM39104 1 0.4235 0.6988 0.824 0.000 0.176
#> GSM39105 1 0.6008 0.4739 0.628 0.000 0.372
#> GSM39106 3 0.5016 0.4867 0.240 0.000 0.760
#> GSM39107 3 0.1163 0.6262 0.028 0.000 0.972
#> GSM39108 3 0.6267 -0.0439 0.452 0.000 0.548
#> GSM39109 2 0.6079 0.5455 0.036 0.748 0.216
#> GSM39110 1 0.6140 0.4117 0.596 0.000 0.404
#> GSM39111 1 0.2448 0.7188 0.924 0.000 0.076
#> GSM39112 3 0.2261 0.6323 0.068 0.000 0.932
#> GSM39113 3 0.0829 0.6174 0.012 0.004 0.984
#> GSM39114 3 0.4235 0.4551 0.000 0.176 0.824
#> GSM39115 1 0.6267 0.2970 0.548 0.000 0.452
#> GSM39148 3 0.6308 -0.1798 0.492 0.000 0.508
#> GSM39149 1 0.6180 0.1039 0.584 0.416 0.000
#> GSM39150 1 0.0892 0.7140 0.980 0.000 0.020
#> GSM39151 1 0.6308 -0.1199 0.508 0.492 0.000
#> GSM39152 1 0.4931 0.4806 0.768 0.232 0.000
#> GSM39153 1 0.3941 0.7070 0.844 0.000 0.156
#> GSM39154 1 0.3412 0.7137 0.876 0.000 0.124
#> GSM39155 1 0.4002 0.7052 0.840 0.000 0.160
#> GSM39156 3 0.5016 0.4846 0.240 0.000 0.760
#> GSM39157 1 0.6062 0.4538 0.616 0.000 0.384
#> GSM39158 1 0.4002 0.7055 0.840 0.000 0.160
#> GSM39159 1 0.2165 0.6739 0.936 0.064 0.000
#> GSM39160 1 0.0424 0.7099 0.992 0.000 0.008
#> GSM39161 1 0.4235 0.5566 0.824 0.176 0.000
#> GSM39162 3 0.6140 0.1065 0.404 0.000 0.596
#> GSM39163 1 0.5363 0.6079 0.724 0.000 0.276
#> GSM39164 1 0.6045 0.4602 0.620 0.000 0.380
#> GSM39165 1 0.1129 0.7004 0.976 0.020 0.004
#> GSM39166 1 0.1529 0.7174 0.960 0.000 0.040
#> GSM39167 1 0.5785 0.5366 0.668 0.000 0.332
#> GSM39168 3 0.6302 -0.1426 0.480 0.000 0.520
#> GSM39169 1 0.5098 0.6423 0.752 0.000 0.248
#> GSM39170 1 0.4452 0.6896 0.808 0.000 0.192
#> GSM39171 1 0.0747 0.7130 0.984 0.000 0.016
#> GSM39172 2 0.6274 0.1950 0.456 0.544 0.000
#> GSM39173 2 0.6111 0.3077 0.396 0.604 0.000
#> GSM39174 1 0.4750 0.6673 0.784 0.000 0.216
#> GSM39175 1 0.1163 0.7160 0.972 0.000 0.028
#> GSM39176 1 0.6079 0.4496 0.612 0.000 0.388
#> GSM39177 1 0.5835 0.2881 0.660 0.340 0.000
#> GSM39178 1 0.2711 0.6524 0.912 0.088 0.000
#> GSM39179 2 0.6309 0.0857 0.500 0.500 0.000
#> GSM39180 2 0.5254 0.4834 0.264 0.736 0.000
#> GSM39181 1 0.2165 0.7191 0.936 0.000 0.064
#> GSM39182 2 0.6045 0.3358 0.380 0.620 0.000
#> GSM39183 1 0.0747 0.7130 0.984 0.000 0.016
#> GSM39184 1 0.3816 0.7084 0.852 0.000 0.148
#> GSM39185 1 0.5098 0.4566 0.752 0.248 0.000
#> GSM39186 1 0.4452 0.6873 0.808 0.000 0.192
#> GSM39187 1 0.6286 0.2586 0.536 0.000 0.464
#> GSM39116 2 0.5968 0.4383 0.000 0.636 0.364
#> GSM39117 2 0.0424 0.6258 0.000 0.992 0.008
#> GSM39118 2 0.4702 0.5940 0.000 0.788 0.212
#> GSM39119 2 0.2796 0.6419 0.000 0.908 0.092
#> GSM39120 3 0.2448 0.6293 0.076 0.000 0.924
#> GSM39121 3 0.0592 0.6061 0.000 0.012 0.988
#> GSM39122 3 0.2356 0.5657 0.000 0.072 0.928
#> GSM39123 2 0.1753 0.6377 0.000 0.952 0.048
#> GSM39124 3 0.5988 0.1447 0.000 0.368 0.632
#> GSM39125 3 0.3340 0.6058 0.120 0.000 0.880
#> GSM39126 3 0.1964 0.5806 0.000 0.056 0.944
#> GSM39127 3 0.6079 0.0955 0.000 0.388 0.612
#> GSM39128 3 0.6008 0.1352 0.000 0.372 0.628
#> GSM39129 2 0.3340 0.6390 0.000 0.880 0.120
#> GSM39130 2 0.1860 0.6385 0.000 0.948 0.052
#> GSM39131 3 0.5988 0.1447 0.000 0.368 0.632
#> GSM39132 3 0.6308 -0.1890 0.000 0.492 0.508
#> GSM39133 2 0.4121 0.6230 0.000 0.832 0.168
#> GSM39134 2 0.3816 0.6318 0.000 0.852 0.148
#> GSM39135 2 0.6062 0.4059 0.000 0.616 0.384
#> GSM39136 2 0.6111 0.3850 0.000 0.604 0.396
#> GSM39137 3 0.4504 0.4349 0.000 0.196 0.804
#> GSM39138 2 0.2959 0.6417 0.000 0.900 0.100
#> GSM39139 2 0.5760 0.4831 0.000 0.672 0.328
#> GSM39140 3 0.4121 0.5715 0.168 0.000 0.832
#> GSM39141 3 0.4291 0.5609 0.180 0.000 0.820
#> GSM39142 3 0.4974 0.4931 0.236 0.000 0.764
#> GSM39143 3 0.4504 0.5445 0.196 0.000 0.804
#> GSM39144 2 0.3941 0.6287 0.000 0.844 0.156
#> GSM39145 2 0.5948 0.4434 0.000 0.640 0.360
#> GSM39146 2 0.6295 0.2170 0.000 0.528 0.472
#> GSM39147 3 0.6295 -0.1377 0.000 0.472 0.528
#> GSM39188 2 0.6295 0.1564 0.472 0.528 0.000
#> GSM39189 1 0.6192 0.0911 0.580 0.420 0.000
#> GSM39190 2 0.6302 0.1375 0.480 0.520 0.000
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM39104 1 0.419 0.6681 0.816 0.032 0.148 0.004
#> GSM39105 1 0.418 0.6659 0.832 0.116 0.044 0.008
#> GSM39106 1 0.592 0.0443 0.492 0.480 0.012 0.016
#> GSM39107 2 0.707 0.3234 0.348 0.536 0.008 0.108
#> GSM39108 1 0.651 0.5207 0.644 0.260 0.080 0.016
#> GSM39109 4 0.765 0.5769 0.064 0.192 0.132 0.612
#> GSM39110 1 0.802 0.3367 0.444 0.220 0.324 0.012
#> GSM39111 1 0.660 0.2967 0.504 0.052 0.432 0.012
#> GSM39112 2 0.590 0.2494 0.388 0.576 0.004 0.032
#> GSM39113 2 0.543 0.4302 0.300 0.668 0.004 0.028
#> GSM39114 2 0.327 0.5303 0.056 0.884 0.004 0.056
#> GSM39115 1 0.363 0.6613 0.868 0.076 0.008 0.048
#> GSM39148 1 0.397 0.6105 0.816 0.164 0.004 0.016
#> GSM39149 3 0.252 0.6648 0.060 0.020 0.916 0.004
#> GSM39150 1 0.592 0.4795 0.652 0.004 0.288 0.056
#> GSM39151 3 0.313 0.6645 0.100 0.008 0.880 0.012
#> GSM39152 3 0.403 0.5780 0.192 0.008 0.796 0.004
#> GSM39153 1 0.303 0.6835 0.888 0.020 0.088 0.004
#> GSM39154 1 0.423 0.6286 0.804 0.004 0.168 0.024
#> GSM39155 1 0.322 0.6805 0.876 0.020 0.100 0.004
#> GSM39156 1 0.577 0.1555 0.540 0.436 0.008 0.016
#> GSM39157 1 0.293 0.6718 0.888 0.096 0.012 0.004
#> GSM39158 1 0.353 0.6637 0.864 0.000 0.056 0.080
#> GSM39159 1 0.663 0.3325 0.564 0.000 0.336 0.100
#> GSM39160 1 0.595 0.4246 0.616 0.000 0.328 0.056
#> GSM39161 1 0.766 0.1759 0.464 0.000 0.260 0.276
#> GSM39162 1 0.489 0.5205 0.732 0.244 0.008 0.016
#> GSM39163 1 0.211 0.6923 0.940 0.024 0.020 0.016
#> GSM39164 1 0.339 0.6731 0.868 0.104 0.024 0.004
#> GSM39165 3 0.546 -0.2105 0.492 0.004 0.496 0.008
#> GSM39166 1 0.503 0.6061 0.768 0.000 0.140 0.092
#> GSM39167 1 0.251 0.6736 0.912 0.072 0.004 0.012
#> GSM39168 1 0.418 0.5996 0.800 0.180 0.008 0.012
#> GSM39169 1 0.258 0.6928 0.916 0.032 0.048 0.004
#> GSM39170 1 0.349 0.6790 0.880 0.020 0.032 0.068
#> GSM39171 1 0.528 0.3544 0.588 0.000 0.400 0.012
#> GSM39172 4 0.729 -0.0655 0.164 0.000 0.340 0.496
#> GSM39173 3 0.260 0.6057 0.004 0.064 0.912 0.020
#> GSM39174 1 0.380 0.6859 0.852 0.044 0.100 0.004
#> GSM39175 1 0.521 0.5173 0.692 0.004 0.280 0.024
#> GSM39176 1 0.208 0.6780 0.932 0.056 0.004 0.008
#> GSM39177 3 0.301 0.6635 0.092 0.008 0.888 0.012
#> GSM39178 1 0.652 0.2633 0.536 0.000 0.384 0.080
#> GSM39179 3 0.209 0.6467 0.024 0.024 0.940 0.012
#> GSM39180 3 0.614 0.3607 0.052 0.008 0.624 0.316
#> GSM39181 1 0.505 0.6035 0.756 0.000 0.068 0.176
#> GSM39182 4 0.451 0.4662 0.120 0.000 0.076 0.804
#> GSM39183 1 0.582 0.5600 0.708 0.000 0.148 0.144
#> GSM39184 1 0.328 0.6633 0.864 0.000 0.116 0.020
#> GSM39185 1 0.783 0.0481 0.412 0.000 0.292 0.296
#> GSM39186 1 0.399 0.6701 0.832 0.032 0.132 0.004
#> GSM39187 1 0.305 0.6543 0.880 0.104 0.004 0.012
#> GSM39116 4 0.590 0.5361 0.000 0.348 0.048 0.604
#> GSM39117 4 0.379 0.6843 0.000 0.044 0.112 0.844
#> GSM39118 2 0.775 -0.3965 0.000 0.388 0.232 0.380
#> GSM39119 4 0.659 0.6283 0.000 0.156 0.216 0.628
#> GSM39120 2 0.530 0.2638 0.388 0.600 0.004 0.008
#> GSM39121 2 0.325 0.5490 0.140 0.852 0.000 0.008
#> GSM39122 2 0.328 0.5511 0.124 0.860 0.000 0.016
#> GSM39123 4 0.258 0.6895 0.000 0.052 0.036 0.912
#> GSM39124 2 0.462 0.4254 0.016 0.792 0.024 0.168
#> GSM39125 1 0.743 0.0559 0.500 0.368 0.016 0.116
#> GSM39126 2 0.320 0.5508 0.136 0.856 0.000 0.008
#> GSM39127 2 0.543 0.1988 0.020 0.640 0.004 0.336
#> GSM39128 2 0.500 0.3953 0.020 0.748 0.016 0.216
#> GSM39129 3 0.751 -0.1172 0.000 0.376 0.440 0.184
#> GSM39130 4 0.339 0.6987 0.000 0.056 0.072 0.872
#> GSM39131 2 0.487 0.3873 0.012 0.760 0.024 0.204
#> GSM39132 2 0.573 -0.0792 0.000 0.576 0.032 0.392
#> GSM39133 4 0.305 0.6912 0.000 0.088 0.028 0.884
#> GSM39134 4 0.691 0.5990 0.000 0.240 0.172 0.588
#> GSM39135 4 0.597 0.5023 0.000 0.368 0.048 0.584
#> GSM39136 4 0.531 0.6267 0.000 0.256 0.044 0.700
#> GSM39137 2 0.427 0.5193 0.068 0.828 0.004 0.100
#> GSM39138 3 0.780 -0.2284 0.000 0.328 0.412 0.260
#> GSM39139 2 0.695 -0.0181 0.000 0.544 0.324 0.132
#> GSM39140 2 0.576 0.0274 0.456 0.520 0.004 0.020
#> GSM39141 1 0.606 0.0880 0.516 0.448 0.008 0.028
#> GSM39142 1 0.619 0.2758 0.584 0.364 0.008 0.044
#> GSM39143 1 0.612 0.1259 0.532 0.428 0.008 0.032
#> GSM39144 3 0.723 -0.0434 0.000 0.392 0.464 0.144
#> GSM39145 2 0.635 0.1298 0.000 0.636 0.252 0.112
#> GSM39146 4 0.559 0.4893 0.000 0.372 0.028 0.600
#> GSM39147 2 0.455 0.3833 0.000 0.804 0.096 0.100
#> GSM39188 3 0.354 0.6514 0.076 0.000 0.864 0.060
#> GSM39189 3 0.415 0.6387 0.132 0.000 0.820 0.048
#> GSM39190 3 0.299 0.6603 0.056 0.008 0.900 0.036
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM39104 5 0.6764 0.1944 0.384 0.080 0.048 0.004 0.484
#> GSM39105 1 0.6011 0.2088 0.556 0.120 0.004 0.000 0.320
#> GSM39106 2 0.6958 0.2829 0.232 0.464 0.016 0.000 0.288
#> GSM39107 2 0.7119 0.5130 0.180 0.588 0.016 0.060 0.156
#> GSM39108 1 0.7125 0.0359 0.396 0.252 0.016 0.000 0.336
#> GSM39109 4 0.8574 0.1889 0.036 0.256 0.072 0.364 0.272
#> GSM39110 5 0.8167 0.1666 0.252 0.184 0.156 0.000 0.408
#> GSM39111 5 0.7597 0.2752 0.244 0.072 0.192 0.004 0.488
#> GSM39112 2 0.6330 0.4543 0.240 0.580 0.004 0.008 0.168
#> GSM39113 2 0.5658 0.5564 0.120 0.676 0.008 0.008 0.188
#> GSM39114 2 0.3059 0.5881 0.004 0.872 0.012 0.020 0.092
#> GSM39115 1 0.5716 0.2716 0.616 0.076 0.016 0.000 0.292
#> GSM39148 1 0.2579 0.5455 0.900 0.064 0.004 0.004 0.028
#> GSM39149 3 0.4919 0.5682 0.024 0.000 0.684 0.024 0.268
#> GSM39150 5 0.6507 0.4422 0.384 0.016 0.080 0.016 0.504
#> GSM39151 3 0.5582 0.5094 0.024 0.000 0.568 0.036 0.372
#> GSM39152 3 0.6644 0.3649 0.068 0.020 0.476 0.024 0.412
#> GSM39153 1 0.3268 0.5076 0.852 0.004 0.028 0.004 0.112
#> GSM39154 1 0.3304 0.4804 0.852 0.000 0.052 0.004 0.092
#> GSM39155 1 0.3500 0.4381 0.808 0.004 0.016 0.000 0.172
#> GSM39156 1 0.6228 0.3485 0.580 0.248 0.004 0.004 0.164
#> GSM39157 1 0.0912 0.5484 0.972 0.012 0.000 0.000 0.016
#> GSM39158 1 0.4686 0.0700 0.636 0.008 0.004 0.008 0.344
#> GSM39159 5 0.6563 0.4063 0.420 0.004 0.060 0.048 0.468
#> GSM39160 5 0.6405 0.4560 0.348 0.000 0.104 0.024 0.524
#> GSM39161 5 0.7252 0.4057 0.356 0.004 0.036 0.164 0.440
#> GSM39162 1 0.4275 0.5116 0.796 0.116 0.008 0.004 0.076
#> GSM39163 1 0.2286 0.4977 0.888 0.000 0.004 0.000 0.108
#> GSM39164 1 0.3058 0.5399 0.876 0.036 0.008 0.004 0.076
#> GSM39165 1 0.6880 -0.1320 0.516 0.004 0.240 0.016 0.224
#> GSM39166 1 0.5786 -0.3569 0.480 0.004 0.024 0.032 0.460
#> GSM39167 1 0.1913 0.5411 0.932 0.016 0.008 0.000 0.044
#> GSM39168 1 0.3821 0.5237 0.824 0.104 0.004 0.004 0.064
#> GSM39169 1 0.2796 0.5012 0.868 0.008 0.008 0.000 0.116
#> GSM39170 1 0.4628 0.0283 0.624 0.004 0.008 0.004 0.360
#> GSM39171 1 0.6625 -0.3492 0.456 0.000 0.128 0.020 0.396
#> GSM39172 4 0.5877 0.2929 0.016 0.000 0.156 0.648 0.180
#> GSM39173 3 0.4085 0.5690 0.004 0.040 0.796 0.008 0.152
#> GSM39174 1 0.2499 0.5341 0.908 0.008 0.028 0.004 0.052
#> GSM39175 1 0.4995 0.2449 0.704 0.000 0.084 0.004 0.208
#> GSM39176 1 0.2452 0.5251 0.896 0.016 0.004 0.000 0.084
#> GSM39177 3 0.5711 0.5313 0.076 0.000 0.660 0.032 0.232
#> GSM39178 5 0.6977 0.5152 0.316 0.004 0.104 0.056 0.520
#> GSM39179 3 0.5308 0.5761 0.016 0.012 0.696 0.048 0.228
#> GSM39180 3 0.7356 0.2474 0.028 0.008 0.408 0.184 0.372
#> GSM39181 1 0.5699 -0.1625 0.552 0.000 0.008 0.068 0.372
#> GSM39182 4 0.2824 0.5774 0.028 0.000 0.016 0.888 0.068
#> GSM39183 5 0.6087 0.3257 0.440 0.008 0.016 0.056 0.480
#> GSM39184 1 0.3449 0.4379 0.812 0.000 0.024 0.000 0.164
#> GSM39185 5 0.7633 0.4572 0.264 0.012 0.068 0.160 0.496
#> GSM39186 1 0.4328 0.4084 0.752 0.012 0.020 0.004 0.212
#> GSM39187 1 0.2152 0.5469 0.920 0.032 0.004 0.000 0.044
#> GSM39116 4 0.5628 0.4612 0.000 0.360 0.032 0.576 0.032
#> GSM39117 4 0.1281 0.6464 0.000 0.012 0.032 0.956 0.000
#> GSM39118 4 0.7133 0.3495 0.000 0.304 0.272 0.408 0.016
#> GSM39119 4 0.5181 0.6179 0.000 0.124 0.148 0.716 0.012
#> GSM39120 2 0.5603 0.5516 0.212 0.660 0.004 0.004 0.120
#> GSM39121 2 0.5374 0.5647 0.220 0.696 0.040 0.004 0.040
#> GSM39122 2 0.4510 0.6180 0.128 0.792 0.016 0.016 0.048
#> GSM39123 4 0.0807 0.6510 0.000 0.012 0.012 0.976 0.000
#> GSM39124 2 0.5346 0.4629 0.008 0.724 0.120 0.132 0.016
#> GSM39125 1 0.7403 -0.0437 0.412 0.392 0.016 0.032 0.148
#> GSM39126 2 0.3619 0.6211 0.096 0.848 0.024 0.008 0.024
#> GSM39127 2 0.4816 0.4271 0.000 0.724 0.028 0.216 0.032
#> GSM39128 2 0.4830 0.4929 0.004 0.760 0.072 0.144 0.020
#> GSM39129 3 0.6176 0.2082 0.000 0.292 0.580 0.108 0.020
#> GSM39130 4 0.0912 0.6534 0.000 0.016 0.012 0.972 0.000
#> GSM39131 2 0.4430 0.5205 0.000 0.784 0.024 0.136 0.056
#> GSM39132 2 0.5078 0.3350 0.000 0.692 0.040 0.244 0.024
#> GSM39133 4 0.2116 0.6663 0.000 0.076 0.004 0.912 0.008
#> GSM39134 4 0.6637 0.4655 0.000 0.308 0.164 0.512 0.016
#> GSM39135 4 0.5495 0.4585 0.000 0.364 0.048 0.576 0.012
#> GSM39136 4 0.5669 0.4128 0.000 0.380 0.040 0.556 0.024
#> GSM39137 2 0.4580 0.5842 0.056 0.808 0.064 0.056 0.016
#> GSM39138 3 0.6461 0.1492 0.000 0.268 0.572 0.132 0.028
#> GSM39139 3 0.5697 -0.0953 0.000 0.432 0.508 0.032 0.028
#> GSM39140 1 0.5599 0.3712 0.620 0.300 0.008 0.004 0.068
#> GSM39141 1 0.5510 0.4227 0.668 0.228 0.008 0.004 0.092
#> GSM39142 1 0.5222 0.4463 0.696 0.196 0.000 0.008 0.100
#> GSM39143 1 0.5601 0.4248 0.668 0.224 0.008 0.008 0.092
#> GSM39144 3 0.5519 0.2311 0.000 0.304 0.624 0.052 0.020
#> GSM39145 2 0.5694 0.1628 0.000 0.504 0.436 0.036 0.024
#> GSM39146 4 0.4838 0.5579 0.000 0.284 0.020 0.676 0.020
#> GSM39147 2 0.5203 0.4298 0.000 0.688 0.240 0.044 0.028
#> GSM39188 3 0.6006 0.5308 0.024 0.000 0.568 0.072 0.336
#> GSM39189 3 0.6287 0.3961 0.036 0.004 0.472 0.052 0.436
#> GSM39190 3 0.4875 0.5513 0.012 0.008 0.636 0.008 0.336
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM39104 6 0.760 -0.1071 0.196 0.000 0.208 0.000 0.256 0.340
#> GSM39105 1 0.716 0.1022 0.368 0.004 0.096 0.000 0.168 0.364
#> GSM39106 6 0.570 0.3722 0.120 0.012 0.084 0.000 0.108 0.676
#> GSM39107 6 0.465 0.4456 0.092 0.016 0.000 0.060 0.064 0.768
#> GSM39108 6 0.710 0.1239 0.256 0.004 0.168 0.000 0.112 0.460
#> GSM39109 6 0.773 -0.0248 0.024 0.016 0.212 0.288 0.064 0.396
#> GSM39110 3 0.714 0.2110 0.144 0.024 0.404 0.000 0.064 0.364
#> GSM39111 3 0.741 0.3123 0.136 0.012 0.436 0.000 0.160 0.256
#> GSM39112 6 0.397 0.4574 0.140 0.020 0.000 0.012 0.036 0.792
#> GSM39113 6 0.325 0.4420 0.056 0.044 0.008 0.012 0.016 0.864
#> GSM39114 6 0.425 0.2902 0.004 0.220 0.004 0.024 0.016 0.732
#> GSM39115 1 0.637 -0.1008 0.384 0.004 0.012 0.000 0.384 0.216
#> GSM39148 1 0.261 0.6923 0.884 0.012 0.000 0.000 0.060 0.044
#> GSM39149 3 0.399 0.6678 0.028 0.092 0.820 0.016 0.028 0.016
#> GSM39150 5 0.707 0.3543 0.184 0.008 0.252 0.000 0.468 0.088
#> GSM39151 3 0.383 0.6578 0.004 0.028 0.824 0.024 0.096 0.024
#> GSM39152 3 0.544 0.5986 0.036 0.036 0.720 0.012 0.128 0.068
#> GSM39153 1 0.332 0.6827 0.848 0.004 0.044 0.000 0.076 0.028
#> GSM39154 1 0.285 0.6716 0.872 0.008 0.052 0.004 0.064 0.000
#> GSM39155 1 0.446 0.5300 0.684 0.000 0.020 0.000 0.264 0.032
#> GSM39156 1 0.567 0.4117 0.592 0.024 0.028 0.000 0.052 0.304
#> GSM39157 1 0.276 0.6880 0.856 0.004 0.000 0.000 0.116 0.024
#> GSM39158 5 0.449 0.4578 0.352 0.004 0.000 0.008 0.616 0.020
#> GSM39159 5 0.528 0.6372 0.120 0.012 0.136 0.016 0.704 0.012
#> GSM39160 5 0.746 0.1608 0.204 0.008 0.328 0.000 0.352 0.108
#> GSM39161 5 0.509 0.6374 0.128 0.012 0.048 0.056 0.740 0.016
#> GSM39162 1 0.221 0.6856 0.912 0.016 0.004 0.000 0.020 0.048
#> GSM39163 1 0.326 0.6232 0.772 0.000 0.000 0.000 0.216 0.012
#> GSM39164 1 0.251 0.6951 0.892 0.000 0.020 0.000 0.060 0.028
#> GSM39165 1 0.647 0.2566 0.536 0.064 0.288 0.000 0.096 0.016
#> GSM39166 5 0.469 0.6648 0.204 0.000 0.048 0.012 0.716 0.020
#> GSM39167 1 0.256 0.6821 0.864 0.008 0.000 0.000 0.120 0.008
#> GSM39168 1 0.173 0.6886 0.924 0.004 0.000 0.000 0.008 0.064
#> GSM39169 1 0.387 0.5652 0.712 0.000 0.004 0.000 0.264 0.020
#> GSM39170 5 0.461 0.5548 0.296 0.000 0.016 0.004 0.656 0.028
#> GSM39171 1 0.691 0.0144 0.432 0.004 0.284 0.000 0.224 0.056
#> GSM39172 4 0.536 0.4465 0.024 0.036 0.216 0.680 0.036 0.008
#> GSM39173 3 0.572 0.5109 0.004 0.280 0.568 0.000 0.136 0.012
#> GSM39174 1 0.253 0.6939 0.884 0.000 0.024 0.000 0.080 0.012
#> GSM39175 1 0.436 0.5836 0.732 0.000 0.108 0.000 0.156 0.004
#> GSM39176 1 0.311 0.6602 0.812 0.004 0.000 0.000 0.168 0.016
#> GSM39177 3 0.604 0.6046 0.092 0.128 0.676 0.052 0.040 0.012
#> GSM39178 5 0.604 0.4903 0.088 0.004 0.232 0.012 0.612 0.052
#> GSM39179 3 0.592 0.6139 0.048 0.124 0.688 0.088 0.036 0.016
#> GSM39180 5 0.661 -0.0253 0.004 0.092 0.320 0.048 0.512 0.024
#> GSM39181 5 0.448 0.6154 0.240 0.000 0.000 0.028 0.700 0.032
#> GSM39182 4 0.361 0.6345 0.016 0.020 0.076 0.840 0.044 0.004
#> GSM39183 5 0.407 0.6835 0.160 0.000 0.036 0.012 0.776 0.016
#> GSM39184 1 0.449 0.5740 0.704 0.004 0.040 0.000 0.236 0.016
#> GSM39185 5 0.425 0.6110 0.072 0.000 0.080 0.032 0.796 0.020
#> GSM39186 1 0.527 0.5366 0.664 0.004 0.048 0.000 0.224 0.060
#> GSM39187 1 0.330 0.6827 0.816 0.000 0.000 0.000 0.128 0.056
#> GSM39116 4 0.597 0.4477 0.000 0.116 0.004 0.552 0.032 0.296
#> GSM39117 4 0.202 0.6872 0.000 0.028 0.036 0.920 0.016 0.000
#> GSM39118 4 0.627 0.1592 0.000 0.412 0.044 0.452 0.016 0.076
#> GSM39119 4 0.558 0.6078 0.000 0.168 0.068 0.684 0.044 0.036
#> GSM39120 6 0.546 0.3979 0.212 0.080 0.004 0.004 0.040 0.660
#> GSM39121 2 0.674 -0.0659 0.252 0.388 0.008 0.004 0.016 0.332
#> GSM39122 6 0.578 0.1269 0.092 0.340 0.000 0.020 0.008 0.540
#> GSM39123 4 0.123 0.6914 0.000 0.004 0.016 0.956 0.024 0.000
#> GSM39124 2 0.525 0.4117 0.040 0.624 0.000 0.056 0.000 0.280
#> GSM39125 6 0.680 0.2253 0.328 0.048 0.000 0.024 0.128 0.472
#> GSM39126 6 0.565 -0.0512 0.092 0.404 0.000 0.008 0.008 0.488
#> GSM39127 6 0.670 0.0501 0.008 0.252 0.004 0.184 0.044 0.508
#> GSM39128 2 0.632 0.1956 0.040 0.464 0.000 0.088 0.016 0.392
#> GSM39129 2 0.631 0.4548 0.000 0.620 0.180 0.032 0.076 0.092
#> GSM39130 4 0.143 0.6925 0.000 0.008 0.024 0.948 0.020 0.000
#> GSM39131 6 0.574 0.1423 0.004 0.268 0.008 0.080 0.032 0.608
#> GSM39132 6 0.685 -0.0763 0.000 0.296 0.004 0.224 0.048 0.428
#> GSM39133 4 0.219 0.6888 0.000 0.004 0.000 0.904 0.032 0.060
#> GSM39134 2 0.593 -0.1706 0.000 0.468 0.020 0.428 0.044 0.040
#> GSM39135 4 0.603 0.4196 0.000 0.224 0.004 0.564 0.024 0.184
#> GSM39136 4 0.675 0.4020 0.000 0.136 0.012 0.512 0.072 0.268
#> GSM39137 2 0.637 0.1660 0.092 0.464 0.000 0.036 0.020 0.388
#> GSM39138 2 0.509 0.4673 0.000 0.732 0.112 0.088 0.048 0.020
#> GSM39139 2 0.282 0.5697 0.000 0.884 0.052 0.024 0.012 0.028
#> GSM39140 1 0.514 0.5174 0.692 0.124 0.008 0.000 0.020 0.156
#> GSM39141 1 0.386 0.6249 0.792 0.056 0.000 0.000 0.020 0.132
#> GSM39142 1 0.412 0.6045 0.740 0.024 0.000 0.000 0.028 0.208
#> GSM39143 1 0.454 0.5290 0.680 0.032 0.000 0.000 0.024 0.264
#> GSM39144 2 0.409 0.4987 0.000 0.784 0.144 0.032 0.028 0.012
#> GSM39145 2 0.381 0.5653 0.000 0.816 0.052 0.016 0.016 0.100
#> GSM39146 4 0.445 0.6189 0.000 0.080 0.008 0.732 0.004 0.176
#> GSM39147 2 0.473 0.5034 0.024 0.716 0.016 0.040 0.000 0.204
#> GSM39188 3 0.505 0.6299 0.004 0.076 0.724 0.084 0.112 0.000
#> GSM39189 3 0.566 0.5547 0.004 0.048 0.652 0.036 0.228 0.032
#> GSM39190 3 0.512 0.5855 0.000 0.116 0.668 0.000 0.196 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) other(p) protocol(p) k
#> CV:NMF 82 0.1504 2.11e-07 1.17e-05 2
#> CV:NMF 46 0.2746 1.45e-07 1.77e-05 3
#> CV:NMF 50 0.6041 2.73e-05 3.51e-08 4
#> CV:NMF 35 0.0502 1.84e-04 5.27e-05 5
#> CV:NMF 47 NA 8.00e-03 1.94e-06 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 8353 rows and 87 columns.
#> Top rows (835, 1670, 2506, 3341, 4176) 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 0.563 0.856 0.911 0.4057 0.550 0.550
#> 3 3 0.461 0.778 0.885 0.1423 0.942 0.895
#> 4 4 0.452 0.758 0.874 0.0554 0.985 0.969
#> 5 5 0.428 0.749 0.867 0.0325 0.999 0.999
#> 6 6 0.433 0.727 0.835 0.0363 1.000 1.000
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
#> GSM39104 1 0.0938 0.93789 0.988 0.012
#> GSM39105 1 0.0000 0.94194 1.000 0.000
#> GSM39106 1 0.4161 0.87110 0.916 0.084
#> GSM39107 1 0.9833 -0.00291 0.576 0.424
#> GSM39108 1 0.2236 0.92054 0.964 0.036
#> GSM39109 1 0.2948 0.90659 0.948 0.052
#> GSM39110 1 0.2603 0.91393 0.956 0.044
#> GSM39111 1 0.2043 0.92394 0.968 0.032
#> GSM39112 1 0.9732 0.08844 0.596 0.404
#> GSM39113 1 0.9833 -0.00291 0.576 0.424
#> GSM39114 2 0.9358 0.68547 0.352 0.648
#> GSM39115 1 0.0000 0.94194 1.000 0.000
#> GSM39148 1 0.0376 0.94173 0.996 0.004
#> GSM39149 1 0.2043 0.91830 0.968 0.032
#> GSM39150 1 0.0376 0.94161 0.996 0.004
#> GSM39151 1 0.2043 0.91830 0.968 0.032
#> GSM39152 1 0.0000 0.94194 1.000 0.000
#> GSM39153 1 0.0000 0.94194 1.000 0.000
#> GSM39154 1 0.0000 0.94194 1.000 0.000
#> GSM39155 1 0.0000 0.94194 1.000 0.000
#> GSM39156 1 0.1843 0.92705 0.972 0.028
#> GSM39157 1 0.0000 0.94194 1.000 0.000
#> GSM39158 1 0.0376 0.94173 0.996 0.004
#> GSM39159 1 0.0376 0.94173 0.996 0.004
#> GSM39160 1 0.0376 0.94161 0.996 0.004
#> GSM39161 1 0.0672 0.94052 0.992 0.008
#> GSM39162 1 0.0376 0.94173 0.996 0.004
#> GSM39163 1 0.0000 0.94194 1.000 0.000
#> GSM39164 1 0.0000 0.94194 1.000 0.000
#> GSM39165 1 0.0000 0.94194 1.000 0.000
#> GSM39166 1 0.0376 0.94173 0.996 0.004
#> GSM39167 1 0.0000 0.94194 1.000 0.000
#> GSM39168 1 0.0376 0.94173 0.996 0.004
#> GSM39169 1 0.0000 0.94194 1.000 0.000
#> GSM39170 1 0.0376 0.94173 0.996 0.004
#> GSM39171 1 0.0376 0.94161 0.996 0.004
#> GSM39172 1 0.1843 0.92806 0.972 0.028
#> GSM39173 1 0.1633 0.93400 0.976 0.024
#> GSM39174 1 0.0000 0.94194 1.000 0.000
#> GSM39175 1 0.0000 0.94194 1.000 0.000
#> GSM39176 1 0.0000 0.94194 1.000 0.000
#> GSM39177 1 0.0376 0.94031 0.996 0.004
#> GSM39178 1 0.0376 0.94161 0.996 0.004
#> GSM39179 1 0.2043 0.91830 0.968 0.032
#> GSM39180 1 0.1633 0.93433 0.976 0.024
#> GSM39181 1 0.0376 0.94173 0.996 0.004
#> GSM39182 1 0.4298 0.87114 0.912 0.088
#> GSM39183 1 0.0376 0.94173 0.996 0.004
#> GSM39184 1 0.0000 0.94194 1.000 0.000
#> GSM39185 1 0.0672 0.94052 0.992 0.008
#> GSM39186 1 0.0000 0.94194 1.000 0.000
#> GSM39187 1 0.0672 0.94055 0.992 0.008
#> GSM39116 2 0.6801 0.87458 0.180 0.820
#> GSM39117 2 0.4161 0.84773 0.084 0.916
#> GSM39118 2 0.6343 0.87309 0.160 0.840
#> GSM39119 2 0.3431 0.84495 0.064 0.936
#> GSM39120 2 0.9977 0.38893 0.472 0.528
#> GSM39121 2 0.8144 0.83435 0.252 0.748
#> GSM39122 2 0.7950 0.84628 0.240 0.760
#> GSM39123 2 0.4161 0.84773 0.084 0.916
#> GSM39124 2 0.7815 0.85302 0.232 0.768
#> GSM39125 2 0.9710 0.58592 0.400 0.600
#> GSM39126 2 0.8608 0.79515 0.284 0.716
#> GSM39127 2 0.7219 0.87141 0.200 0.800
#> GSM39128 2 0.7950 0.84715 0.240 0.760
#> GSM39129 2 0.2043 0.82960 0.032 0.968
#> GSM39130 2 0.4161 0.84773 0.084 0.916
#> GSM39131 2 0.7453 0.86658 0.212 0.788
#> GSM39132 2 0.7299 0.86974 0.204 0.796
#> GSM39133 2 0.3879 0.84700 0.076 0.924
#> GSM39134 2 0.2603 0.83709 0.044 0.956
#> GSM39135 2 0.6887 0.87424 0.184 0.816
#> GSM39136 2 0.6531 0.87468 0.168 0.832
#> GSM39137 2 0.7745 0.85603 0.228 0.772
#> GSM39138 2 0.2043 0.82960 0.032 0.968
#> GSM39139 2 0.2043 0.82960 0.032 0.968
#> GSM39140 1 0.7602 0.65085 0.780 0.220
#> GSM39141 1 0.5946 0.78368 0.856 0.144
#> GSM39142 1 0.6048 0.77714 0.852 0.148
#> GSM39143 1 0.6048 0.77714 0.852 0.148
#> GSM39144 2 0.2043 0.82960 0.032 0.968
#> GSM39145 2 0.5842 0.87127 0.140 0.860
#> GSM39146 2 0.7299 0.87073 0.204 0.796
#> GSM39147 2 0.7453 0.86625 0.212 0.788
#> GSM39188 1 0.1843 0.92192 0.972 0.028
#> GSM39189 1 0.0672 0.94082 0.992 0.008
#> GSM39190 1 0.2423 0.91895 0.960 0.040
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM39104 1 0.1337 0.8782 0.972 0.012 0.016
#> GSM39105 1 0.0592 0.8840 0.988 0.000 0.012
#> GSM39106 1 0.3207 0.8010 0.904 0.084 0.012
#> GSM39107 1 0.6215 0.0252 0.572 0.428 0.000
#> GSM39108 1 0.2297 0.8543 0.944 0.036 0.020
#> GSM39109 1 0.3009 0.8344 0.920 0.052 0.028
#> GSM39110 1 0.2918 0.8380 0.924 0.044 0.032
#> GSM39111 1 0.2176 0.8586 0.948 0.032 0.020
#> GSM39112 1 0.6154 0.1137 0.592 0.408 0.000
#> GSM39113 1 0.6215 0.0252 0.572 0.428 0.000
#> GSM39114 2 0.5859 0.6184 0.344 0.656 0.000
#> GSM39115 1 0.0592 0.8847 0.988 0.000 0.012
#> GSM39148 1 0.0237 0.8851 0.996 0.004 0.000
#> GSM39149 3 0.6215 0.7057 0.428 0.000 0.572
#> GSM39150 1 0.0983 0.8835 0.980 0.004 0.016
#> GSM39151 3 0.5591 0.8181 0.304 0.000 0.696
#> GSM39152 1 0.1031 0.8831 0.976 0.000 0.024
#> GSM39153 1 0.0000 0.8844 1.000 0.000 0.000
#> GSM39154 1 0.0237 0.8841 0.996 0.000 0.004
#> GSM39155 1 0.0000 0.8844 1.000 0.000 0.000
#> GSM39156 1 0.1399 0.8701 0.968 0.028 0.004
#> GSM39157 1 0.0000 0.8844 1.000 0.000 0.000
#> GSM39158 1 0.0892 0.8819 0.980 0.000 0.020
#> GSM39159 1 0.1031 0.8792 0.976 0.000 0.024
#> GSM39160 1 0.0983 0.8835 0.980 0.004 0.016
#> GSM39161 1 0.1163 0.8766 0.972 0.000 0.028
#> GSM39162 1 0.0237 0.8851 0.996 0.004 0.000
#> GSM39163 1 0.0000 0.8844 1.000 0.000 0.000
#> GSM39164 1 0.0424 0.8837 0.992 0.000 0.008
#> GSM39165 1 0.0424 0.8851 0.992 0.000 0.008
#> GSM39166 1 0.0892 0.8819 0.980 0.000 0.020
#> GSM39167 1 0.0000 0.8844 1.000 0.000 0.000
#> GSM39168 1 0.0237 0.8851 0.996 0.004 0.000
#> GSM39169 1 0.0000 0.8844 1.000 0.000 0.000
#> GSM39170 1 0.0892 0.8819 0.980 0.000 0.020
#> GSM39171 1 0.0237 0.8852 0.996 0.004 0.000
#> GSM39172 1 0.2434 0.8545 0.940 0.024 0.036
#> GSM39173 1 0.3845 0.7583 0.872 0.012 0.116
#> GSM39174 1 0.0000 0.8844 1.000 0.000 0.000
#> GSM39175 1 0.0424 0.8841 0.992 0.000 0.008
#> GSM39176 1 0.0000 0.8844 1.000 0.000 0.000
#> GSM39177 1 0.3551 0.7284 0.868 0.000 0.132
#> GSM39178 1 0.0983 0.8844 0.980 0.004 0.016
#> GSM39179 3 0.5733 0.8156 0.324 0.000 0.676
#> GSM39180 1 0.3532 0.7803 0.884 0.008 0.108
#> GSM39181 1 0.0892 0.8819 0.980 0.000 0.020
#> GSM39182 1 0.3765 0.7837 0.888 0.084 0.028
#> GSM39183 1 0.0892 0.8819 0.980 0.000 0.020
#> GSM39184 1 0.0000 0.8844 1.000 0.000 0.000
#> GSM39185 1 0.1163 0.8766 0.972 0.000 0.028
#> GSM39186 1 0.0237 0.8841 0.996 0.000 0.004
#> GSM39187 1 0.0424 0.8848 0.992 0.008 0.000
#> GSM39116 2 0.4235 0.8258 0.176 0.824 0.000
#> GSM39117 2 0.2902 0.7683 0.064 0.920 0.016
#> GSM39118 2 0.4228 0.8186 0.148 0.844 0.008
#> GSM39119 2 0.2527 0.7634 0.044 0.936 0.020
#> GSM39120 2 0.6295 0.3411 0.472 0.528 0.000
#> GSM39121 2 0.5138 0.7751 0.252 0.748 0.000
#> GSM39122 2 0.4931 0.7965 0.232 0.768 0.000
#> GSM39123 2 0.2902 0.7683 0.064 0.920 0.016
#> GSM39124 2 0.4842 0.8041 0.224 0.776 0.000
#> GSM39125 2 0.6126 0.5095 0.400 0.600 0.000
#> GSM39126 2 0.5431 0.7293 0.284 0.716 0.000
#> GSM39127 2 0.4452 0.8236 0.192 0.808 0.000
#> GSM39128 2 0.4931 0.7971 0.232 0.768 0.000
#> GSM39129 2 0.0592 0.7221 0.000 0.988 0.012
#> GSM39130 2 0.2902 0.7683 0.064 0.920 0.016
#> GSM39131 2 0.4605 0.8187 0.204 0.796 0.000
#> GSM39132 2 0.4504 0.8219 0.196 0.804 0.000
#> GSM39133 2 0.2651 0.7692 0.060 0.928 0.012
#> GSM39134 2 0.1289 0.7585 0.032 0.968 0.000
#> GSM39135 2 0.4235 0.8255 0.176 0.824 0.000
#> GSM39136 2 0.4062 0.8244 0.164 0.836 0.000
#> GSM39137 2 0.4796 0.8073 0.220 0.780 0.000
#> GSM39138 2 0.0424 0.7234 0.000 0.992 0.008
#> GSM39139 2 0.0424 0.7234 0.000 0.992 0.008
#> GSM39140 1 0.4796 0.5528 0.780 0.220 0.000
#> GSM39141 1 0.3752 0.7024 0.856 0.144 0.000
#> GSM39142 1 0.3816 0.6951 0.852 0.148 0.000
#> GSM39143 1 0.3816 0.6951 0.852 0.148 0.000
#> GSM39144 2 0.0592 0.7231 0.000 0.988 0.012
#> GSM39145 2 0.3551 0.8166 0.132 0.868 0.000
#> GSM39146 2 0.4555 0.8210 0.200 0.800 0.000
#> GSM39147 2 0.4605 0.8184 0.204 0.796 0.000
#> GSM39188 3 0.6126 0.7703 0.400 0.000 0.600
#> GSM39189 1 0.1411 0.8724 0.964 0.000 0.036
#> GSM39190 1 0.5859 0.0387 0.656 0.000 0.344
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM39104 1 0.1139 0.8938 0.972 0.012 0.008 0.008
#> GSM39105 1 0.0672 0.8979 0.984 0.000 0.008 0.008
#> GSM39106 1 0.2803 0.8276 0.900 0.080 0.008 0.012
#> GSM39107 1 0.5220 0.0341 0.568 0.424 0.000 0.008
#> GSM39108 1 0.2019 0.8718 0.940 0.032 0.004 0.024
#> GSM39109 1 0.2825 0.8497 0.908 0.048 0.008 0.036
#> GSM39110 1 0.2686 0.8552 0.916 0.040 0.012 0.032
#> GSM39111 1 0.1920 0.8754 0.944 0.028 0.004 0.024
#> GSM39112 1 0.5172 0.1208 0.588 0.404 0.000 0.008
#> GSM39113 1 0.5220 0.0341 0.568 0.424 0.000 0.008
#> GSM39114 2 0.4936 0.6264 0.340 0.652 0.000 0.008
#> GSM39115 1 0.0657 0.8988 0.984 0.000 0.004 0.012
#> GSM39148 1 0.0188 0.8995 0.996 0.004 0.000 0.000
#> GSM39149 4 0.7258 -0.2501 0.164 0.000 0.328 0.508
#> GSM39150 1 0.0859 0.8982 0.980 0.004 0.008 0.008
#> GSM39151 3 0.3523 0.5540 0.112 0.000 0.856 0.032
#> GSM39152 1 0.1059 0.8966 0.972 0.000 0.012 0.016
#> GSM39153 1 0.0000 0.8987 1.000 0.000 0.000 0.000
#> GSM39154 1 0.0188 0.8986 0.996 0.000 0.000 0.004
#> GSM39155 1 0.0000 0.8987 1.000 0.000 0.000 0.000
#> GSM39156 1 0.1256 0.8852 0.964 0.028 0.000 0.008
#> GSM39157 1 0.0000 0.8987 1.000 0.000 0.000 0.000
#> GSM39158 1 0.0804 0.8965 0.980 0.000 0.008 0.012
#> GSM39159 1 0.0895 0.8943 0.976 0.000 0.004 0.020
#> GSM39160 1 0.0859 0.8982 0.980 0.004 0.008 0.008
#> GSM39161 1 0.1109 0.8899 0.968 0.000 0.004 0.028
#> GSM39162 1 0.0188 0.8995 0.996 0.004 0.000 0.000
#> GSM39163 1 0.0000 0.8987 1.000 0.000 0.000 0.000
#> GSM39164 1 0.0376 0.8983 0.992 0.000 0.004 0.004
#> GSM39165 1 0.0336 0.8995 0.992 0.000 0.008 0.000
#> GSM39166 1 0.0804 0.8965 0.980 0.000 0.008 0.012
#> GSM39167 1 0.0000 0.8987 1.000 0.000 0.000 0.000
#> GSM39168 1 0.0188 0.8995 0.996 0.004 0.000 0.000
#> GSM39169 1 0.0000 0.8987 1.000 0.000 0.000 0.000
#> GSM39170 1 0.0804 0.8965 0.980 0.000 0.008 0.012
#> GSM39171 1 0.0188 0.8994 0.996 0.004 0.000 0.000
#> GSM39172 1 0.2207 0.8661 0.932 0.024 0.004 0.040
#> GSM39173 1 0.5011 0.5490 0.748 0.004 0.040 0.208
#> GSM39174 1 0.0000 0.8987 1.000 0.000 0.000 0.000
#> GSM39175 1 0.0376 0.8985 0.992 0.000 0.004 0.004
#> GSM39176 1 0.0000 0.8987 1.000 0.000 0.000 0.000
#> GSM39177 1 0.3812 0.7172 0.832 0.000 0.140 0.028
#> GSM39178 1 0.0859 0.8988 0.980 0.004 0.008 0.008
#> GSM39179 3 0.5256 0.3581 0.204 0.000 0.732 0.064
#> GSM39180 1 0.4260 0.6545 0.792 0.008 0.012 0.188
#> GSM39181 1 0.0804 0.8965 0.980 0.000 0.008 0.012
#> GSM39182 1 0.3342 0.8084 0.880 0.080 0.008 0.032
#> GSM39183 1 0.0804 0.8965 0.980 0.000 0.008 0.012
#> GSM39184 1 0.0000 0.8987 1.000 0.000 0.000 0.000
#> GSM39185 1 0.1109 0.8899 0.968 0.000 0.004 0.028
#> GSM39186 1 0.0188 0.8986 0.996 0.000 0.000 0.004
#> GSM39187 1 0.0336 0.8995 0.992 0.008 0.000 0.000
#> GSM39116 2 0.3681 0.8160 0.176 0.816 0.000 0.008
#> GSM39117 2 0.3009 0.7381 0.056 0.892 0.000 0.052
#> GSM39118 2 0.3547 0.8058 0.144 0.840 0.000 0.016
#> GSM39119 2 0.2505 0.7379 0.040 0.920 0.004 0.036
#> GSM39120 2 0.5285 0.3545 0.468 0.524 0.000 0.008
#> GSM39121 2 0.4252 0.7689 0.252 0.744 0.000 0.004
#> GSM39122 2 0.4088 0.7890 0.232 0.764 0.000 0.004
#> GSM39123 2 0.3009 0.7381 0.056 0.892 0.000 0.052
#> GSM39124 2 0.4018 0.7962 0.224 0.772 0.000 0.004
#> GSM39125 2 0.5150 0.5324 0.396 0.596 0.000 0.008
#> GSM39126 2 0.4594 0.7298 0.280 0.712 0.000 0.008
#> GSM39127 2 0.3710 0.8148 0.192 0.804 0.000 0.004
#> GSM39128 2 0.3907 0.7916 0.232 0.768 0.000 0.000
#> GSM39129 2 0.1722 0.6726 0.000 0.944 0.008 0.048
#> GSM39130 2 0.3009 0.7381 0.056 0.892 0.000 0.052
#> GSM39131 2 0.3831 0.8095 0.204 0.792 0.000 0.004
#> GSM39132 2 0.3569 0.8130 0.196 0.804 0.000 0.000
#> GSM39133 2 0.2844 0.7387 0.052 0.900 0.000 0.048
#> GSM39134 2 0.1833 0.7293 0.032 0.944 0.000 0.024
#> GSM39135 2 0.3356 0.8156 0.176 0.824 0.000 0.000
#> GSM39136 2 0.3402 0.8134 0.164 0.832 0.000 0.004
#> GSM39137 2 0.3982 0.7992 0.220 0.776 0.000 0.004
#> GSM39138 2 0.1576 0.6738 0.000 0.948 0.004 0.048
#> GSM39139 2 0.1661 0.6728 0.000 0.944 0.004 0.052
#> GSM39140 1 0.4086 0.6170 0.776 0.216 0.000 0.008
#> GSM39141 1 0.3249 0.7448 0.852 0.140 0.000 0.008
#> GSM39142 1 0.3300 0.7391 0.848 0.144 0.000 0.008
#> GSM39143 1 0.3300 0.7391 0.848 0.144 0.000 0.008
#> GSM39144 2 0.1661 0.6746 0.000 0.944 0.004 0.052
#> GSM39145 2 0.3606 0.8029 0.132 0.844 0.000 0.024
#> GSM39146 2 0.3791 0.8119 0.200 0.796 0.000 0.004
#> GSM39147 2 0.3831 0.8091 0.204 0.792 0.000 0.004
#> GSM39188 3 0.6738 0.3957 0.104 0.000 0.544 0.352
#> GSM39189 1 0.1610 0.8793 0.952 0.000 0.016 0.032
#> GSM39190 4 0.5339 0.2693 0.356 0.000 0.020 0.624
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM39104 1 0.0981 0.9060 0.972 0.012 0.008 0.000 0.008
#> GSM39105 1 0.0566 0.9092 0.984 0.000 0.012 0.000 0.004
#> GSM39106 1 0.2349 0.8518 0.900 0.084 0.012 0.000 0.004
#> GSM39107 1 0.4397 0.0401 0.564 0.432 0.004 0.000 0.000
#> GSM39108 1 0.1728 0.8881 0.940 0.036 0.020 0.000 0.004
#> GSM39109 1 0.2624 0.8682 0.904 0.052 0.028 0.008 0.008
#> GSM39110 1 0.2305 0.8745 0.916 0.044 0.028 0.000 0.012
#> GSM39111 1 0.1646 0.8910 0.944 0.032 0.020 0.000 0.004
#> GSM39112 1 0.4359 0.1247 0.584 0.412 0.004 0.000 0.000
#> GSM39113 1 0.4397 0.0401 0.564 0.432 0.004 0.000 0.000
#> GSM39114 2 0.4118 0.6320 0.336 0.660 0.004 0.000 0.000
#> GSM39115 1 0.0510 0.9099 0.984 0.000 0.016 0.000 0.000
#> GSM39148 1 0.0162 0.9103 0.996 0.004 0.000 0.000 0.000
#> GSM39149 3 0.5482 0.0604 0.084 0.000 0.692 0.028 0.196
#> GSM39150 1 0.0740 0.9094 0.980 0.004 0.008 0.000 0.008
#> GSM39151 5 0.1205 0.0330 0.040 0.000 0.004 0.000 0.956
#> GSM39152 1 0.1209 0.9054 0.964 0.000 0.012 0.012 0.012
#> GSM39153 1 0.0000 0.9096 1.000 0.000 0.000 0.000 0.000
#> GSM39154 1 0.0162 0.9095 0.996 0.000 0.004 0.000 0.000
#> GSM39155 1 0.0000 0.9096 1.000 0.000 0.000 0.000 0.000
#> GSM39156 1 0.1168 0.8970 0.960 0.032 0.008 0.000 0.000
#> GSM39157 1 0.0000 0.9096 1.000 0.000 0.000 0.000 0.000
#> GSM39158 1 0.0727 0.9081 0.980 0.000 0.004 0.012 0.004
#> GSM39159 1 0.0912 0.9050 0.972 0.000 0.012 0.016 0.000
#> GSM39160 1 0.0854 0.9090 0.976 0.004 0.012 0.000 0.008
#> GSM39161 1 0.1117 0.9013 0.964 0.000 0.020 0.016 0.000
#> GSM39162 1 0.0162 0.9103 0.996 0.004 0.000 0.000 0.000
#> GSM39163 1 0.0000 0.9096 1.000 0.000 0.000 0.000 0.000
#> GSM39164 1 0.0290 0.9094 0.992 0.000 0.008 0.000 0.000
#> GSM39165 1 0.0324 0.9105 0.992 0.000 0.000 0.004 0.004
#> GSM39166 1 0.0727 0.9081 0.980 0.000 0.004 0.012 0.004
#> GSM39167 1 0.0000 0.9096 1.000 0.000 0.000 0.000 0.000
#> GSM39168 1 0.0162 0.9103 0.996 0.004 0.000 0.000 0.000
#> GSM39169 1 0.0000 0.9096 1.000 0.000 0.000 0.000 0.000
#> GSM39170 1 0.0727 0.9081 0.980 0.000 0.004 0.012 0.004
#> GSM39171 1 0.0324 0.9108 0.992 0.004 0.004 0.000 0.000
#> GSM39172 1 0.2331 0.8756 0.920 0.028 0.016 0.032 0.004
#> GSM39173 1 0.4313 0.5637 0.712 0.004 0.268 0.008 0.008
#> GSM39174 1 0.0000 0.9096 1.000 0.000 0.000 0.000 0.000
#> GSM39175 1 0.0324 0.9095 0.992 0.000 0.004 0.004 0.000
#> GSM39176 1 0.0000 0.9096 1.000 0.000 0.000 0.000 0.000
#> GSM39177 1 0.4019 0.7548 0.824 0.000 0.036 0.052 0.088
#> GSM39178 1 0.1016 0.9085 0.972 0.004 0.012 0.008 0.004
#> GSM39179 5 0.7218 0.3448 0.128 0.000 0.164 0.144 0.564
#> GSM39180 1 0.4683 0.6647 0.760 0.012 0.168 0.052 0.008
#> GSM39181 1 0.0727 0.9081 0.980 0.000 0.004 0.012 0.004
#> GSM39182 1 0.3236 0.8262 0.868 0.084 0.020 0.024 0.004
#> GSM39183 1 0.0727 0.9081 0.980 0.000 0.004 0.012 0.004
#> GSM39184 1 0.0000 0.9096 1.000 0.000 0.000 0.000 0.000
#> GSM39185 1 0.1117 0.9013 0.964 0.000 0.020 0.016 0.000
#> GSM39186 1 0.0162 0.9095 0.996 0.000 0.004 0.000 0.000
#> GSM39187 1 0.0290 0.9105 0.992 0.008 0.000 0.000 0.000
#> GSM39116 2 0.3203 0.7963 0.168 0.820 0.000 0.012 0.000
#> GSM39117 2 0.3481 0.6736 0.044 0.852 0.020 0.084 0.000
#> GSM39118 2 0.3689 0.7832 0.144 0.816 0.008 0.032 0.000
#> GSM39119 2 0.2379 0.6919 0.028 0.912 0.012 0.048 0.000
#> GSM39120 2 0.4437 0.3482 0.464 0.532 0.004 0.000 0.000
#> GSM39121 2 0.3480 0.7574 0.248 0.752 0.000 0.000 0.000
#> GSM39122 2 0.3336 0.7754 0.228 0.772 0.000 0.000 0.000
#> GSM39123 2 0.3481 0.6736 0.044 0.852 0.020 0.084 0.000
#> GSM39124 2 0.3430 0.7816 0.220 0.776 0.000 0.004 0.000
#> GSM39125 2 0.4310 0.5456 0.392 0.604 0.004 0.000 0.000
#> GSM39126 2 0.3814 0.7224 0.276 0.720 0.004 0.000 0.000
#> GSM39127 2 0.3317 0.7969 0.188 0.804 0.004 0.004 0.000
#> GSM39128 2 0.3491 0.7777 0.228 0.768 0.004 0.000 0.000
#> GSM39129 2 0.3086 0.5920 0.000 0.816 0.004 0.180 0.000
#> GSM39130 2 0.3481 0.6736 0.044 0.852 0.020 0.084 0.000
#> GSM39131 2 0.3109 0.7929 0.200 0.800 0.000 0.000 0.000
#> GSM39132 2 0.3196 0.7955 0.192 0.804 0.004 0.000 0.000
#> GSM39133 2 0.3251 0.6771 0.040 0.864 0.016 0.080 0.000
#> GSM39134 2 0.2932 0.6752 0.020 0.864 0.004 0.112 0.000
#> GSM39135 2 0.3289 0.7966 0.172 0.816 0.004 0.008 0.000
#> GSM39136 2 0.3340 0.7916 0.156 0.824 0.004 0.016 0.000
#> GSM39137 2 0.3242 0.7841 0.216 0.784 0.000 0.000 0.000
#> GSM39138 2 0.2891 0.6000 0.000 0.824 0.000 0.176 0.000
#> GSM39139 2 0.3003 0.5908 0.000 0.812 0.000 0.188 0.000
#> GSM39140 1 0.3430 0.6666 0.776 0.220 0.004 0.000 0.000
#> GSM39141 1 0.2719 0.7834 0.852 0.144 0.004 0.000 0.000
#> GSM39142 1 0.2763 0.7785 0.848 0.148 0.004 0.000 0.000
#> GSM39143 1 0.2763 0.7785 0.848 0.148 0.004 0.000 0.000
#> GSM39144 2 0.3109 0.5792 0.000 0.800 0.000 0.200 0.000
#> GSM39145 2 0.3780 0.7755 0.132 0.808 0.000 0.060 0.000
#> GSM39146 2 0.3196 0.7960 0.192 0.804 0.000 0.004 0.000
#> GSM39147 2 0.3388 0.7927 0.200 0.792 0.000 0.008 0.000
#> GSM39188 4 0.6317 0.0000 0.032 0.000 0.072 0.480 0.416
#> GSM39189 1 0.1893 0.8820 0.936 0.000 0.024 0.028 0.012
#> GSM39190 3 0.5606 0.3392 0.208 0.000 0.672 0.100 0.020
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM39104 1 0.1285 0.9062 0.960 0.012 0.008 NA 0.004 0.008
#> GSM39105 1 0.0798 0.9103 0.976 0.000 0.012 NA 0.004 0.004
#> GSM39106 1 0.2305 0.8568 0.892 0.088 0.012 NA 0.004 0.004
#> GSM39107 1 0.3961 0.0425 0.556 0.440 0.004 NA 0.000 0.000
#> GSM39108 1 0.1866 0.8912 0.932 0.036 0.016 NA 0.004 0.004
#> GSM39109 1 0.2745 0.8717 0.892 0.052 0.020 NA 0.012 0.008
#> GSM39110 1 0.2492 0.8773 0.904 0.044 0.024 NA 0.004 0.012
#> GSM39111 1 0.1792 0.8938 0.936 0.032 0.016 NA 0.004 0.004
#> GSM39112 1 0.3930 0.1222 0.576 0.420 0.004 NA 0.000 0.000
#> GSM39113 1 0.3961 0.0425 0.556 0.440 0.004 NA 0.000 0.000
#> GSM39114 2 0.3668 0.6230 0.328 0.668 0.004 NA 0.000 0.000
#> GSM39115 1 0.0653 0.9117 0.980 0.000 0.012 NA 0.004 0.000
#> GSM39148 1 0.0146 0.9119 0.996 0.004 0.000 NA 0.000 0.000
#> GSM39149 3 0.3793 0.3223 0.068 0.000 0.812 NA 0.036 0.084
#> GSM39150 1 0.1057 0.9094 0.968 0.004 0.004 NA 0.004 0.008
#> GSM39151 6 0.0146 -0.1267 0.004 0.000 0.000 NA 0.000 0.996
#> GSM39152 1 0.1344 0.9070 0.956 0.000 0.012 NA 0.012 0.008
#> GSM39153 1 0.0146 0.9116 0.996 0.000 0.000 NA 0.000 0.000
#> GSM39154 1 0.0146 0.9113 0.996 0.000 0.000 NA 0.000 0.000
#> GSM39155 1 0.0000 0.9113 1.000 0.000 0.000 NA 0.000 0.000
#> GSM39156 1 0.1409 0.9000 0.948 0.032 0.012 NA 0.000 0.000
#> GSM39157 1 0.0000 0.9113 1.000 0.000 0.000 NA 0.000 0.000
#> GSM39158 1 0.0665 0.9103 0.980 0.000 0.000 NA 0.008 0.004
#> GSM39159 1 0.1065 0.9065 0.964 0.000 0.008 NA 0.008 0.000
#> GSM39160 1 0.1171 0.9089 0.964 0.004 0.008 NA 0.004 0.008
#> GSM39161 1 0.1167 0.9031 0.960 0.000 0.012 NA 0.008 0.000
#> GSM39162 1 0.0146 0.9119 0.996 0.004 0.000 NA 0.000 0.000
#> GSM39163 1 0.0000 0.9113 1.000 0.000 0.000 NA 0.000 0.000
#> GSM39164 1 0.0551 0.9108 0.984 0.000 0.008 NA 0.004 0.000
#> GSM39165 1 0.0291 0.9122 0.992 0.000 0.004 NA 0.004 0.000
#> GSM39166 1 0.0665 0.9103 0.980 0.000 0.000 NA 0.008 0.004
#> GSM39167 1 0.0000 0.9113 1.000 0.000 0.000 NA 0.000 0.000
#> GSM39168 1 0.0146 0.9119 0.996 0.004 0.000 NA 0.000 0.000
#> GSM39169 1 0.0000 0.9113 1.000 0.000 0.000 NA 0.000 0.000
#> GSM39170 1 0.0665 0.9103 0.980 0.000 0.000 NA 0.008 0.004
#> GSM39171 1 0.0291 0.9127 0.992 0.004 0.004 NA 0.000 0.000
#> GSM39172 1 0.2619 0.8654 0.896 0.036 0.008 NA 0.012 0.004
#> GSM39173 1 0.4342 0.5791 0.692 0.004 0.260 NA 0.004 0.000
#> GSM39174 1 0.0000 0.9113 1.000 0.000 0.000 NA 0.000 0.000
#> GSM39175 1 0.0405 0.9119 0.988 0.000 0.000 NA 0.004 0.000
#> GSM39176 1 0.0000 0.9113 1.000 0.000 0.000 NA 0.000 0.000
#> GSM39177 1 0.3620 0.7769 0.824 0.000 0.032 NA 0.072 0.072
#> GSM39178 1 0.1241 0.9086 0.960 0.004 0.004 NA 0.008 0.004
#> GSM39179 6 0.8450 0.1043 0.096 0.000 0.244 NA 0.212 0.332
#> GSM39180 1 0.5132 0.6092 0.708 0.016 0.144 NA 0.024 0.000
#> GSM39181 1 0.0665 0.9103 0.980 0.000 0.000 NA 0.008 0.004
#> GSM39182 1 0.3462 0.8128 0.836 0.092 0.004 NA 0.012 0.004
#> GSM39183 1 0.0665 0.9103 0.980 0.000 0.000 NA 0.008 0.004
#> GSM39184 1 0.0000 0.9113 1.000 0.000 0.000 NA 0.000 0.000
#> GSM39185 1 0.1167 0.9031 0.960 0.000 0.012 NA 0.008 0.000
#> GSM39186 1 0.0146 0.9113 0.996 0.000 0.000 NA 0.000 0.000
#> GSM39187 1 0.0260 0.9123 0.992 0.008 0.000 NA 0.000 0.000
#> GSM39116 2 0.2981 0.7610 0.160 0.820 0.000 NA 0.000 0.000
#> GSM39117 2 0.3104 0.5711 0.016 0.824 0.004 NA 0.004 0.000
#> GSM39118 2 0.3663 0.7356 0.128 0.796 0.004 NA 0.000 0.000
#> GSM39119 2 0.2153 0.6118 0.008 0.900 0.004 NA 0.004 0.000
#> GSM39120 2 0.3979 0.3430 0.456 0.540 0.004 NA 0.000 0.000
#> GSM39121 2 0.3215 0.7326 0.240 0.756 0.000 NA 0.000 0.000
#> GSM39122 2 0.3081 0.7471 0.220 0.776 0.000 NA 0.000 0.000
#> GSM39123 2 0.3104 0.5711 0.016 0.824 0.004 NA 0.004 0.000
#> GSM39124 2 0.3230 0.7524 0.212 0.776 0.000 NA 0.000 0.000
#> GSM39125 2 0.3852 0.5321 0.384 0.612 0.004 NA 0.000 0.000
#> GSM39126 2 0.3383 0.7038 0.268 0.728 0.004 NA 0.000 0.000
#> GSM39127 2 0.2989 0.7636 0.176 0.812 0.000 NA 0.004 0.000
#> GSM39128 2 0.3221 0.7498 0.220 0.772 0.000 NA 0.004 0.000
#> GSM39129 2 0.4018 0.3758 0.000 0.580 0.000 NA 0.008 0.000
#> GSM39130 2 0.3104 0.5711 0.016 0.824 0.004 NA 0.004 0.000
#> GSM39131 2 0.2871 0.7614 0.192 0.804 0.000 NA 0.000 0.000
#> GSM39132 2 0.2805 0.7627 0.184 0.812 0.000 NA 0.004 0.000
#> GSM39133 2 0.2948 0.5802 0.016 0.840 0.004 NA 0.004 0.000
#> GSM39134 2 0.3441 0.5912 0.008 0.768 0.004 NA 0.004 0.000
#> GSM39135 2 0.3124 0.7620 0.164 0.816 0.004 NA 0.004 0.000
#> GSM39136 2 0.3271 0.7520 0.144 0.820 0.004 NA 0.004 0.000
#> GSM39137 2 0.3103 0.7545 0.208 0.784 0.000 NA 0.000 0.000
#> GSM39138 2 0.3684 0.4396 0.000 0.628 0.000 NA 0.000 0.000
#> GSM39139 2 0.3774 0.3938 0.000 0.592 0.000 NA 0.000 0.000
#> GSM39140 1 0.3109 0.6655 0.772 0.224 0.004 NA 0.000 0.000
#> GSM39141 1 0.2442 0.7909 0.852 0.144 0.004 NA 0.000 0.000
#> GSM39142 1 0.2482 0.7850 0.848 0.148 0.004 NA 0.000 0.000
#> GSM39143 1 0.2482 0.7850 0.848 0.148 0.004 NA 0.000 0.000
#> GSM39144 2 0.3937 0.3654 0.000 0.572 0.000 NA 0.004 0.000
#> GSM39145 2 0.4575 0.6979 0.124 0.696 0.000 NA 0.000 0.000
#> GSM39146 2 0.3071 0.7633 0.180 0.804 0.000 NA 0.000 0.000
#> GSM39147 2 0.3409 0.7597 0.192 0.780 0.000 NA 0.000 0.000
#> GSM39188 5 0.3816 0.0000 0.016 0.000 0.000 NA 0.688 0.296
#> GSM39189 1 0.1905 0.8869 0.932 0.000 0.020 NA 0.020 0.012
#> GSM39190 3 0.6612 0.4245 0.096 0.000 0.572 NA 0.104 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) other(p) protocol(p) k
#> MAD:hclust 83 0.251 1.04e-11 1.02e-12 2
#> MAD:hclust 82 0.187 1.48e-10 2.61e-11 3
#> MAD:hclust 79 0.237 8.65e-11 1.16e-10 4
#> MAD:hclust 78 0.201 4.63e-12 7.17e-12 5
#> MAD:hclust 74 0.281 4.43e-11 3.86e-13 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 8353 rows and 87 columns.
