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 100 columns.
#> Top rows are extracted by 'SD, CV, MAD, ATC' methods.
#> Subgroups are detected by 'hclust, kmeans, skmeans, pam, mclust, NMF' method.
#> Number of partitions are tried for k = 2, 3, 4, 5, 6.
#> Performed in total 30000 partitions by row resampling.
#>
#> Following methods can be applied to this 'ConsensusPartitionList' object:
#> [1] "cola_report" "collect_classes" "collect_plots" "collect_stats"
#> [5] "colnames" "functional_enrichment" "get_anno_col" "get_anno"
#> [9] "get_classes" "get_matrix" "get_membership" "get_stats"
#> [13] "is_best_k" "is_stable_k" "ncol" "nrow"
#> [17] "rownames" "show" "suggest_best_k" "test_to_known_factors"
#> [21] "top_rows_heatmap" "top_rows_overlap"
#>
#> You can get result for a single method by, e.g. object["SD", "hclust"] or object["SD:hclust"]
#> or a subset of methods by object[c("SD", "CV")], c("hclust", "kmeans")]
The call of run_all_consensus_partition_methods() was:
#> run_all_consensus_partition_methods(data = mat, mc.cores = 4, anno = anno)
Dimension of the input matrix:
mat = get_matrix(res_list)
dim(mat)
#> [1] 8353 100
The density distribution for each sample is visualized as in one column in the following heatmap. The clustering is based on the distance which is the Kolmogorov-Smirnov statistic between two distributions.
library(ComplexHeatmap)
densityHeatmap(mat, top_annotation = HeatmapAnnotation(df = get_anno(res_list),
col = get_anno_col(res_list)), ylab = "value", cluster_columns = TRUE, show_column_names = FALSE,
mc.cores = 4)
Folowing table shows the best k (number of partitions) for each combination
of top-value methods and partition methods. Clicking on the method name in
the table goes to the section for a single combination of methods.
The cola vignette explains the definition of the metrics used for determining the best number of partitions.
suggest_best_k(res_list)
| The best k | 1-PAC | Mean silhouette | Concordance | Optional k | ||
|---|---|---|---|---|---|---|
| CV:skmeans | 2 | 0.958 | 0.958 | 0.982 | ** | |
| CV:NMF | 2 | 0.956 | 0.929 | 0.972 | ** | |
| CV:kmeans | 2 | 0.938 | 0.960 | 0.982 | * | |
| ATC:skmeans | 3 | 0.910 | 0.917 | 0.961 | * | 2 |
| ATC:NMF | 2 | 0.898 | 0.905 | 0.962 | ||
| MAD:skmeans | 2 | 0.843 | 0.914 | 0.962 | ||
| MAD:kmeans | 2 | 0.840 | 0.891 | 0.955 | ||
| SD:NMF | 2 | 0.837 | 0.888 | 0.956 | ||
| MAD:NMF | 2 | 0.816 | 0.884 | 0.952 | ||
| SD:kmeans | 2 | 0.807 | 0.889 | 0.949 | ||
| CV:mclust | 2 | 0.802 | 0.907 | 0.956 | ||
| ATC:pam | 2 | 0.802 | 0.875 | 0.950 | ||
| SD:skmeans | 2 | 0.801 | 0.897 | 0.956 | ||
| SD:mclust | 2 | 0.720 | 0.896 | 0.920 | ||
| ATC:kmeans | 2 | 0.684 | 0.912 | 0.944 | ||
| ATC:mclust | 5 | 0.684 | 0.758 | 0.843 | ||
| MAD:mclust | 2 | 0.471 | 0.891 | 0.894 | ||
| MAD:pam | 2 | 0.461 | 0.819 | 0.903 | ||
| ATC:hclust | 3 | 0.406 | 0.657 | 0.813 | ||
| SD:pam | 2 | 0.251 | 0.748 | 0.848 | ||
| CV:pam | 2 | 0.141 | 0.629 | 0.801 | ||
| CV:hclust | 4 | 0.111 | 0.484 | 0.683 | ||
| SD:hclust | 3 | 0.102 | 0.613 | 0.734 | ||
| MAD:hclust | 2 | 0.084 | 0.711 | 0.802 |
**: 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 0.8374 0.888 0.956 0.502 0.495 0.495
#> CV:NMF 2 0.9563 0.929 0.972 0.503 0.496 0.496
#> MAD:NMF 2 0.8157 0.884 0.952 0.503 0.495 0.495
#> ATC:NMF 2 0.8976 0.905 0.962 0.473 0.515 0.515
#> SD:skmeans 2 0.8009 0.897 0.956 0.505 0.495 0.495
#> CV:skmeans 2 0.9581 0.958 0.982 0.505 0.495 0.495
#> MAD:skmeans 2 0.8432 0.914 0.962 0.505 0.495 0.495
#> ATC:skmeans 2 0.9574 0.939 0.975 0.504 0.496 0.496
#> SD:mclust 2 0.7205 0.896 0.920 0.477 0.495 0.495
#> CV:mclust 2 0.8018 0.907 0.956 0.499 0.495 0.495
#> MAD:mclust 2 0.4710 0.891 0.894 0.469 0.495 0.495
#> ATC:mclust 2 0.7765 0.915 0.954 0.366 0.665 0.665
#> SD:kmeans 2 0.8070 0.889 0.949 0.503 0.497 0.497
#> CV:kmeans 2 0.9380 0.960 0.982 0.505 0.495 0.495
#> MAD:kmeans 2 0.8396 0.891 0.955 0.504 0.496 0.496
#> ATC:kmeans 2 0.6840 0.912 0.944 0.481 0.508 0.508
#> SD:pam 2 0.2508 0.748 0.848 0.459 0.553 0.553
#> CV:pam 2 0.1409 0.629 0.801 0.475 0.519 0.519
#> MAD:pam 2 0.4611 0.819 0.903 0.462 0.535 0.535
#> ATC:pam 2 0.8018 0.875 0.950 0.489 0.515 0.515
#> SD:hclust 2 0.0708 0.468 0.674 0.381 0.515 0.515
#> CV:hclust 2 0.1196 0.608 0.797 0.292 0.904 0.904
#> MAD:hclust 2 0.0844 0.711 0.802 0.434 0.502 0.502
#> ATC:hclust 2 0.3724 0.727 0.860 0.351 0.642 0.642
get_stats(res_list, k = 3)
#> k 1-PAC mean_silhouette concordance area_increased Rand Jaccard
#> SD:NMF 3 0.4719 0.519 0.773 0.323 0.736 0.518
#> CV:NMF 3 0.3978 0.501 0.722 0.309 0.736 0.516
#> MAD:NMF 3 0.4918 0.587 0.802 0.320 0.767 0.561
#> ATC:NMF 3 0.8323 0.889 0.944 0.402 0.708 0.490
#> SD:skmeans 3 0.5624 0.730 0.831 0.306 0.800 0.614
#> CV:skmeans 3 0.5039 0.694 0.823 0.294 0.805 0.625
#> MAD:skmeans 3 0.7380 0.777 0.891 0.300 0.817 0.645
#> ATC:skmeans 3 0.9097 0.917 0.961 0.308 0.762 0.555
#> SD:mclust 3 0.4177 0.608 0.741 0.245 0.867 0.736
#> CV:mclust 3 0.5115 0.740 0.846 0.225 0.870 0.739
#> MAD:mclust 3 0.4943 0.713 0.819 0.282 0.838 0.680
#> ATC:mclust 3 0.3838 0.732 0.845 0.669 0.666 0.512
#> SD:kmeans 3 0.5471 0.651 0.792 0.268 0.829 0.664
#> CV:kmeans 3 0.6414 0.742 0.836 0.257 0.857 0.717
#> MAD:kmeans 3 0.6226 0.760 0.828 0.263 0.847 0.697
#> ATC:kmeans 3 0.4383 0.654 0.787 0.300 0.712 0.501
#> SD:pam 3 0.3828 0.673 0.801 0.428 0.728 0.529
#> CV:pam 3 0.2795 0.596 0.715 0.380 0.714 0.495
#> MAD:pam 3 0.3953 0.714 0.824 0.430 0.746 0.544
#> ATC:pam 3 0.7539 0.824 0.929 0.177 0.916 0.837
#> SD:hclust 3 0.1019 0.613 0.734 0.351 0.666 0.498
#> CV:hclust 3 0.0831 0.458 0.705 0.656 0.594 0.563
#> MAD:hclust 3 0.1149 0.657 0.736 0.211 0.937 0.879
#> ATC:hclust 3 0.4058 0.657 0.813 0.656 0.638 0.477
get_stats(res_list, k = 4)
#> k 1-PAC mean_silhouette concordance area_increased Rand Jaccard
#> SD:NMF 4 0.436 0.4354 0.693 0.1119 0.776 0.455
#> CV:NMF 4 0.436 0.4340 0.693 0.1202 0.802 0.492
#> MAD:NMF 4 0.420 0.3959 0.617 0.1115 0.858 0.620
#> ATC:NMF 4 0.759 0.8176 0.896 0.1328 0.845 0.580
#> SD:skmeans 4 0.457 0.4764 0.686 0.1263 0.877 0.671
#> CV:skmeans 4 0.442 0.3540 0.615 0.1301 0.858 0.619
#> MAD:skmeans 4 0.495 0.4730 0.698 0.1301 0.961 0.888
#> ATC:skmeans 4 0.803 0.8459 0.915 0.1324 0.828 0.546
#> SD:mclust 4 0.508 0.7015 0.801 0.1577 0.897 0.748
#> CV:mclust 4 0.537 0.5828 0.780 0.1071 0.972 0.924
#> MAD:mclust 4 0.670 0.8186 0.869 0.1206 0.895 0.729
#> ATC:mclust 4 0.817 0.8476 0.934 0.0689 0.704 0.418
#> SD:kmeans 4 0.561 0.7386 0.800 0.1196 0.897 0.728
#> CV:kmeans 4 0.511 0.5823 0.750 0.1161 0.941 0.845
#> MAD:kmeans 4 0.549 0.7028 0.742 0.1256 0.936 0.825
#> ATC:kmeans 4 0.598 0.7177 0.809 0.1528 0.838 0.597
#> SD:pam 4 0.473 0.5743 0.764 0.1201 0.848 0.589
#> CV:pam 4 0.404 0.4900 0.724 0.1130 0.838 0.564
#> MAD:pam 4 0.487 0.6446 0.776 0.1178 0.893 0.692
#> ATC:pam 4 0.593 0.6898 0.848 0.1960 0.858 0.685
#> SD:hclust 4 0.131 0.0731 0.482 0.1195 0.571 0.386
#> CV:hclust 4 0.111 0.4841 0.683 0.1561 0.951 0.911
#> MAD:hclust 4 0.187 0.5660 0.700 0.1400 0.958 0.914
#> ATC:hclust 4 0.412 0.4517 0.703 0.1753 0.864 0.681
get_stats(res_list, k = 5)
#> k 1-PAC mean_silhouette concordance area_increased Rand Jaccard
#> SD:NMF 5 0.468 0.428 0.617 0.0693 0.821 0.453
#> CV:NMF 5 0.481 0.414 0.633 0.0713 0.852 0.510
#> MAD:NMF 5 0.477 0.439 0.640 0.0712 0.803 0.422
#> ATC:NMF 5 0.666 0.631 0.788 0.0577 0.915 0.682
#> SD:skmeans 5 0.489 0.433 0.643 0.0696 0.894 0.652
#> CV:skmeans 5 0.455 0.403 0.605 0.0700 0.863 0.564
#> MAD:skmeans 5 0.489 0.372 0.585 0.0715 0.867 0.600
#> ATC:skmeans 5 0.842 0.810 0.894 0.0685 0.888 0.598
#> SD:mclust 5 0.746 0.730 0.864 0.1348 0.854 0.574
#> CV:mclust 5 0.641 0.706 0.816 0.1464 0.797 0.467
#> MAD:mclust 5 0.639 0.692 0.806 0.1333 0.885 0.641
#> ATC:mclust 5 0.684 0.758 0.843 0.1152 0.922 0.787
#> SD:kmeans 5 0.584 0.482 0.693 0.0719 0.921 0.743
#> CV:kmeans 5 0.563 0.451 0.703 0.0684 0.942 0.830
#> MAD:kmeans 5 0.607 0.592 0.703 0.0758 0.902 0.685
#> ATC:kmeans 5 0.646 0.616 0.753 0.0831 0.916 0.715
#> SD:pam 5 0.492 0.357 0.672 0.0417 0.896 0.668
#> CV:pam 5 0.443 0.413 0.704 0.0265 0.988 0.952
#> MAD:pam 5 0.530 0.480 0.745 0.0484 0.964 0.868
#> ATC:pam 5 0.831 0.828 0.919 0.1225 0.818 0.508
#> SD:hclust 5 0.186 0.514 0.674 0.0675 0.575 0.379
#> CV:hclust 5 0.160 0.416 0.667 0.0746 0.923 0.853
#> MAD:hclust 5 0.228 0.506 0.680 0.0792 0.957 0.906
#> ATC:hclust 5 0.451 0.405 0.667 0.0744 0.839 0.574
get_stats(res_list, k = 6)
#> k 1-PAC mean_silhouette concordance area_increased Rand Jaccard
#> SD:NMF 6 0.524 0.385 0.628 0.0445 0.844 0.420
#> CV:NMF 6 0.516 0.359 0.605 0.0434 0.908 0.622
#> MAD:NMF 6 0.527 0.392 0.598 0.0431 0.892 0.564
#> ATC:NMF 6 0.680 0.568 0.755 0.0452 0.903 0.583
#> SD:skmeans 6 0.503 0.324 0.577 0.0420 0.898 0.597
#> CV:skmeans 6 0.485 0.240 0.500 0.0446 0.856 0.484
#> MAD:skmeans 6 0.518 0.293 0.596 0.0413 0.889 0.581
#> ATC:skmeans 6 0.827 0.728 0.864 0.0380 0.940 0.722
#> SD:mclust 6 0.709 0.677 0.816 0.0399 0.922 0.672
#> CV:mclust 6 0.641 0.603 0.766 0.0405 0.943 0.753
#> MAD:mclust 6 0.683 0.628 0.764 0.0596 0.901 0.586
#> ATC:mclust 6 0.685 0.625 0.787 0.0625 0.829 0.500
#> SD:kmeans 6 0.605 0.444 0.650 0.0452 0.921 0.695
#> CV:kmeans 6 0.581 0.413 0.622 0.0421 0.901 0.676
#> MAD:kmeans 6 0.616 0.552 0.715 0.0448 0.983 0.922
#> ATC:kmeans 6 0.709 0.595 0.746 0.0465 0.921 0.678
#> SD:pam 6 0.555 0.446 0.738 0.0239 0.858 0.538
#> CV:pam 6 0.459 0.444 0.698 0.0131 0.972 0.894
#> MAD:pam 6 0.559 0.543 0.746 0.0281 0.921 0.692
#> ATC:pam 6 0.817 0.760 0.887 0.0443 0.954 0.802
#> SD:hclust 6 0.252 0.460 0.640 0.0920 0.886 0.764
#> CV:hclust 6 0.245 0.385 0.656 0.0800 0.953 0.896
#> MAD:hclust 6 0.310 0.358 0.660 0.0691 0.916 0.808
#> ATC:hclust 6 0.525 0.556 0.719 0.0565 0.920 0.710
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 genotype/variation(p) k
#> SD:NMF 93 1.61e-04 2
#> CV:NMF 96 9.23e-05 2
#> MAD:NMF 94 5.70e-05 2
#> ATC:NMF 93 7.87e-05 2
#> SD:skmeans 95 1.56e-05 2
#> CV:skmeans 98 2.90e-05 2
#> MAD:skmeans 97 1.29e-05 2
#> ATC:skmeans 96 1.32e-04 2
#> SD:mclust 97 5.16e-06 2
#> CV:mclust 97 3.40e-06 2
#> MAD:mclust 97 5.16e-06 2
#> ATC:mclust 99 3.10e-08 2
#> SD:kmeans 94 3.38e-05 2
#> CV:kmeans 100 7.47e-05 2
#> MAD:kmeans 94 2.03e-05 2
#> ATC:kmeans 100 1.21e-04 2
#> SD:pam 91 1.05e-04 2
#> CV:pam 78 4.69e-04 2
#> MAD:pam 94 1.39e-04 2
#> ATC:pam 94 2.43e-06 2
#> SD:hclust 74 1.08e-04 2
#> CV:hclust 81 2.79e-02 2
#> MAD:hclust 92 6.92e-05 2
#> ATC:hclust 96 6.13e-02 2
test_to_known_factors(res_list, k = 3)
#> n genotype/variation(p) k
#> SD:NMF 65 1.51e-03 3
#> CV:NMF 62 2.60e-04 3
#> MAD:NMF 74 6.19e-04 3
#> ATC:NMF 97 4.24e-06 3
#> SD:skmeans 88 1.77e-05 3
#> CV:skmeans 81 3.48e-05 3
#> MAD:skmeans 89 2.70e-05 3
#> ATC:skmeans 98 9.99e-06 3
#> SD:mclust 77 1.14e-03 3
#> CV:mclust 94 2.57e-04 3
#> MAD:mclust 92 3.78e-04 3
#> ATC:mclust 89 1.82e-12 3
#> SD:kmeans 81 1.87e-03 3
#> CV:kmeans 90 1.40e-03 3
#> MAD:kmeans 95 2.34e-04 3
#> ATC:kmeans 86 6.61e-04 3
#> SD:pam 83 1.92e-07 3
#> CV:pam 79 3.84e-06 3
#> MAD:pam 89 1.66e-07 3
#> ATC:pam 89 2.91e-06 3
#> SD:hclust 82 1.31e-03 3
#> CV:hclust 60 1.18e-03 3
#> MAD:hclust 90 2.73e-04 3
#> ATC:hclust 79 2.43e-03 3
test_to_known_factors(res_list, k = 4)
#> n genotype/variation(p) k
#> SD:NMF 43 2.60e-02 4
#> CV:NMF 45 4.06e-02 4
#> MAD:NMF 40 2.47e-02 4
#> ATC:NMF 91 2.02e-10 4
#> SD:skmeans 50 3.54e-05 4
#> CV:skmeans 39 5.81e-01 4
#> MAD:skmeans 50 1.06e-05 4
#> ATC:skmeans 93 8.62e-11 4
#> SD:mclust 90 8.84e-06 4
#> CV:mclust 74 8.01e-03 4
#> MAD:mclust 94 1.13e-05 4
#> ATC:mclust 94 4.94e-12 4
#> SD:kmeans 91 1.35e-06 4
#> CV:kmeans 74 3.85e-06 4
#> MAD:kmeans 90 2.55e-07 4
#> ATC:kmeans 86 9.88e-09 4
#> SD:pam 71 3.86e-06 4
#> CV:pam 55 6.51e-03 4
#> MAD:pam 79 1.69e-05 4
#> ATC:pam 82 3.25e-08 4
#> SD:hclust 30 1.06e-02 4
#> CV:hclust 62 1.33e-02 4
#> MAD:hclust 75 2.36e-03 4
#> ATC:hclust 58 5.60e-07 4
test_to_known_factors(res_list, k = 5)
#> n genotype/variation(p) k
#> SD:NMF 53 3.79e-04 5
#> CV:NMF 52 1.00e-02 5
#> MAD:NMF 51 2.55e-03 5
#> ATC:NMF 79 1.26e-11 5
#> SD:skmeans 48 5.90e-07 5
#> CV:skmeans 36 6.53e-05 5
#> MAD:skmeans 37 1.35e-03 5
#> ATC:skmeans 90 5.14e-14 5
#> SD:mclust 85 1.05e-06 5
#> CV:mclust 88 1.97e-07 5
#> MAD:mclust 85 1.61e-06 5
#> ATC:mclust 91 2.60e-13 5
#> SD:kmeans 44 6.46e-03 5
#> CV:kmeans 60 7.39e-10 5
#> MAD:kmeans 70 7.12e-12 5
#> ATC:kmeans 74 1.24e-10 5
#> SD:pam 37 1.81e-01 5
#> CV:pam 45 9.88e-03 5
#> MAD:pam 59 7.29e-03 5
#> ATC:pam 93 7.44e-12 5
#> SD:hclust 64 2.65e-03 5
#> CV:hclust 54 9.59e-03 5
#> MAD:hclust 69 2.86e-03 5
#> ATC:hclust 44 3.87e-06 5
test_to_known_factors(res_list, k = 6)
#> n genotype/variation(p) k
#> SD:NMF 36 1.70e-03 6
#> CV:NMF 33 1.18e-03 6
#> MAD:NMF 41 2.62e-02 6
#> ATC:NMF 71 5.36e-07 6
#> SD:skmeans 30 5.55e-02 6
#> CV:skmeans 3 NA 6
#> MAD:skmeans 27 8.71e-02 6
#> ATC:skmeans 83 1.99e-14 6
#> SD:mclust 73 2.83e-09 6
#> CV:mclust 78 4.07e-05 6
#> MAD:mclust 75 4.67e-09 6
#> ATC:mclust 75 4.12e-13 6
#> SD:kmeans 49 5.34e-08 6
#> CV:kmeans 54 6.55e-06 6
#> MAD:kmeans 68 7.05e-12 6
#> ATC:kmeans 75 2.57e-16 6
#> SD:pam 45 1.72e-01 6
#> CV:pam 45 9.88e-03 6
#> MAD:pam 62 7.42e-04 6
#> ATC:pam 87 8.79e-17 6
#> SD:hclust 58 6.39e-04 6
#> CV:hclust 44 1.90e-03 6
#> MAD:hclust 33 9.63e-03 6
#> ATC:hclust 69 5.43e-13 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 100 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 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 0.0708 0.4681 0.674 0.3807 0.515 0.515
#> 3 3 0.1019 0.6131 0.734 0.3508 0.666 0.498
#> 4 4 0.1306 0.0731 0.482 0.1195 0.571 0.386
#> 5 5 0.1859 0.5141 0.674 0.0675 0.575 0.379
#> 6 6 0.2524 0.4598 0.640 0.0920 0.886 0.764
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
#> GSM25548 1 0.952 0.1252 0.628 0.372
#> GSM25549 1 0.952 0.1379 0.628 0.372
#> GSM25550 1 0.955 0.1244 0.624 0.376
#> GSM25551 1 0.998 -0.3704 0.524 0.476
#> GSM25570 1 0.966 0.0693 0.608 0.392
#> GSM25571 1 0.966 0.0693 0.608 0.392
#> GSM25358 1 0.995 -0.2356 0.540 0.460
#> GSM25359 1 0.995 -0.2356 0.540 0.460
#> GSM25360 1 0.595 0.5915 0.856 0.144
#> GSM25361 1 0.595 0.5915 0.856 0.144
#> GSM25377 2 0.987 -0.1364 0.432 0.568
#> GSM25378 1 0.971 0.3599 0.600 0.400
#> GSM25401 2 0.876 0.6349 0.296 0.704
#> GSM25402 2 0.855 0.6560 0.280 0.720
#> GSM25349 2 0.795 0.6890 0.240 0.760
#> GSM25350 2 0.795 0.6890 0.240 0.760
#> GSM25356 2 0.936 0.5319 0.352 0.648
#> GSM25357 2 0.936 0.5319 0.352 0.648
#> GSM25385 1 0.430 0.6404 0.912 0.088
#> GSM25386 1 0.327 0.6265 0.940 0.060
#> GSM25399 2 0.995 -0.1848 0.460 0.540
#> GSM25400 1 0.861 0.5511 0.716 0.284
#> GSM48659 2 0.958 0.6729 0.380 0.620
#> GSM48660 2 0.833 0.7037 0.264 0.736
#> GSM25409 2 0.996 0.4734 0.464 0.536
#> GSM25410 1 0.327 0.6265 0.940 0.060
#> GSM25426 2 0.988 0.5965 0.436 0.564
#> GSM25427 2 0.998 0.0542 0.472 0.528
#> GSM25540 1 0.988 -0.2147 0.564 0.436
#> GSM25541 1 0.988 -0.2147 0.564 0.436
#> GSM25542 2 0.999 0.5030 0.480 0.520
#> GSM25543 2 0.999 0.4965 0.484 0.516
#> GSM25479 1 0.644 0.6394 0.836 0.164
#> GSM25480 1 0.644 0.6394 0.836 0.164
#> GSM25481 2 0.839 0.6342 0.268 0.732
#> GSM25482 2 0.839 0.6342 0.268 0.732
#> GSM48654 2 0.966 0.6590 0.392 0.608
#> GSM48650 2 0.781 0.6881 0.232 0.768
#> GSM48651 2 0.921 0.6993 0.336 0.664
#> GSM48652 2 0.921 0.6993 0.336 0.664
#> GSM48653 2 0.936 0.6919 0.352 0.648
#> GSM48662 2 0.943 0.6947 0.360 0.640
#> GSM48663 2 0.767 0.6789 0.224 0.776
#> GSM25524 1 0.163 0.6189 0.976 0.024
#> GSM25525 1 0.518 0.6429 0.884 0.116
#> GSM25526 1 0.866 0.5097 0.712 0.288
#> GSM25527 1 0.697 0.6257 0.812 0.188
#> GSM25528 1 0.184 0.6211 0.972 0.028
#> GSM25529 1 0.518 0.6437 0.884 0.116
#> GSM25530 1 0.184 0.6240 0.972 0.028
#> GSM25531 1 0.224 0.6301 0.964 0.036
#> GSM48661 2 0.998 0.5241 0.472 0.528
#> GSM25561 1 0.358 0.6435 0.932 0.068
#> GSM25562 1 0.456 0.6489 0.904 0.096
#> GSM25563 1 0.141 0.6315 0.980 0.020
#> GSM25564 1 0.814 0.5179 0.748 0.252
#> GSM25565 2 0.966 0.6574 0.392 0.608
#> GSM25566 2 0.997 0.5064 0.468 0.532
#> GSM25568 2 0.971 0.6399 0.400 0.600
#> GSM25569 2 0.973 0.6351 0.404 0.596
#> GSM25552 1 0.932 0.2059 0.652 0.348
#> GSM25553 1 0.932 0.2059 0.652 0.348
#> GSM25578 1 0.625 0.6279 0.844 0.156
#> GSM25579 1 0.615 0.6371 0.848 0.152
#> GSM25580 1 0.760 0.5869 0.780 0.220
#> GSM25581 1 0.760 0.5869 0.780 0.220
#> GSM48655 2 0.844 0.7071 0.272 0.728
#> GSM48656 2 0.998 0.5154 0.476 0.524
#> GSM48657 2 0.795 0.6943 0.240 0.760
#> GSM48658 2 0.998 0.5154 0.476 0.524
#> GSM25624 1 0.844 0.5698 0.728 0.272
#> GSM25625 1 0.653 0.6360 0.832 0.168
#> GSM25626 1 0.327 0.6265 0.940 0.060
#> GSM25627 1 0.978 0.0645 0.588 0.412
#> GSM25628 1 0.327 0.6265 0.940 0.060
#> GSM25629 1 1.000 -0.4154 0.508 0.492
#> GSM25630 1 0.204 0.6169 0.968 0.032
#> GSM25631 1 0.939 0.1572 0.644 0.356
#> GSM25632 1 0.456 0.6477 0.904 0.096
#> GSM25633 1 0.722 0.6139 0.800 0.200
#> GSM25634 1 0.781 0.5958 0.768 0.232
#> GSM25635 1 0.714 0.6150 0.804 0.196
#> GSM25656 1 0.373 0.6332 0.928 0.072
#> GSM25657 1 0.634 0.6287 0.840 0.160
#> GSM25658 1 0.781 0.6010 0.768 0.232
#> GSM25659 1 0.775 0.5411 0.772 0.228
#> GSM25660 1 0.730 0.6150 0.796 0.204
#> GSM25661 1 0.671 0.6183 0.824 0.176
#> GSM25662 1 0.987 -0.1419 0.568 0.432
#> GSM25663 1 0.987 -0.1419 0.568 0.432
#> GSM25680 1 0.929 0.1927 0.656 0.344
#> GSM25681 1 0.929 0.1927 0.656 0.344
#> GSM25682 2 0.861 0.7094 0.284 0.716
#> GSM25683 2 0.861 0.7094 0.284 0.716
#> GSM25684 2 0.958 0.6729 0.380 0.620
#> GSM25685 2 0.958 0.6729 0.380 0.620
#> GSM25686 2 0.861 0.7094 0.284 0.716
#> GSM25687 2 0.861 0.7094 0.284 0.716
#> GSM48664 2 0.995 -0.1848 0.460 0.540
#> GSM48665 1 0.730 0.6030 0.796 0.204
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM25548 2 0.754 0.363 0.432 0.528 0.040
#> GSM25549 2 0.763 0.344 0.432 0.524 0.044
#> GSM25550 2 0.745 0.346 0.436 0.528 0.036
#> GSM25551 2 0.637 0.656 0.268 0.704 0.028
#> GSM25570 2 0.742 0.382 0.420 0.544 0.036
#> GSM25571 2 0.742 0.382 0.420 0.544 0.036
#> GSM25358 2 0.738 0.537 0.320 0.628 0.052
#> GSM25359 2 0.738 0.537 0.320 0.628 0.052
#> GSM25360 1 0.558 0.531 0.736 0.256 0.008
#> GSM25361 1 0.558 0.531 0.736 0.256 0.008
#> GSM25377 3 0.426 0.883 0.012 0.140 0.848
#> GSM25378 2 0.999 -0.355 0.332 0.356 0.312
#> GSM25401 2 0.523 0.650 0.068 0.828 0.104
#> GSM25402 2 0.477 0.663 0.052 0.848 0.100
#> GSM25349 2 0.188 0.681 0.004 0.952 0.044
#> GSM25350 2 0.188 0.681 0.004 0.952 0.044
#> GSM25356 2 0.577 0.569 0.024 0.756 0.220
#> GSM25357 2 0.577 0.569 0.024 0.756 0.220
#> GSM25385 1 0.594 0.718 0.788 0.140 0.072
#> GSM25386 1 0.392 0.688 0.856 0.140 0.004
#> GSM25399 3 0.228 0.924 0.008 0.052 0.940
#> GSM25400 1 0.976 0.527 0.428 0.240 0.332
#> GSM48659 2 0.364 0.717 0.124 0.872 0.004
#> GSM48660 2 0.164 0.693 0.016 0.964 0.020
#> GSM25409 2 0.648 0.632 0.244 0.716 0.040
#> GSM25410 1 0.392 0.688 0.856 0.140 0.004
#> GSM25426 2 0.502 0.702 0.192 0.796 0.012
#> GSM25427 2 0.905 0.215 0.160 0.528 0.312
#> GSM25540 2 0.661 0.559 0.356 0.628 0.016
#> GSM25541 2 0.661 0.559 0.356 0.628 0.016
#> GSM25542 2 0.566 0.662 0.264 0.728 0.008
#> GSM25543 2 0.576 0.650 0.276 0.716 0.008
#> GSM25479 1 0.831 0.724 0.632 0.176 0.192
#> GSM25480 1 0.831 0.724 0.632 0.176 0.192
#> GSM25481 2 0.481 0.618 0.028 0.832 0.140
#> GSM25482 2 0.481 0.618 0.028 0.832 0.140
#> GSM48654 2 0.398 0.716 0.144 0.852 0.004
#> GSM48650 2 0.153 0.677 0.004 0.964 0.032
#> GSM48651 2 0.268 0.715 0.076 0.920 0.004
#> GSM48652 2 0.268 0.715 0.076 0.920 0.004
#> GSM48653 2 0.303 0.716 0.092 0.904 0.004
#> GSM48662 2 0.344 0.720 0.088 0.896 0.016
#> GSM48663 2 0.176 0.675 0.004 0.956 0.040
#> GSM25524 1 0.158 0.672 0.964 0.028 0.008
#> GSM25525 1 0.645 0.746 0.764 0.132 0.104
#> GSM25526 1 0.941 0.300 0.448 0.376 0.176
#> GSM25527 1 0.870 0.711 0.588 0.168 0.244
#> GSM25528 1 0.203 0.678 0.952 0.032 0.016
#> GSM25529 1 0.637 0.744 0.768 0.132 0.100
#> GSM25530 1 0.304 0.688 0.920 0.044 0.036
#> GSM25531 1 0.336 0.704 0.908 0.056 0.036
#> GSM48661 2 0.529 0.689 0.228 0.764 0.008
#> GSM25561 1 0.554 0.728 0.808 0.132 0.060
#> GSM25562 1 0.751 0.728 0.696 0.160 0.144
#> GSM25563 1 0.321 0.700 0.900 0.092 0.008
#> GSM25564 1 0.898 0.317 0.496 0.368 0.136
#> GSM25565 2 0.433 0.718 0.144 0.844 0.012
#> GSM25566 2 0.554 0.678 0.236 0.752 0.012
#> GSM25568 2 0.397 0.718 0.132 0.860 0.008
#> GSM25569 2 0.416 0.717 0.144 0.848 0.008
#> GSM25552 2 0.766 0.289 0.452 0.504 0.044
#> GSM25553 2 0.766 0.289 0.452 0.504 0.044
#> GSM25578 1 0.821 0.711 0.628 0.132 0.240
#> GSM25579 1 0.850 0.659 0.612 0.216 0.172
#> GSM25580 1 0.889 0.635 0.532 0.140 0.328
#> GSM25581 1 0.889 0.635 0.532 0.140 0.328
#> GSM48655 2 0.177 0.698 0.024 0.960 0.016
#> GSM48656 2 0.534 0.688 0.232 0.760 0.008
#> GSM48657 2 0.158 0.683 0.008 0.964 0.028
#> GSM48658 2 0.534 0.688 0.232 0.760 0.008
#> GSM25624 1 0.963 0.583 0.460 0.228 0.312
#> GSM25625 1 0.830 0.708 0.632 0.196 0.172
#> GSM25626 1 0.378 0.691 0.864 0.132 0.004
#> GSM25627 2 0.797 0.366 0.372 0.560 0.068
#> GSM25628 1 0.403 0.689 0.856 0.136 0.008
#> GSM25629 2 0.634 0.647 0.264 0.708 0.028
#> GSM25630 1 0.195 0.612 0.952 0.008 0.040
#> GSM25631 2 0.704 0.377 0.444 0.536 0.020
#> GSM25632 1 0.588 0.733 0.788 0.148 0.064
#> GSM25633 1 0.879 0.672 0.552 0.140 0.308
#> GSM25634 1 0.915 0.609 0.496 0.156 0.348
#> GSM25635 1 0.892 0.675 0.548 0.156 0.296
#> GSM25656 1 0.417 0.653 0.872 0.092 0.036
#> GSM25657 1 0.791 0.715 0.656 0.124 0.220
#> GSM25658 1 0.887 0.577 0.560 0.280 0.160
#> GSM25659 1 0.835 0.423 0.568 0.332 0.100
#> GSM25660 1 0.898 0.681 0.548 0.168 0.284
#> GSM25661 1 0.855 0.691 0.588 0.136 0.276
#> GSM25662 2 0.650 0.487 0.396 0.596 0.008
#> GSM25663 2 0.650 0.487 0.396 0.596 0.008
#> GSM25680 2 0.706 0.342 0.464 0.516 0.020
#> GSM25681 2 0.706 0.342 0.464 0.516 0.020
#> GSM25682 2 0.218 0.703 0.032 0.948 0.020
#> GSM25683 2 0.218 0.703 0.032 0.948 0.020
#> GSM25684 2 0.364 0.717 0.124 0.872 0.004
#> GSM25685 2 0.364 0.717 0.124 0.872 0.004
#> GSM25686 2 0.218 0.703 0.032 0.948 0.020
#> GSM25687 2 0.218 0.703 0.032 0.948 0.020
#> GSM48664 3 0.285 0.935 0.012 0.068 0.920
#> GSM48665 1 0.882 0.660 0.552 0.144 0.304
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM25548 2 0.180 0.2854 0.016 0.948 0.004 0.032
#> GSM25549 2 0.171 0.2943 0.020 0.952 0.004 0.024
#> GSM25550 2 0.151 0.2961 0.012 0.960 0.008 0.020
#> GSM25551 2 0.567 -0.2138 0.004 0.676 0.048 0.272
#> GSM25570 2 0.192 0.2819 0.012 0.944 0.008 0.036
#> GSM25571 2 0.192 0.2819 0.012 0.944 0.008 0.036
#> GSM25358 2 0.611 0.0745 0.028 0.696 0.056 0.220
#> GSM25359 2 0.611 0.0745 0.028 0.696 0.056 0.220
#> GSM25360 2 0.558 -0.2998 0.012 0.652 0.316 0.020
#> GSM25361 2 0.558 -0.2998 0.012 0.652 0.316 0.020
#> GSM25377 1 0.428 0.6125 0.800 0.024 0.004 0.172
#> GSM25378 2 0.966 -0.0115 0.280 0.352 0.148 0.220
#> GSM25401 4 0.730 0.7126 0.056 0.400 0.044 0.500
#> GSM25402 4 0.702 0.7389 0.052 0.404 0.032 0.512
#> GSM25349 4 0.543 0.8308 0.016 0.416 0.000 0.568
#> GSM25350 4 0.543 0.8308 0.016 0.416 0.000 0.568
#> GSM25356 4 0.750 0.6004 0.128 0.396 0.012 0.464
#> GSM25357 4 0.750 0.6004 0.128 0.396 0.012 0.464
#> GSM25385 2 0.726 -0.6006 0.060 0.460 0.444 0.036
#> GSM25386 2 0.594 -0.6140 0.000 0.484 0.480 0.036
#> GSM25399 1 0.259 0.6156 0.900 0.004 0.004 0.092
#> GSM25400 2 0.951 -0.1641 0.312 0.364 0.184 0.140
#> GSM48659 2 0.531 -0.5743 0.000 0.576 0.012 0.412
#> GSM48660 4 0.512 0.8149 0.004 0.440 0.000 0.556
#> GSM25409 2 0.492 -0.1205 0.024 0.732 0.004 0.240
#> GSM25410 2 0.594 -0.6140 0.000 0.484 0.480 0.036
#> GSM25426 2 0.620 -0.4797 0.000 0.564 0.060 0.376
#> GSM25427 2 0.923 -0.1043 0.268 0.336 0.076 0.320
#> GSM25540 2 0.479 0.0745 0.008 0.776 0.036 0.180
#> GSM25541 2 0.479 0.0745 0.008 0.776 0.036 0.180
#> GSM25542 2 0.609 -0.3344 0.004 0.608 0.052 0.336
#> GSM25543 2 0.628 -0.3130 0.004 0.600 0.064 0.332
#> GSM25479 2 0.829 -0.3258 0.200 0.484 0.280 0.036
#> GSM25480 2 0.829 -0.3258 0.200 0.484 0.280 0.036
#> GSM25481 4 0.734 0.7171 0.092 0.364 0.024 0.520
#> GSM25482 4 0.734 0.7171 0.092 0.364 0.024 0.520
#> GSM48654 2 0.514 -0.5336 0.000 0.600 0.008 0.392
#> GSM48650 4 0.514 0.8278 0.008 0.392 0.000 0.600
#> GSM48651 2 0.498 -0.6692 0.000 0.536 0.000 0.464
#> GSM48652 2 0.498 -0.6617 0.000 0.540 0.000 0.460
#> GSM48653 2 0.495 -0.6343 0.000 0.556 0.000 0.444
#> GSM48662 2 0.495 -0.6554 0.000 0.560 0.000 0.440
#> GSM48663 4 0.545 0.8279 0.012 0.400 0.004 0.584
#> GSM25524 3 0.546 0.7087 0.004 0.344 0.632 0.020
#> GSM25525 3 0.782 0.5051 0.124 0.396 0.452 0.028
#> GSM25526 2 0.890 0.1187 0.168 0.508 0.168 0.156
#> GSM25527 2 0.885 -0.3010 0.240 0.432 0.268 0.060
#> GSM25528 3 0.560 0.7070 0.012 0.340 0.632 0.016
#> GSM25529 3 0.771 0.5091 0.120 0.408 0.448 0.024
#> GSM25530 3 0.584 0.7130 0.028 0.356 0.608 0.008
#> GSM25531 3 0.597 0.7050 0.032 0.368 0.592 0.008
#> GSM48661 2 0.508 -0.3472 0.000 0.676 0.020 0.304
#> GSM25561 3 0.700 0.6330 0.044 0.444 0.476 0.036
#> GSM25562 2 0.854 -0.4774 0.136 0.476 0.312 0.076
#> GSM25563 3 0.612 0.6618 0.000 0.396 0.552 0.052
#> GSM25564 2 0.784 0.0948 0.124 0.616 0.140 0.120
#> GSM25565 2 0.496 -0.5102 0.000 0.616 0.004 0.380
#> GSM25566 2 0.469 -0.2719 0.000 0.712 0.012 0.276
#> GSM25568 2 0.514 -0.5463 0.000 0.600 0.008 0.392
#> GSM25569 2 0.507 -0.4985 0.000 0.620 0.008 0.372
#> GSM25552 2 0.200 0.3128 0.020 0.944 0.020 0.016
#> GSM25553 2 0.200 0.3128 0.020 0.944 0.020 0.016
#> GSM25578 2 0.861 -0.3557 0.248 0.408 0.308 0.036
#> GSM25579 2 0.747 -0.2338 0.176 0.580 0.224 0.020
#> GSM25580 2 0.895 -0.2814 0.328 0.368 0.248 0.056
#> GSM25581 2 0.895 -0.2814 0.328 0.368 0.248 0.056
#> GSM48655 4 0.510 0.8180 0.004 0.432 0.000 0.564
#> GSM48656 2 0.536 -0.3422 0.004 0.668 0.024 0.304
#> GSM48657 4 0.517 0.8316 0.008 0.404 0.000 0.588
#> GSM48658 2 0.536 -0.3422 0.004 0.668 0.024 0.304
#> GSM25624 2 0.946 -0.1755 0.296 0.376 0.204 0.124
#> GSM25625 2 0.900 -0.3853 0.160 0.436 0.304 0.100
#> GSM25626 3 0.594 0.5888 0.000 0.476 0.488 0.036
#> GSM25627 2 0.797 0.0904 0.052 0.556 0.144 0.248
#> GSM25628 3 0.594 0.5846 0.000 0.476 0.488 0.036
#> GSM25629 2 0.651 -0.2786 0.004 0.588 0.080 0.328
#> GSM25630 3 0.559 0.1319 0.004 0.096 0.732 0.168
#> GSM25631 2 0.396 0.2665 0.012 0.852 0.048 0.088
#> GSM25632 2 0.743 -0.6176 0.072 0.448 0.444 0.036
#> GSM25633 2 0.873 -0.3067 0.312 0.388 0.260 0.040
#> GSM25634 1 0.910 -0.3767 0.348 0.340 0.244 0.068
#> GSM25635 2 0.888 -0.2724 0.288 0.408 0.248 0.056
#> GSM25656 3 0.764 0.3440 0.012 0.256 0.532 0.200
#> GSM25657 2 0.867 -0.4003 0.232 0.392 0.336 0.040
#> GSM25658 2 0.895 -0.1511 0.156 0.480 0.244 0.120
#> GSM25659 2 0.760 -0.0114 0.092 0.608 0.224 0.076
#> GSM25660 2 0.881 -0.2637 0.272 0.432 0.240 0.056
#> GSM25661 2 0.865 -0.3202 0.280 0.404 0.280 0.036
#> GSM25662 2 0.345 0.1771 0.000 0.852 0.020 0.128
#> GSM25663 2 0.345 0.1771 0.000 0.852 0.020 0.128
#> GSM25680 2 0.275 0.2885 0.000 0.904 0.056 0.040
#> GSM25681 2 0.275 0.2885 0.000 0.904 0.056 0.040
#> GSM25682 4 0.497 0.8131 0.000 0.452 0.000 0.548
#> GSM25683 4 0.497 0.8131 0.000 0.452 0.000 0.548
#> GSM25684 2 0.531 -0.5743 0.000 0.576 0.012 0.412
#> GSM25685 2 0.531 -0.5743 0.000 0.576 0.012 0.412
#> GSM25686 4 0.497 0.8131 0.000 0.452 0.000 0.548
#> GSM25687 4 0.497 0.8131 0.000 0.452 0.000 0.548
#> GSM48664 1 0.293 0.6408 0.880 0.012 0.000 0.108
#> GSM48665 2 0.891 -0.2907 0.304 0.380 0.264 0.052
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM25548 2 0.674 0.2803 0.400 0.476 0.044 0.072 0.008
#> GSM25549 2 0.680 0.2478 0.412 0.460 0.044 0.076 0.008
#> GSM25550 2 0.668 0.2387 0.420 0.460 0.036 0.076 0.008
#> GSM25551 2 0.518 0.6273 0.196 0.720 0.048 0.032 0.004
#> GSM25570 2 0.667 0.2809 0.404 0.476 0.036 0.076 0.008
#> GSM25571 2 0.667 0.2809 0.404 0.476 0.036 0.076 0.008
#> GSM25358 2 0.656 0.4619 0.312 0.564 0.036 0.076 0.012
#> GSM25359 2 0.656 0.4619 0.312 0.564 0.036 0.076 0.012
#> GSM25360 1 0.760 0.4787 0.504 0.192 0.236 0.052 0.016
#> GSM25361 1 0.760 0.4787 0.504 0.192 0.236 0.052 0.016
#> GSM25377 5 0.456 0.7656 0.096 0.080 0.000 0.036 0.788
#> GSM25378 1 0.762 0.3996 0.516 0.240 0.008 0.120 0.116
#> GSM25401 2 0.636 0.5634 0.088 0.664 0.016 0.172 0.060
#> GSM25402 2 0.612 0.5787 0.072 0.680 0.016 0.176 0.056
#> GSM25349 2 0.369 0.6440 0.012 0.824 0.012 0.140 0.012
#> GSM25350 2 0.369 0.6440 0.012 0.824 0.012 0.140 0.012
#> GSM25356 2 0.735 0.4841 0.100 0.552 0.012 0.236 0.100
#> GSM25357 2 0.735 0.4841 0.100 0.552 0.012 0.236 0.100
#> GSM25385 1 0.649 0.4917 0.576 0.108 0.284 0.024 0.008
#> GSM25386 1 0.645 0.3995 0.500 0.136 0.352 0.012 0.000
#> GSM25399 5 0.252 0.7525 0.064 0.004 0.008 0.020 0.904
#> GSM25400 1 0.675 0.5087 0.620 0.144 0.004 0.080 0.152
#> GSM48659 2 0.235 0.6889 0.056 0.912 0.016 0.016 0.000
#> GSM48660 2 0.303 0.6571 0.008 0.848 0.000 0.136 0.008
#> GSM25409 2 0.608 0.5572 0.272 0.612 0.008 0.092 0.016
#> GSM25410 1 0.645 0.3995 0.500 0.136 0.352 0.012 0.000
#> GSM25426 2 0.418 0.6768 0.092 0.816 0.064 0.024 0.004
#> GSM25427 2 0.802 0.0634 0.324 0.416 0.008 0.120 0.132
#> GSM25540 2 0.619 0.5120 0.264 0.612 0.068 0.056 0.000
#> GSM25541 2 0.619 0.5120 0.264 0.612 0.068 0.056 0.000
#> GSM25542 2 0.559 0.6227 0.136 0.720 0.084 0.056 0.004
#> GSM25543 2 0.581 0.6096 0.136 0.708 0.092 0.056 0.008
#> GSM25479 1 0.364 0.6396 0.848 0.092 0.008 0.020 0.032
#> GSM25480 1 0.364 0.6396 0.848 0.092 0.008 0.020 0.032
#> GSM25481 2 0.655 0.5281 0.096 0.628 0.004 0.196 0.076
#> GSM25482 2 0.655 0.5281 0.096 0.628 0.004 0.196 0.076
#> GSM48654 2 0.286 0.6871 0.076 0.884 0.016 0.024 0.000
#> GSM48650 2 0.341 0.6201 0.008 0.812 0.000 0.172 0.008
#> GSM48651 2 0.181 0.6886 0.040 0.936 0.000 0.020 0.004
#> GSM48652 2 0.171 0.6886 0.040 0.940 0.000 0.016 0.004
#> GSM48653 2 0.165 0.6885 0.040 0.940 0.000 0.020 0.000
#> GSM48662 2 0.190 0.6920 0.040 0.932 0.004 0.024 0.000
#> GSM48663 2 0.382 0.6177 0.008 0.796 0.004 0.176 0.016
#> GSM25524 1 0.606 0.3375 0.576 0.020 0.340 0.048 0.016
#> GSM25525 1 0.518 0.5935 0.740 0.056 0.164 0.012 0.028
#> GSM25526 1 0.732 0.2986 0.512 0.324 0.072 0.036 0.056
#> GSM25527 1 0.450 0.6349 0.808 0.080 0.024 0.020 0.068
#> GSM25528 1 0.567 0.3764 0.604 0.012 0.332 0.032 0.020
#> GSM25529 1 0.501 0.5949 0.748 0.052 0.164 0.008 0.028
#> GSM25530 1 0.592 0.3969 0.596 0.020 0.328 0.028 0.028
#> GSM25531 1 0.559 0.4352 0.628 0.016 0.308 0.024 0.024
#> GSM48661 2 0.457 0.6563 0.132 0.780 0.048 0.040 0.000
#> GSM25561 1 0.622 0.5152 0.660 0.048 0.208 0.064 0.020
#> GSM25562 1 0.590 0.5933 0.728 0.072 0.088 0.076 0.036
#> GSM25563 1 0.694 0.3225 0.488 0.076 0.376 0.044 0.016
#> GSM25564 1 0.715 0.3270 0.540 0.296 0.072 0.072 0.020
#> GSM25565 2 0.391 0.6882 0.100 0.820 0.012 0.068 0.000
#> GSM25566 2 0.464 0.6423 0.188 0.748 0.020 0.044 0.000
#> GSM25568 2 0.384 0.6790 0.064 0.824 0.012 0.100 0.000
#> GSM25569 2 0.420 0.6758 0.088 0.800 0.012 0.100 0.000
#> GSM25552 1 0.667 -0.2007 0.448 0.436 0.044 0.064 0.008
#> GSM25553 1 0.667 -0.1881 0.452 0.432 0.044 0.064 0.008
#> GSM25578 1 0.328 0.6181 0.868 0.052 0.012 0.004 0.064
#> GSM25579 1 0.432 0.6233 0.788 0.152 0.024 0.004 0.032
#> GSM25580 1 0.439 0.5847 0.796 0.056 0.008 0.016 0.124
#> GSM25581 1 0.439 0.5847 0.796 0.056 0.008 0.016 0.124
#> GSM48655 2 0.315 0.6562 0.012 0.844 0.000 0.136 0.008
#> GSM48656 2 0.477 0.6545 0.136 0.772 0.048 0.040 0.004
#> GSM48657 2 0.344 0.6294 0.012 0.816 0.000 0.164 0.008
#> GSM48658 2 0.477 0.6545 0.136 0.772 0.048 0.040 0.004
#> GSM25624 1 0.621 0.5511 0.684 0.132 0.012 0.072 0.100
#> GSM25625 1 0.680 0.5937 0.656 0.124 0.112 0.060 0.048
#> GSM25626 1 0.639 0.4008 0.508 0.128 0.352 0.012 0.000
#> GSM25627 2 0.703 0.3090 0.308 0.536 0.096 0.032 0.028
#> GSM25628 1 0.636 0.3897 0.504 0.136 0.352 0.008 0.000
#> GSM25629 2 0.510 0.6287 0.172 0.732 0.072 0.020 0.004
#> GSM25630 4 0.625 0.0000 0.080 0.008 0.396 0.504 0.012
#> GSM25631 2 0.674 0.3268 0.360 0.508 0.068 0.060 0.004
#> GSM25632 1 0.627 0.5224 0.592 0.096 0.284 0.020 0.008
#> GSM25633 1 0.414 0.6046 0.820 0.056 0.008 0.020 0.096
#> GSM25634 1 0.540 0.5493 0.744 0.064 0.012 0.056 0.124
#> GSM25635 1 0.430 0.6158 0.816 0.064 0.012 0.024 0.084
#> GSM25656 3 0.479 0.0000 0.036 0.024 0.780 0.132 0.028
#> GSM25657 1 0.438 0.6174 0.820 0.048 0.056 0.016 0.060
#> GSM25658 1 0.702 0.5474 0.596 0.224 0.096 0.028 0.056
#> GSM25659 1 0.647 0.4307 0.600 0.264 0.092 0.032 0.012
#> GSM25660 1 0.449 0.6223 0.804 0.076 0.012 0.024 0.084
#> GSM25661 1 0.354 0.6163 0.852 0.052 0.008 0.008 0.080
#> GSM25662 2 0.619 0.4447 0.328 0.576 0.044 0.044 0.008
#> GSM25663 2 0.619 0.4447 0.328 0.576 0.044 0.044 0.008
#> GSM25680 2 0.711 0.2402 0.380 0.460 0.072 0.084 0.004
#> GSM25681 2 0.711 0.2402 0.380 0.460 0.072 0.084 0.004
#> GSM25682 2 0.326 0.6636 0.020 0.836 0.000 0.140 0.004
#> GSM25683 2 0.326 0.6636 0.020 0.836 0.000 0.140 0.004
#> GSM25684 2 0.235 0.6889 0.056 0.912 0.016 0.016 0.000
#> GSM25685 2 0.235 0.6889 0.056 0.912 0.016 0.016 0.000
#> GSM25686 2 0.326 0.6636 0.020 0.836 0.000 0.140 0.004
#> GSM25687 2 0.326 0.6636 0.020 0.836 0.000 0.140 0.004
#> GSM48664 5 0.304 0.8277 0.100 0.032 0.000 0.004 0.864
#> GSM48665 1 0.410 0.6009 0.820 0.056 0.008 0.016 0.100
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM25548 5 0.744 0.3020 0.300 NA 0.144 0.000 0.436 0.040
#> GSM25549 5 0.751 0.2716 0.316 NA 0.144 0.000 0.416 0.040
#> GSM25550 5 0.743 0.2625 0.324 NA 0.144 0.000 0.416 0.036
#> GSM25551 5 0.512 0.6028 0.152 NA 0.084 0.000 0.712 0.012
#> GSM25570 5 0.736 0.3003 0.312 NA 0.140 0.000 0.436 0.040
#> GSM25571 5 0.736 0.3003 0.312 NA 0.140 0.000 0.436 0.040
#> GSM25358 5 0.697 0.4642 0.248 NA 0.092 0.016 0.544 0.024
#> GSM25359 5 0.697 0.4642 0.248 NA 0.092 0.016 0.544 0.024
#> GSM25360 1 0.767 -0.0527 0.360 NA 0.356 0.000 0.144 0.100
#> GSM25361 1 0.767 -0.0527 0.360 NA 0.356 0.000 0.144 0.100
#> GSM25377 4 0.424 0.7440 0.128 NA 0.004 0.772 0.020 0.000
#> GSM25378 1 0.733 0.3756 0.552 NA 0.024 0.076 0.160 0.036
#> GSM25401 5 0.689 0.4228 0.120 NA 0.024 0.036 0.508 0.016
#> GSM25402 5 0.670 0.4399 0.104 NA 0.024 0.032 0.524 0.016
#> GSM25349 5 0.380 0.5658 0.016 NA 0.000 0.000 0.740 0.012
#> GSM25350 5 0.380 0.5658 0.016 NA 0.000 0.000 0.740 0.012
#> GSM25356 5 0.766 0.3575 0.104 NA 0.008 0.068 0.448 0.076
#> GSM25357 5 0.766 0.3575 0.104 NA 0.008 0.068 0.448 0.076
#> GSM25385 3 0.631 0.4938 0.372 NA 0.492 0.008 0.080 0.036
#> GSM25386 3 0.605 0.6293 0.260 NA 0.576 0.004 0.108 0.052
#> GSM25399 4 0.240 0.6387 0.024 NA 0.000 0.896 0.000 0.016
#> GSM25400 1 0.622 0.4878 0.664 NA 0.016 0.120 0.084 0.028
#> GSM48659 5 0.201 0.6507 0.040 NA 0.008 0.000 0.924 0.012
#> GSM48660 5 0.333 0.5853 0.012 NA 0.000 0.000 0.788 0.008
#> GSM25409 5 0.653 0.5300 0.216 NA 0.072 0.008 0.576 0.008
#> GSM25410 3 0.609 0.6231 0.268 NA 0.568 0.004 0.108 0.052
#> GSM25426 5 0.420 0.6367 0.080 NA 0.068 0.000 0.800 0.024
#> GSM25427 1 0.792 0.0513 0.356 NA 0.008 0.084 0.340 0.044
#> GSM25540 5 0.631 0.5042 0.192 NA 0.136 0.000 0.600 0.024
#> GSM25541 5 0.631 0.5042 0.192 NA 0.136 0.000 0.600 0.024
#> GSM25542 5 0.594 0.5650 0.064 NA 0.128 0.004 0.676 0.036
#> GSM25543 5 0.612 0.5449 0.064 NA 0.140 0.004 0.660 0.040
#> GSM25479 1 0.341 0.5481 0.856 NA 0.040 0.004 0.052 0.036
#> GSM25480 1 0.341 0.5481 0.856 NA 0.040 0.004 0.052 0.036
#> GSM25481 5 0.658 0.3738 0.132 NA 0.004 0.052 0.492 0.004
#> GSM25482 5 0.658 0.3738 0.132 NA 0.004 0.052 0.492 0.004
#> GSM48654 5 0.264 0.6502 0.044 NA 0.040 0.000 0.892 0.008
#> GSM48650 5 0.367 0.5447 0.012 NA 0.000 0.000 0.740 0.008
#> GSM48651 5 0.186 0.6457 0.032 NA 0.000 0.000 0.920 0.000
#> GSM48652 5 0.179 0.6464 0.032 NA 0.000 0.000 0.924 0.000
#> GSM48653 5 0.172 0.6476 0.032 NA 0.000 0.000 0.932 0.004
#> GSM48662 5 0.243 0.6495 0.040 NA 0.004 0.000 0.900 0.012
#> GSM48663 5 0.396 0.5079 0.012 NA 0.000 0.000 0.668 0.004
#> GSM25524 3 0.578 0.5321 0.296 NA 0.588 0.000 0.016 0.060
#> GSM25525 1 0.392 0.3890 0.760 NA 0.196 0.000 0.012 0.028
#> GSM25526 1 0.717 0.3003 0.508 NA 0.088 0.032 0.292 0.036
#> GSM25527 1 0.389 0.5765 0.836 NA 0.040 0.040 0.048 0.020
#> GSM25528 3 0.594 0.4912 0.372 NA 0.516 0.000 0.016 0.060
#> GSM25529 1 0.398 0.3842 0.752 NA 0.200 0.000 0.016 0.032
#> GSM25530 3 0.591 0.5299 0.380 NA 0.524 0.016 0.016 0.032
#> GSM25531 3 0.587 0.4864 0.412 NA 0.496 0.012 0.016 0.036
#> GSM48661 5 0.432 0.6269 0.092 NA 0.084 0.000 0.784 0.020
#> GSM25561 1 0.663 -0.1274 0.476 NA 0.368 0.004 0.056 0.068
#> GSM25562 1 0.638 0.3336 0.608 NA 0.176 0.000 0.076 0.116
#> GSM25563 3 0.635 0.5817 0.272 NA 0.528 0.000 0.064 0.136
#> GSM25564 1 0.706 0.2863 0.492 NA 0.088 0.000 0.284 0.104
#> GSM25565 5 0.412 0.6513 0.060 NA 0.048 0.000 0.804 0.012
#> GSM25566 5 0.480 0.6115 0.136 NA 0.072 0.000 0.740 0.008
#> GSM25568 5 0.468 0.6134 0.012 NA 0.100 0.000 0.752 0.028
#> GSM25569 5 0.500 0.6165 0.028 NA 0.104 0.000 0.736 0.028
#> GSM25552 5 0.759 0.1881 0.340 NA 0.152 0.000 0.384 0.044
#> GSM25553 5 0.759 0.1760 0.344 NA 0.152 0.000 0.380 0.044
#> GSM25578 1 0.213 0.5505 0.920 NA 0.032 0.020 0.008 0.020
#> GSM25579 1 0.462 0.4757 0.744 NA 0.092 0.008 0.136 0.020
#> GSM25580 1 0.242 0.5648 0.900 NA 0.004 0.064 0.004 0.012
#> GSM25581 1 0.242 0.5648 0.900 NA 0.004 0.064 0.004 0.012
#> GSM48655 5 0.344 0.5914 0.020 NA 0.000 0.000 0.788 0.008
#> GSM48656 5 0.448 0.6254 0.092 NA 0.084 0.000 0.776 0.028
#> GSM48657 5 0.370 0.5544 0.016 NA 0.000 0.000 0.744 0.008
#> GSM48658 5 0.448 0.6254 0.092 NA 0.084 0.000 0.776 0.028
#> GSM25624 1 0.534 0.5322 0.740 NA 0.020 0.060 0.076 0.024
#> GSM25625 1 0.683 0.3207 0.596 NA 0.196 0.040 0.088 0.028
#> GSM25626 3 0.603 0.6279 0.268 NA 0.572 0.004 0.108 0.048
#> GSM25627 5 0.727 0.2416 0.292 NA 0.112 0.016 0.488 0.040
#> GSM25628 3 0.602 0.6262 0.256 NA 0.580 0.004 0.112 0.048
#> GSM25629 5 0.511 0.5856 0.160 NA 0.080 0.000 0.712 0.024
#> GSM25630 3 0.545 -0.5176 0.012 NA 0.500 0.004 0.000 0.072
#> GSM25631 5 0.703 0.3391 0.284 NA 0.152 0.000 0.480 0.032
#> GSM25632 1 0.597 -0.3617 0.472 NA 0.420 0.004 0.056 0.040
#> GSM25633 1 0.253 0.5628 0.904 NA 0.016 0.044 0.008 0.016
#> GSM25634 1 0.478 0.5120 0.784 NA 0.028 0.060 0.024 0.052
#> GSM25635 1 0.316 0.5826 0.876 NA 0.016 0.036 0.036 0.020
#> GSM25656 6 0.334 0.0000 0.000 NA 0.196 0.000 0.016 0.784
#> GSM25657 1 0.302 0.5323 0.872 NA 0.068 0.028 0.008 0.020
#> GSM25658 1 0.691 0.4168 0.596 NA 0.108 0.028 0.176 0.048
#> GSM25659 1 0.684 0.3025 0.524 NA 0.148 0.000 0.248 0.040
#> GSM25660 1 0.342 0.5836 0.864 NA 0.020 0.036 0.032 0.024
#> GSM25661 1 0.173 0.5622 0.940 NA 0.012 0.028 0.008 0.008
#> GSM25662 5 0.666 0.4430 0.244 NA 0.136 0.000 0.544 0.028
#> GSM25663 5 0.666 0.4430 0.244 NA 0.136 0.000 0.544 0.028
#> GSM25680 5 0.783 0.2642 0.264 NA 0.188 0.000 0.408 0.060
#> GSM25681 5 0.783 0.2642 0.264 NA 0.188 0.000 0.408 0.060
#> GSM25682 5 0.359 0.5990 0.032 NA 0.000 0.000 0.776 0.004
#> GSM25683 5 0.359 0.5990 0.032 NA 0.000 0.000 0.776 0.004
#> GSM25684 5 0.201 0.6507 0.040 NA 0.008 0.000 0.924 0.012
#> GSM25685 5 0.201 0.6507 0.040 NA 0.008 0.000 0.924 0.012
#> GSM25686 5 0.359 0.5990 0.032 NA 0.000 0.000 0.776 0.004
#> GSM25687 5 0.359 0.5990 0.032 NA 0.000 0.000 0.776 0.004
#> GSM48664 4 0.268 0.7772 0.132 NA 0.000 0.852 0.008 0.000
#> GSM48665 1 0.231 0.5659 0.912 NA 0.008 0.048 0.008 0.012
Heatmaps for the consensus matrix. It visualizes the probability of two samples to be in a same group.
consensus_heatmap(res, k = 2)
consensus_heatmap(res, k = 3)
consensus_heatmap(res, k = 4)
consensus_heatmap(res, k = 5)
consensus_heatmap(res, k = 6)
Heatmaps for the membership of samples in all partitions to see how consistent they are:
membership_heatmap(res, k = 2)
membership_heatmap(res, k = 3)
membership_heatmap(res, k = 4)
membership_heatmap(res, k = 5)
membership_heatmap(res, k = 6)
As soon as we have had the classes for columns, we can look for signatures which are significantly different between classes which can be candidate marks for certain classes. Following are the heatmaps for signatures.
Signature heatmaps where rows are scaled:
get_signatures(res, k = 2)
get_signatures(res, k = 3)
get_signatures(res, k = 4)
get_signatures(res, k = 5)
get_signatures(res, k = 6)
Signature heatmaps where rows are not scaled:
get_signatures(res, k = 2, scale_rows = FALSE)
get_signatures(res, k = 3, scale_rows = FALSE)
get_signatures(res, k = 4, scale_rows = FALSE)
get_signatures(res, k = 5, scale_rows = FALSE)
get_signatures(res, k = 6, scale_rows = FALSE)
Compare the overlap of signatures from different k:
compare_signatures(res)
get_signature() returns a data frame invisibly. TO get the list of signatures, the function
call should be assigned to a variable explicitly. In following code, if plot argument is set
to FALSE, no heatmap is plotted while only the differential analysis is performed.
# code only for demonstration
tb = get_signature(res, k = ..., plot = FALSE)
An example of the output of tb is:
#> which_row fdr mean_1 mean_2 scaled_mean_1 scaled_mean_2 km
#> 1 38 0.042760348 8.373488 9.131774 -0.5533452 0.5164555 1
#> 2 40 0.018707592 7.106213 8.469186 -0.6173731 0.5762149 1
#> 3 55 0.019134737 10.221463 11.207825 -0.6159697 0.5749050 1
#> 4 59 0.006059896 5.921854 7.869574 -0.6899429 0.6439467 1
#> 5 60 0.018055526 8.928898 10.211722 -0.6204761 0.5791110 1
#> 6 98 0.009384629 15.714769 14.887706 0.6635654 -0.6193277 2
...
The columns in tb are:
which_row: row indices corresponding to the input matrix.fdr: FDR for the differential test. mean_x: The mean value in group x.scaled_mean_x: The mean value in group x after rows are scaled.km: Row groups if k-means clustering is applied to rows.UMAP plot which shows how samples are separated.
dimension_reduction(res, k = 2, method = "UMAP")
dimension_reduction(res, k = 3, method = "UMAP")
dimension_reduction(res, k = 4, method = "UMAP")
dimension_reduction(res, k = 5, method = "UMAP")
dimension_reduction(res, k = 6, method = "UMAP")
Following heatmap shows how subgroups are split when increasing k:
collect_classes(res)
Test correlation between subgroups and known annotations. If the known annotation is numeric, one-way ANOVA test is applied, and if the known annotation is discrete, chi-squared contingency table test is applied.
test_to_known_factors(res)
#> n genotype/variation(p) k
#> SD:hclust 74 0.000108 2
#> SD:hclust 82 0.001313 3
#> SD:hclust 30 0.010635 4
#> SD:hclust 64 0.002647 5
#> SD:hclust 58 0.000639 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 100 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 0.807 0.889 0.949 0.5035 0.497 0.497
#> 3 3 0.547 0.651 0.792 0.2681 0.829 0.664
#> 4 4 0.561 0.739 0.800 0.1196 0.897 0.728
#> 5 5 0.584 0.482 0.693 0.0719 0.921 0.743
#> 6 6 0.605 0.444 0.650 0.0452 0.921 0.695
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
#> GSM25548 2 0.1414 0.964 0.020 0.980
#> GSM25549 2 0.1414 0.964 0.020 0.980
#> GSM25550 2 0.1414 0.964 0.020 0.980
#> GSM25551 2 0.0000 0.962 0.000 1.000
#> GSM25570 2 0.1414 0.964 0.020 0.980
#> GSM25571 2 0.1414 0.964 0.020 0.980
#> GSM25358 1 0.5737 0.837 0.864 0.136
#> GSM25359 2 0.3431 0.912 0.064 0.936
#> GSM25360 1 0.0000 0.930 1.000 0.000
#> GSM25361 1 0.6801 0.761 0.820 0.180
#> GSM25377 1 0.1414 0.919 0.980 0.020
#> GSM25378 1 0.4022 0.884 0.920 0.080
#> GSM25401 1 0.8608 0.648 0.716 0.284
#> GSM25402 1 0.6531 0.801 0.832 0.168
#> GSM25349 2 0.0000 0.962 0.000 1.000
#> GSM25350 2 0.0000 0.962 0.000 1.000
#> GSM25356 1 0.4690 0.868 0.900 0.100
#> GSM25357 2 0.0000 0.962 0.000 1.000
#> GSM25385 1 0.0000 0.930 1.000 0.000
#> GSM25386 1 0.0000 0.930 1.000 0.000
#> GSM25399 1 0.1414 0.919 0.980 0.020
#> GSM25400 1 0.0938 0.923 0.988 0.012
#> GSM48659 2 0.1414 0.964 0.020 0.980
#> GSM48660 2 0.0000 0.962 0.000 1.000
#> GSM25409 2 0.0000 0.962 0.000 1.000
#> GSM25410 1 0.0376 0.928 0.996 0.004
#> GSM25426 2 0.0000 0.962 0.000 1.000
#> GSM25427 1 0.4022 0.884 0.920 0.080
#> GSM25540 2 0.9552 0.399 0.376 0.624
#> GSM25541 1 0.9710 0.333 0.600 0.400
#> GSM25542 2 0.0376 0.963 0.004 0.996
#> GSM25543 2 0.0376 0.963 0.004 0.996
#> GSM25479 1 0.0000 0.930 1.000 0.000
#> GSM25480 1 0.0000 0.930 1.000 0.000
#> GSM25481 1 0.9732 0.402 0.596 0.404
#> GSM25482 1 0.9732 0.402 0.596 0.404
#> GSM48654 2 0.1414 0.964 0.020 0.980
#> GSM48650 2 0.0000 0.962 0.000 1.000
#> GSM48651 2 0.0938 0.964 0.012 0.988
#> GSM48652 2 0.1414 0.964 0.020 0.980
#> GSM48653 2 0.1414 0.964 0.020 0.980
#> GSM48662 2 0.1414 0.964 0.020 0.980
#> GSM48663 2 0.0000 0.962 0.000 1.000
#> GSM25524 1 0.0000 0.930 1.000 0.000
#> GSM25525 1 0.0000 0.930 1.000 0.000
#> GSM25526 1 0.0000 0.930 1.000 0.000
#> GSM25527 1 0.0000 0.930 1.000 0.000
#> GSM25528 1 0.0000 0.930 1.000 0.000
#> GSM25529 1 0.0000 0.930 1.000 0.000
#> GSM25530 1 0.0000 0.930 1.000 0.000
#> GSM25531 1 0.0000 0.930 1.000 0.000
#> GSM48661 2 0.1633 0.962 0.024 0.976
#> GSM25561 1 0.0000 0.930 1.000 0.000
#> GSM25562 1 0.0000 0.930 1.000 0.000
#> GSM25563 1 0.0000 0.930 1.000 0.000
#> GSM25564 1 0.8327 0.654 0.736 0.264
#> GSM25565 2 0.0000 0.962 0.000 1.000
#> GSM25566 2 0.0000 0.962 0.000 1.000
#> GSM25568 2 0.8955 0.526 0.312 0.688
#> GSM25569 2 0.1414 0.964 0.020 0.980
#> GSM25552 2 0.1414 0.964 0.020 0.980
#> GSM25553 2 0.4161 0.909 0.084 0.916
#> GSM25578 1 0.0000 0.930 1.000 0.000
#> GSM25579 1 0.0000 0.930 1.000 0.000
#> GSM25580 1 0.0000 0.930 1.000 0.000
#> GSM25581 1 0.0000 0.930 1.000 0.000
#> GSM48655 2 0.0000 0.962 0.000 1.000
#> GSM48656 2 0.1414 0.964 0.020 0.980
#> GSM48657 2 0.0000 0.962 0.000 1.000
#> GSM48658 2 0.1633 0.962 0.024 0.976
#> GSM25624 1 0.0000 0.930 1.000 0.000
#> GSM25625 1 0.0000 0.930 1.000 0.000
#> GSM25626 1 0.0000 0.930 1.000 0.000
#> GSM25627 1 0.9963 0.168 0.536 0.464
#> GSM25628 1 0.2948 0.895 0.948 0.052
#> GSM25629 2 0.7815 0.702 0.232 0.768
#> GSM25630 1 0.0000 0.930 1.000 0.000
#> GSM25631 2 0.4562 0.895 0.096 0.904
#> GSM25632 1 0.0000 0.930 1.000 0.000
#> GSM25633 1 0.0000 0.930 1.000 0.000
#> GSM25634 1 0.0000 0.930 1.000 0.000
#> GSM25635 1 0.0000 0.930 1.000 0.000
#> GSM25656 1 0.9732 0.326 0.596 0.404
#> GSM25657 1 0.0000 0.930 1.000 0.000
#> GSM25658 1 0.0000 0.930 1.000 0.000
#> GSM25659 1 0.0000 0.930 1.000 0.000
#> GSM25660 1 0.0000 0.930 1.000 0.000
#> GSM25661 1 0.0000 0.930 1.000 0.000
#> GSM25662 2 0.1414 0.964 0.020 0.980
#> GSM25663 2 0.1414 0.964 0.020 0.980
#> GSM25680 2 0.1414 0.964 0.020 0.980
#> GSM25681 2 0.1633 0.962 0.024 0.976
#> GSM25682 2 0.0000 0.962 0.000 1.000
#> GSM25683 2 0.0000 0.962 0.000 1.000
#> GSM25684 2 0.1414 0.964 0.020 0.980
#> GSM25685 2 0.1414 0.964 0.020 0.980
#> GSM25686 2 0.0000 0.962 0.000 1.000
#> GSM25687 2 0.0000 0.962 0.000 1.000
#> GSM48664 1 0.1414 0.919 0.980 0.020
#> GSM48665 1 0.1414 0.919 0.980 0.020
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM25548 2 0.2152 0.9196 0.036 0.948 0.016
#> GSM25549 2 0.2152 0.9196 0.036 0.948 0.016
#> GSM25550 2 0.2269 0.9201 0.040 0.944 0.016
#> GSM25551 2 0.1647 0.9241 0.036 0.960 0.004
#> GSM25570 2 0.2152 0.9196 0.036 0.948 0.016
#> GSM25571 2 0.2152 0.9196 0.036 0.948 0.016
#> GSM25358 1 0.7533 0.4011 0.668 0.088 0.244
#> GSM25359 2 0.7124 0.7258 0.088 0.708 0.204
#> GSM25360 3 0.0424 0.6559 0.008 0.000 0.992
#> GSM25361 3 0.4280 0.5830 0.020 0.124 0.856
#> GSM25377 1 0.3619 0.6451 0.864 0.000 0.136
#> GSM25378 1 0.3193 0.6331 0.896 0.004 0.100
#> GSM25401 1 0.6012 0.4466 0.788 0.088 0.124
#> GSM25402 1 0.4413 0.5606 0.852 0.024 0.124
#> GSM25349 2 0.3551 0.9001 0.132 0.868 0.000
#> GSM25350 2 0.3412 0.9034 0.124 0.876 0.000
#> GSM25356 1 0.2599 0.6034 0.932 0.016 0.052
#> GSM25357 2 0.4750 0.8311 0.216 0.784 0.000
#> GSM25385 3 0.2261 0.6280 0.068 0.000 0.932
#> GSM25386 3 0.0892 0.6536 0.020 0.000 0.980
#> GSM25399 1 0.4121 0.6472 0.832 0.000 0.168
#> GSM25400 1 0.4121 0.6472 0.832 0.000 0.168
#> GSM48659 2 0.1765 0.9198 0.040 0.956 0.004
#> GSM48660 2 0.3619 0.9021 0.136 0.864 0.000
#> GSM25409 2 0.3267 0.9064 0.116 0.884 0.000
#> GSM25410 3 0.1163 0.6505 0.028 0.000 0.972
#> GSM25426 2 0.3272 0.9106 0.104 0.892 0.004
#> GSM25427 1 0.3193 0.6331 0.896 0.004 0.100
#> GSM25540 3 0.5356 0.5198 0.020 0.196 0.784
#> GSM25541 3 0.4679 0.5666 0.020 0.148 0.832
#> GSM25542 2 0.5094 0.8521 0.056 0.832 0.112
#> GSM25543 2 0.6192 0.7783 0.060 0.764 0.176
#> GSM25479 1 0.6305 0.4472 0.516 0.000 0.484
#> GSM25480 1 0.6305 0.4472 0.516 0.000 0.484
#> GSM25481 1 0.3039 0.5855 0.920 0.036 0.044
#> GSM25482 1 0.3039 0.5855 0.920 0.036 0.044
#> GSM48654 2 0.1753 0.9201 0.048 0.952 0.000
#> GSM48650 2 0.3619 0.9021 0.136 0.864 0.000
#> GSM48651 2 0.1643 0.9214 0.044 0.956 0.000
#> GSM48652 2 0.1753 0.9209 0.048 0.952 0.000
#> GSM48653 2 0.1989 0.9201 0.048 0.948 0.004
#> GSM48662 2 0.1643 0.9219 0.044 0.956 0.000
#> GSM48663 2 0.3752 0.9006 0.144 0.856 0.000
#> GSM25524 3 0.0892 0.6526 0.020 0.000 0.980
#> GSM25525 3 0.6215 -0.2027 0.428 0.000 0.572
#> GSM25526 3 0.1643 0.6412 0.044 0.000 0.956
#> GSM25527 3 0.6235 -0.2374 0.436 0.000 0.564
#> GSM25528 3 0.4346 0.4834 0.184 0.000 0.816
#> GSM25529 3 0.6168 -0.1489 0.412 0.000 0.588
#> GSM25530 3 0.3941 0.5246 0.156 0.000 0.844
#> GSM25531 3 0.6168 -0.1491 0.412 0.000 0.588
#> GSM48661 2 0.2434 0.9166 0.036 0.940 0.024
#> GSM25561 3 0.1529 0.6445 0.040 0.000 0.960
#> GSM25562 3 0.6308 -0.4225 0.492 0.000 0.508
#> GSM25563 3 0.0592 0.6558 0.012 0.000 0.988
#> GSM25564 3 0.9588 0.0476 0.284 0.240 0.476
#> GSM25565 2 0.0892 0.9261 0.020 0.980 0.000
#> GSM25566 2 0.1411 0.9242 0.036 0.964 0.000
#> GSM25568 2 0.7076 0.6350 0.060 0.684 0.256
#> GSM25569 2 0.1643 0.9213 0.044 0.956 0.000
#> GSM25552 2 0.2152 0.9196 0.036 0.948 0.016
#> GSM25553 2 0.4137 0.8763 0.096 0.872 0.032
#> GSM25578 1 0.6309 0.4138 0.504 0.000 0.496
#> GSM25579 1 0.6305 0.4322 0.516 0.000 0.484
#> GSM25580 1 0.6215 0.5370 0.572 0.000 0.428
#> GSM25581 1 0.6260 0.5183 0.552 0.000 0.448
#> GSM48655 2 0.3267 0.9055 0.116 0.884 0.000
#> GSM48656 2 0.2063 0.9214 0.044 0.948 0.008
#> GSM48657 2 0.3619 0.9021 0.136 0.864 0.000
#> GSM48658 2 0.2152 0.9182 0.036 0.948 0.016
#> GSM25624 1 0.6235 0.5306 0.564 0.000 0.436
#> GSM25625 3 0.1031 0.6534 0.024 0.000 0.976
#> GSM25626 3 0.0592 0.6554 0.012 0.000 0.988
#> GSM25627 3 0.6800 0.4140 0.032 0.308 0.660
#> GSM25628 3 0.3752 0.6006 0.020 0.096 0.884
#> GSM25629 3 0.6717 0.3751 0.020 0.352 0.628
#> GSM25630 3 0.0424 0.6558 0.008 0.000 0.992
#> GSM25631 2 0.5292 0.7527 0.028 0.800 0.172
#> GSM25632 3 0.0892 0.6541 0.020 0.000 0.980
#> GSM25633 1 0.6280 0.4993 0.540 0.000 0.460
#> GSM25634 1 0.6274 0.5069 0.544 0.000 0.456
#> GSM25635 1 0.6267 0.5132 0.548 0.000 0.452
#> GSM25656 3 0.4679 0.5668 0.020 0.148 0.832
#> GSM25657 3 0.6286 -0.3310 0.464 0.000 0.536
#> GSM25658 3 0.4931 0.3885 0.232 0.000 0.768
#> GSM25659 3 0.6282 -0.0287 0.384 0.004 0.612
#> GSM25660 1 0.6274 0.5069 0.544 0.000 0.456
#> GSM25661 1 0.6235 0.5306 0.564 0.000 0.436
#> GSM25662 2 0.1129 0.9235 0.020 0.976 0.004
#> GSM25663 2 0.1491 0.9205 0.016 0.968 0.016
#> GSM25680 2 0.2176 0.9146 0.032 0.948 0.020
#> GSM25681 2 0.2689 0.9084 0.032 0.932 0.036
#> GSM25682 2 0.3192 0.9055 0.112 0.888 0.000
#> GSM25683 2 0.3192 0.9055 0.112 0.888 0.000
#> GSM25684 2 0.1129 0.9235 0.020 0.976 0.004
#> GSM25685 2 0.1453 0.9239 0.024 0.968 0.008
#> GSM25686 2 0.3192 0.9055 0.112 0.888 0.000
#> GSM25687 2 0.3192 0.9055 0.112 0.888 0.000
#> GSM48664 1 0.3816 0.6472 0.852 0.000 0.148
#> GSM48665 1 0.4178 0.6470 0.828 0.000 0.172
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM25548 2 0.3647 0.8158 0.000 0.852 0.040 0.108
#> GSM25549 2 0.3647 0.8158 0.000 0.852 0.040 0.108
#> GSM25550 2 0.3647 0.8158 0.000 0.852 0.040 0.108
#> GSM25551 2 0.3182 0.8387 0.000 0.876 0.028 0.096
#> GSM25570 2 0.3647 0.8158 0.000 0.852 0.040 0.108
#> GSM25571 2 0.3647 0.8158 0.000 0.852 0.040 0.108
#> GSM25358 4 0.8440 0.4515 0.144 0.124 0.176 0.556
#> GSM25359 2 0.7519 0.5684 0.016 0.564 0.232 0.188
#> GSM25360 3 0.3208 0.8250 0.148 0.000 0.848 0.004
#> GSM25361 3 0.6053 0.7361 0.076 0.100 0.748 0.076
#> GSM25377 4 0.5007 0.7620 0.356 0.000 0.008 0.636
#> GSM25378 4 0.4920 0.7702 0.368 0.000 0.004 0.628
#> GSM25401 4 0.5899 0.6358 0.104 0.080 0.060 0.756
#> GSM25402 4 0.5358 0.7337 0.208 0.012 0.044 0.736
#> GSM25349 2 0.5497 0.7296 0.000 0.672 0.044 0.284
#> GSM25350 2 0.5446 0.7380 0.000 0.680 0.044 0.276
#> GSM25356 4 0.4621 0.7838 0.284 0.000 0.008 0.708
#> GSM25357 2 0.5784 0.4765 0.000 0.556 0.032 0.412
#> GSM25385 3 0.3271 0.8272 0.132 0.000 0.856 0.012
#> GSM25386 3 0.3161 0.8289 0.124 0.000 0.864 0.012
#> GSM25399 4 0.5378 0.6542 0.448 0.000 0.012 0.540
#> GSM25400 4 0.5161 0.6278 0.476 0.000 0.004 0.520
#> GSM48659 2 0.3004 0.8372 0.000 0.892 0.048 0.060
#> GSM48660 2 0.4452 0.8197 0.000 0.796 0.048 0.156
#> GSM25409 2 0.4590 0.8048 0.000 0.772 0.036 0.192
#> GSM25410 3 0.3335 0.8289 0.120 0.000 0.860 0.020
#> GSM25426 2 0.5102 0.7940 0.000 0.748 0.064 0.188
#> GSM25427 4 0.4920 0.7702 0.368 0.000 0.004 0.628
#> GSM25540 3 0.5743 0.7139 0.044 0.136 0.756 0.064
#> GSM25541 3 0.6002 0.7178 0.060 0.132 0.744 0.064
#> GSM25542 2 0.6013 0.7130 0.000 0.684 0.196 0.120
#> GSM25543 2 0.7061 0.4735 0.000 0.540 0.312 0.148
#> GSM25479 1 0.0927 0.8098 0.976 0.000 0.016 0.008
#> GSM25480 1 0.1174 0.8083 0.968 0.000 0.020 0.012
#> GSM25481 4 0.4718 0.7782 0.272 0.008 0.004 0.716
#> GSM25482 4 0.4718 0.7782 0.272 0.008 0.004 0.716
#> GSM48654 2 0.2919 0.8360 0.000 0.896 0.044 0.060
#> GSM48650 2 0.5254 0.7817 0.000 0.724 0.056 0.220
#> GSM48651 2 0.3216 0.8336 0.000 0.880 0.044 0.076
#> GSM48652 2 0.3216 0.8336 0.000 0.880 0.044 0.076
#> GSM48653 2 0.3056 0.8346 0.000 0.888 0.040 0.072
#> GSM48662 2 0.2739 0.8376 0.000 0.904 0.036 0.060
#> GSM48663 2 0.5861 0.7179 0.000 0.644 0.060 0.296
#> GSM25524 3 0.5229 0.3756 0.428 0.000 0.564 0.008
#> GSM25525 1 0.1890 0.7988 0.936 0.000 0.056 0.008
#> GSM25526 3 0.5277 0.6585 0.304 0.000 0.668 0.028
#> GSM25527 1 0.1389 0.8069 0.952 0.000 0.048 0.000
#> GSM25528 1 0.4401 0.5287 0.724 0.000 0.272 0.004
#> GSM25529 1 0.2048 0.7935 0.928 0.000 0.064 0.008
#> GSM25530 1 0.5285 -0.0991 0.524 0.000 0.468 0.008
#> GSM25531 1 0.2124 0.7919 0.924 0.000 0.068 0.008
#> GSM48661 2 0.3474 0.8340 0.000 0.868 0.068 0.064
#> GSM25561 3 0.4826 0.7044 0.264 0.000 0.716 0.020
#> GSM25562 1 0.2224 0.8038 0.928 0.000 0.032 0.040
#> GSM25563 3 0.3606 0.8255 0.140 0.000 0.840 0.020
#> GSM25564 1 0.7679 0.2688 0.596 0.236 0.092 0.076
#> GSM25565 2 0.3453 0.8452 0.000 0.868 0.052 0.080
#> GSM25566 2 0.2546 0.8395 0.000 0.912 0.028 0.060
#> GSM25568 2 0.7984 0.4792 0.024 0.516 0.236 0.224
#> GSM25569 2 0.3978 0.8280 0.000 0.836 0.056 0.108
#> GSM25552 2 0.4271 0.8081 0.020 0.836 0.040 0.104
#> GSM25553 2 0.6538 0.6790 0.148 0.700 0.040 0.112
#> GSM25578 1 0.0927 0.8109 0.976 0.000 0.016 0.008
#> GSM25579 1 0.2578 0.7654 0.912 0.000 0.036 0.052
#> GSM25580 1 0.1209 0.7954 0.964 0.000 0.004 0.032
#> GSM25581 1 0.1004 0.8007 0.972 0.000 0.004 0.024
#> GSM48655 2 0.3931 0.8233 0.000 0.832 0.040 0.128
#> GSM48656 2 0.2908 0.8385 0.000 0.896 0.040 0.064
#> GSM48657 2 0.4532 0.8149 0.000 0.792 0.052 0.156
#> GSM48658 2 0.3716 0.8325 0.000 0.852 0.052 0.096
#> GSM25624 1 0.1824 0.7689 0.936 0.000 0.004 0.060
#> GSM25625 3 0.3450 0.8197 0.156 0.000 0.836 0.008
#> GSM25626 3 0.2999 0.8300 0.132 0.004 0.864 0.000
#> GSM25627 3 0.7093 0.5766 0.056 0.240 0.632 0.072
#> GSM25628 3 0.3037 0.8179 0.076 0.036 0.888 0.000
#> GSM25629 3 0.6634 0.5578 0.036 0.268 0.640 0.056
#> GSM25630 3 0.3763 0.8238 0.144 0.000 0.832 0.024
#> GSM25631 2 0.5588 0.7803 0.052 0.772 0.064 0.112
#> GSM25632 3 0.2921 0.8277 0.140 0.000 0.860 0.000
#> GSM25633 1 0.1109 0.7985 0.968 0.000 0.004 0.028
#> GSM25634 1 0.1489 0.7854 0.952 0.000 0.004 0.044
#> GSM25635 1 0.1398 0.7896 0.956 0.000 0.004 0.040
#> GSM25656 3 0.3801 0.7943 0.048 0.060 0.868 0.024
#> GSM25657 1 0.1388 0.8110 0.960 0.000 0.028 0.012
#> GSM25658 1 0.5756 0.1388 0.568 0.000 0.400 0.032
#> GSM25659 1 0.3237 0.7712 0.888 0.008 0.064 0.040
#> GSM25660 1 0.1004 0.8010 0.972 0.000 0.004 0.024
#> GSM25661 1 0.1109 0.7985 0.968 0.000 0.004 0.028
#> GSM25662 2 0.2500 0.8410 0.000 0.916 0.040 0.044
#> GSM25663 2 0.2915 0.8292 0.000 0.892 0.028 0.080
#> GSM25680 2 0.4041 0.8123 0.004 0.840 0.056 0.100
#> GSM25681 2 0.4102 0.8108 0.004 0.836 0.056 0.104
#> GSM25682 2 0.3581 0.8208 0.000 0.852 0.032 0.116
#> GSM25683 2 0.3581 0.8208 0.000 0.852 0.032 0.116
#> GSM25684 2 0.2500 0.8410 0.000 0.916 0.040 0.044
#> GSM25685 2 0.3164 0.8379 0.000 0.884 0.052 0.064
#> GSM25686 2 0.3581 0.8208 0.000 0.852 0.032 0.116
#> GSM25687 2 0.3581 0.8208 0.000 0.852 0.032 0.116
#> GSM48664 4 0.5378 0.6572 0.448 0.000 0.012 0.540
#> GSM48665 1 0.4511 0.2194 0.724 0.000 0.008 0.268
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM25548 2 0.0324 0.44879 0.000 0.992 0.000 0.004 0.004
#> GSM25549 2 0.0324 0.44879 0.000 0.992 0.000 0.004 0.004
#> GSM25550 2 0.0613 0.44376 0.004 0.984 0.000 0.004 0.008
#> GSM25551 2 0.5375 0.06067 0.000 0.568 0.008 0.044 0.380
#> GSM25570 2 0.0324 0.44879 0.000 0.992 0.000 0.004 0.004
#> GSM25571 2 0.0324 0.44879 0.000 0.992 0.000 0.004 0.004
#> GSM25358 4 0.7980 0.38358 0.040 0.248 0.104 0.508 0.100
#> GSM25359 2 0.6972 0.00776 0.000 0.576 0.176 0.076 0.172
#> GSM25360 3 0.2408 0.76501 0.092 0.000 0.892 0.000 0.016
#> GSM25361 3 0.6988 0.50336 0.056 0.356 0.500 0.012 0.076
#> GSM25377 4 0.4483 0.77315 0.156 0.000 0.012 0.768 0.064
#> GSM25378 4 0.4323 0.78157 0.200 0.000 0.004 0.752 0.044
#> GSM25401 4 0.4643 0.63069 0.016 0.000 0.032 0.724 0.228
#> GSM25402 4 0.4741 0.71088 0.048 0.000 0.028 0.752 0.172
#> GSM25349 5 0.6720 0.36732 0.000 0.404 0.020 0.140 0.436
#> GSM25350 5 0.6693 0.36227 0.000 0.408 0.020 0.136 0.436
#> GSM25356 4 0.3915 0.79975 0.112 0.008 0.004 0.820 0.056
#> GSM25357 5 0.6779 0.27835 0.000 0.360 0.000 0.276 0.364
#> GSM25385 3 0.1990 0.76920 0.068 0.000 0.920 0.008 0.004
#> GSM25386 3 0.1788 0.77028 0.056 0.000 0.932 0.008 0.004
#> GSM25399 4 0.4925 0.74795 0.180 0.000 0.016 0.732 0.072
#> GSM25400 4 0.3835 0.70416 0.260 0.000 0.008 0.732 0.000
#> GSM48659 2 0.4898 0.31431 0.000 0.592 0.032 0.000 0.376
#> GSM48660 5 0.5149 0.15540 0.000 0.424 0.004 0.032 0.540
#> GSM25409 2 0.4086 0.25624 0.000 0.788 0.004 0.056 0.152
#> GSM25410 3 0.1901 0.76926 0.056 0.000 0.928 0.012 0.004
#> GSM25426 5 0.6110 0.21332 0.000 0.388 0.028 0.064 0.520
#> GSM25427 4 0.4032 0.78290 0.192 0.000 0.004 0.772 0.032
#> GSM25540 3 0.6669 0.49209 0.020 0.340 0.528 0.016 0.096
#> GSM25541 3 0.6748 0.49323 0.024 0.340 0.524 0.016 0.096
#> GSM25542 2 0.7269 -0.11132 0.000 0.436 0.136 0.060 0.368
#> GSM25543 5 0.7766 0.13891 0.000 0.352 0.232 0.064 0.352
#> GSM25479 1 0.1041 0.79642 0.964 0.000 0.004 0.032 0.000
#> GSM25480 1 0.1116 0.79663 0.964 0.004 0.004 0.028 0.000
#> GSM25481 4 0.4929 0.78674 0.132 0.004 0.008 0.744 0.112
#> GSM25482 4 0.4929 0.78674 0.132 0.004 0.008 0.744 0.112
#> GSM48654 2 0.4902 0.27910 0.000 0.564 0.028 0.000 0.408
#> GSM48650 5 0.5506 0.33276 0.000 0.344 0.004 0.068 0.584
#> GSM48651 2 0.4900 0.17801 0.000 0.512 0.024 0.000 0.464
#> GSM48652 2 0.4897 0.18644 0.000 0.516 0.024 0.000 0.460
#> GSM48653 2 0.4897 0.19667 0.000 0.516 0.024 0.000 0.460
#> GSM48662 2 0.4709 0.29987 0.000 0.612 0.024 0.000 0.364
#> GSM48663 5 0.6099 0.42386 0.000 0.256 0.012 0.136 0.596
#> GSM25524 1 0.6030 0.10686 0.516 0.000 0.392 0.016 0.076
#> GSM25525 1 0.1731 0.78594 0.940 0.000 0.012 0.008 0.040
#> GSM25526 3 0.6278 0.40287 0.344 0.000 0.544 0.032 0.080
#> GSM25527 1 0.1804 0.79597 0.940 0.000 0.024 0.024 0.012
#> GSM25528 1 0.4623 0.65773 0.764 0.000 0.148 0.016 0.072
#> GSM25529 1 0.1808 0.78539 0.936 0.000 0.012 0.008 0.044
#> GSM25530 1 0.5553 0.46402 0.648 0.000 0.264 0.020 0.068
#> GSM25531 1 0.2537 0.77443 0.904 0.000 0.024 0.016 0.056
#> GSM48661 2 0.5171 0.34178 0.000 0.616 0.040 0.008 0.336
#> GSM25561 3 0.5328 0.54654 0.256 0.000 0.660 0.008 0.076
#> GSM25562 1 0.3439 0.75705 0.844 0.000 0.004 0.092 0.060
#> GSM25563 3 0.3568 0.74822 0.080 0.000 0.844 0.012 0.064
#> GSM25564 1 0.7620 0.26197 0.504 0.264 0.056 0.020 0.156
#> GSM25565 2 0.5085 0.24288 0.000 0.632 0.012 0.032 0.324
#> GSM25566 2 0.4576 0.27523 0.000 0.712 0.008 0.032 0.248
#> GSM25568 5 0.8218 0.12886 0.008 0.348 0.180 0.108 0.356
#> GSM25569 2 0.6031 0.13134 0.000 0.552 0.044 0.044 0.360
#> GSM25552 2 0.1074 0.43743 0.016 0.968 0.000 0.004 0.012
#> GSM25553 2 0.2395 0.36453 0.072 0.904 0.000 0.008 0.016
#> GSM25578 1 0.1588 0.79642 0.948 0.000 0.008 0.028 0.016
#> GSM25579 1 0.2297 0.77920 0.912 0.060 0.008 0.020 0.000
#> GSM25580 1 0.2230 0.77044 0.884 0.000 0.000 0.116 0.000
#> GSM25581 1 0.2230 0.77044 0.884 0.000 0.000 0.116 0.000
#> GSM48655 2 0.5109 -0.14466 0.000 0.504 0.000 0.036 0.460
#> GSM48656 2 0.4742 0.34808 0.000 0.648 0.020 0.008 0.324
#> GSM48657 5 0.5077 0.18482 0.000 0.428 0.000 0.036 0.536
#> GSM48658 2 0.4622 0.37268 0.000 0.696 0.028 0.008 0.268
#> GSM25624 1 0.2852 0.72597 0.828 0.000 0.000 0.172 0.000
#> GSM25625 3 0.3446 0.75770 0.112 0.000 0.844 0.016 0.028
#> GSM25626 3 0.1914 0.77028 0.056 0.000 0.928 0.008 0.008
#> GSM25627 3 0.7422 0.41845 0.032 0.120 0.528 0.044 0.276
#> GSM25628 3 0.1996 0.76089 0.032 0.004 0.928 0.000 0.036
#> GSM25629 3 0.7503 0.42233 0.032 0.184 0.524 0.032 0.228
#> GSM25630 3 0.4244 0.72801 0.100 0.000 0.800 0.016 0.084
#> GSM25631 2 0.3105 0.39798 0.016 0.872 0.020 0.004 0.088
#> GSM25632 3 0.2403 0.76630 0.072 0.000 0.904 0.012 0.012
#> GSM25633 1 0.1671 0.78808 0.924 0.000 0.000 0.076 0.000
#> GSM25634 1 0.2561 0.75012 0.856 0.000 0.000 0.144 0.000
#> GSM25635 1 0.2516 0.75299 0.860 0.000 0.000 0.140 0.000
#> GSM25656 3 0.3729 0.72960 0.012 0.024 0.844 0.024 0.096
#> GSM25657 1 0.2568 0.78473 0.904 0.000 0.016 0.048 0.032
#> GSM25658 1 0.6672 0.14609 0.516 0.000 0.344 0.048 0.092
#> GSM25659 1 0.3748 0.74552 0.848 0.056 0.012 0.016 0.068
#> GSM25660 1 0.1952 0.78500 0.912 0.004 0.000 0.084 0.000
#> GSM25661 1 0.1851 0.78385 0.912 0.000 0.000 0.088 0.000
#> GSM25662 2 0.5052 0.29297 0.000 0.600 0.028 0.008 0.364
#> GSM25663 2 0.2899 0.44874 0.000 0.872 0.028 0.004 0.096
#> GSM25680 2 0.1843 0.43958 0.000 0.932 0.008 0.008 0.052
#> GSM25681 2 0.1731 0.43635 0.000 0.940 0.012 0.008 0.040
#> GSM25682 2 0.5058 -0.00561 0.000 0.576 0.000 0.040 0.384
#> GSM25683 2 0.5068 -0.01498 0.000 0.572 0.000 0.040 0.388
#> GSM25684 2 0.5064 0.28859 0.000 0.596 0.028 0.008 0.368
#> GSM25685 2 0.5886 0.12558 0.000 0.500 0.044 0.028 0.428
#> GSM25686 2 0.5058 -0.00561 0.000 0.576 0.000 0.040 0.384
#> GSM25687 2 0.5058 -0.00561 0.000 0.576 0.000 0.040 0.384
#> GSM48664 4 0.4743 0.74882 0.184 0.000 0.012 0.740 0.064
#> GSM48665 1 0.4804 0.32384 0.612 0.000 0.016 0.364 0.008
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM25548 5 0.0000 0.5026 0.000 0.000 0.000 0.000 1.000 NA
#> GSM25549 5 0.0000 0.5026 0.000 0.000 0.000 0.000 1.000 NA
#> GSM25550 5 0.0405 0.5009 0.000 0.008 0.000 0.000 0.988 NA
#> GSM25551 5 0.6412 0.0488 0.000 0.304 0.012 0.024 0.496 NA
#> GSM25570 5 0.0146 0.5028 0.000 0.004 0.000 0.000 0.996 NA
#> GSM25571 5 0.0146 0.5028 0.000 0.004 0.000 0.000 0.996 NA
#> GSM25358 4 0.7883 0.3897 0.032 0.068 0.084 0.456 0.284 NA
#> GSM25359 5 0.6922 0.2458 0.000 0.132 0.124 0.044 0.576 NA
#> GSM25360 3 0.2655 0.7061 0.060 0.000 0.876 0.004 0.000 NA
#> GSM25361 5 0.7521 -0.3211 0.076 0.036 0.352 0.004 0.392 NA
#> GSM25377 4 0.4731 0.7214 0.072 0.016 0.008 0.716 0.000 NA
#> GSM25378 4 0.4230 0.7224 0.164 0.020 0.004 0.768 0.008 NA
#> GSM25401 4 0.5694 0.6203 0.004 0.168 0.032 0.644 0.004 NA
#> GSM25402 4 0.4931 0.6799 0.008 0.124 0.028 0.732 0.004 NA
#> GSM25349 2 0.6527 0.2921 0.000 0.500 0.000 0.084 0.296 NA
#> GSM25350 2 0.6468 0.2900 0.000 0.500 0.000 0.076 0.304 NA
#> GSM25356 4 0.4116 0.7614 0.068 0.048 0.004 0.804 0.004 NA
#> GSM25357 2 0.7348 0.1487 0.000 0.360 0.000 0.300 0.220 NA
#> GSM25385 3 0.1007 0.7280 0.016 0.004 0.968 0.008 0.000 NA
#> GSM25386 3 0.0881 0.7289 0.012 0.000 0.972 0.008 0.000 NA
#> GSM25399 4 0.5086 0.6920 0.084 0.016 0.004 0.660 0.000 NA
#> GSM25400 4 0.3883 0.6760 0.200 0.004 0.000 0.752 0.000 NA
#> GSM48659 2 0.5170 0.2411 0.000 0.484 0.008 0.000 0.444 NA
#> GSM48660 2 0.4281 0.4125 0.000 0.688 0.000 0.016 0.272 NA
#> GSM25409 5 0.4200 0.3384 0.000 0.192 0.000 0.020 0.744 NA
#> GSM25410 3 0.0881 0.7289 0.012 0.000 0.972 0.008 0.000 NA
#> GSM25426 2 0.6777 0.1975 0.000 0.488 0.028 0.028 0.276 NA
#> GSM25427 4 0.3918 0.7164 0.160 0.016 0.000 0.776 0.000 NA
#> GSM25540 3 0.6535 0.2738 0.000 0.056 0.432 0.000 0.368 NA
#> GSM25541 3 0.6765 0.2461 0.008 0.056 0.412 0.000 0.380 NA
#> GSM25542 2 0.7171 0.2303 0.000 0.428 0.104 0.004 0.288 NA
#> GSM25543 2 0.7695 0.1925 0.000 0.396 0.188 0.012 0.220 NA
#> GSM25479 1 0.0964 0.7484 0.968 0.000 0.000 0.012 0.004 NA
#> GSM25480 1 0.1078 0.7487 0.964 0.000 0.000 0.012 0.008 NA
#> GSM25481 4 0.5414 0.7419 0.112 0.076 0.004 0.716 0.020 NA
#> GSM25482 4 0.5414 0.7419 0.112 0.076 0.004 0.716 0.020 NA
#> GSM48654 2 0.4945 0.3055 0.000 0.528 0.004 0.000 0.412 NA
#> GSM48650 2 0.4859 0.3899 0.000 0.692 0.000 0.024 0.204 NA
#> GSM48651 2 0.4435 0.3569 0.000 0.580 0.004 0.000 0.392 NA
#> GSM48652 2 0.4426 0.3570 0.000 0.584 0.004 0.000 0.388 NA
#> GSM48653 2 0.4795 0.3363 0.000 0.560 0.008 0.000 0.392 NA
#> GSM48662 2 0.4760 0.3142 0.000 0.520 0.004 0.000 0.436 NA
#> GSM48663 2 0.5574 0.3974 0.000 0.660 0.000 0.068 0.148 NA
#> GSM25524 1 0.5824 0.2729 0.500 0.004 0.312 0.000 0.000 NA
#> GSM25525 1 0.2595 0.7150 0.836 0.000 0.004 0.000 0.000 NA
#> GSM25526 3 0.7442 0.1193 0.336 0.056 0.400 0.048 0.000 NA
#> GSM25527 1 0.2458 0.7469 0.900 0.008 0.028 0.012 0.000 NA
#> GSM25528 1 0.4432 0.6388 0.708 0.000 0.104 0.000 0.000 NA
#> GSM25529 1 0.2706 0.7136 0.832 0.000 0.008 0.000 0.000 NA
#> GSM25530 1 0.5451 0.5332 0.616 0.000 0.176 0.012 0.000 NA
#> GSM25531 1 0.3309 0.6983 0.788 0.000 0.004 0.016 0.000 NA
#> GSM48661 5 0.5850 -0.2103 0.000 0.436 0.028 0.000 0.440 NA
#> GSM25561 3 0.5574 0.5493 0.192 0.028 0.652 0.012 0.000 NA
#> GSM25562 1 0.4895 0.6666 0.728 0.040 0.004 0.112 0.000 NA
#> GSM25563 3 0.3386 0.7065 0.040 0.032 0.848 0.008 0.000 NA
#> GSM25564 1 0.8311 0.2176 0.432 0.176 0.052 0.032 0.208 NA
#> GSM25565 5 0.5101 -0.0362 0.000 0.396 0.004 0.008 0.540 NA
#> GSM25566 5 0.4680 0.2173 0.000 0.280 0.004 0.004 0.656 NA
#> GSM25568 2 0.8175 0.2176 0.004 0.400 0.116 0.064 0.232 NA
#> GSM25569 2 0.5860 0.3027 0.000 0.472 0.004 0.008 0.384 NA
#> GSM25552 5 0.0976 0.4956 0.008 0.008 0.000 0.000 0.968 NA
#> GSM25553 5 0.2063 0.4633 0.044 0.008 0.000 0.008 0.920 NA
#> GSM25578 1 0.1116 0.7479 0.960 0.000 0.004 0.008 0.000 NA
#> GSM25579 1 0.2568 0.7226 0.876 0.000 0.000 0.012 0.096 NA
#> GSM25580 1 0.3344 0.7134 0.828 0.000 0.008 0.104 0.000 NA
#> GSM25581 1 0.3249 0.7174 0.836 0.000 0.008 0.096 0.000 NA
#> GSM48655 2 0.5117 0.1893 0.000 0.532 0.000 0.012 0.400 NA
#> GSM48656 5 0.5174 -0.2419 0.000 0.460 0.008 0.000 0.468 NA
#> GSM48657 2 0.4797 0.3521 0.000 0.648 0.000 0.012 0.280 NA
#> GSM48658 5 0.5507 -0.0230 0.000 0.356 0.012 0.000 0.532 NA
#> GSM25624 1 0.4284 0.6274 0.728 0.000 0.008 0.200 0.000 NA
#> GSM25625 3 0.3257 0.7025 0.064 0.020 0.860 0.024 0.000 NA
#> GSM25626 3 0.0551 0.7301 0.008 0.004 0.984 0.004 0.000 NA
#> GSM25627 3 0.7974 0.2798 0.004 0.256 0.400 0.056 0.084 NA
#> GSM25628 3 0.1425 0.7288 0.008 0.012 0.952 0.000 0.008 NA
#> GSM25629 3 0.7997 0.2524 0.004 0.196 0.388 0.024 0.176 NA
#> GSM25630 3 0.3714 0.6856 0.044 0.020 0.816 0.008 0.000 NA
#> GSM25631 5 0.3407 0.4211 0.004 0.068 0.008 0.000 0.832 NA
#> GSM25632 3 0.1218 0.7267 0.028 0.000 0.956 0.004 0.000 NA
#> GSM25633 1 0.3140 0.7199 0.844 0.000 0.008 0.092 0.000 NA
#> GSM25634 1 0.3686 0.6904 0.796 0.000 0.008 0.136 0.000 NA
#> GSM25635 1 0.3686 0.6904 0.796 0.000 0.008 0.136 0.000 NA
#> GSM25656 3 0.4660 0.6555 0.004 0.056 0.728 0.012 0.012 NA
#> GSM25657 1 0.2126 0.7434 0.904 0.000 0.004 0.020 0.000 NA
#> GSM25658 1 0.7626 0.1444 0.428 0.056 0.264 0.064 0.000 NA
#> GSM25659 1 0.4886 0.6750 0.748 0.044 0.008 0.020 0.040 NA
#> GSM25660 1 0.3258 0.7204 0.836 0.000 0.008 0.092 0.000 NA
#> GSM25661 1 0.3190 0.7210 0.844 0.000 0.012 0.088 0.000 NA
#> GSM25662 5 0.5659 -0.0802 0.000 0.388 0.012 0.004 0.500 NA
#> GSM25663 5 0.3517 0.4320 0.000 0.096 0.012 0.000 0.820 NA
#> GSM25680 5 0.2044 0.4823 0.000 0.040 0.008 0.004 0.920 NA
#> GSM25681 5 0.1819 0.4865 0.000 0.024 0.008 0.004 0.932 NA
#> GSM25682 5 0.5441 0.0354 0.000 0.412 0.004 0.016 0.504 NA
#> GSM25683 5 0.5441 0.0354 0.000 0.412 0.004 0.016 0.504 NA
#> GSM25684 5 0.5696 -0.0849 0.000 0.388 0.012 0.004 0.496 NA
#> GSM25685 2 0.6741 0.1238 0.000 0.396 0.024 0.020 0.384 NA
#> GSM25686 5 0.5441 0.0354 0.000 0.412 0.004 0.016 0.504 NA
#> GSM25687 5 0.5441 0.0354 0.000 0.412 0.004 0.016 0.504 NA
#> GSM48664 4 0.5043 0.6906 0.092 0.012 0.004 0.664 0.000 NA
#> GSM48665 1 0.5194 0.4395 0.612 0.000 0.012 0.284 0.000 NA
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 genotype/variation(p) k
#> SD:kmeans 94 3.38e-05 2
#> SD:kmeans 81 1.87e-03 3
#> SD:kmeans 91 1.35e-06 4
#> SD:kmeans 44 6.46e-03 5
#> SD:kmeans 49 5.34e-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", "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 100 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 0.801 0.897 0.956 0.5052 0.495 0.495
#> 3 3 0.562 0.730 0.831 0.3057 0.800 0.614
#> 4 4 0.457 0.476 0.686 0.1263 0.877 0.671
#> 5 5 0.489 0.433 0.643 0.0696 0.894 0.652
#> 6 6 0.503 0.324 0.577 0.0420 0.898 0.597
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
#> GSM25548 2 0.0000 0.956 0.000 1.000
#> GSM25549 2 0.0000 0.956 0.000 1.000
#> GSM25550 2 0.0000 0.956 0.000 1.000
#> GSM25551 2 0.0000 0.956 0.000 1.000
#> GSM25570 2 0.0000 0.956 0.000 1.000
#> GSM25571 2 0.0000 0.956 0.000 1.000
#> GSM25358 1 0.5294 0.852 0.880 0.120
#> GSM25359 2 0.3733 0.893 0.072 0.928
#> GSM25360 1 0.0000 0.949 1.000 0.000
#> GSM25361 1 0.9795 0.257 0.584 0.416
#> GSM25377 1 0.0000 0.949 1.000 0.000
#> GSM25378 1 0.3114 0.910 0.944 0.056
#> GSM25401 1 0.6343 0.808 0.840 0.160
#> GSM25402 1 0.4161 0.885 0.916 0.084
#> GSM25349 2 0.0000 0.956 0.000 1.000
#> GSM25350 2 0.0000 0.956 0.000 1.000
#> GSM25356 1 0.4690 0.872 0.900 0.100
#> GSM25357 2 0.0000 0.956 0.000 1.000
#> GSM25385 1 0.0000 0.949 1.000 0.000
#> GSM25386 1 0.0000 0.949 1.000 0.000
#> GSM25399 1 0.0000 0.949 1.000 0.000
#> GSM25400 1 0.0000 0.949 1.000 0.000
#> GSM48659 2 0.0000 0.956 0.000 1.000
#> GSM48660 2 0.0000 0.956 0.000 1.000
#> GSM25409 2 0.0000 0.956 0.000 1.000
#> GSM25410 1 0.0000 0.949 1.000 0.000
#> GSM25426 2 0.0000 0.956 0.000 1.000
#> GSM25427 1 0.3733 0.896 0.928 0.072
#> GSM25540 2 0.8555 0.620 0.280 0.720
#> GSM25541 2 0.9522 0.429 0.372 0.628
#> GSM25542 2 0.0000 0.956 0.000 1.000
#> GSM25543 2 0.0000 0.956 0.000 1.000
#> GSM25479 1 0.0000 0.949 1.000 0.000
#> GSM25480 1 0.0000 0.949 1.000 0.000
#> GSM25481 1 0.9044 0.561 0.680 0.320
#> GSM25482 1 0.8861 0.592 0.696 0.304
#> GSM48654 2 0.0000 0.956 0.000 1.000
#> GSM48650 2 0.0000 0.956 0.000 1.000
#> GSM48651 2 0.0000 0.956 0.000 1.000
#> GSM48652 2 0.0000 0.956 0.000 1.000
#> GSM48653 2 0.0000 0.956 0.000 1.000
#> GSM48662 2 0.0000 0.956 0.000 1.000
#> GSM48663 2 0.0000 0.956 0.000 1.000
#> GSM25524 1 0.0000 0.949 1.000 0.000
#> GSM25525 1 0.0000 0.949 1.000 0.000
#> GSM25526 1 0.0000 0.949 1.000 0.000
#> GSM25527 1 0.0000 0.949 1.000 0.000
#> GSM25528 1 0.0000 0.949 1.000 0.000
#> GSM25529 1 0.0000 0.949 1.000 0.000
#> GSM25530 1 0.0000 0.949 1.000 0.000
#> GSM25531 1 0.0000 0.949 1.000 0.000
#> GSM48661 2 0.0000 0.956 0.000 1.000
#> GSM25561 1 0.0000 0.949 1.000 0.000
#> GSM25562 1 0.0000 0.949 1.000 0.000
#> GSM25563 1 0.0000 0.949 1.000 0.000
#> GSM25564 1 0.8661 0.615 0.712 0.288
#> GSM25565 2 0.0000 0.956 0.000 1.000
#> GSM25566 2 0.0000 0.956 0.000 1.000
#> GSM25568 2 0.9580 0.345 0.380 0.620
#> GSM25569 2 0.0000 0.956 0.000 1.000
#> GSM25552 2 0.0000 0.956 0.000 1.000
#> GSM25553 2 0.3274 0.906 0.060 0.940
#> GSM25578 1 0.0000 0.949 1.000 0.000
#> GSM25579 1 0.0000 0.949 1.000 0.000
#> GSM25580 1 0.0000 0.949 1.000 0.000
#> GSM25581 1 0.0000 0.949 1.000 0.000
#> GSM48655 2 0.0000 0.956 0.000 1.000
#> GSM48656 2 0.0000 0.956 0.000 1.000
#> GSM48657 2 0.0000 0.956 0.000 1.000
#> GSM48658 2 0.0000 0.956 0.000 1.000
#> GSM25624 1 0.0000 0.949 1.000 0.000
#> GSM25625 1 0.0000 0.949 1.000 0.000
#> GSM25626 1 0.0000 0.949 1.000 0.000
#> GSM25627 1 0.9427 0.448 0.640 0.360
#> GSM25628 1 0.5294 0.844 0.880 0.120
#> GSM25629 2 0.7602 0.714 0.220 0.780
#> GSM25630 1 0.0000 0.949 1.000 0.000
#> GSM25631 2 0.5059 0.854 0.112 0.888
#> GSM25632 1 0.0000 0.949 1.000 0.000
#> GSM25633 1 0.0000 0.949 1.000 0.000
#> GSM25634 1 0.0000 0.949 1.000 0.000
#> GSM25635 1 0.0000 0.949 1.000 0.000
#> GSM25656 2 0.9954 0.173 0.460 0.540
#> GSM25657 1 0.0000 0.949 1.000 0.000
#> GSM25658 1 0.0000 0.949 1.000 0.000
#> GSM25659 1 0.0000 0.949 1.000 0.000
#> GSM25660 1 0.0000 0.949 1.000 0.000
#> GSM25661 1 0.0000 0.949 1.000 0.000
#> GSM25662 2 0.0000 0.956 0.000 1.000
#> GSM25663 2 0.0000 0.956 0.000 1.000
#> GSM25680 2 0.0000 0.956 0.000 1.000
#> GSM25681 2 0.0672 0.950 0.008 0.992
#> GSM25682 2 0.0000 0.956 0.000 1.000
#> GSM25683 2 0.0000 0.956 0.000 1.000
#> GSM25684 2 0.0000 0.956 0.000 1.000
#> GSM25685 2 0.0000 0.956 0.000 1.000
#> GSM25686 2 0.0000 0.956 0.000 1.000
#> GSM25687 2 0.0000 0.956 0.000 1.000
#> GSM48664 1 0.0000 0.949 1.000 0.000
#> GSM48665 1 0.0000 0.949 1.000 0.000
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM25548 2 0.2537 0.888152 0.000 0.920 0.080
#> GSM25549 2 0.2537 0.888701 0.000 0.920 0.080
#> GSM25550 2 0.3583 0.875322 0.044 0.900 0.056
#> GSM25551 2 0.2116 0.888453 0.012 0.948 0.040
#> GSM25570 2 0.2356 0.888317 0.000 0.928 0.072
#> GSM25571 2 0.2537 0.887987 0.000 0.920 0.080
#> GSM25358 3 0.9071 0.074373 0.432 0.136 0.432
#> GSM25359 3 0.7505 0.287864 0.044 0.384 0.572
#> GSM25360 3 0.3686 0.749067 0.140 0.000 0.860
#> GSM25361 3 0.2318 0.743109 0.028 0.028 0.944
#> GSM25377 1 0.2152 0.785457 0.948 0.036 0.016
#> GSM25378 1 0.3805 0.741807 0.884 0.092 0.024
#> GSM25401 1 0.9713 0.086359 0.444 0.240 0.316
#> GSM25402 1 0.6526 0.663104 0.760 0.112 0.128
#> GSM25349 2 0.2527 0.865263 0.044 0.936 0.020
#> GSM25350 2 0.2176 0.871593 0.032 0.948 0.020
#> GSM25356 1 0.3910 0.731820 0.876 0.104 0.020
#> GSM25357 2 0.3406 0.843994 0.068 0.904 0.028
#> GSM25385 3 0.5327 0.664734 0.272 0.000 0.728
#> GSM25386 3 0.3619 0.754211 0.136 0.000 0.864
#> GSM25399 1 0.0661 0.808120 0.988 0.004 0.008
#> GSM25400 1 0.1585 0.809276 0.964 0.008 0.028
#> GSM48659 2 0.3267 0.880719 0.000 0.884 0.116
#> GSM48660 2 0.1905 0.875825 0.028 0.956 0.016
#> GSM25409 2 0.2313 0.877919 0.032 0.944 0.024
#> GSM25410 3 0.4683 0.738320 0.140 0.024 0.836
#> GSM25426 2 0.2879 0.878419 0.024 0.924 0.052
#> GSM25427 1 0.3502 0.748724 0.896 0.084 0.020
#> GSM25540 3 0.1647 0.731013 0.004 0.036 0.960
#> GSM25541 3 0.1781 0.742921 0.020 0.020 0.960
#> GSM25542 2 0.6713 0.403690 0.012 0.572 0.416
#> GSM25543 3 0.6937 0.162261 0.020 0.404 0.576
#> GSM25479 1 0.2796 0.806000 0.908 0.000 0.092
#> GSM25480 1 0.2448 0.811884 0.924 0.000 0.076
#> GSM25481 1 0.4280 0.713798 0.856 0.124 0.020
#> GSM25482 1 0.4413 0.710139 0.852 0.124 0.024
#> GSM48654 2 0.3192 0.881250 0.000 0.888 0.112
#> GSM48650 2 0.2050 0.873370 0.028 0.952 0.020
#> GSM48651 2 0.2959 0.884579 0.000 0.900 0.100
#> GSM48652 2 0.3192 0.881250 0.000 0.888 0.112
#> GSM48653 2 0.3192 0.881250 0.000 0.888 0.112
#> GSM48662 2 0.3192 0.881890 0.000 0.888 0.112
#> GSM48663 2 0.2527 0.864840 0.044 0.936 0.020
#> GSM25524 3 0.5706 0.554887 0.320 0.000 0.680
#> GSM25525 1 0.4002 0.763116 0.840 0.000 0.160
#> GSM25526 3 0.4346 0.731019 0.184 0.000 0.816
#> GSM25527 1 0.3879 0.773353 0.848 0.000 0.152
#> GSM25528 1 0.6225 0.228286 0.568 0.000 0.432
#> GSM25529 1 0.4796 0.695741 0.780 0.000 0.220
#> GSM25530 1 0.6309 -0.028407 0.504 0.000 0.496
#> GSM25531 1 0.4842 0.693795 0.776 0.000 0.224
#> GSM48661 2 0.4702 0.818181 0.000 0.788 0.212
#> GSM25561 3 0.5835 0.541458 0.340 0.000 0.660
#> GSM25562 1 0.4121 0.752884 0.832 0.000 0.168
#> GSM25563 3 0.4121 0.742168 0.168 0.000 0.832
#> GSM25564 1 0.9744 0.000843 0.428 0.236 0.336
#> GSM25565 2 0.2599 0.886828 0.016 0.932 0.052
#> GSM25566 2 0.1905 0.884260 0.016 0.956 0.028
#> GSM25568 3 0.9355 0.246418 0.188 0.320 0.492
#> GSM25569 2 0.3340 0.879687 0.000 0.880 0.120
#> GSM25552 2 0.3678 0.881213 0.028 0.892 0.080
#> GSM25553 2 0.8587 0.250119 0.400 0.500 0.100
#> GSM25578 1 0.2448 0.811591 0.924 0.000 0.076
#> GSM25579 1 0.3784 0.795234 0.864 0.004 0.132
#> GSM25580 1 0.1529 0.816321 0.960 0.000 0.040
#> GSM25581 1 0.1964 0.816724 0.944 0.000 0.056
#> GSM48655 2 0.1482 0.879654 0.020 0.968 0.012
#> GSM48656 2 0.3038 0.885595 0.000 0.896 0.104
#> GSM48657 2 0.1620 0.878966 0.024 0.964 0.012
#> GSM48658 2 0.4452 0.840155 0.000 0.808 0.192
#> GSM25624 1 0.1643 0.817867 0.956 0.000 0.044
#> GSM25625 3 0.4504 0.727920 0.196 0.000 0.804
#> GSM25626 3 0.3340 0.754223 0.120 0.000 0.880
#> GSM25627 3 0.4725 0.723468 0.060 0.088 0.852
#> GSM25628 3 0.1529 0.752574 0.040 0.000 0.960
#> GSM25629 3 0.2584 0.728518 0.008 0.064 0.928
#> GSM25630 3 0.4062 0.739939 0.164 0.000 0.836
#> GSM25631 2 0.7075 0.239796 0.020 0.496 0.484
#> GSM25632 3 0.4702 0.713565 0.212 0.000 0.788
#> GSM25633 1 0.1964 0.816762 0.944 0.000 0.056
#> GSM25634 1 0.1753 0.817744 0.952 0.000 0.048
#> GSM25635 1 0.1643 0.817302 0.956 0.000 0.044
#> GSM25656 3 0.1832 0.751692 0.036 0.008 0.956
#> GSM25657 1 0.4062 0.762045 0.836 0.000 0.164
#> GSM25658 3 0.6215 0.267636 0.428 0.000 0.572
#> GSM25659 1 0.6282 0.543217 0.664 0.012 0.324
#> GSM25660 1 0.1753 0.816802 0.952 0.000 0.048
#> GSM25661 1 0.1860 0.817912 0.948 0.000 0.052
#> GSM25662 2 0.3192 0.882850 0.000 0.888 0.112
#> GSM25663 2 0.4555 0.829980 0.000 0.800 0.200
#> GSM25680 2 0.3879 0.869367 0.000 0.848 0.152
#> GSM25681 2 0.4346 0.849256 0.000 0.816 0.184
#> GSM25682 2 0.1636 0.878290 0.020 0.964 0.016
#> GSM25683 2 0.1636 0.878290 0.020 0.964 0.016
#> GSM25684 2 0.3267 0.881141 0.000 0.884 0.116
#> GSM25685 2 0.3267 0.881787 0.000 0.884 0.116
#> GSM25686 2 0.1636 0.878290 0.020 0.964 0.016
#> GSM25687 2 0.1636 0.878290 0.020 0.964 0.016
#> GSM48664 1 0.1453 0.796790 0.968 0.024 0.008
#> GSM48665 1 0.0747 0.804384 0.984 0.016 0.000
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM25548 2 0.530 -0.45407 0.008 0.496 0.000 0.496
#> GSM25549 4 0.528 0.46636 0.008 0.460 0.000 0.532
#> GSM25550 4 0.611 0.47286 0.048 0.428 0.000 0.524
#> GSM25551 2 0.514 0.35539 0.000 0.680 0.024 0.296
#> GSM25570 4 0.527 0.48596 0.008 0.452 0.000 0.540
#> GSM25571 4 0.529 0.43575 0.008 0.480 0.000 0.512
#> GSM25358 1 0.979 0.05760 0.312 0.156 0.252 0.280
#> GSM25359 3 0.871 -0.11690 0.036 0.276 0.368 0.320
#> GSM25360 3 0.365 0.71644 0.108 0.000 0.852 0.040
#> GSM25361 3 0.629 0.59209 0.060 0.016 0.648 0.276
#> GSM25377 1 0.526 0.66565 0.752 0.020 0.036 0.192
#> GSM25378 1 0.623 0.62505 0.696 0.052 0.040 0.212
#> GSM25401 2 0.954 0.03059 0.160 0.368 0.168 0.304
#> GSM25402 1 0.967 0.22688 0.340 0.236 0.140 0.284
#> GSM25349 2 0.469 0.39443 0.012 0.756 0.012 0.220
#> GSM25350 2 0.452 0.37053 0.004 0.728 0.004 0.264
#> GSM25356 1 0.678 0.59594 0.644 0.088 0.028 0.240
#> GSM25357 2 0.563 0.33551 0.028 0.704 0.024 0.244
#> GSM25385 3 0.409 0.68082 0.140 0.000 0.820 0.040
#> GSM25386 3 0.182 0.72394 0.024 0.004 0.948 0.024
#> GSM25399 1 0.376 0.70536 0.848 0.004 0.032 0.116
#> GSM25400 1 0.528 0.68636 0.760 0.004 0.100 0.136
#> GSM48659 2 0.520 0.33311 0.000 0.636 0.016 0.348
#> GSM48660 2 0.302 0.50925 0.000 0.852 0.000 0.148
#> GSM25409 2 0.584 0.11526 0.020 0.592 0.012 0.376
#> GSM25410 3 0.290 0.70953 0.056 0.008 0.904 0.032
#> GSM25426 2 0.484 0.45554 0.000 0.764 0.052 0.184
#> GSM25427 1 0.562 0.64944 0.732 0.040 0.028 0.200
#> GSM25540 3 0.455 0.59910 0.000 0.012 0.732 0.256
#> GSM25541 3 0.568 0.58631 0.032 0.012 0.672 0.284
#> GSM25542 2 0.725 0.15935 0.000 0.536 0.272 0.192
#> GSM25543 2 0.801 -0.05394 0.008 0.388 0.376 0.228
#> GSM25479 1 0.359 0.71850 0.860 0.000 0.088 0.052
#> GSM25480 1 0.395 0.70102 0.840 0.000 0.096 0.064
#> GSM25481 1 0.728 0.55631 0.608 0.120 0.032 0.240
#> GSM25482 1 0.684 0.58353 0.640 0.116 0.020 0.224
#> GSM48654 2 0.472 0.39147 0.000 0.692 0.008 0.300
#> GSM48650 2 0.212 0.51433 0.000 0.924 0.008 0.068
#> GSM48651 2 0.425 0.47571 0.000 0.768 0.012 0.220
#> GSM48652 2 0.460 0.45572 0.000 0.736 0.016 0.248
#> GSM48653 2 0.474 0.43922 0.000 0.704 0.012 0.284
#> GSM48662 2 0.436 0.41142 0.000 0.708 0.000 0.292
#> GSM48663 2 0.324 0.47379 0.004 0.856 0.004 0.136
#> GSM25524 3 0.527 0.52205 0.288 0.000 0.680 0.032
#> GSM25525 1 0.529 0.59842 0.724 0.000 0.216 0.060
#> GSM25526 3 0.436 0.68494 0.136 0.000 0.808 0.056
#> GSM25527 1 0.540 0.57256 0.700 0.000 0.248 0.052
#> GSM25528 3 0.578 -0.00242 0.480 0.000 0.492 0.028
#> GSM25529 1 0.520 0.55829 0.708 0.000 0.252 0.040
#> GSM25530 3 0.566 0.28535 0.396 0.000 0.576 0.028
#> GSM25531 1 0.547 0.45286 0.644 0.000 0.324 0.032
#> GSM48661 2 0.670 0.21251 0.000 0.544 0.100 0.356
#> GSM25561 3 0.531 0.53600 0.280 0.000 0.684 0.036
#> GSM25562 1 0.556 0.65482 0.720 0.000 0.188 0.092
#> GSM25563 3 0.255 0.72697 0.060 0.000 0.912 0.028
#> GSM25564 1 0.975 0.00761 0.360 0.180 0.244 0.216
#> GSM25565 2 0.458 0.48228 0.000 0.768 0.032 0.200
#> GSM25566 2 0.398 0.42899 0.000 0.776 0.004 0.220
#> GSM25568 4 0.960 0.10932 0.124 0.260 0.276 0.340
#> GSM25569 2 0.492 0.28514 0.000 0.628 0.004 0.368
#> GSM25552 4 0.614 0.53150 0.052 0.404 0.000 0.544
#> GSM25553 4 0.816 0.40128 0.220 0.228 0.036 0.516
#> GSM25578 1 0.367 0.70726 0.852 0.000 0.104 0.044
#> GSM25579 1 0.608 0.58501 0.684 0.000 0.164 0.152
#> GSM25580 1 0.289 0.72706 0.896 0.000 0.068 0.036
#> GSM25581 1 0.311 0.72046 0.884 0.000 0.080 0.036
#> GSM48655 2 0.259 0.48903 0.000 0.892 0.004 0.104
#> GSM48656 2 0.469 0.41160 0.004 0.704 0.004 0.288
#> GSM48657 2 0.156 0.51017 0.000 0.944 0.000 0.056
#> GSM48658 2 0.666 -0.02901 0.000 0.464 0.084 0.452
#> GSM25624 1 0.308 0.72665 0.888 0.000 0.064 0.048
#> GSM25625 3 0.376 0.70712 0.144 0.000 0.832 0.024
#> GSM25626 3 0.162 0.72605 0.028 0.000 0.952 0.020
#> GSM25627 3 0.733 0.41960 0.020 0.236 0.592 0.152
#> GSM25628 3 0.229 0.71954 0.004 0.012 0.924 0.060
#> GSM25629 3 0.640 0.52620 0.004 0.132 0.660 0.204
#> GSM25630 3 0.333 0.71013 0.112 0.000 0.864 0.024
#> GSM25631 4 0.772 0.42551 0.052 0.200 0.152 0.596
#> GSM25632 3 0.355 0.70226 0.136 0.000 0.844 0.020
#> GSM25633 1 0.329 0.72068 0.876 0.000 0.080 0.044
#> GSM25634 1 0.249 0.72807 0.912 0.000 0.068 0.020
#> GSM25635 1 0.250 0.72877 0.916 0.000 0.044 0.040
#> GSM25656 3 0.452 0.65468 0.004 0.060 0.808 0.128
#> GSM25657 1 0.444 0.62772 0.764 0.000 0.216 0.020
#> GSM25658 3 0.680 0.29943 0.348 0.004 0.552 0.096
#> GSM25659 1 0.694 0.44178 0.592 0.004 0.260 0.144
#> GSM25660 1 0.332 0.71733 0.876 0.000 0.068 0.056
#> GSM25661 1 0.262 0.72342 0.908 0.000 0.064 0.028
#> GSM25662 2 0.486 0.43556 0.000 0.700 0.016 0.284
#> GSM25663 2 0.661 -0.01192 0.000 0.516 0.084 0.400
#> GSM25680 4 0.545 0.42063 0.000 0.360 0.024 0.616
#> GSM25681 4 0.646 0.51501 0.020 0.276 0.064 0.640
#> GSM25682 2 0.310 0.43372 0.000 0.856 0.004 0.140
#> GSM25683 2 0.289 0.45293 0.000 0.872 0.004 0.124
#> GSM25684 2 0.500 0.35584 0.000 0.660 0.012 0.328
#> GSM25685 2 0.555 0.42924 0.000 0.672 0.048 0.280
#> GSM25686 2 0.305 0.43697 0.000 0.860 0.004 0.136
#> GSM25687 2 0.294 0.44361 0.000 0.868 0.004 0.128
#> GSM48664 1 0.373 0.70211 0.848 0.004 0.028 0.120
#> GSM48665 1 0.324 0.71744 0.880 0.004 0.028 0.088
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM25548 5 0.354 0.62813 0.004 0.176 0.000 0.016 0.804
#> GSM25549 5 0.442 0.64003 0.008 0.168 0.016 0.032 0.776
#> GSM25550 5 0.464 0.62312 0.048 0.132 0.000 0.044 0.776
#> GSM25551 5 0.715 -0.09098 0.000 0.372 0.032 0.180 0.416
#> GSM25570 5 0.293 0.64747 0.004 0.128 0.000 0.012 0.856
#> GSM25571 5 0.342 0.63090 0.004 0.164 0.000 0.016 0.816
#> GSM25358 4 0.878 0.38078 0.176 0.056 0.228 0.428 0.112
#> GSM25359 3 0.865 -0.06496 0.016 0.116 0.304 0.272 0.292
#> GSM25360 3 0.504 0.61594 0.172 0.000 0.736 0.044 0.048
#> GSM25361 3 0.778 0.45241 0.136 0.048 0.504 0.044 0.268
#> GSM25377 4 0.516 0.32979 0.468 0.012 0.008 0.504 0.008
#> GSM25378 4 0.633 0.55258 0.352 0.016 0.036 0.552 0.044
#> GSM25401 4 0.681 0.41289 0.060 0.104 0.140 0.652 0.044
#> GSM25402 4 0.655 0.53697 0.104 0.080 0.116 0.672 0.028
#> GSM25349 2 0.686 0.36496 0.008 0.504 0.008 0.268 0.212
#> GSM25350 2 0.668 0.31054 0.004 0.476 0.000 0.244 0.276
#> GSM25356 4 0.620 0.57862 0.336 0.028 0.016 0.572 0.048
#> GSM25357 2 0.706 0.24762 0.004 0.400 0.012 0.376 0.208
#> GSM25385 3 0.488 0.60739 0.152 0.000 0.736 0.104 0.008
#> GSM25386 3 0.305 0.63840 0.036 0.000 0.876 0.072 0.016
#> GSM25399 1 0.479 0.02036 0.584 0.000 0.024 0.392 0.000
#> GSM25400 1 0.601 -0.23234 0.488 0.000 0.064 0.428 0.020
#> GSM48659 2 0.470 0.39726 0.000 0.720 0.008 0.048 0.224
#> GSM48660 2 0.427 0.52759 0.000 0.784 0.004 0.120 0.092
#> GSM25409 5 0.698 0.00749 0.012 0.348 0.004 0.196 0.440
#> GSM25410 3 0.299 0.62512 0.024 0.000 0.872 0.092 0.012
#> GSM25426 2 0.752 0.35093 0.000 0.468 0.072 0.280 0.180
#> GSM25427 4 0.550 0.45631 0.428 0.004 0.004 0.520 0.044
#> GSM25540 3 0.644 0.51780 0.008 0.084 0.640 0.068 0.200
#> GSM25541 3 0.744 0.51308 0.076 0.056 0.572 0.068 0.228
#> GSM25542 2 0.810 0.18817 0.000 0.420 0.252 0.172 0.156
#> GSM25543 3 0.824 -0.08103 0.000 0.336 0.336 0.160 0.168
#> GSM25479 1 0.384 0.63016 0.832 0.000 0.036 0.096 0.036
#> GSM25480 1 0.469 0.60673 0.776 0.000 0.032 0.112 0.080
#> GSM25481 4 0.638 0.61029 0.272 0.064 0.008 0.604 0.052
#> GSM25482 4 0.628 0.59721 0.324 0.036 0.004 0.568 0.068
#> GSM48654 2 0.303 0.48630 0.000 0.852 0.004 0.016 0.128
#> GSM48650 2 0.450 0.52381 0.000 0.752 0.004 0.176 0.068
#> GSM48651 2 0.277 0.53077 0.000 0.884 0.004 0.036 0.076
#> GSM48652 2 0.230 0.52199 0.000 0.904 0.000 0.024 0.072
#> GSM48653 2 0.362 0.49210 0.000 0.836 0.020 0.032 0.112
#> GSM48662 2 0.422 0.50372 0.000 0.780 0.008 0.052 0.160
#> GSM48663 2 0.532 0.47539 0.000 0.660 0.000 0.228 0.112
#> GSM25524 3 0.547 0.22700 0.428 0.000 0.516 0.052 0.004
#> GSM25525 1 0.424 0.60108 0.792 0.000 0.144 0.040 0.024
#> GSM25526 3 0.673 0.54815 0.188 0.012 0.600 0.168 0.032
#> GSM25527 1 0.528 0.58608 0.724 0.000 0.140 0.108 0.028
#> GSM25528 1 0.528 0.23831 0.584 0.000 0.364 0.048 0.004
#> GSM25529 1 0.391 0.60794 0.812 0.000 0.136 0.032 0.020
#> GSM25530 3 0.579 0.01424 0.460 0.000 0.460 0.076 0.004
#> GSM25531 1 0.538 0.53928 0.672 0.000 0.208 0.116 0.004
#> GSM48661 2 0.597 0.35457 0.000 0.656 0.092 0.044 0.208
#> GSM25561 3 0.625 0.30813 0.372 0.004 0.528 0.072 0.024
#> GSM25562 1 0.625 0.45681 0.596 0.004 0.176 0.216 0.008
#> GSM25563 3 0.434 0.64094 0.104 0.004 0.800 0.076 0.016
#> GSM25564 1 0.970 -0.15614 0.288 0.256 0.128 0.180 0.148
#> GSM25565 2 0.562 0.47656 0.000 0.680 0.024 0.104 0.192
#> GSM25566 2 0.671 0.23422 0.000 0.468 0.020 0.144 0.368
#> GSM25568 2 0.907 0.07670 0.052 0.388 0.188 0.200 0.172
#> GSM25569 2 0.512 0.39290 0.000 0.668 0.004 0.068 0.260
#> GSM25552 5 0.489 0.62866 0.052 0.156 0.004 0.032 0.756
#> GSM25553 5 0.599 0.49827 0.160 0.080 0.012 0.056 0.692
#> GSM25578 1 0.277 0.64042 0.892 0.000 0.052 0.044 0.012
#> GSM25579 1 0.606 0.47079 0.680 0.004 0.084 0.072 0.160
#> GSM25580 1 0.283 0.60092 0.864 0.000 0.012 0.120 0.004
#> GSM25581 1 0.230 0.63211 0.908 0.000 0.020 0.068 0.004
#> GSM48655 2 0.543 0.46001 0.000 0.648 0.000 0.120 0.232
#> GSM48656 2 0.456 0.46131 0.000 0.736 0.004 0.056 0.204
#> GSM48657 2 0.477 0.51228 0.000 0.732 0.000 0.136 0.132
#> GSM48658 2 0.610 0.17618 0.000 0.556 0.060 0.036 0.348
#> GSM25624 1 0.503 0.41035 0.672 0.000 0.060 0.264 0.004
#> GSM25625 3 0.504 0.61242 0.152 0.008 0.732 0.104 0.004
#> GSM25626 3 0.201 0.64269 0.020 0.008 0.932 0.036 0.004
#> GSM25627 3 0.799 0.40352 0.036 0.204 0.488 0.216 0.056
#> GSM25628 3 0.250 0.63649 0.004 0.040 0.912 0.024 0.020
#> GSM25629 3 0.781 0.40120 0.008 0.188 0.508 0.176 0.120
#> GSM25630 3 0.420 0.61599 0.176 0.000 0.776 0.036 0.012
#> GSM25631 5 0.788 0.35830 0.064 0.168 0.176 0.052 0.540
#> GSM25632 3 0.454 0.57426 0.228 0.000 0.724 0.044 0.004
#> GSM25633 1 0.271 0.63657 0.884 0.000 0.044 0.072 0.000
#> GSM25634 1 0.375 0.59585 0.820 0.000 0.036 0.132 0.012
#> GSM25635 1 0.365 0.58814 0.828 0.000 0.020 0.128 0.024
#> GSM25656 3 0.493 0.60931 0.004 0.072 0.776 0.072 0.076
#> GSM25657 1 0.505 0.60231 0.720 0.000 0.160 0.112 0.008
#> GSM25658 3 0.764 0.22246 0.308 0.016 0.404 0.248 0.024
#> GSM25659 1 0.796 0.38989 0.548 0.044 0.136 0.136 0.136
#> GSM25660 1 0.308 0.62224 0.876 0.000 0.024 0.072 0.028
#> GSM25661 1 0.276 0.61691 0.880 0.000 0.024 0.092 0.004
#> GSM25662 2 0.605 0.36967 0.000 0.620 0.036 0.084 0.260
#> GSM25663 5 0.653 0.22597 0.000 0.360 0.076 0.048 0.516
#> GSM25680 5 0.484 0.57331 0.000 0.220 0.024 0.036 0.720
#> GSM25681 5 0.531 0.59546 0.016 0.144 0.056 0.040 0.744
#> GSM25682 2 0.614 0.32319 0.000 0.528 0.000 0.152 0.320
#> GSM25683 2 0.635 0.33750 0.000 0.520 0.004 0.164 0.312
#> GSM25684 2 0.552 0.30550 0.000 0.596 0.008 0.064 0.332
#> GSM25685 2 0.655 0.35296 0.000 0.600 0.052 0.124 0.224
#> GSM25686 2 0.615 0.32575 0.000 0.524 0.000 0.152 0.324
#> GSM25687 2 0.613 0.33703 0.000 0.532 0.000 0.152 0.316
#> GSM48664 1 0.435 0.11764 0.624 0.000 0.008 0.368 0.000
#> GSM48665 1 0.406 0.32000 0.708 0.000 0.000 0.280 0.012
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM25548 5 0.309 0.45553 0.000 0.120 0.000 0.000 0.832 0.048
#> GSM25549 5 0.379 0.46150 0.004 0.132 0.008 0.008 0.804 0.044
#> GSM25550 5 0.411 0.44192 0.020 0.032 0.008 0.056 0.816 0.068
#> GSM25551 5 0.699 -0.14051 0.000 0.236 0.020 0.028 0.384 0.332
#> GSM25570 5 0.274 0.46991 0.000 0.084 0.000 0.012 0.872 0.032
#> GSM25571 5 0.331 0.45844 0.000 0.128 0.000 0.004 0.820 0.048
#> GSM25358 4 0.846 0.09476 0.048 0.032 0.260 0.360 0.088 0.212
#> GSM25359 6 0.844 -0.09351 0.008 0.064 0.284 0.136 0.172 0.336
#> GSM25360 3 0.566 0.56173 0.192 0.012 0.676 0.032 0.028 0.060
#> GSM25361 3 0.862 0.35731 0.156 0.044 0.360 0.036 0.260 0.144
#> GSM25377 4 0.426 0.50988 0.204 0.004 0.012 0.740 0.004 0.036
#> GSM25378 4 0.488 0.56310 0.144 0.000 0.020 0.728 0.016 0.092
#> GSM25401 4 0.779 0.14231 0.028 0.120 0.104 0.416 0.020 0.312
#> GSM25402 4 0.692 0.41931 0.056 0.068 0.068 0.552 0.008 0.248
#> GSM25349 6 0.744 0.18752 0.004 0.336 0.008 0.132 0.136 0.384
#> GSM25350 6 0.738 0.19171 0.004 0.328 0.004 0.088 0.216 0.360
#> GSM25356 4 0.543 0.57195 0.144 0.008 0.008 0.696 0.040 0.104
#> GSM25357 6 0.769 0.32380 0.000 0.212 0.012 0.164 0.220 0.392
#> GSM25385 3 0.523 0.55962 0.120 0.000 0.704 0.128 0.012 0.036
#> GSM25386 3 0.278 0.63478 0.036 0.000 0.884 0.032 0.004 0.044
#> GSM25399 4 0.492 0.14659 0.392 0.000 0.032 0.556 0.000 0.020
#> GSM25400 4 0.557 0.35972 0.312 0.000 0.040 0.584 0.004 0.060
#> GSM48659 2 0.367 0.43188 0.000 0.764 0.008 0.000 0.204 0.024
#> GSM48660 2 0.556 0.24428 0.000 0.628 0.000 0.044 0.100 0.228
#> GSM25409 5 0.797 -0.21449 0.004 0.248 0.016 0.140 0.320 0.272
#> GSM25410 3 0.324 0.63126 0.032 0.000 0.848 0.080 0.000 0.040
#> GSM25426 6 0.693 0.10377 0.000 0.372 0.032 0.040 0.136 0.420
#> GSM25427 4 0.483 0.54578 0.188 0.008 0.004 0.712 0.012 0.076
#> GSM25540 3 0.737 0.51526 0.028 0.068 0.528 0.024 0.168 0.184
#> GSM25541 3 0.850 0.49306 0.104 0.084 0.440 0.036 0.160 0.176
#> GSM25542 2 0.819 0.02474 0.000 0.380 0.184 0.056 0.164 0.216
#> GSM25543 3 0.907 -0.17749 0.024 0.228 0.288 0.092 0.164 0.204
#> GSM25479 1 0.448 0.56753 0.760 0.000 0.024 0.152 0.040 0.024
#> GSM25480 1 0.501 0.55367 0.740 0.000 0.032 0.108 0.088 0.032
#> GSM25481 4 0.525 0.57777 0.076 0.028 0.004 0.716 0.028 0.148
#> GSM25482 4 0.507 0.57787 0.080 0.012 0.000 0.720 0.044 0.144
#> GSM48654 2 0.269 0.47209 0.000 0.868 0.008 0.000 0.100 0.024
#> GSM48650 2 0.558 -0.01192 0.000 0.556 0.000 0.028 0.084 0.332
#> GSM48651 2 0.386 0.42001 0.000 0.788 0.004 0.004 0.076 0.128
#> GSM48652 2 0.307 0.45207 0.000 0.856 0.008 0.004 0.052 0.080
#> GSM48653 2 0.296 0.46181 0.000 0.860 0.012 0.000 0.048 0.080
#> GSM48662 2 0.539 0.40210 0.004 0.672 0.004 0.024 0.156 0.140
#> GSM48663 2 0.661 -0.07014 0.000 0.468 0.000 0.120 0.084 0.328
#> GSM25524 1 0.587 0.10492 0.508 0.000 0.388 0.040 0.016 0.048
#> GSM25525 1 0.352 0.59082 0.844 0.000 0.056 0.060 0.020 0.020
#> GSM25526 3 0.729 0.35171 0.256 0.024 0.452 0.072 0.000 0.196
#> GSM25527 1 0.543 0.55723 0.692 0.000 0.096 0.144 0.012 0.056
#> GSM25528 1 0.493 0.42768 0.668 0.000 0.260 0.028 0.012 0.032
#> GSM25529 1 0.333 0.58381 0.852 0.000 0.076 0.020 0.036 0.016
#> GSM25530 1 0.605 0.40211 0.568 0.000 0.280 0.084 0.004 0.064
#> GSM25531 1 0.479 0.57445 0.744 0.000 0.084 0.120 0.008 0.044
#> GSM48661 2 0.592 0.37447 0.004 0.652 0.096 0.004 0.112 0.132
#> GSM25561 3 0.702 0.21909 0.312 0.008 0.472 0.132 0.016 0.060
#> GSM25562 1 0.721 0.31786 0.488 0.004 0.144 0.256 0.020 0.088
#> GSM25563 3 0.482 0.59688 0.156 0.000 0.724 0.032 0.004 0.084
#> GSM25564 1 0.966 0.00067 0.272 0.196 0.096 0.164 0.104 0.168
#> GSM25565 2 0.693 -0.01390 0.000 0.460 0.020 0.040 0.196 0.284
#> GSM25566 5 0.654 -0.25235 0.000 0.328 0.000 0.020 0.356 0.296
#> GSM25568 2 0.908 0.07354 0.052 0.356 0.184 0.172 0.080 0.156
#> GSM25569 2 0.596 0.34765 0.000 0.612 0.004 0.048 0.168 0.168
#> GSM25552 5 0.421 0.45725 0.024 0.084 0.008 0.016 0.804 0.064
#> GSM25553 5 0.680 0.34101 0.116 0.060 0.036 0.072 0.636 0.080
#> GSM25578 1 0.351 0.57294 0.808 0.000 0.012 0.152 0.008 0.020
#> GSM25579 1 0.664 0.45663 0.612 0.004 0.052 0.124 0.144 0.064
#> GSM25580 1 0.401 0.50612 0.704 0.000 0.016 0.268 0.000 0.012
#> GSM25581 1 0.427 0.54250 0.732 0.000 0.028 0.216 0.008 0.016
#> GSM48655 2 0.644 -0.12946 0.000 0.416 0.000 0.020 0.256 0.308
#> GSM48656 2 0.463 0.43211 0.000 0.716 0.004 0.004 0.144 0.132
#> GSM48657 2 0.594 0.02137 0.000 0.540 0.000 0.024 0.148 0.288
#> GSM48658 2 0.618 0.30446 0.000 0.592 0.052 0.008 0.200 0.148
#> GSM25624 1 0.635 0.07281 0.444 0.000 0.064 0.412 0.012 0.068
#> GSM25625 3 0.628 0.49895 0.168 0.008 0.616 0.104 0.004 0.100
#> GSM25626 3 0.265 0.64236 0.020 0.008 0.892 0.032 0.000 0.048
#> GSM25627 3 0.824 0.30967 0.044 0.164 0.376 0.088 0.028 0.300
#> GSM25628 3 0.361 0.64372 0.036 0.016 0.836 0.008 0.012 0.092
#> GSM25629 3 0.771 0.31907 0.016 0.224 0.388 0.028 0.052 0.292
#> GSM25630 3 0.421 0.60281 0.156 0.000 0.768 0.032 0.004 0.040
#> GSM25631 5 0.824 0.19438 0.096 0.204 0.100 0.012 0.440 0.148
#> GSM25632 3 0.427 0.58720 0.148 0.000 0.768 0.040 0.004 0.040
#> GSM25633 1 0.498 0.54108 0.676 0.000 0.052 0.236 0.004 0.032
#> GSM25634 1 0.541 0.40091 0.584 0.000 0.056 0.320 0.000 0.040
#> GSM25635 1 0.521 0.41893 0.612 0.000 0.040 0.308 0.004 0.036
#> GSM25656 3 0.645 0.59004 0.032 0.104 0.632 0.028 0.036 0.168
#> GSM25657 1 0.521 0.54639 0.684 0.000 0.076 0.196 0.008 0.036
#> GSM25658 1 0.824 -0.11106 0.304 0.040 0.276 0.176 0.000 0.204
#> GSM25659 1 0.793 0.41739 0.532 0.048 0.124 0.120 0.076 0.100
#> GSM25660 1 0.477 0.55649 0.732 0.000 0.032 0.176 0.032 0.028
#> GSM25661 1 0.415 0.51436 0.720 0.000 0.012 0.240 0.004 0.024
#> GSM25662 2 0.587 0.23313 0.000 0.544 0.000 0.012 0.236 0.208
#> GSM25663 5 0.753 0.08912 0.012 0.308 0.052 0.024 0.400 0.204
#> GSM25680 5 0.549 0.36415 0.000 0.236 0.036 0.004 0.636 0.088
#> GSM25681 5 0.571 0.41580 0.020 0.112 0.072 0.012 0.700 0.084
#> GSM25682 5 0.636 -0.24778 0.000 0.272 0.000 0.012 0.388 0.328
#> GSM25683 6 0.660 0.13656 0.000 0.300 0.004 0.016 0.340 0.340
#> GSM25684 2 0.533 0.28010 0.000 0.600 0.004 0.004 0.276 0.116
#> GSM25685 2 0.590 0.22931 0.000 0.596 0.016 0.016 0.144 0.228
#> GSM25686 5 0.654 -0.29547 0.000 0.288 0.000 0.020 0.352 0.340
#> GSM25687 5 0.672 -0.30520 0.000 0.296 0.000 0.032 0.344 0.328
#> GSM48664 4 0.469 0.12758 0.392 0.000 0.020 0.572 0.004 0.012
#> GSM48665 1 0.439 0.18303 0.532 0.000 0.000 0.444 0.000 0.024
Heatmaps for the consensus matrix. It visualizes the probability of two samples to be in a same group.
consensus_heatmap(res, k = 2)
consensus_heatmap(res, k = 3)
consensus_heatmap(res, k = 4)
consensus_heatmap(res, k = 5)
consensus_heatmap(res, k = 6)
Heatmaps for the membership of samples in all partitions to see how consistent they are:
membership_heatmap(res, k = 2)
membership_heatmap(res, k = 3)
membership_heatmap(res, k = 4)
membership_heatmap(res, k = 5)
membership_heatmap(res, k = 6)
As soon as we have had the classes for columns, we can look for signatures which are significantly different between classes which can be candidate marks for certain classes. Following are the heatmaps for signatures.
Signature heatmaps where rows are scaled:
get_signatures(res, k = 2)
get_signatures(res, k = 3)
get_signatures(res, k = 4)
get_signatures(res, k = 5)
get_signatures(res, k = 6)
Signature heatmaps where rows are not scaled:
get_signatures(res, k = 2, scale_rows = FALSE)
get_signatures(res, k = 3, scale_rows = FALSE)
get_signatures(res, k = 4, scale_rows = FALSE)
get_signatures(res, k = 5, scale_rows = FALSE)
get_signatures(res, k = 6, scale_rows = FALSE)
Compare the overlap of signatures from different k:
compare_signatures(res)
get_signature() returns a data frame invisibly. TO get the list of signatures, the function
call should be assigned to a variable explicitly. In following code, if plot argument is set
to FALSE, no heatmap is plotted while only the differential analysis is performed.
# code only for demonstration
tb = get_signature(res, k = ..., plot = FALSE)
An example of the output of tb is:
#> which_row fdr mean_1 mean_2 scaled_mean_1 scaled_mean_2 km
#> 1 38 0.042760348 8.373488 9.131774 -0.5533452 0.5164555 1
#> 2 40 0.018707592 7.106213 8.469186 -0.6173731 0.5762149 1
#> 3 55 0.019134737 10.221463 11.207825 -0.6159697 0.5749050 1
#> 4 59 0.006059896 5.921854 7.869574 -0.6899429 0.6439467 1
#> 5 60 0.018055526 8.928898 10.211722 -0.6204761 0.5791110 1
#> 6 98 0.009384629 15.714769 14.887706 0.6635654 -0.6193277 2
...
The columns in tb are:
which_row: row indices corresponding to the input matrix.fdr: FDR for the differential test. mean_x: The mean value in group x.scaled_mean_x: The mean value in group x after rows are scaled.km: Row groups if k-means clustering is applied to rows.UMAP plot which shows how samples are separated.
dimension_reduction(res, k = 2, method = "UMAP")
dimension_reduction(res, k = 3, method = "UMAP")
dimension_reduction(res, k = 4, method = "UMAP")
dimension_reduction(res, k = 5, method = "UMAP")
dimension_reduction(res, k = 6, method = "UMAP")
Following heatmap shows how subgroups are split when increasing k:
collect_classes(res)
Test correlation between subgroups and known annotations. If the known annotation is numeric, one-way ANOVA test is applied, and if the known annotation is discrete, chi-squared contingency table test is applied.
test_to_known_factors(res)
#> n genotype/variation(p) k
#> SD:skmeans 95 1.56e-05 2
#> SD:skmeans 88 1.77e-05 3
#> SD:skmeans 50 3.54e-05 4
#> SD:skmeans 48 5.90e-07 5
#> SD:skmeans 30 5.55e-02 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 100 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 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 0.251 0.748 0.848 0.4586 0.553 0.553
#> 3 3 0.383 0.673 0.801 0.4277 0.728 0.529
#> 4 4 0.473 0.574 0.764 0.1201 0.848 0.589
#> 5 5 0.492 0.357 0.672 0.0417 0.896 0.668
#> 6 6 0.555 0.446 0.738 0.0239 0.858 0.538
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
#> GSM25548 2 0.0000 0.837 0.000 1.000
#> GSM25549 2 0.0000 0.837 0.000 1.000
#> GSM25550 2 0.0376 0.836 0.004 0.996
#> GSM25551 2 0.4161 0.822 0.084 0.916
#> GSM25570 2 0.0000 0.837 0.000 1.000
#> GSM25571 2 0.0376 0.838 0.004 0.996
#> GSM25358 2 0.5178 0.836 0.116 0.884
#> GSM25359 2 0.6712 0.776 0.176 0.824
#> GSM25360 1 0.9129 0.573 0.672 0.328
#> GSM25361 2 0.6623 0.814 0.172 0.828
#> GSM25377 2 0.8763 0.490 0.296 0.704
#> GSM25378 2 0.6973 0.684 0.188 0.812
#> GSM25401 1 0.5294 0.803 0.880 0.120
#> GSM25402 2 0.9491 0.510 0.368 0.632
#> GSM25349 2 0.3274 0.849 0.060 0.940
#> GSM25350 2 0.2043 0.848 0.032 0.968
#> GSM25356 2 0.4562 0.779 0.096 0.904
#> GSM25357 2 0.1633 0.846 0.024 0.976
#> GSM25385 1 0.6438 0.804 0.836 0.164
#> GSM25386 1 0.7674 0.677 0.776 0.224
#> GSM25399 1 0.3274 0.823 0.940 0.060
#> GSM25400 1 0.7219 0.766 0.800 0.200
#> GSM48659 2 0.5519 0.832 0.128 0.872
#> GSM48660 2 0.5059 0.838 0.112 0.888
#> GSM25409 2 0.1414 0.845 0.020 0.980
#> GSM25410 1 0.9044 0.592 0.680 0.320
#> GSM25426 2 0.9866 0.360 0.432 0.568
#> GSM25427 2 0.1843 0.826 0.028 0.972
#> GSM25540 2 0.9044 0.644 0.320 0.680
#> GSM25541 2 0.9209 0.608 0.336 0.664
#> GSM25542 2 0.5178 0.841 0.116 0.884
#> GSM25543 2 0.5842 0.831 0.140 0.860
#> GSM25479 2 0.9661 0.204 0.392 0.608
#> GSM25480 2 0.8327 0.576 0.264 0.736
#> GSM25481 2 0.1414 0.838 0.020 0.980
#> GSM25482 2 0.1414 0.831 0.020 0.980
#> GSM48654 2 0.4431 0.844 0.092 0.908
#> GSM48650 2 0.3733 0.848 0.072 0.928
#> GSM48651 2 0.5178 0.835 0.116 0.884
#> GSM48652 2 0.2423 0.848 0.040 0.960
#> GSM48653 2 0.8144 0.734 0.252 0.748
#> GSM48662 2 0.2603 0.849 0.044 0.956
#> GSM48663 2 0.4562 0.843 0.096 0.904
#> GSM25524 1 0.1414 0.818 0.980 0.020
#> GSM25525 1 0.8861 0.680 0.696 0.304
#> GSM25526 1 0.2236 0.820 0.964 0.036
#> GSM25527 1 0.7219 0.782 0.800 0.200
#> GSM25528 1 0.2603 0.822 0.956 0.044
#> GSM25529 1 0.5059 0.792 0.888 0.112
#> GSM25530 1 0.4690 0.802 0.900 0.100
#> GSM25531 1 0.0938 0.817 0.988 0.012
#> GSM48661 2 0.8016 0.743 0.244 0.756
#> GSM25561 1 0.8327 0.654 0.736 0.264
#> GSM25562 1 0.9358 0.428 0.648 0.352
#> GSM25563 1 0.6973 0.699 0.812 0.188
#> GSM25564 2 0.6438 0.820 0.164 0.836
#> GSM25565 2 0.5408 0.834 0.124 0.876
#> GSM25566 2 0.0938 0.837 0.012 0.988
#> GSM25568 2 0.3274 0.850 0.060 0.940
#> GSM25569 2 0.0938 0.841 0.012 0.988
#> GSM25552 2 0.0938 0.840 0.012 0.988
#> GSM25553 2 0.1184 0.842 0.016 0.984
#> GSM25578 2 0.8555 0.538 0.280 0.720
#> GSM25579 2 0.4431 0.794 0.092 0.908
#> GSM25580 1 0.7674 0.748 0.776 0.224
#> GSM25581 1 0.7815 0.750 0.768 0.232
#> GSM48655 2 0.2603 0.848 0.044 0.956
#> GSM48656 2 0.5629 0.832 0.132 0.868
#> GSM48657 2 0.4298 0.845 0.088 0.912
#> GSM48658 2 0.7219 0.777 0.200 0.800
#> GSM25624 1 0.9988 0.294 0.520 0.480
#> GSM25625 1 0.2043 0.820 0.968 0.032
#> GSM25626 1 0.1843 0.819 0.972 0.028
#> GSM25627 1 0.2236 0.820 0.964 0.036
#> GSM25628 1 0.5946 0.749 0.856 0.144
#> GSM25629 1 0.3879 0.817 0.924 0.076
#> GSM25630 1 0.2236 0.821 0.964 0.036
#> GSM25631 2 0.2043 0.837 0.032 0.968
#> GSM25632 1 0.1184 0.818 0.984 0.016
#> GSM25633 1 0.8713 0.706 0.708 0.292
#> GSM25634 1 0.4815 0.819 0.896 0.104
#> GSM25635 2 0.8955 0.464 0.312 0.688
#> GSM25656 2 0.9850 0.426 0.428 0.572
#> GSM25657 1 0.5629 0.803 0.868 0.132
#> GSM25658 1 0.1633 0.820 0.976 0.024
#> GSM25659 2 0.8763 0.693 0.296 0.704
#> GSM25660 2 0.8713 0.497 0.292 0.708
#> GSM25661 1 1.0000 0.249 0.504 0.496
#> GSM25662 2 0.5519 0.833 0.128 0.872
#> GSM25663 2 0.5519 0.833 0.128 0.872
#> GSM25680 2 0.0000 0.837 0.000 1.000
#> GSM25681 2 0.0000 0.837 0.000 1.000
#> GSM25682 2 0.4022 0.847 0.080 0.920
#> GSM25683 2 0.5178 0.836 0.116 0.884
#> GSM25684 2 0.5629 0.832 0.132 0.868
#> GSM25685 2 0.9522 0.553 0.372 0.628
#> GSM25686 2 0.4562 0.842 0.096 0.904
#> GSM25687 2 0.2603 0.849 0.044 0.956
#> GSM48664 2 0.9323 0.595 0.348 0.652
#> GSM48665 2 0.7745 0.624 0.228 0.772
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM25548 2 0.0237 0.7962 0.000 0.996 0.004
#> GSM25549 2 0.0237 0.7962 0.000 0.996 0.004
#> GSM25550 2 0.0424 0.7959 0.000 0.992 0.008
#> GSM25551 2 0.3009 0.7734 0.028 0.920 0.052
#> GSM25570 2 0.0237 0.7962 0.000 0.996 0.004
#> GSM25571 2 0.0424 0.7958 0.000 0.992 0.008
#> GSM25358 3 0.4465 0.7746 0.004 0.176 0.820
#> GSM25359 2 0.6294 0.5752 0.020 0.692 0.288
#> GSM25360 1 0.9248 0.1974 0.516 0.188 0.296
#> GSM25361 3 0.6988 0.6367 0.036 0.320 0.644
#> GSM25377 2 0.7214 0.4884 0.324 0.632 0.044
#> GSM25378 2 0.3995 0.7468 0.116 0.868 0.016
#> GSM25401 1 0.7710 0.5750 0.576 0.056 0.368
#> GSM25402 3 0.5180 0.7776 0.032 0.156 0.812
#> GSM25349 2 0.5363 0.5544 0.000 0.724 0.276
#> GSM25350 2 0.3619 0.7404 0.000 0.864 0.136
#> GSM25356 2 0.1315 0.7982 0.008 0.972 0.020
#> GSM25357 2 0.3551 0.7276 0.000 0.868 0.132
#> GSM25385 1 0.2998 0.8188 0.916 0.068 0.016
#> GSM25386 3 0.6452 0.4328 0.264 0.032 0.704
#> GSM25399 1 0.1525 0.8114 0.964 0.004 0.032
#> GSM25400 1 0.6295 0.6490 0.728 0.036 0.236
#> GSM48659 3 0.4452 0.7705 0.000 0.192 0.808
#> GSM48660 3 0.5363 0.7297 0.000 0.276 0.724
#> GSM25409 2 0.2066 0.7857 0.000 0.940 0.060
#> GSM25410 1 0.8915 0.4410 0.572 0.216 0.212
#> GSM25426 3 0.3213 0.7286 0.028 0.060 0.912
#> GSM25427 2 0.2711 0.7586 0.088 0.912 0.000
#> GSM25540 3 0.6192 0.6588 0.060 0.176 0.764
#> GSM25541 3 0.7814 0.5285 0.104 0.244 0.652
#> GSM25542 3 0.5138 0.7329 0.000 0.252 0.748
#> GSM25543 3 0.5810 0.6345 0.000 0.336 0.664
#> GSM25479 2 0.6819 0.0431 0.476 0.512 0.012
#> GSM25480 2 0.2527 0.7867 0.044 0.936 0.020
#> GSM25481 2 0.2804 0.7896 0.016 0.924 0.060
#> GSM25482 2 0.1751 0.7941 0.028 0.960 0.012
#> GSM48654 3 0.4887 0.7583 0.000 0.228 0.772
#> GSM48650 3 0.5835 0.6666 0.000 0.340 0.660
#> GSM48651 3 0.4062 0.7729 0.000 0.164 0.836
#> GSM48652 2 0.5465 0.4955 0.000 0.712 0.288
#> GSM48653 3 0.2063 0.7558 0.008 0.044 0.948
#> GSM48662 2 0.4974 0.6004 0.000 0.764 0.236
#> GSM48663 3 0.4931 0.7569 0.000 0.232 0.768
#> GSM25524 1 0.4235 0.8031 0.824 0.000 0.176
#> GSM25525 1 0.5223 0.7591 0.800 0.176 0.024
#> GSM25526 1 0.5008 0.7961 0.804 0.016 0.180
#> GSM25527 1 0.4551 0.7872 0.844 0.132 0.024
#> GSM25528 1 0.1765 0.8211 0.956 0.004 0.040
#> GSM25529 1 0.2056 0.8211 0.952 0.024 0.024
#> GSM25530 1 0.2446 0.8245 0.936 0.012 0.052
#> GSM25531 1 0.2066 0.8215 0.940 0.000 0.060
#> GSM48661 3 0.1711 0.7480 0.008 0.032 0.960
#> GSM25561 1 0.6757 0.6608 0.736 0.180 0.084
#> GSM25562 3 0.8352 0.1183 0.332 0.100 0.568
#> GSM25563 3 0.3412 0.6527 0.124 0.000 0.876
#> GSM25564 3 0.6682 0.1575 0.008 0.488 0.504
#> GSM25565 3 0.4654 0.7652 0.000 0.208 0.792
#> GSM25566 2 0.0983 0.7948 0.004 0.980 0.016
#> GSM25568 2 0.5621 0.4740 0.000 0.692 0.308
#> GSM25569 2 0.1529 0.7942 0.000 0.960 0.040
#> GSM25552 2 0.1860 0.7877 0.000 0.948 0.052
#> GSM25553 2 0.2356 0.7825 0.000 0.928 0.072
#> GSM25578 2 0.5698 0.6229 0.252 0.736 0.012
#> GSM25579 2 0.1877 0.7946 0.012 0.956 0.032
#> GSM25580 1 0.2173 0.8092 0.944 0.048 0.008
#> GSM25581 1 0.1643 0.8105 0.956 0.044 0.000
#> GSM48655 2 0.6309 -0.2445 0.000 0.504 0.496
#> GSM48656 3 0.4834 0.7697 0.004 0.204 0.792
#> GSM48657 3 0.5760 0.6836 0.000 0.328 0.672
#> GSM48658 3 0.4473 0.7666 0.008 0.164 0.828
#> GSM25624 1 0.5706 0.5087 0.680 0.320 0.000
#> GSM25625 1 0.4805 0.7983 0.812 0.012 0.176
#> GSM25626 1 0.4733 0.7943 0.800 0.004 0.196
#> GSM25627 1 0.5115 0.7936 0.796 0.016 0.188
#> GSM25628 3 0.5785 0.2579 0.332 0.000 0.668
#> GSM25629 1 0.6007 0.7866 0.768 0.048 0.184
#> GSM25630 1 0.3349 0.8239 0.888 0.004 0.108
#> GSM25631 2 0.2414 0.7901 0.020 0.940 0.040
#> GSM25632 1 0.4002 0.8080 0.840 0.000 0.160
#> GSM25633 1 0.3879 0.7507 0.848 0.152 0.000
#> GSM25634 1 0.1905 0.8157 0.956 0.016 0.028
#> GSM25635 2 0.7251 0.4698 0.348 0.612 0.040
#> GSM25656 3 0.1999 0.7180 0.036 0.012 0.952
#> GSM25657 1 0.4477 0.8165 0.864 0.068 0.068
#> GSM25658 1 0.4755 0.7962 0.808 0.008 0.184
#> GSM25659 3 0.7816 0.6323 0.084 0.288 0.628
#> GSM25660 2 0.6252 0.4740 0.344 0.648 0.008
#> GSM25661 1 0.5815 0.5167 0.692 0.304 0.004
#> GSM25662 3 0.3686 0.7737 0.000 0.140 0.860
#> GSM25663 3 0.3816 0.7722 0.000 0.148 0.852
#> GSM25680 2 0.0237 0.7962 0.000 0.996 0.004
#> GSM25681 2 0.0237 0.7962 0.000 0.996 0.004
#> GSM25682 2 0.5905 0.3274 0.000 0.648 0.352
#> GSM25683 3 0.3879 0.7723 0.000 0.152 0.848
#> GSM25684 3 0.4047 0.7736 0.004 0.148 0.848
#> GSM25685 3 0.0424 0.7274 0.008 0.000 0.992
#> GSM25686 3 0.5785 0.6638 0.000 0.332 0.668
#> GSM25687 2 0.5591 0.4481 0.000 0.696 0.304
#> GSM48664 3 0.9083 0.4021 0.320 0.160 0.520
#> GSM48665 2 0.6359 0.4659 0.364 0.628 0.008
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM25548 2 0.0000 0.821028 0.000 1.000 0.000 0.000
#> GSM25549 2 0.0000 0.821028 0.000 1.000 0.000 0.000
#> GSM25550 2 0.0712 0.819621 0.004 0.984 0.004 0.008
#> GSM25551 2 0.4464 0.658825 0.004 0.760 0.224 0.012
#> GSM25570 2 0.0000 0.821028 0.000 1.000 0.000 0.000
#> GSM25571 2 0.0000 0.821028 0.000 1.000 0.000 0.000
#> GSM25358 4 0.2382 0.766400 0.004 0.080 0.004 0.912
#> GSM25359 2 0.7407 0.269563 0.004 0.496 0.344 0.156
#> GSM25360 4 0.7924 0.120145 0.016 0.176 0.356 0.452
#> GSM25361 4 0.7159 0.541035 0.004 0.272 0.160 0.564
#> GSM25377 1 0.3760 0.693690 0.828 0.156 0.012 0.004
#> GSM25378 2 0.4283 0.757336 0.068 0.836 0.084 0.012
#> GSM25401 3 0.6349 0.291979 0.028 0.024 0.564 0.384
#> GSM25402 4 0.2089 0.746764 0.012 0.020 0.028 0.940
#> GSM25349 2 0.5769 0.566742 0.056 0.652 0.000 0.292
#> GSM25350 2 0.4257 0.754874 0.048 0.812 0.000 0.140
#> GSM25356 2 0.2774 0.801147 0.044 0.908 0.004 0.044
#> GSM25357 2 0.4796 0.734228 0.064 0.788 0.004 0.144
#> GSM25385 3 0.6216 -0.005742 0.408 0.028 0.548 0.016
#> GSM25386 4 0.5391 0.373593 0.004 0.012 0.380 0.604
#> GSM25399 1 0.3893 0.681361 0.796 0.000 0.196 0.008
#> GSM25400 1 0.8829 0.064245 0.372 0.060 0.372 0.196
#> GSM48659 4 0.2831 0.765847 0.004 0.120 0.000 0.876
#> GSM48660 4 0.3401 0.753903 0.008 0.152 0.000 0.840
#> GSM25409 2 0.2310 0.809757 0.008 0.920 0.004 0.068
#> GSM25410 3 0.7921 0.092626 0.016 0.172 0.440 0.372
#> GSM25426 4 0.5976 0.214009 0.008 0.024 0.452 0.516
#> GSM25427 2 0.4679 0.340559 0.352 0.648 0.000 0.000
#> GSM25540 3 0.6817 -0.087505 0.000 0.100 0.492 0.408
#> GSM25541 3 0.7307 0.107971 0.004 0.160 0.532 0.304
#> GSM25542 4 0.3172 0.748094 0.000 0.160 0.000 0.840
#> GSM25543 4 0.4360 0.666230 0.008 0.248 0.000 0.744
#> GSM25479 2 0.8141 -0.090083 0.268 0.428 0.292 0.012
#> GSM25480 2 0.1151 0.818734 0.008 0.968 0.024 0.000
#> GSM25481 2 0.2142 0.818501 0.016 0.928 0.000 0.056
#> GSM25482 2 0.3876 0.776794 0.124 0.836 0.000 0.040
#> GSM48654 4 0.3024 0.756778 0.000 0.148 0.000 0.852
#> GSM48650 4 0.5144 0.692131 0.068 0.168 0.004 0.760
#> GSM48651 4 0.1302 0.763541 0.000 0.044 0.000 0.956
#> GSM48652 2 0.4746 0.551726 0.004 0.712 0.008 0.276
#> GSM48653 4 0.2088 0.739606 0.004 0.004 0.064 0.928
#> GSM48662 2 0.3873 0.643837 0.000 0.772 0.000 0.228
#> GSM48663 4 0.3471 0.744111 0.060 0.072 0.000 0.868
#> GSM25524 3 0.1624 0.616776 0.028 0.000 0.952 0.020
#> GSM25525 3 0.6964 0.257906 0.228 0.188 0.584 0.000
#> GSM25526 3 0.0707 0.616888 0.020 0.000 0.980 0.000
#> GSM25527 3 0.6610 0.391372 0.124 0.196 0.664 0.016
#> GSM25528 3 0.6028 -0.197559 0.476 0.004 0.488 0.032
#> GSM25529 1 0.5602 0.209903 0.508 0.020 0.472 0.000
#> GSM25530 3 0.4661 0.326570 0.284 0.004 0.708 0.004
#> GSM25531 1 0.5760 0.248895 0.524 0.000 0.448 0.028
#> GSM48661 4 0.2443 0.747313 0.000 0.024 0.060 0.916
#> GSM25561 1 0.4704 0.677728 0.764 0.028 0.204 0.004
#> GSM25562 3 0.5949 0.397485 0.004 0.068 0.668 0.260
#> GSM25563 4 0.4804 0.390526 0.000 0.000 0.384 0.616
#> GSM25564 4 0.5774 0.107129 0.000 0.464 0.028 0.508
#> GSM25565 4 0.2831 0.767407 0.000 0.120 0.004 0.876
#> GSM25566 2 0.2555 0.809421 0.032 0.920 0.008 0.040
#> GSM25568 2 0.5045 0.529784 0.004 0.680 0.012 0.304
#> GSM25569 2 0.1118 0.819780 0.000 0.964 0.000 0.036
#> GSM25552 2 0.0817 0.818027 0.000 0.976 0.000 0.024
#> GSM25553 2 0.1557 0.813606 0.000 0.944 0.000 0.056
#> GSM25578 1 0.5949 0.564469 0.644 0.288 0.068 0.000
#> GSM25579 2 0.1271 0.821073 0.008 0.968 0.012 0.012
#> GSM25580 1 0.2281 0.737111 0.904 0.000 0.096 0.000
#> GSM25581 1 0.2053 0.742330 0.924 0.004 0.072 0.000
#> GSM48655 4 0.5980 0.262538 0.044 0.396 0.000 0.560
#> GSM48656 4 0.3047 0.767086 0.000 0.116 0.012 0.872
#> GSM48657 4 0.5055 0.698240 0.068 0.160 0.004 0.768
#> GSM48658 4 0.4967 0.727980 0.004 0.104 0.108 0.784
#> GSM25624 1 0.7110 0.429204 0.564 0.236 0.200 0.000
#> GSM25625 3 0.0592 0.617254 0.016 0.000 0.984 0.000
#> GSM25626 3 0.1820 0.616893 0.020 0.000 0.944 0.036
#> GSM25627 3 0.0188 0.615788 0.000 0.000 0.996 0.004
#> GSM25628 3 0.4697 0.241095 0.000 0.000 0.644 0.356
#> GSM25629 3 0.0927 0.612273 0.008 0.016 0.976 0.000
#> GSM25630 3 0.6064 0.400281 0.220 0.000 0.672 0.108
#> GSM25631 2 0.0779 0.819631 0.004 0.980 0.000 0.016
#> GSM25632 3 0.1209 0.613671 0.032 0.000 0.964 0.004
#> GSM25633 1 0.2142 0.745566 0.928 0.016 0.056 0.000
#> GSM25634 1 0.2125 0.742523 0.920 0.004 0.076 0.000
#> GSM25635 1 0.2443 0.739335 0.916 0.060 0.024 0.000
#> GSM25656 4 0.5451 0.205223 0.004 0.008 0.464 0.524
#> GSM25657 3 0.5576 -0.000931 0.444 0.020 0.536 0.000
#> GSM25658 3 0.0707 0.616888 0.020 0.000 0.980 0.000
#> GSM25659 4 0.6028 0.644659 0.004 0.236 0.084 0.676
#> GSM25660 1 0.5520 0.679447 0.744 0.172 0.072 0.012
#> GSM25661 1 0.4514 0.729843 0.812 0.072 0.112 0.004
#> GSM25662 4 0.1296 0.757878 0.004 0.028 0.004 0.964
#> GSM25663 4 0.1557 0.760858 0.000 0.056 0.000 0.944
#> GSM25680 2 0.0188 0.821062 0.000 0.996 0.000 0.004
#> GSM25681 2 0.0000 0.821028 0.000 1.000 0.000 0.000
#> GSM25682 2 0.6576 0.255851 0.068 0.516 0.004 0.412
#> GSM25683 4 0.2088 0.737581 0.064 0.004 0.004 0.928
#> GSM25684 4 0.2156 0.759274 0.004 0.060 0.008 0.928
#> GSM25685 4 0.1978 0.737032 0.004 0.000 0.068 0.928
#> GSM25686 4 0.5540 0.636018 0.068 0.208 0.004 0.720
#> GSM25687 2 0.6343 0.455554 0.068 0.596 0.004 0.332
#> GSM48664 1 0.4535 0.661865 0.816 0.024 0.032 0.128
#> GSM48665 1 0.2494 0.743163 0.916 0.048 0.036 0.000
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM25548 5 0.0000 0.7541 0.000 0.000 0.000 0.000 1.000
#> GSM25549 5 0.0000 0.7541 0.000 0.000 0.000 0.000 1.000
#> GSM25550 5 0.0404 0.7517 0.000 0.000 0.000 0.012 0.988
#> GSM25551 5 0.5314 0.4721 0.000 0.048 0.224 0.036 0.692
#> GSM25570 5 0.0000 0.7541 0.000 0.000 0.000 0.000 1.000
#> GSM25571 5 0.0000 0.7541 0.000 0.000 0.000 0.000 1.000
#> GSM25358 2 0.2166 0.6362 0.000 0.912 0.004 0.012 0.072
#> GSM25359 5 0.7176 -0.0226 0.000 0.176 0.352 0.036 0.436
#> GSM25360 2 0.7176 0.0726 0.340 0.468 0.056 0.000 0.136
#> GSM25361 2 0.6167 0.3216 0.004 0.576 0.140 0.004 0.276
#> GSM25377 1 0.6754 0.4755 0.472 0.000 0.388 0.048 0.092
#> GSM25378 5 0.3778 0.6589 0.124 0.000 0.024 0.028 0.824
#> GSM25401 1 0.7191 -0.3235 0.420 0.384 0.160 0.032 0.004
#> GSM25402 2 0.2830 0.6001 0.044 0.876 0.000 0.080 0.000
#> GSM25349 5 0.5873 -0.0159 0.000 0.112 0.000 0.348 0.540
#> GSM25350 5 0.4455 0.5601 0.000 0.068 0.000 0.188 0.744
#> GSM25356 5 0.3730 0.4515 0.000 0.000 0.000 0.288 0.712
#> GSM25357 5 0.5143 0.0746 0.000 0.048 0.000 0.368 0.584
#> GSM25385 1 0.4685 0.2864 0.724 0.012 0.232 0.024 0.008
#> GSM25386 2 0.5981 0.2199 0.180 0.628 0.180 0.012 0.000
#> GSM25399 1 0.6385 0.4221 0.556 0.008 0.200 0.236 0.000
#> GSM25400 1 0.5439 0.2424 0.700 0.196 0.060 0.000 0.044
#> GSM48659 2 0.1908 0.6308 0.000 0.908 0.000 0.000 0.092
#> GSM48660 2 0.4104 0.5566 0.000 0.788 0.000 0.088 0.124
#> GSM25409 5 0.3011 0.6770 0.000 0.016 0.000 0.140 0.844
#> GSM25410 1 0.7533 -0.1780 0.408 0.384 0.056 0.008 0.144
#> GSM25426 2 0.5251 -0.2988 0.000 0.504 0.456 0.036 0.004
#> GSM25427 5 0.5706 0.3427 0.204 0.000 0.152 0.004 0.640
#> GSM25540 3 0.6739 0.3278 0.024 0.412 0.460 0.012 0.092
#> GSM25541 3 0.7759 0.4580 0.060 0.308 0.464 0.020 0.148
#> GSM25542 2 0.3327 0.5858 0.000 0.828 0.000 0.028 0.144
#> GSM25543 2 0.4755 0.4359 0.000 0.696 0.000 0.060 0.244
#> GSM25479 1 0.6826 0.0416 0.440 0.004 0.116 0.028 0.412
#> GSM25480 5 0.0451 0.7527 0.000 0.008 0.004 0.000 0.988
#> GSM25481 5 0.2840 0.7012 0.004 0.012 0.004 0.108 0.872
#> GSM25482 5 0.5129 0.3195 0.020 0.000 0.024 0.328 0.628
#> GSM48654 2 0.2732 0.5961 0.000 0.840 0.000 0.000 0.160
#> GSM48650 2 0.5238 -0.3252 0.000 0.480 0.000 0.476 0.044
#> GSM48651 2 0.1915 0.6311 0.000 0.928 0.000 0.032 0.040
#> GSM48652 5 0.3661 0.4200 0.000 0.276 0.000 0.000 0.724
#> GSM48653 2 0.0162 0.6280 0.000 0.996 0.000 0.000 0.004
#> GSM48662 5 0.3305 0.5046 0.000 0.224 0.000 0.000 0.776
#> GSM48663 2 0.4897 -0.2307 0.000 0.516 0.000 0.460 0.024
#> GSM25524 1 0.5238 -0.4840 0.484 0.044 0.472 0.000 0.000
#> GSM25525 1 0.5334 0.1007 0.672 0.000 0.148 0.000 0.180
#> GSM25526 1 0.4450 -0.4559 0.508 0.004 0.488 0.000 0.000
#> GSM25527 1 0.6258 -0.0902 0.596 0.004 0.208 0.008 0.184
#> GSM25528 1 0.4489 0.3502 0.768 0.040 0.172 0.016 0.004
#> GSM25529 1 0.4626 0.3493 0.724 0.008 0.236 0.012 0.020
#> GSM25530 1 0.3491 0.0828 0.768 0.000 0.228 0.004 0.000
#> GSM25531 1 0.4010 0.3726 0.784 0.032 0.176 0.008 0.000
#> GSM48661 2 0.1582 0.6344 0.000 0.944 0.028 0.000 0.028
#> GSM25561 1 0.5223 0.4636 0.504 0.008 0.464 0.004 0.020
#> GSM25562 3 0.7909 0.5831 0.200 0.292 0.432 0.012 0.064
#> GSM25563 2 0.5388 0.0831 0.060 0.620 0.312 0.008 0.000
#> GSM25564 2 0.4437 0.0403 0.000 0.532 0.004 0.000 0.464
#> GSM25565 2 0.2536 0.6169 0.000 0.868 0.000 0.004 0.128
#> GSM25566 5 0.3491 0.5800 0.000 0.000 0.004 0.228 0.768
#> GSM25568 5 0.4492 0.3876 0.000 0.296 0.004 0.020 0.680
#> GSM25569 5 0.1211 0.7469 0.000 0.016 0.000 0.024 0.960
#> GSM25552 5 0.0162 0.7534 0.000 0.004 0.000 0.000 0.996
#> GSM25553 5 0.1197 0.7399 0.000 0.048 0.000 0.000 0.952
#> GSM25578 1 0.7136 0.4044 0.436 0.000 0.272 0.020 0.272
#> GSM25579 5 0.0960 0.7500 0.008 0.004 0.016 0.000 0.972
#> GSM25580 1 0.4708 0.4889 0.548 0.000 0.436 0.016 0.000
#> GSM25581 1 0.4897 0.4856 0.516 0.000 0.460 0.024 0.000
#> GSM48655 2 0.6707 -0.4709 0.000 0.388 0.000 0.368 0.244
#> GSM48656 2 0.2230 0.6249 0.000 0.884 0.000 0.000 0.116
#> GSM48657 4 0.5351 0.1511 0.000 0.464 0.000 0.484 0.052
#> GSM48658 2 0.4280 0.5840 0.000 0.796 0.076 0.016 0.112
#> GSM25624 1 0.6493 0.3697 0.520 0.000 0.248 0.004 0.228
#> GSM25625 3 0.4748 0.3746 0.492 0.016 0.492 0.000 0.000
#> GSM25626 1 0.5109 -0.4520 0.504 0.036 0.460 0.000 0.000
#> GSM25627 3 0.5546 0.4743 0.416 0.044 0.528 0.012 0.000
#> GSM25628 3 0.6553 0.5483 0.148 0.356 0.484 0.012 0.000
#> GSM25629 3 0.6092 0.4846 0.388 0.052 0.528 0.028 0.004
#> GSM25630 1 0.7176 -0.0781 0.556 0.116 0.116 0.212 0.000
#> GSM25631 5 0.0162 0.7534 0.000 0.004 0.000 0.000 0.996
#> GSM25632 1 0.4557 -0.4454 0.516 0.008 0.476 0.000 0.000
#> GSM25633 1 0.4897 0.4856 0.516 0.000 0.460 0.024 0.000
#> GSM25634 1 0.4897 0.4856 0.516 0.000 0.460 0.024 0.000
#> GSM25635 1 0.5153 0.4844 0.508 0.000 0.460 0.024 0.008
#> GSM25656 2 0.4735 -0.2825 0.000 0.524 0.460 0.016 0.000
#> GSM25657 1 0.4620 0.2520 0.612 0.000 0.372 0.004 0.012
#> GSM25658 1 0.4450 -0.4559 0.508 0.004 0.488 0.000 0.000
#> GSM25659 2 0.5257 0.4474 0.084 0.680 0.008 0.000 0.228
#> GSM25660 1 0.6908 0.4712 0.480 0.000 0.352 0.040 0.128
#> GSM25661 1 0.6095 0.4885 0.528 0.000 0.380 0.028 0.064
#> GSM25662 2 0.1168 0.6333 0.000 0.960 0.000 0.008 0.032
#> GSM25663 2 0.1628 0.6339 0.000 0.936 0.000 0.008 0.056
#> GSM25680 5 0.0000 0.7541 0.000 0.000 0.000 0.000 1.000
#> GSM25681 5 0.0000 0.7541 0.000 0.000 0.000 0.000 1.000
#> GSM25682 4 0.5736 0.3742 0.000 0.088 0.000 0.512 0.400
#> GSM25683 2 0.4307 -0.2579 0.000 0.504 0.000 0.496 0.000
#> GSM25684 2 0.1043 0.6347 0.000 0.960 0.000 0.000 0.040
#> GSM25685 2 0.0162 0.6264 0.000 0.996 0.004 0.000 0.000
#> GSM25686 4 0.6075 0.4307 0.000 0.356 0.000 0.512 0.132
#> GSM25687 4 0.5808 0.3959 0.000 0.096 0.000 0.512 0.392
#> GSM48664 1 0.6983 0.4488 0.444 0.100 0.412 0.024 0.020
#> GSM48665 1 0.5340 0.4827 0.500 0.000 0.460 0.024 0.016
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM25548 4 0.0000 0.7712 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM25549 4 0.0000 0.7712 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM25550 4 0.0291 0.7700 0.000 0.004 0.000 0.992 0.000 0.004
#> GSM25551 4 0.4428 0.3797 0.000 0.004 0.388 0.588 0.012 0.008
#> GSM25570 4 0.0000 0.7712 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM25571 4 0.0000 0.7712 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM25358 5 0.0790 0.6888 0.000 0.000 0.000 0.032 0.968 0.000
#> GSM25359 3 0.5808 0.0978 0.000 0.004 0.496 0.372 0.116 0.012
#> GSM25360 5 0.7120 0.1866 0.000 0.000 0.216 0.132 0.456 0.196
#> GSM25361 5 0.5771 0.2551 0.000 0.000 0.248 0.244 0.508 0.000
#> GSM25377 1 0.2373 0.5112 0.888 0.004 0.000 0.084 0.000 0.024
#> GSM25378 4 0.3798 0.6947 0.072 0.004 0.044 0.820 0.000 0.060
#> GSM25401 5 0.6437 -0.2014 0.004 0.012 0.368 0.000 0.376 0.240
#> GSM25402 5 0.2197 0.6496 0.000 0.044 0.000 0.000 0.900 0.056
#> GSM25349 4 0.5516 0.2220 0.000 0.368 0.000 0.516 0.108 0.008
#> GSM25350 4 0.4195 0.6292 0.000 0.188 0.000 0.740 0.064 0.008
#> GSM25356 4 0.3563 0.4791 0.000 0.336 0.000 0.664 0.000 0.000
#> GSM25357 4 0.4936 0.2021 0.000 0.408 0.000 0.536 0.048 0.008
#> GSM25385 1 0.6370 0.0780 0.420 0.004 0.356 0.008 0.004 0.208
#> GSM25386 5 0.5153 0.3186 0.000 0.000 0.288 0.000 0.592 0.120
#> GSM25399 6 0.4398 0.0000 0.228 0.028 0.024 0.000 0.004 0.716
#> GSM25400 1 0.7911 -0.0498 0.336 0.000 0.184 0.020 0.172 0.288
#> GSM48659 5 0.1387 0.6838 0.000 0.000 0.000 0.068 0.932 0.000
#> GSM48660 5 0.3324 0.6067 0.000 0.060 0.000 0.112 0.824 0.004
#> GSM25409 4 0.2976 0.7170 0.000 0.124 0.000 0.844 0.020 0.012
#> GSM25410 5 0.7593 -0.0345 0.004 0.000 0.232 0.144 0.352 0.268
#> GSM25426 3 0.3955 0.3821 0.000 0.004 0.668 0.000 0.316 0.012
#> GSM25427 4 0.3807 0.3105 0.368 0.000 0.000 0.628 0.000 0.004
#> GSM25540 3 0.4328 0.4054 0.000 0.000 0.672 0.040 0.284 0.004
#> GSM25541 3 0.4449 0.4277 0.000 0.004 0.712 0.088 0.196 0.000
#> GSM25542 5 0.2623 0.6264 0.000 0.016 0.000 0.132 0.852 0.000
#> GSM25543 5 0.4354 0.4549 0.000 0.052 0.000 0.236 0.704 0.008
#> GSM25479 4 0.7474 -0.1383 0.252 0.004 0.152 0.400 0.000 0.192
#> GSM25480 4 0.0363 0.7695 0.000 0.000 0.012 0.988 0.000 0.000
#> GSM25481 4 0.3768 0.6874 0.008 0.104 0.000 0.796 0.000 0.092
#> GSM25482 4 0.5764 0.4110 0.040 0.292 0.000 0.572 0.000 0.096
#> GSM48654 5 0.2219 0.6427 0.000 0.000 0.000 0.136 0.864 0.000
#> GSM48650 2 0.4860 0.4748 0.000 0.516 0.000 0.040 0.436 0.008
#> GSM48651 5 0.0909 0.6863 0.000 0.012 0.000 0.020 0.968 0.000
#> GSM48652 4 0.3330 0.4832 0.000 0.000 0.000 0.716 0.284 0.000
#> GSM48653 5 0.0291 0.6809 0.000 0.000 0.004 0.004 0.992 0.000
#> GSM48662 4 0.3050 0.5543 0.000 0.000 0.000 0.764 0.236 0.000
#> GSM48663 2 0.4356 0.4768 0.000 0.548 0.000 0.016 0.432 0.004
#> GSM25524 3 0.2930 0.4616 0.020 0.000 0.856 0.000 0.020 0.104
#> GSM25525 3 0.7563 -0.0792 0.200 0.000 0.332 0.180 0.000 0.288
#> GSM25526 3 0.3448 0.4089 0.004 0.000 0.716 0.000 0.000 0.280
#> GSM25527 3 0.7086 0.0783 0.084 0.000 0.432 0.196 0.004 0.284
#> GSM25528 1 0.6189 0.1423 0.468 0.000 0.256 0.000 0.012 0.264
#> GSM25529 1 0.5671 0.1873 0.500 0.000 0.364 0.008 0.000 0.128
#> GSM25530 3 0.6041 -0.0571 0.272 0.000 0.416 0.000 0.000 0.312
#> GSM25531 1 0.6306 0.1756 0.496 0.000 0.220 0.000 0.028 0.256
#> GSM48661 5 0.1625 0.6712 0.000 0.000 0.060 0.012 0.928 0.000
#> GSM25561 1 0.3649 0.4526 0.784 0.000 0.180 0.016 0.004 0.016
#> GSM25562 3 0.4587 0.4464 0.004 0.000 0.712 0.044 0.216 0.024
#> GSM25563 3 0.3851 0.0493 0.000 0.000 0.540 0.000 0.460 0.000
#> GSM25564 5 0.4083 0.0944 0.000 0.000 0.008 0.460 0.532 0.000
#> GSM25565 5 0.1858 0.6734 0.000 0.000 0.000 0.092 0.904 0.004
#> GSM25566 4 0.3368 0.6341 0.000 0.232 0.000 0.756 0.000 0.012
#> GSM25568 4 0.4087 0.4577 0.000 0.004 0.008 0.668 0.312 0.008
#> GSM25569 4 0.1321 0.7637 0.000 0.024 0.000 0.952 0.020 0.004
#> GSM25552 4 0.0000 0.7712 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM25553 4 0.1007 0.7603 0.000 0.000 0.000 0.956 0.044 0.000
#> GSM25578 1 0.4643 0.3063 0.672 0.000 0.012 0.260 0.000 0.056
#> GSM25579 4 0.0976 0.7663 0.008 0.000 0.016 0.968 0.000 0.008
#> GSM25580 1 0.0993 0.5507 0.964 0.000 0.024 0.000 0.000 0.012
#> GSM25581 1 0.0146 0.5518 0.996 0.000 0.004 0.000 0.000 0.000
#> GSM48655 2 0.5971 0.4630 0.000 0.424 0.000 0.232 0.344 0.000
#> GSM48656 5 0.1714 0.6769 0.000 0.000 0.000 0.092 0.908 0.000
#> GSM48657 2 0.4909 0.5282 0.000 0.552 0.000 0.048 0.392 0.008
#> GSM48658 5 0.4222 0.5690 0.000 0.004 0.156 0.084 0.752 0.004
#> GSM25624 1 0.6185 0.1504 0.580 0.000 0.112 0.220 0.000 0.088
#> GSM25625 3 0.3136 0.4391 0.004 0.000 0.768 0.000 0.000 0.228
#> GSM25626 3 0.3833 0.4078 0.004 0.000 0.708 0.000 0.016 0.272
#> GSM25627 3 0.0260 0.4719 0.000 0.000 0.992 0.000 0.008 0.000
#> GSM25628 3 0.2854 0.4623 0.000 0.000 0.792 0.000 0.208 0.000
#> GSM25629 3 0.0551 0.4711 0.000 0.004 0.984 0.000 0.008 0.004
#> GSM25630 2 0.7194 -0.5169 0.008 0.376 0.208 0.000 0.072 0.336
#> GSM25631 4 0.0000 0.7712 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM25632 3 0.3629 0.4119 0.016 0.000 0.724 0.000 0.000 0.260
#> GSM25633 1 0.0291 0.5520 0.992 0.000 0.004 0.000 0.000 0.004
#> GSM25634 1 0.0146 0.5518 0.996 0.000 0.004 0.000 0.000 0.000
#> GSM25635 1 0.0146 0.5504 0.996 0.000 0.000 0.004 0.000 0.000
#> GSM25656 3 0.3531 0.3709 0.000 0.000 0.672 0.000 0.328 0.000
#> GSM25657 1 0.5741 0.0331 0.472 0.000 0.396 0.012 0.000 0.120
#> GSM25658 3 0.3383 0.4176 0.004 0.000 0.728 0.000 0.000 0.268
#> GSM25659 5 0.4959 0.4674 0.000 0.000 0.044 0.224 0.680 0.052
#> GSM25660 1 0.3466 0.4764 0.816 0.004 0.000 0.096 0.000 0.084
#> GSM25661 1 0.3133 0.5049 0.852 0.000 0.016 0.064 0.000 0.068
#> GSM25662 5 0.0458 0.6854 0.000 0.000 0.000 0.016 0.984 0.000
#> GSM25663 5 0.0632 0.6866 0.000 0.000 0.000 0.024 0.976 0.000
#> GSM25680 4 0.0000 0.7712 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM25681 4 0.0000 0.7712 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM25682 2 0.4747 0.2065 0.000 0.584 0.000 0.356 0.060 0.000
#> GSM25683 2 0.3789 0.4947 0.000 0.584 0.000 0.000 0.416 0.000
#> GSM25684 5 0.0458 0.6854 0.000 0.000 0.000 0.016 0.984 0.000
#> GSM25685 5 0.0260 0.6786 0.000 0.000 0.008 0.000 0.992 0.000
#> GSM25686 2 0.5164 0.5648 0.000 0.584 0.000 0.116 0.300 0.000
#> GSM25687 2 0.4786 0.2184 0.000 0.584 0.000 0.352 0.064 0.000
#> GSM48664 1 0.2536 0.4476 0.880 0.004 0.004 0.020 0.092 0.000
#> GSM48665 1 0.0260 0.5482 0.992 0.000 0.000 0.008 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 genotype/variation(p) k
#> SD:pam 91 1.05e-04 2
#> SD:pam 83 1.92e-07 3
#> SD:pam 71 3.86e-06 4
#> SD:pam 37 1.81e-01 5
#> SD:pam 45 1.72e-01 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 100 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 0.720 0.896 0.920 0.4765 0.495 0.495
#> 3 3 0.418 0.608 0.741 0.2448 0.867 0.736
#> 4 4 0.508 0.702 0.801 0.1577 0.897 0.748
#> 5 5 0.746 0.730 0.864 0.1348 0.854 0.574
#> 6 6 0.709 0.677 0.816 0.0399 0.922 0.672
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
#> GSM25548 2 0.2948 0.906 0.052 0.948
#> GSM25549 2 0.3431 0.904 0.064 0.936
#> GSM25550 2 0.3584 0.903 0.068 0.932
#> GSM25551 2 0.0672 0.905 0.008 0.992
#> GSM25570 2 0.3584 0.903 0.068 0.932
#> GSM25571 2 0.3274 0.905 0.060 0.940
#> GSM25358 1 0.3274 0.963 0.940 0.060
#> GSM25359 2 0.8713 0.663 0.292 0.708
#> GSM25360 1 0.2778 0.971 0.952 0.048
#> GSM25361 2 0.9833 0.391 0.424 0.576
#> GSM25377 1 0.2778 0.971 0.952 0.048
#> GSM25378 1 0.2778 0.971 0.952 0.048
#> GSM25401 1 0.3879 0.949 0.924 0.076
#> GSM25402 1 0.2778 0.971 0.952 0.048
#> GSM25349 2 0.0376 0.903 0.004 0.996
#> GSM25350 2 0.0000 0.901 0.000 1.000
#> GSM25356 1 0.2948 0.968 0.948 0.052
#> GSM25357 2 0.2778 0.907 0.048 0.952
#> GSM25385 1 0.2778 0.971 0.952 0.048
#> GSM25386 1 0.2778 0.971 0.952 0.048
#> GSM25399 1 0.2778 0.971 0.952 0.048
#> GSM25400 1 0.2778 0.971 0.952 0.048
#> GSM48659 2 0.2043 0.907 0.032 0.968
#> GSM48660 2 0.0000 0.901 0.000 1.000
#> GSM25409 2 0.0938 0.906 0.012 0.988
#> GSM25410 1 0.2778 0.971 0.952 0.048
#> GSM25426 2 0.3584 0.903 0.068 0.932
#> GSM25427 1 0.3114 0.966 0.944 0.056
#> GSM25540 2 0.9522 0.516 0.372 0.628
#> GSM25541 2 0.9944 0.296 0.456 0.544
#> GSM25542 2 0.8327 0.705 0.264 0.736
#> GSM25543 2 0.8555 0.681 0.280 0.720
#> GSM25479 1 0.0000 0.949 1.000 0.000
#> GSM25480 1 0.0672 0.954 0.992 0.008
#> GSM25481 1 0.7453 0.759 0.788 0.212
#> GSM25482 1 0.7528 0.751 0.784 0.216
#> GSM48654 2 0.0938 0.906 0.012 0.988
#> GSM48650 2 0.3584 0.903 0.068 0.932
#> GSM48651 2 0.0672 0.905 0.008 0.992
#> GSM48652 2 0.0672 0.905 0.008 0.992
#> GSM48653 2 0.0672 0.905 0.008 0.992
#> GSM48662 2 0.0672 0.905 0.008 0.992
#> GSM48663 2 0.3274 0.904 0.060 0.940
#> GSM25524 1 0.2778 0.971 0.952 0.048
#> GSM25525 1 0.0376 0.951 0.996 0.004
#> GSM25526 1 0.2778 0.971 0.952 0.048
#> GSM25527 1 0.0376 0.951 0.996 0.004
#> GSM25528 1 0.2778 0.971 0.952 0.048
#> GSM25529 1 0.0376 0.951 0.996 0.004
#> GSM25530 1 0.2778 0.971 0.952 0.048
#> GSM25531 1 0.2603 0.969 0.956 0.044
#> GSM48661 2 0.4815 0.880 0.104 0.896
#> GSM25561 1 0.2778 0.971 0.952 0.048
#> GSM25562 1 0.2778 0.971 0.952 0.048
#> GSM25563 1 0.2778 0.971 0.952 0.048
#> GSM25564 1 0.4690 0.923 0.900 0.100
#> GSM25565 2 0.0672 0.905 0.008 0.992
#> GSM25566 2 0.0672 0.905 0.008 0.992
#> GSM25568 2 0.9129 0.601 0.328 0.672
#> GSM25569 2 0.0672 0.905 0.008 0.992
#> GSM25552 2 0.3584 0.903 0.068 0.932
#> GSM25553 2 0.4298 0.893 0.088 0.912
#> GSM25578 1 0.0000 0.949 1.000 0.000
#> GSM25579 1 0.2948 0.968 0.948 0.052
#> GSM25580 1 0.0000 0.949 1.000 0.000
#> GSM25581 1 0.0000 0.949 1.000 0.000
#> GSM48655 2 0.0000 0.901 0.000 1.000
#> GSM48656 2 0.2778 0.906 0.048 0.952
#> GSM48657 2 0.0000 0.901 0.000 1.000
#> GSM48658 2 0.5408 0.865 0.124 0.876
#> GSM25624 1 0.0000 0.949 1.000 0.000
#> GSM25625 1 0.2778 0.971 0.952 0.048
#> GSM25626 1 0.2778 0.971 0.952 0.048
#> GSM25627 1 0.3879 0.949 0.924 0.076
#> GSM25628 1 0.2778 0.971 0.952 0.048
#> GSM25629 2 0.9909 0.332 0.444 0.556
#> GSM25630 1 0.2778 0.971 0.952 0.048
#> GSM25631 2 0.6623 0.820 0.172 0.828
#> GSM25632 1 0.2778 0.971 0.952 0.048
#> GSM25633 1 0.0000 0.949 1.000 0.000
#> GSM25634 1 0.0000 0.949 1.000 0.000
#> GSM25635 1 0.0000 0.949 1.000 0.000
#> GSM25656 1 0.4431 0.931 0.908 0.092
#> GSM25657 1 0.0938 0.956 0.988 0.012
#> GSM25658 1 0.2778 0.971 0.952 0.048
#> GSM25659 1 0.2948 0.968 0.948 0.052
#> GSM25660 1 0.0000 0.949 1.000 0.000
#> GSM25661 1 0.0000 0.949 1.000 0.000
#> GSM25662 2 0.0672 0.905 0.008 0.992
#> GSM25663 2 0.5946 0.847 0.144 0.856
#> GSM25680 2 0.3584 0.903 0.068 0.932
#> GSM25681 2 0.3733 0.901 0.072 0.928
#> GSM25682 2 0.0000 0.901 0.000 1.000
#> GSM25683 2 0.0672 0.905 0.008 0.992
#> GSM25684 2 0.0672 0.905 0.008 0.992
#> GSM25685 2 0.3584 0.903 0.068 0.932
#> GSM25686 2 0.0000 0.901 0.000 1.000
#> GSM25687 2 0.0000 0.901 0.000 1.000
#> GSM48664 1 0.2778 0.971 0.952 0.048
#> GSM48665 1 0.2778 0.971 0.952 0.048
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM25548 2 0.5763 0.765 0.008 0.716 0.276
#> GSM25549 2 0.5956 0.767 0.016 0.720 0.264
#> GSM25550 2 0.5473 0.698 0.052 0.808 0.140
#> GSM25551 2 0.6527 0.764 0.020 0.660 0.320
#> GSM25570 2 0.6662 0.748 0.052 0.716 0.232
#> GSM25571 2 0.5992 0.767 0.016 0.716 0.268
#> GSM25358 1 0.8126 0.412 0.564 0.356 0.080
#> GSM25359 2 0.4269 0.541 0.076 0.872 0.052
#> GSM25360 3 0.9951 0.916 0.296 0.324 0.380
#> GSM25361 2 0.4351 0.333 0.168 0.828 0.004
#> GSM25377 1 0.9133 0.443 0.508 0.332 0.160
#> GSM25378 1 0.9133 0.442 0.508 0.332 0.160
#> GSM25401 3 0.8628 0.664 0.116 0.340 0.544
#> GSM25402 1 0.9765 0.238 0.424 0.336 0.240
#> GSM25349 2 0.6905 0.732 0.016 0.544 0.440
#> GSM25350 2 0.6476 0.735 0.004 0.548 0.448
#> GSM25356 1 0.9229 0.437 0.496 0.336 0.168
#> GSM25357 2 0.6875 0.631 0.056 0.700 0.244
#> GSM25385 3 0.9888 0.944 0.272 0.328 0.400
#> GSM25386 3 0.9806 0.939 0.252 0.328 0.420
#> GSM25399 1 0.9133 0.439 0.508 0.332 0.160
#> GSM25400 1 0.6726 0.413 0.644 0.332 0.024
#> GSM48659 2 0.5327 0.765 0.000 0.728 0.272
#> GSM48660 2 0.6215 0.741 0.000 0.572 0.428
#> GSM25409 2 0.7425 0.763 0.052 0.620 0.328
#> GSM25410 3 0.9858 0.945 0.264 0.328 0.408
#> GSM25426 2 0.4642 0.511 0.060 0.856 0.084
#> GSM25427 1 0.9106 0.443 0.508 0.336 0.156
#> GSM25540 2 0.4139 0.420 0.124 0.860 0.016
#> GSM25541 2 0.4293 0.348 0.164 0.832 0.004
#> GSM25542 2 0.4269 0.643 0.052 0.872 0.076
#> GSM25543 2 0.3325 0.529 0.076 0.904 0.020
#> GSM25479 1 0.0661 0.575 0.988 0.004 0.008
#> GSM25480 1 0.2280 0.580 0.940 0.052 0.008
#> GSM25481 1 0.9318 0.413 0.476 0.352 0.172
#> GSM25482 1 0.9318 0.413 0.476 0.352 0.172
#> GSM48654 2 0.5397 0.764 0.000 0.720 0.280
#> GSM48650 2 0.4725 0.509 0.060 0.852 0.088
#> GSM48651 2 0.5678 0.754 0.000 0.684 0.316
#> GSM48652 2 0.5678 0.754 0.000 0.684 0.316
#> GSM48653 2 0.5902 0.756 0.004 0.680 0.316
#> GSM48662 2 0.5678 0.754 0.000 0.684 0.316
#> GSM48663 2 0.4291 0.553 0.008 0.840 0.152
#> GSM25524 3 0.9894 0.941 0.276 0.324 0.400
#> GSM25525 1 0.2496 0.579 0.928 0.068 0.004
#> GSM25526 3 0.9865 0.945 0.268 0.324 0.408
#> GSM25527 1 0.2486 0.579 0.932 0.060 0.008
#> GSM25528 1 0.9951 -0.717 0.380 0.324 0.296
#> GSM25529 1 0.3349 0.568 0.888 0.108 0.004
#> GSM25530 3 0.9931 0.928 0.288 0.324 0.388
#> GSM25531 1 0.7285 0.305 0.632 0.320 0.048
#> GSM48661 2 0.2200 0.660 0.004 0.940 0.056
#> GSM25561 1 0.9602 -0.450 0.460 0.320 0.220
#> GSM25562 1 0.5591 0.430 0.696 0.304 0.000
#> GSM25563 3 0.9833 0.943 0.260 0.324 0.416
#> GSM25564 2 0.6442 -0.401 0.432 0.564 0.004
#> GSM25565 2 0.7013 0.765 0.036 0.640 0.324
#> GSM25566 2 0.6448 0.762 0.016 0.656 0.328
#> GSM25568 2 0.2774 0.534 0.072 0.920 0.008
#> GSM25569 2 0.5678 0.754 0.000 0.684 0.316
#> GSM25552 2 0.2846 0.571 0.056 0.924 0.020
#> GSM25553 2 0.2590 0.540 0.072 0.924 0.004
#> GSM25578 1 0.1289 0.566 0.968 0.000 0.032
#> GSM25579 1 0.6026 0.375 0.624 0.376 0.000
#> GSM25580 1 0.1289 0.570 0.968 0.000 0.032
#> GSM25581 1 0.1163 0.568 0.972 0.000 0.028
#> GSM48655 2 0.6235 0.739 0.000 0.564 0.436
#> GSM48656 2 0.5812 0.767 0.012 0.724 0.264
#> GSM48657 2 0.6235 0.739 0.000 0.564 0.436
#> GSM48658 2 0.4834 0.752 0.004 0.792 0.204
#> GSM25624 1 0.1031 0.581 0.976 0.024 0.000
#> GSM25625 3 0.9865 0.945 0.268 0.324 0.408
#> GSM25626 3 0.9797 0.938 0.252 0.324 0.424
#> GSM25627 3 0.9789 0.886 0.236 0.368 0.396
#> GSM25628 3 0.9745 0.916 0.232 0.348 0.420
#> GSM25629 2 0.8666 -0.502 0.152 0.584 0.264
#> GSM25630 3 0.9894 0.941 0.276 0.324 0.400
#> GSM25631 2 0.2400 0.550 0.064 0.932 0.004
#> GSM25632 3 0.9880 0.944 0.272 0.324 0.404
#> GSM25633 1 0.1289 0.566 0.968 0.000 0.032
#> GSM25634 1 0.1289 0.566 0.968 0.000 0.032
#> GSM25635 1 0.0592 0.573 0.988 0.000 0.012
#> GSM25656 3 0.9745 0.916 0.232 0.348 0.420
#> GSM25657 1 0.5061 0.513 0.784 0.208 0.008
#> GSM25658 3 0.9901 0.940 0.276 0.328 0.396
#> GSM25659 1 0.6209 0.377 0.628 0.368 0.004
#> GSM25660 1 0.1315 0.579 0.972 0.020 0.008
#> GSM25661 1 0.1163 0.568 0.972 0.000 0.028
#> GSM25662 2 0.5497 0.763 0.000 0.708 0.292
#> GSM25663 2 0.5681 0.762 0.016 0.748 0.236
#> GSM25680 2 0.5216 0.765 0.000 0.740 0.260
#> GSM25681 2 0.1781 0.605 0.020 0.960 0.020
#> GSM25682 2 0.6267 0.733 0.000 0.548 0.452
#> GSM25683 2 0.7121 0.741 0.024 0.548 0.428
#> GSM25684 2 0.5560 0.760 0.000 0.700 0.300
#> GSM25685 2 0.2066 0.556 0.060 0.940 0.000
#> GSM25686 2 0.6274 0.731 0.000 0.544 0.456
#> GSM25687 2 0.6274 0.731 0.000 0.544 0.456
#> GSM48664 1 0.9090 0.446 0.512 0.332 0.156
#> GSM48665 1 0.8013 0.443 0.588 0.332 0.080
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM25548 2 0.1743 0.7779 0.000 0.940 0.004 0.056
#> GSM25549 2 0.1970 0.7786 0.000 0.932 0.008 0.060
#> GSM25550 2 0.2773 0.7838 0.000 0.900 0.028 0.072
#> GSM25551 2 0.3037 0.7839 0.000 0.880 0.020 0.100
#> GSM25570 2 0.2335 0.7811 0.000 0.920 0.020 0.060
#> GSM25571 2 0.1824 0.7772 0.000 0.936 0.004 0.060
#> GSM25358 2 0.9425 0.1612 0.236 0.404 0.124 0.236
#> GSM25359 2 0.7456 0.6601 0.096 0.644 0.104 0.156
#> GSM25360 3 0.3743 0.7060 0.160 0.000 0.824 0.016
#> GSM25361 2 0.8207 0.4192 0.136 0.544 0.248 0.072
#> GSM25377 4 0.6157 0.7714 0.232 0.000 0.108 0.660
#> GSM25378 4 0.6877 0.7398 0.280 0.008 0.116 0.596
#> GSM25401 4 0.6904 0.5065 0.008 0.160 0.212 0.620
#> GSM25402 4 0.7445 0.5994 0.052 0.132 0.192 0.624
#> GSM25349 2 0.5696 0.6371 0.012 0.608 0.016 0.364
#> GSM25350 2 0.5285 0.6558 0.012 0.632 0.004 0.352
#> GSM25356 4 0.5857 0.7790 0.172 0.004 0.112 0.712
#> GSM25357 2 0.6936 0.5291 0.012 0.496 0.076 0.416
#> GSM25385 3 0.1022 0.8323 0.032 0.000 0.968 0.000
#> GSM25386 3 0.0000 0.8337 0.000 0.000 1.000 0.000
#> GSM25399 4 0.6248 0.7569 0.252 0.000 0.104 0.644
#> GSM25400 4 0.7088 0.5775 0.392 0.000 0.128 0.480
#> GSM48659 2 0.0937 0.7857 0.000 0.976 0.012 0.012
#> GSM48660 2 0.4978 0.6800 0.012 0.664 0.000 0.324
#> GSM25409 2 0.4946 0.7384 0.012 0.764 0.032 0.192
#> GSM25410 3 0.0188 0.8330 0.000 0.000 0.996 0.004
#> GSM25426 2 0.7293 0.5789 0.016 0.528 0.108 0.348
#> GSM25427 4 0.6333 0.7787 0.232 0.004 0.108 0.656
#> GSM25540 2 0.7972 0.2419 0.100 0.476 0.372 0.052
#> GSM25541 2 0.8028 0.3628 0.120 0.516 0.312 0.052
#> GSM25542 2 0.6455 0.7189 0.060 0.716 0.092 0.132
#> GSM25543 2 0.6760 0.6866 0.084 0.700 0.096 0.120
#> GSM25479 1 0.1109 0.8516 0.968 0.000 0.028 0.004
#> GSM25480 1 0.1576 0.8457 0.948 0.000 0.048 0.004
#> GSM25481 4 0.5602 0.7543 0.104 0.028 0.104 0.764
#> GSM25482 4 0.5714 0.7593 0.112 0.028 0.104 0.756
#> GSM48654 2 0.0657 0.7843 0.000 0.984 0.004 0.012
#> GSM48650 2 0.7173 0.5876 0.016 0.544 0.100 0.340
#> GSM48651 2 0.1716 0.7839 0.000 0.936 0.000 0.064
#> GSM48652 2 0.1474 0.7845 0.000 0.948 0.000 0.052
#> GSM48653 2 0.2021 0.7885 0.000 0.932 0.012 0.056
#> GSM48662 2 0.1022 0.7844 0.000 0.968 0.000 0.032
#> GSM48663 2 0.6744 0.5883 0.012 0.544 0.068 0.376
#> GSM25524 3 0.1302 0.8283 0.044 0.000 0.956 0.000
#> GSM25525 1 0.2888 0.7969 0.872 0.000 0.124 0.004
#> GSM25526 3 0.0000 0.8337 0.000 0.000 1.000 0.000
#> GSM25527 1 0.1940 0.8344 0.924 0.000 0.076 0.000
#> GSM25528 3 0.3444 0.6920 0.184 0.000 0.816 0.000
#> GSM25529 1 0.2888 0.7969 0.872 0.000 0.124 0.004
#> GSM25530 3 0.1557 0.8207 0.056 0.000 0.944 0.000
#> GSM25531 3 0.4972 0.0419 0.456 0.000 0.544 0.000
#> GSM48661 2 0.2830 0.7708 0.000 0.900 0.060 0.040
#> GSM25561 3 0.4543 0.4569 0.324 0.000 0.676 0.000
#> GSM25562 1 0.6212 0.5132 0.684 0.012 0.212 0.092
#> GSM25563 3 0.0592 0.8369 0.016 0.000 0.984 0.000
#> GSM25564 2 0.8068 0.4036 0.228 0.552 0.168 0.052
#> GSM25565 2 0.4137 0.7718 0.008 0.824 0.028 0.140
#> GSM25566 2 0.3366 0.7822 0.008 0.872 0.020 0.100
#> GSM25568 2 0.5223 0.6981 0.088 0.796 0.068 0.048
#> GSM25569 2 0.0592 0.7830 0.000 0.984 0.000 0.016
#> GSM25552 2 0.2623 0.7825 0.000 0.908 0.028 0.064
#> GSM25553 2 0.5298 0.7400 0.060 0.792 0.056 0.092
#> GSM25578 1 0.0817 0.8496 0.976 0.000 0.024 0.000
#> GSM25579 1 0.6135 0.5566 0.724 0.112 0.136 0.028
#> GSM25580 1 0.0921 0.8518 0.972 0.000 0.028 0.000
#> GSM25581 1 0.0817 0.8496 0.976 0.000 0.024 0.000
#> GSM48655 2 0.4914 0.6869 0.012 0.676 0.000 0.312
#> GSM48656 2 0.1297 0.7856 0.000 0.964 0.020 0.016
#> GSM48657 2 0.5018 0.6747 0.012 0.656 0.000 0.332
#> GSM48658 2 0.2224 0.7771 0.000 0.928 0.040 0.032
#> GSM25624 1 0.1302 0.8495 0.956 0.000 0.044 0.000
#> GSM25625 3 0.0592 0.8369 0.016 0.000 0.984 0.000
#> GSM25626 3 0.0000 0.8337 0.000 0.000 1.000 0.000
#> GSM25627 3 0.6037 0.2663 0.000 0.304 0.628 0.068
#> GSM25628 3 0.0188 0.8331 0.000 0.000 0.996 0.004
#> GSM25629 3 0.6320 0.1212 0.008 0.360 0.580 0.052
#> GSM25630 3 0.1022 0.8332 0.032 0.000 0.968 0.000
#> GSM25631 2 0.3169 0.7706 0.004 0.884 0.084 0.028
#> GSM25632 3 0.0817 0.8360 0.024 0.000 0.976 0.000
#> GSM25633 1 0.1022 0.8504 0.968 0.000 0.032 0.000
#> GSM25634 1 0.0817 0.8496 0.976 0.000 0.024 0.000
#> GSM25635 1 0.0921 0.8519 0.972 0.000 0.028 0.000
#> GSM25656 3 0.0524 0.8287 0.000 0.008 0.988 0.004
#> GSM25657 1 0.4356 0.5788 0.708 0.000 0.292 0.000
#> GSM25658 3 0.0524 0.8363 0.008 0.000 0.988 0.004
#> GSM25659 1 0.8424 0.0403 0.452 0.248 0.268 0.032
#> GSM25660 1 0.1305 0.8514 0.960 0.000 0.036 0.004
#> GSM25661 1 0.0817 0.8496 0.976 0.000 0.024 0.000
#> GSM25662 2 0.1722 0.7877 0.000 0.944 0.008 0.048
#> GSM25663 2 0.1677 0.7811 0.000 0.948 0.040 0.012
#> GSM25680 2 0.1452 0.7795 0.000 0.956 0.008 0.036
#> GSM25681 2 0.2021 0.7806 0.000 0.936 0.024 0.040
#> GSM25682 2 0.4999 0.6761 0.012 0.660 0.000 0.328
#> GSM25683 2 0.5662 0.6753 0.012 0.652 0.024 0.312
#> GSM25684 2 0.1022 0.7847 0.000 0.968 0.000 0.032
#> GSM25685 2 0.5976 0.7060 0.004 0.700 0.112 0.184
#> GSM25686 2 0.5057 0.6678 0.012 0.648 0.000 0.340
#> GSM25687 2 0.5018 0.6731 0.012 0.656 0.000 0.332
#> GSM48664 4 0.6375 0.7455 0.272 0.000 0.104 0.624
#> GSM48665 4 0.6735 0.5985 0.388 0.000 0.096 0.516
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM25548 5 0.1082 0.7724 0.000 0.008 0.000 0.028 0.964
#> GSM25549 5 0.0794 0.7699 0.000 0.000 0.000 0.028 0.972
#> GSM25550 5 0.1082 0.7719 0.000 0.000 0.008 0.028 0.964
#> GSM25551 2 0.4416 0.4159 0.000 0.632 0.012 0.000 0.356
#> GSM25570 5 0.0794 0.7699 0.000 0.000 0.000 0.028 0.972
#> GSM25571 5 0.0794 0.7699 0.000 0.000 0.000 0.028 0.972
#> GSM25358 4 0.7589 0.1734 0.056 0.076 0.044 0.452 0.372
#> GSM25359 5 0.4148 0.6383 0.000 0.216 0.028 0.004 0.752
#> GSM25360 3 0.2763 0.8213 0.148 0.000 0.848 0.004 0.000
#> GSM25361 5 0.4181 0.6005 0.016 0.000 0.240 0.008 0.736
#> GSM25377 4 0.1280 0.8643 0.008 0.024 0.008 0.960 0.000
#> GSM25378 4 0.1612 0.8649 0.016 0.024 0.012 0.948 0.000
#> GSM25401 4 0.2859 0.8175 0.000 0.056 0.068 0.876 0.000
#> GSM25402 4 0.2221 0.8396 0.000 0.036 0.052 0.912 0.000
#> GSM25349 2 0.0566 0.8193 0.000 0.984 0.004 0.000 0.012
#> GSM25350 2 0.0404 0.8195 0.000 0.988 0.000 0.000 0.012
#> GSM25356 4 0.1280 0.8643 0.008 0.024 0.008 0.960 0.000
#> GSM25357 2 0.2206 0.7821 0.000 0.912 0.016 0.068 0.004
#> GSM25385 3 0.2011 0.9183 0.088 0.000 0.908 0.004 0.000
#> GSM25386 3 0.1270 0.9300 0.052 0.000 0.948 0.000 0.000
#> GSM25399 4 0.1498 0.8654 0.016 0.024 0.008 0.952 0.000
#> GSM25400 4 0.5560 0.1712 0.440 0.024 0.028 0.508 0.000
#> GSM48659 5 0.3039 0.7024 0.000 0.152 0.012 0.000 0.836
#> GSM48660 2 0.0290 0.8199 0.000 0.992 0.000 0.000 0.008
#> GSM25409 5 0.4883 0.1898 0.000 0.464 0.016 0.004 0.516
#> GSM25410 3 0.1341 0.9285 0.056 0.000 0.944 0.000 0.000
#> GSM25426 2 0.3717 0.7420 0.000 0.836 0.024 0.040 0.100
#> GSM25427 4 0.1393 0.8652 0.012 0.024 0.008 0.956 0.000
#> GSM25540 5 0.4639 0.4692 0.000 0.012 0.344 0.008 0.636
#> GSM25541 5 0.4795 0.5891 0.032 0.008 0.240 0.008 0.712
#> GSM25542 5 0.4589 0.5159 0.000 0.316 0.020 0.004 0.660
#> GSM25543 5 0.3851 0.6957 0.000 0.164 0.036 0.004 0.796
#> GSM25479 1 0.0000 0.8951 1.000 0.000 0.000 0.000 0.000
#> GSM25480 1 0.0955 0.8923 0.968 0.000 0.028 0.000 0.004
#> GSM25481 4 0.1569 0.8586 0.004 0.044 0.008 0.944 0.000
#> GSM25482 4 0.1569 0.8586 0.004 0.044 0.008 0.944 0.000
#> GSM48654 5 0.3171 0.6820 0.000 0.176 0.008 0.000 0.816
#> GSM48650 2 0.1960 0.7935 0.000 0.928 0.020 0.048 0.004
#> GSM48651 2 0.4522 0.2383 0.000 0.552 0.008 0.000 0.440
#> GSM48652 2 0.4538 0.2008 0.000 0.540 0.008 0.000 0.452
#> GSM48653 5 0.4883 0.0197 0.000 0.464 0.016 0.004 0.516
#> GSM48662 5 0.4196 0.3762 0.000 0.356 0.004 0.000 0.640
#> GSM48663 2 0.1205 0.8013 0.000 0.956 0.000 0.040 0.004
#> GSM25524 3 0.2389 0.8612 0.116 0.000 0.880 0.004 0.000
#> GSM25525 1 0.3333 0.8130 0.788 0.000 0.208 0.000 0.004
#> GSM25526 3 0.1205 0.9313 0.040 0.000 0.956 0.004 0.000
#> GSM25527 1 0.1197 0.8932 0.952 0.000 0.048 0.000 0.000
#> GSM25528 1 0.3579 0.7797 0.756 0.000 0.240 0.004 0.000
#> GSM25529 1 0.3333 0.8130 0.788 0.000 0.208 0.000 0.004
#> GSM25530 3 0.4182 0.4518 0.352 0.000 0.644 0.004 0.000
#> GSM25531 1 0.3123 0.8129 0.812 0.000 0.184 0.004 0.000
#> GSM48661 5 0.1503 0.7758 0.000 0.020 0.020 0.008 0.952
#> GSM25561 1 0.3861 0.6834 0.712 0.000 0.284 0.004 0.000
#> GSM25562 1 0.1282 0.8922 0.952 0.000 0.044 0.004 0.000
#> GSM25563 3 0.1197 0.9302 0.048 0.000 0.952 0.000 0.000
#> GSM25564 5 0.5059 0.6270 0.152 0.008 0.096 0.008 0.736
#> GSM25565 2 0.4787 0.1351 0.000 0.548 0.020 0.000 0.432
#> GSM25566 5 0.4655 0.0835 0.000 0.476 0.012 0.000 0.512
#> GSM25568 5 0.1278 0.7752 0.000 0.016 0.020 0.004 0.960
#> GSM25569 5 0.2970 0.6896 0.000 0.168 0.004 0.000 0.828
#> GSM25552 5 0.1082 0.7719 0.000 0.000 0.008 0.028 0.964
#> GSM25553 5 0.2452 0.7481 0.052 0.000 0.012 0.028 0.908
#> GSM25578 1 0.0000 0.8951 1.000 0.000 0.000 0.000 0.000
#> GSM25579 1 0.3545 0.8501 0.832 0.004 0.128 0.004 0.032
#> GSM25580 1 0.0290 0.8935 0.992 0.000 0.000 0.008 0.000
#> GSM25581 1 0.0000 0.8951 1.000 0.000 0.000 0.000 0.000
#> GSM48655 2 0.0290 0.8199 0.000 0.992 0.000 0.000 0.008
#> GSM48656 5 0.0992 0.7739 0.000 0.024 0.008 0.000 0.968
#> GSM48657 2 0.0290 0.8199 0.000 0.992 0.000 0.000 0.008
#> GSM48658 5 0.0981 0.7755 0.000 0.012 0.008 0.008 0.972
#> GSM25624 1 0.0451 0.8964 0.988 0.000 0.008 0.004 0.000
#> GSM25625 3 0.1571 0.9310 0.060 0.000 0.936 0.004 0.000
#> GSM25626 3 0.1043 0.9311 0.040 0.000 0.960 0.000 0.000
#> GSM25627 3 0.1766 0.9229 0.040 0.012 0.940 0.004 0.004
#> GSM25628 3 0.0000 0.9121 0.000 0.000 1.000 0.000 0.000
#> GSM25629 3 0.1492 0.8837 0.000 0.004 0.948 0.008 0.040
#> GSM25630 3 0.1197 0.9207 0.048 0.000 0.952 0.000 0.000
#> GSM25631 5 0.1612 0.7750 0.000 0.012 0.024 0.016 0.948
#> GSM25632 3 0.1544 0.9287 0.068 0.000 0.932 0.000 0.000
#> GSM25633 1 0.0162 0.8958 0.996 0.000 0.000 0.004 0.000
#> GSM25634 1 0.0290 0.8947 0.992 0.000 0.000 0.008 0.000
#> GSM25635 1 0.0404 0.8900 0.988 0.000 0.000 0.012 0.000
#> GSM25656 3 0.0290 0.9171 0.008 0.000 0.992 0.000 0.000
#> GSM25657 1 0.2629 0.8512 0.860 0.000 0.136 0.004 0.000
#> GSM25658 3 0.1282 0.9316 0.044 0.000 0.952 0.004 0.000
#> GSM25659 1 0.4220 0.7947 0.760 0.000 0.200 0.008 0.032
#> GSM25660 1 0.0162 0.8963 0.996 0.000 0.004 0.000 0.000
#> GSM25661 1 0.0000 0.8951 1.000 0.000 0.000 0.000 0.000
#> GSM25662 5 0.4637 0.0778 0.000 0.452 0.012 0.000 0.536
#> GSM25663 5 0.0854 0.7754 0.000 0.012 0.008 0.004 0.976
#> GSM25680 5 0.0854 0.7755 0.000 0.012 0.008 0.004 0.976
#> GSM25681 5 0.0981 0.7762 0.000 0.012 0.008 0.008 0.972
#> GSM25682 2 0.0290 0.8199 0.000 0.992 0.000 0.000 0.008
#> GSM25683 2 0.0693 0.8166 0.000 0.980 0.012 0.000 0.008
#> GSM25684 5 0.4617 0.1405 0.000 0.436 0.012 0.000 0.552
#> GSM25685 2 0.5445 0.3162 0.000 0.564 0.036 0.016 0.384
#> GSM25686 2 0.0290 0.8199 0.000 0.992 0.000 0.000 0.008
#> GSM25687 2 0.0290 0.8199 0.000 0.992 0.000 0.000 0.008
#> GSM48664 4 0.1498 0.8654 0.016 0.024 0.008 0.952 0.000
#> GSM48665 4 0.4692 0.5108 0.320 0.024 0.004 0.652 0.000
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM25548 5 0.4051 -0.6720 0.000 0.008 0.000 0.000 0.560 0.432
#> GSM25549 6 0.3833 0.8770 0.000 0.000 0.000 0.000 0.444 0.556
#> GSM25550 6 0.3838 0.8755 0.000 0.000 0.000 0.000 0.448 0.552
#> GSM25551 5 0.4211 0.4239 0.000 0.364 0.004 0.000 0.616 0.016
#> GSM25570 6 0.3833 0.8770 0.000 0.000 0.000 0.000 0.444 0.556
#> GSM25571 6 0.3843 0.8715 0.000 0.000 0.000 0.000 0.452 0.548
#> GSM25358 4 0.8011 0.2749 0.200 0.052 0.076 0.432 0.224 0.016
#> GSM25359 5 0.6338 0.3888 0.012 0.204 0.040 0.016 0.604 0.124
#> GSM25360 3 0.3486 0.7968 0.128 0.000 0.812 0.008 0.000 0.052
#> GSM25361 5 0.6297 0.1906 0.068 0.000 0.156 0.000 0.560 0.216
#> GSM25377 4 0.0260 0.8720 0.000 0.000 0.000 0.992 0.000 0.008
#> GSM25378 4 0.0146 0.8744 0.004 0.000 0.000 0.996 0.000 0.000
#> GSM25401 4 0.4931 0.6943 0.000 0.024 0.084 0.700 0.004 0.188
#> GSM25402 4 0.2488 0.8133 0.000 0.000 0.076 0.880 0.000 0.044
#> GSM25349 2 0.0436 0.8777 0.000 0.988 0.004 0.004 0.000 0.004
#> GSM25350 2 0.0000 0.8798 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM25356 4 0.0000 0.8743 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM25357 2 0.2458 0.8446 0.000 0.900 0.012 0.052 0.008 0.028
#> GSM25385 3 0.1493 0.8926 0.056 0.000 0.936 0.004 0.000 0.004
#> GSM25386 3 0.0622 0.9010 0.012 0.000 0.980 0.000 0.000 0.008
#> GSM25399 4 0.0806 0.8670 0.000 0.000 0.020 0.972 0.000 0.008
#> GSM25400 4 0.4563 0.3904 0.348 0.000 0.048 0.604 0.000 0.000
#> GSM48659 5 0.2993 0.5455 0.000 0.120 0.008 0.000 0.844 0.028
#> GSM48660 2 0.0000 0.8798 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM25409 2 0.5295 0.3381 0.000 0.640 0.044 0.012 0.268 0.036
#> GSM25410 3 0.1268 0.8967 0.036 0.000 0.952 0.004 0.000 0.008
#> GSM25426 2 0.7291 0.5040 0.000 0.512 0.064 0.072 0.128 0.224
#> GSM25427 4 0.0000 0.8743 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM25540 5 0.5519 0.2918 0.004 0.000 0.176 0.008 0.612 0.200
#> GSM25541 5 0.6707 0.1908 0.084 0.000 0.156 0.008 0.536 0.216
#> GSM25542 5 0.5253 0.5072 0.004 0.152 0.016 0.012 0.692 0.124
#> GSM25543 5 0.6012 0.4476 0.016 0.136 0.056 0.008 0.656 0.128
#> GSM25479 1 0.0000 0.8839 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM25480 1 0.0891 0.8836 0.968 0.000 0.024 0.000 0.000 0.008
#> GSM25481 4 0.0146 0.8740 0.000 0.004 0.000 0.996 0.000 0.000
#> GSM25482 4 0.0146 0.8740 0.000 0.004 0.000 0.996 0.000 0.000
#> GSM48654 5 0.2715 0.5424 0.000 0.112 0.004 0.000 0.860 0.024
#> GSM48650 2 0.5732 0.6442 0.000 0.648 0.048 0.100 0.012 0.192
#> GSM48651 5 0.3979 0.4330 0.000 0.360 0.000 0.000 0.628 0.012
#> GSM48652 5 0.3743 0.4893 0.000 0.252 0.000 0.000 0.724 0.024
#> GSM48653 5 0.3385 0.5386 0.000 0.172 0.004 0.000 0.796 0.028
#> GSM48662 5 0.3791 0.4846 0.000 0.236 0.000 0.000 0.732 0.032
#> GSM48663 2 0.3030 0.8000 0.000 0.848 0.000 0.092 0.004 0.056
#> GSM25524 3 0.2367 0.8503 0.088 0.000 0.888 0.008 0.000 0.016
#> GSM25525 1 0.2706 0.8237 0.832 0.000 0.160 0.000 0.000 0.008
#> GSM25526 3 0.0508 0.9015 0.012 0.000 0.984 0.004 0.000 0.000
#> GSM25527 1 0.0937 0.8817 0.960 0.000 0.040 0.000 0.000 0.000
#> GSM25528 1 0.4452 0.3624 0.548 0.000 0.428 0.008 0.000 0.016
#> GSM25529 1 0.2848 0.8101 0.816 0.000 0.176 0.000 0.000 0.008
#> GSM25530 3 0.2958 0.7789 0.160 0.000 0.824 0.008 0.000 0.008
#> GSM25531 1 0.2948 0.7872 0.804 0.000 0.188 0.008 0.000 0.000
#> GSM48661 5 0.3657 0.4684 0.000 0.024 0.012 0.004 0.788 0.172
#> GSM25561 1 0.4184 0.3472 0.556 0.000 0.432 0.008 0.000 0.004
#> GSM25562 1 0.2744 0.8374 0.864 0.000 0.064 0.072 0.000 0.000
#> GSM25563 3 0.0146 0.8979 0.000 0.000 0.996 0.000 0.000 0.004
#> GSM25564 5 0.7073 0.0516 0.356 0.024 0.128 0.016 0.440 0.036
#> GSM25565 5 0.4198 0.4350 0.000 0.344 0.012 0.004 0.636 0.004
#> GSM25566 5 0.4387 0.3885 0.000 0.392 0.008 0.000 0.584 0.016
#> GSM25568 5 0.4843 0.3976 0.064 0.008 0.048 0.008 0.752 0.120
#> GSM25569 5 0.3254 0.5179 0.000 0.136 0.000 0.000 0.816 0.048
#> GSM25552 6 0.3833 0.8770 0.000 0.000 0.000 0.000 0.444 0.556
#> GSM25553 6 0.4931 0.8135 0.004 0.000 0.020 0.020 0.460 0.496
#> GSM25578 1 0.0000 0.8839 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM25579 1 0.3172 0.8376 0.852 0.000 0.080 0.000 0.036 0.032
#> GSM25580 1 0.0146 0.8830 0.996 0.000 0.000 0.004 0.000 0.000
#> GSM25581 1 0.0000 0.8839 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM48655 2 0.0937 0.8517 0.000 0.960 0.000 0.000 0.040 0.000
#> GSM48656 5 0.2398 0.3640 0.000 0.020 0.000 0.000 0.876 0.104
#> GSM48657 2 0.0000 0.8798 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM48658 5 0.2278 0.4428 0.000 0.000 0.004 0.000 0.868 0.128
#> GSM25624 1 0.0405 0.8846 0.988 0.000 0.004 0.008 0.000 0.000
#> GSM25625 3 0.0622 0.9015 0.012 0.000 0.980 0.008 0.000 0.000
#> GSM25626 3 0.0622 0.9010 0.012 0.000 0.980 0.000 0.000 0.008
#> GSM25627 3 0.6040 0.6424 0.108 0.000 0.656 0.024 0.108 0.104
#> GSM25628 3 0.0713 0.8946 0.000 0.000 0.972 0.000 0.000 0.028
#> GSM25629 3 0.5304 0.5450 0.000 0.000 0.628 0.008 0.172 0.192
#> GSM25630 3 0.0777 0.8980 0.024 0.000 0.972 0.000 0.000 0.004
#> GSM25631 6 0.5027 0.2417 0.000 0.000 0.072 0.000 0.440 0.488
#> GSM25632 3 0.0858 0.9006 0.028 0.000 0.968 0.000 0.000 0.004
#> GSM25633 1 0.0146 0.8844 0.996 0.000 0.000 0.004 0.000 0.000
#> GSM25634 1 0.0146 0.8830 0.996 0.000 0.000 0.004 0.000 0.000
#> GSM25635 1 0.0363 0.8790 0.988 0.000 0.000 0.012 0.000 0.000
#> GSM25656 3 0.0777 0.8948 0.000 0.000 0.972 0.000 0.004 0.024
#> GSM25657 1 0.2020 0.8562 0.896 0.000 0.096 0.008 0.000 0.000
#> GSM25658 3 0.2165 0.8436 0.108 0.000 0.884 0.008 0.000 0.000
#> GSM25659 1 0.4294 0.7697 0.772 0.000 0.116 0.004 0.084 0.024
#> GSM25660 1 0.0146 0.8848 0.996 0.000 0.004 0.000 0.000 0.000
#> GSM25661 1 0.0146 0.8830 0.996 0.000 0.000 0.004 0.000 0.000
#> GSM25662 5 0.2704 0.5454 0.000 0.140 0.000 0.000 0.844 0.016
#> GSM25663 5 0.2051 0.4298 0.000 0.000 0.004 0.004 0.896 0.096
#> GSM25680 5 0.1970 0.3755 0.000 0.000 0.008 0.000 0.900 0.092
#> GSM25681 5 0.4371 -0.2646 0.000 0.000 0.036 0.000 0.620 0.344
#> GSM25682 2 0.0000 0.8798 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM25683 2 0.1204 0.8680 0.000 0.960 0.004 0.004 0.016 0.016
#> GSM25684 5 0.3084 0.5459 0.000 0.132 0.004 0.000 0.832 0.032
#> GSM25685 5 0.6083 0.4183 0.000 0.092 0.064 0.020 0.624 0.200
#> GSM25686 2 0.0000 0.8798 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM25687 2 0.0000 0.8798 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM48664 4 0.0146 0.8737 0.000 0.000 0.000 0.996 0.000 0.004
#> GSM48665 4 0.1141 0.8490 0.052 0.000 0.000 0.948 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 genotype/variation(p) k
#> SD:mclust 97 5.16e-06 2
#> SD:mclust 77 1.14e-03 3
#> SD:mclust 90 8.84e-06 4
#> SD:mclust 85 1.05e-06 5
#> SD:mclust 73 2.83e-09 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 100 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 0.837 0.888 0.956 0.5025 0.495 0.495
#> 3 3 0.472 0.519 0.773 0.3227 0.736 0.518
#> 4 4 0.436 0.435 0.693 0.1119 0.776 0.455
#> 5 5 0.468 0.428 0.617 0.0693 0.821 0.453
#> 6 6 0.524 0.385 0.628 0.0445 0.844 0.420
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
#> GSM25548 2 0.0000 0.9669 0.000 1.000
#> GSM25549 2 0.0000 0.9669 0.000 1.000
#> GSM25550 2 0.0000 0.9669 0.000 1.000
#> GSM25551 2 0.0000 0.9669 0.000 1.000
#> GSM25570 2 0.0000 0.9669 0.000 1.000
#> GSM25571 2 0.0000 0.9669 0.000 1.000
#> GSM25358 2 0.9983 0.0292 0.476 0.524
#> GSM25359 2 0.0000 0.9669 0.000 1.000
#> GSM25360 1 0.0000 0.9380 1.000 0.000
#> GSM25361 1 0.6887 0.7550 0.816 0.184
#> GSM25377 1 0.0000 0.9380 1.000 0.000
#> GSM25378 1 0.4022 0.8722 0.920 0.080
#> GSM25401 2 0.6048 0.8139 0.148 0.852
#> GSM25402 1 0.7376 0.7265 0.792 0.208
#> GSM25349 2 0.0000 0.9669 0.000 1.000
#> GSM25350 2 0.0000 0.9669 0.000 1.000
#> GSM25356 1 0.2423 0.9086 0.960 0.040
#> GSM25357 2 0.0000 0.9669 0.000 1.000
#> GSM25385 1 0.0000 0.9380 1.000 0.000
#> GSM25386 1 0.0376 0.9355 0.996 0.004
#> GSM25399 1 0.0000 0.9380 1.000 0.000
#> GSM25400 1 0.0000 0.9380 1.000 0.000
#> GSM48659 2 0.0000 0.9669 0.000 1.000
#> GSM48660 2 0.0000 0.9669 0.000 1.000
#> GSM25409 2 0.0000 0.9669 0.000 1.000
#> GSM25410 1 0.0672 0.9328 0.992 0.008
#> GSM25426 2 0.0000 0.9669 0.000 1.000
#> GSM25427 1 0.2043 0.9154 0.968 0.032
#> GSM25540 2 0.5737 0.8294 0.136 0.864
#> GSM25541 1 0.9977 0.1241 0.528 0.472
#> GSM25542 2 0.0000 0.9669 0.000 1.000
#> GSM25543 2 0.0000 0.9669 0.000 1.000
#> GSM25479 1 0.0000 0.9380 1.000 0.000
#> GSM25480 1 0.0000 0.9380 1.000 0.000
#> GSM25481 1 0.9983 0.1401 0.524 0.476
#> GSM25482 1 0.9833 0.3029 0.576 0.424
#> GSM48654 2 0.0000 0.9669 0.000 1.000
#> GSM48650 2 0.0000 0.9669 0.000 1.000
#> GSM48651 2 0.0000 0.9669 0.000 1.000
#> GSM48652 2 0.0000 0.9669 0.000 1.000
#> GSM48653 2 0.0000 0.9669 0.000 1.000
#> GSM48662 2 0.0000 0.9669 0.000 1.000
#> GSM48663 2 0.0000 0.9669 0.000 1.000
#> GSM25524 1 0.0000 0.9380 1.000 0.000
#> GSM25525 1 0.0000 0.9380 1.000 0.000
#> GSM25526 1 0.0000 0.9380 1.000 0.000
#> GSM25527 1 0.0000 0.9380 1.000 0.000
#> GSM25528 1 0.0000 0.9380 1.000 0.000
#> GSM25529 1 0.0000 0.9380 1.000 0.000
#> GSM25530 1 0.0000 0.9380 1.000 0.000
#> GSM25531 1 0.0000 0.9380 1.000 0.000
#> GSM48661 2 0.0000 0.9669 0.000 1.000
#> GSM25561 1 0.0000 0.9380 1.000 0.000
#> GSM25562 1 0.0000 0.9380 1.000 0.000
#> GSM25563 1 0.0000 0.9380 1.000 0.000
#> GSM25564 1 0.9833 0.2998 0.576 0.424
#> GSM25565 2 0.0000 0.9669 0.000 1.000
#> GSM25566 2 0.0000 0.9669 0.000 1.000
#> GSM25568 2 0.4815 0.8662 0.104 0.896
#> GSM25569 2 0.0000 0.9669 0.000 1.000
#> GSM25552 2 0.0000 0.9669 0.000 1.000
#> GSM25553 2 0.9209 0.4673 0.336 0.664
#> GSM25578 1 0.0000 0.9380 1.000 0.000
#> GSM25579 1 0.0000 0.9380 1.000 0.000
#> GSM25580 1 0.0000 0.9380 1.000 0.000
#> GSM25581 1 0.0000 0.9380 1.000 0.000
#> GSM48655 2 0.0000 0.9669 0.000 1.000
#> GSM48656 2 0.0000 0.9669 0.000 1.000
#> GSM48657 2 0.0000 0.9669 0.000 1.000
#> GSM48658 2 0.0000 0.9669 0.000 1.000
#> GSM25624 1 0.0000 0.9380 1.000 0.000
#> GSM25625 1 0.0000 0.9380 1.000 0.000
#> GSM25626 1 0.0376 0.9355 0.996 0.004
#> GSM25627 2 0.0938 0.9575 0.012 0.988
#> GSM25628 1 0.9998 0.0501 0.508 0.492
#> GSM25629 2 0.0376 0.9638 0.004 0.996
#> GSM25630 1 0.0000 0.9380 1.000 0.000
#> GSM25631 2 0.3584 0.9060 0.068 0.932
#> GSM25632 1 0.0000 0.9380 1.000 0.000
#> GSM25633 1 0.0000 0.9380 1.000 0.000
#> GSM25634 1 0.0000 0.9380 1.000 0.000
#> GSM25635 1 0.0000 0.9380 1.000 0.000
#> GSM25656 2 0.8144 0.6535 0.252 0.748
#> GSM25657 1 0.0000 0.9380 1.000 0.000
#> GSM25658 1 0.0000 0.9380 1.000 0.000
#> GSM25659 1 0.0000 0.9380 1.000 0.000
#> GSM25660 1 0.0000 0.9380 1.000 0.000
#> GSM25661 1 0.0000 0.9380 1.000 0.000
#> GSM25662 2 0.0000 0.9669 0.000 1.000
#> GSM25663 2 0.0000 0.9669 0.000 1.000
#> GSM25680 2 0.0000 0.9669 0.000 1.000
#> GSM25681 2 0.0000 0.9669 0.000 1.000
#> GSM25682 2 0.0000 0.9669 0.000 1.000
#> GSM25683 2 0.0000 0.9669 0.000 1.000
#> GSM25684 2 0.0000 0.9669 0.000 1.000
#> GSM25685 2 0.0000 0.9669 0.000 1.000
#> GSM25686 2 0.0000 0.9669 0.000 1.000
#> GSM25687 2 0.0000 0.9669 0.000 1.000
#> GSM48664 1 0.0000 0.9380 1.000 0.000
#> GSM48665 1 0.0000 0.9380 1.000 0.000
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM25548 2 0.2537 0.75478 0.000 0.920 0.080
#> GSM25549 2 0.0424 0.76160 0.000 0.992 0.008
#> GSM25550 2 0.3752 0.68250 0.144 0.856 0.000
#> GSM25551 2 0.4842 0.67214 0.000 0.776 0.224
#> GSM25570 2 0.1529 0.75276 0.040 0.960 0.000
#> GSM25571 2 0.1031 0.76188 0.000 0.976 0.024
#> GSM25358 2 0.8338 0.00743 0.400 0.516 0.084
#> GSM25359 2 0.6386 0.41289 0.004 0.584 0.412
#> GSM25360 3 0.4178 0.58392 0.172 0.000 0.828
#> GSM25361 3 0.1315 0.62932 0.020 0.008 0.972
#> GSM25377 1 0.5138 0.57184 0.748 0.252 0.000
#> GSM25378 1 0.6079 0.35743 0.612 0.388 0.000
#> GSM25401 2 0.7279 0.49035 0.292 0.652 0.056
#> GSM25402 1 0.6228 0.37591 0.624 0.372 0.004
#> GSM25349 2 0.3038 0.72167 0.104 0.896 0.000
#> GSM25350 2 0.2796 0.72944 0.092 0.908 0.000
#> GSM25356 1 0.6192 0.28479 0.580 0.420 0.000
#> GSM25357 2 0.1989 0.75070 0.048 0.948 0.004
#> GSM25385 3 0.6267 0.19821 0.452 0.000 0.548
#> GSM25386 3 0.2261 0.62013 0.068 0.000 0.932
#> GSM25399 1 0.1163 0.71593 0.972 0.028 0.000
#> GSM25400 1 0.1919 0.72543 0.956 0.024 0.020
#> GSM48659 2 0.6302 0.30120 0.000 0.520 0.480
#> GSM48660 2 0.1529 0.75294 0.040 0.960 0.000
#> GSM25409 2 0.2796 0.72972 0.092 0.908 0.000
#> GSM25410 3 0.4504 0.57085 0.196 0.000 0.804
#> GSM25426 2 0.5948 0.52556 0.000 0.640 0.360
#> GSM25427 1 0.6026 0.37987 0.624 0.376 0.000
#> GSM25540 3 0.2200 0.61368 0.004 0.056 0.940
#> GSM25541 3 0.1453 0.62958 0.008 0.024 0.968
#> GSM25542 3 0.6307 -0.26719 0.000 0.488 0.512
#> GSM25543 3 0.6260 -0.13800 0.000 0.448 0.552
#> GSM25479 1 0.1989 0.73076 0.948 0.004 0.048
#> GSM25480 1 0.2229 0.73209 0.944 0.012 0.044
#> GSM25481 2 0.6513 -0.03813 0.476 0.520 0.004
#> GSM25482 1 0.6309 0.07191 0.504 0.496 0.000
#> GSM48654 2 0.6168 0.44241 0.000 0.588 0.412
#> GSM48650 2 0.2165 0.75945 0.000 0.936 0.064
#> GSM48651 2 0.4121 0.71496 0.000 0.832 0.168
#> GSM48652 2 0.5291 0.63692 0.000 0.732 0.268
#> GSM48653 2 0.6308 0.26816 0.000 0.508 0.492
#> GSM48662 2 0.2165 0.75900 0.000 0.936 0.064
#> GSM48663 2 0.3030 0.72905 0.092 0.904 0.004
#> GSM25524 3 0.5785 0.42368 0.332 0.000 0.668
#> GSM25525 1 0.5621 0.45320 0.692 0.000 0.308
#> GSM25526 3 0.4291 0.58047 0.180 0.000 0.820
#> GSM25527 1 0.5397 0.50698 0.720 0.000 0.280
#> GSM25528 3 0.6308 0.07757 0.492 0.000 0.508
#> GSM25529 1 0.6111 0.23960 0.604 0.000 0.396
#> GSM25530 3 0.6302 0.11547 0.480 0.000 0.520
#> GSM25531 1 0.5706 0.43004 0.680 0.000 0.320
#> GSM48661 3 0.5591 0.24328 0.000 0.304 0.696
#> GSM25561 3 0.6225 0.24395 0.432 0.000 0.568
#> GSM25562 1 0.4235 0.64741 0.824 0.000 0.176
#> GSM25563 3 0.4605 0.56415 0.204 0.000 0.796
#> GSM25564 3 0.9840 0.12849 0.264 0.320 0.416
#> GSM25565 2 0.3752 0.72753 0.000 0.856 0.144
#> GSM25566 2 0.2796 0.75070 0.000 0.908 0.092
#> GSM25568 2 0.7411 0.34053 0.036 0.548 0.416
#> GSM25569 2 0.3686 0.73007 0.000 0.860 0.140
#> GSM25552 2 0.2772 0.73443 0.080 0.916 0.004
#> GSM25553 2 0.6451 0.10291 0.436 0.560 0.004
#> GSM25578 1 0.3340 0.69697 0.880 0.000 0.120
#> GSM25579 1 0.4733 0.62681 0.800 0.004 0.196
#> GSM25580 1 0.0892 0.73078 0.980 0.000 0.020
#> GSM25581 1 0.2356 0.72184 0.928 0.000 0.072
#> GSM48655 2 0.0424 0.76138 0.000 0.992 0.008
#> GSM48656 2 0.2711 0.75256 0.000 0.912 0.088
#> GSM48657 2 0.1411 0.75382 0.036 0.964 0.000
#> GSM48658 3 0.6079 0.02250 0.000 0.388 0.612
#> GSM25624 1 0.1525 0.73177 0.964 0.004 0.032
#> GSM25625 3 0.5397 0.49373 0.280 0.000 0.720
#> GSM25626 3 0.1753 0.62479 0.048 0.000 0.952
#> GSM25627 3 0.3112 0.57978 0.004 0.096 0.900
#> GSM25628 3 0.0829 0.62972 0.004 0.012 0.984
#> GSM25629 3 0.2878 0.57536 0.000 0.096 0.904
#> GSM25630 3 0.5291 0.50785 0.268 0.000 0.732
#> GSM25631 3 0.3715 0.55131 0.004 0.128 0.868
#> GSM25632 3 0.6215 0.25186 0.428 0.000 0.572
#> GSM25633 1 0.3192 0.70192 0.888 0.000 0.112
#> GSM25634 1 0.2959 0.70871 0.900 0.000 0.100
#> GSM25635 1 0.2599 0.73232 0.932 0.016 0.052
#> GSM25656 3 0.1878 0.62275 0.004 0.044 0.952
#> GSM25657 1 0.5138 0.55144 0.748 0.000 0.252
#> GSM25658 3 0.5098 0.52910 0.248 0.000 0.752
#> GSM25659 3 0.6244 0.22135 0.440 0.000 0.560
#> GSM25660 1 0.2066 0.72761 0.940 0.000 0.060
#> GSM25661 1 0.1765 0.73243 0.956 0.004 0.040
#> GSM25662 2 0.6154 0.44704 0.000 0.592 0.408
#> GSM25663 2 0.5397 0.62061 0.000 0.720 0.280
#> GSM25680 2 0.6286 0.33783 0.000 0.536 0.464
#> GSM25681 3 0.6267 -0.15885 0.000 0.452 0.548
#> GSM25682 2 0.0475 0.76059 0.004 0.992 0.004
#> GSM25683 2 0.2261 0.75855 0.000 0.932 0.068
#> GSM25684 2 0.6045 0.49290 0.000 0.620 0.380
#> GSM25685 3 0.6291 -0.21688 0.000 0.468 0.532
#> GSM25686 2 0.0237 0.76099 0.000 0.996 0.004
#> GSM25687 2 0.1031 0.75668 0.024 0.976 0.000
#> GSM48664 1 0.3816 0.65602 0.852 0.148 0.000
#> GSM48665 1 0.3816 0.65593 0.852 0.148 0.000
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM25548 4 0.3806 0.6732 0.000 0.156 0.020 0.824
#> GSM25549 4 0.1854 0.6988 0.000 0.048 0.012 0.940
#> GSM25550 4 0.1042 0.6789 0.008 0.020 0.000 0.972
#> GSM25551 2 0.3324 0.5287 0.000 0.852 0.012 0.136
#> GSM25570 4 0.1305 0.6959 0.000 0.036 0.004 0.960
#> GSM25571 4 0.2412 0.6986 0.000 0.084 0.008 0.908
#> GSM25358 1 0.6295 0.2651 0.496 0.460 0.024 0.020
#> GSM25359 2 0.5136 0.4501 0.004 0.752 0.188 0.056
#> GSM25360 3 0.0992 0.5996 0.012 0.008 0.976 0.004
#> GSM25361 3 0.2048 0.5952 0.000 0.008 0.928 0.064
#> GSM25377 1 0.3325 0.6870 0.864 0.112 0.000 0.024
#> GSM25378 1 0.4499 0.6581 0.792 0.160 0.000 0.048
#> GSM25401 2 0.4936 0.2426 0.280 0.700 0.020 0.000
#> GSM25402 1 0.5345 0.4194 0.584 0.404 0.008 0.004
#> GSM25349 2 0.5980 0.2107 0.044 0.560 0.000 0.396
#> GSM25350 4 0.5398 0.2342 0.016 0.404 0.000 0.580
#> GSM25356 1 0.4832 0.6405 0.768 0.176 0.000 0.056
#> GSM25357 2 0.5272 0.4445 0.112 0.752 0.000 0.136
#> GSM25385 1 0.7002 0.2950 0.568 0.164 0.268 0.000
#> GSM25386 3 0.7133 0.4698 0.144 0.344 0.512 0.000
#> GSM25399 1 0.1716 0.7075 0.936 0.064 0.000 0.000
#> GSM25400 1 0.1557 0.7095 0.944 0.056 0.000 0.000
#> GSM48659 2 0.7748 0.2576 0.000 0.436 0.304 0.260
#> GSM48660 2 0.5165 0.0773 0.004 0.512 0.000 0.484
#> GSM25409 4 0.4706 0.5801 0.020 0.248 0.000 0.732
#> GSM25410 2 0.7914 -0.3911 0.332 0.356 0.312 0.000
#> GSM25426 2 0.1510 0.5051 0.000 0.956 0.028 0.016
#> GSM25427 1 0.5392 0.6226 0.724 0.072 0.000 0.204
#> GSM25540 3 0.3271 0.5743 0.000 0.132 0.856 0.012
#> GSM25541 3 0.1724 0.6005 0.000 0.032 0.948 0.020
#> GSM25542 2 0.5218 0.4727 0.000 0.736 0.200 0.064
#> GSM25543 2 0.6943 0.2240 0.004 0.540 0.348 0.108
#> GSM25479 1 0.5982 0.6181 0.684 0.000 0.112 0.204
#> GSM25480 1 0.7646 0.2468 0.408 0.000 0.208 0.384
#> GSM25481 1 0.7390 0.3111 0.512 0.204 0.000 0.284
#> GSM25482 1 0.7120 0.1312 0.436 0.128 0.000 0.436
#> GSM48654 2 0.7536 0.3030 0.000 0.488 0.228 0.284
#> GSM48650 2 0.2654 0.5230 0.004 0.888 0.000 0.108
#> GSM48651 2 0.5517 0.4110 0.000 0.648 0.036 0.316
#> GSM48652 2 0.5907 0.4791 0.000 0.680 0.092 0.228
#> GSM48653 2 0.6675 0.4460 0.000 0.616 0.228 0.156
#> GSM48662 4 0.5203 0.4211 0.000 0.348 0.016 0.636
#> GSM48663 2 0.5931 0.0428 0.036 0.504 0.000 0.460
#> GSM25524 3 0.2530 0.5628 0.112 0.000 0.888 0.000
#> GSM25525 3 0.7002 0.1192 0.352 0.000 0.520 0.128
#> GSM25526 3 0.7910 0.2650 0.332 0.308 0.360 0.000
#> GSM25527 1 0.5144 0.6000 0.732 0.000 0.216 0.052
#> GSM25528 3 0.4608 0.3612 0.304 0.000 0.692 0.004
#> GSM25529 3 0.6701 0.2102 0.328 0.000 0.564 0.108
#> GSM25530 1 0.5913 0.3491 0.600 0.048 0.352 0.000
#> GSM25531 1 0.3047 0.6761 0.872 0.012 0.116 0.000
#> GSM48661 3 0.6238 0.3362 0.000 0.276 0.632 0.092
#> GSM25561 3 0.4051 0.4962 0.208 0.004 0.784 0.004
#> GSM25562 1 0.3432 0.6935 0.860 0.012 0.120 0.008
#> GSM25563 3 0.5963 0.5159 0.196 0.116 0.688 0.000
#> GSM25564 3 0.8951 0.2213 0.116 0.152 0.472 0.260
#> GSM25565 2 0.5272 0.4502 0.000 0.680 0.032 0.288
#> GSM25566 2 0.5444 0.2475 0.000 0.560 0.016 0.424
#> GSM25568 3 0.8256 -0.1909 0.020 0.216 0.384 0.380
#> GSM25569 4 0.5632 0.4108 0.000 0.340 0.036 0.624
#> GSM25552 4 0.0712 0.6680 0.004 0.004 0.008 0.984
#> GSM25553 4 0.1733 0.6205 0.028 0.000 0.024 0.948
#> GSM25578 1 0.4938 0.6621 0.772 0.000 0.148 0.080
#> GSM25579 3 0.7661 0.1872 0.212 0.000 0.412 0.376
#> GSM25580 1 0.2142 0.7220 0.928 0.000 0.016 0.056
#> GSM25581 1 0.3691 0.7089 0.856 0.000 0.076 0.068
#> GSM48655 2 0.4977 0.1575 0.000 0.540 0.000 0.460