#> Top rows (835, 1670, 2506, 3341, 4176) 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 0.800 0.885 0.950 0.4413 0.586 0.586
#> 3 3 0.739 0.810 0.907 0.3454 0.810 0.682
#> 4 4 0.578 0.631 0.808 0.1263 0.901 0.779
#> 5 5 0.588 0.527 0.760 0.0836 0.871 0.681
#> 6 6 0.622 0.528 0.711 0.0525 0.895 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
#> GSM39104 1 0.0000 0.9331 1.000 0.000
#> GSM39105 1 0.0000 0.9331 1.000 0.000
#> GSM39106 1 0.0000 0.9331 1.000 0.000
#> GSM39107 1 0.2778 0.8978 0.952 0.048
#> GSM39108 1 0.0000 0.9331 1.000 0.000
#> GSM39109 1 0.6623 0.7878 0.828 0.172
#> GSM39110 1 0.0000 0.9331 1.000 0.000
#> GSM39111 1 0.0000 0.9331 1.000 0.000
#> GSM39112 1 0.0376 0.9305 0.996 0.004
#> GSM39113 1 0.2948 0.8945 0.948 0.052
#> GSM39114 2 0.0672 0.9717 0.008 0.992
#> GSM39115 1 0.0000 0.9331 1.000 0.000
#> GSM39148 1 0.0000 0.9331 1.000 0.000
#> GSM39149 1 0.9000 0.5890 0.684 0.316
#> GSM39150 1 0.0000 0.9331 1.000 0.000
#> GSM39151 1 0.9044 0.5819 0.680 0.320
#> GSM39152 1 0.0000 0.9331 1.000 0.000
#> GSM39153 1 0.0000 0.9331 1.000 0.000
#> GSM39154 1 0.0000 0.9331 1.000 0.000
#> GSM39155 1 0.0000 0.9331 1.000 0.000
#> GSM39156 1 0.0000 0.9331 1.000 0.000
#> GSM39157 1 0.0000 0.9331 1.000 0.000
#> GSM39158 1 0.0000 0.9331 1.000 0.000
#> GSM39159 1 0.0000 0.9331 1.000 0.000
#> GSM39160 1 0.0000 0.9331 1.000 0.000
#> GSM39161 1 0.0000 0.9331 1.000 0.000
#> GSM39162 1 0.0000 0.9331 1.000 0.000
#> GSM39163 1 0.0000 0.9331 1.000 0.000
#> GSM39164 1 0.0000 0.9331 1.000 0.000
#> GSM39165 1 0.0000 0.9331 1.000 0.000
#> GSM39166 1 0.0000 0.9331 1.000 0.000
#> GSM39167 1 0.0000 0.9331 1.000 0.000
#> GSM39168 1 0.0000 0.9331 1.000 0.000
#> GSM39169 1 0.0000 0.9331 1.000 0.000
#> GSM39170 1 0.0000 0.9331 1.000 0.000
#> GSM39171 1 0.0000 0.9331 1.000 0.000
#> GSM39172 1 0.8763 0.6203 0.704 0.296
#> GSM39173 1 0.9087 0.5751 0.676 0.324
#> GSM39174 1 0.0000 0.9331 1.000 0.000
#> GSM39175 1 0.0000 0.9331 1.000 0.000
#> GSM39176 1 0.0000 0.9331 1.000 0.000
#> GSM39177 1 0.0938 0.9255 0.988 0.012
#> GSM39178 1 0.0000 0.9331 1.000 0.000
#> GSM39179 1 0.9044 0.5827 0.680 0.320
#> GSM39180 2 0.9881 0.0992 0.436 0.564
#> GSM39181 1 0.0000 0.9331 1.000 0.000
#> GSM39182 1 0.5842 0.8183 0.860 0.140
#> GSM39183 1 0.0000 0.9331 1.000 0.000
#> GSM39184 1 0.0000 0.9331 1.000 0.000
#> GSM39185 1 0.0000 0.9331 1.000 0.000
#> GSM39186 1 0.0000 0.9331 1.000 0.000
#> GSM39187 1 0.0000 0.9331 1.000 0.000
#> GSM39116 2 0.0000 0.9790 0.000 1.000
#> GSM39117 2 0.0000 0.9790 0.000 1.000
#> GSM39118 2 0.0000 0.9790 0.000 1.000
#> GSM39119 2 0.0000 0.9790 0.000 1.000
#> GSM39120 1 0.0000 0.9331 1.000 0.000
#> GSM39121 1 0.9661 0.3855 0.608 0.392
#> GSM39122 1 0.9954 0.2092 0.540 0.460
#> GSM39123 2 0.0000 0.9790 0.000 1.000
#> GSM39124 2 0.0376 0.9756 0.004 0.996
#> GSM39125 1 0.0000 0.9331 1.000 0.000
#> GSM39126 1 0.9087 0.5399 0.676 0.324
#> GSM39127 2 0.0000 0.9790 0.000 1.000
#> GSM39128 2 0.0000 0.9790 0.000 1.000
#> GSM39129 2 0.0000 0.9790 0.000 1.000
#> GSM39130 2 0.0000 0.9790 0.000 1.000
#> GSM39131 2 0.0000 0.9790 0.000 1.000
#> GSM39132 2 0.0000 0.9790 0.000 1.000
#> GSM39133 2 0.0000 0.9790 0.000 1.000
#> GSM39134 2 0.0000 0.9790 0.000 1.000
#> GSM39135 2 0.0000 0.9790 0.000 1.000
#> GSM39136 2 0.0000 0.9790 0.000 1.000
#> GSM39137 2 0.0376 0.9756 0.004 0.996
#> GSM39138 2 0.0000 0.9790 0.000 1.000
#> GSM39139 2 0.0000 0.9790 0.000 1.000
#> GSM39140 1 0.0000 0.9331 1.000 0.000
#> GSM39141 1 0.0000 0.9331 1.000 0.000
#> GSM39142 1 0.0000 0.9331 1.000 0.000
#> GSM39143 1 0.0000 0.9331 1.000 0.000
#> GSM39144 2 0.0000 0.9790 0.000 1.000
#> GSM39145 2 0.0000 0.9790 0.000 1.000
#> GSM39146 2 0.0000 0.9790 0.000 1.000
#> GSM39147 2 0.0000 0.9790 0.000 1.000
#> GSM39188 1 0.9087 0.5751 0.676 0.324
#> GSM39189 1 0.2236 0.9088 0.964 0.036
#> GSM39190 1 0.9087 0.5751 0.676 0.324
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM39104 1 0.1860 0.86774 0.948 0.000 0.052
#> GSM39105 1 0.0000 0.89378 1.000 0.000 0.000
#> GSM39106 1 0.2096 0.87520 0.944 0.004 0.052
#> GSM39107 1 0.3669 0.82122 0.896 0.064 0.040
#> GSM39108 1 0.1643 0.87285 0.956 0.000 0.044
#> GSM39109 1 0.6934 0.35014 0.624 0.028 0.348
#> GSM39110 1 0.1964 0.87322 0.944 0.000 0.056
#> GSM39111 1 0.1753 0.87176 0.952 0.000 0.048
#> GSM39112 1 0.3572 0.82494 0.900 0.060 0.040
#> GSM39113 1 0.3947 0.81108 0.884 0.076 0.040
#> GSM39114 2 0.1950 0.91315 0.008 0.952 0.040
#> GSM39115 1 0.0000 0.89378 1.000 0.000 0.000
#> GSM39148 1 0.0000 0.89378 1.000 0.000 0.000
#> GSM39149 3 0.3192 0.90784 0.112 0.000 0.888
#> GSM39150 1 0.2066 0.86353 0.940 0.000 0.060
#> GSM39151 3 0.3192 0.90784 0.112 0.000 0.888
#> GSM39152 3 0.3752 0.88901 0.144 0.000 0.856
#> GSM39153 1 0.0237 0.89418 0.996 0.000 0.004
#> GSM39154 1 0.0237 0.89418 0.996 0.000 0.004
#> GSM39155 1 0.0237 0.89418 0.996 0.000 0.004
#> GSM39156 1 0.1031 0.88486 0.976 0.000 0.024
#> GSM39157 1 0.0237 0.89418 0.996 0.000 0.004
#> GSM39158 1 0.0237 0.89418 0.996 0.000 0.004
#> GSM39159 1 0.6062 0.17485 0.616 0.000 0.384
#> GSM39160 1 0.3116 0.81541 0.892 0.000 0.108
#> GSM39161 1 0.6295 -0.19571 0.528 0.000 0.472
#> GSM39162 1 0.0592 0.88994 0.988 0.000 0.012
#> GSM39163 1 0.0237 0.89418 0.996 0.000 0.004
#> GSM39164 1 0.0000 0.89378 1.000 0.000 0.000
#> GSM39165 1 0.5216 0.53889 0.740 0.000 0.260
#> GSM39166 1 0.0592 0.89138 0.988 0.000 0.012
#> GSM39167 1 0.0237 0.89418 0.996 0.000 0.004
#> GSM39168 1 0.0237 0.89277 0.996 0.000 0.004
#> GSM39169 1 0.0237 0.89418 0.996 0.000 0.004
#> GSM39170 1 0.0237 0.89418 0.996 0.000 0.004
#> GSM39171 1 0.1643 0.87510 0.956 0.000 0.044
#> GSM39172 3 0.3192 0.90712 0.112 0.000 0.888
#> GSM39173 3 0.3272 0.90170 0.104 0.004 0.892
#> GSM39174 1 0.0237 0.89418 0.996 0.000 0.004
#> GSM39175 1 0.0237 0.89418 0.996 0.000 0.004
#> GSM39176 1 0.0237 0.89418 0.996 0.000 0.004
#> GSM39177 3 0.3551 0.89811 0.132 0.000 0.868
#> GSM39178 1 0.6260 -0.03381 0.552 0.000 0.448
#> GSM39179 3 0.3116 0.90554 0.108 0.000 0.892
#> GSM39180 3 0.2749 0.85245 0.064 0.012 0.924
#> GSM39181 1 0.0592 0.89138 0.988 0.000 0.012
#> GSM39182 3 0.6260 0.33436 0.448 0.000 0.552
#> GSM39183 1 0.0592 0.89138 0.988 0.000 0.012
#> GSM39184 1 0.0237 0.89418 0.996 0.000 0.004
#> GSM39185 3 0.6308 0.22597 0.492 0.000 0.508
#> GSM39186 1 0.0237 0.89418 0.996 0.000 0.004
#> GSM39187 1 0.0000 0.89378 1.000 0.000 0.000
#> GSM39116 2 0.0237 0.92529 0.000 0.996 0.004
#> GSM39117 2 0.3752 0.87404 0.000 0.856 0.144
#> GSM39118 2 0.1860 0.92098 0.000 0.948 0.052
#> GSM39119 2 0.2261 0.91617 0.000 0.932 0.068
#> GSM39120 1 0.3155 0.84090 0.916 0.044 0.040
#> GSM39121 1 0.7578 0.09533 0.500 0.460 0.040
#> GSM39122 2 0.7578 -0.00285 0.460 0.500 0.040
#> GSM39123 2 0.3752 0.87404 0.000 0.856 0.144
#> GSM39124 2 0.1647 0.91734 0.004 0.960 0.036
#> GSM39125 1 0.2926 0.84723 0.924 0.036 0.040
#> GSM39126 1 0.7464 0.28776 0.560 0.400 0.040
#> GSM39127 2 0.1289 0.92135 0.000 0.968 0.032
#> GSM39128 2 0.1647 0.91734 0.004 0.960 0.036
#> GSM39129 2 0.2261 0.91605 0.000 0.932 0.068
#> GSM39130 2 0.3752 0.87404 0.000 0.856 0.144
#> GSM39131 2 0.1529 0.91864 0.000 0.960 0.040
#> GSM39132 2 0.1289 0.92135 0.000 0.968 0.032
#> GSM39133 2 0.2959 0.90480 0.000 0.900 0.100
#> GSM39134 2 0.2261 0.91605 0.000 0.932 0.068
#> GSM39135 2 0.0237 0.92529 0.000 0.996 0.004
#> GSM39136 2 0.0424 0.92539 0.000 0.992 0.008
#> GSM39137 2 0.2116 0.90971 0.012 0.948 0.040
#> GSM39138 2 0.2261 0.91605 0.000 0.932 0.068
#> GSM39139 2 0.1529 0.92247 0.000 0.960 0.040
#> GSM39140 1 0.1399 0.88052 0.968 0.004 0.028
#> GSM39141 1 0.1399 0.88052 0.968 0.004 0.028
#> GSM39142 1 0.1163 0.88275 0.972 0.000 0.028
#> GSM39143 1 0.1647 0.87519 0.960 0.004 0.036
#> GSM39144 2 0.2261 0.91605 0.000 0.932 0.068
#> GSM39145 2 0.0747 0.92516 0.000 0.984 0.016
#> GSM39146 2 0.1163 0.92228 0.000 0.972 0.028
#> GSM39147 2 0.1411 0.91856 0.000 0.964 0.036
#> GSM39188 3 0.3116 0.90576 0.108 0.000 0.892
#> GSM39189 3 0.3267 0.90687 0.116 0.000 0.884
#> GSM39190 3 0.3192 0.90784 0.112 0.000 0.888
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM39104 1 0.4946 0.7801 0.776 0.008 0.052 0.164
#> GSM39105 1 0.2081 0.8282 0.916 0.000 0.000 0.084
#> GSM39106 1 0.6192 0.7372 0.728 0.104 0.040 0.128
#> GSM39107 1 0.6665 0.4170 0.544 0.360 0.000 0.096
#> GSM39108 1 0.4819 0.7864 0.808 0.028 0.048 0.116
#> GSM39109 1 0.9434 0.3898 0.428 0.220 0.176 0.176
#> GSM39110 1 0.5831 0.7571 0.752 0.080 0.040 0.128
#> GSM39111 1 0.4220 0.7973 0.828 0.004 0.056 0.112
#> GSM39112 1 0.6535 0.5057 0.588 0.312 0.000 0.100
#> GSM39113 1 0.6903 0.3539 0.508 0.380 0.000 0.112
#> GSM39114 2 0.2382 0.4704 0.004 0.912 0.004 0.080
#> GSM39115 1 0.1489 0.8343 0.952 0.000 0.004 0.044
#> GSM39148 1 0.0188 0.8378 0.996 0.004 0.000 0.000
#> GSM39149 3 0.1452 0.9442 0.008 0.000 0.956 0.036
#> GSM39150 1 0.5180 0.7488 0.740 0.000 0.064 0.196
#> GSM39151 3 0.1452 0.9441 0.008 0.000 0.956 0.036
#> GSM39152 3 0.2976 0.9099 0.008 0.000 0.872 0.120
#> GSM39153 1 0.0000 0.8379 1.000 0.000 0.000 0.000
#> GSM39154 1 0.0000 0.8379 1.000 0.000 0.000 0.000
#> GSM39155 1 0.0000 0.8379 1.000 0.000 0.000 0.000
#> GSM39156 1 0.2844 0.8141 0.900 0.052 0.000 0.048
#> GSM39157 1 0.0000 0.8379 1.000 0.000 0.000 0.000
#> GSM39158 1 0.3306 0.7805 0.840 0.000 0.004 0.156
#> GSM39159 1 0.6475 0.5763 0.644 0.000 0.184 0.172
#> GSM39160 1 0.5633 0.7249 0.716 0.000 0.100 0.184
#> GSM39161 1 0.7039 0.4498 0.568 0.000 0.256 0.176
#> GSM39162 1 0.0188 0.8378 0.996 0.004 0.000 0.000
#> GSM39163 1 0.0000 0.8379 1.000 0.000 0.000 0.000
#> GSM39164 1 0.0000 0.8379 1.000 0.000 0.000 0.000
#> GSM39165 1 0.3463 0.7883 0.864 0.000 0.096 0.040
#> GSM39166 1 0.3893 0.7671 0.796 0.000 0.008 0.196
#> GSM39167 1 0.0000 0.8379 1.000 0.000 0.000 0.000
#> GSM39168 1 0.0188 0.8378 0.996 0.004 0.000 0.000
#> GSM39169 1 0.0336 0.8372 0.992 0.000 0.000 0.008
#> GSM39170 1 0.3306 0.7802 0.840 0.000 0.004 0.156
#> GSM39171 1 0.2522 0.8258 0.908 0.000 0.016 0.076
#> GSM39172 3 0.2654 0.9197 0.004 0.000 0.888 0.108
#> GSM39173 3 0.1151 0.9482 0.008 0.000 0.968 0.024
#> GSM39174 1 0.0000 0.8379 1.000 0.000 0.000 0.000
#> GSM39175 1 0.0336 0.8376 0.992 0.000 0.000 0.008
#> GSM39176 1 0.0000 0.8379 1.000 0.000 0.000 0.000
#> GSM39177 3 0.1722 0.9481 0.008 0.000 0.944 0.048
#> GSM39178 1 0.7480 0.4333 0.504 0.000 0.248 0.248
#> GSM39179 3 0.1452 0.9441 0.008 0.000 0.956 0.036
#> GSM39180 3 0.1940 0.9341 0.000 0.000 0.924 0.076
#> GSM39181 1 0.3545 0.7735 0.828 0.000 0.008 0.164
#> GSM39182 1 0.7851 0.1045 0.412 0.004 0.364 0.220
#> GSM39183 1 0.3852 0.7677 0.808 0.000 0.012 0.180
#> GSM39184 1 0.0188 0.8376 0.996 0.000 0.000 0.004
#> GSM39185 1 0.7138 0.4251 0.552 0.000 0.268 0.180
#> GSM39186 1 0.1576 0.8335 0.948 0.000 0.004 0.048
#> GSM39187 1 0.0000 0.8379 1.000 0.000 0.000 0.000
#> GSM39116 2 0.3355 0.2748 0.000 0.836 0.004 0.160
#> GSM39117 4 0.5620 0.8053 0.000 0.416 0.024 0.560
#> GSM39118 2 0.4917 -0.4303 0.000 0.656 0.008 0.336
#> GSM39119 2 0.5268 -0.7170 0.000 0.540 0.008 0.452
#> GSM39120 1 0.5883 0.5660 0.648 0.288 0.000 0.064
#> GSM39121 2 0.5911 0.2731 0.328 0.624 0.004 0.044
#> GSM39122 2 0.5775 0.2981 0.276 0.668 0.004 0.052
#> GSM39123 4 0.5620 0.8053 0.000 0.416 0.024 0.560
#> GSM39124 2 0.0657 0.5114 0.000 0.984 0.004 0.012
#> GSM39125 1 0.5321 0.6584 0.716 0.228 0.000 0.056
#> GSM39126 2 0.6039 0.2625 0.340 0.608 0.004 0.048
#> GSM39127 2 0.1489 0.4875 0.000 0.952 0.004 0.044
#> GSM39128 2 0.0524 0.5118 0.000 0.988 0.004 0.008
#> GSM39129 4 0.5168 0.6927 0.000 0.496 0.004 0.500
#> GSM39130 4 0.5620 0.8053 0.000 0.416 0.024 0.560
#> GSM39131 2 0.0524 0.5124 0.000 0.988 0.004 0.008
#> GSM39132 2 0.1004 0.5038 0.000 0.972 0.004 0.024
#> GSM39133 4 0.5444 0.8016 0.000 0.424 0.016 0.560
#> GSM39134 2 0.5132 -0.6985 0.000 0.548 0.004 0.448
#> GSM39135 2 0.3908 0.1008 0.000 0.784 0.004 0.212
#> GSM39136 2 0.3402 0.2646 0.000 0.832 0.004 0.164
#> GSM39137 2 0.2222 0.4864 0.032 0.932 0.004 0.032
#> GSM39138 4 0.5167 0.7082 0.000 0.488 0.004 0.508
#> GSM39139 2 0.4817 -0.3992 0.000 0.612 0.000 0.388
#> GSM39140 1 0.2943 0.8067 0.892 0.076 0.000 0.032
#> GSM39141 1 0.2489 0.8145 0.912 0.068 0.000 0.020
#> GSM39142 1 0.0937 0.8352 0.976 0.012 0.000 0.012
#> GSM39143 1 0.2489 0.8145 0.912 0.068 0.000 0.020
#> GSM39144 4 0.5167 0.7082 0.000 0.488 0.004 0.508
#> GSM39145 2 0.4477 -0.0802 0.000 0.688 0.000 0.312
#> GSM39146 2 0.1807 0.4750 0.000 0.940 0.008 0.052
#> GSM39147 2 0.0524 0.5117 0.000 0.988 0.004 0.008
#> GSM39188 3 0.1545 0.9438 0.008 0.000 0.952 0.040
#> GSM39189 3 0.2859 0.9168 0.008 0.000 0.880 0.112
#> GSM39190 3 0.1042 0.9486 0.008 0.000 0.972 0.020
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM39104 1 0.5158 0.36344 0.604 0.016 0.012 0.008 0.360
#> GSM39105 1 0.3353 0.62084 0.796 0.000 0.000 0.008 0.196
#> GSM39106 1 0.6324 0.32333 0.564 0.108 0.012 0.008 0.308
#> GSM39107 2 0.6521 0.11320 0.348 0.472 0.000 0.004 0.176
#> GSM39108 1 0.4975 0.50190 0.688 0.028 0.012 0.008 0.264
#> GSM39109 5 0.8112 0.26491 0.252 0.212 0.080 0.016 0.440
#> GSM39110 1 0.5973 0.37187 0.592 0.076 0.012 0.008 0.312
#> GSM39111 1 0.4410 0.49598 0.700 0.000 0.016 0.008 0.276
#> GSM39112 2 0.6596 -0.00925 0.392 0.424 0.000 0.004 0.180
#> GSM39113 2 0.6639 0.11405 0.340 0.472 0.000 0.008 0.180
#> GSM39114 2 0.2112 0.53996 0.004 0.908 0.000 0.004 0.084
#> GSM39115 1 0.2389 0.68461 0.880 0.000 0.000 0.004 0.116
#> GSM39148 1 0.0510 0.73679 0.984 0.000 0.000 0.000 0.016
#> GSM39149 3 0.1670 0.79634 0.000 0.000 0.936 0.052 0.012
#> GSM39150 1 0.4936 0.17549 0.560 0.000 0.016 0.008 0.416
#> GSM39151 3 0.1628 0.78879 0.000 0.000 0.936 0.056 0.008
#> GSM39152 3 0.4252 0.69562 0.000 0.000 0.652 0.008 0.340
#> GSM39153 1 0.0609 0.73688 0.980 0.000 0.000 0.000 0.020
#> GSM39154 1 0.0162 0.73658 0.996 0.000 0.000 0.000 0.004
#> GSM39155 1 0.0290 0.73565 0.992 0.000 0.000 0.000 0.008
#> GSM39156 1 0.2645 0.69053 0.888 0.044 0.000 0.000 0.068
#> GSM39157 1 0.0162 0.73658 0.996 0.000 0.000 0.000 0.004
#> GSM39158 1 0.4229 0.30683 0.704 0.000 0.000 0.020 0.276
#> GSM39159 5 0.6274 0.54552 0.424 0.000 0.080 0.024 0.472
#> GSM39160 1 0.5057 0.06123 0.532 0.000 0.020 0.008 0.440
#> GSM39161 5 0.6552 0.61011 0.384 0.000 0.112 0.024 0.480
#> GSM39162 1 0.0510 0.73679 0.984 0.000 0.000 0.000 0.016
#> GSM39163 1 0.0162 0.73658 0.996 0.000 0.000 0.000 0.004
#> GSM39164 1 0.0290 0.73781 0.992 0.000 0.000 0.000 0.008
#> GSM39165 1 0.4072 0.47067 0.792 0.000 0.048 0.008 0.152
#> GSM39166 1 0.4663 0.10304 0.604 0.000 0.000 0.020 0.376
#> GSM39167 1 0.0290 0.73658 0.992 0.000 0.000 0.000 0.008
#> GSM39168 1 0.0510 0.73679 0.984 0.000 0.000 0.000 0.016
#> GSM39169 1 0.0290 0.73663 0.992 0.000 0.000 0.000 0.008
#> GSM39170 1 0.4360 0.26775 0.680 0.000 0.000 0.020 0.300
#> GSM39171 1 0.2881 0.67317 0.860 0.000 0.012 0.004 0.124
#> GSM39172 3 0.4558 0.73420 0.000 0.000 0.652 0.024 0.324
#> GSM39173 3 0.3745 0.81259 0.000 0.000 0.780 0.024 0.196
#> GSM39174 1 0.0162 0.73658 0.996 0.000 0.000 0.000 0.004
#> GSM39175 1 0.0290 0.73520 0.992 0.000 0.000 0.000 0.008
#> GSM39176 1 0.0290 0.73658 0.992 0.000 0.000 0.000 0.008
#> GSM39177 3 0.2798 0.81956 0.000 0.000 0.852 0.008 0.140
#> GSM39178 5 0.5439 0.57543 0.232 0.000 0.088 0.012 0.668
#> GSM39179 3 0.1582 0.80602 0.000 0.000 0.944 0.028 0.028
#> GSM39180 3 0.4326 0.78118 0.000 0.000 0.708 0.028 0.264
#> GSM39181 1 0.4524 0.16651 0.644 0.000 0.000 0.020 0.336
#> GSM39182 5 0.8295 0.24698 0.176 0.012 0.176 0.184 0.452
#> GSM39183 1 0.4851 0.09458 0.620 0.000 0.008 0.020 0.352
#> GSM39184 1 0.0290 0.73565 0.992 0.000 0.000 0.000 0.008
#> GSM39185 5 0.6502 0.63016 0.356 0.000 0.112 0.024 0.508
#> GSM39186 1 0.1831 0.70686 0.920 0.000 0.000 0.004 0.076
#> GSM39187 1 0.0290 0.73658 0.992 0.000 0.000 0.000 0.008
#> GSM39116 2 0.3300 0.30939 0.000 0.792 0.000 0.204 0.004
#> GSM39117 4 0.3003 0.73856 0.000 0.188 0.000 0.812 0.000
#> GSM39118 2 0.6088 -0.51207 0.000 0.492 0.000 0.380 0.128
#> GSM39119 4 0.6163 0.73334 0.000 0.300 0.000 0.536 0.164
#> GSM39120 1 0.6124 0.06437 0.460 0.412 0.000 0.000 0.128
#> GSM39121 2 0.4481 0.47276 0.232 0.720 0.000 0.000 0.048
#> GSM39122 2 0.4204 0.48987 0.196 0.756 0.000 0.000 0.048
#> GSM39123 4 0.3003 0.73856 0.000 0.188 0.000 0.812 0.000
#> GSM39124 2 0.0324 0.55923 0.000 0.992 0.000 0.004 0.004
#> GSM39125 1 0.5686 0.21076 0.552 0.356 0.000 0.000 0.092
#> GSM39126 2 0.4481 0.47283 0.232 0.720 0.000 0.000 0.048
#> GSM39127 2 0.1197 0.53617 0.000 0.952 0.000 0.048 0.000
#> GSM39128 2 0.0404 0.55727 0.000 0.988 0.000 0.012 0.000
#> GSM39129 4 0.6444 0.68294 0.000 0.308 0.000 0.488 0.204
#> GSM39130 4 0.3003 0.73856 0.000 0.188 0.000 0.812 0.000
#> GSM39131 2 0.0693 0.55851 0.000 0.980 0.000 0.012 0.008
#> GSM39132 2 0.1041 0.54507 0.000 0.964 0.000 0.032 0.004
#> GSM39133 4 0.3003 0.73856 0.000 0.188 0.000 0.812 0.000
#> GSM39134 4 0.6244 0.70504 0.000 0.336 0.000 0.504 0.160
#> GSM39135 2 0.3890 0.17159 0.000 0.736 0.000 0.252 0.012
#> GSM39136 2 0.3430 0.27458 0.000 0.776 0.000 0.220 0.004
#> GSM39137 2 0.1364 0.55726 0.036 0.952 0.000 0.000 0.012
#> GSM39138 4 0.6366 0.70339 0.000 0.284 0.000 0.512 0.204
#> GSM39139 2 0.6526 -0.46632 0.000 0.452 0.000 0.344 0.204
#> GSM39140 1 0.2659 0.68304 0.888 0.060 0.000 0.000 0.052
#> GSM39141 1 0.2304 0.69900 0.908 0.044 0.000 0.000 0.048
#> GSM39142 1 0.1750 0.71686 0.936 0.028 0.000 0.000 0.036
#> GSM39143 1 0.2304 0.69900 0.908 0.044 0.000 0.000 0.048
#> GSM39144 4 0.6407 0.69587 0.000 0.296 0.000 0.500 0.204
#> GSM39145 2 0.6287 -0.30429 0.000 0.520 0.000 0.296 0.184
#> GSM39146 2 0.1478 0.52169 0.000 0.936 0.000 0.064 0.000
#> GSM39147 2 0.0771 0.55240 0.000 0.976 0.000 0.020 0.004
#> GSM39188 3 0.1626 0.79005 0.000 0.000 0.940 0.044 0.016
#> GSM39189 3 0.4585 0.70353 0.000 0.000 0.628 0.020 0.352
#> GSM39190 3 0.3531 0.82014 0.000 0.000 0.816 0.036 0.148
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM39104 5 0.4447 0.4682 0.372 0.004 0.004 0.020 0.600 0.000
#> GSM39105 1 0.4269 -0.1339 0.568 0.000 0.000 0.020 0.412 0.000
#> GSM39106 5 0.5390 0.4316 0.380 0.052 0.004 0.024 0.540 0.000
#> GSM39107 2 0.6474 0.1654 0.224 0.452 0.000 0.016 0.300 0.008
#> GSM39108 5 0.4882 0.3615 0.448 0.008 0.004 0.024 0.512 0.004
#> GSM39109 5 0.5758 0.3422 0.136 0.100 0.064 0.024 0.676 0.000
#> GSM39110 5 0.5388 0.4274 0.392 0.036 0.004 0.028 0.536 0.004
#> GSM39111 5 0.4691 0.4077 0.428 0.004 0.004 0.020 0.540 0.004
#> GSM39112 2 0.6654 0.0634 0.252 0.412 0.000 0.020 0.308 0.008
#> GSM39113 2 0.6451 0.1734 0.216 0.456 0.000 0.016 0.304 0.008
#> GSM39114 2 0.3216 0.5491 0.012 0.828 0.000 0.008 0.140 0.012
#> GSM39115 1 0.3466 0.4629 0.760 0.000 0.000 0.008 0.224 0.008
#> GSM39148 1 0.0291 0.7584 0.992 0.000 0.000 0.004 0.004 0.000
#> GSM39149 3 0.3796 0.7185 0.000 0.000 0.768 0.188 0.012 0.032
#> GSM39150 5 0.5051 0.4576 0.368 0.000 0.012 0.016 0.576 0.028
#> GSM39151 3 0.4397 0.6971 0.000 0.000 0.672 0.284 0.012 0.032
#> GSM39152 3 0.4820 0.6225 0.004 0.000 0.584 0.044 0.364 0.004
#> GSM39153 1 0.0146 0.7596 0.996 0.000 0.000 0.004 0.000 0.000
#> GSM39154 1 0.0000 0.7601 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM39155 1 0.0146 0.7602 0.996 0.000 0.000 0.000 0.004 0.000
#> GSM39156 1 0.3094 0.6246 0.856 0.032 0.000 0.012 0.092 0.008
#> GSM39157 1 0.0146 0.7602 0.996 0.000 0.000 0.000 0.004 0.000
#> GSM39158 1 0.6146 0.1278 0.556 0.000 0.000 0.056 0.260 0.128
#> GSM39159 5 0.7659 0.1898 0.360 0.000 0.072 0.064 0.368 0.136
#> GSM39160 5 0.5311 0.4758 0.352 0.000 0.028 0.016 0.576 0.028
#> GSM39161 5 0.7889 0.2567 0.312 0.000 0.104 0.064 0.384 0.136
#> GSM39162 1 0.0291 0.7584 0.992 0.000 0.000 0.004 0.004 0.000
#> GSM39163 1 0.0146 0.7602 0.996 0.000 0.000 0.000 0.004 0.000
#> GSM39164 1 0.0146 0.7593 0.996 0.000 0.000 0.000 0.004 0.000
#> GSM39165 1 0.2457 0.6469 0.880 0.000 0.036 0.000 0.084 0.000
#> GSM39166 1 0.6671 -0.0908 0.452 0.000 0.004 0.064 0.344 0.136
#> GSM39167 1 0.0146 0.7602 0.996 0.000 0.000 0.000 0.004 0.000
#> GSM39168 1 0.0291 0.7584 0.992 0.000 0.000 0.004 0.004 0.000
#> GSM39169 1 0.0891 0.7480 0.968 0.000 0.000 0.000 0.024 0.008
#> GSM39170 1 0.6355 0.0768 0.528 0.000 0.000 0.064 0.272 0.136
#> GSM39171 1 0.3604 0.4409 0.760 0.000 0.000 0.012 0.216 0.012
#> GSM39172 3 0.4253 0.6980 0.004 0.000 0.680 0.016 0.288 0.012
#> GSM39173 3 0.3058 0.7622 0.000 0.000 0.848 0.024 0.108 0.020
#> GSM39174 1 0.0000 0.7601 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM39175 1 0.0000 0.7601 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM39176 1 0.0146 0.7602 0.996 0.000 0.000 0.000 0.004 0.000
#> GSM39177 3 0.3782 0.7683 0.000 0.000 0.780 0.124 0.096 0.000
#> GSM39178 5 0.5280 0.2566 0.116 0.000 0.068 0.016 0.716 0.084
#> GSM39179 3 0.3328 0.7444 0.000 0.000 0.788 0.192 0.008 0.012
#> GSM39180 3 0.3770 0.7366 0.000 0.000 0.760 0.012 0.204 0.024
#> GSM39181 1 0.6514 0.0492 0.516 0.000 0.004 0.064 0.280 0.136
#> GSM39182 5 0.8039 -0.1635 0.108 0.012 0.244 0.224 0.384 0.028
#> GSM39183 1 0.6686 -0.0119 0.488 0.000 0.008 0.064 0.304 0.136
#> GSM39184 1 0.0146 0.7602 0.996 0.000 0.000 0.000 0.004 0.000
#> GSM39185 5 0.7902 0.2700 0.300 0.000 0.108 0.064 0.392 0.136
#> GSM39186 1 0.2416 0.5865 0.844 0.000 0.000 0.000 0.156 0.000
#> GSM39187 1 0.0146 0.7602 0.996 0.000 0.000 0.000 0.004 0.000
#> GSM39116 2 0.2933 0.4113 0.000 0.852 0.000 0.032 0.008 0.108
#> GSM39117 4 0.5204 0.9979 0.000 0.124 0.000 0.584 0.000 0.292
#> GSM39118 2 0.5070 -0.6337 0.000 0.476 0.000 0.032 0.024 0.468
#> GSM39119 6 0.5740 0.4394 0.000 0.248 0.000 0.144 0.024 0.584
#> GSM39120 2 0.6417 0.0368 0.388 0.404 0.000 0.016 0.184 0.008
#> GSM39121 2 0.4422 0.5304 0.180 0.736 0.000 0.004 0.068 0.012
#> GSM39122 2 0.4309 0.5396 0.160 0.752 0.000 0.004 0.072 0.012
#> GSM39123 4 0.5204 0.9979 0.000 0.124 0.000 0.584 0.000 0.292
#> GSM39124 2 0.0363 0.5758 0.000 0.988 0.000 0.000 0.000 0.012
#> GSM39125 1 0.6003 -0.0344 0.492 0.356 0.000 0.012 0.132 0.008
#> GSM39126 2 0.4731 0.5238 0.180 0.716 0.000 0.008 0.084 0.012
#> GSM39127 2 0.1168 0.5576 0.000 0.956 0.000 0.016 0.000 0.028
#> GSM39128 2 0.0363 0.5753 0.000 0.988 0.000 0.000 0.000 0.012
#> GSM39129 6 0.3463 0.7470 0.000 0.240 0.000 0.004 0.008 0.748
#> GSM39130 4 0.5204 0.9979 0.000 0.124 0.000 0.584 0.000 0.292
#> GSM39131 2 0.0820 0.5687 0.000 0.972 0.000 0.012 0.000 0.016
#> GSM39132 2 0.0909 0.5660 0.000 0.968 0.000 0.012 0.000 0.020
#> GSM39133 4 0.5337 0.9937 0.000 0.124 0.000 0.580 0.004 0.292
#> GSM39134 6 0.4665 0.6600 0.000 0.272 0.000 0.060 0.008 0.660
#> GSM39135 2 0.3534 0.2810 0.000 0.796 0.000 0.036 0.008 0.160
#> GSM39136 2 0.3165 0.3822 0.000 0.836 0.000 0.040 0.008 0.116
#> GSM39137 2 0.1219 0.5779 0.048 0.948 0.000 0.000 0.000 0.004
#> GSM39138 6 0.3081 0.7395 0.000 0.220 0.000 0.004 0.000 0.776
#> GSM39139 6 0.3769 0.6856 0.000 0.356 0.000 0.004 0.000 0.640
#> GSM39140 1 0.2089 0.7058 0.920 0.032 0.000 0.008 0.032 0.008
#> GSM39141 1 0.1936 0.7139 0.928 0.028 0.000 0.008 0.028 0.008
#> GSM39142 1 0.1936 0.7139 0.928 0.028 0.000 0.008 0.028 0.008
#> GSM39143 1 0.1936 0.7139 0.928 0.028 0.000 0.008 0.028 0.008
#> GSM39144 6 0.3190 0.7403 0.000 0.220 0.000 0.008 0.000 0.772
#> GSM39145 6 0.3982 0.5723 0.000 0.460 0.000 0.004 0.000 0.536
#> GSM39146 2 0.1464 0.5475 0.000 0.944 0.000 0.016 0.004 0.036
#> GSM39147 2 0.1075 0.5603 0.000 0.952 0.000 0.000 0.000 0.048
#> GSM39188 3 0.4695 0.7013 0.000 0.000 0.696 0.224 0.028 0.052
#> GSM39189 3 0.4353 0.6614 0.004 0.000 0.640 0.012 0.332 0.012
#> GSM39190 3 0.3508 0.7613 0.000 0.000 0.828 0.060 0.088 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) other(p) protocol(p) k
#> MAD:kmeans 84 0.1832 7.77e-10 1.25e-09 2
#> MAD:kmeans 78 0.0345 3.35e-09 3.56e-09 3
#> MAD:kmeans 65 0.1721 2.60e-05 7.02e-07 4
#> MAD:kmeans 59 0.5672 1.46e-04 1.67e-06 5
#> MAD:kmeans 56 0.4433 7.24e-05 8.45e-07 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 8353 rows and 87 columns.
#> Top rows (835, 1670, 2506, 3341, 4176) 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 0.820 0.893 0.954 0.4990 0.500 0.500
#> 3 3 0.609 0.763 0.885 0.3206 0.785 0.595
#> 4 4 0.539 0.609 0.782 0.1277 0.855 0.617
#> 5 5 0.559 0.542 0.712 0.0672 0.898 0.642
#> 6 6 0.580 0.490 0.675 0.0408 0.909 0.617
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
#> GSM39104 1 0.0000 0.960 1.000 0.000
#> GSM39105 1 0.0000 0.960 1.000 0.000
#> GSM39106 1 0.1633 0.943 0.976 0.024
#> GSM39107 1 0.5294 0.850 0.880 0.120
#> GSM39108 1 0.0000 0.960 1.000 0.000
#> GSM39109 2 0.0672 0.931 0.008 0.992
#> GSM39110 1 0.0000 0.960 1.000 0.000
#> GSM39111 1 0.0000 0.960 1.000 0.000
#> GSM39112 1 0.3733 0.900 0.928 0.072
#> GSM39113 1 0.8267 0.648 0.740 0.260
#> GSM39114 2 0.0000 0.935 0.000 1.000
#> GSM39115 1 0.0000 0.960 1.000 0.000
#> GSM39148 1 0.0000 0.960 1.000 0.000
#> GSM39149 2 0.5178 0.855 0.116 0.884
#> GSM39150 1 0.0000 0.960 1.000 0.000
#> GSM39151 2 0.6801 0.786 0.180 0.820
#> GSM39152 1 0.7950 0.663 0.760 0.240
#> GSM39153 1 0.0000 0.960 1.000 0.000
#> GSM39154 1 0.0000 0.960 1.000 0.000
#> GSM39155 1 0.0000 0.960 1.000 0.000
#> GSM39156 1 0.0000 0.960 1.000 0.000
#> GSM39157 1 0.0000 0.960 1.000 0.000
#> GSM39158 1 0.0000 0.960 1.000 0.000
#> GSM39159 1 0.3274 0.909 0.940 0.060
#> GSM39160 1 0.0000 0.960 1.000 0.000
#> GSM39161 1 0.9170 0.470 0.668 0.332
#> GSM39162 1 0.0000 0.960 1.000 0.000
#> GSM39163 1 0.0000 0.960 1.000 0.000
#> GSM39164 1 0.0000 0.960 1.000 0.000
#> GSM39165 1 0.0000 0.960 1.000 0.000
#> GSM39166 1 0.0000 0.960 1.000 0.000
#> GSM39167 1 0.0000 0.960 1.000 0.000
#> GSM39168 1 0.0000 0.960 1.000 0.000
#> GSM39169 1 0.0000 0.960 1.000 0.000
#> GSM39170 1 0.0000 0.960 1.000 0.000
#> GSM39171 1 0.0000 0.960 1.000 0.000
#> GSM39172 2 0.2423 0.914 0.040 0.960
#> GSM39173 2 0.3431 0.898 0.064 0.936
#> GSM39174 1 0.0000 0.960 1.000 0.000
#> GSM39175 1 0.0000 0.960 1.000 0.000
#> GSM39176 1 0.0000 0.960 1.000 0.000
#> GSM39177 2 0.9896 0.257 0.440 0.560
#> GSM39178 1 0.0000 0.960 1.000 0.000
#> GSM39179 2 0.0672 0.932 0.008 0.992
#> GSM39180 2 0.0000 0.935 0.000 1.000
#> GSM39181 1 0.0000 0.960 1.000 0.000
#> GSM39182 2 0.2778 0.910 0.048 0.952
#> GSM39183 1 0.0000 0.960 1.000 0.000
#> GSM39184 1 0.0000 0.960 1.000 0.000
#> GSM39185 1 0.9866 0.187 0.568 0.432
#> GSM39186 1 0.0000 0.960 1.000 0.000
#> GSM39187 1 0.0000 0.960 1.000 0.000
#> GSM39116 2 0.0000 0.935 0.000 1.000
#> GSM39117 2 0.0000 0.935 0.000 1.000
#> GSM39118 2 0.0000 0.935 0.000 1.000
#> GSM39119 2 0.0000 0.935 0.000 1.000
#> GSM39120 1 0.3879 0.896 0.924 0.076
#> GSM39121 2 0.9491 0.423 0.368 0.632
#> GSM39122 2 0.9358 0.461 0.352 0.648
#> GSM39123 2 0.0000 0.935 0.000 1.000
#> GSM39124 2 0.0000 0.935 0.000 1.000
#> GSM39125 1 0.4562 0.876 0.904 0.096
#> GSM39126 2 0.7602 0.708 0.220 0.780
#> GSM39127 2 0.0000 0.935 0.000 1.000
#> GSM39128 2 0.0000 0.935 0.000 1.000
#> GSM39129 2 0.0000 0.935 0.000 1.000
#> GSM39130 2 0.0000 0.935 0.000 1.000
#> GSM39131 2 0.0000 0.935 0.000 1.000
#> GSM39132 2 0.0000 0.935 0.000 1.000
#> GSM39133 2 0.0000 0.935 0.000 1.000
#> GSM39134 2 0.0000 0.935 0.000 1.000
#> GSM39135 2 0.0000 0.935 0.000 1.000
#> GSM39136 2 0.0000 0.935 0.000 1.000
#> GSM39137 2 0.0000 0.935 0.000 1.000
#> GSM39138 2 0.0000 0.935 0.000 1.000
#> GSM39139 2 0.0000 0.935 0.000 1.000
#> GSM39140 1 0.0672 0.954 0.992 0.008
#> GSM39141 1 0.0000 0.960 1.000 0.000
#> GSM39142 1 0.0000 0.960 1.000 0.000
#> GSM39143 1 0.0000 0.960 1.000 0.000
#> GSM39144 2 0.0000 0.935 0.000 1.000
#> GSM39145 2 0.0000 0.935 0.000 1.000
#> GSM39146 2 0.0000 0.935 0.000 1.000
#> GSM39147 2 0.0000 0.935 0.000 1.000
#> GSM39188 2 0.4690 0.869 0.100 0.900
#> GSM39189 2 0.8327 0.666 0.264 0.736
#> GSM39190 2 0.4431 0.876 0.092 0.908
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM39104 1 0.6204 0.2755 0.576 0.000 0.424
#> GSM39105 1 0.1163 0.8455 0.972 0.000 0.028
#> GSM39106 1 0.7596 0.5963 0.672 0.100 0.228
#> GSM39107 1 0.6140 0.3789 0.596 0.404 0.000
#> GSM39108 1 0.4539 0.7530 0.836 0.016 0.148
#> GSM39109 3 0.6589 0.5134 0.032 0.280 0.688
#> GSM39110 1 0.8056 0.2552 0.532 0.068 0.400
#> GSM39111 1 0.6305 0.0604 0.516 0.000 0.484
#> GSM39112 1 0.4733 0.7176 0.800 0.196 0.004
#> GSM39113 2 0.6468 0.0677 0.444 0.552 0.004
#> GSM39114 2 0.0000 0.8856 0.000 1.000 0.000
#> GSM39115 1 0.0000 0.8533 1.000 0.000 0.000
#> GSM39148 1 0.0000 0.8533 1.000 0.000 0.000
#> GSM39149 3 0.0000 0.8599 0.000 0.000 1.000
#> GSM39150 3 0.6095 0.3365 0.392 0.000 0.608
#> GSM39151 3 0.0000 0.8599 0.000 0.000 1.000
#> GSM39152 3 0.1031 0.8561 0.024 0.000 0.976
#> GSM39153 1 0.0237 0.8530 0.996 0.000 0.004
#> GSM39154 1 0.0424 0.8525 0.992 0.000 0.008
#> GSM39155 1 0.0000 0.8533 1.000 0.000 0.000
#> GSM39156 1 0.1399 0.8457 0.968 0.028 0.004
#> GSM39157 1 0.0000 0.8533 1.000 0.000 0.000
#> GSM39158 1 0.2959 0.7951 0.900 0.000 0.100
#> GSM39159 3 0.5291 0.6442 0.268 0.000 0.732
#> GSM39160 3 0.5327 0.6128 0.272 0.000 0.728
#> GSM39161 3 0.4654 0.7242 0.208 0.000 0.792
#> GSM39162 1 0.0000 0.8533 1.000 0.000 0.000
#> GSM39163 1 0.0000 0.8533 1.000 0.000 0.000
#> GSM39164 1 0.0000 0.8533 1.000 0.000 0.000
#> GSM39165 3 0.6260 0.2426 0.448 0.000 0.552
#> GSM39166 1 0.5650 0.5097 0.688 0.000 0.312
#> GSM39167 1 0.0000 0.8533 1.000 0.000 0.000
#> GSM39168 1 0.0000 0.8533 1.000 0.000 0.000
#> GSM39169 1 0.0892 0.8491 0.980 0.000 0.020
#> GSM39170 1 0.3412 0.7757 0.876 0.000 0.124
#> GSM39171 1 0.5706 0.5296 0.680 0.000 0.320
#> GSM39172 3 0.0237 0.8584 0.000 0.004 0.996
#> GSM39173 3 0.1289 0.8452 0.000 0.032 0.968
#> GSM39174 1 0.0000 0.8533 1.000 0.000 0.000
#> GSM39175 1 0.3192 0.7960 0.888 0.000 0.112
#> GSM39176 1 0.0000 0.8533 1.000 0.000 0.000
#> GSM39177 3 0.0892 0.8582 0.020 0.000 0.980
#> GSM39178 3 0.3482 0.7879 0.128 0.000 0.872
#> GSM39179 3 0.0000 0.8599 0.000 0.000 1.000
#> GSM39180 3 0.1753 0.8315 0.000 0.048 0.952
#> GSM39181 1 0.5254 0.5940 0.736 0.000 0.264
#> GSM39182 3 0.2096 0.8280 0.004 0.052 0.944
#> GSM39183 1 0.6154 0.2680 0.592 0.000 0.408
#> GSM39184 1 0.0424 0.8528 0.992 0.000 0.008
#> GSM39185 3 0.3038 0.8196 0.104 0.000 0.896
#> GSM39186 1 0.0747 0.8502 0.984 0.000 0.016
#> GSM39187 1 0.0000 0.8533 1.000 0.000 0.000
#> GSM39116 2 0.1289 0.8898 0.000 0.968 0.032
#> GSM39117 2 0.4796 0.7998 0.000 0.780 0.220
#> GSM39118 2 0.3267 0.8715 0.000 0.884 0.116
#> GSM39119 2 0.4121 0.8463 0.000 0.832 0.168
#> GSM39120 1 0.5517 0.6379 0.728 0.268 0.004
#> GSM39121 2 0.4504 0.6943 0.196 0.804 0.000
#> GSM39122 2 0.2959 0.8088 0.100 0.900 0.000
#> GSM39123 2 0.4796 0.7998 0.000 0.780 0.220
#> GSM39124 2 0.0000 0.8856 0.000 1.000 0.000
#> GSM39125 1 0.4978 0.6973 0.780 0.216 0.004
#> GSM39126 2 0.1860 0.8533 0.052 0.948 0.000
#> GSM39127 2 0.0237 0.8872 0.000 0.996 0.004
#> GSM39128 2 0.0000 0.8856 0.000 1.000 0.000
#> GSM39129 2 0.4062 0.8496 0.000 0.836 0.164
#> GSM39130 2 0.4796 0.7998 0.000 0.780 0.220
#> GSM39131 2 0.0237 0.8871 0.000 0.996 0.004
#> GSM39132 2 0.0424 0.8881 0.000 0.992 0.008
#> GSM39133 2 0.3941 0.8549 0.000 0.844 0.156
#> GSM39134 2 0.4002 0.8518 0.000 0.840 0.160
#> GSM39135 2 0.1529 0.8901 0.000 0.960 0.040
#> GSM39136 2 0.1529 0.8901 0.000 0.960 0.040
#> GSM39137 2 0.0000 0.8856 0.000 1.000 0.000
#> GSM39138 2 0.4062 0.8489 0.000 0.836 0.164
#> GSM39139 2 0.2711 0.8804 0.000 0.912 0.088
#> GSM39140 1 0.1643 0.8379 0.956 0.044 0.000
#> GSM39141 1 0.1163 0.8454 0.972 0.028 0.000
#> GSM39142 1 0.1031 0.8469 0.976 0.024 0.000
#> GSM39143 1 0.1163 0.8454 0.972 0.028 0.000
#> GSM39144 2 0.3879 0.8565 0.000 0.848 0.152
#> GSM39145 2 0.1753 0.8891 0.000 0.952 0.048
#> GSM39146 2 0.0592 0.8885 0.000 0.988 0.012
#> GSM39147 2 0.0424 0.8881 0.000 0.992 0.008
#> GSM39188 3 0.0000 0.8599 0.000 0.000 1.000
#> GSM39189 3 0.0000 0.8599 0.000 0.000 1.000
#> GSM39190 3 0.0000 0.8599 0.000 0.000 1.000
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM39104 4 0.7205 0.2707 0.200 0.000 0.252 0.548
#> GSM39105 1 0.6454 0.3422 0.544 0.000 0.076 0.380
#> GSM39106 4 0.6039 0.5168 0.140 0.020 0.116 0.724
#> GSM39107 4 0.3948 0.5764 0.096 0.064 0.000 0.840
#> GSM39108 4 0.7042 0.1585 0.352 0.000 0.132 0.516
#> GSM39109 4 0.7928 -0.0133 0.008 0.208 0.380 0.404
#> GSM39110 4 0.6877 0.3975 0.156 0.004 0.232 0.608
#> GSM39111 4 0.7808 0.1091 0.256 0.000 0.344 0.400
#> GSM39112 4 0.3876 0.5576 0.124 0.040 0.000 0.836
#> GSM39113 4 0.3810 0.5808 0.060 0.092 0.000 0.848
#> GSM39114 4 0.4843 0.1198 0.000 0.396 0.000 0.604
#> GSM39115 1 0.4744 0.6017 0.704 0.000 0.012 0.284
#> GSM39148 1 0.1474 0.7628 0.948 0.000 0.000 0.052
#> GSM39149 3 0.1913 0.7608 0.000 0.040 0.940 0.020
#> GSM39150 4 0.7880 -0.0119 0.284 0.000 0.344 0.372
#> GSM39151 3 0.1545 0.7608 0.000 0.040 0.952 0.008
#> GSM39152 3 0.2469 0.7047 0.000 0.000 0.892 0.108
#> GSM39153 1 0.1975 0.7716 0.936 0.000 0.016 0.048
#> GSM39154 1 0.1624 0.7726 0.952 0.000 0.020 0.028
#> GSM39155 1 0.1557 0.7688 0.944 0.000 0.000 0.056
#> GSM39156 1 0.4936 0.5290 0.700 0.000 0.020 0.280
#> GSM39157 1 0.0592 0.7710 0.984 0.000 0.000 0.016
#> GSM39158 1 0.4804 0.6784 0.776 0.000 0.064 0.160
#> GSM39159 3 0.7008 0.3351 0.340 0.004 0.540 0.116
#> GSM39160 3 0.7437 0.2352 0.248 0.000 0.512 0.240
#> GSM39161 3 0.7400 0.4170 0.296 0.016 0.552 0.136
#> GSM39162 1 0.1637 0.7597 0.940 0.000 0.000 0.060
#> GSM39163 1 0.1004 0.7719 0.972 0.000 0.004 0.024
#> GSM39164 1 0.1824 0.7733 0.936 0.000 0.004 0.060
#> GSM39165 1 0.6305 0.1448 0.516 0.000 0.424 0.060
#> GSM39166 1 0.7346 0.3758 0.520 0.000 0.200 0.280
#> GSM39167 1 0.0817 0.7692 0.976 0.000 0.000 0.024
#> GSM39168 1 0.1557 0.7615 0.944 0.000 0.000 0.056
#> GSM39169 1 0.3143 0.7595 0.876 0.000 0.024 0.100
#> GSM39170 1 0.5142 0.6566 0.744 0.000 0.064 0.192
#> GSM39171 1 0.7310 0.3476 0.532 0.000 0.256 0.212
#> GSM39172 3 0.2918 0.7435 0.000 0.116 0.876 0.008
#> GSM39173 3 0.3612 0.7232 0.004 0.144 0.840 0.012
#> GSM39174 1 0.1284 0.7732 0.964 0.000 0.012 0.024
#> GSM39175 1 0.2943 0.7487 0.892 0.000 0.076 0.032
#> GSM39176 1 0.1022 0.7688 0.968 0.000 0.000 0.032
#> GSM39177 3 0.2291 0.7504 0.016 0.016 0.932 0.036
#> GSM39178 3 0.5184 0.5923 0.056 0.000 0.732 0.212
#> GSM39179 3 0.2125 0.7560 0.000 0.076 0.920 0.004
#> GSM39180 3 0.4678 0.6422 0.000 0.232 0.744 0.024
#> GSM39181 1 0.5807 0.6128 0.708 0.000 0.132 0.160
#> GSM39182 3 0.6165 0.5606 0.032 0.268 0.664 0.036
#> GSM39183 1 0.7538 0.3259 0.492 0.000 0.248 0.260
#> GSM39184 1 0.2670 0.7563 0.904 0.000 0.024 0.072
#> GSM39185 3 0.6962 0.5349 0.184 0.020 0.640 0.156
#> GSM39186 1 0.4500 0.6791 0.776 0.000 0.032 0.192
#> GSM39187 1 0.1474 0.7692 0.948 0.000 0.000 0.052
#> GSM39116 2 0.1824 0.8292 0.000 0.936 0.004 0.060
#> GSM39117 2 0.3649 0.7076 0.000 0.796 0.204 0.000
#> GSM39118 2 0.1389 0.8265 0.000 0.952 0.048 0.000
#> GSM39119 2 0.2469 0.7964 0.000 0.892 0.108 0.000
#> GSM39120 4 0.5090 0.4967 0.228 0.044 0.000 0.728
#> GSM39121 4 0.6748 0.2776 0.112 0.328 0.000 0.560
#> GSM39122 4 0.5888 0.0453 0.036 0.424 0.000 0.540
#> GSM39123 2 0.3486 0.7262 0.000 0.812 0.188 0.000
#> GSM39124 2 0.4277 0.6555 0.000 0.720 0.000 0.280
#> GSM39125 4 0.5959 0.3502 0.336 0.032 0.012 0.620
#> GSM39126 4 0.6277 0.2193 0.068 0.360 0.000 0.572
#> GSM39127 2 0.3444 0.7623 0.000 0.816 0.000 0.184
#> GSM39128 2 0.4072 0.6922 0.000 0.748 0.000 0.252
#> GSM39129 2 0.2081 0.8138 0.000 0.916 0.084 0.000
#> GSM39130 2 0.3486 0.7262 0.000 0.812 0.188 0.000
#> GSM39131 2 0.4040 0.6998 0.000 0.752 0.000 0.248
#> GSM39132 2 0.3356 0.7682 0.000 0.824 0.000 0.176
#> GSM39133 2 0.2611 0.8070 0.000 0.896 0.096 0.008
#> GSM39134 2 0.1807 0.8256 0.000 0.940 0.052 0.008
#> GSM39135 2 0.1557 0.8289 0.000 0.944 0.000 0.056
#> GSM39136 2 0.1824 0.8296 0.000 0.936 0.004 0.060
#> GSM39137 2 0.4720 0.5786 0.004 0.672 0.000 0.324
#> GSM39138 2 0.2149 0.8090 0.000 0.912 0.088 0.000
#> GSM39139 2 0.1706 0.8326 0.000 0.948 0.016 0.036
#> GSM39140 1 0.4643 0.4385 0.656 0.000 0.000 0.344
#> GSM39141 1 0.3975 0.5976 0.760 0.000 0.000 0.240
#> GSM39142 1 0.3726 0.6400 0.788 0.000 0.000 0.212
#> GSM39143 1 0.4193 0.5621 0.732 0.000 0.000 0.268
#> GSM39144 2 0.2101 0.8243 0.000 0.928 0.060 0.012
#> GSM39145 2 0.2021 0.8318 0.000 0.932 0.012 0.056
#> GSM39146 2 0.2271 0.8259 0.000 0.916 0.008 0.076
#> GSM39147 2 0.3610 0.7475 0.000 0.800 0.000 0.200
#> GSM39188 3 0.1398 0.7617 0.000 0.040 0.956 0.004
#> GSM39189 3 0.2032 0.7531 0.000 0.028 0.936 0.036
#> GSM39190 3 0.2271 0.7585 0.000 0.076 0.916 0.008
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM39104 4 0.7521 0.2961 0.092 0.000 0.144 0.476 0.288
#> GSM39105 1 0.7596 -0.2145 0.364 0.000 0.052 0.364 0.220
#> GSM39106 5 0.7432 -0.1147 0.088 0.000 0.120 0.348 0.444
#> GSM39107 5 0.3920 0.5623 0.040 0.024 0.000 0.116 0.820
#> GSM39108 4 0.8043 0.1780 0.196 0.000 0.108 0.348 0.348
#> GSM39109 5 0.8785 -0.1007 0.012 0.168 0.244 0.284 0.292
#> GSM39110 4 0.8454 0.1072 0.136 0.012 0.172 0.348 0.332
#> GSM39111 4 0.8191 0.3751 0.164 0.000 0.228 0.412 0.196
#> GSM39112 5 0.4741 0.4684 0.084 0.004 0.008 0.148 0.756
#> GSM39113 5 0.2953 0.5727 0.004 0.028 0.000 0.100 0.868
#> GSM39114 5 0.4273 0.5358 0.000 0.212 0.004 0.036 0.748
#> GSM39115 4 0.6630 0.1691 0.368 0.000 0.012 0.464 0.156
#> GSM39148 1 0.0566 0.7434 0.984 0.000 0.000 0.004 0.012
#> GSM39149 3 0.1547 0.7904 0.000 0.016 0.948 0.032 0.004
#> GSM39150 4 0.6951 0.4586 0.104 0.000 0.196 0.584 0.116
#> GSM39151 3 0.2027 0.7967 0.000 0.024 0.928 0.040 0.008
#> GSM39152 3 0.3838 0.6804 0.000 0.004 0.804 0.148 0.044
#> GSM39153 1 0.2452 0.7432 0.896 0.000 0.016 0.084 0.004
#> GSM39154 1 0.2561 0.7431 0.884 0.000 0.020 0.096 0.000
#> GSM39155 1 0.3922 0.6651 0.780 0.000 0.000 0.180 0.040
#> GSM39156 1 0.4851 0.5843 0.712 0.000 0.000 0.092 0.196
#> GSM39157 1 0.2179 0.7377 0.896 0.000 0.000 0.100 0.004
#> GSM39158 4 0.4516 0.1761 0.416 0.000 0.004 0.576 0.004
#> GSM39159 4 0.7563 0.3137 0.140 0.032 0.304 0.488 0.036
#> GSM39160 4 0.7710 0.3076 0.168 0.000 0.316 0.428 0.088
#> GSM39161 4 0.6350 0.3410 0.092 0.028 0.276 0.596 0.008
#> GSM39162 1 0.0865 0.7419 0.972 0.000 0.000 0.004 0.024
#> GSM39163 1 0.2563 0.7346 0.872 0.000 0.000 0.120 0.008
#> GSM39164 1 0.2873 0.7264 0.856 0.000 0.000 0.128 0.016
#> GSM39165 1 0.6832 0.0490 0.472 0.000 0.300 0.216 0.012
#> GSM39166 4 0.4550 0.5243 0.168 0.000 0.044 0.764 0.024
#> GSM39167 1 0.1197 0.7443 0.952 0.000 0.000 0.048 0.000
#> GSM39168 1 0.0671 0.7429 0.980 0.000 0.000 0.004 0.016
#> GSM39169 1 0.4660 0.5941 0.728 0.000 0.016 0.220 0.036
#> GSM39170 4 0.4960 0.3245 0.352 0.000 0.016 0.616 0.016
#> GSM39171 1 0.7683 -0.1505 0.412 0.000 0.204 0.316 0.068
#> GSM39172 3 0.4668 0.7194 0.000 0.136 0.764 0.084 0.016
#> GSM39173 3 0.4112 0.7601 0.000 0.096 0.812 0.072 0.020
#> GSM39174 1 0.1830 0.7479 0.924 0.000 0.000 0.068 0.008
#> GSM39175 1 0.4584 0.6364 0.752 0.000 0.084 0.160 0.004
#> GSM39176 1 0.1608 0.7443 0.928 0.000 0.000 0.072 0.000
#> GSM39177 3 0.3940 0.7243 0.040 0.016 0.820 0.120 0.004
#> GSM39178 4 0.5871 0.1994 0.044 0.004 0.376 0.552 0.024
#> GSM39179 3 0.2472 0.7993 0.000 0.044 0.908 0.036 0.012
#> GSM39180 3 0.5755 0.6198 0.000 0.212 0.648 0.128 0.012
#> GSM39181 4 0.4728 0.3948 0.296 0.000 0.040 0.664 0.000
#> GSM39182 3 0.8098 0.3429 0.036 0.324 0.424 0.160 0.056
#> GSM39183 4 0.4675 0.5019 0.196 0.000 0.060 0.736 0.008
#> GSM39184 1 0.3972 0.6553 0.764 0.000 0.008 0.212 0.016
#> GSM39185 4 0.6066 0.2494 0.044 0.044 0.324 0.584 0.004
#> GSM39186 1 0.5920 0.3418 0.592 0.000 0.024 0.312 0.072
#> GSM39187 1 0.3064 0.7377 0.856 0.000 0.000 0.108 0.036
#> GSM39116 2 0.3203 0.7528 0.000 0.848 0.008 0.020 0.124
#> GSM39117 2 0.3923 0.6838 0.000 0.812 0.132 0.040 0.016
#> GSM39118 2 0.2011 0.7664 0.000 0.928 0.044 0.008 0.020
#> GSM39119 2 0.2927 0.7428 0.000 0.880 0.080 0.020 0.020
#> GSM39120 5 0.5008 0.5332 0.156 0.020 0.004 0.076 0.744
#> GSM39121 5 0.5126 0.5462 0.092 0.172 0.000 0.016 0.720
#> GSM39122 5 0.4726 0.4914 0.048 0.228 0.000 0.008 0.716
#> GSM39123 2 0.3830 0.6927 0.000 0.820 0.124 0.040 0.016
#> GSM39124 2 0.4735 0.2954 0.000 0.524 0.000 0.016 0.460
#> GSM39125 5 0.6671 0.3361 0.236 0.020 0.008 0.164 0.572
#> GSM39126 5 0.4652 0.5410 0.056 0.188 0.000 0.012 0.744
#> GSM39127 2 0.4313 0.5269 0.000 0.636 0.000 0.008 0.356
#> GSM39128 2 0.4375 0.3944 0.000 0.576 0.000 0.004 0.420
#> GSM39129 2 0.2879 0.7600 0.000 0.880 0.080 0.008 0.032
#> GSM39130 2 0.3923 0.6838 0.000 0.812 0.132 0.040 0.016
#> GSM39131 2 0.4446 0.4378 0.000 0.592 0.000 0.008 0.400
#> GSM39132 2 0.4029 0.5914 0.000 0.680 0.000 0.004 0.316
#> GSM39133 2 0.3100 0.7354 0.000 0.876 0.064 0.040 0.020
#> GSM39134 2 0.1960 0.7673 0.000 0.928 0.048 0.004 0.020
#> GSM39135 2 0.2304 0.7569 0.000 0.892 0.000 0.008 0.100
#> GSM39136 2 0.2727 0.7538 0.000 0.868 0.000 0.016 0.116
#> GSM39137 5 0.5299 -0.0492 0.024 0.420 0.000 0.016 0.540
#> GSM39138 2 0.2046 0.7646 0.000 0.916 0.068 0.000 0.016
#> GSM39139 2 0.3207 0.7630 0.000 0.864 0.040 0.012 0.084
#> GSM39140 1 0.4441 0.5670 0.720 0.000 0.000 0.044 0.236
#> GSM39141 1 0.3241 0.6807 0.832 0.000 0.000 0.024 0.144
#> GSM39142 1 0.2813 0.7111 0.868 0.000 0.000 0.024 0.108
#> GSM39143 1 0.3586 0.6535 0.792 0.000 0.000 0.020 0.188
#> GSM39144 2 0.2535 0.7643 0.000 0.892 0.076 0.000 0.032
#> GSM39145 2 0.2984 0.7471 0.000 0.856 0.016 0.004 0.124
#> GSM39146 2 0.3706 0.7191 0.000 0.792 0.004 0.020 0.184
#> GSM39147 2 0.4047 0.5803 0.000 0.676 0.000 0.004 0.320
#> GSM39188 3 0.1403 0.7976 0.000 0.024 0.952 0.024 0.000
#> GSM39189 3 0.2873 0.7412 0.000 0.000 0.860 0.120 0.020
#> GSM39190 3 0.2857 0.7967 0.000 0.064 0.888 0.028 0.020
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM39104 6 0.692 0.43537 0.052 0.056 0.156 0.000 0.180 0.556
#> GSM39105 6 0.679 0.27020 0.280 0.028 0.036 0.000 0.160 0.496
#> GSM39106 6 0.648 0.51313 0.056 0.132 0.084 0.004 0.088 0.636
#> GSM39107 2 0.567 0.04657 0.040 0.484 0.004 0.004 0.040 0.428
#> GSM39108 6 0.718 0.47342 0.180 0.052 0.116 0.004 0.096 0.552
#> GSM39109 6 0.775 0.26608 0.004 0.088 0.216 0.192 0.048 0.452
#> GSM39110 6 0.721 0.48630 0.120 0.084 0.136 0.004 0.088 0.568
#> GSM39111 6 0.655 0.45406 0.080 0.016 0.208 0.000 0.124 0.572
#> GSM39112 6 0.621 0.11225 0.104 0.360 0.004 0.000 0.044 0.488
#> GSM39113 2 0.536 -0.00203 0.024 0.468 0.008 0.004 0.028 0.468
#> GSM39114 2 0.393 0.51460 0.000 0.764 0.000 0.052 0.008 0.176
#> GSM39115 5 0.682 -0.05642 0.296 0.024 0.008 0.000 0.336 0.336
#> GSM39148 1 0.140 0.74617 0.948 0.004 0.000 0.000 0.024 0.024
#> GSM39149 3 0.324 0.73401 0.000 0.000 0.848 0.060 0.024 0.068
#> GSM39150 5 0.725 -0.05154 0.068 0.016 0.192 0.000 0.388 0.336
#> GSM39151 3 0.300 0.74360 0.000 0.000 0.864 0.064 0.024 0.048
#> GSM39152 3 0.457 0.61958 0.000 0.012 0.752 0.020 0.080 0.136
#> GSM39153 1 0.397 0.73449 0.812 0.016 0.044 0.000 0.036 0.092
#> GSM39154 1 0.390 0.73276 0.812 0.004 0.040 0.000 0.080 0.064
#> GSM39155 1 0.514 0.61341 0.672 0.008 0.008 0.000 0.180 0.132
#> GSM39156 1 0.556 0.59386 0.672 0.080 0.020 0.000 0.044 0.184
#> GSM39157 1 0.360 0.72643 0.812 0.008 0.004 0.000 0.120 0.056
#> GSM39158 5 0.410 0.53110 0.232 0.012 0.004 0.000 0.728 0.024
#> GSM39159 5 0.686 0.43829 0.120 0.004 0.224 0.028 0.552 0.072
#> GSM39160 6 0.793 0.00588 0.112 0.020 0.276 0.004 0.292 0.296
#> GSM39161 5 0.492 0.52464 0.052 0.000 0.156 0.024 0.732 0.036
#> GSM39162 1 0.104 0.74379 0.964 0.008 0.000 0.000 0.004 0.024
#> GSM39163 1 0.371 0.72144 0.800 0.016 0.004 0.000 0.144 0.036
#> GSM39164 1 0.448 0.71370 0.760 0.012 0.016 0.000 0.108 0.104
#> GSM39165 1 0.764 -0.01002 0.396 0.032 0.304 0.008 0.200 0.060
#> GSM39166 5 0.364 0.56716 0.072 0.004 0.020 0.000 0.824 0.080
#> GSM39167 1 0.204 0.74278 0.908 0.004 0.000 0.000 0.072 0.016
#> GSM39168 1 0.174 0.74794 0.932 0.008 0.000 0.000 0.020 0.040
#> GSM39169 1 0.612 0.52984 0.592 0.016 0.036 0.000 0.228 0.128
#> GSM39170 5 0.465 0.53171 0.196 0.004 0.008 0.000 0.708 0.084
#> GSM39171 5 0.797 0.08011 0.304 0.016 0.164 0.004 0.320 0.192
#> GSM39172 3 0.581 0.61858 0.000 0.008 0.608 0.256 0.072 0.056
#> GSM39173 3 0.668 0.60771 0.000 0.052 0.600 0.160 0.088 0.100
#> GSM39174 1 0.352 0.74332 0.820 0.012 0.000 0.000 0.092 0.076
#> GSM39175 1 0.590 0.59635 0.660 0.028 0.096 0.000 0.156 0.060
#> GSM39176 1 0.247 0.74772 0.888 0.012 0.004 0.000 0.084 0.012
#> GSM39177 3 0.408 0.71703 0.016 0.004 0.812 0.040 0.084 0.044
#> GSM39178 5 0.637 0.19277 0.012 0.004 0.292 0.004 0.472 0.216
#> GSM39179 3 0.354 0.74162 0.008 0.012 0.836 0.100 0.032 0.012
#> GSM39180 3 0.650 0.48120 0.000 0.016 0.484 0.332 0.136 0.032
#> GSM39181 5 0.288 0.58434 0.128 0.004 0.008 0.000 0.848 0.012
#> GSM39182 4 0.827 -0.22844 0.056 0.024 0.248 0.420 0.140 0.112
#> GSM39183 5 0.336 0.57780 0.068 0.004 0.020 0.000 0.844 0.064
#> GSM39184 1 0.552 0.59100 0.656 0.024 0.024 0.000 0.216 0.080
#> GSM39185 5 0.445 0.51717 0.020 0.004 0.168 0.036 0.756 0.016
#> GSM39186 1 0.687 0.17894 0.444 0.008 0.044 0.000 0.248 0.256
#> GSM39187 1 0.369 0.73239 0.808 0.024 0.000 0.000 0.120 0.048
#> GSM39116 4 0.413 0.54921 0.000 0.300 0.012 0.676 0.004 0.008
#> GSM39117 4 0.214 0.64962 0.000 0.000 0.064 0.908 0.016 0.012
#> GSM39118 4 0.394 0.68598 0.000 0.164 0.032 0.780 0.012 0.012
#> GSM39119 4 0.255 0.69412 0.000 0.060 0.040 0.888 0.000 0.012
#> GSM39120 2 0.758 -0.04030 0.184 0.420 0.024 0.004 0.088 0.280
#> GSM39121 2 0.446 0.50568 0.072 0.772 0.004 0.028 0.008 0.116
#> GSM39122 2 0.496 0.55086 0.028 0.732 0.004 0.088 0.012 0.136
#> GSM39123 4 0.222 0.65102 0.000 0.004 0.060 0.908 0.016 0.012
#> GSM39124 2 0.427 0.36570 0.000 0.712 0.004 0.240 0.008 0.036
#> GSM39125 2 0.770 -0.14384 0.188 0.348 0.004 0.000 0.236 0.224
#> GSM39126 2 0.498 0.52291 0.044 0.744 0.004 0.048 0.028 0.132
#> GSM39127 2 0.421 0.29032 0.000 0.652 0.000 0.320 0.004 0.024
#> GSM39128 2 0.453 0.34443 0.000 0.656 0.000 0.288 0.004 0.052
#> GSM39129 4 0.486 0.67097 0.000 0.136 0.084 0.736 0.016 0.028
#> GSM39130 4 0.222 0.65102 0.000 0.004 0.060 0.908 0.016 0.012
#> GSM39131 2 0.480 0.29219 0.000 0.632 0.004 0.308 0.008 0.048
#> GSM39132 2 0.434 0.11113 0.000 0.604 0.000 0.372 0.008 0.016
#> GSM39133 4 0.242 0.67349 0.000 0.060 0.020 0.900 0.008 0.012
#> GSM39134 4 0.342 0.69490 0.000 0.148 0.028 0.812 0.004 0.008
#> GSM39135 4 0.432 0.53387 0.000 0.336 0.000 0.636 0.012 0.016
#> GSM39136 4 0.435 0.53006 0.000 0.312 0.008 0.656 0.004 0.020
#> GSM39137 2 0.402 0.47061 0.032 0.780 0.004 0.160 0.004 0.020
#> GSM39138 4 0.418 0.68448 0.000 0.144 0.048 0.776 0.008 0.024
#> GSM39139 4 0.542 0.49680 0.000 0.344 0.032 0.576 0.016 0.032
#> GSM39140 1 0.555 0.55769 0.660 0.136 0.012 0.000 0.028 0.164
#> GSM39141 1 0.402 0.69045 0.792 0.076 0.000 0.000 0.032 0.100
#> GSM39142 1 0.423 0.67145 0.772 0.056 0.000 0.000 0.040 0.132
#> GSM39143 1 0.463 0.64597 0.740 0.072 0.000 0.000 0.044 0.144
#> GSM39144 4 0.468 0.66610 0.000 0.184 0.052 0.728 0.012 0.024
#> GSM39145 4 0.580 0.42066 0.000 0.376 0.040 0.524 0.016 0.044
#> GSM39146 4 0.493 0.38053 0.000 0.376 0.016 0.576 0.012 0.020
#> GSM39147 2 0.464 0.07553 0.000 0.596 0.004 0.364 0.004 0.032
#> GSM39188 3 0.284 0.75285 0.000 0.000 0.868 0.076 0.044 0.012
#> GSM39189 3 0.574 0.61304 0.000 0.004 0.648 0.068 0.108 0.172
#> GSM39190 3 0.456 0.73436 0.000 0.004 0.744 0.136 0.096 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) other(p) protocol(p) k
#> MAD:skmeans 82 8.14e-02 1.64e-07 6.97e-06 2
#> MAD:skmeans 79 2.76e-01 5.54e-09 3.42e-09 3
#> MAD:skmeans 66 2.60e-11 2.90e-18 4.43e-18 4
#> MAD:skmeans 59 1.29e-05 1.25e-09 2.09e-10 5
#> MAD:skmeans 56 2.15e-06 7.30e-09 8.39e-13 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 8353 rows and 87 columns.
#> Top rows (835, 1670, 2506, 3341, 4176) 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 0.629 0.788 0.908 0.2684 0.743 0.743
#> 3 3 0.226 0.570 0.780 1.1910 0.599 0.480
#> 4 4 0.250 0.564 0.741 0.0620 0.987 0.968
#> 5 5 0.261 0.495 0.720 0.0288 0.977 0.943
#> 6 6 0.255 0.501 0.704 0.0137 0.965 0.915
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
#> GSM39104 1 0.1633 0.9084 0.976 0.024
#> GSM39105 1 0.2423 0.9048 0.960 0.040
#> GSM39106 1 0.0672 0.9090 0.992 0.008
#> GSM39107 1 0.2423 0.9048 0.960 0.040
#> GSM39108 1 0.2423 0.9048 0.960 0.040
#> GSM39109 1 0.4161 0.8709 0.916 0.084
#> GSM39110 1 0.1414 0.9088 0.980 0.020
#> GSM39111 1 0.2423 0.9048 0.960 0.040
#> GSM39112 1 0.2423 0.9048 0.960 0.040
#> GSM39113 1 0.2423 0.9048 0.960 0.040
#> GSM39114 1 0.2423 0.9048 0.960 0.040
#> GSM39115 1 0.2423 0.9048 0.960 0.040
#> GSM39148 1 0.0000 0.9078 1.000 0.000
#> GSM39149 1 0.3114 0.8707 0.944 0.056
#> GSM39150 1 0.0000 0.9078 1.000 0.000
#> GSM39151 1 0.4815 0.8203 0.896 0.104
#> GSM39152 1 0.0000 0.9078 1.000 0.000
#> GSM39153 1 0.0000 0.9078 1.000 0.000
#> GSM39154 1 0.0376 0.9085 0.996 0.004
#> GSM39155 1 0.2423 0.9048 0.960 0.040
#> GSM39156 1 0.0000 0.9078 1.000 0.000
#> GSM39157 1 0.2423 0.9048 0.960 0.040
#> GSM39158 1 0.0000 0.9078 1.000 0.000
#> GSM39159 1 0.0000 0.9078 1.000 0.000
#> GSM39160 1 0.0000 0.9078 1.000 0.000
#> GSM39161 1 0.0000 0.9078 1.000 0.000
#> GSM39162 1 0.0000 0.9078 1.000 0.000
#> GSM39163 1 0.0672 0.9091 0.992 0.008
#> GSM39164 1 0.0000 0.9078 1.000 0.000
#> GSM39165 1 0.0000 0.9078 1.000 0.000
#> GSM39166 1 0.0000 0.9078 1.000 0.000
#> GSM39167 1 0.0000 0.9078 1.000 0.000
#> GSM39168 1 0.0000 0.9078 1.000 0.000
#> GSM39169 1 0.0000 0.9078 1.000 0.000
#> GSM39170 1 0.0000 0.9078 1.000 0.000
#> GSM39171 1 0.2423 0.9048 0.960 0.040
#> GSM39172 1 0.8955 0.3676 0.688 0.312
#> GSM39173 1 0.0000 0.9078 1.000 0.000
#> GSM39174 1 0.0672 0.9091 0.992 0.008
#> GSM39175 1 0.0000 0.9078 1.000 0.000
#> GSM39176 1 0.0000 0.9078 1.000 0.000
#> GSM39177 1 0.0000 0.9078 1.000 0.000
#> GSM39178 1 0.0000 0.9078 1.000 0.000
#> GSM39179 1 0.9580 0.1052 0.620 0.380
#> GSM39180 1 0.9977 -0.3117 0.528 0.472
#> GSM39181 1 0.1414 0.9088 0.980 0.020
#> GSM39182 1 0.5294 0.7864 0.880 0.120
#> GSM39183 1 0.0672 0.9091 0.992 0.008
#> GSM39184 1 0.2423 0.9048 0.960 0.040
#> GSM39185 1 0.0938 0.9092 0.988 0.012
#> GSM39186 1 0.2236 0.9058 0.964 0.036
#> GSM39187 1 0.0000 0.9078 1.000 0.000
#> GSM39116 2 0.9954 0.3844 0.460 0.540
#> GSM39117 2 0.0000 0.7236 0.000 1.000
#> GSM39118 2 0.9933 0.4074 0.452 0.548
#> GSM39119 2 0.6623 0.7615 0.172 0.828
#> GSM39120 1 0.1414 0.9089 0.980 0.020
#> GSM39121 1 0.2423 0.9048 0.960 0.040
#> GSM39122 1 0.2423 0.9048 0.960 0.040
#> GSM39123 2 0.0000 0.7236 0.000 1.000
#> GSM39124 1 0.2603 0.9026 0.956 0.044
#> GSM39125 1 0.1414 0.9089 0.980 0.020
#> GSM39126 1 0.0672 0.9090 0.992 0.008
#> GSM39127 1 0.9209 0.3931 0.664 0.336
#> GSM39128 1 0.0000 0.9078 1.000 0.000
#> GSM39129 2 0.6531 0.7616 0.168 0.832
#> GSM39130 2 0.0000 0.7236 0.000 1.000
#> GSM39131 1 0.2778 0.9008 0.952 0.048
#> GSM39132 1 0.6247 0.7788 0.844 0.156
#> GSM39133 2 0.0376 0.7251 0.004 0.996
#> GSM39134 2 0.6973 0.7600 0.188 0.812
#> GSM39135 2 0.9983 0.3305 0.476 0.524
#> GSM39136 2 0.9209 0.6377 0.336 0.664
#> GSM39137 1 0.2423 0.9048 0.960 0.040
#> GSM39138 2 0.7950 0.7418 0.240 0.760
#> GSM39139 1 0.9248 0.3614 0.660 0.340
#> GSM39140 1 0.2423 0.9048 0.960 0.040
#> GSM39141 1 0.2423 0.9048 0.960 0.040
#> GSM39142 1 0.2423 0.9048 0.960 0.040
#> GSM39143 1 0.2423 0.9048 0.960 0.040
#> GSM39144 2 0.9248 0.6325 0.340 0.660
#> GSM39145 1 0.9358 0.3423 0.648 0.352
#> GSM39146 1 0.8713 0.5148 0.708 0.292
#> GSM39147 1 0.2423 0.9048 0.960 0.040
#> GSM39188 1 0.9754 -0.0344 0.592 0.408
#> GSM39189 1 0.6887 0.6797 0.816 0.184
#> GSM39190 1 0.9686 0.1724 0.604 0.396
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM39104 2 0.6307 -0.0128 0.488 0.512 0.000
#> GSM39105 2 0.3482 0.7056 0.128 0.872 0.000
#> GSM39106 1 0.5529 0.6292 0.704 0.296 0.000
#> GSM39107 2 0.1411 0.7028 0.036 0.964 0.000
#> GSM39108 2 0.4555 0.6848 0.200 0.800 0.000
#> GSM39109 2 0.3207 0.7204 0.084 0.904 0.012
#> GSM39110 1 0.4605 0.5757 0.796 0.204 0.000
#> GSM39111 2 0.4452 0.6732 0.192 0.808 0.000
#> GSM39112 2 0.2066 0.7130 0.060 0.940 0.000
#> GSM39113 2 0.1964 0.7125 0.056 0.944 0.000
#> GSM39114 2 0.0000 0.6853 0.000 1.000 0.000
#> GSM39115 2 0.5216 0.6097 0.260 0.740 0.000
#> GSM39148 1 0.0747 0.7289 0.984 0.016 0.000
#> GSM39149 1 0.5803 0.6461 0.736 0.248 0.016
#> GSM39150 1 0.5058 0.6425 0.756 0.244 0.000
#> GSM39151 2 0.8296 0.1346 0.412 0.508 0.080
#> GSM39152 1 0.4346 0.6931 0.816 0.184 0.000
#> GSM39153 1 0.0424 0.7281 0.992 0.008 0.000
#> GSM39154 1 0.2165 0.7265 0.936 0.064 0.000
#> GSM39155 1 0.6215 -0.0179 0.572 0.428 0.000
#> GSM39156 1 0.2356 0.7231 0.928 0.072 0.000
#> GSM39157 2 0.5948 0.5153 0.360 0.640 0.000
#> GSM39158 1 0.3816 0.7129 0.852 0.148 0.000
#> GSM39159 1 0.5968 0.4409 0.636 0.364 0.000
#> GSM39160 1 0.5397 0.6170 0.720 0.280 0.000
#> GSM39161 1 0.4605 0.6693 0.796 0.204 0.000
#> GSM39162 1 0.0892 0.7288 0.980 0.020 0.000
#> GSM39163 1 0.4702 0.6404 0.788 0.212 0.000
#> GSM39164 1 0.0892 0.7294 0.980 0.020 0.000
#> GSM39165 1 0.1753 0.7262 0.952 0.048 0.000
#> GSM39166 1 0.5926 0.4943 0.644 0.356 0.000
#> GSM39167 1 0.0892 0.7288 0.980 0.020 0.000
#> GSM39168 1 0.0892 0.7288 0.980 0.020 0.000
#> GSM39169 1 0.0747 0.7289 0.984 0.016 0.000
#> GSM39170 1 0.1163 0.7301 0.972 0.028 0.000
#> GSM39171 2 0.6267 0.2647 0.452 0.548 0.000
#> GSM39172 1 0.7419 0.5537 0.680 0.088 0.232
#> GSM39173 1 0.0892 0.7288 0.980 0.020 0.000
#> GSM39174 1 0.2165 0.7181 0.936 0.064 0.000
#> GSM39175 1 0.2878 0.7276 0.904 0.096 0.000
#> GSM39176 1 0.0424 0.7281 0.992 0.008 0.000
#> GSM39177 1 0.6079 0.4054 0.612 0.388 0.000
#> GSM39178 1 0.6168 0.3837 0.588 0.412 0.000
#> GSM39179 1 0.8827 0.0903 0.496 0.120 0.384
#> GSM39180 2 0.9857 0.0214 0.252 0.380 0.368
#> GSM39181 1 0.5058 0.6164 0.756 0.244 0.000
#> GSM39182 1 0.7794 0.3838 0.572 0.368 0.060
#> GSM39183 1 0.5760 0.5391 0.672 0.328 0.000
#> GSM39184 2 0.6225 0.3576 0.432 0.568 0.000
#> GSM39185 1 0.5706 0.5461 0.680 0.320 0.000
#> GSM39186 2 0.6235 0.2983 0.436 0.564 0.000
#> GSM39187 1 0.4235 0.6740 0.824 0.176 0.000
#> GSM39116 2 0.5529 0.3748 0.000 0.704 0.296
#> GSM39117 3 0.0000 0.7732 0.000 0.000 1.000
#> GSM39118 2 0.6252 0.0732 0.000 0.556 0.444
#> GSM39119 3 0.4452 0.7406 0.000 0.192 0.808
#> GSM39120 2 0.5988 0.3867 0.368 0.632 0.000
#> GSM39121 2 0.5529 0.5502 0.296 0.704 0.000
#> GSM39122 2 0.4702 0.6576 0.212 0.788 0.000
#> GSM39123 3 0.0000 0.7732 0.000 0.000 1.000
#> GSM39124 2 0.4931 0.6524 0.232 0.768 0.000
#> GSM39125 2 0.4121 0.6469 0.168 0.832 0.000
#> GSM39126 1 0.6280 0.1705 0.540 0.460 0.000
#> GSM39127 2 0.5028 0.6230 0.040 0.828 0.132
#> GSM39128 2 0.6045 0.1207 0.380 0.620 0.000
#> GSM39129 3 0.4912 0.7370 0.008 0.196 0.796
#> GSM39130 3 0.0000 0.7732 0.000 0.000 1.000
#> GSM39131 2 0.1289 0.6965 0.032 0.968 0.000
#> GSM39132 2 0.4174 0.7054 0.092 0.872 0.036
#> GSM39133 3 0.0424 0.7749 0.000 0.008 0.992
#> GSM39134 3 0.5875 0.7424 0.072 0.136 0.792
#> GSM39135 2 0.7043 0.1509 0.024 0.576 0.400
#> GSM39136 3 0.6204 0.3908 0.000 0.424 0.576
#> GSM39137 2 0.3816 0.7126 0.148 0.852 0.000
#> GSM39138 3 0.6510 0.6625 0.156 0.088 0.756
#> GSM39139 2 0.5823 0.6572 0.064 0.792 0.144
#> GSM39140 2 0.5650 0.5645 0.312 0.688 0.000
#> GSM39141 2 0.4555 0.6906 0.200 0.800 0.000
#> GSM39142 2 0.4346 0.6873 0.184 0.816 0.000
#> GSM39143 2 0.3412 0.7073 0.124 0.876 0.000
#> GSM39144 3 0.6111 0.4078 0.000 0.396 0.604
#> GSM39145 2 0.4805 0.6118 0.012 0.812 0.176
#> GSM39146 2 0.2269 0.6847 0.016 0.944 0.040
#> GSM39147 2 0.3038 0.7143 0.104 0.896 0.000
#> GSM39188 1 0.9311 0.2155 0.468 0.168 0.364
#> GSM39189 1 0.7788 0.5475 0.632 0.284 0.084
#> GSM39190 2 0.6423 0.5616 0.044 0.728 0.228
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM39104 2 0.6557 0.0548 0.448 0.476 NA 0.000
#> GSM39105 2 0.2984 0.6994 0.084 0.888 NA 0.000
#> GSM39106 1 0.5599 0.5994 0.672 0.276 NA 0.000
#> GSM39107 2 0.3427 0.6915 0.028 0.860 NA 0.000
#> GSM39108 2 0.4121 0.6764 0.184 0.796 NA 0.000
#> GSM39109 2 0.3726 0.7136 0.060 0.864 NA 0.008
#> GSM39110 1 0.3528 0.5679 0.808 0.192 NA 0.000
#> GSM39111 2 0.3547 0.6844 0.144 0.840 NA 0.000
#> GSM39112 2 0.3523 0.6913 0.032 0.856 NA 0.000
#> GSM39113 2 0.3485 0.6902 0.028 0.856 NA 0.000
#> GSM39114 2 0.2868 0.6758 0.000 0.864 NA 0.000
#> GSM39115 2 0.5219 0.6018 0.216 0.728 NA 0.000
#> GSM39148 1 0.0469 0.7131 0.988 0.012 NA 0.000
#> GSM39149 1 0.5572 0.6138 0.692 0.260 NA 0.008
#> GSM39150 1 0.6139 0.5919 0.656 0.244 NA 0.000
#> GSM39151 2 0.7154 0.2144 0.364 0.540 NA 0.052
#> GSM39152 1 0.3870 0.6611 0.788 0.208 NA 0.000
#> GSM39153 1 0.0336 0.7130 0.992 0.008 NA 0.000
#> GSM39154 1 0.2081 0.7027 0.916 0.084 NA 0.000
#> GSM39155 1 0.5337 0.0210 0.564 0.424 NA 0.000
#> GSM39156 1 0.2149 0.6990 0.912 0.088 NA 0.000
#> GSM39157 2 0.4730 0.4715 0.364 0.636 NA 0.000
#> GSM39158 1 0.5416 0.6656 0.740 0.148 NA 0.000
#> GSM39159 1 0.5527 0.4407 0.616 0.356 NA 0.000
#> GSM39160 1 0.5321 0.5768 0.672 0.296 NA 0.000
#> GSM39161 1 0.6083 0.6079 0.672 0.216 NA 0.000
#> GSM39162 1 0.0921 0.7113 0.972 0.028 NA 0.000
#> GSM39163 1 0.3688 0.6238 0.792 0.208 NA 0.000
#> GSM39164 1 0.0707 0.7135 0.980 0.020 NA 0.000
#> GSM39165 1 0.1557 0.7068 0.944 0.056 NA 0.000
#> GSM39166 1 0.6831 0.4420 0.536 0.352 NA 0.000
#> GSM39167 1 0.0921 0.7113 0.972 0.028 NA 0.000
#> GSM39168 1 0.0921 0.7113 0.972 0.028 NA 0.000
#> GSM39169 1 0.0592 0.7129 0.984 0.016 NA 0.000
#> GSM39170 1 0.3144 0.7016 0.884 0.044 NA 0.000
#> GSM39171 2 0.5894 0.2770 0.392 0.568 NA 0.000
#> GSM39172 1 0.6990 0.5387 0.632 0.120 NA 0.224
#> GSM39173 1 0.0921 0.7113 0.972 0.028 NA 0.000
#> GSM39174 1 0.1637 0.7049 0.940 0.060 NA 0.000
#> GSM39175 1 0.2216 0.7115 0.908 0.092 NA 0.000
#> GSM39176 1 0.0000 0.7124 1.000 0.000 NA 0.000
#> GSM39177 1 0.6163 0.3326 0.532 0.416 NA 0.000
#> GSM39178 1 0.6857 0.3464 0.492 0.404 NA 0.000
#> GSM39179 1 0.7977 0.1390 0.496 0.112 NA 0.344
#> GSM39180 2 0.8898 0.0787 0.252 0.376 NA 0.320
#> GSM39181 1 0.6147 0.5995 0.664 0.224 NA 0.000
#> GSM39182 1 0.6189 0.3584 0.568 0.372 NA 0.060
#> GSM39183 1 0.6607 0.5392 0.592 0.296 NA 0.000
#> GSM39184 2 0.6186 0.3779 0.352 0.584 NA 0.000
#> GSM39185 1 0.6729 0.4957 0.572 0.312 NA 0.000
#> GSM39186 2 0.6123 0.3212 0.372 0.572 NA 0.000
#> GSM39187 1 0.3569 0.6381 0.804 0.196 NA 0.000
#> GSM39116 2 0.6440 0.4842 0.000 0.644 NA 0.208
#> GSM39117 4 0.0592 0.7685 0.000 0.000 NA 0.984
#> GSM39118 2 0.4955 0.1904 0.000 0.556 NA 0.444
#> GSM39119 4 0.3356 0.6948 0.000 0.176 NA 0.824
#> GSM39120 2 0.6275 0.4286 0.328 0.596 NA 0.000
#> GSM39121 2 0.4936 0.4805 0.340 0.652 NA 0.000
#> GSM39122 2 0.4872 0.6177 0.244 0.728 NA 0.000
#> GSM39123 4 0.0592 0.7685 0.000 0.000 NA 0.984
#> GSM39124 2 0.4262 0.6240 0.236 0.756 NA 0.000
#> GSM39125 2 0.5007 0.6266 0.172 0.760 NA 0.000
#> GSM39126 1 0.6690 0.2478 0.548 0.352 NA 0.000
#> GSM39127 2 0.5478 0.6489 0.036 0.752 NA 0.036
#> GSM39128 2 0.7942 0.0787 0.368 0.440 NA 0.016
#> GSM39129 4 0.3768 0.7514 0.000 0.008 NA 0.808
#> GSM39130 4 0.0592 0.7685 0.000 0.000 NA 0.984
#> GSM39131 2 0.4218 0.6582 0.012 0.796 NA 0.008
#> GSM39132 2 0.5476 0.6619 0.056 0.744 NA 0.016
#> GSM39133 4 0.0336 0.7698 0.000 0.008 NA 0.992
#> GSM39134 4 0.4428 0.7204 0.068 0.124 NA 0.808
#> GSM39135 2 0.6671 0.3075 0.020 0.576 NA 0.348
#> GSM39136 4 0.6562 0.2216 0.000 0.404 NA 0.516
#> GSM39137 2 0.3105 0.6912 0.140 0.856 NA 0.000
#> GSM39138 4 0.4985 0.6688 0.152 0.080 NA 0.768
#> GSM39139 2 0.5333 0.6881 0.068 0.792 NA 0.076
#> GSM39140 2 0.4250 0.5848 0.276 0.724 NA 0.000
#> GSM39141 2 0.3311 0.6772 0.172 0.828 NA 0.000
#> GSM39142 2 0.3528 0.6720 0.192 0.808 NA 0.000
#> GSM39143 2 0.2469 0.6967 0.108 0.892 NA 0.000
#> GSM39144 4 0.6566 0.4960 0.000 0.288 NA 0.600
#> GSM39145 2 0.4549 0.6625 0.016 0.820 NA 0.108
#> GSM39146 2 0.2761 0.6902 0.012 0.908 NA 0.016
#> GSM39147 2 0.3333 0.7065 0.088 0.872 NA 0.000
#> GSM39188 4 0.9204 0.2548 0.280 0.072 NA 0.332
#> GSM39189 1 0.7611 0.5275 0.568 0.288 NA 0.056
#> GSM39190 2 0.5230 0.5896 0.028 0.744 NA 0.208
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM39104 1 0.6619 -0.05658 0.444 0.420 0.108 0.000 NA
#> GSM39105 2 0.3018 0.69603 0.084 0.872 0.008 0.000 NA
#> GSM39106 1 0.5939 0.54534 0.640 0.240 0.084 0.000 NA
#> GSM39107 2 0.3160 0.66761 0.004 0.808 0.188 0.000 NA
#> GSM39108 2 0.3805 0.66692 0.184 0.784 0.000 0.000 NA
#> GSM39109 2 0.4153 0.69595 0.052 0.812 0.104 0.000 NA
#> GSM39110 1 0.3109 0.53952 0.800 0.200 0.000 0.000 NA
#> GSM39111 2 0.3412 0.67482 0.152 0.820 0.000 0.000 NA
#> GSM39112 2 0.3160 0.66761 0.004 0.808 0.188 0.000 NA
#> GSM39113 2 0.3160 0.66761 0.004 0.808 0.188 0.000 NA
#> GSM39114 2 0.3353 0.66525 0.000 0.796 0.196 0.000 NA
#> GSM39115 2 0.4763 0.59843 0.212 0.712 0.000 0.000 NA
#> GSM39148 1 0.0404 0.63587 0.988 0.012 0.000 0.000 NA
#> GSM39149 1 0.5487 0.56852 0.668 0.252 0.016 0.008 NA
#> GSM39150 1 0.5504 0.53183 0.644 0.224 0.000 0.000 NA
#> GSM39151 2 0.7178 0.17760 0.348 0.500 0.036 0.036 NA
#> GSM39152 1 0.3548 0.62411 0.796 0.188 0.004 0.000 NA
#> GSM39153 1 0.0510 0.63535 0.984 0.016 0.000 0.000 NA
#> GSM39154 1 0.1956 0.63403 0.916 0.076 0.000 0.000 NA
#> GSM39155 1 0.5201 0.00872 0.532 0.424 0.000 0.000 NA
#> GSM39156 1 0.1851 0.62951 0.912 0.088 0.000 0.000 NA
#> GSM39157 2 0.4045 0.47161 0.356 0.644 0.000 0.000 NA
#> GSM39158 1 0.5025 0.55721 0.704 0.124 0.000 0.000 NA
#> GSM39159 1 0.5260 0.40852 0.592 0.348 0.000 0.000 NA
#> GSM39160 1 0.4780 0.56447 0.672 0.280 0.000 0.000 NA
#> GSM39161 1 0.5546 0.52405 0.648 0.176 0.000 0.000 NA
#> GSM39162 1 0.0794 0.63461 0.972 0.028 0.000 0.000 NA
#> GSM39163 1 0.3663 0.57752 0.776 0.208 0.000 0.000 NA
#> GSM39164 1 0.0609 0.63675 0.980 0.020 0.000 0.000 NA
#> GSM39165 1 0.1341 0.63302 0.944 0.056 0.000 0.000 NA
#> GSM39166 1 0.6351 0.43161 0.500 0.316 0.000 0.000 NA
#> GSM39167 1 0.0794 0.63461 0.972 0.028 0.000 0.000 NA
#> GSM39168 1 0.0794 0.63461 0.972 0.028 0.000 0.000 NA
#> GSM39169 1 0.0404 0.63587 0.988 0.012 0.000 0.000 NA
#> GSM39170 1 0.3141 0.59974 0.852 0.040 0.000 0.000 NA
#> GSM39171 2 0.5378 0.22622 0.392 0.548 0.000 0.000 NA
#> GSM39172 1 0.6089 0.34778 0.644 0.100 0.000 0.212 NA
#> GSM39173 1 0.0955 0.63446 0.968 0.028 0.000 0.000 NA
#> GSM39174 1 0.1410 0.63988 0.940 0.060 0.000 0.000 NA
#> GSM39175 1 0.1851 0.64876 0.912 0.088 0.000 0.000 NA
#> GSM39176 1 0.0162 0.63383 0.996 0.004 0.000 0.000 NA
#> GSM39177 1 0.5492 0.37014 0.536 0.396 0.000 0.000 NA
#> GSM39178 1 0.6229 0.32492 0.464 0.392 0.000 0.000 NA
#> GSM39179 1 0.7835 -0.41102 0.448 0.064 0.016 0.308 NA
#> GSM39180 2 0.8203 -0.22675 0.240 0.348 0.004 0.312 NA
#> GSM39181 1 0.5817 0.50990 0.612 0.204 0.000 0.000 NA
#> GSM39182 1 0.5386 0.36428 0.564 0.372 0.000 0.064 NA
#> GSM39183 1 0.6191 0.47539 0.536 0.292 0.000 0.000 NA
#> GSM39184 2 0.5613 0.41625 0.308 0.592 0.000 0.000 NA
#> GSM39185 1 0.6373 0.46975 0.532 0.280 0.004 0.000 NA
#> GSM39186 2 0.5498 0.33642 0.356 0.568 0.000 0.000 NA
#> GSM39187 1 0.3318 0.59227 0.800 0.192 0.000 0.000 NA
#> GSM39116 2 0.6133 0.49970 0.000 0.648 0.136 0.176 NA
#> GSM39117 4 0.0451 0.59437 0.000 0.000 0.008 0.988 NA
#> GSM39118 2 0.4410 0.25661 0.000 0.556 0.004 0.440 NA
#> GSM39119 4 0.3086 0.44998 0.000 0.180 0.004 0.816 NA
#> GSM39120 2 0.5903 0.39259 0.332 0.548 0.120 0.000 NA
#> GSM39121 2 0.4268 0.47798 0.344 0.648 0.008 0.000 NA
#> GSM39122 2 0.4223 0.62039 0.248 0.724 0.028 0.000 NA
#> GSM39123 4 0.0451 0.59437 0.000 0.000 0.008 0.988 NA
#> GSM39124 2 0.4095 0.63831 0.220 0.752 0.024 0.000 NA
#> GSM39125 2 0.4845 0.62514 0.148 0.724 0.128 0.000 NA
#> GSM39126 1 0.6133 0.25992 0.544 0.292 0.164 0.000 NA
#> GSM39127 2 0.5087 0.63369 0.032 0.748 0.164 0.016 NA
#> GSM39128 2 0.7513 0.06386 0.356 0.428 0.164 0.012 NA
#> GSM39129 4 0.3795 0.50837 0.000 0.000 0.192 0.780 NA
#> GSM39130 4 0.0451 0.59437 0.000 0.000 0.008 0.988 NA
#> GSM39131 2 0.4130 0.64738 0.008 0.740 0.240 0.004 NA
#> GSM39132 2 0.5383 0.63565 0.052 0.728 0.168 0.012 NA
#> GSM39133 4 0.0290 0.59765 0.000 0.008 0.000 0.992 NA
#> GSM39134 4 0.4017 0.41159 0.068 0.128 0.004 0.800 NA
#> GSM39135 2 0.6301 0.40344 0.024 0.604 0.056 0.288 NA
#> GSM39136 4 0.6665 0.04736 0.000 0.428 0.092 0.440 NA
#> GSM39137 2 0.2719 0.68660 0.144 0.852 0.004 0.000 NA
#> GSM39138 4 0.4519 0.11473 0.148 0.100 0.000 0.752 NA
#> GSM39139 2 0.4598 0.67066 0.044 0.812 0.064 0.040 NA
#> GSM39140 2 0.3774 0.54887 0.296 0.704 0.000 0.000 NA
#> GSM39141 2 0.2929 0.67051 0.180 0.820 0.000 0.000 NA
#> GSM39142 2 0.3039 0.67194 0.192 0.808 0.000 0.000 NA
#> GSM39143 2 0.2230 0.69187 0.116 0.884 0.000 0.000 NA
#> GSM39144 4 0.5346 0.25944 0.000 0.028 0.016 0.552 NA
#> GSM39145 2 0.3922 0.66591 0.012 0.844 0.044 0.060 NA
#> GSM39146 2 0.2251 0.67490 0.000 0.916 0.052 0.008 NA
#> GSM39147 2 0.2418 0.69385 0.044 0.912 0.024 0.000 NA
#> GSM39188 3 0.7863 0.00000 0.200 0.052 0.432 0.300 NA
#> GSM39189 1 0.6894 0.46736 0.552 0.256 0.000 0.056 NA
#> GSM39190 2 0.6322 0.53840 0.016 0.652 0.028 0.160 NA
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM39104 2 0.6332 0.14021 0.416 0.420 0.120 0.000 NA 0.004
#> GSM39105 2 0.3093 0.67887 0.092 0.852 0.008 0.000 NA 0.004
#> GSM39106 1 0.5632 0.57082 0.636 0.228 0.084 0.000 NA 0.008
#> GSM39107 2 0.2762 0.66604 0.000 0.804 0.196 0.000 NA 0.000
#> GSM39108 2 0.3800 0.66025 0.168 0.776 0.000 0.000 NA 0.008
#> GSM39109 2 0.4243 0.69422 0.048 0.792 0.108 0.004 NA 0.008
#> GSM39110 1 0.2902 0.55015 0.800 0.196 0.000 0.000 NA 0.000
#> GSM39111 2 0.3414 0.66363 0.140 0.812 0.000 0.000 NA 0.008
#> GSM39112 2 0.2762 0.66604 0.000 0.804 0.196 0.000 NA 0.000
#> GSM39113 2 0.2762 0.66604 0.000 0.804 0.196 0.000 NA 0.000
#> GSM39114 2 0.2933 0.66523 0.000 0.796 0.200 0.000 NA 0.000
#> GSM39115 2 0.4711 0.61043 0.192 0.704 0.004 0.000 NA 0.008
#> GSM39148 1 0.0363 0.68389 0.988 0.012 0.000 0.000 NA 0.000
#> GSM39149 1 0.5920 0.58201 0.624 0.236 0.016 0.004 NA 0.064
#> GSM39150 1 0.5157 0.58354 0.636 0.204 0.000 0.000 NA 0.004
#> GSM39151 2 0.7300 0.13655 0.320 0.424 0.008 0.024 NA 0.176
#> GSM39152 1 0.3327 0.65508 0.792 0.188 0.004 0.000 NA 0.004
#> GSM39153 1 0.0508 0.68325 0.984 0.012 0.000 0.000 NA 0.000
#> GSM39154 1 0.1643 0.68280 0.924 0.068 0.000 0.000 NA 0.000
#> GSM39155 1 0.4863 -0.03526 0.528 0.412 0.000 0.000 NA 0.000
#> GSM39156 1 0.1663 0.67696 0.912 0.088 0.000 0.000 NA 0.000
#> GSM39157 2 0.3769 0.47093 0.356 0.640 0.000 0.000 NA 0.000
#> GSM39158 1 0.4587 0.61385 0.688 0.108 0.000 0.000 NA 0.000
#> GSM39159 1 0.4795 0.42466 0.604 0.324 0.000 0.000 NA 0.000
#> GSM39160 1 0.4673 0.57555 0.660 0.264 0.000 0.000 NA 0.004
#> GSM39161 1 0.5040 0.58972 0.636 0.152 0.000 0.000 NA 0.000
#> GSM39162 1 0.0632 0.68282 0.976 0.024 0.000 0.000 NA 0.000
#> GSM39163 1 0.3403 0.58810 0.768 0.212 0.000 0.000 NA 0.000
#> GSM39164 1 0.0547 0.68452 0.980 0.020 0.000 0.000 NA 0.000
#> GSM39165 1 0.1152 0.68323 0.952 0.044 0.000 0.000 NA 0.000
#> GSM39166 1 0.5830 0.44210 0.488 0.284 0.000 0.000 NA 0.000
#> GSM39167 1 0.0632 0.68282 0.976 0.024 0.000 0.000 NA 0.000
#> GSM39168 1 0.0632 0.68282 0.976 0.024 0.000 0.000 NA 0.000
#> GSM39169 1 0.0363 0.68389 0.988 0.012 0.000 0.000 NA 0.000
#> GSM39170 1 0.2706 0.65749 0.852 0.024 0.000 0.000 NA 0.000
#> GSM39171 2 0.5096 0.26833 0.388 0.536 0.000 0.000 NA 0.004
#> GSM39172 1 0.5486 0.48673 0.648 0.088 0.000 0.208 NA 0.000
#> GSM39173 1 0.0777 0.68283 0.972 0.024 0.000 0.000 NA 0.000
#> GSM39174 1 0.1471 0.68322 0.932 0.064 0.000 0.000 NA 0.000
#> GSM39175 1 0.1663 0.69750 0.912 0.088 0.000 0.000 NA 0.000
#> GSM39176 1 0.0146 0.68237 0.996 0.000 0.000 0.000 NA 0.000
#> GSM39177 1 0.4978 0.36154 0.532 0.396 0.000 0.000 NA 0.000
#> GSM39178 1 0.5719 0.33092 0.460 0.372 0.000 0.000 NA 0.000
#> GSM39179 1 0.7459 -0.28162 0.416 0.036 0.020 0.268 NA 0.020
#> GSM39180 2 0.7527 -0.00372 0.244 0.332 0.004 0.316 NA 0.004
#> GSM39181 1 0.5388 0.54987 0.584 0.188 0.000 0.000 NA 0.000
#> GSM39182 1 0.5018 0.35510 0.556 0.372 0.000 0.068 NA 0.000
#> GSM39183 1 0.5688 0.48562 0.524 0.264 0.000 0.000 NA 0.000
#> GSM39184 2 0.5282 0.41351 0.304 0.568 0.000 0.000 NA 0.000
#> GSM39185 1 0.5879 0.46188 0.508 0.260 0.004 0.000 NA 0.000
#> GSM39186 2 0.5098 0.33568 0.352 0.556 0.000 0.000 NA 0.000
#> GSM39187 1 0.2948 0.62668 0.804 0.188 0.000 0.000 NA 0.000
#> GSM39116 2 0.5453 0.53530 0.000 0.668 0.112 0.172 NA 0.044
#> GSM39117 4 0.0260 0.52486 0.000 0.000 0.008 0.992 NA 0.000
#> GSM39118 2 0.3838 0.28819 0.000 0.552 0.000 0.448 NA 0.000
#> GSM39119 4 0.2664 0.33568 0.000 0.184 0.000 0.816 NA 0.000
#> GSM39120 2 0.5257 0.40848 0.328 0.556 0.116 0.000 NA 0.000
#> GSM39121 2 0.3742 0.47288 0.348 0.648 0.004 0.000 NA 0.000
#> GSM39122 2 0.3817 0.61096 0.252 0.720 0.028 0.000 NA 0.000
#> GSM39123 4 0.0260 0.52486 0.000 0.000 0.008 0.992 NA 0.000
#> GSM39124 2 0.3753 0.62971 0.220 0.748 0.028 0.000 NA 0.004
#> GSM39125 2 0.4451 0.61395 0.148 0.724 0.124 0.000 NA 0.000
#> GSM39126 1 0.5492 0.27689 0.552 0.280 0.168 0.000 NA 0.000
#> GSM39127 2 0.4469 0.64440 0.032 0.768 0.140 0.012 NA 0.044
#> GSM39128 2 0.6599 0.05981 0.360 0.452 0.140 0.008 NA 0.036
#> GSM39129 4 0.3695 -0.17034 0.000 0.000 0.000 0.624 NA 0.000
#> GSM39130 4 0.0260 0.52486 0.000 0.000 0.008 0.992 NA 0.000
#> GSM39131 2 0.3388 0.65366 0.004 0.764 0.224 0.004 NA 0.004
#> GSM39132 2 0.4546 0.64974 0.040 0.760 0.144 0.008 NA 0.044
#> GSM39133 4 0.0260 0.52537 0.000 0.008 0.000 0.992 NA 0.000
#> GSM39134 4 0.3508 0.38845 0.068 0.132 0.000 0.800 NA 0.000
#> GSM39135 2 0.5677 0.44710 0.024 0.612 0.044 0.284 NA 0.032
#> GSM39136 2 0.5979 -0.01263 0.000 0.448 0.072 0.432 NA 0.044
#> GSM39137 2 0.2558 0.66722 0.156 0.840 0.004 0.000 NA 0.000
#> GSM39138 4 0.4683 0.29614 0.108 0.096 0.052 0.744 NA 0.000
#> GSM39139 2 0.4072 0.67763 0.048 0.820 0.056 0.028 NA 0.044
#> GSM39140 2 0.3428 0.53808 0.304 0.696 0.000 0.000 NA 0.000
#> GSM39141 2 0.2730 0.65104 0.192 0.808 0.000 0.000 NA 0.000
#> GSM39142 2 0.2793 0.65438 0.200 0.800 0.000 0.000 NA 0.000
#> GSM39143 2 0.2135 0.67337 0.128 0.872 0.000 0.000 NA 0.000
#> GSM39144 6 0.4322 0.00000 0.000 0.008 0.008 0.472 NA 0.512
#> GSM39145 2 0.3396 0.67156 0.012 0.856 0.032 0.056 NA 0.040
#> GSM39146 2 0.1821 0.67481 0.000 0.928 0.040 0.008 NA 0.024
#> GSM39147 2 0.2164 0.69295 0.044 0.912 0.028 0.000 NA 0.016
#> GSM39188 3 0.5758 0.00000 0.112 0.008 0.584 0.280 NA 0.008
#> GSM39189 1 0.6567 0.51450 0.532 0.252 0.000 0.064 NA 0.008
#> GSM39190 2 0.7248 0.43584 0.016 0.548 0.040 0.076 NA 0.196
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) other(p) protocol(p) k
#> MAD:pam 76 0.31534 1.11e-04 3.79e-05 2
#> MAD:pam 65 0.01086 1.89e-10 3.81e-08 3
#> MAD:pam 62 0.00755 1.18e-10 3.24e-08 4
#> MAD:pam 56 0.02012 8.84e-10 1.93e-07 5
#> MAD:pam 56 0.02048 7.12e-09 3.69e-07 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 8353 rows and 87 columns.