#> GSM48656 4 0.5041 0.5959 0.000 0.232 0.040 0.728
#> GSM48657 2 0.5220 0.2245 0.008 0.568 0.000 0.424
#> GSM48658 3 0.6788 0.2558 0.000 0.144 0.592 0.264
#> GSM25624 1 0.2473 0.7215 0.908 0.000 0.012 0.080
#> GSM25625 3 0.6971 0.2330 0.372 0.120 0.508 0.000
#> GSM25626 3 0.7148 0.4596 0.140 0.364 0.496 0.000
#> GSM25627 2 0.4677 0.2648 0.040 0.768 0.192 0.000
#> GSM25628 3 0.4608 0.4998 0.004 0.304 0.692 0.000
#> GSM25629 3 0.4998 0.2603 0.000 0.488 0.512 0.000
#> GSM25630 3 0.4060 0.5728 0.112 0.048 0.836 0.004
#> GSM25631 3 0.4540 0.4847 0.004 0.008 0.740 0.248
#> GSM25632 1 0.6637 0.2938 0.572 0.104 0.324 0.000
#> GSM25633 1 0.2706 0.7135 0.900 0.000 0.080 0.020
#> GSM25634 1 0.2363 0.7187 0.920 0.000 0.056 0.024
#> GSM25635 1 0.4227 0.7008 0.820 0.000 0.060 0.120
#> GSM25656 3 0.4819 0.4635 0.004 0.344 0.652 0.000
#> GSM25657 1 0.2737 0.7007 0.888 0.000 0.104 0.008
#> GSM25658 3 0.7849 0.2279 0.352 0.268 0.380 0.000
#> GSM25659 3 0.5515 0.5143 0.116 0.000 0.732 0.152
#> GSM25660 1 0.6887 0.4890 0.560 0.000 0.132 0.308
#> GSM25661 1 0.2919 0.7200 0.896 0.000 0.044 0.060
#> GSM25662 2 0.5031 0.5244 0.000 0.768 0.092 0.140
#> GSM25663 4 0.7185 0.2757 0.000 0.284 0.176 0.540
#> GSM25680 3 0.7249 -0.0538 0.000 0.144 0.444 0.412
#> GSM25681 3 0.5937 0.0768 0.000 0.036 0.492 0.472
#> GSM25682 2 0.5097 0.2312 0.004 0.568 0.000 0.428
#> GSM25683 2 0.3528 0.5038 0.000 0.808 0.000 0.192
#> GSM25684 2 0.6465 0.4634 0.000 0.636 0.136 0.228
#> GSM25685 2 0.3554 0.4933 0.000 0.844 0.136 0.020
#> GSM25686 2 0.4955 0.1902 0.000 0.556 0.000 0.444
#> GSM25687 2 0.5132 0.1802 0.004 0.548 0.000 0.448
#> GSM48664 1 0.2313 0.7140 0.924 0.044 0.000 0.032
#> GSM48665 1 0.2867 0.7152 0.884 0.012 0.000 0.104
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM25548 5 0.495 0.45955 0.000 0.256 0.000 0.068 0.676
#> GSM25549 5 0.434 0.51059 0.004 0.268 0.000 0.020 0.708
#> GSM25550 5 0.394 0.57263 0.016 0.208 0.000 0.008 0.768
#> GSM25551 4 0.481 0.55521 0.004 0.208 0.004 0.724 0.060
#> GSM25570 5 0.384 0.57944 0.008 0.196 0.000 0.016 0.780
#> GSM25571 5 0.430 0.54710 0.000 0.216 0.000 0.044 0.740
#> GSM25358 4 0.570 0.22885 0.328 0.052 0.000 0.596 0.024
#> GSM25359 4 0.546 0.55128 0.008 0.160 0.064 0.724 0.044
#> GSM25360 3 0.383 0.63728 0.040 0.004 0.844 0.056 0.056
#> GSM25361 3 0.515 0.55544 0.028 0.008 0.720 0.040 0.204
#> GSM25377 1 0.634 0.55432 0.684 0.140 0.032 0.088 0.056
#> GSM25378 1 0.548 0.60645 0.700 0.036 0.004 0.200 0.060
#> GSM25401 4 0.506 0.54281 0.164 0.120 0.000 0.712 0.004
#> GSM25402 1 0.710 0.13543 0.464 0.100 0.020 0.384 0.032
#> GSM25349 2 0.367 0.57813 0.008 0.848 0.012 0.064 0.068
#> GSM25350 2 0.466 0.54407 0.008 0.768 0.012 0.060 0.152
#> GSM25356 1 0.566 0.59475 0.700 0.040 0.004 0.172 0.084
#> GSM25357 4 0.565 0.40741 0.024 0.316 0.000 0.608 0.052
#> GSM25385 1 0.788 0.20480 0.436 0.024 0.216 0.284 0.040
#> GSM25386 3 0.803 0.48862 0.084 0.152 0.532 0.180 0.052
#> GSM25399 1 0.452 0.63113 0.816 0.056 0.028 0.052 0.048
#> GSM25400 1 0.287 0.66329 0.880 0.012 0.008 0.092 0.008
#> GSM48659 2 0.803 0.26977 0.000 0.424 0.256 0.192 0.128
#> GSM48660 2 0.253 0.58181 0.008 0.912 0.020 0.020 0.040
#> GSM25409 2 0.492 0.23452 0.000 0.556 0.000 0.028 0.416
#> GSM25410 3 0.893 0.40551 0.152 0.172 0.420 0.200 0.056
#> GSM25426 4 0.390 0.57368 0.000 0.216 0.012 0.764 0.008
#> GSM25427 1 0.713 0.47983 0.576 0.216 0.016 0.056 0.136
#> GSM25540 3 0.474 0.59707 0.004 0.020 0.760 0.160 0.056
#> GSM25541 3 0.552 0.59724 0.020 0.008 0.712 0.120 0.140
#> GSM25542 2 0.701 0.11816 0.000 0.476 0.332 0.156 0.036
#> GSM25543 2 0.685 0.00524 0.000 0.468 0.376 0.116 0.040
#> GSM25479 1 0.467 0.48829 0.656 0.004 0.016 0.004 0.320
#> GSM25480 5 0.557 0.02246 0.380 0.000 0.044 0.016 0.560
#> GSM25481 1 0.801 0.20452 0.392 0.352 0.012 0.084 0.160
#> GSM25482 1 0.797 0.13107 0.384 0.296 0.008 0.060 0.252
#> GSM48654 2 0.589 0.50341 0.000 0.656 0.224 0.068 0.052
#> GSM48650 2 0.476 0.02347 0.000 0.552 0.012 0.432 0.004
#> GSM48651 2 0.418 0.56680 0.000 0.812 0.052 0.100 0.036
#> GSM48652 2 0.472 0.53313 0.000 0.768 0.072 0.132 0.028
#> GSM48653 2 0.670 0.35484 0.000 0.572 0.188 0.204 0.036
#> GSM48662 2 0.353 0.55746 0.000 0.820 0.016 0.012 0.152
#> GSM48663 2 0.361 0.55481 0.016 0.860 0.020 0.048 0.056
#> GSM25524 3 0.595 0.56104 0.132 0.000 0.688 0.100 0.080
#> GSM25525 1 0.737 0.17793 0.408 0.000 0.256 0.032 0.304
#> GSM25526 4 0.575 0.24550 0.252 0.000 0.124 0.620 0.004
#> GSM25527 1 0.562 0.60214 0.712 0.000 0.060 0.100 0.128
#> GSM25528 3 0.632 0.28786 0.320 0.000 0.560 0.036 0.084
#> GSM25529 1 0.743 0.01157 0.332 0.000 0.320 0.028 0.320
#> GSM25530 1 0.538 0.51080 0.692 0.000 0.196 0.096 0.016
#> GSM25531 1 0.320 0.64965 0.864 0.000 0.052 0.076 0.008
#> GSM48661 3 0.600 0.44253 0.000 0.228 0.636 0.108 0.028
#> GSM25561 3 0.602 0.56487 0.144 0.076 0.704 0.044 0.032
#> GSM25562 1 0.825 0.26551 0.484 0.184 0.212 0.052 0.068
#> GSM25563 3 0.677 0.59002 0.096 0.096 0.664 0.100 0.044
#> GSM25564 2 0.878 0.08388 0.148 0.400 0.280 0.052 0.120
#> GSM25565 2 0.563 0.53279 0.000 0.684 0.052 0.204 0.060
#> GSM25566 2 0.660 0.42410 0.000 0.532 0.012 0.228 0.228
#> GSM25568 2 0.696 0.10032 0.016 0.528 0.328 0.064 0.064
#> GSM25569 2 0.448 0.55290 0.000 0.776 0.068 0.016 0.140
#> GSM25552 5 0.358 0.57781 0.004 0.204 0.000 0.008 0.784
#> GSM25553 5 0.426 0.56157 0.020 0.192 0.016 0.004 0.768
#> GSM25578 1 0.380 0.64427 0.812 0.000 0.032 0.012 0.144
#> GSM25579 5 0.616 0.22993 0.248 0.000 0.128 0.020 0.604
#> GSM25580 1 0.263 0.67404 0.904 0.004 0.024 0.016 0.052
#> GSM25581 1 0.366 0.65975 0.832 0.000 0.044 0.012 0.112
#> GSM48655 2 0.526 0.53349 0.000 0.680 0.000 0.144 0.176
#> GSM48656 2 0.437 0.46074 0.000 0.712 0.024 0.004 0.260
#> GSM48657 2 0.491 0.54317 0.008 0.736 0.000 0.136 0.120
#> GSM48658 3 0.745 0.19337 0.000 0.260 0.496 0.084 0.160
#> GSM25624 1 0.412 0.66442 0.816 0.008 0.024 0.036 0.116
#> GSM25625 1 0.734 -0.13211 0.348 0.004 0.340 0.292 0.016
#> GSM25626 3 0.749 0.40490 0.080 0.072 0.464 0.360 0.024
#> GSM25627 4 0.376 0.59300 0.032 0.096 0.036 0.836 0.000
#> GSM25628 3 0.520 0.57328 0.012 0.060 0.692 0.232 0.004
#> GSM25629 4 0.501 0.50277 0.012 0.072 0.148 0.752 0.016
#> GSM25630 3 0.523 0.61276 0.072 0.092 0.768 0.032 0.036
#> GSM25631 3 0.578 0.24022 0.008 0.052 0.544 0.008 0.388
#> GSM25632 1 0.723 0.23279 0.504 0.008 0.288 0.160 0.040
#> GSM25633 1 0.267 0.66991 0.900 0.000 0.036 0.020 0.044
#> GSM25634 1 0.292 0.67373 0.892 0.004 0.048 0.028 0.028
#> GSM25635 1 0.535 0.59343 0.700 0.012 0.016 0.056 0.216
#> GSM25656 3 0.627 0.35369 0.004 0.092 0.496 0.396 0.012
#> GSM25657 1 0.223 0.66726 0.920 0.000 0.032 0.036 0.012
#> GSM25658 4 0.646 0.13690 0.304 0.008 0.148 0.536 0.004
#> GSM25659 3 0.709 0.41827 0.160 0.016 0.560 0.036 0.228
#> GSM25660 5 0.557 -0.21849 0.464 0.004 0.040 0.008 0.484
#> GSM25661 1 0.280 0.66847 0.880 0.000 0.016 0.012 0.092
#> GSM25662 4 0.593 0.33020 0.000 0.352 0.052 0.564 0.032
#> GSM25663 5 0.750 0.11396 0.000 0.312 0.076 0.156 0.456
#> GSM25680 5 0.788 0.30952 0.000 0.172 0.204 0.152 0.472
#> GSM25681 5 0.606 0.35651 0.004 0.080 0.244 0.036 0.636
#> GSM25682 2 0.643 0.36258 0.000 0.504 0.000 0.272 0.224
#> GSM25683 4 0.574 0.16114 0.000 0.400 0.004 0.520 0.076
#> GSM25684 4 0.668 0.15642 0.000 0.380 0.036 0.480 0.104
#> GSM25685 4 0.416 0.57692 0.000 0.200 0.024 0.764 0.012
#> GSM25686 2 0.618 0.40337 0.000 0.552 0.000 0.252 0.196
#> GSM25687 2 0.620 0.43341 0.000 0.552 0.000 0.212 0.236
#> GSM48664 1 0.441 0.64801 0.820 0.056 0.020 0.056 0.048
#> GSM48665 1 0.406 0.65758 0.832 0.040 0.008 0.044 0.076
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM25548 5 0.168 0.6214 0.004 0.016 0.000 0.040 0.936 0.004
#> GSM25549 5 0.191 0.6160 0.008 0.032 0.004 0.012 0.932 0.012
#> GSM25550 5 0.208 0.5985 0.020 0.040 0.004 0.000 0.920 0.016
#> GSM25551 4 0.305 0.5593 0.000 0.008 0.020 0.856 0.100 0.016
#> GSM25570 5 0.149 0.6125 0.008 0.008 0.000 0.008 0.948 0.028
#> GSM25571 5 0.248 0.6192 0.008 0.008 0.004 0.056 0.900 0.024
#> GSM25358 4 0.800 0.0287 0.152 0.016 0.300 0.364 0.152 0.016
#> GSM25359 3 0.717 -0.0153 0.008 0.028 0.412 0.292 0.240 0.020
#> GSM25360 3 0.529 0.2299 0.036 0.016 0.536 0.004 0.008 0.400
#> GSM25361 6 0.570 0.3169 0.016 0.020 0.200 0.004 0.116 0.644
#> GSM25377 1 0.607 0.4874 0.580 0.228 0.152 0.008 0.000 0.032
#> GSM25378 1 0.695 0.5563 0.596 0.040 0.144 0.120 0.084 0.016
#> GSM25401 4 0.496 0.4499 0.188 0.044 0.052 0.708 0.000 0.008
#> GSM25402 1 0.705 0.1116 0.424 0.096 0.108 0.356 0.000 0.016
#> GSM25349 2 0.634 0.4216 0.004 0.576 0.124 0.084 0.212 0.000
#> GSM25350 2 0.670 0.1216 0.004 0.428 0.144 0.048 0.372 0.004
#> GSM25356 1 0.710 0.5490 0.600 0.048 0.112 0.092 0.120 0.028
#> GSM25357 4 0.554 0.4557 0.012 0.044 0.088 0.688 0.160 0.008
#> GSM25385 3 0.589 0.3961 0.224 0.012 0.624 0.100 0.004 0.036
#> GSM25386 3 0.333 0.5221 0.028 0.032 0.864 0.044 0.004 0.028
#> GSM25399 1 0.492 0.6057 0.712 0.152 0.096 0.000 0.000 0.040
#> GSM25400 1 0.367 0.6704 0.832 0.016 0.052 0.076 0.000 0.024
#> GSM48659 4 0.802 -0.1472 0.000 0.272 0.032 0.336 0.136 0.224
#> GSM48660 2 0.418 0.5428 0.004 0.788 0.036 0.072 0.100 0.000
#> GSM25409 5 0.466 0.4579 0.004 0.236 0.028 0.028 0.700 0.004
#> GSM25410 3 0.367 0.5189 0.048 0.036 0.840 0.060 0.008 0.008
#> GSM25426 4 0.189 0.5624 0.000 0.024 0.044 0.924 0.008 0.000
#> GSM25427 1 0.797 0.2524 0.404 0.264 0.144 0.008 0.148 0.032
#> GSM25540 6 0.628 -0.0678 0.000 0.036 0.384 0.064 0.032 0.484
#> GSM25541 6 0.581 0.1553 0.008 0.016 0.300 0.036 0.048 0.592
#> GSM25542 3 0.531 0.4216 0.000 0.232 0.660 0.068 0.016 0.024
#> GSM25543 3 0.509 0.3917 0.000 0.280 0.644 0.040 0.012 0.024
#> GSM25479 1 0.499 0.6001 0.708 0.020 0.004 0.004 0.100 0.164
#> GSM25480 1 0.699 0.3104 0.476 0.020 0.024 0.012 0.256 0.212
#> GSM25481 2 0.757 0.1281 0.292 0.440 0.108 0.020 0.128 0.012
#> GSM25482 2 0.780 0.0499 0.308 0.352 0.080 0.024 0.228 0.008
#> GSM48654 2 0.713 0.4210 0.000 0.536 0.048 0.208 0.096 0.112
#> GSM48650 4 0.532 -0.0397 0.000 0.392 0.028 0.536 0.040 0.004
#> GSM48651 2 0.584 0.4502 0.000 0.616 0.016 0.244 0.084 0.040
#> GSM48652 2 0.599 0.4099 0.000 0.580 0.028 0.288 0.072 0.032
#> GSM48653 2 0.731 0.1593 0.000 0.392 0.044 0.368 0.056 0.140
#> GSM48662 2 0.536 0.5122 0.000 0.692 0.016 0.088 0.164 0.040
#> GSM48663 2 0.444 0.5282 0.008 0.784 0.060 0.056 0.088 0.004
#> GSM25524 6 0.494 0.3471 0.112 0.016 0.136 0.016 0.000 0.720
#> GSM25525 6 0.571 -0.0644 0.392 0.008 0.008 0.000 0.100 0.492
#> GSM25526 4 0.691 0.1894 0.256 0.008 0.112 0.500 0.000 0.124
#> GSM25527 1 0.486 0.5663 0.680 0.012 0.008 0.024 0.020 0.256
#> GSM25528 6 0.622 0.2481 0.320 0.020 0.148 0.004 0.004 0.504
#> GSM25529 6 0.553 0.0357 0.368 0.000 0.020 0.000 0.084 0.528
#> GSM25530 1 0.566 0.5376 0.668 0.036 0.144 0.020 0.000 0.132
#> GSM25531 1 0.353 0.6769 0.836 0.016 0.036 0.020 0.000 0.092
#> GSM48661 6 0.785 0.0710 0.000 0.228 0.192 0.152 0.028 0.400
#> GSM25561 3 0.644 0.3995 0.080 0.148 0.564 0.004 0.000 0.204
#> GSM25562 2 0.687 -0.1532 0.372 0.408 0.148 0.008 0.000 0.064
#> GSM25563 3 0.594 0.4106 0.036 0.144 0.608 0.008 0.000 0.204
#> GSM25564 2 0.851 0.0849 0.176 0.380 0.040 0.076 0.064 0.264
#> GSM25565 5 0.808 -0.0495 0.000 0.240 0.236 0.208 0.296 0.020
#> GSM25566 5 0.662 0.3709 0.000 0.120 0.068 0.308 0.496 0.008
#> GSM25568 2 0.506 0.2753 0.000 0.644 0.284 0.016 0.028 0.028
#> GSM25569 2 0.627 0.4621 0.000 0.604 0.104 0.052 0.212 0.028
#> GSM25552 5 0.200 0.5957 0.004 0.044 0.004 0.000 0.920 0.028
#> GSM25553 5 0.310 0.5723 0.012 0.064 0.012 0.004 0.868 0.040
#> GSM25578 1 0.442 0.6548 0.772 0.008 0.032 0.004 0.052 0.132
#> GSM25579 5 0.681 -0.2401 0.248 0.008 0.020 0.004 0.368 0.352
#> GSM25580 1 0.294 0.6975 0.876 0.052 0.024 0.000 0.008 0.040
#> GSM25581 1 0.374 0.6846 0.820 0.016 0.028 0.000 0.028 0.108
#> GSM48655 5 0.687 0.0353 0.004 0.316 0.032 0.248 0.396 0.004
#> GSM48656 2 0.665 0.4275 0.008 0.560 0.024 0.088 0.256 0.064
#> GSM48657 2 0.613 0.3484 0.012 0.516 0.008 0.316 0.144 0.004
#> GSM48658 6 0.716 0.2273 0.000 0.164 0.060 0.112 0.112 0.552
#> GSM25624 1 0.563 0.6568 0.712 0.020 0.096 0.016 0.080 0.076
#> GSM25625 3 0.777 0.1940 0.288 0.012 0.340 0.204 0.000 0.156
#> GSM25626 3 0.535 0.4838 0.044 0.012 0.692 0.160 0.000 0.092
#> GSM25627 4 0.379 0.5272 0.020 0.028 0.068 0.828 0.000 0.056
#> GSM25628 3 0.658 0.3466 0.008 0.080 0.520 0.108 0.000 0.284
#> GSM25629 4 0.434 0.4971 0.008 0.012 0.064 0.752 0.000 0.164
#> GSM25630 3 0.626 0.3388 0.036 0.148 0.524 0.004 0.000 0.288
#> GSM25631 6 0.506 0.3833 0.004 0.028 0.056 0.008 0.212 0.692
#> GSM25632 3 0.619 0.2965 0.324 0.012 0.528 0.040 0.000 0.096
#> GSM25633 1 0.351 0.6910 0.840 0.016 0.056 0.000 0.016 0.072
#> GSM25634 1 0.413 0.6738 0.792 0.028 0.108 0.000 0.008 0.064
#> GSM25635 1 0.624 0.6094 0.644 0.020 0.056 0.016 0.160 0.104
#> GSM25656 3 0.723 0.2849 0.012 0.080 0.464 0.252 0.004 0.188
#> GSM25657 1 0.326 0.6821 0.860 0.036 0.040 0.012 0.000 0.052
#> GSM25658 4 0.712 0.1471 0.276 0.028 0.048 0.456 0.000 0.192
#> GSM25659 6 0.587 0.4133 0.116 0.120 0.020 0.024 0.036 0.684
#> GSM25660 1 0.642 0.4339 0.520 0.016 0.020 0.000 0.248 0.196
#> GSM25661 1 0.316 0.6890 0.860 0.020 0.020 0.000 0.020 0.080
#> GSM25662 4 0.478 0.4680 0.000 0.100 0.028 0.756 0.088 0.028
#> GSM25663 5 0.567 0.5482 0.000 0.044 0.044 0.164 0.680 0.068
#> GSM25680 5 0.659 0.4657 0.004 0.028 0.084 0.104 0.600 0.180
#> GSM25681 5 0.577 0.4584 0.008 0.020 0.112 0.028 0.664 0.168
#> GSM25682 5 0.612 0.4323 0.000 0.100 0.048 0.284 0.560 0.008
#> GSM25683 4 0.618 0.1689 0.000 0.064 0.084 0.556 0.288 0.008
#> GSM25684 4 0.607 0.3478 0.000 0.136 0.016 0.636 0.144 0.068
#> GSM25685 4 0.252 0.5543 0.000 0.048 0.020 0.900 0.016 0.016
#> GSM25686 5 0.613 0.4143 0.000 0.140 0.032 0.284 0.540 0.004
#> GSM25687 5 0.602 0.3960 0.000 0.156 0.020 0.280 0.540 0.004
#> GSM48664 1 0.451 0.6392 0.764 0.120 0.072 0.000 0.008 0.036
#> GSM48665 1 0.456 0.6846 0.792 0.068 0.044 0.012 0.060 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 genotype/variation(p) k
#> SD:NMF 93 0.000161 2
#> SD:NMF 65 0.001513 3
#> SD:NMF 43 0.026032 4
#> SD:NMF 53 0.000379 5
#> SD:NMF 36 0.001702 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 100 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 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 0.1196 0.608 0.797 0.2919 0.904 0.904
#> 3 3 0.0831 0.458 0.705 0.6555 0.594 0.563
#> 4 4 0.1106 0.484 0.683 0.1561 0.951 0.911
#> 5 5 0.1597 0.416 0.667 0.0746 0.923 0.853
#> 6 6 0.2446 0.385 0.656 0.0800 0.953 0.896
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
#> GSM25548 2 0.416 0.75855 0.084 0.916
#> GSM25549 2 0.430 0.75930 0.088 0.912
#> GSM25550 2 0.430 0.75883 0.088 0.912
#> GSM25551 2 0.388 0.75996 0.076 0.924
#> GSM25570 2 0.388 0.75839 0.076 0.924
#> GSM25571 2 0.388 0.75839 0.076 0.924
#> GSM25358 2 0.574 0.76026 0.136 0.864
#> GSM25359 2 0.529 0.76348 0.120 0.880
#> GSM25360 2 0.653 0.74517 0.168 0.832
#> GSM25361 2 0.625 0.75523 0.156 0.844
#> GSM25377 1 0.753 0.65256 0.784 0.216
#> GSM25378 2 0.891 0.54894 0.308 0.692
#> GSM25401 2 0.767 0.69313 0.224 0.776
#> GSM25402 2 0.745 0.71365 0.212 0.788
#> GSM25349 2 0.443 0.75329 0.092 0.908
#> GSM25350 2 0.430 0.75407 0.088 0.912
#> GSM25356 2 0.795 0.66180 0.240 0.760
#> GSM25357 2 0.781 0.67075 0.232 0.768
#> GSM25385 2 0.802 0.69494 0.244 0.756
#> GSM25386 2 0.714 0.72306 0.196 0.804
#> GSM25399 1 0.584 0.66524 0.860 0.140
#> GSM25400 2 0.949 0.41916 0.368 0.632
#> GSM48659 2 0.224 0.75598 0.036 0.964
#> GSM48660 2 0.278 0.75020 0.048 0.952
#> GSM25409 2 0.343 0.76408 0.064 0.936
#> GSM25410 2 0.722 0.72643 0.200 0.800
#> GSM25426 2 0.388 0.75996 0.076 0.924
#> GSM25427 2 0.971 0.29253 0.400 0.600
#> GSM25540 2 0.373 0.76432 0.072 0.928
#> GSM25541 2 0.373 0.76432 0.072 0.928
#> GSM25542 2 0.518 0.75125 0.116 0.884
#> GSM25543 2 0.634 0.73751 0.160 0.840
#> GSM25479 2 0.981 0.21055 0.420 0.580
#> GSM25480 2 0.988 0.18073 0.436 0.564
#> GSM25481 2 0.871 0.53552 0.292 0.708
#> GSM25482 2 0.871 0.53552 0.292 0.708
#> GSM48654 2 0.242 0.75253 0.040 0.960
#> GSM48650 2 0.311 0.75329 0.056 0.944
#> GSM48651 2 0.260 0.74974 0.044 0.956
#> GSM48652 2 0.295 0.75114 0.052 0.948
#> GSM48653 2 0.242 0.75218 0.040 0.960
#> GSM48662 2 0.242 0.75197 0.040 0.960
#> GSM48663 2 0.430 0.74875 0.088 0.912
#> GSM25524 2 0.921 0.55100 0.336 0.664
#> GSM25525 2 0.996 0.09594 0.464 0.536
#> GSM25526 2 0.861 0.62484 0.284 0.716
#> GSM25527 2 0.963 0.38155 0.388 0.612
#> GSM25528 2 0.925 0.54790 0.340 0.660
#> GSM25529 2 0.999 0.02589 0.484 0.516
#> GSM25530 2 0.909 0.56270 0.324 0.676
#> GSM25531 2 0.929 0.52149 0.344 0.656
#> GSM48661 2 0.278 0.75754 0.048 0.952
#> GSM25561 2 0.760 0.71994 0.220 0.780
#> GSM25562 2 0.886 0.62064 0.304 0.696
#> GSM25563 2 0.808 0.68663 0.248 0.752
#> GSM25564 2 0.781 0.70280 0.232 0.768
#> GSM25565 2 0.311 0.76284 0.056 0.944
#> GSM25566 2 0.163 0.75616 0.024 0.976
#> GSM25568 2 0.469 0.74006 0.100 0.900
#> GSM25569 2 0.311 0.75734 0.056 0.944
#> GSM25552 2 0.574 0.75467 0.136 0.864
#> GSM25553 2 0.662 0.73945 0.172 0.828
#> GSM25578 2 0.994 0.05658 0.456 0.544
#> GSM25579 2 0.932 0.51237 0.348 0.652
#> GSM25580 2 0.999 -0.09871 0.484 0.516
#> GSM25581 2 0.998 -0.05934 0.476 0.524
#> GSM48655 2 0.295 0.75097 0.052 0.948
#> GSM48656 2 0.260 0.75563 0.044 0.956
#> GSM48657 2 0.295 0.75097 0.052 0.948
#> GSM48658 2 0.242 0.75608 0.040 0.960
#> GSM25624 2 0.966 0.32333 0.392 0.608
#> GSM25625 2 0.876 0.59709 0.296 0.704
#> GSM25626 2 0.689 0.73273 0.184 0.816
#> GSM25627 2 0.855 0.62739 0.280 0.720
#> GSM25628 2 0.671 0.73605 0.176 0.824
#> GSM25629 2 0.388 0.75996 0.076 0.924
#> GSM25630 2 0.876 0.63051 0.296 0.704
#> GSM25631 2 0.388 0.76125 0.076 0.924
#> GSM25632 2 0.980 0.38246 0.416 0.584
#> GSM25633 2 0.995 -0.00802 0.460 0.540
#> GSM25634 2 0.978 0.28765 0.412 0.588
#> GSM25635 2 0.978 0.24352 0.412 0.588
#> GSM25656 2 0.738 0.72138 0.208 0.792
#> GSM25657 1 1.000 0.08240 0.504 0.496
#> GSM25658 2 0.866 0.61832 0.288 0.712
#> GSM25659 2 0.738 0.70611 0.208 0.792
#> GSM25660 2 0.985 0.20530 0.428 0.572
#> GSM25661 2 0.994 0.03749 0.456 0.544
#> GSM25662 2 0.373 0.76061 0.072 0.928
#> GSM25663 2 0.373 0.76061 0.072 0.928
#> GSM25680 2 0.529 0.75951 0.120 0.880
#> GSM25681 2 0.541 0.75990 0.124 0.876
#> GSM25682 2 0.204 0.75327 0.032 0.968
#> GSM25683 2 0.204 0.75327 0.032 0.968
#> GSM25684 2 0.224 0.75644 0.036 0.964
#> GSM25685 2 0.358 0.76388 0.068 0.932
#> GSM25686 2 0.204 0.75327 0.032 0.968
#> GSM25687 2 0.204 0.75327 0.032 0.968
#> GSM48664 1 0.574 0.66479 0.864 0.136
#> GSM48665 1 1.000 0.11134 0.508 0.492
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM25548 2 0.585 0.6659 0.140 0.792 0.068
#> GSM25549 2 0.621 0.6636 0.152 0.772 0.076
#> GSM25550 2 0.591 0.6680 0.144 0.788 0.068
#> GSM25551 2 0.475 0.6874 0.076 0.852 0.072
#> GSM25570 2 0.591 0.6635 0.144 0.788 0.068
#> GSM25571 2 0.591 0.6635 0.144 0.788 0.068
#> GSM25358 2 0.670 0.6501 0.164 0.744 0.092
#> GSM25359 2 0.651 0.6664 0.156 0.756 0.088
#> GSM25360 2 0.924 0.1519 0.264 0.528 0.208
#> GSM25361 2 0.866 0.3611 0.244 0.592 0.164
#> GSM25377 1 0.850 0.1190 0.544 0.104 0.352
#> GSM25378 1 0.828 0.2942 0.472 0.452 0.076
#> GSM25401 2 0.811 0.4227 0.272 0.620 0.108
#> GSM25402 2 0.781 0.4931 0.244 0.652 0.104
#> GSM25349 2 0.415 0.6816 0.044 0.876 0.080
#> GSM25350 2 0.425 0.6825 0.048 0.872 0.080
#> GSM25356 2 0.747 0.5152 0.216 0.684 0.100
#> GSM25357 2 0.736 0.5271 0.212 0.692 0.096
#> GSM25385 2 0.914 0.0524 0.360 0.488 0.152
#> GSM25386 2 0.946 -0.0486 0.248 0.500 0.252
#> GSM25399 1 0.753 0.0868 0.564 0.044 0.392
#> GSM25400 1 0.776 0.4140 0.564 0.380 0.056
#> GSM48659 2 0.231 0.7132 0.032 0.944 0.024
#> GSM48660 2 0.253 0.6989 0.020 0.936 0.044
#> GSM25409 2 0.408 0.7162 0.072 0.880 0.048
#> GSM25410 2 0.933 0.0822 0.268 0.516 0.216
#> GSM25426 2 0.475 0.6874 0.076 0.852 0.072
#> GSM25427 1 0.801 0.3678 0.524 0.412 0.064
#> GSM25540 2 0.507 0.7001 0.096 0.836 0.068
#> GSM25541 2 0.507 0.7001 0.096 0.836 0.068
#> GSM25542 2 0.585 0.6551 0.060 0.788 0.152
#> GSM25543 2 0.773 0.5123 0.132 0.676 0.192
#> GSM25479 1 0.644 0.5332 0.696 0.276 0.028
#> GSM25480 1 0.638 0.5336 0.712 0.256 0.032
#> GSM25481 2 0.785 0.1081 0.384 0.556 0.060
#> GSM25482 2 0.785 0.1081 0.384 0.556 0.060
#> GSM48654 2 0.192 0.7082 0.020 0.956 0.024
#> GSM48650 2 0.280 0.6959 0.016 0.924 0.060
#> GSM48651 2 0.227 0.6986 0.016 0.944 0.040
#> GSM48652 2 0.238 0.6983 0.016 0.940 0.044
#> GSM48653 2 0.231 0.7025 0.024 0.944 0.032
#> GSM48662 2 0.243 0.7055 0.024 0.940 0.036
#> GSM48663 2 0.398 0.6807 0.048 0.884 0.068
#> GSM25524 1 0.929 -0.3296 0.476 0.168 0.356
#> GSM25525 1 0.653 0.4553 0.744 0.188 0.068
#> GSM25526 2 0.862 0.0999 0.388 0.508 0.104
#> GSM25527 1 0.768 0.4511 0.608 0.328 0.064
#> GSM25528 1 0.920 -0.2488 0.504 0.168 0.328
#> GSM25529 1 0.639 0.4716 0.752 0.184 0.064
#> GSM25530 1 0.952 -0.0822 0.488 0.232 0.280
#> GSM25531 1 0.892 0.2171 0.560 0.268 0.172
#> GSM48661 2 0.269 0.7160 0.036 0.932 0.032
#> GSM25561 1 0.994 -0.4125 0.364 0.356 0.280
#> GSM25562 1 0.972 -0.0830 0.416 0.360 0.224
#> GSM25563 3 0.996 0.2807 0.292 0.336 0.372
#> GSM25564 2 0.868 0.2254 0.340 0.540 0.120
#> GSM25565 2 0.378 0.7164 0.044 0.892 0.064
#> GSM25566 2 0.218 0.7145 0.032 0.948 0.020
#> GSM25568 2 0.482 0.6916 0.064 0.848 0.088
#> GSM25569 2 0.348 0.7160 0.048 0.904 0.048
#> GSM25552 2 0.680 0.6181 0.204 0.724 0.072
#> GSM25553 2 0.759 0.4645 0.300 0.632 0.068
#> GSM25578 1 0.573 0.5347 0.752 0.228 0.020
#> GSM25579 1 0.798 0.3204 0.536 0.400 0.064
#> GSM25580 1 0.645 0.5253 0.736 0.212 0.052
#> GSM25581 1 0.672 0.5310 0.720 0.220 0.060
#> GSM48655 2 0.266 0.7022 0.024 0.932 0.044
#> GSM48656 2 0.266 0.7161 0.044 0.932 0.024
#> GSM48657 2 0.285 0.6945 0.020 0.924 0.056
#> GSM48658 2 0.253 0.7162 0.044 0.936 0.020
#> GSM25624 1 0.785 0.4662 0.588 0.344 0.068
#> GSM25625 2 0.894 -0.2494 0.432 0.444 0.124
#> GSM25626 2 0.919 0.1521 0.256 0.536 0.208
#> GSM25627 2 0.859 0.1278 0.376 0.520 0.104
#> GSM25628 2 0.911 0.2156 0.244 0.548 0.208
#> GSM25629 2 0.475 0.6874 0.076 0.852 0.072
#> GSM25630 3 0.917 0.3241 0.372 0.152 0.476
#> GSM25631 2 0.560 0.6748 0.136 0.804 0.060
#> GSM25632 1 0.873 0.1852 0.580 0.260 0.160
#> GSM25633 1 0.661 0.5393 0.716 0.236 0.048
#> GSM25634 1 0.873 0.3369 0.568 0.288 0.144
#> GSM25635 1 0.753 0.5152 0.624 0.316 0.060
#> GSM25656 2 0.913 -0.0436 0.168 0.520 0.312
#> GSM25657 1 0.734 0.5032 0.688 0.224 0.088
#> GSM25658 2 0.863 0.0807 0.392 0.504 0.104
#> GSM25659 2 0.784 0.2091 0.380 0.560 0.060
#> GSM25660 1 0.700 0.5275 0.672 0.280 0.048
#> GSM25661 1 0.577 0.5348 0.756 0.220 0.024
#> GSM25662 2 0.560 0.6783 0.136 0.804 0.060
#> GSM25663 2 0.560 0.6783 0.136 0.804 0.060
#> GSM25680 2 0.669 0.6449 0.148 0.748 0.104
#> GSM25681 2 0.688 0.6338 0.156 0.736 0.108
#> GSM25682 2 0.244 0.7079 0.032 0.940 0.028
#> GSM25683 2 0.244 0.7079 0.032 0.940 0.028
#> GSM25684 2 0.219 0.7120 0.028 0.948 0.024
#> GSM25685 2 0.348 0.7082 0.048 0.904 0.048
#> GSM25686 2 0.244 0.7079 0.032 0.940 0.028
#> GSM25687 2 0.244 0.7079 0.032 0.940 0.028
#> GSM48664 1 0.733 0.0911 0.576 0.036 0.388
#> GSM48665 1 0.715 0.5192 0.696 0.228 0.076
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM25548 2 0.577 0.62696 0.160 0.732 0.096 0.012
#> GSM25549 2 0.607 0.62244 0.168 0.716 0.096 0.020
#> GSM25550 2 0.575 0.62771 0.164 0.728 0.100 0.008
#> GSM25551 2 0.449 0.64702 0.056 0.820 0.112 0.012
#> GSM25570 2 0.571 0.62551 0.160 0.736 0.092 0.012
#> GSM25571 2 0.571 0.62551 0.160 0.736 0.092 0.012
#> GSM25358 2 0.632 0.61967 0.148 0.696 0.140 0.016
#> GSM25359 2 0.609 0.63416 0.132 0.716 0.136 0.016
#> GSM25360 2 0.880 0.13128 0.260 0.444 0.236 0.060
#> GSM25361 2 0.852 0.30581 0.252 0.496 0.192 0.060
#> GSM25377 4 0.729 0.82538 0.404 0.076 0.028 0.492
#> GSM25378 1 0.739 0.42299 0.524 0.364 0.072 0.040
#> GSM25401 2 0.797 0.33033 0.260 0.540 0.160 0.040
#> GSM25402 2 0.794 0.35980 0.248 0.556 0.148 0.048
#> GSM25349 2 0.462 0.62446 0.024 0.824 0.084 0.068
#> GSM25350 2 0.455 0.62448 0.024 0.828 0.080 0.068
#> GSM25356 2 0.751 0.45620 0.220 0.616 0.080 0.084
#> GSM25357 2 0.743 0.46823 0.216 0.624 0.080 0.080
#> GSM25385 2 0.831 -0.04818 0.352 0.396 0.232 0.020
#> GSM25386 2 0.805 -0.13121 0.212 0.396 0.380 0.012
#> GSM25399 4 0.578 0.90237 0.412 0.032 0.000 0.556
#> GSM25400 1 0.708 0.53601 0.596 0.296 0.044 0.064
#> GSM48659 2 0.209 0.67685 0.020 0.940 0.028 0.012
#> GSM48660 2 0.310 0.64994 0.016 0.896 0.064 0.024
#> GSM25409 2 0.426 0.68435 0.072 0.844 0.060 0.024
#> GSM25410 2 0.800 0.03506 0.232 0.420 0.340 0.008
#> GSM25426 2 0.449 0.64702 0.056 0.820 0.112 0.012
#> GSM25427 1 0.735 0.44082 0.556 0.324 0.036 0.084
#> GSM25540 2 0.462 0.66871 0.096 0.816 0.076 0.012
#> GSM25541 2 0.462 0.66871 0.096 0.816 0.076 0.012
#> GSM25542 2 0.552 0.61825 0.040 0.756 0.164 0.040
#> GSM25543 2 0.723 0.41707 0.080 0.608 0.264 0.048
#> GSM25479 1 0.530 0.59361 0.760 0.176 0.032 0.032
#> GSM25480 1 0.526 0.58572 0.772 0.156 0.032 0.040
#> GSM25481 2 0.751 0.00850 0.396 0.488 0.040 0.076
#> GSM25482 2 0.751 0.00850 0.396 0.488 0.040 0.076
#> GSM48654 2 0.201 0.67011 0.008 0.940 0.040 0.012
#> GSM48650 2 0.320 0.64717 0.008 0.888 0.072 0.032
#> GSM48651 2 0.260 0.65107 0.004 0.912 0.064 0.020
#> GSM48652 2 0.281 0.64985 0.004 0.904 0.064 0.028
#> GSM48653 2 0.241 0.65551 0.004 0.920 0.060 0.016
#> GSM48662 2 0.222 0.66385 0.008 0.932 0.044 0.016
#> GSM48663 2 0.441 0.62602 0.024 0.836 0.076 0.064
#> GSM25524 1 0.862 0.12313 0.488 0.064 0.240 0.208
#> GSM25525 1 0.586 0.47671 0.760 0.092 0.080 0.068
#> GSM25526 2 0.818 -0.01183 0.376 0.432 0.160 0.032
#> GSM25527 1 0.661 0.59517 0.660 0.240 0.052 0.048
#> GSM25528 1 0.858 0.18157 0.500 0.068 0.236 0.196
#> GSM25529 1 0.550 0.49376 0.780 0.092 0.076 0.052
#> GSM25530 1 0.883 0.30735 0.508 0.124 0.204 0.164
#> GSM25531 1 0.782 0.51110 0.608 0.160 0.152 0.080
#> GSM48661 2 0.259 0.68223 0.028 0.920 0.040 0.012
#> GSM25561 1 0.928 -0.00675 0.360 0.236 0.316 0.088
#> GSM25562 1 0.932 0.23664 0.408 0.256 0.228 0.108
#> GSM25563 3 0.863 0.20416 0.220 0.240 0.480 0.060
#> GSM25564 2 0.820 -0.05197 0.372 0.444 0.144 0.040
#> GSM25565 2 0.365 0.68244 0.044 0.872 0.068 0.016
#> GSM25566 2 0.210 0.68101 0.020 0.936 0.040 0.004
#> GSM25568 2 0.528 0.64093 0.056 0.792 0.096 0.056
#> GSM25569 2 0.377 0.68057 0.040 0.864 0.080 0.016
#> GSM25552 2 0.655 0.57808 0.220 0.664 0.096 0.020
#> GSM25553 2 0.739 0.33839 0.336 0.540 0.096 0.028
#> GSM25578 1 0.485 0.55361 0.796 0.140 0.020 0.044
#> GSM25579 1 0.726 0.47286 0.564 0.320 0.084 0.032
#> GSM25580 1 0.488 0.49824 0.796 0.124 0.012 0.068
#> GSM25581 1 0.507 0.51355 0.784 0.128 0.012 0.076
#> GSM48655 2 0.289 0.66600 0.020 0.908 0.048 0.024
#> GSM48656 2 0.256 0.68255 0.036 0.920 0.036 0.008
#> GSM48657 2 0.338 0.64443 0.016 0.884 0.068 0.032
#> GSM48658 2 0.241 0.68258 0.036 0.924 0.036 0.004
#> GSM25624 1 0.674 0.56416 0.640 0.260 0.044 0.056
#> GSM25625 1 0.825 0.27690 0.444 0.364 0.152 0.040
#> GSM25626 2 0.802 0.08639 0.216 0.440 0.332 0.012
#> GSM25627 2 0.816 0.01672 0.364 0.444 0.160 0.032
#> GSM25628 2 0.794 0.12547 0.204 0.456 0.328 0.012
#> GSM25629 2 0.449 0.64702 0.056 0.820 0.112 0.012
#> GSM25630 3 0.742 0.06332 0.164 0.032 0.608 0.196
#> GSM25631 2 0.553 0.64038 0.156 0.752 0.076 0.016
#> GSM25632 1 0.799 0.45961 0.564 0.168 0.216 0.052
#> GSM25633 1 0.514 0.54448 0.784 0.132 0.020 0.064
#> GSM25634 1 0.846 0.45575 0.552 0.176 0.116 0.156
#> GSM25635 1 0.605 0.58547 0.684 0.244 0.024 0.048
#> GSM25656 3 0.716 0.35998 0.024 0.340 0.552 0.084
#> GSM25657 1 0.651 0.48102 0.696 0.156 0.032 0.116
#> GSM25658 2 0.818 -0.03091 0.380 0.428 0.160 0.032
#> GSM25659 2 0.705 -0.07009 0.456 0.460 0.056 0.028
#> GSM25660 1 0.595 0.59656 0.708 0.212 0.024 0.056
#> GSM25661 1 0.405 0.53876 0.836 0.124 0.012 0.028
#> GSM25662 2 0.526 0.64126 0.144 0.760 0.092 0.004
#> GSM25663 2 0.531 0.63903 0.148 0.756 0.092 0.004
#> GSM25680 2 0.647 0.60801 0.152 0.696 0.124 0.028
#> GSM25681 2 0.673 0.59214 0.168 0.676 0.124 0.032
#> GSM25682 2 0.250 0.66855 0.020 0.924 0.040 0.016
#> GSM25683 2 0.250 0.66855 0.020 0.924 0.040 0.016
#> GSM25684 2 0.194 0.67611 0.016 0.944 0.032 0.008
#> GSM25685 2 0.281 0.67407 0.028 0.904 0.064 0.004
#> GSM25686 2 0.250 0.66855 0.020 0.924 0.040 0.016
#> GSM25687 2 0.250 0.66855 0.020 0.924 0.040 0.016
#> GSM48664 4 0.581 0.90073 0.428 0.024 0.004 0.544
#> GSM48665 1 0.594 0.50286 0.720 0.160 0.012 0.108
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM25548 2 0.531 0.5109 0.156 0.676 0.168 0.000 0.000
#> GSM25549 2 0.566 0.4903 0.168 0.648 0.180 0.000 0.004
#> GSM25550 2 0.534 0.5103 0.156 0.672 0.172 0.000 0.000
#> GSM25551 2 0.422 0.5564 0.036 0.780 0.168 0.000 0.016
#> GSM25570 2 0.527 0.5091 0.160 0.680 0.160 0.000 0.000
#> GSM25571 2 0.527 0.5091 0.160 0.680 0.160 0.000 0.000
#> GSM25358 2 0.563 0.4775 0.132 0.644 0.220 0.004 0.000
#> GSM25359 2 0.535 0.5143 0.108 0.672 0.216 0.004 0.000
#> GSM25360 2 0.838 -0.2836 0.232 0.380 0.292 0.068 0.028
#> GSM25361 2 0.811 -0.1088 0.244 0.420 0.260 0.048 0.028
#> GSM25377 5 0.644 0.8345 0.352 0.056 0.032 0.016 0.544
#> GSM25378 1 0.702 0.2995 0.488 0.336 0.124 0.000 0.052
#> GSM25401 2 0.715 0.0805 0.224 0.500 0.236 0.000 0.040
#> GSM25402 2 0.730 0.1370 0.204 0.520 0.224 0.008 0.044
#> GSM25349 2 0.368 0.5845 0.012 0.832 0.120 0.004 0.032
#> GSM25350 2 0.363 0.5850 0.012 0.836 0.116 0.004 0.032
#> GSM25356 2 0.702 0.3319 0.168 0.600 0.136 0.008 0.088
#> GSM25357 2 0.698 0.3413 0.164 0.604 0.140 0.008 0.084
#> GSM25385 2 0.766 -0.3996 0.308 0.328 0.328 0.008 0.028
#> GSM25386 3 0.758 0.5501 0.152 0.324 0.460 0.048 0.016
#> GSM25399 5 0.478 0.8768 0.348 0.016 0.004 0.004 0.628
#> GSM25400 1 0.693 0.4247 0.544 0.280 0.088 0.000 0.088
#> GSM48659 2 0.213 0.6349 0.024 0.920 0.052 0.000 0.004
#> GSM48660 2 0.233 0.6107 0.004 0.904 0.076 0.000 0.016
#> GSM25409 2 0.408 0.6255 0.064 0.800 0.128 0.000 0.008
#> GSM25410 3 0.731 0.5147 0.192 0.352 0.424 0.024 0.008
#> GSM25426 2 0.422 0.5564 0.036 0.780 0.168 0.000 0.016
#> GSM25427 1 0.709 0.3673 0.504 0.316 0.076 0.000 0.104
#> GSM25540 2 0.444 0.5792 0.088 0.756 0.156 0.000 0.000
#> GSM25541 2 0.444 0.5792 0.088 0.756 0.156 0.000 0.000
#> GSM25542 2 0.484 0.5228 0.036 0.720 0.220 0.000 0.024
#> GSM25543 2 0.634 0.1785 0.060 0.556 0.340 0.008 0.036
#> GSM25479 1 0.494 0.5499 0.780 0.112 0.048 0.028 0.032
#> GSM25480 1 0.489 0.5385 0.788 0.096 0.044 0.044 0.028
#> GSM25481 2 0.703 0.0336 0.356 0.488 0.076 0.004 0.076
#> GSM25482 2 0.703 0.0336 0.356 0.488 0.076 0.004 0.076
#> GSM48654 2 0.203 0.6293 0.012 0.924 0.056 0.000 0.008
#> GSM48650 2 0.245 0.6080 0.004 0.896 0.084 0.000 0.016
#> GSM48651 2 0.194 0.6131 0.000 0.920 0.068 0.000 0.012
#> GSM48652 2 0.205 0.6139 0.000 0.916 0.068 0.000 0.016
#> GSM48653 2 0.192 0.6177 0.004 0.924 0.064 0.000 0.008
#> GSM48662 2 0.157 0.6258 0.000 0.936 0.060 0.000 0.004
#> GSM48663 2 0.353 0.5843 0.012 0.844 0.108 0.004 0.032
#> GSM25524 1 0.809 0.1559 0.472 0.036 0.160 0.264 0.068
#> GSM25525 1 0.496 0.4599 0.772 0.064 0.024 0.120 0.020
#> GSM25526 2 0.753 -0.1490 0.332 0.388 0.236 0.000 0.044
#> GSM25527 1 0.640 0.5397 0.652 0.176 0.116 0.016 0.040
#> GSM25528 1 0.799 0.2435 0.496 0.044 0.164 0.240 0.056
#> GSM25529 1 0.462 0.4722 0.792 0.064 0.024 0.108 0.012
#> GSM25530 1 0.825 0.3059 0.520 0.076 0.156 0.168 0.080
#> GSM25531 1 0.711 0.4684 0.624 0.104 0.160 0.064 0.048
#> GSM48661 2 0.252 0.6340 0.024 0.900 0.068 0.000 0.008
#> GSM25561 3 0.931 -0.0344 0.308 0.132 0.312 0.164 0.084
#> GSM25562 1 0.934 0.1041 0.364 0.200 0.200 0.080 0.156
#> GSM25563 3 0.865 0.2930 0.164 0.160 0.472 0.144 0.060
#> GSM25564 2 0.824 -0.1509 0.344 0.404 0.140 0.040 0.072
#> GSM25565 2 0.332 0.6246 0.044 0.848 0.104 0.000 0.004
#> GSM25566 2 0.252 0.6334 0.024 0.896 0.076 0.000 0.004
#> GSM25568 2 0.450 0.5898 0.036 0.784 0.148 0.012 0.020
#> GSM25569 2 0.348 0.6267 0.032 0.844 0.108 0.000 0.016
#> GSM25552 2 0.632 0.3995 0.232 0.592 0.160 0.004 0.012
#> GSM25553 2 0.719 0.1109 0.336 0.464 0.164 0.008 0.028
#> GSM25578 1 0.367 0.5056 0.852 0.080 0.012 0.032 0.024
#> GSM25579 1 0.677 0.3370 0.576 0.260 0.120 0.024 0.020
#> GSM25580 1 0.506 0.4770 0.764 0.096 0.012 0.028 0.100
#> GSM25581 1 0.528 0.4845 0.756 0.096 0.012 0.048 0.088
#> GSM48655 2 0.216 0.6244 0.012 0.916 0.064 0.000 0.008
#> GSM48656 2 0.250 0.6330 0.032 0.900 0.064 0.000 0.004
#> GSM48657 2 0.251 0.6053 0.004 0.892 0.088 0.000 0.016
#> GSM48658 2 0.234 0.6330 0.032 0.904 0.064 0.000 0.000
#> GSM25624 1 0.660 0.5012 0.612 0.240 0.064 0.012 0.072
#> GSM25625 1 0.771 -0.0445 0.420 0.324 0.196 0.008 0.052
#> GSM25626 3 0.744 0.4842 0.180 0.368 0.412 0.024 0.016
#> GSM25627 2 0.752 -0.1410 0.328 0.392 0.236 0.000 0.044
#> GSM25628 3 0.736 0.4525 0.164 0.384 0.412 0.024 0.016
#> GSM25629 2 0.422 0.5564 0.036 0.780 0.168 0.000 0.016
#> GSM25630 4 0.389 0.0000 0.016 0.008 0.160 0.804 0.012
#> GSM25631 2 0.512 0.5317 0.148 0.696 0.156 0.000 0.000
#> GSM25632 1 0.745 0.3844 0.556 0.124 0.224 0.072 0.024
#> GSM25633 1 0.427 0.5090 0.820 0.088 0.020 0.020 0.052
#> GSM25634 1 0.886 0.2817 0.448 0.116 0.096 0.128 0.212
#> GSM25635 1 0.544 0.5544 0.700 0.196 0.044 0.000 0.060
#> GSM25656 3 0.744 -0.3969 0.008 0.128 0.560 0.136 0.168
#> GSM25657 1 0.538 0.4297 0.740 0.100 0.032 0.012 0.116
#> GSM25658 2 0.753 -0.1573 0.336 0.384 0.236 0.000 0.044
#> GSM25659 1 0.688 0.0201 0.452 0.404 0.108 0.012 0.024
#> GSM25660 1 0.554 0.5620 0.720 0.160 0.052 0.008 0.060
#> GSM25661 1 0.334 0.4966 0.868 0.072 0.008 0.016 0.036
#> GSM25662 2 0.504 0.5402 0.136 0.716 0.144 0.000 0.004
#> GSM25663 2 0.508 0.5370 0.140 0.712 0.144 0.000 0.004
#> GSM25680 2 0.616 0.4334 0.152 0.616 0.216 0.004 0.012
#> GSM25681 2 0.632 0.4046 0.168 0.596 0.220 0.004 0.012
#> GSM25682 2 0.196 0.6299 0.012 0.928 0.052 0.000 0.008
#> GSM25683 2 0.196 0.6299 0.012 0.928 0.052 0.000 0.008
#> GSM25684 2 0.203 0.6337 0.020 0.924 0.052 0.000 0.004
#> GSM25685 2 0.268 0.6201 0.028 0.880 0.092 0.000 0.000
#> GSM25686 2 0.188 0.6294 0.012 0.932 0.048 0.000 0.008
#> GSM25687 2 0.188 0.6294 0.012 0.932 0.048 0.000 0.008
#> GSM48664 5 0.476 0.8883 0.380 0.012 0.008 0.000 0.600
#> GSM48665 1 0.489 0.4671 0.748 0.116 0.016 0.000 0.120
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM25548 5 0.515 0.42805 0.132 0.000 0.236 0.000 0.628 0.004
#> GSM25549 5 0.544 0.40475 0.136 0.000 0.248 0.000 0.604 0.012
#> GSM25550 5 0.517 0.42757 0.132 0.000 0.240 0.000 0.624 0.004
#> GSM25551 5 0.392 0.52691 0.020 0.000 0.204 0.016 0.756 0.004
#> GSM25570 5 0.514 0.42692 0.136 0.000 0.228 0.000 0.632 0.004
#> GSM25571 5 0.514 0.42692 0.136 0.000 0.228 0.000 0.632 0.004
#> GSM25358 5 0.548 0.40025 0.108 0.004 0.272 0.004 0.604 0.008
#> GSM25359 5 0.514 0.44554 0.088 0.004 0.272 0.000 0.628 0.008
#> GSM25360 3 0.788 0.41988 0.176 0.052 0.400 0.020 0.304 0.048
#> GSM25361 3 0.758 0.29477 0.188 0.032 0.380 0.024 0.344 0.032
#> GSM25377 4 0.564 0.80485 0.348 0.016 0.024 0.568 0.028 0.016
#> GSM25378 1 0.672 0.25339 0.508 0.000 0.164 0.048 0.264 0.016
#> GSM25401 5 0.691 -0.03770 0.200 0.000 0.288 0.048 0.452 0.012
#> GSM25402 5 0.700 0.00687 0.188 0.008 0.272 0.052 0.472 0.008
#> GSM25349 5 0.401 0.55429 0.012 0.004 0.136 0.024 0.796 0.028
#> GSM25350 5 0.393 0.55654 0.012 0.004 0.136 0.024 0.800 0.024
#> GSM25356 5 0.688 0.25128 0.144 0.004 0.196 0.076 0.556 0.024
#> GSM25357 5 0.679 0.26070 0.136 0.004 0.204 0.076 0.560 0.020
#> GSM25385 3 0.724 0.36075 0.244 0.016 0.436 0.032 0.256 0.016
#> GSM25386 3 0.679 0.50794 0.080 0.060 0.552 0.008 0.260 0.040
#> GSM25399 4 0.424 0.75951 0.264 0.008 0.016 0.700 0.012 0.000
#> GSM25400 1 0.669 0.39586 0.560 0.000 0.124 0.096 0.204 0.016
#> GSM48659 5 0.189 0.62031 0.020 0.000 0.056 0.004 0.920 0.000
#> GSM48660 5 0.255 0.59406 0.008 0.000 0.076 0.008 0.888 0.020
#> GSM25409 5 0.370 0.59895 0.056 0.000 0.144 0.008 0.792 0.000
#> GSM25410 3 0.668 0.55644 0.112 0.048 0.528 0.004 0.284 0.024
#> GSM25426 5 0.392 0.52691 0.020 0.000 0.204 0.016 0.756 0.004
#> GSM25427 1 0.698 0.34594 0.524 0.000 0.096 0.112 0.240 0.028
#> GSM25540 5 0.427 0.52596 0.080 0.000 0.204 0.000 0.716 0.000
#> GSM25541 5 0.427 0.52596 0.080 0.000 0.204 0.000 0.716 0.000
#> GSM25542 5 0.442 0.44989 0.004 0.004 0.276 0.016 0.684 0.016
#> GSM25543 5 0.565 0.05749 0.020 0.008 0.404 0.020 0.516 0.032
#> GSM25479 1 0.398 0.52683 0.828 0.016 0.060 0.032 0.048 0.016
#> GSM25480 1 0.409 0.52015 0.824 0.024 0.064 0.020 0.036 0.032
#> GSM25481 5 0.717 -0.04696 0.372 0.004 0.108 0.072 0.416 0.028
#> GSM25482 5 0.717 -0.04696 0.372 0.004 0.108 0.072 0.416 0.028
#> GSM48654 5 0.175 0.61735 0.016 0.000 0.044 0.004 0.932 0.004
#> GSM48650 5 0.274 0.59082 0.008 0.000 0.084 0.008 0.876 0.024
#> GSM48651 5 0.198 0.60489 0.008 0.000 0.064 0.008 0.916 0.004
#> GSM48652 5 0.246 0.59880 0.008 0.000 0.064 0.008 0.896 0.024
#> GSM48653 5 0.175 0.61003 0.008 0.000 0.056 0.004 0.928 0.004
#> GSM48662 5 0.179 0.61641 0.008 0.000 0.068 0.000 0.920 0.004
#> GSM48663 5 0.394 0.55263 0.016 0.004 0.128 0.024 0.804 0.024
#> GSM25524 1 0.848 0.03041 0.368 0.220 0.208 0.136 0.016 0.052
#> GSM25525 1 0.413 0.45785 0.808 0.104 0.028 0.028 0.016 0.016
#> GSM25526 5 0.723 -0.26770 0.312 0.000 0.296 0.052 0.328 0.012
#> GSM25527 1 0.585 0.51280 0.664 0.008 0.140 0.060 0.120 0.008
#> GSM25528 1 0.792 0.18279 0.444 0.200 0.188 0.124 0.012 0.032
#> GSM25529 1 0.384 0.46802 0.824 0.100 0.028 0.020 0.016 0.012
#> GSM25530 1 0.800 0.22056 0.480 0.116 0.200 0.128 0.028 0.048
#> GSM25531 1 0.713 0.39229 0.564 0.056 0.224 0.064 0.052 0.040
#> GSM48661 5 0.241 0.61304 0.012 0.000 0.108 0.000 0.876 0.004
#> GSM25561 3 0.902 -0.11368 0.204 0.168 0.356 0.064 0.056 0.152
#> GSM25562 3 0.909 -0.01035 0.284 0.052 0.288 0.072 0.132 0.172
#> GSM25563 3 0.761 -0.11739 0.068 0.144 0.516 0.008 0.092 0.172
#> GSM25564 5 0.805 -0.29114 0.332 0.024 0.176 0.028 0.352 0.088
#> GSM25565 5 0.331 0.59187 0.032 0.000 0.140 0.004 0.820 0.004
#> GSM25566 5 0.241 0.61382 0.016 0.000 0.100 0.004 0.880 0.000
#> GSM25568 5 0.478 0.54547 0.020 0.008 0.188 0.016 0.728 0.040
#> GSM25569 5 0.341 0.59788 0.012 0.000 0.160 0.008 0.808 0.012
#> GSM25552 5 0.609 0.28133 0.196 0.000 0.236 0.004 0.544 0.020
#> GSM25553 5 0.693 -0.11037 0.296 0.004 0.260 0.012 0.404 0.024
#> GSM25578 1 0.276 0.49993 0.896 0.028 0.028 0.020 0.020 0.008
#> GSM25579 1 0.634 0.26143 0.580 0.016 0.176 0.024 0.196 0.008
#> GSM25580 1 0.419 0.45931 0.812 0.032 0.020 0.088 0.024 0.024
#> GSM25581 1 0.445 0.46109 0.800 0.044 0.016 0.076 0.028 0.036
#> GSM48655 5 0.226 0.60609 0.008 0.000 0.068 0.000 0.900 0.024
#> GSM48656 5 0.249 0.61163 0.020 0.000 0.100 0.000 0.876 0.004
#> GSM48657 5 0.279 0.58793 0.008 0.000 0.088 0.008 0.872 0.024
#> GSM48658 5 0.235 0.61160 0.020 0.000 0.100 0.000 0.880 0.000
#> GSM25624 1 0.634 0.47570 0.624 0.012 0.108 0.076 0.168 0.012
#> GSM25625 1 0.761 -0.21272 0.360 0.008 0.288 0.060 0.264 0.020
#> GSM25626 3 0.640 0.55337 0.100 0.036 0.544 0.008 0.296 0.016
#> GSM25627 5 0.724 -0.26679 0.308 0.000 0.300 0.052 0.328 0.012
#> GSM25628 3 0.631 0.51870 0.084 0.036 0.536 0.008 0.320 0.016
#> GSM25629 5 0.392 0.52691 0.020 0.000 0.204 0.016 0.756 0.004
#> GSM25630 2 0.193 0.00000 0.000 0.916 0.068 0.008 0.004 0.004
#> GSM25631 5 0.497 0.45989 0.120 0.000 0.224 0.000 0.652 0.004
#> GSM25632 1 0.723 0.37516 0.536 0.080 0.244 0.040 0.076 0.024
#> GSM25633 1 0.361 0.49196 0.848 0.016 0.040 0.060 0.024 0.012
#> GSM25634 1 0.879 0.02738 0.368 0.076 0.132 0.132 0.040 0.252
#> GSM25635 1 0.523 0.52700 0.712 0.000 0.072 0.084 0.124 0.008
#> GSM25656 6 0.576 0.00000 0.000 0.060 0.192 0.008 0.096 0.644
#> GSM25657 1 0.454 0.41079 0.764 0.012 0.044 0.128 0.052 0.000
#> GSM25658 5 0.724 -0.27375 0.316 0.000 0.296 0.052 0.324 0.012
#> GSM25659 1 0.669 -0.12785 0.444 0.004 0.176 0.016 0.340 0.020
#> GSM25660 1 0.531 0.53858 0.728 0.008 0.088 0.068 0.092 0.016
#> GSM25661 1 0.236 0.49375 0.916 0.016 0.016 0.016 0.016 0.020
#> GSM25662 5 0.490 0.46973 0.112 0.000 0.208 0.000 0.672 0.008
#> GSM25663 5 0.494 0.46562 0.116 0.000 0.208 0.000 0.668 0.008
#> GSM25680 5 0.581 0.29996 0.116 0.000 0.308 0.004 0.552 0.020
#> GSM25681 5 0.584 0.24468 0.120 0.000 0.332 0.004 0.528 0.016
#> GSM25682 5 0.194 0.61642 0.016 0.000 0.056 0.004 0.920 0.004
#> GSM25683 5 0.194 0.61642 0.016 0.000 0.056 0.004 0.920 0.004
#> GSM25684 5 0.183 0.61925 0.020 0.000 0.052 0.004 0.924 0.000
#> GSM25685 5 0.262 0.60288 0.024 0.000 0.104 0.004 0.868 0.000
#> GSM25686 5 0.188 0.61619 0.016 0.000 0.052 0.004 0.924 0.004
#> GSM25687 5 0.188 0.61619 0.016 0.000 0.052 0.004 0.924 0.004
#> GSM48664 4 0.420 0.84006 0.344 0.000 0.004 0.636 0.012 0.004
#> GSM48665 1 0.416 0.45560 0.776 0.000 0.024 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 genotype/variation(p) k
#> CV:hclust 81 0.02792 2
#> CV:hclust 60 0.00118 3
#> CV:hclust 62 0.01331 4
#> CV:hclust 54 0.00959 5
#> CV:hclust 44 0.00190 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 100 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 0.938 0.960 0.982 0.5051 0.495 0.495
#> 3 3 0.641 0.742 0.836 0.2575 0.857 0.717
#> 4 4 0.511 0.582 0.750 0.1161 0.941 0.845
#> 5 5 0.563 0.451 0.703 0.0684 0.942 0.830
#> 6 6 0.581 0.413 0.622 0.0421 0.901 0.676
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
#> GSM25548 2 0.0000 0.979 0.000 1.000
#> GSM25549 2 0.0000 0.979 0.000 1.000
#> GSM25550 2 0.0000 0.979 0.000 1.000
#> GSM25551 2 0.0000 0.979 0.000 1.000
#> GSM25570 2 0.0000 0.979 0.000 1.000
#> GSM25571 2 0.0000 0.979 0.000 1.000
#> GSM25358 1 0.1633 0.963 0.976 0.024
#> GSM25359 2 0.0000 0.979 0.000 1.000
#> GSM25360 1 0.0000 0.984 1.000 0.000
#> GSM25361 2 0.9087 0.540 0.324 0.676
#> GSM25377 1 0.0000 0.984 1.000 0.000
#> GSM25378 1 0.0000 0.984 1.000 0.000
#> GSM25401 1 0.8555 0.621 0.720 0.280
#> GSM25402 1 0.0000 0.984 1.000 0.000
#> GSM25349 2 0.0000 0.979 0.000 1.000
#> GSM25350 2 0.0000 0.979 0.000 1.000
#> GSM25356 1 0.0000 0.984 1.000 0.000
#> GSM25357 2 0.0000 0.979 0.000 1.000
#> GSM25385 1 0.0000 0.984 1.000 0.000
#> GSM25386 1 0.0000 0.984 1.000 0.000
#> GSM25399 1 0.0000 0.984 1.000 0.000
#> GSM25400 1 0.0000 0.984 1.000 0.000
#> GSM48659 2 0.0000 0.979 0.000 1.000
#> GSM48660 2 0.0000 0.979 0.000 1.000
#> GSM25409 2 0.0000 0.979 0.000 1.000
#> GSM25410 1 0.0000 0.984 1.000 0.000
#> GSM25426 2 0.0000 0.979 0.000 1.000
#> GSM25427 1 0.0376 0.980 0.996 0.004
#> GSM25540 2 0.0000 0.979 0.000 1.000
#> GSM25541 2 0.2948 0.935 0.052 0.948
#> GSM25542 2 0.0000 0.979 0.000 1.000
#> GSM25543 2 0.0000 0.979 0.000 1.000
#> GSM25479 1 0.0000 0.984 1.000 0.000
#> GSM25480 1 0.0000 0.984 1.000 0.000
#> GSM25481 1 0.1414 0.967 0.980 0.020
#> GSM25482 1 0.1414 0.967 0.980 0.020
#> GSM48654 2 0.0000 0.979 0.000 1.000
#> GSM48650 2 0.0000 0.979 0.000 1.000
#> GSM48651 2 0.0000 0.979 0.000 1.000
#> GSM48652 2 0.0000 0.979 0.000 1.000
#> GSM48653 2 0.0000 0.979 0.000 1.000
#> GSM48662 2 0.0000 0.979 0.000 1.000
#> GSM48663 2 0.0000 0.979 0.000 1.000
#> GSM25524 1 0.0000 0.984 1.000 0.000
#> GSM25525 1 0.0000 0.984 1.000 0.000
#> GSM25526 1 0.0000 0.984 1.000 0.000
#> GSM25527 1 0.0000 0.984 1.000 0.000
#> GSM25528 1 0.0000 0.984 1.000 0.000
#> GSM25529 1 0.0000 0.984 1.000 0.000
#> GSM25530 1 0.0000 0.984 1.000 0.000
#> GSM25531 1 0.0000 0.984 1.000 0.000
#> GSM48661 2 0.0000 0.979 0.000 1.000
#> GSM25561 1 0.0000 0.984 1.000 0.000
#> GSM25562 1 0.0000 0.984 1.000 0.000
#> GSM25563 1 0.0000 0.984 1.000 0.000
#> GSM25564 1 0.6801 0.778 0.820 0.180
#> GSM25565 2 0.0000 0.979 0.000 1.000
#> GSM25566 2 0.0000 0.979 0.000 1.000
#> GSM25568 1 0.8016 0.682 0.756 0.244
#> GSM25569 2 0.0000 0.979 0.000 1.000
#> GSM25552 2 0.0000 0.979 0.000 1.000
#> GSM25553 2 0.8207 0.668 0.256 0.744
#> GSM25578 1 0.0000 0.984 1.000 0.000
#> GSM25579 1 0.0000 0.984 1.000 0.000
#> GSM25580 1 0.0000 0.984 1.000 0.000
#> GSM25581 1 0.0000 0.984 1.000 0.000
#> GSM48655 2 0.0000 0.979 0.000 1.000
#> GSM48656 2 0.0000 0.979 0.000 1.000
#> GSM48657 2 0.0000 0.979 0.000 1.000
#> GSM48658 2 0.0000 0.979 0.000 1.000
#> GSM25624 1 0.0000 0.984 1.000 0.000
#> GSM25625 1 0.0000 0.984 1.000 0.000
#> GSM25626 1 0.0000 0.984 1.000 0.000
#> GSM25627 2 0.5294 0.863 0.120 0.880
#> GSM25628 2 0.5629 0.850 0.132 0.868
#> GSM25629 2 0.0000 0.979 0.000 1.000
#> GSM25630 1 0.0000 0.984 1.000 0.000
#> GSM25631 2 0.0000 0.979 0.000 1.000
#> GSM25632 1 0.0000 0.984 1.000 0.000
#> GSM25633 1 0.0000 0.984 1.000 0.000
#> GSM25634 1 0.0000 0.984 1.000 0.000
#> GSM25635 1 0.0000 0.984 1.000 0.000
#> GSM25656 2 0.2603 0.941 0.044 0.956
#> GSM25657 1 0.0000 0.984 1.000 0.000
#> GSM25658 1 0.0000 0.984 1.000 0.000
#> GSM25659 1 0.0000 0.984 1.000 0.000
#> GSM25660 1 0.0000 0.984 1.000 0.000
#> GSM25661 1 0.0000 0.984 1.000 0.000
#> GSM25662 2 0.0000 0.979 0.000 1.000
#> GSM25663 2 0.0000 0.979 0.000 1.000
#> GSM25680 2 0.0000 0.979 0.000 1.000
#> GSM25681 2 0.5294 0.863 0.120 0.880
#> GSM25682 2 0.0000 0.979 0.000 1.000
#> GSM25683 2 0.0000 0.979 0.000 1.000
#> GSM25684 2 0.0000 0.979 0.000 1.000
#> GSM25685 2 0.0000 0.979 0.000 1.000
#> GSM25686 2 0.0000 0.979 0.000 1.000
#> GSM25687 2 0.0000 0.979 0.000 1.000
#> GSM48664 1 0.0000 0.984 1.000 0.000
#> GSM48665 1 0.0000 0.984 1.000 0.000
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM25548 2 0.1964 0.9109 0.000 0.944 0.056
#> GSM25549 2 0.2165 0.9090 0.000 0.936 0.064
#> GSM25550 2 0.2165 0.9089 0.000 0.936 0.064
#> GSM25551 2 0.0747 0.9179 0.000 0.984 0.016
#> GSM25570 2 0.1964 0.9109 0.000 0.944 0.056
#> GSM25571 2 0.1964 0.9109 0.000 0.944 0.056
#> GSM25358 1 0.9111 0.0504 0.472 0.144 0.384
#> GSM25359 2 0.4399 0.7893 0.000 0.812 0.188
#> GSM25360 3 0.5254 0.7052 0.264 0.000 0.736
#> GSM25361 3 0.7016 0.6643 0.156 0.116 0.728
#> GSM25377 1 0.4062 0.7519 0.836 0.000 0.164
#> GSM25378 1 0.4504 0.7276 0.804 0.000 0.196
#> GSM25401 3 0.8848 0.1094 0.372 0.124 0.504
#> GSM25402 1 0.6280 0.2928 0.540 0.000 0.460
#> GSM25349 2 0.2796 0.8987 0.000 0.908 0.092
#> GSM25350 2 0.2711 0.9014 0.000 0.912 0.088
#> GSM25356 1 0.4062 0.7471 0.836 0.000 0.164
#> GSM25357 2 0.1315 0.9153 0.008 0.972 0.020
#> GSM25385 3 0.5678 0.6549 0.316 0.000 0.684
#> GSM25386 3 0.4834 0.7121 0.204 0.004 0.792
#> GSM25399 1 0.3551 0.7643 0.868 0.000 0.132
#> GSM25400 1 0.3816 0.7579 0.852 0.000 0.148
#> GSM48659 2 0.1753 0.9201 0.000 0.952 0.048
#> GSM48660 2 0.1643 0.9129 0.000 0.956 0.044
#> GSM25409 2 0.2711 0.9084 0.000 0.912 0.088
#> GSM25410 3 0.5016 0.6903 0.240 0.000 0.760
#> GSM25426 2 0.1529 0.9137 0.000 0.960 0.040
#> GSM25427 1 0.4861 0.7220 0.800 0.008 0.192
#> GSM25540 2 0.5968 0.5168 0.000 0.636 0.364
#> GSM25541 2 0.8028 0.3172 0.072 0.560 0.368
#> GSM25542 2 0.3551 0.8661 0.000 0.868 0.132
#> GSM25543 3 0.6140 0.0802 0.000 0.404 0.596
#> GSM25479 1 0.1643 0.7927 0.956 0.000 0.044
#> GSM25480 1 0.2625 0.7769 0.916 0.000 0.084
#> GSM25481 1 0.5109 0.7087 0.780 0.008 0.212
#> GSM25482 1 0.5109 0.7087 0.780 0.008 0.212
#> GSM48654 2 0.1753 0.9189 0.000 0.952 0.048
#> GSM48650 2 0.1411 0.9153 0.000 0.964 0.036
#> GSM48651 2 0.1411 0.9177 0.000 0.964 0.036
#> GSM48652 2 0.1411 0.9177 0.000 0.964 0.036
#> GSM48653 2 0.1529 0.9177 0.000 0.960 0.040
#> GSM48662 2 0.1289 0.9181 0.000 0.968 0.032
#> GSM48663 2 0.2356 0.8986 0.000 0.928 0.072
#> GSM25524 3 0.5926 0.6326 0.356 0.000 0.644
#> GSM25525 1 0.2711 0.7730 0.912 0.000 0.088
#> GSM25526 1 0.6235 -0.0848 0.564 0.000 0.436
#> GSM25527 1 0.1163 0.7965 0.972 0.000 0.028
#> GSM25528 1 0.5138 0.5581 0.748 0.000 0.252
#> GSM25529 1 0.2878 0.7680 0.904 0.000 0.096
#> GSM25530 1 0.4452 0.6650 0.808 0.000 0.192
#> GSM25531 1 0.3038 0.7656 0.896 0.000 0.104
#> GSM48661 2 0.1753 0.9193 0.000 0.952 0.048
#> GSM25561 3 0.6062 0.5779 0.384 0.000 0.616
#> GSM25562 1 0.3551 0.7516 0.868 0.000 0.132
#> GSM25563 3 0.5363 0.6993 0.276 0.000 0.724
#> GSM25564 1 0.8587 0.2207 0.604 0.220 0.176
#> GSM25565 2 0.1860 0.9193 0.000 0.948 0.052
#> GSM25566 2 0.0892 0.9194 0.000 0.980 0.020
#> GSM25568 3 0.7493 0.6289 0.136 0.168 0.696
#> GSM25569 2 0.2356 0.9152 0.000 0.928 0.072
#> GSM25552 2 0.2356 0.9086 0.000 0.928 0.072
#> GSM25553 2 0.7772 0.5794 0.196 0.672 0.132
#> GSM25578 1 0.2356 0.7810 0.928 0.000 0.072
#> GSM25579 1 0.3116 0.7685 0.892 0.000 0.108
#> GSM25580 1 0.0747 0.7971 0.984 0.000 0.016
#> GSM25581 1 0.1031 0.7973 0.976 0.000 0.024
#> GSM48655 2 0.1289 0.9157 0.000 0.968 0.032
#> GSM48656 2 0.1289 0.9187 0.000 0.968 0.032
#> GSM48657 2 0.1289 0.9157 0.000 0.968 0.032
#> GSM48658 2 0.1753 0.9193 0.000 0.952 0.048
#> GSM25624 1 0.2066 0.7909 0.940 0.000 0.060
#> GSM25625 3 0.6252 0.4475 0.444 0.000 0.556
#> GSM25626 3 0.5365 0.7104 0.252 0.004 0.744
#> GSM25627 2 0.7389 0.1618 0.036 0.556 0.408
#> GSM25628 3 0.7186 0.6150 0.080 0.224 0.696
#> GSM25629 2 0.5497 0.6023 0.000 0.708 0.292
#> GSM25630 3 0.5621 0.6773 0.308 0.000 0.692
#> GSM25631 2 0.2448 0.9097 0.000 0.924 0.076
#> GSM25632 3 0.5497 0.6936 0.292 0.000 0.708
#> GSM25633 1 0.1411 0.7977 0.964 0.000 0.036
#> GSM25634 1 0.1411 0.7980 0.964 0.000 0.036
#> GSM25635 1 0.1289 0.7942 0.968 0.000 0.032
#> GSM25656 3 0.7164 0.5759 0.064 0.256 0.680
#> GSM25657 1 0.1753 0.7968 0.952 0.000 0.048
#> GSM25658 1 0.5988 0.1979 0.632 0.000 0.368
#> GSM25659 1 0.4974 0.6027 0.764 0.000 0.236
#> GSM25660 1 0.1289 0.7969 0.968 0.000 0.032
#> GSM25661 1 0.1031 0.7966 0.976 0.000 0.024
#> GSM25662 2 0.0592 0.9180 0.000 0.988 0.012
#> GSM25663 2 0.1289 0.9174 0.000 0.968 0.032
#> GSM25680 2 0.2448 0.9058 0.000 0.924 0.076
#> GSM25681 2 0.5816 0.7708 0.056 0.788 0.156
#> GSM25682 2 0.0424 0.9180 0.000 0.992 0.008
#> GSM25683 2 0.0424 0.9180 0.000 0.992 0.008
#> GSM25684 2 0.0747 0.9180 0.000 0.984 0.016
#> GSM25685 2 0.1163 0.9163 0.000 0.972 0.028
#> GSM25686 2 0.0592 0.9179 0.000 0.988 0.012
#> GSM25687 2 0.0592 0.9179 0.000 0.988 0.012
#> GSM48664 1 0.3551 0.7643 0.868 0.000 0.132
#> GSM48665 1 0.3412 0.7664 0.876 0.000 0.124
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM25548 2 0.4267 0.7920 0.000 0.788 0.024 0.188
#> GSM25549 2 0.4459 0.7908 0.000 0.780 0.032 0.188
#> GSM25550 2 0.4465 0.7882 0.004 0.776 0.020 0.200
#> GSM25551 2 0.3453 0.8023 0.000 0.868 0.052 0.080
#> GSM25570 2 0.4267 0.7920 0.000 0.788 0.024 0.188
#> GSM25571 2 0.4267 0.7920 0.000 0.788 0.024 0.188
#> GSM25358 4 0.9391 0.2608 0.216 0.120 0.260 0.404
#> GSM25359 2 0.7135 0.5797 0.000 0.560 0.240 0.200
#> GSM25360 3 0.3757 0.7479 0.152 0.000 0.828 0.020
#> GSM25361 3 0.7150 0.6109 0.116 0.052 0.652 0.180
#> GSM25377 4 0.5668 0.4193 0.444 0.000 0.024 0.532
#> GSM25378 4 0.5938 0.4561 0.480 0.000 0.036 0.484
#> GSM25401 4 0.9173 0.1460 0.140 0.136 0.300 0.424
#> GSM25402 4 0.8155 0.3538 0.272 0.012 0.292 0.424
#> GSM25349 2 0.4711 0.7698 0.000 0.740 0.024 0.236
#> GSM25350 2 0.4678 0.7727 0.000 0.744 0.024 0.232
#> GSM25356 4 0.5771 0.4655 0.460 0.000 0.028 0.512
#> GSM25357 2 0.3392 0.7974 0.000 0.856 0.020 0.124
#> GSM25385 3 0.4215 0.7018 0.104 0.000 0.824 0.072
#> GSM25386 3 0.2256 0.7578 0.056 0.000 0.924 0.020
#> GSM25399 1 0.5511 -0.3588 0.500 0.000 0.016 0.484
#> GSM25400 1 0.6000 -0.4541 0.508 0.000 0.040 0.452
#> GSM48659 2 0.2483 0.8346 0.000 0.916 0.032 0.052
#> GSM48660 2 0.2737 0.8122 0.000 0.888 0.008 0.104
#> GSM25409 2 0.4353 0.8024 0.000 0.756 0.012 0.232
#> GSM25410 3 0.2928 0.7464 0.052 0.000 0.896 0.052
#> GSM25426 2 0.4419 0.7728 0.000 0.812 0.104 0.084
#> GSM25427 4 0.5392 0.4716 0.460 0.000 0.012 0.528
#> GSM25540 2 0.7558 0.3424 0.000 0.444 0.360 0.196
#> GSM25541 2 0.7799 0.3321 0.008 0.444 0.360 0.188
#> GSM25542 2 0.5979 0.7365 0.000 0.692 0.136 0.172
#> GSM25543 3 0.7173 0.4156 0.000 0.216 0.556 0.228
#> GSM25479 1 0.1706 0.5912 0.948 0.000 0.016 0.036
#> GSM25480 1 0.0804 0.5916 0.980 0.000 0.012 0.008
#> GSM25481 4 0.6046 0.5143 0.420 0.012 0.024 0.544
#> GSM25482 4 0.6046 0.5143 0.420 0.012 0.024 0.544
#> GSM48654 2 0.2699 0.8286 0.000 0.904 0.028 0.068
#> GSM48650 2 0.2799 0.8159 0.000 0.884 0.008 0.108
#> GSM48651 2 0.2271 0.8247 0.000 0.916 0.008 0.076
#> GSM48652 2 0.2271 0.8247 0.000 0.916 0.008 0.076
#> GSM48653 2 0.2965 0.8276 0.000 0.892 0.036 0.072
#> GSM48662 2 0.2610 0.8301 0.000 0.900 0.012 0.088
#> GSM48663 2 0.4019 0.7609 0.000 0.792 0.012 0.196
#> GSM25524 1 0.5511 -0.2576 0.500 0.000 0.484 0.016
#> GSM25525 1 0.1398 0.5877 0.956 0.000 0.040 0.004
#> GSM25526 1 0.7202 -0.0244 0.464 0.000 0.396 0.140
#> GSM25527 1 0.3143 0.5588 0.876 0.000 0.024 0.100
#> GSM25528 1 0.4095 0.4662 0.792 0.000 0.192 0.016
#> GSM25529 1 0.1722 0.5846 0.944 0.000 0.048 0.008
#> GSM25530 1 0.3547 0.5197 0.840 0.000 0.144 0.016
#> GSM25531 1 0.2198 0.5750 0.920 0.000 0.072 0.008
#> GSM48661 2 0.3266 0.8348 0.000 0.876 0.040 0.084
#> GSM25561 3 0.5442 0.5346 0.336 0.000 0.636 0.028
#> GSM25562 1 0.5247 0.4691 0.752 0.000 0.100 0.148
#> GSM25563 3 0.3708 0.7434 0.148 0.000 0.832 0.020
#> GSM25564 1 0.9112 -0.0160 0.472 0.212 0.136 0.180
#> GSM25565 2 0.3308 0.8344 0.000 0.872 0.036 0.092
#> GSM25566 2 0.2722 0.8339 0.000 0.904 0.032 0.064
#> GSM25568 3 0.8252 0.4934 0.064 0.140 0.524 0.272
#> GSM25569 2 0.4281 0.8118 0.000 0.792 0.028 0.180
#> GSM25552 2 0.4527 0.7891 0.008 0.780 0.020 0.192
#> GSM25553 2 0.7955 0.5656 0.160 0.552 0.044 0.244
#> GSM25578 1 0.1059 0.5930 0.972 0.000 0.016 0.012
#> GSM25579 1 0.3674 0.5176 0.848 0.000 0.036 0.116
#> GSM25580 1 0.3895 0.4866 0.804 0.000 0.012 0.184
#> GSM25581 1 0.2999 0.5507 0.864 0.000 0.004 0.132
#> GSM48655 2 0.1902 0.8246 0.000 0.932 0.004 0.064
#> GSM48656 2 0.2675 0.8310 0.000 0.892 0.008 0.100
#> GSM48657 2 0.2266 0.8203 0.000 0.912 0.004 0.084
#> GSM48658 2 0.4088 0.8255 0.000 0.820 0.040 0.140
#> GSM25624 1 0.4599 0.3586 0.736 0.000 0.016 0.248
#> GSM25625 3 0.5732 0.5013 0.264 0.000 0.672 0.064
#> GSM25626 3 0.2363 0.7538 0.056 0.000 0.920 0.024
#> GSM25627 2 0.8016 0.0027 0.040 0.436 0.404 0.120
#> GSM25628 3 0.2495 0.7289 0.012 0.036 0.924 0.028
#> GSM25629 2 0.6585 0.4846 0.000 0.584 0.312 0.104
#> GSM25630 3 0.4706 0.6903 0.224 0.000 0.748 0.028
#> GSM25631 2 0.4881 0.7851 0.000 0.756 0.048 0.196
#> GSM25632 3 0.4010 0.7410 0.156 0.000 0.816 0.028
#> GSM25633 1 0.3048 0.5581 0.876 0.000 0.016 0.108
#> GSM25634 1 0.3852 0.5002 0.808 0.000 0.012 0.180
#> GSM25635 1 0.4472 0.4172 0.760 0.000 0.020 0.220
#> GSM25656 3 0.3485 0.6950 0.004 0.076 0.872 0.048
#> GSM25657 1 0.2773 0.5806 0.900 0.000 0.028 0.072
#> GSM25658 1 0.7072 0.0415 0.524 0.000 0.336 0.140
#> GSM25659 1 0.4106 0.5244 0.832 0.000 0.084 0.084
#> GSM25660 1 0.2714 0.5619 0.884 0.000 0.004 0.112
#> GSM25661 1 0.2714 0.5631 0.884 0.000 0.004 0.112
#> GSM25662 2 0.1520 0.8298 0.000 0.956 0.020 0.024
#> GSM25663 2 0.3245 0.8245 0.000 0.872 0.028 0.100
#> GSM25680 2 0.5035 0.7817 0.000 0.748 0.056 0.196
#> GSM25681 2 0.7638 0.6628 0.076 0.616 0.112 0.196
#> GSM25682 2 0.1004 0.8299 0.000 0.972 0.004 0.024
#> GSM25683 2 0.1004 0.8299 0.000 0.972 0.004 0.024
#> GSM25684 2 0.1724 0.8296 0.000 0.948 0.032 0.020
#> GSM25685 2 0.3691 0.7948 0.000 0.856 0.076 0.068
#> GSM25686 2 0.1004 0.8299 0.000 0.972 0.004 0.024
#> GSM25687 2 0.1004 0.8299 0.000 0.972 0.004 0.024
#> GSM48664 1 0.5604 -0.3423 0.504 0.000 0.020 0.476
#> GSM48665 1 0.5452 -0.2912 0.556 0.000 0.016 0.428
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM25548 2 0.4286 0.30946 0.000 0.652 0.004 0.004 0.340
#> GSM25549 2 0.4389 0.27525 0.000 0.624 0.004 0.004 0.368
#> GSM25550 2 0.4402 0.26754 0.000 0.620 0.004 0.004 0.372
#> GSM25551 2 0.3367 0.51768 0.000 0.840 0.016 0.016 0.128
#> GSM25570 2 0.4302 0.30614 0.000 0.648 0.004 0.004 0.344
#> GSM25571 2 0.4286 0.30946 0.000 0.652 0.004 0.004 0.340
#> GSM25358 4 0.8806 0.18254 0.064 0.076 0.252 0.392 0.216
#> GSM25359 2 0.6797 -0.70974 0.000 0.404 0.200 0.008 0.388
#> GSM25360 3 0.4017 0.67975 0.116 0.000 0.812 0.016 0.056
#> GSM25361 3 0.7528 0.01774 0.136 0.060 0.412 0.008 0.384
#> GSM25377 4 0.4845 0.54878 0.124 0.000 0.016 0.752 0.108
#> GSM25378 4 0.5314 0.58226 0.256 0.000 0.036 0.672 0.036
#> GSM25401 4 0.8732 0.16770 0.072 0.096 0.276 0.420 0.136
#> GSM25402 4 0.7542 0.32568 0.116 0.004 0.284 0.492 0.104
#> GSM25349 2 0.5973 0.38450 0.000 0.588 0.020 0.084 0.308
#> GSM25350 2 0.5902 0.38748 0.000 0.588 0.016 0.084 0.312
#> GSM25356 4 0.4626 0.60663 0.224 0.000 0.020 0.728 0.028
#> GSM25357 2 0.5293 0.39142 0.000 0.720 0.024 0.136 0.120
#> GSM25385 3 0.3452 0.66927 0.056 0.000 0.856 0.068 0.020
#> GSM25386 3 0.1300 0.69093 0.028 0.000 0.956 0.016 0.000
#> GSM25399 4 0.5341 0.49806 0.204 0.000 0.012 0.688 0.096
#> GSM25400 4 0.4992 0.51441 0.320 0.000 0.028 0.640 0.012
#> GSM48659 2 0.2806 0.61833 0.000 0.844 0.004 0.000 0.152
#> GSM48660 2 0.2921 0.59749 0.000 0.844 0.004 0.004 0.148
#> GSM25409 2 0.4419 0.35030 0.000 0.644 0.004 0.008 0.344
#> GSM25410 3 0.2178 0.68722 0.024 0.000 0.920 0.048 0.008
#> GSM25426 2 0.4560 0.41094 0.000 0.764 0.060 0.016 0.160
#> GSM25427 4 0.4874 0.57635 0.244 0.000 0.012 0.700 0.044
#> GSM25540 5 0.7193 0.96479 0.012 0.312 0.268 0.004 0.404
#> GSM25541 5 0.7241 0.96523 0.016 0.304 0.260 0.004 0.416
#> GSM25542 2 0.6616 0.19360 0.000 0.564 0.124 0.040 0.272
#> GSM25543 3 0.7121 0.11971 0.000 0.120 0.468 0.060 0.352
#> GSM25479 1 0.2866 0.62733 0.872 0.000 0.004 0.100 0.024
#> GSM25480 1 0.2494 0.63180 0.904 0.000 0.008 0.056 0.032
#> GSM25481 4 0.5654 0.59327 0.212 0.008 0.020 0.680 0.080
#> GSM25482 4 0.5654 0.59327 0.212 0.008 0.020 0.680 0.080
#> GSM48654 2 0.2763 0.61248 0.000 0.848 0.004 0.000 0.148
#> GSM48650 2 0.3044 0.58798 0.000 0.840 0.004 0.008 0.148
#> GSM48651 2 0.2471 0.61276 0.000 0.864 0.000 0.000 0.136
#> GSM48652 2 0.2583 0.61409 0.000 0.864 0.004 0.000 0.132
#> GSM48653 2 0.2806 0.61137 0.000 0.844 0.004 0.000 0.152
#> GSM48662 2 0.3048 0.61936 0.000 0.820 0.004 0.000 0.176
#> GSM48663 2 0.5620 0.41323 0.000 0.664 0.016 0.104 0.216
#> GSM25524 1 0.5223 0.26728 0.628 0.000 0.316 0.008 0.048
#> GSM25525 1 0.0955 0.62639 0.968 0.000 0.004 0.000 0.028
#> GSM25526 1 0.7777 0.00745 0.400 0.000 0.352 0.128 0.120
#> GSM25527 1 0.3694 0.58479 0.796 0.000 0.000 0.172 0.032
#> GSM25528 1 0.3925 0.55268 0.816 0.000 0.124 0.020 0.040
#> GSM25529 1 0.1082 0.62554 0.964 0.000 0.008 0.000 0.028
#> GSM25530 1 0.3623 0.57858 0.848 0.000 0.072 0.052 0.028
#> GSM25531 1 0.2196 0.61704 0.916 0.000 0.004 0.056 0.024
#> GSM48661 2 0.3171 0.61389 0.000 0.816 0.008 0.000 0.176
#> GSM25561 3 0.6392 0.52693 0.264 0.000 0.588 0.036 0.112
#> GSM25562 1 0.6483 0.37858 0.596 0.000 0.040 0.232 0.132
#> GSM25563 3 0.4640 0.67335 0.100 0.000 0.776 0.024 0.100
#> GSM25564 1 0.9226 0.03207 0.356 0.152 0.072 0.156 0.264
#> GSM25565 2 0.3757 0.56708 0.000 0.772 0.020 0.000 0.208
#> GSM25566 2 0.2806 0.57262 0.000 0.844 0.004 0.000 0.152
#> GSM25568 3 0.8763 0.24096 0.044 0.100 0.360 0.184 0.312
#> GSM25569 2 0.5174 0.48344 0.000 0.624 0.012 0.036 0.328
#> GSM25552 2 0.4900 0.25386 0.020 0.616 0.004 0.004 0.356
#> GSM25553 2 0.6984 -0.28344 0.096 0.432 0.012 0.036 0.424
#> GSM25578 1 0.1591 0.63250 0.940 0.000 0.004 0.052 0.004
#> GSM25579 1 0.3371 0.58395 0.848 0.000 0.008 0.040 0.104
#> GSM25580 1 0.5151 0.44683 0.644 0.000 0.008 0.300 0.048
#> GSM25581 1 0.4284 0.54404 0.736 0.000 0.000 0.224 0.040
#> GSM48655 2 0.1965 0.62185 0.000 0.904 0.000 0.000 0.096
#> GSM48656 2 0.2891 0.61923 0.000 0.824 0.000 0.000 0.176
#> GSM48657 2 0.2127 0.61263 0.000 0.892 0.000 0.000 0.108
#> GSM48658 2 0.3607 0.58372 0.000 0.752 0.004 0.000 0.244
#> GSM25624 1 0.5441 0.19710 0.536 0.000 0.008 0.412 0.044
#> GSM25625 3 0.5469 0.52178 0.184 0.000 0.704 0.064 0.048
#> GSM25626 3 0.1569 0.68934 0.032 0.000 0.948 0.012 0.008
#> GSM25627 3 0.8729 -0.16924 0.048 0.332 0.360 0.104 0.156
#> GSM25628 3 0.3020 0.64685 0.016 0.024 0.880 0.004 0.076
#> GSM25629 2 0.7126 -0.37418 0.008 0.484 0.292 0.020 0.196
#> GSM25630 3 0.5858 0.61990 0.180 0.000 0.668 0.032 0.120
#> GSM25631 2 0.5022 0.17512 0.004 0.584 0.016 0.008 0.388
#> GSM25632 3 0.2756 0.69725 0.092 0.000 0.880 0.004 0.024
#> GSM25633 1 0.3977 0.56568 0.764 0.000 0.000 0.204 0.032
#> GSM25634 1 0.5257 0.44067 0.640 0.000 0.008 0.296 0.056
#> GSM25635 1 0.5016 0.35549 0.608 0.000 0.000 0.348 0.044
#> GSM25656 3 0.4444 0.61436 0.008 0.028 0.780 0.024 0.160
#> GSM25657 1 0.4579 0.55479 0.744 0.000 0.008 0.192 0.056
#> GSM25658 1 0.7718 0.05898 0.440 0.000 0.312 0.128 0.120
#> GSM25659 1 0.4185 0.56827 0.812 0.004 0.020 0.060 0.104
#> GSM25660 1 0.4193 0.55511 0.748 0.000 0.000 0.212 0.040
#> GSM25661 1 0.4134 0.54913 0.744 0.000 0.000 0.224 0.032
#> GSM25662 2 0.1026 0.61544 0.000 0.968 0.004 0.004 0.024
#> GSM25663 2 0.3171 0.52986 0.000 0.816 0.008 0.000 0.176
#> GSM25680 2 0.4995 0.17645 0.000 0.584 0.028 0.004 0.384
#> GSM25681 2 0.6598 -0.16331 0.080 0.484 0.036 0.004 0.396
#> GSM25682 2 0.1116 0.61822 0.000 0.964 0.004 0.004 0.028
#> GSM25683 2 0.0932 0.61781 0.000 0.972 0.004 0.004 0.020
#> GSM25684 2 0.1788 0.60712 0.000 0.932 0.008 0.004 0.056
#> GSM25685 2 0.4112 0.45155 0.000 0.800 0.048 0.016 0.136
#> GSM25686 2 0.1116 0.61822 0.000 0.964 0.004 0.004 0.028
#> GSM25687 2 0.1202 0.61910 0.000 0.960 0.004 0.004 0.032
#> GSM48664 4 0.5419 0.48908 0.208 0.000 0.012 0.680 0.100
#> GSM48665 4 0.5516 0.20658 0.388 0.000 0.008 0.552 0.052
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM25548 5 0.4097 -0.1045 0.000 0.008 0.000 0.000 0.500 0.492
#> GSM25549 6 0.3993 0.0406 0.000 0.004 0.000 0.000 0.476 0.520
#> GSM25550 6 0.4227 0.0334 0.000 0.004 0.000 0.008 0.488 0.500
#> GSM25551 5 0.5340 0.3949 0.000 0.268 0.008 0.000 0.600 0.124
#> GSM25570 5 0.4096 -0.0899 0.000 0.008 0.000 0.000 0.508 0.484
#> GSM25571 5 0.4095 -0.0815 0.000 0.008 0.000 0.000 0.512 0.480
#> GSM25358 4 0.7999 -0.1351 0.044 0.148 0.168 0.444 0.008 0.188
#> GSM25359 6 0.7286 0.3430 0.000 0.148 0.156 0.004 0.256 0.436
#> GSM25360 3 0.3166 0.6971 0.060 0.036 0.856 0.000 0.000 0.048
#> GSM25361 6 0.7317 -0.0831 0.076 0.088 0.316 0.012 0.040 0.468
#> GSM25377 4 0.6005 0.5149 0.132 0.228 0.012 0.596 0.000 0.032
#> GSM25378 4 0.4624 0.5900 0.188 0.060 0.024 0.724 0.000 0.004
#> GSM25401 2 0.7627 0.2858 0.020 0.380 0.176 0.348 0.044 0.032
#> GSM25402 4 0.7089 -0.1078 0.048 0.244 0.208 0.476 0.004 0.020
#> GSM25349 5 0.6257 0.2247 0.000 0.092 0.012 0.052 0.540 0.304
#> GSM25350 5 0.6254 0.2183 0.000 0.088 0.012 0.052 0.532 0.316
#> GSM25356 4 0.4484 0.6106 0.148 0.072 0.008 0.752 0.000 0.020
#> GSM25357 5 0.6714 0.2587 0.000 0.172 0.000 0.180 0.528 0.120
#> GSM25385 3 0.3003 0.6569 0.012 0.052 0.864 0.068 0.000 0.004
#> GSM25386 3 0.0924 0.7140 0.004 0.008 0.972 0.008 0.000 0.008
#> GSM25399 4 0.6021 0.4974 0.152 0.240 0.000 0.568 0.000 0.040
#> GSM25400 4 0.4761 0.5818 0.212 0.088 0.012 0.688 0.000 0.000
#> GSM48659 5 0.2821 0.6068 0.000 0.020 0.004 0.004 0.856 0.116
#> GSM48660 5 0.1462 0.6016 0.000 0.008 0.000 0.000 0.936 0.056
#> GSM25409 5 0.4549 0.1039 0.000 0.028 0.000 0.004 0.552 0.416
#> GSM25410 3 0.1693 0.7050 0.004 0.020 0.932 0.044 0.000 0.000
#> GSM25426 5 0.5984 0.3373 0.000 0.280 0.032 0.004 0.560 0.124
#> GSM25427 4 0.3921 0.5844 0.184 0.028 0.004 0.768 0.000 0.016
#> GSM25540 6 0.7133 0.3876 0.000 0.100 0.204 0.004 0.240 0.452
#> GSM25541 6 0.7079 0.3911 0.000 0.088 0.216 0.004 0.240 0.452
#> GSM25542 5 0.6957 0.0332 0.000 0.096 0.112 0.012 0.456 0.324
#> GSM25543 6 0.7674 -0.2311 0.000 0.160 0.316 0.028 0.116 0.380
#> GSM25479 1 0.3040 0.6329 0.868 0.028 0.016 0.072 0.000 0.016
#> GSM25480 1 0.2713 0.6372 0.888 0.012 0.012 0.048 0.000 0.040
#> GSM25481 4 0.5880 0.5659 0.140 0.076 0.012 0.664 0.004 0.104
#> GSM25482 4 0.5880 0.5659 0.140 0.076 0.012 0.664 0.004 0.104
#> GSM48654 5 0.1913 0.6120 0.000 0.012 0.004 0.004 0.920 0.060
#> GSM48650 5 0.2691 0.5954 0.000 0.088 0.000 0.008 0.872 0.032
#> GSM48651 5 0.1225 0.6159 0.000 0.012 0.000 0.000 0.952 0.036
#> GSM48652 5 0.1225 0.6161 0.000 0.012 0.000 0.000 0.952 0.036
#> GSM48653 5 0.1862 0.6143 0.000 0.016 0.008 0.004 0.928 0.044
#> GSM48662 5 0.1444 0.6117 0.000 0.000 0.000 0.000 0.928 0.072
#> GSM48663 5 0.5535 0.2869 0.000 0.096 0.000 0.048 0.636 0.220
#> GSM25524 1 0.5472 0.3478 0.616 0.064 0.276 0.004 0.000 0.040
#> GSM25525 1 0.1965 0.6325 0.924 0.040 0.008 0.004 0.000 0.024
#> GSM25526 2 0.8007 0.3940 0.232 0.296 0.284 0.168 0.000 0.020
#> GSM25527 1 0.4179 0.5548 0.748 0.044 0.008 0.192 0.000 0.008
#> GSM25528 1 0.4434 0.5601 0.772 0.060 0.116 0.008 0.000 0.044
#> GSM25529 1 0.2239 0.6306 0.912 0.040 0.016 0.004 0.000 0.028
#> GSM25530 1 0.3981 0.5942 0.820 0.064 0.048 0.028 0.000 0.040
#> GSM25531 1 0.3131 0.6177 0.868 0.052 0.024 0.016 0.000 0.040
#> GSM48661 5 0.2698 0.6061 0.000 0.020 0.008 0.004 0.872 0.096
#> GSM25561 3 0.6968 0.5131 0.188 0.152 0.536 0.020 0.000 0.104
#> GSM25562 1 0.8014 0.1218 0.396 0.200 0.052 0.232 0.000 0.120
#> GSM25563 3 0.5668 0.6274 0.040 0.156 0.672 0.024 0.000 0.108
#> GSM25564 1 0.9460 -0.0302 0.264 0.176 0.044 0.132 0.204 0.180
#> GSM25565 5 0.3958 0.5389 0.000 0.024 0.008 0.012 0.760 0.196
#> GSM25566 5 0.3834 0.5165 0.000 0.024 0.004 0.000 0.728 0.244
#> GSM25568 6 0.8564 -0.3165 0.012 0.176 0.256 0.112 0.088 0.356
#> GSM25569 5 0.5137 0.3258 0.000 0.068 0.004 0.012 0.612 0.304
#> GSM25552 5 0.4357 -0.1432 0.004 0.004 0.000 0.008 0.492 0.492
#> GSM25553 6 0.6602 0.3179 0.076 0.040 0.020 0.028 0.276 0.560
#> GSM25578 1 0.1332 0.6405 0.952 0.012 0.000 0.028 0.000 0.008
#> GSM25579 1 0.3346 0.5797 0.816 0.008 0.000 0.036 0.000 0.140
#> GSM25580 1 0.5358 0.4139 0.604 0.072 0.016 0.300 0.000 0.008
#> GSM25581 1 0.4748 0.5051 0.676 0.056 0.008 0.252 0.000 0.008
#> GSM48655 5 0.1151 0.6202 0.000 0.012 0.000 0.000 0.956 0.032
#> GSM48656 5 0.1471 0.6120 0.000 0.000 0.000 0.004 0.932 0.064
#> GSM48657 5 0.1176 0.6181 0.000 0.024 0.000 0.000 0.956 0.020
#> GSM48658 5 0.3130 0.5489 0.000 0.008 0.004 0.004 0.808 0.176
#> GSM25624 4 0.5632 0.1086 0.408 0.072 0.012 0.496 0.000 0.012
#> GSM25625 3 0.5942 0.3400 0.076 0.132 0.648 0.132 0.000 0.012
#> GSM25626 3 0.2118 0.6928 0.004 0.048 0.916 0.012 0.000 0.020
#> GSM25627 2 0.8417 0.4332 0.008 0.356 0.272 0.092 0.180 0.092
#> GSM25628 3 0.3473 0.6189 0.000 0.092 0.832 0.004 0.016 0.056
#> GSM25629 2 0.7703 0.1935 0.000 0.332 0.232 0.008 0.284 0.144
#> GSM25630 3 0.6555 0.5811 0.068 0.196 0.588 0.032 0.000 0.116
#> GSM25631 6 0.4534 0.1335 0.000 0.004 0.008 0.012 0.460 0.516
#> GSM25632 3 0.2190 0.7091 0.032 0.032 0.916 0.008 0.000 0.012
#> GSM25633 1 0.4486 0.5349 0.712 0.052 0.004 0.220 0.000 0.012
#> GSM25634 1 0.5463 0.3487 0.572 0.080 0.008 0.328 0.000 0.012
#> GSM25635 1 0.5278 0.3071 0.560 0.056 0.008 0.364 0.000 0.012
#> GSM25656 3 0.5935 0.4949 0.000 0.228 0.596 0.024 0.012 0.140
#> GSM25657 1 0.5054 0.5524 0.720 0.104 0.016 0.132 0.000 0.028
#> GSM25658 2 0.8043 0.3632 0.256 0.300 0.244 0.180 0.000 0.020
#> GSM25659 1 0.5362 0.5137 0.700 0.072 0.012 0.072 0.000 0.144
#> GSM25660 1 0.4329 0.5357 0.720 0.040 0.004 0.224 0.000 0.012
#> GSM25661 1 0.4233 0.5369 0.724 0.040 0.004 0.224 0.000 0.008
#> GSM25662 5 0.3448 0.5954 0.000 0.072 0.004 0.000 0.816 0.108
#> GSM25663 5 0.4053 0.3797 0.000 0.020 0.004 0.000 0.676 0.300
#> GSM25680 6 0.4080 0.1270 0.000 0.008 0.000 0.000 0.456 0.536
#> GSM25681 6 0.5607 0.2561 0.060 0.016 0.016 0.000 0.376 0.532
#> GSM25682 5 0.3356 0.5894 0.000 0.052 0.000 0.000 0.808 0.140
#> GSM25683 5 0.3395 0.5899 0.000 0.060 0.000 0.000 0.808 0.132
#> GSM25684 5 0.3703 0.5871 0.000 0.072 0.004 0.000 0.792 0.132
#> GSM25685 5 0.5350 0.3931 0.000 0.264 0.016 0.000 0.612 0.108
#> GSM25686 5 0.3356 0.5894 0.000 0.052 0.000 0.000 0.808 0.140
#> GSM25687 5 0.3316 0.5914 0.000 0.052 0.000 0.000 0.812 0.136
#> GSM48664 4 0.5880 0.4976 0.144 0.236 0.000 0.584 0.000 0.036
#> GSM48665 4 0.5705 0.1847 0.368 0.088 0.004 0.520 0.000 0.020
Heatmaps for the consensus matrix. It visualizes the probability of two samples to be in a same group.
consensus_heatmap(res, k = 2)
consensus_heatmap(res, k = 3)
consensus_heatmap(res, k = 4)
consensus_heatmap(res, k = 5)
consensus_heatmap(res, k = 6)
Heatmaps for the membership of samples in all partitions to see how consistent they are:
membership_heatmap(res, k = 2)
membership_heatmap(res, k = 3)
membership_heatmap(res, k = 4)
membership_heatmap(res, k = 5)
membership_heatmap(res, k = 6)
As soon as we have had the classes for columns, we can look for signatures which are significantly different between classes which can be candidate marks for certain classes. Following are the heatmaps for signatures.
Signature heatmaps where rows are scaled:
get_signatures(res, k = 2)
get_signatures(res, k = 3)
get_signatures(res, k = 4)
get_signatures(res, k = 5)
get_signatures(res, k = 6)
Signature heatmaps where rows are not scaled:
get_signatures(res, k = 2, scale_rows = FALSE)
get_signatures(res, k = 3, scale_rows = FALSE)
get_signatures(res, k = 4, scale_rows = FALSE)
get_signatures(res, k = 5, scale_rows = FALSE)
get_signatures(res, k = 6, scale_rows = FALSE)
Compare the overlap of signatures from different k:
compare_signatures(res)
get_signature() returns a data frame invisibly. TO get the list of signatures, the function
call should be assigned to a variable explicitly. In following code, if plot argument is set
to FALSE, no heatmap is plotted while only the differential analysis is performed.
# code only for demonstration
tb = get_signature(res, k = ..., plot = FALSE)
An example of the output of tb is:
#> which_row fdr mean_1 mean_2 scaled_mean_1 scaled_mean_2 km
#> 1 38 0.042760348 8.373488 9.131774 -0.5533452 0.5164555 1
#> 2 40 0.018707592 7.106213 8.469186 -0.6173731 0.5762149 1
#> 3 55 0.019134737 10.221463 11.207825 -0.6159697 0.5749050 1
#> 4 59 0.006059896 5.921854 7.869574 -0.6899429 0.6439467 1
#> 5 60 0.018055526 8.928898 10.211722 -0.6204761 0.5791110 1
#> 6 98 0.009384629 15.714769 14.887706 0.6635654 -0.6193277 2
...
The columns in tb are:
which_row: row indices corresponding to the input matrix.fdr: FDR for the differential test. mean_x: The mean value in group x.scaled_mean_x: The mean value in group x after rows are scaled.km: Row groups if k-means clustering is applied to rows.UMAP plot which shows how samples are separated.
dimension_reduction(res, k = 2, method = "UMAP")
dimension_reduction(res, k = 3, method = "UMAP")
dimension_reduction(res, k = 4, method = "UMAP")
dimension_reduction(res, k = 5, method = "UMAP")
dimension_reduction(res, k = 6, method = "UMAP")
Following heatmap shows how subgroups are split when increasing k:
collect_classes(res)
Test correlation between subgroups and known annotations. If the known annotation is numeric, one-way ANOVA test is applied, and if the known annotation is discrete, chi-squared contingency table test is applied.
test_to_known_factors(res)
#> n genotype/variation(p) k
#> CV:kmeans 100 7.47e-05 2
#> CV:kmeans 90 1.40e-03 3
#> CV:kmeans 74 3.85e-06 4
#> CV:kmeans 60 7.39e-10 5
#> CV:kmeans 54 6.55e-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["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 100 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 0.958 0.958 0.982 0.5054 0.495 0.495
#> 3 3 0.504 0.694 0.823 0.2937 0.805 0.625
#> 4 4 0.442 0.354 0.615 0.1301 0.858 0.619
#> 5 5 0.455 0.403 0.605 0.0700 0.863 0.564
#> 6 6 0.485 0.240 0.500 0.0446 0.856 0.484
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
#> GSM25548 2 0.0000 0.972 0.000 1.000
#> GSM25549 2 0.0000 0.972 0.000 1.000
#> GSM25550 2 0.0672 0.967 0.008 0.992
#> GSM25551 2 0.0000 0.972 0.000 1.000
#> GSM25570 2 0.0000 0.972 0.000 1.000
#> GSM25571 2 0.0000 0.972 0.000 1.000
#> GSM25358 1 0.1633 0.967 0.976 0.024
#> GSM25359 2 0.0000 0.972 0.000 1.000
#> GSM25360 1 0.0000 0.989 1.000 0.000
#> GSM25361 2 0.9393 0.471 0.356 0.644
#> GSM25377 1 0.0000 0.989 1.000 0.000
#> GSM25378 1 0.0000 0.989 1.000 0.000
#> GSM25401 1 0.6887 0.774 0.816 0.184
#> GSM25402 1 0.0000 0.989 1.000 0.000
#> GSM25349 2 0.0000 0.972 0.000 1.000
#> GSM25350 2 0.0000 0.972 0.000 1.000
#> GSM25356 1 0.0000 0.989 1.000 0.000
#> GSM25357 2 0.0376 0.970 0.004 0.996
#> GSM25385 1 0.0000 0.989 1.000 0.000
#> GSM25386 1 0.0000 0.989 1.000 0.000
#> GSM25399 1 0.0000 0.989 1.000 0.000
#> GSM25400 1 0.0000 0.989 1.000 0.000
#> GSM48659 2 0.0000 0.972 0.000 1.000
#> GSM48660 2 0.0000 0.972 0.000 1.000
#> GSM25409 2 0.0000 0.972 0.000 1.000
#> GSM25410 1 0.0000 0.989 1.000 0.000
#> GSM25426 2 0.0000 0.972 0.000 1.000
#> GSM25427 1 0.0000 0.989 1.000 0.000
#> GSM25540 2 0.0000 0.972 0.000 1.000
#> GSM25541 2 0.1633 0.953 0.024 0.976
#> GSM25542 2 0.0000 0.972 0.000 1.000
#> GSM25543 2 0.0000 0.972 0.000 1.000
#> GSM25479 1 0.0000 0.989 1.000 0.000
#> GSM25480 1 0.0000 0.989 1.000 0.000
#> GSM25481 1 0.0000 0.989 1.000 0.000
#> GSM25482 1 0.0000 0.989 1.000 0.000
#> GSM48654 2 0.0000 0.972 0.000 1.000
#> GSM48650 2 0.0000 0.972 0.000 1.000
#> GSM48651 2 0.0000 0.972 0.000 1.000
#> GSM48652 2 0.0000 0.972 0.000 1.000
#> GSM48653 2 0.0000 0.972 0.000 1.000
#> GSM48662 2 0.0000 0.972 0.000 1.000
#> GSM48663 2 0.0000 0.972 0.000 1.000
#> GSM25524 1 0.0000 0.989 1.000 0.000
#> GSM25525 1 0.0000 0.989 1.000 0.000
#> GSM25526 1 0.0000 0.989 1.000 0.000
#> GSM25527 1 0.0000 0.989 1.000 0.000
#> GSM25528 1 0.0000 0.989 1.000 0.000
#> GSM25529 1 0.0000 0.989 1.000 0.000
#> GSM25530 1 0.0000 0.989 1.000 0.000
#> GSM25531 1 0.0000 0.989 1.000 0.000
#> GSM48661 2 0.0000 0.972 0.000 1.000
#> GSM25561 1 0.0000 0.989 1.000 0.000
#> GSM25562 1 0.0000 0.989 1.000 0.000
#> GSM25563 1 0.0000 0.989 1.000 0.000
#> GSM25564 1 0.3733 0.917 0.928 0.072
#> GSM25565 2 0.0000 0.972 0.000 1.000
#> GSM25566 2 0.0000 0.972 0.000 1.000
#> GSM25568 1 0.7528 0.723 0.784 0.216
#> GSM25569 2 0.0000 0.972 0.000 1.000
#> GSM25552 2 0.0000 0.972 0.000 1.000
#> GSM25553 2 0.9358 0.478 0.352 0.648
#> GSM25578 1 0.0000 0.989 1.000 0.000
#> GSM25579 1 0.0000 0.989 1.000 0.000
#> GSM25580 1 0.0000 0.989 1.000 0.000
#> GSM25581 1 0.0000 0.989 1.000 0.000
#> GSM48655 2 0.0000 0.972 0.000 1.000
#> GSM48656 2 0.0000 0.972 0.000 1.000
#> GSM48657 2 0.0000 0.972 0.000 1.000
#> GSM48658 2 0.0000 0.972 0.000 1.000
#> GSM25624 1 0.0000 0.989 1.000 0.000
#> GSM25625 1 0.0000 0.989 1.000 0.000
#> GSM25626 1 0.0000 0.989 1.000 0.000
#> GSM25627 2 0.9000 0.546 0.316 0.684
#> GSM25628 2 0.6343 0.808 0.160 0.840
#> GSM25629 2 0.0000 0.972 0.000 1.000
#> GSM25630 1 0.0000 0.989 1.000 0.000
#> GSM25631 2 0.0000 0.972 0.000 1.000
#> GSM25632 1 0.0000 0.989 1.000 0.000
#> GSM25633 1 0.0000 0.989 1.000 0.000
#> GSM25634 1 0.0000 0.989 1.000 0.000
#> GSM25635 1 0.0000 0.989 1.000 0.000
#> GSM25656 2 0.1633 0.953 0.024 0.976
#> GSM25657 1 0.0000 0.989 1.000 0.000
#> GSM25658 1 0.0000 0.989 1.000 0.000
#> GSM25659 1 0.0000 0.989 1.000 0.000
#> GSM25660 1 0.0000 0.989 1.000 0.000
#> GSM25661 1 0.0000 0.989 1.000 0.000
#> GSM25662 2 0.0000 0.972 0.000 1.000
#> GSM25663 2 0.0000 0.972 0.000 1.000
#> GSM25680 2 0.0000 0.972 0.000 1.000
#> GSM25681 2 0.4939 0.870 0.108 0.892
#> GSM25682 2 0.0000 0.972 0.000 1.000
#> GSM25683 2 0.0000 0.972 0.000 1.000
#> GSM25684 2 0.0000 0.972 0.000 1.000
#> GSM25685 2 0.0000 0.972 0.000 1.000
#> GSM25686 2 0.0000 0.972 0.000 1.000
#> GSM25687 2 0.0000 0.972 0.000 1.000
#> GSM48664 1 0.0000 0.989 1.000 0.000
#> GSM48665 1 0.0000 0.989 1.000 0.000
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM25548 2 0.2772 0.87240 0.004 0.916 0.080
#> GSM25549 2 0.3349 0.86820 0.004 0.888 0.108
#> GSM25550 2 0.4749 0.83699 0.040 0.844 0.116
#> GSM25551 2 0.4062 0.81167 0.000 0.836 0.164
#> GSM25570 2 0.3129 0.86817 0.008 0.904 0.088
#> GSM25571 2 0.3112 0.86972 0.004 0.900 0.096
#> GSM25358 3 0.9059 0.33610 0.408 0.136 0.456
#> GSM25359 3 0.6912 -0.01080 0.016 0.444 0.540
#> GSM25360 3 0.5733 0.44249 0.324 0.000 0.676
#> GSM25361 3 0.7661 0.57928 0.172 0.144 0.684
#> GSM25377 1 0.2356 0.83315 0.928 0.000 0.072
#> GSM25378 1 0.2945 0.80195 0.908 0.004 0.088
#> GSM25401 3 0.8693 0.53809 0.232 0.176 0.592
#> GSM25402 1 0.7656 0.23975 0.572 0.052 0.376
#> GSM25349 2 0.3375 0.87340 0.008 0.892 0.100
#> GSM25350 2 0.3459 0.87526 0.012 0.892 0.096
#> GSM25356 1 0.3043 0.80689 0.908 0.008 0.084
#> GSM25357 2 0.5521 0.76076 0.032 0.788 0.180
#> GSM25385 3 0.6062 0.39254 0.384 0.000 0.616
#> GSM25386 3 0.4413 0.59185 0.160 0.008 0.832
#> GSM25399 1 0.1529 0.83257 0.960 0.000 0.040
#> GSM25400 1 0.2878 0.80575 0.904 0.000 0.096
#> GSM48659 2 0.2537 0.88320 0.000 0.920 0.080
#> GSM48660 2 0.1753 0.88332 0.000 0.952 0.048
#> GSM25409 2 0.4540 0.84942 0.028 0.848 0.124
#> GSM25410 3 0.5325 0.56236 0.248 0.004 0.748
#> GSM25426 2 0.5058 0.71872 0.000 0.756 0.244
#> GSM25427 1 0.2492 0.81172 0.936 0.016 0.048
#> GSM25540 3 0.5502 0.50058 0.008 0.248 0.744
#> GSM25541 3 0.5891 0.61612 0.052 0.168 0.780
#> GSM25542 2 0.5948 0.52909 0.000 0.640 0.360
#> GSM25543 3 0.6881 0.20613 0.020 0.388 0.592
#> GSM25479 1 0.2066 0.83736 0.940 0.000 0.060
#> GSM25480 1 0.2448 0.83180 0.924 0.000 0.076
#> GSM25481 1 0.4994 0.71672 0.836 0.052 0.112
#> GSM25482 1 0.3356 0.78903 0.908 0.036 0.056
#> GSM48654 2 0.2711 0.88181 0.000 0.912 0.088
#> GSM48650 2 0.2165 0.88172 0.000 0.936 0.064
#> GSM48651 2 0.1964 0.88298 0.000 0.944 0.056
#> GSM48652 2 0.1529 0.88391 0.000 0.960 0.040
#> GSM48653 2 0.2448 0.88090 0.000 0.924 0.076
#> GSM48662 2 0.1860 0.88812 0.000 0.948 0.052
#> GSM48663 2 0.2200 0.88166 0.004 0.940 0.056
#> GSM25524 1 0.6154 0.30680 0.592 0.000 0.408
#> GSM25525 1 0.3038 0.81961 0.896 0.000 0.104
#> GSM25526 3 0.6189 0.43948 0.364 0.004 0.632
#> GSM25527 1 0.2959 0.82586 0.900 0.000 0.100
#> GSM25528 1 0.5058 0.68586 0.756 0.000 0.244
#> GSM25529 1 0.3340 0.81200 0.880 0.000 0.120
#> GSM25530 1 0.5058 0.68181 0.756 0.000 0.244
#> GSM25531 1 0.3551 0.80860 0.868 0.000 0.132
#> GSM48661 2 0.3340 0.87017 0.000 0.880 0.120
#> GSM25561 1 0.5988 0.41907 0.632 0.000 0.368
#> GSM25562 1 0.3941 0.79057 0.844 0.000 0.156
#> GSM25563 3 0.5882 0.40145 0.348 0.000 0.652
#> GSM25564 1 0.9678 -0.16598 0.444 0.228 0.328
#> GSM25565 2 0.3340 0.87786 0.000 0.880 0.120
#> GSM25566 2 0.2165 0.88881 0.000 0.936 0.064
#> GSM25568 3 0.9233 0.45926 0.268 0.204 0.528
#> GSM25569 2 0.2959 0.88551 0.000 0.900 0.100
#> GSM25552 2 0.4519 0.84203 0.032 0.852 0.116
#> GSM25553 2 0.9707 -0.09075 0.352 0.424 0.224
#> GSM25578 1 0.1860 0.83562 0.948 0.000 0.052
#> GSM25579 1 0.4235 0.77279 0.824 0.000 0.176
#> GSM25580 1 0.0747 0.83513 0.984 0.000 0.016
#> GSM25581 1 0.1411 0.83643 0.964 0.000 0.036
#> GSM48655 2 0.0592 0.88385 0.000 0.988 0.012
#> GSM48656 2 0.2066 0.88730 0.000 0.940 0.060
#> GSM48657 2 0.1529 0.88286 0.000 0.960 0.040
#> GSM48658 2 0.2959 0.88152 0.000 0.900 0.100
#> GSM25624 1 0.0424 0.83239 0.992 0.000 0.008
#> GSM25625 3 0.6192 0.34378 0.420 0.000 0.580
#> GSM25626 3 0.4539 0.60242 0.148 0.016 0.836
#> GSM25627 3 0.6388 0.50521 0.024 0.284 0.692
#> GSM25628 3 0.3845 0.63337 0.012 0.116 0.872
#> GSM25629 3 0.5929 0.42568 0.004 0.320 0.676
#> GSM25630 3 0.6252 0.17904 0.444 0.000 0.556
#> GSM25631 2 0.6849 0.47395 0.020 0.600 0.380
#> GSM25632 3 0.6307 0.02807 0.488 0.000 0.512
#> GSM25633 1 0.2448 0.83345 0.924 0.000 0.076
#> GSM25634 1 0.1643 0.83782 0.956 0.000 0.044
#> GSM25635 1 0.1163 0.83393 0.972 0.000 0.028
#> GSM25656 3 0.3682 0.63088 0.008 0.116 0.876
#> GSM25657 1 0.2878 0.83311 0.904 0.000 0.096
#> GSM25658 1 0.6500 0.00126 0.532 0.004 0.464
#> GSM25659 1 0.5404 0.68317 0.740 0.004 0.256
#> GSM25660 1 0.1411 0.83699 0.964 0.000 0.036
#> GSM25661 1 0.1289 0.83597 0.968 0.000 0.032
#> GSM25662 2 0.1529 0.88445 0.000 0.960 0.040
#> GSM25663 2 0.2796 0.87904 0.000 0.908 0.092
#> GSM25680 2 0.4784 0.82180 0.004 0.796 0.200
#> GSM25681 2 0.7491 0.53639 0.056 0.620 0.324
#> GSM25682 2 0.1031 0.88364 0.000 0.976 0.024
#> GSM25683 2 0.1163 0.88374 0.000 0.972 0.028
#> GSM25684 2 0.2165 0.88275 0.000 0.936 0.064
#> GSM25685 2 0.4605 0.77343 0.000 0.796 0.204
#> GSM25686 2 0.0892 0.88354 0.000 0.980 0.020
#> GSM25687 2 0.1289 0.88281 0.000 0.968 0.032
#> GSM48664 1 0.1031 0.83053 0.976 0.000 0.024
#> GSM48665 1 0.0892 0.82663 0.980 0.000 0.020
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM25548 2 0.281 0.36339 0.000 0.868 0.000 0.132
#> GSM25549 2 0.339 0.36438 0.000 0.852 0.016 0.132
#> GSM25550 2 0.424 0.33734 0.052 0.844 0.024 0.080
#> GSM25551 2 0.614 -0.00231 0.000 0.496 0.048 0.456
#> GSM25570 2 0.182 0.37021 0.000 0.936 0.004 0.060
#> GSM25571 2 0.227 0.36805 0.000 0.912 0.004 0.084
#> GSM25358 3 0.943 0.30362 0.288 0.108 0.368 0.236
#> GSM25359 2 0.820 0.03856 0.012 0.400 0.268 0.320
#> GSM25360 3 0.616 0.54343 0.196 0.040 0.708 0.056
#> GSM25361 3 0.914 0.41906 0.148 0.304 0.424 0.124
#> GSM25377 1 0.430 0.76403 0.824 0.004 0.112 0.060
#> GSM25378 1 0.674 0.61658 0.684 0.040 0.152 0.124
#> GSM25401 3 0.806 0.28135 0.104 0.052 0.424 0.420
#> GSM25402 3 0.815 0.11636 0.376 0.012 0.376 0.236
#> GSM25349 4 0.642 -0.05337 0.004 0.432 0.056 0.508
#> GSM25350 2 0.578 0.13176 0.000 0.560 0.032 0.408
#> GSM25356 1 0.543 0.71327 0.760 0.016 0.148 0.076
#> GSM25357 4 0.687 0.04434 0.016 0.420 0.064 0.500
#> GSM25385 3 0.525 0.52444 0.236 0.004 0.720 0.040
#> GSM25386 3 0.401 0.60571 0.080 0.008 0.848 0.064
#> GSM25399 1 0.371 0.76995 0.848 0.000 0.112 0.040
#> GSM25400 1 0.543 0.69028 0.728 0.000 0.188 0.084
#> GSM48659 2 0.515 -0.03563 0.000 0.536 0.004 0.460
#> GSM48660 4 0.534 0.15987 0.000 0.424 0.012 0.564
#> GSM25409 2 0.639 0.23110 0.028 0.624 0.040 0.308
#> GSM25410 3 0.430 0.60541 0.092 0.032 0.840 0.036
#> GSM25426 4 0.655 0.16155 0.000 0.304 0.104 0.592
#> GSM25427 1 0.546 0.71156 0.776 0.032 0.100 0.092
#> GSM25540 3 0.790 0.22665 0.004 0.312 0.432 0.252
#> GSM25541 3 0.829 0.34475 0.032 0.300 0.468 0.200
#> GSM25542 4 0.761 0.06533 0.000 0.276 0.248 0.476
#> GSM25543 4 0.837 -0.12007 0.024 0.224 0.372 0.380
#> GSM25479 1 0.300 0.78160 0.892 0.004 0.080 0.024
#> GSM25480 1 0.406 0.76586 0.848 0.036 0.096 0.020
#> GSM25481 1 0.782 0.51267 0.612 0.088 0.148 0.152
#> GSM25482 1 0.614 0.66815 0.740 0.064 0.116 0.080
#> GSM48654 4 0.541 0.07932 0.000 0.480 0.012 0.508
#> GSM48650 4 0.530 0.20822 0.000 0.372 0.016 0.612
#> GSM48651 4 0.508 0.17539 0.000 0.420 0.004 0.576
#> GSM48652 4 0.487 0.20234 0.000 0.404 0.000 0.596
#> GSM48653 4 0.505 0.18891 0.000 0.408 0.004 0.588
#> GSM48662 2 0.499 -0.09181 0.000 0.520 0.000 0.480
#> GSM48663 4 0.545 0.16707 0.000 0.388 0.020 0.592
#> GSM25524 1 0.595 0.02269 0.488 0.004 0.480 0.028
#> GSM25525 1 0.418 0.73661 0.824 0.016 0.140 0.020
#> GSM25526 3 0.663 0.51484 0.252 0.000 0.612 0.136
#> GSM25527 1 0.369 0.76685 0.860 0.004 0.088 0.048
#> GSM25528 1 0.513 0.55478 0.680 0.004 0.300 0.016
#> GSM25529 1 0.425 0.71606 0.800 0.008 0.176 0.016
#> GSM25530 1 0.499 0.52476 0.672 0.004 0.316 0.008
#> GSM25531 1 0.385 0.73100 0.820 0.000 0.160 0.020
#> GSM48661 4 0.621 0.02287 0.000 0.472 0.052 0.476
#> GSM25561 3 0.634 0.01697 0.468 0.012 0.484 0.036
#> GSM25562 1 0.552 0.60440 0.684 0.000 0.264 0.052
#> GSM25563 3 0.558 0.50242 0.248 0.004 0.696 0.052
#> GSM25564 1 0.943 -0.14371 0.392 0.140 0.176 0.292
#> GSM25565 2 0.585 -0.02091 0.000 0.508 0.032 0.460
#> GSM25566 2 0.532 0.08170 0.000 0.572 0.012 0.416
#> GSM25568 4 0.969 -0.31699 0.220 0.148 0.304 0.328
#> GSM25569 2 0.544 0.08874 0.000 0.560 0.016 0.424
#> GSM25552 2 0.484 0.33057 0.040 0.808 0.036 0.116
#> GSM25553 2 0.828 0.14436 0.212 0.556 0.088 0.144
#> GSM25578 1 0.238 0.77538 0.916 0.004 0.072 0.008
#> GSM25579 1 0.653 0.59945 0.696 0.132 0.140 0.032
#> GSM25580 1 0.234 0.78072 0.920 0.000 0.060 0.020
#> GSM25581 1 0.206 0.77805 0.932 0.000 0.052 0.016
#> GSM48655 2 0.516 -0.10533 0.000 0.516 0.004 0.480
#> GSM48656 2 0.516 -0.08560 0.000 0.524 0.004 0.472
#> GSM48657 4 0.498 0.11857 0.000 0.464 0.000 0.536
#> GSM48658 2 0.562 0.10060 0.000 0.560 0.024 0.416
#> GSM25624 1 0.432 0.75712 0.824 0.008 0.120 0.048
#> GSM25625 3 0.582 0.48361 0.296 0.004 0.652 0.048
#> GSM25626 3 0.395 0.60927 0.072 0.004 0.848 0.076
#> GSM25627 3 0.706 0.25993 0.024 0.064 0.488 0.424
#> GSM25628 3 0.592 0.52067 0.016 0.064 0.704 0.216
#> GSM25629 4 0.725 0.06644 0.000 0.160 0.336 0.504
#> GSM25630 3 0.565 0.43841 0.296 0.000 0.656 0.048
#> GSM25631 2 0.744 0.20175 0.044 0.616 0.136 0.204
#> GSM25632 3 0.551 0.32811 0.352 0.000 0.620 0.028
#> GSM25633 1 0.310 0.78013 0.876 0.000 0.104 0.020
#> GSM25634 1 0.273 0.78112 0.900 0.004 0.084 0.012
#> GSM25635 1 0.283 0.77376 0.900 0.000 0.060 0.040
#> GSM25656 3 0.699 0.45540 0.024 0.080 0.584 0.312
#> GSM25657 1 0.420 0.74235 0.808 0.000 0.156 0.036
#> GSM25658 3 0.758 0.21113 0.384 0.008 0.456 0.152
#> GSM25659 1 0.715 0.53414 0.632 0.052 0.232 0.084
#> GSM25660 1 0.222 0.78067 0.924 0.000 0.060 0.016
#> GSM25661 1 0.226 0.77806 0.924 0.000 0.056 0.020
#> GSM25662 4 0.569 0.08467 0.000 0.460 0.024 0.516
#> GSM25663 2 0.602 0.23404 0.000 0.632 0.068 0.300
#> GSM25680 2 0.508 0.31903 0.004 0.740 0.040 0.216
#> GSM25681 2 0.702 0.23867 0.056 0.668 0.112 0.164
#> GSM25682 2 0.502 0.07087 0.000 0.600 0.004 0.396
#> GSM25683 2 0.524 -0.00584 0.000 0.556 0.008 0.436
#> GSM25684 2 0.538 -0.04697 0.000 0.540 0.012 0.448
#> GSM25685 4 0.610 0.15705 0.000 0.364 0.056 0.580
#> GSM25686 2 0.503 0.06982 0.000 0.596 0.004 0.400
#> GSM25687 2 0.504 0.07276 0.000 0.592 0.004 0.404
#> GSM48664 1 0.322 0.77016 0.880 0.000 0.076 0.044
#> GSM48665 1 0.241 0.77040 0.916 0.000 0.064 0.020
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM25548 5 0.507 0.39151 0.000 0.296 0.008 0.044 0.652
#> GSM25549 5 0.528 0.49663 0.008 0.224 0.024 0.044 0.700
#> GSM25550 5 0.532 0.50148 0.016 0.200 0.028 0.040 0.716
#> GSM25551 2 0.722 0.11401 0.000 0.356 0.016 0.316 0.312
#> GSM25570 5 0.448 0.48897 0.000 0.224 0.008 0.036 0.732
#> GSM25571 5 0.511 0.42729 0.000 0.268 0.008 0.056 0.668
#> GSM25358 4 0.887 0.10857 0.180 0.036 0.256 0.380 0.148
#> GSM25359 5 0.893 -0.03461 0.020 0.176 0.224 0.284 0.296
#> GSM25360 3 0.554 0.46235 0.160 0.004 0.716 0.056 0.064
#> GSM25361 3 0.847 0.17264 0.100 0.064 0.404 0.096 0.336
#> GSM25377 1 0.627 0.63423 0.636 0.008 0.124 0.204 0.028
#> GSM25378 1 0.677 0.52138 0.584 0.004 0.124 0.236 0.052
#> GSM25401 4 0.689 0.36896 0.056 0.140 0.176 0.612 0.016
#> GSM25402 4 0.849 0.08382 0.220 0.088 0.244 0.416 0.032
#> GSM25349 2 0.668 0.37420 0.012 0.584 0.024 0.140 0.240
#> GSM25350 2 0.674 0.30630 0.004 0.536 0.024 0.144 0.292
#> GSM25356 1 0.601 0.58030 0.612 0.008 0.060 0.292 0.028
#> GSM25357 2 0.756 0.25530 0.016 0.424 0.032 0.348 0.180
#> GSM25385 3 0.597 0.36075 0.156 0.000 0.644 0.180 0.020
#> GSM25386 3 0.391 0.42743 0.048 0.004 0.832 0.092 0.024
#> GSM25399 1 0.501 0.66577 0.728 0.000 0.096 0.164 0.012
#> GSM25400 1 0.633 0.57176 0.616 0.004 0.128 0.224 0.028
#> GSM48659 2 0.585 0.41235 0.000 0.596 0.012 0.092 0.300
#> GSM48660 2 0.334 0.54801 0.000 0.856 0.008 0.064 0.072
#> GSM25409 2 0.726 0.05096 0.016 0.432 0.032 0.128 0.392
#> GSM25410 3 0.546 0.40452 0.100 0.004 0.720 0.144 0.032
#> GSM25426 4 0.680 -0.32940 0.000 0.408 0.024 0.428 0.140
#> GSM25427 1 0.610 0.61802 0.680 0.036 0.064 0.188 0.032
#> GSM25540 5 0.840 0.02982 0.008 0.108 0.300 0.244 0.340
#> GSM25541 3 0.861 -0.01519 0.032 0.084 0.332 0.224 0.328
#> GSM25542 2 0.798 0.25865 0.004 0.476 0.152 0.188 0.180
#> GSM25543 3 0.849 -0.02190 0.008 0.244 0.368 0.136 0.244
#> GSM25479 1 0.492 0.68209 0.764 0.000 0.104 0.088 0.044
#> GSM25480 1 0.620 0.63049 0.660 0.000 0.164 0.088 0.088
#> GSM25481 1 0.777 0.44553 0.532 0.080 0.072 0.252 0.064
#> GSM25482 1 0.745 0.49595 0.564 0.060 0.064 0.240 0.072
#> GSM48654 2 0.467 0.51397 0.000 0.748 0.012 0.064 0.176
#> GSM48650 2 0.407 0.55735 0.000 0.784 0.004 0.164 0.048
#> GSM48651 2 0.372 0.56823 0.000 0.828 0.008 0.060 0.104
#> GSM48652 2 0.323 0.55536 0.000 0.852 0.000 0.060 0.088
#> GSM48653 2 0.500 0.52100 0.000 0.732 0.012 0.108 0.148
#> GSM48662 2 0.392 0.52134 0.000 0.796 0.004 0.044 0.156
#> GSM48663 2 0.437 0.51052 0.008 0.796 0.008 0.092 0.096
#> GSM25524 3 0.571 0.04710 0.368 0.000 0.564 0.040 0.028
#> GSM25525 1 0.547 0.59548 0.684 0.000 0.216 0.072 0.028
#> GSM25526 4 0.712 -0.00702 0.152 0.008 0.404 0.412 0.024
#> GSM25527 1 0.557 0.63246 0.696 0.000 0.156 0.120 0.028
#> GSM25528 1 0.561 0.32130 0.524 0.000 0.416 0.048 0.012
#> GSM25529 1 0.546 0.62296 0.692 0.000 0.200 0.080 0.028
#> GSM25530 1 0.615 0.39313 0.512 0.000 0.384 0.088 0.016
#> GSM25531 1 0.568 0.60423 0.660 0.000 0.212 0.112 0.016
#> GSM48661 2 0.651 0.40234 0.000 0.616 0.068 0.108 0.208
#> GSM25561 3 0.603 0.18647 0.336 0.000 0.556 0.096 0.012
#> GSM25562 1 0.624 0.55282 0.624 0.016 0.248 0.092 0.020
#> GSM25563 3 0.533 0.45284 0.212 0.000 0.692 0.076 0.020
#> GSM25564 1 0.963 -0.04373 0.316 0.204 0.156 0.216 0.108
#> GSM25565 2 0.659 0.43881 0.000 0.576 0.036 0.144 0.244
#> GSM25566 2 0.676 0.30996 0.000 0.492 0.024 0.148 0.336
#> GSM25568 3 0.974 0.03103 0.132 0.224 0.300 0.188 0.156
#> GSM25569 2 0.611 0.36622 0.000 0.604 0.028 0.096 0.272
#> GSM25552 5 0.541 0.49570 0.032 0.208 0.024 0.028 0.708
#> GSM25553 5 0.734 0.42017 0.128 0.132 0.072 0.056 0.612
#> GSM25578 1 0.470 0.67500 0.768 0.000 0.136 0.068 0.028
#> GSM25579 1 0.760 0.41375 0.508 0.004 0.216 0.088 0.184
#> GSM25580 1 0.311 0.69090 0.872 0.000 0.076 0.036 0.016
#> GSM25581 1 0.344 0.69273 0.852 0.000 0.076 0.060 0.012
#> GSM48655 2 0.442 0.53798 0.000 0.756 0.004 0.060 0.180
#> GSM48656 2 0.457 0.49383 0.004 0.752 0.008 0.048 0.188
#> GSM48657 2 0.381 0.55668 0.000 0.816 0.004 0.064 0.116
#> GSM48658 2 0.652 0.23271 0.000 0.524 0.024 0.120 0.332
#> GSM25624 1 0.526 0.66085 0.732 0.000 0.108 0.124 0.036
#> GSM25625 3 0.670 0.28538 0.216 0.004 0.560 0.200 0.020
#> GSM25626 3 0.564 0.33101 0.052 0.024 0.704 0.192 0.028
#> GSM25627 4 0.746 0.35429 0.012 0.140 0.228 0.540 0.080
#> GSM25628 3 0.683 0.12296 0.004 0.080 0.576 0.256 0.084
#> GSM25629 4 0.787 0.21963 0.000 0.176 0.204 0.472 0.148
#> GSM25630 3 0.530 0.44617 0.228 0.004 0.692 0.056 0.020
#> GSM25631 5 0.720 0.42191 0.020 0.176 0.124 0.084 0.596
#> GSM25632 3 0.571 0.41605 0.252 0.000 0.640 0.092 0.016
#> GSM25633 1 0.405 0.68330 0.800 0.000 0.144 0.040 0.016
#> GSM25634 1 0.424 0.68765 0.804 0.000 0.104 0.068 0.024
#> GSM25635 1 0.468 0.68001 0.772 0.000 0.092 0.112 0.024
#> GSM25656 3 0.803 0.07686 0.016 0.108 0.472 0.256 0.148
#> GSM25657 1 0.556 0.65263 0.680 0.000 0.184 0.120 0.016
#> GSM25658 4 0.767 0.04062 0.268 0.016 0.292 0.400 0.024
#> GSM25659 1 0.842 0.21247 0.416 0.020 0.276 0.160 0.128
#> GSM25660 1 0.544 0.66949 0.724 0.000 0.136 0.060 0.080
#> GSM25661 1 0.418 0.69071 0.804 0.000 0.116 0.060 0.020
#> GSM25662 2 0.630 0.46358 0.000 0.584 0.012 0.200 0.204
#> GSM25663 5 0.696 -0.04730 0.000 0.412 0.056 0.100 0.432
#> GSM25680 5 0.611 0.42892 0.000 0.232 0.040 0.096 0.632
#> GSM25681 5 0.638 0.49765 0.012 0.108 0.152 0.064 0.664
#> GSM25682 2 0.577 0.44131 0.000 0.608 0.004 0.116 0.272
#> GSM25683 2 0.573 0.47172 0.000 0.632 0.004 0.140 0.224
#> GSM25684 2 0.593 0.44017 0.000 0.592 0.004 0.132 0.272
#> GSM25685 2 0.676 0.29008 0.000 0.436 0.016 0.388 0.160
#> GSM25686 2 0.560 0.45054 0.000 0.620 0.000 0.120 0.260
#> GSM25687 2 0.542 0.47034 0.000 0.644 0.000 0.112 0.244
#> GSM48664 1 0.451 0.67440 0.776 0.000 0.064 0.140 0.020
#> GSM48665 1 0.386 0.67835 0.832 0.000 0.052 0.088 0.028
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM25548 6 0.510 0.27789 0.000 0.052 0.000 0.012 0.424 0.512
#> GSM25549 6 0.637 0.39674 0.012 0.108 0.020 0.036 0.232 0.592
#> GSM25550 6 0.635 0.44823 0.020 0.072 0.016 0.068 0.204 0.620
#> GSM25551 5 0.676 0.23435 0.000 0.232 0.060 0.028 0.544 0.136
#> GSM25570 6 0.534 0.36083 0.004 0.052 0.008 0.012 0.356 0.568
#> GSM25571 6 0.488 0.25325 0.000 0.048 0.004 0.000 0.448 0.500
#> GSM25358 4 0.933 0.00170 0.068 0.140 0.232 0.312 0.152 0.096
#> GSM25359 6 0.893 0.12122 0.016 0.192 0.204 0.076 0.252 0.260
#> GSM25360 3 0.725 0.43528 0.212 0.068 0.524 0.072 0.000 0.124
#> GSM25361 6 0.914 -0.14401 0.192 0.140 0.248 0.080 0.044 0.296
#> GSM25377 4 0.569 0.05764 0.400 0.008 0.060 0.504 0.000 0.028
#> GSM25378 4 0.677 0.33705 0.248 0.020 0.072 0.556 0.012 0.092
#> GSM25401 4 0.870 -0.10764 0.032 0.164 0.276 0.300 0.192 0.036
#> GSM25402 4 0.795 0.20008 0.120 0.096 0.224 0.480 0.044 0.036
#> GSM25349 5 0.809 0.10795 0.000 0.236 0.060 0.148 0.400 0.156
#> GSM25350 5 0.776 0.07959 0.000 0.260 0.036 0.084 0.376 0.244
#> GSM25356 4 0.696 0.33946 0.264 0.032 0.052 0.548 0.036 0.068
#> GSM25357 5 0.722 0.27448 0.012 0.172 0.040 0.092 0.556 0.128
#> GSM25385 3 0.642 0.37698 0.192 0.032 0.576 0.176 0.004 0.020
#> GSM25386 3 0.507 0.48497 0.080 0.060 0.744 0.084 0.000 0.032
#> GSM25399 1 0.562 0.16619 0.492 0.004 0.076 0.412 0.004 0.012
#> GSM25400 4 0.691 0.12698 0.352 0.044 0.148 0.436 0.004 0.016
#> GSM48659 2 0.636 0.09708 0.000 0.412 0.008 0.012 0.380 0.188
#> GSM48660 5 0.633 0.01504 0.000 0.352 0.028 0.048 0.508 0.064
#> GSM25409 6 0.812 0.01115 0.020 0.152 0.044 0.092 0.336 0.356
#> GSM25410 3 0.543 0.46002 0.072 0.040 0.724 0.120 0.016 0.028
#> GSM25426 5 0.654 0.21300 0.000 0.268 0.096 0.040 0.552 0.044
#> GSM25427 4 0.680 0.31699 0.260 0.040 0.068 0.560 0.020 0.052
#> GSM25540 2 0.812 -0.11193 0.016 0.320 0.260 0.020 0.108 0.276
#> GSM25541 6 0.862 0.03221 0.040 0.252 0.268 0.068 0.060 0.312
#> GSM25542 5 0.796 -0.00389 0.000 0.280 0.132 0.048 0.384 0.156
#> GSM25543 3 0.907 -0.03480 0.048 0.272 0.296 0.076 0.120 0.188
#> GSM25479 1 0.514 0.46229 0.684 0.000 0.056 0.192 0.000 0.068
#> GSM25480 1 0.582 0.41156 0.648 0.012 0.052 0.168 0.000 0.120
#> GSM25481 4 0.836 0.35244 0.196 0.072 0.084 0.472 0.076 0.100
#> GSM25482 4 0.713 0.32537 0.268 0.060 0.036 0.532 0.032 0.072
#> GSM48654 2 0.589 0.16726 0.000 0.484 0.020 0.000 0.372 0.124
#> GSM48650 5 0.497 0.29039 0.000 0.280 0.016 0.032 0.652 0.020
#> GSM48651 5 0.532 -0.06169 0.000 0.440 0.016 0.008 0.492 0.044
#> GSM48652 5 0.496 -0.09011 0.000 0.456 0.012 0.000 0.492 0.040
#> GSM48653 2 0.529 0.13807 0.000 0.540 0.020 0.000 0.380 0.060
#> GSM48662 2 0.650 0.12430 0.000 0.452 0.008 0.032 0.352 0.156
#> GSM48663 5 0.715 0.06994 0.000 0.312 0.028 0.092 0.460 0.108
#> GSM25524 1 0.666 0.06824 0.480 0.032 0.336 0.120 0.000 0.032
#> GSM25525 1 0.392 0.51092 0.804 0.004 0.080 0.088 0.000 0.024
#> GSM25526 3 0.828 0.24807 0.208 0.132 0.424 0.168 0.024 0.044
#> GSM25527 1 0.550 0.44435 0.684 0.020 0.096 0.160 0.000 0.040
#> GSM25528 1 0.515 0.44335 0.660 0.004 0.224 0.096 0.000 0.016
#> GSM25529 1 0.377 0.51039 0.820 0.008 0.080 0.068 0.000 0.024
#> GSM25530 1 0.555 0.44851 0.644 0.012 0.180 0.148 0.000 0.016
#> GSM25531 1 0.555 0.47739 0.660 0.012 0.136 0.164 0.000 0.028
#> GSM48661 2 0.633 0.23072 0.000 0.508 0.032 0.008 0.308 0.144
#> GSM25561 1 0.742 0.02908 0.376 0.036 0.344 0.188 0.000 0.056
#> GSM25562 1 0.697 0.30267 0.516 0.056 0.124 0.260 0.000 0.044
#> GSM25563 3 0.723 0.36848 0.224 0.060 0.512 0.144 0.000 0.060
#> GSM25564 1 0.982 -0.17192 0.252 0.164 0.116 0.184 0.144 0.140
#> GSM25565 5 0.680 0.06616 0.000 0.336 0.044 0.044 0.480 0.096
#> GSM25566 5 0.648 0.26946 0.000 0.160 0.036 0.032 0.576 0.196
#> GSM25568 2 0.955 -0.16266 0.124 0.252 0.248 0.152 0.068 0.156
#> GSM25569 2 0.721 0.04806 0.000 0.360 0.024 0.036 0.316 0.264
#> GSM25552 6 0.646 0.42201 0.048 0.092 0.012 0.040 0.188 0.620
#> GSM25553 6 0.795 0.35445 0.140 0.092 0.044 0.104 0.092 0.528
#> GSM25578 1 0.417 0.50109 0.772 0.004 0.036 0.152 0.000 0.036
#> GSM25579 1 0.691 0.30794 0.572 0.056 0.080 0.116 0.000 0.176
#> GSM25580 1 0.449 0.43953 0.680 0.004 0.048 0.264 0.000 0.004
#> GSM25581 1 0.478 0.46198 0.712 0.012 0.056 0.200 0.000 0.020
#> GSM48655 5 0.470 0.33500 0.000 0.140 0.004 0.028 0.736 0.092
#> GSM48656 2 0.655 0.11682 0.000 0.404 0.016 0.020 0.396 0.164
#> GSM48657 5 0.495 0.29651 0.000 0.196 0.000 0.048 0.696 0.060
#> GSM48658 2 0.686 0.20391 0.004 0.460 0.028 0.016 0.288 0.204
#> GSM25624 4 0.658 0.01731 0.388 0.016 0.092 0.452 0.004 0.048
#> GSM25625 3 0.726 0.31025 0.236 0.060 0.492 0.164 0.000 0.048
#> GSM25626 3 0.507 0.48727 0.084 0.052 0.760 0.056 0.016 0.032
#> GSM25627 3 0.871 0.16848 0.036 0.276 0.344 0.132 0.156 0.056
#> GSM25628 3 0.668 0.36546 0.020 0.256 0.568 0.028 0.048 0.080
#> GSM25629 2 0.804 0.01517 0.004 0.388 0.204 0.060 0.260 0.084
#> GSM25630 3 0.666 0.29309 0.292 0.040 0.520 0.112 0.000 0.036
#> GSM25631 6 0.788 0.23634 0.080 0.260 0.080 0.040 0.064 0.476
#> GSM25632 3 0.610 0.21586 0.356 0.020 0.504 0.104 0.000 0.016
#> GSM25633 1 0.466 0.45969 0.688 0.004 0.064 0.236 0.000 0.008
#> GSM25634 1 0.563 0.37317 0.584 0.012 0.080 0.304 0.000 0.020
#> GSM25635 1 0.576 0.34060 0.580 0.016 0.052 0.316 0.004 0.032
#> GSM25656 3 0.844 0.13880 0.040 0.336 0.344 0.056 0.096 0.128
#> GSM25657 1 0.551 0.43682 0.632 0.012 0.124 0.220 0.000 0.012
#> GSM25658 3 0.818 0.06176 0.268 0.152 0.312 0.236 0.004 0.028
#> GSM25659 1 0.791 0.21132 0.496 0.132 0.092 0.180 0.016 0.084
#> GSM25660 1 0.514 0.44977 0.700 0.012 0.036 0.180 0.000 0.072
#> GSM25661 1 0.404 0.47750 0.744 0.004 0.028 0.212 0.000 0.012
#> GSM25662 5 0.555 0.27387 0.000 0.228 0.028 0.008 0.636 0.100
#> GSM25663 5 0.698 0.12077 0.008 0.184 0.036 0.016 0.468 0.288
#> GSM25680 6 0.712 0.26771 0.012 0.204 0.052 0.008 0.260 0.464
#> GSM25681 6 0.712 0.39935 0.044 0.116 0.076 0.044 0.116 0.604
#> GSM25682 5 0.312 0.39610 0.000 0.040 0.008 0.000 0.840 0.112
#> GSM25683 5 0.303 0.40058 0.000 0.052 0.008 0.000 0.852 0.088
#> GSM25684 5 0.575 0.21539 0.000 0.232 0.036 0.004 0.612 0.116
#> GSM25685 5 0.620 0.18613 0.000 0.312 0.068 0.024 0.548 0.048
#> GSM25686 5 0.235 0.39606 0.000 0.008 0.000 0.000 0.868 0.124
#> GSM25687 5 0.307 0.39316 0.000 0.032 0.000 0.008 0.840 0.120
#> GSM48664 1 0.503 0.09049 0.480 0.016 0.024 0.472 0.000 0.008
#> GSM48665 1 0.511 0.18751 0.536 0.008 0.016 0.412 0.004 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 genotype/variation(p) k
#> CV:skmeans 98 2.90e-05 2
#> CV:skmeans 81 3.48e-05 3
#> CV:skmeans 39 5.81e-01 4
#> CV:skmeans 36 6.53e-05 5
#> CV:skmeans 3 NA 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 100 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.
#> 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 0.141 0.629 0.801 0.4750 0.519 0.519
#> 3 3 0.280 0.596 0.715 0.3798 0.714 0.495
#> 4 4 0.404 0.490 0.724 0.1130 0.838 0.564
#> 5 5 0.443 0.413 0.704 0.0265 0.988 0.952
#> 6 6 0.459 0.444 0.698 0.0131 0.972 0.894
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
#> GSM25548 2 0.4161 0.77409 0.084 0.916
#> GSM25549 2 0.5059 0.77295 0.112 0.888
#> GSM25550 2 0.3431 0.76931 0.064 0.936
#> GSM25551 2 0.9970 0.37705 0.468 0.532
#> GSM25570 2 0.3431 0.76931 0.064 0.936
#> GSM25571 2 0.3733 0.77143 0.072 0.928
#> GSM25358 2 0.5178 0.75815 0.116 0.884
#> GSM25359 2 0.9944 0.40872 0.456 0.544
#> GSM25360 2 0.9248 0.28603 0.340 0.660
#> GSM25361 2 0.9209 0.53837 0.336 0.664
#> GSM25377 1 0.9635 0.49328 0.612 0.388
#> GSM25378 2 1.0000 -0.03854 0.500 0.500
#> GSM25401 1 0.4690 0.72970 0.900 0.100
#> GSM25402 2 0.9881 0.23816 0.436 0.564
#> GSM25349 2 0.4431 0.77864 0.092 0.908
#> GSM25350 2 0.4298 0.77960 0.088 0.912
#> GSM25356 2 0.9993 -0.00792 0.484 0.516
#> GSM25357 2 0.6343 0.77792 0.160 0.840
#> GSM25385 1 0.6048 0.73634 0.852 0.148
#> GSM25386 1 0.9922 0.20483 0.552 0.448
#> GSM25399 1 0.7453 0.69679 0.788 0.212
#> GSM25400 1 0.8955 0.62857 0.688 0.312
#> GSM48659 2 0.2423 0.78965 0.040 0.960
#> GSM48660 2 0.0672 0.78145 0.008 0.992
#> GSM25409 2 0.4939 0.77324 0.108 0.892
#> GSM25410 1 0.9996 0.33290 0.512 0.488
#> GSM25426 2 1.0000 0.11657 0.496 0.504
#> GSM25427 2 0.6801 0.74568 0.180 0.820
#> GSM25540 2 0.9933 0.29641 0.452 0.548
#> GSM25541 2 0.9993 0.20315 0.484 0.516
#> GSM25542 2 0.3584 0.78061 0.068 0.932
#> GSM25543 2 0.7453 0.72331 0.212 0.788
#> GSM25479 1 0.8763 0.54952 0.704 0.296
#> GSM25480 1 0.9686 0.32536 0.604 0.396
#> GSM25481 2 0.4022 0.77323 0.080 0.920
#> GSM25482 2 0.8144 0.53496 0.252 0.748
#> GSM48654 2 0.0672 0.78097 0.008 0.992
#> GSM48650 2 0.5178 0.77163 0.116 0.884
#> GSM48651 2 0.1184 0.78319 0.016 0.984
#> GSM48652 2 0.5842 0.76846 0.140 0.860
#> GSM48653 2 0.7745 0.63587 0.228 0.772
#> GSM48662 2 0.4939 0.79136 0.108 0.892
#> GSM48663 2 0.2043 0.78579 0.032 0.968
#> GSM25524 1 0.4161 0.72454 0.916 0.084
#> GSM25525 1 0.6048 0.71348 0.852 0.148
#> GSM25526 1 0.3733 0.72359 0.928 0.072
#> GSM25527 1 0.5629 0.73863 0.868 0.132
#> GSM25528 1 0.4431 0.73565 0.908 0.092
#> GSM25529 1 0.0938 0.72705 0.988 0.012
#> GSM25530 1 0.0938 0.72597 0.988 0.012
#> GSM25531 1 0.5294 0.74032 0.880 0.120
#> GSM48661 2 0.5178 0.75264 0.116 0.884
#> GSM25561 1 0.7139 0.68473 0.804 0.196
#> GSM25562 1 0.8499 0.57467 0.724 0.276
#> GSM25563 1 0.9460 0.36073 0.636 0.364
#> GSM25564 2 0.3879 0.78470 0.076 0.924
#> GSM25565 2 0.2603 0.78522 0.044 0.956
#> GSM25566 2 0.9170 0.60520 0.332 0.668
#> GSM25568 2 0.6048 0.77048 0.148 0.852
#> GSM25569 2 0.6048 0.78344 0.148 0.852
#> GSM25552 2 0.3431 0.76931 0.064 0.936
#> GSM25553 2 0.3584 0.77001 0.068 0.932
#> GSM25578 1 0.9775 0.32219 0.588 0.412
#> GSM25579 2 0.9209 0.53948 0.336 0.664
#> GSM25580 1 0.8327 0.64238 0.736 0.264
#> GSM25581 1 0.6343 0.71259 0.840 0.160
#> GSM48655 2 0.1414 0.78357 0.020 0.980
#> GSM48656 2 0.1633 0.78514 0.024 0.976
#> GSM48657 2 0.2236 0.78553 0.036 0.964
#> GSM48658 2 0.7602 0.68995 0.220 0.780
#> GSM25624 1 0.9209 0.55951 0.664 0.336
#> GSM25625 1 0.2778 0.73073 0.952 0.048
#> GSM25626 1 0.6438 0.71487 0.836 0.164
#> GSM25627 1 0.4298 0.72489 0.912 0.088
#> GSM25628 1 0.9358 0.38681 0.648 0.352
#> GSM25629 1 0.4690 0.72444 0.900 0.100
#> GSM25630 1 0.9358 0.59282 0.648 0.352
#> GSM25631 2 0.7602 0.71635 0.220 0.780
#> GSM25632 1 0.3431 0.72128 0.936 0.064
#> GSM25633 1 0.7219 0.68986 0.800 0.200
#> GSM25634 1 0.7299 0.71832 0.796 0.204
#> GSM25635 1 0.9963 0.25364 0.536 0.464
#> GSM25656 2 0.9732 0.39714 0.404 0.596
#> GSM25657 1 0.2423 0.73421 0.960 0.040
#> GSM25658 1 0.3584 0.72187 0.932 0.068
#> GSM25659 2 0.6438 0.75150 0.164 0.836
#> GSM25660 1 0.9881 0.29556 0.564 0.436
#> GSM25661 1 0.8861 0.58877 0.696 0.304
#> GSM25662 2 0.4298 0.77076 0.088 0.912
#> GSM25663 2 0.3431 0.78202 0.064 0.936
#> GSM25680 2 0.6048 0.76248 0.148 0.852
#> GSM25681 2 0.6712 0.74317 0.176 0.824
#> GSM25682 2 0.2423 0.78044 0.040 0.960
#> GSM25683 2 0.0376 0.77980 0.004 0.996
#> GSM25684 2 0.3431 0.77904 0.064 0.936
#> GSM25685 2 0.9608 0.43570 0.384 0.616
#> GSM25686 2 0.0000 0.78006 0.000 1.000
#> GSM25687 2 0.2778 0.77887 0.048 0.952
#> GSM48664 2 0.9963 0.01176 0.464 0.536
#> GSM48665 1 0.9087 0.48310 0.676 0.324
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM25548 2 0.2878 0.7405 0.000 0.904 0.096
#> GSM25549 2 0.3112 0.7442 0.004 0.900 0.096
#> GSM25550 2 0.3116 0.7387 0.000 0.892 0.108
#> GSM25551 2 0.8984 0.4060 0.212 0.564 0.224
#> GSM25570 2 0.2959 0.7389 0.000 0.900 0.100
#> GSM25571 2 0.3112 0.7411 0.004 0.900 0.096
#> GSM25358 3 0.5408 0.7078 0.052 0.136 0.812
#> GSM25359 2 0.7610 0.5377 0.216 0.676 0.108
#> GSM25360 3 1.0000 0.0431 0.332 0.332 0.336
#> GSM25361 3 0.8810 0.5688 0.172 0.252 0.576
#> GSM25377 1 0.6597 0.6157 0.664 0.312 0.024
#> GSM25378 2 0.6836 0.6378 0.240 0.704 0.056
#> GSM25401 1 0.5012 0.7034 0.788 0.008 0.204
#> GSM25402 3 0.7139 0.6242 0.244 0.068 0.688
#> GSM25349 2 0.6396 0.5368 0.016 0.664 0.320
#> GSM25350 2 0.5269 0.6912 0.016 0.784 0.200
#> GSM25356 2 0.7816 0.5665 0.288 0.628 0.084
#> GSM25357 2 0.7410 0.4111 0.040 0.576 0.384
#> GSM25385 1 0.4708 0.7574 0.844 0.120 0.036
#> GSM25386 3 0.5680 0.6032 0.212 0.024 0.764
#> GSM25399 1 0.5442 0.7185 0.812 0.056 0.132
#> GSM25400 1 0.6448 0.3934 0.636 0.012 0.352
#> GSM48659 3 0.4663 0.6990 0.016 0.156 0.828
#> GSM48660 3 0.5138 0.6377 0.000 0.252 0.748
#> GSM25409 2 0.3678 0.7493 0.028 0.892 0.080
#> GSM25410 1 0.9294 0.2251 0.484 0.172 0.344
#> GSM25426 3 0.6586 0.5426 0.216 0.056 0.728
#> GSM25427 2 0.2550 0.7319 0.024 0.936 0.040
#> GSM25540 3 0.9537 0.3012 0.224 0.296 0.480
#> GSM25541 3 0.9461 0.3660 0.280 0.224 0.496
#> GSM25542 3 0.4645 0.7036 0.008 0.176 0.816
#> GSM25543 3 0.7266 0.6359 0.080 0.232 0.688
#> GSM25479 1 0.6686 0.3568 0.612 0.372 0.016
#> GSM25480 2 0.5315 0.6542 0.216 0.772 0.012
#> GSM25481 2 0.6275 0.4835 0.008 0.644 0.348
#> GSM25482 2 0.4848 0.7360 0.036 0.836 0.128
#> GSM48654 3 0.2625 0.7053 0.000 0.084 0.916
#> GSM48650 3 0.6447 0.6708 0.060 0.196 0.744
#> GSM48651 3 0.3941 0.7008 0.000 0.156 0.844
#> GSM48652 3 0.7722 0.1568 0.048 0.432 0.520
#> GSM48653 3 0.0475 0.6922 0.004 0.004 0.992
#> GSM48662 2 0.7274 0.1015 0.028 0.520 0.452
#> GSM48663 3 0.4399 0.6855 0.000 0.188 0.812
#> GSM25524 1 0.5159 0.7182 0.820 0.040 0.140
#> GSM25525 1 0.3644 0.7550 0.872 0.124 0.004
#> GSM25526 1 0.2301 0.7602 0.936 0.004 0.060
#> GSM25527 1 0.4172 0.7635 0.868 0.104 0.028
#> GSM25528 1 0.2945 0.7654 0.908 0.088 0.004
#> GSM25529 1 0.2339 0.7653 0.940 0.048 0.012
#> GSM25530 1 0.1411 0.7604 0.964 0.036 0.000
#> GSM25531 1 0.3310 0.7633 0.908 0.028 0.064
#> GSM48661 3 0.3918 0.7106 0.004 0.140 0.856
#> GSM25561 1 0.6154 0.6357 0.752 0.204 0.044
#> GSM25562 1 0.7708 0.1392 0.528 0.048 0.424
#> GSM25563 3 0.7246 0.4782 0.300 0.052 0.648
#> GSM25564 3 0.6507 0.6033 0.028 0.284 0.688
#> GSM25565 3 0.3941 0.7013 0.000 0.156 0.844
#> GSM25566 2 0.7031 0.5961 0.196 0.716 0.088
#> GSM25568 3 0.7694 0.5476 0.068 0.316 0.616
#> GSM25569 2 0.5692 0.7136 0.040 0.784 0.176
#> GSM25552 2 0.3879 0.7231 0.000 0.848 0.152
#> GSM25553 2 0.3686 0.7295 0.000 0.860 0.140
#> GSM25578 2 0.4555 0.6290 0.200 0.800 0.000
#> GSM25579 2 0.3528 0.7189 0.092 0.892 0.016
#> GSM25580 1 0.5105 0.7402 0.828 0.124 0.048
#> GSM25581 1 0.4002 0.7352 0.840 0.160 0.000
#> GSM48655 3 0.5733 0.5255 0.000 0.324 0.676
#> GSM48656 3 0.5138 0.6447 0.000 0.252 0.748
#> GSM48657 3 0.3715 0.7113 0.004 0.128 0.868
#> GSM48658 3 0.7330 0.6603 0.092 0.216 0.692
#> GSM25624 1 0.5810 0.5526 0.664 0.336 0.000
#> GSM25625 1 0.3590 0.7568 0.896 0.028 0.076
#> GSM25626 1 0.4399 0.7441 0.812 0.000 0.188
#> GSM25627 1 0.6106 0.6669 0.756 0.044 0.200
#> GSM25628 3 0.6834 0.4931 0.260 0.048 0.692
#> GSM25629 1 0.6722 0.6375 0.720 0.060 0.220
#> GSM25630 1 0.6297 0.6908 0.756 0.060 0.184
#> GSM25631 2 0.4253 0.7408 0.048 0.872 0.080
#> GSM25632 1 0.2400 0.7591 0.932 0.004 0.064
#> GSM25633 1 0.4702 0.7112 0.788 0.212 0.000
#> GSM25634 1 0.4836 0.7432 0.848 0.072 0.080
#> GSM25635 2 0.6835 0.4352 0.284 0.676 0.040
#> GSM25656 3 0.6633 0.5519 0.212 0.060 0.728
#> GSM25657 1 0.3875 0.7642 0.888 0.044 0.068
#> GSM25658 1 0.2584 0.7586 0.928 0.008 0.064
#> GSM25659 3 0.8886 0.3526 0.132 0.352 0.516
#> GSM25660 2 0.6168 0.1014 0.412 0.588 0.000
#> GSM25661 1 0.6057 0.5511 0.656 0.340 0.004
#> GSM25662 3 0.2261 0.7056 0.000 0.068 0.932
#> GSM25663 3 0.4555 0.6768 0.000 0.200 0.800
#> GSM25680 2 0.4324 0.7342 0.028 0.860 0.112
#> GSM25681 2 0.3045 0.7447 0.020 0.916 0.064
#> GSM25682 3 0.6295 0.0572 0.000 0.472 0.528
#> GSM25683 3 0.3038 0.7060 0.000 0.104 0.896
#> GSM25684 3 0.1964 0.7048 0.000 0.056 0.944
#> GSM25685 3 0.5402 0.6089 0.180 0.028 0.792
#> GSM25686 3 0.4399 0.6926 0.000 0.188 0.812
#> GSM25687 2 0.6309 0.0535 0.000 0.504 0.496
#> GSM48664 1 0.9566 -0.0432 0.424 0.196 0.380
#> GSM48665 1 0.6495 0.1694 0.536 0.460 0.004
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM25548 2 0.0000 0.78506 0.000 1.000 0.000 0.000
#> GSM25549 2 0.0000 0.78506 0.000 1.000 0.000 0.000
#> GSM25550 2 0.0657 0.78681 0.000 0.984 0.004 0.012
#> GSM25551 3 0.5511 0.30866 0.004 0.284 0.676 0.036
#> GSM25570 2 0.0000 0.78506 0.000 1.000 0.000 0.000
#> GSM25571 2 0.0000 0.78506 0.000 1.000 0.000 0.000
#> GSM25358 4 0.3885 0.73407 0.056 0.056 0.024 0.864
#> GSM25359 3 0.5746 0.16419 0.004 0.368 0.600 0.028
#> GSM25360 4 0.8882 0.19633 0.224 0.304 0.060 0.412
#> GSM25361 4 0.7397 0.27014 0.004 0.144 0.392 0.460
#> GSM25377 1 0.3791 0.59609 0.848 0.120 0.012 0.020
#> GSM25378 2 0.6415 0.63251 0.120 0.716 0.116 0.048
#> GSM25401 3 0.5904 0.23923 0.344 0.004 0.612 0.040
#> GSM25402 4 0.5200 0.60610 0.072 0.000 0.184 0.744
#> GSM25349 2 0.5669 0.45282 0.004 0.600 0.024 0.372
#> GSM25350 2 0.4225 0.71522 0.000 0.792 0.024 0.184
#> GSM25356 2 0.7016 0.58725 0.164 0.672 0.072 0.092
#> GSM25357 2 0.7135 0.48741 0.020 0.592 0.112 0.276
#> GSM25385 3 0.7768 -0.00127 0.388 0.076 0.480 0.056
#> GSM25386 4 0.6597 0.27686 0.060 0.008 0.420 0.512
#> GSM25399 1 0.2542 0.59716 0.904 0.000 0.084 0.012
#> GSM25400 1 0.7092 0.19153 0.528 0.012 0.096 0.364
#> GSM48659 4 0.5159 0.69237 0.000 0.088 0.156 0.756
#> GSM48660 4 0.2654 0.72133 0.000 0.108 0.004 0.888
#> GSM25409 2 0.1059 0.78707 0.000 0.972 0.012 0.016
#> GSM25410 4 0.9317 -0.19354 0.352 0.128 0.160 0.360
#> GSM25426 3 0.4331 0.32482 0.000 0.000 0.712 0.288
#> GSM25427 2 0.3997 0.66850 0.164 0.816 0.012 0.008
#> GSM25540 3 0.6066 0.39019 0.004 0.116 0.692 0.188
#> GSM25541 3 0.6586 0.41638 0.032 0.088 0.676 0.204
#> GSM25542 4 0.4106 0.73689 0.000 0.084 0.084 0.832
#> GSM25543 4 0.6555 0.61721 0.000 0.156 0.212 0.632
#> GSM25479 1 0.8223 0.20850 0.428 0.296 0.260 0.016
#> GSM25480 2 0.4540 0.66175 0.032 0.772 0.196 0.000
#> GSM25481 2 0.4905 0.45197 0.004 0.632 0.000 0.364
#> GSM25482 2 0.3221 0.77264 0.020 0.876 0.004 0.100
#> GSM48654 4 0.1792 0.72769 0.000 0.000 0.068 0.932
#> GSM48650 4 0.6472 0.61968 0.000 0.148 0.212 0.640
#> GSM48651 4 0.1798 0.73899 0.000 0.040 0.016 0.944
#> GSM48652 2 0.7621 -0.10950 0.000 0.420 0.204 0.376
#> GSM48653 4 0.2281 0.72363 0.000 0.000 0.096 0.904
#> GSM48662 2 0.6878 0.09016 0.004 0.504 0.092 0.400
#> GSM48663 4 0.1042 0.73439 0.000 0.020 0.008 0.972
#> GSM25524 3 0.5062 0.32878 0.284 0.000 0.692 0.024
#> GSM25525 1 0.7220 0.37145 0.532 0.176 0.292 0.000
#> GSM25526 1 0.5000 -0.05860 0.500 0.000 0.500 0.000
#> GSM25527 1 0.7265 0.25656 0.524 0.128 0.340 0.008
#> GSM25528 1 0.5022 0.48423 0.708 0.028 0.264 0.000
#> GSM25529 1 0.5212 0.22981 0.572 0.008 0.420 0.000
#> GSM25530 1 0.4746 0.41551 0.688 0.008 0.304 0.000
#> GSM25531 1 0.4994 0.51711 0.744 0.000 0.208 0.048
#> GSM48661 4 0.2500 0.74236 0.000 0.044 0.040 0.916
#> GSM25561 1 0.6614 0.36788 0.608 0.056 0.312 0.024
#> GSM25562 3 0.7471 0.40252 0.184 0.008 0.540 0.268
#> GSM25563 4 0.6292 0.23023 0.060 0.000 0.416 0.524
#> GSM25564 4 0.5101 0.64757 0.004 0.228 0.036 0.732
#> GSM25565 4 0.2699 0.74498 0.000 0.068 0.028 0.904
#> GSM25566 2 0.5961 0.52255 0.004 0.636 0.308 0.052
#> GSM25568 4 0.6716 0.56236 0.004 0.252 0.128 0.616
#> GSM25569 2 0.4213 0.75750 0.004 0.832 0.072 0.092
#> GSM25552 2 0.1940 0.77738 0.000 0.924 0.000 0.076
#> GSM25553 2 0.1637 0.78562 0.000 0.940 0.000 0.060
#> GSM25578 2 0.5842 0.07045 0.448 0.520 0.032 0.000
#> GSM25579 2 0.2266 0.76794 0.000 0.912 0.084 0.004
#> GSM25580 1 0.0188 0.59885 0.996 0.000 0.004 0.000
#> GSM25581 1 0.0895 0.60512 0.976 0.020 0.004 0.000
#> GSM48655 4 0.5528 0.60639 0.000 0.236 0.064 0.700
#> GSM48656 4 0.3105 0.72411 0.000 0.140 0.004 0.856
#> GSM48657 4 0.2578 0.73901 0.000 0.036 0.052 0.912
#> GSM48658 4 0.6502 0.61723 0.004 0.124 0.228 0.644
#> GSM25624 1 0.6522 0.45706 0.608 0.280 0.112 0.000
#> GSM25625 3 0.5311 0.17900 0.392 0.008 0.596 0.004
#> GSM25626 3 0.6275 -0.00791 0.460 0.000 0.484 0.056
#> GSM25627 3 0.2988 0.45855 0.112 0.000 0.876 0.012
#> GSM25628 3 0.4401 0.35061 0.004 0.000 0.724 0.272
#> GSM25629 3 0.2662 0.46568 0.084 0.000 0.900 0.016
#> GSM25630 1 0.8119 0.34245 0.540 0.080 0.276 0.104
#> GSM25631 2 0.0707 0.78284 0.000 0.980 0.020 0.000
#> GSM25632 3 0.4998 -0.00662 0.488 0.000 0.512 0.000
#> GSM25633 1 0.2413 0.61351 0.916 0.064 0.020 0.000
#> GSM25634 1 0.1674 0.60453 0.952 0.004 0.012 0.032
#> GSM25635 1 0.6439 0.37315 0.628 0.300 0.036 0.036
#> GSM25656 3 0.4356 0.31246 0.000 0.000 0.708 0.292
#> GSM25657 3 0.6224 0.01910 0.436 0.044 0.516 0.004
#> GSM25658 3 0.4996 0.00145 0.484 0.000 0.516 0.000
#> GSM25659 4 0.7842 0.44771 0.088 0.284 0.072 0.556
#> GSM25660 1 0.4542 0.53146 0.768 0.208 0.020 0.004
#> GSM25661 1 0.2125 0.61086 0.932 0.052 0.012 0.004
#> GSM25662 4 0.1474 0.72918 0.000 0.000 0.052 0.948
#> GSM25663 4 0.1854 0.73852 0.000 0.048 0.012 0.940
#> GSM25680 2 0.1938 0.78016 0.000 0.936 0.052 0.012
#> GSM25681 2 0.0000 0.78506 0.000 1.000 0.000 0.000
#> GSM25682 4 0.4920 0.27393 0.000 0.368 0.004 0.628
#> GSM25683 4 0.1297 0.73588 0.000 0.016 0.020 0.964
#> GSM25684 4 0.2868 0.71061 0.000 0.000 0.136 0.864
#> GSM25685 4 0.4713 0.51194 0.000 0.000 0.360 0.640
#> GSM25686 4 0.1970 0.74092 0.000 0.060 0.008 0.932
#> GSM25687 4 0.5472 0.01591 0.000 0.440 0.016 0.544
#> GSM48664 1 0.7895 0.31226 0.572 0.092 0.084 0.252
#> GSM48665 1 0.4825 0.55346 0.800 0.120 0.068 0.012
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM25548 5 0.0000 0.76294 0.000 0.000 0.000 0.000 1.000
#> GSM25549 5 0.0000 0.76294 0.000 0.000 0.000 0.000 1.000
#> GSM25550 5 0.1106 0.76543 0.000 0.012 0.000 0.024 0.964
#> GSM25551 3 0.5241 0.29415 0.000 0.016 0.672 0.056 0.256
#> GSM25570 5 0.0000 0.76294 0.000 0.000 0.000 0.000 1.000
#> GSM25571 5 0.0000 0.76294 0.000 0.000 0.000 0.000 1.000
#> GSM25358 2 0.3514 0.72479 0.044 0.868 0.024 0.020 0.044
#> GSM25359 3 0.5136 0.20186 0.000 0.016 0.624 0.028 0.332
#> GSM25360 2 0.7712 0.07966 0.216 0.412 0.068 0.000 0.304
#> GSM25361 2 0.6062 0.28468 0.000 0.464 0.416 0.000 0.120
#> GSM25377 1 0.4631 0.34706 0.780 0.012 0.008 0.108 0.092
#> GSM25378 5 0.6183 0.59093 0.116 0.028 0.140 0.032 0.684
#> GSM25401 3 0.6147 0.17571 0.324 0.040 0.572 0.064 0.000
#> GSM25402 2 0.4400 0.62055 0.060 0.744 0.196 0.000 0.000
#> GSM25349 5 0.5867 0.47520 0.008 0.336 0.024 0.044 0.588
#> GSM25350 5 0.4806 0.70391 0.000 0.144 0.016 0.088 0.752
#> GSM25356 5 0.7180 0.52478 0.164 0.072 0.092 0.052 0.620
#> GSM25357 5 0.7124 0.47037 0.008 0.232 0.068 0.128 0.564
#> GSM25385 3 0.7040 0.02468 0.360 0.044 0.504 0.040 0.052
#> GSM25386 2 0.6414 0.32170 0.040 0.508 0.392 0.052 0.008
#> GSM25399 1 0.4955 0.11934 0.720 0.008 0.084 0.188 0.000
#> GSM25400 1 0.6321 -0.16974 0.524 0.340 0.124 0.000 0.012
#> GSM48659 2 0.5196 0.68053 0.000 0.744 0.120 0.084 0.052
#> GSM48660 2 0.2616 0.71010 0.000 0.880 0.000 0.020 0.100
#> GSM25409 5 0.1442 0.76522 0.000 0.012 0.004 0.032 0.952
#> GSM25410 2 0.8097 -0.35800 0.336 0.360 0.176 0.000 0.128
#> GSM25426 3 0.5203 0.27972 0.000 0.272 0.648 0.080 0.000
#> GSM25427 5 0.4092 0.62006 0.164 0.008 0.008 0.028 0.792
#> GSM25540 3 0.5690 0.37623 0.004 0.188 0.684 0.024 0.100
#> GSM25541 3 0.5736 0.40080 0.016 0.192 0.692 0.024 0.076
#> GSM25542 2 0.4124 0.72966 0.000 0.820 0.068 0.040 0.072
#> GSM25543 2 0.6277 0.62541 0.000 0.632 0.180 0.040 0.148
#> GSM25479 1 0.7424 -0.01203 0.412 0.008 0.272 0.020 0.288
#> GSM25480 5 0.3710 0.66110 0.024 0.000 0.192 0.000 0.784
#> GSM25481 5 0.4747 0.46238 0.000 0.352 0.000 0.028 0.620
#> GSM25482 5 0.4244 0.72905 0.016 0.084 0.000 0.100 0.800
#> GSM48654 2 0.1934 0.71791 0.000 0.928 0.052 0.016 0.004
#> GSM48650 2 0.6690 0.62282 0.000 0.620 0.148 0.100 0.132
#> GSM48651 2 0.1651 0.72031 0.000 0.944 0.012 0.008 0.036
#> GSM48652 5 0.7526 -0.15082 0.000 0.376 0.140 0.080 0.404
#> GSM48653 2 0.2754 0.71053 0.000 0.880 0.080 0.040 0.000
#> GSM48662 5 0.6552 0.08090 0.000 0.392 0.064 0.056 0.488
#> GSM48663 2 0.1408 0.71399 0.000 0.948 0.000 0.044 0.008
#> GSM25524 3 0.4241 0.30529 0.264 0.008 0.716 0.012 0.000
#> GSM25525 1 0.6484 -0.10790 0.504 0.000 0.304 0.004 0.188
#> GSM25526 3 0.4305 0.03497 0.488 0.000 0.512 0.000 0.000
#> GSM25527 1 0.6227 -0.16299 0.516 0.008 0.356 0.000 0.120
#> GSM25528 1 0.5104 0.15814 0.672 0.000 0.272 0.028 0.028
#> GSM25529 1 0.4895 0.08579 0.528 0.000 0.452 0.012 0.008
#> GSM25530 1 0.4127 0.03475 0.680 0.000 0.312 0.000 0.008
#> GSM25531 1 0.4384 0.19196 0.728 0.044 0.228 0.000 0.000
#> GSM48661 2 0.2237 0.72615 0.000 0.916 0.040 0.004 0.040
#> GSM25561 1 0.6528 0.20441 0.580 0.024 0.304 0.044 0.048
#> GSM25562 3 0.6027 0.35490 0.144 0.248 0.600 0.000 0.008
#> GSM25563 2 0.5472 0.25259 0.044 0.512 0.436 0.008 0.000
#> GSM25564 2 0.4593 0.64263 0.004 0.728 0.040 0.004 0.224
#> GSM25565 2 0.2437 0.72810 0.000 0.904 0.032 0.004 0.060
#> GSM25566 5 0.6110 0.56381 0.000 0.044 0.264 0.076 0.616
#> GSM25568 2 0.5946 0.56906 0.000 0.612 0.140 0.008 0.240
#> GSM25569 5 0.4303 0.73949 0.000 0.072 0.080 0.040 0.808
#> GSM25552 5 0.1638 0.76012 0.000 0.064 0.000 0.004 0.932
#> GSM25553 5 0.1357 0.76465 0.000 0.048 0.000 0.004 0.948
#> GSM25578 5 0.6187 -0.12299 0.444 0.000 0.032 0.060 0.464
#> GSM25579 5 0.1671 0.74961 0.000 0.000 0.076 0.000 0.924
#> GSM25580 1 0.2179 0.37687 0.896 0.000 0.004 0.100 0.000
#> GSM25581 1 0.2305 0.38191 0.896 0.000 0.000 0.092 0.012
#> GSM48655 2 0.6080 0.59498 0.000 0.648 0.048 0.096 0.208
#> GSM48656 2 0.2674 0.71084 0.000 0.856 0.000 0.004 0.140
#> GSM48657 2 0.3965 0.70409 0.000 0.784 0.028 0.180 0.008
#> GSM48658 2 0.6259 0.60690 0.000 0.620 0.228 0.040 0.112
#> GSM25624 1 0.6047 -0.00875 0.596 0.000 0.124 0.012 0.268
#> GSM25625 3 0.4088 0.21983 0.368 0.000 0.632 0.000 0.000
#> GSM25626 3 0.5750 -0.02726 0.448 0.056 0.484 0.012 0.000
#> GSM25627 3 0.2349 0.39978 0.084 0.004 0.900 0.012 0.000
#> GSM25628 3 0.3934 0.35426 0.000 0.244 0.740 0.016 0.000
#> GSM25629 3 0.2456 0.40105 0.064 0.008 0.904 0.024 0.000
#> GSM25630 4 0.8020 0.00000 0.376 0.068 0.128 0.396 0.032
#> GSM25631 5 0.0404 0.76301 0.000 0.000 0.012 0.000 0.988
#> GSM25632 3 0.4300 0.06071 0.476 0.000 0.524 0.000 0.000
#> GSM25633 1 0.3523 0.39041 0.844 0.000 0.012 0.096 0.048
#> GSM25634 1 0.2756 0.37229 0.880 0.024 0.004 0.092 0.000
#> GSM25635 1 0.6580 0.22064 0.600 0.028 0.016 0.108 0.248
#> GSM25656 3 0.5096 0.28030 0.000 0.272 0.656 0.072 0.000
#> GSM25657 3 0.6071 0.03630 0.400 0.004 0.520 0.036 0.040
#> GSM25658 3 0.4297 0.06625 0.472 0.000 0.528 0.000 0.000
#> GSM25659 2 0.6773 0.45462 0.084 0.560 0.080 0.000 0.276
#> GSM25660 1 0.5123 0.32719 0.728 0.004 0.012 0.096 0.160
#> GSM25661 1 0.3108 0.39042 0.876 0.004 0.012 0.072 0.036
#> GSM25662 2 0.1444 0.71618 0.000 0.948 0.040 0.012 0.000
#> GSM25663 2 0.1787 0.72053 0.000 0.940 0.016 0.012 0.032
#> GSM25680 5 0.2390 0.75251 0.000 0.012 0.044 0.032 0.912
#> GSM25681 5 0.0000 0.76294 0.000 0.000 0.000 0.000 1.000
#> GSM25682 2 0.6090 0.16687 0.000 0.516 0.000 0.136 0.348
#> GSM25683 2 0.3250 0.70338 0.000 0.820 0.008 0.168 0.004
#> GSM25684 2 0.3749 0.68869 0.000 0.816 0.104 0.080 0.000
#> GSM25685 2 0.5265 0.54733 0.000 0.636 0.284 0.080 0.000
#> GSM25686 2 0.3366 0.70226 0.000 0.828 0.000 0.140 0.032
#> GSM25687 2 0.6487 -0.05898 0.000 0.432 0.004 0.160 0.404
#> GSM48664 1 0.8128 0.01543 0.516 0.228 0.080 0.092 0.084
#> GSM48665 1 0.5417 0.35552 0.740 0.004 0.068 0.100 0.088
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM25548 2 0.0000 0.7719 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM25549 2 0.0000 0.7719 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM25550 2 0.0993 0.7747 0.000 0.964 0.000 0.000 0.012 0.024
#> GSM25551 3 0.5037 0.3164 0.000 0.228 0.660 0.000 0.016 0.096
#> GSM25570 2 0.0000 0.7719 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM25571 2 0.0000 0.7719 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM25358 5 0.3156 0.7202 0.044 0.044 0.020 0.000 0.868 0.024
#> GSM25359 3 0.5037 0.2676 0.000 0.304 0.616 0.000 0.016 0.064
#> GSM25360 5 0.7278 0.1906 0.160 0.304 0.096 0.004 0.428 0.008
#> GSM25361 5 0.5756 0.2765 0.000 0.120 0.416 0.000 0.452 0.012
#> GSM25377 1 0.4820 0.2741 0.768 0.088 0.020 0.060 0.008 0.056
#> GSM25378 2 0.5654 0.6017 0.088 0.680 0.164 0.004 0.024 0.040
#> GSM25401 3 0.5574 0.3836 0.188 0.000 0.668 0.020 0.040 0.084
#> GSM25402 5 0.4006 0.6304 0.052 0.000 0.200 0.000 0.744 0.004
#> GSM25349 2 0.5413 0.4837 0.004 0.580 0.024 0.004 0.336 0.052
#> GSM25350 2 0.4410 0.7077 0.000 0.744 0.016 0.000 0.144 0.096
#> GSM25356 2 0.6697 0.5550 0.108 0.624 0.120 0.008 0.072 0.068
#> GSM25357 2 0.6398 0.4801 0.008 0.560 0.060 0.000 0.232 0.140
#> GSM25385 3 0.6451 0.2315 0.264 0.048 0.584 0.020 0.044 0.040
#> GSM25386 5 0.5933 0.3169 0.028 0.008 0.388 0.000 0.492 0.084
#> GSM25399 4 0.4353 0.0000 0.384 0.000 0.028 0.588 0.000 0.000
#> GSM25400 1 0.6842 -0.0321 0.448 0.012 0.160 0.028 0.340 0.012
#> GSM48659 5 0.4711 0.6786 0.000 0.048 0.112 0.000 0.740 0.100
#> GSM48660 5 0.2263 0.7038 0.000 0.100 0.000 0.000 0.884 0.016
#> GSM25409 2 0.1514 0.7737 0.000 0.944 0.004 0.004 0.012 0.036
#> GSM25410 5 0.7632 -0.1502 0.256 0.136 0.236 0.008 0.364 0.000
#> GSM25426 3 0.4999 0.3143 0.000 0.000 0.632 0.000 0.240 0.128
#> GSM25427 2 0.3433 0.5933 0.200 0.780 0.008 0.008 0.004 0.000
#> GSM25540 3 0.5417 0.3958 0.004 0.080 0.676 0.000 0.176 0.064
#> GSM25541 3 0.5422 0.4096 0.012 0.076 0.688 0.000 0.164 0.060
#> GSM25542 5 0.3779 0.7225 0.000 0.072 0.064 0.000 0.816 0.048
#> GSM25543 5 0.5782 0.6326 0.000 0.148 0.168 0.000 0.628 0.056
#> GSM25479 1 0.7213 0.0602 0.340 0.296 0.312 0.012 0.008 0.032
#> GSM25480 2 0.3284 0.6721 0.020 0.784 0.196 0.000 0.000 0.000
#> GSM25481 2 0.4822 0.4887 0.000 0.608 0.000 0.016 0.336 0.040
#> GSM25482 2 0.4370 0.7243 0.008 0.776 0.004 0.020 0.080 0.112
#> GSM48654 5 0.1826 0.7113 0.000 0.004 0.052 0.000 0.924 0.020
#> GSM48650 5 0.6065 0.6275 0.000 0.128 0.140 0.000 0.616 0.116
#> GSM48651 5 0.1370 0.7137 0.000 0.036 0.012 0.000 0.948 0.004
#> GSM48652 2 0.6821 -0.1445 0.000 0.404 0.128 0.000 0.372 0.096
#> GSM48653 5 0.2554 0.7053 0.000 0.000 0.076 0.000 0.876 0.048
#> GSM48662 2 0.5968 0.0976 0.000 0.488 0.060 0.000 0.384 0.068
#> GSM48663 5 0.1196 0.7069 0.000 0.008 0.000 0.000 0.952 0.040
#> GSM25524 3 0.3393 0.4565 0.140 0.000 0.820 0.020 0.008 0.012
#> GSM25525 1 0.6750 0.0786 0.396 0.180 0.380 0.028 0.000 0.016
#> GSM25526 3 0.4607 0.2965 0.356 0.000 0.604 0.028 0.000 0.012
#> GSM25527 3 0.6636 0.0057 0.388 0.132 0.432 0.028 0.008 0.012
#> GSM25528 1 0.5098 0.1761 0.596 0.024 0.340 0.032 0.000 0.008
#> GSM25529 3 0.4540 -0.0493 0.452 0.008 0.524 0.008 0.000 0.008
#> GSM25530 1 0.4906 0.1196 0.572 0.008 0.380 0.028 0.000 0.012
#> GSM25531 1 0.5233 0.2220 0.620 0.000 0.300 0.024 0.044 0.012
#> GSM48661 5 0.2009 0.7201 0.000 0.040 0.040 0.000 0.916 0.004
#> GSM25561 1 0.6702 0.1224 0.528 0.028 0.296 0.104 0.024 0.020
#> GSM25562 3 0.5557 0.4058 0.108 0.008 0.636 0.008 0.228 0.012
#> GSM25563 5 0.5299 0.2451 0.040 0.000 0.432 0.004 0.500 0.024
#> GSM25564 5 0.3987 0.6411 0.004 0.224 0.040 0.000 0.732 0.000
#> GSM25565 5 0.2046 0.7219 0.000 0.060 0.032 0.000 0.908 0.000
#> GSM25566 2 0.5627 0.5689 0.000 0.608 0.256 0.000 0.044 0.092
#> GSM25568 5 0.5534 0.5659 0.000 0.240 0.136 0.004 0.608 0.012
#> GSM25569 2 0.4020 0.7480 0.000 0.804 0.076 0.004 0.072 0.044
#> GSM25552 2 0.1387 0.7691 0.000 0.932 0.000 0.000 0.068 0.000
#> GSM25553 2 0.1141 0.7737 0.000 0.948 0.000 0.000 0.052 0.000
#> GSM25578 1 0.4917 0.1086 0.548 0.404 0.032 0.012 0.000 0.004
#> GSM25579 2 0.1556 0.7577 0.000 0.920 0.080 0.000 0.000 0.000
#> GSM25580 1 0.1890 0.3351 0.916 0.000 0.024 0.060 0.000 0.000
#> GSM25581 1 0.1858 0.3397 0.924 0.012 0.012 0.052 0.000 0.000
#> GSM48655 5 0.5552 0.5972 0.000 0.200 0.044 0.000 0.640 0.116
#> GSM48656 5 0.2260 0.7045 0.000 0.140 0.000 0.000 0.860 0.000
#> GSM48657 5 0.3720 0.6899 0.000 0.008 0.020 0.008 0.776 0.188
#> GSM48658 5 0.5736 0.6088 0.000 0.108 0.224 0.000 0.616 0.052
#> GSM25624 1 0.6645 0.0747 0.496 0.276 0.176 0.040 0.000 0.012
#> GSM25625 3 0.3987 0.4134 0.236 0.000 0.728 0.024 0.000 0.012
#> GSM25626 3 0.5669 0.2940 0.316 0.000 0.584 0.028 0.044 0.028
#> GSM25627 3 0.1442 0.4657 0.012 0.000 0.944 0.000 0.004 0.040
#> GSM25628 3 0.4007 0.3787 0.000 0.000 0.728 0.000 0.220 0.052
#> GSM25629 3 0.1524 0.4552 0.000 0.000 0.932 0.000 0.008 0.060
#> GSM25630 6 0.6101 0.0000 0.232 0.008 0.036 0.072 0.032 0.620
#> GSM25631 2 0.0363 0.7720 0.000 0.988 0.012 0.000 0.000 0.000
#> GSM25632 3 0.4634 0.3152 0.344 0.000 0.616 0.024 0.004 0.012
#> GSM25633 1 0.2771 0.3540 0.888 0.040 0.024 0.036 0.000 0.012
#> GSM25634 1 0.3325 0.3089 0.832 0.000 0.028 0.120 0.016 0.004
#> GSM25635 1 0.4986 0.2203 0.704 0.188 0.012 0.072 0.024 0.000
#> GSM25656 3 0.4888 0.3178 0.000 0.000 0.644 0.000 0.240 0.116
#> GSM25657 3 0.5724 0.2681 0.284 0.036 0.604 0.060 0.004 0.012
#> GSM25658 3 0.4543 0.3224 0.336 0.000 0.624 0.028 0.000 0.012
#> GSM25659 5 0.6173 0.4560 0.076 0.276 0.084 0.004 0.560 0.000
#> GSM25660 1 0.3536 0.3164 0.836 0.100 0.012 0.028 0.004 0.020
#> GSM25661 1 0.1508 0.3378 0.948 0.012 0.016 0.020 0.004 0.000
#> GSM25662 5 0.1196 0.7095 0.000 0.000 0.040 0.000 0.952 0.008
#> GSM25663 5 0.1503 0.7141 0.000 0.032 0.016 0.000 0.944 0.008
#> GSM25680 2 0.2122 0.7610 0.000 0.912 0.040 0.000 0.008 0.040
#> GSM25681 2 0.0000 0.7719 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM25682 5 0.5686 0.1425 0.000 0.348 0.000 0.004 0.500 0.148
#> GSM25683 5 0.3206 0.6891 0.000 0.004 0.008 0.008 0.808 0.172
#> GSM25684 5 0.3473 0.6836 0.000 0.000 0.096 0.000 0.808 0.096
#> GSM25685 5 0.4851 0.5634 0.000 0.000 0.272 0.000 0.632 0.096
#> GSM25686 5 0.3352 0.6871 0.000 0.032 0.000 0.008 0.812 0.148
#> GSM25687 5 0.6091 -0.0886 0.000 0.404 0.004 0.008 0.416 0.168
#> GSM48664 1 0.7679 -0.0259 0.496 0.060 0.080 0.140 0.212 0.012
#> GSM48665 1 0.3818 0.3312 0.824 0.040 0.060 0.068 0.004 0.004
Heatmaps for the consensus matrix. It visualizes the probability of two samples to be in a same group.
consensus_heatmap(res, k = 2)
consensus_heatmap(res, k = 3)
consensus_heatmap(res, k = 4)
consensus_heatmap(res, k = 5)
consensus_heatmap(res, k = 6)
Heatmaps for the membership of samples in all partitions to see how consistent they are:
membership_heatmap(res, k = 2)
membership_heatmap(res, k = 3)
membership_heatmap(res, k = 4)
membership_heatmap(res, k = 5)
membership_heatmap(res, k = 6)
As soon as we have had the classes for columns, we can look for signatures which are significantly different between classes which can be candidate marks for certain classes. Following are the heatmaps for signatures.
Signature heatmaps where rows are scaled:
get_signatures(res, k = 2)
get_signatures(res, k = 3)
get_signatures(res, k = 4)
get_signatures(res, k = 5)
get_signatures(res, k = 6)
Signature heatmaps where rows are not scaled:
get_signatures(res, k = 2, scale_rows = FALSE)
get_signatures(res, k = 3, scale_rows = FALSE)
get_signatures(res, k = 4, scale_rows = FALSE)
get_signatures(res, k = 5, scale_rows = FALSE)
get_signatures(res, k = 6, scale_rows = FALSE)
Compare the overlap of signatures from different k:
compare_signatures(res)
get_signature() returns a data frame invisibly. TO get the list of signatures, the function
call should be assigned to a variable explicitly. In following code, if plot argument is set
to FALSE, no heatmap is plotted while only the differential analysis is performed.
# code only for demonstration
tb = get_signature(res, k = ..., plot = FALSE)
An example of the output of tb is:
#> which_row fdr mean_1 mean_2 scaled_mean_1 scaled_mean_2 km
#> 1 38 0.042760348 8.373488 9.131774 -0.5533452 0.5164555 1
#> 2 40 0.018707592 7.106213 8.469186 -0.6173731 0.5762149 1
#> 3 55 0.019134737 10.221463 11.207825 -0.6159697 0.5749050 1
#> 4 59 0.006059896 5.921854 7.869574 -0.6899429 0.6439467 1
#> 5 60 0.018055526 8.928898 10.211722 -0.6204761 0.5791110 1
#> 6 98 0.009384629 15.714769 14.887706 0.6635654 -0.6193277 2
...
The columns in tb are:
which_row: row indices corresponding to the input matrix.fdr: FDR for the differential test. mean_x: The mean value in group x.scaled_mean_x: The mean value in group x after rows are scaled.km: Row groups if k-means clustering is applied to rows.UMAP plot which shows how samples are separated.
dimension_reduction(res, k = 2, method = "UMAP")
dimension_reduction(res, k = 3, method = "UMAP")
dimension_reduction(res, k = 4, method = "UMAP")
dimension_reduction(res, k = 5, method = "UMAP")
dimension_reduction(res, k = 6, method = "UMAP")
Following heatmap shows how subgroups are split when increasing k:
collect_classes(res)
Test correlation between subgroups and known annotations. If the known annotation is numeric, one-way ANOVA test is applied, and if the known annotation is discrete, chi-squared contingency table test is applied.
test_to_known_factors(res)
#> n genotype/variation(p) k
#> CV:pam 78 4.69e-04 2
#> CV:pam 79 3.84e-06 3
#> CV:pam 55 6.51e-03 4
#> CV:pam 45 9.88e-03 5
#> CV:pam 45 9.88e-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["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 100 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 0.802 0.907 0.956 0.4988 0.495 0.495
#> 3 3 0.512 0.740 0.846 0.2247 0.870 0.739
#> 4 4 0.537 0.583 0.780 0.1071 0.972 0.924
#> 5 5 0.641 0.706 0.816 0.1464 0.797 0.467
#> 6 6 0.641 0.603 0.766 0.0405 0.943 0.753
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
#> GSM25548 2 0.0000 0.932 0.000 1.000
#> GSM25549 2 0.0000 0.932 0.000 1.000
#> GSM25550 2 0.2778 0.914 0.048 0.952
#> GSM25551 2 0.0938 0.929 0.012 0.988
#> GSM25570 2 0.1414 0.926 0.020 0.980
#> GSM25571 2 0.0000 0.932 0.000 1.000
#> GSM25358 1 0.2236 0.942 0.964 0.036
#> GSM25359 2 0.6801 0.803 0.180 0.820
#> GSM25360 1 0.0000 0.974 1.000 0.000
#> GSM25361 2 0.9909 0.307 0.444 0.556
#> GSM25377 1 0.0000 0.974 1.000 0.000
#> GSM25378 1 0.0000 0.974 1.000 0.000
#> GSM25401 1 0.7528 0.710 0.784 0.216
#> GSM25402 1 0.0000 0.974 1.000 0.000
#> GSM25349 2 0.0376 0.931 0.004 0.996
#> GSM25350 2 0.0000 0.932 0.000 1.000
#> GSM25356 1 0.0000 0.974 1.000 0.000
#> GSM25357 2 0.4939 0.874 0.108 0.892
#> GSM25385 1 0.0000 0.974 1.000 0.000
#> GSM25386 1 0.0000 0.974 1.000 0.000
#> GSM25399 1 0.0000 0.974 1.000 0.000
#> GSM25400 1 0.0000 0.974 1.000 0.000
#> GSM48659 2 0.0000 0.932 0.000 1.000
#> GSM48660 2 0.0000 0.932 0.000 1.000
#> GSM25409 2 0.0376 0.931 0.004 0.996
#> GSM25410 1 0.0000 0.974 1.000 0.000
#> GSM25426 2 0.3733 0.901 0.072 0.928
#> GSM25427 1 0.0376 0.971 0.996 0.004
#> GSM25540 2 0.6973 0.793 0.188 0.812
#> GSM25541 2 0.7883 0.736 0.236 0.764
#> GSM25542 2 0.6247 0.827 0.156 0.844
#> GSM25543 2 0.7056 0.790 0.192 0.808
#> GSM25479 1 0.0000 0.974 1.000 0.000
#> GSM25480 1 0.0000 0.974 1.000 0.000
#> GSM25481 1 0.0376 0.971 0.996 0.004
#> GSM25482 1 0.0000 0.974 1.000 0.000
#> GSM48654 2 0.0000 0.932 0.000 1.000
#> GSM48650 2 0.2603 0.916 0.044 0.956
#> GSM48651 2 0.0000 0.932 0.000 1.000
#> GSM48652 2 0.0000 0.932 0.000 1.000
#> GSM48653 2 0.0000 0.932 0.000 1.000
#> GSM48662 2 0.0000 0.932 0.000 1.000
#> GSM48663 2 0.2603 0.916 0.044 0.956
#> GSM25524 1 0.0000 0.974 1.000 0.000
#> GSM25525 1 0.0000 0.974 1.000 0.000
#> GSM25526 1 0.0000 0.974 1.000 0.000
#> GSM25527 1 0.0000 0.974 1.000 0.000
#> GSM25528 1 0.0000 0.974 1.000 0.000
#> GSM25529 1 0.0000 0.974 1.000 0.000
#> GSM25530 1 0.0000 0.974 1.000 0.000
#> GSM25531 1 0.0000 0.974 1.000 0.000
#> GSM48661 2 0.0376 0.931 0.004 0.996
#> GSM25561 1 0.0000 0.974 1.000 0.000
#> GSM25562 1 0.0000 0.974 1.000 0.000
#> GSM25563 1 0.0000 0.974 1.000 0.000
#> GSM25564 1 0.3733 0.904 0.928 0.072
#> GSM25565 2 0.0000 0.932 0.000 1.000
#> GSM25566 2 0.0000 0.932 0.000 1.000
#> GSM25568 2 0.9988 0.152 0.480 0.520
#> GSM25569 2 0.0000 0.932 0.000 1.000
#> GSM25552 2 0.2948 0.912 0.052 0.948
#> GSM25553 2 0.7674 0.743 0.224 0.776
#> GSM25578 1 0.0000 0.974 1.000 0.000
#> GSM25579 1 0.0938 0.964 0.988 0.012
#> GSM25580 1 0.0000 0.974 1.000 0.000
#> GSM25581 1 0.0000 0.974 1.000 0.000
#> GSM48655 2 0.0000 0.932 0.000 1.000
#> GSM48656 2 0.0000 0.932 0.000 1.000
#> GSM48657 2 0.0000 0.932 0.000 1.000
#> GSM48658 2 0.0000 0.932 0.000 1.000
#> GSM25624 1 0.0000 0.974 1.000 0.000
#> GSM25625 1 0.0000 0.974 1.000 0.000
#> GSM25626 1 0.0000 0.974 1.000 0.000
#> GSM25627 1 0.8386 0.618 0.732 0.268
#> GSM25628 1 0.8207 0.641 0.744 0.256
#> GSM25629 2 0.9686 0.413 0.396 0.604
#> GSM25630 1 0.0000 0.974 1.000 0.000
#> GSM25631 2 0.3584 0.903 0.068 0.932
#> GSM25632 1 0.0000 0.974 1.000 0.000
#> GSM25633 1 0.0000 0.974 1.000 0.000
#> GSM25634 1 0.0000 0.974 1.000 0.000
#> GSM25635 1 0.0000 0.974 1.000 0.000
#> GSM25656 1 0.8499 0.602 0.724 0.276
#> GSM25657 1 0.0000 0.974 1.000 0.000
#> GSM25658 1 0.0000 0.974 1.000 0.000
#> GSM25659 1 0.1414 0.957 0.980 0.020
#> GSM25660 1 0.0000 0.974 1.000 0.000
#> GSM25661 1 0.0000 0.974 1.000 0.000
#> GSM25662 2 0.0000 0.932 0.000 1.000
#> GSM25663 2 0.0000 0.932 0.000 1.000
#> GSM25680 2 0.0672 0.930 0.008 0.992
#> GSM25681 2 0.6048 0.837 0.148 0.852
#> GSM25682 2 0.0000 0.932 0.000 1.000
#> GSM25683 2 0.0000 0.932 0.000 1.000
#> GSM25684 2 0.0000 0.932 0.000 1.000
#> GSM25685 2 0.3431 0.906 0.064 0.936
#> GSM25686 2 0.0000 0.932 0.000 1.000
#> GSM25687 2 0.0000 0.932 0.000 1.000
#> GSM48664 1 0.0000 0.974 1.000 0.000
#> GSM48665 1 0.0000 0.974 1.000 0.000
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM25548 2 0.0592 0.8816 0.000 0.988 0.012
#> GSM25549 2 0.0983 0.8804 0.004 0.980 0.016
#> GSM25550 2 0.4475 0.8082 0.144 0.840 0.016
#> GSM25551 2 0.0000 0.8830 0.000 1.000 0.000
#> GSM25570 2 0.1170 0.8797 0.008 0.976 0.016
#> GSM25571 2 0.0983 0.8804 0.004 0.980 0.016
#> GSM25358 1 0.2955 0.6304 0.912 0.080 0.008
#> GSM25359 2 0.5219 0.7627 0.196 0.788 0.016
#> GSM25360 3 0.6228 0.7763 0.372 0.004 0.624
#> GSM25361 2 0.8597 0.4207 0.292 0.576 0.132
#> GSM25377 1 0.0424 0.7211 0.992 0.000 0.008
#> GSM25378 1 0.0000 0.7202 1.000 0.000 0.000
#> GSM25401 3 0.8379 0.7659 0.352 0.096 0.552
#> GSM25402 1 0.4233 0.5079 0.836 0.004 0.160
#> GSM25349 2 0.0983 0.8809 0.004 0.980 0.016
#> GSM25350 2 0.0237 0.8824 0.000 0.996 0.004
#> GSM25356 1 0.0424 0.7211 0.992 0.000 0.008
#> GSM25357 2 0.5610 0.7586 0.196 0.776 0.028
#> GSM25385 3 0.6244 0.7714 0.440 0.000 0.560
#> GSM25386 3 0.5896 0.8186 0.292 0.008 0.700
#> GSM25399 1 0.0424 0.7211 0.992 0.000 0.008
#> GSM25400 1 0.0237 0.7193 0.996 0.000 0.004
#> GSM48659 2 0.0237 0.8829 0.000 0.996 0.004
#> GSM48660 2 0.0475 0.8832 0.004 0.992 0.004
#> GSM25409 2 0.1170 0.8791 0.016 0.976 0.008
#> GSM25410 3 0.6192 0.7979 0.420 0.000 0.580
#> GSM25426 2 0.5318 0.7489 0.204 0.780 0.016
#> GSM25427 1 0.0237 0.7202 0.996 0.000 0.004
#> GSM25540 2 0.5939 0.7165 0.224 0.748 0.028
#> GSM25541 2 0.6490 0.6718 0.256 0.708 0.036
#> GSM25542 2 0.4723 0.7903 0.160 0.824 0.016
#> GSM25543 2 0.6264 0.6920 0.244 0.724 0.032
#> GSM25479 1 0.5138 0.7156 0.748 0.000 0.252
#> GSM25480 1 0.5621 0.6955 0.692 0.000 0.308
#> GSM25481 1 0.0661 0.7197 0.988 0.004 0.008
#> GSM25482 1 0.0829 0.7189 0.984 0.004 0.012
#> GSM48654 2 0.0237 0.8832 0.004 0.996 0.000
#> GSM48650 2 0.4605 0.7585 0.204 0.796 0.000
#> GSM48651 2 0.0000 0.8830 0.000 1.000 0.000
#> GSM48652 2 0.0000 0.8830 0.000 1.000 0.000
#> GSM48653 2 0.0237 0.8832 0.004 0.996 0.000
#> GSM48662 2 0.0000 0.8830 0.000 1.000 0.000
#> GSM48663 2 0.4883 0.7574 0.208 0.788 0.004
#> GSM25524 3 0.5760 0.8156 0.328 0.000 0.672
#> GSM25525 1 0.5810 0.6777 0.664 0.000 0.336
#> GSM25526 3 0.6373 0.8022 0.408 0.004 0.588
#> GSM25527 1 0.4504 0.7311 0.804 0.000 0.196
#> GSM25528 1 0.6180 -0.2314 0.584 0.000 0.416
#> GSM25529 1 0.5988 0.6514 0.632 0.000 0.368
#> GSM25530 3 0.6244 0.6946 0.440 0.000 0.560
#> GSM25531 1 0.4178 0.5673 0.828 0.000 0.172
#> GSM48661 2 0.0747 0.8813 0.016 0.984 0.000
#> GSM25561 1 0.6079 -0.1296 0.612 0.000 0.388
#> GSM25562 1 0.3038 0.6996 0.896 0.000 0.104
#> GSM25563 3 0.5621 0.8193 0.308 0.000 0.692
#> GSM25564 1 0.8160 0.0931 0.608 0.288 0.104
#> GSM25565 2 0.0237 0.8829 0.000 0.996 0.004
#> GSM25566 2 0.0000 0.8830 0.000 1.000 0.000
#> GSM25568 2 0.6597 0.5937 0.312 0.664 0.024
#> GSM25569 2 0.0000 0.8830 0.000 1.000 0.000
#> GSM25552 2 0.5551 0.7349 0.224 0.760 0.016
#> GSM25553 2 0.7509 0.5439 0.300 0.636 0.064
#> GSM25578 1 0.5431 0.7046 0.716 0.000 0.284
#> GSM25579 1 0.6157 0.5398 0.780 0.092 0.128
#> GSM25580 1 0.4750 0.7104 0.784 0.000 0.216
#> GSM25581 1 0.5098 0.7133 0.752 0.000 0.248
#> GSM48655 2 0.0592 0.8810 0.000 0.988 0.012
#> GSM48656 2 0.0000 0.8830 0.000 1.000 0.000
#> GSM48657 2 0.0424 0.8819 0.000 0.992 0.008
#> GSM48658 2 0.0000 0.8830 0.000 1.000 0.000
#> GSM25624 1 0.4605 0.7255 0.796 0.000 0.204
#> GSM25625 3 0.6154 0.7967 0.408 0.000 0.592
#> GSM25626 3 0.6527 0.8285 0.320 0.020 0.660
#> GSM25627 3 0.8614 0.7027 0.228 0.172 0.600
#> GSM25628 3 0.8221 0.7509 0.248 0.128 0.624
#> GSM25629 3 0.9561 0.4958 0.216 0.316 0.468
#> GSM25630 3 0.5835 0.8103 0.340 0.000 0.660
#> GSM25631 2 0.5360 0.7409 0.220 0.768 0.012
#> GSM25632 3 0.5926 0.8146 0.356 0.000 0.644
#> GSM25633 1 0.5098 0.7157 0.752 0.000 0.248
#> GSM25634 1 0.5098 0.7160 0.752 0.000 0.248
#> GSM25635 1 0.4654 0.7116 0.792 0.000 0.208
#> GSM25656 3 0.7979 0.7604 0.248 0.112 0.640
#> GSM25657 1 0.2796 0.7099 0.908 0.000 0.092
#> GSM25658 3 0.6783 0.8100 0.396 0.016 0.588
#> GSM25659 1 0.6662 0.4597 0.736 0.072 0.192
#> GSM25660 1 0.5291 0.7121 0.732 0.000 0.268
#> GSM25661 1 0.5178 0.7127 0.744 0.000 0.256
#> GSM25662 2 0.0000 0.8830 0.000 1.000 0.000
#> GSM25663 2 0.0000 0.8830 0.000 1.000 0.000
#> GSM25680 2 0.1015 0.8820 0.012 0.980 0.008
#> GSM25681 2 0.6927 0.6627 0.240 0.700 0.060
#> GSM25682 2 0.1643 0.8686 0.000 0.956 0.044
#> GSM25683 2 0.1529 0.8703 0.000 0.960 0.040
#> GSM25684 2 0.0000 0.8830 0.000 1.000 0.000
#> GSM25685 2 0.4834 0.7562 0.204 0.792 0.004
#> GSM25686 2 0.1643 0.8686 0.000 0.956 0.044
#> GSM25687 2 0.1643 0.8686 0.000 0.956 0.044
#> GSM48664 1 0.0424 0.7211 0.992 0.000 0.008
#> GSM48665 1 0.0237 0.7229 0.996 0.000 0.004
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM25548 2 0.2714 0.8677 0.004 0.884 0.000 0.112
#> GSM25549 2 0.2999 0.8578 0.004 0.864 0.000 0.132
#> GSM25550 2 0.4213 0.8423 0.028 0.824 0.012 0.136
#> GSM25551 2 0.2593 0.8797 0.000 0.904 0.016 0.080
#> GSM25570 2 0.3182 0.8566 0.004 0.860 0.004 0.132
#> GSM25571 2 0.3128 0.8585 0.004 0.864 0.004 0.128
#> GSM25358 1 0.8514 -0.0895 0.432 0.376 0.096 0.096
#> GSM25359 2 0.3943 0.8561 0.032 0.856 0.024 0.088
#> GSM25360 3 0.6775 0.5240 0.328 0.004 0.568 0.100
#> GSM25361 2 0.9649 -0.0109 0.252 0.356 0.140 0.252
#> GSM25377 4 0.5756 0.6640 0.400 0.000 0.032 0.568
#> GSM25378 1 0.6411 -0.2774 0.584 0.020 0.040 0.356
#> GSM25401 3 0.7997 0.1853 0.056 0.104 0.504 0.336
#> GSM25402 4 0.8719 0.4347 0.308 0.036 0.288 0.368
#> GSM25349 2 0.1396 0.8887 0.004 0.960 0.004 0.032
#> GSM25350 2 0.1296 0.8879 0.004 0.964 0.004 0.028
#> GSM25356 1 0.5842 -0.4561 0.520 0.000 0.032 0.448
#> GSM25357 2 0.4447 0.8390 0.016 0.820 0.040 0.124
#> GSM25385 3 0.5696 0.6103 0.232 0.000 0.692 0.076
#> GSM25386 3 0.2021 0.6551 0.056 0.000 0.932 0.012
#> GSM25399 4 0.5969 0.6776 0.392 0.000 0.044 0.564
#> GSM25400 1 0.5531 0.1791 0.716 0.008 0.052 0.224
#> GSM48659 2 0.1743 0.8889 0.000 0.940 0.004 0.056
#> GSM48660 2 0.1109 0.8870 0.000 0.968 0.004 0.028
#> GSM25409 2 0.2099 0.8860 0.020 0.936 0.004 0.040
#> GSM25410 3 0.4786 0.6279 0.132 0.008 0.796 0.064
#> GSM25426 2 0.6439 0.6313 0.000 0.648 0.180 0.172
#> GSM25427 1 0.5800 -0.3850 0.548 0.000 0.032 0.420
#> GSM25540 2 0.6788 0.7245 0.040 0.680 0.132 0.148
#> GSM25541 2 0.7110 0.7118 0.072 0.668 0.108 0.152
#> GSM25542 2 0.4233 0.8498 0.040 0.848 0.040 0.072
#> GSM25543 2 0.5719 0.8022 0.044 0.764 0.092 0.100
#> GSM25479 1 0.1118 0.5316 0.964 0.000 0.000 0.036
#> GSM25480 1 0.2635 0.5087 0.904 0.000 0.020 0.076
#> GSM25481 1 0.6215 -0.4700 0.512 0.008 0.036 0.444
#> GSM25482 1 0.5838 -0.4493 0.524 0.000 0.032 0.444
#> GSM48654 2 0.0779 0.8901 0.000 0.980 0.004 0.016
#> GSM48650 2 0.3450 0.8659 0.004 0.872 0.040 0.084
#> GSM48651 2 0.1209 0.8894 0.000 0.964 0.004 0.032
#> GSM48652 2 0.1109 0.8883 0.000 0.968 0.004 0.028
#> GSM48653 2 0.1890 0.8886 0.000 0.936 0.008 0.056
#> GSM48662 2 0.0336 0.8904 0.000 0.992 0.000 0.008
#> GSM48663 2 0.1909 0.8871 0.004 0.940 0.008 0.048
#> GSM25524 3 0.6015 0.6055 0.268 0.000 0.652 0.080
#> GSM25525 1 0.3796 0.4807 0.848 0.000 0.056 0.096
#> GSM25526 3 0.3709 0.6240 0.100 0.004 0.856 0.040
#> GSM25527 1 0.2002 0.5186 0.936 0.000 0.020 0.044
#> GSM25528 3 0.6605 0.3213 0.440 0.000 0.480 0.080
#> GSM25529 1 0.4424 0.4542 0.812 0.000 0.088 0.100
#> GSM25530 3 0.6592 0.4549 0.392 0.000 0.524 0.084
#> GSM25531 1 0.5511 0.3401 0.720 0.000 0.196 0.084
#> GSM48661 2 0.1635 0.8889 0.000 0.948 0.008 0.044
#> GSM25561 1 0.6412 0.0199 0.572 0.000 0.348 0.080
#> GSM25562 1 0.3497 0.5059 0.876 0.008 0.060 0.056
#> GSM25563 3 0.5327 0.6419 0.220 0.000 0.720 0.060
#> GSM25564 1 0.8523 -0.0240 0.404 0.404 0.104 0.088
#> GSM25565 2 0.1004 0.8905 0.000 0.972 0.004 0.024
#> GSM25566 2 0.0000 0.8894 0.000 1.000 0.000 0.000
#> GSM25568 2 0.7072 0.6998 0.132 0.672 0.068 0.128
#> GSM25569 2 0.1109 0.8904 0.000 0.968 0.004 0.028
#> GSM25552 2 0.4777 0.8257 0.036 0.796 0.020 0.148
#> GSM25553 2 0.7126 0.6287 0.192 0.644 0.040 0.124
#> GSM25578 1 0.1970 0.5213 0.932 0.000 0.008 0.060
#> GSM25579 1 0.6535 0.3638 0.708 0.068 0.076 0.148
#> GSM25580 1 0.2345 0.4695 0.900 0.000 0.000 0.100
#> GSM25581 1 0.1022 0.5275 0.968 0.000 0.000 0.032
#> GSM48655 2 0.1209 0.8873 0.000 0.964 0.004 0.032
#> GSM48656 2 0.0707 0.8905 0.000 0.980 0.000 0.020
#> GSM48657 2 0.1305 0.8861 0.000 0.960 0.004 0.036
#> GSM48658 2 0.2198 0.8868 0.000 0.920 0.008 0.072
#> GSM25624 1 0.2271 0.4886 0.916 0.000 0.008 0.076
#> GSM25625 3 0.5435 0.6400 0.204 0.004 0.728 0.064
#> GSM25626 3 0.2510 0.6477 0.064 0.008 0.916 0.012
#> GSM25627 3 0.5361 0.5032 0.012 0.184 0.748 0.056
#> GSM25628 3 0.3602 0.6134 0.012 0.072 0.872 0.044
#> GSM25629 3 0.6634 0.2054 0.004 0.384 0.536 0.076
#> GSM25630 3 0.6040 0.6022 0.272 0.000 0.648 0.080
#> GSM25631 2 0.4574 0.8426 0.012 0.808 0.044 0.136
#> GSM25632 3 0.5267 0.6405 0.240 0.000 0.712 0.048
#> GSM25633 1 0.1004 0.5271 0.972 0.000 0.004 0.024
#> GSM25634 1 0.2179 0.5044 0.924 0.000 0.012 0.064
#> GSM25635 1 0.3047 0.4375 0.872 0.000 0.012 0.116
#> GSM25656 3 0.3529 0.6148 0.012 0.068 0.876 0.044
#> GSM25657 1 0.3840 0.4611 0.844 0.000 0.104 0.052
#> GSM25658 3 0.3463 0.6292 0.096 0.004 0.868 0.032
#> GSM25659 1 0.6695 0.3575 0.696 0.056 0.104 0.144
#> GSM25660 1 0.1584 0.5277 0.952 0.000 0.012 0.036
#> GSM25661 1 0.1211 0.5248 0.960 0.000 0.000 0.040
#> GSM25662 2 0.1398 0.8887 0.000 0.956 0.004 0.040
#> GSM25663 2 0.1732 0.8899 0.008 0.948 0.004 0.040
#> GSM25680 2 0.3102 0.8671 0.004 0.872 0.008 0.116
#> GSM25681 2 0.5558 0.7863 0.088 0.748 0.012 0.152
#> GSM25682 2 0.2131 0.8828 0.000 0.932 0.032 0.036
#> GSM25683 2 0.2319 0.8824 0.000 0.924 0.040 0.036
#> GSM25684 2 0.0895 0.8896 0.000 0.976 0.004 0.020
#> GSM25685 2 0.6284 0.6543 0.000 0.664 0.172 0.164
#> GSM25686 2 0.2224 0.8814 0.000 0.928 0.032 0.040
#> GSM25687 2 0.2131 0.8828 0.000 0.932 0.032 0.036
#> GSM48664 1 0.5850 -0.5017 0.512 0.000 0.032 0.456
#> GSM48665 1 0.5600 -0.2577 0.596 0.000 0.028 0.376
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM25548 5 0.1571 0.8083 0.000 0.060 0.000 0.004 0.936
#> GSM25549 5 0.1205 0.8071 0.000 0.040 0.000 0.004 0.956
#> GSM25550 5 0.1604 0.8066 0.004 0.044 0.004 0.004 0.944
#> GSM25551 2 0.2300 0.8866 0.000 0.904 0.024 0.000 0.072
#> GSM25570 5 0.1430 0.8065 0.000 0.052 0.000 0.004 0.944
#> GSM25571 5 0.1502 0.8075 0.000 0.056 0.000 0.004 0.940
#> GSM25358 4 0.9033 0.1194 0.076 0.144 0.116 0.364 0.300
#> GSM25359 5 0.5272 0.3720 0.000 0.396 0.052 0.000 0.552
#> GSM25360 1 0.4597 0.0645 0.564 0.000 0.424 0.000 0.012
#> GSM25361 5 0.3639 0.7403 0.128 0.016 0.028 0.000 0.828
#> GSM25377 4 0.1357 0.8374 0.048 0.000 0.004 0.948 0.000
#> GSM25378 4 0.1697 0.8352 0.060 0.000 0.008 0.932 0.000
#> GSM25401 3 0.6199 0.2574 0.024 0.044 0.528 0.388 0.016
#> GSM25402 4 0.4713 0.4213 0.044 0.000 0.280 0.676 0.000
#> GSM25349 2 0.1956 0.8912 0.000 0.916 0.008 0.000 0.076
#> GSM25350 2 0.2488 0.8717 0.000 0.872 0.004 0.000 0.124
#> GSM25356 4 0.1282 0.8397 0.044 0.000 0.004 0.952 0.000
#> GSM25357 2 0.2740 0.8739 0.000 0.888 0.044 0.004 0.064
#> GSM25385 3 0.4457 0.7409 0.116 0.000 0.760 0.124 0.000
#> GSM25386 3 0.2148 0.8126 0.048 0.004 0.924 0.008 0.016
#> GSM25399 4 0.1357 0.8374 0.048 0.000 0.004 0.948 0.000
#> GSM25400 4 0.4622 -0.0614 0.440 0.000 0.012 0.548 0.000
#> GSM48659 5 0.3630 0.7112 0.000 0.204 0.016 0.000 0.780
#> GSM48660 2 0.1892 0.8904 0.000 0.916 0.004 0.000 0.080
#> GSM25409 5 0.4491 0.5052 0.008 0.364 0.004 0.000 0.624
#> GSM25410 3 0.3446 0.7930 0.048 0.000 0.840 0.108 0.004
#> GSM25426 2 0.3692 0.8119 0.000 0.840 0.084 0.020 0.056
#> GSM25427 4 0.2330 0.8119 0.088 0.004 0.004 0.900 0.004
#> GSM25540 5 0.3340 0.7778 0.004 0.032 0.124 0.000 0.840
#> GSM25541 5 0.2861 0.7973 0.016 0.024 0.076 0.000 0.884
#> GSM25542 5 0.5961 -0.0528 0.000 0.452 0.092 0.004 0.452
#> GSM25543 5 0.4975 0.7391 0.012 0.100 0.140 0.004 0.744
#> GSM25479 1 0.3132 0.7536 0.820 0.000 0.008 0.172 0.000
#> GSM25480 1 0.2331 0.7632 0.900 0.000 0.000 0.080 0.020
#> GSM25481 4 0.1443 0.8393 0.044 0.000 0.004 0.948 0.004
#> GSM25482 4 0.1443 0.8393 0.044 0.000 0.004 0.948 0.004
#> GSM48654 5 0.4981 0.2667 0.000 0.412 0.024 0.004 0.560
#> GSM48650 2 0.2313 0.8635 0.000 0.916 0.040 0.012 0.032
#> GSM48651 2 0.2848 0.8658 0.000 0.840 0.004 0.000 0.156
#> GSM48652 2 0.3039 0.8665 0.000 0.836 0.012 0.000 0.152
#> GSM48653 2 0.3890 0.8424 0.000 0.792 0.036 0.004 0.168
#> GSM48662 2 0.4383 0.2571 0.000 0.572 0.004 0.000 0.424
#> GSM48663 2 0.1831 0.8906 0.000 0.920 0.004 0.000 0.076
#> GSM25524 1 0.4449 -0.1099 0.512 0.000 0.484 0.004 0.000
#> GSM25525 1 0.0579 0.7456 0.984 0.000 0.008 0.000 0.008
#> GSM25526 3 0.2735 0.8051 0.036 0.000 0.880 0.084 0.000
#> GSM25527 1 0.3635 0.7140 0.748 0.000 0.004 0.248 0.000
#> GSM25528 1 0.2179 0.7090 0.896 0.000 0.100 0.004 0.000
#> GSM25529 1 0.0693 0.7458 0.980 0.000 0.008 0.000 0.012
#> GSM25530 1 0.2953 0.6843 0.844 0.000 0.144 0.012 0.000
#> GSM25531 1 0.3214 0.7523 0.844 0.000 0.036 0.120 0.000
#> GSM48661 5 0.3897 0.7274 0.000 0.204 0.028 0.000 0.768
#> GSM25561 1 0.1831 0.7283 0.920 0.000 0.076 0.004 0.000
#> GSM25562 1 0.3810 0.7544 0.788 0.000 0.036 0.176 0.000
#> GSM25563 3 0.4183 0.5087 0.324 0.000 0.668 0.008 0.000
#> GSM25564 5 0.6523 0.5406 0.196 0.016 0.052 0.096 0.640
#> GSM25565 2 0.4016 0.6785 0.000 0.716 0.012 0.000 0.272
#> GSM25566 2 0.3550 0.7323 0.000 0.760 0.004 0.000 0.236
#> GSM25568 5 0.3646 0.7998 0.040 0.072 0.040 0.000 0.848
#> GSM25569 5 0.3550 0.7161 0.000 0.236 0.004 0.000 0.760
#> GSM25552 5 0.1925 0.8050 0.012 0.036 0.012 0.004 0.936
#> GSM25553 5 0.3126 0.7697 0.088 0.028 0.016 0.000 0.868
#> GSM25578 1 0.2477 0.7648 0.892 0.000 0.008 0.092 0.008
#> GSM25579 1 0.1921 0.7460 0.932 0.000 0.012 0.012 0.044
#> GSM25580 1 0.3906 0.6625 0.704 0.000 0.004 0.292 0.000
#> GSM25581 1 0.3318 0.7451 0.800 0.000 0.008 0.192 0.000
#> GSM48655 2 0.1638 0.8888 0.000 0.932 0.000 0.004 0.064
#> GSM48656 5 0.3861 0.6787 0.000 0.284 0.004 0.000 0.712
#> GSM48657 2 0.1768 0.8897 0.000 0.924 0.000 0.004 0.072
#> GSM48658 5 0.2806 0.7809 0.000 0.152 0.004 0.000 0.844
#> GSM25624 1 0.3684 0.6828 0.720 0.000 0.000 0.280 0.000
#> GSM25625 3 0.3802 0.7943 0.096 0.000 0.820 0.080 0.004
#> GSM25626 3 0.2429 0.8181 0.028 0.004 0.916 0.032 0.020
#> GSM25627 3 0.2674 0.7875 0.004 0.044 0.900 0.008 0.044
#> GSM25628 3 0.1883 0.7977 0.000 0.012 0.932 0.008 0.048
#> GSM25629 3 0.3457 0.7488 0.000 0.064 0.848 0.008 0.080
#> GSM25630 1 0.4451 -0.1432 0.504 0.000 0.492 0.004 0.000
#> GSM25631 5 0.1673 0.8081 0.008 0.032 0.016 0.000 0.944
#> GSM25632 3 0.4366 0.4961 0.320 0.000 0.664 0.016 0.000
#> GSM25633 1 0.3231 0.7478 0.800 0.000 0.004 0.196 0.000
#> GSM25634 1 0.3934 0.6925 0.716 0.000 0.008 0.276 0.000
#> GSM25635 1 0.4166 0.5768 0.648 0.000 0.004 0.348 0.000
#> GSM25656 3 0.2171 0.7986 0.004 0.020 0.924 0.008 0.044
#> GSM25657 1 0.3461 0.7311 0.772 0.000 0.004 0.224 0.000
#> GSM25658 3 0.2835 0.8069 0.036 0.000 0.880 0.080 0.004
#> GSM25659 1 0.1780 0.7452 0.940 0.000 0.024 0.008 0.028
#> GSM25660 1 0.3209 0.7506 0.812 0.000 0.000 0.180 0.008
#> GSM25661 1 0.3455 0.7350 0.784 0.000 0.008 0.208 0.000
#> GSM25662 2 0.2462 0.8847 0.000 0.880 0.008 0.000 0.112
#> GSM25663 5 0.3366 0.7489 0.000 0.212 0.004 0.000 0.784
#> GSM25680 5 0.1168 0.8074 0.000 0.032 0.008 0.000 0.960
#> GSM25681 5 0.1653 0.8055 0.028 0.024 0.004 0.000 0.944
#> GSM25682 2 0.1205 0.8780 0.000 0.956 0.000 0.004 0.040
#> GSM25683 2 0.1644 0.8859 0.000 0.940 0.008 0.004 0.048
#> GSM25684 2 0.3209 0.8417 0.000 0.812 0.008 0.000 0.180
#> GSM25685 2 0.4003 0.8042 0.000 0.820 0.088 0.020 0.072
#> GSM25686 2 0.1205 0.8780 0.000 0.956 0.000 0.004 0.040
#> GSM25687 2 0.1205 0.8780 0.000 0.956 0.000 0.004 0.040
#> GSM48664 4 0.1270 0.8395 0.052 0.000 0.000 0.948 0.000
#> GSM48665 4 0.1478 0.8325 0.064 0.000 0.000 0.936 0.000
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM25548 5 0.2925 0.6793 0.000 0.148 0.004 0.000 0.832 0.016
#> GSM25549 5 0.1390 0.7109 0.000 0.032 0.004 0.000 0.948 0.016
#> GSM25550 5 0.1341 0.7042 0.000 0.024 0.000 0.000 0.948 0.028
#> GSM25551 2 0.4070 0.6182 0.000 0.776 0.016 0.004 0.148 0.056
#> GSM25570 5 0.1564 0.7070 0.000 0.040 0.000 0.000 0.936 0.024
#> GSM25571 5 0.2715 0.6941 0.000 0.112 0.004 0.000 0.860 0.024
#> GSM25358 4 0.8553 -0.0097 0.132 0.228 0.120 0.380 0.132 0.008
#> GSM25359 2 0.6206 -0.0723 0.000 0.424 0.036 0.000 0.412 0.128
#> GSM25360 3 0.6401 0.3211 0.292 0.000 0.452 0.000 0.024 0.232
#> GSM25361 5 0.4593 0.6121 0.060 0.004 0.024 0.000 0.728 0.184
#> GSM25377 4 0.0547 0.8478 0.020 0.000 0.000 0.980 0.000 0.000
#> GSM25378 4 0.0458 0.8476 0.016 0.000 0.000 0.984 0.000 0.000
#> GSM25401 3 0.6529 0.0732 0.004 0.016 0.444 0.228 0.004 0.304
#> GSM25402 4 0.4281 0.4766 0.016 0.000 0.272 0.688 0.000 0.024
#> GSM25349 2 0.0603 0.6542 0.000 0.980 0.004 0.000 0.016 0.000
#> GSM25350 2 0.1075 0.6586 0.000 0.952 0.000 0.000 0.048 0.000
#> GSM25356 4 0.0260 0.8493 0.008 0.000 0.000 0.992 0.000 0.000
#> GSM25357 2 0.2763 0.5963 0.008 0.892 0.048 0.012 0.016 0.024
#> GSM25385 3 0.3792 0.6576 0.052 0.000 0.780 0.160 0.000 0.008
#> GSM25386 3 0.1015 0.6690 0.012 0.000 0.968 0.004 0.004 0.012
#> GSM25399 4 0.0458 0.8474 0.016 0.000 0.000 0.984 0.000 0.000
#> GSM25400 4 0.4127 0.1913 0.364 0.000 0.012 0.620 0.000 0.004
#> GSM48659 5 0.4949 0.4767 0.000 0.248 0.004 0.000 0.644 0.104
#> GSM48660 2 0.0458 0.6552 0.000 0.984 0.000 0.000 0.016 0.000
#> GSM25409 2 0.3841 0.4520 0.000 0.616 0.000 0.000 0.380 0.004
#> GSM25410 3 0.2876 0.6705 0.016 0.000 0.844 0.132 0.000 0.008
#> GSM25426 6 0.5815 0.7166 0.000 0.304 0.080 0.008 0.036 0.572
#> GSM25427 4 0.0508 0.8446 0.012 0.004 0.000 0.984 0.000 0.000
#> GSM25540 5 0.4754 0.6179 0.000 0.020 0.052 0.000 0.668 0.260
#> GSM25541 5 0.4070 0.6721 0.008 0.016 0.032 0.000 0.764 0.180
#> GSM25542 2 0.7062 0.1499 0.000 0.412 0.088 0.000 0.292 0.208
#> GSM25543 5 0.6511 0.5568 0.004 0.116 0.112 0.004 0.580 0.184
#> GSM25479 1 0.2020 0.8008 0.896 0.000 0.000 0.096 0.000 0.008
#> GSM25480 1 0.1693 0.7965 0.936 0.000 0.000 0.032 0.020 0.012
#> GSM25481 4 0.0260 0.8481 0.008 0.000 0.000 0.992 0.000 0.000
#> GSM25482 4 0.0260 0.8481 0.008 0.000 0.000 0.992 0.000 0.000
#> GSM48654 5 0.5765 0.1098 0.000 0.352 0.008 0.000 0.496 0.144
#> GSM48650 2 0.5274 -0.4004 0.000 0.548 0.036 0.008 0.024 0.384
#> GSM48651 2 0.3857 0.6180 0.000 0.768 0.000 0.000 0.152 0.080
#> GSM48652 2 0.4923 0.5656 0.000 0.660 0.008 0.000 0.232 0.100
#> GSM48653 2 0.6176 0.3916 0.000 0.516 0.032 0.000 0.288 0.164
#> GSM48662 2 0.3899 0.4986 0.000 0.628 0.000 0.000 0.364 0.008
#> GSM48663 2 0.1218 0.6402 0.000 0.956 0.000 0.004 0.012 0.028
#> GSM25524 3 0.5731 0.3750 0.288 0.000 0.508 0.000 0.000 0.204
#> GSM25525 1 0.2604 0.7448 0.872 0.000 0.028 0.000 0.004 0.096
#> GSM25526 3 0.2662 0.6720 0.012 0.000 0.868 0.108 0.004 0.008
#> GSM25527 1 0.3488 0.7552 0.764 0.000 0.016 0.216 0.000 0.004
#> GSM25528 1 0.5529 0.4021 0.560 0.000 0.228 0.000 0.000 0.212
#> GSM25529 1 0.2798 0.7328 0.852 0.000 0.036 0.000 0.000 0.112
#> GSM25530 1 0.5842 0.2756 0.520 0.000 0.292 0.008 0.000 0.180
#> GSM25531 1 0.4327 0.7537 0.772 0.000 0.072 0.108 0.000 0.048
#> GSM48661 5 0.5965 0.5646 0.000 0.180 0.028 0.000 0.564 0.228
#> GSM25561 1 0.5330 0.4718 0.612 0.000 0.208 0.004 0.000 0.176
#> GSM25562 1 0.3910 0.7790 0.784 0.000 0.044 0.148 0.000 0.024
#> GSM25563 3 0.4840 0.5685 0.152 0.000 0.680 0.004 0.000 0.164
#> GSM25564 5 0.7421 0.3044 0.276 0.076 0.064 0.088 0.488 0.008
#> GSM25565 2 0.4564 0.5626 0.000 0.656 0.012 0.000 0.292 0.040
#> GSM25566 2 0.3975 0.6063 0.000 0.716 0.000 0.000 0.244 0.040
#> GSM25568 5 0.4725 0.6605 0.096 0.040 0.040 0.000 0.768 0.056
#> GSM25569 5 0.4535 0.4454 0.000 0.296 0.000 0.000 0.644 0.060
#> GSM25552 5 0.1485 0.7039 0.000 0.024 0.004 0.000 0.944 0.028
#> GSM25553 5 0.3348 0.6544 0.112 0.016 0.008 0.000 0.836 0.028
#> GSM25578 1 0.1265 0.8001 0.948 0.000 0.000 0.044 0.000 0.008
#> GSM25579 1 0.4001 0.7223 0.800 0.000 0.020 0.008 0.092 0.080
#> GSM25580 1 0.2664 0.7718 0.816 0.000 0.000 0.184 0.000 0.000
#> GSM25581 1 0.2053 0.7984 0.888 0.000 0.000 0.108 0.000 0.004
#> GSM48655 2 0.0603 0.6526 0.000 0.980 0.000 0.000 0.016 0.004
#> GSM48656 5 0.3804 0.2087 0.000 0.424 0.000 0.000 0.576 0.000
#> GSM48657 2 0.0363 0.6532 0.000 0.988 0.000 0.000 0.012 0.000
#> GSM48658 5 0.4184 0.6904 0.000 0.124 0.004 0.000 0.752 0.120
#> GSM25624 1 0.2854 0.7551 0.792 0.000 0.000 0.208 0.000 0.000
#> GSM25625 3 0.3175 0.6812 0.032 0.000 0.844 0.108 0.004 0.012
#> GSM25626 3 0.1843 0.6770 0.016 0.000 0.932 0.032 0.004 0.016
#> GSM25627 3 0.4103 0.5647 0.084 0.000 0.768 0.000 0.012 0.136
#> GSM25628 3 0.2346 0.6238 0.000 0.000 0.868 0.000 0.008 0.124
#> GSM25629 3 0.4316 0.4322 0.000 0.016 0.700 0.000 0.032 0.252
#> GSM25630 3 0.5731 0.3901 0.276 0.000 0.512 0.000 0.000 0.212
#> GSM25631 5 0.2611 0.6861 0.000 0.008 0.012 0.000 0.864 0.116
#> GSM25632 3 0.4635 0.5929 0.148 0.000 0.712 0.008 0.000 0.132
#> GSM25633 1 0.2300 0.7953 0.856 0.000 0.000 0.144 0.000 0.000
#> GSM25634 1 0.3471 0.7716 0.784 0.000 0.020 0.188 0.000 0.008
#> GSM25635 1 0.3288 0.6850 0.724 0.000 0.000 0.276 0.000 0.000
#> GSM25656 3 0.2212 0.6253 0.000 0.000 0.880 0.000 0.008 0.112
#> GSM25657 1 0.3947 0.7607 0.756 0.000 0.032 0.196 0.000 0.016
#> GSM25658 3 0.3957 0.6274 0.100 0.000 0.788 0.100 0.004 0.008
#> GSM25659 1 0.3508 0.7324 0.832 0.000 0.032 0.000 0.064 0.072
#> GSM25660 1 0.2255 0.8009 0.892 0.000 0.000 0.088 0.016 0.004
#> GSM25661 1 0.2191 0.7941 0.876 0.000 0.000 0.120 0.000 0.004
#> GSM25662 2 0.4506 0.5971 0.000 0.704 0.004 0.000 0.204 0.088
#> GSM25663 5 0.4215 0.6026 0.000 0.244 0.000 0.000 0.700 0.056
#> GSM25680 5 0.2563 0.7197 0.000 0.040 0.004 0.000 0.880 0.076
#> GSM25681 5 0.2418 0.6948 0.008 0.004 0.008 0.000 0.884 0.096
#> GSM25682 2 0.0260 0.6386 0.000 0.992 0.000 0.000 0.000 0.008
#> GSM25683 2 0.1679 0.6412 0.000 0.936 0.008 0.000 0.028 0.028
#> GSM25684 2 0.4948 0.5469 0.000 0.636 0.004 0.000 0.264 0.096
#> GSM25685 6 0.6047 0.7315 0.000 0.160 0.092 0.004 0.120 0.624
#> GSM25686 2 0.0260 0.6386 0.000 0.992 0.000 0.000 0.000 0.008
#> GSM25687 2 0.0260 0.6386 0.000 0.992 0.000 0.000 0.000 0.008
#> GSM48664 4 0.0458 0.8492 0.016 0.000 0.000 0.984 0.000 0.000
#> GSM48665 4 0.0363 0.8498 0.012 0.000 0.000 0.988 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 genotype/variation(p) k
#> CV:mclust 97 3.40e-06 2
#> CV:mclust 94 2.57e-04 3
#> CV:mclust 74 8.01e-03 4
#> CV:mclust 88 1.97e-07 5
#> CV:mclust 78 4.07e-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 100 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 0.956 0.929 0.972 0.5032 0.496 0.496
#> 3 3 0.398 0.501 0.722 0.3085 0.736 0.516
#> 4 4 0.436 0.434 0.693 0.1202 0.802 0.492
#> 5 5 0.481 0.414 0.633 0.0713 0.852 0.510
#> 6 6 0.516 0.359 0.605 0.0434 0.908 0.622
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
#> GSM25548 2 0.0000 0.973 0.000 1.000
#> GSM25549 2 0.0000 0.973 0.000 1.000
#> GSM25550 2 0.0000 0.973 0.000 1.000
#> GSM25551 2 0.0000 0.973 0.000 1.000
#> GSM25570 2 0.0000 0.973 0.000 1.000
#> GSM25571 2 0.0000 0.973 0.000 1.000
#> GSM25358 1 0.9427 0.441 0.640 0.360
#> GSM25359 2 0.0000 0.973 0.000 1.000
#> GSM25360 1 0.0000 0.967 1.000 0.000
#> GSM25361 2 0.9896 0.199 0.440 0.560
#> GSM25377 1 0.0000 0.967 1.000 0.000
#> GSM25378 1 0.0672 0.962 0.992 0.008
#> GSM25401 2 0.7299 0.732 0.204 0.796
#> GSM25402 1 0.3114 0.921 0.944 0.056
#> GSM25349 2 0.0000 0.973 0.000 1.000
#> GSM25350 2 0.0000 0.973 0.000 1.000
#> GSM25356 1 0.0000 0.967 1.000 0.000
#> GSM25357 2 0.0000 0.973 0.000 1.000
#> GSM25385 1 0.0000 0.967 1.000 0.000
#> GSM25386 1 0.4815 0.871 0.896 0.104
#> GSM25399 1 0.0000 0.967 1.000 0.000
#> GSM25400 1 0.0000 0.967 1.000 0.000
#> GSM48659 2 0.0000 0.973 0.000 1.000
#> GSM48660 2 0.0000 0.973 0.000 1.000
#> GSM25409 2 0.0000 0.973 0.000 1.000
#> GSM25410 1 0.2236 0.940 0.964 0.036
#> GSM25426 2 0.0000 0.973 0.000 1.000
#> GSM25427 1 0.0938 0.960 0.988 0.012
#> GSM25540 2 0.0000 0.973 0.000 1.000
#> GSM25541 2 0.4022 0.898 0.080 0.920
#> GSM25542 2 0.0000 0.973 0.000 1.000
#> GSM25543 2 0.0000 0.973 0.000 1.000
#> GSM25479 1 0.0000 0.967 1.000 0.000
#> GSM25480 1 0.0000 0.967 1.000 0.000
#> GSM25481 1 0.3114 0.923 0.944 0.056
#> GSM25482 1 0.0938 0.960 0.988 0.012
#> GSM48654 2 0.0000 0.973 0.000 1.000
#> GSM48650 2 0.0000 0.973 0.000 1.000
#> GSM48651 2 0.0000 0.973 0.000 1.000
#> GSM48652 2 0.0000 0.973 0.000 1.000
#> GSM48653 2 0.0000 0.973 0.000 1.000
#> GSM48662 2 0.0000 0.973 0.000 1.000
#> GSM48663 2 0.0000 0.973 0.000 1.000
#> GSM25524 1 0.0000 0.967 1.000 0.000
#> GSM25525 1 0.0000 0.967 1.000 0.000
#> GSM25526 1 0.0000 0.967 1.000 0.000
#> GSM25527 1 0.0000 0.967 1.000 0.000
#> GSM25528 1 0.0000 0.967 1.000 0.000
#> GSM25529 1 0.0000 0.967 1.000 0.000
#> GSM25530 1 0.0000 0.967 1.000 0.000
#> GSM25531 1 0.0000 0.967 1.000 0.000
#> GSM48661 2 0.0000 0.973 0.000 1.000
#> GSM25561 1 0.0000 0.967 1.000 0.000
#> GSM25562 1 0.0000 0.967 1.000 0.000
#> GSM25563 1 0.0000 0.967 1.000 0.000
#> GSM25564 1 0.9087 0.523 0.676 0.324
#> GSM25565 2 0.0000 0.973 0.000 1.000
#> GSM25566 2 0.0000 0.973 0.000 1.000
#> GSM25568 2 0.9922 0.170 0.448 0.552
#> GSM25569 2 0.0000 0.973 0.000 1.000
#> GSM25552 2 0.0672 0.967 0.008 0.992
#> GSM25553 1 0.9983 0.087 0.524 0.476
#> GSM25578 1 0.0000 0.967 1.000 0.000
#> GSM25579 1 0.0000 0.967 1.000 0.000
#> GSM25580 1 0.0000 0.967 1.000 0.000
#> GSM25581 1 0.0000 0.967 1.000 0.000
#> GSM48655 2 0.0000 0.973 0.000 1.000
#> GSM48656 2 0.0000 0.973 0.000 1.000
#> GSM48657 2 0.0000 0.973 0.000 1.000
#> GSM48658 2 0.0000 0.973 0.000 1.000
#> GSM25624 1 0.0000 0.967 1.000 0.000
#> GSM25625 1 0.0000 0.967 1.000 0.000
#> GSM25626 1 0.1633 0.951 0.976 0.024
#> GSM25627 2 0.0000 0.973 0.000 1.000
#> GSM25628 2 0.4022 0.898 0.080 0.920
#> GSM25629 2 0.0000 0.973 0.000 1.000
#> GSM25630 1 0.0000 0.967 1.000 0.000
#> GSM25631 2 0.0376 0.970 0.004 0.996
#> GSM25632 1 0.0000 0.967 1.000 0.000
#> GSM25633 1 0.0000 0.967 1.000 0.000
#> GSM25634 1 0.0000 0.967 1.000 0.000
#> GSM25635 1 0.0000 0.967 1.000 0.000
#> GSM25656 2 0.1843 0.949 0.028 0.972
#> GSM25657 1 0.0000 0.967 1.000 0.000
#> GSM25658 1 0.0000 0.967 1.000 0.000
#> GSM25659 1 0.0376 0.965 0.996 0.004
#> GSM25660 1 0.0000 0.967 1.000 0.000
#> GSM25661 1 0.0000 0.967 1.000 0.000
#> GSM25662 2 0.0000 0.973 0.000 1.000
#> GSM25663 2 0.0000 0.973 0.000 1.000
#> GSM25680 2 0.0000 0.973 0.000 1.000
#> GSM25681 2 0.2236 0.942 0.036 0.964
#> GSM25682 2 0.0000 0.973 0.000 1.000
#> GSM25683 2 0.0000 0.973 0.000 1.000
#> GSM25684 2 0.0000 0.973 0.000 1.000
#> GSM25685 2 0.0000 0.973 0.000 1.000
#> GSM25686 2 0.0000 0.973 0.000 1.000
#> GSM25687 2 0.0000 0.973 0.000 1.000
#> GSM48664 1 0.0000 0.967 1.000 0.000
#> GSM48665 1 0.0000 0.967 1.000 0.000
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM25548 2 0.518 0.6254 0.000 0.744 0.256
#> GSM25549 2 0.418 0.6843 0.000 0.828 0.172
#> GSM25550 2 0.429 0.5912 0.064 0.872 0.064
#> GSM25551 2 0.631 0.0327 0.000 0.504 0.496
#> GSM25570 2 0.176 0.6722 0.004 0.956 0.040
#> GSM25571 2 0.378 0.6953 0.004 0.864 0.132
#> GSM25358 1 0.906 0.3463 0.544 0.276 0.180
#> GSM25359 3 0.559 0.4014 0.000 0.304 0.696
#> GSM25360 3 0.631 -0.3684 0.488 0.000 0.512
#> GSM25361 3 0.385 0.5105 0.108 0.016 0.876
#> GSM25377 1 0.503 0.7009 0.828 0.132 0.040
#> GSM25378 1 0.703 0.5288 0.660 0.296 0.044
#> GSM25401 3 0.929 0.3142 0.184 0.312 0.504
#> GSM25402 1 0.702 0.6716 0.728 0.156 0.116
#> GSM25349 2 0.295 0.6800 0.020 0.920 0.060
#> GSM25350 2 0.177 0.6654 0.016 0.960 0.024
#> GSM25356 1 0.731 0.5259 0.648 0.296 0.056
#> GSM25357 2 0.550 0.5720 0.000 0.708 0.292
#> GSM25385 1 0.642 0.4649 0.572 0.004 0.424
#> GSM25386 3 0.516 0.3660 0.216 0.008 0.776
#> GSM25399 1 0.231 0.7615 0.944 0.032 0.024
#> GSM25400 1 0.219 0.7698 0.948 0.024 0.028
#> GSM48659 3 0.628 0.0519 0.000 0.460 0.540
#> GSM48660 2 0.220 0.6853 0.004 0.940 0.056
#> GSM25409 2 0.113 0.6727 0.004 0.976 0.020
#> GSM25410 3 0.623 0.1243 0.340 0.008 0.652
#> GSM25426 3 0.611 0.2426 0.000 0.396 0.604
#> GSM25427 1 0.783 0.2180 0.500 0.448 0.052
#> GSM25540 3 0.311 0.5355 0.004 0.096 0.900
#> GSM25541 3 0.287 0.5393 0.008 0.076 0.916
#> GSM25542 3 0.593 0.3388 0.000 0.356 0.644
#> GSM25543 3 0.550 0.4145 0.000 0.292 0.708
#> GSM25479 1 0.177 0.7710 0.960 0.016 0.024
#> GSM25480 1 0.437 0.7641 0.860 0.032 0.108
#> GSM25481 2 0.776 0.1439 0.360 0.580 0.060
#> GSM25482 2 0.764 0.1130 0.372 0.576 0.052
#> GSM48654 2 0.625 0.2422 0.000 0.556 0.444
#> GSM48650 2 0.571 0.5398 0.000 0.680 0.320
#> GSM48651 2 0.573 0.5328 0.000 0.676 0.324
#> GSM48652 2 0.586 0.4884 0.000 0.656 0.344
#> GSM48653 3 0.625 0.0934 0.000 0.444 0.556
#> GSM48662 2 0.388 0.6921 0.000 0.848 0.152
#> GSM48663 2 0.253 0.6443 0.020 0.936 0.044
#> GSM25524 1 0.626 0.4644 0.552 0.000 0.448
#> GSM25525 1 0.412 0.7495 0.832 0.000 0.168
#> GSM25526 3 0.631 0.1449 0.328 0.012 0.660
#> GSM25527 1 0.319 0.7631 0.888 0.000 0.112
#> GSM25528 1 0.556 0.6556 0.700 0.000 0.300
#> GSM25529 1 0.445 0.7385 0.808 0.000 0.192
#> GSM25530 1 0.543 0.6616 0.716 0.000 0.284
#> GSM25531 1 0.394 0.7454 0.844 0.000 0.156
#> GSM48661 3 0.571 0.3787 0.000 0.320 0.680
#> GSM25561 1 0.579 0.6187 0.668 0.000 0.332
#> GSM25562 1 0.327 0.7659 0.884 0.000 0.116
#> GSM25563 3 0.622 -0.2169 0.432 0.000 0.568
#> GSM25564 1 0.930 0.4561 0.512 0.292 0.196
#> GSM25565 2 0.586 0.4914 0.000 0.656 0.344
#> GSM25566 2 0.586 0.4911 0.000 0.656 0.344
#> GSM25568 2 0.941 0.0971 0.240 0.508 0.252
#> GSM25569 2 0.489 0.6448 0.000 0.772 0.228
#> GSM25552 2 0.445 0.5922 0.040 0.860 0.100
#> GSM25553 2 0.754 0.4036 0.216 0.680 0.104
#> GSM25578 1 0.245 0.7716 0.924 0.000 0.076
#> GSM25579 1 0.566 0.7398 0.772 0.028 0.200
#> GSM25580 1 0.269 0.7544 0.932 0.036 0.032
#> GSM25581 1 0.140 0.7700 0.968 0.004 0.028
#> GSM48655 2 0.375 0.6943 0.000 0.856 0.144
#> GSM48656 2 0.304 0.7006 0.000 0.896 0.104
#> GSM48657 2 0.263 0.7004 0.000 0.916 0.084
#> GSM48658 3 0.629 0.0502 0.000 0.464 0.536
#> GSM25624 1 0.231 0.7657 0.944 0.024 0.032
#> GSM25625 1 0.630 0.3691 0.520 0.000 0.480
#> GSM25626 3 0.540 0.3036 0.256 0.004 0.740
#> GSM25627 3 0.479 0.5383 0.056 0.096 0.848
#> GSM25628 3 0.414 0.5397 0.096 0.032 0.872
#> GSM25629 3 0.392 0.5190 0.004 0.140 0.856
#> GSM25630 1 0.623 0.4753 0.564 0.000 0.436
#> GSM25631 3 0.628 0.2593 0.004 0.384 0.612
#> GSM25632 1 0.603 0.5579 0.624 0.000 0.376
#> GSM25633 1 0.226 0.7707 0.932 0.000 0.068
#> GSM25634 1 0.175 0.7703 0.952 0.000 0.048
#> GSM25635 1 0.232 0.7631 0.944 0.028 0.028
#> GSM25656 3 0.347 0.5443 0.056 0.040 0.904
#> GSM25657 1 0.304 0.7660 0.896 0.000 0.104
#> GSM25658 3 0.623 -0.1756 0.436 0.000 0.564
#> GSM25659 1 0.613 0.6744 0.700 0.016 0.284
#> GSM25660 1 0.378 0.7535 0.892 0.044 0.064
#> GSM25661 1 0.132 0.7703 0.972 0.008 0.020
#> GSM25662 3 0.630 -0.0264 0.000 0.484 0.516
#> GSM25663 2 0.565 0.5306 0.000 0.688 0.312
#> GSM25680 3 0.621 0.1363 0.000 0.428 0.572
#> GSM25681 3 0.638 0.3138 0.012 0.340 0.648
#> GSM25682 2 0.355 0.6976 0.000 0.868 0.132
#> GSM25683 2 0.529 0.5984 0.000 0.732 0.268
#> GSM25684 3 0.631 -0.0977 0.000 0.500 0.500
#> GSM25685 3 0.597 0.3102 0.000 0.364 0.636
#> GSM25686 2 0.355 0.6981 0.000 0.868 0.132
#> GSM25687 2 0.304 0.7000 0.000 0.896 0.104
#> GSM48664 1 0.531 0.6930 0.816 0.136 0.048
#> GSM48665 1 0.590 0.6617 0.776 0.176 0.048
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM25548 2 0.578 0.6288 0.000 0.704 0.108 0.188
#> GSM25549 2 0.480 0.6085 0.000 0.776 0.160 0.064
#> GSM25550 2 0.169 0.6370 0.008 0.952 0.032 0.008
#> GSM25551 4 0.279 0.4931 0.004 0.088 0.012 0.896
#> GSM25570 2 0.326 0.6492 0.004 0.884 0.068 0.044
#> GSM25571 2 0.479 0.6687 0.004 0.792 0.072 0.132
#> GSM25358 1 0.609 0.2376 0.508 0.036 0.004 0.452
#> GSM25359 4 0.476 0.4817 0.000 0.064 0.156 0.780
#> GSM25360 3 0.304 0.5590 0.100 0.000 0.880 0.020
#> GSM25361 3 0.239 0.5364 0.008 0.016 0.924 0.052
#> GSM25377 1 0.435 0.7082 0.844 0.060 0.044 0.052
#> GSM25378 1 0.440 0.6727 0.812 0.112 0.000 0.076
#> GSM25401 4 0.474 0.2881 0.240 0.008 0.012 0.740
#> GSM25402 1 0.648 0.4230 0.572 0.028 0.032 0.368
#> GSM25349 2 0.609 0.5341 0.052 0.608 0.004 0.336
#> GSM25350 2 0.444 0.6811 0.020 0.788 0.008 0.184
#> GSM25356 1 0.443 0.6753 0.824 0.084 0.008 0.084
#> GSM25357 4 0.486 0.3782 0.060 0.172 0.000 0.768
#> GSM25385 1 0.690 0.3962 0.592 0.000 0.224 0.184
#> GSM25386 3 0.643 0.4159 0.072 0.008 0.612 0.308
#> GSM25399 1 0.345 0.7161 0.884 0.048 0.020 0.048
#> GSM25400 1 0.249 0.7193 0.916 0.016 0.004 0.064
#> GSM48659 4 0.785 0.1306 0.000 0.300 0.296 0.404
#> GSM48660 2 0.545 0.6233 0.012 0.676 0.020 0.292
#> GSM25409 2 0.307 0.6909 0.004 0.868 0.004 0.124
#> GSM25410 3 0.804 0.1893 0.340 0.004 0.368 0.288
#> GSM25426 4 0.111 0.5050 0.004 0.028 0.000 0.968
#> GSM25427 1 0.573 0.4211 0.564 0.412 0.012 0.012
#> GSM25540 3 0.457 0.4148 0.000 0.008 0.716 0.276
#> GSM25541 3 0.358 0.4882 0.000 0.004 0.816 0.180
#> GSM25542 4 0.659 0.4464 0.000 0.160 0.212 0.628
#> GSM25543 3 0.694 0.1179 0.000 0.128 0.540 0.332
#> GSM25479 1 0.441 0.7116 0.812 0.080 0.108 0.000