#> Top rows (835, 1670, 2506, 3341, 4176) 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 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 0.578 0.718 0.882 0.46895 0.543 0.543
#> 3 3 0.531 0.747 0.833 0.32412 0.759 0.586
#> 4 4 0.503 0.560 0.699 0.11683 0.851 0.621
#> 5 5 0.576 0.621 0.777 0.07484 0.910 0.699
#> 6 6 0.715 0.677 0.776 0.00145 0.807 0.476
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
#> GSM39104 1 0.0000 0.8208 1.000 0.000
#> GSM39105 1 0.0000 0.8208 1.000 0.000
#> GSM39106 1 0.0938 0.8177 0.988 0.012
#> GSM39107 2 0.9933 0.0198 0.452 0.548
#> GSM39108 1 0.0000 0.8208 1.000 0.000
#> GSM39109 2 0.9998 -0.2450 0.492 0.508
#> GSM39110 1 0.6801 0.7111 0.820 0.180
#> GSM39111 1 0.1633 0.8127 0.976 0.024
#> GSM39112 1 0.9491 0.4467 0.632 0.368
#> GSM39113 2 0.9686 0.2217 0.396 0.604
#> GSM39114 2 0.0000 0.9259 0.000 1.000
#> GSM39115 1 0.0000 0.8208 1.000 0.000
#> GSM39148 1 0.0000 0.8208 1.000 0.000
#> GSM39149 1 0.9993 0.2782 0.516 0.484
#> GSM39150 1 0.0000 0.8208 1.000 0.000
#> GSM39151 1 0.9993 0.2782 0.516 0.484
#> GSM39152 1 0.9993 0.2782 0.516 0.484
#> GSM39153 1 0.0000 0.8208 1.000 0.000
#> GSM39154 1 0.0000 0.8208 1.000 0.000
#> GSM39155 1 0.0000 0.8208 1.000 0.000
#> GSM39156 1 0.0376 0.8197 0.996 0.004
#> GSM39157 1 0.0000 0.8208 1.000 0.000
#> GSM39158 1 0.0000 0.8208 1.000 0.000
#> GSM39159 1 0.9170 0.5398 0.668 0.332
#> GSM39160 1 0.0000 0.8208 1.000 0.000
#> GSM39161 1 0.9993 0.2782 0.516 0.484
#> GSM39162 1 0.0376 0.8197 0.996 0.004
#> GSM39163 1 0.0000 0.8208 1.000 0.000
#> GSM39164 1 0.0000 0.8208 1.000 0.000
#> GSM39165 1 0.4562 0.7759 0.904 0.096
#> GSM39166 1 0.0376 0.8200 0.996 0.004
#> GSM39167 1 0.0000 0.8208 1.000 0.000
#> GSM39168 1 0.0000 0.8208 1.000 0.000
#> GSM39169 1 0.0000 0.8208 1.000 0.000
#> GSM39170 1 0.0000 0.8208 1.000 0.000
#> GSM39171 1 0.0000 0.8208 1.000 0.000
#> GSM39172 1 0.9996 0.2699 0.512 0.488
#> GSM39173 1 0.9996 0.2699 0.512 0.488
#> GSM39174 1 0.0000 0.8208 1.000 0.000
#> GSM39175 1 0.0000 0.8208 1.000 0.000
#> GSM39176 1 0.0000 0.8208 1.000 0.000
#> GSM39177 1 0.9993 0.2782 0.516 0.484
#> GSM39178 1 0.8386 0.6241 0.732 0.268
#> GSM39179 1 0.9993 0.2782 0.516 0.484
#> GSM39180 1 0.9996 0.2699 0.512 0.488
#> GSM39181 1 0.6887 0.7079 0.816 0.184
#> GSM39182 1 0.9996 0.2699 0.512 0.488
#> GSM39183 1 0.5178 0.7618 0.884 0.116
#> GSM39184 1 0.0000 0.8208 1.000 0.000
#> GSM39185 1 0.9993 0.2782 0.516 0.484
#> GSM39186 1 0.0000 0.8208 1.000 0.000
#> GSM39187 1 0.0000 0.8208 1.000 0.000
#> GSM39116 2 0.0000 0.9259 0.000 1.000
#> GSM39117 2 0.0000 0.9259 0.000 1.000
#> GSM39118 2 0.0000 0.9259 0.000 1.000
#> GSM39119 2 0.0000 0.9259 0.000 1.000
#> GSM39120 1 0.4298 0.7823 0.912 0.088
#> GSM39121 2 0.4562 0.8255 0.096 0.904
#> GSM39122 2 0.4562 0.8255 0.096 0.904
#> GSM39123 2 0.0000 0.9259 0.000 1.000
#> GSM39124 2 0.0000 0.9259 0.000 1.000
#> GSM39125 1 0.4431 0.7805 0.908 0.092
#> GSM39126 2 0.6531 0.7177 0.168 0.832
#> GSM39127 2 0.0000 0.9259 0.000 1.000
#> GSM39128 2 0.0000 0.9259 0.000 1.000
#> GSM39129 2 0.0000 0.9259 0.000 1.000
#> GSM39130 2 0.0000 0.9259 0.000 1.000
#> GSM39131 2 0.0000 0.9259 0.000 1.000
#> GSM39132 2 0.0000 0.9259 0.000 1.000
#> GSM39133 2 0.0000 0.9259 0.000 1.000
#> GSM39134 2 0.0000 0.9259 0.000 1.000
#> GSM39135 2 0.0000 0.9259 0.000 1.000
#> GSM39136 2 0.0000 0.9259 0.000 1.000
#> GSM39137 2 0.0376 0.9225 0.004 0.996
#> GSM39138 2 0.0000 0.9259 0.000 1.000
#> GSM39139 2 0.0000 0.9259 0.000 1.000
#> GSM39140 1 0.0672 0.8180 0.992 0.008
#> GSM39141 1 0.0672 0.8180 0.992 0.008
#> GSM39142 1 0.0376 0.8197 0.996 0.004
#> GSM39143 1 0.0672 0.8191 0.992 0.008
#> GSM39144 2 0.0000 0.9259 0.000 1.000
#> GSM39145 2 0.0000 0.9259 0.000 1.000
#> GSM39146 2 0.0000 0.9259 0.000 1.000
#> GSM39147 2 0.0000 0.9259 0.000 1.000
#> GSM39188 1 0.9993 0.2782 0.516 0.484
#> GSM39189 1 0.9993 0.2782 0.516 0.484
#> GSM39190 1 0.9993 0.2782 0.516 0.484
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM39104 1 0.1964 0.847 0.944 0.000 0.056
#> GSM39105 1 0.5465 0.561 0.712 0.000 0.288
#> GSM39106 3 0.5497 0.656 0.292 0.000 0.708
#> GSM39107 3 0.4399 0.662 0.044 0.092 0.864
#> GSM39108 3 0.6244 0.359 0.440 0.000 0.560
#> GSM39109 3 0.7308 0.582 0.296 0.056 0.648
#> GSM39110 3 0.5926 0.550 0.356 0.000 0.644
#> GSM39111 1 0.2356 0.845 0.928 0.000 0.072
#> GSM39112 3 0.5891 0.734 0.200 0.036 0.764
#> GSM39113 3 0.3797 0.679 0.056 0.052 0.892
#> GSM39114 3 0.3551 0.574 0.000 0.132 0.868
#> GSM39115 1 0.0592 0.846 0.988 0.000 0.012
#> GSM39148 1 0.5216 0.515 0.740 0.000 0.260
#> GSM39149 1 0.6027 0.789 0.776 0.164 0.060
#> GSM39150 1 0.1860 0.847 0.948 0.000 0.052
#> GSM39151 1 0.6027 0.789 0.776 0.164 0.060
#> GSM39152 1 0.6027 0.789 0.776 0.164 0.060
#> GSM39153 1 0.1964 0.829 0.944 0.000 0.056
#> GSM39154 1 0.0237 0.848 0.996 0.000 0.004
#> GSM39155 1 0.3038 0.787 0.896 0.000 0.104
#> GSM39156 3 0.5138 0.731 0.252 0.000 0.748
#> GSM39157 1 0.2537 0.808 0.920 0.000 0.080
#> GSM39158 1 0.0237 0.848 0.996 0.000 0.004
#> GSM39159 1 0.1585 0.851 0.964 0.028 0.008
#> GSM39160 1 0.1753 0.847 0.952 0.000 0.048
#> GSM39161 1 0.5202 0.761 0.772 0.220 0.008
#> GSM39162 3 0.6260 0.448 0.448 0.000 0.552
#> GSM39163 1 0.0747 0.845 0.984 0.000 0.016
#> GSM39164 1 0.3116 0.782 0.892 0.000 0.108
#> GSM39165 1 0.1182 0.852 0.976 0.012 0.012
#> GSM39166 1 0.0237 0.849 0.996 0.004 0.000
#> GSM39167 1 0.1411 0.837 0.964 0.000 0.036
#> GSM39168 1 0.5785 0.316 0.668 0.000 0.332
#> GSM39169 1 0.3551 0.750 0.868 0.000 0.132
#> GSM39170 1 0.0237 0.848 0.996 0.000 0.004
#> GSM39171 1 0.1860 0.847 0.948 0.000 0.052
#> GSM39172 1 0.6535 0.750 0.728 0.220 0.052
#> GSM39173 1 0.6027 0.789 0.776 0.164 0.060
#> GSM39174 1 0.2796 0.798 0.908 0.000 0.092
#> GSM39175 1 0.0237 0.848 0.996 0.000 0.004
#> GSM39176 1 0.0592 0.846 0.988 0.000 0.012
#> GSM39177 1 0.5932 0.790 0.780 0.164 0.056
#> GSM39178 1 0.3791 0.837 0.892 0.060 0.048
#> GSM39179 1 0.6027 0.789 0.776 0.164 0.060
#> GSM39180 1 0.6586 0.751 0.728 0.216 0.056
#> GSM39181 1 0.1529 0.847 0.960 0.040 0.000
#> GSM39182 1 0.6012 0.759 0.748 0.220 0.032
#> GSM39183 1 0.1031 0.849 0.976 0.024 0.000
#> GSM39184 1 0.0424 0.847 0.992 0.000 0.008
#> GSM39185 1 0.5202 0.761 0.772 0.220 0.008
#> GSM39186 1 0.4235 0.689 0.824 0.000 0.176
#> GSM39187 1 0.2959 0.794 0.900 0.000 0.100
#> GSM39116 2 0.4346 0.890 0.000 0.816 0.184
#> GSM39117 2 0.0237 0.801 0.004 0.996 0.000
#> GSM39118 2 0.4002 0.899 0.000 0.840 0.160
#> GSM39119 2 0.4002 0.899 0.000 0.840 0.160
#> GSM39120 3 0.5461 0.741 0.244 0.008 0.748
#> GSM39121 3 0.2261 0.630 0.000 0.068 0.932
#> GSM39122 3 0.2537 0.623 0.000 0.080 0.920
#> GSM39123 2 0.0237 0.801 0.004 0.996 0.000
#> GSM39124 3 0.3941 0.544 0.000 0.156 0.844
#> GSM39125 3 0.5502 0.739 0.248 0.008 0.744
#> GSM39126 3 0.2356 0.628 0.000 0.072 0.928
#> GSM39127 3 0.5733 0.200 0.000 0.324 0.676
#> GSM39128 3 0.4842 0.433 0.000 0.224 0.776
#> GSM39129 2 0.4002 0.899 0.000 0.840 0.160
#> GSM39130 2 0.0237 0.801 0.004 0.996 0.000
#> GSM39131 3 0.4931 0.414 0.000 0.232 0.768
#> GSM39132 2 0.5905 0.733 0.000 0.648 0.352
#> GSM39133 2 0.0475 0.804 0.004 0.992 0.004
#> GSM39134 2 0.4002 0.899 0.000 0.840 0.160
#> GSM39135 2 0.4702 0.872 0.000 0.788 0.212
#> GSM39136 2 0.4504 0.883 0.000 0.804 0.196
#> GSM39137 3 0.2959 0.607 0.000 0.100 0.900
#> GSM39138 2 0.4002 0.899 0.000 0.840 0.160
#> GSM39139 2 0.4002 0.899 0.000 0.840 0.160
#> GSM39140 3 0.5502 0.739 0.248 0.008 0.744
#> GSM39141 3 0.5502 0.739 0.248 0.008 0.744
#> GSM39142 3 0.5216 0.731 0.260 0.000 0.740
#> GSM39143 3 0.5502 0.739 0.248 0.008 0.744
#> GSM39144 2 0.4002 0.899 0.000 0.840 0.160
#> GSM39145 2 0.4062 0.898 0.000 0.836 0.164
#> GSM39146 2 0.6126 0.631 0.000 0.600 0.400
#> GSM39147 2 0.6008 0.701 0.000 0.628 0.372
#> GSM39188 1 0.6083 0.787 0.772 0.168 0.060
#> GSM39189 1 0.6083 0.787 0.772 0.168 0.060
#> GSM39190 1 0.6083 0.787 0.772 0.168 0.060
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM39104 1 0.7536 0.662 0.488 0.284 0.228 0.000
#> GSM39105 1 0.6843 0.661 0.532 0.356 0.112 0.000
#> GSM39106 2 0.6377 0.237 0.256 0.632 0.112 0.000
#> GSM39107 2 0.4142 0.570 0.080 0.844 0.012 0.064
#> GSM39108 2 0.7328 -0.378 0.392 0.452 0.156 0.000
#> GSM39109 2 0.7223 0.240 0.100 0.528 0.356 0.016
#> GSM39110 2 0.7381 -0.141 0.324 0.512 0.160 0.004
#> GSM39111 1 0.7692 0.631 0.456 0.272 0.272 0.000
#> GSM39112 2 0.4274 0.548 0.116 0.832 0.028 0.024
#> GSM39113 2 0.3558 0.578 0.048 0.872 0.008 0.072
#> GSM39114 2 0.5500 -0.323 0.016 0.520 0.000 0.464
#> GSM39115 1 0.6641 0.741 0.600 0.276 0.124 0.000
#> GSM39148 1 0.5085 0.679 0.616 0.376 0.008 0.000
#> GSM39149 3 0.1716 0.881 0.064 0.000 0.936 0.000
#> GSM39150 1 0.6187 0.320 0.516 0.052 0.432 0.000
#> GSM39151 3 0.1716 0.881 0.064 0.000 0.936 0.000
#> GSM39152 3 0.2011 0.873 0.080 0.000 0.920 0.000
#> GSM39153 1 0.6005 0.738 0.616 0.324 0.060 0.000
#> GSM39154 1 0.6350 0.746 0.612 0.296 0.092 0.000
#> GSM39155 1 0.5848 0.731 0.616 0.336 0.048 0.000
#> GSM39156 2 0.5168 0.351 0.248 0.712 0.040 0.000
#> GSM39157 1 0.5666 0.722 0.616 0.348 0.036 0.000
#> GSM39158 1 0.6536 0.537 0.580 0.096 0.324 0.000
#> GSM39159 3 0.4761 0.353 0.372 0.000 0.628 0.000
#> GSM39160 1 0.5604 0.168 0.504 0.020 0.476 0.000
#> GSM39161 3 0.4086 0.689 0.216 0.000 0.776 0.008
#> GSM39162 1 0.5236 0.577 0.560 0.432 0.008 0.000
#> GSM39163 1 0.6141 0.743 0.616 0.312 0.072 0.000
#> GSM39164 1 0.5599 0.717 0.616 0.352 0.032 0.000
#> GSM39165 1 0.5493 0.192 0.528 0.016 0.456 0.000
#> GSM39166 1 0.5088 0.266 0.572 0.004 0.424 0.000
#> GSM39167 1 0.6052 0.741 0.616 0.320 0.064 0.000
#> GSM39168 1 0.5138 0.656 0.600 0.392 0.008 0.000
#> GSM39169 1 0.5682 0.716 0.612 0.352 0.036 0.000
#> GSM39170 1 0.6761 0.719 0.608 0.224 0.168 0.000
#> GSM39171 1 0.7412 0.641 0.504 0.200 0.296 0.000
#> GSM39172 3 0.1042 0.847 0.020 0.000 0.972 0.008
#> GSM39173 3 0.1716 0.881 0.064 0.000 0.936 0.000
#> GSM39174 1 0.5666 0.720 0.616 0.348 0.036 0.000
#> GSM39175 1 0.6501 0.743 0.616 0.268 0.116 0.000
#> GSM39176 1 0.6295 0.746 0.616 0.296 0.088 0.000
#> GSM39177 3 0.2081 0.874 0.084 0.000 0.916 0.000
#> GSM39178 3 0.4898 0.166 0.416 0.000 0.584 0.000
#> GSM39179 3 0.1716 0.881 0.064 0.000 0.936 0.000
#> GSM39180 3 0.0336 0.840 0.000 0.000 0.992 0.008
#> GSM39181 1 0.4998 0.123 0.512 0.000 0.488 0.000
#> GSM39182 3 0.1639 0.843 0.036 0.004 0.952 0.008
#> GSM39183 1 0.4977 0.185 0.540 0.000 0.460 0.000
#> GSM39184 1 0.6661 0.741 0.604 0.264 0.132 0.000
#> GSM39185 3 0.3681 0.740 0.176 0.000 0.816 0.008
#> GSM39186 1 0.5599 0.717 0.616 0.352 0.032 0.000
#> GSM39187 1 0.6039 0.721 0.596 0.348 0.056 0.000
#> GSM39116 4 0.4019 0.713 0.012 0.196 0.000 0.792
#> GSM39117 4 0.6197 0.600 0.364 0.024 0.024 0.588
#> GSM39118 4 0.1545 0.764 0.008 0.040 0.000 0.952
#> GSM39119 4 0.1716 0.762 0.064 0.000 0.000 0.936
#> GSM39120 2 0.4105 0.514 0.156 0.812 0.032 0.000
#> GSM39121 2 0.2654 0.544 0.004 0.888 0.000 0.108
#> GSM39122 2 0.2714 0.540 0.004 0.884 0.000 0.112
#> GSM39123 4 0.6237 0.601 0.376 0.024 0.024 0.576
#> GSM39124 2 0.5691 -0.342 0.024 0.508 0.000 0.468
#> GSM39125 2 0.4423 0.499 0.168 0.792 0.040 0.000
#> GSM39126 2 0.3099 0.562 0.020 0.876 0.000 0.104
#> GSM39127 4 0.4999 0.590 0.012 0.328 0.000 0.660
#> GSM39128 2 0.6147 -0.366 0.048 0.488 0.000 0.464
#> GSM39129 4 0.1389 0.767 0.048 0.000 0.000 0.952
#> GSM39130 4 0.6197 0.600 0.364 0.024 0.024 0.588
#> GSM39131 2 0.6147 -0.368 0.048 0.488 0.000 0.464
#> GSM39132 4 0.5127 0.553 0.012 0.356 0.000 0.632
#> GSM39133 4 0.5839 0.613 0.376 0.020 0.012 0.592
#> GSM39134 4 0.1118 0.768 0.036 0.000 0.000 0.964
#> GSM39135 4 0.4175 0.700 0.012 0.212 0.000 0.776
#> GSM39136 4 0.3577 0.732 0.012 0.156 0.000 0.832
#> GSM39137 2 0.3842 0.475 0.036 0.836 0.000 0.128
#> GSM39138 4 0.1389 0.767 0.048 0.000 0.000 0.952
#> GSM39139 4 0.1798 0.768 0.040 0.016 0.000 0.944
#> GSM39140 2 0.4707 0.449 0.204 0.760 0.036 0.000
#> GSM39141 2 0.4323 0.485 0.184 0.788 0.028 0.000
#> GSM39142 2 0.4910 0.317 0.276 0.704 0.020 0.000
#> GSM39143 2 0.4579 0.459 0.200 0.768 0.032 0.000
#> GSM39144 4 0.1389 0.767 0.048 0.000 0.000 0.952
#> GSM39145 4 0.3032 0.746 0.008 0.124 0.000 0.868
#> GSM39146 4 0.5865 0.573 0.048 0.340 0.000 0.612
#> GSM39147 4 0.5290 0.482 0.012 0.404 0.000 0.584
#> GSM39188 3 0.1716 0.881 0.064 0.000 0.936 0.000
#> GSM39189 3 0.1792 0.880 0.068 0.000 0.932 0.000
#> GSM39190 3 0.1716 0.881 0.064 0.000 0.936 0.000
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM39104 1 0.4170 0.7871 0.820 0.020 0.080 0.008 0.072
#> GSM39105 1 0.3889 0.8079 0.808 0.008 0.032 0.004 0.148
#> GSM39106 5 0.4852 0.7071 0.116 0.020 0.096 0.004 0.764
#> GSM39107 5 0.3715 0.6923 0.036 0.136 0.004 0.004 0.820
#> GSM39108 5 0.6810 0.3045 0.336 0.020 0.132 0.008 0.504
#> GSM39109 5 0.7733 0.2713 0.084 0.192 0.244 0.004 0.476
#> GSM39110 5 0.6645 0.5393 0.216 0.036 0.152 0.004 0.592
#> GSM39111 1 0.5286 0.6978 0.736 0.036 0.140 0.004 0.084
#> GSM39112 5 0.2977 0.7240 0.052 0.060 0.004 0.004 0.880
#> GSM39113 5 0.3768 0.6782 0.028 0.156 0.004 0.004 0.808
#> GSM39114 5 0.4713 0.0611 0.000 0.440 0.000 0.016 0.544
#> GSM39115 1 0.2777 0.8395 0.864 0.000 0.016 0.000 0.120
#> GSM39148 1 0.2179 0.8419 0.888 0.000 0.000 0.000 0.112
#> GSM39149 3 0.1798 0.8472 0.064 0.004 0.928 0.000 0.004
#> GSM39150 1 0.1697 0.8047 0.932 0.008 0.060 0.000 0.000
#> GSM39151 3 0.1544 0.8500 0.068 0.000 0.932 0.000 0.000
#> GSM39152 3 0.3360 0.7894 0.168 0.012 0.816 0.000 0.004
#> GSM39153 1 0.1965 0.8454 0.904 0.000 0.000 0.000 0.096
#> GSM39154 1 0.2352 0.8477 0.896 0.004 0.008 0.000 0.092
#> GSM39155 1 0.2516 0.8318 0.860 0.000 0.000 0.000 0.140
#> GSM39156 5 0.3291 0.7360 0.120 0.000 0.040 0.000 0.840
#> GSM39157 1 0.2690 0.8209 0.844 0.000 0.000 0.000 0.156
#> GSM39158 1 0.1243 0.8168 0.960 0.004 0.028 0.000 0.008
#> GSM39159 1 0.6356 0.2605 0.572 0.184 0.232 0.000 0.012
#> GSM39160 1 0.1943 0.8003 0.924 0.020 0.056 0.000 0.000
#> GSM39161 1 0.7334 -0.0283 0.468 0.196 0.300 0.020 0.016
#> GSM39162 1 0.2970 0.8060 0.828 0.000 0.004 0.000 0.168
#> GSM39163 1 0.2233 0.8445 0.892 0.000 0.004 0.000 0.104
#> GSM39164 1 0.2233 0.8437 0.892 0.000 0.004 0.000 0.104
#> GSM39165 1 0.2670 0.7764 0.888 0.016 0.088 0.004 0.004
#> GSM39166 1 0.1911 0.7987 0.932 0.028 0.036 0.000 0.004
#> GSM39167 1 0.2329 0.8397 0.876 0.000 0.000 0.000 0.124
#> GSM39168 1 0.2488 0.8399 0.872 0.000 0.004 0.000 0.124
#> GSM39169 1 0.2233 0.8431 0.892 0.000 0.004 0.000 0.104
#> GSM39170 1 0.1243 0.8338 0.960 0.004 0.008 0.000 0.028
#> GSM39171 1 0.2331 0.8044 0.908 0.008 0.068 0.000 0.016
#> GSM39172 3 0.6574 0.5680 0.252 0.152 0.572 0.008 0.016
#> GSM39173 3 0.1704 0.8498 0.068 0.000 0.928 0.000 0.004
#> GSM39174 1 0.2179 0.8418 0.888 0.000 0.000 0.000 0.112
#> GSM39175 1 0.0865 0.8334 0.972 0.000 0.004 0.000 0.024
#> GSM39176 1 0.2280 0.8424 0.880 0.000 0.000 0.000 0.120
#> GSM39177 3 0.2798 0.8242 0.140 0.008 0.852 0.000 0.000
#> GSM39178 1 0.4456 0.6461 0.768 0.072 0.152 0.000 0.008
#> GSM39179 3 0.1544 0.8500 0.068 0.000 0.932 0.000 0.000
#> GSM39180 3 0.4548 0.7201 0.092 0.108 0.780 0.000 0.020
#> GSM39181 1 0.3856 0.7389 0.840 0.056 0.076 0.020 0.008
#> GSM39182 3 0.7572 0.4649 0.280 0.196 0.472 0.020 0.032
#> GSM39183 1 0.3334 0.7540 0.864 0.048 0.072 0.004 0.012
#> GSM39184 1 0.1952 0.8469 0.912 0.000 0.004 0.000 0.084
#> GSM39185 1 0.7171 -0.1640 0.432 0.196 0.348 0.008 0.016
#> GSM39186 1 0.2516 0.8319 0.860 0.000 0.000 0.000 0.140
#> GSM39187 1 0.2773 0.8142 0.836 0.000 0.000 0.000 0.164
#> GSM39116 2 0.5498 0.4449 0.000 0.580 0.000 0.340 0.080
#> GSM39117 4 0.1043 0.7942 0.000 0.040 0.000 0.960 0.000
#> GSM39118 2 0.4392 0.3544 0.000 0.612 0.000 0.380 0.008
#> GSM39119 4 0.4291 -0.2331 0.000 0.464 0.000 0.536 0.000
#> GSM39120 5 0.1894 0.7430 0.072 0.000 0.008 0.000 0.920
#> GSM39121 5 0.3779 0.5712 0.000 0.236 0.000 0.012 0.752
#> GSM39122 5 0.3642 0.5813 0.000 0.232 0.000 0.008 0.760
#> GSM39123 4 0.1121 0.7957 0.000 0.044 0.000 0.956 0.000
#> GSM39124 2 0.4740 0.1061 0.000 0.516 0.000 0.016 0.468
#> GSM39125 5 0.2054 0.7430 0.072 0.004 0.008 0.000 0.916
#> GSM39126 5 0.3578 0.6083 0.004 0.204 0.000 0.008 0.784
#> GSM39127 2 0.6211 0.4692 0.000 0.548 0.000 0.248 0.204
#> GSM39128 2 0.5542 0.1765 0.000 0.500 0.000 0.068 0.432
#> GSM39129 2 0.4283 0.1955 0.000 0.544 0.000 0.456 0.000
#> GSM39130 4 0.0963 0.7965 0.000 0.036 0.000 0.964 0.000
#> GSM39131 2 0.5348 0.0937 0.000 0.492 0.000 0.052 0.456
#> GSM39132 2 0.5775 0.4792 0.000 0.608 0.000 0.148 0.244
#> GSM39133 4 0.1792 0.7722 0.000 0.084 0.000 0.916 0.000
#> GSM39134 2 0.4294 0.2136 0.000 0.532 0.000 0.468 0.000
#> GSM39135 2 0.5498 0.4459 0.000 0.580 0.000 0.340 0.080
#> GSM39136 2 0.5562 0.3846 0.000 0.520 0.000 0.408 0.072
#> GSM39137 5 0.4836 0.4048 0.000 0.304 0.000 0.044 0.652
#> GSM39138 2 0.4283 0.1955 0.000 0.544 0.000 0.456 0.000
#> GSM39139 2 0.4249 0.2488 0.000 0.568 0.000 0.432 0.000
#> GSM39140 5 0.2707 0.7436 0.100 0.000 0.024 0.000 0.876
#> GSM39141 5 0.2740 0.7439 0.096 0.000 0.028 0.000 0.876
#> GSM39142 5 0.4021 0.6931 0.200 0.000 0.036 0.000 0.764
#> GSM39143 5 0.2959 0.7422 0.100 0.000 0.036 0.000 0.864
#> GSM39144 2 0.4283 0.1955 0.000 0.544 0.000 0.456 0.000
#> GSM39145 2 0.4836 0.4119 0.000 0.628 0.000 0.336 0.036
#> GSM39146 2 0.5902 0.4735 0.000 0.600 0.000 0.208 0.192
#> GSM39147 2 0.5606 0.4405 0.000 0.600 0.000 0.104 0.296
#> GSM39188 3 0.1544 0.8500 0.068 0.000 0.932 0.000 0.000
#> GSM39189 3 0.2719 0.8368 0.068 0.048 0.884 0.000 0.000
#> GSM39190 3 0.1894 0.8503 0.072 0.008 0.920 0.000 0.000
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM39104 1 0.3451 0.6977 0.840 0.000 0.052 0.076 0.028 0.004
#> GSM39105 1 0.1873 0.7487 0.924 0.000 0.020 0.048 0.008 0.000
#> GSM39106 1 0.7766 0.3894 0.496 0.156 0.032 0.152 0.144 0.020
#> GSM39107 2 0.6654 -0.0519 0.344 0.468 0.000 0.076 0.104 0.008
#> GSM39108 1 0.6585 0.5350 0.640 0.092 0.064 0.116 0.076 0.012
#> GSM39109 5 0.7382 -0.1138 0.392 0.040 0.060 0.096 0.396 0.016
#> GSM39110 1 0.7592 0.4333 0.544 0.108 0.080 0.140 0.112 0.016
#> GSM39111 1 0.4008 0.6595 0.792 0.000 0.092 0.088 0.028 0.000
#> GSM39112 1 0.7446 0.3338 0.456 0.236 0.000 0.132 0.156 0.020
#> GSM39113 2 0.5617 0.4031 0.164 0.664 0.000 0.080 0.088 0.004
#> GSM39114 2 0.1003 0.7416 0.000 0.964 0.000 0.020 0.000 0.016
#> GSM39115 1 0.0520 0.7696 0.984 0.000 0.000 0.008 0.008 0.000
#> GSM39148 1 0.0622 0.7696 0.980 0.000 0.000 0.012 0.008 0.000
#> GSM39149 3 0.0260 0.9745 0.000 0.000 0.992 0.008 0.000 0.000
#> GSM39150 1 0.1353 0.7596 0.952 0.000 0.024 0.012 0.012 0.000
#> GSM39151 3 0.0405 0.9755 0.004 0.000 0.988 0.008 0.000 0.000
#> GSM39152 3 0.1536 0.9392 0.024 0.000 0.944 0.020 0.012 0.000
#> GSM39153 1 0.0146 0.7685 0.996 0.000 0.000 0.004 0.000 0.000
#> GSM39154 1 0.0260 0.7680 0.992 0.000 0.000 0.000 0.008 0.000
#> GSM39155 1 0.0405 0.7692 0.988 0.000 0.000 0.008 0.004 0.000
#> GSM39156 1 0.7441 0.3981 0.500 0.172 0.008 0.132 0.168 0.020
#> GSM39157 1 0.0291 0.7693 0.992 0.000 0.000 0.004 0.004 0.000
#> GSM39158 1 0.0806 0.7658 0.972 0.000 0.000 0.008 0.020 0.000
#> GSM39159 5 0.4458 0.5548 0.352 0.000 0.040 0.000 0.608 0.000
#> GSM39160 1 0.1788 0.7445 0.928 0.000 0.028 0.004 0.040 0.000
#> GSM39161 5 0.3669 0.6381 0.208 0.000 0.028 0.000 0.760 0.004
#> GSM39162 1 0.1237 0.7644 0.956 0.000 0.004 0.020 0.020 0.000
#> GSM39163 1 0.0405 0.7681 0.988 0.000 0.000 0.004 0.008 0.000
#> GSM39164 1 0.0000 0.7680 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM39165 1 0.2103 0.7447 0.916 0.000 0.040 0.024 0.020 0.000
#> GSM39166 1 0.1644 0.7193 0.920 0.000 0.000 0.004 0.076 0.000
#> GSM39167 1 0.0405 0.7690 0.988 0.000 0.000 0.008 0.004 0.000
#> GSM39168 1 0.0912 0.7679 0.972 0.004 0.004 0.008 0.012 0.000
#> GSM39169 1 0.0260 0.7693 0.992 0.000 0.000 0.000 0.008 0.000
#> GSM39170 1 0.0405 0.7682 0.988 0.000 0.000 0.004 0.008 0.000
#> GSM39171 1 0.1821 0.7473 0.928 0.000 0.040 0.008 0.024 0.000
#> GSM39172 5 0.4002 0.5711 0.068 0.000 0.188 0.000 0.744 0.000
#> GSM39173 3 0.0146 0.9754 0.000 0.000 0.996 0.000 0.004 0.000
#> GSM39174 1 0.0000 0.7680 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM39175 1 0.0146 0.7679 0.996 0.000 0.000 0.000 0.004 0.000
#> GSM39176 1 0.0520 0.7689 0.984 0.000 0.000 0.008 0.008 0.000
#> GSM39177 3 0.0870 0.9722 0.012 0.000 0.972 0.004 0.012 0.000
#> GSM39178 1 0.5224 -0.3518 0.468 0.000 0.092 0.000 0.440 0.000
#> GSM39179 3 0.0260 0.9748 0.000 0.000 0.992 0.000 0.008 0.000
#> GSM39180 5 0.4176 0.1799 0.016 0.000 0.404 0.000 0.580 0.000
#> GSM39181 1 0.3265 0.4695 0.748 0.000 0.000 0.000 0.248 0.004
#> GSM39182 5 0.3745 0.6204 0.100 0.000 0.116 0.000 0.784 0.000
#> GSM39183 1 0.3464 0.2769 0.688 0.000 0.000 0.000 0.312 0.000
#> GSM39184 1 0.0146 0.7679 0.996 0.000 0.000 0.000 0.004 0.000
#> GSM39185 5 0.3618 0.6434 0.192 0.000 0.040 0.000 0.768 0.000
#> GSM39186 1 0.0146 0.7685 0.996 0.000 0.000 0.004 0.000 0.000
#> GSM39187 1 0.0520 0.7695 0.984 0.000 0.000 0.008 0.008 0.000
#> GSM39116 2 0.4820 0.5460 0.000 0.652 0.000 0.088 0.004 0.256
#> GSM39117 4 0.3341 0.9722 0.000 0.012 0.000 0.776 0.004 0.208
#> GSM39118 6 0.2755 0.8147 0.000 0.140 0.000 0.012 0.004 0.844
#> GSM39119 6 0.3689 0.7701 0.000 0.068 0.000 0.136 0.004 0.792
#> GSM39120 1 0.7520 0.3574 0.468 0.204 0.004 0.124 0.180 0.020
#> GSM39121 2 0.0909 0.7272 0.000 0.968 0.000 0.020 0.012 0.000
#> GSM39122 2 0.0909 0.7272 0.000 0.968 0.000 0.020 0.012 0.000
#> GSM39123 4 0.3504 0.9707 0.000 0.024 0.000 0.776 0.004 0.196
#> GSM39124 2 0.0777 0.7446 0.000 0.972 0.000 0.004 0.000 0.024
#> GSM39125 1 0.7525 0.3671 0.476 0.204 0.008 0.112 0.180 0.020
#> GSM39126 2 0.1341 0.7149 0.000 0.948 0.000 0.028 0.024 0.000
#> GSM39127 2 0.4569 0.6140 0.000 0.700 0.000 0.096 0.004 0.200
#> GSM39128 2 0.1528 0.7442 0.000 0.936 0.000 0.016 0.000 0.048
#> GSM39129 6 0.1082 0.8951 0.000 0.040 0.000 0.004 0.000 0.956
#> GSM39130 4 0.3230 0.9708 0.000 0.012 0.000 0.776 0.000 0.212
#> GSM39131 2 0.1367 0.7451 0.000 0.944 0.000 0.012 0.000 0.044
#> GSM39132 2 0.4152 0.6051 0.000 0.712 0.000 0.044 0.004 0.240
#> GSM39133 4 0.3543 0.9517 0.000 0.032 0.000 0.768 0.000 0.200
#> GSM39134 6 0.2696 0.8508 0.000 0.048 0.000 0.076 0.004 0.872
#> GSM39135 2 0.4688 0.5267 0.000 0.644 0.000 0.064 0.004 0.288
#> GSM39136 2 0.5167 0.5320 0.000 0.632 0.000 0.148 0.004 0.216
#> GSM39137 2 0.0725 0.7397 0.000 0.976 0.000 0.012 0.000 0.012
#> GSM39138 6 0.1257 0.8881 0.000 0.028 0.000 0.020 0.000 0.952
#> GSM39139 6 0.1075 0.8945 0.000 0.048 0.000 0.000 0.000 0.952
#> GSM39140 1 0.7577 0.3707 0.476 0.184 0.008 0.132 0.180 0.020
#> GSM39141 1 0.7526 0.3807 0.484 0.180 0.008 0.128 0.180 0.020
#> GSM39142 1 0.7013 0.4444 0.556 0.172 0.008 0.104 0.140 0.020
#> GSM39143 1 0.7548 0.3759 0.480 0.184 0.008 0.128 0.180 0.020
#> GSM39144 6 0.1082 0.8951 0.000 0.040 0.000 0.004 0.000 0.956
#> GSM39145 6 0.2165 0.8514 0.000 0.108 0.000 0.008 0.000 0.884
#> GSM39146 2 0.4091 0.6314 0.000 0.732 0.000 0.052 0.004 0.212
#> GSM39147 2 0.2667 0.7129 0.000 0.852 0.000 0.020 0.000 0.128
#> GSM39188 3 0.0291 0.9768 0.004 0.000 0.992 0.000 0.004 0.000
#> GSM39189 3 0.1074 0.9568 0.012 0.000 0.960 0.000 0.028 0.000
#> GSM39190 3 0.0520 0.9756 0.008 0.000 0.984 0.000 0.008 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) other(p) protocol(p) k
#> MAD:mclust 69 0.20904 9.91e-10 1.26e-08 2
#> MAD:mclust 81 0.00209 5.35e-13 6.46e-10 3
#> MAD:mclust 63 0.00776 4.26e-10 5.61e-09 4
#> MAD:mclust 61 0.07864 1.02e-06 9.16e-08 5
#> MAD:mclust 70 0.38169 2.95e-08 7.73e-08 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 8353 rows and 87 columns.
#> Top rows (835, 1670, 2506, 3341, 4176) 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 0.768 0.883 0.949 0.4956 0.502 0.502
#> 3 3 0.413 0.417 0.637 0.3163 0.628 0.396
#> 4 4 0.410 0.438 0.694 0.0962 0.761 0.463
#> 5 5 0.492 0.424 0.642 0.0977 0.782 0.414
#> 6 6 0.606 0.513 0.689 0.0469 0.914 0.651
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
#> GSM39104 1 0.0000 0.951 1.000 0.000
#> GSM39105 1 0.0000 0.951 1.000 0.000
#> GSM39106 1 0.0376 0.948 0.996 0.004
#> GSM39107 1 0.8713 0.589 0.708 0.292
#> GSM39108 1 0.0000 0.951 1.000 0.000
#> GSM39109 2 0.7219 0.749 0.200 0.800
#> GSM39110 1 0.0000 0.951 1.000 0.000
#> GSM39111 1 0.0000 0.951 1.000 0.000
#> GSM39112 1 0.3879 0.890 0.924 0.076
#> GSM39113 1 0.9795 0.285 0.584 0.416
#> GSM39114 2 0.0000 0.934 0.000 1.000
#> GSM39115 1 0.0000 0.951 1.000 0.000
#> GSM39148 1 0.0000 0.951 1.000 0.000
#> GSM39149 2 0.7815 0.712 0.232 0.768
#> GSM39150 1 0.0000 0.951 1.000 0.000
#> GSM39151 2 0.7674 0.722 0.224 0.776
#> GSM39152 1 0.2423 0.921 0.960 0.040
#> GSM39153 1 0.0000 0.951 1.000 0.000
#> GSM39154 1 0.0000 0.951 1.000 0.000
#> GSM39155 1 0.0000 0.951 1.000 0.000
#> GSM39156 1 0.0000 0.951 1.000 0.000
#> GSM39157 1 0.0000 0.951 1.000 0.000
#> GSM39158 1 0.0000 0.951 1.000 0.000
#> GSM39159 1 0.1414 0.937 0.980 0.020
#> GSM39160 1 0.0000 0.951 1.000 0.000
#> GSM39161 1 0.5178 0.842 0.884 0.116
#> GSM39162 1 0.0000 0.951 1.000 0.000
#> GSM39163 1 0.0000 0.951 1.000 0.000
#> GSM39164 1 0.0000 0.951 1.000 0.000
#> GSM39165 1 0.0000 0.951 1.000 0.000
#> GSM39166 1 0.0000 0.951 1.000 0.000
#> GSM39167 1 0.0000 0.951 1.000 0.000
#> GSM39168 1 0.0000 0.951 1.000 0.000
#> GSM39169 1 0.0000 0.951 1.000 0.000
#> GSM39170 1 0.0000 0.951 1.000 0.000
#> GSM39171 1 0.0000 0.951 1.000 0.000
#> GSM39172 2 0.1184 0.927 0.016 0.984
#> GSM39173 2 0.1414 0.924 0.020 0.980
#> GSM39174 1 0.0000 0.951 1.000 0.000
#> GSM39175 1 0.0000 0.951 1.000 0.000
#> GSM39176 1 0.0000 0.951 1.000 0.000
#> GSM39177 1 0.9608 0.344 0.616 0.384
#> GSM39178 1 0.0000 0.951 1.000 0.000
#> GSM39179 2 0.0938 0.929 0.012 0.988
#> GSM39180 2 0.0000 0.934 0.000 1.000
#> GSM39181 1 0.0000 0.951 1.000 0.000
#> GSM39182 2 0.3879 0.884 0.076 0.924
#> GSM39183 1 0.0000 0.951 1.000 0.000
#> GSM39184 1 0.0000 0.951 1.000 0.000
#> GSM39185 1 0.8909 0.531 0.692 0.308
#> GSM39186 1 0.0000 0.951 1.000 0.000
#> GSM39187 1 0.0000 0.951 1.000 0.000
#> GSM39116 2 0.0000 0.934 0.000 1.000
#> GSM39117 2 0.0000 0.934 0.000 1.000
#> GSM39118 2 0.0000 0.934 0.000 1.000
#> GSM39119 2 0.0000 0.934 0.000 1.000
#> GSM39120 1 0.7376 0.733 0.792 0.208
#> GSM39121 2 0.9460 0.439 0.364 0.636
#> GSM39122 2 0.8386 0.642 0.268 0.732
#> GSM39123 2 0.0000 0.934 0.000 1.000
#> GSM39124 2 0.0000 0.934 0.000 1.000
#> GSM39125 1 0.7056 0.754 0.808 0.192
#> GSM39126 2 0.7453 0.732 0.212 0.788
#> GSM39127 2 0.0000 0.934 0.000 1.000
#> GSM39128 2 0.0000 0.934 0.000 1.000
#> GSM39129 2 0.0000 0.934 0.000 1.000
#> GSM39130 2 0.0000 0.934 0.000 1.000
#> GSM39131 2 0.0000 0.934 0.000 1.000
#> GSM39132 2 0.0000 0.934 0.000 1.000
#> GSM39133 2 0.0000 0.934 0.000 1.000
#> GSM39134 2 0.0000 0.934 0.000 1.000
#> GSM39135 2 0.0000 0.934 0.000 1.000
#> GSM39136 2 0.0000 0.934 0.000 1.000
#> GSM39137 2 0.0376 0.932 0.004 0.996
#> GSM39138 2 0.0000 0.934 0.000 1.000
#> GSM39139 2 0.0000 0.934 0.000 1.000
#> GSM39140 1 0.4298 0.877 0.912 0.088
#> GSM39141 1 0.0376 0.948 0.996 0.004
#> GSM39142 1 0.0000 0.951 1.000 0.000
#> GSM39143 1 0.0000 0.951 1.000 0.000
#> GSM39144 2 0.0000 0.934 0.000 1.000
#> GSM39145 2 0.0000 0.934 0.000 1.000
#> GSM39146 2 0.0000 0.934 0.000 1.000
#> GSM39147 2 0.0000 0.934 0.000 1.000
#> GSM39188 2 0.5519 0.836 0.128 0.872
#> GSM39189 2 0.9833 0.303 0.424 0.576
#> GSM39190 2 0.4161 0.878 0.084 0.916
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM39104 1 0.4842 0.5702 0.776 0.000 0.224
#> GSM39105 1 0.6192 0.3288 0.580 0.000 0.420
#> GSM39106 3 0.4931 0.4731 0.232 0.000 0.768
#> GSM39107 3 0.0983 0.5914 0.016 0.004 0.980
#> GSM39108 1 0.6305 0.1673 0.516 0.000 0.484
#> GSM39109 2 0.8977 0.3027 0.188 0.560 0.252
#> GSM39110 3 0.6286 -0.0636 0.464 0.000 0.536
#> GSM39111 1 0.4235 0.5888 0.824 0.000 0.176
#> GSM39112 3 0.1860 0.6126 0.052 0.000 0.948
#> GSM39113 3 0.0983 0.5747 0.004 0.016 0.980
#> GSM39114 3 0.4452 0.3535 0.000 0.192 0.808
#> GSM39115 1 0.6307 0.1623 0.512 0.000 0.488
#> GSM39148 3 0.6286 -0.0680 0.464 0.000 0.536
#> GSM39149 1 0.6140 0.2739 0.596 0.404 0.000
#> GSM39150 1 0.2448 0.6058 0.924 0.000 0.076
#> GSM39151 1 0.6225 0.2316 0.568 0.432 0.000
#> GSM39152 1 0.4702 0.5001 0.788 0.212 0.000
#> GSM39153 1 0.5529 0.5196 0.704 0.000 0.296
#> GSM39154 1 0.5058 0.5593 0.756 0.000 0.244
#> GSM39155 1 0.5497 0.5234 0.708 0.000 0.292
#> GSM39156 3 0.4452 0.5239 0.192 0.000 0.808
#> GSM39157 1 0.6244 0.2872 0.560 0.000 0.440
#> GSM39158 1 0.5560 0.5168 0.700 0.000 0.300
#> GSM39159 1 0.3192 0.5619 0.888 0.112 0.000
#> GSM39160 1 0.1643 0.6012 0.956 0.000 0.044
#> GSM39161 1 0.4931 0.4817 0.768 0.232 0.000
#> GSM39162 3 0.5926 0.2315 0.356 0.000 0.644
#> GSM39163 1 0.5810 0.4712 0.664 0.000 0.336
#> GSM39164 1 0.6111 0.3791 0.604 0.000 0.396
#> GSM39165 1 0.0829 0.5936 0.984 0.004 0.012
#> GSM39166 1 0.3192 0.6061 0.888 0.000 0.112
#> GSM39167 1 0.6244 0.2893 0.560 0.000 0.440
#> GSM39168 3 0.6204 0.0549 0.424 0.000 0.576
#> GSM39169 1 0.5529 0.5168 0.704 0.000 0.296
#> GSM39170 1 0.5138 0.5556 0.748 0.000 0.252
#> GSM39171 1 0.3192 0.6061 0.888 0.000 0.112
#> GSM39172 1 0.6305 0.1231 0.516 0.484 0.000
#> GSM39173 2 0.6192 0.0434 0.420 0.580 0.000
#> GSM39174 1 0.5760 0.4796 0.672 0.000 0.328
#> GSM39175 1 0.3500 0.6064 0.880 0.004 0.116
#> GSM39176 1 0.6095 0.3868 0.608 0.000 0.392
#> GSM39177 1 0.5529 0.4167 0.704 0.296 0.000
#> GSM39178 1 0.2796 0.5676 0.908 0.092 0.000
#> GSM39179 1 0.6302 0.1319 0.520 0.480 0.000
#> GSM39180 2 0.5621 0.3157 0.308 0.692 0.000
#> GSM39181 1 0.3116 0.6065 0.892 0.000 0.108
#> GSM39182 1 0.6309 0.0944 0.504 0.496 0.000
#> GSM39183 1 0.2537 0.6063 0.920 0.000 0.080
#> GSM39184 1 0.4887 0.5674 0.772 0.000 0.228
#> GSM39185 1 0.5431 0.4313 0.716 0.284 0.000
#> GSM39186 1 0.5650 0.5023 0.688 0.000 0.312
#> GSM39187 3 0.6299 -0.1048 0.476 0.000 0.524
#> GSM39116 2 0.6026 0.5764 0.000 0.624 0.376
#> GSM39117 2 0.1163 0.6653 0.028 0.972 0.000
#> GSM39118 2 0.4605 0.6977 0.000 0.796 0.204
#> GSM39119 2 0.2165 0.7162 0.000 0.936 0.064
#> GSM39120 3 0.2301 0.6129 0.060 0.004 0.936
#> GSM39121 3 0.2261 0.5238 0.000 0.068 0.932
#> GSM39122 3 0.4002 0.4172 0.000 0.160 0.840
#> GSM39123 2 0.1129 0.6952 0.004 0.976 0.020
#> GSM39124 3 0.5882 0.0124 0.000 0.348 0.652
#> GSM39125 3 0.3038 0.5934 0.104 0.000 0.896
#> GSM39126 3 0.3340 0.4710 0.000 0.120 0.880
#> GSM39127 3 0.6291 -0.3169 0.000 0.468 0.532
#> GSM39128 3 0.6140 -0.1570 0.000 0.404 0.596
#> GSM39129 2 0.2711 0.7221 0.000 0.912 0.088
#> GSM39130 2 0.0892 0.6973 0.000 0.980 0.020
#> GSM39131 3 0.6204 -0.2105 0.000 0.424 0.576
#> GSM39132 2 0.6309 0.3545 0.000 0.504 0.496
#> GSM39133 2 0.3551 0.7235 0.000 0.868 0.132
#> GSM39134 2 0.3551 0.7236 0.000 0.868 0.132
#> GSM39135 2 0.6095 0.5545 0.000 0.608 0.392
#> GSM39136 2 0.6008 0.5809 0.000 0.628 0.372
#> GSM39137 3 0.5497 0.1610 0.000 0.292 0.708
#> GSM39138 2 0.2448 0.7197 0.000 0.924 0.076
#> GSM39139 2 0.5760 0.6216 0.000 0.672 0.328
#> GSM39140 3 0.3686 0.5728 0.140 0.000 0.860
#> GSM39141 3 0.3816 0.5673 0.148 0.000 0.852
#> GSM39142 3 0.4750 0.4944 0.216 0.000 0.784
#> GSM39143 3 0.4002 0.5572 0.160 0.000 0.840
#> GSM39144 2 0.3482 0.7240 0.000 0.872 0.128
#> GSM39145 2 0.5968 0.5906 0.000 0.636 0.364
#> GSM39146 2 0.6244 0.4707 0.000 0.560 0.440
#> GSM39147 3 0.6299 -0.3444 0.000 0.476 0.524
#> GSM39188 1 0.6295 0.1549 0.528 0.472 0.000
#> GSM39189 1 0.6008 0.3188 0.628 0.372 0.000
#> GSM39190 1 0.6299 0.1471 0.524 0.476 0.000
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM39104 1 0.606 0.6711 0.720 0.064 0.180 0.036
#> GSM39105 1 0.637 0.5928 0.692 0.200 0.076 0.032
#> GSM39106 2 0.707 -0.0364 0.440 0.476 0.044 0.040
#> GSM39107 2 0.691 0.4606 0.292 0.592 0.012 0.104
#> GSM39108 1 0.760 0.4599 0.552 0.296 0.120 0.032
#> GSM39109 4 0.993 0.0121 0.280 0.200 0.240 0.280
#> GSM39110 1 0.783 0.2177 0.452 0.400 0.116 0.032
#> GSM39111 1 0.707 0.5563 0.596 0.080 0.292 0.032
#> GSM39112 2 0.535 0.4698 0.272 0.692 0.004 0.032
#> GSM39113 2 0.573 0.5062 0.216 0.712 0.012 0.060
#> GSM39114 2 0.329 0.4532 0.036 0.884 0.008 0.072
#> GSM39115 1 0.448 0.6528 0.820 0.120 0.016 0.044
#> GSM39148 1 0.441 0.5440 0.756 0.232 0.004 0.008
#> GSM39149 3 0.272 0.6714 0.052 0.028 0.912 0.008
#> GSM39150 1 0.520 0.6024 0.700 0.000 0.264 0.036
#> GSM39151 3 0.256 0.6851 0.068 0.004 0.912 0.016
#> GSM39152 3 0.402 0.5415 0.224 0.000 0.772 0.004
#> GSM39153 1 0.296 0.7214 0.896 0.028 0.072 0.004
#> GSM39154 1 0.322 0.7023 0.864 0.004 0.124 0.008
#> GSM39155 1 0.259 0.7220 0.904 0.016 0.080 0.000
#> GSM39156 2 0.567 0.0404 0.472 0.508 0.004 0.016
#> GSM39157 1 0.304 0.6573 0.876 0.112 0.008 0.004
#> GSM39158 1 0.288 0.7033 0.904 0.008 0.028 0.060
#> GSM39159 1 0.627 0.4261 0.620 0.000 0.292 0.088
#> GSM39160 1 0.532 0.5513 0.660 0.000 0.312 0.028
#> GSM39161 1 0.731 0.2632 0.528 0.000 0.200 0.272
#> GSM39162 1 0.495 0.3459 0.648 0.344 0.000 0.008
#> GSM39163 1 0.209 0.7130 0.940 0.020 0.028 0.012
#> GSM39164 1 0.426 0.6657 0.824 0.128 0.040 0.008
#> GSM39165 1 0.547 0.3742 0.576 0.004 0.408 0.012
#> GSM39166 1 0.448 0.6598 0.804 0.000 0.128 0.068
#> GSM39167 1 0.248 0.6802 0.904 0.088 0.008 0.000
#> GSM39168 1 0.490 0.4391 0.688 0.300 0.004 0.008
#> GSM39169 1 0.350 0.7190 0.860 0.036 0.104 0.000
#> GSM39170 1 0.232 0.7115 0.928 0.004 0.032 0.036
#> GSM39171 1 0.552 0.5678 0.660 0.008 0.308 0.024
#> GSM39172 3 0.762 0.3810 0.228 0.000 0.464 0.308
#> GSM39173 3 0.310 0.6130 0.012 0.060 0.896 0.032
#> GSM39174 1 0.368 0.7170 0.856 0.060 0.084 0.000
#> GSM39175 1 0.453 0.6261 0.752 0.004 0.232 0.012
#> GSM39176 1 0.201 0.6918 0.932 0.060 0.004 0.004
#> GSM39177 3 0.424 0.6483 0.176 0.000 0.796 0.028
#> GSM39178 1 0.625 0.3404 0.544 0.000 0.396 0.060
#> GSM39179 3 0.327 0.6692 0.048 0.004 0.884 0.064
#> GSM39180 3 0.546 0.5601 0.064 0.000 0.708 0.228
#> GSM39181 1 0.572 0.5676 0.700 0.000 0.088 0.212
#> GSM39182 4 0.610 0.1882 0.180 0.000 0.140 0.680
#> GSM39183 1 0.579 0.5766 0.708 0.000 0.124 0.168
#> GSM39184 1 0.354 0.7032 0.852 0.008 0.128 0.012
#> GSM39185 1 0.761 0.1252 0.476 0.000 0.260 0.264
#> GSM39186 1 0.417 0.7156 0.840 0.052 0.096 0.012
#> GSM39187 1 0.378 0.6237 0.828 0.156 0.008 0.008
#> GSM39116 4 0.506 0.5758 0.000 0.284 0.024 0.692
#> GSM39117 4 0.310 0.5888 0.000 0.020 0.104 0.876
#> GSM39118 4 0.777 0.4544 0.000 0.328 0.252 0.420
#> GSM39119 4 0.669 0.5131 0.000 0.144 0.248 0.608
#> GSM39120 2 0.554 0.4860 0.276 0.680 0.004 0.040
#> GSM39121 2 0.352 0.4955 0.112 0.856 0.000 0.032
#> GSM39122 2 0.393 0.4755 0.068 0.856 0.012 0.064
#> GSM39123 4 0.248 0.6374 0.000 0.052 0.032 0.916
#> GSM39124 2 0.489 0.2548 0.008 0.732 0.016 0.244
#> GSM39125 1 0.766 -0.2113 0.424 0.408 0.008 0.160
#> GSM39126 2 0.405 0.4766 0.072 0.852 0.016 0.060
#> GSM39127 2 0.552 -0.1363 0.004 0.556 0.012 0.428
#> GSM39128 2 0.513 0.1406 0.000 0.668 0.020 0.312
#> GSM39129 3 0.741 -0.0359 0.000 0.256 0.516 0.228
#> GSM39130 4 0.293 0.6331 0.000 0.048 0.056 0.896
#> GSM39131 2 0.503 0.1222 0.000 0.672 0.016 0.312
#> GSM39132 2 0.563 -0.0998 0.000 0.588 0.028 0.384
#> GSM39133 4 0.284 0.6471 0.000 0.088 0.020 0.892
#> GSM39134 4 0.744 0.5062 0.000 0.240 0.248 0.512
#> GSM39135 4 0.568 0.4265 0.000 0.404 0.028 0.568
#> GSM39136 4 0.439 0.5988 0.000 0.236 0.012 0.752
#> GSM39137 2 0.501 0.3646 0.032 0.772 0.020 0.176
#> GSM39138 3 0.757 -0.1312 0.000 0.228 0.480 0.292
#> GSM39139 2 0.743 -0.1994 0.000 0.496 0.308 0.196
#> GSM39140 2 0.540 0.2568 0.404 0.580 0.000 0.016
#> GSM39141 2 0.559 0.1419 0.456 0.524 0.000 0.020
#> GSM39142 1 0.571 0.1101 0.556 0.416 0.000 0.028
#> GSM39143 2 0.560 0.0902 0.476 0.504 0.000 0.020
#> GSM39144 3 0.755 -0.1133 0.000 0.332 0.464 0.204
#> GSM39145 2 0.724 -0.1488 0.000 0.536 0.276 0.188
#> GSM39146 4 0.507 0.5211 0.000 0.320 0.016 0.664
#> GSM39147 2 0.528 0.2002 0.000 0.736 0.072 0.192
#> GSM39188 3 0.322 0.6852 0.076 0.000 0.880 0.044
#> GSM39189 3 0.476 0.5835 0.220 0.000 0.748 0.032
#> GSM39190 3 0.327 0.6870 0.084 0.004 0.880 0.032
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM39104 5 0.519 0.3818 0.080 0.000 0.244 0.004 0.672
#> GSM39105 5 0.456 0.5299 0.152 0.000 0.100 0.000 0.748
#> GSM39106 5 0.395 0.5896 0.060 0.060 0.048 0.000 0.832
#> GSM39107 5 0.631 0.4107 0.068 0.220 0.000 0.084 0.628
#> GSM39108 5 0.485 0.5563 0.108 0.020 0.116 0.000 0.756
#> GSM39109 5 0.627 0.3743 0.008 0.020 0.168 0.172 0.632
#> GSM39110 5 0.525 0.5376 0.068 0.044 0.160 0.000 0.728
#> GSM39111 5 0.539 0.2809 0.064 0.004 0.328 0.000 0.604
#> GSM39112 5 0.505 0.4991 0.076 0.200 0.000 0.012 0.712
#> GSM39113 5 0.582 0.4501 0.044 0.228 0.008 0.052 0.668
#> GSM39114 5 0.573 0.1464 0.016 0.372 0.000 0.056 0.556
#> GSM39115 5 0.592 0.1495 0.340 0.004 0.036 0.040 0.580
#> GSM39148 1 0.438 0.5697 0.780 0.076 0.004 0.004 0.136
#> GSM39149 3 0.347 0.6445 0.012 0.068 0.860 0.008 0.052
#> GSM39150 5 0.698 -0.0160 0.184 0.000 0.328 0.024 0.464
#> GSM39151 3 0.311 0.6584 0.012 0.032 0.884 0.016 0.056
#> GSM39152 3 0.482 0.5037 0.056 0.004 0.700 0.000 0.240
#> GSM39153 1 0.337 0.6359 0.860 0.004 0.064 0.008 0.064
#> GSM39154 1 0.350 0.6406 0.860 0.024 0.076 0.008 0.032
#> GSM39155 1 0.506 0.5772 0.720 0.000 0.116 0.008 0.156
#> GSM39156 1 0.644 0.2658 0.516 0.152 0.004 0.004 0.324
#> GSM39157 1 0.247 0.6346 0.912 0.040 0.012 0.004 0.032
#> GSM39158 1 0.515 0.5782 0.744 0.000 0.040 0.096 0.120
#> GSM39159 1 0.761 0.3126 0.524 0.008 0.220 0.144 0.104
#> GSM39160 3 0.700 0.0952 0.188 0.000 0.400 0.020 0.392
#> GSM39161 1 0.797 0.2483 0.452 0.008 0.148 0.280 0.112
#> GSM39162 1 0.503 0.5187 0.716 0.120 0.000 0.004 0.160
#> GSM39163 1 0.278 0.6387 0.896 0.000 0.032 0.036 0.036
#> GSM39164 1 0.481 0.5779 0.740 0.028 0.044 0.000 0.188
#> GSM39165 1 0.587 0.4487 0.636 0.052 0.268 0.004 0.040
#> GSM39166 1 0.746 0.3220 0.520 0.000 0.188 0.100 0.192
#> GSM39167 1 0.220 0.6367 0.924 0.036 0.004 0.008 0.028
#> GSM39168 1 0.491 0.5146 0.712 0.080 0.004 0.000 0.204
#> GSM39169 1 0.497 0.5867 0.740 0.004 0.084 0.012 0.160
#> GSM39170 1 0.628 0.5162 0.660 0.004 0.092 0.076 0.168
#> GSM39171 3 0.716 0.1070 0.360 0.000 0.388 0.020 0.232
#> GSM39172 4 0.660 -0.2407 0.080 0.008 0.432 0.452 0.028
#> GSM39173 3 0.454 0.4002 0.008 0.268 0.704 0.008 0.012
#> GSM39174 1 0.347 0.6377 0.856 0.012 0.072 0.004 0.056
#> GSM39175 1 0.412 0.6093 0.792 0.000 0.144 0.008 0.056
#> GSM39176 1 0.128 0.6423 0.960 0.004 0.000 0.016 0.020
#> GSM39177 3 0.468 0.6332 0.112 0.084 0.780 0.012 0.012
#> GSM39178 3 0.765 0.2930 0.204 0.000 0.452 0.076 0.268
#> GSM39179 3 0.405 0.6410 0.020 0.084 0.832 0.048 0.016
#> GSM39180 3 0.768 0.4223 0.080 0.112 0.564 0.196 0.048
#> GSM39181 1 0.712 0.4200 0.544 0.004 0.064 0.256 0.132
#> GSM39182 4 0.316 0.5534 0.052 0.000 0.060 0.872 0.016
#> GSM39183 1 0.774 0.3585 0.508 0.004 0.132 0.196 0.160
#> GSM39184 1 0.486 0.6104 0.768 0.000 0.108 0.044 0.080
#> GSM39185 1 0.837 0.0651 0.356 0.008 0.192 0.320 0.124
#> GSM39186 1 0.631 0.3107 0.536 0.000 0.128 0.012 0.324
#> GSM39187 1 0.286 0.6310 0.888 0.036 0.000 0.016 0.060
#> GSM39116 4 0.562 0.3227 0.000 0.328 0.008 0.592 0.072
#> GSM39117 4 0.228 0.6529 0.000 0.060 0.032 0.908 0.000
#> GSM39118 2 0.644 0.1142 0.000 0.508 0.104 0.364 0.024
#> GSM39119 4 0.562 0.4089 0.000 0.280 0.112 0.608 0.000
#> GSM39120 5 0.669 0.2384 0.220 0.316 0.000 0.004 0.460
#> GSM39121 2 0.621 0.1989 0.256 0.564 0.000 0.004 0.176
#> GSM39122 2 0.554 0.3402 0.092 0.656 0.000 0.012 0.240
#> GSM39123 4 0.196 0.6536 0.000 0.052 0.012 0.928 0.008
#> GSM39124 2 0.536 0.4907 0.048 0.736 0.004 0.132 0.080
#> GSM39125 1 0.773 0.2760 0.492 0.168 0.000 0.136 0.204
#> GSM39126 2 0.526 0.3887 0.144 0.680 0.000 0.000 0.176
#> GSM39127 2 0.586 0.2589 0.012 0.568 0.000 0.340 0.080
#> GSM39128 2 0.526 0.4647 0.036 0.716 0.000 0.184 0.064
#> GSM39129 2 0.591 0.3206 0.000 0.540 0.380 0.056 0.024
#> GSM39130 4 0.201 0.6567 0.004 0.056 0.016 0.924 0.000
#> GSM39131 2 0.624 0.3627 0.004 0.584 0.004 0.224 0.184
#> GSM39132 2 0.555 0.3276 0.000 0.616 0.004 0.292 0.088
#> GSM39133 4 0.228 0.6451 0.000 0.096 0.004 0.896 0.004
#> GSM39134 2 0.606 0.0887 0.000 0.508 0.128 0.364 0.000
#> GSM39135 2 0.499 -0.0299 0.000 0.512 0.008 0.464 0.016
#> GSM39136 4 0.497 0.3914 0.000 0.320 0.000 0.632 0.048
#> GSM39137 2 0.468 0.4901 0.120 0.776 0.000 0.036 0.068
#> GSM39138 2 0.566 0.3467 0.000 0.560 0.348 0.092 0.000
#> GSM39139 2 0.440 0.4482 0.000 0.724 0.240 0.032 0.004
#> GSM39140 1 0.625 0.3187 0.544 0.292 0.000 0.004 0.160
#> GSM39141 1 0.615 0.3733 0.580 0.224 0.000 0.004 0.192
#> GSM39142 1 0.601 0.3889 0.588 0.148 0.000 0.004 0.260
#> GSM39143 1 0.630 0.3118 0.536 0.164 0.000 0.004 0.296
#> GSM39144 2 0.545 0.3267 0.000 0.560 0.384 0.048 0.008
#> GSM39145 2 0.442 0.4654 0.000 0.740 0.216 0.036 0.008
#> GSM39146 4 0.512 0.3455 0.000 0.336 0.004 0.616 0.044
#> GSM39147 2 0.259 0.5208 0.004 0.908 0.024 0.024 0.040
#> GSM39188 3 0.228 0.6690 0.032 0.016 0.924 0.020 0.008
#> GSM39189 3 0.521 0.5689 0.056 0.000 0.728 0.048 0.168
#> GSM39190 3 0.317 0.6671 0.036 0.044 0.884 0.016 0.020
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM39104 6 0.3789 0.5561 0.040 0.004 0.072 0.000 0.064 0.820
#> GSM39105 6 0.3920 0.5752 0.136 0.000 0.024 0.000 0.052 0.788
#> GSM39106 6 0.3304 0.5986 0.036 0.064 0.012 0.000 0.032 0.856
#> GSM39107 6 0.6201 0.3946 0.044 0.280 0.000 0.048 0.052 0.576
#> GSM39108 6 0.3799 0.5764 0.076 0.008 0.080 0.000 0.020 0.816
#> GSM39109 6 0.4672 0.4630 0.004 0.016 0.124 0.112 0.004 0.740
#> GSM39110 6 0.3447 0.5541 0.048 0.012 0.108 0.000 0.004 0.828
#> GSM39111 6 0.4669 0.4289 0.036 0.000 0.208 0.000 0.048 0.708
#> GSM39112 6 0.4800 0.5171 0.100 0.212 0.000 0.000 0.008 0.680
#> GSM39113 6 0.4607 0.4783 0.028 0.256 0.000 0.016 0.012 0.688
#> GSM39114 6 0.5017 0.1693 0.004 0.416 0.000 0.032 0.016 0.532
#> GSM39115 6 0.6008 0.0748 0.156 0.008 0.000 0.004 0.376 0.456
#> GSM39148 1 0.1908 0.7409 0.924 0.012 0.000 0.000 0.044 0.020
#> GSM39149 3 0.3955 0.7357 0.008 0.052 0.812 0.008 0.020 0.100
#> GSM39150 6 0.6078 0.2852 0.032 0.004 0.164 0.000 0.232 0.568
#> GSM39151 3 0.4512 0.7235 0.004 0.012 0.756 0.020 0.052 0.156
#> GSM39152 3 0.5115 0.4617 0.004 0.000 0.560 0.000 0.080 0.356
#> GSM39153 1 0.3674 0.7289 0.836 0.004 0.060 0.008 0.060 0.032
#> GSM39154 1 0.3101 0.7342 0.868 0.008 0.036 0.008 0.068 0.012
#> GSM39155 1 0.5328 0.4370 0.564 0.004 0.016 0.000 0.352 0.064
#> GSM39156 1 0.3533 0.6900 0.824 0.036 0.008 0.004 0.008 0.120
#> GSM39157 1 0.3261 0.7141 0.780 0.016 0.000 0.000 0.204 0.000
#> GSM39158 5 0.3053 0.6788 0.172 0.004 0.000 0.000 0.812 0.012
#> GSM39159 5 0.3620 0.7512 0.092 0.008 0.048 0.008 0.832 0.012
#> GSM39160 6 0.7143 0.0462 0.064 0.012 0.260 0.008 0.180 0.476
#> GSM39161 5 0.3611 0.7328 0.060 0.024 0.028 0.032 0.848 0.008
#> GSM39162 1 0.1503 0.7321 0.944 0.032 0.000 0.000 0.008 0.016
#> GSM39163 1 0.3941 0.5889 0.660 0.004 0.004 0.000 0.328 0.004
#> GSM39164 1 0.3183 0.7382 0.852 0.000 0.012 0.004 0.068 0.064
#> GSM39165 1 0.5832 0.5146 0.624 0.044 0.232 0.000 0.084 0.016
#> GSM39166 5 0.2993 0.7540 0.080 0.000 0.016 0.004 0.864 0.036
#> GSM39167 1 0.3121 0.7193 0.804 0.012 0.000 0.000 0.180 0.004
#> GSM39168 1 0.1464 0.7336 0.944 0.016 0.000 0.000 0.004 0.036
#> GSM39169 1 0.5228 0.4530 0.572 0.004 0.016 0.000 0.352 0.056
#> GSM39170 5 0.3336 0.7232 0.132 0.008 0.000 0.004 0.824 0.032
#> GSM39171 5 0.7980 0.0806 0.244 0.012 0.224 0.000 0.288 0.232
#> GSM39172 4 0.6306 0.2308 0.024 0.020 0.292 0.572 0.048 0.044
#> GSM39173 3 0.4772 0.4706 0.000 0.200 0.704 0.000 0.064 0.032
#> GSM39174 1 0.3484 0.7245 0.812 0.004 0.024 0.008 0.148 0.004
#> GSM39175 1 0.4896 0.6783 0.732 0.008 0.076 0.008 0.152 0.024
#> GSM39176 1 0.3829 0.6664 0.720 0.008 0.000 0.004 0.260 0.008
#> GSM39177 3 0.5003 0.6883 0.068 0.048 0.764 0.016 0.076 0.028
#> GSM39178 5 0.6108 0.3042 0.016 0.008 0.204 0.004 0.568 0.200
#> GSM39179 3 0.5397 0.6891 0.068 0.024 0.740 0.064 0.028 0.076
#> GSM39180 5 0.6643 -0.0312 0.000 0.092 0.340 0.052 0.488 0.028
#> GSM39181 5 0.3407 0.7376 0.108 0.016 0.000 0.040 0.832 0.004
#> GSM39182 4 0.3593 0.6162 0.008 0.016 0.064 0.844 0.052 0.016
#> GSM39183 5 0.2773 0.7567 0.076 0.004 0.004 0.020 0.880 0.016
#> GSM39184 1 0.5423 0.4938 0.576 0.008 0.036 0.000 0.340 0.040
#> GSM39185 5 0.2881 0.7305 0.040 0.028 0.020 0.028 0.884 0.000
#> GSM39186 1 0.6694 0.2587 0.472 0.008 0.028 0.004 0.212 0.276
#> GSM39187 1 0.3624 0.7146 0.780 0.016 0.000 0.004 0.188 0.012
#> GSM39116 4 0.5088 0.4576 0.000 0.296 0.004 0.624 0.016 0.060
#> GSM39117 4 0.1109 0.6862 0.000 0.004 0.012 0.964 0.016 0.004
#> GSM39118 4 0.6432 0.2014 0.000 0.352 0.132 0.472 0.012 0.032
#> GSM39119 4 0.5177 0.5479 0.000 0.172 0.104 0.688 0.032 0.004
#> GSM39120 6 0.6701 0.1543 0.180 0.356 0.004 0.000 0.044 0.416
#> GSM39121 2 0.4891 0.1873 0.436 0.516 0.004 0.000 0.004 0.040
#> GSM39122 2 0.5703 0.4676 0.164 0.644 0.016 0.024 0.000 0.152
#> GSM39123 4 0.0748 0.6879 0.000 0.004 0.004 0.976 0.016 0.000
#> GSM39124 2 0.4936 0.5063 0.104 0.732 0.016 0.120 0.000 0.028
#> GSM39125 1 0.8206 0.0124 0.308 0.224 0.004 0.032 0.284 0.148
#> GSM39126 2 0.4448 0.5106 0.176 0.732 0.008 0.004 0.000 0.080
#> GSM39127 2 0.5887 0.2533 0.004 0.592 0.008 0.272 0.032 0.092
#> GSM39128 2 0.4925 0.4687 0.060 0.740 0.004 0.144 0.028 0.024
#> GSM39129 2 0.5565 0.2321 0.000 0.488 0.432 0.016 0.040 0.024
#> GSM39130 4 0.0982 0.6879 0.000 0.004 0.004 0.968 0.020 0.004
#> GSM39131 2 0.5815 0.3508 0.000 0.604 0.004 0.152 0.028 0.212
#> GSM39132 2 0.4988 0.3881 0.000 0.700 0.012 0.192 0.020 0.076
#> GSM39133 4 0.1599 0.6856 0.000 0.028 0.000 0.940 0.024 0.008
#> GSM39134 2 0.6788 0.0539 0.000 0.428 0.172 0.340 0.056 0.004
#> GSM39135 4 0.4779 0.1802 0.000 0.464 0.012 0.500 0.020 0.004
#> GSM39136 4 0.5222 0.4481 0.000 0.312 0.004 0.608 0.044 0.032
#> GSM39137 2 0.4582 0.5194 0.176 0.744 0.016 0.044 0.008 0.012
#> GSM39138 2 0.6023 0.2574 0.000 0.484 0.396 0.052 0.060 0.008
#> GSM39139 2 0.4853 0.4125 0.004 0.636 0.308 0.008 0.036 0.008
#> GSM39140 1 0.3168 0.6664 0.820 0.148 0.000 0.000 0.004 0.028
#> GSM39141 1 0.3010 0.6912 0.848 0.112 0.000 0.004 0.004 0.032
#> GSM39142 1 0.3039 0.7196 0.860 0.052 0.000 0.000 0.020 0.068
#> GSM39143 1 0.3647 0.6860 0.812 0.104 0.000 0.000 0.016 0.068
#> GSM39144 2 0.5633 0.2430 0.000 0.492 0.424 0.024 0.040 0.020
#> GSM39145 2 0.4898 0.4640 0.000 0.656 0.280 0.020 0.028 0.016
#> GSM39146 4 0.3350 0.6317 0.004 0.156 0.000 0.812 0.012 0.016
#> GSM39147 2 0.4694 0.5341 0.060 0.772 0.100 0.044 0.008 0.016
#> GSM39188 3 0.4027 0.7483 0.004 0.016 0.816 0.032 0.056 0.076
#> GSM39189 3 0.6396 0.5598 0.012 0.016 0.568 0.032 0.112 0.260
#> GSM39190 3 0.5208 0.6876 0.000 0.080 0.704 0.004 0.144 0.068
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) other(p) protocol(p) k
#> MAD:NMF 83 1.38e-01 1.27e-07 2.42e-06 2
#> MAD:NMF 46 5.94e-02 5.30e-08 1.71e-05 3
#> MAD:NMF 47 1.09e-02 6.19e-07 3.24e-09 4
#> MAD:NMF 36 2.89e-07 7.84e-08 8.44e-11 5
#> MAD:NMF 51 8.65e-10 3.73e-10 4.97e-14 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 8353 rows and 87 columns.