#> GSM25480 1 0.701 0.4640 0.560 0.156 0.284 0.000
#> GSM25481 1 0.648 0.3909 0.552 0.368 0.000 0.080
#> GSM25482 1 0.612 0.3127 0.516 0.436 0.000 0.048
#> GSM48654 4 0.780 0.1087 0.000 0.320 0.264 0.416
#> GSM48650 4 0.454 0.3946 0.016 0.204 0.008 0.772
#> GSM48651 4 0.693 -0.0452 0.000 0.396 0.112 0.492
#> GSM48652 4 0.688 0.1499 0.000 0.340 0.120 0.540
#> GSM48653 4 0.705 0.3846 0.000 0.196 0.232 0.572
#> GSM48662 2 0.502 0.6821 0.004 0.760 0.052 0.184
#> GSM48663 2 0.512 0.6486 0.032 0.736 0.008 0.224
#> GSM25524 3 0.496 0.4762 0.204 0.000 0.748 0.048
#> GSM25525 1 0.619 0.2873 0.516 0.052 0.432 0.000
#> GSM25526 4 0.703 -0.0749 0.364 0.000 0.128 0.508
#> GSM25527 1 0.280 0.7241 0.892 0.004 0.096 0.008
#> GSM25528 3 0.487 0.1786 0.356 0.004 0.640 0.000
#> GSM25529 1 0.578 0.2295 0.500 0.028 0.472 0.000
#> GSM25530 1 0.499 0.6249 0.744 0.000 0.208 0.048
#> GSM25531 1 0.295 0.7119 0.888 0.000 0.088 0.024
#> GSM48661 3 0.724 -0.0947 0.000 0.144 0.456 0.400
#> GSM25561 3 0.506 0.3600 0.288 0.004 0.692 0.016
#> GSM25562 1 0.470 0.6628 0.764 0.016 0.208 0.012
#> GSM25563 3 0.591 0.4727 0.208 0.000 0.688 0.104
#> GSM25564 3 0.926 0.1060 0.240 0.336 0.340 0.084
#> GSM25565 4 0.685 0.0321 0.000 0.376 0.108 0.516
#> GSM25566 4 0.628 0.0606 0.000 0.360 0.068 0.572
#> GSM25568 2 0.705 -0.0123 0.028 0.460 0.456 0.056
#> GSM25569 2 0.613 0.6193 0.000 0.668 0.116 0.216
#> GSM25552 2 0.259 0.6269 0.012 0.908 0.076 0.004
#> GSM25553 2 0.384 0.5538 0.036 0.836 0.128 0.000
#> GSM25578 1 0.370 0.6972 0.828 0.016 0.156 0.000
#> GSM25579 3 0.713 0.2501 0.256 0.152 0.584 0.008
#> GSM25580 1 0.276 0.7338 0.904 0.048 0.048 0.000
#> GSM25581 1 0.320 0.7278 0.880 0.040 0.080 0.000
#> GSM48655 2 0.506 0.4909 0.004 0.584 0.000 0.412
#> GSM48656 2 0.352 0.6933 0.000 0.856 0.032 0.112
#> GSM48657 2 0.539 0.5859 0.012 0.636 0.008 0.344
#> GSM48658 3 0.789 -0.2444 0.000 0.292 0.368 0.340
#> GSM25624 1 0.250 0.7341 0.916 0.040 0.044 0.000
#> GSM25625 1 0.737 0.2518 0.524 0.000 0.244 0.232
#> GSM25626 4 0.745 -0.2190 0.184 0.000 0.344 0.472
#> GSM25627 4 0.324 0.4168 0.052 0.000 0.068 0.880
#> GSM25628 3 0.546 0.1597 0.008 0.004 0.504 0.484
#> GSM25629 4 0.327 0.3916 0.000 0.000 0.168 0.832
#> GSM25630 3 0.443 0.5103 0.168 0.004 0.796 0.032
#> GSM25631 3 0.556 0.4032 0.004 0.236 0.704 0.056
#> GSM25632 1 0.619 0.4985 0.652 0.000 0.244 0.104
#> GSM25633 1 0.240 0.7280 0.904 0.004 0.092 0.000
#> GSM25634 1 0.187 0.7305 0.928 0.000 0.072 0.000
#> GSM25635 1 0.300 0.7347 0.896 0.064 0.036 0.004
#> GSM25656 4 0.544 -0.1136 0.004 0.008 0.456 0.532
#> GSM25657 1 0.225 0.7283 0.920 0.000 0.068 0.012
#> GSM25658 4 0.671 -0.2456 0.444 0.000 0.088 0.468
#> GSM25659 3 0.517 0.4215 0.216 0.044 0.736 0.004
#> GSM25660 1 0.692 0.5323 0.588 0.232 0.180 0.000
#> GSM25661 1 0.297 0.7317 0.892 0.036 0.072 0.000
#> GSM25662 4 0.492 0.4487 0.000 0.164 0.068 0.768
#> GSM25663 2 0.712 0.4218 0.004 0.552 0.140 0.304
#> GSM25680 3 0.771 -0.0643 0.000 0.300 0.448 0.252
#> GSM25681 3 0.624 0.3459 0.004 0.220 0.668 0.108
#> GSM25682 2 0.504 0.5141 0.000 0.592 0.004 0.404
#> GSM25683 4 0.494 0.1450 0.000 0.340 0.008 0.652
#> GSM25684 4 0.605 0.3576 0.000 0.240 0.096 0.664
#> GSM25685 4 0.247 0.5196 0.000 0.028 0.056 0.916
#> GSM25686 2 0.508 0.4831 0.000 0.576 0.004 0.420
#> GSM25687 2 0.468 0.5853 0.000 0.648 0.000 0.352
#> GSM48664 1 0.320 0.7173 0.892 0.064 0.016 0.028
#> GSM48665 1 0.252 0.7173 0.904 0.088 0.004 0.004
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM25548 5 0.497 0.48396 0.000 0.196 0.020 0.060 0.724
#> GSM25549 5 0.445 0.52730 0.000 0.196 0.040 0.012 0.752
#> GSM25550 5 0.414 0.49725 0.012 0.244 0.004 0.004 0.736
#> GSM25551 4 0.289 0.54790 0.004 0.068 0.004 0.884 0.040
#> GSM25570 5 0.351 0.53334 0.004 0.196 0.000 0.008 0.792
#> GSM25571 5 0.430 0.52362 0.004 0.188 0.004 0.040 0.764
#> GSM25358 4 0.546 0.04772 0.380 0.012 0.008 0.572 0.028
#> GSM25359 4 0.477 0.50541 0.000 0.068 0.068 0.780 0.084
#> GSM25360 3 0.444 0.59802 0.064 0.008 0.808 0.036 0.084
#> GSM25361 3 0.468 0.48734 0.004 0.004 0.716 0.040 0.236
#> GSM25377 1 0.564 0.55206 0.672 0.236 0.056 0.008 0.028
#> GSM25378 1 0.463 0.65778 0.764 0.028 0.000 0.160 0.048
#> GSM25401 4 0.403 0.50424 0.168 0.024 0.004 0.792 0.012
#> GSM25402 1 0.638 0.45711 0.592 0.080 0.024 0.288 0.016
#> GSM25349 2 0.495 0.56007 0.032 0.780 0.020 0.096 0.072
#> GSM25350 2 0.472 0.53270 0.024 0.784 0.020 0.044 0.128
#> GSM25356 1 0.412 0.68608 0.820 0.060 0.004 0.092 0.024
#> GSM25357 4 0.533 0.43196 0.044 0.216 0.012 0.704 0.024
#> GSM25385 4 0.729 -0.20178 0.360 0.000 0.228 0.384 0.028
#> GSM25386 3 0.563 0.60122 0.040 0.148 0.724 0.068 0.020
#> GSM25399 1 0.416 0.64880 0.800 0.144 0.028 0.004 0.024
#> GSM25400 1 0.320 0.69374 0.848 0.020 0.000 0.124 0.008
#> GSM48659 2 0.840 0.31514 0.000 0.304 0.144 0.284 0.268
#> GSM48660 2 0.256 0.55797 0.008 0.908 0.052 0.016 0.016
#> GSM25409 2 0.550 0.13104 0.008 0.500 0.004 0.036 0.452
#> GSM25410 3 0.754 0.52317 0.192 0.128 0.560 0.100 0.020
#> GSM25426 4 0.223 0.53583 0.000 0.104 0.000 0.892 0.004
#> GSM25427 1 0.589 0.47638 0.592 0.308 0.004 0.008 0.088
#> GSM25540 3 0.574 0.52972 0.000 0.020 0.664 0.196 0.120
#> GSM25541 3 0.602 0.46369 0.004 0.012 0.624 0.124 0.236
#> GSM25542 2 0.644 0.20361 0.000 0.520 0.336 0.128 0.016
#> GSM25543 3 0.518 0.11663 0.000 0.472 0.496 0.012 0.020
#> GSM25479 1 0.402 0.65980 0.772 0.000 0.024 0.008 0.196
#> GSM25480 5 0.656 -0.18254 0.400 0.004 0.112 0.016 0.468
#> GSM25481 1 0.676 0.41852 0.540 0.324 0.012 0.040 0.084
#> GSM25482 1 0.673 0.42773 0.552 0.280 0.008 0.024 0.136
#> GSM48654 2 0.726 0.53376 0.000 0.544 0.220 0.124 0.112
#> GSM48650 4 0.500 -0.10568 0.000 0.476 0.016 0.500 0.008
#> GSM48651 2 0.583 0.55127 0.000 0.676 0.096 0.184 0.044
#> GSM48652 2 0.595 0.51448 0.000 0.644 0.108 0.220 0.028
#> GSM48653 2 0.715 0.38320 0.000 0.488 0.188 0.284 0.040
#> GSM48662 2 0.481 0.59487 0.000 0.772 0.068 0.048 0.112
#> GSM48663 2 0.275 0.53029 0.024 0.904 0.016 0.016 0.040
#> GSM25524 3 0.624 0.54696 0.124 0.004 0.672 0.080 0.120
#> GSM25525 1 0.743 0.25992 0.424 0.000 0.244 0.040 0.292
#> GSM25526 4 0.519 0.36088 0.208 0.000 0.052 0.708 0.032
#> GSM25527 1 0.512 0.66337 0.740 0.000 0.036 0.144 0.080
#> GSM25528 3 0.592 0.43753 0.240 0.004 0.640 0.020 0.096
#> GSM25529 1 0.756 0.16802 0.368 0.000 0.284 0.040 0.308
#> GSM25530 1 0.565 0.50656 0.652 0.000 0.244 0.084 0.020
#> GSM25531 1 0.425 0.67783 0.808 0.000 0.068 0.092 0.032
#> GSM48661 3 0.739 -0.22474 0.000 0.396 0.400 0.132 0.072
#> GSM25561 3 0.498 0.57604 0.156 0.088 0.740 0.004 0.012
#> GSM25562 1 0.736 -0.02862 0.392 0.272 0.308 0.000 0.028
#> GSM25563 3 0.589 0.59551 0.092 0.156 0.700 0.024 0.028
#> GSM25564 2 0.757 -0.01392 0.128 0.496 0.288 0.012 0.076
#> GSM25565 2 0.696 0.51101 0.000 0.552 0.080 0.260 0.108
#> GSM25566 4 0.702 -0.30612 0.000 0.376 0.016 0.396 0.212
#> GSM25568 2 0.578 -0.10052 0.008 0.516 0.424 0.016 0.036
#> GSM25569 2 0.692 0.53506 0.000 0.564 0.132 0.068 0.236
#> GSM25552 5 0.335 0.54001 0.004 0.192 0.000 0.004 0.800
#> GSM25553 5 0.407 0.52155 0.012 0.204 0.012 0.004 0.768
#> GSM25578 1 0.542 0.62522 0.700 0.000 0.096 0.024 0.180
#> GSM25579 5 0.615 0.26767 0.120 0.000 0.220 0.032 0.628
#> GSM25580 1 0.242 0.70703 0.912 0.036 0.016 0.000 0.036
#> GSM25581 1 0.346 0.70012 0.844 0.004 0.040 0.004 0.108
#> GSM48655 2 0.563 0.56121 0.000 0.660 0.008 0.188 0.144
#> GSM48656 2 0.507 0.45585 0.000 0.668 0.028 0.024 0.280
#> GSM48657 2 0.473 0.57655 0.004 0.744 0.000 0.144 0.108
#> GSM48658 2 0.842 0.33517 0.000 0.332 0.232 0.164 0.272
#> GSM25624 1 0.362 0.69996 0.836 0.004 0.016 0.024 0.120
#> GSM25625 1 0.726 0.10546 0.372 0.000 0.240 0.364 0.024
#> GSM25626 3 0.712 0.26257 0.120 0.056 0.420 0.404 0.000
#> GSM25627 4 0.190 0.55781 0.020 0.028 0.016 0.936 0.000
#> GSM25628 3 0.585 0.47682 0.004 0.104 0.604 0.284 0.004
#> GSM25629 4 0.241 0.52200 0.000 0.008 0.056 0.908 0.028
#> GSM25630 3 0.487 0.59723 0.072 0.148 0.756 0.004 0.020
#> GSM25631 5 0.606 0.17180 0.000 0.064 0.400 0.024 0.512
#> GSM25632 1 0.664 0.19256 0.492 0.000 0.356 0.128 0.024
#> GSM25633 1 0.318 0.69605 0.872 0.000 0.068 0.024 0.036
#> GSM25634 1 0.316 0.69073 0.872 0.024 0.084 0.008 0.012
#> GSM25635 1 0.427 0.68695 0.792 0.024 0.000 0.044 0.140
#> GSM25656 3 0.632 0.44761 0.004 0.200 0.588 0.200 0.008
#> GSM25657 1 0.336 0.68664 0.856 0.016 0.100 0.024 0.004
#> GSM25658 4 0.520 0.22247 0.296 0.000 0.040 0.648 0.016
#> GSM25659 3 0.681 0.47180 0.124 0.020 0.600 0.036 0.220
#> GSM25660 5 0.534 -0.15736 0.424 0.004 0.028 0.008 0.536
#> GSM25661 1 0.314 0.70178 0.856 0.000 0.032 0.004 0.108
#> GSM25662 4 0.621 0.12653 0.000 0.300 0.040 0.584 0.076
#> GSM25663 5 0.692 -0.06708 0.000 0.324 0.020 0.188 0.468
#> GSM25680 5 0.746 0.33416 0.000 0.120 0.200 0.152 0.528
#> GSM25681 5 0.578 0.34306 0.000 0.044 0.292 0.044 0.620
#> GSM25682 2 0.665 0.43719 0.000 0.484 0.004 0.256 0.256
#> GSM25683 4 0.592 -0.00532 0.000 0.364 0.008 0.540 0.088
#> GSM25684 4 0.695 0.02221 0.000 0.272 0.044 0.528 0.156
#> GSM25685 4 0.334 0.52547 0.000 0.104 0.028 0.852 0.016
#> GSM25686 2 0.674 0.41385 0.000 0.460 0.004 0.284 0.252
#> GSM25687 2 0.662 0.39213 0.000 0.484 0.004 0.220 0.292
#> GSM48664 1 0.360 0.66327 0.824 0.144 0.012 0.004 0.016
#> GSM48665 1 0.307 0.69429 0.872 0.088 0.004 0.008 0.028
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM25548 6 0.370 0.60562 0.000 0.128 0.000 0.020 0.048 0.804
#> GSM25549 6 0.337 0.63940 0.000 0.080 0.012 0.036 0.024 0.848
#> GSM25550 6 0.308 0.62292 0.012 0.100 0.004 0.032 0.000 0.852
#> GSM25551 5 0.249 0.56077 0.000 0.068 0.000 0.008 0.888 0.036
#> GSM25570 6 0.262 0.64553 0.004 0.092 0.000 0.024 0.004 0.876
#> GSM25571 6 0.316 0.64097 0.000 0.092 0.000 0.020 0.040 0.848
#> GSM25358 5 0.738 0.21777 0.248 0.012 0.136 0.064 0.500 0.040
#> GSM25359 5 0.750 0.21609 0.008 0.048 0.220 0.072 0.496 0.156
#> GSM25360 3 0.558 0.27995 0.032 0.012 0.692 0.172 0.032 0.060
#> GSM25361 4 0.711 0.31375 0.012 0.040 0.308 0.460 0.020 0.160
#> GSM25377 1 0.708 0.39520 0.492 0.224 0.132 0.144 0.008 0.000
#> GSM25378 1 0.604 0.57878 0.656 0.012 0.020 0.076 0.172 0.064
#> GSM25401 5 0.441 0.47994 0.192 0.036 0.012 0.020 0.740 0.000
#> GSM25402 1 0.726 0.29149 0.460 0.064 0.080 0.080 0.316 0.000
#> GSM25349 2 0.652 0.41606 0.008 0.628 0.068 0.072 0.060 0.164
#> GSM25350 2 0.710 0.29299 0.008 0.516 0.112 0.076 0.028 0.260
#> GSM25356 1 0.642 0.58128 0.648 0.024 0.052 0.104 0.136 0.036
#> GSM25357 5 0.642 0.41494 0.016 0.188 0.044 0.080 0.628 0.044
#> GSM25385 3 0.696 0.24640 0.244 0.004 0.420 0.056 0.276 0.000
#> GSM25386 3 0.339 0.47447 0.012 0.040 0.852 0.020 0.072 0.004
#> GSM25399 1 0.611 0.54693 0.636 0.148 0.084 0.116 0.016 0.000
#> GSM25400 1 0.439 0.61329 0.764 0.000 0.032 0.052 0.144 0.008
#> GSM48659 2 0.839 0.15846 0.000 0.304 0.060 0.228 0.256 0.152
#> GSM48660 2 0.342 0.49240 0.000 0.840 0.056 0.064 0.000 0.040
#> GSM25409 6 0.523 0.18730 0.004 0.384 0.000 0.040 0.024 0.548
#> GSM25410 3 0.539 0.44728 0.076 0.028 0.724 0.072 0.096 0.004
#> GSM25426 5 0.229 0.56064 0.000 0.076 0.000 0.020 0.896 0.008
#> GSM25427 1 0.844 0.35810 0.404 0.200 0.092 0.152 0.008 0.144
#> GSM25540 3 0.774 -0.08608 0.000 0.080 0.380 0.304 0.192 0.044
#> GSM25541 4 0.786 0.19385 0.012 0.052 0.304 0.420 0.108 0.104
#> GSM25542 3 0.654 0.29550 0.000 0.300 0.532 0.060 0.076 0.032
#> GSM25543 3 0.508 0.38913 0.000 0.244 0.668 0.052 0.012 0.024
#> GSM25479 1 0.561 0.58036 0.684 0.008 0.040 0.112 0.012 0.144
#> GSM25480 1 0.704 0.21855 0.404 0.004 0.036 0.152 0.028 0.376
#> GSM25481 1 0.877 0.16001 0.328 0.308 0.080 0.128 0.040 0.116
#> GSM25482 1 0.858 0.31937 0.396 0.216 0.060 0.120 0.032 0.176
#> GSM48654 2 0.728 0.42750 0.000 0.544 0.136 0.128 0.128 0.064
#> GSM48650 2 0.483 0.12360 0.000 0.524 0.012 0.024 0.436 0.004
#> GSM48651 2 0.473 0.50086 0.000 0.756 0.036 0.052 0.128 0.028
#> GSM48652 2 0.494 0.47768 0.000 0.724 0.028 0.064 0.164 0.020
#> GSM48653 2 0.685 0.34702 0.000 0.540 0.080 0.160 0.200 0.020
#> GSM48662 2 0.421 0.50563 0.000 0.792 0.020 0.056 0.024 0.108
#> GSM48663 2 0.465 0.45848 0.008 0.768 0.056 0.096 0.004 0.068
#> GSM25524 4 0.672 0.21898 0.152 0.004 0.332 0.464 0.036 0.012
#> GSM25525 1 0.693 0.18112 0.444 0.000 0.040 0.312 0.020 0.184
#> GSM25526 5 0.502 0.36789 0.236 0.000 0.020 0.072 0.668 0.004
#> GSM25527 1 0.408 0.61621 0.796 0.000 0.012 0.112 0.056 0.024
#> GSM25528 4 0.688 0.11271 0.284 0.000 0.332 0.348 0.012 0.024
#> GSM25529 1 0.709 0.21581 0.456 0.000 0.052 0.284 0.024 0.184
#> GSM25530 1 0.522 0.53520 0.684 0.000 0.144 0.132 0.040 0.000
#> GSM25531 1 0.343 0.62724 0.840 0.000 0.024 0.080 0.052 0.004
#> GSM48661 2 0.815 -0.00886 0.000 0.340 0.228 0.252 0.136 0.044
#> GSM25561 3 0.521 0.38459 0.072 0.060 0.708 0.152 0.004 0.004
#> GSM25562 2 0.796 -0.20798 0.260 0.272 0.260 0.196 0.012 0.000
#> GSM25563 3 0.494 0.40805 0.036 0.096 0.736 0.120 0.008 0.004
#> GSM25564 2 0.815 0.02975 0.096 0.432 0.144 0.252 0.028 0.048
#> GSM25565 2 0.848 0.28821 0.004 0.360 0.148 0.076 0.216 0.196
#> GSM25566 5 0.697 -0.16629 0.000 0.296 0.024 0.016 0.340 0.324
#> GSM25568 2 0.709 0.01647 0.024 0.440 0.348 0.140 0.016 0.032
#> GSM25569 2 0.656 0.39305 0.000 0.580 0.108 0.060 0.036 0.216
#> GSM25552 6 0.208 0.62622 0.008 0.056 0.000 0.024 0.000 0.912
#> GSM25553 6 0.299 0.59697 0.024 0.060 0.008 0.036 0.000 0.872
#> GSM25578 1 0.462 0.59549 0.764 0.000 0.032 0.072 0.020 0.112
#> GSM25579 6 0.655 -0.02615 0.204 0.000 0.032 0.240 0.012 0.512
#> GSM25580 1 0.423 0.64238 0.808 0.036 0.032 0.076 0.004 0.044
#> GSM25581 1 0.384 0.63537 0.824 0.016 0.036 0.052 0.000 0.072
#> GSM48655 2 0.606 0.39879 0.000 0.592 0.012 0.028 0.168 0.200
#> GSM48656 2 0.463 0.41955 0.000 0.708 0.004 0.052 0.020 0.216
#> GSM48657 2 0.515 0.48798 0.000 0.704 0.004 0.048 0.148 0.096
#> GSM48658 4 0.840 -0.11058 0.000 0.300 0.092 0.316 0.136 0.156
#> GSM25624 1 0.456 0.63488 0.772 0.004 0.064 0.048 0.008 0.104
#> GSM25625 1 0.703 0.02500 0.372 0.000 0.192 0.084 0.352 0.000
#> GSM25626 3 0.634 0.36334 0.088 0.020 0.556 0.056 0.280 0.000
#> GSM25627 5 0.313 0.54526 0.056 0.028 0.016 0.032 0.868 0.000
#> GSM25628 3 0.682 0.30869 0.008 0.080 0.516 0.132 0.260 0.004
#> GSM25629 5 0.349 0.53705 0.012 0.044 0.016 0.092 0.836 0.000
#> GSM25630 3 0.518 0.33137 0.024 0.112 0.664 0.200 0.000 0.000
#> GSM25631 4 0.689 0.26287 0.004 0.064 0.132 0.428 0.008 0.364
#> GSM25632 3 0.609 0.16647 0.340 0.000 0.512 0.056 0.092 0.000
#> GSM25633 1 0.316 0.63109 0.860 0.004 0.056 0.064 0.008 0.008
#> GSM25634 1 0.486 0.61439 0.732 0.016 0.144 0.092 0.008 0.008
#> GSM25635 1 0.526 0.61444 0.716 0.008 0.032 0.064 0.024 0.156
#> GSM25656 3 0.710 0.30037 0.000 0.156 0.508 0.188 0.136 0.012
#> GSM25657 1 0.450 0.61915 0.760 0.032 0.096 0.108 0.004 0.000
#> GSM25658 5 0.595 0.25197 0.284 0.008 0.016 0.124 0.564 0.004
#> GSM25659 4 0.765 0.34146 0.176 0.088 0.104 0.532 0.020 0.080
#> GSM25660 1 0.577 0.33384 0.488 0.000 0.016 0.116 0.000 0.380
#> GSM25661 1 0.361 0.63441 0.832 0.012 0.020 0.052 0.000 0.084
#> GSM25662 5 0.555 0.22139 0.000 0.332 0.012 0.044 0.576 0.036
#> GSM25663 6 0.655 0.27430 0.000 0.268 0.016 0.044 0.144 0.528
#> GSM25680 6 0.745 0.30736 0.000 0.096 0.056 0.192 0.152 0.504
#> GSM25681 6 0.727 0.16823 0.012 0.048 0.176 0.180 0.052 0.532
#> GSM25682 2 0.666 0.03394 0.000 0.380 0.012 0.020 0.208 0.380
#> GSM25683 5 0.664 0.04298 0.000 0.336 0.024 0.028 0.468 0.144
#> GSM25684 5 0.629 0.13686 0.000 0.328 0.012 0.052 0.520 0.088
#> GSM25685 5 0.304 0.53047 0.000 0.128 0.000 0.032 0.836 0.004
#> GSM25686 2 0.675 0.05539 0.000 0.392 0.012 0.028 0.200 0.368
#> GSM25687 6 0.659 -0.09406 0.000 0.400 0.016 0.028 0.148 0.408
#> GSM48664 1 0.643 0.55491 0.620 0.128 0.072 0.152 0.020 0.008
#> GSM48665 1 0.443 0.63704 0.792 0.024 0.024 0.104 0.016 0.040
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 genotype/variation(p) k
#> CV:NMF 96 9.23e-05 2
#> CV:NMF 62 2.60e-04 3
#> CV:NMF 45 4.06e-02 4
#> CV:NMF 52 1.00e-02 5
#> CV:NMF 33 1.18e-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["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 100 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 0.0844 0.711 0.802 0.4339 0.502 0.502
#> 3 3 0.1149 0.657 0.736 0.2112 0.937 0.879
#> 4 4 0.1875 0.566 0.700 0.1400 0.958 0.914
#> 5 5 0.2282 0.506 0.680 0.0792 0.957 0.906
#> 6 6 0.3097 0.358 0.660 0.0691 0.916 0.808
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
#> GSM25548 2 0.767 0.7119 0.224 0.776
#> GSM25549 2 0.767 0.7061 0.224 0.776
#> GSM25550 2 0.795 0.7039 0.240 0.760
#> GSM25551 2 0.482 0.7969 0.104 0.896
#> GSM25570 2 0.767 0.7061 0.224 0.776
#> GSM25571 2 0.767 0.7061 0.224 0.776
#> GSM25358 2 0.844 0.6565 0.272 0.728
#> GSM25359 2 0.844 0.6565 0.272 0.728
#> GSM25360 1 0.998 0.4079 0.524 0.476
#> GSM25361 1 0.998 0.4079 0.524 0.476
#> GSM25377 1 0.373 0.6920 0.928 0.072
#> GSM25378 1 1.000 0.0528 0.504 0.496
#> GSM25401 2 0.753 0.6801 0.216 0.784
#> GSM25402 2 0.844 0.6025 0.272 0.728
#> GSM25349 2 0.358 0.7745 0.068 0.932
#> GSM25350 2 0.358 0.7745 0.068 0.932
#> GSM25356 2 0.753 0.6894 0.216 0.784
#> GSM25357 2 0.753 0.6894 0.216 0.784
#> GSM25385 1 0.827 0.7912 0.740 0.260
#> GSM25386 1 0.891 0.7455 0.692 0.308
#> GSM25399 1 0.311 0.6813 0.944 0.056
#> GSM25400 1 0.886 0.6545 0.696 0.304
#> GSM48659 2 0.358 0.8010 0.068 0.932
#> GSM48660 2 0.278 0.7880 0.048 0.952
#> GSM25409 2 0.644 0.7884 0.164 0.836
#> GSM25410 1 0.895 0.7405 0.688 0.312
#> GSM25426 2 0.456 0.7884 0.096 0.904
#> GSM25427 2 0.992 0.1762 0.448 0.552
#> GSM25540 2 0.781 0.6894 0.232 0.768
#> GSM25541 2 0.781 0.6894 0.232 0.768
#> GSM25542 2 0.680 0.7373 0.180 0.820
#> GSM25543 2 0.689 0.7339 0.184 0.816
#> GSM25479 1 0.760 0.8091 0.780 0.220
#> GSM25480 1 0.760 0.8091 0.780 0.220
#> GSM25481 2 0.850 0.5912 0.276 0.724
#> GSM25482 2 0.850 0.5912 0.276 0.724
#> GSM48654 2 0.416 0.8033 0.084 0.916
#> GSM48650 2 0.343 0.7778 0.064 0.936
#> GSM48651 2 0.204 0.7899 0.032 0.968
#> GSM48652 2 0.224 0.7901 0.036 0.964
#> GSM48653 2 0.184 0.7930 0.028 0.972
#> GSM48662 2 0.242 0.7946 0.040 0.960
#> GSM48663 2 0.373 0.7751 0.072 0.928
#> GSM25524 1 0.730 0.7874 0.796 0.204
#> GSM25525 1 0.775 0.8098 0.772 0.228
#> GSM25526 1 0.978 0.5376 0.588 0.412
#> GSM25527 1 0.745 0.8075 0.788 0.212
#> GSM25528 1 0.745 0.7932 0.788 0.212
#> GSM25529 1 0.760 0.8125 0.780 0.220
#> GSM25530 1 0.529 0.7643 0.880 0.120
#> GSM25531 1 0.584 0.7946 0.860 0.140
#> GSM48661 2 0.430 0.7995 0.088 0.912
#> GSM25561 1 0.861 0.7811 0.716 0.284
#> GSM25562 1 0.909 0.7305 0.676 0.324
#> GSM25563 1 0.850 0.7799 0.724 0.276
#> GSM25564 1 0.997 0.3534 0.532 0.468
#> GSM25565 2 0.506 0.7893 0.112 0.888
#> GSM25566 2 0.469 0.8014 0.100 0.900
#> GSM25568 2 0.311 0.7994 0.056 0.944
#> GSM25569 2 0.295 0.8001 0.052 0.948
#> GSM25552 2 0.802 0.6905 0.244 0.756
#> GSM25553 2 0.802 0.6905 0.244 0.756
#> GSM25578 1 0.680 0.8102 0.820 0.180
#> GSM25579 1 0.788 0.7988 0.764 0.236
#> GSM25580 1 0.615 0.7917 0.848 0.152
#> GSM25581 1 0.615 0.7917 0.848 0.152
#> GSM48655 2 0.278 0.7872 0.048 0.952
#> GSM48656 2 0.482 0.8016 0.104 0.896
#> GSM48657 2 0.343 0.7778 0.064 0.936
#> GSM48658 2 0.443 0.7975 0.092 0.908
#> GSM25624 1 0.808 0.7516 0.752 0.248
#> GSM25625 1 0.900 0.7383 0.684 0.316
#> GSM25626 1 0.881 0.7538 0.700 0.300
#> GSM25627 2 1.000 -0.2993 0.500 0.500
#> GSM25628 1 0.886 0.7500 0.696 0.304
#> GSM25629 2 0.952 0.2575 0.372 0.628
#> GSM25630 1 0.760 0.7884 0.780 0.220
#> GSM25631 2 0.753 0.7164 0.216 0.784
#> GSM25632 1 0.760 0.8137 0.780 0.220
#> GSM25633 1 0.653 0.8084 0.832 0.168
#> GSM25634 1 0.662 0.8051 0.828 0.172
#> GSM25635 1 0.680 0.8002 0.820 0.180
#> GSM25656 1 0.827 0.7700 0.740 0.260
#> GSM25657 1 0.563 0.7965 0.868 0.132
#> GSM25658 1 0.929 0.6713 0.656 0.344
#> GSM25659 1 0.985 0.4908 0.572 0.428
#> GSM25660 1 0.697 0.8082 0.812 0.188
#> GSM25661 1 0.625 0.8028 0.844 0.156
#> GSM25662 2 0.722 0.7390 0.200 0.800
#> GSM25663 2 0.722 0.7390 0.200 0.800
#> GSM25680 2 0.767 0.6940 0.224 0.776
#> GSM25681 2 0.767 0.6940 0.224 0.776
#> GSM25682 2 0.327 0.7904 0.060 0.940
#> GSM25683 2 0.327 0.7904 0.060 0.940
#> GSM25684 2 0.343 0.8014 0.064 0.936
#> GSM25685 2 0.358 0.8021 0.068 0.932
#> GSM25686 2 0.327 0.7904 0.060 0.940
#> GSM25687 2 0.327 0.7904 0.060 0.940
#> GSM48664 1 0.311 0.6813 0.944 0.056
#> GSM48665 1 0.563 0.7845 0.868 0.132
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM25548 2 0.568 0.7103 0.212 0.764 0.024
#> GSM25549 2 0.576 0.7043 0.208 0.764 0.028
#> GSM25550 2 0.606 0.7019 0.224 0.744 0.032
#> GSM25551 2 0.482 0.7771 0.088 0.848 0.064
#> GSM25570 2 0.580 0.7030 0.212 0.760 0.028
#> GSM25571 2 0.580 0.7030 0.212 0.760 0.028
#> GSM25358 2 0.710 0.6327 0.240 0.692 0.068
#> GSM25359 2 0.710 0.6327 0.240 0.692 0.068
#> GSM25360 1 0.906 0.3329 0.452 0.412 0.136
#> GSM25361 1 0.910 0.3435 0.452 0.408 0.140
#> GSM25377 3 0.706 0.9569 0.352 0.032 0.616
#> GSM25378 2 0.858 -0.0763 0.444 0.460 0.096
#> GSM25401 2 0.718 0.6077 0.184 0.712 0.104
#> GSM25402 2 0.780 0.5318 0.216 0.664 0.120
#> GSM25349 2 0.315 0.7699 0.040 0.916 0.044
#> GSM25350 2 0.315 0.7699 0.040 0.916 0.044
#> GSM25356 2 0.651 0.6685 0.156 0.756 0.088
#> GSM25357 2 0.651 0.6685 0.156 0.756 0.088
#> GSM25385 1 0.803 0.6601 0.656 0.176 0.168
#> GSM25386 1 0.895 0.6199 0.568 0.220 0.212
#> GSM25399 3 0.678 0.9739 0.364 0.020 0.616
#> GSM25400 1 0.811 0.4725 0.624 0.264 0.112
#> GSM48659 2 0.309 0.7927 0.072 0.912 0.016
#> GSM48660 2 0.206 0.7810 0.024 0.952 0.024
#> GSM25409 2 0.475 0.7853 0.144 0.832 0.024
#> GSM25410 1 0.891 0.6224 0.572 0.220 0.208
#> GSM25426 2 0.457 0.7608 0.072 0.860 0.068
#> GSM25427 2 0.870 0.2019 0.360 0.524 0.116
#> GSM25540 2 0.635 0.6908 0.212 0.740 0.048
#> GSM25541 2 0.635 0.6908 0.212 0.740 0.048
#> GSM25542 2 0.547 0.7348 0.168 0.796 0.036
#> GSM25543 2 0.541 0.7329 0.172 0.796 0.032
#> GSM25479 1 0.560 0.6597 0.800 0.148 0.052
#> GSM25480 1 0.560 0.6597 0.800 0.148 0.052
#> GSM25481 2 0.732 0.5810 0.184 0.704 0.112
#> GSM25482 2 0.732 0.5810 0.184 0.704 0.112
#> GSM48654 2 0.344 0.7962 0.088 0.896 0.016
#> GSM48650 2 0.269 0.7707 0.032 0.932 0.036
#> GSM48651 2 0.164 0.7836 0.020 0.964 0.016
#> GSM48652 2 0.178 0.7833 0.020 0.960 0.020
#> GSM48653 2 0.205 0.7881 0.028 0.952 0.020
#> GSM48662 2 0.218 0.7885 0.032 0.948 0.020
#> GSM48663 2 0.304 0.7700 0.040 0.920 0.040
#> GSM25524 1 0.701 0.5214 0.652 0.040 0.308
#> GSM25525 1 0.589 0.6525 0.796 0.104 0.100
#> GSM25526 1 0.865 0.5367 0.556 0.320 0.124
#> GSM25527 1 0.658 0.6433 0.756 0.136 0.108
#> GSM25528 1 0.709 0.5587 0.676 0.056 0.268
#> GSM25529 1 0.596 0.6437 0.792 0.096 0.112
#> GSM25530 1 0.634 0.5307 0.716 0.032 0.252
#> GSM25531 1 0.567 0.5993 0.800 0.060 0.140
#> GSM48661 2 0.367 0.7926 0.092 0.888 0.020
#> GSM25561 1 0.876 0.6498 0.588 0.196 0.216
#> GSM25562 1 0.813 0.6510 0.632 0.244 0.124
#> GSM25563 1 0.849 0.5880 0.592 0.132 0.276
#> GSM25564 1 0.877 0.3721 0.500 0.384 0.116
#> GSM25565 2 0.420 0.7841 0.112 0.864 0.024
#> GSM25566 2 0.393 0.7955 0.092 0.880 0.028
#> GSM25568 2 0.378 0.7966 0.064 0.892 0.044
#> GSM25569 2 0.308 0.7977 0.060 0.916 0.024
#> GSM25552 2 0.614 0.6863 0.232 0.736 0.032
#> GSM25553 2 0.614 0.6863 0.232 0.736 0.032
#> GSM25578 1 0.542 0.6255 0.820 0.100 0.080
#> GSM25579 1 0.632 0.6487 0.764 0.160 0.076
#> GSM25580 1 0.604 0.5610 0.788 0.100 0.112
#> GSM25581 1 0.604 0.5610 0.788 0.100 0.112
#> GSM48655 2 0.192 0.7822 0.024 0.956 0.020
#> GSM48656 2 0.385 0.7945 0.108 0.876 0.016
#> GSM48657 2 0.256 0.7737 0.036 0.936 0.028
#> GSM48658 2 0.369 0.7909 0.100 0.884 0.016
#> GSM25624 1 0.679 0.5777 0.728 0.196 0.076
#> GSM25625 1 0.834 0.6552 0.620 0.236 0.144
#> GSM25626 1 0.891 0.6238 0.572 0.204 0.224
#> GSM25627 1 0.874 0.2758 0.464 0.428 0.108
#> GSM25628 1 0.884 0.6201 0.580 0.204 0.216
#> GSM25629 2 0.805 0.2335 0.340 0.580 0.080
#> GSM25630 1 0.769 0.4828 0.596 0.060 0.344
#> GSM25631 2 0.580 0.7095 0.212 0.760 0.028
#> GSM25632 1 0.639 0.6610 0.768 0.120 0.112
#> GSM25633 1 0.524 0.6120 0.828 0.100 0.072
#> GSM25634 1 0.625 0.6116 0.776 0.116 0.108
#> GSM25635 1 0.586 0.6010 0.796 0.120 0.084
#> GSM25656 1 0.880 0.5288 0.564 0.152 0.284
#> GSM25657 1 0.561 0.5768 0.808 0.072 0.120
#> GSM25658 1 0.823 0.6101 0.620 0.256 0.124
#> GSM25659 1 0.807 0.4898 0.564 0.360 0.076
#> GSM25660 1 0.571 0.6211 0.804 0.116 0.080
#> GSM25661 1 0.490 0.5975 0.844 0.092 0.064
#> GSM25662 2 0.544 0.7386 0.192 0.784 0.024
#> GSM25663 2 0.544 0.7386 0.192 0.784 0.024
#> GSM25680 2 0.620 0.6978 0.208 0.748 0.044
#> GSM25681 2 0.630 0.6903 0.208 0.744 0.048
#> GSM25682 2 0.231 0.7846 0.032 0.944 0.024
#> GSM25683 2 0.231 0.7846 0.032 0.944 0.024
#> GSM25684 2 0.300 0.7930 0.068 0.916 0.016
#> GSM25685 2 0.371 0.7920 0.076 0.892 0.032
#> GSM25686 2 0.231 0.7846 0.032 0.944 0.024
#> GSM25687 2 0.231 0.7846 0.032 0.944 0.024
#> GSM48664 3 0.678 0.9751 0.364 0.020 0.616
#> GSM48665 1 0.610 0.5611 0.784 0.096 0.120
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM25548 2 0.591 0.6562 0.236 0.696 0.044 0.024
#> GSM25549 2 0.617 0.6422 0.244 0.680 0.040 0.036
#> GSM25550 2 0.641 0.6408 0.244 0.668 0.044 0.044
#> GSM25551 2 0.566 0.7362 0.092 0.768 0.096 0.044
#> GSM25570 2 0.625 0.6404 0.244 0.676 0.044 0.036
#> GSM25571 2 0.625 0.6404 0.244 0.676 0.044 0.036
#> GSM25358 2 0.716 0.5851 0.228 0.616 0.132 0.024
#> GSM25359 2 0.716 0.5851 0.228 0.616 0.132 0.024
#> GSM25360 1 0.876 0.2824 0.444 0.300 0.192 0.064
#> GSM25361 1 0.877 0.2825 0.444 0.296 0.196 0.064
#> GSM25377 4 0.507 0.9213 0.308 0.012 0.004 0.676
#> GSM25378 1 0.738 0.1580 0.492 0.384 0.016 0.108
#> GSM25401 2 0.776 0.5116 0.196 0.608 0.112 0.084
#> GSM25402 2 0.804 0.4561 0.220 0.576 0.088 0.116
#> GSM25349 2 0.381 0.7213 0.016 0.864 0.048 0.072
#> GSM25350 2 0.381 0.7213 0.016 0.864 0.048 0.072
#> GSM25356 2 0.738 0.5648 0.148 0.640 0.060 0.152
#> GSM25357 2 0.738 0.5648 0.148 0.640 0.060 0.152
#> GSM25385 1 0.764 0.2219 0.536 0.108 0.320 0.036
#> GSM25386 1 0.758 0.0645 0.432 0.136 0.420 0.012
#> GSM25399 4 0.443 0.9464 0.304 0.000 0.000 0.696
#> GSM25400 1 0.665 0.4759 0.676 0.192 0.032 0.100
#> GSM48659 2 0.366 0.7604 0.064 0.872 0.048 0.016
#> GSM48660 2 0.247 0.7451 0.008 0.920 0.016 0.056
#> GSM25409 2 0.524 0.7494 0.144 0.776 0.024 0.056
#> GSM25410 1 0.751 0.0819 0.436 0.140 0.416 0.008
#> GSM25426 2 0.568 0.7209 0.076 0.768 0.104 0.052
#> GSM25427 2 0.811 0.0658 0.392 0.432 0.036 0.140
#> GSM25540 2 0.647 0.6433 0.204 0.668 0.116 0.012
#> GSM25541 2 0.647 0.6433 0.204 0.668 0.116 0.012
#> GSM25542 2 0.599 0.7018 0.148 0.732 0.092 0.028
#> GSM25543 2 0.607 0.6989 0.152 0.728 0.088 0.032
#> GSM25479 1 0.473 0.5964 0.824 0.064 0.044 0.068
#> GSM25480 1 0.473 0.5964 0.824 0.064 0.044 0.068
#> GSM25481 2 0.727 0.5142 0.216 0.624 0.040 0.120
#> GSM25482 2 0.727 0.5142 0.216 0.624 0.040 0.120
#> GSM48654 2 0.374 0.7631 0.076 0.868 0.028 0.028
#> GSM48650 2 0.289 0.7306 0.008 0.900 0.020 0.072
#> GSM48651 2 0.195 0.7509 0.012 0.940 0.004 0.044
#> GSM48652 2 0.197 0.7491 0.008 0.940 0.008 0.044
#> GSM48653 2 0.262 0.7555 0.016 0.920 0.028 0.036
#> GSM48662 2 0.293 0.7544 0.024 0.908 0.028 0.040
#> GSM48663 2 0.390 0.7214 0.020 0.860 0.044 0.076
#> GSM25524 1 0.669 0.0666 0.532 0.004 0.384 0.080
#> GSM25525 1 0.512 0.5476 0.780 0.032 0.152 0.036
#> GSM25526 1 0.784 0.4529 0.580 0.236 0.120 0.064
#> GSM25527 1 0.503 0.5956 0.808 0.068 0.048 0.076
#> GSM25528 1 0.647 0.1707 0.576 0.004 0.348 0.072
#> GSM25529 1 0.455 0.5520 0.808 0.024 0.144 0.024
#> GSM25530 1 0.657 0.2364 0.604 0.000 0.280 0.116
#> GSM25531 1 0.559 0.4703 0.740 0.012 0.172 0.076
#> GSM48661 2 0.388 0.7573 0.096 0.852 0.044 0.008
#> GSM25561 1 0.775 0.2457 0.552 0.076 0.300 0.072
#> GSM25562 1 0.696 0.5282 0.680 0.140 0.112 0.068
#> GSM25563 3 0.734 -0.1549 0.440 0.052 0.460 0.048
#> GSM25564 1 0.839 0.3765 0.500 0.288 0.148 0.064
#> GSM25565 2 0.470 0.7495 0.116 0.812 0.052 0.020
#> GSM25566 2 0.425 0.7624 0.104 0.836 0.044 0.016
#> GSM25568 2 0.579 0.7452 0.092 0.764 0.084 0.060
#> GSM25569 2 0.509 0.7581 0.084 0.804 0.060 0.052
#> GSM25552 2 0.667 0.6175 0.260 0.644 0.048 0.048
#> GSM25553 2 0.667 0.6175 0.260 0.644 0.048 0.048
#> GSM25578 1 0.393 0.5861 0.864 0.040 0.056 0.040
#> GSM25579 1 0.506 0.5982 0.804 0.092 0.056 0.048
#> GSM25580 1 0.333 0.5702 0.876 0.032 0.004 0.088
#> GSM25581 1 0.333 0.5702 0.876 0.032 0.004 0.088
#> GSM48655 2 0.222 0.7465 0.008 0.928 0.008 0.056
#> GSM48656 2 0.403 0.7614 0.100 0.848 0.032 0.020
#> GSM48657 2 0.295 0.7330 0.012 0.900 0.020 0.068
#> GSM48658 2 0.402 0.7553 0.104 0.840 0.052 0.004
#> GSM25624 1 0.498 0.5622 0.792 0.128 0.016 0.064
#> GSM25625 1 0.746 0.4817 0.608 0.160 0.196 0.036
#> GSM25626 1 0.767 0.0782 0.440 0.124 0.416 0.020
#> GSM25627 1 0.830 0.3286 0.480 0.332 0.128 0.060
#> GSM25628 1 0.744 0.0415 0.436 0.120 0.432 0.012
#> GSM25629 2 0.802 0.1956 0.344 0.492 0.116 0.048
#> GSM25630 3 0.491 0.4352 0.116 0.012 0.796 0.076
#> GSM25631 2 0.638 0.6534 0.236 0.672 0.064 0.028
#> GSM25632 1 0.552 0.5390 0.756 0.036 0.164 0.044
#> GSM25633 1 0.347 0.5813 0.884 0.024 0.040 0.052
#> GSM25634 1 0.374 0.5812 0.864 0.028 0.020 0.088
#> GSM25635 1 0.346 0.5887 0.880 0.048 0.012 0.060
#> GSM25656 3 0.649 0.4139 0.092 0.088 0.720 0.100
#> GSM25657 1 0.448 0.5671 0.828 0.020 0.056 0.096
#> GSM25658 1 0.755 0.5112 0.628 0.180 0.116 0.076
#> GSM25659 1 0.753 0.4529 0.588 0.260 0.104 0.048
#> GSM25660 1 0.416 0.5987 0.852 0.052 0.032 0.064
#> GSM25661 1 0.319 0.5810 0.896 0.024 0.028 0.052
#> GSM25662 2 0.565 0.7110 0.188 0.736 0.052 0.024
#> GSM25663 2 0.565 0.7110 0.188 0.736 0.052 0.024
#> GSM25680 2 0.678 0.6384 0.228 0.656 0.072 0.044
#> GSM25681 2 0.699 0.6192 0.236 0.640 0.072 0.052
#> GSM25682 2 0.255 0.7487 0.016 0.916 0.008 0.060
#> GSM25683 2 0.255 0.7487 0.016 0.916 0.008 0.060
#> GSM25684 2 0.358 0.7607 0.060 0.876 0.048 0.016
#> GSM25685 2 0.431 0.7595 0.068 0.840 0.072 0.020
#> GSM25686 2 0.255 0.7487 0.016 0.916 0.008 0.060
#> GSM25687 2 0.255 0.7487 0.016 0.916 0.008 0.060
#> GSM48664 4 0.450 0.9510 0.316 0.000 0.000 0.684
#> GSM48665 1 0.389 0.5739 0.852 0.036 0.012 0.100
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM25548 2 0.614 0.6144 0.188 0.656 0.080 0.076 0.000
#> GSM25549 2 0.636 0.5981 0.196 0.636 0.084 0.084 0.000
#> GSM25550 2 0.652 0.5951 0.192 0.624 0.080 0.104 0.000
#> GSM25551 2 0.556 0.7011 0.068 0.740 0.108 0.068 0.016
#> GSM25570 2 0.641 0.5957 0.196 0.632 0.084 0.088 0.000
#> GSM25571 2 0.641 0.5957 0.196 0.632 0.084 0.088 0.000
#> GSM25358 2 0.682 0.5158 0.164 0.592 0.188 0.052 0.004
#> GSM25359 2 0.682 0.5158 0.164 0.592 0.188 0.052 0.004
#> GSM25360 1 0.848 -0.1209 0.348 0.264 0.268 0.108 0.012
#> GSM25361 1 0.848 -0.1228 0.348 0.260 0.272 0.108 0.012
#> GSM25377 5 0.480 0.8716 0.244 0.012 0.000 0.040 0.704
#> GSM25378 1 0.742 0.1855 0.484 0.340 0.020 0.084 0.072
#> GSM25401 2 0.807 0.4245 0.180 0.520 0.096 0.164 0.040
#> GSM25402 2 0.831 0.3610 0.208 0.484 0.068 0.180 0.060
#> GSM25349 2 0.379 0.6441 0.004 0.784 0.008 0.196 0.008
#> GSM25350 2 0.379 0.6441 0.004 0.784 0.008 0.196 0.008
#> GSM25356 2 0.677 0.4250 0.072 0.536 0.008 0.328 0.056
#> GSM25357 2 0.677 0.4250 0.072 0.536 0.008 0.328 0.056
#> GSM25385 3 0.670 0.4872 0.416 0.104 0.452 0.012 0.016
#> GSM25386 3 0.628 0.6338 0.304 0.132 0.552 0.012 0.000
#> GSM25399 5 0.378 0.8865 0.236 0.000 0.000 0.012 0.752
#> GSM25400 1 0.671 0.4019 0.648 0.176 0.052 0.044 0.080
#> GSM48659 2 0.301 0.7222 0.028 0.880 0.068 0.024 0.000
#> GSM48660 2 0.207 0.7006 0.000 0.896 0.000 0.104 0.000
#> GSM25409 2 0.526 0.7149 0.104 0.748 0.036 0.104 0.008
#> GSM25410 3 0.655 0.6246 0.304 0.136 0.540 0.016 0.004
#> GSM25426 2 0.554 0.6839 0.052 0.740 0.120 0.068 0.020
#> GSM25427 2 0.803 0.0118 0.376 0.384 0.020 0.136 0.084
#> GSM25540 2 0.599 0.6159 0.136 0.656 0.176 0.032 0.000
#> GSM25541 2 0.599 0.6159 0.136 0.656 0.176 0.032 0.000
#> GSM25542 2 0.559 0.6653 0.100 0.716 0.120 0.064 0.000
#> GSM25543 2 0.565 0.6611 0.100 0.712 0.120 0.068 0.000
#> GSM25479 1 0.440 0.5344 0.824 0.044 0.040 0.052 0.040
#> GSM25480 1 0.440 0.5344 0.824 0.044 0.040 0.052 0.040
#> GSM25481 2 0.726 0.4372 0.200 0.540 0.000 0.176 0.084
#> GSM25482 2 0.726 0.4372 0.200 0.540 0.000 0.176 0.084
#> GSM48654 2 0.324 0.7244 0.032 0.872 0.044 0.052 0.000
#> GSM48650 2 0.289 0.6676 0.000 0.836 0.000 0.160 0.004
#> GSM48651 2 0.205 0.7081 0.004 0.912 0.004 0.080 0.000
#> GSM48652 2 0.173 0.7062 0.000 0.920 0.000 0.080 0.000
#> GSM48653 2 0.243 0.7120 0.004 0.900 0.020 0.076 0.000
#> GSM48662 2 0.284 0.7103 0.012 0.880 0.020 0.088 0.000
#> GSM48663 2 0.355 0.6366 0.000 0.772 0.000 0.220 0.008
#> GSM25524 3 0.672 0.2778 0.404 0.000 0.464 0.064 0.068
#> GSM25525 1 0.491 0.4497 0.764 0.020 0.152 0.036 0.028
#> GSM25526 1 0.772 0.2450 0.528 0.216 0.168 0.044 0.044
#> GSM25527 1 0.470 0.5273 0.804 0.044 0.072 0.028 0.052
#> GSM25528 1 0.662 -0.2779 0.464 0.000 0.412 0.056 0.068
#> GSM25529 1 0.443 0.4701 0.800 0.016 0.124 0.024 0.036
#> GSM25530 1 0.705 -0.1605 0.480 0.000 0.332 0.044 0.144
#> GSM25531 1 0.624 0.2151 0.632 0.008 0.240 0.044 0.076
#> GSM48661 2 0.357 0.7210 0.052 0.848 0.080 0.020 0.000
#> GSM25561 1 0.769 -0.2466 0.464 0.060 0.348 0.084 0.044
#> GSM25562 1 0.714 0.3474 0.624 0.104 0.124 0.116 0.032
#> GSM25563 3 0.714 0.5109 0.332 0.032 0.512 0.092 0.032
#> GSM25564 1 0.821 0.1496 0.444 0.260 0.168 0.112 0.016
#> GSM25565 2 0.445 0.7168 0.064 0.800 0.084 0.052 0.000
#> GSM25566 2 0.421 0.7297 0.064 0.816 0.072 0.048 0.000
#> GSM25568 2 0.564 0.6781 0.044 0.724 0.064 0.148 0.020
#> GSM25569 2 0.499 0.6999 0.036 0.760 0.056 0.140 0.008
#> GSM25552 2 0.667 0.5709 0.216 0.600 0.076 0.108 0.000
#> GSM25553 2 0.667 0.5709 0.216 0.600 0.076 0.108 0.000
#> GSM25578 1 0.319 0.5329 0.884 0.028 0.040 0.016 0.032
#> GSM25579 1 0.442 0.5283 0.820 0.072 0.040 0.036 0.032
#> GSM25580 1 0.259 0.5438 0.900 0.020 0.004 0.008 0.068
#> GSM25581 1 0.259 0.5438 0.900 0.020 0.004 0.008 0.068
#> GSM48655 2 0.218 0.6994 0.000 0.896 0.000 0.100 0.004
#> GSM48656 2 0.371 0.7272 0.056 0.844 0.068 0.032 0.000
#> GSM48657 2 0.301 0.6666 0.000 0.836 0.004 0.156 0.004
#> GSM48658 2 0.372 0.7202 0.060 0.836 0.088 0.016 0.000
#> GSM25624 1 0.496 0.4949 0.776 0.120 0.036 0.024 0.044
#> GSM25625 1 0.696 0.1564 0.556 0.160 0.244 0.024 0.016
#> GSM25626 3 0.639 0.6269 0.316 0.124 0.544 0.004 0.012
#> GSM25627 1 0.807 0.1033 0.420 0.324 0.172 0.044 0.040
#> GSM25628 3 0.640 0.6355 0.312 0.116 0.552 0.016 0.004
#> GSM25629 2 0.756 0.2171 0.288 0.488 0.160 0.044 0.020
#> GSM25630 3 0.590 -0.5071 0.056 0.000 0.660 0.216 0.068
#> GSM25631 2 0.644 0.6149 0.180 0.632 0.120 0.068 0.000
#> GSM25632 1 0.522 0.3753 0.704 0.024 0.228 0.028 0.016
#> GSM25633 1 0.335 0.5347 0.872 0.020 0.064 0.016 0.028
#> GSM25634 1 0.374 0.5299 0.852 0.016 0.048 0.020 0.064
#> GSM25635 1 0.295 0.5474 0.892 0.032 0.020 0.008 0.048
#> GSM25656 4 0.764 0.0000 0.040 0.056 0.304 0.500 0.100
#> GSM25657 1 0.409 0.5250 0.824 0.016 0.052 0.012 0.096
#> GSM25658 1 0.746 0.3429 0.584 0.160 0.152 0.056 0.048
#> GSM25659 1 0.728 0.2671 0.556 0.240 0.120 0.068 0.016
#> GSM25660 1 0.318 0.5520 0.884 0.032 0.036 0.012 0.036
#> GSM25661 1 0.228 0.5399 0.920 0.012 0.020 0.004 0.044
#> GSM25662 2 0.526 0.6805 0.144 0.732 0.080 0.044 0.000
#> GSM25663 2 0.526 0.6805 0.144 0.732 0.080 0.044 0.000
#> GSM25680 2 0.665 0.5888 0.184 0.616 0.108 0.092 0.000
#> GSM25681 2 0.679 0.5678 0.192 0.600 0.112 0.096 0.000
#> GSM25682 2 0.257 0.7028 0.008 0.880 0.000 0.108 0.004
#> GSM25683 2 0.257 0.7028 0.008 0.880 0.000 0.108 0.004
#> GSM25684 2 0.283 0.7212 0.020 0.888 0.068 0.024 0.000
#> GSM25685 2 0.340 0.7196 0.024 0.848 0.108 0.020 0.000
#> GSM25686 2 0.257 0.7028 0.008 0.880 0.000 0.108 0.004
#> GSM25687 2 0.257 0.7028 0.008 0.880 0.000 0.108 0.004
#> GSM48664 5 0.393 0.9015 0.276 0.000 0.000 0.008 0.716
#> GSM48665 1 0.322 0.5442 0.864 0.024 0.008 0.008 0.096
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM25548 5 0.604 0.45381 0.152 0.140 0.072 0.000 0.628 0.008
#> GSM25549 5 0.615 0.44485 0.156 0.144 0.076 0.000 0.616 0.008
#> GSM25550 5 0.648 0.42574 0.156 0.184 0.072 0.000 0.576 0.012
#> GSM25551 5 0.482 0.42418 0.036 0.100 0.104 0.008 0.748 0.004
#> GSM25570 5 0.618 0.44191 0.156 0.148 0.076 0.000 0.612 0.008
#> GSM25571 5 0.618 0.44191 0.156 0.148 0.076 0.000 0.612 0.008
#> GSM25358 5 0.623 0.35751 0.072 0.092 0.256 0.000 0.572 0.008
#> GSM25359 5 0.623 0.35751 0.072 0.092 0.256 0.000 0.572 0.008
#> GSM25360 3 0.843 0.22439 0.264 0.136 0.280 0.016 0.272 0.032
#> GSM25361 3 0.847 0.22582 0.264 0.136 0.280 0.016 0.268 0.036
#> GSM25377 4 0.417 0.82924 0.168 0.072 0.000 0.752 0.004 0.004
#> GSM25378 1 0.694 -0.01148 0.488 0.180 0.016 0.048 0.264 0.004
#> GSM25401 5 0.793 -0.58095 0.140 0.276 0.104 0.060 0.416 0.004
#> GSM25402 5 0.816 -0.63167 0.172 0.292 0.068 0.076 0.380 0.012
#> GSM25349 5 0.400 -0.14455 0.004 0.328 0.000 0.000 0.656 0.012
#> GSM25350 5 0.400 -0.14455 0.004 0.328 0.000 0.000 0.656 0.012
#> GSM25356 2 0.615 1.00000 0.052 0.496 0.008 0.040 0.388 0.016
#> GSM25357 2 0.615 1.00000 0.052 0.496 0.008 0.040 0.388 0.016
#> GSM25385 3 0.590 0.49140 0.268 0.016 0.584 0.020 0.112 0.000
#> GSM25386 3 0.484 0.55121 0.164 0.008 0.688 0.000 0.140 0.000
#> GSM25399 4 0.296 0.81645 0.132 0.020 0.008 0.840 0.000 0.000
#> GSM25400 1 0.662 0.46963 0.624 0.084 0.048 0.092 0.144 0.008
#> GSM48659 5 0.250 0.49926 0.016 0.044 0.032 0.000 0.900 0.008
#> GSM48660 5 0.270 0.31721 0.000 0.188 0.000 0.000 0.812 0.000
#> GSM25409 5 0.497 0.47221 0.072 0.180 0.024 0.004 0.712 0.008
#> GSM25410 3 0.514 0.54998 0.168 0.012 0.672 0.000 0.144 0.004
#> GSM25426 5 0.475 0.38202 0.020 0.100 0.116 0.012 0.748 0.004
#> GSM25427 1 0.748 -0.35931 0.376 0.252 0.016 0.056 0.292 0.008
#> GSM25540 5 0.549 0.48122 0.080 0.088 0.148 0.000 0.680 0.004
#> GSM25541 5 0.549 0.48122 0.080 0.088 0.148 0.000 0.680 0.004
#> GSM25542 5 0.515 0.47217 0.068 0.108 0.096 0.000 0.720 0.008
#> GSM25543 5 0.519 0.47129 0.068 0.112 0.096 0.000 0.716 0.008
#> GSM25479 1 0.404 0.60780 0.828 0.044 0.040 0.036 0.040 0.012
#> GSM25480 1 0.404 0.60780 0.828 0.044 0.040 0.036 0.040 0.012
#> GSM25481 5 0.703 -0.67471 0.184 0.320 0.004 0.076 0.416 0.000
#> GSM25482 5 0.703 -0.67471 0.184 0.320 0.004 0.076 0.416 0.000
#> GSM48654 5 0.266 0.47732 0.016 0.076 0.012 0.000 0.884 0.012
#> GSM48650 5 0.324 0.09941 0.000 0.268 0.000 0.000 0.732 0.000
#> GSM48651 5 0.252 0.36928 0.000 0.152 0.004 0.000 0.844 0.000
#> GSM48652 5 0.242 0.36351 0.000 0.156 0.000 0.000 0.844 0.000
#> GSM48653 5 0.250 0.39946 0.000 0.116 0.004 0.000 0.868 0.012
#> GSM48662 5 0.267 0.37861 0.008 0.156 0.000 0.000 0.836 0.000
#> GSM48663 5 0.367 -0.23106 0.000 0.368 0.000 0.000 0.632 0.000
#> GSM25524 3 0.703 0.37722 0.224 0.092 0.548 0.048 0.004 0.084
#> GSM25525 1 0.467 0.54197 0.748 0.020 0.168 0.032 0.012 0.020
#> GSM25526 1 0.790 0.17149 0.440 0.080 0.196 0.056 0.220 0.008
#> GSM25527 1 0.441 0.60368 0.800 0.028 0.064 0.060 0.040 0.008
#> GSM25528 3 0.697 0.29283 0.320 0.076 0.484 0.052 0.000 0.068
#> GSM25529 1 0.435 0.56113 0.784 0.016 0.128 0.040 0.012 0.020
#> GSM25530 3 0.777 0.23947 0.324 0.068 0.392 0.152 0.004 0.060
#> GSM25531 1 0.721 0.00937 0.484 0.068 0.300 0.108 0.016 0.024
#> GSM48661 5 0.271 0.51259 0.020 0.052 0.032 0.000 0.888 0.008
#> GSM25561 1 0.787 -0.14880 0.396 0.068 0.344 0.036 0.036 0.120
#> GSM25562 1 0.727 0.40882 0.588 0.096 0.132 0.032 0.092 0.060
#> GSM25563 3 0.688 0.43437 0.224 0.036 0.560 0.016 0.040 0.124
#> GSM25564 1 0.805 0.13282 0.424 0.116 0.128 0.016 0.268 0.048
#> GSM25565 5 0.361 0.52236 0.040 0.092 0.036 0.000 0.828 0.004
#> GSM25566 5 0.382 0.51058 0.044 0.084 0.044 0.000 0.820 0.008
#> GSM25568 5 0.555 0.33554 0.028 0.244 0.032 0.004 0.648 0.044
#> GSM25569 5 0.480 0.40850 0.024 0.232 0.024 0.000 0.696 0.024
#> GSM25552 5 0.650 0.40780 0.172 0.176 0.068 0.000 0.572 0.012
#> GSM25553 5 0.650 0.40780 0.172 0.176 0.068 0.000 0.572 0.012
#> GSM25578 1 0.275 0.61151 0.892 0.012 0.044 0.028 0.020 0.004
#> GSM25579 1 0.401 0.59378 0.820 0.028 0.052 0.024 0.072 0.004
#> GSM25580 1 0.226 0.62151 0.908 0.020 0.008 0.056 0.008 0.000
#> GSM25581 1 0.226 0.62151 0.908 0.020 0.008 0.056 0.008 0.000
#> GSM48655 5 0.270 0.31024 0.000 0.188 0.000 0.000 0.812 0.000
#> GSM48656 5 0.279 0.51311 0.028 0.064 0.024 0.000 0.880 0.004
#> GSM48657 5 0.329 0.08024 0.000 0.276 0.000 0.000 0.724 0.000
#> GSM48658 5 0.290 0.51487 0.028 0.056 0.036 0.000 0.876 0.004
#> GSM25624 1 0.490 0.56757 0.760 0.056 0.032 0.048 0.100 0.004
#> GSM25625 1 0.675 0.23131 0.504 0.040 0.276 0.024 0.156 0.000
#> GSM25626 3 0.503 0.55384 0.176 0.000 0.672 0.012 0.140 0.000
#> GSM25627 5 0.793 -0.20472 0.332 0.080 0.208 0.044 0.332 0.004
#> GSM25628 3 0.549 0.54470 0.168 0.016 0.664 0.000 0.132 0.020
#> GSM25629 5 0.702 0.19168 0.212 0.076 0.188 0.012 0.508 0.004
#> GSM25630 3 0.694 -0.38670 0.024 0.196 0.504 0.040 0.004 0.232
#> GSM25631 5 0.608 0.47076 0.144 0.136 0.076 0.000 0.632 0.012
#> GSM25632 1 0.584 0.30822 0.604 0.044 0.284 0.024 0.008 0.036
#> GSM25633 1 0.382 0.60929 0.840 0.036 0.044 0.044 0.012 0.024
#> GSM25634 1 0.399 0.60429 0.820 0.020 0.044 0.084 0.008 0.024
#> GSM25635 1 0.277 0.62391 0.892 0.020 0.020 0.044 0.020 0.004
#> GSM25656 6 0.387 0.00000 0.008 0.032 0.100 0.000 0.048 0.812
#> GSM25657 1 0.408 0.59992 0.804 0.024 0.052 0.104 0.008 0.008
#> GSM25658 1 0.753 0.35243 0.532 0.096 0.128 0.060 0.172 0.012
#> GSM25659 1 0.717 0.26758 0.536 0.088 0.092 0.016 0.236 0.032
#> GSM25660 1 0.309 0.62920 0.876 0.024 0.040 0.036 0.020 0.004
#> GSM25661 1 0.245 0.62121 0.904 0.008 0.020 0.052 0.008 0.008
#> GSM25662 5 0.483 0.50809 0.116 0.100 0.044 0.000 0.736 0.004
#> GSM25663 5 0.483 0.50809 0.116 0.100 0.044 0.000 0.736 0.004
#> GSM25680 5 0.629 0.44937 0.144 0.140 0.084 0.000 0.616 0.016
#> GSM25681 5 0.655 0.42691 0.152 0.144 0.092 0.000 0.592 0.020
#> GSM25682 5 0.284 0.32806 0.004 0.188 0.000 0.000 0.808 0.000
#> GSM25683 5 0.284 0.32806 0.004 0.188 0.000 0.000 0.808 0.000
#> GSM25684 5 0.230 0.49643 0.008 0.044 0.032 0.000 0.908 0.008
#> GSM25685 5 0.298 0.48175 0.012 0.044 0.068 0.000 0.868 0.008
#> GSM25686 5 0.284 0.32806 0.004 0.188 0.000 0.000 0.808 0.000
#> GSM25687 5 0.284 0.32806 0.004 0.188 0.000 0.000 0.808 0.000
#> GSM48664 4 0.311 0.85932 0.196 0.012 0.000 0.792 0.000 0.000
#> GSM48665 1 0.298 0.62402 0.868 0.024 0.016 0.080 0.012 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 genotype/variation(p) k
#> MAD:hclust 92 6.92e-05 2
#> MAD:hclust 90 2.73e-04 3
#> MAD:hclust 75 2.36e-03 4
#> MAD:hclust 69 2.86e-03 5
#> MAD:hclust 33 9.63e-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["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 100 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 0.840 0.891 0.955 0.5042 0.496 0.496
#> 3 3 0.623 0.760 0.828 0.2632 0.847 0.697
#> 4 4 0.549 0.703 0.742 0.1256 0.936 0.825
#> 5 5 0.607 0.592 0.703 0.0758 0.902 0.685
#> 6 6 0.616 0.552 0.715 0.0448 0.983 0.922
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
#> GSM25548 2 0.0000 0.9580 0.000 1.000
#> GSM25549 2 0.0000 0.9580 0.000 1.000
#> GSM25550 2 0.0000 0.9580 0.000 1.000
#> GSM25551 2 0.0000 0.9580 0.000 1.000
#> GSM25570 2 0.0000 0.9580 0.000 1.000
#> GSM25571 2 0.0000 0.9580 0.000 1.000
#> GSM25358 1 0.2423 0.9227 0.960 0.040
#> GSM25359 2 0.6887 0.7595 0.184 0.816
#> GSM25360 1 0.0000 0.9444 1.000 0.000
#> GSM25361 1 0.9608 0.3550 0.616 0.384
#> GSM25377 1 0.0376 0.9446 0.996 0.004
#> GSM25378 1 0.2043 0.9270 0.968 0.032
#> GSM25401 1 0.5737 0.8290 0.864 0.136
#> GSM25402 1 0.4562 0.8713 0.904 0.096
#> GSM25349 2 0.0000 0.9580 0.000 1.000
#> GSM25350 2 0.0000 0.9580 0.000 1.000
#> GSM25356 1 0.4298 0.8791 0.912 0.088
#> GSM25357 2 0.0376 0.9553 0.004 0.996
#> GSM25385 1 0.0000 0.9444 1.000 0.000
#> GSM25386 1 0.0000 0.9444 1.000 0.000
#> GSM25399 1 0.0376 0.9446 0.996 0.004
#> GSM25400 1 0.0376 0.9446 0.996 0.004
#> GSM48659 2 0.0376 0.9568 0.004 0.996
#> GSM48660 2 0.0000 0.9580 0.000 1.000
#> GSM25409 2 0.0000 0.9580 0.000 1.000
#> GSM25410 1 0.0000 0.9444 1.000 0.000
#> GSM25426 2 0.0376 0.9568 0.004 0.996
#> GSM25427 1 0.2043 0.9270 0.968 0.032
#> GSM25540 2 0.9286 0.4882 0.344 0.656
#> GSM25541 2 0.9970 0.1344 0.468 0.532
#> GSM25542 2 0.0376 0.9568 0.004 0.996
#> GSM25543 2 0.0376 0.9568 0.004 0.996
#> GSM25479 1 0.0376 0.9446 0.996 0.004
#> GSM25480 1 0.0376 0.9446 0.996 0.004
#> GSM25481 1 0.8555 0.6283 0.720 0.280
#> GSM25482 1 0.8555 0.6283 0.720 0.280
#> GSM48654 2 0.0376 0.9568 0.004 0.996
#> GSM48650 2 0.0000 0.9580 0.000 1.000
#> GSM48651 2 0.0000 0.9580 0.000 1.000
#> GSM48652 2 0.0376 0.9568 0.004 0.996
#> GSM48653 2 0.0376 0.9568 0.004 0.996
#> GSM48662 2 0.0000 0.9580 0.000 1.000
#> GSM48663 2 0.0000 0.9580 0.000 1.000
#> GSM25524 1 0.0000 0.9444 1.000 0.000
#> GSM25525 1 0.0376 0.9446 0.996 0.004
#> GSM25526 1 0.0000 0.9444 1.000 0.000
#> GSM25527 1 0.0376 0.9446 0.996 0.004
#> GSM25528 1 0.0000 0.9444 1.000 0.000
#> GSM25529 1 0.0000 0.9444 1.000 0.000
#> GSM25530 1 0.0000 0.9444 1.000 0.000
#> GSM25531 1 0.0000 0.9444 1.000 0.000
#> GSM48661 2 0.0376 0.9568 0.004 0.996
#> GSM25561 1 0.0000 0.9444 1.000 0.000
#> GSM25562 1 0.0376 0.9446 0.996 0.004
#> GSM25563 1 0.0000 0.9444 1.000 0.000
#> GSM25564 1 0.8909 0.5805 0.692 0.308
#> GSM25565 2 0.0000 0.9580 0.000 1.000
#> GSM25566 2 0.0000 0.9580 0.000 1.000
#> GSM25568 2 0.9170 0.4629 0.332 0.668
#> GSM25569 2 0.0000 0.9580 0.000 1.000
#> GSM25552 2 0.0000 0.9580 0.000 1.000
#> GSM25553 2 0.0000 0.9580 0.000 1.000
#> GSM25578 1 0.0376 0.9446 0.996 0.004
#> GSM25579 1 0.0376 0.9446 0.996 0.004
#> GSM25580 1 0.0376 0.9446 0.996 0.004
#> GSM25581 1 0.0376 0.9446 0.996 0.004
#> GSM48655 2 0.0000 0.9580 0.000 1.000
#> GSM48656 2 0.0000 0.9580 0.000 1.000
#> GSM48657 2 0.0000 0.9580 0.000 1.000
#> GSM48658 2 0.0376 0.9568 0.004 0.996
#> GSM25624 1 0.0376 0.9446 0.996 0.004
#> GSM25625 1 0.0000 0.9444 1.000 0.000
#> GSM25626 1 0.0000 0.9444 1.000 0.000
#> GSM25627 1 0.2603 0.9153 0.956 0.044
#> GSM25628 1 0.9460 0.4067 0.636 0.364
#> GSM25629 2 0.9000 0.5471 0.316 0.684
#> GSM25630 1 0.0000 0.9444 1.000 0.000
#> GSM25631 2 0.5737 0.8253 0.136 0.864
#> GSM25632 1 0.0000 0.9444 1.000 0.000
#> GSM25633 1 0.0376 0.9446 0.996 0.004
#> GSM25634 1 0.0376 0.9446 0.996 0.004
#> GSM25635 1 0.0376 0.9446 0.996 0.004
#> GSM25656 1 1.0000 -0.0317 0.504 0.496
#> GSM25657 1 0.0000 0.9444 1.000 0.000
#> GSM25658 1 0.0000 0.9444 1.000 0.000
#> GSM25659 1 0.0000 0.9444 1.000 0.000
#> GSM25660 1 0.0376 0.9446 0.996 0.004
#> GSM25661 1 0.0376 0.9446 0.996 0.004
#> GSM25662 2 0.0376 0.9568 0.004 0.996
#> GSM25663 2 0.0376 0.9568 0.004 0.996
#> GSM25680 2 0.0376 0.9568 0.004 0.996
#> GSM25681 2 0.0000 0.9580 0.000 1.000
#> GSM25682 2 0.0000 0.9580 0.000 1.000
#> GSM25683 2 0.0000 0.9580 0.000 1.000
#> GSM25684 2 0.0376 0.9568 0.004 0.996
#> GSM25685 2 0.0376 0.9568 0.004 0.996
#> GSM25686 2 0.0000 0.9580 0.000 1.000
#> GSM25687 2 0.0000 0.9580 0.000 1.000
#> GSM48664 1 0.0376 0.9446 0.996 0.004
#> GSM48665 1 0.0376 0.9446 0.996 0.004
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM25548 2 0.2711 0.927 0.000 0.912 0.088
#> GSM25549 2 0.2796 0.926 0.000 0.908 0.092
#> GSM25550 2 0.2537 0.928 0.000 0.920 0.080
#> GSM25551 2 0.1860 0.936 0.000 0.948 0.052
#> GSM25570 2 0.2711 0.927 0.000 0.912 0.088
#> GSM25571 2 0.2711 0.927 0.000 0.912 0.088
#> GSM25358 1 0.7015 0.604 0.696 0.064 0.240
#> GSM25359 2 0.6882 0.719 0.096 0.732 0.172
#> GSM25360 3 0.6282 0.849 0.384 0.004 0.612
#> GSM25361 3 0.7637 0.777 0.284 0.076 0.640
#> GSM25377 1 0.5291 0.651 0.732 0.000 0.268
#> GSM25378 1 0.5797 0.639 0.712 0.008 0.280
#> GSM25401 1 0.7178 0.508 0.512 0.024 0.464
#> GSM25402 1 0.6950 0.565 0.572 0.020 0.408
#> GSM25349 2 0.2448 0.934 0.000 0.924 0.076
#> GSM25350 2 0.2356 0.935 0.000 0.928 0.072
#> GSM25356 1 0.6284 0.619 0.680 0.016 0.304
#> GSM25357 2 0.3340 0.912 0.000 0.880 0.120
#> GSM25385 3 0.6291 0.793 0.468 0.000 0.532
#> GSM25386 3 0.6330 0.854 0.396 0.004 0.600
#> GSM25399 1 0.5216 0.654 0.740 0.000 0.260
#> GSM25400 1 0.5216 0.654 0.740 0.000 0.260
#> GSM48659 2 0.2261 0.933 0.000 0.932 0.068
#> GSM48660 2 0.1753 0.932 0.000 0.952 0.048
#> GSM25409 2 0.2261 0.933 0.000 0.932 0.068
#> GSM25410 3 0.6359 0.853 0.404 0.004 0.592
#> GSM25426 2 0.1860 0.939 0.000 0.948 0.052
#> GSM25427 1 0.5797 0.639 0.712 0.008 0.280
#> GSM25540 3 0.7810 0.760 0.268 0.092 0.640
#> GSM25541 3 0.7637 0.780 0.284 0.076 0.640
#> GSM25542 2 0.3192 0.912 0.000 0.888 0.112
#> GSM25543 2 0.3941 0.875 0.000 0.844 0.156
#> GSM25479 1 0.0747 0.691 0.984 0.000 0.016
#> GSM25480 1 0.0747 0.691 0.984 0.000 0.016
#> GSM25481 1 0.6823 0.610 0.668 0.036 0.296
#> GSM25482 1 0.6823 0.610 0.668 0.036 0.296
#> GSM48654 2 0.2356 0.932 0.000 0.928 0.072
#> GSM48650 2 0.2066 0.930 0.000 0.940 0.060
#> GSM48651 2 0.1753 0.936 0.000 0.952 0.048
#> GSM48652 2 0.2261 0.933 0.000 0.932 0.068
#> GSM48653 2 0.2448 0.931 0.000 0.924 0.076
#> GSM48662 2 0.1643 0.935 0.000 0.956 0.044
#> GSM48663 2 0.1860 0.931 0.000 0.948 0.052
#> GSM25524 3 0.6260 0.832 0.448 0.000 0.552
#> GSM25525 1 0.3412 0.558 0.876 0.000 0.124
#> GSM25526 3 0.6260 0.828 0.448 0.000 0.552
#> GSM25527 1 0.2261 0.638 0.932 0.000 0.068
#> GSM25528 1 0.6026 -0.419 0.624 0.000 0.376
#> GSM25529 1 0.3482 0.551 0.872 0.000 0.128
#> GSM25530 1 0.6295 -0.693 0.528 0.000 0.472
#> GSM25531 1 0.3192 0.575 0.888 0.000 0.112
#> GSM48661 2 0.2537 0.930 0.000 0.920 0.080
#> GSM25561 3 0.6308 0.765 0.492 0.000 0.508
#> GSM25562 1 0.2066 0.658 0.940 0.000 0.060
#> GSM25563 3 0.6168 0.852 0.412 0.000 0.588
#> GSM25564 1 0.8804 0.193 0.584 0.204 0.212
#> GSM25565 2 0.1529 0.940 0.000 0.960 0.040
#> GSM25566 2 0.0892 0.939 0.000 0.980 0.020
#> GSM25568 2 0.7298 0.669 0.088 0.692 0.220
#> GSM25569 2 0.2165 0.936 0.000 0.936 0.064
#> GSM25552 2 0.2796 0.926 0.000 0.908 0.092
#> GSM25553 2 0.3670 0.919 0.020 0.888 0.092
#> GSM25578 1 0.0747 0.689 0.984 0.000 0.016
#> GSM25579 1 0.2066 0.657 0.940 0.000 0.060
#> GSM25580 1 0.0424 0.698 0.992 0.000 0.008
#> GSM25581 1 0.0424 0.698 0.992 0.000 0.008
#> GSM48655 2 0.1529 0.934 0.000 0.960 0.040
#> GSM48656 2 0.1964 0.936 0.000 0.944 0.056
#> GSM48657 2 0.1860 0.932 0.000 0.948 0.052
#> GSM48658 2 0.2165 0.934 0.000 0.936 0.064
#> GSM25624 1 0.0237 0.697 0.996 0.000 0.004
#> GSM25625 3 0.6280 0.812 0.460 0.000 0.540
#> GSM25626 3 0.6386 0.852 0.412 0.004 0.584
#> GSM25627 3 0.7406 0.833 0.360 0.044 0.596
#> GSM25628 3 0.7588 0.808 0.312 0.064 0.624
#> GSM25629 3 0.8311 0.730 0.252 0.132 0.616
#> GSM25630 3 0.6244 0.837 0.440 0.000 0.560
#> GSM25631 2 0.7245 0.695 0.168 0.712 0.120
#> GSM25632 3 0.6267 0.831 0.452 0.000 0.548
#> GSM25633 1 0.0000 0.696 1.000 0.000 0.000
#> GSM25634 1 0.0000 0.696 1.000 0.000 0.000
#> GSM25635 1 0.0237 0.697 0.996 0.000 0.004
#> GSM25656 3 0.7902 0.779 0.280 0.092 0.628
#> GSM25657 1 0.1860 0.659 0.948 0.000 0.052
#> GSM25658 1 0.5621 -0.055 0.692 0.000 0.308
#> GSM25659 1 0.4399 0.440 0.812 0.000 0.188
#> GSM25660 1 0.0000 0.696 1.000 0.000 0.000
#> GSM25661 1 0.0000 0.696 1.000 0.000 0.000
#> GSM25662 2 0.1753 0.938 0.000 0.952 0.048
#> GSM25663 2 0.2261 0.938 0.000 0.932 0.068
#> GSM25680 2 0.3340 0.922 0.000 0.880 0.120
#> GSM25681 2 0.3267 0.924 0.000 0.884 0.116
#> GSM25682 2 0.1753 0.934 0.000 0.952 0.048
#> GSM25683 2 0.1753 0.934 0.000 0.952 0.048
#> GSM25684 2 0.2066 0.936 0.000 0.940 0.060
#> GSM25685 2 0.2356 0.935 0.000 0.928 0.072
#> GSM25686 2 0.1860 0.933 0.000 0.948 0.052
#> GSM25687 2 0.1860 0.933 0.000 0.948 0.052
#> GSM48664 1 0.5216 0.653 0.740 0.000 0.260
#> GSM48665 1 0.5216 0.654 0.740 0.000 0.260
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM25548 2 0.3679 0.805 0.000 0.856 0.060 0.084
#> GSM25549 2 0.3679 0.805 0.000 0.856 0.060 0.084
#> GSM25550 2 0.3828 0.804 0.000 0.848 0.068 0.084
#> GSM25551 2 0.3182 0.819 0.000 0.876 0.028 0.096
#> GSM25570 2 0.3679 0.805 0.000 0.856 0.060 0.084
#> GSM25571 2 0.3679 0.805 0.000 0.856 0.060 0.084
#> GSM25358 4 0.9386 0.487 0.320 0.120 0.188 0.372
#> GSM25359 2 0.7316 0.445 0.012 0.556 0.292 0.140
#> GSM25360 3 0.3249 0.858 0.140 0.000 0.852 0.008
#> GSM25361 3 0.6367 0.746 0.084 0.084 0.728 0.104
#> GSM25377 4 0.5372 0.771 0.444 0.000 0.012 0.544
#> GSM25378 4 0.5415 0.790 0.436 0.004 0.008 0.552
#> GSM25401 4 0.7455 0.576 0.196 0.024 0.188 0.592
#> GSM25402 4 0.7422 0.659 0.276 0.016 0.148 0.560
#> GSM25349 2 0.4880 0.801 0.000 0.760 0.052 0.188
#> GSM25350 2 0.4920 0.800 0.000 0.756 0.052 0.192
#> GSM25356 4 0.5573 0.796 0.396 0.012 0.008 0.584
#> GSM25357 2 0.4283 0.729 0.000 0.740 0.004 0.256
#> GSM25385 3 0.3969 0.846 0.180 0.000 0.804 0.016
#> GSM25386 3 0.3377 0.860 0.140 0.000 0.848 0.012
#> GSM25399 4 0.5396 0.742 0.464 0.000 0.012 0.524
#> GSM25400 1 0.5409 -0.728 0.496 0.000 0.012 0.492
#> GSM48659 2 0.4964 0.811 0.000 0.764 0.068 0.168
#> GSM48660 2 0.4957 0.807 0.000 0.748 0.048 0.204
#> GSM25409 2 0.4070 0.809 0.000 0.824 0.044 0.132
#> GSM25410 3 0.3711 0.858 0.140 0.000 0.836 0.024
#> GSM25426 2 0.3659 0.817 0.000 0.840 0.024 0.136
#> GSM25427 4 0.5427 0.784 0.444 0.004 0.008 0.544
#> GSM25540 3 0.6310 0.763 0.084 0.084 0.732 0.100
#> GSM25541 3 0.6433 0.758 0.088 0.088 0.724 0.100
#> GSM25542 2 0.7507 0.473 0.000 0.480 0.316 0.204
#> GSM25543 2 0.7613 0.365 0.000 0.428 0.368 0.204
#> GSM25479 1 0.1004 0.710 0.972 0.000 0.024 0.004
#> GSM25480 1 0.0817 0.711 0.976 0.000 0.024 0.000
#> GSM25481 4 0.5883 0.787 0.388 0.016 0.016 0.580
#> GSM25482 4 0.5883 0.787 0.388 0.016 0.016 0.580
#> GSM48654 2 0.4893 0.809 0.000 0.768 0.064 0.168
#> GSM48650 2 0.5102 0.799 0.000 0.732 0.048 0.220
#> GSM48651 2 0.4789 0.811 0.000 0.772 0.056 0.172
#> GSM48652 2 0.4789 0.811 0.000 0.772 0.056 0.172
#> GSM48653 2 0.4937 0.810 0.000 0.764 0.064 0.172
#> GSM48662 2 0.4370 0.818 0.000 0.800 0.044 0.156
#> GSM48663 2 0.5608 0.782 0.000 0.684 0.060 0.256
#> GSM25524 1 0.5604 -0.199 0.504 0.000 0.476 0.020
#> GSM25525 1 0.2255 0.707 0.920 0.000 0.068 0.012
#> GSM25526 3 0.5957 0.529 0.364 0.000 0.588 0.048
#> GSM25527 1 0.2739 0.711 0.904 0.000 0.060 0.036
#> GSM25528 1 0.4737 0.517 0.728 0.000 0.252 0.020
#> GSM25529 1 0.2473 0.700 0.908 0.000 0.080 0.012
#> GSM25530 1 0.5587 0.201 0.600 0.000 0.372 0.028
#> GSM25531 1 0.2773 0.699 0.900 0.000 0.072 0.028
#> GSM48661 2 0.5355 0.797 0.000 0.736 0.084 0.180
#> GSM25561 3 0.4983 0.751 0.272 0.000 0.704 0.024
#> GSM25562 1 0.2759 0.700 0.904 0.000 0.044 0.052
#> GSM25563 3 0.3853 0.854 0.160 0.000 0.820 0.020
#> GSM25564 1 0.8330 0.198 0.544 0.224 0.152 0.080
#> GSM25565 2 0.4070 0.832 0.000 0.824 0.044 0.132
#> GSM25566 2 0.1389 0.828 0.000 0.952 0.000 0.048
#> GSM25568 2 0.7968 0.551 0.012 0.472 0.280 0.236
#> GSM25569 2 0.5496 0.803 0.000 0.724 0.088 0.188
#> GSM25552 2 0.3966 0.806 0.000 0.840 0.072 0.088
#> GSM25553 2 0.4346 0.801 0.004 0.824 0.076 0.096
#> GSM25578 1 0.0469 0.713 0.988 0.000 0.012 0.000
#> GSM25579 1 0.2750 0.689 0.908 0.004 0.056 0.032
#> GSM25580 1 0.1661 0.690 0.944 0.000 0.004 0.052
#> GSM25581 1 0.1743 0.688 0.940 0.000 0.004 0.056
#> GSM48655 2 0.3856 0.819 0.000 0.832 0.032 0.136
#> GSM48656 2 0.4746 0.814 0.000 0.776 0.056 0.168
#> GSM48657 2 0.4881 0.806 0.000 0.756 0.048 0.196
#> GSM48658 2 0.5535 0.797 0.000 0.720 0.088 0.192
#> GSM25624 1 0.1902 0.680 0.932 0.000 0.004 0.064
#> GSM25625 3 0.3895 0.846 0.184 0.000 0.804 0.012
#> GSM25626 3 0.3300 0.859 0.144 0.000 0.848 0.008
#> GSM25627 3 0.6903 0.795 0.132 0.080 0.688 0.100
#> GSM25628 3 0.3981 0.844 0.100 0.040 0.848 0.012
#> GSM25629 3 0.6545 0.769 0.088 0.108 0.716 0.088
#> GSM25630 3 0.4004 0.850 0.164 0.000 0.812 0.024
#> GSM25631 2 0.7246 0.708 0.068 0.656 0.124 0.152
#> GSM25632 3 0.3937 0.841 0.188 0.000 0.800 0.012
#> GSM25633 1 0.1576 0.696 0.948 0.000 0.004 0.048
#> GSM25634 1 0.2048 0.681 0.928 0.000 0.008 0.064
#> GSM25635 1 0.2542 0.659 0.904 0.000 0.012 0.084
#> GSM25656 3 0.4858 0.828 0.084 0.052 0.816 0.048
#> GSM25657 1 0.2224 0.709 0.928 0.000 0.040 0.032
#> GSM25658 1 0.5599 0.437 0.664 0.000 0.288 0.048
#> GSM25659 1 0.3974 0.659 0.844 0.008 0.108 0.040
#> GSM25660 1 0.1576 0.693 0.948 0.000 0.004 0.048
#> GSM25661 1 0.1489 0.689 0.952 0.000 0.004 0.044
#> GSM25662 2 0.3856 0.830 0.000 0.832 0.032 0.136
#> GSM25663 2 0.3754 0.821 0.000 0.852 0.064 0.084
#> GSM25680 2 0.4469 0.801 0.000 0.808 0.080 0.112
#> GSM25681 2 0.4411 0.802 0.000 0.812 0.080 0.108
#> GSM25682 2 0.2888 0.808 0.000 0.872 0.004 0.124
#> GSM25683 2 0.2888 0.808 0.000 0.872 0.004 0.124
#> GSM25684 2 0.3984 0.828 0.000 0.828 0.040 0.132
#> GSM25685 2 0.4322 0.821 0.000 0.804 0.044 0.152
#> GSM25686 2 0.2888 0.808 0.000 0.872 0.004 0.124
#> GSM25687 2 0.2888 0.808 0.000 0.872 0.004 0.124
#> GSM48664 4 0.5399 0.741 0.468 0.000 0.012 0.520
#> GSM48665 1 0.5300 -0.544 0.580 0.000 0.012 0.408
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM25548 5 0.4367 0.7966 0.000 0.416 0.000 0.004 0.580
#> GSM25549 5 0.4367 0.7966 0.000 0.416 0.000 0.004 0.580
#> GSM25550 5 0.4341 0.7918 0.000 0.404 0.000 0.004 0.592
#> GSM25551 2 0.6048 -0.2373 0.000 0.516 0.036 0.048 0.400
#> GSM25570 5 0.4367 0.7966 0.000 0.416 0.000 0.004 0.580
#> GSM25571 5 0.4359 0.7947 0.000 0.412 0.000 0.004 0.584
#> GSM25358 4 0.7689 0.5744 0.100 0.024 0.136 0.548 0.192
#> GSM25359 5 0.6917 0.2726 0.000 0.204 0.192 0.048 0.556
#> GSM25360 3 0.2206 0.8366 0.068 0.000 0.912 0.004 0.016
#> GSM25361 3 0.5980 0.6761 0.044 0.028 0.616 0.016 0.296
#> GSM25377 4 0.4289 0.8348 0.176 0.000 0.000 0.760 0.064
#> GSM25378 4 0.3850 0.8431 0.172 0.000 0.004 0.792 0.032
#> GSM25401 4 0.5562 0.6851 0.056 0.020 0.116 0.740 0.068
#> GSM25402 4 0.5353 0.7709 0.120 0.000 0.096 0.732 0.052
#> GSM25349 2 0.5763 0.2581 0.000 0.600 0.004 0.108 0.288
#> GSM25350 2 0.5798 0.2463 0.000 0.592 0.004 0.108 0.296
#> GSM25356 4 0.3368 0.8452 0.156 0.000 0.000 0.820 0.024
#> GSM25357 2 0.6824 -0.1159 0.000 0.428 0.012 0.188 0.372
#> GSM25385 3 0.2060 0.8359 0.052 0.000 0.924 0.008 0.016
#> GSM25386 3 0.1412 0.8377 0.036 0.000 0.952 0.004 0.008
#> GSM25399 4 0.4725 0.8195 0.200 0.000 0.000 0.720 0.080
#> GSM25400 4 0.4173 0.8131 0.224 0.000 0.012 0.748 0.016
#> GSM48659 2 0.2824 0.4792 0.000 0.880 0.024 0.008 0.088
#> GSM48660 2 0.1628 0.5246 0.000 0.936 0.000 0.008 0.056
#> GSM25409 5 0.5039 0.5565 0.000 0.456 0.000 0.032 0.512
#> GSM25410 3 0.1455 0.8375 0.032 0.000 0.952 0.008 0.008
#> GSM25426 2 0.6298 0.1867 0.000 0.580 0.052 0.068 0.300
#> GSM25427 4 0.3888 0.8429 0.176 0.000 0.004 0.788 0.032
#> GSM25540 3 0.5079 0.7314 0.012 0.028 0.696 0.016 0.248
#> GSM25541 3 0.5129 0.7265 0.012 0.028 0.688 0.016 0.256
#> GSM25542 2 0.7169 0.2210 0.000 0.536 0.204 0.064 0.196
#> GSM25543 2 0.7732 0.0975 0.000 0.408 0.276 0.064 0.252
#> GSM25479 1 0.1498 0.8135 0.952 0.000 0.008 0.024 0.016
#> GSM25480 1 0.1405 0.8133 0.956 0.000 0.008 0.020 0.016
#> GSM25481 4 0.4060 0.8428 0.156 0.000 0.004 0.788 0.052
#> GSM25482 4 0.4060 0.8428 0.156 0.000 0.004 0.788 0.052
#> GSM48654 2 0.2027 0.5109 0.000 0.928 0.024 0.008 0.040
#> GSM48650 2 0.3086 0.5142 0.000 0.864 0.004 0.040 0.092
#> GSM48651 2 0.0579 0.5284 0.000 0.984 0.008 0.008 0.000
#> GSM48652 2 0.0693 0.5280 0.000 0.980 0.012 0.008 0.000
#> GSM48653 2 0.1597 0.5193 0.000 0.948 0.020 0.008 0.024
#> GSM48662 2 0.1492 0.5207 0.000 0.948 0.008 0.004 0.040
#> GSM48663 2 0.3946 0.4783 0.000 0.800 0.000 0.080 0.120
#> GSM25524 1 0.5792 0.3006 0.568 0.000 0.356 0.024 0.052
#> GSM25525 1 0.2305 0.8004 0.916 0.000 0.028 0.012 0.044
#> GSM25526 3 0.6289 0.4178 0.320 0.000 0.564 0.040 0.076
#> GSM25527 1 0.2263 0.8090 0.920 0.000 0.020 0.024 0.036
#> GSM25528 1 0.4249 0.7172 0.800 0.000 0.120 0.024 0.056
#> GSM25529 1 0.2305 0.8004 0.916 0.000 0.028 0.012 0.044
#> GSM25530 1 0.5576 0.6161 0.688 0.000 0.200 0.040 0.072
#> GSM25531 1 0.3071 0.7929 0.880 0.000 0.032 0.032 0.056
#> GSM48661 2 0.3525 0.4542 0.000 0.836 0.032 0.012 0.120
#> GSM25561 3 0.5111 0.6993 0.200 0.000 0.716 0.028 0.056
#> GSM25562 1 0.3434 0.7936 0.860 0.000 0.028 0.056 0.056
#> GSM25563 3 0.3152 0.8263 0.052 0.000 0.876 0.028 0.044
#> GSM25564 1 0.7115 0.4255 0.588 0.212 0.064 0.020 0.116
#> GSM25565 2 0.4460 0.3323 0.000 0.748 0.016 0.032 0.204
#> GSM25566 2 0.5033 -0.2346 0.000 0.568 0.004 0.028 0.400
#> GSM25568 2 0.7497 0.2014 0.000 0.512 0.152 0.108 0.228
#> GSM25569 2 0.3921 0.4465 0.000 0.812 0.012 0.048 0.128
#> GSM25552 5 0.4350 0.7945 0.000 0.408 0.000 0.004 0.588
#> GSM25553 5 0.4460 0.7853 0.004 0.392 0.000 0.004 0.600
#> GSM25578 1 0.0740 0.8128 0.980 0.000 0.004 0.008 0.008
#> GSM25579 1 0.1651 0.8109 0.944 0.000 0.012 0.008 0.036
#> GSM25580 1 0.3099 0.7594 0.848 0.000 0.000 0.124 0.028
#> GSM25581 1 0.2964 0.7616 0.856 0.000 0.000 0.120 0.024
#> GSM48655 2 0.3807 0.4316 0.000 0.792 0.004 0.028 0.176
#> GSM48656 2 0.2796 0.4936 0.000 0.868 0.008 0.008 0.116
#> GSM48657 2 0.2953 0.5094 0.000 0.868 0.004 0.028 0.100
#> GSM48658 2 0.3900 0.3938 0.000 0.788 0.020 0.012 0.180
#> GSM25624 1 0.3730 0.7045 0.800 0.000 0.004 0.168 0.028
#> GSM25625 3 0.2466 0.8307 0.076 0.000 0.900 0.012 0.012
#> GSM25626 3 0.1329 0.8367 0.032 0.000 0.956 0.008 0.004
#> GSM25627 3 0.6727 0.7019 0.064 0.100 0.664 0.048 0.124
#> GSM25628 3 0.2127 0.8290 0.016 0.016 0.932 0.016 0.020
#> GSM25629 3 0.5952 0.7229 0.012 0.092 0.692 0.044 0.160
#> GSM25630 3 0.3581 0.8189 0.068 0.000 0.852 0.036 0.044
#> GSM25631 5 0.6346 0.4666 0.040 0.368 0.044 0.012 0.536
#> GSM25632 3 0.2409 0.8317 0.060 0.000 0.908 0.020 0.012
#> GSM25633 1 0.2388 0.7892 0.900 0.000 0.000 0.072 0.028
#> GSM25634 1 0.3409 0.7329 0.824 0.000 0.000 0.144 0.032
#> GSM25635 1 0.3574 0.7103 0.804 0.000 0.000 0.168 0.028
#> GSM25656 3 0.4366 0.7954 0.012 0.028 0.804 0.036 0.120
#> GSM25657 1 0.1498 0.8140 0.952 0.000 0.008 0.024 0.016
#> GSM25658 1 0.5884 0.5557 0.640 0.000 0.248 0.036 0.076
#> GSM25659 1 0.2912 0.7864 0.876 0.000 0.028 0.008 0.088
#> GSM25660 1 0.2540 0.7809 0.888 0.000 0.000 0.088 0.024
#> GSM25661 1 0.2597 0.7779 0.884 0.000 0.000 0.092 0.024
#> GSM25662 2 0.4932 0.2382 0.000 0.708 0.032 0.028 0.232
#> GSM25663 5 0.4800 0.5295 0.000 0.476 0.004 0.012 0.508
#> GSM25680 5 0.4262 0.7475 0.000 0.440 0.000 0.000 0.560
#> GSM25681 5 0.4390 0.7455 0.000 0.428 0.004 0.000 0.568
#> GSM25682 2 0.5303 0.0387 0.000 0.604 0.012 0.040 0.344
#> GSM25683 2 0.5303 0.0387 0.000 0.604 0.012 0.040 0.344
#> GSM25684 2 0.4892 0.2544 0.000 0.720 0.036 0.028 0.216
#> GSM25685 2 0.5537 0.2827 0.000 0.672 0.048 0.044 0.236
#> GSM25686 2 0.5303 0.0387 0.000 0.604 0.012 0.040 0.344
#> GSM25687 2 0.5303 0.0387 0.000 0.604 0.012 0.040 0.344
#> GSM48664 4 0.4707 0.8162 0.212 0.000 0.000 0.716 0.072
#> GSM48665 4 0.5171 0.3343 0.456 0.000 0.000 0.504 0.040
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM25548 5 0.3309 0.7311 0.000 0.280 0.000 0.000 0.720 NA
#> GSM25549 5 0.3288 0.7310 0.000 0.276 0.000 0.000 0.724 NA
#> GSM25550 5 0.3565 0.7301 0.000 0.276 0.000 0.004 0.716 NA
#> GSM25551 5 0.5790 0.2223 0.000 0.376 0.012 0.012 0.508 NA
#> GSM25570 5 0.3309 0.7311 0.000 0.280 0.000 0.000 0.720 NA
#> GSM25571 5 0.3309 0.7311 0.000 0.280 0.000 0.000 0.720 NA
#> GSM25358 4 0.7032 0.5322 0.048 0.008 0.104 0.556 0.216 NA
#> GSM25359 5 0.6071 0.3932 0.000 0.100 0.152 0.012 0.636 NA
#> GSM25360 3 0.2557 0.7582 0.036 0.000 0.892 0.004 0.012 NA
#> GSM25361 3 0.6758 0.5348 0.024 0.028 0.480 0.004 0.316 NA
#> GSM25377 4 0.4203 0.7801 0.072 0.000 0.000 0.768 0.024 NA
#> GSM25378 4 0.3029 0.7955 0.104 0.000 0.004 0.852 0.008 NA
#> GSM25401 4 0.6040 0.6259 0.032 0.004 0.124 0.660 0.052 NA
#> GSM25402 4 0.5439 0.7080 0.044 0.004 0.104 0.716 0.032 NA
#> GSM25349 2 0.6968 0.2088 0.000 0.452 0.008 0.068 0.292 NA
#> GSM25350 2 0.6904 0.2176 0.000 0.468 0.008 0.068 0.284 NA
#> GSM25356 4 0.2981 0.8003 0.064 0.000 0.000 0.864 0.020 NA
#> GSM25357 5 0.7110 0.1537 0.000 0.276 0.000 0.212 0.416 NA
#> GSM25385 3 0.1483 0.7590 0.036 0.000 0.944 0.008 0.000 NA
#> GSM25386 3 0.0520 0.7636 0.008 0.000 0.984 0.000 0.000 NA
#> GSM25399 4 0.4668 0.7575 0.080 0.000 0.000 0.712 0.020 NA
#> GSM25400 4 0.3800 0.7599 0.160 0.000 0.004 0.776 0.000 NA
#> GSM48659 2 0.3138 0.4996 0.000 0.840 0.004 0.000 0.096 NA
#> GSM48660 2 0.1890 0.5288 0.000 0.924 0.000 0.008 0.044 NA
#> GSM25409 5 0.5077 0.5803 0.000 0.328 0.000 0.028 0.600 NA
#> GSM25410 3 0.0622 0.7637 0.012 0.000 0.980 0.000 0.000 NA
#> GSM25426 2 0.6385 -0.0075 0.000 0.432 0.020 0.020 0.404 NA
#> GSM25427 4 0.3200 0.7960 0.104 0.000 0.004 0.844 0.012 NA
#> GSM25540 3 0.6376 0.5811 0.008 0.036 0.536 0.004 0.280 NA
#> GSM25541 3 0.6376 0.5811 0.008 0.036 0.536 0.004 0.280 NA
#> GSM25542 2 0.7235 0.2775 0.000 0.448 0.116 0.008 0.164 NA
#> GSM25543 2 0.7556 0.2118 0.000 0.372 0.176 0.004 0.172 NA
#> GSM25479 1 0.1851 0.7834 0.928 0.000 0.000 0.036 0.012 NA
#> GSM25480 1 0.1857 0.7834 0.928 0.000 0.000 0.032 0.012 NA
#> GSM25481 4 0.3875 0.7912 0.064 0.004 0.004 0.820 0.040 NA
#> GSM25482 4 0.3875 0.7912 0.064 0.004 0.004 0.820 0.040 NA
#> GSM48654 2 0.2134 0.5250 0.000 0.904 0.000 0.000 0.052 NA
#> GSM48650 2 0.3659 0.5119 0.000 0.820 0.000 0.032 0.088 NA
#> GSM48651 2 0.0665 0.5386 0.000 0.980 0.000 0.004 0.008 NA
#> GSM48652 2 0.0653 0.5383 0.000 0.980 0.000 0.004 0.004 NA
#> GSM48653 2 0.1901 0.5310 0.000 0.924 0.008 0.000 0.028 NA
#> GSM48662 2 0.1666 0.5329 0.000 0.936 0.000 0.008 0.036 NA
#> GSM48663 2 0.5290 0.4577 0.000 0.704 0.004 0.092 0.088 NA
#> GSM25524 1 0.5803 0.3125 0.536 0.000 0.304 0.008 0.004 NA
#> GSM25525 1 0.2339 0.7557 0.880 0.000 0.004 0.004 0.004 NA
#> GSM25526 3 0.6722 0.2447 0.328 0.000 0.472 0.024 0.036 NA
#> GSM25527 1 0.3255 0.7727 0.844 0.000 0.020 0.052 0.000 NA
#> GSM25528 1 0.3491 0.7212 0.804 0.000 0.036 0.004 0.004 NA
#> GSM25529 1 0.2689 0.7504 0.864 0.000 0.016 0.004 0.004 NA
#> GSM25530 1 0.5153 0.6221 0.660 0.000 0.088 0.028 0.000 NA
#> GSM25531 1 0.4158 0.7021 0.740 0.000 0.020 0.036 0.000 NA
#> GSM48661 2 0.3690 0.4834 0.000 0.804 0.012 0.000 0.116 NA
#> GSM25561 3 0.5365 0.6152 0.192 0.000 0.656 0.008 0.016 NA
#> GSM25562 1 0.4441 0.7094 0.752 0.000 0.004 0.080 0.020 NA
#> GSM25563 3 0.3123 0.7420 0.020 0.000 0.844 0.004 0.016 NA
#> GSM25564 1 0.7896 0.3101 0.464 0.248 0.056 0.028 0.064 NA
#> GSM25565 2 0.5040 0.2942 0.000 0.656 0.004 0.008 0.236 NA
#> GSM25566 2 0.5052 -0.2244 0.000 0.500 0.004 0.008 0.444 NA
#> GSM25568 2 0.7489 0.2744 0.004 0.448 0.056 0.048 0.164 NA
#> GSM25569 2 0.4888 0.4275 0.000 0.676 0.004 0.004 0.108 NA
#> GSM25552 5 0.4083 0.7174 0.000 0.284 0.000 0.008 0.688 NA
#> GSM25553 5 0.4194 0.7087 0.004 0.264 0.000 0.008 0.700 NA
#> GSM25578 1 0.1218 0.7838 0.956 0.000 0.000 0.028 0.004 NA
#> GSM25579 1 0.1630 0.7850 0.940 0.000 0.000 0.020 0.016 NA
#> GSM25580 1 0.3594 0.7375 0.804 0.000 0.008 0.144 0.004 NA
#> GSM25581 1 0.3594 0.7375 0.804 0.000 0.008 0.144 0.004 NA
#> GSM48655 2 0.4159 0.3782 0.000 0.732 0.000 0.016 0.216 NA
#> GSM48656 2 0.2445 0.5194 0.000 0.872 0.000 0.000 0.108 NA
#> GSM48657 2 0.3453 0.4982 0.000 0.828 0.000 0.024 0.104 NA
#> GSM48658 2 0.4056 0.4341 0.000 0.748 0.004 0.000 0.184 NA
#> GSM25624 1 0.4100 0.7017 0.752 0.000 0.008 0.176 0.000 NA
#> GSM25625 3 0.3253 0.7317 0.088 0.000 0.848 0.012 0.008 NA
#> GSM25626 3 0.1026 0.7621 0.012 0.000 0.968 0.008 0.004 NA
#> GSM25627 3 0.7652 0.5477 0.048 0.104 0.536 0.036 0.100 NA
#> GSM25628 3 0.2107 0.7603 0.004 0.016 0.920 0.004 0.012 NA
#> GSM25629 3 0.7214 0.5652 0.020 0.116 0.520 0.004 0.164 NA
#> GSM25630 3 0.3684 0.7293 0.052 0.000 0.808 0.004 0.012 NA
#> GSM25631 5 0.5916 0.4185 0.032 0.260 0.016 0.000 0.596 NA
#> GSM25632 3 0.1480 0.7607 0.040 0.000 0.940 0.000 0.000 NA
#> GSM25633 1 0.3025 0.7662 0.856 0.000 0.008 0.092 0.004 NA
#> GSM25634 1 0.3922 0.7315 0.784 0.000 0.008 0.140 0.004 NA
#> GSM25635 1 0.3954 0.7122 0.764 0.000 0.008 0.172 0.000 NA
#> GSM25656 3 0.5164 0.7073 0.008 0.024 0.704 0.004 0.112 NA
#> GSM25657 1 0.2521 0.7766 0.896 0.000 0.008 0.028 0.012 NA
#> GSM25658 1 0.6292 0.4292 0.552 0.000 0.256 0.024 0.020 NA
#> GSM25659 1 0.3363 0.7453 0.828 0.000 0.008 0.008 0.032 NA
#> GSM25660 1 0.2904 0.7599 0.852 0.000 0.008 0.112 0.000 NA
#> GSM25661 1 0.3046 0.7589 0.848 0.000 0.008 0.112 0.004 NA
#> GSM25662 2 0.5155 0.1855 0.000 0.588 0.004 0.004 0.324 NA
#> GSM25663 5 0.4648 0.4136 0.000 0.408 0.000 0.000 0.548 NA
#> GSM25680 5 0.3695 0.6894 0.000 0.244 0.000 0.000 0.732 NA
#> GSM25681 5 0.3731 0.6978 0.000 0.240 0.004 0.000 0.736 NA
#> GSM25682 2 0.5232 -0.0834 0.000 0.500 0.000 0.016 0.428 NA
#> GSM25683 2 0.5232 -0.0834 0.000 0.500 0.000 0.016 0.428 NA
#> GSM25684 2 0.5259 0.1417 0.000 0.564 0.004 0.004 0.344 NA
#> GSM25685 2 0.5865 0.1445 0.000 0.524 0.016 0.008 0.344 NA
#> GSM25686 2 0.5298 -0.0659 0.000 0.504 0.000 0.020 0.420 NA
#> GSM25687 2 0.5302 -0.0665 0.000 0.500 0.000 0.020 0.424 NA
#> GSM48664 4 0.4861 0.7428 0.108 0.000 0.000 0.700 0.020 NA
#> GSM48665 4 0.5290 0.2720 0.392 0.000 0.000 0.520 0.008 NA
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 genotype/variation(p) k
#> MAD:kmeans 94 2.03e-05 2
#> MAD:kmeans 95 2.34e-04 3
#> MAD:kmeans 90 2.55e-07 4
#> MAD:kmeans 70 7.12e-12 5
#> MAD:kmeans 68 7.05e-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["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 100 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 0.843 0.914 0.962 0.5052 0.495 0.495
#> 3 3 0.738 0.777 0.891 0.2996 0.817 0.645
#> 4 4 0.495 0.473 0.698 0.1301 0.961 0.888
#> 5 5 0.489 0.372 0.585 0.0715 0.867 0.600
#> 6 6 0.518 0.293 0.596 0.0413 0.889 0.581
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
#> GSM25548 2 0.0000 0.955 0.000 1.000
#> GSM25549 2 0.0000 0.955 0.000 1.000
#> GSM25550 2 0.0000 0.955 0.000 1.000
#> GSM25551 2 0.0000 0.955 0.000 1.000
#> GSM25570 2 0.0000 0.955 0.000 1.000
#> GSM25571 2 0.0000 0.955 0.000 1.000
#> GSM25358 1 0.0672 0.959 0.992 0.008
#> GSM25359 2 0.6887 0.778 0.184 0.816
#> GSM25360 1 0.0000 0.964 1.000 0.000
#> GSM25361 2 0.9881 0.286 0.436 0.564
#> GSM25377 1 0.0000 0.964 1.000 0.000
#> GSM25378 1 0.0672 0.958 0.992 0.008
#> GSM25401 1 0.6343 0.806 0.840 0.160
#> GSM25402 1 0.4562 0.880 0.904 0.096
#> GSM25349 2 0.0000 0.955 0.000 1.000
#> GSM25350 2 0.0000 0.955 0.000 1.000
#> GSM25356 1 0.4562 0.880 0.904 0.096
#> GSM25357 2 0.0000 0.955 0.000 1.000
#> GSM25385 1 0.0000 0.964 1.000 0.000
#> GSM25386 1 0.0000 0.964 1.000 0.000
#> GSM25399 1 0.0000 0.964 1.000 0.000
#> GSM25400 1 0.0000 0.964 1.000 0.000
#> GSM48659 2 0.0000 0.955 0.000 1.000
#> GSM48660 2 0.0000 0.955 0.000 1.000
#> GSM25409 2 0.0000 0.955 0.000 1.000
#> GSM25410 1 0.0000 0.964 1.000 0.000
#> GSM25426 2 0.0000 0.955 0.000 1.000
#> GSM25427 1 0.0938 0.955 0.988 0.012
#> GSM25540 2 0.7674 0.726 0.224 0.776
#> GSM25541 2 0.8763 0.615 0.296 0.704
#> GSM25542 2 0.0000 0.955 0.000 1.000
#> GSM25543 2 0.0672 0.949 0.008 0.992
#> GSM25479 1 0.0000 0.964 1.000 0.000
#> GSM25480 1 0.0000 0.964 1.000 0.000
#> GSM25481 1 0.7815 0.708 0.768 0.232
#> GSM25482 1 0.7815 0.708 0.768 0.232
#> GSM48654 2 0.0000 0.955 0.000 1.000
#> GSM48650 2 0.0000 0.955 0.000 1.000
#> GSM48651 2 0.0000 0.955 0.000 1.000
#> GSM48652 2 0.0000 0.955 0.000 1.000
#> GSM48653 2 0.0000 0.955 0.000 1.000
#> GSM48662 2 0.0000 0.955 0.000 1.000
#> GSM48663 2 0.0000 0.955 0.000 1.000
#> GSM25524 1 0.0000 0.964 1.000 0.000
#> GSM25525 1 0.0000 0.964 1.000 0.000
#> GSM25526 1 0.0000 0.964 1.000 0.000
#> GSM25527 1 0.0000 0.964 1.000 0.000
#> GSM25528 1 0.0000 0.964 1.000 0.000
#> GSM25529 1 0.0000 0.964 1.000 0.000
#> GSM25530 1 0.0000 0.964 1.000 0.000
#> GSM25531 1 0.0000 0.964 1.000 0.000
#> GSM48661 2 0.0000 0.955 0.000 1.000
#> GSM25561 1 0.0000 0.964 1.000 0.000
#> GSM25562 1 0.0000 0.964 1.000 0.000
#> GSM25563 1 0.0000 0.964 1.000 0.000
#> GSM25564 1 0.9323 0.495 0.652 0.348
#> GSM25565 2 0.0000 0.955 0.000 1.000
#> GSM25566 2 0.0000 0.955 0.000 1.000
#> GSM25568 2 0.8955 0.528 0.312 0.688
#> GSM25569 2 0.0000 0.955 0.000 1.000
#> GSM25552 2 0.0000 0.955 0.000 1.000
#> GSM25553 2 0.0000 0.955 0.000 1.000
#> GSM25578 1 0.0000 0.964 1.000 0.000
#> GSM25579 1 0.0000 0.964 1.000 0.000
#> GSM25580 1 0.0000 0.964 1.000 0.000
#> GSM25581 1 0.0000 0.964 1.000 0.000
#> GSM48655 2 0.0000 0.955 0.000 1.000
#> GSM48656 2 0.0000 0.955 0.000 1.000
#> GSM48657 2 0.0000 0.955 0.000 1.000
#> GSM48658 2 0.0000 0.955 0.000 1.000
#> GSM25624 1 0.0000 0.964 1.000 0.000
#> GSM25625 1 0.0000 0.964 1.000 0.000
#> GSM25626 1 0.0000 0.964 1.000 0.000
#> GSM25627 1 0.3733 0.904 0.928 0.072
#> GSM25628 1 0.9635 0.314 0.612 0.388
#> GSM25629 2 0.7453 0.743 0.212 0.788
#> GSM25630 1 0.0000 0.964 1.000 0.000
#> GSM25631 2 0.5294 0.848 0.120 0.880
#> GSM25632 1 0.0000 0.964 1.000 0.000
#> GSM25633 1 0.0000 0.964 1.000 0.000
#> GSM25634 1 0.0000 0.964 1.000 0.000
#> GSM25635 1 0.0000 0.964 1.000 0.000
#> GSM25656 2 0.9000 0.578 0.316 0.684
#> GSM25657 1 0.0000 0.964 1.000 0.000
#> GSM25658 1 0.0000 0.964 1.000 0.000
#> GSM25659 1 0.0000 0.964 1.000 0.000
#> GSM25660 1 0.0000 0.964 1.000 0.000
#> GSM25661 1 0.0000 0.964 1.000 0.000
#> GSM25662 2 0.0000 0.955 0.000 1.000
#> GSM25663 2 0.0000 0.955 0.000 1.000
#> GSM25680 2 0.0000 0.955 0.000 1.000
#> GSM25681 2 0.0000 0.955 0.000 1.000
#> GSM25682 2 0.0000 0.955 0.000 1.000
#> GSM25683 2 0.0000 0.955 0.000 1.000
#> GSM25684 2 0.0000 0.955 0.000 1.000
#> GSM25685 2 0.0000 0.955 0.000 1.000
#> GSM25686 2 0.0000 0.955 0.000 1.000
#> GSM25687 2 0.0000 0.955 0.000 1.000
#> GSM48664 1 0.0000 0.964 1.000 0.000
#> GSM48665 1 0.0000 0.964 1.000 0.000
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM25548 2 0.0424 0.91946 0.000 0.992 0.008
#> GSM25549 2 0.0424 0.91946 0.000 0.992 0.008
#> GSM25550 2 0.1453 0.91107 0.024 0.968 0.008
#> GSM25551 2 0.0592 0.92009 0.000 0.988 0.012
#> GSM25570 2 0.0424 0.91946 0.000 0.992 0.008
#> GSM25571 2 0.0424 0.91946 0.000 0.992 0.008
#> GSM25358 1 0.8010 0.15668 0.548 0.068 0.384
#> GSM25359 3 0.7187 -0.09290 0.024 0.480 0.496
#> GSM25360 3 0.1529 0.83121 0.040 0.000 0.960
#> GSM25361 3 0.2879 0.80878 0.024 0.052 0.924
#> GSM25377 1 0.0237 0.86208 0.996 0.000 0.004
#> GSM25378 1 0.0848 0.85749 0.984 0.008 0.008
#> GSM25401 1 0.7260 0.46158 0.636 0.048 0.316
#> GSM25402 1 0.4953 0.74383 0.808 0.016 0.176
#> GSM25349 2 0.1774 0.91766 0.016 0.960 0.024
#> GSM25350 2 0.1015 0.91910 0.008 0.980 0.012
#> GSM25356 1 0.1267 0.85062 0.972 0.024 0.004
#> GSM25357 2 0.2063 0.89882 0.044 0.948 0.008
#> GSM25385 3 0.4702 0.73182 0.212 0.000 0.788
#> GSM25386 3 0.1643 0.83207 0.044 0.000 0.956
#> GSM25399 1 0.0000 0.86278 1.000 0.000 0.000
#> GSM25400 1 0.0747 0.86698 0.984 0.000 0.016
#> GSM48659 2 0.1753 0.91480 0.000 0.952 0.048
#> GSM48660 2 0.0661 0.92084 0.004 0.988 0.008
#> GSM25409 2 0.1170 0.91886 0.008 0.976 0.016
#> GSM25410 3 0.1964 0.82943 0.056 0.000 0.944
#> GSM25426 2 0.2448 0.89536 0.000 0.924 0.076
#> GSM25427 1 0.0848 0.85719 0.984 0.008 0.008
#> GSM25540 3 0.0983 0.82320 0.004 0.016 0.980
#> GSM25541 3 0.1129 0.82191 0.004 0.020 0.976
#> GSM25542 2 0.6754 0.34756 0.012 0.556 0.432
#> GSM25543 2 0.6682 0.16936 0.008 0.504 0.488
#> GSM25479 1 0.1529 0.86715 0.960 0.000 0.040
#> GSM25480 1 0.1289 0.86833 0.968 0.000 0.032
#> GSM25481 1 0.1453 0.84875 0.968 0.024 0.008
#> GSM25482 1 0.1453 0.84875 0.968 0.024 0.008
#> GSM48654 2 0.1964 0.91169 0.000 0.944 0.056
#> GSM48650 2 0.0829 0.92109 0.004 0.984 0.012
#> GSM48651 2 0.1289 0.91908 0.000 0.968 0.032
#> GSM48652 2 0.1753 0.91482 0.000 0.952 0.048
#> GSM48653 2 0.1860 0.91361 0.000 0.948 0.052
#> GSM48662 2 0.0424 0.92048 0.000 0.992 0.008
#> GSM48663 2 0.0983 0.91864 0.016 0.980 0.004
#> GSM25524 3 0.5397 0.59727 0.280 0.000 0.720
#> GSM25525 1 0.4178 0.78634 0.828 0.000 0.172
#> GSM25526 3 0.3192 0.80063 0.112 0.000 0.888
#> GSM25527 1 0.3686 0.81607 0.860 0.000 0.140
#> GSM25528 1 0.6244 0.21592 0.560 0.000 0.440
#> GSM25529 1 0.4555 0.75415 0.800 0.000 0.200
#> GSM25530 3 0.6309 -0.05676 0.500 0.000 0.500
#> GSM25531 1 0.4605 0.75021 0.796 0.000 0.204
#> GSM48661 2 0.4842 0.75432 0.000 0.776 0.224
#> GSM25561 3 0.5650 0.56143 0.312 0.000 0.688
#> GSM25562 1 0.3816 0.80002 0.852 0.000 0.148
#> GSM25563 3 0.1643 0.83078 0.044 0.000 0.956
#> GSM25564 1 0.9901 0.02424 0.404 0.296 0.300
#> GSM25565 2 0.1643 0.91737 0.000 0.956 0.044
#> GSM25566 2 0.0424 0.92100 0.000 0.992 0.008
#> GSM25568 2 0.8984 -0.00646 0.128 0.436 0.436
#> GSM25569 2 0.1643 0.91853 0.000 0.956 0.044
#> GSM25552 2 0.1015 0.91687 0.012 0.980 0.008
#> GSM25553 2 0.5643 0.67834 0.220 0.760 0.020
#> GSM25578 1 0.1529 0.86745 0.960 0.000 0.040
#> GSM25579 1 0.3551 0.82754 0.868 0.000 0.132
#> GSM25580 1 0.0892 0.86651 0.980 0.000 0.020
#> GSM25581 1 0.1289 0.86818 0.968 0.000 0.032
#> GSM48655 2 0.0000 0.91987 0.000 1.000 0.000
#> GSM48656 2 0.1529 0.91847 0.000 0.960 0.040
#> GSM48657 2 0.0475 0.92015 0.004 0.992 0.004
#> GSM48658 2 0.4178 0.82060 0.000 0.828 0.172
#> GSM25624 1 0.1411 0.86866 0.964 0.000 0.036
#> GSM25625 3 0.4504 0.74332 0.196 0.000 0.804
#> GSM25626 3 0.1163 0.83142 0.028 0.000 0.972
#> GSM25627 3 0.1751 0.82913 0.028 0.012 0.960
#> GSM25628 3 0.0848 0.82723 0.008 0.008 0.984
#> GSM25629 3 0.0829 0.82586 0.004 0.012 0.984
#> GSM25630 3 0.2165 0.82525 0.064 0.000 0.936
#> GSM25631 2 0.6540 0.39567 0.008 0.584 0.408
#> GSM25632 3 0.4605 0.73234 0.204 0.000 0.796
#> GSM25633 1 0.2066 0.86096 0.940 0.000 0.060
#> GSM25634 1 0.1411 0.86872 0.964 0.000 0.036
#> GSM25635 1 0.1289 0.86834 0.968 0.000 0.032
#> GSM25656 3 0.0983 0.82268 0.004 0.016 0.980
#> GSM25657 1 0.3551 0.81926 0.868 0.000 0.132
#> GSM25658 3 0.6260 0.14991 0.448 0.000 0.552
#> GSM25659 1 0.5882 0.51864 0.652 0.000 0.348
#> GSM25660 1 0.1529 0.86791 0.960 0.000 0.040
#> GSM25661 1 0.1163 0.86764 0.972 0.000 0.028
#> GSM25662 2 0.1529 0.91690 0.000 0.960 0.040
#> GSM25663 2 0.2711 0.87478 0.000 0.912 0.088
#> GSM25680 2 0.1643 0.91752 0.000 0.956 0.044
#> GSM25681 2 0.1643 0.91775 0.000 0.956 0.044
#> GSM25682 2 0.0000 0.91987 0.000 1.000 0.000
#> GSM25683 2 0.0000 0.91987 0.000 1.000 0.000
#> GSM25684 2 0.1411 0.91767 0.000 0.964 0.036
#> GSM25685 2 0.2356 0.90429 0.000 0.928 0.072
#> GSM25686 2 0.0000 0.91987 0.000 1.000 0.000
#> GSM25687 2 0.0000 0.91987 0.000 1.000 0.000
#> GSM48664 1 0.0237 0.86208 0.996 0.000 0.004
#> GSM48665 1 0.0000 0.86278 1.000 0.000 0.000
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM25548 2 0.4677 0.3821 0.000 0.680 0.004 0.316
#> GSM25549 2 0.4819 0.3661 0.000 0.652 0.004 0.344
#> GSM25550 2 0.5628 0.3246 0.032 0.644 0.004 0.320
#> GSM25551 2 0.3812 0.4695 0.000 0.832 0.028 0.140
#> GSM25570 2 0.4608 0.3739 0.000 0.692 0.004 0.304
#> GSM25571 2 0.4560 0.3848 0.000 0.700 0.004 0.296
#> GSM25358 1 0.9701 0.0511 0.360 0.184 0.272 0.184
#> GSM25359 3 0.8047 -0.2212 0.012 0.368 0.408 0.212
#> GSM25360 3 0.2586 0.7185 0.040 0.000 0.912 0.048
#> GSM25361 3 0.6784 0.4491 0.040 0.052 0.616 0.292
#> GSM25377 1 0.4244 0.7273 0.800 0.000 0.032 0.168
#> GSM25378 1 0.4719 0.7193 0.792 0.016 0.032 0.160
#> GSM25401 1 0.9832 0.1065 0.312 0.168 0.272 0.248
#> GSM25402 1 0.8709 0.4153 0.480 0.068 0.228 0.224
#> GSM25349 2 0.5701 0.3307 0.020 0.628 0.012 0.340
#> GSM25350 2 0.4917 0.3932 0.008 0.656 0.000 0.336
#> GSM25356 1 0.5504 0.7011 0.756 0.060 0.024 0.160
#> GSM25357 2 0.4954 0.3970 0.028 0.772 0.020 0.180
#> GSM25385 3 0.5221 0.6321 0.208 0.000 0.732 0.060
#> GSM25386 3 0.1545 0.7060 0.008 0.000 0.952 0.040
#> GSM25399 1 0.3793 0.7531 0.844 0.000 0.044 0.112
#> GSM25400 1 0.4499 0.7419 0.804 0.000 0.072 0.124
#> GSM48659 2 0.5594 0.1762 0.000 0.520 0.020 0.460
#> GSM48660 2 0.4720 0.3436 0.004 0.672 0.000 0.324
#> GSM25409 2 0.4648 0.4403 0.016 0.748 0.004 0.232
#> GSM25410 3 0.1575 0.7091 0.012 0.004 0.956 0.028
#> GSM25426 2 0.4919 0.4129 0.000 0.752 0.048 0.200
#> GSM25427 1 0.4224 0.7272 0.808 0.008 0.020 0.164
#> GSM25540 3 0.4290 0.5799 0.000 0.016 0.772 0.212
#> GSM25541 3 0.5438 0.5850 0.024 0.028 0.728 0.220
#> GSM25542 2 0.7888 -0.4355 0.000 0.368 0.288 0.344
#> GSM25543 4 0.7971 0.3911 0.004 0.252 0.364 0.380
#> GSM25479 1 0.2751 0.7596 0.904 0.000 0.056 0.040
#> GSM25480 1 0.3903 0.7545 0.844 0.000 0.076 0.080
#> GSM25481 1 0.6540 0.6224 0.660 0.080 0.024 0.236
#> GSM25482 1 0.6056 0.6588 0.700 0.068 0.020 0.212
#> GSM48654 2 0.5673 0.1469 0.000 0.528 0.024 0.448
#> GSM48650 2 0.4406 0.3729 0.000 0.700 0.000 0.300
#> GSM48651 2 0.4746 0.3206 0.000 0.632 0.000 0.368
#> GSM48652 2 0.4830 0.2622 0.000 0.608 0.000 0.392
#> GSM48653 2 0.5673 0.1713 0.000 0.528 0.024 0.448
#> GSM48662 2 0.5178 0.2693 0.004 0.600 0.004 0.392
#> GSM48663 2 0.5024 0.3163 0.008 0.632 0.000 0.360
#> GSM25524 3 0.5537 0.5653 0.256 0.000 0.688 0.056
#> GSM25525 1 0.4938 0.6914 0.772 0.000 0.148 0.080
#> GSM25526 3 0.4781 0.6475 0.212 0.000 0.752 0.036
#> GSM25527 1 0.4312 0.7241 0.812 0.000 0.132 0.056
#> GSM25528 1 0.6207 0.0796 0.496 0.000 0.452 0.052
#> GSM25529 1 0.5254 0.6312 0.724 0.000 0.220 0.056
#> GSM25530 3 0.6247 0.0900 0.428 0.000 0.516 0.056
#> GSM25531 1 0.4994 0.6624 0.744 0.000 0.208 0.048
#> GSM48661 4 0.7151 0.1455 0.000 0.420 0.132 0.448
#> GSM25561 3 0.5732 0.5585 0.264 0.000 0.672 0.064
#> GSM25562 1 0.5905 0.6848 0.700 0.000 0.156 0.144
#> GSM25563 3 0.2500 0.7213 0.044 0.000 0.916 0.040
#> GSM25564 1 0.9769 -0.2220 0.320 0.212 0.168 0.300
#> GSM25565 2 0.4673 0.4242 0.000 0.700 0.008 0.292
#> GSM25566 2 0.3751 0.4994 0.000 0.800 0.004 0.196
#> GSM25568 4 0.8455 0.3696 0.068 0.224 0.188 0.520
#> GSM25569 2 0.5126 0.2443 0.000 0.552 0.004 0.444
#> GSM25552 2 0.6034 0.2808 0.036 0.592 0.008 0.364
#> GSM25553 2 0.7531 0.0772 0.120 0.460 0.016 0.404
#> GSM25578 1 0.2983 0.7556 0.892 0.000 0.068 0.040
#> GSM25579 1 0.6392 0.6250 0.676 0.008 0.152 0.164
#> GSM25580 1 0.1733 0.7637 0.948 0.000 0.024 0.028
#> GSM25581 1 0.2222 0.7607 0.924 0.000 0.060 0.016
#> GSM48655 2 0.3400 0.4813 0.000 0.820 0.000 0.180
#> GSM48656 2 0.5435 0.1968 0.000 0.564 0.016 0.420
#> GSM48657 2 0.3764 0.4377 0.000 0.784 0.000 0.216
#> GSM48658 4 0.6624 0.0287 0.008 0.400 0.064 0.528
#> GSM25624 1 0.3004 0.7639 0.892 0.000 0.048 0.060
#> GSM25625 3 0.4365 0.6943 0.188 0.000 0.784 0.028
#> GSM25626 3 0.1624 0.7133 0.020 0.000 0.952 0.028
#> GSM25627 3 0.6667 0.5415 0.060 0.072 0.688 0.180
#> GSM25628 3 0.2662 0.6823 0.000 0.016 0.900 0.084
#> GSM25629 3 0.5298 0.5550 0.004 0.072 0.748 0.176
#> GSM25630 3 0.2644 0.7224 0.060 0.000 0.908 0.032
#> GSM25631 4 0.8475 0.2865 0.060 0.216 0.216 0.508
#> GSM25632 3 0.4194 0.6901 0.172 0.000 0.800 0.028
#> GSM25633 1 0.2843 0.7528 0.892 0.000 0.088 0.020
#> GSM25634 1 0.2256 0.7634 0.924 0.000 0.056 0.020
#> GSM25635 1 0.1837 0.7639 0.944 0.000 0.028 0.028
#> GSM25656 3 0.4274 0.6220 0.000 0.044 0.808 0.148
#> GSM25657 1 0.4880 0.6907 0.760 0.000 0.188 0.052
#> GSM25658 3 0.6327 0.0205 0.444 0.000 0.496 0.060
#> GSM25659 1 0.7054 0.4751 0.572 0.000 0.232 0.196
#> GSM25660 1 0.2739 0.7604 0.904 0.000 0.060 0.036
#> GSM25661 1 0.2408 0.7641 0.920 0.000 0.044 0.036
#> GSM25662 2 0.4220 0.4342 0.000 0.748 0.004 0.248
#> GSM25663 2 0.5596 0.3445 0.000 0.696 0.068 0.236
#> GSM25680 2 0.5487 0.2976 0.000 0.580 0.020 0.400
#> GSM25681 2 0.6347 0.2168 0.000 0.524 0.064 0.412
#> GSM25682 2 0.0817 0.5070 0.000 0.976 0.000 0.024
#> GSM25683 2 0.1022 0.5064 0.000 0.968 0.000 0.032
#> GSM25684 2 0.4222 0.4232 0.000 0.728 0.000 0.272
#> GSM25685 2 0.5446 0.3586 0.000 0.680 0.044 0.276
#> GSM25686 2 0.0707 0.5043 0.000 0.980 0.000 0.020
#> GSM25687 2 0.1022 0.5079 0.000 0.968 0.000 0.032
#> GSM48664 1 0.2741 0.7509 0.892 0.000 0.012 0.096
#> GSM48665 1 0.2101 0.7571 0.928 0.000 0.012 0.060
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM25548 5 0.559 0.4000 0.000 0.280 0.004 0.096 0.620
#> GSM25549 5 0.619 0.3754 0.000 0.328 0.008 0.124 0.540
#> GSM25550 5 0.681 0.3597 0.008 0.284 0.008 0.192 0.508
#> GSM25551 5 0.447 0.4096 0.000 0.132 0.012 0.080 0.776
#> GSM25570 5 0.590 0.3811 0.000 0.312 0.004 0.112 0.572
#> GSM25571 5 0.553 0.4006 0.000 0.268 0.004 0.096 0.632
#> GSM25358 4 0.901 0.3207 0.180 0.064 0.204 0.412 0.140
#> GSM25359 3 0.827 0.0510 0.000 0.176 0.364 0.164 0.296
#> GSM25360 3 0.389 0.6682 0.100 0.032 0.828 0.040 0.000
#> GSM25361 3 0.771 0.5524 0.104 0.184 0.568 0.092 0.052
#> GSM25377 4 0.499 0.4946 0.452 0.008 0.016 0.524 0.000
#> GSM25378 4 0.512 0.5891 0.380 0.008 0.012 0.588 0.012
#> GSM25401 4 0.718 0.4384 0.088 0.036 0.140 0.620 0.116
#> GSM25402 4 0.693 0.5406 0.168 0.032 0.144 0.616 0.040
#> GSM25349 5 0.687 0.0114 0.000 0.352 0.016 0.184 0.448
#> GSM25350 5 0.638 0.1096 0.000 0.384 0.012 0.120 0.484
#> GSM25356 4 0.565 0.6283 0.332 0.024 0.000 0.596 0.048
#> GSM25357 5 0.565 0.2903 0.004 0.104 0.020 0.188 0.684
#> GSM25385 3 0.599 0.5326 0.188 0.020 0.640 0.152 0.000
#> GSM25386 3 0.300 0.6640 0.040 0.016 0.880 0.064 0.000
#> GSM25399 1 0.497 -0.1824 0.564 0.000 0.032 0.404 0.000
#> GSM25400 4 0.574 0.3319 0.432 0.004 0.072 0.492 0.000
#> GSM48659 2 0.518 0.4104 0.000 0.588 0.028 0.012 0.372
#> GSM48660 5 0.544 -0.3612 0.000 0.468 0.004 0.048 0.480
#> GSM25409 5 0.672 0.2740 0.004 0.300 0.020 0.148 0.528
#> GSM25410 3 0.297 0.6633 0.032 0.016 0.880 0.072 0.000
#> GSM25426 5 0.631 0.2114 0.000 0.200 0.052 0.116 0.632
#> GSM25427 4 0.529 0.5549 0.428 0.020 0.004 0.536 0.012
#> GSM25540 3 0.528 0.6133 0.008 0.136 0.744 0.068 0.044
#> GSM25541 3 0.569 0.6243 0.040 0.160 0.716 0.064 0.020
#> GSM25542 2 0.790 0.2870 0.000 0.364 0.284 0.072 0.280
#> GSM25543 3 0.817 -0.2080 0.004 0.312 0.368 0.100 0.216
#> GSM25479 1 0.371 0.5915 0.816 0.004 0.044 0.136 0.000
#> GSM25480 1 0.450 0.5889 0.772 0.016 0.064 0.148 0.000
#> GSM25481 4 0.611 0.6308 0.320 0.044 0.004 0.584 0.048
#> GSM25482 4 0.565 0.6237 0.368 0.036 0.000 0.568 0.028
#> GSM48654 2 0.483 0.5279 0.000 0.644 0.024 0.008 0.324
#> GSM48650 5 0.524 -0.1830 0.000 0.372 0.004 0.044 0.580
#> GSM48651 2 0.508 0.3557 0.000 0.496 0.008 0.020 0.476
#> GSM48652 2 0.487 0.4740 0.000 0.588 0.016 0.008 0.388
#> GSM48653 2 0.561 0.5237 0.000 0.612 0.040 0.032 0.316
#> GSM48662 2 0.519 0.4355 0.000 0.596 0.008 0.036 0.360
#> GSM48663 5 0.609 -0.2368 0.000 0.416 0.000 0.124 0.460
#> GSM25524 3 0.589 0.0956 0.432 0.008 0.484 0.076 0.000
#> GSM25525 1 0.422 0.5828 0.796 0.012 0.120 0.072 0.000
#> GSM25526 3 0.701 0.3233 0.296 0.020 0.488 0.192 0.004
#> GSM25527 1 0.413 0.5916 0.804 0.012 0.108 0.076 0.000
#> GSM25528 1 0.522 0.4390 0.644 0.004 0.288 0.064 0.000
#> GSM25529 1 0.377 0.5970 0.828 0.012 0.104 0.056 0.000
#> GSM25530 1 0.627 0.2565 0.520 0.008 0.344 0.128 0.000
#> GSM25531 1 0.508 0.5472 0.712 0.004 0.144 0.140 0.000
#> GSM48661 2 0.626 0.4934 0.000 0.632 0.136 0.040 0.192
#> GSM25561 3 0.631 0.3185 0.360 0.020 0.520 0.100 0.000
#> GSM25562 1 0.642 0.2922 0.596 0.036 0.128 0.240 0.000
#> GSM25563 3 0.390 0.6666 0.096 0.028 0.828 0.048 0.000
#> GSM25564 1 0.977 -0.1018 0.292 0.232 0.176 0.156 0.144
#> GSM25565 5 0.606 -0.1902 0.000 0.408 0.036 0.048 0.508
#> GSM25566 5 0.509 0.3167 0.000 0.288 0.008 0.048 0.656
#> GSM25568 2 0.814 0.3019 0.048 0.516 0.168 0.180 0.088
#> GSM25569 2 0.530 0.4198 0.000 0.660 0.020 0.048 0.272
#> GSM25552 5 0.660 0.3415 0.004 0.352 0.004 0.164 0.476
#> GSM25553 5 0.833 0.2410 0.076 0.316 0.020 0.224 0.364
#> GSM25578 1 0.230 0.6010 0.908 0.000 0.040 0.052 0.000
#> GSM25579 1 0.608 0.4889 0.692 0.064 0.076 0.152 0.016
#> GSM25580 1 0.249 0.5512 0.872 0.000 0.004 0.124 0.000
#> GSM25581 1 0.319 0.5754 0.852 0.004 0.032 0.112 0.000
#> GSM48655 5 0.480 0.1421 0.000 0.272 0.000 0.052 0.676
#> GSM48656 2 0.543 0.5109 0.004 0.644 0.024 0.036 0.292
#> GSM48657 5 0.548 -0.1395 0.000 0.352 0.004 0.064 0.580
#> GSM48658 2 0.653 0.4641 0.008 0.624 0.096 0.056 0.216
#> GSM25624 1 0.535 0.3644 0.660 0.016 0.060 0.264 0.000
#> GSM25625 3 0.601 0.4837 0.252 0.008 0.600 0.140 0.000
#> GSM25626 3 0.302 0.6634 0.036 0.008 0.872 0.084 0.000
#> GSM25627 3 0.855 0.4130 0.080 0.124 0.468 0.236 0.092
#> GSM25628 3 0.321 0.6632 0.012 0.076 0.872 0.032 0.008
#> GSM25629 3 0.776 0.4099 0.020 0.232 0.520 0.132 0.096
#> GSM25630 3 0.409 0.6602 0.104 0.020 0.812 0.064 0.000
#> GSM25631 2 0.847 -0.0150 0.068 0.492 0.144 0.104 0.192
#> GSM25632 3 0.556 0.5287 0.236 0.012 0.656 0.096 0.000
#> GSM25633 1 0.339 0.5974 0.852 0.008 0.060 0.080 0.000
#> GSM25634 1 0.389 0.5591 0.804 0.008 0.040 0.148 0.000
#> GSM25635 1 0.401 0.5294 0.788 0.016 0.024 0.172 0.000
#> GSM25656 3 0.569 0.6320 0.024 0.136 0.724 0.080 0.036
#> GSM25657 1 0.427 0.5838 0.784 0.004 0.116 0.096 0.000
#> GSM25658 1 0.724 0.1177 0.416 0.012 0.320 0.244 0.008
#> GSM25659 1 0.751 0.3778 0.560 0.096 0.176 0.152 0.016
#> GSM25660 1 0.312 0.5897 0.860 0.016 0.016 0.108 0.000
#> GSM25661 1 0.284 0.5621 0.868 0.004 0.016 0.112 0.000
#> GSM25662 5 0.515 0.1670 0.000 0.268 0.024 0.036 0.672
#> GSM25663 5 0.669 0.2885 0.004 0.216 0.116 0.060 0.604
#> GSM25680 5 0.667 0.3120 0.000 0.352 0.036 0.108 0.504
#> GSM25681 5 0.728 0.2996 0.000 0.336 0.076 0.120 0.468
#> GSM25682 5 0.208 0.4017 0.000 0.032 0.004 0.040 0.924
#> GSM25683 5 0.183 0.3918 0.000 0.028 0.004 0.032 0.936
#> GSM25684 5 0.484 0.1355 0.000 0.304 0.024 0.012 0.660
#> GSM25685 5 0.588 0.0880 0.000 0.288 0.040 0.056 0.616
#> GSM25686 5 0.175 0.3994 0.000 0.028 0.000 0.036 0.936
#> GSM25687 5 0.280 0.3910 0.000 0.068 0.004 0.044 0.884
#> GSM48664 1 0.425 -0.1091 0.624 0.004 0.000 0.372 0.000
#> GSM48665 1 0.380 0.2195 0.700 0.000 0.000 0.300 0.000
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM25548 5 0.282 0.4186 0.000 0.096 0.004 0.004 0.864 0.032
#> GSM25549 5 0.421 0.4269 0.004 0.132 0.000 0.028 0.776 0.060
#> GSM25550 5 0.433 0.4269 0.020 0.056 0.008 0.032 0.800 0.084
#> GSM25551 5 0.699 -0.4109 0.000 0.232 0.032 0.028 0.460 0.248
#> GSM25570 5 0.245 0.4399 0.004 0.068 0.000 0.004 0.892 0.032
#> GSM25571 5 0.270 0.4284 0.000 0.092 0.004 0.004 0.872 0.028
#> GSM25358 4 0.864 0.1265 0.088 0.020 0.216 0.376 0.108 0.192
#> GSM25359 3 0.853 -0.0568 0.004 0.108 0.284 0.084 0.276 0.244
#> GSM25360 3 0.574 0.5567 0.168 0.004 0.672 0.032 0.032 0.092
#> GSM25361 3 0.852 0.3905 0.108 0.056 0.368 0.024 0.220 0.224
#> GSM25377 4 0.452 0.5321 0.200 0.004 0.020 0.724 0.000 0.052
#> GSM25378 4 0.560 0.5396 0.196 0.004 0.040 0.664 0.012 0.084
#> GSM25401 4 0.731 0.2691 0.048 0.032 0.164 0.496 0.016 0.244
#> GSM25402 4 0.634 0.4433 0.072 0.008 0.136 0.620 0.012 0.152
#> GSM25349 2 0.774 0.0917 0.004 0.416 0.016 0.140 0.208 0.216
#> GSM25350 2 0.721 0.1001 0.000 0.444 0.004 0.116 0.248 0.188
#> GSM25356 4 0.409 0.5700 0.144 0.000 0.028 0.784 0.032 0.012
#> GSM25357 5 0.815 -0.3704 0.004 0.268 0.024 0.200 0.324 0.180
#> GSM25385 3 0.661 0.4509 0.172 0.000 0.568 0.160 0.012 0.088
#> GSM25386 3 0.403 0.6031 0.064 0.004 0.812 0.036 0.008 0.076
#> GSM25399 4 0.452 0.3449 0.336 0.000 0.032 0.624 0.000 0.008
#> GSM25400 4 0.630 0.2816 0.344 0.000 0.064 0.504 0.008 0.080
#> GSM48659 2 0.607 0.2813 0.000 0.592 0.020 0.020 0.204 0.164
#> GSM48660 2 0.335 0.3982 0.000 0.844 0.000 0.036 0.064 0.056
#> GSM25409 5 0.788 -0.0318 0.020 0.272 0.020 0.108 0.416 0.164
#> GSM25410 3 0.390 0.6083 0.036 0.000 0.820 0.052 0.016 0.076
#> GSM25426 6 0.759 0.0000 0.000 0.332 0.044 0.048 0.244 0.332
#> GSM25427 4 0.511 0.5527 0.176 0.008 0.028 0.712 0.012 0.064
#> GSM25540 3 0.641 0.5227 0.020 0.064 0.576 0.000 0.104 0.236
#> GSM25541 3 0.703 0.5363 0.068 0.052 0.528 0.004 0.088 0.260
#> GSM25542 2 0.761 0.0558 0.000 0.404 0.220 0.032 0.080 0.264
#> GSM25543 3 0.864 0.0123 0.016 0.268 0.316 0.072 0.112 0.216
#> GSM25479 1 0.466 0.5192 0.716 0.000 0.032 0.208 0.008 0.036
#> GSM25480 1 0.566 0.5281 0.684 0.000 0.048 0.152 0.036 0.080
#> GSM25481 4 0.557 0.5477 0.112 0.044 0.020 0.716 0.024 0.084
#> GSM25482 4 0.573 0.5496 0.156 0.032 0.012 0.688 0.036 0.076
#> GSM48654 2 0.436 0.4169 0.000 0.760 0.016 0.004 0.100 0.120
#> GSM48650 2 0.456 0.2448 0.000 0.752 0.000 0.044 0.096 0.108
#> GSM48651 2 0.347 0.3841 0.000 0.820 0.000 0.012 0.112 0.056
#> GSM48652 2 0.314 0.4207 0.000 0.848 0.004 0.004 0.072 0.072
#> GSM48653 2 0.425 0.3939 0.000 0.768 0.008 0.012 0.076 0.136
#> GSM48662 2 0.418 0.4135 0.000 0.752 0.000 0.008 0.160 0.080
#> GSM48663 2 0.600 0.2758 0.000 0.632 0.004 0.120 0.096 0.148
#> GSM25524 1 0.589 0.0907 0.508 0.000 0.368 0.052 0.000 0.072
#> GSM25525 1 0.385 0.5700 0.816 0.000 0.088 0.032 0.008 0.056
#> GSM25526 3 0.697 0.2523 0.304 0.004 0.428 0.068 0.000 0.196
#> GSM25527 1 0.464 0.5545 0.752 0.000 0.076 0.092 0.000 0.080
#> GSM25528 1 0.478 0.5095 0.692 0.000 0.224 0.040 0.000 0.044
#> GSM25529 1 0.358 0.5682 0.832 0.000 0.080 0.032 0.004 0.052
#> GSM25530 1 0.591 0.3759 0.572 0.000 0.272 0.108 0.000 0.048
#> GSM25531 1 0.473 0.5359 0.736 0.000 0.100 0.120 0.000 0.044
#> GSM48661 2 0.594 0.2961 0.004 0.612 0.072 0.012 0.052 0.248
#> GSM25561 3 0.674 0.1432 0.344 0.000 0.452 0.108 0.004 0.092
#> GSM25562 1 0.740 0.2045 0.460 0.012 0.132 0.276 0.012 0.108
#> GSM25563 3 0.531 0.5650 0.128 0.000 0.692 0.084 0.000 0.096
#> GSM25564 1 0.959 -0.0732 0.244 0.204 0.092 0.144 0.088 0.228
#> GSM25565 2 0.670 0.1514 0.000 0.496 0.024 0.024 0.220 0.236
#> GSM25566 2 0.639 -0.2777 0.000 0.416 0.000 0.036 0.388 0.160
#> GSM25568 2 0.919 0.1175 0.060 0.332 0.152 0.160 0.080 0.216
#> GSM25569 2 0.574 0.3773 0.000 0.640 0.016 0.024 0.164 0.156
#> GSM25552 5 0.497 0.4134 0.020 0.092 0.000 0.024 0.728 0.136
#> GSM25553 5 0.606 0.3650 0.064 0.048 0.004 0.080 0.668 0.136
#> GSM25578 1 0.391 0.5518 0.776 0.000 0.040 0.164 0.000 0.020
#> GSM25579 1 0.644 0.4620 0.644 0.004 0.068 0.072 0.100 0.112
#> GSM25580 1 0.408 0.4233 0.688 0.000 0.008 0.284 0.000 0.020
#> GSM25581 1 0.392 0.5037 0.748 0.000 0.020 0.212 0.000 0.020
#> GSM48655 2 0.551 0.0580 0.000 0.600 0.000 0.016 0.252 0.132
#> GSM48656 2 0.484 0.4191 0.004 0.736 0.024 0.008 0.096 0.132
#> GSM48657 2 0.505 0.1960 0.000 0.692 0.000 0.032 0.168 0.108
#> GSM48658 2 0.708 0.2712 0.024 0.524 0.080 0.004 0.148 0.220
#> GSM25624 1 0.635 0.0196 0.468 0.004 0.048 0.388 0.008 0.084
#> GSM25625 3 0.614 0.4441 0.248 0.004 0.580 0.084 0.000 0.084
#> GSM25626 3 0.343 0.6088 0.060 0.000 0.832 0.020 0.000 0.088
#> GSM25627 3 0.823 0.2891 0.096 0.100 0.368 0.092 0.016 0.328
#> GSM25628 3 0.417 0.6073 0.016 0.028 0.776 0.008 0.012 0.160
#> GSM25629 3 0.695 0.3126 0.028 0.124 0.424 0.016 0.024 0.384
#> GSM25630 3 0.450 0.5645 0.148 0.000 0.748 0.056 0.000 0.048
#> GSM25631 5 0.835 0.1883 0.084 0.168 0.108 0.020 0.440 0.180
#> GSM25632 3 0.542 0.3819 0.296 0.000 0.596 0.080 0.000 0.028
#> GSM25633 1 0.452 0.5115 0.724 0.000 0.048 0.196 0.000 0.032
#> GSM25634 1 0.533 0.3903 0.624 0.000 0.040 0.280 0.004 0.052
#> GSM25635 1 0.498 0.3470 0.608 0.000 0.020 0.324 0.000 0.048
#> GSM25656 3 0.611 0.5476 0.020 0.060 0.600 0.020 0.032 0.268
#> GSM25657 1 0.530 0.4658 0.632 0.000 0.096 0.248 0.000 0.024
#> GSM25658 1 0.771 0.1072 0.376 0.004 0.244 0.180 0.004 0.192
#> GSM25659 1 0.765 0.3738 0.540 0.024 0.136 0.120 0.060 0.120
#> GSM25660 1 0.412 0.5279 0.768 0.000 0.020 0.168 0.008 0.036
#> GSM25661 1 0.454 0.4665 0.696 0.000 0.016 0.236 0.000 0.052
#> GSM25662 2 0.662 -0.2415 0.000 0.468 0.020 0.020 0.308 0.184
#> GSM25663 5 0.720 -0.0699 0.004 0.288 0.056 0.008 0.404 0.240
#> GSM25680 5 0.483 0.3571 0.004 0.160 0.016 0.004 0.720 0.096
#> GSM25681 5 0.515 0.3843 0.016 0.080 0.044 0.016 0.744 0.100
#> GSM25682 5 0.585 -0.1920 0.000 0.400 0.000 0.016 0.460 0.124
#> GSM25683 5 0.591 -0.2268 0.000 0.424 0.000 0.020 0.436 0.120
#> GSM25684 2 0.622 -0.2173 0.000 0.476 0.008 0.012 0.332 0.172
#> GSM25685 2 0.697 -0.5126 0.000 0.416 0.028 0.020 0.260 0.276
#> GSM25686 5 0.596 -0.1904 0.000 0.416 0.000 0.020 0.436 0.128
#> GSM25687 5 0.593 -0.1841 0.000 0.408 0.000 0.020 0.448 0.124
#> GSM48664 4 0.433 0.3310 0.336 0.000 0.004 0.636 0.004 0.020
#> GSM48665 4 0.433 0.0522 0.464 0.000 0.000 0.516 0.000 0.020
Heatmaps for the consensus matrix. It visualizes the probability of two samples to be in a same group.
consensus_heatmap(res, k = 2)
consensus_heatmap(res, k = 3)
consensus_heatmap(res, k = 4)
consensus_heatmap(res, k = 5)
consensus_heatmap(res, k = 6)
Heatmaps for the membership of samples in all partitions to see how consistent they are:
membership_heatmap(res, k = 2)
membership_heatmap(res, k = 3)
membership_heatmap(res, k = 4)
membership_heatmap(res, k = 5)
membership_heatmap(res, k = 6)
As soon as we have had the classes for columns, we can look for signatures which are significantly different between classes which can be candidate marks for certain classes. Following are the heatmaps for signatures.
Signature heatmaps where rows are scaled:
get_signatures(res, k = 2)
get_signatures(res, k = 3)
get_signatures(res, k = 4)
get_signatures(res, k = 5)
get_signatures(res, k = 6)
Signature heatmaps where rows are not scaled:
get_signatures(res, k = 2, scale_rows = FALSE)
get_signatures(res, k = 3, scale_rows = FALSE)
get_signatures(res, k = 4, scale_rows = FALSE)
get_signatures(res, k = 5, scale_rows = FALSE)
get_signatures(res, k = 6, scale_rows = FALSE)
Compare the overlap of signatures from different k:
compare_signatures(res)
get_signature() returns a data frame invisibly. TO get the list of signatures, the function
call should be assigned to a variable explicitly. In following code, if plot argument is set
to FALSE, no heatmap is plotted while only the differential analysis is performed.
# code only for demonstration
tb = get_signature(res, k = ..., plot = FALSE)
An example of the output of tb is:
#> which_row fdr mean_1 mean_2 scaled_mean_1 scaled_mean_2 km
#> 1 38 0.042760348 8.373488 9.131774 -0.5533452 0.5164555 1
#> 2 40 0.018707592 7.106213 8.469186 -0.6173731 0.5762149 1
#> 3 55 0.019134737 10.221463 11.207825 -0.6159697 0.5749050 1
#> 4 59 0.006059896 5.921854 7.869574 -0.6899429 0.6439467 1
#> 5 60 0.018055526 8.928898 10.211722 -0.6204761 0.5791110 1
#> 6 98 0.009384629 15.714769 14.887706 0.6635654 -0.6193277 2
...
The columns in tb are:
which_row: row indices corresponding to the input matrix.fdr: FDR for the differential test. mean_x: The mean value in group x.scaled_mean_x: The mean value in group x after rows are scaled.km: Row groups if k-means clustering is applied to rows.UMAP plot which shows how samples are separated.
dimension_reduction(res, k = 2, method = "UMAP")
dimension_reduction(res, k = 3, method = "UMAP")
dimension_reduction(res, k = 4, method = "UMAP")
dimension_reduction(res, k = 5, method = "UMAP")
dimension_reduction(res, k = 6, method = "UMAP")
Following heatmap shows how subgroups are split when increasing k:
collect_classes(res)
Test correlation between subgroups and known annotations. If the known annotation is numeric, one-way ANOVA test is applied, and if the known annotation is discrete, chi-squared contingency table test is applied.
test_to_known_factors(res)
#> n genotype/variation(p) k
#> MAD:skmeans 97 1.29e-05 2
#> MAD:skmeans 89 2.70e-05 3
#> MAD:skmeans 50 1.06e-05 4
#> MAD:skmeans 37 1.35e-03 5
#> MAD:skmeans 27 8.71e-02 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 100 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 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 0.461 0.819 0.903 0.4615 0.535 0.535
#> 3 3 0.395 0.714 0.824 0.4297 0.746 0.544
#> 4 4 0.487 0.645 0.776 0.1178 0.893 0.692
#> 5 5 0.530 0.480 0.745 0.0484 0.964 0.868
#> 6 6 0.559 0.543 0.746 0.0281 0.921 0.692
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
#> GSM25548 2 0.0000 0.912 0.000 1.000
#> GSM25549 2 0.0000 0.912 0.000 1.000
#> GSM25550 2 0.0000 0.912 0.000 1.000
#> GSM25551 2 0.7139 0.767 0.196 0.804
#> GSM25570 2 0.0000 0.912 0.000 1.000
#> GSM25571 2 0.0000 0.912 0.000 1.000
#> GSM25358 2 0.0938 0.912 0.012 0.988
#> GSM25359 2 0.3733 0.879 0.072 0.928
#> GSM25360 1 0.9286 0.622 0.656 0.344
#> GSM25361 2 0.4939 0.855 0.108 0.892
#> GSM25377 2 0.8267 0.659 0.260 0.740
#> GSM25378 2 0.7219 0.742 0.200 0.800
#> GSM25401 1 0.3274 0.859 0.940 0.060
#> GSM25402 1 0.9732 0.401 0.596 0.404
#> GSM25349 2 0.0672 0.913 0.008 0.992
#> GSM25350 2 0.0000 0.912 0.000 1.000
#> GSM25356 2 0.5946 0.809 0.144 0.856
#> GSM25357 2 0.0376 0.912 0.004 0.996
#> GSM25385 1 0.2236 0.859 0.964 0.036
#> GSM25386 1 0.7745 0.749 0.772 0.228
#> GSM25399 1 0.2948 0.856 0.948 0.052
#> GSM25400 1 0.6531 0.814 0.832 0.168
#> GSM48659 2 0.1414 0.910 0.020 0.980
#> GSM48660 2 0.0672 0.912 0.008 0.992
#> GSM25409 2 0.0376 0.912 0.004 0.996
#> GSM25410 1 0.6343 0.832 0.840 0.160
#> GSM25426 2 0.9491 0.413 0.368 0.632
#> GSM25427 2 0.3431 0.876 0.064 0.936
#> GSM25540 2 0.8207 0.691 0.256 0.744
#> GSM25541 2 0.9896 0.164 0.440 0.560
#> GSM25542 2 0.1843 0.907 0.028 0.972
#> GSM25543 2 0.0938 0.913 0.012 0.988
#> GSM25479 1 0.9129 0.543 0.672 0.328
#> GSM25480 2 0.7815 0.707 0.232 0.768
#> GSM25481 2 0.0672 0.910 0.008 0.992
#> GSM25482 2 0.6048 0.806 0.148 0.852
#> GSM48654 2 0.0938 0.912 0.012 0.988
#> GSM48650 2 0.0376 0.912 0.004 0.996
#> GSM48651 2 0.1184 0.912 0.016 0.984
#> GSM48652 2 0.1843 0.906 0.028 0.972
#> GSM48653 2 0.5519 0.834 0.128 0.872
#> GSM48662 2 0.0000 0.912 0.000 1.000
#> GSM48663 2 0.0672 0.912 0.008 0.992
#> GSM25524 1 0.5842 0.820 0.860 0.140
#> GSM25525 1 0.7815 0.733 0.768 0.232
#> GSM25526 1 0.3114 0.860 0.944 0.056
#> GSM25527 1 0.3584 0.851 0.932 0.068
#> GSM25528 1 0.0672 0.853 0.992 0.008
#> GSM25529 1 0.1414 0.857 0.980 0.020
#> GSM25530 1 0.0000 0.851 1.000 0.000
#> GSM25531 1 0.0000 0.851 1.000 0.000
#> GSM48661 2 0.4298 0.870 0.088 0.912
#> GSM25561 1 0.6712 0.791 0.824 0.176
#> GSM25562 1 0.9866 0.335 0.568 0.432
#> GSM25563 1 0.7745 0.740 0.772 0.228
#> GSM25564 2 0.2236 0.903 0.036 0.964
#> GSM25565 2 0.0672 0.912 0.008 0.992
#> GSM25566 2 0.1184 0.911 0.016 0.984
#> GSM25568 2 0.1184 0.912 0.016 0.984
#> GSM25569 2 0.0376 0.913 0.004 0.996
#> GSM25552 2 0.0000 0.912 0.000 1.000
#> GSM25553 2 0.0376 0.911 0.004 0.996
#> GSM25578 2 0.7883 0.697 0.236 0.764
#> GSM25579 2 0.5059 0.857 0.112 0.888
#> GSM25580 1 0.3879 0.851 0.924 0.076
#> GSM25581 1 0.2043 0.856 0.968 0.032
#> GSM48655 2 0.0376 0.912 0.004 0.996
#> GSM48656 2 0.0938 0.912 0.012 0.988
#> GSM48657 2 0.1633 0.909 0.024 0.976
#> GSM48658 2 0.5059 0.850 0.112 0.888
#> GSM25624 1 0.8763 0.629 0.704 0.296
#> GSM25625 1 0.3114 0.860 0.944 0.056
#> GSM25626 1 0.3114 0.860 0.944 0.056
#> GSM25627 1 0.3114 0.860 0.944 0.056
#> GSM25628 1 0.4815 0.844 0.896 0.104
#> GSM25629 1 0.3114 0.860 0.944 0.056
#> GSM25630 1 0.0938 0.855 0.988 0.012
#> GSM25631 2 0.0938 0.911 0.012 0.988
#> GSM25632 1 0.0376 0.853 0.996 0.004
#> GSM25633 1 0.5842 0.827 0.860 0.140
#> GSM25634 1 0.1184 0.856 0.984 0.016
#> GSM25635 1 0.8608 0.676 0.716 0.284
#> GSM25656 2 0.8016 0.690 0.244 0.756
#> GSM25657 1 0.3114 0.858 0.944 0.056
#> GSM25658 1 0.2948 0.860 0.948 0.052
#> GSM25659 2 0.6148 0.808 0.152 0.848
#> GSM25660 2 0.9491 0.413 0.368 0.632
#> GSM25661 1 0.9209 0.564 0.664 0.336
#> GSM25662 2 0.0672 0.912 0.008 0.992
#> GSM25663 2 0.0672 0.912 0.008 0.992
#> GSM25680 2 0.0672 0.912 0.008 0.992
#> GSM25681 2 0.0376 0.912 0.004 0.996
#> GSM25682 2 0.0376 0.912 0.004 0.996
#> GSM25683 2 0.0672 0.912 0.008 0.992
#> GSM25684 2 0.0672 0.912 0.008 0.992
#> GSM25685 2 0.6048 0.816 0.148 0.852
#> GSM25686 2 0.0672 0.912 0.008 0.992
#> GSM25687 2 0.0376 0.912 0.004 0.996
#> GSM48664 2 0.4690 0.857 0.100 0.900
#> GSM48665 2 0.9522 0.439 0.372 0.628
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM25548 2 0.2356 0.825 0.000 0.928 0.072
#> GSM25549 2 0.0892 0.828 0.000 0.980 0.020
#> GSM25550 2 0.2165 0.826 0.000 0.936 0.064
#> GSM25551 2 0.4807 0.762 0.092 0.848 0.060
#> GSM25570 2 0.0747 0.827 0.000 0.984 0.016
#> GSM25571 2 0.0592 0.826 0.000 0.988 0.012
#> GSM25358 3 0.4654 0.773 0.000 0.208 0.792
#> GSM25359 2 0.3618 0.804 0.012 0.884 0.104
#> GSM25360 1 0.9030 0.118 0.492 0.140 0.368
#> GSM25361 3 0.6090 0.700 0.020 0.264 0.716
#> GSM25377 2 0.6699 0.641 0.256 0.700 0.044
#> GSM25378 2 0.3116 0.794 0.108 0.892 0.000
#> GSM25401 1 0.6113 0.676 0.688 0.012 0.300
#> GSM25402 3 0.7451 0.237 0.396 0.040 0.564
#> GSM25349 2 0.3941 0.768 0.000 0.844 0.156
#> GSM25350 2 0.2261 0.829 0.000 0.932 0.068
#> GSM25356 2 0.1636 0.824 0.020 0.964 0.016
#> GSM25357 2 0.2711 0.808 0.000 0.912 0.088
#> GSM25385 1 0.0475 0.830 0.992 0.004 0.004
#> GSM25386 3 0.5881 0.534 0.256 0.016 0.728
#> GSM25399 1 0.1832 0.827 0.956 0.008 0.036
#> GSM25400 1 0.7363 0.516 0.656 0.064 0.280
#> GSM48659 3 0.4978 0.747 0.004 0.216 0.780
#> GSM48660 3 0.5363 0.729 0.000 0.276 0.724
#> GSM25409 2 0.1964 0.825 0.000 0.944 0.056
#> GSM25410 1 0.6644 0.752 0.752 0.108 0.140
#> GSM25426 3 0.2806 0.755 0.040 0.032 0.928
#> GSM25427 2 0.4137 0.788 0.096 0.872 0.032
#> GSM25540 3 0.7043 0.665 0.136 0.136 0.728
#> GSM25541 3 0.6374 0.681 0.132 0.100 0.768
#> GSM25542 3 0.4291 0.786 0.000 0.180 0.820
#> GSM25543 3 0.6126 0.617 0.004 0.352 0.644
#> GSM25479 1 0.6180 0.479 0.660 0.332 0.008
#> GSM25480 2 0.2492 0.819 0.048 0.936 0.016
#> GSM25481 2 0.2496 0.828 0.004 0.928 0.068
#> GSM25482 2 0.1905 0.822 0.028 0.956 0.016
#> GSM48654 3 0.3816 0.791 0.000 0.148 0.852
#> GSM48650 3 0.6291 0.382 0.000 0.468 0.532
#> GSM48651 3 0.3551 0.788 0.000 0.132 0.868
#> GSM48652 2 0.6252 0.496 0.008 0.648 0.344
#> GSM48653 3 0.1129 0.776 0.004 0.020 0.976
#> GSM48662 2 0.6126 0.258 0.000 0.600 0.400
#> GSM48663 3 0.4504 0.775 0.000 0.196 0.804
#> GSM25524 1 0.5958 0.662 0.692 0.008 0.300
#> GSM25525 1 0.5219 0.737 0.788 0.196 0.016
#> GSM25526 1 0.4261 0.811 0.848 0.012 0.140
#> GSM25527 1 0.1411 0.833 0.964 0.036 0.000
#> GSM25528 1 0.0475 0.830 0.992 0.004 0.004
#> GSM25529 1 0.3263 0.835 0.912 0.048 0.040
#> GSM25530 1 0.2063 0.833 0.948 0.008 0.044
#> GSM25531 1 0.0475 0.830 0.992 0.004 0.004
#> GSM48661 3 0.1964 0.783 0.000 0.056 0.944
#> GSM25561 1 0.6981 0.723 0.732 0.132 0.136
#> GSM25562 3 0.7821 0.520 0.224 0.116 0.660
#> GSM25563 3 0.3349 0.709 0.108 0.004 0.888
#> GSM25564 3 0.6696 0.526 0.020 0.348 0.632
#> GSM25565 3 0.4887 0.750 0.000 0.228 0.772
#> GSM25566 2 0.2682 0.819 0.004 0.920 0.076
#> GSM25568 2 0.5431 0.589 0.000 0.716 0.284
#> GSM25569 2 0.2356 0.819 0.000 0.928 0.072
#> GSM25552 2 0.2878 0.818 0.000 0.904 0.096
#> GSM25553 2 0.2625 0.821 0.000 0.916 0.084
#> GSM25578 2 0.4702 0.706 0.212 0.788 0.000
#> GSM25579 2 0.3481 0.826 0.044 0.904 0.052
#> GSM25580 1 0.0661 0.829 0.988 0.008 0.004
#> GSM25581 1 0.0424 0.830 0.992 0.008 0.000
#> GSM48655 2 0.5948 0.284 0.000 0.640 0.360
#> GSM48656 3 0.4504 0.780 0.000 0.196 0.804
#> GSM48657 3 0.4654 0.769 0.000 0.208 0.792
#> GSM48658 3 0.1989 0.780 0.004 0.048 0.948
#> GSM25624 1 0.5848 0.604 0.720 0.268 0.012
#> GSM25625 1 0.4261 0.811 0.848 0.012 0.140
#> GSM25626 1 0.4110 0.809 0.844 0.004 0.152
#> GSM25627 1 0.4261 0.811 0.848 0.012 0.140
#> GSM25628 1 0.6404 0.610 0.644 0.012 0.344
#> GSM25629 1 0.4805 0.797 0.812 0.012 0.176
#> GSM25630 1 0.2066 0.837 0.940 0.000 0.060
#> GSM25631 2 0.2400 0.825 0.004 0.932 0.064
#> GSM25632 1 0.3030 0.827 0.904 0.004 0.092
#> GSM25633 1 0.2651 0.819 0.928 0.060 0.012
#> GSM25634 1 0.0829 0.831 0.984 0.004 0.012
#> GSM25635 1 0.4733 0.705 0.800 0.196 0.004
#> GSM25656 3 0.3009 0.763 0.028 0.052 0.920
#> GSM25657 1 0.2806 0.838 0.928 0.032 0.040
#> GSM25658 1 0.4195 0.813 0.852 0.012 0.136
#> GSM25659 3 0.8223 0.551 0.108 0.288 0.604
#> GSM25660 2 0.6247 0.430 0.376 0.620 0.004
#> GSM25661 1 0.5178 0.625 0.744 0.256 0.000
#> GSM25662 3 0.3116 0.786 0.000 0.108 0.892
#> GSM25663 3 0.3482 0.786 0.000 0.128 0.872
#> GSM25680 2 0.1964 0.831 0.000 0.944 0.056
#> GSM25681 2 0.0592 0.826 0.000 0.988 0.012
#> GSM25682 2 0.4887 0.661 0.000 0.772 0.228
#> GSM25683 3 0.4504 0.774 0.000 0.196 0.804
#> GSM25684 3 0.3192 0.786 0.000 0.112 0.888
#> GSM25685 3 0.0475 0.767 0.004 0.004 0.992
#> GSM25686 3 0.5835 0.638 0.000 0.340 0.660
#> GSM25687 2 0.5882 0.376 0.000 0.652 0.348
#> GSM48664 3 0.9241 0.339 0.164 0.352 0.484
#> GSM48665 2 0.6154 0.378 0.408 0.592 0.000
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM25548 2 0.0817 0.8252 0.000 0.976 0.000 0.024
#> GSM25549 2 0.0000 0.8264 0.000 1.000 0.000 0.000
#> GSM25550 2 0.0817 0.8252 0.000 0.976 0.000 0.024
#> GSM25551 2 0.4253 0.7184 0.016 0.776 0.208 0.000
#> GSM25570 2 0.0000 0.8264 0.000 1.000 0.000 0.000
#> GSM25571 2 0.0000 0.8264 0.000 1.000 0.000 0.000
#> GSM25358 4 0.3400 0.7276 0.000 0.180 0.000 0.820
#> GSM25359 2 0.5522 0.7089 0.000 0.716 0.204 0.080
#> GSM25360 4 0.8389 -0.0385 0.108 0.076 0.364 0.452
#> GSM25361 4 0.6357 0.6414 0.000 0.232 0.124 0.644
#> GSM25377 1 0.3819 0.6889 0.816 0.172 0.004 0.008
#> GSM25378 2 0.2335 0.8101 0.060 0.920 0.020 0.000
#> GSM25401 3 0.6635 0.6046 0.152 0.000 0.620 0.228
#> GSM25402 4 0.6634 0.4237 0.164 0.000 0.212 0.624
#> GSM25349 2 0.4307 0.7617 0.000 0.784 0.024 0.192
#> GSM25350 2 0.1661 0.8281 0.000 0.944 0.004 0.052
#> GSM25356 2 0.3396 0.8135 0.016 0.884 0.036 0.064
#> GSM25357 2 0.4633 0.7626 0.000 0.780 0.048 0.172
#> GSM25385 3 0.5168 0.3057 0.492 0.000 0.504 0.004
#> GSM25386 4 0.6467 0.5308 0.080 0.008 0.292 0.620
#> GSM25399 1 0.1474 0.7169 0.948 0.000 0.052 0.000
#> GSM25400 1 0.7545 0.3953 0.616 0.060 0.204 0.120
#> GSM48659 4 0.5140 0.7245 0.000 0.144 0.096 0.760
#> GSM48660 4 0.3706 0.7079 0.000 0.112 0.040 0.848
#> GSM25409 2 0.3399 0.8016 0.000 0.868 0.040 0.092
#> GSM25410 3 0.8010 0.6109 0.136 0.116 0.600 0.148
#> GSM25426 4 0.5415 0.5246 0.004 0.008 0.436 0.552
#> GSM25427 2 0.5004 0.2446 0.392 0.604 0.000 0.004
#> GSM25540 4 0.8145 0.3982 0.060 0.104 0.376 0.460
#> GSM25541 4 0.6744 0.4221 0.008 0.068 0.456 0.468
#> GSM25542 4 0.1940 0.7595 0.000 0.076 0.000 0.924
#> GSM25543 4 0.5113 0.5817 0.000 0.264 0.032 0.704
#> GSM25479 1 0.6750 0.4048 0.612 0.208 0.180 0.000
#> GSM25480 2 0.2837 0.8194 0.028 0.912 0.036 0.024
#> GSM25481 2 0.2335 0.8202 0.020 0.920 0.000 0.060
#> GSM25482 2 0.6338 0.7319 0.140 0.716 0.040 0.104
#> GSM48654 4 0.3486 0.7607 0.000 0.044 0.092 0.864
#> GSM48650 4 0.6619 0.4406 0.000 0.332 0.100 0.568
#> GSM48651 4 0.1854 0.7582 0.000 0.048 0.012 0.940
#> GSM48652 2 0.6378 0.5297 0.000 0.628 0.108 0.264
#> GSM48653 4 0.2530 0.7465 0.000 0.000 0.112 0.888
#> GSM48662 2 0.5352 0.3981 0.000 0.596 0.016 0.388
#> GSM48663 4 0.1913 0.7406 0.000 0.020 0.040 0.940
#> GSM25524 3 0.6071 0.6511 0.144 0.000 0.684 0.172
#> GSM25525 3 0.7529 0.3540 0.344 0.196 0.460 0.000
#> GSM25526 3 0.3074 0.7527 0.152 0.000 0.848 0.000
#> GSM25527 3 0.6188 0.4799 0.396 0.056 0.548 0.000
#> GSM25528 1 0.4991 0.0429 0.608 0.000 0.388 0.004
#> GSM25529 1 0.5755 0.3719 0.624 0.044 0.332 0.000
#> GSM25530 3 0.4713 0.5951 0.360 0.000 0.640 0.000
#> GSM25531 1 0.4134 0.4605 0.740 0.000 0.260 0.000
#> GSM48661 4 0.3056 0.7586 0.000 0.072 0.040 0.888
#> GSM25561 1 0.5937 0.6124 0.712 0.088 0.188 0.012
#> GSM25562 4 0.8018 0.3017 0.036 0.128 0.404 0.432
#> GSM25563 4 0.5364 0.5954 0.028 0.000 0.320 0.652
#> GSM25564 4 0.6066 0.5009 0.028 0.312 0.024 0.636
#> GSM25565 4 0.4050 0.7197 0.000 0.168 0.024 0.808
#> GSM25566 2 0.4171 0.7867 0.000 0.824 0.060 0.116
#> GSM25568 2 0.4456 0.6125 0.000 0.716 0.004 0.280
#> GSM25569 2 0.3612 0.8040 0.000 0.856 0.044 0.100
#> GSM25552 2 0.1389 0.8249 0.000 0.952 0.000 0.048
#> GSM25553 2 0.0921 0.8254 0.000 0.972 0.000 0.028
#> GSM25578 1 0.5078 0.6092 0.700 0.272 0.028 0.000
#> GSM25579 2 0.4242 0.8043 0.068 0.848 0.036 0.048
#> GSM25580 1 0.0000 0.7391 1.000 0.000 0.000 0.000
#> GSM25581 1 0.0000 0.7391 1.000 0.000 0.000 0.000
#> GSM48655 2 0.6044 0.2933 0.000 0.528 0.044 0.428
#> GSM48656 4 0.3047 0.7531 0.000 0.116 0.012 0.872
#> GSM48657 4 0.3071 0.7255 0.000 0.068 0.044 0.888
#> GSM48658 4 0.3501 0.7528 0.000 0.020 0.132 0.848
#> GSM25624 1 0.6193 0.4354 0.672 0.148 0.180 0.000
#> GSM25625 3 0.2868 0.7487 0.136 0.000 0.864 0.000
#> GSM25626 3 0.3597 0.7542 0.148 0.000 0.836 0.016
#> GSM25627 3 0.3024 0.7525 0.148 0.000 0.852 0.000
#> GSM25628 3 0.4583 0.6433 0.076 0.004 0.808 0.112
#> GSM25629 3 0.1302 0.6940 0.044 0.000 0.956 0.000
#> GSM25630 3 0.5558 0.6324 0.324 0.000 0.640 0.036
#> GSM25631 2 0.0817 0.8252 0.000 0.976 0.000 0.024
#> GSM25632 3 0.3448 0.7521 0.168 0.000 0.828 0.004
#> GSM25633 1 0.1661 0.7418 0.944 0.052 0.000 0.004
#> GSM25634 1 0.1938 0.7217 0.936 0.000 0.052 0.012
#> GSM25635 1 0.0000 0.7391 1.000 0.000 0.000 0.000
#> GSM25656 4 0.5636 0.5316 0.000 0.024 0.424 0.552
#> GSM25657 3 0.5535 0.4917 0.420 0.020 0.560 0.000
#> GSM25658 3 0.3123 0.7527 0.156 0.000 0.844 0.000
#> GSM25659 4 0.7499 0.4887 0.124 0.268 0.032 0.576
#> GSM25660 1 0.3672 0.6918 0.824 0.164 0.012 0.000
#> GSM25661 1 0.1474 0.7426 0.948 0.052 0.000 0.000
#> GSM25662 4 0.2843 0.7529 0.000 0.020 0.088 0.892
#> GSM25663 4 0.0707 0.7569 0.000 0.020 0.000 0.980
#> GSM25680 2 0.0592 0.8263 0.000 0.984 0.000 0.016
#> GSM25681 2 0.0000 0.8264 0.000 1.000 0.000 0.000
#> GSM25682 2 0.5717 0.5997 0.000 0.632 0.044 0.324
#> GSM25683 4 0.2111 0.7376 0.000 0.024 0.044 0.932
#> GSM25684 4 0.2949 0.7527 0.000 0.024 0.088 0.888
#> GSM25685 4 0.2704 0.7456 0.000 0.000 0.124 0.876
#> GSM25686 4 0.4552 0.6337 0.000 0.172 0.044 0.784
#> GSM25687 2 0.6055 0.3328 0.000 0.520 0.044 0.436
#> GSM48664 1 0.4245 0.6921 0.832 0.104 0.008 0.056
#> GSM48665 1 0.0336 0.7418 0.992 0.008 0.000 0.000
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM25548 5 0.0162 0.7186 0.000 0.004 0.000 0.000 0.996
#> GSM25549 5 0.0000 0.7186 0.000 0.000 0.000 0.000 1.000
#> GSM25550 5 0.0162 0.7186 0.000 0.004 0.000 0.000 0.996
#> GSM25551 5 0.5701 0.4834 0.000 0.060 0.100 0.136 0.704
#> GSM25570 5 0.0000 0.7186 0.000 0.000 0.000 0.000 1.000
#> GSM25571 5 0.0000 0.7186 0.000 0.000 0.000 0.000 1.000
#> GSM25358 2 0.3243 0.5273 0.000 0.812 0.004 0.004 0.180
#> GSM25359 5 0.6062 0.3249 0.000 0.012 0.116 0.288 0.584
#> GSM25360 3 0.5501 0.1072 0.000 0.444 0.492 0.000 0.064
#> GSM25361 2 0.6190 0.4471 0.000 0.636 0.056 0.088 0.220
#> GSM25377 1 0.3357 0.7170 0.836 0.000 0.016 0.012 0.136
#> GSM25378 5 0.2214 0.6980 0.052 0.000 0.028 0.004 0.916
#> GSM25401 3 0.3966 0.5747 0.008 0.224 0.756 0.012 0.000
#> GSM25402 2 0.5732 0.2676 0.020 0.544 0.388 0.048 0.000
#> GSM25349 5 0.5358 0.3954 0.000 0.092 0.008 0.228 0.672
#> GSM25350 5 0.1981 0.7061 0.000 0.028 0.000 0.048 0.924
#> GSM25356 5 0.4518 0.3383 0.016 0.000 0.004 0.320 0.660
#> GSM25357 5 0.5310 -0.2412 0.000 0.040 0.004 0.448 0.508
#> GSM25385 3 0.4507 0.2800 0.412 0.004 0.580 0.004 0.000
#> GSM25386 2 0.5532 0.4670 0.000 0.644 0.256 0.092 0.008
#> GSM25399 1 0.5109 0.6226 0.696 0.000 0.132 0.172 0.000
#> GSM25400 1 0.7298 0.2060 0.476 0.116 0.336 0.004 0.068
#> GSM48659 2 0.2694 0.5231 0.004 0.864 0.000 0.004 0.128
#> GSM48660 2 0.4959 0.3327 0.000 0.684 0.000 0.240 0.076
#> GSM25409 5 0.2648 0.6332 0.000 0.000 0.000 0.152 0.848
#> GSM25410 3 0.5632 0.6202 0.012 0.152 0.712 0.028 0.096
#> GSM25426 2 0.6600 0.3122 0.004 0.496 0.264 0.236 0.000
#> GSM25427 5 0.4630 0.2156 0.416 0.004 0.000 0.008 0.572
#> GSM25540 2 0.8008 0.1733 0.004 0.360 0.332 0.228 0.076
#> GSM25541 2 0.7607 0.2739 0.004 0.436 0.284 0.228 0.048
#> GSM25542 2 0.3558 0.5412 0.000 0.828 0.000 0.108 0.064
#> GSM25543 2 0.5779 0.2703 0.000 0.624 0.008 0.116 0.252
#> GSM25479 1 0.7146 0.2824 0.476 0.004 0.288 0.024 0.208
#> GSM25480 5 0.2656 0.6985 0.028 0.000 0.012 0.064 0.896
#> GSM25481 5 0.3757 0.6134 0.008 0.040 0.000 0.136 0.816
#> GSM25482 5 0.6030 0.3198 0.196 0.000 0.000 0.224 0.580
#> GSM48654 2 0.1787 0.5800 0.004 0.936 0.000 0.044 0.016
#> GSM48650 2 0.6646 -0.3107 0.004 0.444 0.000 0.356 0.196
#> GSM48651 2 0.3003 0.5474 0.000 0.864 0.000 0.092 0.044
#> GSM48652 5 0.4614 0.2707 0.004 0.356 0.008 0.004 0.628
#> GSM48653 2 0.0000 0.5828 0.000 1.000 0.000 0.000 0.000
#> GSM48662 5 0.5300 0.1217 0.000 0.328 0.000 0.068 0.604
#> GSM48663 2 0.4341 0.0918 0.000 0.592 0.000 0.404 0.004
#> GSM25524 3 0.5398 0.6396 0.032 0.144 0.716 0.108 0.000
#> GSM25525 3 0.5541 0.4684 0.164 0.000 0.648 0.000 0.188
#> GSM25526 3 0.0404 0.7031 0.012 0.000 0.988 0.000 0.000
#> GSM25527 3 0.4303 0.5788 0.192 0.000 0.752 0.000 0.056
#> GSM25528 3 0.4420 0.1183 0.448 0.004 0.548 0.000 0.000
#> GSM25529 1 0.7051 0.3075 0.524 0.008 0.300 0.124 0.044
#> GSM25530 3 0.3231 0.6047 0.196 0.000 0.800 0.004 0.000
#> GSM25531 1 0.4437 0.1439 0.532 0.000 0.464 0.004 0.000
#> GSM48661 2 0.1831 0.5776 0.000 0.920 0.000 0.004 0.076
#> GSM25561 1 0.4788 0.6033 0.748 0.004 0.100 0.144 0.004
#> GSM25562 2 0.8183 0.2206 0.000 0.352 0.316 0.208 0.124
#> GSM25563 2 0.5476 0.4785 0.004 0.664 0.204 0.128 0.000
#> GSM25564 2 0.4552 0.2318 0.004 0.668 0.020 0.000 0.308
#> GSM25565 2 0.4049 0.4895 0.000 0.780 0.000 0.056 0.164
#> GSM25566 5 0.3873 0.5507 0.000 0.012 0.008 0.212 0.768
#> GSM25568 5 0.4016 0.3960 0.000 0.272 0.000 0.012 0.716
#> GSM25569 5 0.3387 0.6363 0.000 0.032 0.004 0.128 0.836
#> GSM25552 5 0.1121 0.7094 0.000 0.044 0.000 0.000 0.956
#> GSM25553 5 0.0404 0.7179 0.000 0.012 0.000 0.000 0.988
#> GSM25578 1 0.4025 0.5939 0.700 0.000 0.008 0.000 0.292
#> GSM25579 5 0.3759 0.6733 0.072 0.024 0.024 0.028 0.852
#> GSM25580 1 0.0290 0.7612 0.992 0.000 0.008 0.000 0.000
#> GSM25581 1 0.0290 0.7615 0.992 0.000 0.008 0.000 0.000
#> GSM48655 5 0.6673 -0.4820 0.000 0.244 0.000 0.332 0.424
#> GSM48656 2 0.3575 0.5463 0.000 0.824 0.000 0.056 0.120
#> GSM48657 2 0.5175 -0.1952 0.000 0.496 0.000 0.464 0.040
#> GSM48658 2 0.2597 0.5872 0.000 0.904 0.040 0.036 0.020
#> GSM25624 1 0.5772 0.5029 0.652 0.000 0.188 0.012 0.148
#> GSM25625 3 0.2642 0.6777 0.008 0.008 0.880 0.104 0.000
#> GSM25626 3 0.0968 0.7050 0.012 0.012 0.972 0.004 0.000
#> GSM25627 3 0.2857 0.6762 0.008 0.012 0.868 0.112 0.000
#> GSM25628 3 0.5932 0.4821 0.004 0.140 0.620 0.232 0.004
#> GSM25629 3 0.5499 0.5352 0.004 0.112 0.652 0.232 0.000
#> GSM25630 3 0.4588 0.6333 0.128 0.012 0.768 0.092 0.000
#> GSM25631 5 0.0162 0.7186 0.000 0.004 0.000 0.000 0.996
#> GSM25632 3 0.1082 0.7031 0.028 0.000 0.964 0.008 0.000
#> GSM25633 1 0.0451 0.7626 0.988 0.000 0.004 0.000 0.008
#> GSM25634 1 0.1341 0.7465 0.944 0.000 0.056 0.000 0.000
#> GSM25635 1 0.0162 0.7606 0.996 0.000 0.004 0.000 0.000
#> GSM25656 2 0.6785 0.3265 0.004 0.504 0.256 0.228 0.008
#> GSM25657 3 0.4218 0.4635 0.324 0.000 0.668 0.004 0.004
#> GSM25658 3 0.0404 0.7031 0.012 0.000 0.988 0.000 0.000
#> GSM25659 2 0.6904 0.2033 0.088 0.568 0.060 0.012 0.272
#> GSM25660 1 0.3106 0.7131 0.840 0.000 0.020 0.000 0.140
#> GSM25661 1 0.0486 0.7616 0.988 0.000 0.004 0.004 0.004
#> GSM25662 2 0.0451 0.5819 0.000 0.988 0.000 0.008 0.004
#> GSM25663 2 0.2416 0.5511 0.000 0.888 0.000 0.100 0.012
#> GSM25680 5 0.0162 0.7185 0.000 0.000 0.000 0.004 0.996
#> GSM25681 5 0.0000 0.7186 0.000 0.000 0.000 0.000 1.000
#> GSM25682 4 0.6118 0.4050 0.000 0.128 0.000 0.468 0.404
#> GSM25683 2 0.4440 -0.0732 0.000 0.528 0.000 0.468 0.004
#> GSM25684 2 0.0727 0.5830 0.004 0.980 0.000 0.004 0.012
#> GSM25685 2 0.1026 0.5845 0.004 0.968 0.024 0.004 0.000
#> GSM25686 4 0.5816 0.1171 0.000 0.440 0.000 0.468 0.092
#> GSM25687 4 0.6469 0.5722 0.000 0.196 0.000 0.468 0.336
#> GSM48664 1 0.1059 0.7565 0.968 0.020 0.000 0.004 0.008
#> GSM48665 1 0.0609 0.7606 0.980 0.000 0.020 0.000 0.000
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM25548 4 0.0000 0.75083 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM25549 4 0.0000 0.75083 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM25550 4 0.0146 0.75089 0.000 0.004 0.000 0.996 0.000 0.000
#> GSM25551 4 0.4230 0.51813 0.000 0.004 0.024 0.676 0.004 0.292
#> GSM25570 4 0.0000 0.75083 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM25571 4 0.0000 0.75083 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM25358 5 0.2656 0.66655 0.000 0.008 0.000 0.120 0.860 0.012
#> GSM25359 4 0.6513 0.31771 0.000 0.124 0.056 0.528 0.012 0.280
#> GSM25360 5 0.4974 0.04130 0.000 0.000 0.440 0.056 0.500 0.004
#> GSM25361 5 0.4914 0.47183 0.000 0.000 0.008 0.152 0.680 0.160
#> GSM25377 1 0.3235 0.71612 0.836 0.016 0.016 0.124 0.000 0.008
#> GSM25378 4 0.2495 0.72827 0.048 0.008 0.024 0.900 0.000 0.020
#> GSM25401 3 0.3370 0.49299 0.000 0.004 0.772 0.000 0.212 0.012
#> GSM25402 5 0.5222 0.18732 0.000 0.052 0.400 0.000 0.528 0.020
#> GSM25349 4 0.5466 0.44578 0.000 0.212 0.000 0.640 0.112 0.036
#> GSM25350 4 0.2186 0.73517 0.000 0.056 0.000 0.908 0.024 0.012
#> GSM25356 4 0.4646 0.19957 0.016 0.396 0.000 0.568 0.000 0.020
#> GSM25357 2 0.4841 0.38843 0.000 0.576 0.000 0.372 0.040 0.012
#> GSM25385 3 0.4658 0.37165 0.384 0.000 0.568 0.000 0.000 0.048
#> GSM25386 5 0.5521 0.23336 0.000 0.000 0.188 0.004 0.580 0.228
#> GSM25399 1 0.6552 0.35649 0.468 0.056 0.160 0.000 0.000 0.316
#> GSM25400 1 0.6576 0.00774 0.440 0.008 0.392 0.052 0.104 0.004
#> GSM48659 5 0.4003 0.60145 0.000 0.000 0.000 0.124 0.760 0.116
#> GSM48660 5 0.4516 0.38642 0.000 0.260 0.000 0.072 0.668 0.000
#> GSM25409 4 0.2558 0.67280 0.000 0.156 0.000 0.840 0.004 0.000
#> GSM25410 3 0.4536 0.56567 0.000 0.000 0.748 0.068 0.140 0.044
#> GSM25426 6 0.5411 0.70817 0.000 0.000 0.148 0.000 0.296 0.556
#> GSM25427 4 0.4668 0.20727 0.412 0.012 0.000 0.556 0.008 0.012
#> GSM25540 6 0.6676 0.71795 0.000 0.000 0.208 0.056 0.268 0.468
#> GSM25541 6 0.6123 0.72158 0.000 0.000 0.156 0.032 0.288 0.524
#> GSM25542 5 0.3107 0.64259 0.000 0.116 0.000 0.052 0.832 0.000
#> GSM25543 5 0.5084 0.38231 0.000 0.116 0.000 0.232 0.644 0.008
#> GSM25479 1 0.6773 0.26736 0.468 0.000 0.256 0.208 0.000 0.068
#> GSM25480 4 0.2811 0.73017 0.028 0.048 0.008 0.884 0.000 0.032
#> GSM25481 4 0.5125 0.49098 0.004 0.228 0.000 0.668 0.028 0.072
#> GSM25482 4 0.6746 0.13872 0.156 0.340 0.000 0.432 0.000 0.072
#> GSM48654 5 0.2593 0.67870 0.000 0.036 0.000 0.012 0.884 0.068
#> GSM48650 2 0.6591 0.41922 0.000 0.452 0.000 0.124 0.348 0.076
#> GSM48651 5 0.1320 0.69006 0.000 0.016 0.000 0.036 0.948 0.000
#> GSM48652 4 0.5046 0.38199 0.000 0.000 0.000 0.620 0.256 0.124
#> GSM48653 5 0.0146 0.68694 0.000 0.000 0.000 0.000 0.996 0.004
#> GSM48662 4 0.4798 0.15978 0.000 0.060 0.000 0.564 0.376 0.000
#> GSM48663 5 0.3868 -0.21177 0.000 0.492 0.000 0.000 0.508 0.000
#> GSM25524 3 0.5598 0.29571 0.024 0.000 0.616 0.000 0.160 0.200
#> GSM25525 3 0.4416 0.57785 0.124 0.000 0.716 0.160 0.000 0.000
#> GSM25526 3 0.0000 0.67509 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM25527 3 0.2837 0.68591 0.088 0.000 0.856 0.056 0.000 0.000
#> GSM25528 3 0.3782 0.30681 0.412 0.000 0.588 0.000 0.000 0.000
#> GSM25529 1 0.6525 0.23141 0.484 0.000 0.252 0.044 0.000 0.220
#> GSM25530 3 0.2482 0.68305 0.148 0.000 0.848 0.000 0.000 0.004
#> GSM25531 3 0.3982 0.11462 0.460 0.000 0.536 0.000 0.000 0.004
#> GSM48661 5 0.0547 0.68962 0.000 0.000 0.000 0.020 0.980 0.000
#> GSM25561 1 0.3695 0.58929 0.732 0.000 0.024 0.000 0.000 0.244
#> GSM25562 6 0.7278 0.61352 0.000 0.000 0.232 0.104 0.304 0.360
#> GSM25563 5 0.5068 0.12303 0.004 0.000 0.104 0.000 0.620 0.272
#> GSM25564 5 0.3543 0.39955 0.004 0.000 0.004 0.272 0.720 0.000
#> GSM25565 5 0.3045 0.64195 0.000 0.060 0.000 0.100 0.840 0.000
#> GSM25566 4 0.3838 0.56654 0.000 0.240 0.000 0.732 0.008 0.020
#> GSM25568 4 0.3606 0.47878 0.000 0.004 0.000 0.708 0.284 0.004
#> GSM25569 4 0.3235 0.67315 0.000 0.136 0.000 0.824 0.032 0.008
#> GSM25552 4 0.1387 0.73583 0.000 0.000 0.000 0.932 0.068 0.000
#> GSM25553 4 0.0260 0.75062 0.000 0.000 0.000 0.992 0.008 0.000
#> GSM25578 1 0.3528 0.59108 0.700 0.000 0.004 0.296 0.000 0.000
#> GSM25579 4 0.3790 0.70199 0.072 0.016 0.004 0.832 0.040 0.036
#> GSM25580 1 0.0146 0.75853 0.996 0.000 0.004 0.000 0.000 0.000
#> GSM25581 1 0.0260 0.75914 0.992 0.000 0.008 0.000 0.000 0.000
#> GSM48655 2 0.5941 0.40684 0.000 0.408 0.000 0.376 0.216 0.000
#> GSM48656 5 0.2672 0.67461 0.000 0.052 0.000 0.080 0.868 0.000
#> GSM48657 2 0.4312 0.46036 0.000 0.604 0.000 0.028 0.368 0.000
#> GSM48658 5 0.2469 0.66330 0.000 0.012 0.028 0.004 0.896 0.060
#> GSM25624 1 0.5386 0.47443 0.640 0.008 0.196 0.148 0.000 0.008
#> GSM25625 3 0.2491 0.53127 0.000 0.000 0.836 0.000 0.000 0.164
#> GSM25626 3 0.0363 0.67162 0.000 0.000 0.988 0.000 0.000 0.012
#> GSM25627 3 0.3126 0.38930 0.000 0.000 0.752 0.000 0.000 0.248
#> GSM25628 6 0.4899 0.49137 0.000 0.000 0.404 0.000 0.064 0.532
#> GSM25629 6 0.4205 0.40303 0.000 0.000 0.420 0.000 0.016 0.564
#> GSM25630 3 0.5029 0.62228 0.056 0.208 0.692 0.000 0.008 0.036
#> GSM25631 4 0.0146 0.75082 0.000 0.000 0.000 0.996 0.000 0.004
#> GSM25632 3 0.0806 0.67492 0.008 0.000 0.972 0.000 0.000 0.020
#> GSM25633 1 0.0291 0.75978 0.992 0.000 0.004 0.004 0.000 0.000
#> GSM25634 1 0.1387 0.73325 0.932 0.000 0.068 0.000 0.000 0.000
#> GSM25635 1 0.0000 0.75771 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM25656 6 0.5475 0.70651 0.000 0.000 0.148 0.000 0.316 0.536
#> GSM25657 3 0.3452 0.60521 0.256 0.000 0.736 0.004 0.000 0.004
#> GSM25658 3 0.0000 0.67509 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM25659 5 0.5419 0.39702 0.084 0.000 0.040 0.220 0.652 0.004
#> GSM25660 1 0.2846 0.70876 0.840 0.000 0.004 0.140 0.000 0.016
#> GSM25661 1 0.0260 0.75886 0.992 0.000 0.000 0.000 0.000 0.008
#> GSM25662 5 0.0000 0.68708 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM25663 5 0.0692 0.68706 0.000 0.020 0.000 0.004 0.976 0.000
#> GSM25680 4 0.0000 0.75083 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM25681 4 0.0146 0.75083 0.000 0.004 0.000 0.996 0.000 0.000
#> GSM25682 2 0.4433 0.42348 0.000 0.616 0.000 0.344 0.040 0.000
#> GSM25683 2 0.3717 0.41173 0.000 0.616 0.000 0.000 0.384 0.000
#> GSM25684 5 0.2003 0.65414 0.000 0.000 0.000 0.000 0.884 0.116
#> GSM25685 5 0.2454 0.63071 0.000 0.000 0.000 0.000 0.840 0.160
#> GSM25686 2 0.4814 0.55730 0.000 0.616 0.000 0.080 0.304 0.000
#> GSM25687 2 0.5048 0.58199 0.000 0.616 0.000 0.264 0.120 0.000
#> GSM48664 1 0.0717 0.75559 0.976 0.008 0.000 0.000 0.016 0.000
#> GSM48665 1 0.0692 0.75711 0.976 0.004 0.020 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 genotype/variation(p) k
#> MAD:pam 94 1.39e-04 2
#> MAD:pam 89 1.66e-07 3
#> MAD:pam 79 1.69e-05 4
#> MAD:pam 59 7.29e-03 5
#> MAD:pam 62 7.42e-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["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 100 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 0.471 0.891 0.894 0.4688 0.495 0.495
#> 3 3 0.494 0.713 0.819 0.2822 0.838 0.680
#> 4 4 0.670 0.819 0.869 0.1206 0.895 0.729
#> 5 5 0.639 0.692 0.806 0.1333 0.885 0.641
#> 6 6 0.683 0.628 0.764 0.0596 0.901 0.586
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
#> GSM25548 2 0.0672 0.943 0.008 0.992
#> GSM25549 2 0.0938 0.943 0.012 0.988
#> GSM25550 2 0.1633 0.939 0.024 0.976
#> GSM25551 2 0.0000 0.942 0.000 1.000
#> GSM25570 2 0.1414 0.941 0.020 0.980
#> GSM25571 2 0.0938 0.943 0.012 0.988
#> GSM25358 1 0.5842 0.918 0.860 0.140
#> GSM25359 2 0.5408 0.852 0.124 0.876
#> GSM25360 1 0.5519 0.927 0.872 0.128
#> GSM25361 2 0.9358 0.429 0.352 0.648
#> GSM25377 1 0.5408 0.929 0.876 0.124
#> GSM25378 1 0.5408 0.929 0.876 0.124
#> GSM25401 1 0.7528 0.846 0.784 0.216
#> GSM25402 1 0.6148 0.910 0.848 0.152
#> GSM25349 2 0.0000 0.942 0.000 1.000
#> GSM25350 2 0.0000 0.942 0.000 1.000
#> GSM25356 1 0.5842 0.919 0.860 0.140
#> GSM25357 2 0.0000 0.942 0.000 1.000
#> GSM25385 1 0.5408 0.929 0.876 0.124
#> GSM25386 1 0.5519 0.927 0.872 0.128
#> GSM25399 1 0.5408 0.929 0.876 0.124
#> GSM25400 1 0.5408 0.929 0.876 0.124
#> GSM48659 2 0.0672 0.943 0.008 0.992
#> GSM48660 2 0.0000 0.942 0.000 1.000
#> GSM25409 2 0.0672 0.943 0.008 0.992
#> GSM25410 1 0.5408 0.929 0.876 0.124
#> GSM25426 2 0.1184 0.942 0.016 0.984
#> GSM25427 1 0.5408 0.929 0.876 0.124
#> GSM25540 2 0.7815 0.696 0.232 0.768
#> GSM25541 2 0.9427 0.406 0.360 0.640
#> GSM25542 2 0.3431 0.911 0.064 0.936
#> GSM25543 2 0.3879 0.900 0.076 0.924
#> GSM25479 1 0.0000 0.878 1.000 0.000
#> GSM25480 1 0.1184 0.887 0.984 0.016
#> GSM25481 1 0.8499 0.762 0.724 0.276
#> GSM25482 1 0.8144 0.799 0.748 0.252
#> GSM48654 2 0.0672 0.943 0.008 0.992
#> GSM48650 2 0.1414 0.941 0.020 0.980
#> GSM48651 2 0.0000 0.942 0.000 1.000
#> GSM48652 2 0.0000 0.942 0.000 1.000
#> GSM48653 2 0.0000 0.942 0.000 1.000
#> GSM48662 2 0.0000 0.942 0.000 1.000
#> GSM48663 2 0.1414 0.941 0.020 0.980
#> GSM25524 1 0.5408 0.929 0.876 0.124
#> GSM25525 1 0.0672 0.883 0.992 0.008
#> GSM25526 1 0.5408 0.929 0.876 0.124
#> GSM25527 1 0.0376 0.880 0.996 0.004
#> GSM25528 1 0.5408 0.929 0.876 0.124
#> GSM25529 1 0.0672 0.883 0.992 0.008
#> GSM25530 1 0.5408 0.929 0.876 0.124
#> GSM25531 1 0.4431 0.919 0.908 0.092
#> GSM48661 2 0.2778 0.925 0.048 0.952
#> GSM25561 1 0.5408 0.929 0.876 0.124
#> GSM25562 1 0.5178 0.927 0.884 0.116
#> GSM25563 1 0.5408 0.929 0.876 0.124
#> GSM25564 1 0.9286 0.634 0.656 0.344
#> GSM25565 2 0.0000 0.942 0.000 1.000
#> GSM25566 2 0.0000 0.942 0.000 1.000
#> GSM25568 2 0.7376 0.727 0.208 0.792
#> GSM25569 2 0.0000 0.942 0.000 1.000
#> GSM25552 2 0.2236 0.933 0.036 0.964
#> GSM25553 2 0.2948 0.922 0.052 0.948
#> GSM25578 1 0.0000 0.878 1.000 0.000
#> GSM25579 1 0.5178 0.927 0.884 0.116
#> GSM25580 1 0.0000 0.878 1.000 0.000
#> GSM25581 1 0.0000 0.878 1.000 0.000
#> GSM48655 2 0.0000 0.942 0.000 1.000
#> GSM48656 2 0.1633 0.939 0.024 0.976
#> GSM48657 2 0.0000 0.942 0.000 1.000
#> GSM48658 2 0.2778 0.925 0.048 0.952
#> GSM25624 1 0.1843 0.893 0.972 0.028
#> GSM25625 1 0.5408 0.929 0.876 0.124
#> GSM25626 1 0.5408 0.929 0.876 0.124
#> GSM25627 1 0.7745 0.832 0.772 0.228
#> GSM25628 1 0.8016 0.801 0.756 0.244
#> GSM25629 2 0.9833 0.156 0.424 0.576
#> GSM25630 1 0.5408 0.929 0.876 0.124
#> GSM25631 2 0.5737 0.837 0.136 0.864
#> GSM25632 1 0.5408 0.929 0.876 0.124
#> GSM25633 1 0.0000 0.878 1.000 0.000
#> GSM25634 1 0.0000 0.878 1.000 0.000
#> GSM25635 1 0.0000 0.878 1.000 0.000
#> GSM25656 1 0.8207 0.783 0.744 0.256
#> GSM25657 1 0.1633 0.891 0.976 0.024
#> GSM25658 1 0.5408 0.929 0.876 0.124
#> GSM25659 1 0.5629 0.924 0.868 0.132
#> GSM25660 1 0.0000 0.878 1.000 0.000
#> GSM25661 1 0.0000 0.878 1.000 0.000
#> GSM25662 2 0.0376 0.943 0.004 0.996
#> GSM25663 2 0.2603 0.928 0.044 0.956
#> GSM25680 2 0.1414 0.941 0.020 0.980
#> GSM25681 2 0.2236 0.933 0.036 0.964
#> GSM25682 2 0.0000 0.942 0.000 1.000
#> GSM25683 2 0.0000 0.942 0.000 1.000
#> GSM25684 2 0.0376 0.943 0.004 0.996
#> GSM25685 2 0.0938 0.943 0.012 0.988
#> GSM25686 2 0.0000 0.942 0.000 1.000
#> GSM25687 2 0.0000 0.942 0.000 1.000
#> GSM48664 1 0.5408 0.929 0.876 0.124
#> GSM48665 1 0.5408 0.929 0.876 0.124
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM25548 2 0.1643 0.8937 0.000 0.956 0.044
#> GSM25549 2 0.1753 0.8932 0.000 0.952 0.048
#> GSM25550 2 0.2066 0.8907 0.000 0.940 0.060
#> GSM25551 2 0.0000 0.8908 0.000 1.000 0.000
#> GSM25570 2 0.2066 0.8902 0.000 0.940 0.060
#> GSM25571 2 0.1753 0.8932 0.000 0.952 0.048
#> GSM25358 1 0.9385 0.3622 0.484 0.188 0.328
#> GSM25359 2 0.4605 0.7696 0.000 0.796 0.204
#> GSM25360 3 0.3805 0.7074 0.092 0.024 0.884
#> GSM25361 3 0.7043 0.1014 0.020 0.448 0.532
#> GSM25377 1 0.7222 0.6187 0.696 0.084 0.220
#> GSM25378 1 0.7263 0.6199 0.692 0.084 0.224
#> GSM25401 3 0.8399 0.5026 0.188 0.188 0.624
#> GSM25402 1 0.8291 0.5194 0.580 0.100 0.320
#> GSM25349 2 0.0892 0.8936 0.000 0.980 0.020
#> GSM25350 2 0.0747 0.8872 0.000 0.984 0.016
#> GSM25356 1 0.7222 0.6187 0.696 0.084 0.220
#> GSM25357 2 0.1399 0.8868 0.004 0.968 0.028
#> GSM25385 3 0.5493 0.6049 0.232 0.012 0.756
#> GSM25386 3 0.1620 0.7067 0.024 0.012 0.964
#> GSM25399 1 0.7569 0.6184 0.664 0.088 0.248
#> GSM25400 1 0.7782 0.6073 0.648 0.096 0.256
#> GSM48659 2 0.1289 0.8943 0.000 0.968 0.032
#> GSM48660 2 0.0747 0.8839 0.000 0.984 0.016
#> GSM25409 2 0.1163 0.8948 0.000 0.972 0.028
#> GSM25410 3 0.2845 0.7175 0.068 0.012 0.920
#> GSM25426 2 0.4452 0.7851 0.000 0.808 0.192
#> GSM25427 1 0.7298 0.6183 0.692 0.088 0.220
#> GSM25540 2 0.6489 0.2124 0.004 0.540 0.456
#> GSM25541 3 0.6398 0.2116 0.004 0.416 0.580
#> GSM25542 2 0.4555 0.7704 0.000 0.800 0.200
#> GSM25543 2 0.4750 0.7523 0.000 0.784 0.216
#> GSM25479 1 0.2939 0.7055 0.916 0.012 0.072
#> GSM25480 1 0.3120 0.7089 0.908 0.012 0.080
#> GSM25481 1 0.7263 0.6149 0.692 0.084 0.224
#> GSM25482 1 0.7222 0.6187 0.696 0.084 0.220
#> GSM48654 2 0.1289 0.8943 0.000 0.968 0.032
#> GSM48650 2 0.4452 0.7851 0.000 0.808 0.192
#> GSM48651 2 0.0000 0.8908 0.000 1.000 0.000
#> GSM48652 2 0.0000 0.8908 0.000 1.000 0.000
#> GSM48653 2 0.0237 0.8909 0.000 0.996 0.004
#> GSM48662 2 0.0237 0.8893 0.000 0.996 0.004
#> GSM48663 2 0.4605 0.7884 0.000 0.796 0.204
#> GSM25524 3 0.4805 0.6743 0.176 0.012 0.812
#> GSM25525 1 0.3989 0.6982 0.864 0.012 0.124
#> GSM25526 3 0.4059 0.7034 0.128 0.012 0.860
#> GSM25527 1 0.3459 0.6985 0.892 0.012 0.096
#> GSM25528 1 0.6825 0.2414 0.500 0.012 0.488
#> GSM25529 1 0.4059 0.6956 0.860 0.012 0.128
#> GSM25530 3 0.5775 0.5625 0.260 0.012 0.728
#> GSM25531 1 0.5884 0.6389 0.716 0.012 0.272
#> GSM48661 2 0.4555 0.7750 0.000 0.800 0.200
#> GSM25561 3 0.6675 0.0846 0.404 0.012 0.584
#> GSM25562 1 0.7451 0.6157 0.636 0.060 0.304
#> GSM25563 3 0.2550 0.7160 0.056 0.012 0.932
#> GSM25564 3 0.9364 0.2406 0.172 0.372 0.456
#> GSM25565 2 0.0237 0.8909 0.000 0.996 0.004
#> GSM25566 2 0.0424 0.8874 0.000 0.992 0.008
#> GSM25568 2 0.6016 0.6815 0.020 0.724 0.256
#> GSM25569 2 0.0000 0.8908 0.000 1.000 0.000
#> GSM25552 2 0.3267 0.8611 0.000 0.884 0.116
#> GSM25553 2 0.5536 0.7246 0.012 0.752 0.236
#> GSM25578 1 0.2845 0.7041 0.920 0.012 0.068
#> GSM25579 1 0.8162 0.5510 0.568 0.084 0.348
#> GSM25580 1 0.2339 0.7002 0.940 0.012 0.048
#> GSM25581 1 0.2749 0.7041 0.924 0.012 0.064
#> GSM48655 2 0.0892 0.8826 0.000 0.980 0.020
#> GSM48656 2 0.1411 0.8939 0.000 0.964 0.036
#> GSM48657 2 0.0892 0.8846 0.000 0.980 0.020
#> GSM48658 2 0.4121 0.8099 0.000 0.832 0.168
#> GSM25624 1 0.3377 0.7114 0.896 0.012 0.092
#> GSM25625 3 0.4575 0.6883 0.160 0.012 0.828
#> GSM25626 3 0.2550 0.7170 0.056 0.012 0.932
#> GSM25627 3 0.5094 0.6639 0.040 0.136 0.824
#> GSM25628 3 0.1525 0.7003 0.004 0.032 0.964
#> GSM25629 3 0.3879 0.6474 0.000 0.152 0.848
#> GSM25630 3 0.4692 0.6860 0.168 0.012 0.820
#> GSM25631 2 0.4974 0.7395 0.000 0.764 0.236
#> GSM25632 3 0.4968 0.6678 0.188 0.012 0.800
#> GSM25633 1 0.2845 0.7041 0.920 0.012 0.068
#> GSM25634 1 0.2845 0.7041 0.920 0.012 0.068
#> GSM25635 1 0.2651 0.7050 0.928 0.012 0.060
#> GSM25656 3 0.2165 0.6992 0.000 0.064 0.936
#> GSM25657 1 0.3377 0.7066 0.896 0.012 0.092
#> GSM25658 3 0.6247 0.6183 0.212 0.044 0.744
#> GSM25659 1 0.8786 0.3535 0.464 0.112 0.424
#> GSM25660 1 0.2845 0.7060 0.920 0.012 0.068
#> GSM25661 1 0.2446 0.7016 0.936 0.012 0.052
#> GSM25662 2 0.1163 0.8947 0.000 0.972 0.028
#> GSM25663 2 0.2625 0.8741 0.000 0.916 0.084
#> GSM25680 2 0.1964 0.8884 0.000 0.944 0.056
#> GSM25681 2 0.3686 0.8419 0.000 0.860 0.140
#> GSM25682 2 0.0892 0.8826 0.000 0.980 0.020
#> GSM25683 2 0.0747 0.8840 0.000 0.984 0.016
#> GSM25684 2 0.0747 0.8940 0.000 0.984 0.016
#> GSM25685 2 0.4291 0.7984 0.000 0.820 0.180
#> GSM25686 2 0.0892 0.8826 0.000 0.980 0.020
#> GSM25687 2 0.0892 0.8826 0.000 0.980 0.020
#> GSM48664 1 0.7222 0.6230 0.696 0.084 0.220
#> GSM48665 1 0.7344 0.6235 0.684 0.084 0.232
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM25548 2 0.2928 0.9066 0.004 0.896 0.024 0.076
#> GSM25549 2 0.2984 0.9007 0.000 0.888 0.028 0.084
#> GSM25550 2 0.3169 0.9023 0.004 0.884 0.028 0.084
#> GSM25551 2 0.1590 0.9192 0.008 0.956 0.008 0.028
#> GSM25570 2 0.2984 0.9007 0.000 0.888 0.028 0.084
#> GSM25571 2 0.2882 0.9025 0.000 0.892 0.024 0.084
#> GSM25358 2 0.7326 0.5696 0.100 0.640 0.192 0.068
#> GSM25359 2 0.2413 0.9190 0.004 0.924 0.036 0.036
#> GSM25360 3 0.4375 0.7898 0.144 0.036 0.812 0.008
#> GSM25361 2 0.8189 0.0516 0.100 0.472 0.360 0.068
#> GSM25377 4 0.3741 0.8708 0.108 0.004 0.036 0.852
#> GSM25378 4 0.3770 0.8721 0.104 0.004 0.040 0.852
#> GSM25401 4 0.8355 0.2589 0.052 0.144 0.348 0.456
#> GSM25402 4 0.7427 0.5207 0.096 0.036 0.304 0.564
#> GSM25349 2 0.2060 0.9169 0.000 0.932 0.016 0.052
#> GSM25350 2 0.2578 0.9098 0.000 0.912 0.036 0.052
#> GSM25356 4 0.3709 0.8709 0.100 0.004 0.040 0.856
#> GSM25357 2 0.2483 0.9159 0.000 0.916 0.032 0.052
#> GSM25385 3 0.3726 0.7728 0.212 0.000 0.788 0.000
#> GSM25386 3 0.2197 0.7939 0.080 0.004 0.916 0.000
#> GSM25399 4 0.4689 0.8289 0.184 0.004 0.036 0.776
#> GSM25400 4 0.6708 0.6709 0.256 0.016 0.096 0.632
#> GSM48659 2 0.1229 0.9210 0.008 0.968 0.020 0.004
#> GSM48660 2 0.2319 0.9026 0.000 0.924 0.040 0.036
#> GSM25409 2 0.1890 0.9200 0.000 0.936 0.008 0.056
#> GSM25410 3 0.2466 0.8006 0.096 0.004 0.900 0.000
#> GSM25426 2 0.3127 0.9006 0.008 0.892 0.068 0.032
#> GSM25427 4 0.3917 0.8712 0.108 0.004 0.044 0.844
#> GSM25540 3 0.7268 0.3804 0.048 0.356 0.540 0.056
#> GSM25541 3 0.7437 0.4263 0.056 0.328 0.552 0.064
#> GSM25542 2 0.2170 0.9180 0.008 0.936 0.028 0.028
#> GSM25543 2 0.2463 0.9157 0.008 0.924 0.036 0.032
#> GSM25479 1 0.0188 0.8982 0.996 0.000 0.004 0.000
#> GSM25480 1 0.0336 0.8985 0.992 0.000 0.008 0.000
#> GSM25481 4 0.3647 0.8689 0.096 0.004 0.040 0.860
#> GSM25482 4 0.3709 0.8709 0.100 0.004 0.040 0.856
#> GSM48654 2 0.0992 0.9225 0.008 0.976 0.012 0.004
#> GSM48650 2 0.2877 0.9112 0.008 0.904 0.060 0.028
#> GSM48651 2 0.0927 0.9220 0.008 0.976 0.000 0.016
#> GSM48652 2 0.0804 0.9220 0.008 0.980 0.000 0.012
#> GSM48653 2 0.0672 0.9220 0.008 0.984 0.000 0.008
#> GSM48662 2 0.0992 0.9226 0.008 0.976 0.004 0.012
#> GSM48663 2 0.2722 0.9067 0.000 0.904 0.064 0.032
#> GSM25524 3 0.4188 0.7449 0.244 0.004 0.752 0.000
#> GSM25525 1 0.1474 0.8832 0.948 0.000 0.052 0.000
#> GSM25526 3 0.2831 0.8050 0.120 0.004 0.876 0.000
#> GSM25527 1 0.1209 0.8934 0.964 0.000 0.032 0.004
#> GSM25528 1 0.5167 -0.1687 0.508 0.004 0.488 0.000
#> GSM25529 1 0.1474 0.8832 0.948 0.000 0.052 0.000
#> GSM25530 3 0.4134 0.7229 0.260 0.000 0.740 0.000
#> GSM25531 1 0.3311 0.7516 0.828 0.000 0.172 0.000
#> GSM48661 2 0.2261 0.9185 0.008 0.932 0.024 0.036
#> GSM25561 3 0.4697 0.5532 0.356 0.000 0.644 0.000
#> GSM25562 1 0.5932 0.6452 0.728 0.016 0.140 0.116
#> GSM25563 3 0.2831 0.8073 0.120 0.004 0.876 0.000
#> GSM25564 2 0.6538 0.6149 0.100 0.676 0.200 0.024
#> GSM25565 2 0.1229 0.9234 0.008 0.968 0.004 0.020
#> GSM25566 2 0.0927 0.9207 0.000 0.976 0.008 0.016
#> GSM25568 2 0.3574 0.8979 0.008 0.872 0.056 0.064
#> GSM25569 2 0.0927 0.9237 0.008 0.976 0.000 0.016
#> GSM25552 2 0.3082 0.9009 0.000 0.884 0.032 0.084
#> GSM25553 2 0.3886 0.8916 0.020 0.860 0.040 0.080
#> GSM25578 1 0.0188 0.8982 0.996 0.000 0.004 0.000
#> GSM25579 1 0.2198 0.8689 0.920 0.008 0.072 0.000
#> GSM25580 1 0.0469 0.8929 0.988 0.000 0.000 0.012
#> GSM25581 1 0.0188 0.8970 0.996 0.000 0.000 0.004
#> GSM48655 2 0.2500 0.8992 0.000 0.916 0.044 0.040
#> GSM48656 2 0.1854 0.9203 0.008 0.948 0.020 0.024
#> GSM48657 2 0.2319 0.9026 0.000 0.924 0.040 0.036
#> GSM48658 2 0.2153 0.9189 0.008 0.936 0.020 0.036
#> GSM25624 1 0.0336 0.8953 0.992 0.000 0.000 0.008
#> GSM25625 3 0.3052 0.8039 0.136 0.004 0.860 0.000
#> GSM25626 3 0.2530 0.8017 0.100 0.004 0.896 0.000
#> GSM25627 3 0.6492 0.4935 0.088 0.276 0.628 0.008
#> GSM25628 3 0.2670 0.7752 0.052 0.040 0.908 0.000
#> GSM25629 3 0.5285 0.6319 0.052 0.184 0.752 0.012
#> GSM25630 3 0.3626 0.7929 0.184 0.004 0.812 0.000
#> GSM25631 2 0.3450 0.9015 0.016 0.880 0.032 0.072
#> GSM25632 3 0.3569 0.7845 0.196 0.000 0.804 0.000
#> GSM25633 1 0.0188 0.8970 0.996 0.000 0.000 0.004
#> GSM25634 1 0.0188 0.8970 0.996 0.000 0.000 0.004
#> GSM25635 1 0.0921 0.8834 0.972 0.000 0.000 0.028
#> GSM25656 3 0.2466 0.7816 0.056 0.028 0.916 0.000
#> GSM25657 1 0.1576 0.8857 0.948 0.000 0.048 0.004
#> GSM25658 3 0.3351 0.8034 0.148 0.008 0.844 0.000
#> GSM25659 1 0.5527 0.6554 0.740 0.080 0.172 0.008
#> GSM25660 1 0.0336 0.8985 0.992 0.000 0.008 0.000
#> GSM25661 1 0.0336 0.8953 0.992 0.000 0.000 0.008
#> GSM25662 2 0.0859 0.9231 0.008 0.980 0.004 0.008
#> GSM25663 2 0.2505 0.9152 0.008 0.920 0.020 0.052
#> GSM25680 2 0.2528 0.9114 0.008 0.908 0.004 0.080
#> GSM25681 2 0.2803 0.9088 0.012 0.900 0.008 0.080
#> GSM25682 2 0.2675 0.8955 0.000 0.908 0.048 0.044
#> GSM25683 2 0.2670 0.8968 0.000 0.908 0.052 0.040
#> GSM25684 2 0.0672 0.9228 0.008 0.984 0.000 0.008
#> GSM25685 2 0.2546 0.9133 0.008 0.920 0.044 0.028
#> GSM25686 2 0.2675 0.8955 0.000 0.908 0.048 0.044
#> GSM25687 2 0.2675 0.8955 0.000 0.908 0.048 0.044
#> GSM48664 4 0.3725 0.8671 0.120 0.004 0.028 0.848
#> GSM48665 4 0.4419 0.8380 0.176 0.004 0.028 0.792
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM25548 5 0.2612 0.70885 0.000 0.124 0.008 0.000 0.868
#> GSM25549 5 0.0727 0.65974 0.000 0.012 0.004 0.004 0.980
#> GSM25550 5 0.0727 0.65764 0.000 0.012 0.004 0.004 0.980
#> GSM25551 2 0.5300 0.29098 0.000 0.604 0.068 0.000 0.328
#> GSM25570 5 0.0613 0.65780 0.000 0.008 0.004 0.004 0.984
#> GSM25571 5 0.0833 0.66147 0.000 0.016 0.004 0.004 0.976
#> GSM25358 4 0.8993 0.27203 0.100 0.104 0.224 0.424 0.148
#> GSM25359 5 0.5447 0.64522 0.000 0.280 0.084 0.004 0.632
#> GSM25360 3 0.3578 0.69871 0.204 0.008 0.784 0.004 0.000
#> GSM25361 5 0.5030 0.33911 0.016 0.012 0.356 0.004 0.612
#> GSM25377 4 0.0162 0.84642 0.004 0.000 0.000 0.996 0.000
#> GSM25378 4 0.0162 0.84642 0.004 0.000 0.000 0.996 0.000
#> GSM25401 4 0.4643 0.55601 0.012 0.012 0.320 0.656 0.000
#> GSM25402 4 0.4497 0.63235 0.028 0.008 0.248 0.716 0.000
#> GSM25349 2 0.2629 0.78570 0.000 0.860 0.004 0.000 0.136
#> GSM25350 2 0.2074 0.80505 0.000 0.896 0.000 0.000 0.104
#> GSM25356 4 0.0162 0.84642 0.004 0.000 0.000 0.996 0.000
#> GSM25357 2 0.3289 0.80400 0.000 0.860 0.036 0.016 0.088
#> GSM25385 3 0.3768 0.72525 0.228 0.004 0.760 0.008 0.000
#> GSM25386 3 0.1628 0.78052 0.056 0.008 0.936 0.000 0.000
#> GSM25399 4 0.2390 0.81172 0.084 0.000 0.020 0.896 0.000
#> GSM25400 4 0.5714 0.41601 0.312 0.000 0.108 0.580 0.000
#> GSM48659 5 0.4719 0.69658 0.000 0.248 0.056 0.000 0.696
#> GSM48660 2 0.1478 0.81495 0.000 0.936 0.000 0.000 0.064
#> GSM25409 5 0.4262 0.42739 0.000 0.440 0.000 0.000 0.560
#> GSM25410 3 0.2017 0.78604 0.080 0.008 0.912 0.000 0.000
#> GSM25426 2 0.4328 0.71898 0.000 0.780 0.108 0.004 0.108
#> GSM25427 4 0.0290 0.84577 0.008 0.000 0.000 0.992 0.000
#> GSM25540 3 0.4552 0.31941 0.000 0.012 0.632 0.004 0.352
#> GSM25541 3 0.4692 0.00981 0.000 0.008 0.528 0.004 0.460
#> GSM25542 5 0.5596 0.53938 0.000 0.376 0.068 0.004 0.552
#> GSM25543 5 0.5849 0.64067 0.000 0.264 0.128 0.004 0.604
#> GSM25479 1 0.0290 0.90162 0.992 0.000 0.008 0.000 0.000
#> GSM25480 1 0.0771 0.90187 0.976 0.000 0.020 0.004 0.000
#> GSM25481 4 0.0162 0.84642 0.004 0.000 0.000 0.996 0.000
#> GSM25482 4 0.0162 0.84642 0.004 0.000 0.000 0.996 0.000
#> GSM48654 5 0.4927 0.65968 0.000 0.296 0.052 0.000 0.652
#> GSM48650 2 0.3400 0.77610 0.000 0.848 0.076 0.004 0.072
#> GSM48651 2 0.4603 0.45330 0.000 0.668 0.032 0.000 0.300
#> GSM48652 2 0.5236 -0.28288 0.000 0.492 0.044 0.000 0.464
#> GSM48653 5 0.5193 0.56756 0.000 0.364 0.052 0.000 0.584
#> GSM48662 5 0.4497 0.50064 0.000 0.424 0.008 0.000 0.568
#> GSM48663 2 0.1831 0.81552 0.000 0.920 0.000 0.004 0.076
#> GSM25524 3 0.4517 0.49897 0.372 0.008 0.616 0.004 0.000
#> GSM25525 1 0.2753 0.84111 0.856 0.008 0.136 0.000 0.000
#> GSM25526 3 0.2463 0.78817 0.100 0.008 0.888 0.004 0.000
#> GSM25527 1 0.1764 0.89026 0.928 0.000 0.064 0.008 0.000
#> GSM25528 1 0.3937 0.65926 0.736 0.008 0.252 0.004 0.000
#> GSM25529 1 0.2753 0.84111 0.856 0.008 0.136 0.000 0.000
#> GSM25530 3 0.4524 0.40082 0.420 0.004 0.572 0.004 0.000
#> GSM25531 1 0.3044 0.82330 0.840 0.004 0.148 0.008 0.000
#> GSM48661 5 0.4741 0.71282 0.000 0.204 0.068 0.004 0.724
#> GSM25561 3 0.4698 0.21621 0.468 0.008 0.520 0.004 0.000
#> GSM25562 1 0.3536 0.79514 0.812 0.000 0.156 0.032 0.000
#> GSM25563 3 0.2694 0.78489 0.128 0.004 0.864 0.004 0.000
#> GSM25564 5 0.7050 0.52268 0.084 0.116 0.224 0.004 0.572
#> GSM25565 5 0.4968 0.38268 0.000 0.456 0.028 0.000 0.516
#> GSM25566 5 0.4300 0.37745 0.000 0.476 0.000 0.000 0.524
#> GSM25568 5 0.4806 0.71741 0.004 0.156 0.092 0.004 0.744
#> GSM25569 5 0.4777 0.67227 0.000 0.292 0.044 0.000 0.664
#> GSM25552 5 0.0613 0.65780 0.000 0.008 0.004 0.004 0.984
#> GSM25553 5 0.0854 0.66023 0.000 0.012 0.008 0.004 0.976
#> GSM25578 1 0.0451 0.90017 0.988 0.000 0.004 0.008 0.000
#> GSM25579 1 0.3209 0.81725 0.812 0.008 0.180 0.000 0.000
#> GSM25580 1 0.0510 0.89755 0.984 0.000 0.000 0.016 0.000
#> GSM25581 1 0.0290 0.89854 0.992 0.000 0.000 0.008 0.000
#> GSM48655 2 0.1197 0.81156 0.000 0.952 0.000 0.000 0.048
#> GSM48656 5 0.3988 0.70181 0.000 0.252 0.016 0.000 0.732
#> GSM48657 2 0.1544 0.81463 0.000 0.932 0.000 0.000 0.068
#> GSM48658 5 0.4109 0.72445 0.000 0.148 0.060 0.004 0.788
#> GSM25624 1 0.0510 0.89888 0.984 0.000 0.000 0.016 0.000
#> GSM25625 3 0.2338 0.79057 0.112 0.000 0.884 0.004 0.000
#> GSM25626 3 0.2017 0.78604 0.080 0.008 0.912 0.000 0.000
#> GSM25627 3 0.2764 0.77872 0.072 0.020 0.892 0.004 0.012
#> GSM25628 3 0.0451 0.74681 0.000 0.008 0.988 0.000 0.004
#> GSM25629 3 0.1074 0.74226 0.000 0.012 0.968 0.004 0.016
#> GSM25630 3 0.3487 0.73494 0.212 0.008 0.780 0.000 0.000
#> GSM25631 5 0.4049 0.67527 0.004 0.052 0.140 0.004 0.800
#> GSM25632 3 0.3579 0.71744 0.240 0.004 0.756 0.000 0.000
#> GSM25633 1 0.0510 0.89888 0.984 0.000 0.000 0.016 0.000
#> GSM25634 1 0.0693 0.90215 0.980 0.000 0.012 0.008 0.000
#> GSM25635 1 0.1043 0.88355 0.960 0.000 0.000 0.040 0.000
#> GSM25656 3 0.0960 0.74621 0.000 0.016 0.972 0.004 0.008
#> GSM25657 1 0.1857 0.88932 0.928 0.004 0.060 0.008 0.000
#> GSM25658 3 0.2818 0.78628 0.128 0.008 0.860 0.004 0.000
#> GSM25659 1 0.3972 0.75999 0.764 0.008 0.212 0.000 0.016
#> GSM25660 1 0.0898 0.90284 0.972 0.000 0.020 0.008 0.000
#> GSM25661 1 0.0290 0.89854 0.992 0.000 0.000 0.008 0.000
#> GSM25662 5 0.5175 0.49232 0.000 0.408 0.044 0.000 0.548
#> GSM25663 5 0.4077 0.72398 0.000 0.172 0.044 0.004 0.780
#> GSM25680 5 0.3622 0.72242 0.000 0.124 0.056 0.000 0.820
#> GSM25681 5 0.2824 0.71372 0.000 0.096 0.032 0.000 0.872
#> GSM25682 2 0.0510 0.79307 0.000 0.984 0.000 0.000 0.016
#> GSM25683 2 0.1818 0.80684 0.000 0.932 0.024 0.000 0.044
#> GSM25684 5 0.5131 0.57655 0.000 0.364 0.048 0.000 0.588
#> GSM25685 2 0.5543 0.51034 0.000 0.648 0.116 0.004 0.232
#> GSM25686 2 0.0510 0.79307 0.000 0.984 0.000 0.000 0.016
#> GSM25687 2 0.0510 0.79307 0.000 0.984 0.000 0.000 0.016
#> GSM48664 4 0.0290 0.84582 0.008 0.000 0.000 0.992 0.000
#> GSM48665 4 0.2338 0.80042 0.112 0.000 0.004 0.884 0.000
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM25548 6 0.2920 0.6410 0.000 0.008 0.004 0.000 0.168 0.820
#> GSM25549 6 0.0405 0.7689 0.000 0.008 0.000 0.000 0.004 0.988
#> GSM25550 6 0.0291 0.7667 0.000 0.004 0.004 0.000 0.000 0.992
#> GSM25551 5 0.6051 0.2091 0.000 0.412 0.008 0.004 0.416 0.160
#> GSM25570 6 0.0260 0.7675 0.000 0.008 0.000 0.000 0.000 0.992
#> GSM25571 6 0.0622 0.7692 0.000 0.008 0.000 0.000 0.012 0.980
#> GSM25358 4 0.7475 0.2868 0.060 0.040 0.116 0.444 0.324 0.016
#> GSM25359 5 0.6548 0.4853 0.004 0.216 0.032 0.004 0.504 0.240
#> GSM25360 3 0.4499 0.6829 0.168 0.000 0.716 0.000 0.112 0.004
#> GSM25361 5 0.6100 0.0928 0.016 0.000 0.184 0.000 0.484 0.316
#> GSM25377 4 0.0146 0.8719 0.004 0.000 0.000 0.996 0.000 0.000
#> GSM25378 4 0.0000 0.8728 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM25401 4 0.4614 0.6063 0.000 0.000 0.228 0.676 0.096 0.000
#> GSM25402 4 0.4431 0.6456 0.004 0.000 0.200 0.712 0.084 0.000
#> GSM25349 2 0.2356 0.7151 0.000 0.884 0.008 0.004 0.004 0.100
#> GSM25350 2 0.1493 0.7499 0.000 0.936 0.004 0.000 0.004 0.056
#> GSM25356 4 0.0000 0.8728 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM25357 2 0.2722 0.7288 0.000 0.876 0.008 0.016 0.088 0.012
#> GSM25385 3 0.1714 0.7888 0.092 0.000 0.908 0.000 0.000 0.000
#> GSM25386 3 0.1267 0.7980 0.000 0.000 0.940 0.000 0.060 0.000
#> GSM25399 4 0.1151 0.8548 0.032 0.000 0.012 0.956 0.000 0.000
#> GSM25400 4 0.4650 0.5861 0.224 0.000 0.080 0.688 0.008 0.000
#> GSM48659 5 0.5525 0.5684 0.000 0.120 0.012 0.000 0.568 0.300
#> GSM48660 2 0.0405 0.7608 0.000 0.988 0.000 0.000 0.004 0.008
#> GSM25409 2 0.4822 0.3936 0.000 0.656 0.012 0.000 0.068 0.264
#> GSM25410 3 0.1387 0.7913 0.000 0.000 0.932 0.000 0.068 0.000
#> GSM25426 5 0.5543 -0.1739 0.000 0.428 0.072 0.004 0.480 0.016
#> GSM25427 4 0.0000 0.8728 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM25540 5 0.5046 0.2811 0.000 0.000 0.256 0.000 0.620 0.124
#> GSM25541 5 0.5495 0.2397 0.000 0.000 0.304 0.000 0.540 0.156
#> GSM25542 5 0.5957 0.5502 0.000 0.088 0.068 0.000 0.576 0.268
#> GSM25543 5 0.5557 0.5482 0.004 0.068 0.040 0.000 0.604 0.284
#> GSM25479 1 0.0363 0.8880 0.988 0.000 0.000 0.012 0.000 0.000
#> GSM25480 1 0.0363 0.8880 0.988 0.000 0.000 0.012 0.000 0.000
#> GSM25481 4 0.0000 0.8728 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM25482 4 0.0000 0.8728 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM48654 5 0.5439 0.5699 0.000 0.112 0.012 0.000 0.580 0.296
#> GSM48650 2 0.4464 0.6115 0.000 0.720 0.028 0.032 0.216 0.004
#> GSM48651 2 0.5948 -0.1664 0.000 0.456 0.000 0.000 0.284 0.260
#> GSM48652 5 0.5886 0.4907 0.000 0.236 0.000 0.000 0.472 0.292
#> GSM48653 5 0.5539 0.5681 0.000 0.136 0.008 0.000 0.564 0.292
#> GSM48662 2 0.6083 -0.3837 0.000 0.388 0.000 0.000 0.280 0.332
#> GSM48663 2 0.2739 0.7127 0.000 0.872 0.000 0.084 0.032 0.012
#> GSM25524 3 0.4655 0.5029 0.300 0.000 0.632 0.000 0.068 0.000
#> GSM25525 1 0.3341 0.8053 0.816 0.000 0.116 0.000 0.068 0.000
#> GSM25526 3 0.1471 0.7915 0.004 0.000 0.932 0.000 0.064 0.000
#> GSM25527 1 0.2122 0.8757 0.900 0.000 0.076 0.024 0.000 0.000
#> GSM25528 1 0.4847 0.3930 0.588 0.000 0.340 0.000 0.072 0.000
#> GSM25529 1 0.3341 0.8053 0.816 0.000 0.116 0.000 0.068 0.000
#> GSM25530 3 0.3189 0.6395 0.236 0.000 0.760 0.000 0.004 0.000
#> GSM25531 1 0.2805 0.8248 0.828 0.000 0.160 0.012 0.000 0.000
#> GSM48661 5 0.4308 0.5334 0.000 0.028 0.012 0.000 0.680 0.280
#> GSM25561 3 0.4929 0.1254 0.428 0.000 0.508 0.000 0.064 0.000
#> GSM25562 1 0.3274 0.8017 0.824 0.000 0.080 0.096 0.000 0.000
#> GSM25563 3 0.1498 0.8009 0.028 0.000 0.940 0.000 0.032 0.000
#> GSM25564 5 0.8204 0.1591 0.200 0.028 0.184 0.004 0.328 0.256
#> GSM25565 5 0.5944 0.5081 0.000 0.244 0.000 0.000 0.452 0.304
#> GSM25566 2 0.6100 -0.4082 0.000 0.384 0.000 0.000 0.308 0.308
#> GSM25568 5 0.5254 0.3711 0.004 0.012 0.056 0.000 0.524 0.404
#> GSM25569 5 0.5649 0.4761 0.000 0.132 0.004 0.000 0.464 0.400
#> GSM25552 6 0.0146 0.7613 0.000 0.000 0.000 0.000 0.004 0.996
#> GSM25553 6 0.1411 0.7515 0.000 0.000 0.004 0.000 0.060 0.936
#> GSM25578 1 0.0363 0.8880 0.988 0.000 0.000 0.012 0.000 0.000
#> GSM25579 1 0.3469 0.8023 0.812 0.000 0.120 0.000 0.064 0.004
#> GSM25580 1 0.1444 0.8643 0.928 0.000 0.000 0.072 0.000 0.000
#> GSM25581 1 0.0632 0.8877 0.976 0.000 0.000 0.024 0.000 0.000
#> GSM48655 2 0.0405 0.7605 0.000 0.988 0.000 0.000 0.008 0.004
#> GSM48656 5 0.5700 0.4094 0.000 0.160 0.000 0.000 0.436 0.404
#> GSM48657 2 0.0622 0.7604 0.000 0.980 0.000 0.000 0.008 0.012
#> GSM48658 5 0.4179 0.4967 0.000 0.016 0.008 0.000 0.652 0.324
#> GSM25624 1 0.1124 0.8872 0.956 0.000 0.008 0.036 0.000 0.000
#> GSM25625 3 0.0993 0.8011 0.012 0.000 0.964 0.000 0.024 0.000
#> GSM25626 3 0.1387 0.7907 0.000 0.000 0.932 0.000 0.068 0.000
#> GSM25627 3 0.3769 0.4840 0.004 0.000 0.640 0.000 0.356 0.000
#> GSM25628 3 0.2527 0.7857 0.000 0.000 0.832 0.000 0.168 0.000
#> GSM25629 3 0.3843 0.3944 0.000 0.000 0.548 0.000 0.452 0.000
#> GSM25630 3 0.2745 0.7829 0.068 0.000 0.864 0.000 0.068 0.000
#> GSM25631 6 0.4683 0.2813 0.000 0.000 0.064 0.000 0.320 0.616
#> GSM25632 3 0.1471 0.7966 0.064 0.000 0.932 0.000 0.004 0.000
#> GSM25633 1 0.0632 0.8877 0.976 0.000 0.000 0.024 0.000 0.000
#> GSM25634 1 0.0790 0.8861 0.968 0.000 0.000 0.032 0.000 0.000
#> GSM25635 1 0.1957 0.8333 0.888 0.000 0.000 0.112 0.000 0.000
#> GSM25656 3 0.2597 0.7790 0.000 0.000 0.824 0.000 0.176 0.000
#> GSM25657 1 0.1802 0.8740 0.916 0.000 0.072 0.012 0.000 0.000
#> GSM25658 3 0.2801 0.7859 0.068 0.000 0.860 0.000 0.072 0.000
#> GSM25659 1 0.4357 0.7170 0.732 0.000 0.156 0.000 0.108 0.004
#> GSM25660 1 0.0458 0.8882 0.984 0.000 0.000 0.016 0.000 0.000
#> GSM25661 1 0.0937 0.8825 0.960 0.000 0.000 0.040 0.000 0.000
#> GSM25662 5 0.5422 0.5704 0.000 0.136 0.004 0.000 0.572 0.288
#> GSM25663 5 0.5448 0.4755 0.000 0.092 0.012 0.000 0.532 0.364
#> GSM25680 6 0.4289 0.2509 0.000 0.008 0.020 0.000 0.332 0.640
#> GSM25681 6 0.3541 0.5273 0.000 0.000 0.020 0.000 0.232 0.748
#> GSM25682 2 0.0000 0.7585 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM25683 2 0.2417 0.7343 0.000 0.888 0.008 0.004 0.088 0.012
#> GSM25684 5 0.5389 0.5695 0.000 0.132 0.004 0.000 0.576 0.288
#> GSM25685 5 0.5130 0.4330 0.000 0.124 0.080 0.000 0.708 0.088
#> GSM25686 2 0.0146 0.7567 0.000 0.996 0.000 0.000 0.004 0.000
#> GSM25687 2 0.0000 0.7585 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM48664 4 0.0000 0.8728 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM48665 4 0.0458 0.8672 0.016 0.000 0.000 0.984 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 genotype/variation(p) k
#> MAD:mclust 97 5.16e-06 2
#> MAD:mclust 92 3.78e-04 3
#> MAD:mclust 94 1.13e-05 4
#> MAD:mclust 85 1.61e-06 5
#> MAD:mclust 75 4.67e-09 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 100 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 0.816 0.884 0.952 0.5034 0.495 0.495
#> 3 3 0.492 0.587 0.802 0.3202 0.767 0.561
#> 4 4 0.420 0.396 0.617 0.1115 0.858 0.620
#> 5 5 0.477 0.439 0.640 0.0712 0.803 0.422
#> 6 6 0.527 0.392 0.598 0.0431 0.892 0.564
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
#> GSM25548 2 0.0000 0.965 0.000 1.000
#> GSM25549 2 0.0000 0.965 0.000 1.000
#> GSM25550 2 0.0000 0.965 0.000 1.000
#> GSM25551 2 0.0000 0.965 0.000 1.000
#> GSM25570 2 0.0000 0.965 0.000 1.000
#> GSM25571 2 0.0000 0.965 0.000 1.000
#> GSM25358 1 0.7056 0.756 0.808 0.192
#> GSM25359 2 0.4161 0.890 0.084 0.916
#> GSM25360 1 0.0672 0.927 0.992 0.008
#> GSM25361 1 0.9170 0.504 0.668 0.332
#> GSM25377 1 0.0000 0.931 1.000 0.000
#> GSM25378 1 0.3733 0.879 0.928 0.072
#> GSM25401 1 0.9732 0.372 0.596 0.404
#> GSM25402 1 0.4161 0.869 0.916 0.084
#> GSM25349 2 0.0000 0.965 0.000 1.000
#> GSM25350 2 0.0000 0.965 0.000 1.000
#> GSM25356 1 0.3879 0.876 0.924 0.076
#> GSM25357 2 0.0000 0.965 0.000 1.000
#> GSM25385 1 0.0000 0.931 1.000 0.000
#> GSM25386 1 0.2236 0.907 0.964 0.036
#> GSM25399 1 0.0000 0.931 1.000 0.000
#> GSM25400 1 0.0000 0.931 1.000 0.000
#> GSM48659 2 0.0000 0.965 0.000 1.000
#> GSM48660 2 0.0000 0.965 0.000 1.000
#> GSM25409 2 0.0000 0.965 0.000 1.000
#> GSM25410 1 0.1633 0.916 0.976 0.024
#> GSM25426 2 0.0000 0.965 0.000 1.000
#> GSM25427 1 0.1633 0.916 0.976 0.024
#> GSM25540 2 0.8443 0.620 0.272 0.728
#> GSM25541 1 0.9954 0.154 0.540 0.460
#> GSM25542 2 0.0000 0.965 0.000 1.000
#> GSM25543 2 0.1184 0.952 0.016 0.984
#> GSM25479 1 0.0000 0.931 1.000 0.000
#> GSM25480 1 0.0000 0.931 1.000 0.000
#> GSM25481 1 0.9427 0.473 0.640 0.360
#> GSM25482 1 0.8713 0.604 0.708 0.292
#> GSM48654 2 0.0000 0.965 0.000 1.000
#> GSM48650 2 0.0000 0.965 0.000 1.000
#> GSM48651 2 0.0000 0.965 0.000 1.000
#> GSM48652 2 0.0000 0.965 0.000 1.000
#> GSM48653 2 0.0000 0.965 0.000 1.000
#> GSM48662 2 0.0000 0.965 0.000 1.000
#> GSM48663 2 0.0000 0.965 0.000 1.000
#> GSM25524 1 0.0000 0.931 1.000 0.000
#> GSM25525 1 0.0000 0.931 1.000 0.000
#> GSM25526 1 0.0000 0.931 1.000 0.000
#> GSM25527 1 0.0000 0.931 1.000 0.000
#> GSM25528 1 0.0000 0.931 1.000 0.000
#> GSM25529 1 0.0000 0.931 1.000 0.000
#> GSM25530 1 0.0000 0.931 1.000 0.000
#> GSM25531 1 0.0000 0.931 1.000 0.000
#> GSM48661 2 0.0000 0.965 0.000 1.000
#> GSM25561 1 0.0000 0.931 1.000 0.000
#> GSM25562 1 0.0000 0.931 1.000 0.000
#> GSM25563 1 0.0672 0.927 0.992 0.008
#> GSM25564 1 0.9866 0.301 0.568 0.432
#> GSM25565 2 0.0000 0.965 0.000 1.000
#> GSM25566 2 0.0000 0.965 0.000 1.000
#> GSM25568 2 0.5629 0.833 0.132 0.868
#> GSM25569 2 0.0000 0.965 0.000 1.000
#> GSM25552 2 0.0000 0.965 0.000 1.000
#> GSM25553 2 0.5059 0.857 0.112 0.888
#> GSM25578 1 0.0000 0.931 1.000 0.000
#> GSM25579 1 0.0000 0.931 1.000 0.000
#> GSM25580 1 0.0000 0.931 1.000 0.000
#> GSM25581 1 0.0000 0.931 1.000 0.000
#> GSM48655 2 0.0000 0.965 0.000 1.000
#> GSM48656 2 0.0000 0.965 0.000 1.000
#> GSM48657 2 0.0000 0.965 0.000 1.000
#> GSM48658 2 0.0000 0.965 0.000 1.000
#> GSM25624 1 0.0000 0.931 1.000 0.000
#> GSM25625 1 0.0000 0.931 1.000 0.000
#> GSM25626 1 0.0672 0.927 0.992 0.008
#> GSM25627 2 0.9000 0.519 0.316 0.684
#> GSM25628 1 0.9944 0.167 0.544 0.456
#> GSM25629 2 0.3733 0.903 0.072 0.928
#> GSM25630 1 0.0000 0.931 1.000 0.000
#> GSM25631 2 0.5408 0.846 0.124 0.876
#> GSM25632 1 0.0000 0.931 1.000 0.000
#> GSM25633 1 0.0000 0.931 1.000 0.000
#> GSM25634 1 0.0000 0.931 1.000 0.000
#> GSM25635 1 0.0000 0.931 1.000 0.000
#> GSM25656 2 0.9833 0.244 0.424 0.576
#> GSM25657 1 0.0000 0.931 1.000 0.000
#> GSM25658 1 0.0000 0.931 1.000 0.000
#> GSM25659 1 0.0000 0.931 1.000 0.000
#> GSM25660 1 0.0000 0.931 1.000 0.000
#> GSM25661 1 0.0000 0.931 1.000 0.000
#> GSM25662 2 0.0000 0.965 0.000 1.000
#> GSM25663 2 0.0000 0.965 0.000 1.000
#> GSM25680 2 0.0000 0.965 0.000 1.000
#> GSM25681 2 0.0000 0.965 0.000 1.000
#> GSM25682 2 0.0000 0.965 0.000 1.000
#> GSM25683 2 0.0000 0.965 0.000 1.000
#> GSM25684 2 0.0000 0.965 0.000 1.000
#> GSM25685 2 0.0000 0.965 0.000 1.000
#> GSM25686 2 0.0000 0.965 0.000 1.000
#> GSM25687 2 0.0000 0.965 0.000 1.000
#> GSM48664 1 0.0000 0.931 1.000 0.000
#> GSM48665 1 0.0000 0.931 1.000 0.000
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM25548 2 0.1860 0.8110 0.000 0.948 0.052
#> GSM25549 2 0.1337 0.8107 0.012 0.972 0.016
#> GSM25550 2 0.3686 0.7114 0.140 0.860 0.000
#> GSM25551 2 0.4399 0.7536 0.000 0.812 0.188
#> GSM25570 2 0.1031 0.8020 0.024 0.976 0.000
#> GSM25571 2 0.1289 0.8131 0.000 0.968 0.032
#> GSM25358 1 0.7970 0.5301 0.596 0.324 0.080
#> GSM25359 2 0.6291 0.2889 0.000 0.532 0.468
#> GSM25360 3 0.3116 0.6545 0.108 0.000 0.892
#> GSM25361 3 0.0829 0.6916 0.012 0.004 0.984
#> GSM25377 1 0.4399 0.6389 0.812 0.188 0.000
#> GSM25378 1 0.5254 0.5881 0.736 0.264 0.000
#> GSM25401 1 0.8538 0.3472 0.520 0.380 0.100
#> GSM25402 1 0.6488 0.6324 0.744 0.192 0.064
#> GSM25349 2 0.2537 0.7717 0.080 0.920 0.000
#> GSM25350 2 0.2165 0.7826 0.064 0.936 0.000
#> GSM25356 1 0.5785 0.4949 0.668 0.332 0.000
#> GSM25357 2 0.1860 0.7898 0.052 0.948 0.000
#> GSM25385 3 0.6215 0.1250 0.428 0.000 0.572
#> GSM25386 3 0.1753 0.6825 0.048 0.000 0.952
#> GSM25399 1 0.1031 0.7120 0.976 0.024 0.000
#> GSM25400 1 0.1411 0.7265 0.964 0.000 0.036
#> GSM48659 2 0.6308 0.2979 0.000 0.508 0.492
#> GSM48660 2 0.1620 0.8065 0.024 0.964 0.012
#> GSM25409 2 0.2261 0.7808 0.068 0.932 0.000
#> GSM25410 3 0.2959 0.6595 0.100 0.000 0.900
#> GSM25426 2 0.5988 0.5586 0.000 0.632 0.368
#> GSM25427 1 0.5529 0.5477 0.704 0.296 0.000
#> GSM25540 3 0.1411 0.6882 0.000 0.036 0.964
#> GSM25541 3 0.0747 0.6933 0.000 0.016 0.984
#> GSM25542 3 0.6095 0.0576 0.000 0.392 0.608
#> GSM25543 3 0.5785 0.2562 0.000 0.332 0.668
#> GSM25479 1 0.1753 0.7271 0.952 0.000 0.048
#> GSM25480 1 0.2096 0.7270 0.944 0.004 0.052
#> GSM25481 1 0.6274 0.2433 0.544 0.456 0.000
#> GSM25482 1 0.6260 0.2637 0.552 0.448 0.000
#> GSM48654 2 0.6045 0.5365 0.000 0.620 0.380
#> GSM48650 2 0.1529 0.8128 0.000 0.960 0.040
#> GSM48651 2 0.3619 0.7855 0.000 0.864 0.136
#> GSM48652 2 0.4555 0.7452 0.000 0.800 0.200
#> GSM48653 2 0.6286 0.3698 0.000 0.536 0.464
#> GSM48662 2 0.1411 0.8132 0.000 0.964 0.036
#> GSM48663 2 0.2356 0.7779 0.072 0.928 0.000
#> GSM25524 3 0.5529 0.4482 0.296 0.000 0.704
#> GSM25525 1 0.4931 0.6244 0.768 0.000 0.232
#> GSM25526 3 0.4555 0.5773 0.200 0.000 0.800
#> GSM25527 1 0.5178 0.6004 0.744 0.000 0.256
#> GSM25528 1 0.6299 0.1585 0.524 0.000 0.476
#> GSM25529 1 0.5621 0.5297 0.692 0.000 0.308
#> GSM25530 1 0.6308 0.1063 0.508 0.000 0.492
#> GSM25531 1 0.5363 0.5755 0.724 0.000 0.276
#> GSM48661 3 0.5291 0.3886 0.000 0.268 0.732
#> GSM25561 3 0.6126 0.2069 0.400 0.000 0.600
#> GSM25562 1 0.4555 0.6558 0.800 0.000 0.200
#> GSM25563 3 0.4235 0.6037 0.176 0.000 0.824
#> GSM25564 1 0.9986 0.0877 0.352 0.308 0.340
#> GSM25565 2 0.4002 0.7714 0.000 0.840 0.160
#> GSM25566 2 0.2878 0.7999 0.000 0.904 0.096
#> GSM25568 3 0.6823 -0.1386 0.012 0.484 0.504
#> GSM25569 2 0.3551 0.7860 0.000 0.868 0.132
#> GSM25552 2 0.2356 0.7778 0.072 0.928 0.000
#> GSM25553 2 0.5465 0.4632 0.288 0.712 0.000
#> GSM25578 1 0.2796 0.7149 0.908 0.000 0.092
#> GSM25579 1 0.5254 0.5896 0.736 0.000 0.264
#> GSM25580 1 0.1289 0.7250 0.968 0.000 0.032
#> GSM25581 1 0.2066 0.7239 0.940 0.000 0.060
#> GSM48655 2 0.0747 0.8117 0.000 0.984 0.016
#> GSM48656 2 0.2796 0.8027 0.000 0.908 0.092
#> GSM48657 2 0.1399 0.8019 0.028 0.968 0.004
#> GSM48658 3 0.5760 0.2573 0.000 0.328 0.672
#> GSM25624 1 0.1753 0.7256 0.952 0.000 0.048
#> GSM25625 3 0.5363 0.4780 0.276 0.000 0.724
#> GSM25626 3 0.1643 0.6836 0.044 0.000 0.956
#> GSM25627 3 0.1877 0.6923 0.012 0.032 0.956
#> GSM25628 3 0.1015 0.6932 0.008 0.012 0.980
#> GSM25629 3 0.1964 0.6791 0.000 0.056 0.944
#> GSM25630 3 0.4931 0.5408 0.232 0.000 0.768
#> GSM25631 3 0.4504 0.5346 0.000 0.196 0.804
#> GSM25632 3 0.6215 0.1201 0.428 0.000 0.572
#> GSM25633 1 0.3192 0.7071 0.888 0.000 0.112
#> GSM25634 1 0.3267 0.7057 0.884 0.000 0.116
#> GSM25635 1 0.1753 0.7263 0.952 0.000 0.048
#> GSM25656 3 0.0747 0.6930 0.000 0.016 0.984
#> GSM25657 1 0.4796 0.6360 0.780 0.000 0.220
#> GSM25658 3 0.5560 0.4432 0.300 0.000 0.700
#> GSM25659 1 0.6442 0.2751 0.564 0.004 0.432
#> GSM25660 1 0.2280 0.7273 0.940 0.008 0.052
#> GSM25661 1 0.1411 0.7261 0.964 0.000 0.036
#> GSM25662 2 0.5968 0.5609 0.000 0.636 0.364
#> GSM25663 2 0.5178 0.6879 0.000 0.744 0.256
#> GSM25680 2 0.6267 0.3950 0.000 0.548 0.452
#> GSM25681 2 0.6252 0.3986 0.000 0.556 0.444
#> GSM25682 2 0.0592 0.8112 0.000 0.988 0.012
#> GSM25683 2 0.1753 0.8115 0.000 0.952 0.048
#> GSM25684 2 0.5465 0.6610 0.000 0.712 0.288
#> GSM25685 3 0.6280 -0.1990 0.000 0.460 0.540
#> GSM25686 2 0.0983 0.8114 0.004 0.980 0.016
#> GSM25687 2 0.1129 0.8046 0.020 0.976 0.004
#> GSM48664 1 0.2959 0.6863 0.900 0.100 0.000
#> GSM48665 1 0.2711 0.6917 0.912 0.088 0.000
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM25548 2 0.506 0.49992 0.000 0.648 0.012 0.340
#> GSM25549 2 0.524 0.40347 0.004 0.576 0.004 0.416
#> GSM25550 2 0.584 0.30730 0.032 0.520 0.000 0.448
#> GSM25551 2 0.573 0.43868 0.000 0.616 0.344 0.040
#> GSM25570 2 0.513 0.36529 0.004 0.552 0.000 0.444
#> GSM25571 2 0.535 0.43399 0.000 0.596 0.016 0.388
#> GSM25358 1 0.812 0.43178 0.552 0.176 0.216 0.056
#> GSM25359 3 0.604 0.01164 0.004 0.416 0.544 0.036
#> GSM25360 3 0.600 -0.00267 0.040 0.000 0.512 0.448
#> GSM25361 4 0.591 -0.00948 0.012 0.016 0.456 0.516
#> GSM25377 1 0.377 0.65094 0.868 0.064 0.020 0.048
#> GSM25378 1 0.421 0.64030 0.840 0.096 0.016 0.048
#> GSM25401 1 0.903 0.04091 0.356 0.260 0.324 0.060
#> GSM25402 1 0.783 0.46329 0.588 0.156 0.200 0.056
#> GSM25349 2 0.468 0.64498 0.084 0.816 0.016 0.084
#> GSM25350 2 0.411 0.64668 0.032 0.812 0.000 0.156
#> GSM25356 1 0.484 0.60917 0.792 0.136 0.008 0.064
#> GSM25357 2 0.690 0.47847 0.100 0.668 0.184 0.048
#> GSM25385 1 0.671 0.29775 0.528 0.000 0.376 0.096
#> GSM25386 3 0.520 0.36876 0.088 0.000 0.752 0.160
#> GSM25399 1 0.359 0.66047 0.880 0.040 0.044 0.036
#> GSM25400 1 0.335 0.66314 0.888 0.028 0.060 0.024
#> GSM48659 2 0.700 0.33766 0.000 0.508 0.368 0.124
#> GSM48660 2 0.418 0.67598 0.028 0.848 0.044 0.080
#> GSM25409 2 0.440 0.59859 0.016 0.760 0.000 0.224
#> GSM25410 3 0.574 0.33308 0.232 0.008 0.700 0.060
#> GSM25426 3 0.628 -0.17891 0.000 0.464 0.480 0.056
#> GSM25427 1 0.521 0.57169 0.756 0.140 0.000 0.104
#> GSM25540 3 0.517 0.18817 0.000 0.012 0.620 0.368
#> GSM25541 3 0.537 0.11636 0.004 0.008 0.576 0.412
#> GSM25542 3 0.611 0.01203 0.000 0.392 0.556 0.052
#> GSM25543 3 0.688 0.16467 0.000 0.340 0.540 0.120
#> GSM25479 1 0.542 0.50694 0.680 0.004 0.032 0.284
#> GSM25480 1 0.685 0.20236 0.492 0.032 0.040 0.436
#> GSM25481 1 0.683 0.42581 0.604 0.256 0.004 0.136
#> GSM25482 1 0.704 0.39612 0.592 0.224 0.004 0.180
#> GSM48654 2 0.641 0.46588 0.000 0.592 0.320 0.088
#> GSM48650 2 0.649 0.38730 0.016 0.600 0.328 0.056
#> GSM48651 2 0.401 0.65593 0.000 0.816 0.156 0.028
#> GSM48652 2 0.508 0.59559 0.000 0.716 0.248 0.036
#> GSM48653 2 0.621 0.41792 0.000 0.576 0.360 0.064
#> GSM48662 2 0.360 0.68329 0.000 0.848 0.028 0.124
#> GSM48663 2 0.482 0.65611 0.048 0.808 0.028 0.116
#> GSM25524 3 0.711 -0.10082 0.128 0.000 0.456 0.416
#> GSM25525 4 0.712 0.16447 0.368 0.000 0.136 0.496
#> GSM25526 3 0.625 0.15110 0.340 0.004 0.596 0.060
#> GSM25527 1 0.623 0.48921 0.668 0.000 0.148 0.184
#> GSM25528 4 0.788 0.19377 0.332 0.000 0.288 0.380
#> GSM25529 4 0.750 0.21430 0.344 0.000 0.192 0.464
#> GSM25530 1 0.670 0.38577 0.612 0.000 0.232 0.156
#> GSM25531 1 0.460 0.60682 0.796 0.000 0.132 0.072
#> GSM48661 3 0.685 0.33783 0.000 0.204 0.600 0.196
#> GSM25561 4 0.768 0.06720 0.216 0.000 0.384 0.400
#> GSM25562 1 0.509 0.60146 0.772 0.004 0.084 0.140
#> GSM25563 3 0.722 0.10932 0.180 0.000 0.536 0.284
#> GSM25564 4 0.940 0.20877 0.120 0.256 0.220 0.404
#> GSM25565 2 0.435 0.63014 0.000 0.780 0.196 0.024
#> GSM25566 2 0.390 0.67254 0.000 0.832 0.132 0.036
#> GSM25568 2 0.803 0.10423 0.012 0.412 0.368 0.208
#> GSM25569 2 0.471 0.66645 0.000 0.788 0.072 0.140
#> GSM25552 4 0.551 -0.32946 0.016 0.488 0.000 0.496
#> GSM25553 4 0.630 -0.20667 0.060 0.420 0.000 0.520
#> GSM25578 1 0.441 0.59736 0.780 0.000 0.028 0.192
#> GSM25579 4 0.687 0.37077 0.216 0.016 0.132 0.636
#> GSM25580 1 0.241 0.66016 0.896 0.000 0.000 0.104
#> GSM25581 1 0.280 0.65973 0.884 0.000 0.008 0.108
#> GSM48655 2 0.293 0.69196 0.000 0.896 0.056 0.048
#> GSM48656 2 0.476 0.64805 0.000 0.772 0.052 0.176
#> GSM48657 2 0.297 0.68240 0.028 0.904 0.016 0.052
#> GSM48658 3 0.761 0.18385 0.000 0.260 0.476 0.264
#> GSM25624 1 0.310 0.65838 0.872 0.008 0.004 0.116
#> GSM25625 3 0.734 0.03640 0.376 0.000 0.464 0.160
#> GSM25626 3 0.477 0.40030 0.116 0.012 0.804 0.068
#> GSM25627 3 0.695 0.38156 0.104 0.192 0.660 0.044
#> GSM25628 3 0.427 0.41127 0.008 0.036 0.820 0.136
#> GSM25629 3 0.430 0.44138 0.012 0.116 0.828 0.044
#> GSM25630 3 0.726 0.02985 0.152 0.000 0.480 0.368
#> GSM25631 4 0.655 0.15512 0.000 0.096 0.328 0.576
#> GSM25632 1 0.734 0.20015 0.504 0.000 0.316 0.180
#> GSM25633 1 0.398 0.64846 0.836 0.000 0.056 0.108
#> GSM25634 1 0.276 0.66307 0.904 0.000 0.044 0.052
#> GSM25635 1 0.310 0.66058 0.872 0.004 0.008 0.116
#> GSM25656 3 0.379 0.43224 0.012 0.040 0.860 0.088
#> GSM25657 1 0.510 0.60090 0.764 0.000 0.100 0.136
#> GSM25658 1 0.632 0.17356 0.480 0.004 0.468 0.048
#> GSM25659 4 0.708 0.31306 0.168 0.000 0.276 0.556
#> GSM25660 1 0.588 0.37787 0.580 0.020 0.012 0.388
#> GSM25661 1 0.316 0.65472 0.868 0.008 0.004 0.120
#> GSM25662 2 0.531 0.38363 0.000 0.576 0.412 0.012
#> GSM25663 2 0.636 0.59907 0.000 0.656 0.176 0.168
#> GSM25680 2 0.782 0.15897 0.000 0.376 0.256 0.368
#> GSM25681 4 0.766 0.10097 0.000 0.296 0.244 0.460
#> GSM25682 2 0.203 0.69150 0.000 0.936 0.028 0.036
#> GSM25683 2 0.443 0.59869 0.000 0.772 0.204 0.024
#> GSM25684 2 0.567 0.53802 0.000 0.652 0.300 0.048
#> GSM25685 3 0.559 -0.13829 0.000 0.456 0.524 0.020
#> GSM25686 2 0.130 0.69002 0.000 0.964 0.016 0.020
#> GSM25687 2 0.158 0.68963 0.000 0.948 0.004 0.048
#> GSM48664 1 0.258 0.66101 0.912 0.052 0.000 0.036
#> GSM48665 1 0.249 0.66390 0.916 0.036 0.000 0.048
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM25548 5 0.485 0.57371 0.000 0.196 0.000 0.092 0.712
#> GSM25549 5 0.414 0.61864 0.000 0.200 0.008 0.028 0.764
#> GSM25550 5 0.386 0.63708 0.024 0.156 0.000 0.016 0.804
#> GSM25551 4 0.407 0.50560 0.000 0.136 0.000 0.788 0.076
#> GSM25570 5 0.360 0.63876 0.000 0.180 0.000 0.024 0.796
#> GSM25571 5 0.440 0.63070 0.000 0.152 0.008 0.068 0.772
#> GSM25358 4 0.637 0.08630 0.344 0.016 0.016 0.548 0.076
#> GSM25359 4 0.566 0.48511 0.004 0.072 0.084 0.720 0.120
#> GSM25360 3 0.203 0.63967 0.044 0.000 0.928 0.016 0.012
#> GSM25361 3 0.379 0.58776 0.020 0.008 0.824 0.016 0.132
#> GSM25377 1 0.618 0.58896 0.668 0.168 0.004 0.072 0.088
#> GSM25378 1 0.480 0.64887 0.752 0.028 0.000 0.164 0.056
#> GSM25401 4 0.471 0.46195 0.208 0.056 0.000 0.728 0.008
#> GSM25402 1 0.594 0.27886 0.536 0.052 0.004 0.388 0.020
#> GSM25349 2 0.539 0.51904 0.044 0.724 0.000 0.096 0.136
#> GSM25350 2 0.500 0.35963 0.012 0.672 0.000 0.040 0.276
#> GSM25356 1 0.585 0.63025 0.696 0.048 0.004 0.128 0.124
#> GSM25357 4 0.585 0.42928 0.020 0.196 0.004 0.664 0.116
#> GSM25385 1 0.708 0.41208 0.536 0.004 0.168 0.248 0.044
#> GSM25386 3 0.788 0.54571 0.060 0.152 0.532 0.204 0.052
#> GSM25399 1 0.476 0.64652 0.784 0.092 0.004 0.044 0.076
#> GSM25400 1 0.323 0.68000 0.872 0.024 0.004 0.072 0.028
#> GSM48659 2 0.818 0.30759 0.000 0.372 0.300 0.196 0.132
#> GSM48660 2 0.215 0.59683 0.000 0.916 0.000 0.048 0.036
#> GSM25409 5 0.580 0.13514 0.008 0.432 0.000 0.068 0.492
#> GSM25410 3 0.925 0.36533 0.188 0.148 0.344 0.256 0.064
#> GSM25426 4 0.311 0.50898 0.000 0.144 0.004 0.840 0.012
#> GSM25427 1 0.671 0.54907 0.608 0.188 0.004 0.056 0.144
#> GSM25540 3 0.240 0.63171 0.000 0.036 0.912 0.040 0.012
#> GSM25541 3 0.259 0.63555 0.012 0.008 0.908 0.028 0.044
#> GSM25542 2 0.670 0.32327 0.000 0.572 0.228 0.160 0.040
#> GSM25543 2 0.645 0.22068 0.000 0.572 0.292 0.088 0.048
#> GSM25479 1 0.533 0.59244 0.672 0.000 0.052 0.024 0.252
#> GSM25480 5 0.639 -0.26435 0.420 0.000 0.084 0.028 0.468
#> GSM25481 1 0.696 0.43585 0.536 0.288 0.000 0.084 0.092
#> GSM25482 1 0.687 0.41710 0.556 0.236 0.000 0.048 0.160
#> GSM48654 2 0.590 0.54353 0.000 0.652 0.228 0.076 0.044
#> GSM48650 2 0.466 0.20303 0.004 0.556 0.000 0.432 0.008
#> GSM48651 2 0.412 0.58341 0.000 0.804 0.032 0.132 0.032
#> GSM48652 2 0.438 0.58521 0.000 0.792 0.080 0.108 0.020
#> GSM48653 2 0.628 0.50970 0.000 0.616 0.196 0.160 0.028
#> GSM48662 2 0.363 0.58175 0.000 0.836 0.028 0.024 0.112
#> GSM48663 2 0.402 0.53750 0.036 0.824 0.000 0.052 0.088
#> GSM25524 3 0.391 0.58590 0.108 0.000 0.824 0.032 0.036
#> GSM25525 1 0.729 0.22556 0.396 0.000 0.300 0.024 0.280
#> GSM25526 4 0.685 0.03447 0.316 0.008 0.168 0.496 0.012
#> GSM25527 1 0.562 0.62354 0.712 0.000 0.124 0.060 0.104
#> GSM25528 3 0.558 0.27703 0.304 0.000 0.620 0.020 0.056
#> GSM25529 1 0.702 0.21494 0.432 0.000 0.348 0.020 0.200
#> GSM25530 1 0.608 0.48014 0.624 0.000 0.248 0.092 0.036
#> GSM25531 1 0.387 0.66116 0.824 0.000 0.084 0.080 0.012
#> GSM48661 3 0.583 0.01114 0.000 0.440 0.488 0.056 0.016
#> GSM25561 3 0.686 0.60628 0.100 0.132 0.652 0.044 0.072
#> GSM25562 1 0.852 0.28278 0.448 0.272 0.092 0.068 0.120
#> GSM25563 3 0.683 0.60822 0.092 0.144 0.652 0.048 0.064
#> GSM25564 2 0.899 -0.00747 0.124 0.332 0.308 0.048 0.188
#> GSM25565 2 0.576 0.54713 0.000 0.684 0.080 0.184 0.052
#> GSM25566 2 0.681 0.16318 0.000 0.416 0.004 0.340 0.240
#> GSM25568 2 0.627 0.29539 0.004 0.612 0.260 0.040 0.084
#> GSM25569 2 0.515 0.54364 0.000 0.724 0.080 0.024 0.172
#> GSM25552 5 0.377 0.62429 0.008 0.200 0.000 0.012 0.780
#> GSM25553 5 0.405 0.62174 0.016 0.176 0.012 0.008 0.788
#> GSM25578 1 0.549 0.62187 0.696 0.000 0.092 0.028 0.184
#> GSM25579 5 0.683 0.12534 0.204 0.000 0.268 0.020 0.508
#> GSM25580 1 0.262 0.69606 0.900 0.008 0.012 0.012 0.068
#> GSM25581 1 0.281 0.69466 0.896 0.016 0.036 0.004 0.048
#> GSM48655 2 0.576 0.46081 0.000 0.620 0.000 0.200 0.180
#> GSM48656 2 0.466 0.55275 0.000 0.764 0.060 0.024 0.152
#> GSM48657 2 0.562 0.48051 0.004 0.648 0.000 0.208 0.140
#> GSM48658 3 0.661 0.15816 0.004 0.324 0.540 0.036 0.096
#> GSM25624 1 0.283 0.69522 0.892 0.004 0.032 0.012 0.060
#> GSM25625 1 0.702 0.01970 0.396 0.000 0.392 0.192 0.020
#> GSM25626 3 0.814 0.38027 0.132 0.076 0.436 0.320 0.036
#> GSM25627 4 0.460 0.47019 0.072 0.048 0.080 0.796 0.004
#> GSM25628 3 0.625 0.55856 0.008 0.124 0.640 0.200 0.028
#> GSM25629 4 0.582 0.35580 0.024 0.080 0.224 0.664 0.008
#> GSM25630 3 0.658 0.60248 0.080 0.160 0.664 0.044 0.052
#> GSM25631 3 0.483 0.42447 0.004 0.052 0.692 0.000 0.252
#> GSM25632 1 0.732 0.34662 0.512 0.004 0.272 0.148 0.064
#> GSM25633 1 0.303 0.68838 0.880 0.000 0.064 0.024 0.032
#> GSM25634 1 0.274 0.69522 0.900 0.004 0.028 0.048 0.020
#> GSM25635 1 0.439 0.67202 0.796 0.008 0.056 0.016 0.124
#> GSM25656 3 0.729 0.40929 0.008 0.152 0.492 0.304 0.044
#> GSM25657 1 0.310 0.68979 0.884 0.008 0.056 0.028 0.024
#> GSM25658 4 0.651 -0.14933 0.420 0.016 0.084 0.468 0.012
#> GSM25659 3 0.637 0.38821 0.192 0.008 0.612 0.016 0.172
#> GSM25660 1 0.605 0.35848 0.508 0.004 0.080 0.008 0.400
#> GSM25661 1 0.300 0.69165 0.872 0.000 0.032 0.008 0.088
#> GSM25662 4 0.583 0.24484 0.000 0.340 0.020 0.576 0.064
#> GSM25663 5 0.780 0.12689 0.000 0.308 0.092 0.180 0.420
#> GSM25680 5 0.688 0.48080 0.000 0.104 0.180 0.120 0.596
#> GSM25681 5 0.606 0.52217 0.020 0.048 0.180 0.068 0.684
#> GSM25682 4 0.667 -0.00482 0.000 0.336 0.000 0.424 0.240
#> GSM25683 4 0.564 0.31128 0.000 0.276 0.000 0.608 0.116
#> GSM25684 4 0.657 0.23566 0.000 0.304 0.028 0.540 0.128
#> GSM25685 4 0.406 0.48915 0.000 0.176 0.020 0.784 0.020
#> GSM25686 4 0.664 0.01229 0.000 0.340 0.000 0.428 0.232
#> GSM25687 2 0.675 0.04927 0.000 0.380 0.000 0.360 0.260
#> GSM48664 1 0.437 0.65755 0.800 0.104 0.000 0.040 0.056
#> GSM48665 1 0.367 0.67774 0.844 0.080 0.000 0.028 0.048
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM25548 5 0.322 0.6817 0.000 0.060 0.004 0.052 0.856 0.028
#> GSM25549 5 0.226 0.6817 0.000 0.048 0.004 0.008 0.908 0.032
#> GSM25550 5 0.233 0.6732 0.024 0.056 0.000 0.004 0.904 0.012
#> GSM25551 4 0.346 0.4866 0.004 0.032 0.008 0.828 0.120 0.008
#> GSM25570 5 0.250 0.6819 0.004 0.044 0.004 0.004 0.896 0.048
#> GSM25571 5 0.336 0.6712 0.004 0.028 0.000 0.040 0.844 0.084
#> GSM25358 4 0.844 0.2174 0.148 0.024 0.216 0.404 0.156 0.052
#> GSM25359 4 0.751 0.1059 0.012 0.032 0.324 0.344 0.260 0.028
#> GSM25360 3 0.560 0.2696 0.008 0.008 0.516 0.024 0.040 0.404
#> GSM25361 6 0.640 0.0660 0.012 0.012 0.324 0.012 0.140 0.500
#> GSM25377 1 0.731 0.4617 0.540 0.188 0.144 0.032 0.020 0.076
#> GSM25378 1 0.661 0.5245 0.620 0.016 0.072 0.168 0.088 0.036
#> GSM25401 4 0.391 0.4943 0.148 0.024 0.016 0.792 0.000 0.020
#> GSM25402 4 0.738 -0.0904 0.376 0.096 0.084 0.392 0.000 0.052
#> GSM25349 2 0.665 0.4490 0.028 0.584 0.144 0.028 0.196 0.020
#> GSM25350 2 0.623 0.2657 0.012 0.512 0.120 0.004 0.332 0.020
#> GSM25356 1 0.801 0.4306 0.516 0.052 0.104 0.116 0.152 0.060
#> GSM25357 4 0.721 0.1457 0.016 0.128 0.072 0.488 0.280 0.016
#> GSM25385 3 0.749 0.3129 0.248 0.020 0.476 0.156 0.016 0.084
#> GSM25386 3 0.473 0.5703 0.036 0.032 0.780 0.080 0.012 0.060