#> Top rows (835, 1670, 2506, 3341, 4176) 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 1.000 0.984 0.994 0.0699 0.933 0.933
#> 3 3 0.668 0.743 0.911 3.1322 0.875 0.866
#> 4 4 0.661 -0.508 0.765 0.0922 0.733 0.710
#> 5 5 0.599 0.757 0.889 0.1770 0.627 0.567
#> 6 6 0.565 0.733 0.847 0.1696 0.962 0.937
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
#> GSM39104 1 0.0000 0.996 1.000 0.000
#> GSM39105 1 0.0000 0.996 1.000 0.000
#> GSM39106 1 0.0000 0.996 1.000 0.000
#> GSM39107 1 0.0000 0.996 1.000 0.000
#> GSM39108 1 0.0000 0.996 1.000 0.000
#> GSM39109 1 0.0000 0.996 1.000 0.000
#> GSM39110 1 0.0000 0.996 1.000 0.000
#> GSM39111 1 0.0000 0.996 1.000 0.000
#> GSM39112 1 0.0000 0.996 1.000 0.000
#> GSM39113 1 0.0000 0.996 1.000 0.000
#> GSM39114 1 0.0000 0.996 1.000 0.000
#> GSM39115 1 0.0000 0.996 1.000 0.000
#> GSM39148 1 0.0000 0.996 1.000 0.000
#> GSM39149 1 0.0000 0.996 1.000 0.000
#> GSM39150 1 0.0000 0.996 1.000 0.000
#> GSM39151 1 0.0000 0.996 1.000 0.000
#> GSM39152 1 0.0000 0.996 1.000 0.000
#> GSM39153 1 0.0000 0.996 1.000 0.000
#> GSM39154 1 0.0000 0.996 1.000 0.000
#> GSM39155 1 0.0000 0.996 1.000 0.000
#> GSM39156 1 0.0000 0.996 1.000 0.000
#> GSM39157 1 0.0000 0.996 1.000 0.000
#> GSM39158 1 0.0000 0.996 1.000 0.000
#> GSM39159 1 0.0000 0.996 1.000 0.000
#> GSM39160 1 0.0000 0.996 1.000 0.000
#> GSM39161 1 0.0000 0.996 1.000 0.000
#> GSM39162 1 0.0000 0.996 1.000 0.000
#> GSM39163 1 0.0000 0.996 1.000 0.000
#> GSM39164 1 0.0000 0.996 1.000 0.000
#> GSM39165 1 0.0000 0.996 1.000 0.000
#> GSM39166 1 0.0000 0.996 1.000 0.000
#> GSM39167 1 0.0000 0.996 1.000 0.000
#> GSM39168 1 0.0000 0.996 1.000 0.000
#> GSM39169 1 0.0000 0.996 1.000 0.000
#> GSM39170 1 0.0000 0.996 1.000 0.000
#> GSM39171 1 0.0000 0.996 1.000 0.000
#> GSM39172 1 0.0000 0.996 1.000 0.000
#> GSM39173 1 0.0000 0.996 1.000 0.000
#> GSM39174 1 0.0000 0.996 1.000 0.000
#> GSM39175 1 0.0000 0.996 1.000 0.000
#> GSM39176 1 0.0000 0.996 1.000 0.000
#> GSM39177 1 0.0000 0.996 1.000 0.000
#> GSM39178 1 0.0000 0.996 1.000 0.000
#> GSM39179 1 0.0000 0.996 1.000 0.000
#> GSM39180 1 0.0000 0.996 1.000 0.000
#> GSM39181 1 0.0000 0.996 1.000 0.000
#> GSM39182 1 0.0000 0.996 1.000 0.000
#> GSM39183 1 0.0000 0.996 1.000 0.000
#> GSM39184 1 0.0000 0.996 1.000 0.000
#> GSM39185 1 0.0000 0.996 1.000 0.000
#> GSM39186 1 0.0000 0.996 1.000 0.000
#> GSM39187 1 0.0000 0.996 1.000 0.000
#> GSM39116 1 0.0000 0.996 1.000 0.000
#> GSM39117 2 0.0000 0.878 0.000 1.000
#> GSM39118 1 0.0000 0.996 1.000 0.000
#> GSM39119 1 0.0672 0.987 0.992 0.008
#> GSM39120 1 0.0000 0.996 1.000 0.000
#> GSM39121 1 0.0000 0.996 1.000 0.000
#> GSM39122 1 0.0000 0.996 1.000 0.000
#> GSM39123 2 0.7950 0.681 0.240 0.760
#> GSM39124 1 0.0000 0.996 1.000 0.000
#> GSM39125 1 0.0000 0.996 1.000 0.000
#> GSM39126 1 0.0000 0.996 1.000 0.000
#> GSM39127 1 0.0000 0.996 1.000 0.000
#> GSM39128 1 0.0000 0.996 1.000 0.000
#> GSM39129 1 0.0000 0.996 1.000 0.000
#> GSM39130 2 0.0000 0.878 0.000 1.000
#> GSM39131 1 0.0000 0.996 1.000 0.000
#> GSM39132 1 0.0000 0.996 1.000 0.000
#> GSM39133 1 0.8861 0.506 0.696 0.304
#> GSM39134 1 0.0000 0.996 1.000 0.000
#> GSM39135 1 0.0000 0.996 1.000 0.000
#> GSM39136 1 0.0938 0.983 0.988 0.012
#> GSM39137 1 0.0000 0.996 1.000 0.000
#> GSM39138 1 0.0000 0.996 1.000 0.000
#> GSM39139 1 0.0000 0.996 1.000 0.000
#> GSM39140 1 0.0000 0.996 1.000 0.000
#> GSM39141 1 0.0000 0.996 1.000 0.000
#> GSM39142 1 0.0000 0.996 1.000 0.000
#> GSM39143 1 0.0000 0.996 1.000 0.000
#> GSM39144 1 0.0000 0.996 1.000 0.000
#> GSM39145 1 0.0000 0.996 1.000 0.000
#> GSM39146 1 0.0000 0.996 1.000 0.000
#> GSM39147 1 0.0000 0.996 1.000 0.000
#> GSM39188 1 0.0000 0.996 1.000 0.000
#> GSM39189 1 0.0000 0.996 1.000 0.000
#> GSM39190 1 0.0000 0.996 1.000 0.000
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM39104 1 0.0000 0.8945 1.000 0.000 0.000
#> GSM39105 1 0.0000 0.8945 1.000 0.000 0.000
#> GSM39106 1 0.0000 0.8945 1.000 0.000 0.000
#> GSM39107 1 0.0000 0.8945 1.000 0.000 0.000
#> GSM39108 1 0.0000 0.8945 1.000 0.000 0.000
#> GSM39109 1 0.0000 0.8945 1.000 0.000 0.000
#> GSM39110 1 0.0000 0.8945 1.000 0.000 0.000
#> GSM39111 1 0.0000 0.8945 1.000 0.000 0.000
#> GSM39112 1 0.0000 0.8945 1.000 0.000 0.000
#> GSM39113 1 0.0000 0.8945 1.000 0.000 0.000
#> GSM39114 1 0.6305 0.0976 0.516 0.484 0.000
#> GSM39115 1 0.0000 0.8945 1.000 0.000 0.000
#> GSM39148 1 0.0000 0.8945 1.000 0.000 0.000
#> GSM39149 1 0.0000 0.8945 1.000 0.000 0.000
#> GSM39150 1 0.0000 0.8945 1.000 0.000 0.000
#> GSM39151 1 0.0000 0.8945 1.000 0.000 0.000
#> GSM39152 1 0.0000 0.8945 1.000 0.000 0.000
#> GSM39153 1 0.0000 0.8945 1.000 0.000 0.000
#> GSM39154 1 0.0000 0.8945 1.000 0.000 0.000
#> GSM39155 1 0.0000 0.8945 1.000 0.000 0.000
#> GSM39156 1 0.0000 0.8945 1.000 0.000 0.000
#> GSM39157 1 0.0000 0.8945 1.000 0.000 0.000
#> GSM39158 1 0.0000 0.8945 1.000 0.000 0.000
#> GSM39159 1 0.0000 0.8945 1.000 0.000 0.000
#> GSM39160 1 0.0000 0.8945 1.000 0.000 0.000
#> GSM39161 1 0.0000 0.8945 1.000 0.000 0.000
#> GSM39162 1 0.0000 0.8945 1.000 0.000 0.000
#> GSM39163 1 0.0000 0.8945 1.000 0.000 0.000
#> GSM39164 1 0.0000 0.8945 1.000 0.000 0.000
#> GSM39165 1 0.0000 0.8945 1.000 0.000 0.000
#> GSM39166 1 0.0000 0.8945 1.000 0.000 0.000
#> GSM39167 1 0.0000 0.8945 1.000 0.000 0.000
#> GSM39168 1 0.0000 0.8945 1.000 0.000 0.000
#> GSM39169 1 0.0000 0.8945 1.000 0.000 0.000
#> GSM39170 1 0.0000 0.8945 1.000 0.000 0.000
#> GSM39171 1 0.0000 0.8945 1.000 0.000 0.000
#> GSM39172 1 0.0000 0.8945 1.000 0.000 0.000
#> GSM39173 1 0.6026 0.3767 0.624 0.376 0.000
#> GSM39174 1 0.0000 0.8945 1.000 0.000 0.000
#> GSM39175 1 0.0000 0.8945 1.000 0.000 0.000
#> GSM39176 1 0.0000 0.8945 1.000 0.000 0.000
#> GSM39177 1 0.0000 0.8945 1.000 0.000 0.000
#> GSM39178 1 0.0000 0.8945 1.000 0.000 0.000
#> GSM39179 1 0.0000 0.8945 1.000 0.000 0.000
#> GSM39180 1 0.0000 0.8945 1.000 0.000 0.000
#> GSM39181 1 0.0000 0.8945 1.000 0.000 0.000
#> GSM39182 1 0.0000 0.8945 1.000 0.000 0.000
#> GSM39183 1 0.0000 0.8945 1.000 0.000 0.000
#> GSM39184 1 0.0000 0.8945 1.000 0.000 0.000
#> GSM39185 1 0.0000 0.8945 1.000 0.000 0.000
#> GSM39186 1 0.0000 0.8945 1.000 0.000 0.000
#> GSM39187 1 0.0000 0.8945 1.000 0.000 0.000
#> GSM39116 1 0.6307 0.0848 0.512 0.488 0.000
#> GSM39117 3 0.0000 0.7729 0.000 0.000 1.000
#> GSM39118 1 0.6307 0.0848 0.512 0.488 0.000
#> GSM39119 2 0.6680 -0.1219 0.484 0.508 0.008
#> GSM39120 1 0.0000 0.8945 1.000 0.000 0.000
#> GSM39121 1 0.0000 0.8945 1.000 0.000 0.000
#> GSM39122 1 0.0000 0.8945 1.000 0.000 0.000
#> GSM39123 3 0.6208 0.5160 0.192 0.052 0.756
#> GSM39124 1 0.6305 0.0976 0.516 0.484 0.000
#> GSM39125 1 0.0000 0.8945 1.000 0.000 0.000
#> GSM39126 1 0.0000 0.8945 1.000 0.000 0.000
#> GSM39127 1 0.6307 0.0848 0.512 0.488 0.000
#> GSM39128 1 0.4555 0.6878 0.800 0.200 0.000
#> GSM39129 2 0.0424 0.7457 0.008 0.992 0.000
#> GSM39130 3 0.0000 0.7729 0.000 0.000 1.000
#> GSM39131 1 0.6307 0.0848 0.512 0.488 0.000
#> GSM39132 1 0.6305 0.0976 0.516 0.484 0.000
#> GSM39133 1 0.7246 0.4595 0.648 0.052 0.300
#> GSM39134 2 0.0424 0.7457 0.008 0.992 0.000
#> GSM39135 1 0.6307 0.0848 0.512 0.488 0.000
#> GSM39136 1 0.6587 0.2436 0.568 0.424 0.008
#> GSM39137 1 0.6267 0.1897 0.548 0.452 0.000
#> GSM39138 2 0.0424 0.7457 0.008 0.992 0.000
#> GSM39139 2 0.0424 0.7457 0.008 0.992 0.000
#> GSM39140 1 0.0000 0.8945 1.000 0.000 0.000
#> GSM39141 1 0.0000 0.8945 1.000 0.000 0.000
#> GSM39142 1 0.0000 0.8945 1.000 0.000 0.000
#> GSM39143 1 0.0000 0.8945 1.000 0.000 0.000
#> GSM39144 2 0.0424 0.7457 0.008 0.992 0.000
#> GSM39145 1 0.6252 0.2111 0.556 0.444 0.000
#> GSM39146 1 0.5650 0.5088 0.688 0.312 0.000
#> GSM39147 1 0.6305 0.0976 0.516 0.484 0.000
#> GSM39188 1 0.0000 0.8945 1.000 0.000 0.000
#> GSM39189 1 0.0000 0.8945 1.000 0.000 0.000
#> GSM39190 1 0.0000 0.8945 1.000 0.000 0.000
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM39104 1 0.5000 -0.8781 0.504 0.000 0.496 0.000
#> GSM39105 1 0.5000 -0.8781 0.504 0.000 0.496 0.000
#> GSM39106 1 0.5000 -0.8781 0.504 0.000 0.496 0.000
#> GSM39107 1 0.5000 -0.8781 0.504 0.000 0.496 0.000
#> GSM39108 1 0.5000 -0.8781 0.504 0.000 0.496 0.000
#> GSM39109 1 0.5000 -0.8781 0.504 0.000 0.496 0.000
#> GSM39110 1 0.5000 -0.8781 0.504 0.000 0.496 0.000
#> GSM39111 1 0.5000 -0.8781 0.504 0.000 0.496 0.000
#> GSM39112 1 0.5000 -0.8781 0.504 0.000 0.496 0.000
#> GSM39113 1 0.5000 -0.8781 0.504 0.000 0.496 0.000
#> GSM39114 1 0.0188 0.1216 0.996 0.000 0.004 0.000
#> GSM39115 1 0.5000 -0.8781 0.504 0.000 0.496 0.000
#> GSM39148 1 0.5000 -0.8781 0.504 0.000 0.496 0.000
#> GSM39149 1 0.5000 -0.8781 0.504 0.000 0.496 0.000
#> GSM39150 1 0.5000 -0.8781 0.504 0.000 0.496 0.000
#> GSM39151 1 0.5000 -0.8781 0.504 0.000 0.496 0.000
#> GSM39152 1 0.5000 -0.8781 0.504 0.000 0.496 0.000
#> GSM39153 1 0.5000 -0.8781 0.504 0.000 0.496 0.000
#> GSM39154 1 0.5000 -0.8781 0.504 0.000 0.496 0.000
#> GSM39155 1 0.5000 -0.8781 0.504 0.000 0.496 0.000
#> GSM39156 1 0.5000 -0.8781 0.504 0.000 0.496 0.000
#> GSM39157 1 0.5000 -0.8781 0.504 0.000 0.496 0.000
#> GSM39158 3 0.5167 1.0000 0.488 0.000 0.508 0.004
#> GSM39159 1 0.5000 -0.8781 0.504 0.000 0.496 0.000
#> GSM39160 1 0.5000 -0.8781 0.504 0.000 0.496 0.000
#> GSM39161 3 0.5167 1.0000 0.488 0.000 0.508 0.004
#> GSM39162 1 0.5000 -0.8781 0.504 0.000 0.496 0.000
#> GSM39163 1 0.5000 -0.8781 0.504 0.000 0.496 0.000
#> GSM39164 1 0.5000 -0.8781 0.504 0.000 0.496 0.000
#> GSM39165 1 0.5000 -0.8781 0.504 0.000 0.496 0.000
#> GSM39166 3 0.5167 1.0000 0.488 0.000 0.508 0.004
#> GSM39167 1 0.5000 -0.8781 0.504 0.000 0.496 0.000
#> GSM39168 1 0.5000 -0.8781 0.504 0.000 0.496 0.000
#> GSM39169 1 0.5000 -0.8781 0.504 0.000 0.496 0.000
#> GSM39170 1 0.5000 -0.8781 0.504 0.000 0.496 0.000
#> GSM39171 1 0.5000 -0.8781 0.504 0.000 0.496 0.000
#> GSM39172 1 0.5000 -0.8781 0.504 0.000 0.496 0.000
#> GSM39173 1 0.2530 0.0797 0.888 0.000 0.112 0.000
#> GSM39174 1 0.5000 -0.8781 0.504 0.000 0.496 0.000
#> GSM39175 1 0.5000 -0.8781 0.504 0.000 0.496 0.000
#> GSM39176 1 0.5000 -0.8781 0.504 0.000 0.496 0.000
#> GSM39177 1 0.5000 -0.8781 0.504 0.000 0.496 0.000
#> GSM39178 1 0.5000 -0.8781 0.504 0.000 0.496 0.000
#> GSM39179 1 0.5000 -0.8781 0.504 0.000 0.496 0.000
#> GSM39180 4 0.7890 -0.2697 0.044 0.100 0.416 0.440
#> GSM39181 3 0.5167 1.0000 0.488 0.000 0.508 0.004
#> GSM39182 1 0.5000 -0.8781 0.504 0.000 0.496 0.000
#> GSM39183 3 0.5167 1.0000 0.488 0.000 0.508 0.004
#> GSM39184 1 0.5000 -0.8781 0.504 0.000 0.496 0.000
#> GSM39185 3 0.5167 1.0000 0.488 0.000 0.508 0.004
#> GSM39186 1 0.5000 -0.8781 0.504 0.000 0.496 0.000
#> GSM39187 1 0.5000 -0.8781 0.504 0.000 0.496 0.000
#> GSM39116 1 0.0188 0.1128 0.996 0.004 0.000 0.000
#> GSM39117 4 0.4972 0.2578 0.000 0.456 0.000 0.544
#> GSM39118 1 0.0188 0.1128 0.996 0.004 0.000 0.000
#> GSM39119 1 0.5288 -0.1729 0.776 0.112 0.096 0.016
#> GSM39120 1 0.5000 -0.8781 0.504 0.000 0.496 0.000
#> GSM39121 1 0.5000 -0.8781 0.504 0.000 0.496 0.000
#> GSM39122 1 0.5000 -0.8781 0.504 0.000 0.496 0.000
#> GSM39123 2 0.0000 -0.4830 0.000 1.000 0.000 0.000
#> GSM39124 1 0.0188 0.1216 0.996 0.000 0.004 0.000
#> GSM39125 1 0.5000 -0.8781 0.504 0.000 0.496 0.000
#> GSM39126 1 0.5000 -0.8781 0.504 0.000 0.496 0.000
#> GSM39127 1 0.0000 0.1140 1.000 0.000 0.000 0.000
#> GSM39128 1 0.4643 -0.2555 0.656 0.000 0.344 0.000
#> GSM39129 1 0.5512 -0.4762 0.496 0.000 0.488 0.016
#> GSM39130 4 0.4972 0.2578 0.000 0.456 0.000 0.544
#> GSM39131 1 0.0000 0.1140 1.000 0.000 0.000 0.000
#> GSM39132 1 0.0188 0.1216 0.996 0.000 0.004 0.000
#> GSM39133 2 0.6359 -0.1161 0.396 0.544 0.056 0.004
#> GSM39134 1 0.5512 -0.4762 0.496 0.000 0.488 0.016
#> GSM39135 1 0.0000 0.1140 1.000 0.000 0.000 0.000
#> GSM39136 1 0.5136 -0.1107 0.768 0.144 0.084 0.004
#> GSM39137 1 0.1118 0.1391 0.964 0.000 0.036 0.000
#> GSM39138 1 0.5512 -0.4762 0.496 0.000 0.488 0.016
#> GSM39139 1 0.5512 -0.4762 0.496 0.000 0.488 0.016
#> GSM39140 1 0.5000 -0.8781 0.504 0.000 0.496 0.000
#> GSM39141 1 0.5000 -0.8781 0.504 0.000 0.496 0.000
#> GSM39142 1 0.5000 -0.8781 0.504 0.000 0.496 0.000
#> GSM39143 1 0.5000 -0.8781 0.504 0.000 0.496 0.000
#> GSM39144 1 0.5512 -0.4762 0.496 0.000 0.488 0.016
#> GSM39145 1 0.1302 0.1343 0.956 0.000 0.044 0.000
#> GSM39146 1 0.3444 -0.0154 0.816 0.000 0.184 0.000
#> GSM39147 1 0.0188 0.1216 0.996 0.000 0.004 0.000
#> GSM39188 1 0.5000 -0.8781 0.504 0.000 0.496 0.000
#> GSM39189 1 0.5000 -0.8781 0.504 0.000 0.496 0.000
#> GSM39190 1 0.5000 -0.8781 0.504 0.000 0.496 0.000
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM39104 1 0.0324 0.9470 0.992 0.004 0.00 0.000 0.004
#> GSM39105 1 0.0324 0.9470 0.992 0.004 0.00 0.000 0.004
#> GSM39106 1 0.0324 0.9470 0.992 0.004 0.00 0.000 0.004
#> GSM39107 1 0.0324 0.9470 0.992 0.004 0.00 0.000 0.004
#> GSM39108 1 0.0324 0.9470 0.992 0.004 0.00 0.000 0.004
#> GSM39109 1 0.0324 0.9470 0.992 0.004 0.00 0.000 0.004
#> GSM39110 1 0.0324 0.9470 0.992 0.004 0.00 0.000 0.004
#> GSM39111 1 0.0324 0.9470 0.992 0.004 0.00 0.000 0.004
#> GSM39112 1 0.0324 0.9470 0.992 0.004 0.00 0.000 0.004
#> GSM39113 1 0.0324 0.9470 0.992 0.004 0.00 0.000 0.004
#> GSM39114 2 0.4171 0.6543 0.396 0.604 0.00 0.000 0.000
#> GSM39115 1 0.0324 0.9470 0.992 0.004 0.00 0.000 0.004
#> GSM39148 1 0.0000 0.9479 1.000 0.000 0.00 0.000 0.000
#> GSM39149 1 0.0609 0.9373 0.980 0.000 0.00 0.000 0.020
#> GSM39150 1 0.0510 0.9404 0.984 0.000 0.00 0.000 0.016
#> GSM39151 1 0.0703 0.9337 0.976 0.000 0.00 0.000 0.024
#> GSM39152 1 0.0609 0.9373 0.980 0.000 0.00 0.000 0.020
#> GSM39153 1 0.0000 0.9479 1.000 0.000 0.00 0.000 0.000
#> GSM39154 1 0.0000 0.9479 1.000 0.000 0.00 0.000 0.000
#> GSM39155 1 0.0162 0.9471 0.996 0.000 0.00 0.000 0.004
#> GSM39156 1 0.0000 0.9479 1.000 0.000 0.00 0.000 0.000
#> GSM39157 1 0.0000 0.9479 1.000 0.000 0.00 0.000 0.000
#> GSM39158 1 0.3542 0.7553 0.840 0.112 0.02 0.000 0.028
#> GSM39159 1 0.0162 0.9465 0.996 0.000 0.00 0.000 0.004
#> GSM39160 1 0.0404 0.9431 0.988 0.000 0.00 0.000 0.012
#> GSM39161 1 0.3542 0.7553 0.840 0.112 0.02 0.000 0.028
#> GSM39162 1 0.0000 0.9479 1.000 0.000 0.00 0.000 0.000
#> GSM39163 1 0.0000 0.9479 1.000 0.000 0.00 0.000 0.000
#> GSM39164 1 0.0000 0.9479 1.000 0.000 0.00 0.000 0.000
#> GSM39165 1 0.0000 0.9479 1.000 0.000 0.00 0.000 0.000
#> GSM39166 1 0.3542 0.7553 0.840 0.112 0.02 0.000 0.028
#> GSM39167 1 0.0000 0.9479 1.000 0.000 0.00 0.000 0.000
#> GSM39168 1 0.0000 0.9479 1.000 0.000 0.00 0.000 0.000
#> GSM39169 1 0.0162 0.9471 0.996 0.000 0.00 0.000 0.004
#> GSM39170 1 0.0000 0.9479 1.000 0.000 0.00 0.000 0.000
#> GSM39171 1 0.0404 0.9431 0.988 0.000 0.00 0.000 0.012
#> GSM39172 1 0.0609 0.9356 0.980 0.000 0.00 0.000 0.020
#> GSM39173 1 0.4299 -0.1555 0.608 0.388 0.00 0.000 0.004
#> GSM39174 1 0.0000 0.9479 1.000 0.000 0.00 0.000 0.000
#> GSM39175 1 0.0000 0.9479 1.000 0.000 0.00 0.000 0.000
#> GSM39176 1 0.0000 0.9479 1.000 0.000 0.00 0.000 0.000
#> GSM39177 1 0.0162 0.9468 0.996 0.000 0.00 0.000 0.004
#> GSM39178 1 0.0404 0.9431 0.988 0.000 0.00 0.000 0.012
#> GSM39179 1 0.0703 0.9337 0.976 0.000 0.00 0.000 0.024
#> GSM39180 3 0.0000 0.0000 0.000 0.000 1.00 0.000 0.000
#> GSM39181 1 0.3542 0.7553 0.840 0.112 0.02 0.000 0.028
#> GSM39182 1 0.0609 0.9356 0.980 0.000 0.00 0.000 0.020
#> GSM39183 1 0.3542 0.7553 0.840 0.112 0.02 0.000 0.028
#> GSM39184 1 0.0162 0.9471 0.996 0.000 0.00 0.000 0.004
#> GSM39185 1 0.3542 0.7553 0.840 0.112 0.02 0.000 0.028
#> GSM39186 1 0.0162 0.9471 0.996 0.000 0.00 0.000 0.004
#> GSM39187 1 0.0000 0.9479 1.000 0.000 0.00 0.000 0.000
#> GSM39116 2 0.4310 0.6530 0.392 0.604 0.00 0.000 0.004
#> GSM39117 4 0.0000 0.7508 0.000 0.000 0.00 1.000 0.000
#> GSM39118 2 0.4310 0.6530 0.392 0.604 0.00 0.000 0.004
#> GSM39119 2 0.6031 0.3357 0.244 0.576 0.00 0.000 0.180
#> GSM39120 1 0.0162 0.9464 0.996 0.004 0.00 0.000 0.000
#> GSM39121 1 0.0162 0.9464 0.996 0.004 0.00 0.000 0.000
#> GSM39122 1 0.0162 0.9464 0.996 0.004 0.00 0.000 0.000
#> GSM39123 5 0.4273 -0.5371 0.000 0.000 0.00 0.448 0.552
#> GSM39124 2 0.4171 0.6543 0.396 0.604 0.00 0.000 0.000
#> GSM39125 1 0.0162 0.9464 0.996 0.004 0.00 0.000 0.000
#> GSM39126 1 0.0162 0.9464 0.996 0.004 0.00 0.000 0.000
#> GSM39127 2 0.4201 0.6460 0.408 0.592 0.00 0.000 0.000
#> GSM39128 1 0.3789 0.5350 0.760 0.224 0.00 0.000 0.016
#> GSM39129 2 0.2179 0.0686 0.000 0.888 0.00 0.000 0.112
#> GSM39130 4 0.3424 0.7434 0.000 0.000 0.00 0.760 0.240
#> GSM39131 2 0.4201 0.6460 0.408 0.592 0.00 0.000 0.000
#> GSM39132 2 0.4171 0.6543 0.396 0.604 0.00 0.000 0.000
#> GSM39133 5 0.6022 0.1054 0.252 0.112 0.02 0.000 0.616
#> GSM39134 2 0.2179 0.0686 0.000 0.888 0.00 0.000 0.112
#> GSM39135 2 0.4161 0.6535 0.392 0.608 0.00 0.000 0.000
#> GSM39136 2 0.6940 0.3214 0.280 0.484 0.02 0.000 0.216
#> GSM39137 2 0.4283 0.5852 0.456 0.544 0.00 0.000 0.000
#> GSM39138 2 0.2179 0.0686 0.000 0.888 0.00 0.000 0.112
#> GSM39139 2 0.2179 0.0686 0.000 0.888 0.00 0.000 0.112
#> GSM39140 1 0.0000 0.9479 1.000 0.000 0.00 0.000 0.000
#> GSM39141 1 0.0000 0.9479 1.000 0.000 0.00 0.000 0.000
#> GSM39142 1 0.0000 0.9479 1.000 0.000 0.00 0.000 0.000
#> GSM39143 1 0.0000 0.9479 1.000 0.000 0.00 0.000 0.000
#> GSM39144 2 0.2179 0.0686 0.000 0.888 0.00 0.000 0.112
#> GSM39145 2 0.4256 0.6051 0.436 0.564 0.00 0.000 0.000
#> GSM39146 1 0.4201 -0.1732 0.592 0.408 0.00 0.000 0.000
#> GSM39147 2 0.4171 0.6543 0.396 0.604 0.00 0.000 0.000
#> GSM39188 1 0.0703 0.9337 0.976 0.000 0.00 0.000 0.024
#> GSM39189 1 0.0703 0.9337 0.976 0.000 0.00 0.000 0.024
#> GSM39190 1 0.0703 0.9337 0.976 0.000 0.00 0.000 0.024
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM39104 1 0.0713 0.8756 0.972 0.028 0 0.000 0.000 0.000
#> GSM39105 1 0.0713 0.8756 0.972 0.028 0 0.000 0.000 0.000
#> GSM39106 1 0.0713 0.8756 0.972 0.028 0 0.000 0.000 0.000
#> GSM39107 1 0.0713 0.8756 0.972 0.028 0 0.000 0.000 0.000
#> GSM39108 1 0.0713 0.8756 0.972 0.028 0 0.000 0.000 0.000
#> GSM39109 1 0.0713 0.8756 0.972 0.028 0 0.000 0.000 0.000
#> GSM39110 1 0.0713 0.8756 0.972 0.028 0 0.000 0.000 0.000
#> GSM39111 1 0.0713 0.8756 0.972 0.028 0 0.000 0.000 0.000
#> GSM39112 1 0.0713 0.8756 0.972 0.028 0 0.000 0.000 0.000
#> GSM39113 1 0.0713 0.8756 0.972 0.028 0 0.000 0.000 0.000
#> GSM39114 2 0.5814 0.8727 0.364 0.448 0 0.000 0.000 0.188
#> GSM39115 1 0.0713 0.8756 0.972 0.028 0 0.000 0.000 0.000
#> GSM39148 1 0.0632 0.8712 0.976 0.024 0 0.000 0.000 0.000
#> GSM39149 1 0.2048 0.8183 0.880 0.120 0 0.000 0.000 0.000
#> GSM39150 1 0.1501 0.8510 0.924 0.076 0 0.000 0.000 0.000
#> GSM39151 1 0.2178 0.8073 0.868 0.132 0 0.000 0.000 0.000
#> GSM39152 1 0.1765 0.8379 0.904 0.096 0 0.000 0.000 0.000
#> GSM39153 1 0.0547 0.8756 0.980 0.020 0 0.000 0.000 0.000
#> GSM39154 1 0.0713 0.8756 0.972 0.028 0 0.000 0.000 0.000
#> GSM39155 1 0.0146 0.8770 0.996 0.004 0 0.000 0.000 0.000
#> GSM39156 1 0.0547 0.8756 0.980 0.020 0 0.000 0.000 0.000
#> GSM39157 1 0.0458 0.8766 0.984 0.016 0 0.000 0.000 0.000
#> GSM39158 1 0.3464 0.5144 0.688 0.312 0 0.000 0.000 0.000
#> GSM39159 1 0.0865 0.8760 0.964 0.036 0 0.000 0.000 0.000
#> GSM39160 1 0.1141 0.8660 0.948 0.052 0 0.000 0.000 0.000
#> GSM39161 1 0.3482 0.5062 0.684 0.316 0 0.000 0.000 0.000
#> GSM39162 1 0.0632 0.8712 0.976 0.024 0 0.000 0.000 0.000
#> GSM39163 1 0.0713 0.8756 0.972 0.028 0 0.000 0.000 0.000
#> GSM39164 1 0.0260 0.8771 0.992 0.008 0 0.000 0.000 0.000
#> GSM39165 1 0.0547 0.8756 0.980 0.020 0 0.000 0.000 0.000
#> GSM39166 1 0.3464 0.5144 0.688 0.312 0 0.000 0.000 0.000
#> GSM39167 1 0.0547 0.8756 0.980 0.020 0 0.000 0.000 0.000
#> GSM39168 1 0.0632 0.8712 0.976 0.024 0 0.000 0.000 0.000
#> GSM39169 1 0.0260 0.8763 0.992 0.008 0 0.000 0.000 0.000
#> GSM39170 1 0.0632 0.8712 0.976 0.024 0 0.000 0.000 0.000
#> GSM39171 1 0.1007 0.8706 0.956 0.044 0 0.000 0.000 0.000
#> GSM39172 1 0.2003 0.8152 0.884 0.116 0 0.000 0.000 0.000
#> GSM39173 1 0.5057 -0.3632 0.560 0.352 0 0.000 0.000 0.088
#> GSM39174 1 0.0458 0.8766 0.984 0.016 0 0.000 0.000 0.000
#> GSM39175 1 0.0547 0.8756 0.980 0.020 0 0.000 0.000 0.000
#> GSM39176 1 0.0713 0.8756 0.972 0.028 0 0.000 0.000 0.000
#> GSM39177 1 0.1141 0.8643 0.948 0.052 0 0.000 0.000 0.000
#> GSM39178 1 0.1204 0.8641 0.944 0.056 0 0.000 0.000 0.000
#> GSM39179 1 0.2135 0.8094 0.872 0.128 0 0.000 0.000 0.000
#> GSM39180 3 0.0000 0.0000 0.000 0.000 1 0.000 0.000 0.000
#> GSM39181 1 0.3464 0.5144 0.688 0.312 0 0.000 0.000 0.000
#> GSM39182 1 0.2003 0.8152 0.884 0.116 0 0.000 0.000 0.000
#> GSM39183 1 0.3464 0.5144 0.688 0.312 0 0.000 0.000 0.000
#> GSM39184 1 0.0146 0.8770 0.996 0.004 0 0.000 0.000 0.000
#> GSM39185 1 0.3482 0.5062 0.684 0.316 0 0.000 0.000 0.000
#> GSM39186 1 0.0865 0.8734 0.964 0.036 0 0.000 0.000 0.000
#> GSM39187 1 0.0547 0.8756 0.980 0.020 0 0.000 0.000 0.000
#> GSM39116 2 0.5814 0.8713 0.364 0.448 0 0.000 0.000 0.188
#> GSM39117 4 0.5296 0.4040 0.000 0.100 0 0.452 0.448 0.000
#> GSM39118 2 0.5814 0.8713 0.364 0.448 0 0.000 0.000 0.188
#> GSM39119 2 0.5717 -0.0544 0.088 0.628 0 0.000 0.072 0.212
#> GSM39120 1 0.0790 0.8678 0.968 0.032 0 0.000 0.000 0.000
#> GSM39121 1 0.0865 0.8654 0.964 0.036 0 0.000 0.000 0.000
#> GSM39122 1 0.0865 0.8654 0.964 0.036 0 0.000 0.000 0.000
#> GSM39123 5 0.0000 -0.1327 0.000 0.000 0 0.000 1.000 0.000
#> GSM39124 2 0.5814 0.8727 0.364 0.448 0 0.000 0.000 0.188
#> GSM39125 1 0.0790 0.8678 0.968 0.032 0 0.000 0.000 0.000
#> GSM39126 1 0.0790 0.8678 0.968 0.032 0 0.000 0.000 0.000
#> GSM39127 2 0.5848 0.8583 0.380 0.428 0 0.000 0.000 0.192
#> GSM39128 1 0.4383 0.4029 0.716 0.176 0 0.000 0.000 0.108
#> GSM39129 6 0.0000 1.0000 0.000 0.000 0 0.000 0.000 1.000
#> GSM39130 4 0.0000 0.3971 0.000 0.000 0 1.000 0.000 0.000
#> GSM39131 2 0.5848 0.8583 0.380 0.428 0 0.000 0.000 0.192
#> GSM39132 2 0.5814 0.8727 0.364 0.448 0 0.000 0.000 0.188
#> GSM39133 5 0.4933 0.2561 0.064 0.432 0 0.000 0.504 0.000
#> GSM39134 6 0.0000 1.0000 0.000 0.000 0 0.000 0.000 1.000
#> GSM39135 2 0.5833 0.8714 0.364 0.444 0 0.000 0.000 0.192
#> GSM39136 2 0.5467 -0.1489 0.084 0.676 0 0.000 0.104 0.136
#> GSM39137 1 0.5725 -0.8040 0.420 0.416 0 0.000 0.000 0.164
#> GSM39138 6 0.0000 1.0000 0.000 0.000 0 0.000 0.000 1.000
#> GSM39139 6 0.0000 1.0000 0.000 0.000 0 0.000 0.000 1.000
#> GSM39140 1 0.0713 0.8697 0.972 0.028 0 0.000 0.000 0.000
#> GSM39141 1 0.0632 0.8712 0.976 0.024 0 0.000 0.000 0.000
#> GSM39142 1 0.0632 0.8712 0.976 0.024 0 0.000 0.000 0.000
#> GSM39143 1 0.0632 0.8712 0.976 0.024 0 0.000 0.000 0.000
#> GSM39144 6 0.0000 1.0000 0.000 0.000 0 0.000 0.000 1.000
#> GSM39145 2 0.5642 0.8328 0.388 0.460 0 0.000 0.000 0.152
#> GSM39146 1 0.5391 -0.4062 0.552 0.308 0 0.000 0.000 0.140
#> GSM39147 2 0.5814 0.8727 0.364 0.448 0 0.000 0.000 0.188
#> GSM39188 1 0.2219 0.8071 0.864 0.136 0 0.000 0.000 0.000
#> GSM39189 1 0.2135 0.8094 0.872 0.128 0 0.000 0.000 0.000
#> GSM39190 1 0.2135 0.8094 0.872 0.128 0 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) other(p) protocol(p) k
#> ATC:hclust 87 1.000 2.02e-01 1.02e-02 2
#> ATC:hclust 72 0.444 2.30e-03 2.86e-04 3
#> ATC:hclust 6 NA NA NA 4
#> ATC:hclust 75 0.634 3.16e-04 5.29e-06 5
#> ATC:hclust 76 0.492 4.42e-05 7.40e-05 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 8353 rows and 87 columns.
#> Top rows (835, 1670, 2506, 3341, 4176) 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 0.847 0.941 0.965 0.3771 0.607 0.607
#> 3 3 0.610 0.729 0.843 0.3623 0.919 0.870
#> 4 4 0.581 0.833 0.884 0.2674 0.764 0.581
#> 5 5 0.778 0.642 0.828 0.1047 0.994 0.983
#> 6 6 0.727 0.553 0.759 0.0544 0.946 0.842
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
#> GSM39104 1 0.000 0.981 1.000 0.000
#> GSM39105 1 0.000 0.981 1.000 0.000
#> GSM39106 1 0.000 0.981 1.000 0.000
#> GSM39107 1 0.000 0.981 1.000 0.000
#> GSM39108 1 0.000 0.981 1.000 0.000
#> GSM39109 1 0.000 0.981 1.000 0.000
#> GSM39110 1 0.000 0.981 1.000 0.000
#> GSM39111 1 0.000 0.981 1.000 0.000
#> GSM39112 1 0.000 0.981 1.000 0.000
#> GSM39113 1 0.000 0.981 1.000 0.000
#> GSM39114 2 0.518 0.934 0.116 0.884
#> GSM39115 1 0.000 0.981 1.000 0.000
#> GSM39148 1 0.000 0.981 1.000 0.000
#> GSM39149 1 0.000 0.981 1.000 0.000
#> GSM39150 1 0.000 0.981 1.000 0.000
#> GSM39151 1 0.000 0.981 1.000 0.000
#> GSM39152 1 0.000 0.981 1.000 0.000
#> GSM39153 1 0.000 0.981 1.000 0.000
#> GSM39154 1 0.000 0.981 1.000 0.000
#> GSM39155 1 0.000 0.981 1.000 0.000
#> GSM39156 1 0.000 0.981 1.000 0.000
#> GSM39157 1 0.000 0.981 1.000 0.000
#> GSM39158 1 0.000 0.981 1.000 0.000
#> GSM39159 1 0.000 0.981 1.000 0.000
#> GSM39160 1 0.000 0.981 1.000 0.000
#> GSM39161 1 0.000 0.981 1.000 0.000
#> GSM39162 1 0.000 0.981 1.000 0.000
#> GSM39163 1 0.000 0.981 1.000 0.000
#> GSM39164 1 0.000 0.981 1.000 0.000
#> GSM39165 1 0.000 0.981 1.000 0.000
#> GSM39166 1 0.000 0.981 1.000 0.000
#> GSM39167 1 0.000 0.981 1.000 0.000
#> GSM39168 1 0.000 0.981 1.000 0.000
#> GSM39169 1 0.000 0.981 1.000 0.000
#> GSM39170 1 0.000 0.981 1.000 0.000
#> GSM39171 1 0.000 0.981 1.000 0.000
#> GSM39172 1 0.000 0.981 1.000 0.000
#> GSM39173 2 0.552 0.922 0.128 0.872
#> GSM39174 1 0.000 0.981 1.000 0.000
#> GSM39175 1 0.000 0.981 1.000 0.000
#> GSM39176 1 0.000 0.981 1.000 0.000
#> GSM39177 1 0.000 0.981 1.000 0.000
#> GSM39178 1 0.000 0.981 1.000 0.000
#> GSM39179 1 0.000 0.981 1.000 0.000
#> GSM39180 2 0.373 0.947 0.072 0.928
#> GSM39181 1 0.000 0.981 1.000 0.000
#> GSM39182 1 0.000 0.981 1.000 0.000
#> GSM39183 1 0.000 0.981 1.000 0.000
#> GSM39184 1 0.000 0.981 1.000 0.000
#> GSM39185 1 0.000 0.981 1.000 0.000
#> GSM39186 1 0.000 0.981 1.000 0.000
#> GSM39187 1 0.000 0.981 1.000 0.000
#> GSM39116 2 0.518 0.934 0.116 0.884
#> GSM39117 2 0.000 0.910 0.000 1.000
#> GSM39118 2 0.518 0.934 0.116 0.884
#> GSM39119 2 0.327 0.948 0.060 0.940
#> GSM39120 1 0.000 0.981 1.000 0.000
#> GSM39121 1 0.000 0.981 1.000 0.000
#> GSM39122 1 0.000 0.981 1.000 0.000
#> GSM39123 2 0.000 0.910 0.000 1.000
#> GSM39124 2 0.518 0.934 0.116 0.884
#> GSM39125 1 0.000 0.981 1.000 0.000
#> GSM39126 1 0.000 0.981 1.000 0.000
#> GSM39127 2 0.518 0.934 0.116 0.884
#> GSM39128 1 0.814 0.614 0.748 0.252
#> GSM39129 2 0.327 0.948 0.060 0.940
#> GSM39130 2 0.000 0.910 0.000 1.000
#> GSM39131 2 0.946 0.534 0.364 0.636
#> GSM39132 2 0.327 0.948 0.060 0.940
#> GSM39133 2 0.000 0.910 0.000 1.000
#> GSM39134 2 0.327 0.948 0.060 0.940
#> GSM39135 2 0.518 0.934 0.116 0.884
#> GSM39136 2 0.327 0.948 0.060 0.940
#> GSM39137 1 0.999 -0.104 0.516 0.484
#> GSM39138 2 0.327 0.948 0.060 0.940
#> GSM39139 2 0.327 0.948 0.060 0.940
#> GSM39140 1 0.000 0.981 1.000 0.000
#> GSM39141 1 0.000 0.981 1.000 0.000
#> GSM39142 1 0.000 0.981 1.000 0.000
#> GSM39143 1 0.000 0.981 1.000 0.000
#> GSM39144 2 0.327 0.948 0.060 0.940
#> GSM39145 2 0.518 0.934 0.116 0.884
#> GSM39146 1 0.932 0.387 0.652 0.348
#> GSM39147 2 0.373 0.947 0.072 0.928
#> GSM39188 1 0.000 0.981 1.000 0.000
#> GSM39189 1 0.000 0.981 1.000 0.000
#> GSM39190 1 0.000 0.981 1.000 0.000
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM39104 1 0.207 0.8389 0.940 0.000 0.060
#> GSM39105 1 0.196 0.8393 0.944 0.000 0.056
#> GSM39106 1 0.311 0.8357 0.916 0.028 0.056
#> GSM39107 1 0.158 0.8407 0.964 0.028 0.008
#> GSM39108 1 0.311 0.8357 0.916 0.028 0.056
#> GSM39109 1 0.311 0.8357 0.916 0.028 0.056
#> GSM39110 1 0.311 0.8357 0.916 0.028 0.056
#> GSM39111 1 0.285 0.8378 0.924 0.020 0.056
#> GSM39112 1 0.311 0.8357 0.916 0.028 0.056
#> GSM39113 1 0.311 0.8357 0.916 0.028 0.056
#> GSM39114 2 0.304 0.6585 0.104 0.896 0.000
#> GSM39115 1 0.298 0.8368 0.920 0.024 0.056
#> GSM39148 1 0.103 0.8408 0.976 0.024 0.000
#> GSM39149 1 0.604 0.6972 0.620 0.000 0.380
#> GSM39150 1 0.604 0.6972 0.620 0.000 0.380
#> GSM39151 1 0.606 0.6940 0.616 0.000 0.384
#> GSM39152 1 0.604 0.6972 0.620 0.000 0.380
#> GSM39153 1 0.000 0.8459 1.000 0.000 0.000
#> GSM39154 1 0.000 0.8459 1.000 0.000 0.000
#> GSM39155 1 0.000 0.8459 1.000 0.000 0.000
#> GSM39156 1 0.000 0.8459 1.000 0.000 0.000
#> GSM39157 1 0.103 0.8408 0.976 0.024 0.000
#> GSM39158 1 0.450 0.7830 0.804 0.000 0.196
#> GSM39159 1 0.568 0.7172 0.684 0.000 0.316
#> GSM39160 1 0.604 0.6972 0.620 0.000 0.380
#> GSM39161 1 0.579 0.7084 0.668 0.000 0.332
#> GSM39162 1 0.103 0.8408 0.976 0.024 0.000
#> GSM39163 1 0.000 0.8459 1.000 0.000 0.000
#> GSM39164 1 0.000 0.8459 1.000 0.000 0.000
#> GSM39165 1 0.497 0.7623 0.764 0.000 0.236
#> GSM39166 1 0.604 0.6972 0.620 0.000 0.380
#> GSM39167 1 0.000 0.8459 1.000 0.000 0.000
#> GSM39168 1 0.103 0.8408 0.976 0.024 0.000
#> GSM39169 1 0.103 0.8408 0.976 0.024 0.000
#> GSM39170 1 0.103 0.8408 0.976 0.024 0.000
#> GSM39171 1 0.196 0.8393 0.944 0.000 0.056
#> GSM39172 1 0.606 0.6940 0.616 0.000 0.384
#> GSM39173 2 0.428 0.6324 0.072 0.872 0.056
#> GSM39174 1 0.103 0.8408 0.976 0.024 0.000
#> GSM39175 1 0.103 0.8439 0.976 0.000 0.024
#> GSM39176 1 0.000 0.8459 1.000 0.000 0.000
#> GSM39177 1 0.573 0.7117 0.676 0.000 0.324
#> GSM39178 1 0.604 0.6972 0.620 0.000 0.380
#> GSM39179 1 0.606 0.6940 0.616 0.000 0.384
#> GSM39180 2 0.923 0.0874 0.196 0.524 0.280
#> GSM39181 1 0.576 0.7112 0.672 0.000 0.328
#> GSM39182 1 0.576 0.7090 0.672 0.000 0.328
#> GSM39183 1 0.579 0.7105 0.668 0.000 0.332
#> GSM39184 1 0.000 0.8459 1.000 0.000 0.000
#> GSM39185 1 0.815 0.6180 0.580 0.088 0.332
#> GSM39186 1 0.196 0.8393 0.944 0.000 0.056
#> GSM39187 1 0.000 0.8459 1.000 0.000 0.000
#> GSM39116 2 0.263 0.6762 0.084 0.916 0.000
#> GSM39117 3 0.611 1.0000 0.000 0.396 0.604
#> GSM39118 2 0.287 0.6729 0.076 0.916 0.008
#> GSM39119 2 0.263 0.5286 0.000 0.916 0.084
#> GSM39120 1 0.116 0.8391 0.972 0.028 0.000
#> GSM39121 1 0.175 0.8280 0.952 0.048 0.000
#> GSM39122 1 0.175 0.8280 0.952 0.048 0.000
#> GSM39123 3 0.611 1.0000 0.000 0.396 0.604
#> GSM39124 2 0.263 0.6762 0.084 0.916 0.000
#> GSM39125 1 0.000 0.8459 1.000 0.000 0.000
#> GSM39126 1 0.271 0.7935 0.912 0.088 0.000
#> GSM39127 2 0.263 0.6762 0.084 0.916 0.000
#> GSM39128 2 0.620 0.2860 0.424 0.576 0.000
#> GSM39129 2 0.327 0.5008 0.000 0.884 0.116
#> GSM39130 3 0.611 1.0000 0.000 0.396 0.604
#> GSM39131 2 0.470 0.5272 0.212 0.788 0.000
#> GSM39132 2 0.000 0.6059 0.000 1.000 0.000
#> GSM39133 2 0.475 0.2258 0.000 0.784 0.216
#> GSM39134 2 0.327 0.5008 0.000 0.884 0.116
#> GSM39135 2 0.263 0.6762 0.084 0.916 0.000
#> GSM39136 2 0.116 0.5789 0.000 0.972 0.028
#> GSM39137 2 0.573 0.3962 0.324 0.676 0.000
#> GSM39138 2 0.327 0.5008 0.000 0.884 0.116
#> GSM39139 2 0.164 0.5746 0.000 0.956 0.044
#> GSM39140 1 0.175 0.8280 0.952 0.048 0.000
#> GSM39141 1 0.116 0.8391 0.972 0.028 0.000
#> GSM39142 1 0.103 0.8408 0.976 0.024 0.000
#> GSM39143 1 0.116 0.8391 0.972 0.028 0.000
#> GSM39144 2 0.327 0.5008 0.000 0.884 0.116
#> GSM39145 2 0.263 0.6762 0.084 0.916 0.000
#> GSM39146 2 0.631 0.2157 0.488 0.512 0.000
#> GSM39147 2 0.254 0.6751 0.080 0.920 0.000
#> GSM39188 1 0.606 0.6940 0.616 0.000 0.384
#> GSM39189 1 0.606 0.6940 0.616 0.000 0.384
#> GSM39190 1 0.606 0.6940 0.616 0.000 0.384
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM39104 1 0.4301 0.837 0.816 0.000 0.120 0.064
#> GSM39105 1 0.4428 0.839 0.816 0.004 0.116 0.064
#> GSM39106 1 0.4179 0.845 0.832 0.004 0.104 0.060
#> GSM39107 1 0.3431 0.873 0.876 0.004 0.060 0.060
#> GSM39108 1 0.4179 0.845 0.832 0.004 0.104 0.060
#> GSM39109 1 0.4254 0.841 0.828 0.004 0.104 0.064
#> GSM39110 1 0.4179 0.845 0.832 0.004 0.104 0.060
#> GSM39111 1 0.4428 0.839 0.816 0.004 0.116 0.064
#> GSM39112 1 0.4179 0.845 0.832 0.004 0.104 0.060
#> GSM39113 1 0.4179 0.845 0.832 0.004 0.104 0.060
#> GSM39114 2 0.1635 0.815 0.008 0.948 0.000 0.044
#> GSM39115 1 0.4353 0.842 0.820 0.004 0.116 0.060
#> GSM39148 1 0.0000 0.921 1.000 0.000 0.000 0.000
#> GSM39149 3 0.3280 0.872 0.124 0.000 0.860 0.016
#> GSM39150 3 0.4552 0.835 0.128 0.000 0.800 0.072
#> GSM39151 3 0.3224 0.870 0.120 0.000 0.864 0.016
#> GSM39152 3 0.3032 0.873 0.124 0.000 0.868 0.008
#> GSM39153 1 0.0469 0.920 0.988 0.000 0.012 0.000
#> GSM39154 1 0.0469 0.920 0.988 0.000 0.012 0.000
#> GSM39155 1 0.0657 0.920 0.984 0.000 0.012 0.004
#> GSM39156 1 0.0469 0.920 0.988 0.000 0.012 0.000
#> GSM39157 1 0.0188 0.921 0.996 0.000 0.004 0.000
#> GSM39158 3 0.5571 0.616 0.396 0.000 0.580 0.024
#> GSM39159 3 0.4262 0.823 0.236 0.000 0.756 0.008
#> GSM39160 3 0.4428 0.839 0.124 0.000 0.808 0.068
#> GSM39161 3 0.4238 0.845 0.176 0.000 0.796 0.028
#> GSM39162 1 0.0000 0.921 1.000 0.000 0.000 0.000
#> GSM39163 1 0.0469 0.920 0.988 0.000 0.012 0.000
#> GSM39164 1 0.0469 0.920 0.988 0.000 0.012 0.000
#> GSM39165 3 0.5168 0.405 0.492 0.000 0.504 0.004
#> GSM39166 3 0.3787 0.869 0.124 0.000 0.840 0.036
#> GSM39167 1 0.0657 0.919 0.984 0.000 0.012 0.004
#> GSM39168 1 0.0000 0.921 1.000 0.000 0.000 0.000
#> GSM39169 1 0.0188 0.921 0.996 0.000 0.000 0.004
#> GSM39170 1 0.0000 0.921 1.000 0.000 0.000 0.000
#> GSM39171 1 0.4301 0.837 0.816 0.000 0.120 0.064
#> GSM39172 3 0.3052 0.874 0.136 0.000 0.860 0.004
#> GSM39173 2 0.1909 0.809 0.008 0.940 0.048 0.004
#> GSM39174 1 0.0188 0.921 0.996 0.000 0.004 0.000
#> GSM39175 1 0.0779 0.917 0.980 0.000 0.016 0.004
#> GSM39176 1 0.0336 0.921 0.992 0.000 0.008 0.000
#> GSM39177 3 0.4053 0.833 0.228 0.000 0.768 0.004
#> GSM39178 3 0.3161 0.872 0.124 0.000 0.864 0.012
#> GSM39179 3 0.3161 0.873 0.124 0.000 0.864 0.012
#> GSM39180 3 0.5122 0.548 0.016 0.208 0.748 0.028
#> GSM39181 3 0.4576 0.821 0.232 0.000 0.748 0.020
#> GSM39182 3 0.3972 0.849 0.204 0.000 0.788 0.008
#> GSM39183 3 0.4204 0.851 0.192 0.000 0.788 0.020
#> GSM39184 1 0.0657 0.920 0.984 0.000 0.012 0.004
#> GSM39185 3 0.5037 0.746 0.100 0.072 0.800 0.028
#> GSM39186 1 0.4428 0.839 0.816 0.004 0.116 0.064
#> GSM39187 1 0.0657 0.919 0.984 0.000 0.012 0.004
#> GSM39116 2 0.0804 0.834 0.008 0.980 0.012 0.000
#> GSM39117 4 0.2704 0.993 0.000 0.124 0.000 0.876
#> GSM39118 2 0.0804 0.834 0.008 0.980 0.012 0.000
#> GSM39119 2 0.1767 0.812 0.000 0.944 0.012 0.044
#> GSM39120 1 0.0188 0.919 0.996 0.000 0.004 0.000
#> GSM39121 1 0.1004 0.900 0.972 0.024 0.004 0.000
#> GSM39122 1 0.1209 0.896 0.964 0.032 0.004 0.000
#> GSM39123 4 0.3280 0.987 0.000 0.124 0.016 0.860
#> GSM39124 2 0.0336 0.835 0.008 0.992 0.000 0.000
#> GSM39125 1 0.0657 0.917 0.984 0.000 0.012 0.004
#> GSM39126 1 0.3249 0.759 0.852 0.140 0.008 0.000
#> GSM39127 2 0.0937 0.834 0.012 0.976 0.012 0.000
#> GSM39128 2 0.5134 0.416 0.320 0.664 0.012 0.004
#> GSM39129 2 0.4780 0.702 0.000 0.788 0.096 0.116
#> GSM39130 4 0.2704 0.993 0.000 0.124 0.000 0.876
#> GSM39131 2 0.2473 0.784 0.080 0.908 0.012 0.000
#> GSM39132 2 0.0188 0.833 0.004 0.996 0.000 0.000
#> GSM39133 2 0.5883 0.367 0.000 0.640 0.060 0.300
#> GSM39134 2 0.4780 0.702 0.000 0.788 0.096 0.116
#> GSM39135 2 0.0804 0.834 0.008 0.980 0.012 0.000
#> GSM39136 2 0.1082 0.831 0.004 0.972 0.020 0.004
#> GSM39137 2 0.2928 0.752 0.108 0.880 0.012 0.000
#> GSM39138 2 0.4780 0.702 0.000 0.788 0.096 0.116
#> GSM39139 2 0.3080 0.774 0.000 0.880 0.096 0.024
#> GSM39140 1 0.0000 0.921 1.000 0.000 0.000 0.000
#> GSM39141 1 0.0000 0.921 1.000 0.000 0.000 0.000
#> GSM39142 1 0.0000 0.921 1.000 0.000 0.000 0.000
#> GSM39143 1 0.0000 0.921 1.000 0.000 0.000 0.000
#> GSM39144 2 0.4780 0.702 0.000 0.788 0.096 0.116
#> GSM39145 2 0.0336 0.835 0.008 0.992 0.000 0.000
#> GSM39146 2 0.5217 0.316 0.380 0.608 0.012 0.000
#> GSM39147 2 0.0336 0.835 0.008 0.992 0.000 0.000
#> GSM39188 3 0.3166 0.867 0.116 0.000 0.868 0.016
#> GSM39189 3 0.3280 0.873 0.124 0.000 0.860 0.016
#> GSM39190 3 0.3280 0.872 0.124 0.000 0.860 0.016
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM39104 1 0.1828 0.401 0.936 0.000 0.028 0.004 0.032
#> GSM39105 1 0.1582 0.417 0.944 0.000 0.028 0.000 0.028
#> GSM39106 1 0.0703 0.442 0.976 0.000 0.024 0.000 0.000
#> GSM39107 1 0.0000 0.453 1.000 0.000 0.000 0.000 0.000
#> GSM39108 1 0.0703 0.442 0.976 0.000 0.024 0.000 0.000
#> GSM39109 1 0.1300 0.423 0.956 0.000 0.028 0.000 0.016
#> GSM39110 1 0.0703 0.442 0.976 0.000 0.024 0.000 0.000
#> GSM39111 1 0.1493 0.418 0.948 0.000 0.028 0.000 0.024
#> GSM39112 1 0.0703 0.442 0.976 0.000 0.024 0.000 0.000
#> GSM39113 1 0.0865 0.438 0.972 0.000 0.024 0.000 0.004
#> GSM39114 2 0.1469 0.786 0.036 0.948 0.000 0.000 0.016
#> GSM39115 1 0.1106 0.435 0.964 0.000 0.024 0.000 0.012
#> GSM39148 1 0.4101 0.680 0.628 0.000 0.000 0.000 0.372
#> GSM39149 3 0.3479 0.749 0.080 0.000 0.836 0.000 0.084
#> GSM39150 3 0.5657 0.429 0.380 0.000 0.544 0.004 0.072
#> GSM39151 3 0.1704 0.782 0.004 0.000 0.928 0.000 0.068
#> GSM39152 3 0.2575 0.781 0.012 0.000 0.884 0.004 0.100
#> GSM39153 1 0.4150 0.671 0.612 0.000 0.000 0.000 0.388
#> GSM39154 1 0.4171 0.664 0.604 0.000 0.000 0.000 0.396
#> GSM39155 1 0.4138 0.664 0.616 0.000 0.000 0.000 0.384
#> GSM39156 1 0.4150 0.671 0.612 0.000 0.000 0.000 0.388
#> GSM39157 1 0.4101 0.680 0.628 0.000 0.000 0.000 0.372
#> GSM39158 3 0.6505 0.295 0.112 0.000 0.584 0.044 0.260
#> GSM39159 3 0.3752 0.696 0.048 0.000 0.804 0.000 0.148
#> GSM39160 3 0.5468 0.448 0.368 0.000 0.568 0.004 0.060
#> GSM39161 3 0.3809 0.758 0.016 0.000 0.824 0.044 0.116
#> GSM39162 1 0.4101 0.680 0.628 0.000 0.000 0.000 0.372
#> GSM39163 1 0.4161 0.668 0.608 0.000 0.000 0.000 0.392
#> GSM39164 1 0.4161 0.668 0.608 0.000 0.000 0.000 0.392
#> GSM39165 5 0.6638 0.000 0.224 0.000 0.364 0.000 0.412
#> GSM39166 3 0.5866 0.692 0.132 0.000 0.684 0.048 0.136
#> GSM39167 1 0.4201 0.648 0.592 0.000 0.000 0.000 0.408
#> GSM39168 1 0.4101 0.680 0.628 0.000 0.000 0.000 0.372
#> GSM39169 1 0.4045 0.675 0.644 0.000 0.000 0.000 0.356
#> GSM39170 1 0.4101 0.680 0.628 0.000 0.000 0.000 0.372
#> GSM39171 1 0.2067 0.406 0.924 0.000 0.028 0.004 0.044
#> GSM39172 3 0.0671 0.788 0.004 0.000 0.980 0.000 0.016
#> GSM39173 2 0.1216 0.795 0.000 0.960 0.020 0.000 0.020
#> GSM39174 1 0.4101 0.680 0.628 0.000 0.000 0.000 0.372
#> GSM39175 1 0.4375 0.626 0.576 0.000 0.004 0.000 0.420
#> GSM39176 1 0.4138 0.673 0.616 0.000 0.000 0.000 0.384
#> GSM39177 3 0.2740 0.767 0.028 0.000 0.876 0.000 0.096
#> GSM39178 3 0.4068 0.712 0.144 0.000 0.792 0.004 0.060
#> GSM39179 3 0.1831 0.781 0.004 0.000 0.920 0.000 0.076
#> GSM39180 3 0.5674 0.645 0.000 0.132 0.700 0.044 0.124
#> GSM39181 3 0.4708 0.703 0.028 0.000 0.752 0.044 0.176
#> GSM39182 3 0.2370 0.763 0.040 0.000 0.904 0.000 0.056
#> GSM39183 3 0.3951 0.757 0.016 0.000 0.812 0.044 0.128
#> GSM39184 1 0.4138 0.664 0.616 0.000 0.000 0.000 0.384
#> GSM39185 3 0.4328 0.746 0.012 0.012 0.796 0.044 0.136
#> GSM39186 1 0.1911 0.400 0.932 0.000 0.028 0.004 0.036
#> GSM39187 1 0.4201 0.648 0.592 0.000 0.000 0.000 0.408
#> GSM39116 2 0.0486 0.798 0.004 0.988 0.004 0.000 0.004
#> GSM39117 4 0.1341 1.000 0.000 0.056 0.000 0.944 0.000
#> GSM39118 2 0.0324 0.799 0.000 0.992 0.004 0.000 0.004
#> GSM39119 2 0.0703 0.793 0.000 0.976 0.000 0.024 0.000
#> GSM39120 1 0.4101 0.680 0.628 0.000 0.000 0.000 0.372
#> GSM39121 1 0.4101 0.680 0.628 0.000 0.000 0.000 0.372
#> GSM39122 1 0.4101 0.680 0.628 0.000 0.000 0.000 0.372
#> GSM39123 4 0.1341 1.000 0.000 0.056 0.000 0.944 0.000
#> GSM39124 2 0.0510 0.799 0.000 0.984 0.000 0.000 0.016
#> GSM39125 1 0.4171 0.658 0.604 0.000 0.000 0.000 0.396
#> GSM39126 1 0.5717 0.496 0.540 0.092 0.000 0.000 0.368
#> GSM39127 2 0.0486 0.798 0.004 0.988 0.004 0.000 0.004
#> GSM39128 2 0.4873 0.435 0.068 0.688 0.000 0.000 0.244
#> GSM39129 2 0.5304 0.477 0.000 0.560 0.000 0.056 0.384
#> GSM39130 4 0.1341 1.000 0.000 0.056 0.000 0.944 0.000
#> GSM39131 2 0.1412 0.779 0.008 0.952 0.004 0.000 0.036
#> GSM39132 2 0.0510 0.799 0.000 0.984 0.000 0.000 0.016
#> GSM39133 2 0.7025 0.250 0.000 0.568 0.216 0.124 0.092
#> GSM39134 2 0.5304 0.477 0.000 0.560 0.000 0.056 0.384
#> GSM39135 2 0.0324 0.799 0.000 0.992 0.004 0.000 0.004
#> GSM39136 2 0.0324 0.799 0.000 0.992 0.004 0.000 0.004
#> GSM39137 2 0.1484 0.770 0.008 0.944 0.000 0.000 0.048
#> GSM39138 2 0.5304 0.477 0.000 0.560 0.000 0.056 0.384
#> GSM39139 2 0.4403 0.526 0.000 0.608 0.000 0.008 0.384
#> GSM39140 1 0.4101 0.680 0.628 0.000 0.000 0.000 0.372
#> GSM39141 1 0.4101 0.680 0.628 0.000 0.000 0.000 0.372
#> GSM39142 1 0.4101 0.680 0.628 0.000 0.000 0.000 0.372
#> GSM39143 1 0.4101 0.680 0.628 0.000 0.000 0.000 0.372
#> GSM39144 2 0.5304 0.477 0.000 0.560 0.000 0.056 0.384
#> GSM39145 2 0.0510 0.799 0.000 0.984 0.000 0.000 0.016
#> GSM39146 2 0.4868 0.489 0.084 0.720 0.004 0.000 0.192
#> GSM39147 2 0.0510 0.799 0.000 0.984 0.000 0.000 0.016
#> GSM39188 3 0.1638 0.782 0.004 0.000 0.932 0.000 0.064
#> GSM39189 3 0.1569 0.790 0.008 0.000 0.944 0.004 0.044
#> GSM39190 3 0.1704 0.781 0.004 0.000 0.928 0.000 0.068
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM39104 1 0.6476 0.36366 0.352 0.004 0.004 0.344 0.004 0.292
#> GSM39105 1 0.6637 0.37697 0.372 0.008 0.004 0.328 0.008 0.280
#> GSM39106 1 0.6305 0.38713 0.372 0.008 0.000 0.340 0.000 0.280
#> GSM39107 1 0.6305 0.38713 0.372 0.008 0.000 0.340 0.000 0.280
#> GSM39108 1 0.6305 0.38713 0.372 0.008 0.000 0.340 0.000 0.280
#> GSM39109 1 0.6554 0.37305 0.356 0.008 0.004 0.348 0.004 0.280
#> GSM39110 1 0.6422 0.38671 0.372 0.008 0.000 0.340 0.004 0.276
#> GSM39111 1 0.6548 0.37721 0.372 0.008 0.004 0.332 0.004 0.280
#> GSM39112 1 0.6305 0.38713 0.372 0.008 0.000 0.340 0.000 0.280
#> GSM39113 1 0.6308 0.38049 0.364 0.008 0.000 0.348 0.000 0.280
#> GSM39114 2 0.1464 0.85181 0.000 0.944 0.000 0.016 0.004 0.036
#> GSM39115 1 0.6417 0.38503 0.384 0.008 0.000 0.324 0.004 0.280
#> GSM39148 1 0.0405 0.72876 0.988 0.000 0.000 0.008 0.000 0.004
#> GSM39149 3 0.1464 0.49443 0.000 0.000 0.944 0.036 0.004 0.016
#> GSM39150 6 0.7184 -0.31340 0.004 0.000 0.300 0.312 0.064 0.320
#> GSM39151 3 0.0725 0.50966 0.000 0.000 0.976 0.000 0.012 0.012
#> GSM39152 3 0.1275 0.51044 0.000 0.000 0.956 0.012 0.016 0.016
#> GSM39153 1 0.0551 0.72352 0.984 0.000 0.004 0.000 0.008 0.004
#> GSM39154 1 0.0922 0.71932 0.968 0.000 0.004 0.000 0.024 0.004
#> GSM39155 1 0.2170 0.70630 0.908 0.000 0.000 0.060 0.016 0.016
#> GSM39156 1 0.0436 0.72483 0.988 0.000 0.004 0.000 0.004 0.004
#> GSM39157 1 0.0146 0.72674 0.996 0.000 0.000 0.000 0.004 0.000
#> GSM39158 5 0.7101 0.26634 0.164 0.012 0.360 0.028 0.412 0.024
#> GSM39159 3 0.5867 -0.11960 0.288 0.012 0.544 0.000 0.152 0.004
#> GSM39160 3 0.7062 -0.00236 0.000 0.000 0.328 0.296 0.064 0.312
#> GSM39161 3 0.4274 -0.35161 0.000 0.012 0.552 0.000 0.432 0.004
#> GSM39162 1 0.0405 0.72876 0.988 0.000 0.000 0.008 0.000 0.004
#> GSM39163 1 0.0748 0.72132 0.976 0.000 0.004 0.000 0.016 0.004
#> GSM39164 1 0.0146 0.72674 0.996 0.000 0.000 0.000 0.004 0.000
#> GSM39165 1 0.4501 0.31918 0.684 0.000 0.256 0.004 0.052 0.004
#> GSM39166 3 0.6115 -0.32423 0.000 0.000 0.440 0.108 0.412 0.040
#> GSM39167 1 0.1364 0.70447 0.944 0.000 0.004 0.000 0.048 0.004
#> GSM39168 1 0.0405 0.72876 0.988 0.000 0.000 0.008 0.000 0.004
#> GSM39169 1 0.1297 0.72305 0.948 0.000 0.000 0.040 0.000 0.012
#> GSM39170 1 0.0922 0.72667 0.968 0.000 0.000 0.024 0.004 0.004
#> GSM39171 1 0.7019 0.36718 0.380 0.000 0.016 0.312 0.032 0.260
#> GSM39172 3 0.2030 0.48286 0.000 0.000 0.908 0.000 0.064 0.028
#> GSM39173 2 0.1364 0.85323 0.000 0.944 0.004 0.000 0.004 0.048
#> GSM39174 1 0.0000 0.72744 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM39175 1 0.1969 0.68836 0.920 0.000 0.020 0.004 0.052 0.004
#> GSM39176 1 0.0653 0.72194 0.980 0.000 0.004 0.000 0.012 0.004
#> GSM39177 3 0.1003 0.49199 0.028 0.000 0.964 0.000 0.004 0.004
#> GSM39178 3 0.5964 0.16317 0.000 0.000 0.604 0.096 0.084 0.216
#> GSM39179 3 0.0603 0.50778 0.000 0.000 0.980 0.000 0.016 0.004
#> GSM39180 5 0.6026 0.34052 0.000 0.092 0.308 0.004 0.548 0.048
#> GSM39181 3 0.5993 -0.41276 0.076 0.012 0.476 0.012 0.412 0.012
#> GSM39182 3 0.4884 0.17768 0.160 0.000 0.704 0.000 0.112 0.024
#> GSM39183 3 0.4978 -0.32905 0.004 0.012 0.540 0.012 0.416 0.016
#> GSM39184 1 0.2483 0.70132 0.896 0.000 0.004 0.060 0.024 0.016
#> GSM39185 5 0.4315 0.23999 0.000 0.012 0.492 0.000 0.492 0.004
#> GSM39186 1 0.6875 0.34556 0.348 0.008 0.004 0.324 0.020 0.296
#> GSM39187 1 0.1429 0.70192 0.940 0.000 0.004 0.000 0.052 0.004
#> GSM39116 2 0.0603 0.86186 0.000 0.980 0.000 0.000 0.016 0.004
#> GSM39117 4 0.4124 0.96705 0.000 0.008 0.000 0.648 0.332 0.012
#> GSM39118 2 0.0603 0.86186 0.000 0.980 0.000 0.000 0.016 0.004
#> GSM39119 2 0.1498 0.84338 0.000 0.940 0.000 0.000 0.032 0.028
#> GSM39120 1 0.1180 0.72580 0.960 0.000 0.004 0.024 0.004 0.008
#> GSM39121 1 0.1003 0.72648 0.964 0.000 0.000 0.028 0.004 0.004
#> GSM39122 1 0.1553 0.72351 0.944 0.008 0.000 0.032 0.004 0.012
#> GSM39123 4 0.4576 0.93280 0.000 0.008 0.000 0.556 0.412 0.024
#> GSM39124 2 0.1082 0.85568 0.000 0.956 0.000 0.000 0.004 0.040
#> GSM39125 1 0.2033 0.70184 0.916 0.000 0.004 0.020 0.056 0.004
#> GSM39126 1 0.3144 0.60179 0.832 0.136 0.000 0.020 0.004 0.008
#> GSM39127 2 0.0146 0.86471 0.000 0.996 0.000 0.000 0.000 0.004
#> GSM39128 2 0.3323 0.54221 0.204 0.780 0.000 0.000 0.008 0.008
#> GSM39129 6 0.3684 0.70632 0.000 0.372 0.000 0.000 0.000 0.628
#> GSM39130 4 0.4124 0.96705 0.000 0.008 0.000 0.648 0.332 0.012
#> GSM39131 2 0.0405 0.86278 0.008 0.988 0.000 0.000 0.000 0.004
#> GSM39132 2 0.1152 0.85422 0.000 0.952 0.000 0.000 0.004 0.044
#> GSM39133 2 0.5248 0.18344 0.000 0.496 0.012 0.024 0.444 0.024
#> GSM39134 6 0.3684 0.70632 0.000 0.372 0.000 0.000 0.000 0.628
#> GSM39135 2 0.0146 0.86471 0.000 0.996 0.000 0.000 0.000 0.004
#> GSM39136 2 0.1049 0.84994 0.000 0.960 0.000 0.000 0.032 0.008
#> GSM39137 2 0.0551 0.86351 0.008 0.984 0.000 0.000 0.004 0.004
#> GSM39138 6 0.3684 0.70632 0.000 0.372 0.000 0.000 0.000 0.628
#> GSM39139 6 0.3706 0.69419 0.000 0.380 0.000 0.000 0.000 0.620
#> GSM39140 1 0.1003 0.72648 0.964 0.000 0.000 0.028 0.004 0.004
#> GSM39141 1 0.0858 0.72742 0.968 0.000 0.000 0.028 0.000 0.004
#> GSM39142 1 0.0405 0.72884 0.988 0.000 0.000 0.008 0.000 0.004
#> GSM39143 1 0.0858 0.72742 0.968 0.000 0.000 0.028 0.000 0.004
#> GSM39144 6 0.3684 0.70632 0.000 0.372 0.000 0.000 0.000 0.628
#> GSM39145 2 0.1152 0.85422 0.000 0.952 0.000 0.000 0.004 0.044
#> GSM39146 2 0.3035 0.63282 0.148 0.828 0.000 0.000 0.008 0.016
#> GSM39147 2 0.1152 0.85422 0.000 0.952 0.000 0.000 0.004 0.044
#> GSM39188 3 0.0405 0.50757 0.000 0.000 0.988 0.000 0.008 0.004
#> GSM39189 3 0.2817 0.47109 0.000 0.000 0.868 0.008 0.072 0.052
#> GSM39190 3 0.0260 0.50940 0.000 0.000 0.992 0.000 0.008 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) other(p) protocol(p) k
#> ATC:kmeans 85 0.2206 6.68e-07 8.53e-07 2
#> ATC:kmeans 82 0.3221 7.22e-06 1.70e-06 3
#> ATC:kmeans 83 0.0125 3.24e-07 3.76e-08 4
#> ATC:kmeans 62 0.3640 1.78e-04 1.18e-05 5
#> ATC:kmeans 56 0.6365 1.93e-02 3.86e-03 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 8353 rows and 87 columns.
#> Top rows (835, 1670, 2506, 3341, 4176) 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 1.000 0.985 0.994 0.4639 0.536 0.536
#> 3 3 0.975 0.916 0.958 0.4150 0.776 0.595
#> 4 4 0.787 0.871 0.915 0.1317 0.867 0.637
#> 5 5 0.773 0.786 0.865 0.0615 0.966 0.866
#> 6 6 0.793 0.637 0.807 0.0419 0.942 0.752
suggest_best_k() suggests the best \(k\) based on these statistics. The rules are as follows:
suggest_best_k(res)
#> [1] 3
#> attr(,"optional")
#> [1] 2
There is also optional best \(k\) = 2 that is worth to check.
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
#> GSM39104 1 0.000 0.996 1.000 0.000
#> GSM39105 1 0.000 0.996 1.000 0.000
#> GSM39106 1 0.000 0.996 1.000 0.000
#> GSM39107 1 0.000 0.996 1.000 0.000
#> GSM39108 1 0.000 0.996 1.000 0.000
#> GSM39109 1 0.000 0.996 1.000 0.000
#> GSM39110 1 0.000 0.996 1.000 0.000
#> GSM39111 1 0.000 0.996 1.000 0.000
#> GSM39112 1 0.000 0.996 1.000 0.000
#> GSM39113 1 0.000 0.996 1.000 0.000
#> GSM39114 2 0.000 0.989 0.000 1.000
#> GSM39115 1 0.000 0.996 1.000 0.000
#> GSM39148 1 0.000 0.996 1.000 0.000
#> GSM39149 1 0.000 0.996 1.000 0.000
#> GSM39150 1 0.000 0.996 1.000 0.000
#> GSM39151 2 0.876 0.575 0.296 0.704
#> GSM39152 1 0.000 0.996 1.000 0.000
#> GSM39153 1 0.000 0.996 1.000 0.000
#> GSM39154 1 0.000 0.996 1.000 0.000
#> GSM39155 1 0.000 0.996 1.000 0.000
#> GSM39156 1 0.000 0.996 1.000 0.000
#> GSM39157 1 0.000 0.996 1.000 0.000
#> GSM39158 1 0.000 0.996 1.000 0.000
#> GSM39159 1 0.000 0.996 1.000 0.000
#> GSM39160 1 0.000 0.996 1.000 0.000
#> GSM39161 2 0.000 0.989 0.000 1.000
#> GSM39162 1 0.000 0.996 1.000 0.000
#> GSM39163 1 0.000 0.996 1.000 0.000
#> GSM39164 1 0.000 0.996 1.000 0.000
#> GSM39165 1 0.000 0.996 1.000 0.000
#> GSM39166 1 0.000 0.996 1.000 0.000
#> GSM39167 1 0.000 0.996 1.000 0.000
#> GSM39168 1 0.000 0.996 1.000 0.000
#> GSM39169 1 0.000 0.996 1.000 0.000
#> GSM39170 1 0.000 0.996 1.000 0.000
#> GSM39171 1 0.000 0.996 1.000 0.000
#> GSM39172 1 0.000 0.996 1.000 0.000
#> GSM39173 2 0.000 0.989 0.000 1.000
#> GSM39174 1 0.000 0.996 1.000 0.000
#> GSM39175 1 0.000 0.996 1.000 0.000
#> GSM39176 1 0.000 0.996 1.000 0.000
#> GSM39177 1 0.000 0.996 1.000 0.000
#> GSM39178 1 0.000 0.996 1.000 0.000
#> GSM39179 1 0.000 0.996 1.000 0.000
#> GSM39180 2 0.000 0.989 0.000 1.000
#> GSM39181 1 0.000 0.996 1.000 0.000
#> GSM39182 1 0.000 0.996 1.000 0.000
#> GSM39183 1 0.000 0.996 1.000 0.000
#> GSM39184 1 0.000 0.996 1.000 0.000
#> GSM39185 2 0.000 0.989 0.000 1.000
#> GSM39186 1 0.000 0.996 1.000 0.000
#> GSM39187 1 0.000 0.996 1.000 0.000
#> GSM39116 2 0.000 0.989 0.000 1.000
#> GSM39117 2 0.000 0.989 0.000 1.000
#> GSM39118 2 0.000 0.989 0.000 1.000
#> GSM39119 2 0.000 0.989 0.000 1.000
#> GSM39120 1 0.000 0.996 1.000 0.000
#> GSM39121 1 0.000 0.996 1.000 0.000
#> GSM39122 1 0.000 0.996 1.000 0.000
#> GSM39123 2 0.000 0.989 0.000 1.000
#> GSM39124 2 0.000 0.989 0.000 1.000
#> GSM39125 1 0.000 0.996 1.000 0.000
#> GSM39126 2 0.118 0.974 0.016 0.984
#> GSM39127 2 0.000 0.989 0.000 1.000
#> GSM39128 2 0.000 0.989 0.000 1.000
#> GSM39129 2 0.000 0.989 0.000 1.000
#> GSM39130 2 0.000 0.989 0.000 1.000
#> GSM39131 2 0.000 0.989 0.000 1.000
#> GSM39132 2 0.000 0.989 0.000 1.000
#> GSM39133 2 0.000 0.989 0.000 1.000
#> GSM39134 2 0.000 0.989 0.000 1.000
#> GSM39135 2 0.000 0.989 0.000 1.000
#> GSM39136 2 0.000 0.989 0.000 1.000
#> GSM39137 2 0.000 0.989 0.000 1.000
#> GSM39138 2 0.000 0.989 0.000 1.000
#> GSM39139 2 0.000 0.989 0.000 1.000
#> GSM39140 1 0.000 0.996 1.000 0.000
#> GSM39141 1 0.000 0.996 1.000 0.000
#> GSM39142 1 0.000 0.996 1.000 0.000
#> GSM39143 1 0.000 0.996 1.000 0.000
#> GSM39144 2 0.000 0.989 0.000 1.000
#> GSM39145 2 0.000 0.989 0.000 1.000
#> GSM39146 2 0.000 0.989 0.000 1.000
#> GSM39147 2 0.000 0.989 0.000 1.000
#> GSM39188 2 0.000 0.989 0.000 1.000
#> GSM39189 1 0.000 0.996 1.000 0.000
#> GSM39190 1 0.788 0.686 0.764 0.236
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM39104 1 0.2711 0.912 0.912 0.000 0.088
#> GSM39105 1 0.2711 0.912 0.912 0.000 0.088
#> GSM39106 1 0.2711 0.912 0.912 0.000 0.088
#> GSM39107 1 0.2711 0.912 0.912 0.000 0.088
#> GSM39108 1 0.2711 0.912 0.912 0.000 0.088
#> GSM39109 1 0.2711 0.912 0.912 0.000 0.088
#> GSM39110 1 0.2711 0.912 0.912 0.000 0.088
#> GSM39111 1 0.2711 0.912 0.912 0.000 0.088
#> GSM39112 1 0.2711 0.912 0.912 0.000 0.088
#> GSM39113 1 0.2711 0.912 0.912 0.000 0.088
#> GSM39114 2 0.0000 0.960 0.000 1.000 0.000
#> GSM39115 1 0.2711 0.912 0.912 0.000 0.088
#> GSM39148 1 0.0000 0.946 1.000 0.000 0.000
#> GSM39149 3 0.0000 0.960 0.000 0.000 1.000
#> GSM39150 3 0.0237 0.958 0.004 0.000 0.996
#> GSM39151 3 0.0000 0.960 0.000 0.000 1.000
#> GSM39152 3 0.0000 0.960 0.000 0.000 1.000
#> GSM39153 1 0.0000 0.946 1.000 0.000 0.000
#> GSM39154 1 0.0000 0.946 1.000 0.000 0.000
#> GSM39155 1 0.0000 0.946 1.000 0.000 0.000
#> GSM39156 1 0.0000 0.946 1.000 0.000 0.000
#> GSM39157 1 0.0000 0.946 1.000 0.000 0.000
#> GSM39158 3 0.2711 0.921 0.088 0.000 0.912
#> GSM39159 3 0.2711 0.921 0.088 0.000 0.912
#> GSM39160 3 0.0000 0.960 0.000 0.000 1.000
#> GSM39161 3 0.2056 0.942 0.024 0.024 0.952
#> GSM39162 1 0.0000 0.946 1.000 0.000 0.000
#> GSM39163 1 0.0000 0.946 1.000 0.000 0.000
#> GSM39164 1 0.0000 0.946 1.000 0.000 0.000
#> GSM39165 3 0.3267 0.900 0.116 0.000 0.884
#> GSM39166 3 0.0000 0.960 0.000 0.000 1.000
#> GSM39167 1 0.0000 0.946 1.000 0.000 0.000
#> GSM39168 1 0.0000 0.946 1.000 0.000 0.000
#> GSM39169 1 0.0000 0.946 1.000 0.000 0.000
#> GSM39170 1 0.0000 0.946 1.000 0.000 0.000
#> GSM39171 1 0.6192 0.393 0.580 0.000 0.420
#> GSM39172 3 0.0000 0.960 0.000 0.000 1.000
#> GSM39173 2 0.0000 0.960 0.000 1.000 0.000
#> GSM39174 1 0.0000 0.946 1.000 0.000 0.000
#> GSM39175 1 0.5760 0.490 0.672 0.000 0.328
#> GSM39176 1 0.0000 0.946 1.000 0.000 0.000
#> GSM39177 3 0.2711 0.921 0.088 0.000 0.912
#> GSM39178 3 0.0000 0.960 0.000 0.000 1.000
#> GSM39179 3 0.0000 0.960 0.000 0.000 1.000
#> GSM39180 2 0.5760 0.505 0.000 0.672 0.328
#> GSM39181 3 0.2711 0.921 0.088 0.000 0.912
#> GSM39182 3 0.2711 0.921 0.088 0.000 0.912
#> GSM39183 3 0.0000 0.960 0.000 0.000 1.000
#> GSM39184 1 0.0000 0.946 1.000 0.000 0.000
#> GSM39185 3 0.3038 0.876 0.000 0.104 0.896
#> GSM39186 1 0.5058 0.742 0.756 0.000 0.244
#> GSM39187 1 0.0237 0.944 0.996 0.000 0.004
#> GSM39116 2 0.0000 0.960 0.000 1.000 0.000
#> GSM39117 2 0.1529 0.933 0.000 0.960 0.040
#> GSM39118 2 0.0000 0.960 0.000 1.000 0.000
#> GSM39119 2 0.0000 0.960 0.000 1.000 0.000
#> GSM39120 1 0.0000 0.946 1.000 0.000 0.000
#> GSM39121 1 0.0000 0.946 1.000 0.000 0.000
#> GSM39122 1 0.0000 0.946 1.000 0.000 0.000
#> GSM39123 2 0.1529 0.933 0.000 0.960 0.040
#> GSM39124 2 0.0000 0.960 0.000 1.000 0.000
#> GSM39125 1 0.0237 0.944 0.996 0.000 0.004
#> GSM39126 2 0.6267 0.199 0.452 0.548 0.000
#> GSM39127 2 0.0000 0.960 0.000 1.000 0.000
#> GSM39128 2 0.0000 0.960 0.000 1.000 0.000
#> GSM39129 2 0.0000 0.960 0.000 1.000 0.000
#> GSM39130 2 0.1529 0.933 0.000 0.960 0.040
#> GSM39131 2 0.0000 0.960 0.000 1.000 0.000
#> GSM39132 2 0.0000 0.960 0.000 1.000 0.000
#> GSM39133 2 0.1529 0.933 0.000 0.960 0.040
#> GSM39134 2 0.0000 0.960 0.000 1.000 0.000
#> GSM39135 2 0.0000 0.960 0.000 1.000 0.000
#> GSM39136 2 0.0000 0.960 0.000 1.000 0.000
#> GSM39137 2 0.0000 0.960 0.000 1.000 0.000
#> GSM39138 2 0.0000 0.960 0.000 1.000 0.000
#> GSM39139 2 0.0000 0.960 0.000 1.000 0.000
#> GSM39140 1 0.0000 0.946 1.000 0.000 0.000
#> GSM39141 1 0.0000 0.946 1.000 0.000 0.000
#> GSM39142 1 0.0000 0.946 1.000 0.000 0.000
#> GSM39143 1 0.0000 0.946 1.000 0.000 0.000
#> GSM39144 2 0.0000 0.960 0.000 1.000 0.000
#> GSM39145 2 0.0000 0.960 0.000 1.000 0.000
#> GSM39146 2 0.0000 0.960 0.000 1.000 0.000
#> GSM39147 2 0.0000 0.960 0.000 1.000 0.000
#> GSM39188 3 0.0000 0.960 0.000 0.000 1.000
#> GSM39189 3 0.0000 0.960 0.000 0.000 1.000
#> GSM39190 3 0.0000 0.960 0.000 0.000 1.000
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM39104 4 0.2831 0.9552 0.120 0.000 0.004 0.876
#> GSM39105 4 0.2760 0.9564 0.128 0.000 0.000 0.872
#> GSM39106 4 0.2868 0.9548 0.136 0.000 0.000 0.864
#> GSM39107 4 0.3074 0.9451 0.152 0.000 0.000 0.848
#> GSM39108 4 0.2868 0.9551 0.136 0.000 0.000 0.864
#> GSM39109 4 0.2704 0.9565 0.124 0.000 0.000 0.876
#> GSM39110 4 0.2921 0.9527 0.140 0.000 0.000 0.860
#> GSM39111 4 0.2760 0.9564 0.128 0.000 0.000 0.872
#> GSM39112 4 0.2868 0.9551 0.136 0.000 0.000 0.864
#> GSM39113 4 0.2814 0.9558 0.132 0.000 0.000 0.868
#> GSM39114 2 0.0336 0.9157 0.000 0.992 0.000 0.008
#> GSM39115 4 0.2814 0.9555 0.132 0.000 0.000 0.868
#> GSM39148 1 0.0000 0.9552 1.000 0.000 0.000 0.000
#> GSM39149 3 0.4730 0.4234 0.000 0.000 0.636 0.364
#> GSM39150 4 0.2999 0.8030 0.004 0.000 0.132 0.864
#> GSM39151 3 0.0188 0.8859 0.000 0.000 0.996 0.004
#> GSM39152 3 0.1637 0.8670 0.000 0.000 0.940 0.060
#> GSM39153 1 0.0336 0.9563 0.992 0.000 0.000 0.008
#> GSM39154 1 0.0469 0.9551 0.988 0.000 0.000 0.012
#> GSM39155 1 0.3942 0.6759 0.764 0.000 0.000 0.236
#> GSM39156 1 0.0336 0.9563 0.992 0.000 0.000 0.008
#> GSM39157 1 0.0336 0.9563 0.992 0.000 0.000 0.008
#> GSM39158 3 0.2197 0.8616 0.048 0.000 0.928 0.024
#> GSM39159 3 0.0657 0.8848 0.012 0.000 0.984 0.004
#> GSM39160 4 0.3486 0.7342 0.000 0.000 0.188 0.812
#> GSM39161 3 0.0921 0.8792 0.000 0.000 0.972 0.028
#> GSM39162 1 0.0000 0.9552 1.000 0.000 0.000 0.000
#> GSM39163 1 0.0469 0.9551 0.988 0.000 0.000 0.012
#> GSM39164 1 0.0592 0.9531 0.984 0.000 0.000 0.016
#> GSM39165 3 0.5404 0.0656 0.476 0.000 0.512 0.012
#> GSM39166 3 0.2704 0.8254 0.000 0.000 0.876 0.124
#> GSM39167 1 0.0469 0.9551 0.988 0.000 0.000 0.012
#> GSM39168 1 0.0336 0.9563 0.992 0.000 0.000 0.008
#> GSM39169 1 0.3444 0.7667 0.816 0.000 0.000 0.184
#> GSM39170 1 0.0188 0.9542 0.996 0.000 0.000 0.004
#> GSM39171 4 0.3525 0.9270 0.100 0.000 0.040 0.860
#> GSM39172 3 0.0188 0.8856 0.000 0.000 0.996 0.004
#> GSM39173 2 0.0000 0.9194 0.000 1.000 0.000 0.000
#> GSM39174 1 0.0336 0.9563 0.992 0.000 0.000 0.008
#> GSM39175 1 0.2222 0.8994 0.924 0.000 0.060 0.016
#> GSM39176 1 0.0336 0.9563 0.992 0.000 0.000 0.008
#> GSM39177 3 0.2480 0.8337 0.088 0.000 0.904 0.008
#> GSM39178 3 0.3024 0.7935 0.000 0.000 0.852 0.148
#> GSM39179 3 0.0469 0.8853 0.000 0.000 0.988 0.012
#> GSM39180 3 0.6327 0.4783 0.000 0.228 0.648 0.124
#> GSM39181 3 0.0804 0.8857 0.008 0.000 0.980 0.012
#> GSM39182 3 0.0592 0.8840 0.000 0.000 0.984 0.016
#> GSM39183 3 0.0336 0.8851 0.000 0.000 0.992 0.008
#> GSM39184 1 0.3400 0.7712 0.820 0.000 0.000 0.180
#> GSM39185 3 0.2888 0.8160 0.000 0.004 0.872 0.124
#> GSM39186 4 0.3160 0.9445 0.108 0.000 0.020 0.872
#> GSM39187 1 0.0469 0.9551 0.988 0.000 0.000 0.012
#> GSM39116 2 0.2149 0.8928 0.000 0.912 0.000 0.088
#> GSM39117 2 0.6685 0.5492 0.000 0.592 0.284 0.124
#> GSM39118 2 0.2149 0.8925 0.000 0.912 0.000 0.088
#> GSM39119 2 0.2149 0.8925 0.000 0.912 0.000 0.088
#> GSM39120 1 0.0188 0.9542 0.996 0.000 0.000 0.004
#> GSM39121 1 0.0188 0.9538 0.996 0.004 0.000 0.000
#> GSM39122 1 0.0524 0.9506 0.988 0.008 0.000 0.004
#> GSM39123 2 0.6685 0.5492 0.000 0.592 0.284 0.124
#> GSM39124 2 0.0000 0.9194 0.000 1.000 0.000 0.000
#> GSM39125 1 0.0188 0.9542 0.996 0.000 0.000 0.004
#> GSM39126 1 0.3626 0.7448 0.812 0.184 0.000 0.004
#> GSM39127 2 0.0000 0.9194 0.000 1.000 0.000 0.000
#> GSM39128 2 0.0000 0.9194 0.000 1.000 0.000 0.000
#> GSM39129 2 0.0469 0.9186 0.000 0.988 0.000 0.012
#> GSM39130 2 0.6685 0.5492 0.000 0.592 0.284 0.124
#> GSM39131 2 0.0000 0.9194 0.000 1.000 0.000 0.000
#> GSM39132 2 0.0000 0.9194 0.000 1.000 0.000 0.000
#> GSM39133 2 0.6685 0.5492 0.000 0.592 0.284 0.124
#> GSM39134 2 0.0469 0.9186 0.000 0.988 0.000 0.012
#> GSM39135 2 0.0336 0.9191 0.000 0.992 0.000 0.008
#> GSM39136 2 0.2281 0.8883 0.000 0.904 0.000 0.096
#> GSM39137 2 0.0000 0.9194 0.000 1.000 0.000 0.000
#> GSM39138 2 0.0469 0.9186 0.000 0.988 0.000 0.012
#> GSM39139 2 0.0000 0.9194 0.000 1.000 0.000 0.000
#> GSM39140 1 0.0657 0.9463 0.984 0.012 0.000 0.004
#> GSM39141 1 0.0000 0.9552 1.000 0.000 0.000 0.000
#> GSM39142 1 0.0336 0.9563 0.992 0.000 0.000 0.008
#> GSM39143 1 0.0188 0.9545 0.996 0.000 0.000 0.004
#> GSM39144 2 0.0469 0.9186 0.000 0.988 0.000 0.012
#> GSM39145 2 0.0000 0.9194 0.000 1.000 0.000 0.000
#> GSM39146 2 0.1716 0.9007 0.000 0.936 0.000 0.064
#> GSM39147 2 0.0000 0.9194 0.000 1.000 0.000 0.000
#> GSM39188 3 0.0469 0.8860 0.000 0.000 0.988 0.012
#> GSM39189 3 0.1637 0.8657 0.000 0.000 0.940 0.060
#> GSM39190 3 0.0188 0.8859 0.000 0.000 0.996 0.004
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM39104 5 0.1012 0.906 0.020 0.000 0.012 0.000 0.968
#> GSM39105 5 0.1282 0.897 0.044 0.000 0.000 0.004 0.952
#> GSM39106 5 0.0703 0.906 0.024 0.000 0.000 0.000 0.976
#> GSM39107 5 0.1282 0.893 0.044 0.000 0.004 0.000 0.952
#> GSM39108 5 0.0703 0.906 0.024 0.000 0.000 0.000 0.976
#> GSM39109 5 0.0609 0.907 0.020 0.000 0.000 0.000 0.980
#> GSM39110 5 0.0963 0.895 0.036 0.000 0.000 0.000 0.964
#> GSM39111 5 0.0880 0.904 0.032 0.000 0.000 0.000 0.968
#> GSM39112 5 0.0703 0.906 0.024 0.000 0.000 0.000 0.976
#> GSM39113 5 0.0609 0.907 0.020 0.000 0.000 0.000 0.980
#> GSM39114 2 0.0290 0.919 0.000 0.992 0.000 0.000 0.008
#> GSM39115 5 0.0963 0.903 0.036 0.000 0.000 0.000 0.964
#> GSM39148 1 0.0865 0.885 0.972 0.000 0.000 0.004 0.024
#> GSM39149 3 0.3106 0.646 0.000 0.000 0.840 0.020 0.140
#> GSM39150 5 0.3123 0.777 0.000 0.000 0.160 0.012 0.828
#> GSM39151 3 0.1638 0.709 0.000 0.000 0.932 0.064 0.004
#> GSM39152 3 0.1168 0.713 0.000 0.000 0.960 0.032 0.008
#> GSM39153 1 0.1612 0.884 0.948 0.000 0.016 0.024 0.012
#> GSM39154 1 0.3474 0.848 0.856 0.000 0.052 0.068 0.024
#> GSM39155 1 0.5903 0.610 0.636 0.000 0.044 0.064 0.256
#> GSM39156 1 0.1806 0.883 0.940 0.000 0.016 0.028 0.016
#> GSM39157 1 0.0854 0.887 0.976 0.000 0.004 0.008 0.012
#> GSM39158 3 0.5905 0.503 0.044 0.000 0.532 0.392 0.032
#> GSM39159 3 0.5318 0.590 0.052 0.000 0.616 0.324 0.008
#> GSM39160 5 0.4811 0.191 0.000 0.000 0.452 0.020 0.528
#> GSM39161 3 0.4425 0.443 0.000 0.000 0.544 0.452 0.004
#> GSM39162 1 0.0865 0.885 0.972 0.000 0.000 0.004 0.024
#> GSM39163 1 0.2568 0.871 0.904 0.000 0.032 0.048 0.016
#> GSM39164 1 0.2086 0.881 0.928 0.000 0.016 0.028 0.028
#> GSM39165 3 0.5760 0.425 0.280 0.000 0.620 0.084 0.016
#> GSM39166 3 0.5289 0.591 0.000 0.000 0.616 0.312 0.072
#> GSM39167 1 0.3656 0.840 0.844 0.000 0.052 0.080 0.024
#> GSM39168 1 0.0955 0.886 0.968 0.000 0.000 0.004 0.028
#> GSM39169 1 0.3635 0.694 0.748 0.000 0.000 0.004 0.248
#> GSM39170 1 0.1243 0.881 0.960 0.000 0.004 0.008 0.028
#> GSM39171 5 0.6011 0.547 0.072 0.000 0.268 0.040 0.620
#> GSM39172 3 0.2690 0.667 0.000 0.000 0.844 0.156 0.000
#> GSM39173 2 0.0000 0.924 0.000 1.000 0.000 0.000 0.000
#> GSM39174 1 0.0404 0.887 0.988 0.000 0.000 0.000 0.012
#> GSM39175 1 0.5629 0.632 0.668 0.000 0.220 0.088 0.024
#> GSM39176 1 0.1612 0.884 0.948 0.000 0.016 0.024 0.012
#> GSM39177 3 0.2139 0.693 0.052 0.000 0.916 0.032 0.000
#> GSM39178 3 0.3527 0.671 0.000 0.000 0.828 0.056 0.116
#> GSM39179 3 0.1608 0.705 0.000 0.000 0.928 0.072 0.000
#> GSM39180 4 0.3437 0.811 0.000 0.048 0.120 0.832 0.000
#> GSM39181 3 0.5343 0.541 0.036 0.000 0.572 0.380 0.012
#> GSM39182 3 0.4791 0.135 0.012 0.000 0.524 0.460 0.004
#> GSM39183 3 0.4575 0.555 0.004 0.000 0.596 0.392 0.008
#> GSM39184 1 0.5792 0.697 0.680 0.000 0.052 0.080 0.188
#> GSM39185 4 0.3333 0.478 0.000 0.000 0.208 0.788 0.004
#> GSM39186 5 0.1808 0.893 0.044 0.000 0.012 0.008 0.936
#> GSM39187 1 0.3839 0.833 0.832 0.000 0.056 0.088 0.024
#> GSM39116 2 0.3913 0.579 0.000 0.676 0.000 0.324 0.000
#> GSM39117 4 0.4139 0.890 0.000 0.132 0.084 0.784 0.000
#> GSM39118 2 0.3561 0.688 0.000 0.740 0.000 0.260 0.000
#> GSM39119 2 0.3003 0.778 0.000 0.812 0.000 0.188 0.000
#> GSM39120 1 0.1710 0.874 0.940 0.000 0.004 0.016 0.040
#> GSM39121 1 0.1870 0.873 0.936 0.004 0.004 0.016 0.040
#> GSM39122 1 0.3328 0.826 0.848 0.012 0.004 0.016 0.120
#> GSM39123 4 0.4139 0.890 0.000 0.132 0.084 0.784 0.000
#> GSM39124 2 0.0000 0.924 0.000 1.000 0.000 0.000 0.000
#> GSM39125 1 0.4245 0.827 0.800 0.000 0.048 0.124 0.028
#> GSM39126 1 0.5045 0.556 0.672 0.280 0.004 0.016 0.028
#> GSM39127 2 0.0290 0.923 0.000 0.992 0.000 0.008 0.000
#> GSM39128 2 0.0898 0.912 0.000 0.972 0.000 0.020 0.008
#> GSM39129 2 0.0404 0.922 0.000 0.988 0.000 0.012 0.000
#> GSM39130 4 0.4139 0.890 0.000 0.132 0.084 0.784 0.000
#> GSM39131 2 0.0000 0.924 0.000 1.000 0.000 0.000 0.000
#> GSM39132 2 0.0000 0.924 0.000 1.000 0.000 0.000 0.000
#> GSM39133 4 0.4139 0.890 0.000 0.132 0.084 0.784 0.000
#> GSM39134 2 0.0404 0.922 0.000 0.988 0.000 0.012 0.000
#> GSM39135 2 0.0703 0.917 0.000 0.976 0.000 0.024 0.000
#> GSM39136 2 0.3857 0.601 0.000 0.688 0.000 0.312 0.000
#> GSM39137 2 0.0000 0.924 0.000 1.000 0.000 0.000 0.000
#> GSM39138 2 0.0404 0.922 0.000 0.988 0.000 0.012 0.000
#> GSM39139 2 0.0000 0.924 0.000 1.000 0.000 0.000 0.000
#> GSM39140 1 0.1996 0.871 0.932 0.008 0.004 0.016 0.040
#> GSM39141 1 0.1116 0.882 0.964 0.000 0.004 0.004 0.028
#> GSM39142 1 0.0794 0.888 0.972 0.000 0.000 0.000 0.028
#> GSM39143 1 0.1365 0.882 0.952 0.000 0.004 0.004 0.040
#> GSM39144 2 0.0404 0.922 0.000 0.988 0.000 0.012 0.000
#> GSM39145 2 0.0000 0.924 0.000 1.000 0.000 0.000 0.000
#> GSM39146 2 0.2929 0.786 0.000 0.820 0.000 0.180 0.000
#> GSM39147 2 0.0000 0.924 0.000 1.000 0.000 0.000 0.000
#> GSM39188 3 0.2230 0.690 0.000 0.000 0.884 0.116 0.000
#> GSM39189 3 0.2230 0.704 0.000 0.000 0.912 0.044 0.044
#> GSM39190 3 0.2074 0.694 0.000 0.000 0.896 0.104 0.000
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM39104 6 0.0146 0.9143 0.000 0.000 0.004 0.000 0.000 0.996
#> GSM39105 6 0.0951 0.9097 0.004 0.000 0.008 0.000 0.020 0.968
#> GSM39106 6 0.0520 0.9133 0.008 0.000 0.000 0.000 0.008 0.984
#> GSM39107 6 0.0820 0.9088 0.016 0.000 0.000 0.000 0.012 0.972
#> GSM39108 6 0.0405 0.9163 0.004 0.000 0.000 0.000 0.008 0.988
#> GSM39109 6 0.0000 0.9160 0.000 0.000 0.000 0.000 0.000 1.000
#> GSM39110 6 0.1003 0.9033 0.020 0.000 0.000 0.000 0.016 0.964
#> GSM39111 6 0.0260 0.9163 0.000 0.000 0.000 0.000 0.008 0.992
#> GSM39112 6 0.0405 0.9163 0.004 0.000 0.000 0.000 0.008 0.988
#> GSM39113 6 0.0405 0.9163 0.004 0.000 0.000 0.000 0.008 0.988
#> GSM39114 2 0.0653 0.9000 0.000 0.980 0.000 0.004 0.004 0.012
#> GSM39115 6 0.0653 0.9145 0.004 0.000 0.004 0.000 0.012 0.980
#> GSM39148 1 0.0405 0.6950 0.988 0.000 0.000 0.000 0.004 0.008
#> GSM39149 3 0.1196 0.7949 0.000 0.000 0.952 0.000 0.008 0.040
#> GSM39150 6 0.3651 0.6940 0.000 0.000 0.180 0.000 0.048 0.772
#> GSM39151 3 0.1053 0.8066 0.000 0.000 0.964 0.020 0.012 0.004
#> GSM39152 3 0.1124 0.7937 0.000 0.000 0.956 0.000 0.036 0.008
#> GSM39153 1 0.3586 0.6353 0.720 0.000 0.000 0.000 0.268 0.012
#> GSM39154 1 0.4192 0.4701 0.572 0.000 0.000 0.000 0.412 0.016
#> GSM39155 1 0.6010 0.2430 0.412 0.000 0.004 0.000 0.384 0.200
#> GSM39156 1 0.3717 0.6284 0.708 0.000 0.000 0.000 0.276 0.016
#> GSM39157 1 0.3110 0.6679 0.792 0.000 0.000 0.000 0.196 0.012
#> GSM39158 5 0.4566 0.5101 0.048 0.000 0.116 0.084 0.752 0.000
#> GSM39159 5 0.5742 0.3149 0.032 0.000 0.352 0.088 0.528 0.000
#> GSM39160 3 0.4855 0.4069 0.000 0.000 0.596 0.000 0.076 0.328
#> GSM39161 5 0.5941 0.1825 0.000 0.000 0.316 0.236 0.448 0.000
#> GSM39162 1 0.0405 0.6950 0.988 0.000 0.000 0.000 0.004 0.008
#> GSM39163 1 0.3984 0.5699 0.648 0.000 0.000 0.000 0.336 0.016
#> GSM39164 1 0.3652 0.6356 0.720 0.000 0.000 0.000 0.264 0.016
#> GSM39165 5 0.6141 0.2180 0.236 0.000 0.304 0.000 0.452 0.008
#> GSM39166 5 0.5713 0.2103 0.000 0.000 0.356 0.068 0.532 0.044
#> GSM39167 1 0.4212 0.4519 0.560 0.000 0.000 0.000 0.424 0.016
#> GSM39168 1 0.0820 0.6967 0.972 0.000 0.000 0.000 0.016 0.012
#> GSM39169 1 0.3511 0.5434 0.760 0.000 0.000 0.000 0.024 0.216
#> GSM39170 1 0.1732 0.6653 0.920 0.000 0.000 0.004 0.072 0.004
#> GSM39171 6 0.6958 -0.0861 0.072 0.000 0.200 0.000 0.352 0.376
#> GSM39172 3 0.2019 0.7788 0.000 0.000 0.900 0.088 0.012 0.000
#> GSM39173 2 0.0405 0.9036 0.000 0.988 0.004 0.008 0.000 0.000
#> GSM39174 1 0.2019 0.6963 0.900 0.000 0.000 0.000 0.088 0.012
#> GSM39175 5 0.5500 -0.2982 0.440 0.000 0.088 0.000 0.460 0.012
#> GSM39176 1 0.3564 0.6373 0.724 0.000 0.000 0.000 0.264 0.012
#> GSM39177 3 0.3142 0.6597 0.044 0.000 0.840 0.008 0.108 0.000
#> GSM39178 3 0.4186 0.6282 0.000 0.000 0.756 0.016 0.164 0.064
#> GSM39179 3 0.0993 0.8054 0.000 0.000 0.964 0.024 0.012 0.000
#> GSM39180 4 0.1599 0.6945 0.000 0.008 0.028 0.940 0.024 0.000
#> GSM39181 5 0.4519 0.4979 0.020 0.000 0.152 0.092 0.736 0.000
#> GSM39182 3 0.5405 0.2076 0.020 0.000 0.480 0.436 0.064 0.000
#> GSM39183 5 0.5008 0.3742 0.000 0.000 0.280 0.108 0.612 0.000
#> GSM39184 1 0.5634 0.2897 0.444 0.000 0.004 0.000 0.424 0.128
#> GSM39185 4 0.4660 0.1916 0.000 0.000 0.056 0.600 0.344 0.000
#> GSM39186 6 0.1970 0.8709 0.000 0.000 0.028 0.000 0.060 0.912
#> GSM39187 1 0.4238 0.4188 0.540 0.000 0.000 0.000 0.444 0.016
#> GSM39116 4 0.4328 0.0152 0.000 0.460 0.000 0.520 0.020 0.000
#> GSM39117 4 0.1261 0.7330 0.000 0.024 0.024 0.952 0.000 0.000
#> GSM39118 2 0.4101 0.2587 0.000 0.580 0.000 0.408 0.012 0.000
#> GSM39119 2 0.3565 0.5454 0.000 0.692 0.000 0.304 0.004 0.000
#> GSM39120 1 0.2531 0.6266 0.856 0.000 0.000 0.012 0.132 0.000
#> GSM39121 1 0.2841 0.6137 0.848 0.012 0.000 0.012 0.128 0.000
#> GSM39122 1 0.4228 0.5643 0.784 0.028 0.000 0.016 0.128 0.044
#> GSM39123 4 0.1261 0.7330 0.000 0.024 0.024 0.952 0.000 0.000
#> GSM39124 2 0.0291 0.9033 0.000 0.992 0.000 0.004 0.004 0.000
#> GSM39125 5 0.4192 -0.3181 0.412 0.000 0.000 0.016 0.572 0.000
#> GSM39126 1 0.5634 0.2726 0.572 0.280 0.000 0.016 0.132 0.000
#> GSM39127 2 0.1549 0.8870 0.000 0.936 0.000 0.044 0.020 0.000
#> GSM39128 2 0.2179 0.8614 0.000 0.900 0.000 0.036 0.064 0.000
#> GSM39129 2 0.0865 0.8982 0.000 0.964 0.000 0.036 0.000 0.000
#> GSM39130 4 0.1261 0.7330 0.000 0.024 0.024 0.952 0.000 0.000
#> GSM39131 2 0.1257 0.8947 0.000 0.952 0.000 0.028 0.020 0.000
#> GSM39132 2 0.0291 0.9033 0.000 0.992 0.000 0.004 0.004 0.000
#> GSM39133 4 0.1261 0.7330 0.000 0.024 0.024 0.952 0.000 0.000
#> GSM39134 2 0.0790 0.9001 0.000 0.968 0.000 0.032 0.000 0.000
#> GSM39135 2 0.1745 0.8762 0.000 0.920 0.000 0.068 0.012 0.000
#> GSM39136 4 0.4165 0.0390 0.000 0.452 0.000 0.536 0.012 0.000
#> GSM39137 2 0.0520 0.9017 0.000 0.984 0.000 0.008 0.008 0.000
#> GSM39138 2 0.0790 0.9001 0.000 0.968 0.000 0.032 0.000 0.000
#> GSM39139 2 0.0146 0.9040 0.000 0.996 0.000 0.004 0.000 0.000
#> GSM39140 1 0.2257 0.6337 0.876 0.000 0.000 0.008 0.116 0.000
#> GSM39141 1 0.0891 0.6882 0.968 0.000 0.000 0.000 0.024 0.008
#> GSM39142 1 0.2384 0.6962 0.884 0.000 0.000 0.000 0.084 0.032
#> GSM39143 1 0.1176 0.6870 0.956 0.000 0.000 0.000 0.020 0.024
#> GSM39144 2 0.0790 0.9001 0.000 0.968 0.000 0.032 0.000 0.000
#> GSM39145 2 0.0000 0.9039 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM39146 2 0.4011 0.5228 0.000 0.672 0.000 0.304 0.024 0.000
#> GSM39147 2 0.0000 0.9039 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM39188 3 0.1082 0.8037 0.000 0.000 0.956 0.040 0.004 0.000
#> GSM39189 3 0.1983 0.7854 0.000 0.000 0.916 0.012 0.060 0.012
#> GSM39190 3 0.1007 0.8039 0.000 0.000 0.956 0.044 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) other(p) protocol(p) k
#> ATC:skmeans 87 7.15e-02 3.03e-07 7.30e-07 2
#> ATC:skmeans 84 1.49e-03 3.65e-09 3.07e-09 3
#> ATC:skmeans 84 2.71e-11 4.43e-16 4.88e-16 4
#> ATC:skmeans 82 2.54e-11 3.98e-14 6.24e-15 5
#> ATC:skmeans 66 3.12e-09 1.04e-10 3.18e-11 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 8353 rows and 87 columns.