#> GSM25399 1 0.622 0.5443 0.656 0.136 0.092 0.036 0.008 0.072
#> GSM25400 1 0.460 0.6067 0.788 0.032 0.040 0.068 0.008 0.064
#> GSM48659 2 0.839 0.2646 0.000 0.304 0.092 0.208 0.104 0.292
#> GSM48660 2 0.327 0.5844 0.004 0.848 0.056 0.008 0.080 0.004
#> GSM25409 5 0.506 0.5297 0.016 0.244 0.028 0.032 0.676 0.004
#> GSM25410 3 0.538 0.5456 0.060 0.040 0.724 0.116 0.004 0.056
#> GSM25426 4 0.216 0.5126 0.000 0.056 0.008 0.908 0.028 0.000
#> GSM25427 1 0.787 0.4205 0.492 0.200 0.104 0.016 0.120 0.068
#> GSM25540 3 0.547 0.2651 0.004 0.016 0.500 0.040 0.012 0.428
#> GSM25541 6 0.510 -0.1758 0.016 0.008 0.432 0.012 0.012 0.520
#> GSM25542 3 0.483 0.4155 0.000 0.248 0.676 0.052 0.004 0.020
#> GSM25543 3 0.430 0.4600 0.000 0.220 0.728 0.012 0.028 0.012
#> GSM25479 1 0.542 0.5066 0.668 0.004 0.020 0.008 0.120 0.180
#> GSM25480 1 0.716 0.1741 0.428 0.016 0.044 0.004 0.276 0.232
#> GSM25481 1 0.840 0.1587 0.348 0.348 0.080 0.060 0.112 0.052
#> GSM25482 1 0.799 0.3458 0.436 0.256 0.048 0.032 0.172 0.056
#> GSM48654 2 0.674 0.5309 0.000 0.588 0.156 0.056 0.060 0.140
#> GSM48650 2 0.563 0.3545 0.000 0.548 0.036 0.356 0.052 0.008
#> GSM48651 2 0.511 0.5833 0.000 0.732 0.052 0.120 0.072 0.024
#> GSM48652 2 0.559 0.5830 0.000 0.704 0.072 0.112 0.068 0.044
#> GSM48653 2 0.716 0.4872 0.000 0.532 0.108 0.180 0.036 0.144
#> GSM48662 2 0.381 0.5810 0.000 0.804 0.076 0.004 0.104 0.012
#> GSM48663 2 0.459 0.5356 0.008 0.764 0.068 0.016 0.128 0.016
#> GSM25524 6 0.609 0.1429 0.136 0.000 0.264 0.020 0.016 0.564
#> GSM25525 6 0.632 -0.0927 0.400 0.004 0.024 0.008 0.120 0.444
#> GSM25526 4 0.588 0.2525 0.284 0.004 0.036 0.572 0.000 0.104
#> GSM25527 1 0.460 0.5450 0.720 0.000 0.016 0.040 0.016 0.208
#> GSM25528 6 0.640 0.1285 0.368 0.000 0.220 0.000 0.020 0.392
#> GSM25529 1 0.588 0.0436 0.444 0.000 0.032 0.000 0.092 0.432
#> GSM25530 1 0.656 0.4210 0.584 0.020 0.168 0.056 0.004 0.168
#> GSM25531 1 0.469 0.5974 0.772 0.024 0.040 0.064 0.004 0.096
#> GSM48661 2 0.754 0.0739 0.000 0.324 0.284 0.080 0.016 0.296
#> GSM25561 3 0.540 0.5107 0.068 0.068 0.696 0.004 0.008 0.156
#> GSM25562 2 0.837 -0.1834 0.304 0.356 0.164 0.032 0.040 0.104
#> GSM25563 3 0.485 0.5438 0.036 0.056 0.744 0.016 0.004 0.144
#> GSM25564 2 0.767 0.1262 0.120 0.436 0.032 0.048 0.044 0.320
#> GSM25565 2 0.734 0.3847 0.000 0.456 0.216 0.120 0.196 0.012
#> GSM25566 5 0.681 0.3038 0.000 0.240 0.032 0.272 0.444 0.012
#> GSM25568 2 0.592 0.3024 0.020 0.560 0.328 0.008 0.020 0.064
#> GSM25569 2 0.596 0.5457 0.000 0.632 0.164 0.020 0.148 0.036
#> GSM25552 5 0.295 0.6545 0.008 0.076 0.004 0.000 0.864 0.048
#> GSM25553 5 0.342 0.6348 0.012 0.064 0.024 0.000 0.848 0.052
#> GSM25578 1 0.501 0.5328 0.704 0.004 0.024 0.004 0.084 0.180
#> GSM25579 6 0.672 0.1906 0.224 0.004 0.024 0.004 0.356 0.388
#> GSM25580 1 0.306 0.6268 0.868 0.036 0.004 0.004 0.020 0.068
#> GSM25581 1 0.316 0.6184 0.860 0.012 0.012 0.008 0.020 0.088
#> GSM48655 2 0.637 0.2229 0.000 0.512 0.036 0.144 0.300 0.008
#> GSM48656 2 0.503 0.5845 0.004 0.744 0.072 0.020 0.112 0.048
#> GSM48657 2 0.550 0.4984 0.008 0.656 0.004 0.152 0.164 0.016
#> GSM48658 6 0.723 -0.3006 0.000 0.356 0.144 0.048 0.044 0.408
#> GSM25624 1 0.403 0.6221 0.824 0.020 0.032 0.012 0.048 0.064
#> GSM25625 1 0.773 -0.1627 0.308 0.004 0.248 0.272 0.000 0.168
#> GSM25626 3 0.623 0.4839 0.064 0.020 0.596 0.232 0.000 0.088
#> GSM25627 4 0.465 0.4886 0.052 0.064 0.024 0.772 0.000 0.088
#> GSM25628 3 0.602 0.4911 0.004 0.072 0.608 0.104 0.000 0.212
#> GSM25629 4 0.486 0.4363 0.028 0.028 0.040 0.724 0.000 0.180
#> GSM25630 3 0.552 0.5166 0.048 0.072 0.676 0.012 0.004 0.188
#> GSM25631 6 0.604 0.2768 0.012 0.056 0.132 0.000 0.176 0.624
#> GSM25632 3 0.647 0.2403 0.332 0.004 0.488 0.064 0.000 0.112
#> GSM25633 1 0.327 0.6066 0.844 0.004 0.032 0.012 0.004 0.104
#> GSM25634 1 0.423 0.6165 0.808 0.028 0.084 0.020 0.012 0.048
#> GSM25635 1 0.436 0.6126 0.796 0.012 0.016 0.024 0.084 0.068
#> GSM25656 3 0.662 0.4793 0.004 0.108 0.588 0.144 0.012 0.144
#> GSM25657 1 0.464 0.5981 0.780 0.040 0.076 0.020 0.008 0.076
#> GSM25658 4 0.607 0.0207 0.384 0.020 0.016 0.484 0.000 0.096
#> GSM25659 6 0.584 0.3736 0.200 0.044 0.036 0.008 0.052 0.660
#> GSM25660 1 0.585 0.3679 0.564 0.008 0.008 0.000 0.248 0.172
#> GSM25661 1 0.326 0.6137 0.844 0.008 0.012 0.000 0.036 0.100
#> GSM25662 4 0.616 0.2611 0.000 0.220 0.036 0.600 0.120 0.024
#> GSM25663 5 0.707 0.5092 0.000 0.204 0.076 0.108 0.548 0.064
#> GSM25680 5 0.568 0.5139 0.000 0.016 0.044 0.088 0.652 0.200
#> GSM25681 5 0.573 0.4682 0.012 0.008 0.088 0.048 0.668 0.176
#> GSM25682 5 0.642 0.4140 0.000 0.248 0.024 0.240 0.484 0.004
#> GSM25683 4 0.677 -0.1277 0.000 0.228 0.032 0.424 0.308 0.008
#> GSM25684 4 0.628 0.1821 0.000 0.196 0.004 0.536 0.232 0.032
#> GSM25685 4 0.334 0.4819 0.000 0.088 0.016 0.848 0.020 0.028
#> GSM25686 5 0.640 0.4024 0.000 0.220 0.020 0.288 0.468 0.004
#> GSM25687 5 0.641 0.4028 0.000 0.272 0.016 0.228 0.476 0.008
#> GSM48664 1 0.595 0.5585 0.676 0.156 0.068 0.024 0.020 0.056
#> GSM48665 1 0.410 0.6229 0.820 0.068 0.016 0.016 0.036 0.044
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 genotype/variation(p) k
#> MAD:NMF 94 0.000057 2
#> MAD:NMF 74 0.000619 3
#> MAD:NMF 40 0.024680 4
#> MAD:NMF 51 0.002546 5
#> MAD:NMF 41 0.026226 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 100 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 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 0.372 0.727 0.860 0.3507 0.642 0.642
#> 3 3 0.406 0.657 0.813 0.6565 0.638 0.477
#> 4 4 0.412 0.452 0.703 0.1753 0.864 0.681
#> 5 5 0.451 0.405 0.667 0.0744 0.839 0.574
#> 6 6 0.525 0.556 0.719 0.0565 0.920 0.710
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
#> GSM25548 2 0.1184 0.835 0.016 0.984
#> GSM25549 2 0.1184 0.835 0.016 0.984
#> GSM25550 2 0.1184 0.835 0.016 0.984
#> GSM25551 2 0.1184 0.835 0.016 0.984
#> GSM25570 2 0.1184 0.835 0.016 0.984
#> GSM25571 2 0.1184 0.835 0.016 0.984
#> GSM25358 1 0.9522 0.533 0.628 0.372
#> GSM25359 1 0.9522 0.533 0.628 0.372
#> GSM25360 1 0.0938 0.739 0.988 0.012
#> GSM25361 1 0.9248 0.605 0.660 0.340
#> GSM25377 2 0.9393 0.525 0.356 0.644
#> GSM25378 2 0.6887 0.752 0.184 0.816
#> GSM25401 2 0.8016 0.698 0.244 0.756
#> GSM25402 2 0.8016 0.698 0.244 0.756
#> GSM25349 2 0.1414 0.830 0.020 0.980
#> GSM25350 2 0.1414 0.830 0.020 0.980
#> GSM25356 2 0.0672 0.835 0.008 0.992
#> GSM25357 2 0.0376 0.833 0.004 0.996
#> GSM25385 1 0.8144 0.702 0.748 0.252
#> GSM25386 1 0.0938 0.739 0.988 0.012
#> GSM25399 2 0.9393 0.525 0.356 0.644
#> GSM25400 2 0.6887 0.752 0.184 0.816
#> GSM48659 2 0.0000 0.832 0.000 1.000
#> GSM48660 2 0.0938 0.827 0.012 0.988
#> GSM25409 2 0.1414 0.830 0.020 0.980
#> GSM25410 1 0.0938 0.739 0.988 0.012
#> GSM25426 2 0.0376 0.833 0.004 0.996
#> GSM25427 2 0.6887 0.752 0.184 0.816
#> GSM25540 1 0.9248 0.605 0.660 0.340
#> GSM25541 1 0.9248 0.605 0.660 0.340
#> GSM25542 1 0.9970 0.237 0.532 0.468
#> GSM25543 1 0.9970 0.237 0.532 0.468
#> GSM25479 2 0.7602 0.718 0.220 0.780
#> GSM25480 2 0.7602 0.718 0.220 0.780
#> GSM25481 2 0.1633 0.831 0.024 0.976
#> GSM25482 2 0.1633 0.831 0.024 0.976
#> GSM48654 2 0.1633 0.832 0.024 0.976
#> GSM48650 2 0.0938 0.827 0.012 0.988
#> GSM48651 2 0.0938 0.827 0.012 0.988
#> GSM48652 2 0.0938 0.827 0.012 0.988
#> GSM48653 2 0.0938 0.827 0.012 0.988
#> GSM48662 2 0.0938 0.827 0.012 0.988
#> GSM48663 2 0.0938 0.827 0.012 0.988
#> GSM25524 2 0.9710 0.357 0.400 0.600
#> GSM25525 2 0.7815 0.706 0.232 0.768
#> GSM25526 1 0.9833 0.406 0.576 0.424
#> GSM25527 2 0.9129 0.563 0.328 0.672
#> GSM25528 1 0.8443 0.689 0.728 0.272
#> GSM25529 2 0.8763 0.627 0.296 0.704
#> GSM25530 1 0.8386 0.692 0.732 0.268
#> GSM25531 1 0.8386 0.692 0.732 0.268
#> GSM48661 2 0.1633 0.832 0.024 0.976
#> GSM25561 1 0.3274 0.742 0.940 0.060
#> GSM25562 2 0.8608 0.644 0.284 0.716
#> GSM25563 1 0.2423 0.742 0.960 0.040
#> GSM25564 2 0.3733 0.816 0.072 0.928
#> GSM25565 2 0.1414 0.830 0.020 0.980
#> GSM25566 2 0.1414 0.830 0.020 0.980
#> GSM25568 2 0.0376 0.833 0.004 0.996
#> GSM25569 2 0.0000 0.832 0.000 1.000
#> GSM25552 2 0.1414 0.830 0.020 0.980
#> GSM25553 2 0.1414 0.830 0.020 0.980
#> GSM25578 2 0.8763 0.625 0.296 0.704
#> GSM25579 2 0.8763 0.625 0.296 0.704
#> GSM25580 2 0.8763 0.625 0.296 0.704
#> GSM25581 2 0.8763 0.625 0.296 0.704
#> GSM48655 2 0.0000 0.832 0.000 1.000
#> GSM48656 2 0.1633 0.832 0.024 0.976
#> GSM48657 2 0.0000 0.832 0.000 1.000
#> GSM48658 2 0.1633 0.832 0.024 0.976
#> GSM25624 2 0.3584 0.820 0.068 0.932
#> GSM25625 1 0.0938 0.739 0.988 0.012
#> GSM25626 1 0.0938 0.739 0.988 0.012
#> GSM25627 2 0.8081 0.688 0.248 0.752
#> GSM25628 1 0.0938 0.739 0.988 0.012
#> GSM25629 2 0.8608 0.642 0.284 0.716
#> GSM25630 1 0.0938 0.739 0.988 0.012
#> GSM25631 2 0.8555 0.648 0.280 0.720
#> GSM25632 1 0.0938 0.739 0.988 0.012
#> GSM25633 2 0.8608 0.642 0.284 0.716
#> GSM25634 2 0.8608 0.642 0.284 0.716
#> GSM25635 2 0.8608 0.642 0.284 0.716
#> GSM25656 1 0.7376 0.719 0.792 0.208
#> GSM25657 2 0.8713 0.631 0.292 0.708
#> GSM25658 2 0.6973 0.749 0.188 0.812
#> GSM25659 2 0.7528 0.725 0.216 0.784
#> GSM25660 2 0.8661 0.638 0.288 0.712
#> GSM25661 2 0.8661 0.638 0.288 0.712
#> GSM25662 2 0.2603 0.829 0.044 0.956
#> GSM25663 2 0.2603 0.829 0.044 0.956
#> GSM25680 2 0.1184 0.835 0.016 0.984
#> GSM25681 2 0.1184 0.835 0.016 0.984
#> GSM25682 2 0.0376 0.833 0.004 0.996
#> GSM25683 2 0.0376 0.833 0.004 0.996
#> GSM25684 2 0.0000 0.832 0.000 1.000
#> GSM25685 2 0.0000 0.832 0.000 1.000
#> GSM25686 2 0.0376 0.833 0.004 0.996
#> GSM25687 2 0.0376 0.833 0.004 0.996
#> GSM48664 2 0.9393 0.525 0.356 0.644
#> GSM48665 2 0.9393 0.525 0.356 0.644
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM25548 2 0.2448 0.8962 0.076 0.924 0.000
#> GSM25549 2 0.2448 0.8962 0.076 0.924 0.000
#> GSM25550 2 0.2448 0.8962 0.076 0.924 0.000
#> GSM25551 2 0.2448 0.8962 0.076 0.924 0.000
#> GSM25570 2 0.2448 0.8962 0.076 0.924 0.000
#> GSM25571 2 0.2448 0.8962 0.076 0.924 0.000
#> GSM25358 1 0.6516 -0.3545 0.516 0.004 0.480
#> GSM25359 1 0.6516 -0.3545 0.516 0.004 0.480
#> GSM25360 3 0.0892 0.8197 0.020 0.000 0.980
#> GSM25361 3 0.7493 0.2969 0.480 0.036 0.484
#> GSM25377 1 0.1491 0.5437 0.968 0.016 0.016
#> GSM25378 1 0.6299 0.2381 0.524 0.476 0.000
#> GSM25401 1 0.6521 0.4637 0.644 0.340 0.016
#> GSM25402 1 0.6521 0.4637 0.644 0.340 0.016
#> GSM25349 2 0.4345 0.8579 0.136 0.848 0.016
#> GSM25350 2 0.4345 0.8579 0.136 0.848 0.016
#> GSM25356 2 0.2261 0.8984 0.068 0.932 0.000
#> GSM25357 2 0.2165 0.8983 0.064 0.936 0.000
#> GSM25385 3 0.5859 0.5946 0.344 0.000 0.656
#> GSM25386 3 0.0892 0.8197 0.020 0.000 0.980
#> GSM25399 1 0.1491 0.5437 0.968 0.016 0.016
#> GSM25400 1 0.6299 0.2381 0.524 0.476 0.000
#> GSM48659 2 0.0661 0.8747 0.008 0.988 0.004
#> GSM48660 2 0.1620 0.8874 0.024 0.964 0.012
#> GSM25409 2 0.4345 0.8579 0.136 0.848 0.016
#> GSM25410 3 0.0892 0.8197 0.020 0.000 0.980
#> GSM25426 2 0.2165 0.8983 0.064 0.936 0.000
#> GSM25427 1 0.6299 0.2381 0.524 0.476 0.000
#> GSM25540 3 0.7493 0.2969 0.480 0.036 0.484
#> GSM25541 3 0.7493 0.2969 0.480 0.036 0.484
#> GSM25542 1 0.9030 0.0331 0.492 0.140 0.368
#> GSM25543 1 0.9030 0.0331 0.492 0.140 0.368
#> GSM25479 1 0.6608 0.4273 0.560 0.432 0.008
#> GSM25480 1 0.6608 0.4273 0.560 0.432 0.008
#> GSM25481 2 0.4539 0.8494 0.148 0.836 0.016
#> GSM25482 2 0.4539 0.8494 0.148 0.836 0.016
#> GSM48654 2 0.3983 0.7583 0.144 0.852 0.004
#> GSM48650 2 0.1620 0.8861 0.024 0.964 0.012
#> GSM48651 2 0.1620 0.8874 0.024 0.964 0.012
#> GSM48652 2 0.1620 0.8874 0.024 0.964 0.012
#> GSM48653 2 0.1620 0.8874 0.024 0.964 0.012
#> GSM48662 2 0.1620 0.8874 0.024 0.964 0.012
#> GSM48663 2 0.2269 0.8870 0.040 0.944 0.016
#> GSM25524 1 0.4586 0.5008 0.856 0.048 0.096
#> GSM25525 1 0.6745 0.4315 0.560 0.428 0.012
#> GSM25526 1 0.7267 0.2277 0.668 0.064 0.268
#> GSM25527 1 0.5402 0.6848 0.792 0.180 0.028
#> GSM25528 1 0.6274 -0.3214 0.544 0.000 0.456
#> GSM25529 1 0.5414 0.6945 0.772 0.212 0.016
#> GSM25530 1 0.6291 -0.3475 0.532 0.000 0.468
#> GSM25531 1 0.6291 -0.3475 0.532 0.000 0.468
#> GSM48661 2 0.3983 0.7583 0.144 0.852 0.004
#> GSM25561 3 0.2537 0.7996 0.080 0.000 0.920
#> GSM25562 1 0.5061 0.7008 0.784 0.208 0.008
#> GSM25563 3 0.1860 0.8113 0.052 0.000 0.948
#> GSM25564 2 0.3686 0.8293 0.140 0.860 0.000
#> GSM25565 2 0.4345 0.8579 0.136 0.848 0.016
#> GSM25566 2 0.4345 0.8579 0.136 0.848 0.016
#> GSM25568 2 0.0983 0.8723 0.016 0.980 0.004
#> GSM25569 2 0.0661 0.8747 0.008 0.988 0.004
#> GSM25552 2 0.4345 0.8579 0.136 0.848 0.016
#> GSM25553 2 0.4345 0.8579 0.136 0.848 0.016
#> GSM25578 1 0.5072 0.6993 0.792 0.196 0.012
#> GSM25579 1 0.5072 0.6993 0.792 0.196 0.012
#> GSM25580 1 0.5072 0.6993 0.792 0.196 0.012
#> GSM25581 1 0.5072 0.6993 0.792 0.196 0.012
#> GSM48655 2 0.0661 0.8747 0.008 0.988 0.004
#> GSM48656 2 0.3983 0.7583 0.144 0.852 0.004
#> GSM48657 2 0.0661 0.8747 0.008 0.988 0.004
#> GSM48658 2 0.3983 0.7583 0.144 0.852 0.004
#> GSM25624 2 0.4978 0.6639 0.216 0.780 0.004
#> GSM25625 3 0.0892 0.8197 0.020 0.000 0.980
#> GSM25626 3 0.0892 0.8197 0.020 0.000 0.980
#> GSM25627 1 0.6082 0.6479 0.692 0.296 0.012
#> GSM25628 3 0.0892 0.8197 0.020 0.000 0.980
#> GSM25629 1 0.5681 0.6957 0.748 0.236 0.016
#> GSM25630 3 0.0892 0.8197 0.020 0.000 0.980
#> GSM25631 1 0.5723 0.6952 0.744 0.240 0.016
#> GSM25632 3 0.0892 0.8197 0.020 0.000 0.980
#> GSM25633 1 0.5681 0.6957 0.748 0.236 0.016
#> GSM25634 1 0.5681 0.6957 0.748 0.236 0.016
#> GSM25635 1 0.5681 0.6957 0.748 0.236 0.016
#> GSM25656 3 0.6381 0.5758 0.340 0.012 0.648
#> GSM25657 1 0.5072 0.6985 0.792 0.196 0.012
#> GSM25658 1 0.6235 0.3709 0.564 0.436 0.000
#> GSM25659 1 0.6192 0.4324 0.580 0.420 0.000
#> GSM25660 1 0.5171 0.7006 0.784 0.204 0.012
#> GSM25661 1 0.4912 0.6994 0.796 0.196 0.008
#> GSM25662 2 0.5325 0.6842 0.248 0.748 0.004
#> GSM25663 2 0.5325 0.6842 0.248 0.748 0.004
#> GSM25680 2 0.2448 0.8962 0.076 0.924 0.000
#> GSM25681 2 0.2448 0.8962 0.076 0.924 0.000
#> GSM25682 2 0.2165 0.8983 0.064 0.936 0.000
#> GSM25683 2 0.2165 0.8983 0.064 0.936 0.000
#> GSM25684 2 0.0661 0.8747 0.008 0.988 0.004
#> GSM25685 2 0.0661 0.8747 0.008 0.988 0.004
#> GSM25686 2 0.2165 0.8983 0.064 0.936 0.000
#> GSM25687 2 0.2165 0.8983 0.064 0.936 0.000
#> GSM48664 1 0.1491 0.5437 0.968 0.016 0.016
#> GSM48665 1 0.1491 0.5437 0.968 0.016 0.016
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM25548 2 0.1118 0.562 0.036 0.964 0.000 0.000
#> GSM25549 2 0.1118 0.562 0.036 0.964 0.000 0.000
#> GSM25550 2 0.1118 0.562 0.036 0.964 0.000 0.000
#> GSM25551 2 0.1118 0.562 0.036 0.964 0.000 0.000
#> GSM25570 2 0.1118 0.562 0.036 0.964 0.000 0.000
#> GSM25571 2 0.1118 0.562 0.036 0.964 0.000 0.000
#> GSM25358 3 0.6909 0.336 0.456 0.012 0.460 0.072
#> GSM25359 3 0.6909 0.336 0.456 0.012 0.460 0.072
#> GSM25360 3 0.0000 0.756 0.000 0.000 1.000 0.000
#> GSM25361 3 0.7063 0.288 0.456 0.032 0.460 0.052
#> GSM25377 1 0.5110 0.480 0.656 0.016 0.000 0.328
#> GSM25378 2 0.6840 -0.149 0.432 0.468 0.000 0.100
#> GSM25401 2 0.7906 -0.206 0.348 0.352 0.000 0.300
#> GSM25402 2 0.7906 -0.206 0.348 0.352 0.000 0.300
#> GSM25349 2 0.6084 0.525 0.092 0.656 0.000 0.252
#> GSM25350 2 0.6084 0.525 0.092 0.656 0.000 0.252
#> GSM25356 2 0.0779 0.565 0.016 0.980 0.000 0.004
#> GSM25357 2 0.0592 0.565 0.016 0.984 0.000 0.000
#> GSM25385 3 0.5896 0.563 0.296 0.004 0.648 0.052
#> GSM25386 3 0.0000 0.756 0.000 0.000 1.000 0.000
#> GSM25399 1 0.5110 0.480 0.656 0.016 0.000 0.328
#> GSM25400 2 0.6840 -0.149 0.432 0.468 0.000 0.100
#> GSM48659 2 0.5028 -0.306 0.004 0.596 0.000 0.400
#> GSM48660 2 0.5207 0.487 0.028 0.680 0.000 0.292
#> GSM25409 2 0.6084 0.525 0.092 0.656 0.000 0.252
#> GSM25410 3 0.0000 0.756 0.000 0.000 1.000 0.000
#> GSM25426 2 0.0927 0.560 0.016 0.976 0.000 0.008
#> GSM25427 2 0.6840 -0.149 0.432 0.468 0.000 0.100
#> GSM25540 3 0.7063 0.288 0.456 0.032 0.460 0.052
#> GSM25541 3 0.7063 0.288 0.456 0.032 0.460 0.052
#> GSM25542 1 0.8235 -0.068 0.480 0.084 0.348 0.088
#> GSM25543 1 0.8235 -0.068 0.480 0.084 0.348 0.088
#> GSM25479 1 0.5440 0.421 0.596 0.384 0.000 0.020
#> GSM25480 1 0.5440 0.421 0.596 0.384 0.000 0.020
#> GSM25481 2 0.4562 0.559 0.056 0.792 0.000 0.152
#> GSM25482 2 0.4562 0.559 0.056 0.792 0.000 0.152
#> GSM48654 4 0.7246 0.960 0.144 0.408 0.000 0.448
#> GSM48650 2 0.5384 0.472 0.028 0.648 0.000 0.324
#> GSM48651 2 0.5207 0.487 0.028 0.680 0.000 0.292
#> GSM48652 2 0.5207 0.487 0.028 0.680 0.000 0.292
#> GSM48653 2 0.5207 0.487 0.028 0.680 0.000 0.292
#> GSM48662 2 0.5207 0.487 0.028 0.680 0.000 0.292
#> GSM48663 2 0.5404 0.477 0.028 0.644 0.000 0.328
#> GSM25524 1 0.5233 0.532 0.784 0.032 0.056 0.128
#> GSM25525 1 0.5699 0.422 0.588 0.380 0.000 0.032
#> GSM25526 1 0.6892 0.245 0.648 0.036 0.224 0.092
#> GSM25527 1 0.4107 0.696 0.832 0.128 0.012 0.028
#> GSM25528 1 0.6818 -0.264 0.504 0.008 0.412 0.076
#> GSM25529 1 0.4199 0.685 0.804 0.164 0.000 0.032
#> GSM25530 1 0.6833 -0.293 0.492 0.008 0.424 0.076
#> GSM25531 1 0.6833 -0.293 0.492 0.008 0.424 0.076
#> GSM48661 4 0.7246 0.960 0.144 0.408 0.000 0.448
#> GSM25561 3 0.2282 0.742 0.052 0.000 0.924 0.024
#> GSM25562 1 0.4872 0.691 0.776 0.148 0.000 0.076
#> GSM25563 3 0.1610 0.750 0.032 0.000 0.952 0.016
#> GSM25564 2 0.6429 0.478 0.160 0.648 0.000 0.192
#> GSM25565 2 0.6084 0.525 0.092 0.656 0.000 0.252
#> GSM25566 2 0.6084 0.525 0.092 0.656 0.000 0.252
#> GSM25568 2 0.5475 0.228 0.036 0.656 0.000 0.308
#> GSM25569 2 0.5383 0.270 0.036 0.672 0.000 0.292
#> GSM25552 2 0.6084 0.525 0.092 0.656 0.000 0.252
#> GSM25553 2 0.6084 0.525 0.092 0.656 0.000 0.252
#> GSM25578 1 0.3300 0.703 0.848 0.144 0.000 0.008
#> GSM25579 1 0.3300 0.703 0.848 0.144 0.000 0.008
#> GSM25580 1 0.3249 0.702 0.852 0.140 0.000 0.008
#> GSM25581 1 0.3249 0.702 0.852 0.140 0.000 0.008
#> GSM48655 2 0.5659 -0.409 0.032 0.600 0.000 0.368
#> GSM48656 4 0.7246 0.960 0.144 0.408 0.000 0.448
#> GSM48657 2 0.5659 -0.409 0.032 0.600 0.000 0.368
#> GSM48658 4 0.7246 0.960 0.144 0.408 0.000 0.448
#> GSM25624 4 0.7683 0.836 0.216 0.384 0.000 0.400
#> GSM25625 3 0.0804 0.755 0.012 0.000 0.980 0.008
#> GSM25626 3 0.0000 0.756 0.000 0.000 1.000 0.000
#> GSM25627 1 0.5589 0.640 0.724 0.192 0.004 0.080
#> GSM25628 3 0.0000 0.756 0.000 0.000 1.000 0.000
#> GSM25629 1 0.4239 0.686 0.808 0.160 0.004 0.028
#> GSM25630 3 0.0000 0.756 0.000 0.000 1.000 0.000
#> GSM25631 1 0.4285 0.684 0.804 0.164 0.004 0.028
#> GSM25632 3 0.0000 0.756 0.000 0.000 1.000 0.000
#> GSM25633 1 0.4239 0.686 0.808 0.160 0.004 0.028
#> GSM25634 1 0.4239 0.686 0.808 0.160 0.004 0.028
#> GSM25635 1 0.4239 0.686 0.808 0.160 0.004 0.028
#> GSM25656 3 0.6021 0.544 0.320 0.004 0.624 0.052
#> GSM25657 1 0.3501 0.703 0.848 0.132 0.000 0.020
#> GSM25658 1 0.6783 0.316 0.512 0.388 0.000 0.100
#> GSM25659 1 0.5971 0.410 0.584 0.368 0.000 0.048
#> GSM25660 1 0.3763 0.700 0.832 0.144 0.000 0.024
#> GSM25661 1 0.5722 0.670 0.716 0.136 0.000 0.148
#> GSM25662 2 0.4467 0.299 0.172 0.788 0.000 0.040
#> GSM25663 2 0.4467 0.299 0.172 0.788 0.000 0.040
#> GSM25680 2 0.1118 0.562 0.036 0.964 0.000 0.000
#> GSM25681 2 0.1118 0.562 0.036 0.964 0.000 0.000
#> GSM25682 2 0.0592 0.565 0.016 0.984 0.000 0.000
#> GSM25683 2 0.0592 0.565 0.016 0.984 0.000 0.000
#> GSM25684 2 0.5040 -0.310 0.008 0.628 0.000 0.364
#> GSM25685 2 0.5040 -0.310 0.008 0.628 0.000 0.364
#> GSM25686 2 0.0592 0.565 0.016 0.984 0.000 0.000
#> GSM25687 2 0.0592 0.565 0.016 0.984 0.000 0.000
#> GSM48664 1 0.5110 0.480 0.656 0.016 0.000 0.328
#> GSM48665 1 0.5110 0.480 0.656 0.016 0.000 0.328
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM25548 5 0.0609 0.5532 0.020 0.000 0.000 0.000 0.980
#> GSM25549 5 0.0609 0.5532 0.020 0.000 0.000 0.000 0.980
#> GSM25550 5 0.0609 0.5532 0.020 0.000 0.000 0.000 0.980
#> GSM25551 5 0.0609 0.5532 0.020 0.000 0.000 0.000 0.980
#> GSM25570 5 0.0609 0.5532 0.020 0.000 0.000 0.000 0.980
#> GSM25571 5 0.0609 0.5532 0.020 0.000 0.000 0.000 0.980
#> GSM25358 1 0.6957 -0.0213 0.472 0.044 0.396 0.068 0.020
#> GSM25359 1 0.6957 -0.0213 0.472 0.044 0.396 0.068 0.020
#> GSM25360 3 0.0290 0.8788 0.008 0.000 0.992 0.000 0.000
#> GSM25361 1 0.6952 0.0875 0.476 0.036 0.396 0.060 0.032
#> GSM25377 4 0.3563 0.7650 0.208 0.000 0.000 0.780 0.012
#> GSM25378 5 0.6420 -0.0871 0.400 0.024 0.000 0.096 0.480
#> GSM25401 4 0.6520 0.4563 0.052 0.072 0.000 0.524 0.352
#> GSM25402 4 0.6520 0.4563 0.052 0.072 0.000 0.524 0.352
#> GSM25349 5 0.6596 0.0720 0.072 0.392 0.000 0.052 0.484
#> GSM25350 5 0.6596 0.0720 0.072 0.392 0.000 0.052 0.484
#> GSM25356 5 0.0162 0.5479 0.000 0.000 0.000 0.004 0.996
#> GSM25357 5 0.0000 0.5478 0.000 0.000 0.000 0.000 1.000
#> GSM25385 3 0.5539 0.4401 0.316 0.012 0.616 0.052 0.004
#> GSM25386 3 0.0451 0.8790 0.008 0.004 0.988 0.000 0.000
#> GSM25399 4 0.3563 0.7650 0.208 0.000 0.000 0.780 0.012
#> GSM25400 5 0.6420 -0.0871 0.400 0.024 0.000 0.096 0.480
#> GSM48659 2 0.6437 0.2115 0.024 0.460 0.000 0.096 0.420
#> GSM48660 2 0.4210 0.3806 0.000 0.588 0.000 0.000 0.412
#> GSM25409 5 0.6596 0.0720 0.072 0.392 0.000 0.052 0.484
#> GSM25410 3 0.0451 0.8790 0.008 0.004 0.988 0.000 0.000
#> GSM25426 5 0.0290 0.5457 0.000 0.008 0.000 0.000 0.992
#> GSM25427 5 0.6420 -0.0871 0.400 0.024 0.000 0.096 0.480
#> GSM25540 1 0.6952 0.0875 0.476 0.036 0.396 0.060 0.032
#> GSM25541 1 0.6952 0.0875 0.476 0.036 0.396 0.060 0.032
#> GSM25542 1 0.7656 0.2046 0.496 0.088 0.312 0.048 0.056
#> GSM25543 1 0.7656 0.2046 0.496 0.088 0.312 0.048 0.056
#> GSM25479 1 0.5857 0.3863 0.556 0.040 0.000 0.036 0.368
#> GSM25480 1 0.5857 0.3863 0.556 0.040 0.000 0.036 0.368
#> GSM25481 5 0.4572 0.4004 0.016 0.168 0.000 0.056 0.760
#> GSM25482 5 0.4572 0.4004 0.016 0.168 0.000 0.056 0.760
#> GSM48654 2 0.8319 0.2913 0.168 0.336 0.000 0.180 0.316
#> GSM48650 2 0.4620 0.3604 0.008 0.616 0.000 0.008 0.368
#> GSM48651 2 0.4210 0.3806 0.000 0.588 0.000 0.000 0.412
#> GSM48652 2 0.4210 0.3806 0.000 0.588 0.000 0.000 0.412
#> GSM48653 2 0.4210 0.3806 0.000 0.588 0.000 0.000 0.412
#> GSM48662 2 0.4210 0.3806 0.000 0.588 0.000 0.000 0.412
#> GSM48663 2 0.4723 0.3242 0.008 0.612 0.000 0.012 0.368
#> GSM25524 1 0.4622 0.4495 0.800 0.036 0.028 0.104 0.032
#> GSM25525 1 0.5600 0.3819 0.576 0.044 0.000 0.020 0.360
#> GSM25526 1 0.6471 0.4021 0.668 0.052 0.164 0.080 0.036
#> GSM25527 1 0.3545 0.5959 0.832 0.008 0.004 0.024 0.132
#> GSM25528 1 0.6719 0.0916 0.524 0.048 0.340 0.084 0.004
#> GSM25529 1 0.4071 0.5801 0.788 0.020 0.000 0.024 0.168
#> GSM25530 1 0.6749 0.0577 0.512 0.048 0.352 0.084 0.004
#> GSM25531 1 0.6749 0.0577 0.512 0.048 0.352 0.084 0.004
#> GSM48661 2 0.8319 0.2913 0.168 0.336 0.000 0.180 0.316
#> GSM25561 3 0.3268 0.8181 0.068 0.032 0.868 0.032 0.000
#> GSM25562 1 0.4897 0.5547 0.728 0.004 0.000 0.112 0.156
#> GSM25563 3 0.2511 0.8438 0.044 0.024 0.908 0.024 0.000
#> GSM25564 5 0.6334 -0.0322 0.120 0.360 0.000 0.012 0.508
#> GSM25565 5 0.6596 0.0720 0.072 0.392 0.000 0.052 0.484
#> GSM25566 5 0.6596 0.0720 0.072 0.392 0.000 0.052 0.484
#> GSM25568 2 0.6017 0.3352 0.020 0.480 0.000 0.064 0.436
#> GSM25569 2 0.5974 0.3226 0.020 0.464 0.000 0.060 0.456
#> GSM25552 5 0.6596 0.0720 0.072 0.392 0.000 0.052 0.484
#> GSM25553 5 0.6596 0.0720 0.072 0.392 0.000 0.052 0.484
#> GSM25578 1 0.3730 0.6026 0.808 0.004 0.000 0.036 0.152
#> GSM25579 1 0.3730 0.6026 0.808 0.004 0.000 0.036 0.152
#> GSM25580 1 0.3688 0.6026 0.812 0.004 0.000 0.036 0.148
#> GSM25581 1 0.3688 0.6026 0.812 0.004 0.000 0.036 0.148
#> GSM48655 5 0.7207 -0.2615 0.052 0.380 0.000 0.140 0.428
#> GSM48656 2 0.8319 0.2913 0.168 0.336 0.000 0.180 0.316
#> GSM48657 5 0.7207 -0.2615 0.052 0.380 0.000 0.140 0.428
#> GSM48658 2 0.8319 0.2913 0.168 0.336 0.000 0.180 0.316
#> GSM25624 2 0.8402 0.2346 0.228 0.312 0.000 0.156 0.304
#> GSM25625 3 0.1200 0.8694 0.016 0.012 0.964 0.008 0.000
#> GSM25626 3 0.0162 0.8775 0.000 0.004 0.996 0.000 0.000
#> GSM25627 1 0.5110 0.5646 0.728 0.080 0.000 0.024 0.168
#> GSM25628 3 0.0162 0.8775 0.000 0.004 0.996 0.000 0.000
#> GSM25629 1 0.3421 0.6047 0.816 0.016 0.000 0.004 0.164
#> GSM25630 3 0.0162 0.8775 0.000 0.004 0.996 0.000 0.000
#> GSM25631 1 0.3461 0.6037 0.812 0.016 0.000 0.004 0.168
#> GSM25632 3 0.0162 0.8775 0.000 0.004 0.996 0.000 0.000
#> GSM25633 1 0.3421 0.6047 0.816 0.016 0.000 0.004 0.164
#> GSM25634 1 0.3421 0.6047 0.816 0.016 0.000 0.004 0.164
#> GSM25635 1 0.3421 0.6047 0.816 0.016 0.000 0.004 0.164
#> GSM25656 3 0.6293 0.3292 0.340 0.040 0.556 0.060 0.004
#> GSM25657 1 0.4088 0.5927 0.792 0.004 0.000 0.064 0.140
#> GSM25658 1 0.6420 0.2391 0.480 0.024 0.000 0.096 0.400
#> GSM25659 1 0.6235 0.3545 0.544 0.040 0.000 0.064 0.352
#> GSM25660 1 0.4019 0.5948 0.792 0.004 0.000 0.052 0.152
#> GSM25661 1 0.5544 0.4832 0.660 0.004 0.000 0.192 0.144
#> GSM25662 5 0.3898 0.3996 0.160 0.016 0.000 0.024 0.800
#> GSM25663 5 0.3898 0.3996 0.160 0.016 0.000 0.024 0.800
#> GSM25680 5 0.0609 0.5532 0.020 0.000 0.000 0.000 0.980
#> GSM25681 5 0.0609 0.5532 0.020 0.000 0.000 0.000 0.980
#> GSM25682 5 0.0000 0.5478 0.000 0.000 0.000 0.000 1.000
#> GSM25683 5 0.0000 0.5478 0.000 0.000 0.000 0.000 1.000
#> GSM25684 5 0.6120 -0.0952 0.020 0.336 0.000 0.088 0.556
#> GSM25685 5 0.6120 -0.0952 0.020 0.336 0.000 0.088 0.556
#> GSM25686 5 0.0000 0.5478 0.000 0.000 0.000 0.000 1.000
#> GSM25687 5 0.0000 0.5478 0.000 0.000 0.000 0.000 1.000
#> GSM48664 4 0.3628 0.7637 0.216 0.000 0.000 0.772 0.012
#> GSM48665 4 0.3628 0.7637 0.216 0.000 0.000 0.772 0.012
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM25548 5 0.0806 0.75423 0.020 0.008 0.000 0.000 0.972 0.000
#> GSM25549 5 0.0806 0.75423 0.020 0.008 0.000 0.000 0.972 0.000
#> GSM25550 5 0.0806 0.75423 0.020 0.008 0.000 0.000 0.972 0.000
#> GSM25551 5 0.0692 0.75354 0.020 0.004 0.000 0.000 0.976 0.000
#> GSM25570 5 0.0806 0.75423 0.020 0.008 0.000 0.000 0.972 0.000
#> GSM25571 5 0.0806 0.75423 0.020 0.008 0.000 0.000 0.972 0.000
#> GSM25358 1 0.6806 0.00994 0.460 0.024 0.376 0.064 0.012 0.064
#> GSM25359 1 0.6806 0.00994 0.460 0.024 0.376 0.064 0.012 0.064
#> GSM25360 3 0.0291 0.86083 0.004 0.000 0.992 0.000 0.000 0.004
#> GSM25361 1 0.6736 0.14568 0.480 0.012 0.360 0.056 0.024 0.068
#> GSM25377 4 0.0865 0.74809 0.036 0.000 0.000 0.964 0.000 0.000
#> GSM25378 5 0.6561 0.02773 0.388 0.036 0.000 0.092 0.452 0.032
#> GSM25401 4 0.5861 0.44665 0.012 0.084 0.000 0.540 0.340 0.024
#> GSM25402 4 0.5861 0.44665 0.012 0.084 0.000 0.540 0.340 0.024
#> GSM25349 2 0.6196 0.60619 0.072 0.512 0.000 0.028 0.356 0.032
#> GSM25350 2 0.6196 0.60619 0.072 0.512 0.000 0.028 0.356 0.032
#> GSM25356 5 0.0291 0.74658 0.000 0.004 0.000 0.004 0.992 0.000
#> GSM25357 5 0.0000 0.74791 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM25385 3 0.5396 0.42895 0.300 0.016 0.612 0.040 0.000 0.032
#> GSM25386 3 0.0291 0.86133 0.004 0.000 0.992 0.000 0.000 0.004
#> GSM25399 4 0.0865 0.74809 0.036 0.000 0.000 0.964 0.000 0.000
#> GSM25400 5 0.6561 0.02773 0.388 0.036 0.000 0.092 0.452 0.032
#> GSM48659 2 0.6095 -0.24222 0.004 0.392 0.000 0.000 0.224 0.380
#> GSM48660 2 0.3431 0.65621 0.000 0.756 0.000 0.000 0.228 0.016
#> GSM25409 2 0.6196 0.60619 0.072 0.512 0.000 0.028 0.356 0.032
#> GSM25410 3 0.0291 0.86133 0.004 0.000 0.992 0.000 0.000 0.004
#> GSM25426 5 0.0260 0.74608 0.000 0.000 0.000 0.000 0.992 0.008
#> GSM25427 5 0.6561 0.02773 0.388 0.036 0.000 0.092 0.452 0.032
#> GSM25540 1 0.6736 0.14568 0.480 0.012 0.360 0.056 0.024 0.068
#> GSM25541 1 0.6736 0.14568 0.480 0.012 0.360 0.056 0.024 0.068
#> GSM25542 1 0.7043 0.20304 0.492 0.044 0.312 0.012 0.040 0.100
#> GSM25543 1 0.7043 0.20304 0.492 0.044 0.312 0.012 0.040 0.100
#> GSM25479 1 0.5676 0.44937 0.596 0.068 0.000 0.028 0.292 0.016
#> GSM25480 1 0.5676 0.44937 0.596 0.068 0.000 0.028 0.292 0.016
#> GSM25481 5 0.4711 0.42181 0.008 0.188 0.000 0.040 0.724 0.040
#> GSM25482 5 0.4711 0.42181 0.008 0.188 0.000 0.040 0.724 0.040
#> GSM48654 6 0.4276 0.86563 0.104 0.000 0.000 0.000 0.168 0.728
#> GSM48650 2 0.2738 0.64293 0.000 0.820 0.000 0.000 0.176 0.004
#> GSM48651 2 0.3431 0.65621 0.000 0.756 0.000 0.000 0.228 0.016
#> GSM48652 2 0.3431 0.65621 0.000 0.756 0.000 0.000 0.228 0.016
#> GSM48653 2 0.3431 0.65621 0.000 0.756 0.000 0.000 0.228 0.016
#> GSM48662 2 0.3431 0.65621 0.000 0.756 0.000 0.000 0.228 0.016
#> GSM48663 2 0.3333 0.65187 0.000 0.784 0.000 0.000 0.192 0.024
#> GSM25524 1 0.4584 0.53472 0.784 0.036 0.032 0.068 0.004 0.076
#> GSM25525 1 0.5451 0.43917 0.604 0.080 0.000 0.004 0.288 0.024
#> GSM25526 1 0.5972 0.43194 0.676 0.036 0.140 0.064 0.008 0.076
#> GSM25527 1 0.2992 0.64729 0.868 0.020 0.004 0.008 0.084 0.016
#> GSM25528 1 0.6988 0.12208 0.488 0.036 0.300 0.064 0.000 0.112
#> GSM25529 1 0.3701 0.63196 0.816 0.028 0.000 0.012 0.120 0.024
#> GSM25530 1 0.7019 0.08974 0.476 0.036 0.312 0.064 0.000 0.112
#> GSM25531 1 0.7019 0.08974 0.476 0.036 0.312 0.064 0.000 0.112
#> GSM48661 6 0.4276 0.86563 0.104 0.000 0.000 0.000 0.168 0.728
#> GSM25561 3 0.3413 0.79392 0.064 0.008 0.848 0.036 0.000 0.044
#> GSM25562 1 0.4094 0.61672 0.772 0.012 0.000 0.108 0.108 0.000
#> GSM25563 3 0.2677 0.82028 0.040 0.008 0.892 0.028 0.000 0.032
#> GSM25564 2 0.6067 0.35114 0.136 0.440 0.000 0.004 0.404 0.016
#> GSM25565 2 0.6196 0.60619 0.072 0.512 0.000 0.028 0.356 0.032
#> GSM25566 2 0.6196 0.60619 0.072 0.512 0.000 0.028 0.356 0.032
#> GSM25568 2 0.5667 0.34592 0.000 0.532 0.000 0.000 0.240 0.228
#> GSM25569 2 0.5561 0.38199 0.000 0.552 0.000 0.000 0.244 0.204
#> GSM25552 2 0.6196 0.60619 0.072 0.512 0.000 0.028 0.356 0.032
#> GSM25553 2 0.6196 0.60619 0.072 0.512 0.000 0.028 0.356 0.032
#> GSM25578 1 0.3053 0.65229 0.852 0.008 0.000 0.036 0.100 0.004
#> GSM25579 1 0.3053 0.65229 0.852 0.008 0.000 0.036 0.100 0.004
#> GSM25580 1 0.3005 0.65211 0.856 0.008 0.000 0.036 0.096 0.004
#> GSM25581 1 0.3005 0.65211 0.856 0.008 0.000 0.036 0.096 0.004
#> GSM48655 6 0.6109 0.68389 0.036 0.200 0.000 0.000 0.208 0.556
#> GSM48656 6 0.4276 0.86563 0.104 0.000 0.000 0.000 0.168 0.728
#> GSM48657 6 0.6109 0.68389 0.036 0.200 0.000 0.000 0.208 0.556
#> GSM48658 6 0.4276 0.86563 0.104 0.000 0.000 0.000 0.168 0.728
#> GSM25624 6 0.5215 0.75588 0.192 0.008 0.000 0.000 0.160 0.640
#> GSM25625 3 0.1823 0.85311 0.016 0.012 0.932 0.004 0.000 0.036
#> GSM25626 3 0.0777 0.85581 0.000 0.004 0.972 0.000 0.000 0.024
#> GSM25627 1 0.4371 0.58875 0.732 0.004 0.000 0.000 0.116 0.148
#> GSM25628 3 0.0777 0.85581 0.000 0.004 0.972 0.000 0.000 0.024
#> GSM25629 1 0.2678 0.65309 0.860 0.004 0.000 0.000 0.116 0.020
#> GSM25630 3 0.0777 0.85581 0.000 0.004 0.972 0.000 0.000 0.024
#> GSM25631 1 0.2723 0.65245 0.856 0.004 0.000 0.000 0.120 0.020
#> GSM25632 3 0.0777 0.85581 0.000 0.004 0.972 0.000 0.000 0.024
#> GSM25633 1 0.2678 0.65309 0.860 0.004 0.000 0.000 0.116 0.020
#> GSM25634 1 0.2678 0.65309 0.860 0.004 0.000 0.000 0.116 0.020
#> GSM25635 1 0.2678 0.65309 0.860 0.004 0.000 0.000 0.116 0.020
#> GSM25656 3 0.6199 0.26241 0.340 0.012 0.520 0.056 0.000 0.072
#> GSM25657 1 0.3348 0.64597 0.832 0.012 0.000 0.060 0.096 0.000
#> GSM25658 1 0.6539 0.18391 0.472 0.036 0.000 0.092 0.368 0.032
#> GSM25659 1 0.6149 0.40076 0.560 0.060 0.000 0.060 0.300 0.020
#> GSM25660 1 0.3259 0.64716 0.836 0.012 0.000 0.048 0.104 0.000
#> GSM25661 1 0.4742 0.55817 0.696 0.012 0.000 0.196 0.096 0.000
#> GSM25662 5 0.3719 0.59886 0.160 0.004 0.000 0.016 0.792 0.028
#> GSM25663 5 0.3719 0.59886 0.160 0.004 0.000 0.016 0.792 0.028
#> GSM25680 5 0.0806 0.75423 0.020 0.008 0.000 0.000 0.972 0.000
#> GSM25681 5 0.0806 0.75423 0.020 0.008 0.000 0.000 0.972 0.000
#> GSM25682 5 0.0000 0.74791 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM25683 5 0.0000 0.74791 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM25684 5 0.5988 -0.15971 0.004 0.208 0.000 0.000 0.456 0.332
#> GSM25685 5 0.5988 -0.15971 0.004 0.208 0.000 0.000 0.456 0.332
#> GSM25686 5 0.0000 0.74791 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM25687 5 0.0000 0.74791 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM48664 4 0.1219 0.75001 0.048 0.004 0.000 0.948 0.000 0.000
#> GSM48665 4 0.1219 0.75001 0.048 0.004 0.000 0.948 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 genotype/variation(p) k
#> ATC:hclust 96 6.13e-02 2
#> ATC:hclust 79 2.43e-03 3
#> ATC:hclust 58 5.60e-07 4
#> ATC:hclust 44 3.87e-06 5
#> ATC:hclust 69 5.43e-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["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 100 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 0.684 0.912 0.944 0.4813 0.508 0.508
#> 3 3 0.438 0.654 0.787 0.2997 0.712 0.501
#> 4 4 0.598 0.718 0.809 0.1528 0.838 0.597
#> 5 5 0.646 0.616 0.753 0.0831 0.916 0.715
#> 6 6 0.709 0.595 0.746 0.0465 0.921 0.678
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
#> GSM25548 2 0.0938 0.945 0.012 0.988
#> GSM25549 2 0.1414 0.944 0.020 0.980
#> GSM25550 2 0.1414 0.944 0.020 0.980
#> GSM25551 2 0.0938 0.945 0.012 0.988
#> GSM25570 2 0.1414 0.944 0.020 0.980
#> GSM25571 2 0.0938 0.945 0.012 0.988
#> GSM25358 1 0.2423 0.946 0.960 0.040
#> GSM25359 1 0.1414 0.945 0.980 0.020
#> GSM25360 1 0.1184 0.944 0.984 0.016
#> GSM25361 1 0.1843 0.946 0.972 0.028
#> GSM25377 2 0.1843 0.942 0.028 0.972
#> GSM25378 2 0.1843 0.942 0.028 0.972
#> GSM25401 2 0.1843 0.942 0.028 0.972
#> GSM25402 2 0.1843 0.942 0.028 0.972
#> GSM25349 2 0.0672 0.942 0.008 0.992
#> GSM25350 2 0.0672 0.942 0.008 0.992
#> GSM25356 2 0.1843 0.942 0.028 0.972
#> GSM25357 2 0.0938 0.945 0.012 0.988
#> GSM25385 1 0.1184 0.944 0.984 0.016
#> GSM25386 1 0.1184 0.944 0.984 0.016
#> GSM25399 1 0.7602 0.756 0.780 0.220
#> GSM25400 1 0.7528 0.763 0.784 0.216
#> GSM48659 2 0.0000 0.942 0.000 1.000
#> GSM48660 2 0.0000 0.942 0.000 1.000
#> GSM25409 2 0.1843 0.942 0.028 0.972
#> GSM25410 1 0.1184 0.944 0.984 0.016
#> GSM25426 2 0.0938 0.945 0.012 0.988
#> GSM25427 2 0.1843 0.942 0.028 0.972
#> GSM25540 1 0.0376 0.943 0.996 0.004
#> GSM25541 1 0.1633 0.945 0.976 0.024
#> GSM25542 1 0.2423 0.946 0.960 0.040
#> GSM25543 1 0.2423 0.946 0.960 0.040
#> GSM25479 2 0.6438 0.831 0.164 0.836
#> GSM25480 2 0.6438 0.831 0.164 0.836
#> GSM25481 2 0.1843 0.942 0.028 0.972
#> GSM25482 2 0.1843 0.942 0.028 0.972
#> GSM48654 2 0.0672 0.940 0.008 0.992
#> GSM48650 2 0.0000 0.942 0.000 1.000
#> GSM48651 2 0.0000 0.942 0.000 1.000
#> GSM48652 2 0.0000 0.942 0.000 1.000
#> GSM48653 2 0.0000 0.942 0.000 1.000
#> GSM48662 2 0.0000 0.942 0.000 1.000
#> GSM48663 2 0.0000 0.942 0.000 1.000
#> GSM25524 1 0.1843 0.944 0.972 0.028
#> GSM25525 2 0.6438 0.831 0.164 0.836
#> GSM25526 1 0.0672 0.943 0.992 0.008
#> GSM25527 1 0.4298 0.924 0.912 0.088
#> GSM25528 1 0.0000 0.941 1.000 0.000
#> GSM25529 1 0.4298 0.924 0.912 0.088
#> GSM25530 1 0.0000 0.941 1.000 0.000
#> GSM25531 1 0.0000 0.941 1.000 0.000
#> GSM48661 1 0.2948 0.943 0.948 0.052
#> GSM25561 1 0.1184 0.944 0.984 0.016
#> GSM25562 2 0.9000 0.599 0.316 0.684
#> GSM25563 1 0.1184 0.944 0.984 0.016
#> GSM25564 2 0.1633 0.942 0.024 0.976
#> GSM25565 2 0.0938 0.945 0.012 0.988
#> GSM25566 2 0.1414 0.944 0.020 0.980
#> GSM25568 2 0.0000 0.942 0.000 1.000
#> GSM25569 2 0.0000 0.942 0.000 1.000
#> GSM25552 2 0.1843 0.942 0.028 0.972
#> GSM25553 2 0.1843 0.942 0.028 0.972
#> GSM25578 1 0.4431 0.921 0.908 0.092
#> GSM25579 2 0.9129 0.575 0.328 0.672
#> GSM25580 1 0.4298 0.924 0.912 0.088
#> GSM25581 1 0.4298 0.924 0.912 0.088
#> GSM48655 2 0.0000 0.942 0.000 1.000
#> GSM48656 2 0.6623 0.795 0.172 0.828
#> GSM48657 2 0.0000 0.942 0.000 1.000
#> GSM48658 1 0.5178 0.922 0.884 0.116
#> GSM25624 2 0.6887 0.796 0.184 0.816
#> GSM25625 1 0.1184 0.944 0.984 0.016
#> GSM25626 1 0.1184 0.944 0.984 0.016
#> GSM25627 1 0.5178 0.922 0.884 0.116
#> GSM25628 1 0.1184 0.944 0.984 0.016
#> GSM25629 1 0.2423 0.946 0.960 0.040
#> GSM25630 1 0.1184 0.944 0.984 0.016
#> GSM25631 1 0.4562 0.923 0.904 0.096
#> GSM25632 1 0.1184 0.944 0.984 0.016
#> GSM25633 1 0.4298 0.924 0.912 0.088
#> GSM25634 1 0.4298 0.924 0.912 0.088
#> GSM25635 1 0.4298 0.924 0.912 0.088
#> GSM25656 1 0.1184 0.944 0.984 0.016
#> GSM25657 1 0.4298 0.924 0.912 0.088
#> GSM25658 2 0.9087 0.583 0.324 0.676
#> GSM25659 2 0.5178 0.878 0.116 0.884
#> GSM25660 1 0.4431 0.921 0.908 0.092
#> GSM25661 2 0.9087 0.583 0.324 0.676
#> GSM25662 2 0.0938 0.945 0.012 0.988
#> GSM25663 1 0.6801 0.828 0.820 0.180
#> GSM25680 2 0.0938 0.945 0.012 0.988
#> GSM25681 2 0.0938 0.945 0.012 0.988
#> GSM25682 2 0.0938 0.945 0.012 0.988
#> GSM25683 2 0.0938 0.945 0.012 0.988
#> GSM25684 2 0.0938 0.945 0.012 0.988
#> GSM25685 2 0.0938 0.945 0.012 0.988
#> GSM25686 2 0.0938 0.945 0.012 0.988
#> GSM25687 2 0.0938 0.945 0.012 0.988
#> GSM48664 2 0.8327 0.693 0.264 0.736
#> GSM48665 2 0.5294 0.874 0.120 0.880
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM25548 2 0.6008 0.7941 0.372 0.628 0.000
#> GSM25549 2 0.6008 0.7941 0.372 0.628 0.000
#> GSM25550 2 0.6008 0.7941 0.372 0.628 0.000
#> GSM25551 2 0.6008 0.7941 0.372 0.628 0.000
#> GSM25570 2 0.6008 0.7941 0.372 0.628 0.000
#> GSM25571 2 0.6008 0.7941 0.372 0.628 0.000
#> GSM25358 1 0.5785 0.5575 0.668 0.000 0.332
#> GSM25359 1 0.6302 0.1952 0.520 0.000 0.480
#> GSM25360 3 0.0000 0.8338 0.000 0.000 1.000
#> GSM25361 1 0.6244 0.3101 0.560 0.000 0.440
#> GSM25377 1 0.5397 0.4293 0.720 0.280 0.000
#> GSM25378 2 0.6308 0.6422 0.492 0.508 0.000
#> GSM25401 1 0.5968 0.1709 0.636 0.364 0.000
#> GSM25402 1 0.3267 0.5182 0.884 0.116 0.000
#> GSM25349 2 0.0892 0.7121 0.020 0.980 0.000
#> GSM25350 2 0.0892 0.7121 0.020 0.980 0.000
#> GSM25356 2 0.6045 0.7925 0.380 0.620 0.000
#> GSM25357 2 0.6008 0.7941 0.372 0.628 0.000
#> GSM25385 3 0.0000 0.8338 0.000 0.000 1.000
#> GSM25386 3 0.0000 0.8338 0.000 0.000 1.000
#> GSM25399 1 0.3755 0.6799 0.872 0.120 0.008
#> GSM25400 1 0.1647 0.6942 0.960 0.004 0.036
#> GSM48659 2 0.3412 0.7219 0.124 0.876 0.000
#> GSM48660 2 0.0892 0.7121 0.020 0.980 0.000
#> GSM25409 2 0.5291 0.7683 0.268 0.732 0.000
#> GSM25410 3 0.0000 0.8338 0.000 0.000 1.000
#> GSM25426 2 0.6008 0.7941 0.372 0.628 0.000
#> GSM25427 2 0.6299 0.6826 0.476 0.524 0.000
#> GSM25540 3 0.5216 0.5552 0.260 0.000 0.740
#> GSM25541 1 0.6008 0.4550 0.628 0.000 0.372
#> GSM25542 3 0.8230 0.3830 0.280 0.112 0.608
#> GSM25543 3 0.9001 0.2138 0.332 0.148 0.520
#> GSM25479 1 0.1636 0.6973 0.964 0.020 0.016
#> GSM25480 1 0.1636 0.6973 0.964 0.020 0.016
#> GSM25481 2 0.6079 0.7907 0.388 0.612 0.000
#> GSM25482 2 0.6079 0.7907 0.388 0.612 0.000
#> GSM48654 2 0.5595 0.5783 0.228 0.756 0.016
#> GSM48650 2 0.0892 0.7121 0.020 0.980 0.000
#> GSM48651 2 0.0892 0.7121 0.020 0.980 0.000
#> GSM48652 2 0.0892 0.7121 0.020 0.980 0.000
#> GSM48653 2 0.0892 0.7121 0.020 0.980 0.000
#> GSM48662 2 0.0892 0.7121 0.020 0.980 0.000
#> GSM48663 2 0.0892 0.7121 0.020 0.980 0.000
#> GSM25524 1 0.5397 0.6202 0.720 0.000 0.280
#> GSM25525 1 0.1905 0.6795 0.956 0.028 0.016
#> GSM25526 3 0.5859 0.3761 0.344 0.000 0.656
#> GSM25527 1 0.5363 0.6220 0.724 0.000 0.276
#> GSM25528 3 0.5859 0.3761 0.344 0.000 0.656
#> GSM25529 1 0.5397 0.6202 0.720 0.000 0.280
#> GSM25530 3 0.2448 0.7801 0.076 0.000 0.924
#> GSM25531 1 0.6291 0.2212 0.532 0.000 0.468
#> GSM48661 3 0.9904 0.0971 0.316 0.284 0.400
#> GSM25561 3 0.0000 0.8338 0.000 0.000 1.000
#> GSM25562 1 0.3482 0.6764 0.872 0.128 0.000
#> GSM25563 3 0.0000 0.8338 0.000 0.000 1.000
#> GSM25564 2 0.5291 0.7683 0.268 0.732 0.000
#> GSM25565 2 0.5291 0.7683 0.268 0.732 0.000
#> GSM25566 2 0.5291 0.7683 0.268 0.732 0.000
#> GSM25568 2 0.1964 0.7179 0.056 0.944 0.000
#> GSM25569 2 0.0592 0.7128 0.012 0.988 0.000
#> GSM25552 2 0.5291 0.7683 0.268 0.732 0.000
#> GSM25553 2 0.5291 0.7683 0.268 0.732 0.000
#> GSM25578 1 0.3038 0.7085 0.896 0.000 0.104
#> GSM25579 1 0.3028 0.7090 0.920 0.032 0.048
#> GSM25580 1 0.3340 0.7059 0.880 0.000 0.120
#> GSM25581 1 0.5397 0.6202 0.720 0.000 0.280
#> GSM48655 2 0.3482 0.7198 0.128 0.872 0.000
#> GSM48656 2 0.6095 -0.0487 0.392 0.608 0.000
#> GSM48657 2 0.1031 0.7091 0.024 0.976 0.000
#> GSM48658 1 0.8171 0.5223 0.644 0.172 0.184
#> GSM25624 1 0.4802 0.6412 0.824 0.156 0.020
#> GSM25625 3 0.0000 0.8338 0.000 0.000 1.000
#> GSM25626 3 0.0000 0.8338 0.000 0.000 1.000
#> GSM25627 1 0.8125 0.5280 0.648 0.172 0.180
#> GSM25628 3 0.0000 0.8338 0.000 0.000 1.000
#> GSM25629 1 0.5988 0.4577 0.632 0.000 0.368
#> GSM25630 3 0.0000 0.8338 0.000 0.000 1.000
#> GSM25631 1 0.4605 0.6726 0.796 0.000 0.204
#> GSM25632 3 0.0000 0.8338 0.000 0.000 1.000
#> GSM25633 1 0.5363 0.6220 0.724 0.000 0.276
#> GSM25634 1 0.5363 0.6220 0.724 0.000 0.276
#> GSM25635 1 0.5363 0.6220 0.724 0.000 0.276
#> GSM25656 3 0.0000 0.8338 0.000 0.000 1.000
#> GSM25657 1 0.6294 0.5999 0.692 0.020 0.288
#> GSM25658 1 0.1337 0.6947 0.972 0.012 0.016
#> GSM25659 1 0.1711 0.6958 0.960 0.032 0.008
#> GSM25660 1 0.3116 0.7078 0.892 0.000 0.108
#> GSM25661 1 0.3454 0.6892 0.888 0.104 0.008
#> GSM25662 1 0.6309 -0.6249 0.504 0.496 0.000
#> GSM25663 1 0.3116 0.7078 0.892 0.000 0.108
#> GSM25680 2 0.6008 0.7941 0.372 0.628 0.000
#> GSM25681 2 0.6008 0.7941 0.372 0.628 0.000
#> GSM25682 2 0.6008 0.7941 0.372 0.628 0.000
#> GSM25683 2 0.6008 0.7941 0.372 0.628 0.000
#> GSM25684 2 0.6008 0.7941 0.372 0.628 0.000
#> GSM25685 2 0.6008 0.7941 0.372 0.628 0.000
#> GSM25686 2 0.5988 0.7943 0.368 0.632 0.000
#> GSM25687 2 0.5988 0.7943 0.368 0.632 0.000
#> GSM48664 1 0.3482 0.6764 0.872 0.128 0.000
#> GSM48665 1 0.3482 0.6764 0.872 0.128 0.000
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM25548 2 0.0000 0.83619 0.000 1.000 0.000 0.000
#> GSM25549 2 0.0000 0.83619 0.000 1.000 0.000 0.000
#> GSM25550 2 0.0000 0.83619 0.000 1.000 0.000 0.000
#> GSM25551 2 0.0000 0.83619 0.000 1.000 0.000 0.000
#> GSM25570 2 0.0000 0.83619 0.000 1.000 0.000 0.000
#> GSM25571 2 0.0000 0.83619 0.000 1.000 0.000 0.000
#> GSM25358 1 0.7347 0.61705 0.616 0.232 0.048 0.104
#> GSM25359 1 0.7943 0.60959 0.604 0.156 0.136 0.104
#> GSM25360 3 0.0000 0.96980 0.000 0.000 1.000 0.000
#> GSM25361 1 0.4484 0.76618 0.812 0.004 0.064 0.120
#> GSM25377 1 0.6595 0.51115 0.608 0.124 0.000 0.268
#> GSM25378 2 0.5277 0.65196 0.132 0.752 0.000 0.116
#> GSM25401 1 0.7834 -0.13749 0.372 0.368 0.000 0.260
#> GSM25402 2 0.6823 0.46099 0.244 0.596 0.000 0.160
#> GSM25349 4 0.4868 0.71196 0.012 0.304 0.000 0.684
#> GSM25350 4 0.4868 0.71196 0.012 0.304 0.000 0.684
#> GSM25356 2 0.2589 0.77640 0.000 0.884 0.000 0.116
#> GSM25357 2 0.0000 0.83619 0.000 1.000 0.000 0.000
#> GSM25385 3 0.0921 0.96133 0.000 0.000 0.972 0.028
#> GSM25386 3 0.0000 0.96980 0.000 0.000 1.000 0.000
#> GSM25399 1 0.4059 0.71798 0.788 0.012 0.000 0.200
#> GSM25400 1 0.3958 0.77542 0.836 0.052 0.000 0.112
#> GSM48659 4 0.5236 0.62590 0.008 0.432 0.000 0.560
#> GSM48660 4 0.4973 0.73194 0.008 0.348 0.000 0.644
#> GSM25409 2 0.5969 0.43882 0.044 0.564 0.000 0.392
#> GSM25410 3 0.0000 0.96980 0.000 0.000 1.000 0.000
#> GSM25426 2 0.0000 0.83619 0.000 1.000 0.000 0.000
#> GSM25427 2 0.5452 0.64822 0.108 0.736 0.000 0.156
#> GSM25540 1 0.6616 0.49265 0.584 0.000 0.308 0.108
#> GSM25541 1 0.4254 0.77175 0.824 0.004 0.052 0.120
#> GSM25542 4 0.8735 -0.00735 0.296 0.044 0.252 0.408
#> GSM25543 4 0.8181 -0.05675 0.344 0.044 0.140 0.472
#> GSM25479 1 0.3009 0.79156 0.892 0.056 0.000 0.052
#> GSM25480 1 0.3009 0.79156 0.892 0.056 0.000 0.052
#> GSM25481 2 0.2611 0.78771 0.008 0.896 0.000 0.096
#> GSM25482 2 0.2342 0.79682 0.008 0.912 0.000 0.080
#> GSM48654 4 0.5811 0.62470 0.116 0.180 0.000 0.704
#> GSM48650 4 0.4955 0.73184 0.008 0.344 0.000 0.648
#> GSM48651 4 0.4973 0.73194 0.008 0.348 0.000 0.644
#> GSM48652 4 0.4837 0.73209 0.004 0.348 0.000 0.648
#> GSM48653 4 0.4973 0.73194 0.008 0.348 0.000 0.644
#> GSM48662 4 0.4955 0.73184 0.008 0.344 0.000 0.648
#> GSM48663 4 0.4955 0.73184 0.008 0.344 0.000 0.648
#> GSM25524 1 0.2686 0.80248 0.916 0.040 0.012 0.032
#> GSM25525 1 0.4467 0.72386 0.788 0.172 0.000 0.040
#> GSM25526 1 0.6684 0.43998 0.560 0.000 0.336 0.104
#> GSM25527 1 0.2689 0.79973 0.916 0.036 0.012 0.036
#> GSM25528 1 0.5897 0.41378 0.588 0.000 0.368 0.044
#> GSM25529 1 0.2484 0.80103 0.924 0.040 0.012 0.024
#> GSM25530 3 0.4365 0.71792 0.188 0.000 0.784 0.028
#> GSM25531 1 0.3745 0.77442 0.852 0.000 0.088 0.060
#> GSM48661 4 0.5578 0.30283 0.312 0.040 0.000 0.648
#> GSM25561 3 0.1022 0.95937 0.000 0.000 0.968 0.032
#> GSM25562 1 0.4059 0.71798 0.788 0.012 0.000 0.200
#> GSM25563 3 0.0921 0.96139 0.000 0.000 0.972 0.028
#> GSM25564 2 0.6488 0.43410 0.104 0.604 0.000 0.292
#> GSM25565 2 0.5013 0.45279 0.020 0.688 0.000 0.292
#> GSM25566 2 0.5038 0.44882 0.020 0.684 0.000 0.296
#> GSM25568 4 0.4453 0.69863 0.012 0.244 0.000 0.744
#> GSM25569 4 0.5018 0.73335 0.012 0.332 0.000 0.656
#> GSM25552 2 0.5663 0.62272 0.060 0.676 0.000 0.264
#> GSM25553 2 0.5989 0.60000 0.080 0.656 0.000 0.264
#> GSM25578 1 0.1909 0.80026 0.940 0.048 0.004 0.008
#> GSM25579 1 0.1854 0.79973 0.940 0.048 0.000 0.012
#> GSM25580 1 0.1677 0.80130 0.948 0.040 0.012 0.000
#> GSM25581 1 0.2484 0.80103 0.924 0.040 0.012 0.024
#> GSM48655 4 0.5582 0.64603 0.024 0.400 0.000 0.576
#> GSM48656 4 0.5219 0.47185 0.244 0.044 0.000 0.712
#> GSM48657 4 0.5137 0.72677 0.024 0.296 0.000 0.680
#> GSM48658 1 0.4948 0.34527 0.560 0.000 0.000 0.440
#> GSM25624 1 0.4327 0.70305 0.768 0.016 0.000 0.216
#> GSM25625 3 0.0000 0.96980 0.000 0.000 1.000 0.000
#> GSM25626 3 0.0000 0.96980 0.000 0.000 1.000 0.000
#> GSM25627 1 0.4564 0.56921 0.672 0.000 0.000 0.328
#> GSM25628 3 0.0000 0.96980 0.000 0.000 1.000 0.000
#> GSM25629 1 0.4335 0.74655 0.796 0.000 0.036 0.168
#> GSM25630 3 0.0000 0.96980 0.000 0.000 1.000 0.000
#> GSM25631 1 0.3708 0.76162 0.832 0.020 0.000 0.148
#> GSM25632 3 0.0000 0.96980 0.000 0.000 1.000 0.000
#> GSM25633 1 0.3367 0.78345 0.876 0.020 0.012 0.092
#> GSM25634 1 0.3734 0.77552 0.852 0.020 0.012 0.116
#> GSM25635 1 0.4008 0.76690 0.832 0.020 0.012 0.136
#> GSM25656 3 0.1022 0.95937 0.000 0.000 0.968 0.032
#> GSM25657 1 0.1953 0.80045 0.944 0.012 0.012 0.032
#> GSM25658 1 0.3521 0.78282 0.864 0.052 0.000 0.084
#> GSM25659 1 0.4780 0.74152 0.788 0.116 0.000 0.096
#> GSM25660 1 0.1909 0.80026 0.940 0.048 0.004 0.008
#> GSM25661 1 0.3625 0.74529 0.828 0.012 0.000 0.160
#> GSM25662 2 0.2589 0.76631 0.044 0.912 0.000 0.044
#> GSM25663 1 0.5354 0.73813 0.752 0.152 0.004 0.092
#> GSM25680 2 0.0000 0.83619 0.000 1.000 0.000 0.000
#> GSM25681 2 0.0000 0.83619 0.000 1.000 0.000 0.000
#> GSM25682 2 0.0000 0.83619 0.000 1.000 0.000 0.000
#> GSM25683 2 0.0000 0.83619 0.000 1.000 0.000 0.000
#> GSM25684 2 0.0000 0.83619 0.000 1.000 0.000 0.000
#> GSM25685 2 0.0000 0.83619 0.000 1.000 0.000 0.000
#> GSM25686 2 0.0000 0.83619 0.000 1.000 0.000 0.000
#> GSM25687 2 0.0000 0.83619 0.000 1.000 0.000 0.000
#> GSM48664 1 0.4059 0.71798 0.788 0.012 0.000 0.200
#> GSM48665 1 0.4253 0.70855 0.776 0.016 0.000 0.208
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM25548 4 0.0324 0.9230 0.000 0.004 0.000 0.992 0.004
#> GSM25549 4 0.0324 0.9230 0.000 0.004 0.000 0.992 0.004
#> GSM25550 4 0.0324 0.9230 0.000 0.004 0.000 0.992 0.004
#> GSM25551 4 0.0162 0.9235 0.000 0.004 0.000 0.996 0.000
#> GSM25570 4 0.0324 0.9230 0.000 0.004 0.000 0.992 0.004
#> GSM25571 4 0.0324 0.9230 0.000 0.004 0.000 0.992 0.004
#> GSM25358 1 0.5864 0.6015 0.708 0.040 0.016 0.116 0.120
#> GSM25359 1 0.6093 0.6176 0.708 0.040 0.056 0.068 0.128
#> GSM25360 3 0.0324 0.9319 0.004 0.000 0.992 0.000 0.004
#> GSM25361 1 0.4792 0.6391 0.724 0.048 0.008 0.004 0.216
#> GSM25377 5 0.6776 0.3632 0.320 0.156 0.000 0.024 0.500
#> GSM25378 4 0.4891 0.6009 0.068 0.004 0.000 0.704 0.224
#> GSM25401 5 0.7170 0.4046 0.276 0.160 0.000 0.056 0.508
#> GSM25402 5 0.7018 0.2374 0.284 0.008 0.000 0.328 0.380
#> GSM25349 2 0.3946 0.6208 0.000 0.800 0.000 0.120 0.080
#> GSM25350 2 0.3946 0.6208 0.000 0.800 0.000 0.120 0.080
#> GSM25356 4 0.2848 0.7956 0.000 0.004 0.000 0.840 0.156
#> GSM25357 4 0.0404 0.9230 0.000 0.012 0.000 0.988 0.000
#> GSM25385 3 0.2204 0.9078 0.008 0.036 0.920 0.000 0.036
#> GSM25386 3 0.0162 0.9322 0.004 0.000 0.996 0.000 0.000
#> GSM25399 5 0.4829 0.0470 0.480 0.020 0.000 0.000 0.500
#> GSM25400 1 0.4189 0.5381 0.736 0.012 0.000 0.012 0.240
#> GSM48659 2 0.5751 0.4928 0.000 0.540 0.000 0.364 0.096
#> GSM48660 2 0.2471 0.7160 0.000 0.864 0.000 0.136 0.000
#> GSM25409 5 0.6686 0.2541 0.004 0.376 0.000 0.200 0.420
#> GSM25410 3 0.0162 0.9322 0.004 0.000 0.996 0.000 0.000
#> GSM25426 4 0.0404 0.9230 0.000 0.012 0.000 0.988 0.000
#> GSM25427 4 0.5125 0.5450 0.060 0.008 0.000 0.672 0.260
#> GSM25540 1 0.6180 0.5666 0.644 0.040 0.168 0.000 0.148
#> GSM25541 1 0.3804 0.6716 0.796 0.044 0.000 0.000 0.160
#> GSM25542 5 0.8520 -0.0428 0.196 0.240 0.196 0.004 0.364
#> GSM25543 5 0.7324 -0.0185 0.212 0.324 0.028 0.004 0.432
#> GSM25479 1 0.2672 0.6334 0.872 0.004 0.000 0.008 0.116
#> GSM25480 1 0.2672 0.6334 0.872 0.004 0.000 0.008 0.116
#> GSM25481 4 0.3631 0.7743 0.012 0.024 0.000 0.820 0.144
#> GSM25482 4 0.3629 0.7842 0.012 0.028 0.000 0.824 0.136
#> GSM48654 2 0.5816 0.3742 0.040 0.512 0.000 0.028 0.420
#> GSM48650 2 0.2471 0.7160 0.000 0.864 0.000 0.136 0.000
#> GSM48651 2 0.2471 0.7160 0.000 0.864 0.000 0.136 0.000
#> GSM48652 2 0.2864 0.7146 0.000 0.852 0.000 0.136 0.012
#> GSM48653 2 0.2471 0.7160 0.000 0.864 0.000 0.136 0.000
#> GSM48662 2 0.2471 0.7160 0.000 0.864 0.000 0.136 0.000
#> GSM48663 2 0.2471 0.7160 0.000 0.864 0.000 0.136 0.000
#> GSM25524 1 0.2676 0.6901 0.884 0.036 0.000 0.000 0.080
#> GSM25525 1 0.3359 0.6289 0.848 0.004 0.000 0.052 0.096
#> GSM25526 1 0.5916 0.5769 0.664 0.032 0.172 0.000 0.132
#> GSM25527 1 0.0404 0.7000 0.988 0.000 0.000 0.000 0.012
#> GSM25528 1 0.5559 0.5722 0.688 0.036 0.200 0.000 0.076
#> GSM25529 1 0.0324 0.6995 0.992 0.004 0.000 0.000 0.004
#> GSM25530 3 0.5796 0.4572 0.284 0.036 0.624 0.000 0.056
#> GSM25531 1 0.3643 0.6807 0.848 0.036 0.044 0.000 0.072
#> GSM48661 5 0.6626 -0.2707 0.144 0.396 0.004 0.008 0.448
#> GSM25561 3 0.2791 0.8891 0.016 0.036 0.892 0.000 0.056
#> GSM25562 1 0.4907 -0.1302 0.488 0.024 0.000 0.000 0.488
#> GSM25563 3 0.2036 0.9113 0.008 0.028 0.928 0.000 0.036
#> GSM25564 5 0.7554 0.2343 0.052 0.360 0.000 0.208 0.380
#> GSM25565 2 0.6815 -0.1975 0.004 0.424 0.000 0.248 0.324
#> GSM25566 5 0.6844 0.2201 0.004 0.364 0.000 0.244 0.388
#> GSM25568 2 0.5006 0.4889 0.000 0.624 0.000 0.048 0.328
#> GSM25569 2 0.4255 0.6881 0.000 0.776 0.000 0.128 0.096
#> GSM25552 5 0.7346 0.3156 0.036 0.324 0.000 0.220 0.420
#> GSM25553 5 0.7389 0.3186 0.040 0.324 0.000 0.216 0.420
#> GSM25578 1 0.0880 0.6908 0.968 0.000 0.000 0.000 0.032
#> GSM25579 1 0.1341 0.6803 0.944 0.000 0.000 0.000 0.056
#> GSM25580 1 0.0794 0.6920 0.972 0.000 0.000 0.000 0.028
#> GSM25581 1 0.0290 0.6994 0.992 0.000 0.000 0.000 0.008
#> GSM48655 2 0.6128 0.5395 0.000 0.564 0.000 0.204 0.232
#> GSM48656 2 0.5880 0.3371 0.064 0.504 0.004 0.008 0.420
#> GSM48657 2 0.5032 0.6136 0.000 0.688 0.000 0.092 0.220
#> GSM48658 1 0.6242 0.3021 0.444 0.124 0.004 0.000 0.428
#> GSM25624 1 0.6103 0.3402 0.456 0.092 0.004 0.004 0.444
#> GSM25625 3 0.0000 0.9320 0.000 0.000 1.000 0.000 0.000
#> GSM25626 3 0.0000 0.9320 0.000 0.000 1.000 0.000 0.000
#> GSM25627 1 0.6030 0.3364 0.472 0.100 0.004 0.000 0.424
#> GSM25628 3 0.0000 0.9320 0.000 0.000 1.000 0.000 0.000
#> GSM25629 1 0.5024 0.5511 0.628 0.040 0.004 0.000 0.328
#> GSM25630 3 0.0000 0.9320 0.000 0.000 1.000 0.000 0.000
#> GSM25631 1 0.4550 0.5899 0.692 0.028 0.004 0.000 0.276
#> GSM25632 3 0.0000 0.9320 0.000 0.000 1.000 0.000 0.000
#> GSM25633 1 0.3006 0.6728 0.836 0.004 0.004 0.000 0.156
#> GSM25634 1 0.3611 0.6430 0.780 0.008 0.004 0.000 0.208
#> GSM25635 1 0.4311 0.5998 0.712 0.020 0.004 0.000 0.264
#> GSM25656 3 0.2494 0.8976 0.008 0.032 0.904 0.000 0.056
#> GSM25657 1 0.0404 0.6973 0.988 0.000 0.000 0.000 0.012
#> GSM25658 1 0.3250 0.5906 0.820 0.004 0.000 0.008 0.168
#> GSM25659 1 0.4067 0.4782 0.748 0.004 0.000 0.020 0.228
#> GSM25660 1 0.0880 0.6908 0.968 0.000 0.000 0.000 0.032
#> GSM25661 1 0.4341 0.1530 0.592 0.004 0.000 0.000 0.404
#> GSM25662 4 0.2082 0.8686 0.016 0.032 0.000 0.928 0.024
#> GSM25663 1 0.3629 0.6842 0.824 0.028 0.000 0.012 0.136
#> GSM25680 4 0.0162 0.9228 0.000 0.000 0.000 0.996 0.004
#> GSM25681 4 0.0162 0.9228 0.000 0.000 0.000 0.996 0.004
#> GSM25682 4 0.0510 0.9226 0.000 0.016 0.000 0.984 0.000
#> GSM25683 4 0.0404 0.9230 0.000 0.012 0.000 0.988 0.000
#> GSM25684 4 0.0510 0.9226 0.000 0.016 0.000 0.984 0.000
#> GSM25685 4 0.0404 0.9230 0.000 0.012 0.000 0.988 0.000
#> GSM25686 4 0.0510 0.9226 0.000 0.016 0.000 0.984 0.000
#> GSM25687 4 0.0510 0.9226 0.000 0.016 0.000 0.984 0.000
#> GSM48664 5 0.4980 0.0473 0.484 0.028 0.000 0.000 0.488
#> GSM48665 1 0.5049 -0.1498 0.484 0.032 0.000 0.000 0.484
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM25548 5 0.1720 0.8260 0.000 0.032 0.000 0.040 0.928 0.000
#> GSM25549 5 0.2201 0.8152 0.000 0.052 0.000 0.048 0.900 0.000
#> GSM25550 5 0.2134 0.8171 0.000 0.052 0.000 0.044 0.904 0.000
#> GSM25551 5 0.0000 0.8326 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM25570 5 0.1720 0.8260 0.000 0.032 0.000 0.040 0.928 0.000
#> GSM25571 5 0.1720 0.8260 0.000 0.032 0.000 0.040 0.928 0.000
#> GSM25358 1 0.6563 0.5616 0.604 0.204 0.016 0.092 0.056 0.028
#> GSM25359 1 0.6825 0.5481 0.588 0.204 0.028 0.100 0.048 0.032
#> GSM25360 3 0.0000 0.8431 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM25361 1 0.6303 0.5444 0.592 0.196 0.008 0.124 0.000 0.080
#> GSM25377 4 0.3496 0.7049 0.140 0.052 0.000 0.804 0.004 0.000
#> GSM25378 5 0.6503 0.2878 0.036 0.128 0.000 0.332 0.488 0.016
#> GSM25401 4 0.3676 0.6973 0.120 0.052 0.000 0.808 0.020 0.000
#> GSM25402 4 0.6934 0.4812 0.148 0.116 0.000 0.528 0.196 0.012
#> GSM25349 2 0.5425 0.7099 0.000 0.636 0.000 0.076 0.048 0.240
#> GSM25350 2 0.5425 0.7099 0.000 0.636 0.000 0.076 0.048 0.240
#> GSM25356 5 0.4637 0.5257 0.000 0.064 0.000 0.308 0.628 0.000
#> GSM25357 5 0.0405 0.8317 0.000 0.004 0.000 0.008 0.988 0.000
#> GSM25385 3 0.4719 0.7296 0.012 0.164 0.732 0.072 0.000 0.020
#> GSM25386 3 0.0000 0.8431 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM25399 4 0.3052 0.6873 0.216 0.004 0.000 0.780 0.000 0.000
#> GSM25400 1 0.6619 0.1269 0.456 0.176 0.000 0.328 0.020 0.020
#> GSM48659 5 0.6317 -0.2566 0.000 0.156 0.000 0.032 0.412 0.400
#> GSM48660 2 0.4956 0.7883 0.000 0.592 0.000 0.004 0.072 0.332
#> GSM25409 4 0.5108 0.2277 0.000 0.436 0.000 0.484 0.080 0.000
#> GSM25410 3 0.0000 0.8431 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM25426 5 0.0508 0.8309 0.000 0.004 0.000 0.012 0.984 0.000
#> GSM25427 5 0.6402 0.1526 0.024 0.128 0.000 0.392 0.440 0.016
#> GSM25540 1 0.6776 0.5407 0.584 0.176 0.048 0.108 0.000 0.084
#> GSM25541 1 0.5662 0.5929 0.652 0.160 0.000 0.108 0.000 0.080
#> GSM25542 6 0.8394 0.2866 0.144 0.248 0.152 0.096 0.000 0.360
#> GSM25543 6 0.7829 0.3389 0.148 0.312 0.024 0.168 0.000 0.348
#> GSM25479 1 0.4229 0.5491 0.744 0.064 0.000 0.180 0.000 0.012
#> GSM25480 1 0.4229 0.5491 0.744 0.064 0.000 0.180 0.000 0.012
#> GSM25481 5 0.5760 0.4280 0.008 0.120 0.000 0.300 0.560 0.012
#> GSM25482 5 0.5455 0.4797 0.008 0.116 0.000 0.280 0.592 0.004
#> GSM48654 6 0.1637 0.5187 0.056 0.004 0.000 0.004 0.004 0.932
#> GSM48650 2 0.4956 0.7883 0.000 0.592 0.000 0.004 0.072 0.332
#> GSM48651 2 0.4956 0.7883 0.000 0.592 0.000 0.004 0.072 0.332
#> GSM48652 2 0.4835 0.7834 0.000 0.592 0.000 0.000 0.072 0.336
#> GSM48653 2 0.4956 0.7883 0.000 0.592 0.000 0.004 0.072 0.332
#> GSM48662 2 0.4956 0.7883 0.000 0.592 0.000 0.004 0.072 0.332
#> GSM48663 2 0.4956 0.7883 0.000 0.592 0.000 0.004 0.072 0.332
#> GSM25524 1 0.5310 0.6355 0.660 0.216 0.000 0.088 0.004 0.032
#> GSM25525 1 0.4266 0.6049 0.776 0.072 0.000 0.124 0.012 0.016
#> GSM25526 1 0.5812 0.6018 0.664 0.168 0.048 0.088 0.000 0.032
#> GSM25527 1 0.0458 0.7005 0.984 0.000 0.000 0.016 0.000 0.000
#> GSM25528 1 0.5704 0.5961 0.656 0.196 0.048 0.080 0.000 0.020
#> GSM25529 1 0.1268 0.7057 0.952 0.036 0.000 0.008 0.000 0.004
#> GSM25530 3 0.7439 0.0940 0.312 0.200 0.388 0.080 0.000 0.020
#> GSM25531 1 0.4666 0.6333 0.716 0.184 0.000 0.076 0.000 0.024
#> GSM48661 6 0.2346 0.5618 0.124 0.008 0.000 0.000 0.000 0.868
#> GSM25561 3 0.5206 0.7014 0.028 0.172 0.700 0.080 0.000 0.020
#> GSM25562 4 0.3298 0.6764 0.236 0.008 0.000 0.756 0.000 0.000
#> GSM25563 3 0.4036 0.7560 0.004 0.136 0.780 0.068 0.000 0.012
#> GSM25564 4 0.5669 0.3333 0.016 0.360 0.000 0.516 0.108 0.000
#> GSM25565 2 0.5966 -0.1744 0.000 0.452 0.000 0.404 0.120 0.024
#> GSM25566 2 0.5475 -0.2185 0.000 0.460 0.000 0.416 0.124 0.000
#> GSM25568 6 0.4019 0.1212 0.000 0.216 0.000 0.028 0.016 0.740
#> GSM25569 2 0.5486 0.6750 0.000 0.508 0.000 0.024 0.068 0.400
#> GSM25552 4 0.5524 0.3540 0.012 0.364 0.000 0.524 0.100 0.000
#> GSM25553 4 0.5524 0.3540 0.012 0.364 0.000 0.524 0.100 0.000
#> GSM25578 1 0.1398 0.6972 0.940 0.008 0.000 0.052 0.000 0.000
#> GSM25579 1 0.1524 0.6949 0.932 0.008 0.000 0.060 0.000 0.000
#> GSM25580 1 0.1265 0.6989 0.948 0.008 0.000 0.044 0.000 0.000
#> GSM25581 1 0.0458 0.7015 0.984 0.000 0.000 0.016 0.000 0.000
#> GSM48655 6 0.4132 0.2011 0.000 0.104 0.000 0.004 0.136 0.756
#> GSM48656 6 0.1411 0.5229 0.060 0.004 0.000 0.000 0.000 0.936
#> GSM48657 6 0.4045 -0.0602 0.000 0.268 0.000 0.004 0.028 0.700
#> GSM48658 6 0.3874 0.4004 0.356 0.000 0.000 0.008 0.000 0.636
#> GSM25624 6 0.4867 0.3559 0.340 0.048 0.000 0.012 0.000 0.600
#> GSM25625 3 0.0000 0.8431 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM25626 3 0.0000 0.8431 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM25627 6 0.3887 0.3926 0.360 0.000 0.000 0.008 0.000 0.632
#> GSM25628 3 0.0000 0.8431 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM25629 1 0.4923 0.4523 0.656 0.028 0.000 0.052 0.000 0.264
#> GSM25630 3 0.0000 0.8431 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM25631 1 0.3809 0.5168 0.732 0.004 0.000 0.024 0.000 0.240
#> GSM25632 3 0.0000 0.8431 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM25633 1 0.2070 0.6598 0.892 0.000 0.000 0.008 0.000 0.100
#> GSM25634 1 0.2553 0.6282 0.848 0.000 0.000 0.008 0.000 0.144
#> GSM25635 1 0.3012 0.5780 0.796 0.000 0.000 0.008 0.000 0.196
#> GSM25656 3 0.4771 0.7216 0.004 0.156 0.724 0.092 0.000 0.024
#> GSM25657 1 0.2234 0.6701 0.872 0.004 0.000 0.124 0.000 0.000
#> GSM25658 1 0.5036 0.3918 0.648 0.076 0.000 0.260 0.004 0.012
#> GSM25659 1 0.5087 0.3066 0.616 0.064 0.000 0.304 0.004 0.012
#> GSM25660 1 0.1265 0.6989 0.948 0.008 0.000 0.044 0.000 0.000
#> GSM25661 4 0.3607 0.5001 0.348 0.000 0.000 0.652 0.000 0.000
#> GSM25662 5 0.1635 0.8053 0.016 0.012 0.000 0.012 0.944 0.016
#> GSM25663 1 0.4562 0.6510 0.752 0.124 0.000 0.068 0.000 0.056
#> GSM25680 5 0.1720 0.8260 0.000 0.032 0.000 0.040 0.928 0.000
#> GSM25681 5 0.1720 0.8260 0.000 0.032 0.000 0.040 0.928 0.000
#> GSM25682 5 0.0146 0.8326 0.000 0.004 0.000 0.000 0.996 0.000
#> GSM25683 5 0.0291 0.8319 0.000 0.004 0.000 0.004 0.992 0.000
#> GSM25684 5 0.0260 0.8316 0.000 0.000 0.000 0.008 0.992 0.000
#> GSM25685 5 0.0260 0.8316 0.000 0.000 0.000 0.008 0.992 0.000
#> GSM25686 5 0.0146 0.8326 0.000 0.004 0.000 0.000 0.996 0.000
#> GSM25687 5 0.0146 0.8326 0.000 0.004 0.000 0.000 0.996 0.000
#> GSM48664 4 0.3287 0.6907 0.220 0.012 0.000 0.768 0.000 0.000
#> GSM48665 4 0.3320 0.6959 0.212 0.016 0.000 0.772 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 genotype/variation(p) k
#> ATC:kmeans 100 1.21e-04 2
#> ATC:kmeans 86 6.61e-04 3
#> ATC:kmeans 86 9.88e-09 4
#> ATC:kmeans 74 1.24e-10 5
#> ATC:kmeans 75 2.57e-16 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 100 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 0.957 0.939 0.975 0.5041 0.496 0.496
#> 3 3 0.910 0.917 0.961 0.3078 0.762 0.555
#> 4 4 0.803 0.846 0.915 0.1324 0.828 0.546
#> 5 5 0.842 0.810 0.894 0.0685 0.888 0.598
#> 6 6 0.827 0.728 0.864 0.0380 0.940 0.722
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
#> GSM25548 2 0.0000 0.961 0.000 1.000
#> GSM25549 2 0.0000 0.961 0.000 1.000
#> GSM25550 2 0.0000 0.961 0.000 1.000
#> GSM25551 2 0.0000 0.961 0.000 1.000
#> GSM25570 2 0.0000 0.961 0.000 1.000
#> GSM25571 2 0.0000 0.961 0.000 1.000
#> GSM25358 1 0.0000 0.987 1.000 0.000
#> GSM25359 1 0.0000 0.987 1.000 0.000
#> GSM25360 1 0.0000 0.987 1.000 0.000
#> GSM25361 1 0.0000 0.987 1.000 0.000
#> GSM25377 2 0.0000 0.961 0.000 1.000
#> GSM25378 2 0.0000 0.961 0.000 1.000
#> GSM25401 2 0.0000 0.961 0.000 1.000
#> GSM25402 2 0.0000 0.961 0.000 1.000
#> GSM25349 2 0.0000 0.961 0.000 1.000
#> GSM25350 2 0.0000 0.961 0.000 1.000
#> GSM25356 2 0.0000 0.961 0.000 1.000
#> GSM25357 2 0.0000 0.961 0.000 1.000
#> GSM25385 1 0.0000 0.987 1.000 0.000
#> GSM25386 1 0.0000 0.987 1.000 0.000
#> GSM25399 1 0.0000 0.987 1.000 0.000
#> GSM25400 1 0.0000 0.987 1.000 0.000
#> GSM48659 2 0.0000 0.961 0.000 1.000
#> GSM48660 2 0.0000 0.961 0.000 1.000
#> GSM25409 2 0.0000 0.961 0.000 1.000
#> GSM25410 1 0.0000 0.987 1.000 0.000
#> GSM25426 2 0.0000 0.961 0.000 1.000
#> GSM25427 2 0.0000 0.961 0.000 1.000
#> GSM25540 1 0.0000 0.987 1.000 0.000
#> GSM25541 1 0.0000 0.987 1.000 0.000
#> GSM25542 1 0.0938 0.977 0.988 0.012
#> GSM25543 1 0.0938 0.977 0.988 0.012
#> GSM25479 2 0.9661 0.387 0.392 0.608
#> GSM25480 2 0.9393 0.473 0.356 0.644
#> GSM25481 2 0.0000 0.961 0.000 1.000
#> GSM25482 2 0.0000 0.961 0.000 1.000
#> GSM48654 2 0.4298 0.879 0.088 0.912
#> GSM48650 2 0.0000 0.961 0.000 1.000
#> GSM48651 2 0.0000 0.961 0.000 1.000
#> GSM48652 2 0.0000 0.961 0.000 1.000
#> GSM48653 2 0.0000 0.961 0.000 1.000
#> GSM48662 2 0.0000 0.961 0.000 1.000
#> GSM48663 2 0.0000 0.961 0.000 1.000
#> GSM25524 1 0.0000 0.987 1.000 0.000
#> GSM25525 2 0.9248 0.508 0.340 0.660
#> GSM25526 1 0.0000 0.987 1.000 0.000
#> GSM25527 1 0.0000 0.987 1.000 0.000
#> GSM25528 1 0.0000 0.987 1.000 0.000
#> GSM25529 1 0.0000 0.987 1.000 0.000
#> GSM25530 1 0.0000 0.987 1.000 0.000
#> GSM25531 1 0.0000 0.987 1.000 0.000
#> GSM48661 1 0.0000 0.987 1.000 0.000
#> GSM25561 1 0.0000 0.987 1.000 0.000
#> GSM25562 1 0.0000 0.987 1.000 0.000
#> GSM25563 1 0.0000 0.987 1.000 0.000
#> GSM25564 2 0.0000 0.961 0.000 1.000
#> GSM25565 2 0.0000 0.961 0.000 1.000
#> GSM25566 2 0.0000 0.961 0.000 1.000
#> GSM25568 2 0.0000 0.961 0.000 1.000
#> GSM25569 2 0.0000 0.961 0.000 1.000
#> GSM25552 2 0.0000 0.961 0.000 1.000
#> GSM25553 2 0.0000 0.961 0.000 1.000
#> GSM25578 1 0.0000 0.987 1.000 0.000
#> GSM25579 1 0.0000 0.987 1.000 0.000
#> GSM25580 1 0.0000 0.987 1.000 0.000
#> GSM25581 1 0.0000 0.987 1.000 0.000
#> GSM48655 2 0.0000 0.961 0.000 1.000
#> GSM48656 1 0.9129 0.482 0.672 0.328
#> GSM48657 2 0.0000 0.961 0.000 1.000
#> GSM48658 1 0.0000 0.987 1.000 0.000
#> GSM25624 2 0.9954 0.188 0.460 0.540
#> GSM25625 1 0.0000 0.987 1.000 0.000
#> GSM25626 1 0.0000 0.987 1.000 0.000
#> GSM25627 1 0.0000 0.987 1.000 0.000
#> GSM25628 1 0.0000 0.987 1.000 0.000
#> GSM25629 1 0.0000 0.987 1.000 0.000
#> GSM25630 1 0.0000 0.987 1.000 0.000
#> GSM25631 1 0.0000 0.987 1.000 0.000
#> GSM25632 1 0.0000 0.987 1.000 0.000
#> GSM25633 1 0.0000 0.987 1.000 0.000
#> GSM25634 1 0.0000 0.987 1.000 0.000
#> GSM25635 1 0.0000 0.987 1.000 0.000
#> GSM25656 1 0.0000 0.987 1.000 0.000
#> GSM25657 1 0.0000 0.987 1.000 0.000
#> GSM25658 1 0.0672 0.980 0.992 0.008
#> GSM25659 2 0.3274 0.909 0.060 0.940
#> GSM25660 1 0.0000 0.987 1.000 0.000
#> GSM25661 1 0.0672 0.980 0.992 0.008
#> GSM25662 2 0.0000 0.961 0.000 1.000
#> GSM25663 1 0.0000 0.987 1.000 0.000
#> GSM25680 2 0.0000 0.961 0.000 1.000
#> GSM25681 2 0.0000 0.961 0.000 1.000
#> GSM25682 2 0.0000 0.961 0.000 1.000
#> GSM25683 2 0.0000 0.961 0.000 1.000
#> GSM25684 2 0.0000 0.961 0.000 1.000
#> GSM25685 2 0.0000 0.961 0.000 1.000
#> GSM25686 2 0.0000 0.961 0.000 1.000
#> GSM25687 2 0.0000 0.961 0.000 1.000
#> GSM48664 1 0.7056 0.749 0.808 0.192
#> GSM48665 2 0.7745 0.705 0.228 0.772
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM25548 2 0.0000 0.9849 0.000 1.000 0.000
#> GSM25549 2 0.0000 0.9849 0.000 1.000 0.000
#> GSM25550 2 0.0000 0.9849 0.000 1.000 0.000
#> GSM25551 2 0.0000 0.9849 0.000 1.000 0.000
#> GSM25570 2 0.0000 0.9849 0.000 1.000 0.000
#> GSM25571 2 0.0000 0.9849 0.000 1.000 0.000
#> GSM25358 3 0.0000 0.9607 0.000 0.000 1.000
#> GSM25359 3 0.0000 0.9607 0.000 0.000 1.000
#> GSM25360 3 0.0000 0.9607 0.000 0.000 1.000
#> GSM25361 3 0.0000 0.9607 0.000 0.000 1.000
#> GSM25377 1 0.3686 0.8018 0.860 0.140 0.000
#> GSM25378 2 0.4452 0.7512 0.192 0.808 0.000
#> GSM25401 1 0.5016 0.6738 0.760 0.240 0.000
#> GSM25402 1 0.5138 0.6699 0.748 0.252 0.000
#> GSM25349 2 0.0592 0.9834 0.012 0.988 0.000
#> GSM25350 2 0.0592 0.9834 0.012 0.988 0.000
#> GSM25356 2 0.0237 0.9829 0.004 0.996 0.000
#> GSM25357 2 0.0000 0.9849 0.000 1.000 0.000
#> GSM25385 3 0.0000 0.9607 0.000 0.000 1.000
#> GSM25386 3 0.0000 0.9607 0.000 0.000 1.000
#> GSM25399 1 0.0237 0.9137 0.996 0.000 0.004
#> GSM25400 1 0.1860 0.8913 0.948 0.000 0.052
#> GSM48659 2 0.0000 0.9849 0.000 1.000 0.000
#> GSM48660 2 0.0592 0.9834 0.012 0.988 0.000
#> GSM25409 2 0.0592 0.9834 0.012 0.988 0.000
#> GSM25410 3 0.0000 0.9607 0.000 0.000 1.000
#> GSM25426 2 0.0000 0.9849 0.000 1.000 0.000
#> GSM25427 2 0.4399 0.7575 0.188 0.812 0.000
#> GSM25540 3 0.0000 0.9607 0.000 0.000 1.000
#> GSM25541 3 0.0000 0.9607 0.000 0.000 1.000
#> GSM25542 3 0.0424 0.9553 0.008 0.000 0.992
#> GSM25543 3 0.0424 0.9553 0.008 0.000 0.992
#> GSM25479 1 0.0424 0.9146 0.992 0.000 0.008
#> GSM25480 1 0.0475 0.9137 0.992 0.004 0.004
#> GSM25481 2 0.0000 0.9849 0.000 1.000 0.000
#> GSM25482 2 0.0000 0.9849 0.000 1.000 0.000
#> GSM48654 3 0.4808 0.7479 0.008 0.188 0.804
#> GSM48650 2 0.0592 0.9834 0.012 0.988 0.000
#> GSM48651 2 0.0592 0.9834 0.012 0.988 0.000
#> GSM48652 2 0.0592 0.9834 0.012 0.988 0.000
#> GSM48653 2 0.0592 0.9834 0.012 0.988 0.000
#> GSM48662 2 0.0592 0.9834 0.012 0.988 0.000
#> GSM48663 2 0.0592 0.9834 0.012 0.988 0.000
#> GSM25524 1 0.5733 0.5657 0.676 0.000 0.324
#> GSM25525 1 0.1860 0.8874 0.948 0.052 0.000
#> GSM25526 3 0.0000 0.9607 0.000 0.000 1.000
#> GSM25527 1 0.0892 0.9122 0.980 0.000 0.020
#> GSM25528 3 0.4346 0.7497 0.184 0.000 0.816
#> GSM25529 1 0.0747 0.9136 0.984 0.000 0.016
#> GSM25530 3 0.0237 0.9583 0.004 0.000 0.996
#> GSM25531 1 0.6204 0.3122 0.576 0.000 0.424
#> GSM48661 3 0.0237 0.9580 0.004 0.000 0.996
#> GSM25561 3 0.0000 0.9607 0.000 0.000 1.000
#> GSM25562 1 0.0000 0.9120 1.000 0.000 0.000
#> GSM25563 3 0.0000 0.9607 0.000 0.000 1.000
#> GSM25564 2 0.0592 0.9834 0.012 0.988 0.000
#> GSM25565 2 0.0592 0.9834 0.012 0.988 0.000
#> GSM25566 2 0.0592 0.9834 0.012 0.988 0.000
#> GSM25568 2 0.1453 0.9642 0.008 0.968 0.024
#> GSM25569 2 0.0424 0.9840 0.008 0.992 0.000
#> GSM25552 2 0.0424 0.9840 0.008 0.992 0.000
#> GSM25553 2 0.0424 0.9840 0.008 0.992 0.000
#> GSM25578 1 0.0592 0.9143 0.988 0.000 0.012
#> GSM25579 1 0.0424 0.9146 0.992 0.000 0.008
#> GSM25580 1 0.0592 0.9143 0.988 0.000 0.012
#> GSM25581 1 0.0892 0.9122 0.980 0.000 0.020
#> GSM48655 2 0.0237 0.9844 0.004 0.996 0.000
#> GSM48656 3 0.5339 0.8200 0.096 0.080 0.824
#> GSM48657 2 0.0592 0.9834 0.012 0.988 0.000
#> GSM48658 3 0.2711 0.8973 0.088 0.000 0.912
#> GSM25624 1 0.2096 0.8821 0.944 0.052 0.004
#> GSM25625 3 0.0000 0.9607 0.000 0.000 1.000
#> GSM25626 3 0.0000 0.9607 0.000 0.000 1.000
#> GSM25627 3 0.2878 0.8902 0.096 0.000 0.904
#> GSM25628 3 0.0000 0.9607 0.000 0.000 1.000
#> GSM25629 3 0.2066 0.9175 0.060 0.000 0.940
#> GSM25630 3 0.0000 0.9607 0.000 0.000 1.000
#> GSM25631 3 0.4931 0.7085 0.232 0.000 0.768
#> GSM25632 3 0.0000 0.9607 0.000 0.000 1.000
#> GSM25633 1 0.0892 0.9122 0.980 0.000 0.020
#> GSM25634 1 0.0747 0.9136 0.984 0.000 0.016
#> GSM25635 1 0.6305 0.0295 0.516 0.000 0.484
#> GSM25656 3 0.0000 0.9607 0.000 0.000 1.000
#> GSM25657 1 0.0747 0.9136 0.984 0.000 0.016
#> GSM25658 1 0.0424 0.9146 0.992 0.000 0.008
#> GSM25659 1 0.0237 0.9124 0.996 0.004 0.000
#> GSM25660 1 0.0592 0.9143 0.988 0.000 0.012
#> GSM25661 1 0.0424 0.9146 0.992 0.000 0.008
#> GSM25662 2 0.0000 0.9849 0.000 1.000 0.000
#> GSM25663 3 0.0000 0.9607 0.000 0.000 1.000
#> GSM25680 2 0.0000 0.9849 0.000 1.000 0.000
#> GSM25681 2 0.0000 0.9849 0.000 1.000 0.000
#> GSM25682 2 0.0000 0.9849 0.000 1.000 0.000
#> GSM25683 2 0.0000 0.9849 0.000 1.000 0.000
#> GSM25684 2 0.0000 0.9849 0.000 1.000 0.000
#> GSM25685 2 0.0000 0.9849 0.000 1.000 0.000
#> GSM25686 2 0.0000 0.9849 0.000 1.000 0.000
#> GSM25687 2 0.0000 0.9849 0.000 1.000 0.000
#> GSM48664 1 0.0237 0.9137 0.996 0.000 0.004
#> GSM48665 1 0.0000 0.9120 1.000 0.000 0.000
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM25548 2 0.0000 0.9131 0.000 1.000 0.000 0.000
#> GSM25549 2 0.0000 0.9131 0.000 1.000 0.000 0.000
#> GSM25550 2 0.0000 0.9131 0.000 1.000 0.000 0.000
#> GSM25551 2 0.0000 0.9131 0.000 1.000 0.000 0.000
#> GSM25570 2 0.0000 0.9131 0.000 1.000 0.000 0.000
#> GSM25571 2 0.0000 0.9131 0.000 1.000 0.000 0.000
#> GSM25358 3 0.0000 0.9649 0.000 0.000 1.000 0.000
#> GSM25359 3 0.0000 0.9649 0.000 0.000 1.000 0.000
#> GSM25360 3 0.0000 0.9649 0.000 0.000 1.000 0.000
#> GSM25361 3 0.0000 0.9649 0.000 0.000 1.000 0.000
#> GSM25377 1 0.2255 0.8650 0.920 0.012 0.000 0.068
#> GSM25378 2 0.0469 0.9021 0.012 0.988 0.000 0.000
#> GSM25401 1 0.6396 0.2779 0.564 0.360 0.000 0.076
#> GSM25402 2 0.4730 0.3782 0.364 0.636 0.000 0.000
#> GSM25349 4 0.3688 0.8529 0.000 0.208 0.000 0.792
#> GSM25350 4 0.3688 0.8529 0.000 0.208 0.000 0.792
#> GSM25356 2 0.0000 0.9131 0.000 1.000 0.000 0.000
#> GSM25357 2 0.0000 0.9131 0.000 1.000 0.000 0.000
#> GSM25385 3 0.0000 0.9649 0.000 0.000 1.000 0.000
#> GSM25386 3 0.0000 0.9649 0.000 0.000 1.000 0.000
#> GSM25399 1 0.0000 0.9228 1.000 0.000 0.000 0.000
#> GSM25400 1 0.1706 0.8960 0.948 0.036 0.016 0.000
#> GSM48659 4 0.4250 0.7876 0.000 0.276 0.000 0.724
#> GSM48660 4 0.3688 0.8529 0.000 0.208 0.000 0.792
#> GSM25409 2 0.4761 0.4211 0.004 0.664 0.000 0.332
#> GSM25410 3 0.0000 0.9649 0.000 0.000 1.000 0.000
#> GSM25426 2 0.0000 0.9131 0.000 1.000 0.000 0.000
#> GSM25427 2 0.0592 0.8981 0.016 0.984 0.000 0.000
#> GSM25540 3 0.0000 0.9649 0.000 0.000 1.000 0.000
#> GSM25541 3 0.0336 0.9584 0.008 0.000 0.992 0.000
#> GSM25542 3 0.0000 0.9649 0.000 0.000 1.000 0.000
#> GSM25543 3 0.0000 0.9649 0.000 0.000 1.000 0.000
#> GSM25479 1 0.0000 0.9228 1.000 0.000 0.000 0.000
#> GSM25480 1 0.0000 0.9228 1.000 0.000 0.000 0.000
#> GSM25481 2 0.0000 0.9131 0.000 1.000 0.000 0.000
#> GSM25482 2 0.0000 0.9131 0.000 1.000 0.000 0.000
#> GSM48654 4 0.0000 0.7581 0.000 0.000 0.000 1.000
#> GSM48650 4 0.3688 0.8529 0.000 0.208 0.000 0.792
#> GSM48651 4 0.3688 0.8529 0.000 0.208 0.000 0.792
#> GSM48652 4 0.3688 0.8529 0.000 0.208 0.000 0.792
#> GSM48653 4 0.3688 0.8529 0.000 0.208 0.000 0.792
#> GSM48662 4 0.3688 0.8529 0.000 0.208 0.000 0.792
#> GSM48663 4 0.3688 0.8529 0.000 0.208 0.000 0.792
#> GSM25524 3 0.2814 0.8273 0.132 0.000 0.868 0.000
#> GSM25525 1 0.0817 0.9107 0.976 0.024 0.000 0.000
#> GSM25526 3 0.0000 0.9649 0.000 0.000 1.000 0.000
#> GSM25527 1 0.2408 0.8727 0.896 0.000 0.000 0.104
#> GSM25528 3 0.0188 0.9616 0.004 0.000 0.996 0.000
#> GSM25529 1 0.0469 0.9219 0.988 0.000 0.000 0.012
#> GSM25530 3 0.0000 0.9649 0.000 0.000 1.000 0.000
#> GSM25531 3 0.4972 0.1194 0.456 0.000 0.544 0.000
#> GSM48661 4 0.3649 0.5874 0.000 0.000 0.204 0.796
#> GSM25561 3 0.0000 0.9649 0.000 0.000 1.000 0.000
#> GSM25562 1 0.0000 0.9228 1.000 0.000 0.000 0.000
#> GSM25563 3 0.0000 0.9649 0.000 0.000 1.000 0.000
#> GSM25564 4 0.4328 0.8094 0.008 0.244 0.000 0.748
#> GSM25565 2 0.4941 0.0671 0.000 0.564 0.000 0.436
#> GSM25566 2 0.4585 0.4266 0.000 0.668 0.000 0.332
#> GSM25568 4 0.2760 0.8279 0.000 0.128 0.000 0.872
#> GSM25569 4 0.3688 0.8529 0.000 0.208 0.000 0.792
#> GSM25552 2 0.3266 0.7376 0.000 0.832 0.000 0.168
#> GSM25553 2 0.3933 0.6857 0.008 0.792 0.000 0.200
#> GSM25578 1 0.0469 0.9219 0.988 0.000 0.000 0.012
#> GSM25579 1 0.0336 0.9225 0.992 0.000 0.000 0.008
#> GSM25580 1 0.0469 0.9219 0.988 0.000 0.000 0.012
#> GSM25581 1 0.0927 0.9184 0.976 0.000 0.008 0.016
#> GSM48655 4 0.4072 0.8028 0.000 0.252 0.000 0.748
#> GSM48656 4 0.0000 0.7581 0.000 0.000 0.000 1.000
#> GSM48657 4 0.3266 0.8440 0.000 0.168 0.000 0.832
#> GSM48658 4 0.4307 0.6329 0.048 0.000 0.144 0.808
#> GSM25624 4 0.2973 0.6310 0.144 0.000 0.000 0.856
#> GSM25625 3 0.0000 0.9649 0.000 0.000 1.000 0.000
#> GSM25626 3 0.0000 0.9649 0.000 0.000 1.000 0.000
#> GSM25627 4 0.4356 0.6288 0.048 0.000 0.148 0.804
#> GSM25628 3 0.0000 0.9649 0.000 0.000 1.000 0.000
#> GSM25629 3 0.4485 0.7429 0.028 0.000 0.772 0.200
#> GSM25630 3 0.0000 0.9649 0.000 0.000 1.000 0.000
#> GSM25631 1 0.7413 0.4331 0.516 0.000 0.252 0.232
#> GSM25632 3 0.0000 0.9649 0.000 0.000 1.000 0.000
#> GSM25633 1 0.3610 0.8013 0.800 0.000 0.000 0.200
#> GSM25634 1 0.3610 0.8013 0.800 0.000 0.000 0.200
#> GSM25635 1 0.5522 0.7246 0.716 0.000 0.080 0.204
#> GSM25656 3 0.0000 0.9649 0.000 0.000 1.000 0.000
#> GSM25657 1 0.0469 0.9219 0.988 0.000 0.000 0.012
#> GSM25658 1 0.0000 0.9228 1.000 0.000 0.000 0.000
#> GSM25659 1 0.0000 0.9228 1.000 0.000 0.000 0.000
#> GSM25660 1 0.0336 0.9225 0.992 0.000 0.000 0.008
#> GSM25661 1 0.0000 0.9228 1.000 0.000 0.000 0.000
#> GSM25662 2 0.0000 0.9131 0.000 1.000 0.000 0.000
#> GSM25663 3 0.0000 0.9649 0.000 0.000 1.000 0.000
#> GSM25680 2 0.0000 0.9131 0.000 1.000 0.000 0.000
#> GSM25681 2 0.0000 0.9131 0.000 1.000 0.000 0.000
#> GSM25682 2 0.0000 0.9131 0.000 1.000 0.000 0.000
#> GSM25683 2 0.0000 0.9131 0.000 1.000 0.000 0.000
#> GSM25684 2 0.0000 0.9131 0.000 1.000 0.000 0.000
#> GSM25685 2 0.0000 0.9131 0.000 1.000 0.000 0.000
#> GSM25686 2 0.0000 0.9131 0.000 1.000 0.000 0.000
#> GSM25687 2 0.0000 0.9131 0.000 1.000 0.000 0.000
#> GSM48664 1 0.0000 0.9228 1.000 0.000 0.000 0.000
#> GSM48665 1 0.0000 0.9228 1.000 0.000 0.000 0.000
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM25548 4 0.0162 0.937 0.000 0.000 0.000 0.996 0.004
#> GSM25549 4 0.0324 0.935 0.000 0.004 0.000 0.992 0.004
#> GSM25550 4 0.0324 0.935 0.000 0.004 0.000 0.992 0.004
#> GSM25551 4 0.0000 0.937 0.000 0.000 0.000 1.000 0.000
#> GSM25570 4 0.0162 0.937 0.000 0.000 0.000 0.996 0.004
#> GSM25571 4 0.0162 0.937 0.000 0.000 0.000 0.996 0.004
#> GSM25358 3 0.0000 0.979 0.000 0.000 1.000 0.000 0.000
#> GSM25359 3 0.0000 0.979 0.000 0.000 1.000 0.000 0.000
#> GSM25360 3 0.0000 0.979 0.000 0.000 1.000 0.000 0.000
#> GSM25361 3 0.0000 0.979 0.000 0.000 1.000 0.000 0.000
#> GSM25377 1 0.5927 0.365 0.540 0.340 0.000 0.000 0.120
#> GSM25378 4 0.3753 0.821 0.044 0.020 0.000 0.832 0.104
#> GSM25401 1 0.6199 0.178 0.468 0.408 0.000 0.004 0.120
#> GSM25402 1 0.6951 0.242 0.484 0.048 0.000 0.348 0.120
#> GSM25349 2 0.0794 0.896 0.000 0.972 0.000 0.028 0.000
#> GSM25350 2 0.0794 0.896 0.000 0.972 0.000 0.028 0.000
#> GSM25356 4 0.1670 0.901 0.000 0.012 0.000 0.936 0.052
#> GSM25357 4 0.0000 0.937 0.000 0.000 0.000 1.000 0.000
#> GSM25385 3 0.0000 0.979 0.000 0.000 1.000 0.000 0.000
#> GSM25386 3 0.0000 0.979 0.000 0.000 1.000 0.000 0.000
#> GSM25399 1 0.3527 0.736 0.828 0.056 0.000 0.000 0.116
#> GSM25400 1 0.3641 0.748 0.840 0.016 0.024 0.008 0.112
#> GSM48659 4 0.5052 0.407 0.000 0.340 0.000 0.612 0.048
#> GSM48660 2 0.1485 0.898 0.000 0.948 0.000 0.032 0.020
#> GSM25409 2 0.3929 0.805 0.036 0.820 0.000 0.028 0.116
#> GSM25410 3 0.0000 0.979 0.000 0.000 1.000 0.000 0.000
#> GSM25426 4 0.0000 0.937 0.000 0.000 0.000 1.000 0.000
#> GSM25427 4 0.4440 0.786 0.072 0.028 0.000 0.792 0.108
#> GSM25540 3 0.0000 0.979 0.000 0.000 1.000 0.000 0.000
#> GSM25541 3 0.1200 0.951 0.012 0.008 0.964 0.000 0.016
#> GSM25542 3 0.0000 0.979 0.000 0.000 1.000 0.000 0.000
#> GSM25543 3 0.0162 0.976 0.000 0.004 0.996 0.000 0.000
#> GSM25479 1 0.0703 0.797 0.976 0.000 0.000 0.000 0.024
#> GSM25480 1 0.0703 0.797 0.976 0.000 0.000 0.000 0.024
#> GSM25481 4 0.5015 0.739 0.028 0.120 0.000 0.748 0.104
#> GSM25482 4 0.4700 0.761 0.020 0.112 0.000 0.768 0.100
#> GSM48654 5 0.3074 0.701 0.000 0.196 0.000 0.000 0.804
#> GSM48650 2 0.1485 0.898 0.000 0.948 0.000 0.032 0.020
#> GSM48651 2 0.1485 0.898 0.000 0.948 0.000 0.032 0.020
#> GSM48652 2 0.1485 0.898 0.000 0.948 0.000 0.032 0.020
#> GSM48653 2 0.1485 0.898 0.000 0.948 0.000 0.032 0.020
#> GSM48662 2 0.1485 0.898 0.000 0.948 0.000 0.032 0.020
#> GSM48663 2 0.1485 0.898 0.000 0.948 0.000 0.032 0.020
#> GSM25524 3 0.3596 0.714 0.212 0.000 0.776 0.000 0.012
#> GSM25525 1 0.2754 0.758 0.884 0.004 0.000 0.080 0.032
#> GSM25526 3 0.2079 0.901 0.064 0.000 0.916 0.000 0.020
#> GSM25527 1 0.2522 0.757 0.880 0.012 0.000 0.000 0.108
#> GSM25528 3 0.1571 0.923 0.060 0.000 0.936 0.000 0.004
#> GSM25529 1 0.2006 0.785 0.916 0.012 0.000 0.000 0.072
#> GSM25530 3 0.0000 0.979 0.000 0.000 1.000 0.000 0.000
#> GSM25531 1 0.5384 0.147 0.512 0.012 0.444 0.000 0.032
#> GSM48661 5 0.3779 0.716 0.000 0.144 0.052 0.000 0.804
#> GSM25561 3 0.0000 0.979 0.000 0.000 1.000 0.000 0.000
#> GSM25562 1 0.3741 0.726 0.816 0.076 0.000 0.000 0.108
#> GSM25563 3 0.0000 0.979 0.000 0.000 1.000 0.000 0.000
#> GSM25564 2 0.1596 0.888 0.012 0.948 0.000 0.028 0.012
#> GSM25565 2 0.2570 0.857 0.000 0.888 0.000 0.028 0.084
#> GSM25566 2 0.3037 0.842 0.004 0.864 0.000 0.032 0.100
#> GSM25568 2 0.4457 0.273 0.000 0.620 0.000 0.012 0.368
#> GSM25569 2 0.2067 0.873 0.000 0.920 0.000 0.032 0.048
#> GSM25552 2 0.4602 0.776 0.040 0.784 0.000 0.064 0.112
#> GSM25553 2 0.4341 0.790 0.044 0.800 0.000 0.044 0.112
#> GSM25578 1 0.1942 0.787 0.920 0.012 0.000 0.000 0.068
#> GSM25579 1 0.1877 0.788 0.924 0.012 0.000 0.000 0.064
#> GSM25580 1 0.2006 0.785 0.916 0.012 0.000 0.000 0.072
#> GSM25581 1 0.2166 0.784 0.912 0.012 0.004 0.000 0.072
#> GSM48655 5 0.6371 0.408 0.000 0.268 0.000 0.216 0.516
#> GSM48656 5 0.3074 0.701 0.000 0.196 0.000 0.000 0.804
#> GSM48657 5 0.4713 0.247 0.000 0.440 0.000 0.016 0.544
#> GSM48658 5 0.2783 0.749 0.012 0.116 0.004 0.000 0.868
#> GSM25624 5 0.3055 0.758 0.072 0.064 0.000 0.000 0.864
#> GSM25625 3 0.0000 0.979 0.000 0.000 1.000 0.000 0.000
#> GSM25626 3 0.0000 0.979 0.000 0.000 1.000 0.000 0.000
#> GSM25627 5 0.2844 0.756 0.028 0.092 0.004 0.000 0.876
#> GSM25628 3 0.0000 0.979 0.000 0.000 1.000 0.000 0.000
#> GSM25629 5 0.4108 0.696 0.068 0.012 0.116 0.000 0.804
#> GSM25630 3 0.0000 0.979 0.000 0.000 1.000 0.000 0.000
#> GSM25631 5 0.3163 0.703 0.164 0.012 0.000 0.000 0.824
#> GSM25632 3 0.0000 0.979 0.000 0.000 1.000 0.000 0.000
#> GSM25633 5 0.4655 0.134 0.476 0.012 0.000 0.000 0.512
#> GSM25634 5 0.4339 0.489 0.336 0.012 0.000 0.000 0.652
#> GSM25635 5 0.3686 0.670 0.204 0.012 0.004 0.000 0.780
#> GSM25656 3 0.0000 0.979 0.000 0.000 1.000 0.000 0.000
#> GSM25657 1 0.1942 0.787 0.920 0.012 0.000 0.000 0.068
#> GSM25658 1 0.1012 0.795 0.968 0.012 0.000 0.000 0.020