#> Top rows (835, 1670, 2506, 3341, 4176) 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 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 0.858 0.896 0.957 0.3858 0.607 0.607
#> 3 3 0.768 0.843 0.937 0.1070 0.984 0.974
#> 4 4 0.582 0.631 0.823 0.4854 0.773 0.616
#> 5 5 0.554 0.691 0.794 0.0746 0.893 0.742
#> 6 6 0.537 0.474 0.716 0.0481 0.856 0.618
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
#> GSM39104 1 0.0000 0.9697 1.000 0.000
#> GSM39105 1 0.0000 0.9697 1.000 0.000
#> GSM39106 1 0.0000 0.9697 1.000 0.000
#> GSM39107 1 0.0000 0.9697 1.000 0.000
#> GSM39108 1 0.0000 0.9697 1.000 0.000
#> GSM39109 1 0.0000 0.9697 1.000 0.000
#> GSM39110 1 0.0000 0.9697 1.000 0.000
#> GSM39111 1 0.0000 0.9697 1.000 0.000
#> GSM39112 1 0.0000 0.9697 1.000 0.000
#> GSM39113 1 0.0000 0.9697 1.000 0.000
#> GSM39114 2 0.0000 0.9027 0.000 1.000
#> GSM39115 1 0.0000 0.9697 1.000 0.000
#> GSM39148 1 0.0000 0.9697 1.000 0.000
#> GSM39149 1 0.0000 0.9697 1.000 0.000
#> GSM39150 1 0.0000 0.9697 1.000 0.000
#> GSM39151 1 0.0672 0.9634 0.992 0.008
#> GSM39152 1 0.0000 0.9697 1.000 0.000
#> GSM39153 1 0.0000 0.9697 1.000 0.000
#> GSM39154 1 0.0000 0.9697 1.000 0.000
#> GSM39155 1 0.0000 0.9697 1.000 0.000
#> GSM39156 1 0.0000 0.9697 1.000 0.000
#> GSM39157 1 0.0000 0.9697 1.000 0.000
#> GSM39158 1 0.0000 0.9697 1.000 0.000
#> GSM39159 1 0.0000 0.9697 1.000 0.000
#> GSM39160 1 0.0000 0.9697 1.000 0.000
#> GSM39161 1 0.4298 0.8871 0.912 0.088
#> GSM39162 1 0.0000 0.9697 1.000 0.000
#> GSM39163 1 0.0000 0.9697 1.000 0.000
#> GSM39164 1 0.0000 0.9697 1.000 0.000
#> GSM39165 1 0.0000 0.9697 1.000 0.000
#> GSM39166 1 0.0000 0.9697 1.000 0.000
#> GSM39167 1 0.0000 0.9697 1.000 0.000
#> GSM39168 1 0.0000 0.9697 1.000 0.000
#> GSM39169 1 0.0000 0.9697 1.000 0.000
#> GSM39170 1 0.0000 0.9697 1.000 0.000
#> GSM39171 1 0.0000 0.9697 1.000 0.000
#> GSM39172 1 0.0000 0.9697 1.000 0.000
#> GSM39173 2 0.0000 0.9027 0.000 1.000
#> GSM39174 1 0.0000 0.9697 1.000 0.000
#> GSM39175 1 0.0000 0.9697 1.000 0.000
#> GSM39176 1 0.0000 0.9697 1.000 0.000
#> GSM39177 1 0.0000 0.9697 1.000 0.000
#> GSM39178 1 0.0000 0.9697 1.000 0.000
#> GSM39179 1 0.0000 0.9697 1.000 0.000
#> GSM39180 1 0.9944 0.0564 0.544 0.456
#> GSM39181 1 0.0000 0.9697 1.000 0.000
#> GSM39182 1 0.0000 0.9697 1.000 0.000
#> GSM39183 1 0.3114 0.9205 0.944 0.056
#> GSM39184 1 0.0000 0.9697 1.000 0.000
#> GSM39185 1 0.6048 0.8125 0.852 0.148
#> GSM39186 1 0.0000 0.9697 1.000 0.000
#> GSM39187 1 0.0000 0.9697 1.000 0.000
#> GSM39116 2 0.5178 0.8215 0.116 0.884
#> GSM39117 2 0.0000 0.9027 0.000 1.000
#> GSM39118 2 0.8386 0.6472 0.268 0.732
#> GSM39119 2 0.0000 0.9027 0.000 1.000
#> GSM39120 1 0.0000 0.9697 1.000 0.000
#> GSM39121 1 0.3274 0.9169 0.940 0.060
#> GSM39122 1 0.3114 0.9206 0.944 0.056
#> GSM39123 2 0.9732 0.3825 0.404 0.596
#> GSM39124 2 0.0000 0.9027 0.000 1.000
#> GSM39125 1 0.0000 0.9697 1.000 0.000
#> GSM39126 1 0.5519 0.8414 0.872 0.128
#> GSM39127 2 0.1633 0.8890 0.024 0.976
#> GSM39128 1 0.9087 0.4842 0.676 0.324
#> GSM39129 2 0.0000 0.9027 0.000 1.000
#> GSM39130 2 0.0000 0.9027 0.000 1.000
#> GSM39131 2 0.9933 0.2464 0.452 0.548
#> GSM39132 2 0.0000 0.9027 0.000 1.000
#> GSM39133 2 0.8207 0.6657 0.256 0.744
#> GSM39134 2 0.0000 0.9027 0.000 1.000
#> GSM39135 2 0.0000 0.9027 0.000 1.000
#> GSM39136 2 0.0000 0.9027 0.000 1.000
#> GSM39137 2 0.9988 0.1516 0.480 0.520
#> GSM39138 2 0.0000 0.9027 0.000 1.000
#> GSM39139 2 0.0000 0.9027 0.000 1.000
#> GSM39140 1 0.3274 0.9169 0.940 0.060
#> GSM39141 1 0.0000 0.9697 1.000 0.000
#> GSM39142 1 0.0000 0.9697 1.000 0.000
#> GSM39143 1 0.0000 0.9697 1.000 0.000
#> GSM39144 2 0.0000 0.9027 0.000 1.000
#> GSM39145 2 0.0000 0.9027 0.000 1.000
#> GSM39146 1 0.7745 0.6868 0.772 0.228
#> GSM39147 2 0.0000 0.9027 0.000 1.000
#> GSM39188 1 0.4298 0.8868 0.912 0.088
#> GSM39189 1 0.0000 0.9697 1.000 0.000
#> GSM39190 1 0.0000 0.9697 1.000 0.000
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM39104 1 0.0000 0.9477 1.000 0.000 0.000
#> GSM39105 1 0.0000 0.9477 1.000 0.000 0.000
#> GSM39106 1 0.0237 0.9476 0.996 0.000 0.004
#> GSM39107 1 0.0237 0.9476 0.996 0.000 0.004
#> GSM39108 1 0.0237 0.9476 0.996 0.000 0.004
#> GSM39109 1 0.0000 0.9477 1.000 0.000 0.000
#> GSM39110 1 0.0000 0.9477 1.000 0.000 0.000
#> GSM39111 1 0.0000 0.9477 1.000 0.000 0.000
#> GSM39112 1 0.0237 0.9476 0.996 0.000 0.004
#> GSM39113 1 0.0237 0.9476 0.996 0.000 0.004
#> GSM39114 2 0.0000 0.7898 0.000 1.000 0.000
#> GSM39115 1 0.0000 0.9477 1.000 0.000 0.000
#> GSM39148 1 0.0237 0.9476 0.996 0.000 0.004
#> GSM39149 1 0.0000 0.9477 1.000 0.000 0.000
#> GSM39150 1 0.1411 0.9331 0.964 0.000 0.036
#> GSM39151 1 0.1964 0.9232 0.944 0.000 0.056
#> GSM39152 1 0.1964 0.9232 0.944 0.000 0.056
#> GSM39153 1 0.0000 0.9477 1.000 0.000 0.000
#> GSM39154 1 0.0237 0.9476 0.996 0.000 0.004
#> GSM39155 1 0.0000 0.9477 1.000 0.000 0.000
#> GSM39156 1 0.0000 0.9477 1.000 0.000 0.000
#> GSM39157 1 0.0237 0.9476 0.996 0.000 0.004
#> GSM39158 1 0.1964 0.9232 0.944 0.000 0.056
#> GSM39159 1 0.0892 0.9411 0.980 0.000 0.020
#> GSM39160 1 0.1411 0.9331 0.964 0.000 0.036
#> GSM39161 1 0.4458 0.8530 0.864 0.080 0.056
#> GSM39162 1 0.0237 0.9476 0.996 0.000 0.004
#> GSM39163 1 0.0000 0.9477 1.000 0.000 0.000
#> GSM39164 1 0.0000 0.9477 1.000 0.000 0.000
#> GSM39165 1 0.0000 0.9477 1.000 0.000 0.000
#> GSM39166 1 0.1964 0.9232 0.944 0.000 0.056
#> GSM39167 1 0.0237 0.9476 0.996 0.000 0.004
#> GSM39168 1 0.0237 0.9476 0.996 0.000 0.004
#> GSM39169 1 0.0000 0.9477 1.000 0.000 0.000
#> GSM39170 1 0.0237 0.9476 0.996 0.000 0.004
#> GSM39171 1 0.0000 0.9477 1.000 0.000 0.000
#> GSM39172 1 0.1031 0.9396 0.976 0.000 0.024
#> GSM39173 2 0.0000 0.7898 0.000 1.000 0.000
#> GSM39174 1 0.0237 0.9476 0.996 0.000 0.004
#> GSM39175 1 0.0000 0.9477 1.000 0.000 0.000
#> GSM39176 1 0.0237 0.9476 0.996 0.000 0.004
#> GSM39177 1 0.0000 0.9477 1.000 0.000 0.000
#> GSM39178 1 0.1964 0.9232 0.944 0.000 0.056
#> GSM39179 1 0.0747 0.9428 0.984 0.000 0.016
#> GSM39180 1 0.7853 0.2674 0.556 0.384 0.060
#> GSM39181 1 0.1964 0.9232 0.944 0.000 0.056
#> GSM39182 1 0.0237 0.9476 0.996 0.000 0.004
#> GSM39183 1 0.2384 0.9185 0.936 0.008 0.056
#> GSM39184 1 0.0000 0.9477 1.000 0.000 0.000
#> GSM39185 1 0.6203 0.7222 0.760 0.184 0.056
#> GSM39186 1 0.0592 0.9442 0.988 0.000 0.012
#> GSM39187 1 0.0000 0.9477 1.000 0.000 0.000
#> GSM39116 2 0.5982 0.3912 0.328 0.668 0.004
#> GSM39117 3 0.1964 0.9422 0.000 0.056 0.944
#> GSM39118 2 0.2959 0.6903 0.100 0.900 0.000
#> GSM39119 2 0.0000 0.7898 0.000 1.000 0.000
#> GSM39120 1 0.0237 0.9476 0.996 0.000 0.004
#> GSM39121 1 0.3851 0.8228 0.860 0.136 0.004
#> GSM39122 1 0.3851 0.8228 0.860 0.136 0.004
#> GSM39123 3 0.2165 0.9000 0.000 0.064 0.936
#> GSM39124 2 0.0000 0.7898 0.000 1.000 0.000
#> GSM39125 1 0.1031 0.9415 0.976 0.000 0.024
#> GSM39126 1 0.4978 0.7124 0.780 0.216 0.004
#> GSM39127 2 0.3192 0.6767 0.112 0.888 0.000
#> GSM39128 1 0.6079 0.3501 0.612 0.388 0.000
#> GSM39129 2 0.2066 0.7609 0.000 0.940 0.060
#> GSM39130 3 0.1753 0.9455 0.000 0.048 0.952
#> GSM39131 2 0.6495 0.1343 0.460 0.536 0.004
#> GSM39132 2 0.0000 0.7898 0.000 1.000 0.000
#> GSM39133 2 0.8779 0.0126 0.112 0.472 0.416
#> GSM39134 2 0.2066 0.7609 0.000 0.940 0.060
#> GSM39135 2 0.0000 0.7898 0.000 1.000 0.000
#> GSM39136 2 0.0000 0.7898 0.000 1.000 0.000
#> GSM39137 2 0.6228 0.3458 0.372 0.624 0.004
#> GSM39138 2 0.2066 0.7609 0.000 0.940 0.060
#> GSM39139 2 0.2066 0.7609 0.000 0.940 0.060
#> GSM39140 1 0.3851 0.8228 0.860 0.136 0.004
#> GSM39141 1 0.0237 0.9476 0.996 0.000 0.004
#> GSM39142 1 0.0237 0.9476 0.996 0.000 0.004
#> GSM39143 1 0.0237 0.9476 0.996 0.000 0.004
#> GSM39144 2 0.2066 0.7609 0.000 0.940 0.060
#> GSM39145 2 0.0000 0.7898 0.000 1.000 0.000
#> GSM39146 1 0.6359 0.2850 0.592 0.404 0.004
#> GSM39147 2 0.0000 0.7898 0.000 1.000 0.000
#> GSM39188 1 0.4868 0.8298 0.844 0.100 0.056
#> GSM39189 1 0.1964 0.9232 0.944 0.000 0.056
#> GSM39190 1 0.1031 0.9396 0.976 0.000 0.024
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM39104 1 0.4855 0.5351 0.600 0.000 0.400 0.0
#> GSM39105 1 0.4790 0.5511 0.620 0.000 0.380 0.0
#> GSM39106 1 0.4661 0.5651 0.652 0.000 0.348 0.0
#> GSM39107 1 0.3172 0.6763 0.840 0.000 0.160 0.0
#> GSM39108 1 0.4624 0.5902 0.660 0.000 0.340 0.0
#> GSM39109 1 0.4817 0.5410 0.612 0.000 0.388 0.0
#> GSM39110 1 0.0188 0.6953 0.996 0.000 0.004 0.0
#> GSM39111 1 0.4817 0.5410 0.612 0.000 0.388 0.0
#> GSM39112 1 0.2408 0.6865 0.896 0.000 0.104 0.0
#> GSM39113 1 0.4776 0.5548 0.624 0.000 0.376 0.0
#> GSM39114 2 0.0000 0.8998 0.000 1.000 0.000 0.0
#> GSM39115 1 0.4790 0.5511 0.620 0.000 0.380 0.0
#> GSM39148 1 0.0000 0.6938 1.000 0.000 0.000 0.0
#> GSM39149 1 0.4817 0.5410 0.612 0.000 0.388 0.0
#> GSM39150 3 0.3311 0.6807 0.172 0.000 0.828 0.0
#> GSM39151 3 0.1557 0.7467 0.056 0.000 0.944 0.0
#> GSM39152 3 0.1557 0.7463 0.056 0.000 0.944 0.0
#> GSM39153 1 0.0707 0.6955 0.980 0.000 0.020 0.0
#> GSM39154 1 0.4585 0.5908 0.668 0.000 0.332 0.0
#> GSM39155 1 0.4790 0.5511 0.620 0.000 0.380 0.0
#> GSM39156 1 0.1792 0.6982 0.932 0.000 0.068 0.0
#> GSM39157 1 0.2281 0.6927 0.904 0.000 0.096 0.0
#> GSM39158 3 0.1302 0.7452 0.044 0.000 0.956 0.0
#> GSM39159 1 0.4996 0.3069 0.516 0.000 0.484 0.0
#> GSM39160 3 0.3123 0.6814 0.156 0.000 0.844 0.0
#> GSM39161 3 0.2647 0.6695 0.120 0.000 0.880 0.0
#> GSM39162 1 0.0000 0.6938 1.000 0.000 0.000 0.0
#> GSM39163 1 0.4730 0.5772 0.636 0.000 0.364 0.0
#> GSM39164 1 0.4103 0.6308 0.744 0.000 0.256 0.0
#> GSM39165 1 0.4431 0.5981 0.696 0.000 0.304 0.0
#> GSM39166 3 0.1302 0.7452 0.044 0.000 0.956 0.0
#> GSM39167 1 0.4331 0.6096 0.712 0.000 0.288 0.0
#> GSM39168 1 0.0000 0.6938 1.000 0.000 0.000 0.0
#> GSM39169 1 0.0188 0.6953 0.996 0.000 0.004 0.0
#> GSM39170 1 0.0469 0.6928 0.988 0.000 0.012 0.0
#> GSM39171 1 0.4843 0.5376 0.604 0.000 0.396 0.0
#> GSM39172 3 0.4996 -0.1856 0.484 0.000 0.516 0.0
#> GSM39173 2 0.2345 0.8196 0.100 0.900 0.000 0.0
#> GSM39174 1 0.0707 0.6959 0.980 0.000 0.020 0.0
#> GSM39175 1 0.4843 0.5412 0.604 0.000 0.396 0.0
#> GSM39176 1 0.3356 0.6687 0.824 0.000 0.176 0.0
#> GSM39177 1 0.4304 0.6323 0.716 0.000 0.284 0.0
#> GSM39178 3 0.1302 0.7452 0.044 0.000 0.956 0.0
#> GSM39179 3 0.5000 -0.3330 0.500 0.000 0.500 0.0
#> GSM39180 3 0.6844 0.0416 0.260 0.152 0.588 0.0
#> GSM39181 3 0.1022 0.7455 0.032 0.000 0.968 0.0
#> GSM39182 1 0.1211 0.6934 0.960 0.000 0.040 0.0
#> GSM39183 3 0.1022 0.7455 0.032 0.000 0.968 0.0
#> GSM39184 1 0.4790 0.5511 0.620 0.000 0.380 0.0
#> GSM39185 3 0.1833 0.6907 0.024 0.032 0.944 0.0
#> GSM39186 3 0.4661 0.2906 0.348 0.000 0.652 0.0
#> GSM39187 1 0.4643 0.5893 0.656 0.000 0.344 0.0
#> GSM39116 2 0.5090 0.4898 0.324 0.660 0.016 0.0
#> GSM39117 4 0.0000 0.8365 0.000 0.000 0.000 1.0
#> GSM39118 2 0.2408 0.8462 0.044 0.920 0.036 0.0
#> GSM39119 2 0.0000 0.8998 0.000 1.000 0.000 0.0
#> GSM39120 1 0.0469 0.6928 0.988 0.000 0.012 0.0
#> GSM39121 1 0.1004 0.6793 0.972 0.024 0.004 0.0
#> GSM39122 1 0.1936 0.6770 0.940 0.032 0.028 0.0
#> GSM39123 4 0.0000 0.8365 0.000 0.000 0.000 1.0
#> GSM39124 2 0.0000 0.8998 0.000 1.000 0.000 0.0
#> GSM39125 1 0.3400 0.5868 0.820 0.000 0.180 0.0
#> GSM39126 1 0.1854 0.6639 0.940 0.048 0.012 0.0
#> GSM39127 2 0.2021 0.8572 0.056 0.932 0.012 0.0
#> GSM39128 1 0.5310 -0.0562 0.576 0.412 0.012 0.0
#> GSM39129 2 0.1022 0.8890 0.000 0.968 0.032 0.0
#> GSM39130 4 0.0000 0.8365 0.000 0.000 0.000 1.0
#> GSM39131 2 0.4898 0.5907 0.260 0.716 0.024 0.0
#> GSM39132 2 0.0000 0.8998 0.000 1.000 0.000 0.0
#> GSM39133 4 0.7902 0.2459 0.004 0.364 0.232 0.4
#> GSM39134 2 0.1022 0.8890 0.000 0.968 0.032 0.0
#> GSM39135 2 0.0000 0.8998 0.000 1.000 0.000 0.0
#> GSM39136 2 0.0000 0.8998 0.000 1.000 0.000 0.0
#> GSM39137 2 0.4281 0.7023 0.180 0.792 0.028 0.0
#> GSM39138 2 0.1022 0.8890 0.000 0.968 0.032 0.0
#> GSM39139 2 0.1022 0.8890 0.000 0.968 0.032 0.0
#> GSM39140 1 0.0817 0.6775 0.976 0.024 0.000 0.0
#> GSM39141 1 0.0921 0.6957 0.972 0.000 0.028 0.0
#> GSM39142 1 0.0921 0.6957 0.972 0.000 0.028 0.0
#> GSM39143 1 0.0921 0.6957 0.972 0.000 0.028 0.0
#> GSM39144 2 0.1022 0.8890 0.000 0.968 0.032 0.0
#> GSM39145 2 0.0188 0.8987 0.004 0.996 0.000 0.0
#> GSM39146 1 0.5977 -0.0891 0.528 0.432 0.040 0.0
#> GSM39147 2 0.0000 0.8998 0.000 1.000 0.000 0.0
#> GSM39188 3 0.1022 0.7455 0.032 0.000 0.968 0.0
#> GSM39189 3 0.1118 0.7443 0.036 0.000 0.964 0.0
#> GSM39190 3 0.4996 -0.3030 0.484 0.000 0.516 0.0
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM39104 1 0.4718 0.6957 0.540 0.000 0.016 0.444 0.000
#> GSM39105 1 0.4278 0.7001 0.548 0.000 0.000 0.452 0.000
#> GSM39106 1 0.3715 0.7062 0.736 0.000 0.004 0.260 0.000
#> GSM39107 1 0.3586 0.7203 0.736 0.000 0.000 0.264 0.000
#> GSM39108 1 0.4101 0.7285 0.628 0.000 0.000 0.372 0.000
#> GSM39109 1 0.4632 0.6957 0.540 0.000 0.012 0.448 0.000
#> GSM39110 1 0.0566 0.6975 0.984 0.000 0.004 0.012 0.000
#> GSM39111 1 0.4637 0.6918 0.536 0.000 0.012 0.452 0.000
#> GSM39112 1 0.3074 0.7056 0.804 0.000 0.000 0.196 0.000
#> GSM39113 1 0.4262 0.7045 0.560 0.000 0.000 0.440 0.000
#> GSM39114 2 0.2127 0.7134 0.000 0.892 0.000 0.000 0.108
#> GSM39115 1 0.4273 0.7016 0.552 0.000 0.000 0.448 0.000
#> GSM39148 1 0.0000 0.6926 1.000 0.000 0.000 0.000 0.000
#> GSM39149 1 0.4637 0.6918 0.536 0.000 0.012 0.452 0.000
#> GSM39150 3 0.5400 0.7424 0.096 0.000 0.632 0.272 0.000
#> GSM39151 3 0.4415 0.6445 0.004 0.000 0.552 0.444 0.000
#> GSM39152 3 0.4229 0.8336 0.020 0.000 0.704 0.276 0.000
#> GSM39153 1 0.0671 0.6986 0.980 0.000 0.004 0.016 0.000
#> GSM39154 1 0.3861 0.7312 0.712 0.000 0.004 0.284 0.000
#> GSM39155 1 0.4262 0.7061 0.560 0.000 0.000 0.440 0.000
#> GSM39156 1 0.1831 0.7200 0.920 0.000 0.004 0.076 0.000
#> GSM39157 1 0.2763 0.7119 0.848 0.000 0.004 0.148 0.000
#> GSM39158 3 0.3689 0.8467 0.004 0.000 0.740 0.256 0.000
#> GSM39159 1 0.5839 0.6112 0.560 0.000 0.116 0.324 0.000
#> GSM39160 3 0.5446 0.7406 0.100 0.000 0.628 0.272 0.000
#> GSM39161 3 0.4096 0.8281 0.004 0.020 0.744 0.232 0.000
#> GSM39162 1 0.0000 0.6926 1.000 0.000 0.000 0.000 0.000
#> GSM39163 1 0.4288 0.7282 0.612 0.000 0.004 0.384 0.000
#> GSM39164 1 0.3663 0.7175 0.776 0.000 0.016 0.208 0.000
#> GSM39165 1 0.4065 0.6974 0.720 0.000 0.016 0.264 0.000
#> GSM39166 3 0.3689 0.8467 0.004 0.000 0.740 0.256 0.000
#> GSM39167 1 0.3766 0.7067 0.728 0.000 0.004 0.268 0.000
#> GSM39168 1 0.0000 0.6926 1.000 0.000 0.000 0.000 0.000
#> GSM39169 1 0.0771 0.7014 0.976 0.000 0.004 0.020 0.000
#> GSM39170 1 0.0162 0.6920 0.996 0.000 0.004 0.000 0.000
#> GSM39171 1 0.4718 0.6957 0.540 0.000 0.016 0.444 0.000
#> GSM39172 1 0.6470 0.4313 0.536 0.008 0.192 0.264 0.000
#> GSM39173 2 0.3209 0.6576 0.180 0.812 0.000 0.000 0.008
#> GSM39174 1 0.0451 0.6972 0.988 0.000 0.004 0.008 0.000
#> GSM39175 1 0.4610 0.7110 0.596 0.000 0.016 0.388 0.000
#> GSM39176 1 0.3010 0.7251 0.824 0.000 0.004 0.172 0.000
#> GSM39177 1 0.4227 0.7471 0.692 0.000 0.016 0.292 0.000
#> GSM39178 3 0.3814 0.8454 0.004 0.000 0.720 0.276 0.000
#> GSM39179 1 0.5970 0.6064 0.524 0.000 0.120 0.356 0.000
#> GSM39180 3 0.6696 0.0377 0.184 0.360 0.448 0.008 0.000
#> GSM39181 3 0.3612 0.8453 0.000 0.000 0.732 0.268 0.000
#> GSM39182 1 0.2193 0.7030 0.920 0.008 0.028 0.044 0.000
#> GSM39183 3 0.3534 0.8453 0.000 0.000 0.744 0.256 0.000
#> GSM39184 1 0.4273 0.7026 0.552 0.000 0.000 0.448 0.000
#> GSM39185 3 0.4429 0.5583 0.000 0.192 0.744 0.064 0.000
#> GSM39186 4 0.6638 -0.4075 0.272 0.000 0.276 0.452 0.000
#> GSM39187 1 0.4171 0.7248 0.604 0.000 0.000 0.396 0.000
#> GSM39116 2 0.2824 0.7352 0.020 0.864 0.000 0.116 0.000
#> GSM39117 4 0.6200 0.4728 0.000 0.000 0.256 0.548 0.196
#> GSM39118 2 0.1638 0.7611 0.004 0.932 0.000 0.064 0.000
#> GSM39119 2 0.0404 0.7548 0.000 0.988 0.000 0.000 0.012
#> GSM39120 1 0.0162 0.6920 0.996 0.000 0.004 0.000 0.000
#> GSM39121 1 0.0609 0.7001 0.980 0.000 0.000 0.020 0.000
#> GSM39122 1 0.3010 0.7041 0.824 0.004 0.000 0.172 0.000
#> GSM39123 4 0.6466 0.4705 0.000 0.008 0.252 0.540 0.200
#> GSM39124 2 0.2127 0.7134 0.000 0.892 0.000 0.000 0.108
#> GSM39125 1 0.5348 0.6480 0.656 0.000 0.112 0.232 0.000
#> GSM39126 1 0.4225 0.2448 0.632 0.364 0.004 0.000 0.000
#> GSM39127 2 0.2011 0.7563 0.004 0.908 0.000 0.088 0.000
#> GSM39128 2 0.3759 0.6130 0.220 0.764 0.000 0.016 0.000
#> GSM39129 5 0.3561 1.0000 0.000 0.260 0.000 0.000 0.740
#> GSM39130 4 0.6494 0.4588 0.000 0.000 0.252 0.492 0.256
#> GSM39131 2 0.3846 0.7001 0.056 0.800 0.000 0.144 0.000
#> GSM39132 2 0.2127 0.7134 0.000 0.892 0.000 0.000 0.108
#> GSM39133 2 0.7317 0.3806 0.000 0.540 0.096 0.188 0.176
#> GSM39134 5 0.3561 1.0000 0.000 0.260 0.000 0.000 0.740
#> GSM39135 2 0.0000 0.7568 0.000 1.000 0.000 0.000 0.000
#> GSM39136 2 0.0000 0.7568 0.000 1.000 0.000 0.000 0.000
#> GSM39137 2 0.4600 0.6807 0.008 0.748 0.000 0.180 0.064
#> GSM39138 5 0.3561 1.0000 0.000 0.260 0.000 0.000 0.740
#> GSM39139 5 0.3561 1.0000 0.000 0.260 0.000 0.000 0.740
#> GSM39140 1 0.0000 0.6926 1.000 0.000 0.000 0.000 0.000
#> GSM39141 1 0.2852 0.7043 0.828 0.000 0.000 0.172 0.000
#> GSM39142 1 0.2929 0.7038 0.820 0.000 0.000 0.180 0.000
#> GSM39143 1 0.2929 0.7038 0.820 0.000 0.000 0.180 0.000
#> GSM39144 5 0.3561 1.0000 0.000 0.260 0.000 0.000 0.740
#> GSM39145 2 0.2127 0.7134 0.000 0.892 0.000 0.000 0.108
#> GSM39146 2 0.4444 0.6188 0.072 0.748 0.000 0.180 0.000
#> GSM39147 2 0.2127 0.7134 0.000 0.892 0.000 0.000 0.108
#> GSM39188 3 0.3586 0.8465 0.000 0.000 0.736 0.264 0.000
#> GSM39189 3 0.3661 0.8458 0.000 0.000 0.724 0.276 0.000
#> GSM39190 1 0.6102 0.5729 0.440 0.000 0.124 0.436 0.000
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM39104 5 0.4101 0.04241 0.408 0.000 0.012 0.000 0.580 0.000
#> GSM39105 5 0.4116 0.03284 0.416 0.000 0.012 0.000 0.572 0.000
#> GSM39106 1 0.3729 0.53057 0.692 0.000 0.012 0.000 0.296 0.000
#> GSM39107 1 0.3445 0.60726 0.744 0.000 0.012 0.000 0.244 0.000
#> GSM39108 1 0.3647 0.52074 0.640 0.000 0.000 0.000 0.360 0.000
#> GSM39109 5 0.3993 0.05457 0.400 0.000 0.008 0.000 0.592 0.000
#> GSM39110 1 0.0458 0.70748 0.984 0.000 0.000 0.000 0.016 0.000
#> GSM39111 5 0.4101 0.04630 0.408 0.000 0.012 0.000 0.580 0.000
#> GSM39112 1 0.3717 0.35529 0.616 0.000 0.000 0.000 0.384 0.000
#> GSM39113 5 0.4123 0.02478 0.420 0.000 0.012 0.000 0.568 0.000
#> GSM39114 2 0.6044 0.78973 0.000 0.396 0.000 0.000 0.348 0.256
#> GSM39115 5 0.4116 0.03284 0.416 0.000 0.012 0.000 0.572 0.000
#> GSM39148 1 0.0000 0.70490 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM39149 5 0.4084 0.05545 0.400 0.000 0.012 0.000 0.588 0.000
#> GSM39150 3 0.4093 0.64072 0.012 0.000 0.584 0.000 0.404 0.000
#> GSM39151 3 0.3864 0.53930 0.000 0.000 0.520 0.000 0.480 0.000
#> GSM39152 3 0.3695 0.67873 0.000 0.000 0.624 0.000 0.376 0.000
#> GSM39153 1 0.0603 0.70898 0.980 0.000 0.004 0.000 0.016 0.000
#> GSM39154 1 0.3534 0.62318 0.740 0.000 0.016 0.000 0.244 0.000
#> GSM39155 1 0.4161 0.32848 0.540 0.000 0.012 0.000 0.448 0.000
#> GSM39156 1 0.1588 0.70897 0.924 0.000 0.004 0.000 0.072 0.000
#> GSM39157 1 0.2442 0.66757 0.852 0.000 0.004 0.000 0.144 0.000
#> GSM39158 3 0.1501 0.74107 0.000 0.000 0.924 0.000 0.076 0.000
#> GSM39159 1 0.5787 0.34754 0.504 0.000 0.244 0.000 0.252 0.000
#> GSM39160 3 0.3867 0.70031 0.012 0.000 0.660 0.000 0.328 0.000
#> GSM39161 3 0.1501 0.74107 0.000 0.000 0.924 0.000 0.076 0.000
#> GSM39162 1 0.0000 0.70490 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM39163 1 0.3807 0.51531 0.628 0.000 0.004 0.000 0.368 0.000
#> GSM39164 1 0.2964 0.64041 0.792 0.000 0.004 0.000 0.204 0.000
#> GSM39165 1 0.3320 0.62893 0.772 0.000 0.016 0.000 0.212 0.000
#> GSM39166 3 0.1556 0.74149 0.000 0.000 0.920 0.000 0.080 0.000
#> GSM39167 1 0.3290 0.63399 0.776 0.000 0.016 0.000 0.208 0.000
#> GSM39168 1 0.0000 0.70490 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM39169 1 0.0632 0.70997 0.976 0.000 0.000 0.000 0.024 0.000
#> GSM39170 1 0.0000 0.70490 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM39171 5 0.4116 0.03107 0.416 0.000 0.012 0.000 0.572 0.000
#> GSM39172 1 0.5573 0.33089 0.524 0.000 0.312 0.000 0.164 0.000
#> GSM39173 2 0.6087 0.62942 0.176 0.412 0.000 0.000 0.400 0.012
#> GSM39174 1 0.0260 0.70799 0.992 0.000 0.000 0.000 0.008 0.000
#> GSM39175 1 0.4026 0.53224 0.636 0.000 0.016 0.000 0.348 0.000
#> GSM39176 1 0.2738 0.65870 0.820 0.000 0.004 0.000 0.176 0.000
#> GSM39177 1 0.3541 0.64083 0.728 0.000 0.012 0.000 0.260 0.000
#> GSM39178 3 0.3515 0.70852 0.000 0.000 0.676 0.000 0.324 0.000
#> GSM39179 1 0.5758 0.33536 0.492 0.000 0.196 0.000 0.312 0.000
#> GSM39180 3 0.5516 0.29859 0.172 0.104 0.660 0.000 0.064 0.000
#> GSM39181 3 0.1863 0.73804 0.000 0.000 0.896 0.000 0.104 0.000
#> GSM39182 1 0.2263 0.69966 0.896 0.000 0.048 0.000 0.056 0.000
#> GSM39183 3 0.1501 0.74107 0.000 0.000 0.924 0.000 0.076 0.000
#> GSM39184 5 0.4136 -0.00748 0.428 0.000 0.012 0.000 0.560 0.000
#> GSM39185 3 0.1588 0.63062 0.000 0.072 0.924 0.000 0.004 0.000
#> GSM39186 5 0.5318 -0.05874 0.148 0.000 0.272 0.000 0.580 0.000
#> GSM39187 1 0.4037 0.48987 0.608 0.000 0.012 0.000 0.380 0.000
#> GSM39116 5 0.5002 -0.69868 0.000 0.412 0.000 0.000 0.516 0.072
#> GSM39117 4 0.1444 0.70980 0.000 0.072 0.000 0.928 0.000 0.000
#> GSM39118 5 0.5442 -0.77050 0.000 0.412 0.000 0.000 0.468 0.120
#> GSM39119 2 0.5981 0.78752 0.000 0.404 0.004 0.000 0.400 0.192
#> GSM39120 1 0.0146 0.70453 0.996 0.000 0.000 0.000 0.004 0.000
#> GSM39121 1 0.0547 0.70769 0.980 0.000 0.000 0.000 0.020 0.000
#> GSM39122 1 0.2703 0.64541 0.824 0.004 0.000 0.000 0.172 0.000
#> GSM39123 4 0.1267 0.72908 0.000 0.000 0.060 0.940 0.000 0.000
#> GSM39124 2 0.6044 0.78973 0.000 0.396 0.000 0.000 0.348 0.256
#> GSM39125 1 0.5429 0.40939 0.576 0.000 0.236 0.000 0.188 0.000
#> GSM39126 1 0.4368 0.30228 0.656 0.296 0.000 0.000 0.048 0.000
#> GSM39127 5 0.5372 -0.73671 0.004 0.412 0.000 0.000 0.488 0.096
#> GSM39128 2 0.5891 0.62469 0.172 0.412 0.000 0.000 0.412 0.004
#> GSM39129 6 0.0000 1.00000 0.000 0.000 0.000 0.000 0.000 1.000
#> GSM39130 2 0.3782 -0.65620 0.000 0.588 0.000 0.412 0.000 0.000
#> GSM39131 5 0.4670 -0.63305 0.036 0.412 0.000 0.000 0.548 0.004
#> GSM39132 2 0.6044 0.78973 0.000 0.396 0.000 0.000 0.348 0.256
#> GSM39133 4 0.6136 0.52531 0.000 0.120 0.288 0.540 0.052 0.000
#> GSM39134 6 0.0000 1.00000 0.000 0.000 0.000 0.000 0.000 1.000
#> GSM39135 2 0.5838 0.78674 0.000 0.412 0.000 0.000 0.400 0.188
#> GSM39136 2 0.5838 0.78674 0.000 0.412 0.000 0.000 0.400 0.188
#> GSM39137 5 0.4531 -0.60706 0.000 0.408 0.000 0.000 0.556 0.036
#> GSM39138 6 0.0000 1.00000 0.000 0.000 0.000 0.000 0.000 1.000
#> GSM39139 6 0.0000 1.00000 0.000 0.000 0.000 0.000 0.000 1.000
#> GSM39140 1 0.0000 0.70490 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM39141 1 0.2562 0.64586 0.828 0.000 0.000 0.000 0.172 0.000
#> GSM39142 1 0.2664 0.64058 0.816 0.000 0.000 0.000 0.184 0.000
#> GSM39143 1 0.2664 0.64058 0.816 0.000 0.000 0.000 0.184 0.000
#> GSM39144 6 0.0000 1.00000 0.000 0.000 0.000 0.000 0.000 1.000
#> GSM39145 2 0.6044 0.78973 0.000 0.396 0.000 0.000 0.348 0.256
#> GSM39146 5 0.4093 -0.58749 0.012 0.404 0.000 0.000 0.584 0.000
#> GSM39147 2 0.6044 0.78973 0.000 0.396 0.000 0.000 0.348 0.256
#> GSM39188 3 0.2053 0.74632 0.000 0.004 0.888 0.000 0.108 0.000
#> GSM39189 3 0.3464 0.71236 0.000 0.000 0.688 0.000 0.312 0.000
#> GSM39190 5 0.5152 -0.06422 0.400 0.000 0.088 0.000 0.512 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) other(p) protocol(p) k
#> ATC:pam 82 0.2992 8.44e-07 5.96e-07 2
#> ATC:pam 80 0.3809 2.38e-05 7.41e-06 3
#> ATC:pam 77 0.0551 1.07e-05 2.05e-06 4
#> ATC:pam 79 0.0655 7.50e-06 2.53e-06 5
#> ATC:pam 60 0.5450 3.78e-03 4.26e-04 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 8353 rows and 87 columns.
#> Top rows (835, 1670, 2506, 3341, 4176) 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 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 0.975 0.965 0.983 0.46871 0.524 0.524
#> 3 3 0.603 0.748 0.832 0.29991 0.732 0.536
#> 4 4 0.827 0.875 0.927 0.16281 0.875 0.678
#> 5 5 0.829 0.848 0.912 -0.00469 0.915 0.748
#> 6 6 0.722 0.728 0.793 0.08116 0.904 0.679
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
#> GSM39104 1 0.0000 0.998 1.000 0.000
#> GSM39105 1 0.0000 0.998 1.000 0.000
#> GSM39106 1 0.0000 0.998 1.000 0.000
#> GSM39107 2 0.9358 0.505 0.352 0.648
#> GSM39108 1 0.0000 0.998 1.000 0.000
#> GSM39109 1 0.0672 0.990 0.992 0.008
#> GSM39110 1 0.4562 0.888 0.904 0.096
#> GSM39111 1 0.0000 0.998 1.000 0.000
#> GSM39112 1 0.0000 0.998 1.000 0.000
#> GSM39113 1 0.0000 0.998 1.000 0.000
#> GSM39114 2 0.0000 0.958 0.000 1.000
#> GSM39115 1 0.0000 0.998 1.000 0.000
#> GSM39148 1 0.0000 0.998 1.000 0.000
#> GSM39149 1 0.0000 0.998 1.000 0.000
#> GSM39150 1 0.0000 0.998 1.000 0.000
#> GSM39151 1 0.0000 0.998 1.000 0.000
#> GSM39152 1 0.0000 0.998 1.000 0.000
#> GSM39153 1 0.0000 0.998 1.000 0.000
#> GSM39154 1 0.0000 0.998 1.000 0.000
#> GSM39155 1 0.0000 0.998 1.000 0.000
#> GSM39156 1 0.0000 0.998 1.000 0.000
#> GSM39157 1 0.0000 0.998 1.000 0.000
#> GSM39158 1 0.0000 0.998 1.000 0.000
#> GSM39159 1 0.0000 0.998 1.000 0.000
#> GSM39160 1 0.0000 0.998 1.000 0.000
#> GSM39161 1 0.0000 0.998 1.000 0.000
#> GSM39162 1 0.0000 0.998 1.000 0.000
#> GSM39163 1 0.0000 0.998 1.000 0.000
#> GSM39164 1 0.0000 0.998 1.000 0.000
#> GSM39165 1 0.0000 0.998 1.000 0.000
#> GSM39166 1 0.0000 0.998 1.000 0.000
#> GSM39167 1 0.0000 0.998 1.000 0.000
#> GSM39168 1 0.0000 0.998 1.000 0.000
#> GSM39169 1 0.0000 0.998 1.000 0.000
#> GSM39170 1 0.0000 0.998 1.000 0.000
#> GSM39171 1 0.0000 0.998 1.000 0.000
#> GSM39172 1 0.0000 0.998 1.000 0.000
#> GSM39173 2 0.0000 0.958 0.000 1.000
#> GSM39174 1 0.0000 0.998 1.000 0.000
#> GSM39175 1 0.0000 0.998 1.000 0.000
#> GSM39176 1 0.0000 0.998 1.000 0.000
#> GSM39177 1 0.0000 0.998 1.000 0.000
#> GSM39178 1 0.0000 0.998 1.000 0.000
#> GSM39179 1 0.0000 0.998 1.000 0.000
#> GSM39180 2 0.2778 0.927 0.048 0.952
#> GSM39181 1 0.0000 0.998 1.000 0.000
#> GSM39182 2 0.5946 0.837 0.144 0.856
#> GSM39183 1 0.0000 0.998 1.000 0.000
#> GSM39184 1 0.0000 0.998 1.000 0.000
#> GSM39185 2 0.9460 0.480 0.364 0.636
#> GSM39186 1 0.0000 0.998 1.000 0.000
#> GSM39187 1 0.0000 0.998 1.000 0.000
#> GSM39116 2 0.0000 0.958 0.000 1.000
#> GSM39117 2 0.0000 0.958 0.000 1.000
#> GSM39118 2 0.0000 0.958 0.000 1.000
#> GSM39119 2 0.0000 0.958 0.000 1.000
#> GSM39120 1 0.0000 0.998 1.000 0.000
#> GSM39121 2 0.5629 0.855 0.132 0.868
#> GSM39122 2 0.5408 0.862 0.124 0.876
#> GSM39123 2 0.0000 0.958 0.000 1.000
#> GSM39124 2 0.0000 0.958 0.000 1.000
#> GSM39125 1 0.0000 0.998 1.000 0.000
#> GSM39126 2 0.1843 0.941 0.028 0.972
#> GSM39127 2 0.0000 0.958 0.000 1.000
#> GSM39128 2 0.0000 0.958 0.000 1.000
#> GSM39129 2 0.0000 0.958 0.000 1.000
#> GSM39130 2 0.0000 0.958 0.000 1.000
#> GSM39131 2 0.0000 0.958 0.000 1.000
#> GSM39132 2 0.0000 0.958 0.000 1.000
#> GSM39133 2 0.0000 0.958 0.000 1.000
#> GSM39134 2 0.0000 0.958 0.000 1.000
#> GSM39135 2 0.0000 0.958 0.000 1.000
#> GSM39136 2 0.0000 0.958 0.000 1.000
#> GSM39137 2 0.0000 0.958 0.000 1.000
#> GSM39138 2 0.0000 0.958 0.000 1.000
#> GSM39139 2 0.0000 0.958 0.000 1.000
#> GSM39140 2 0.6247 0.829 0.156 0.844
#> GSM39141 1 0.0000 0.998 1.000 0.000
#> GSM39142 1 0.0000 0.998 1.000 0.000
#> GSM39143 1 0.0000 0.998 1.000 0.000
#> GSM39144 2 0.0000 0.958 0.000 1.000
#> GSM39145 2 0.0000 0.958 0.000 1.000
#> GSM39146 2 0.0000 0.958 0.000 1.000
#> GSM39147 2 0.0000 0.958 0.000 1.000
#> GSM39188 1 0.0000 0.998 1.000 0.000
#> GSM39189 1 0.0000 0.998 1.000 0.000
#> GSM39190 1 0.0000 0.998 1.000 0.000
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM39104 1 0.2625 0.7605 0.916 0.000 0.084
#> GSM39105 1 0.0747 0.8058 0.984 0.000 0.016
#> GSM39106 1 0.1529 0.7970 0.960 0.000 0.040
#> GSM39107 1 0.7756 -0.0450 0.564 0.056 0.380
#> GSM39108 1 0.1163 0.8025 0.972 0.000 0.028
#> GSM39109 1 0.7325 -0.0308 0.576 0.036 0.388
#> GSM39110 1 0.3456 0.7503 0.904 0.036 0.060
#> GSM39111 1 0.1643 0.8010 0.956 0.000 0.044
#> GSM39112 1 0.3192 0.7268 0.888 0.000 0.112
#> GSM39113 1 0.7245 0.0201 0.596 0.036 0.368
#> GSM39114 2 0.0747 0.9197 0.000 0.984 0.016
#> GSM39115 1 0.6879 0.0558 0.616 0.024 0.360
#> GSM39148 1 0.1031 0.8053 0.976 0.000 0.024
#> GSM39149 3 0.6291 0.6572 0.468 0.000 0.532
#> GSM39150 3 0.6079 0.8292 0.388 0.000 0.612
#> GSM39151 3 0.5621 0.8987 0.308 0.000 0.692
#> GSM39152 3 0.6062 0.8344 0.384 0.000 0.616
#> GSM39153 1 0.0592 0.8065 0.988 0.000 0.012
#> GSM39154 1 0.1163 0.7959 0.972 0.000 0.028
#> GSM39155 1 0.1031 0.7989 0.976 0.000 0.024
#> GSM39156 1 0.0592 0.8065 0.988 0.000 0.012
#> GSM39157 1 0.0592 0.8065 0.988 0.000 0.012
#> GSM39158 1 0.5733 0.2544 0.676 0.000 0.324
#> GSM39159 3 0.5905 0.8653 0.352 0.000 0.648
#> GSM39160 3 0.6204 0.7646 0.424 0.000 0.576
#> GSM39161 3 0.5591 0.8967 0.304 0.000 0.696
#> GSM39162 1 0.1289 0.7998 0.968 0.000 0.032
#> GSM39163 1 0.1031 0.7989 0.976 0.000 0.024
#> GSM39164 1 0.1163 0.8058 0.972 0.000 0.028
#> GSM39165 1 0.4974 0.4793 0.764 0.000 0.236
#> GSM39166 3 0.5706 0.8922 0.320 0.000 0.680
#> GSM39167 1 0.0592 0.8067 0.988 0.000 0.012
#> GSM39168 1 0.1289 0.7998 0.968 0.000 0.032
#> GSM39169 1 0.0892 0.8064 0.980 0.000 0.020
#> GSM39170 1 0.3116 0.7248 0.892 0.000 0.108
#> GSM39171 1 0.1753 0.8010 0.952 0.000 0.048
#> GSM39172 3 0.5591 0.8967 0.304 0.000 0.696
#> GSM39173 2 0.2796 0.9173 0.000 0.908 0.092
#> GSM39174 1 0.0592 0.8065 0.988 0.000 0.012
#> GSM39175 1 0.1643 0.7968 0.956 0.000 0.044
#> GSM39176 1 0.1031 0.7990 0.976 0.000 0.024
#> GSM39177 1 0.6215 -0.3336 0.572 0.000 0.428
#> GSM39178 3 0.5760 0.8880 0.328 0.000 0.672
#> GSM39179 3 0.5650 0.8979 0.312 0.000 0.688
#> GSM39180 3 0.7297 0.7110 0.188 0.108 0.704
#> GSM39181 3 0.5810 0.8828 0.336 0.000 0.664
#> GSM39182 3 0.9452 0.5796 0.232 0.268 0.500
#> GSM39183 3 0.5621 0.8973 0.308 0.000 0.692
#> GSM39184 1 0.1163 0.8044 0.972 0.000 0.028
#> GSM39185 3 0.6927 0.8212 0.240 0.060 0.700
#> GSM39186 1 0.1289 0.8063 0.968 0.000 0.032
#> GSM39187 1 0.0592 0.8047 0.988 0.000 0.012
#> GSM39116 2 0.1163 0.9203 0.000 0.972 0.028
#> GSM39117 2 0.5327 0.8232 0.000 0.728 0.272
#> GSM39118 2 0.2066 0.9201 0.000 0.940 0.060
#> GSM39119 2 0.3551 0.8980 0.000 0.868 0.132
#> GSM39120 1 0.1860 0.7958 0.948 0.000 0.052
#> GSM39121 1 0.7773 0.3821 0.612 0.316 0.072
#> GSM39122 1 0.7944 0.3309 0.580 0.348 0.072
#> GSM39123 2 0.5327 0.8232 0.000 0.728 0.272
#> GSM39124 2 0.0747 0.9197 0.000 0.984 0.016
#> GSM39125 1 0.5988 0.0614 0.632 0.000 0.368
#> GSM39126 2 0.7558 0.5074 0.284 0.644 0.072
#> GSM39127 2 0.0892 0.9201 0.000 0.980 0.020
#> GSM39128 2 0.1529 0.9100 0.000 0.960 0.040
#> GSM39129 2 0.3412 0.8996 0.000 0.876 0.124
#> GSM39130 2 0.5327 0.8232 0.000 0.728 0.272
#> GSM39131 2 0.1289 0.9140 0.000 0.968 0.032
#> GSM39132 2 0.0747 0.9197 0.000 0.984 0.016
#> GSM39133 2 0.5327 0.8232 0.000 0.728 0.272
#> GSM39134 2 0.2448 0.9139 0.000 0.924 0.076
#> GSM39135 2 0.1163 0.9203 0.000 0.972 0.028
#> GSM39136 2 0.2711 0.9122 0.000 0.912 0.088
#> GSM39137 2 0.1163 0.9156 0.000 0.972 0.028
#> GSM39138 2 0.2448 0.9139 0.000 0.924 0.076
#> GSM39139 2 0.2448 0.9139 0.000 0.924 0.076
#> GSM39140 1 0.7545 0.4308 0.652 0.272 0.076
#> GSM39141 1 0.0892 0.8061 0.980 0.000 0.020
#> GSM39142 1 0.2878 0.7426 0.904 0.000 0.096
#> GSM39143 1 0.1031 0.8051 0.976 0.000 0.024
#> GSM39144 2 0.2448 0.9139 0.000 0.924 0.076
#> GSM39145 2 0.0892 0.9203 0.000 0.980 0.020
#> GSM39146 2 0.2261 0.9015 0.000 0.932 0.068
#> GSM39147 2 0.0747 0.9197 0.000 0.984 0.016
#> GSM39188 3 0.5621 0.8987 0.308 0.000 0.692
#> GSM39189 3 0.5621 0.8987 0.308 0.000 0.692
#> GSM39190 3 0.5621 0.8987 0.308 0.000 0.692
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM39104 1 0.1867 0.9201 0.928 0.000 0.072 0.000
#> GSM39105 1 0.0188 0.9298 0.996 0.000 0.004 0.000
#> GSM39106 1 0.0817 0.9334 0.976 0.000 0.024 0.000
#> GSM39107 1 0.7231 0.0654 0.464 0.144 0.392 0.000
#> GSM39108 1 0.0895 0.9339 0.976 0.004 0.020 0.000
#> GSM39109 3 0.6600 0.1816 0.396 0.084 0.520 0.000
#> GSM39110 1 0.4332 0.8169 0.816 0.112 0.072 0.000
#> GSM39111 1 0.1211 0.9322 0.960 0.000 0.040 0.000
#> GSM39112 1 0.2500 0.9197 0.916 0.044 0.040 0.000
#> GSM39113 1 0.5109 0.7424 0.744 0.060 0.196 0.000
#> GSM39114 2 0.0188 0.9619 0.000 0.996 0.000 0.004
#> GSM39115 1 0.3899 0.8553 0.840 0.052 0.108 0.000
#> GSM39148 1 0.2131 0.9269 0.932 0.036 0.032 0.000
#> GSM39149 3 0.2216 0.8722 0.092 0.000 0.908 0.000
#> GSM39150 3 0.1474 0.8969 0.052 0.000 0.948 0.000
#> GSM39151 3 0.0188 0.9062 0.004 0.000 0.996 0.000
#> GSM39152 3 0.2011 0.8807 0.080 0.000 0.920 0.000
#> GSM39153 1 0.0592 0.9327 0.984 0.000 0.016 0.000
#> GSM39154 1 0.0000 0.9282 1.000 0.000 0.000 0.000
#> GSM39155 1 0.0000 0.9282 1.000 0.000 0.000 0.000
#> GSM39156 1 0.0469 0.9322 0.988 0.000 0.012 0.000
#> GSM39157 1 0.1452 0.9321 0.956 0.008 0.036 0.000
#> GSM39158 1 0.4500 0.5889 0.684 0.000 0.316 0.000
#> GSM39159 3 0.2647 0.8412 0.120 0.000 0.880 0.000
#> GSM39160 3 0.2081 0.8780 0.084 0.000 0.916 0.000
#> GSM39161 3 0.0188 0.9039 0.004 0.000 0.996 0.000
#> GSM39162 1 0.2131 0.9277 0.932 0.032 0.036 0.000
#> GSM39163 1 0.0336 0.9311 0.992 0.000 0.008 0.000
#> GSM39164 1 0.0000 0.9282 1.000 0.000 0.000 0.000
#> GSM39165 1 0.4164 0.6594 0.736 0.000 0.264 0.000
#> GSM39166 3 0.0592 0.9092 0.016 0.000 0.984 0.000
#> GSM39167 1 0.0336 0.9311 0.992 0.000 0.008 0.000
#> GSM39168 1 0.2131 0.9277 0.932 0.032 0.036 0.000
#> GSM39169 1 0.1624 0.9313 0.952 0.020 0.028 0.000
#> GSM39170 1 0.1398 0.9323 0.956 0.004 0.040 0.000
#> GSM39171 1 0.0817 0.9332 0.976 0.000 0.024 0.000
#> GSM39172 3 0.0469 0.9084 0.012 0.000 0.988 0.000
#> GSM39173 4 0.3870 0.7998 0.000 0.208 0.004 0.788
#> GSM39174 1 0.1305 0.9317 0.960 0.004 0.036 0.000
#> GSM39175 1 0.0817 0.9328 0.976 0.000 0.024 0.000
#> GSM39176 1 0.0592 0.9331 0.984 0.000 0.016 0.000
#> GSM39177 3 0.3873 0.7199 0.228 0.000 0.772 0.000
#> GSM39178 3 0.0817 0.9087 0.024 0.000 0.976 0.000
#> GSM39179 3 0.0469 0.9088 0.012 0.000 0.988 0.000
#> GSM39180 4 0.6421 0.2443 0.004 0.056 0.432 0.508
#> GSM39181 3 0.1118 0.9013 0.036 0.000 0.964 0.000
#> GSM39182 3 0.4019 0.7956 0.008 0.076 0.848 0.068
#> GSM39183 3 0.0469 0.9084 0.012 0.000 0.988 0.000
#> GSM39184 1 0.0592 0.9331 0.984 0.000 0.016 0.000
#> GSM39185 3 0.1892 0.8665 0.004 0.036 0.944 0.016
#> GSM39186 1 0.0707 0.9331 0.980 0.000 0.020 0.000
#> GSM39187 1 0.0000 0.9282 1.000 0.000 0.000 0.000
#> GSM39116 2 0.0657 0.9625 0.000 0.984 0.004 0.012
#> GSM39117 4 0.0000 0.8719 0.000 0.000 0.000 1.000
#> GSM39118 2 0.1389 0.9408 0.000 0.952 0.000 0.048
#> GSM39119 4 0.2530 0.8948 0.000 0.112 0.000 0.888
#> GSM39120 1 0.1305 0.9324 0.960 0.004 0.036 0.000
#> GSM39121 2 0.1890 0.9053 0.056 0.936 0.008 0.000
#> GSM39122 2 0.1151 0.9397 0.024 0.968 0.008 0.000
#> GSM39123 4 0.0000 0.8719 0.000 0.000 0.000 1.000
#> GSM39124 2 0.0336 0.9628 0.000 0.992 0.000 0.008
#> GSM39125 1 0.3743 0.8370 0.824 0.016 0.160 0.000
#> GSM39126 2 0.0657 0.9516 0.012 0.984 0.004 0.000
#> GSM39127 2 0.0469 0.9631 0.000 0.988 0.000 0.012
#> GSM39128 2 0.0592 0.9627 0.000 0.984 0.000 0.016
#> GSM39129 4 0.2530 0.8948 0.000 0.112 0.000 0.888
#> GSM39130 4 0.0000 0.8719 0.000 0.000 0.000 1.000
#> GSM39131 2 0.0592 0.9627 0.000 0.984 0.000 0.016
#> GSM39132 2 0.0469 0.9631 0.000 0.988 0.000 0.012
#> GSM39133 4 0.0188 0.8733 0.000 0.004 0.000 0.996
#> GSM39134 4 0.2589 0.8946 0.000 0.116 0.000 0.884
#> GSM39135 2 0.0469 0.9631 0.000 0.988 0.000 0.012
#> GSM39136 2 0.3751 0.7476 0.000 0.800 0.004 0.196
#> GSM39137 2 0.0188 0.9619 0.000 0.996 0.000 0.004
#> GSM39138 4 0.2589 0.8946 0.000 0.116 0.000 0.884
#> GSM39139 4 0.2647 0.8932 0.000 0.120 0.000 0.880
#> GSM39140 2 0.2546 0.8535 0.092 0.900 0.008 0.000
#> GSM39141 1 0.2036 0.9278 0.936 0.032 0.032 0.000
#> GSM39142 1 0.1635 0.9311 0.948 0.008 0.044 0.000
#> GSM39143 1 0.2021 0.9295 0.936 0.024 0.040 0.000
#> GSM39144 4 0.2589 0.8946 0.000 0.116 0.000 0.884
#> GSM39145 2 0.0817 0.9551 0.000 0.976 0.000 0.024
#> GSM39146 2 0.0779 0.9620 0.000 0.980 0.004 0.016
#> GSM39147 2 0.0336 0.9628 0.000 0.992 0.000 0.008
#> GSM39188 3 0.0188 0.9062 0.004 0.000 0.996 0.000
#> GSM39189 3 0.0469 0.9088 0.012 0.000 0.988 0.000
#> GSM39190 3 0.0188 0.9062 0.004 0.000 0.996 0.000
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM39104 1 0.1579 0.9191 0.944 0.000 0.024 0.032 0.000
#> GSM39105 1 0.0963 0.9211 0.964 0.000 0.000 0.036 0.000
#> GSM39106 1 0.1281 0.9277 0.956 0.000 0.012 0.032 0.000
#> GSM39107 1 0.3067 0.9067 0.876 0.016 0.040 0.068 0.000
#> GSM39108 1 0.1704 0.9217 0.928 0.000 0.004 0.068 0.000
#> GSM39109 1 0.2696 0.9149 0.896 0.012 0.040 0.052 0.000
#> GSM39110 1 0.3274 0.8989 0.868 0.048 0.024 0.060 0.000
#> GSM39111 1 0.1195 0.9232 0.960 0.000 0.028 0.012 0.000
#> GSM39112 1 0.2463 0.9106 0.888 0.008 0.004 0.100 0.000
#> GSM39113 1 0.2741 0.9136 0.892 0.012 0.032 0.064 0.000
#> GSM39114 2 0.0162 0.8914 0.000 0.996 0.000 0.000 0.004
#> GSM39115 1 0.2199 0.9187 0.916 0.016 0.008 0.060 0.000
#> GSM39148 1 0.2339 0.9119 0.892 0.004 0.004 0.100 0.000
#> GSM39149 3 0.0955 0.8573 0.028 0.000 0.968 0.004 0.000
#> GSM39150 3 0.0880 0.8575 0.032 0.000 0.968 0.000 0.000
#> GSM39151 3 0.1300 0.8534 0.016 0.000 0.956 0.028 0.000
#> GSM39152 3 0.0880 0.8575 0.032 0.000 0.968 0.000 0.000
#> GSM39153 1 0.0963 0.9211 0.964 0.000 0.000 0.036 0.000
#> GSM39154 1 0.0963 0.9211 0.964 0.000 0.000 0.036 0.000
#> GSM39155 1 0.0963 0.9211 0.964 0.000 0.000 0.036 0.000
#> GSM39156 1 0.0963 0.9211 0.964 0.000 0.000 0.036 0.000
#> GSM39157 1 0.2228 0.9133 0.900 0.004 0.004 0.092 0.000
#> GSM39158 1 0.1970 0.9109 0.924 0.004 0.060 0.012 0.000
#> GSM39159 3 0.4552 0.1249 0.468 0.000 0.524 0.008 0.000
#> GSM39160 3 0.0880 0.8575 0.032 0.000 0.968 0.000 0.000
#> GSM39161 3 0.1300 0.8534 0.016 0.000 0.956 0.028 0.000
#> GSM39162 1 0.2228 0.9133 0.900 0.004 0.004 0.092 0.000
#> GSM39163 1 0.0963 0.9211 0.964 0.000 0.000 0.036 0.000
#> GSM39164 1 0.0880 0.9220 0.968 0.000 0.000 0.032 0.000
#> GSM39165 3 0.3932 0.5027 0.328 0.000 0.672 0.000 0.000
#> GSM39166 3 0.1012 0.8564 0.020 0.000 0.968 0.012 0.000
#> GSM39167 1 0.0963 0.9211 0.964 0.000 0.000 0.036 0.000
#> GSM39168 1 0.2228 0.9133 0.900 0.004 0.004 0.092 0.000
#> GSM39169 1 0.1928 0.9203 0.920 0.004 0.004 0.072 0.000
#> GSM39170 1 0.0609 0.9280 0.980 0.000 0.000 0.020 0.000
#> GSM39171 1 0.1117 0.9246 0.964 0.000 0.016 0.020 0.000
#> GSM39172 3 0.1386 0.8541 0.016 0.000 0.952 0.032 0.000
#> GSM39173 2 0.1544 0.8490 0.000 0.932 0.000 0.000 0.068
#> GSM39174 1 0.1704 0.9218 0.928 0.000 0.004 0.068 0.000
#> GSM39175 1 0.1041 0.9228 0.964 0.000 0.004 0.032 0.000
#> GSM39176 1 0.0703 0.9246 0.976 0.000 0.000 0.024 0.000
#> GSM39177 3 0.0963 0.8558 0.036 0.000 0.964 0.000 0.000
#> GSM39178 3 0.1012 0.8564 0.020 0.000 0.968 0.012 0.000
#> GSM39179 3 0.1117 0.8549 0.016 0.000 0.964 0.020 0.000
#> GSM39180 3 0.5496 0.5725 0.016 0.100 0.712 0.012 0.160
#> GSM39181 3 0.4641 0.1702 0.456 0.000 0.532 0.012 0.000
#> GSM39182 3 0.2734 0.7591 0.100 0.004 0.880 0.012 0.004
#> GSM39183 3 0.1012 0.8564 0.020 0.000 0.968 0.012 0.000
#> GSM39184 1 0.0290 0.9262 0.992 0.000 0.000 0.008 0.000
#> GSM39185 3 0.3113 0.7843 0.016 0.080 0.876 0.020 0.008
#> GSM39186 1 0.1041 0.9228 0.964 0.000 0.004 0.032 0.000
#> GSM39187 1 0.0963 0.9211 0.964 0.000 0.000 0.036 0.000
#> GSM39116 2 0.0566 0.8919 0.000 0.984 0.012 0.000 0.004
#> GSM39117 4 0.2813 0.9591 0.000 0.000 0.000 0.832 0.168
#> GSM39118 2 0.0807 0.8896 0.000 0.976 0.012 0.000 0.012
#> GSM39119 2 0.3550 0.6570 0.000 0.760 0.000 0.004 0.236
#> GSM39120 1 0.0566 0.9279 0.984 0.000 0.004 0.012 0.000
#> GSM39121 1 0.5604 0.0638 0.472 0.456 0.000 0.072 0.000
#> GSM39122 2 0.5459 0.2829 0.360 0.568 0.000 0.072 0.000
#> GSM39123 4 0.2813 0.9591 0.000 0.000 0.000 0.832 0.168
#> GSM39124 2 0.0162 0.8914 0.000 0.996 0.000 0.000 0.004
#> GSM39125 1 0.1461 0.9233 0.952 0.004 0.028 0.016 0.000
#> GSM39126 2 0.3752 0.4666 0.292 0.708 0.000 0.000 0.000
#> GSM39127 2 0.0566 0.8919 0.000 0.984 0.012 0.000 0.004
#> GSM39128 2 0.0566 0.8919 0.000 0.984 0.012 0.000 0.004
#> GSM39129 5 0.0162 0.9876 0.000 0.000 0.000 0.004 0.996
#> GSM39130 4 0.2813 0.9591 0.000 0.000 0.000 0.832 0.168
#> GSM39131 2 0.0000 0.8912 0.000 1.000 0.000 0.000 0.000
#> GSM39132 2 0.0865 0.8870 0.000 0.972 0.004 0.000 0.024
#> GSM39133 4 0.4199 0.8744 0.000 0.056 0.000 0.764 0.180
#> GSM39134 5 0.0000 0.9903 0.000 0.000 0.000 0.000 1.000
#> GSM39135 2 0.0566 0.8919 0.000 0.984 0.012 0.000 0.004
#> GSM39136 2 0.2561 0.7763 0.000 0.856 0.000 0.000 0.144
#> GSM39137 2 0.0000 0.8912 0.000 1.000 0.000 0.000 0.000
#> GSM39138 5 0.0000 0.9903 0.000 0.000 0.000 0.000 1.000
#> GSM39139 5 0.0609 0.9661 0.000 0.020 0.000 0.000 0.980
#> GSM39140 1 0.4645 0.7204 0.724 0.204 0.000 0.072 0.000
#> GSM39141 1 0.2339 0.9119 0.892 0.004 0.004 0.100 0.000
#> GSM39142 1 0.2568 0.9093 0.888 0.016 0.004 0.092 0.000
#> GSM39143 1 0.2339 0.9119 0.892 0.004 0.004 0.100 0.000
#> GSM39144 5 0.0000 0.9903 0.000 0.000 0.000 0.000 1.000
#> GSM39145 2 0.0290 0.8903 0.000 0.992 0.000 0.000 0.008
#> GSM39146 2 0.0566 0.8919 0.000 0.984 0.012 0.000 0.004
#> GSM39147 2 0.0162 0.8914 0.000 0.996 0.000 0.000 0.004
#> GSM39188 3 0.1300 0.8534 0.016 0.000 0.956 0.028 0.000
#> GSM39189 3 0.1300 0.8534 0.016 0.000 0.956 0.028 0.000
#> GSM39190 3 0.1300 0.8534 0.016 0.000 0.956 0.028 0.000
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM39104 5 0.4718 0.7590 0.316 0.000 0.068 0.000 0.616 0.000
#> GSM39105 5 0.3782 0.8786 0.412 0.000 0.000 0.000 0.588 0.000
#> GSM39106 1 0.3351 0.2944 0.712 0.000 0.000 0.000 0.288 0.000
#> GSM39107 1 0.2513 0.6531 0.852 0.000 0.008 0.000 0.140 0.000
#> GSM39108 1 0.2219 0.6540 0.864 0.000 0.000 0.000 0.136 0.000
#> GSM39109 1 0.2841 0.6377 0.824 0.000 0.012 0.000 0.164 0.000
#> GSM39110 1 0.3168 0.6425 0.828 0.056 0.000 0.000 0.116 0.000
#> GSM39111 5 0.3851 0.7955 0.460 0.000 0.000 0.000 0.540 0.000
#> GSM39112 1 0.0291 0.6710 0.992 0.004 0.004 0.000 0.000 0.000
#> GSM39113 1 0.2513 0.6531 0.852 0.000 0.008 0.000 0.140 0.000
#> GSM39114 2 0.2219 0.8604 0.000 0.864 0.000 0.000 0.136 0.000
#> GSM39115 1 0.2300 0.6468 0.856 0.000 0.000 0.000 0.144 0.000
#> GSM39148 1 0.0000 0.6773 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM39149 3 0.1398 0.8451 0.008 0.000 0.940 0.000 0.052 0.000
#> GSM39150 3 0.1890 0.8387 0.024 0.000 0.916 0.000 0.060 0.000
#> GSM39151 3 0.1956 0.8239 0.000 0.000 0.908 0.004 0.080 0.008
#> GSM39152 3 0.2099 0.8264 0.008 0.000 0.904 0.004 0.080 0.004
#> GSM39153 5 0.3804 0.8716 0.424 0.000 0.000 0.000 0.576 0.000
#> GSM39154 5 0.3804 0.8718 0.424 0.000 0.000 0.000 0.576 0.000
#> GSM39155 5 0.3765 0.8789 0.404 0.000 0.000 0.000 0.596 0.000
#> GSM39156 5 0.3747 0.8776 0.396 0.000 0.000 0.000 0.604 0.000
#> GSM39157 1 0.0000 0.6773 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM39158 5 0.4873 0.7572 0.420 0.000 0.060 0.000 0.520 0.000
#> GSM39159 3 0.5303 0.4073 0.196 0.000 0.600 0.000 0.204 0.000
#> GSM39160 3 0.1196 0.8445 0.008 0.000 0.952 0.000 0.040 0.000
#> GSM39161 3 0.0862 0.8402 0.000 0.000 0.972 0.004 0.016 0.008
#> GSM39162 1 0.0000 0.6773 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM39163 5 0.3727 0.8735 0.388 0.000 0.000 0.000 0.612 0.000
#> GSM39164 5 0.3838 0.8393 0.448 0.000 0.000 0.000 0.552 0.000
#> GSM39165 3 0.4583 0.6182 0.128 0.000 0.696 0.000 0.176 0.000
#> GSM39166 3 0.1625 0.8395 0.012 0.000 0.928 0.000 0.060 0.000
#> GSM39167 5 0.3727 0.8735 0.388 0.000 0.000 0.000 0.612 0.000
#> GSM39168 1 0.0000 0.6773 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM39169 1 0.2300 0.6466 0.856 0.000 0.000 0.000 0.144 0.000
#> GSM39170 1 0.3023 0.4794 0.768 0.000 0.000 0.000 0.232 0.000
#> GSM39171 5 0.4587 0.7878 0.356 0.000 0.048 0.000 0.596 0.000
#> GSM39172 3 0.0810 0.8433 0.004 0.000 0.976 0.004 0.008 0.008
#> GSM39173 2 0.2553 0.8515 0.000 0.848 0.000 0.000 0.144 0.008
#> GSM39174 1 0.1863 0.6683 0.896 0.000 0.000 0.000 0.104 0.000
#> GSM39175 5 0.4609 0.8124 0.364 0.000 0.048 0.000 0.588 0.000
#> GSM39176 1 0.3833 -0.4112 0.556 0.000 0.000 0.000 0.444 0.000
#> GSM39177 3 0.2106 0.8336 0.032 0.000 0.904 0.000 0.064 0.000
#> GSM39178 3 0.1524 0.8394 0.008 0.000 0.932 0.000 0.060 0.000
#> GSM39179 3 0.2211 0.8249 0.008 0.000 0.900 0.004 0.080 0.008
#> GSM39180 3 0.4586 0.6625 0.012 0.148 0.732 0.000 0.104 0.004
#> GSM39181 3 0.4500 0.5529 0.248 0.000 0.676 0.000 0.076 0.000
#> GSM39182 3 0.4107 0.6939 0.168 0.016 0.760 0.000 0.056 0.000
#> GSM39183 3 0.1625 0.8395 0.012 0.000 0.928 0.000 0.060 0.000
#> GSM39184 1 0.3862 -0.6754 0.524 0.000 0.000 0.000 0.476 0.000
#> GSM39185 3 0.3100 0.7754 0.012 0.108 0.848 0.004 0.028 0.000
#> GSM39186 5 0.4037 0.8546 0.380 0.000 0.012 0.000 0.608 0.000
#> GSM39187 5 0.3727 0.8735 0.388 0.000 0.000 0.000 0.612 0.000
#> GSM39116 2 0.0000 0.8872 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM39117 4 0.0146 0.9415 0.000 0.000 0.000 0.996 0.000 0.004
#> GSM39118 2 0.0260 0.8850 0.000 0.992 0.000 0.000 0.008 0.000
#> GSM39119 2 0.3426 0.7967 0.000 0.764 0.000 0.004 0.220 0.012
#> GSM39120 1 0.3101 0.4445 0.756 0.000 0.000 0.000 0.244 0.000
#> GSM39121 1 0.3979 0.0855 0.540 0.456 0.000 0.000 0.004 0.000
#> GSM39122 2 0.3999 -0.0338 0.496 0.500 0.000 0.000 0.004 0.000
#> GSM39123 4 0.0146 0.9415 0.000 0.000 0.000 0.996 0.000 0.004
#> GSM39124 2 0.2219 0.8604 0.000 0.864 0.000 0.000 0.136 0.000
#> GSM39125 1 0.3774 0.0690 0.664 0.000 0.008 0.000 0.328 0.000
#> GSM39126 2 0.0692 0.8699 0.020 0.976 0.000 0.000 0.004 0.000
#> GSM39127 2 0.0000 0.8872 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM39128 2 0.0000 0.8872 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM39129 6 0.0665 0.9878 0.000 0.008 0.000 0.004 0.008 0.980
#> GSM39130 4 0.0146 0.9415 0.000 0.000 0.000 0.996 0.000 0.004
#> GSM39131 2 0.0000 0.8872 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM39132 2 0.2219 0.8604 0.000 0.864 0.000 0.000 0.136 0.000
#> GSM39133 4 0.3161 0.8188 0.000 0.076 0.000 0.840 0.080 0.004
#> GSM39134 6 0.0260 0.9938 0.000 0.008 0.000 0.000 0.000 0.992
#> GSM39135 2 0.0000 0.8872 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM39136 2 0.1753 0.8403 0.000 0.912 0.000 0.000 0.084 0.004
#> GSM39137 2 0.0000 0.8872 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM39138 6 0.0260 0.9938 0.000 0.008 0.000 0.000 0.000 0.992
#> GSM39139 6 0.0458 0.9857 0.000 0.016 0.000 0.000 0.000 0.984
#> GSM39140 1 0.3830 0.3301 0.620 0.376 0.000 0.000 0.004 0.000
#> GSM39141 1 0.0000 0.6773 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM39142 1 0.0146 0.6778 0.996 0.000 0.000 0.000 0.004 0.000
#> GSM39143 1 0.0000 0.6773 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM39144 6 0.0260 0.9938 0.000 0.008 0.000 0.000 0.000 0.992
#> GSM39145 2 0.2219 0.8604 0.000 0.864 0.000 0.000 0.136 0.000
#> GSM39146 2 0.0000 0.8872 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM39147 2 0.2219 0.8604 0.000 0.864 0.000 0.000 0.136 0.000
#> GSM39188 3 0.2009 0.8232 0.000 0.000 0.904 0.004 0.084 0.008
#> GSM39189 3 0.2211 0.8249 0.008 0.000 0.900 0.004 0.080 0.008
#> GSM39190 3 0.1956 0.8239 0.000 0.000 0.908 0.004 0.080 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) other(p) protocol(p) k
#> ATC:mclust 86 0.20580 3.97e-09 1.18e-08 2
#> ATC:mclust 76 0.01616 2.11e-09 8.51e-10 3
#> ATC:mclust 84 0.01389 2.13e-09 2.74e-09 4
#> ATC:mclust 82 0.01786 2.46e-07 3.55e-09 5
#> ATC:mclust 77 0.00356 7.56e-09 2.82e-10 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 8353 rows and 87 columns.
#> Top rows (835, 1670, 2506, 3341, 4176) 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 0.999 0.957 0.980 0.430 0.558 0.558
#> 3 3 0.854 0.877 0.950 0.143 0.906 0.842
#> 4 4 0.547 0.747 0.860 0.335 0.661 0.447
#> 5 5 0.587 0.706 0.807 0.156 0.769 0.428
#> 6 6 0.602 0.583 0.772 0.043 0.963 0.844
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
#> GSM39104 1 0.0000 0.996 1.000 0.000
#> GSM39105 1 0.0000 0.996 1.000 0.000
#> GSM39106 1 0.0000 0.996 1.000 0.000
#> GSM39107 1 0.0000 0.996 1.000 0.000
#> GSM39108 1 0.0000 0.996 1.000 0.000
#> GSM39109 1 0.0000 0.996 1.000 0.000
#> GSM39110 1 0.0000 0.996 1.000 0.000
#> GSM39111 1 0.0000 0.996 1.000 0.000
#> GSM39112 1 0.0000 0.996 1.000 0.000
#> GSM39113 1 0.0000 0.996 1.000 0.000
#> GSM39114 2 0.0000 0.943 0.000 1.000
#> GSM39115 1 0.0000 0.996 1.000 0.000
#> GSM39148 1 0.0000 0.996 1.000 0.000
#> GSM39149 1 0.0000 0.996 1.000 0.000
#> GSM39150 1 0.0000 0.996 1.000 0.000
#> GSM39151 1 0.4562 0.888 0.904 0.096
#> GSM39152 1 0.0000 0.996 1.000 0.000
#> GSM39153 1 0.0000 0.996 1.000 0.000
#> GSM39154 1 0.0000 0.996 1.000 0.000
#> GSM39155 1 0.0000 0.996 1.000 0.000
#> GSM39156 1 0.0000 0.996 1.000 0.000
#> GSM39157 1 0.0000 0.996 1.000 0.000
#> GSM39158 1 0.0000 0.996 1.000 0.000
#> GSM39159 1 0.0000 0.996 1.000 0.000
#> GSM39160 1 0.0000 0.996 1.000 0.000
#> GSM39161 1 0.1633 0.972 0.976 0.024
#> GSM39162 1 0.0000 0.996 1.000 0.000
#> GSM39163 1 0.0000 0.996 1.000 0.000
#> GSM39164 1 0.0000 0.996 1.000 0.000
#> GSM39165 1 0.0000 0.996 1.000 0.000
#> GSM39166 1 0.0000 0.996 1.000 0.000
#> GSM39167 1 0.0000 0.996 1.000 0.000
#> GSM39168 1 0.0000 0.996 1.000 0.000
#> GSM39169 1 0.0000 0.996 1.000 0.000
#> GSM39170 1 0.0000 0.996 1.000 0.000
#> GSM39171 1 0.0000 0.996 1.000 0.000
#> GSM39172 1 0.0000 0.996 1.000 0.000
#> GSM39173 2 0.0000 0.943 0.000 1.000
#> GSM39174 1 0.0000 0.996 1.000 0.000
#> GSM39175 1 0.0000 0.996 1.000 0.000
#> GSM39176 1 0.0000 0.996 1.000 0.000
#> GSM39177 1 0.0000 0.996 1.000 0.000
#> GSM39178 1 0.0000 0.996 1.000 0.000
#> GSM39179 1 0.0000 0.996 1.000 0.000
#> GSM39180 2 0.0000 0.943 0.000 1.000
#> GSM39181 1 0.0000 0.996 1.000 0.000
#> GSM39182 1 0.0000 0.996 1.000 0.000
#> GSM39183 1 0.0000 0.996 1.000 0.000
#> GSM39184 1 0.0000 0.996 1.000 0.000
#> GSM39185 2 0.7299 0.760 0.204 0.796
#> GSM39186 1 0.0000 0.996 1.000 0.000
#> GSM39187 1 0.0000 0.996 1.000 0.000
#> GSM39116 2 0.0000 0.943 0.000 1.000
#> GSM39117 2 0.0000 0.943 0.000 1.000
#> GSM39118 2 0.0000 0.943 0.000 1.000
#> GSM39119 2 0.0000 0.943 0.000 1.000
#> GSM39120 1 0.0000 0.996 1.000 0.000
#> GSM39121 1 0.0000 0.996 1.000 0.000
#> GSM39122 1 0.0000 0.996 1.000 0.000
#> GSM39123 2 0.0000 0.943 0.000 1.000
#> GSM39124 2 0.0000 0.943 0.000 1.000
#> GSM39125 1 0.0000 0.996 1.000 0.000
#> GSM39126 1 0.0672 0.988 0.992 0.008
#> GSM39127 2 0.0376 0.941 0.004 0.996
#> GSM39128 2 0.8713 0.633 0.292 0.708
#> GSM39129 2 0.0000 0.943 0.000 1.000
#> GSM39130 2 0.0000 0.943 0.000 1.000
#> GSM39131 2 0.2423 0.916 0.040 0.960
#> GSM39132 2 0.0000 0.943 0.000 1.000
#> GSM39133 2 0.0000 0.943 0.000 1.000
#> GSM39134 2 0.0000 0.943 0.000 1.000
#> GSM39135 2 0.0000 0.943 0.000 1.000
#> GSM39136 2 0.0000 0.943 0.000 1.000
#> GSM39137 2 0.7602 0.740 0.220 0.780
#> GSM39138 2 0.0000 0.943 0.000 1.000
#> GSM39139 2 0.0000 0.943 0.000 1.000
#> GSM39140 1 0.0000 0.996 1.000 0.000
#> GSM39141 1 0.0000 0.996 1.000 0.000
#> GSM39142 1 0.0000 0.996 1.000 0.000
#> GSM39143 1 0.0000 0.996 1.000 0.000
#> GSM39144 2 0.0000 0.943 0.000 1.000
#> GSM39145 2 0.0000 0.943 0.000 1.000
#> GSM39146 2 0.9000 0.589 0.316 0.684
#> GSM39147 2 0.0000 0.943 0.000 1.000
#> GSM39188 2 0.9866 0.314 0.432 0.568
#> GSM39189 1 0.0000 0.996 1.000 0.000
#> GSM39190 1 0.4161 0.903 0.916 0.084
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM39104 1 0.0000 0.9464 1.000 0.000 0.000
#> GSM39105 1 0.0000 0.9464 1.000 0.000 0.000
#> GSM39106 1 0.0000 0.9464 1.000 0.000 0.000
#> GSM39107 1 0.0000 0.9464 1.000 0.000 0.000
#> GSM39108 1 0.0000 0.9464 1.000 0.000 0.000
#> GSM39109 1 0.0000 0.9464 1.000 0.000 0.000
#> GSM39110 1 0.1411 0.9216 0.964 0.036 0.000
#> GSM39111 1 0.0000 0.9464 1.000 0.000 0.000
#> GSM39112 1 0.0892 0.9342 0.980 0.020 0.000
#> GSM39113 1 0.0000 0.9464 1.000 0.000 0.000
#> GSM39114 2 0.0237 0.9088 0.004 0.996 0.000
#> GSM39115 1 0.0000 0.9464 1.000 0.000 0.000
#> GSM39148 1 0.0747 0.9370 0.984 0.016 0.000
#> GSM39149 1 0.0000 0.9464 1.000 0.000 0.000
#> GSM39150 1 0.0000 0.9464 1.000 0.000 0.000
#> GSM39151 1 0.2711 0.8819 0.912 0.000 0.088
#> GSM39152 1 0.0424 0.9423 0.992 0.000 0.008
#> GSM39153 1 0.0000 0.9464 1.000 0.000 0.000
#> GSM39154 1 0.0000 0.9464 1.000 0.000 0.000
#> GSM39155 1 0.0000 0.9464 1.000 0.000 0.000
#> GSM39156 1 0.0000 0.9464 1.000 0.000 0.000
#> GSM39157 1 0.0000 0.9464 1.000 0.000 0.000
#> GSM39158 1 0.0000 0.9464 1.000 0.000 0.000
#> GSM39159 1 0.0000 0.9464 1.000 0.000 0.000
#> GSM39160 1 0.0000 0.9464 1.000 0.000 0.000
#> GSM39161 1 0.4555 0.7591 0.800 0.000 0.200
#> GSM39162 1 0.0237 0.9444 0.996 0.004 0.000
#> GSM39163 1 0.0000 0.9464 1.000 0.000 0.000
#> GSM39164 1 0.0000 0.9464 1.000 0.000 0.000
#> GSM39165 1 0.0000 0.9464 1.000 0.000 0.000
#> GSM39166 1 0.0424 0.9423 0.992 0.000 0.008
#> GSM39167 1 0.0000 0.9464 1.000 0.000 0.000
#> GSM39168 1 0.0237 0.9444 0.996 0.004 0.000
#> GSM39169 1 0.0747 0.9370 0.984 0.016 0.000
#> GSM39170 1 0.0000 0.9464 1.000 0.000 0.000
#> GSM39171 1 0.0000 0.9464 1.000 0.000 0.000
#> GSM39172 1 0.3879 0.8180 0.848 0.000 0.152
#> GSM39173 2 0.0237 0.9088 0.004 0.996 0.000
#> GSM39174 1 0.0000 0.9464 1.000 0.000 0.000
#> GSM39175 1 0.0000 0.9464 1.000 0.000 0.000
#> GSM39176 1 0.0000 0.9464 1.000 0.000 0.000
#> GSM39177 1 0.0000 0.9464 1.000 0.000 0.000
#> GSM39178 1 0.0237 0.9445 0.996 0.000 0.004
#> GSM39179 1 0.0747 0.9374 0.984 0.000 0.016
#> GSM39180 3 0.0829 0.9770 0.012 0.004 0.984
#> GSM39181 1 0.0000 0.9464 1.000 0.000 0.000
#> GSM39182 1 0.2625 0.8850 0.916 0.000 0.084
#> GSM39183 1 0.0424 0.9423 0.992 0.000 0.008
#> GSM39184 1 0.0000 0.9464 1.000 0.000 0.000
#> GSM39185 1 0.6045 0.4426 0.620 0.000 0.380
#> GSM39186 1 0.0000 0.9464 1.000 0.000 0.000
#> GSM39187 1 0.0000 0.9464 1.000 0.000 0.000
#> GSM39116 2 0.0237 0.9088 0.004 0.996 0.000
#> GSM39117 3 0.0237 0.9943 0.000 0.004 0.996
#> GSM39118 2 0.1964 0.8826 0.000 0.944 0.056
#> GSM39119 2 0.4702 0.7152 0.000 0.788 0.212
#> GSM39120 1 0.0000 0.9464 1.000 0.000 0.000
#> GSM39121 1 0.6244 0.2137 0.560 0.440 0.000
#> GSM39122 1 0.6307 0.0417 0.512 0.488 0.000
#> GSM39123 3 0.0237 0.9943 0.000 0.004 0.996
#> GSM39124 2 0.0237 0.9088 0.004 0.996 0.000
#> GSM39125 1 0.0000 0.9464 1.000 0.000 0.000
#> GSM39126 2 0.4605 0.6202 0.204 0.796 0.000
#> GSM39127 2 0.0237 0.9088 0.004 0.996 0.000
#> GSM39128 2 0.3116 0.7770 0.108 0.892 0.000
#> GSM39129 2 0.2165 0.8768 0.000 0.936 0.064
#> GSM39130 3 0.0237 0.9943 0.000 0.004 0.996
#> GSM39131 2 0.0237 0.9088 0.004 0.996 0.000
#> GSM39132 2 0.0000 0.9078 0.000 1.000 0.000
#> GSM39133 3 0.0237 0.9943 0.000 0.004 0.996
#> GSM39134 2 0.1964 0.8823 0.000 0.944 0.056
#> GSM39135 2 0.0000 0.9078 0.000 1.000 0.000
#> GSM39136 2 0.3551 0.8144 0.000 0.868 0.132
#> GSM39137 2 0.0237 0.9088 0.004 0.996 0.000
#> GSM39138 2 0.1643 0.8895 0.000 0.956 0.044
#> GSM39139 2 0.0000 0.9078 0.000 1.000 0.000
#> GSM39140 1 0.5016 0.6728 0.760 0.240 0.000
#> GSM39141 1 0.1289 0.9250 0.968 0.032 0.000
#> GSM39142 1 0.0000 0.9464 1.000 0.000 0.000
#> GSM39143 1 0.0747 0.9370 0.984 0.016 0.000
#> GSM39144 2 0.0592 0.9036 0.000 0.988 0.012
#> GSM39145 2 0.0237 0.9088 0.004 0.996 0.000
#> GSM39146 2 0.6168 0.2685 0.412 0.588 0.000
#> GSM39147 2 0.0000 0.9078 0.000 1.000 0.000
#> GSM39188 1 0.6295 0.1912 0.528 0.000 0.472
#> GSM39189 1 0.3116 0.8629 0.892 0.000 0.108
#> GSM39190 1 0.3752 0.8267 0.856 0.000 0.144
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM39104 3 0.2469 0.8479 0.108 0.000 0.892 0.000
#> GSM39105 1 0.3801 0.7765 0.780 0.000 0.220 0.000
#> GSM39106 1 0.3123 0.8207 0.844 0.000 0.156 0.000
#> GSM39107 1 0.1584 0.8326 0.952 0.000 0.036 0.012
#> GSM39108 1 0.2011 0.8341 0.920 0.000 0.080 0.000
#> GSM39109 1 0.2149 0.8316 0.912 0.000 0.088 0.000
#> GSM39110 1 0.3464 0.8296 0.860 0.032 0.108 0.000
#> GSM39111 1 0.4877 0.4687 0.592 0.000 0.408 0.000
#> GSM39112 1 0.1398 0.8244 0.956 0.000 0.040 0.004
#> GSM39113 1 0.1722 0.8269 0.944 0.000 0.048 0.008
#> GSM39114 2 0.4855 0.3636 0.400 0.600 0.000 0.000
#> GSM39115 1 0.2011 0.8347 0.920 0.000 0.080 0.000
#> GSM39148 1 0.1209 0.8346 0.964 0.004 0.032 0.000
#> GSM39149 3 0.1488 0.8772 0.032 0.012 0.956 0.000
#> GSM39150 3 0.1474 0.8800 0.052 0.000 0.948 0.000
#> GSM39151 3 0.1724 0.8561 0.020 0.032 0.948 0.000
#> GSM39152 3 0.1004 0.8832 0.024 0.004 0.972 0.000
#> GSM39153 1 0.4989 0.2271 0.528 0.000 0.472 0.000
#> GSM39154 1 0.3942 0.7418 0.764 0.000 0.236 0.000
#> GSM39155 1 0.3486 0.7995 0.812 0.000 0.188 0.000
#> GSM39156 1 0.4356 0.6965 0.708 0.000 0.292 0.000
#> GSM39157 1 0.1474 0.8365 0.948 0.000 0.052 0.000
#> GSM39158 3 0.4972 0.0307 0.456 0.000 0.544 0.000
#> GSM39159 3 0.2589 0.8525 0.116 0.000 0.884 0.000
#> GSM39160 3 0.1302 0.8818 0.044 0.000 0.956 0.000
#> GSM39161 3 0.2140 0.8707 0.052 0.008 0.932 0.008
#> GSM39162 1 0.1474 0.8388 0.948 0.000 0.052 0.000
#> GSM39163 1 0.4522 0.6245 0.680 0.000 0.320 0.000
#> GSM39164 1 0.4741 0.6576 0.668 0.004 0.328 0.000
#> GSM39165 3 0.1637 0.8805 0.060 0.000 0.940 0.000
#> GSM39166 3 0.1302 0.8818 0.044 0.000 0.956 0.000
#> GSM39167 1 0.4608 0.6491 0.692 0.004 0.304 0.000
#> GSM39168 1 0.1389 0.8396 0.952 0.000 0.048 0.000
#> GSM39169 1 0.2831 0.8340 0.876 0.004 0.120 0.000
#> GSM39170 1 0.1940 0.8364 0.924 0.000 0.076 0.000
#> GSM39171 3 0.2704 0.8297 0.124 0.000 0.876 0.000
#> GSM39172 3 0.1929 0.8791 0.036 0.000 0.940 0.024
#> GSM39173 2 0.1109 0.8014 0.004 0.968 0.028 0.000
#> GSM39174 1 0.2469 0.8333 0.892 0.000 0.108 0.000
#> GSM39175 3 0.2216 0.8682 0.092 0.000 0.908 0.000
#> GSM39176 1 0.3311 0.7979 0.828 0.000 0.172 0.000
#> GSM39177 3 0.1489 0.8826 0.044 0.004 0.952 0.000
#> GSM39178 3 0.1302 0.8818 0.044 0.000 0.956 0.000
#> GSM39179 3 0.1209 0.8837 0.032 0.004 0.964 0.000
#> GSM39180 3 0.6789 0.1213 0.020 0.052 0.504 0.424
#> GSM39181 3 0.3172 0.8116 0.160 0.000 0.840 0.000
#> GSM39182 3 0.6359 0.6318 0.132 0.000 0.648 0.220
#> GSM39183 3 0.1940 0.8758 0.076 0.000 0.924 0.000
#> GSM39184 1 0.3074 0.8217 0.848 0.000 0.152 0.000
#> GSM39185 3 0.4618 0.7663 0.052 0.004 0.796 0.148
#> GSM39186 1 0.5000 0.1825 0.500 0.000 0.500 0.000
#> GSM39187 1 0.4431 0.6372 0.696 0.000 0.304 0.000
#> GSM39116 1 0.3570 0.7527 0.860 0.092 0.000 0.048
#> GSM39117 4 0.0188 0.8217 0.000 0.000 0.004 0.996
#> GSM39118 2 0.3107 0.7806 0.036 0.884 0.000 0.080
#> GSM39119 2 0.2867 0.7541 0.000 0.884 0.012 0.104
#> GSM39120 1 0.1792 0.8373 0.932 0.000 0.068 0.000
#> GSM39121 1 0.2197 0.7962 0.916 0.080 0.004 0.000
#> GSM39122 1 0.2408 0.7845 0.896 0.104 0.000 0.000
#> GSM39123 4 0.1474 0.8252 0.052 0.000 0.000 0.948
#> GSM39124 2 0.4103 0.5709 0.256 0.744 0.000 0.000
#> GSM39125 1 0.2408 0.8291 0.896 0.000 0.104 0.000
#> GSM39126 1 0.3529 0.7600 0.836 0.152 0.012 0.000
#> GSM39127 1 0.3688 0.6921 0.792 0.208 0.000 0.000
#> GSM39128 1 0.3300 0.7591 0.848 0.144 0.008 0.000
#> GSM39129 2 0.2363 0.7764 0.000 0.920 0.024 0.056
#> GSM39130 4 0.0188 0.8241 0.004 0.000 0.000 0.996
#> GSM39131 1 0.4855 0.3240 0.600 0.400 0.000 0.000
#> GSM39132 2 0.1489 0.8039 0.044 0.952 0.000 0.004
#> GSM39133 4 0.1792 0.8155 0.068 0.000 0.000 0.932
#> GSM39134 2 0.1938 0.7922 0.000 0.936 0.012 0.052
#> GSM39135 2 0.5247 0.5138 0.284 0.684 0.000 0.032
#> GSM39136 4 0.6611 -0.0787 0.080 0.456 0.000 0.464
#> GSM39137 1 0.4008 0.6528 0.756 0.244 0.000 0.000
#> GSM39138 2 0.1706 0.7952 0.000 0.948 0.016 0.036
#> GSM39139 2 0.0524 0.8081 0.008 0.988 0.000 0.004
#> GSM39140 1 0.2542 0.8014 0.904 0.084 0.012 0.000
#> GSM39141 1 0.0469 0.8263 0.988 0.000 0.012 0.000
#> GSM39142 1 0.0592 0.8312 0.984 0.000 0.016 0.000
#> GSM39143 1 0.0188 0.8269 0.996 0.000 0.004 0.000
#> GSM39144 2 0.1247 0.8057 0.004 0.968 0.012 0.016
#> GSM39145 2 0.1118 0.8070 0.036 0.964 0.000 0.000
#> GSM39146 1 0.1978 0.7964 0.928 0.068 0.000 0.004
#> GSM39147 2 0.1743 0.7975 0.056 0.940 0.000 0.004
#> GSM39188 3 0.2400 0.8281 0.004 0.044 0.924 0.028
#> GSM39189 3 0.1443 0.8807 0.028 0.004 0.960 0.008
#> GSM39190 3 0.1486 0.8612 0.008 0.024 0.960 0.008
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM39104 5 0.4707 0.106 0.020 0.000 0.392 0.000 0.588
#> GSM39105 5 0.2209 0.784 0.032 0.000 0.056 0.000 0.912
#> GSM39106 5 0.2554 0.812 0.072 0.000 0.036 0.000 0.892
#> GSM39107 5 0.3048 0.780 0.176 0.000 0.004 0.000 0.820
#> GSM39108 5 0.1830 0.814 0.068 0.000 0.008 0.000 0.924
#> GSM39109 5 0.1568 0.802 0.036 0.000 0.020 0.000 0.944
#> GSM39110 5 0.2518 0.813 0.080 0.016 0.008 0.000 0.896
#> GSM39111 5 0.3391 0.623 0.012 0.000 0.188 0.000 0.800
#> GSM39112 5 0.1502 0.806 0.056 0.000 0.000 0.004 0.940
#> GSM39113 5 0.1798 0.810 0.064 0.000 0.004 0.004 0.928
#> GSM39114 5 0.4225 0.396 0.004 0.364 0.000 0.000 0.632
#> GSM39115 5 0.1965 0.808 0.052 0.000 0.024 0.000 0.924
#> GSM39148 1 0.2228 0.788 0.900 0.004 0.004 0.000 0.092
#> GSM39149 3 0.2732 0.761 0.000 0.000 0.840 0.000 0.160
#> GSM39150 3 0.4696 0.372 0.016 0.000 0.556 0.000 0.428
#> GSM39151 3 0.2069 0.773 0.000 0.012 0.912 0.000 0.076
#> GSM39152 3 0.2209 0.793 0.032 0.000 0.912 0.000 0.056
#> GSM39153 1 0.2927 0.779 0.868 0.000 0.092 0.000 0.040
#> GSM39154 1 0.2903 0.796 0.872 0.000 0.048 0.000 0.080
#> GSM39155 5 0.3805 0.758 0.184 0.000 0.032 0.000 0.784
#> GSM39156 1 0.3995 0.762 0.788 0.000 0.060 0.000 0.152
#> GSM39157 1 0.1892 0.793 0.916 0.000 0.004 0.000 0.080
#> GSM39158 1 0.5365 0.653 0.664 0.000 0.204 0.000 0.132
#> GSM39159 1 0.4485 0.553 0.680 0.000 0.292 0.000 0.028
#> GSM39160 3 0.3759 0.732 0.016 0.000 0.764 0.000 0.220
#> GSM39161 3 0.3970 0.669 0.220 0.004 0.760 0.004 0.012
#> GSM39162 1 0.1768 0.792 0.924 0.000 0.004 0.000 0.072
#> GSM39163 1 0.2770 0.786 0.880 0.000 0.076 0.000 0.044
#> GSM39164 1 0.5957 0.534 0.588 0.000 0.176 0.000 0.236
#> GSM39165 3 0.4946 0.355 0.368 0.000 0.596 0.000 0.036
#> GSM39166 3 0.3655 0.776 0.036 0.000 0.804 0.000 0.160
#> GSM39167 1 0.2989 0.789 0.868 0.000 0.072 0.000 0.060
#> GSM39168 1 0.4288 0.356 0.612 0.000 0.004 0.000 0.384
#> GSM39169 5 0.2731 0.809 0.104 0.004 0.016 0.000 0.876
#> GSM39170 1 0.0510 0.786 0.984 0.000 0.000 0.000 0.016
#> GSM39171 3 0.5106 0.277 0.036 0.000 0.508 0.000 0.456
#> GSM39172 3 0.2352 0.764 0.092 0.000 0.896 0.004 0.008
#> GSM39173 2 0.1124 0.890 0.000 0.960 0.036 0.000 0.004
#> GSM39174 1 0.1768 0.795 0.924 0.000 0.004 0.000 0.072
#> GSM39175 1 0.5095 0.276 0.560 0.000 0.400 0.000 0.040
#> GSM39176 1 0.1195 0.793 0.960 0.000 0.028 0.000 0.012
#> GSM39177 3 0.3673 0.744 0.140 0.012 0.820 0.000 0.028
#> GSM39178 3 0.3229 0.787 0.032 0.000 0.840 0.000 0.128
#> GSM39179 3 0.1444 0.779 0.040 0.000 0.948 0.000 0.012
#> GSM39180 4 0.7511 0.411 0.200 0.020 0.276 0.476 0.028
#> GSM39181 1 0.4503 0.569 0.696 0.000 0.268 0.000 0.036
#> GSM39182 1 0.6102 0.480 0.632 0.000 0.128 0.212 0.028
#> GSM39183 3 0.4355 0.672 0.224 0.000 0.732 0.000 0.044
#> GSM39184 5 0.3409 0.788 0.144 0.000 0.032 0.000 0.824
#> GSM39185 1 0.6380 0.390 0.572 0.004 0.296 0.104 0.024
#> GSM39186 5 0.3061 0.697 0.020 0.000 0.136 0.000 0.844
#> GSM39187 1 0.3442 0.790 0.836 0.000 0.060 0.000 0.104
#> GSM39116 5 0.7092 0.451 0.108 0.132 0.000 0.188 0.572
#> GSM39117 4 0.0324 0.831 0.000 0.004 0.000 0.992 0.004
#> GSM39118 2 0.2528 0.872 0.012 0.908 0.008 0.056 0.016
#> GSM39119 2 0.1997 0.877 0.000 0.924 0.036 0.040 0.000
#> GSM39120 1 0.0794 0.782 0.972 0.000 0.000 0.000 0.028
#> GSM39121 1 0.3714 0.740 0.812 0.056 0.000 0.000 0.132
#> GSM39122 5 0.5013 0.667 0.232 0.084 0.000 0.000 0.684
#> GSM39123 4 0.0566 0.832 0.004 0.000 0.000 0.984 0.012
#> GSM39124 2 0.2532 0.835 0.012 0.892 0.000 0.008 0.088
#> GSM39125 1 0.1568 0.796 0.944 0.000 0.020 0.000 0.036
#> GSM39126 1 0.2654 0.777 0.888 0.064 0.000 0.000 0.048
#> GSM39127 1 0.5596 0.335 0.552 0.376 0.000 0.004 0.068
#> GSM39128 1 0.3969 0.751 0.808 0.096 0.000 0.004 0.092
#> GSM39129 2 0.1914 0.869 0.000 0.924 0.060 0.016 0.000
#> GSM39130 4 0.0000 0.832 0.000 0.000 0.000 1.000 0.000
#> GSM39131 2 0.3695 0.700 0.164 0.800 0.000 0.000 0.036
#> GSM39132 2 0.1012 0.889 0.000 0.968 0.000 0.020 0.012
#> GSM39133 4 0.0898 0.829 0.008 0.000 0.000 0.972 0.020
#> GSM39134 2 0.1106 0.891 0.000 0.964 0.024 0.012 0.000
#> GSM39135 2 0.2937 0.844 0.016 0.884 0.000 0.040 0.060
#> GSM39136 4 0.5091 0.385 0.004 0.328 0.000 0.624 0.044
#> GSM39137 2 0.5714 0.452 0.212 0.624 0.000 0.000 0.164
#> GSM39138 2 0.1251 0.888 0.000 0.956 0.036 0.008 0.000
#> GSM39139 2 0.0566 0.893 0.000 0.984 0.012 0.000 0.004
#> GSM39140 1 0.2438 0.772 0.900 0.040 0.000 0.000 0.060
#> GSM39141 1 0.3093 0.730 0.824 0.008 0.000 0.000 0.168
#> GSM39142 5 0.4270 0.530 0.336 0.004 0.004 0.000 0.656
#> GSM39143 5 0.4147 0.595 0.316 0.008 0.000 0.000 0.676
#> GSM39144 2 0.0865 0.893 0.000 0.972 0.024 0.004 0.000
#> GSM39145 2 0.0290 0.892 0.000 0.992 0.000 0.000 0.008
#> GSM39146 1 0.5501 0.613 0.684 0.112 0.000 0.016 0.188
#> GSM39147 2 0.0671 0.891 0.000 0.980 0.000 0.004 0.016
#> GSM39188 3 0.1026 0.750 0.004 0.024 0.968 0.000 0.004
#> GSM39189 3 0.2144 0.783 0.020 0.000 0.912 0.000 0.068
#> GSM39190 3 0.1701 0.776 0.028 0.012 0.944 0.000 0.016
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM39104 6 0.5673 0.3608 0.052 0.000 0.280 0.000 0.076 0.592
#> GSM39105 6 0.2169 0.7071 0.012 0.000 0.080 0.000 0.008 0.900
#> GSM39106 6 0.3811 0.6975 0.116 0.000 0.040 0.000 0.040 0.804
#> GSM39107 6 0.5158 0.4612 0.144 0.000 0.004 0.000 0.220 0.632
#> GSM39108 6 0.1845 0.7216 0.052 0.000 0.028 0.000 0.000 0.920
#> GSM39109 6 0.1528 0.7053 0.012 0.000 0.028 0.000 0.016 0.944
#> GSM39110 6 0.3256 0.7099 0.104 0.008 0.028 0.000 0.016 0.844
#> GSM39111 6 0.2592 0.6915 0.016 0.000 0.116 0.000 0.004 0.864
#> GSM39112 6 0.1003 0.6878 0.004 0.004 0.000 0.000 0.028 0.964
#> GSM39113 6 0.1138 0.6989 0.012 0.000 0.004 0.000 0.024 0.960
#> GSM39114 6 0.4099 0.3141 0.000 0.372 0.000 0.000 0.016 0.612
#> GSM39115 6 0.0984 0.7053 0.012 0.000 0.008 0.000 0.012 0.968
#> GSM39148 1 0.2066 0.6500 0.908 0.000 0.000 0.000 0.052 0.040
#> GSM39149 3 0.3525 0.6462 0.016 0.004 0.816 0.000 0.032 0.132
#> GSM39150 3 0.6177 0.1190 0.040 0.000 0.428 0.000 0.116 0.416
#> GSM39151 3 0.2008 0.6551 0.004 0.004 0.920 0.000 0.032 0.040
#> GSM39152 3 0.2231 0.6700 0.068 0.000 0.900 0.000 0.004 0.028
#> GSM39153 1 0.2485 0.6640 0.884 0.000 0.084 0.000 0.008 0.024
#> GSM39154 1 0.3530 0.6424 0.824 0.000 0.104 0.000 0.044 0.028
#> GSM39155 6 0.5380 0.4847 0.240 0.000 0.124 0.000 0.016 0.620
#> GSM39156 1 0.3817 0.6415 0.800 0.000 0.120 0.000 0.024 0.056
#> GSM39157 1 0.1313 0.6658 0.952 0.000 0.004 0.000 0.016 0.028
#> GSM39158 1 0.6071 0.4456 0.600 0.004 0.172 0.000 0.172 0.052
#> GSM39159 1 0.4771 0.4743 0.652 0.000 0.248 0.000 0.100 0.000
#> GSM39160 3 0.4764 0.5633 0.028 0.000 0.664 0.000 0.040 0.268
#> GSM39161 3 0.5569 0.3130 0.320 0.000 0.520 0.000 0.160 0.000
#> GSM39162 1 0.2145 0.6469 0.900 0.000 0.000 0.000 0.072 0.028
#> GSM39163 1 0.2288 0.6640 0.900 0.000 0.068 0.000 0.016 0.016
#> GSM39164 1 0.4919 0.5277 0.676 0.000 0.216 0.000 0.016 0.092
#> GSM39165 1 0.4684 0.0832 0.520 0.000 0.444 0.000 0.028 0.008
#> GSM39166 3 0.5879 0.5909 0.096 0.000 0.632 0.000 0.112 0.160
#> GSM39167 1 0.3074 0.6515 0.856 0.000 0.080 0.000 0.044 0.020
#> GSM39168 1 0.4266 0.5060 0.712 0.000 0.012 0.000 0.040 0.236
#> GSM39169 6 0.2563 0.7160 0.076 0.008 0.016 0.000 0.012 0.888
#> GSM39170 1 0.3984 0.3965 0.648 0.000 0.000 0.000 0.336 0.016
#> GSM39171 3 0.5371 0.3288 0.088 0.000 0.512 0.000 0.008 0.392
#> GSM39172 3 0.4307 0.5232 0.072 0.000 0.704 0.000 0.224 0.000
#> GSM39173 2 0.1074 0.8929 0.000 0.960 0.028 0.000 0.012 0.000
#> GSM39174 1 0.1970 0.6585 0.920 0.000 0.008 0.000 0.044 0.028
#> GSM39175 1 0.5106 0.3647 0.600 0.000 0.324 0.000 0.052 0.024
#> GSM39176 1 0.1578 0.6636 0.936 0.000 0.012 0.000 0.048 0.004
#> GSM39177 3 0.4027 0.4765 0.308 0.000 0.672 0.000 0.012 0.008
#> GSM39178 3 0.3893 0.6636 0.056 0.000 0.796 0.000 0.028 0.120
#> GSM39179 3 0.2830 0.6471 0.064 0.000 0.864 0.000 0.068 0.004
#> GSM39180 5 0.4722 0.2392 0.052 0.000 0.116 0.072 0.752 0.008
#> GSM39181 1 0.5159 0.5071 0.664 0.000 0.164 0.000 0.156 0.016
#> GSM39182 1 0.6500 -0.0129 0.460 0.000 0.036 0.168 0.332 0.004
#> GSM39183 3 0.6095 0.2840 0.344 0.004 0.476 0.000 0.164 0.012
#> GSM39184 6 0.6523 0.2562 0.312 0.000 0.136 0.000 0.068 0.484
#> GSM39185 5 0.6310 0.0126 0.356 0.000 0.184 0.024 0.436 0.000
#> GSM39186 6 0.2711 0.6904 0.012 0.000 0.116 0.000 0.012 0.860
#> GSM39187 1 0.4170 0.6137 0.776 0.000 0.120 0.000 0.076 0.028
#> GSM39116 5 0.7038 0.0862 0.044 0.044 0.000 0.120 0.424 0.368
#> GSM39117 4 0.0363 0.8480 0.000 0.000 0.000 0.988 0.012 0.000
#> GSM39118 2 0.4025 0.7856 0.000 0.804 0.004 0.064 0.048 0.080
#> GSM39119 2 0.2058 0.8705 0.000 0.916 0.012 0.048 0.024 0.000
#> GSM39120 1 0.4268 0.2093 0.556 0.000 0.004 0.000 0.428 0.012
#> GSM39121 1 0.4543 0.5568 0.756 0.112 0.000 0.000 0.076 0.056
#> GSM39122 6 0.5203 0.5020 0.188 0.052 0.000 0.000 0.080 0.680
#> GSM39123 4 0.0806 0.8495 0.000 0.000 0.000 0.972 0.020 0.008
#> GSM39124 2 0.1333 0.8839 0.000 0.944 0.000 0.000 0.008 0.048
#> GSM39125 1 0.1956 0.6500 0.908 0.000 0.000 0.004 0.080 0.008
#> GSM39126 1 0.3962 0.5528 0.772 0.128 0.000 0.000 0.096 0.004
#> GSM39127 2 0.5530 0.2795 0.328 0.564 0.000 0.000 0.080 0.028
#> GSM39128 1 0.4652 0.4735 0.680 0.252 0.000 0.000 0.048 0.020
#> GSM39129 2 0.2366 0.8573 0.000 0.900 0.056 0.024 0.020 0.000
#> GSM39130 4 0.0363 0.8471 0.000 0.000 0.000 0.988 0.012 0.000
#> GSM39131 2 0.2281 0.8624 0.048 0.908 0.000 0.004 0.028 0.012
#> GSM39132 2 0.0696 0.8959 0.004 0.980 0.000 0.004 0.008 0.004
#> GSM39133 4 0.1124 0.8451 0.000 0.000 0.000 0.956 0.036 0.008
#> GSM39134 2 0.0909 0.8949 0.000 0.968 0.000 0.020 0.012 0.000
#> GSM39135 2 0.2567 0.8465 0.004 0.876 0.000 0.012 0.008 0.100
#> GSM39136 4 0.5974 0.4417 0.000 0.252 0.000 0.584 0.092 0.072
#> GSM39137 2 0.3505 0.7683 0.092 0.824 0.000 0.000 0.016 0.068
#> GSM39138 2 0.0964 0.8940 0.000 0.968 0.016 0.004 0.012 0.000
#> GSM39139 2 0.0146 0.8967 0.000 0.996 0.000 0.000 0.004 0.000
#> GSM39140 1 0.5116 0.3353 0.604 0.044 0.000 0.000 0.320 0.032
#> GSM39141 1 0.4140 0.5376 0.744 0.000 0.000 0.000 0.152 0.104
#> GSM39142 6 0.4371 0.3291 0.392 0.000 0.000 0.000 0.028 0.580
#> GSM39143 6 0.4793 0.4406 0.288 0.000 0.000 0.000 0.084 0.628
#> GSM39144 2 0.0551 0.8961 0.000 0.984 0.004 0.004 0.008 0.000
#> GSM39145 2 0.0405 0.8972 0.000 0.988 0.000 0.000 0.008 0.004
#> GSM39146 1 0.6783 0.2293 0.524 0.244 0.000 0.012 0.084 0.136
#> GSM39147 2 0.0696 0.8959 0.004 0.980 0.000 0.004 0.008 0.004
#> GSM39188 3 0.1296 0.6255 0.000 0.004 0.948 0.000 0.044 0.004
#> GSM39189 3 0.3318 0.6194 0.020 0.000 0.824 0.000 0.132 0.024
#> GSM39190 3 0.1706 0.6541 0.024 0.004 0.936 0.000 0.032 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) other(p) protocol(p) k
#> ATC:NMF 86 1.28e-01 8.56e-08 4.44e-07 2
#> ATC:NMF 82 1.64e-01 1.20e-07 1.62e-08 3
#> ATC:NMF 79 5.19e-02 9.69e-06 9.32e-06 4
#> ATC:NMF 73 1.03e-07 1.98e-09 8.01e-11 5
#> ATC:NMF 60 1.41e-08 2.51e-10 2.40e-11 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