#> GSM25659 1 0.0671 0.797 0.980 0.004 0.000 0.000 0.016
#> GSM25660 1 0.2006 0.785 0.916 0.012 0.000 0.000 0.072
#> GSM25661 1 0.0798 0.796 0.976 0.008 0.000 0.000 0.016
#> GSM25662 4 0.0000 0.937 0.000 0.000 0.000 1.000 0.000
#> GSM25663 3 0.0451 0.971 0.000 0.000 0.988 0.008 0.004
#> GSM25680 4 0.0162 0.937 0.000 0.000 0.000 0.996 0.004
#> GSM25681 4 0.0162 0.937 0.000 0.000 0.000 0.996 0.004
#> GSM25682 4 0.0000 0.937 0.000 0.000 0.000 1.000 0.000
#> GSM25683 4 0.0000 0.937 0.000 0.000 0.000 1.000 0.000
#> GSM25684 4 0.0000 0.937 0.000 0.000 0.000 1.000 0.000
#> GSM25685 4 0.0000 0.937 0.000 0.000 0.000 1.000 0.000
#> GSM25686 4 0.0000 0.937 0.000 0.000 0.000 1.000 0.000
#> GSM25687 4 0.0000 0.937 0.000 0.000 0.000 1.000 0.000
#> GSM48664 1 0.3359 0.741 0.840 0.052 0.000 0.000 0.108
#> GSM48665 1 0.2853 0.761 0.876 0.052 0.000 0.000 0.072
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM25548 5 0.0000 0.9332 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM25549 5 0.0146 0.9299 0.000 0.004 0.000 0.000 0.996 0.000
#> GSM25550 5 0.0146 0.9299 0.000 0.004 0.000 0.000 0.996 0.000
#> GSM25551 5 0.0000 0.9332 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM25570 5 0.0000 0.9332 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM25571 5 0.0000 0.9332 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM25358 3 0.0713 0.9347 0.000 0.000 0.972 0.028 0.000 0.000
#> GSM25359 3 0.0458 0.9421 0.000 0.000 0.984 0.016 0.000 0.000
#> GSM25360 3 0.0146 0.9448 0.000 0.000 0.996 0.000 0.000 0.004
#> GSM25361 3 0.1452 0.9280 0.000 0.020 0.948 0.020 0.000 0.012
#> GSM25377 4 0.3554 0.5948 0.040 0.112 0.000 0.820 0.000 0.028
#> GSM25378 4 0.4075 0.4614 0.004 0.004 0.000 0.668 0.312 0.012
#> GSM25401 4 0.2704 0.6078 0.012 0.100 0.000 0.868 0.000 0.020
#> GSM25402 4 0.2143 0.6213 0.016 0.008 0.000 0.916 0.048 0.012
#> GSM25349 2 0.1251 0.8599 0.000 0.956 0.000 0.024 0.012 0.008
#> GSM25350 2 0.1138 0.8589 0.000 0.960 0.000 0.024 0.012 0.004
#> GSM25356 5 0.3854 0.0239 0.000 0.000 0.000 0.464 0.536 0.000
#> GSM25357 5 0.0260 0.9275 0.000 0.000 0.000 0.008 0.992 0.000
#> GSM25385 3 0.0000 0.9449 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM25386 3 0.0000 0.9449 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM25399 4 0.4832 0.4545 0.228 0.060 0.000 0.684 0.000 0.028
#> GSM25400 4 0.2967 0.5433 0.136 0.004 0.008 0.840 0.000 0.012
#> GSM48659 5 0.5270 0.0546 0.000 0.404 0.000 0.000 0.496 0.100
#> GSM48660 2 0.1858 0.8674 0.000 0.912 0.000 0.000 0.012 0.076
#> GSM25409 2 0.3659 0.7214 0.000 0.780 0.000 0.180 0.012 0.028
#> GSM25410 3 0.0000 0.9449 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM25426 5 0.0260 0.9275 0.000 0.000 0.000 0.008 0.992 0.000
#> GSM25427 4 0.3159 0.6137 0.004 0.004 0.000 0.812 0.168 0.012
#> GSM25540 3 0.0909 0.9372 0.000 0.000 0.968 0.020 0.000 0.012
#> GSM25541 3 0.3434 0.8169 0.136 0.008 0.820 0.024 0.000 0.012
#> GSM25542 3 0.0993 0.9361 0.000 0.000 0.964 0.024 0.000 0.012
#> GSM25543 3 0.1700 0.9215 0.000 0.028 0.936 0.024 0.000 0.012
#> GSM25479 1 0.2755 0.6957 0.844 0.004 0.000 0.140 0.000 0.012
#> GSM25480 1 0.2755 0.6957 0.844 0.004 0.000 0.140 0.000 0.012
#> GSM25481 4 0.5684 0.4112 0.000 0.148 0.000 0.536 0.308 0.008
#> GSM25482 4 0.5529 0.3612 0.000 0.148 0.000 0.516 0.336 0.000
#> GSM48654 6 0.1411 0.7628 0.000 0.060 0.000 0.004 0.000 0.936
#> GSM48650 2 0.1858 0.8674 0.000 0.912 0.000 0.000 0.012 0.076
#> GSM48651 2 0.1858 0.8674 0.000 0.912 0.000 0.000 0.012 0.076
#> GSM48652 2 0.1858 0.8674 0.000 0.912 0.000 0.000 0.012 0.076
#> GSM48653 2 0.1858 0.8674 0.000 0.912 0.000 0.000 0.012 0.076
#> GSM48662 2 0.1858 0.8674 0.000 0.912 0.000 0.000 0.012 0.076
#> GSM48663 2 0.1858 0.8674 0.000 0.912 0.000 0.000 0.012 0.076
#> GSM25524 3 0.5038 0.4122 0.316 0.004 0.596 0.084 0.000 0.000
#> GSM25525 1 0.3846 0.6595 0.784 0.008 0.000 0.164 0.032 0.012
#> GSM25526 3 0.3129 0.7962 0.152 0.000 0.820 0.024 0.000 0.004
#> GSM25527 1 0.0622 0.7529 0.980 0.000 0.000 0.012 0.000 0.008
#> GSM25528 3 0.2909 0.8159 0.136 0.000 0.836 0.028 0.000 0.000
#> GSM25529 1 0.0777 0.7505 0.972 0.000 0.000 0.024 0.000 0.004
#> GSM25530 3 0.0891 0.9332 0.008 0.000 0.968 0.024 0.000 0.000
#> GSM25531 1 0.4379 0.2228 0.576 0.000 0.396 0.028 0.000 0.000
#> GSM48661 6 0.1464 0.7631 0.000 0.036 0.016 0.004 0.000 0.944
#> GSM25561 3 0.0000 0.9449 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM25562 4 0.5864 0.0585 0.408 0.092 0.000 0.468 0.000 0.032
#> GSM25563 3 0.0000 0.9449 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM25564 2 0.2278 0.8549 0.000 0.904 0.000 0.032 0.012 0.052
#> GSM25565 2 0.2313 0.8203 0.000 0.884 0.000 0.100 0.012 0.004
#> GSM25566 2 0.2752 0.8037 0.000 0.864 0.000 0.104 0.012 0.020
#> GSM25568 6 0.3999 0.0414 0.000 0.496 0.000 0.004 0.000 0.500
#> GSM25569 2 0.1913 0.8641 0.000 0.908 0.000 0.000 0.012 0.080
#> GSM25552 2 0.4405 0.6160 0.000 0.696 0.000 0.252 0.024 0.028
#> GSM25553 2 0.4383 0.6192 0.004 0.700 0.000 0.252 0.016 0.028
#> GSM25578 1 0.0146 0.7560 0.996 0.000 0.000 0.000 0.000 0.004
#> GSM25579 1 0.0653 0.7551 0.980 0.004 0.000 0.012 0.000 0.004
#> GSM25580 1 0.0146 0.7560 0.996 0.000 0.000 0.000 0.000 0.004
#> GSM25581 1 0.0291 0.7554 0.992 0.000 0.000 0.004 0.000 0.004
#> GSM48655 6 0.4895 0.5464 0.000 0.228 0.000 0.000 0.124 0.648
#> GSM48656 6 0.1285 0.7645 0.000 0.052 0.000 0.004 0.000 0.944
#> GSM48657 6 0.3592 0.4608 0.000 0.344 0.000 0.000 0.000 0.656
#> GSM48658 6 0.1562 0.7675 0.024 0.032 0.000 0.004 0.000 0.940
#> GSM25624 6 0.1606 0.7533 0.056 0.008 0.000 0.004 0.000 0.932
#> GSM25625 3 0.0260 0.9445 0.000 0.000 0.992 0.000 0.000 0.008
#> GSM25626 3 0.0146 0.9448 0.000 0.000 0.996 0.000 0.000 0.004
#> GSM25627 6 0.1462 0.7603 0.056 0.008 0.000 0.000 0.000 0.936
#> GSM25628 3 0.0146 0.9448 0.000 0.000 0.996 0.000 0.000 0.004
#> GSM25629 6 0.3776 0.6348 0.196 0.000 0.048 0.000 0.000 0.756
#> GSM25630 3 0.0146 0.9448 0.000 0.000 0.996 0.000 0.000 0.004
#> GSM25631 6 0.3101 0.6162 0.244 0.000 0.000 0.000 0.000 0.756
#> GSM25632 3 0.0000 0.9449 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM25633 1 0.3023 0.5571 0.768 0.000 0.000 0.000 0.000 0.232
#> GSM25634 1 0.3695 0.2646 0.624 0.000 0.000 0.000 0.000 0.376
#> GSM25635 6 0.3782 0.3152 0.412 0.000 0.000 0.000 0.000 0.588
#> GSM25656 3 0.0622 0.9411 0.000 0.000 0.980 0.008 0.000 0.012
#> GSM25657 1 0.1334 0.7430 0.948 0.000 0.000 0.032 0.000 0.020
#> GSM25658 1 0.4392 0.1375 0.504 0.004 0.000 0.476 0.000 0.016
#> GSM25659 1 0.4150 0.5775 0.720 0.016 0.000 0.236 0.000 0.028
#> GSM25660 1 0.0146 0.7560 0.996 0.000 0.000 0.000 0.000 0.004
#> GSM25661 1 0.4778 0.4501 0.652 0.036 0.000 0.284 0.000 0.028
#> GSM25662 5 0.0000 0.9332 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM25663 3 0.2137 0.9152 0.004 0.020 0.924 0.020 0.020 0.012
#> GSM25680 5 0.0000 0.9332 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM25681 5 0.0000 0.9332 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM25682 5 0.0000 0.9332 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM25683 5 0.0000 0.9332 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM25684 5 0.0000 0.9332 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM25685 5 0.0000 0.9332 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM25686 5 0.0000 0.9332 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM25687 5 0.0000 0.9332 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM48664 4 0.5662 0.0110 0.428 0.076 0.000 0.468 0.000 0.028
#> GSM48665 1 0.5705 -0.0584 0.456 0.080 0.000 0.436 0.000 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 genotype/variation(p) k
#> ATC:skmeans 96 1.32e-04 2
#> ATC:skmeans 98 9.99e-06 3
#> ATC:skmeans 93 8.62e-11 4
#> ATC:skmeans 90 5.14e-14 5
#> ATC:skmeans 83 1.99e-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", "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 100 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 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 0.802 0.875 0.950 0.4887 0.515 0.515
#> 3 3 0.754 0.824 0.929 0.1766 0.916 0.837
#> 4 4 0.593 0.690 0.848 0.1960 0.858 0.685
#> 5 5 0.831 0.828 0.919 0.1225 0.818 0.508
#> 6 6 0.817 0.760 0.887 0.0443 0.954 0.802
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
#> GSM25548 2 0.0000 0.9605 0.000 1.000
#> GSM25549 2 0.0000 0.9605 0.000 1.000
#> GSM25550 2 0.0000 0.9605 0.000 1.000
#> GSM25551 2 0.0000 0.9605 0.000 1.000
#> GSM25570 2 0.0000 0.9605 0.000 1.000
#> GSM25571 2 0.0000 0.9605 0.000 1.000
#> GSM25358 1 0.0000 0.9349 1.000 0.000
#> GSM25359 1 0.0000 0.9349 1.000 0.000
#> GSM25360 1 0.0000 0.9349 1.000 0.000
#> GSM25361 1 0.0000 0.9349 1.000 0.000
#> GSM25377 1 0.5408 0.8267 0.876 0.124
#> GSM25378 2 0.0000 0.9605 0.000 1.000
#> GSM25401 1 0.8608 0.6160 0.716 0.284
#> GSM25402 1 0.4690 0.8506 0.900 0.100
#> GSM25349 2 0.0000 0.9605 0.000 1.000
#> GSM25350 2 0.9970 0.0406 0.468 0.532
#> GSM25356 2 0.0000 0.9605 0.000 1.000
#> GSM25357 2 0.0000 0.9605 0.000 1.000
#> GSM25385 1 0.0000 0.9349 1.000 0.000
#> GSM25386 1 0.0000 0.9349 1.000 0.000
#> GSM25399 1 0.0000 0.9349 1.000 0.000
#> GSM25400 1 0.0000 0.9349 1.000 0.000
#> GSM48659 2 0.0000 0.9605 0.000 1.000
#> GSM48660 2 0.0000 0.9605 0.000 1.000
#> GSM25409 1 0.9866 0.2881 0.568 0.432
#> GSM25410 1 0.0000 0.9349 1.000 0.000
#> GSM25426 2 0.0000 0.9605 0.000 1.000
#> GSM25427 2 0.0000 0.9605 0.000 1.000
#> GSM25540 1 0.0000 0.9349 1.000 0.000
#> GSM25541 1 0.0000 0.9349 1.000 0.000
#> GSM25542 1 0.0000 0.9349 1.000 0.000
#> GSM25543 1 0.0000 0.9349 1.000 0.000
#> GSM25479 1 0.9909 0.2163 0.556 0.444
#> GSM25480 1 0.9795 0.3280 0.584 0.416
#> GSM25481 2 0.0000 0.9605 0.000 1.000
#> GSM25482 2 0.0000 0.9605 0.000 1.000
#> GSM48654 1 0.9044 0.5533 0.680 0.320
#> GSM48650 2 0.0000 0.9605 0.000 1.000
#> GSM48651 2 0.0000 0.9605 0.000 1.000
#> GSM48652 2 0.0000 0.9605 0.000 1.000
#> GSM48653 2 0.2948 0.9144 0.052 0.948
#> GSM48662 2 0.4161 0.8814 0.084 0.916
#> GSM48663 2 0.6801 0.7574 0.180 0.820
#> GSM25524 1 0.0000 0.9349 1.000 0.000
#> GSM25525 2 0.0376 0.9573 0.004 0.996
#> GSM25526 1 0.0000 0.9349 1.000 0.000
#> GSM25527 1 0.0000 0.9349 1.000 0.000
#> GSM25528 1 0.0000 0.9349 1.000 0.000
#> GSM25529 1 0.0000 0.9349 1.000 0.000
#> GSM25530 1 0.0000 0.9349 1.000 0.000
#> GSM25531 1 0.0000 0.9349 1.000 0.000
#> GSM48661 1 0.0376 0.9322 0.996 0.004
#> GSM25561 1 0.0000 0.9349 1.000 0.000
#> GSM25562 1 0.0000 0.9349 1.000 0.000
#> GSM25563 1 0.0000 0.9349 1.000 0.000
#> GSM25564 2 0.0000 0.9605 0.000 1.000
#> GSM25565 1 0.9170 0.5313 0.668 0.332
#> GSM25566 2 0.6148 0.7976 0.152 0.848
#> GSM25568 2 0.0000 0.9605 0.000 1.000
#> GSM25569 2 0.0000 0.9605 0.000 1.000
#> GSM25552 2 0.9922 0.1176 0.448 0.552
#> GSM25553 1 0.9850 0.2996 0.572 0.428
#> GSM25578 1 0.0000 0.9349 1.000 0.000
#> GSM25579 1 0.0000 0.9349 1.000 0.000
#> GSM25580 1 0.0000 0.9349 1.000 0.000
#> GSM25581 1 0.0000 0.9349 1.000 0.000
#> GSM48655 2 0.0000 0.9605 0.000 1.000
#> GSM48656 1 0.0938 0.9265 0.988 0.012
#> GSM48657 2 0.0000 0.9605 0.000 1.000
#> GSM48658 1 0.0000 0.9349 1.000 0.000
#> GSM25624 2 0.0376 0.9573 0.004 0.996
#> GSM25625 1 0.0000 0.9349 1.000 0.000
#> GSM25626 1 0.0000 0.9349 1.000 0.000
#> GSM25627 1 0.0000 0.9349 1.000 0.000
#> GSM25628 1 0.0000 0.9349 1.000 0.000
#> GSM25629 1 0.0000 0.9349 1.000 0.000
#> GSM25630 1 0.0000 0.9349 1.000 0.000
#> GSM25631 1 0.0000 0.9349 1.000 0.000
#> GSM25632 1 0.0000 0.9349 1.000 0.000
#> GSM25633 1 0.0000 0.9349 1.000 0.000
#> GSM25634 1 0.0000 0.9349 1.000 0.000
#> GSM25635 1 0.0000 0.9349 1.000 0.000
#> GSM25656 1 0.0000 0.9349 1.000 0.000
#> GSM25657 1 0.0000 0.9349 1.000 0.000
#> GSM25658 1 0.0000 0.9349 1.000 0.000
#> GSM25659 1 0.9170 0.5308 0.668 0.332
#> GSM25660 1 0.0000 0.9349 1.000 0.000
#> GSM25661 1 0.0000 0.9349 1.000 0.000
#> GSM25662 1 0.9170 0.5313 0.668 0.332
#> GSM25663 1 0.0376 0.9322 0.996 0.004
#> GSM25680 2 0.0000 0.9605 0.000 1.000
#> GSM25681 2 0.0000 0.9605 0.000 1.000
#> GSM25682 2 0.0000 0.9605 0.000 1.000
#> GSM25683 2 0.0000 0.9605 0.000 1.000
#> GSM25684 2 0.0000 0.9605 0.000 1.000
#> GSM25685 2 0.0000 0.9605 0.000 1.000
#> GSM25686 2 0.0000 0.9605 0.000 1.000
#> GSM25687 2 0.0000 0.9605 0.000 1.000
#> GSM48664 1 0.0000 0.9349 1.000 0.000
#> GSM48665 1 0.1633 0.9175 0.976 0.024
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM25548 2 0.0000 0.950 0.000 1.000 0.000
#> GSM25549 2 0.0000 0.950 0.000 1.000 0.000
#> GSM25550 2 0.0000 0.950 0.000 1.000 0.000
#> GSM25551 2 0.0000 0.950 0.000 1.000 0.000
#> GSM25570 2 0.0000 0.950 0.000 1.000 0.000
#> GSM25571 2 0.0000 0.950 0.000 1.000 0.000
#> GSM25358 1 0.0237 0.870 0.996 0.000 0.004
#> GSM25359 1 0.0237 0.870 0.996 0.000 0.004
#> GSM25360 3 0.5948 0.308 0.360 0.000 0.640
#> GSM25361 1 0.1031 0.863 0.976 0.000 0.024
#> GSM25377 1 0.3267 0.786 0.884 0.116 0.000
#> GSM25378 2 0.0000 0.950 0.000 1.000 0.000
#> GSM25401 1 0.5754 0.591 0.700 0.296 0.004
#> GSM25402 1 0.3193 0.801 0.896 0.100 0.004
#> GSM25349 2 0.0000 0.950 0.000 1.000 0.000
#> GSM25350 2 0.6442 0.107 0.432 0.564 0.004
#> GSM25356 2 0.0000 0.950 0.000 1.000 0.000
#> GSM25357 2 0.0000 0.950 0.000 1.000 0.000
#> GSM25385 1 0.5882 0.498 0.652 0.000 0.348
#> GSM25386 3 0.0000 0.938 0.000 0.000 1.000
#> GSM25399 1 0.0000 0.870 1.000 0.000 0.000
#> GSM25400 1 0.0237 0.870 0.996 0.000 0.004
#> GSM48659 2 0.0000 0.950 0.000 1.000 0.000
#> GSM48660 2 0.0000 0.950 0.000 1.000 0.000
#> GSM25409 1 0.6505 0.226 0.528 0.468 0.004
#> GSM25410 3 0.0000 0.938 0.000 0.000 1.000
#> GSM25426 2 0.0000 0.950 0.000 1.000 0.000
#> GSM25427 2 0.0000 0.950 0.000 1.000 0.000
#> GSM25540 1 0.1031 0.862 0.976 0.000 0.024
#> GSM25541 1 0.0237 0.870 0.996 0.000 0.004
#> GSM25542 1 0.2590 0.832 0.924 0.004 0.072
#> GSM25543 1 0.0237 0.870 0.996 0.000 0.004
#> GSM25479 1 0.6154 0.327 0.592 0.408 0.000
#> GSM25480 1 0.6026 0.439 0.624 0.376 0.000
#> GSM25481 2 0.0000 0.950 0.000 1.000 0.000
#> GSM25482 2 0.0000 0.950 0.000 1.000 0.000
#> GSM48654 1 0.6298 0.440 0.608 0.388 0.004
#> GSM48650 2 0.0000 0.950 0.000 1.000 0.000
#> GSM48651 2 0.0000 0.950 0.000 1.000 0.000
#> GSM48652 2 0.0000 0.950 0.000 1.000 0.000
#> GSM48653 2 0.1860 0.899 0.052 0.948 0.000
#> GSM48662 2 0.2772 0.863 0.080 0.916 0.004
#> GSM48663 2 0.4465 0.732 0.176 0.820 0.004
#> GSM25524 1 0.0000 0.870 1.000 0.000 0.000
#> GSM25525 2 0.1643 0.909 0.044 0.956 0.000
#> GSM25526 1 0.0747 0.865 0.984 0.000 0.016
#> GSM25527 1 0.0000 0.870 1.000 0.000 0.000
#> GSM25528 1 0.0892 0.863 0.980 0.000 0.020
#> GSM25529 1 0.0000 0.870 1.000 0.000 0.000
#> GSM25530 1 0.5216 0.639 0.740 0.000 0.260
#> GSM25531 1 0.0000 0.870 1.000 0.000 0.000
#> GSM48661 1 0.3461 0.820 0.900 0.024 0.076
#> GSM25561 1 0.2625 0.823 0.916 0.000 0.084
#> GSM25562 1 0.0237 0.870 0.996 0.000 0.004
#> GSM25563 1 0.5560 0.581 0.700 0.000 0.300
#> GSM25564 2 0.0000 0.950 0.000 1.000 0.000
#> GSM25565 1 0.6330 0.423 0.600 0.396 0.004
#> GSM25566 2 0.3879 0.772 0.152 0.848 0.000
#> GSM25568 2 0.0000 0.950 0.000 1.000 0.000
#> GSM25569 2 0.0000 0.950 0.000 1.000 0.000
#> GSM25552 2 0.6398 0.167 0.416 0.580 0.004
#> GSM25553 1 0.6500 0.238 0.532 0.464 0.004
#> GSM25578 1 0.0000 0.870 1.000 0.000 0.000
#> GSM25579 1 0.0000 0.870 1.000 0.000 0.000
#> GSM25580 1 0.0000 0.870 1.000 0.000 0.000
#> GSM25581 1 0.0000 0.870 1.000 0.000 0.000
#> GSM48655 2 0.0000 0.950 0.000 1.000 0.000
#> GSM48656 1 0.0661 0.868 0.988 0.008 0.004
#> GSM48657 2 0.0000 0.950 0.000 1.000 0.000
#> GSM48658 1 0.0237 0.870 0.996 0.000 0.004
#> GSM25624 2 0.1860 0.899 0.052 0.948 0.000
#> GSM25625 3 0.0000 0.938 0.000 0.000 1.000
#> GSM25626 3 0.0000 0.938 0.000 0.000 1.000
#> GSM25627 1 0.0000 0.870 1.000 0.000 0.000
#> GSM25628 3 0.0000 0.938 0.000 0.000 1.000
#> GSM25629 1 0.0237 0.870 0.996 0.000 0.004
#> GSM25630 3 0.0237 0.936 0.004 0.000 0.996
#> GSM25631 1 0.0237 0.870 0.996 0.000 0.004
#> GSM25632 3 0.0237 0.936 0.004 0.000 0.996
#> GSM25633 1 0.0000 0.870 1.000 0.000 0.000
#> GSM25634 1 0.0000 0.870 1.000 0.000 0.000
#> GSM25635 1 0.0000 0.870 1.000 0.000 0.000
#> GSM25656 1 0.3551 0.787 0.868 0.000 0.132
#> GSM25657 1 0.0000 0.870 1.000 0.000 0.000
#> GSM25658 1 0.0000 0.870 1.000 0.000 0.000
#> GSM25659 1 0.6057 0.529 0.656 0.340 0.004
#> GSM25660 1 0.0000 0.870 1.000 0.000 0.000
#> GSM25661 1 0.0000 0.870 1.000 0.000 0.000
#> GSM25662 1 0.6330 0.423 0.600 0.396 0.004
#> GSM25663 1 0.1399 0.856 0.968 0.028 0.004
#> GSM25680 2 0.0000 0.950 0.000 1.000 0.000
#> GSM25681 2 0.0000 0.950 0.000 1.000 0.000
#> GSM25682 2 0.0000 0.950 0.000 1.000 0.000
#> GSM25683 2 0.0000 0.950 0.000 1.000 0.000
#> GSM25684 2 0.0000 0.950 0.000 1.000 0.000
#> GSM25685 2 0.0000 0.950 0.000 1.000 0.000
#> GSM25686 2 0.0000 0.950 0.000 1.000 0.000
#> GSM25687 2 0.0000 0.950 0.000 1.000 0.000
#> GSM48664 1 0.0000 0.870 1.000 0.000 0.000
#> GSM48665 1 0.0892 0.862 0.980 0.020 0.000
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM25548 2 0.0000 0.83990 0.000 1.000 0.000 0.000
#> GSM25549 2 0.0000 0.83990 0.000 1.000 0.000 0.000
#> GSM25550 2 0.0000 0.83990 0.000 1.000 0.000 0.000
#> GSM25551 2 0.0000 0.83990 0.000 1.000 0.000 0.000
#> GSM25570 2 0.0000 0.83990 0.000 1.000 0.000 0.000
#> GSM25571 2 0.0000 0.83990 0.000 1.000 0.000 0.000
#> GSM25358 1 0.3428 0.79191 0.844 0.012 0.000 0.144
#> GSM25359 1 0.2973 0.79498 0.856 0.000 0.000 0.144
#> GSM25360 3 0.4673 0.65977 0.132 0.000 0.792 0.076
#> GSM25361 1 0.4677 0.75682 0.768 0.000 0.040 0.192
#> GSM25377 1 0.5847 0.18654 0.560 0.036 0.000 0.404
#> GSM25378 2 0.0000 0.83990 0.000 1.000 0.000 0.000
#> GSM25401 4 0.7250 0.10074 0.336 0.160 0.000 0.504
#> GSM25402 1 0.5999 0.45730 0.552 0.044 0.000 0.404
#> GSM25349 4 0.3024 0.74004 0.000 0.148 0.000 0.852
#> GSM25350 4 0.1978 0.71209 0.004 0.068 0.000 0.928
#> GSM25356 2 0.0188 0.83697 0.000 0.996 0.000 0.004
#> GSM25357 2 0.0000 0.83990 0.000 1.000 0.000 0.000
#> GSM25385 1 0.6702 0.45841 0.544 0.000 0.356 0.100
#> GSM25386 3 0.0000 0.95637 0.000 0.000 1.000 0.000
#> GSM25399 1 0.0000 0.81927 1.000 0.000 0.000 0.000
#> GSM25400 1 0.2973 0.79498 0.856 0.000 0.000 0.144
#> GSM48659 2 0.0921 0.81574 0.000 0.972 0.000 0.028
#> GSM48660 4 0.3975 0.64006 0.000 0.240 0.000 0.760
#> GSM25409 4 0.3471 0.68616 0.060 0.072 0.000 0.868
#> GSM25410 3 0.0000 0.95637 0.000 0.000 1.000 0.000
#> GSM25426 2 0.4605 0.29520 0.000 0.664 0.000 0.336
#> GSM25427 2 0.1716 0.77893 0.000 0.936 0.000 0.064
#> GSM25540 1 0.3658 0.79149 0.836 0.000 0.020 0.144
#> GSM25541 1 0.2973 0.79498 0.856 0.000 0.000 0.144
#> GSM25542 1 0.6182 0.39883 0.520 0.000 0.052 0.428
#> GSM25543 1 0.4925 0.45748 0.572 0.000 0.000 0.428
#> GSM25479 1 0.3837 0.59224 0.776 0.224 0.000 0.000
#> GSM25480 1 0.4182 0.64686 0.796 0.180 0.000 0.024
#> GSM25481 2 0.0000 0.83990 0.000 1.000 0.000 0.000
#> GSM25482 2 0.0000 0.83990 0.000 1.000 0.000 0.000
#> GSM48654 2 0.7667 -0.03293 0.224 0.440 0.000 0.336
#> GSM48650 4 0.2973 0.74001 0.000 0.144 0.000 0.856
#> GSM48651 4 0.2973 0.74001 0.000 0.144 0.000 0.856
#> GSM48652 4 0.3123 0.73636 0.000 0.156 0.000 0.844
#> GSM48653 4 0.2868 0.74184 0.000 0.136 0.000 0.864
#> GSM48662 4 0.2589 0.73847 0.000 0.116 0.000 0.884
#> GSM48663 4 0.1867 0.71559 0.000 0.072 0.000 0.928
#> GSM25524 1 0.0336 0.81954 0.992 0.000 0.000 0.008
#> GSM25525 2 0.4250 0.51153 0.276 0.724 0.000 0.000
#> GSM25526 1 0.2197 0.81536 0.928 0.000 0.024 0.048
#> GSM25527 1 0.0000 0.81927 1.000 0.000 0.000 0.000
#> GSM25528 1 0.1733 0.81621 0.948 0.000 0.028 0.024
#> GSM25529 1 0.0000 0.81927 1.000 0.000 0.000 0.000
#> GSM25530 1 0.4164 0.63340 0.736 0.000 0.264 0.000
#> GSM25531 1 0.0000 0.81927 1.000 0.000 0.000 0.000
#> GSM48661 1 0.8335 0.52436 0.552 0.212 0.092 0.144
#> GSM25561 1 0.4114 0.77751 0.828 0.000 0.112 0.060
#> GSM25562 1 0.4830 0.47642 0.608 0.000 0.000 0.392
#> GSM25563 1 0.6708 0.57336 0.592 0.000 0.280 0.128
#> GSM25564 2 0.0592 0.82761 0.000 0.984 0.000 0.016
#> GSM25565 4 0.7439 0.20999 0.296 0.204 0.000 0.500
#> GSM25566 2 0.6130 0.00966 0.052 0.548 0.000 0.400
#> GSM25568 2 0.5000 -0.17700 0.000 0.500 0.000 0.500
#> GSM25569 4 0.3486 0.71925 0.000 0.188 0.000 0.812
#> GSM25552 4 0.7466 0.28004 0.176 0.388 0.000 0.436
#> GSM25553 4 0.7731 0.35814 0.248 0.316 0.000 0.436
#> GSM25578 1 0.0000 0.81927 1.000 0.000 0.000 0.000
#> GSM25579 1 0.0469 0.82018 0.988 0.000 0.000 0.012
#> GSM25580 1 0.0000 0.81927 1.000 0.000 0.000 0.000
#> GSM25581 1 0.0000 0.81927 1.000 0.000 0.000 0.000
#> GSM48655 2 0.0000 0.83990 0.000 1.000 0.000 0.000
#> GSM48656 1 0.4920 0.59602 0.628 0.004 0.000 0.368
#> GSM48657 2 0.5000 -0.08312 0.000 0.504 0.000 0.496
#> GSM48658 1 0.2973 0.79498 0.856 0.000 0.000 0.144
#> GSM25624 2 0.4679 0.38868 0.352 0.648 0.000 0.000
#> GSM25625 3 0.0000 0.95637 0.000 0.000 1.000 0.000
#> GSM25626 3 0.0000 0.95637 0.000 0.000 1.000 0.000
#> GSM25627 1 0.1302 0.81753 0.956 0.000 0.000 0.044
#> GSM25628 3 0.0000 0.95637 0.000 0.000 1.000 0.000
#> GSM25629 1 0.2973 0.79498 0.856 0.000 0.000 0.144
#> GSM25630 3 0.0000 0.95637 0.000 0.000 1.000 0.000
#> GSM25631 1 0.5063 0.73737 0.768 0.108 0.000 0.124
#> GSM25632 3 0.0000 0.95637 0.000 0.000 1.000 0.000
#> GSM25633 1 0.0000 0.81927 1.000 0.000 0.000 0.000
#> GSM25634 1 0.0000 0.81927 1.000 0.000 0.000 0.000
#> GSM25635 1 0.0000 0.81927 1.000 0.000 0.000 0.000
#> GSM25656 1 0.6945 0.51852 0.552 0.000 0.312 0.136
#> GSM25657 1 0.0000 0.81927 1.000 0.000 0.000 0.000
#> GSM25658 1 0.1940 0.81373 0.924 0.000 0.000 0.076
#> GSM25659 1 0.7137 0.33436 0.536 0.304 0.000 0.160
#> GSM25660 1 0.0000 0.81927 1.000 0.000 0.000 0.000
#> GSM25661 1 0.0000 0.81927 1.000 0.000 0.000 0.000
#> GSM25662 2 0.6955 0.17085 0.296 0.560 0.000 0.144
#> GSM25663 1 0.6295 0.62614 0.660 0.196 0.000 0.144
#> GSM25680 2 0.0000 0.83990 0.000 1.000 0.000 0.000
#> GSM25681 2 0.0000 0.83990 0.000 1.000 0.000 0.000
#> GSM25682 2 0.0000 0.83990 0.000 1.000 0.000 0.000
#> GSM25683 2 0.0000 0.83990 0.000 1.000 0.000 0.000
#> GSM25684 2 0.0000 0.83990 0.000 1.000 0.000 0.000
#> GSM25685 2 0.0000 0.83990 0.000 1.000 0.000 0.000
#> GSM25686 2 0.0000 0.83990 0.000 1.000 0.000 0.000
#> GSM25687 2 0.0000 0.83990 0.000 1.000 0.000 0.000
#> GSM48664 1 0.0188 0.81839 0.996 0.000 0.000 0.004
#> GSM48665 1 0.2125 0.78114 0.920 0.004 0.000 0.076
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM25548 4 0.0000 0.9550 0.000 0.000 0.000 1.000 0.000
#> GSM25549 4 0.0000 0.9550 0.000 0.000 0.000 1.000 0.000
#> GSM25550 4 0.0000 0.9550 0.000 0.000 0.000 1.000 0.000
#> GSM25551 4 0.0000 0.9550 0.000 0.000 0.000 1.000 0.000
#> GSM25570 4 0.0000 0.9550 0.000 0.000 0.000 1.000 0.000
#> GSM25571 4 0.0000 0.9550 0.000 0.000 0.000 1.000 0.000
#> GSM25358 5 0.1544 0.8754 0.068 0.000 0.000 0.000 0.932
#> GSM25359 5 0.1544 0.8754 0.068 0.000 0.000 0.000 0.932
#> GSM25360 3 0.4088 0.3031 0.000 0.000 0.632 0.000 0.368
#> GSM25361 5 0.1544 0.8754 0.068 0.000 0.000 0.000 0.932
#> GSM25377 1 0.4297 0.7053 0.764 0.072 0.000 0.000 0.164
#> GSM25378 4 0.0000 0.9550 0.000 0.000 0.000 1.000 0.000
#> GSM25401 5 0.1544 0.8606 0.000 0.068 0.000 0.000 0.932
#> GSM25402 5 0.1800 0.8718 0.020 0.048 0.000 0.000 0.932
#> GSM25349 2 0.0404 0.9255 0.000 0.988 0.000 0.012 0.000
#> GSM25350 2 0.0404 0.9245 0.000 0.988 0.000 0.000 0.012
#> GSM25356 4 0.0000 0.9550 0.000 0.000 0.000 1.000 0.000
#> GSM25357 4 0.0000 0.9550 0.000 0.000 0.000 1.000 0.000
#> GSM25385 5 0.4755 0.6555 0.060 0.000 0.244 0.000 0.696
#> GSM25386 3 0.0000 0.9279 0.000 0.000 1.000 0.000 0.000
#> GSM25399 1 0.0609 0.8612 0.980 0.000 0.000 0.000 0.020
#> GSM25400 5 0.1544 0.8754 0.068 0.000 0.000 0.000 0.932
#> GSM48659 4 0.0290 0.9492 0.000 0.008 0.000 0.992 0.000
#> GSM48660 2 0.0000 0.9330 0.000 1.000 0.000 0.000 0.000
#> GSM25409 2 0.3336 0.6439 0.000 0.772 0.000 0.000 0.228
#> GSM25410 3 0.0000 0.9279 0.000 0.000 1.000 0.000 0.000
#> GSM25426 4 0.1544 0.8914 0.000 0.068 0.000 0.932 0.000
#> GSM25427 4 0.0162 0.9521 0.000 0.004 0.000 0.996 0.000
#> GSM25540 5 0.1544 0.8754 0.068 0.000 0.000 0.000 0.932
#> GSM25541 5 0.1608 0.8742 0.072 0.000 0.000 0.000 0.928
#> GSM25542 5 0.1697 0.8651 0.008 0.060 0.000 0.000 0.932
#> GSM25543 5 0.1697 0.8651 0.008 0.060 0.000 0.000 0.932
#> GSM25479 1 0.0000 0.8682 1.000 0.000 0.000 0.000 0.000
#> GSM25480 1 0.0290 0.8638 0.992 0.008 0.000 0.000 0.000
#> GSM25481 4 0.0000 0.9550 0.000 0.000 0.000 1.000 0.000
#> GSM25482 4 0.0000 0.9550 0.000 0.000 0.000 1.000 0.000
#> GSM48654 5 0.2179 0.7881 0.000 0.112 0.000 0.000 0.888
#> GSM48650 2 0.0000 0.9330 0.000 1.000 0.000 0.000 0.000
#> GSM48651 2 0.0000 0.9330 0.000 1.000 0.000 0.000 0.000
#> GSM48652 2 0.0162 0.9307 0.000 0.996 0.000 0.004 0.000
#> GSM48653 2 0.0000 0.9330 0.000 1.000 0.000 0.000 0.000
#> GSM48662 2 0.0000 0.9330 0.000 1.000 0.000 0.000 0.000
#> GSM48663 2 0.0000 0.9330 0.000 1.000 0.000 0.000 0.000
#> GSM25524 1 0.1608 0.8337 0.928 0.000 0.000 0.000 0.072
#> GSM25525 1 0.4161 0.3789 0.608 0.000 0.000 0.392 0.000
#> GSM25526 1 0.5005 0.5532 0.660 0.000 0.064 0.000 0.276
#> GSM25527 1 0.0000 0.8682 1.000 0.000 0.000 0.000 0.000
#> GSM25528 1 0.3608 0.7598 0.824 0.000 0.064 0.000 0.112
#> GSM25529 1 0.0000 0.8682 1.000 0.000 0.000 0.000 0.000
#> GSM25530 1 0.3961 0.6337 0.736 0.000 0.248 0.000 0.016
#> GSM25531 1 0.0000 0.8682 1.000 0.000 0.000 0.000 0.000
#> GSM48661 5 0.0000 0.8610 0.000 0.000 0.000 0.000 1.000
#> GSM25561 1 0.5449 0.3186 0.556 0.000 0.068 0.000 0.376
#> GSM25562 5 0.4035 0.7735 0.156 0.060 0.000 0.000 0.784
#> GSM25563 5 0.3868 0.7903 0.060 0.000 0.140 0.000 0.800
#> GSM25564 4 0.0000 0.9550 0.000 0.000 0.000 1.000 0.000
#> GSM25565 5 0.1544 0.8606 0.000 0.068 0.000 0.000 0.932
#> GSM25566 4 0.3994 0.7318 0.000 0.068 0.000 0.792 0.140
#> GSM25568 4 0.3336 0.6941 0.000 0.228 0.000 0.772 0.000
#> GSM25569 2 0.3177 0.6855 0.000 0.792 0.000 0.208 0.000
#> GSM25552 4 0.5542 0.0215 0.000 0.068 0.000 0.500 0.432
#> GSM25553 5 0.5513 0.2417 0.000 0.068 0.000 0.408 0.524
#> GSM25578 1 0.0000 0.8682 1.000 0.000 0.000 0.000 0.000
#> GSM25579 1 0.2773 0.7539 0.836 0.000 0.000 0.000 0.164
#> GSM25580 1 0.0000 0.8682 1.000 0.000 0.000 0.000 0.000
#> GSM25581 1 0.0000 0.8682 1.000 0.000 0.000 0.000 0.000
#> GSM48655 4 0.0510 0.9439 0.000 0.000 0.000 0.984 0.016
#> GSM48656 5 0.1908 0.8080 0.000 0.092 0.000 0.000 0.908
#> GSM48657 2 0.1544 0.8711 0.000 0.932 0.000 0.000 0.068
#> GSM48658 5 0.0000 0.8610 0.000 0.000 0.000 0.000 1.000
#> GSM25624 1 0.5086 0.5020 0.636 0.000 0.000 0.304 0.060
#> GSM25625 3 0.0000 0.9279 0.000 0.000 1.000 0.000 0.000
#> GSM25626 3 0.0000 0.9279 0.000 0.000 1.000 0.000 0.000
#> GSM25627 1 0.2852 0.7634 0.828 0.000 0.000 0.000 0.172
#> GSM25628 3 0.0000 0.9279 0.000 0.000 1.000 0.000 0.000
#> GSM25629 5 0.1478 0.8754 0.064 0.000 0.000 0.000 0.936
#> GSM25630 3 0.0000 0.9279 0.000 0.000 1.000 0.000 0.000
#> GSM25631 5 0.2648 0.8230 0.152 0.000 0.000 0.000 0.848
#> GSM25632 3 0.0000 0.9279 0.000 0.000 1.000 0.000 0.000
#> GSM25633 1 0.0000 0.8682 1.000 0.000 0.000 0.000 0.000
#> GSM25634 1 0.0000 0.8682 1.000 0.000 0.000 0.000 0.000
#> GSM25635 1 0.0000 0.8682 1.000 0.000 0.000 0.000 0.000
#> GSM25656 5 0.2900 0.8331 0.028 0.000 0.108 0.000 0.864
#> GSM25657 1 0.0000 0.8682 1.000 0.000 0.000 0.000 0.000
#> GSM25658 1 0.4256 0.2183 0.564 0.000 0.000 0.000 0.436
#> GSM25659 5 0.4985 0.3954 0.392 0.012 0.000 0.016 0.580
#> GSM25660 1 0.0000 0.8682 1.000 0.000 0.000 0.000 0.000
#> GSM25661 1 0.0000 0.8682 1.000 0.000 0.000 0.000 0.000
#> GSM25662 5 0.1544 0.8469 0.000 0.000 0.000 0.068 0.932
#> GSM25663 5 0.1818 0.8739 0.044 0.000 0.000 0.024 0.932
#> GSM25680 4 0.0000 0.9550 0.000 0.000 0.000 1.000 0.000
#> GSM25681 4 0.0000 0.9550 0.000 0.000 0.000 1.000 0.000
#> GSM25682 4 0.0000 0.9550 0.000 0.000 0.000 1.000 0.000
#> GSM25683 4 0.0000 0.9550 0.000 0.000 0.000 1.000 0.000
#> GSM25684 4 0.0000 0.9550 0.000 0.000 0.000 1.000 0.000
#> GSM25685 4 0.0000 0.9550 0.000 0.000 0.000 1.000 0.000
#> GSM25686 4 0.0000 0.9550 0.000 0.000 0.000 1.000 0.000
#> GSM25687 4 0.0000 0.9550 0.000 0.000 0.000 1.000 0.000
#> GSM48664 1 0.0162 0.8670 0.996 0.000 0.000 0.000 0.004
#> GSM48665 1 0.0290 0.8638 0.992 0.008 0.000 0.000 0.000
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM25548 5 0.0000 0.91249 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM25549 5 0.0000 0.91249 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM25550 5 0.0000 0.91249 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM25551 5 0.0000 0.91249 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM25570 5 0.0000 0.91249 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM25571 5 0.0000 0.91249 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM25358 4 0.1141 0.83916 0.052 0.000 0.000 0.948 0.000 0.000
#> GSM25359 4 0.0632 0.84469 0.024 0.000 0.000 0.976 0.000 0.000
#> GSM25360 3 0.3765 0.26132 0.000 0.000 0.596 0.404 0.000 0.000
#> GSM25361 4 0.0632 0.84469 0.024 0.000 0.000 0.976 0.000 0.000
#> GSM25377 1 0.7030 -0.09102 0.424 0.332 0.000 0.132 0.004 0.108
#> GSM25378 5 0.0000 0.91249 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM25401 4 0.5788 0.20450 0.020 0.324 0.000 0.544 0.004 0.108
#> GSM25402 4 0.1003 0.83100 0.028 0.004 0.000 0.964 0.000 0.004
#> GSM25349 2 0.2669 0.78744 0.000 0.864 0.000 0.024 0.004 0.108
#> GSM25350 2 0.2669 0.78744 0.000 0.864 0.000 0.024 0.004 0.108
#> GSM25356 5 0.0603 0.90095 0.000 0.004 0.000 0.000 0.980 0.016
#> GSM25357 5 0.0146 0.91191 0.000 0.000 0.000 0.000 0.996 0.004
#> GSM25385 4 0.2912 0.66080 0.000 0.000 0.216 0.784 0.000 0.000
#> GSM25386 3 0.0000 0.91898 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM25399 1 0.0632 0.85591 0.976 0.000 0.000 0.024 0.000 0.000
#> GSM25400 4 0.1141 0.83916 0.052 0.000 0.000 0.948 0.000 0.000
#> GSM48659 5 0.0692 0.89820 0.000 0.020 0.000 0.000 0.976 0.004
#> GSM48660 2 0.1663 0.74056 0.000 0.912 0.000 0.000 0.088 0.000
#> GSM25409 2 0.4457 0.67778 0.000 0.720 0.000 0.168 0.004 0.108
#> GSM25410 3 0.0000 0.91898 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM25426 5 0.5763 0.14989 0.000 0.324 0.000 0.024 0.540 0.112
#> GSM25427 5 0.2454 0.81297 0.000 0.008 0.000 0.020 0.884 0.088
#> GSM25540 4 0.0632 0.84469 0.024 0.000 0.000 0.976 0.000 0.000
#> GSM25541 4 0.0632 0.84469 0.024 0.000 0.000 0.976 0.000 0.000
#> GSM25542 4 0.0632 0.84469 0.024 0.000 0.000 0.976 0.000 0.000
#> GSM25543 4 0.0632 0.84469 0.024 0.000 0.000 0.976 0.000 0.000
#> GSM25479 1 0.0260 0.85910 0.992 0.000 0.000 0.000 0.008 0.000
#> GSM25480 1 0.0622 0.85273 0.980 0.000 0.000 0.012 0.008 0.000
#> GSM25481 5 0.0000 0.91249 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM25482 5 0.0146 0.91191 0.000 0.000 0.000 0.000 0.996 0.004
#> GSM48654 6 0.2053 0.89322 0.000 0.000 0.000 0.108 0.004 0.888
#> GSM48650 2 0.0000 0.81869 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM48651 2 0.0000 0.81869 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM48652 2 0.0000 0.81869 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM48653 2 0.0000 0.81869 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM48662 2 0.0000 0.81869 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM48663 2 0.0000 0.81869 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM25524 1 0.1387 0.82924 0.932 0.000 0.000 0.068 0.000 0.000
#> GSM25525 1 0.3797 0.27998 0.580 0.000 0.000 0.000 0.420 0.000
#> GSM25526 1 0.4317 0.60649 0.688 0.000 0.060 0.252 0.000 0.000
#> GSM25527 1 0.0000 0.86352 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM25528 1 0.3183 0.75524 0.828 0.000 0.060 0.112 0.000 0.000
#> GSM25529 1 0.0000 0.86352 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM25530 1 0.3766 0.65590 0.748 0.000 0.212 0.040 0.000 0.000
#> GSM25531 1 0.0713 0.85075 0.972 0.000 0.000 0.028 0.000 0.000
#> GSM48661 6 0.1957 0.89198 0.000 0.000 0.000 0.112 0.000 0.888
#> GSM25561 1 0.4808 0.39098 0.576 0.000 0.064 0.360 0.000 0.000
#> GSM25562 4 0.5549 0.35849 0.164 0.304 0.000 0.532 0.000 0.000
#> GSM25563 4 0.2389 0.76135 0.008 0.000 0.128 0.864 0.000 0.000
#> GSM25564 5 0.0508 0.90361 0.000 0.000 0.000 0.004 0.984 0.012
#> GSM25565 4 0.5795 0.17525 0.000 0.324 0.000 0.540 0.028 0.108
#> GSM25566 5 0.6539 -0.00448 0.000 0.324 0.000 0.088 0.480 0.108
#> GSM25568 2 0.5524 0.49272 0.000 0.568 0.000 0.008 0.288 0.136
#> GSM25569 2 0.3929 0.74347 0.000 0.792 0.000 0.020 0.112 0.076
#> GSM25552 5 0.7324 -0.29449 0.000 0.320 0.000 0.236 0.336 0.108
#> GSM25553 2 0.7369 0.21919 0.000 0.324 0.000 0.292 0.276 0.108
#> GSM25578 1 0.0000 0.86352 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM25579 1 0.2562 0.73076 0.828 0.000 0.000 0.172 0.000 0.000
#> GSM25580 1 0.0000 0.86352 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM25581 1 0.0000 0.86352 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM48655 5 0.2178 0.79116 0.000 0.000 0.000 0.000 0.868 0.132
#> GSM48656 6 0.2053 0.89338 0.000 0.004 0.000 0.108 0.000 0.888
#> GSM48657 6 0.2135 0.81320 0.000 0.128 0.000 0.000 0.000 0.872
#> GSM48658 6 0.2053 0.89313 0.004 0.000 0.000 0.108 0.000 0.888
#> GSM25624 6 0.3163 0.75912 0.040 0.000 0.000 0.000 0.140 0.820
#> GSM25625 3 0.0260 0.91187 0.000 0.000 0.992 0.008 0.000 0.000
#> GSM25626 3 0.0000 0.91898 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM25627 6 0.2260 0.78240 0.140 0.000 0.000 0.000 0.000 0.860
#> GSM25628 3 0.0000 0.91898 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM25629 4 0.1168 0.83923 0.028 0.000 0.000 0.956 0.000 0.016
#> GSM25630 3 0.0000 0.91898 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM25631 4 0.2178 0.78570 0.132 0.000 0.000 0.868 0.000 0.000
#> GSM25632 3 0.0000 0.91898 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM25633 1 0.0000 0.86352 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM25634 1 0.0000 0.86352 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM25635 1 0.0000 0.86352 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM25656 4 0.1950 0.81241 0.024 0.000 0.064 0.912 0.000 0.000
#> GSM25657 1 0.0000 0.86352 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM25658 1 0.3789 0.28467 0.584 0.000 0.000 0.416 0.000 0.000
#> GSM25659 4 0.3915 0.54060 0.304 0.000 0.000 0.680 0.008 0.008
#> GSM25660 1 0.0000 0.86352 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM25661 1 0.0000 0.86352 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM25662 4 0.1267 0.80637 0.000 0.000 0.000 0.940 0.060 0.000
#> GSM25663 4 0.1075 0.84056 0.048 0.000 0.000 0.952 0.000 0.000
#> GSM25680 5 0.0000 0.91249 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM25681 5 0.0000 0.91249 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM25682 5 0.0146 0.91191 0.000 0.000 0.000 0.000 0.996 0.004
#> GSM25683 5 0.0146 0.91191 0.000 0.000 0.000 0.000 0.996 0.004
#> GSM25684 5 0.0146 0.91191 0.000 0.000 0.000 0.000 0.996 0.004
#> GSM25685 5 0.0146 0.91191 0.000 0.000 0.000 0.000 0.996 0.004
#> GSM25686 5 0.0146 0.91191 0.000 0.000 0.000 0.000 0.996 0.004
#> GSM25687 5 0.0146 0.91191 0.000 0.000 0.000 0.000 0.996 0.004
#> GSM48664 1 0.0458 0.85872 0.984 0.000 0.000 0.016 0.000 0.000
#> GSM48665 1 0.0993 0.84342 0.964 0.000 0.000 0.024 0.000 0.012
Heatmaps for the consensus matrix. It visualizes the probability of two samples to be in a same group.
consensus_heatmap(res, k = 2)
consensus_heatmap(res, k = 3)
consensus_heatmap(res, k = 4)
consensus_heatmap(res, k = 5)
consensus_heatmap(res, k = 6)
Heatmaps for the membership of samples in all partitions to see how consistent they are:
membership_heatmap(res, k = 2)
membership_heatmap(res, k = 3)
membership_heatmap(res, k = 4)
membership_heatmap(res, k = 5)
membership_heatmap(res, k = 6)
As soon as we have had the classes for columns, we can look for signatures which are significantly different between classes which can be candidate marks for certain classes. Following are the heatmaps for signatures.
Signature heatmaps where rows are scaled:
get_signatures(res, k = 2)
get_signatures(res, k = 3)
get_signatures(res, k = 4)
get_signatures(res, k = 5)
get_signatures(res, k = 6)
Signature heatmaps where rows are not scaled:
get_signatures(res, k = 2, scale_rows = FALSE)
get_signatures(res, k = 3, scale_rows = FALSE)
get_signatures(res, k = 4, scale_rows = FALSE)
get_signatures(res, k = 5, scale_rows = FALSE)
get_signatures(res, k = 6, scale_rows = FALSE)
Compare the overlap of signatures from different k:
compare_signatures(res)
get_signature() returns a data frame invisibly. TO get the list of signatures, the function
call should be assigned to a variable explicitly. In following code, if plot argument is set
to FALSE, no heatmap is plotted while only the differential analysis is performed.
# code only for demonstration
tb = get_signature(res, k = ..., plot = FALSE)
An example of the output of tb is:
#> which_row fdr mean_1 mean_2 scaled_mean_1 scaled_mean_2 km
#> 1 38 0.042760348 8.373488 9.131774 -0.5533452 0.5164555 1
#> 2 40 0.018707592 7.106213 8.469186 -0.6173731 0.5762149 1
#> 3 55 0.019134737 10.221463 11.207825 -0.6159697 0.5749050 1
#> 4 59 0.006059896 5.921854 7.869574 -0.6899429 0.6439467 1
#> 5 60 0.018055526 8.928898 10.211722 -0.6204761 0.5791110 1
#> 6 98 0.009384629 15.714769 14.887706 0.6635654 -0.6193277 2
...
The columns in tb are:
which_row: row indices corresponding to the input matrix.fdr: FDR for the differential test. mean_x: The mean value in group x.scaled_mean_x: The mean value in group x after rows are scaled.km: Row groups if k-means clustering is applied to rows.UMAP plot which shows how samples are separated.
dimension_reduction(res, k = 2, method = "UMAP")
dimension_reduction(res, k = 3, method = "UMAP")
dimension_reduction(res, k = 4, method = "UMAP")
dimension_reduction(res, k = 5, method = "UMAP")
dimension_reduction(res, k = 6, method = "UMAP")
Following heatmap shows how subgroups are split when increasing k:
collect_classes(res)
Test correlation between subgroups and known annotations. If the known annotation is numeric, one-way ANOVA test is applied, and if the known annotation is discrete, chi-squared contingency table test is applied.
test_to_known_factors(res)
#> n genotype/variation(p) k
#> ATC:pam 94 2.43e-06 2
#> ATC:pam 89 2.91e-06 3
#> ATC:pam 82 3.25e-08 4
#> ATC:pam 93 7.44e-12 5
#> ATC:pam 87 8.79e-17 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 100 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 5.
#>
#> 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 0.776 0.915 0.954 0.3658 0.665 0.665
#> 3 3 0.384 0.732 0.845 0.6695 0.666 0.512
#> 4 4 0.817 0.848 0.934 0.0689 0.704 0.418
#> 5 5 0.684 0.758 0.843 0.1152 0.922 0.787
#> 6 6 0.685 0.625 0.787 0.0625 0.829 0.500
suggest_best_k() suggests the best \(k\) based on these statistics. The rules are as follows:
suggest_best_k(res)
#> [1] 5
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
#> GSM25548 2 0.0938 0.974 0.012 0.988
#> GSM25549 2 0.2043 0.967 0.032 0.968
#> GSM25550 2 0.2043 0.967 0.032 0.968
#> GSM25551 2 0.0000 0.975 0.000 1.000
#> GSM25570 2 0.2043 0.967 0.032 0.968
#> GSM25571 2 0.0000 0.975 0.000 1.000
#> GSM25358 1 0.0000 0.945 1.000 0.000
#> GSM25359 1 0.0000 0.945 1.000 0.000
#> GSM25360 1 0.0000 0.945 1.000 0.000
#> GSM25361 1 0.0000 0.945 1.000 0.000
#> GSM25377 1 0.8081 0.728 0.752 0.248
#> GSM25378 2 0.2236 0.964 0.036 0.964
#> GSM25401 1 0.8144 0.726 0.748 0.252
#> GSM25402 1 0.9248 0.574 0.660 0.340
#> GSM25349 1 0.0672 0.944 0.992 0.008
#> GSM25350 1 0.0672 0.944 0.992 0.008
#> GSM25356 2 0.2236 0.964 0.036 0.964
#> GSM25357 2 0.0000 0.975 0.000 1.000
#> GSM25385 1 0.0000 0.945 1.000 0.000
#> GSM25386 1 0.0000 0.945 1.000 0.000
#> GSM25399 1 0.4562 0.886 0.904 0.096
#> GSM25400 1 0.0672 0.943 0.992 0.008
#> GSM48659 1 0.0672 0.944 0.992 0.008
#> GSM48660 1 0.0672 0.944 0.992 0.008
#> GSM25409 1 0.7299 0.787 0.796 0.204
#> GSM25410 1 0.0000 0.945 1.000 0.000
#> GSM25426 2 0.0000 0.975 0.000 1.000
#> GSM25427 2 0.7056 0.751 0.192 0.808
#> GSM25540 1 0.0000 0.945 1.000 0.000
#> GSM25541 1 0.0000 0.945 1.000 0.000
#> GSM25542 1 0.0000 0.945 1.000 0.000
#> GSM25543 1 0.0000 0.945 1.000 0.000
#> GSM25479 1 0.7453 0.775 0.788 0.212
#> GSM25480 1 0.7602 0.765 0.780 0.220
#> GSM25481 2 0.2043 0.967 0.032 0.968
#> GSM25482 2 0.2043 0.967 0.032 0.968
#> GSM48654 1 0.0672 0.944 0.992 0.008
#> GSM48650 1 0.0672 0.944 0.992 0.008
#> GSM48651 1 0.0672 0.944 0.992 0.008
#> GSM48652 1 0.0672 0.944 0.992 0.008
#> GSM48653 1 0.0672 0.944 0.992 0.008
#> GSM48662 1 0.0672 0.944 0.992 0.008
#> GSM48663 1 0.0672 0.944 0.992 0.008
#> GSM25524 1 0.0000 0.945 1.000 0.000
#> GSM25525 1 0.9608 0.482 0.616 0.384
#> GSM25526 1 0.0000 0.945 1.000 0.000
#> GSM25527 1 0.0000 0.945 1.000 0.000
#> GSM25528 1 0.0000 0.945 1.000 0.000
#> GSM25529 1 0.0000 0.945 1.000 0.000
#> GSM25530 1 0.0000 0.945 1.000 0.000
#> GSM25531 1 0.0000 0.945 1.000 0.000
#> GSM48661 1 0.0672 0.944 0.992 0.008
#> GSM25561 1 0.0000 0.945 1.000 0.000
#> GSM25562 1 0.0938 0.941 0.988 0.012
#> GSM25563 1 0.0000 0.945 1.000 0.000
#> GSM25564 1 0.2603 0.927 0.956 0.044
#> GSM25565 1 0.3733 0.906 0.928 0.072
#> GSM25566 1 0.5519 0.861 0.872 0.128
#> GSM25568 1 0.0672 0.944 0.992 0.008
#> GSM25569 1 0.0672 0.944 0.992 0.008
#> GSM25552 1 0.8555 0.688 0.720 0.280
#> GSM25553 1 0.8443 0.700 0.728 0.272
#> GSM25578 1 0.3274 0.913 0.940 0.060
#> GSM25579 1 0.3431 0.911 0.936 0.064
#> GSM25580 1 0.2948 0.918 0.948 0.052
#> GSM25581 1 0.0000 0.945 1.000 0.000
#> GSM48655 1 0.0672 0.944 0.992 0.008
#> GSM48656 1 0.0672 0.944 0.992 0.008
#> GSM48657 1 0.0672 0.944 0.992 0.008
#> GSM48658 1 0.0672 0.944 0.992 0.008
#> GSM25624 1 0.0672 0.944 0.992 0.008
#> GSM25625 1 0.0000 0.945 1.000 0.000
#> GSM25626 1 0.0000 0.945 1.000 0.000
#> GSM25627 1 0.0672 0.944 0.992 0.008
#> GSM25628 1 0.0000 0.945 1.000 0.000
#> GSM25629 1 0.0000 0.945 1.000 0.000
#> GSM25630 1 0.0000 0.945 1.000 0.000
#> GSM25631 1 0.0000 0.945 1.000 0.000
#> GSM25632 1 0.0000 0.945 1.000 0.000
#> GSM25633 1 0.0000 0.945 1.000 0.000
#> GSM25634 1 0.0000 0.945 1.000 0.000
#> GSM25635 1 0.0000 0.945 1.000 0.000
#> GSM25656 1 0.0000 0.945 1.000 0.000
#> GSM25657 1 0.0000 0.945 1.000 0.000
#> GSM25658 1 0.6438 0.827 0.836 0.164
#> GSM25659 1 0.7745 0.755 0.772 0.228
#> GSM25660 1 0.3733 0.904 0.928 0.072
#> GSM25661 1 0.7219 0.789 0.800 0.200
#> GSM25662 1 0.1184 0.942 0.984 0.016
#> GSM25663 1 0.0000 0.945 1.000 0.000
#> GSM25680 2 0.0000 0.975 0.000 1.000
#> GSM25681 2 0.0672 0.975 0.008 0.992
#> GSM25682 2 0.0000 0.975 0.000 1.000
#> GSM25683 2 0.0000 0.975 0.000 1.000
#> GSM25684 2 0.0000 0.975 0.000 1.000
#> GSM25685 2 0.0000 0.975 0.000 1.000
#> GSM25686 2 0.0000 0.975 0.000 1.000
#> GSM25687 2 0.0000 0.975 0.000 1.000
#> GSM48664 1 0.7299 0.784 0.796 0.204
#> GSM48665 1 0.6343 0.830 0.840 0.160
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM25548 2 0.0237 0.8335 0.004 0.996 0.000
#> GSM25549 2 0.6305 0.1473 0.484 0.516 0.000
#> GSM25550 2 0.6302 0.1621 0.480 0.520 0.000
#> GSM25551 2 0.0237 0.8335 0.004 0.996 0.000
#> GSM25570 2 0.6225 0.2907 0.432 0.568 0.000
#> GSM25571 2 0.0237 0.8335 0.004 0.996 0.000
#> GSM25358 1 0.2796 0.7917 0.908 0.000 0.092
#> GSM25359 1 0.2261 0.8145 0.932 0.000 0.068
#> GSM25360 3 0.3116 0.8213 0.108 0.000 0.892
#> GSM25361 3 0.6095 0.6302 0.392 0.000 0.608
#> GSM25377 1 0.3412 0.7694 0.876 0.124 0.000
#> GSM25378 1 0.6111 0.2631 0.604 0.396 0.000
#> GSM25401 1 0.3752 0.7560 0.856 0.144 0.000
#> GSM25402 1 0.4047 0.7576 0.848 0.148 0.004
#> GSM25349 3 0.4139 0.8097 0.016 0.124 0.860
#> GSM25350 3 0.4139 0.8097 0.016 0.124 0.860
#> GSM25356 1 0.6267 0.0582 0.548 0.452 0.000
#> GSM25357 2 0.0237 0.8335 0.004 0.996 0.000
#> GSM25385 3 0.3340 0.8172 0.120 0.000 0.880
#> GSM25386 3 0.3340 0.8172 0.120 0.000 0.880
#> GSM25399 1 0.0237 0.8477 0.996 0.004 0.000
#> GSM25400 1 0.0892 0.8430 0.980 0.000 0.020
#> GSM48659 3 0.4139 0.8097 0.016 0.124 0.860
#> GSM48660 3 0.4139 0.8097 0.016 0.124 0.860
#> GSM25409 1 0.4291 0.7155 0.820 0.180 0.000
#> GSM25410 3 0.3340 0.8172 0.120 0.000 0.880
#> GSM25426 2 0.0237 0.8335 0.004 0.996 0.000
#> GSM25427 1 0.5529 0.5243 0.704 0.296 0.000
#> GSM25540 3 0.5497 0.7607 0.292 0.000 0.708
#> GSM25541 3 0.6307 0.4369 0.488 0.000 0.512
#> GSM25542 3 0.5115 0.8070 0.228 0.004 0.768
#> GSM25543 3 0.4974 0.8024 0.236 0.000 0.764
#> GSM25479 1 0.0747 0.8458 0.984 0.016 0.000
#> GSM25480 1 0.0747 0.8458 0.984 0.016 0.000
#> GSM25481 2 0.6308 0.1142 0.492 0.508 0.000
#> GSM25482 2 0.6302 0.1621 0.480 0.520 0.000
#> GSM48654 3 0.4569 0.8219 0.068 0.072 0.860
#> GSM48650 3 0.4139 0.8097 0.016 0.124 0.860
#> GSM48651 3 0.4139 0.8097 0.016 0.124 0.860
#> GSM48652 3 0.4139 0.8097 0.016 0.124 0.860
#> GSM48653 3 0.4139 0.8097 0.016 0.124 0.860
#> GSM48662 3 0.4139 0.8097 0.016 0.124 0.860
#> GSM48663 3 0.4139 0.8097 0.016 0.124 0.860
#> GSM25524 1 0.0237 0.8483 0.996 0.000 0.004
#> GSM25525 1 0.4504 0.7010 0.804 0.196 0.000
#> GSM25526 3 0.4291 0.8053 0.180 0.000 0.820
#> GSM25527 1 0.3482 0.7330 0.872 0.000 0.128
#> GSM25528 1 0.1289 0.8410 0.968 0.000 0.032
#> GSM25529 1 0.0000 0.8480 1.000 0.000 0.000
#> GSM25530 1 0.6140 0.3230 0.596 0.000 0.404
#> GSM25531 1 0.0592 0.8458 0.988 0.000 0.012
#> GSM48661 3 0.3851 0.8152 0.136 0.004 0.860
#> GSM25561 3 0.3412 0.8178 0.124 0.000 0.876
#> GSM25562 1 0.1289 0.8346 0.968 0.000 0.032
#> GSM25563 3 0.3340 0.8172 0.120 0.000 0.880
#> GSM25564 3 0.9018 0.3562 0.412 0.132 0.456
#> GSM25565 3 0.7273 0.7913 0.156 0.132 0.712
#> GSM25566 3 0.8924 0.5327 0.336 0.140 0.524
#> GSM25568 3 0.4139 0.8097 0.016 0.124 0.860
#> GSM25569 3 0.4139 0.8097 0.016 0.124 0.860
#> GSM25552 1 0.4452 0.6993 0.808 0.192 0.000
#> GSM25553 1 0.4291 0.7155 0.820 0.180 0.000
#> GSM25578 1 0.0000 0.8480 1.000 0.000 0.000
#> GSM25579 1 0.0000 0.8480 1.000 0.000 0.000
#> GSM25580 1 0.0000 0.8480 1.000 0.000 0.000
#> GSM25581 1 0.0000 0.8480 1.000 0.000 0.000
#> GSM48655 3 0.4139 0.8097 0.016 0.124 0.860
#> GSM48656 3 0.3965 0.8163 0.132 0.008 0.860
#> GSM48657 3 0.4139 0.8097 0.016 0.124 0.860
#> GSM48658 3 0.3918 0.8161 0.140 0.004 0.856
#> GSM25624 3 0.5277 0.8255 0.180 0.024 0.796
#> GSM25625 3 0.2959 0.8234 0.100 0.000 0.900
#> GSM25626 3 0.2959 0.8234 0.100 0.000 0.900
#> GSM25627 3 0.4110 0.8177 0.152 0.004 0.844
#> GSM25628 3 0.2959 0.8234 0.100 0.000 0.900
#> GSM25629 3 0.4974 0.8024 0.236 0.000 0.764
#> GSM25630 3 0.2959 0.8234 0.100 0.000 0.900
#> GSM25631 3 0.5327 0.7784 0.272 0.000 0.728
#> GSM25632 3 0.2959 0.8234 0.100 0.000 0.900
#> GSM25633 3 0.6274 0.5076 0.456 0.000 0.544
#> GSM25634 3 0.6192 0.5811 0.420 0.000 0.580
#> GSM25635 3 0.5431 0.7693 0.284 0.000 0.716
#> GSM25656 3 0.2959 0.8234 0.100 0.000 0.900
#> GSM25657 1 0.0747 0.8441 0.984 0.000 0.016
#> GSM25658 1 0.0237 0.8477 0.996 0.004 0.000
#> GSM25659 1 0.3267 0.7799 0.884 0.116 0.000
#> GSM25660 1 0.0000 0.8480 1.000 0.000 0.000
#> GSM25661 1 0.0592 0.8472 0.988 0.012 0.000
#> GSM25662 3 0.7433 0.7783 0.132 0.168 0.700
#> GSM25663 1 0.6244 -0.2571 0.560 0.000 0.440
#> GSM25680 2 0.0237 0.8335 0.004 0.996 0.000
#> GSM25681 2 0.0237 0.8335 0.004 0.996 0.000
#> GSM25682 2 0.0237 0.8335 0.004 0.996 0.000
#> GSM25683 2 0.0237 0.8335 0.004 0.996 0.000
#> GSM25684 2 0.0237 0.8335 0.004 0.996 0.000
#> GSM25685 2 0.0237 0.8335 0.004 0.996 0.000
#> GSM25686 2 0.0237 0.8335 0.004 0.996 0.000
#> GSM25687 2 0.0237 0.8335 0.004 0.996 0.000
#> GSM48664 1 0.0592 0.8472 0.988 0.012 0.000
#> GSM48665 1 0.0661 0.8488 0.988 0.008 0.004
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM25548 2 0.0707 0.9416 0.020 0.980 0.000 0.000
#> GSM25549 1 0.3764 0.7428 0.784 0.216 0.000 0.000
#> GSM25550 1 0.3801 0.7374 0.780 0.220 0.000 0.000
#> GSM25551 2 0.0707 0.9416 0.020 0.980 0.000 0.000
#> GSM25570 2 0.4998 -0.0577 0.488 0.512 0.000 0.000
#> GSM25571 2 0.0707 0.9416 0.020 0.980 0.000 0.000
#> GSM25358 1 0.1824 0.8913 0.936 0.004 0.060 0.000
#> GSM25359 1 0.3448 0.7829 0.828 0.004 0.168 0.000
#> GSM25360 3 0.0000 0.8566 0.000 0.000 1.000 0.000
#> GSM25361 1 0.0707 0.9147 0.980 0.020 0.000 0.000
#> GSM25377 1 0.0336 0.9196 0.992 0.008 0.000 0.000
#> GSM25378 1 0.0921 0.9143 0.972 0.028 0.000 0.000
#> GSM25401 1 0.0336 0.9196 0.992 0.008 0.000 0.000
#> GSM25402 1 0.0336 0.9202 0.992 0.008 0.000 0.000
#> GSM25349 4 0.0336 0.9341 0.000 0.008 0.000 0.992
#> GSM25350 4 0.0336 0.9341 0.000 0.008 0.000 0.992
#> GSM25356 1 0.1474 0.9013 0.948 0.052 0.000 0.000
#> GSM25357 2 0.0707 0.9416 0.020 0.980 0.000 0.000
#> GSM25385 3 0.3873 0.6947 0.228 0.000 0.772 0.000
#> GSM25386 3 0.0000 0.8566 0.000 0.000 1.000 0.000
#> GSM25399 1 0.0000 0.9209 1.000 0.000 0.000 0.000
#> GSM25400 1 0.0524 0.9195 0.988 0.004 0.008 0.000
#> GSM48659 4 0.0469 0.9311 0.000 0.012 0.000 0.988
#> GSM48660 4 0.0336 0.9341 0.000 0.008 0.000 0.992
#> GSM25409 1 0.1022 0.9123 0.968 0.032 0.000 0.000
#> GSM25410 3 0.0000 0.8566 0.000 0.000 1.000 0.000
#> GSM25426 2 0.0707 0.9416 0.020 0.980 0.000 0.000
#> GSM25427 1 0.0469 0.9191 0.988 0.012 0.000 0.000
#> GSM25540 1 0.0707 0.9147 0.980 0.020 0.000 0.000
#> GSM25541 1 0.0707 0.9147 0.980 0.020 0.000 0.000
#> GSM25542 1 0.6292 0.2119 0.548 0.044 0.008 0.400
#> GSM25543 1 0.5832 0.4683 0.640 0.044 0.004 0.312
#> GSM25479 1 0.0000 0.9209 1.000 0.000 0.000 0.000
#> GSM25480 1 0.0000 0.9209 1.000 0.000 0.000 0.000
#> GSM25481 1 0.3688 0.7532 0.792 0.208 0.000 0.000
#> GSM25482 1 0.3801 0.7374 0.780 0.220 0.000 0.000
#> GSM48654 4 0.0469 0.9252 0.000 0.012 0.000 0.988
#> GSM48650 4 0.0336 0.9341 0.000 0.008 0.000 0.992
#> GSM48651 4 0.0336 0.9341 0.000 0.008 0.000 0.992
#> GSM48652 4 0.0336 0.9341 0.000 0.008 0.000 0.992
#> GSM48653 4 0.0336 0.9341 0.000 0.008 0.000 0.992
#> GSM48662 4 0.0336 0.9341 0.000 0.008 0.000 0.992
#> GSM48663 4 0.0336 0.9341 0.000 0.008 0.000 0.992
#> GSM25524 1 0.0779 0.9178 0.980 0.004 0.016 0.000
#> GSM25525 1 0.0469 0.9191 0.988 0.012 0.000 0.000
#> GSM25526 3 0.4843 0.3891 0.396 0.000 0.604 0.000
#> GSM25527 1 0.0707 0.9147 0.980 0.020 0.000 0.000
#> GSM25528 1 0.4079 0.7540 0.800 0.020 0.180 0.000
#> GSM25529 1 0.0000 0.9209 1.000 0.000 0.000 0.000
#> GSM25530 1 0.5606 -0.0616 0.500 0.020 0.480 0.000
#> GSM25531 1 0.3991 0.7646 0.808 0.020 0.172 0.000
#> GSM48661 4 0.0707 0.9209 0.000 0.020 0.000 0.980
#> GSM25561 3 0.4134 0.6589 0.260 0.000 0.740 0.000
#> GSM25562 1 0.0000 0.9209 1.000 0.000 0.000 0.000
#> GSM25563 3 0.2868 0.7788 0.136 0.000 0.864 0.000
#> GSM25564 1 0.1022 0.9123 0.968 0.032 0.000 0.000
#> GSM25565 1 0.1022 0.9123 0.968 0.032 0.000 0.000
#> GSM25566 1 0.1022 0.9123 0.968 0.032 0.000 0.000
#> GSM25568 4 0.0592 0.9275 0.000 0.016 0.000 0.984
#> GSM25569 4 0.0336 0.9341 0.000 0.008 0.000 0.992
#> GSM25552 1 0.1022 0.9123 0.968 0.032 0.000 0.000
#> GSM25553 1 0.1022 0.9123 0.968 0.032 0.000 0.000
#> GSM25578 1 0.0000 0.9209 1.000 0.000 0.000 0.000
#> GSM25579 1 0.0000 0.9209 1.000 0.000 0.000 0.000
#> GSM25580 1 0.0000 0.9209 1.000 0.000 0.000 0.000
#> GSM25581 1 0.0707 0.9147 0.980 0.020 0.000 0.000
#> GSM48655 4 0.0336 0.9341 0.000 0.008 0.000 0.992
#> GSM48656 4 0.0707 0.9209 0.000 0.020 0.000 0.980
#> GSM48657 4 0.0336 0.9341 0.000 0.008 0.000 0.992
#> GSM48658 4 0.1520 0.9011 0.020 0.024 0.000 0.956
#> GSM25624 4 0.6108 0.1449 0.424 0.048 0.000 0.528
#> GSM25625 3 0.0469 0.8478 0.000 0.012 0.988 0.000
#> GSM25626 3 0.0000 0.8566 0.000 0.000 1.000 0.000
#> GSM25627 4 0.5393 0.5251 0.268 0.044 0.000 0.688
#> GSM25628 3 0.0000 0.8566 0.000 0.000 1.000 0.000
#> GSM25629 1 0.4954 0.6928 0.772 0.020 0.028 0.180
#> GSM25630 3 0.0000 0.8566 0.000 0.000 1.000 0.000
#> GSM25631 1 0.1211 0.9133 0.960 0.040 0.000 0.000
#> GSM25632 3 0.0000 0.8566 0.000 0.000 1.000 0.000
#> GSM25633 1 0.0707 0.9147 0.980 0.020 0.000 0.000
#> GSM25634 1 0.0707 0.9147 0.980 0.020 0.000 0.000
#> GSM25635 1 0.0707 0.9147 0.980 0.020 0.000 0.000
#> GSM25656 3 0.2011 0.8177 0.080 0.000 0.920 0.000
#> GSM25657 1 0.0707 0.9147 0.980 0.020 0.000 0.000
#> GSM25658 1 0.0000 0.9209 1.000 0.000 0.000 0.000
#> GSM25659 1 0.0000 0.9209 1.000 0.000 0.000 0.000
#> GSM25660 1 0.0000 0.9209 1.000 0.000 0.000 0.000
#> GSM25661 1 0.0000 0.9209 1.000 0.000 0.000 0.000
#> GSM25662 1 0.2973 0.8242 0.856 0.144 0.000 0.000
#> GSM25663 1 0.0000 0.9209 1.000 0.000 0.000 0.000
#> GSM25680 2 0.0707 0.9416 0.020 0.980 0.000 0.000
#> GSM25681 2 0.0707 0.9416 0.020 0.980 0.000 0.000
#> GSM25682 2 0.0707 0.9416 0.020 0.980 0.000 0.000
#> GSM25683 2 0.0707 0.9416 0.020 0.980 0.000 0.000
#> GSM25684 2 0.0707 0.9416 0.020 0.980 0.000 0.000
#> GSM25685 2 0.0707 0.9416 0.020 0.980 0.000 0.000
#> GSM25686 2 0.0707 0.9416 0.020 0.980 0.000 0.000
#> GSM25687 2 0.0707 0.9416 0.020 0.980 0.000 0.000
#> GSM48664 1 0.0000 0.9209 1.000 0.000 0.000 0.000
#> GSM48665 1 0.0000 0.9209 1.000 0.000 0.000 0.000
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM25548 5 0.2424 0.7824 0.000 0.000 0.000 0.132 0.868
#> GSM25549 4 0.4972 0.9342 0.068 0.000 0.000 0.672 0.260
#> GSM25550 4 0.4995 0.9352 0.068 0.000 0.000 0.668 0.264
#> GSM25551 5 0.0000 0.9772 0.000 0.000 0.000 0.000 1.000
#> GSM25570 4 0.4484 0.8394 0.024 0.000 0.000 0.668 0.308
#> GSM25571 5 0.0000 0.9772 0.000 0.000 0.000 0.000 1.000
#> GSM25358 1 0.4791 0.7220 0.740 0.000 0.072 0.176 0.012
#> GSM25359 1 0.4444 0.7247 0.748 0.000 0.072 0.180 0.000
#> GSM25360 3 0.0162 0.7683 0.000 0.000 0.996 0.004 0.000
#> GSM25361 1 0.3003 0.7619 0.812 0.000 0.000 0.188 0.000
#> GSM25377 1 0.2017 0.7467 0.912 0.000 0.000 0.080 0.008
#> GSM25378 1 0.6376 0.2079 0.516 0.000 0.000 0.264 0.220
#> GSM25401 1 0.2843 0.7031 0.848 0.000 0.000 0.144 0.008
#> GSM25402 1 0.3075 0.7370 0.860 0.000 0.000 0.092 0.048
#> GSM25349 2 0.0510 0.9115 0.000 0.984 0.000 0.000 0.016
#> GSM25350 2 0.0510 0.9115 0.000 0.984 0.000 0.000 0.016
#> GSM25356 4 0.5887 0.7975 0.156 0.000 0.000 0.592 0.252
#> GSM25357 5 0.0000 0.9772 0.000 0.000 0.000 0.000 1.000
#> GSM25385 3 0.4662 0.6992 0.096 0.000 0.736 0.168 0.000
#> GSM25386 3 0.0000 0.7689 0.000 0.000 1.000 0.000 0.000
#> GSM25399 1 0.0324 0.7824 0.992 0.000 0.000 0.004 0.004
#> GSM25400 1 0.1041 0.7822 0.964 0.000 0.000 0.032 0.004
#> GSM48659 2 0.3143 0.6780 0.000 0.796 0.000 0.000 0.204
#> GSM48660 2 0.0510 0.9115 0.000 0.984 0.000 0.000 0.016
#> GSM25409 1 0.4429 0.6064 0.744 0.000 0.000 0.192 0.064
#> GSM25410 3 0.0000 0.7689 0.000 0.000 1.000 0.000 0.000
#> GSM25426 5 0.0000 0.9772 0.000 0.000 0.000 0.000 1.000
#> GSM25427 1 0.5820 0.4552 0.612 0.000 0.000 0.196 0.192
#> GSM25540 1 0.3710 0.7549 0.784 0.000 0.024 0.192 0.000
#> GSM25541 1 0.3003 0.7619 0.812 0.000 0.000 0.188 0.000
#> GSM25542 1 0.7418 0.1814 0.384 0.320 0.032 0.264 0.000
#> GSM25543 1 0.6436 0.4410 0.504 0.232 0.000 0.264 0.000
#> GSM25479 1 0.1502 0.7601 0.940 0.000 0.000 0.056 0.004
#> GSM25480 1 0.1892 0.7461 0.916 0.000 0.000 0.080 0.004
#> GSM25481 4 0.5028 0.9332 0.072 0.000 0.000 0.668 0.260
#> GSM25482 4 0.4995 0.9352 0.068 0.000 0.000 0.668 0.264
#> GSM48654 2 0.2012 0.8658 0.060 0.920 0.000 0.020 0.000
#> GSM48650 2 0.0510 0.9115 0.000 0.984 0.000 0.000 0.016
#> GSM48651 2 0.0510 0.9115 0.000 0.984 0.000 0.000 0.016
#> GSM48652 2 0.0510 0.9115 0.000 0.984 0.000 0.000 0.016
#> GSM48653 2 0.0510 0.9115 0.000 0.984 0.000 0.000 0.016
#> GSM48662 2 0.0510 0.9115 0.000 0.984 0.000 0.000 0.016
#> GSM48663 2 0.0510 0.9115 0.000 0.984 0.000 0.000 0.016
#> GSM25524 1 0.3109 0.7622 0.800 0.000 0.000 0.200 0.000
#> GSM25525 1 0.2997 0.6975 0.840 0.000 0.000 0.148 0.012
#> GSM25526 3 0.6052 0.5478 0.248 0.000 0.572 0.180 0.000
#> GSM25527 1 0.2813 0.7652 0.832 0.000 0.000 0.168 0.000
#> GSM25528 1 0.4960 0.6891 0.708 0.000 0.112 0.180 0.000
#> GSM25529 1 0.2648 0.7704 0.848 0.000 0.000 0.152 0.000
#> GSM25530 3 0.6349 0.2383 0.360 0.000 0.472 0.168 0.000
#> GSM25531 1 0.4502 0.7226 0.744 0.000 0.076 0.180 0.000
#> GSM48661 2 0.2915 0.8027 0.116 0.860 0.000 0.024 0.000
#> GSM25561 3 0.5258 0.6736 0.140 0.000 0.680 0.180 0.000
#> GSM25562 1 0.0162 0.7817 0.996 0.000 0.000 0.004 0.000
#> GSM25563 3 0.5218 0.6762 0.136 0.000 0.684 0.180 0.000
#> GSM25564 1 0.4455 0.6077 0.744 0.000 0.000 0.188 0.068
#> GSM25565 1 0.7803 0.2024 0.484 0.144 0.000 0.192 0.180
#> GSM25566 1 0.6286 0.4686 0.640 0.060 0.000 0.192 0.108
#> GSM25568 2 0.2012 0.8658 0.060 0.920 0.000 0.020 0.000
#> GSM25569 2 0.0510 0.9115 0.000 0.984 0.000 0.000 0.016
#> GSM25552 1 0.4547 0.6008 0.736 0.000 0.000 0.192 0.072
#> GSM25553 1 0.4036 0.6584 0.788 0.000 0.000 0.144 0.068
#> GSM25578 1 0.0000 0.7823 1.000 0.000 0.000 0.000 0.000
#> GSM25579 1 0.0162 0.7817 0.996 0.000 0.000 0.004 0.000
#> GSM25580 1 0.0000 0.7823 1.000 0.000 0.000 0.000 0.000
#> GSM25581 1 0.2773 0.7666 0.836 0.000 0.000 0.164 0.000
#> GSM48655 2 0.0510 0.9115 0.000 0.984 0.000 0.000 0.016
#> GSM48656 2 0.2012 0.8658 0.060 0.920 0.000 0.020 0.000
#> GSM48657 2 0.0510 0.9115 0.000 0.984 0.000 0.000 0.016
#> GSM48658 2 0.4884 0.6925 0.128 0.720 0.000 0.152 0.000
#> GSM25624 1 0.5737 0.0422 0.460 0.456 0.000 0.084 0.000
#> GSM25625 3 0.1502 0.7425 0.004 0.000 0.940 0.056 0.000
#> GSM25626 3 0.0000 0.7689 0.000 0.000 1.000 0.000 0.000
#> GSM25627 2 0.6180 0.4398 0.220 0.556 0.000 0.224 0.000
#> GSM25628 3 0.0000 0.7689 0.000 0.000 1.000 0.000 0.000
#> GSM25629 1 0.4960 0.6776 0.668 0.064 0.000 0.268 0.000
#> GSM25630 3 0.0000 0.7689 0.000 0.000 1.000 0.000 0.000
#> GSM25631 1 0.3750 0.7428 0.756 0.012 0.000 0.232 0.000
#> GSM25632 3 0.0000 0.7689 0.000 0.000 1.000 0.000 0.000
#> GSM25633 1 0.2966 0.7635 0.816 0.000 0.000 0.184 0.000
#> GSM25634 1 0.3039 0.7660 0.808 0.000 0.000 0.192 0.000
#> GSM25635 1 0.3177 0.7572 0.792 0.000 0.000 0.208 0.000
#> GSM25656 3 0.5440 0.6580 0.156 0.000 0.660 0.184 0.000
#> GSM25657 1 0.0703 0.7839 0.976 0.000 0.000 0.024 0.000
#> GSM25658 1 0.0579 0.7811 0.984 0.000 0.000 0.008 0.008
#> GSM25659 1 0.2563 0.7137 0.872 0.000 0.000 0.120 0.008
#> GSM25660 1 0.0000 0.7823 1.000 0.000 0.000 0.000 0.000
#> GSM25661 1 0.0324 0.7814 0.992 0.000 0.000 0.004 0.004
#> GSM25662 1 0.4928 0.5612 0.692 0.004 0.008 0.040 0.256
#> GSM25663 1 0.3039 0.7639 0.808 0.000 0.000 0.192 0.000
#> GSM25680 5 0.0290 0.9682 0.000 0.000 0.000 0.008 0.992
#> GSM25681 5 0.1121 0.9218 0.000 0.000 0.000 0.044 0.956
#> GSM25682 5 0.0000 0.9772 0.000 0.000 0.000 0.000 1.000
#> GSM25683 5 0.0000 0.9772 0.000 0.000 0.000 0.000 1.000
#> GSM25684 5 0.0000 0.9772 0.000 0.000 0.000 0.000 1.000
#> GSM25685 5 0.0000 0.9772 0.000 0.000 0.000 0.000 1.000
#> GSM25686 5 0.0000 0.9772 0.000 0.000 0.000 0.000 1.000
#> GSM25687 5 0.0000 0.9772 0.000 0.000 0.000 0.000 1.000
#> GSM48664 1 0.0451 0.7817 0.988 0.000 0.000 0.008 0.004
#> GSM48665 1 0.0324 0.7814 0.992 0.000 0.000 0.004 0.004
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM25548 5 0.1556 0.735149 0.000 0.000 0.000 0.080 0.920 0.000
#> GSM25549 5 0.4338 0.593351 0.000 0.000 0.000 0.484 0.496 0.020
#> GSM25550 5 0.4338 0.593351 0.000 0.000 0.000 0.484 0.496 0.020
#> GSM25551 5 0.0000 0.751679 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM25570 5 0.4337 0.595014 0.000 0.000 0.000 0.480 0.500 0.020
#> GSM25571 5 0.0000 0.751679 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM25358 4 0.5304 0.408410 0.388 0.000 0.004 0.516 0.000 0.092
#> GSM25359 4 0.5397 0.419885 0.384 0.000 0.008 0.516 0.000 0.092
#> GSM25360 3 0.1967 0.893915 0.000 0.000 0.904 0.084 0.000 0.012
#> GSM25361 1 0.5241 0.209159 0.568 0.000 0.000 0.312 0.000 0.120
#> GSM25377 1 0.0603 0.732509 0.980 0.000 0.000 0.000 0.004 0.016
#> GSM25378 5 0.5811 0.274982 0.360 0.000 0.000 0.136 0.492 0.012
#> GSM25401 1 0.2776 0.676410 0.860 0.000 0.000 0.052 0.088 0.000
#> GSM25402 1 0.4756 0.555765 0.696 0.000 0.000 0.060 0.216 0.028
#> GSM25349 2 0.0000 0.915257 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM25350 2 0.0000 0.915257 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM25356 5 0.5550 0.575281 0.088 0.000 0.000 0.400 0.496 0.016
#> GSM25357 5 0.0000 0.751679 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM25385 4 0.5745 0.472078 0.036 0.000 0.388 0.500 0.000 0.076
#> GSM25386 3 0.0260 0.962616 0.000 0.000 0.992 0.008 0.000 0.000
#> GSM25399 1 0.1010 0.733085 0.960 0.000 0.000 0.004 0.000 0.036
#> GSM25400 1 0.3327 0.645851 0.820 0.000 0.000 0.092 0.000 0.088
#> GSM48659 2 0.5152 -0.000294 0.000 0.468 0.000 0.000 0.448 0.084
#> GSM48660 2 0.0000 0.915257 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM25409 1 0.4313 0.594623 0.728 0.000 0.000 0.148 0.124 0.000
#> GSM25410 3 0.0260 0.962616 0.000 0.000 0.992 0.008 0.000 0.000
#> GSM25426 5 0.0000 0.751679 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM25427 5 0.5664 0.225253 0.384 0.000 0.000 0.112 0.492 0.012
#> GSM25540 4 0.5589 0.419900 0.380 0.000 0.012 0.504 0.000 0.104
#> GSM25541 1 0.4791 0.422358 0.652 0.000 0.000 0.244 0.000 0.104
#> GSM25542 6 0.4980 0.592052 0.132 0.120 0.012 0.020 0.000 0.716
#> GSM25543 6 0.4712 0.389296 0.288 0.040 0.000 0.020 0.000 0.652
#> GSM25479 1 0.0622 0.731781 0.980 0.000 0.000 0.008 0.000 0.012
#> GSM25480 1 0.0820 0.729175 0.972 0.000 0.000 0.016 0.000 0.012
#> GSM25481 5 0.4338 0.593351 0.000 0.000 0.000 0.484 0.496 0.020
#> GSM25482 5 0.4338 0.593351 0.000 0.000 0.000 0.484 0.496 0.020
#> GSM48654 6 0.3659 0.483838 0.000 0.364 0.000 0.000 0.000 0.636
#> GSM48650 2 0.0000 0.915257 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM48651 2 0.0000 0.915257 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM48652 2 0.0000 0.915257 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM48653 2 0.0000 0.915257 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM48662 2 0.0000 0.915257 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM48663 2 0.0000 0.915257 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM25524 1 0.5201 -0.118102 0.500 0.000 0.000 0.408 0.000 0.092
#> GSM25525 1 0.3660 0.649318 0.800 0.000 0.000 0.096 0.100 0.004
#> GSM25526 4 0.6077 0.563558 0.056 0.000 0.336 0.516 0.000 0.092
#> GSM25527 1 0.4414 0.508724 0.704 0.000 0.000 0.204 0.000 0.092
#> GSM25528 4 0.6116 0.508977 0.332 0.000 0.060 0.516 0.000 0.092
#> GSM25529 1 0.4570 0.468255 0.680 0.000 0.000 0.228 0.000 0.092
#> GSM25530 4 0.6309 0.566762 0.104 0.000 0.328 0.500 0.000 0.068
#> GSM25531 4 0.5909 0.471441 0.356 0.000 0.040 0.512 0.000 0.092
#> GSM48661 6 0.3531 0.527423 0.000 0.328 0.000 0.000 0.000 0.672
#> GSM25561 4 0.5798 0.528268 0.032 0.000 0.360 0.516 0.000 0.092
#> GSM25562 1 0.0000 0.734410 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM25563 4 0.5691 0.507131 0.024 0.000 0.372 0.512 0.000 0.092
#> GSM25564 1 0.4390 0.587426 0.720 0.000 0.000 0.148 0.132 0.000
#> GSM25565 5 0.6385 0.117259 0.400 0.044 0.000 0.140 0.416 0.000
#> GSM25566 1 0.6159 0.276579 0.536 0.040 0.000 0.152 0.272 0.000
#> GSM25568 6 0.3659 0.483838 0.000 0.364 0.000 0.000 0.000 0.636
#> GSM25569 2 0.1610 0.841542 0.000 0.916 0.000 0.000 0.000 0.084
#> GSM25552 1 0.4348 0.592163 0.724 0.000 0.000 0.152 0.124 0.000
#> GSM25553 1 0.3873 0.627237 0.772 0.000 0.000 0.124 0.104 0.000
#> GSM25578 1 0.0146 0.734353 0.996 0.000 0.000 0.000 0.000 0.004
#> GSM25579 1 0.0146 0.734353 0.996 0.000 0.000 0.000 0.000 0.004
#> GSM25580 1 0.0146 0.734393 0.996 0.000 0.000 0.000 0.000 0.004
#> GSM25581 1 0.4520 0.482356 0.688 0.000 0.000 0.220 0.000 0.092
#> GSM48655 2 0.2060 0.830110 0.000 0.900 0.000 0.000 0.016 0.084
#> GSM48656 6 0.3659 0.483838 0.000 0.364 0.000 0.000 0.000 0.636
#> GSM48657 2 0.0363 0.907264 0.000 0.988 0.000 0.000 0.000 0.012
#> GSM48658 6 0.1610 0.593804 0.000 0.084 0.000 0.000 0.000 0.916
#> GSM25624 6 0.5166 0.276085 0.384 0.092 0.000 0.000 0.000 0.524
#> GSM25625 3 0.2488 0.874182 0.000 0.000 0.880 0.076 0.000 0.044
#> GSM25626 3 0.0000 0.963598 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM25627 6 0.1745 0.599827 0.012 0.068 0.000 0.000 0.000 0.920
#> GSM25628 3 0.0000 0.963598 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM25629 6 0.5712 0.095223 0.328 0.012 0.004 0.116 0.000 0.540
#> GSM25630 3 0.0000 0.963598 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM25631 1 0.5222 0.364042 0.584 0.000 0.000 0.128 0.000 0.288
#> GSM25632 3 0.0000 0.963598 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM25633 1 0.4707 0.484358 0.676 0.000 0.000 0.204 0.000 0.120
#> GSM25634 1 0.4001 0.603228 0.760 0.000 0.000 0.112 0.000 0.128
#> GSM25635 1 0.5399 0.073357 0.528 0.000 0.000 0.344 0.000 0.128
#> GSM25656 4 0.5865 0.529800 0.032 0.000 0.356 0.512 0.000 0.100
#> GSM25657 1 0.3006 0.661959 0.844 0.000 0.000 0.064 0.000 0.092
#> GSM25658 1 0.0146 0.734353 0.996 0.000 0.000 0.000 0.000 0.004
#> GSM25659 1 0.0909 0.727678 0.968 0.000 0.000 0.020 0.000 0.012
#> GSM25660 1 0.0146 0.734393 0.996 0.000 0.000 0.000 0.000 0.004
#> GSM25661 1 0.0458 0.732495 0.984 0.000 0.000 0.000 0.000 0.016
#> GSM25662 5 0.6376 0.189891 0.336 0.004 0.004 0.056 0.504 0.096
#> GSM25663 1 0.4358 0.520582 0.712 0.000 0.000 0.196 0.000 0.092
#> GSM25680 5 0.0000 0.751679 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM25681 5 0.0632 0.747918 0.000 0.000 0.000 0.024 0.976 0.000
#> GSM25682 5 0.0000 0.751679 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM25683 5 0.0000 0.751679 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM25684 5 0.0000 0.751679 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM25685 5 0.0000 0.751679 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM25686 5 0.0000 0.751679 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM25687 5 0.0000 0.751679 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM48664 1 0.0458 0.732495 0.984 0.000 0.000 0.000 0.000 0.016
#> GSM48665 1 0.0458 0.732495 0.984 0.000 0.000 0.000 0.000 0.016
Heatmaps for the consensus matrix. It visualizes the probability of two samples to be in a same group.
consensus_heatmap(res, k = 2)
consensus_heatmap(res, k = 3)
consensus_heatmap(res, k = 4)
consensus_heatmap(res, k = 5)
consensus_heatmap(res, k = 6)
Heatmaps for the membership of samples in all partitions to see how consistent they are:
membership_heatmap(res, k = 2)
membership_heatmap(res, k = 3)
membership_heatmap(res, k = 4)
membership_heatmap(res, k = 5)
membership_heatmap(res, k = 6)
As soon as we have had the classes for columns, we can look for signatures which are significantly different between classes which can be candidate marks for certain classes. Following are the heatmaps for signatures.
Signature heatmaps where rows are scaled:
get_signatures(res, k = 2)
get_signatures(res, k = 3)
get_signatures(res, k = 4)
get_signatures(res, k = 5)
get_signatures(res, k = 6)
Signature heatmaps where rows are not scaled:
get_signatures(res, k = 2, scale_rows = FALSE)
get_signatures(res, k = 3, scale_rows = FALSE)
get_signatures(res, k = 4, scale_rows = FALSE)
get_signatures(res, k = 5, scale_rows = FALSE)
get_signatures(res, k = 6, scale_rows = FALSE)
Compare the overlap of signatures from different k:
compare_signatures(res)
get_signature() returns a data frame invisibly. TO get the list of signatures, the function
call should be assigned to a variable explicitly. In following code, if plot argument is set
to FALSE, no heatmap is plotted while only the differential analysis is performed.
# code only for demonstration
tb = get_signature(res, k = ..., plot = FALSE)
An example of the output of tb is:
#> which_row fdr mean_1 mean_2 scaled_mean_1 scaled_mean_2 km
#> 1 38 0.042760348 8.373488 9.131774 -0.5533452 0.5164555 1
#> 2 40 0.018707592 7.106213 8.469186 -0.6173731 0.5762149 1
#> 3 55 0.019134737 10.221463 11.207825 -0.6159697 0.5749050 1
#> 4 59 0.006059896 5.921854 7.869574 -0.6899429 0.6439467 1
#> 5 60 0.018055526 8.928898 10.211722 -0.6204761 0.5791110 1
#> 6 98 0.009384629 15.714769 14.887706 0.6635654 -0.6193277 2
...
The columns in tb are:
which_row: row indices corresponding to the input matrix.fdr: FDR for the differential test. mean_x: The mean value in group x.scaled_mean_x: The mean value in group x after rows are scaled.km: Row groups if k-means clustering is applied to rows.UMAP plot which shows how samples are separated.
dimension_reduction(res, k = 2, method = "UMAP")
dimension_reduction(res, k = 3, method = "UMAP")
dimension_reduction(res, k = 4, method = "UMAP")
dimension_reduction(res, k = 5, method = "UMAP")
dimension_reduction(res, k = 6, method = "UMAP")
Following heatmap shows how subgroups are split when increasing k:
collect_classes(res)
Test correlation between subgroups and known annotations. If the known annotation is numeric, one-way ANOVA test is applied, and if the known annotation is discrete, chi-squared contingency table test is applied.
test_to_known_factors(res)
#> n genotype/variation(p) k
#> ATC:mclust 99 3.10e-08 2
#> ATC:mclust 89 1.82e-12 3
#> ATC:mclust 94 4.94e-12 4
#> ATC:mclust 91 2.60e-13 5
#> ATC:mclust 75 4.12e-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["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 100 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 0.898 0.905 0.962 0.4728 0.515 0.515
#> 3 3 0.832 0.889 0.944 0.4016 0.708 0.490
#> 4 4 0.759 0.818 0.896 0.1328 0.845 0.580
#> 5 5 0.666 0.631 0.788 0.0577 0.915 0.682
#> 6 6 0.680 0.568 0.755 0.0452 0.903 0.583
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
#> GSM25548 2 0.0000 0.9833 0.000 1.000
#> GSM25549 2 0.0000 0.9833 0.000 1.000
#> GSM25550 2 0.0000 0.9833 0.000 1.000
#> GSM25551 2 0.0000 0.9833 0.000 1.000
#> GSM25570 2 0.0000 0.9833 0.000 1.000
#> GSM25571 2 0.0000 0.9833 0.000 1.000
#> GSM25358 1 0.0000 0.9217 1.000 0.000
#> GSM25359 1 0.0000 0.9217 1.000 0.000
#> GSM25360 1 0.0000 0.9217 1.000 0.000
#> GSM25361 1 0.0000 0.9217 1.000 0.000
#> GSM25377 2 0.0000 0.9833 0.000 1.000
#> GSM25378 2 0.0000 0.9833 0.000 1.000
#> GSM25401 2 0.0000 0.9833 0.000 1.000
#> GSM25402 2 0.0000 0.9833 0.000 1.000
#> GSM25349 2 0.0000 0.9833 0.000 1.000
#> GSM25350 2 0.0000 0.9833 0.000 1.000
#> GSM25356 2 0.0000 0.9833 0.000 1.000
#> GSM25357 2 0.0000 0.9833 0.000 1.000
#> GSM25385 1 0.0000 0.9217 1.000 0.000
#> GSM25386 1 0.0000 0.9217 1.000 0.000
#> GSM25399 1 1.0000 0.1149 0.500 0.500
#> GSM25400 1 0.9754 0.3941 0.592 0.408
#> GSM48659 2 0.0000 0.9833 0.000 1.000
#> GSM48660 2 0.0000 0.9833 0.000 1.000
#> GSM25409 2 0.0000 0.9833 0.000 1.000
#> GSM25410 1 0.0000 0.9217 1.000 0.000
#> GSM25426 2 0.0000 0.9833 0.000 1.000
#> GSM25427 2 0.0000 0.9833 0.000 1.000
#> GSM25540 1 0.0000 0.9217 1.000 0.000
#> GSM25541 1 0.0000 0.9217 1.000 0.000
#> GSM25542 1 0.0376 0.9198 0.996 0.004
#> GSM25543 1 0.3584 0.8742 0.932 0.068
#> GSM25479 2 0.0000 0.9833 0.000 1.000
#> GSM25480 2 0.0000 0.9833 0.000 1.000
#> GSM25481 2 0.0000 0.9833 0.000 1.000
#> GSM25482 2 0.0000 0.9833 0.000 1.000
#> GSM48654 2 0.0000 0.9833 0.000 1.000
#> GSM48650 2 0.0000 0.9833 0.000 1.000
#> GSM48651 2 0.0000 0.9833 0.000 1.000
#> GSM48652 2 0.0000 0.9833 0.000 1.000
#> GSM48653 2 0.0000 0.9833 0.000 1.000
#> GSM48662 2 0.0000 0.9833 0.000 1.000
#> GSM48663 2 0.0000 0.9833 0.000 1.000
#> GSM25524 1 0.0938 0.9150 0.988 0.012
#> GSM25525 2 0.0000 0.9833 0.000 1.000
#> GSM25526 1 0.0000 0.9217 1.000 0.000
#> GSM25527 1 0.0672 0.9175 0.992 0.008
#> GSM25528 1 0.0000 0.9217 1.000 0.000
#> GSM25529 1 0.0000 0.9217 1.000 0.000
#> GSM25530 1 0.0000 0.9217 1.000 0.000
#> GSM25531 1 0.0000 0.9217 1.000 0.000
#> GSM48661 1 0.0000 0.9217 1.000 0.000
#> GSM25561 1 0.0000 0.9217 1.000 0.000
#> GSM25562 2 0.0376 0.9795 0.004 0.996
#> GSM25563 1 0.0000 0.9217 1.000 0.000
#> GSM25564 2 0.0000 0.9833 0.000 1.000
#> GSM25565 2 0.0000 0.9833 0.000 1.000
#> GSM25566 2 0.0000 0.9833 0.000 1.000
#> GSM25568 2 0.0000 0.9833 0.000 1.000
#> GSM25569 2 0.0000 0.9833 0.000 1.000
#> GSM25552 2 0.0000 0.9833 0.000 1.000
#> GSM25553 2 0.0000 0.9833 0.000 1.000
#> GSM25578 2 0.9988 -0.0783 0.480 0.520
#> GSM25579 2 0.0376 0.9795 0.004 0.996
#> GSM25580 1 0.8661 0.6272 0.712 0.288
#> GSM25581 1 0.0000 0.9217 1.000 0.000
#> GSM48655 2 0.0000 0.9833 0.000 1.000
#> GSM48656 2 0.0000 0.9833 0.000 1.000
#> GSM48657 2 0.0000 0.9833 0.000 1.000
#> GSM48658 1 0.9491 0.4856 0.632 0.368
#> GSM25624 2 0.0000 0.9833 0.000 1.000
#> GSM25625 1 0.0000 0.9217 1.000 0.000
#> GSM25626 1 0.0000 0.9217 1.000 0.000
#> GSM25627 1 0.8763 0.6150 0.704 0.296
#> GSM25628 1 0.0000 0.9217 1.000 0.000
#> GSM25629 1 0.0000 0.9217 1.000 0.000
#> GSM25630 1 0.0000 0.9217 1.000 0.000
#> GSM25631 1 0.9983 0.2005 0.524 0.476
#> GSM25632 1 0.0000 0.9217 1.000 0.000
#> GSM25633 1 0.0000 0.9217 1.000 0.000
#> GSM25634 1 0.5294 0.8285 0.880 0.120
#> GSM25635 1 0.0376 0.9198 0.996 0.004
#> GSM25656 1 0.0000 0.9217 1.000 0.000
#> GSM25657 1 0.0000 0.9217 1.000 0.000
#> GSM25658 2 0.2603 0.9359 0.044 0.956
#> GSM25659 2 0.0000 0.9833 0.000 1.000
#> GSM25660 1 0.9608 0.4512 0.616 0.384
#> GSM25661 2 0.0938 0.9714 0.012 0.988
#> GSM25662 2 0.0000 0.9833 0.000 1.000
#> GSM25663 2 0.9209 0.4188 0.336 0.664
#> GSM25680 2 0.0000 0.9833 0.000 1.000
#> GSM25681 2 0.0000 0.9833 0.000 1.000
#> GSM25682 2 0.0000 0.9833 0.000 1.000
#> GSM25683 2 0.0000 0.9833 0.000 1.000
#> GSM25684 2 0.0000 0.9833 0.000 1.000
#> GSM25685 2 0.0000 0.9833 0.000 1.000
#> GSM25686 2 0.0000 0.9833 0.000 1.000
#> GSM25687 2 0.0000 0.9833 0.000 1.000
#> GSM48664 2 0.0000 0.9833 0.000 1.000
#> GSM48665 2 0.0000 0.9833 0.000 1.000
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM25548 2 0.0000 0.955 0.000 1.000 0.000
#> GSM25549 2 0.1529 0.946 0.040 0.960 0.000
#> GSM25550 2 0.1964 0.937 0.056 0.944 0.000
#> GSM25551 2 0.0000 0.955 0.000 1.000 0.000
#> GSM25570 2 0.0747 0.954 0.016 0.984 0.000
#> GSM25571 2 0.0000 0.955 0.000 1.000 0.000
#> GSM25358 3 0.0661 0.943 0.004 0.008 0.988
#> GSM25359 3 0.0475 0.944 0.004 0.004 0.992
#> GSM25360 3 0.0000 0.944 0.000 0.000 1.000
#> GSM25361 1 0.5560 0.597 0.700 0.000 0.300
#> GSM25377 1 0.0424 0.913 0.992 0.008 0.000
#> GSM25378 2 0.5465 0.646 0.288 0.712 0.000
#> GSM25401 1 0.0747 0.910 0.984 0.016 0.000
#> GSM25402 1 0.1163 0.904 0.972 0.028 0.000
#> GSM25349 2 0.3482 0.879 0.128 0.872 0.000
#> GSM25350 2 0.4062 0.841 0.164 0.836 0.000
#> GSM25356 2 0.5591 0.616 0.304 0.696 0.000
#> GSM25357 2 0.0000 0.955 0.000 1.000 0.000
#> GSM25385 3 0.0237 0.944 0.004 0.000 0.996
#> GSM25386 3 0.0000 0.944 0.000 0.000 1.000
#> GSM25399 1 0.0237 0.915 0.996 0.000 0.004
#> GSM25400 1 0.1163 0.911 0.972 0.000 0.028
#> GSM48659 2 0.0237 0.954 0.000 0.996 0.004
#> GSM48660 2 0.0747 0.954 0.016 0.984 0.000
#> GSM25409 1 0.2165 0.878 0.936 0.064 0.000
#> GSM25410 3 0.0424 0.942 0.000 0.008 0.992
#> GSM25426 2 0.0000 0.955 0.000 1.000 0.000
#> GSM25427 1 0.3879 0.791 0.848 0.152 0.000
#> GSM25540 3 0.1529 0.925 0.040 0.000 0.960
#> GSM25541 1 0.3116 0.855 0.892 0.000 0.108
#> GSM25542 3 0.2261 0.903 0.000 0.068 0.932
#> GSM25543 3 0.0747 0.938 0.000 0.016 0.984
#> GSM25479 1 0.0000 0.915 1.000 0.000 0.000
#> GSM25480 1 0.0000 0.915 1.000 0.000 0.000
#> GSM25481 2 0.2537 0.921 0.080 0.920 0.000
#> GSM25482 2 0.1964 0.937 0.056 0.944 0.000
#> GSM48654 2 0.1529 0.930 0.000 0.960 0.040
#> GSM48650 2 0.1289 0.949 0.032 0.968 0.000
#> GSM48651 2 0.1289 0.949 0.032 0.968 0.000
#> GSM48652 2 0.0424 0.955 0.008 0.992 0.000
#> GSM48653 2 0.0747 0.954 0.016 0.984 0.000
#> GSM48662 2 0.0747 0.954 0.016 0.984 0.000
#> GSM48663 2 0.0892 0.953 0.020 0.980 0.000
#> GSM25524 1 0.6291 0.154 0.532 0.000 0.468
#> GSM25525 1 0.0237 0.914 0.996 0.004 0.000
#> GSM25526 3 0.0424 0.943 0.008 0.000 0.992
#> GSM25527 1 0.2448 0.883 0.924 0.000 0.076
#> GSM25528 3 0.3412 0.844 0.124 0.000 0.876
#> GSM25529 1 0.1860 0.898 0.948 0.000 0.052
#> GSM25530 3 0.1643 0.922 0.044 0.000 0.956
#> GSM25531 1 0.5988 0.457 0.632 0.000 0.368
#> GSM48661 3 0.4796 0.738 0.000 0.220 0.780
#> GSM25561 3 0.0424 0.943 0.008 0.000 0.992
#> GSM25562 1 0.0000 0.915 1.000 0.000 0.000
#> GSM25563 3 0.0000 0.944 0.000 0.000 1.000
#> GSM25564 1 0.5621 0.516 0.692 0.308 0.000
#> GSM25565 2 0.1643 0.944 0.044 0.956 0.000
#> GSM25566 2 0.4062 0.840 0.164 0.836 0.000
#> GSM25568 2 0.1411 0.933 0.000 0.964 0.036
#> GSM25569 2 0.0237 0.955 0.004 0.996 0.000
#> GSM25552 1 0.2165 0.878 0.936 0.064 0.000
#> GSM25553 1 0.1163 0.903 0.972 0.028 0.000
#> GSM25578 1 0.0892 0.912 0.980 0.000 0.020
#> GSM25579 1 0.0000 0.915 1.000 0.000 0.000
#> GSM25580 1 0.0892 0.912 0.980 0.000 0.020
#> GSM25581 1 0.2066 0.893 0.940 0.000 0.060
#> GSM48655 2 0.0237 0.954 0.000 0.996 0.004
#> GSM48656 2 0.1170 0.950 0.008 0.976 0.016
#> GSM48657 2 0.0424 0.955 0.008 0.992 0.000
#> GSM48658 3 0.4178 0.795 0.000 0.172 0.828
#> GSM25624 2 0.3686 0.867 0.140 0.860 0.000
#> GSM25625 3 0.0237 0.943 0.000 0.004 0.996
#> GSM25626 3 0.0000 0.944 0.000 0.000 1.000
#> GSM25627 3 0.2261 0.900 0.000 0.068 0.932
#> GSM25628 3 0.0000 0.944 0.000 0.000 1.000
#> GSM25629 3 0.0747 0.940 0.016 0.000 0.984
#> GSM25630 3 0.0237 0.944 0.004 0.000 0.996
#> GSM25631 3 0.6696 0.391 0.348 0.020 0.632
#> GSM25632 3 0.0000 0.944 0.000 0.000 1.000
#> GSM25633 1 0.2448 0.882 0.924 0.000 0.076
#> GSM25634 1 0.1411 0.906 0.964 0.000 0.036
#> GSM25635 3 0.2796 0.880 0.092 0.000 0.908
#> GSM25656 3 0.0237 0.944 0.004 0.000 0.996
#> GSM25657 1 0.1289 0.908 0.968 0.000 0.032
#> GSM25658 1 0.0000 0.915 1.000 0.000 0.000
#> GSM25659 1 0.0237 0.914 0.996 0.004 0.000
#> GSM25660 1 0.1031 0.911 0.976 0.000 0.024
#> GSM25661 1 0.0000 0.915 1.000 0.000 0.000
#> GSM25662 2 0.0424 0.952 0.000 0.992 0.008
#> GSM25663 1 0.6001 0.751 0.772 0.052 0.176
#> GSM25680 2 0.0000 0.955 0.000 1.000 0.000
#> GSM25681 2 0.0000 0.955 0.000 1.000 0.000
#> GSM25682 2 0.0000 0.955 0.000 1.000 0.000
#> GSM25683 2 0.0237 0.954 0.000 0.996 0.004
#> GSM25684 2 0.0000 0.955 0.000 1.000 0.000
#> GSM25685 2 0.0747 0.947 0.000 0.984 0.016
#> GSM25686 2 0.0000 0.955 0.000 1.000 0.000
#> GSM25687 2 0.0000 0.955 0.000 1.000 0.000
#> GSM48664 1 0.0000 0.915 1.000 0.000 0.000
#> GSM48665 1 0.0000 0.915 1.000 0.000 0.000
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM25548 2 0.1474 0.917 0.000 0.948 0.000 0.052
#> GSM25549 2 0.2859 0.869 0.008 0.880 0.000 0.112
#> GSM25550 2 0.1975 0.918 0.016 0.936 0.000 0.048
#> GSM25551 2 0.0921 0.919 0.000 0.972 0.000 0.028
#> GSM25570 2 0.2125 0.904 0.004 0.920 0.000 0.076
#> GSM25571 2 0.1022 0.920 0.000 0.968 0.000 0.032
#> GSM25358 2 0.4050 0.757 0.016 0.824 0.148 0.012
#> GSM25359 2 0.5630 0.345 0.012 0.600 0.376 0.012
#> GSM25360 3 0.0188 0.900 0.004 0.000 0.996 0.000
#> GSM25361 3 0.4872 0.414 0.356 0.000 0.640 0.004
#> GSM25377 1 0.0707 0.886 0.980 0.000 0.000 0.020
#> GSM25378 2 0.1406 0.885 0.024 0.960 0.000 0.016
#> GSM25401 1 0.0779 0.887 0.980 0.004 0.000 0.016
#> GSM25402 2 0.3829 0.762 0.152 0.828 0.004 0.016
#> GSM25349 4 0.1305 0.867 0.036 0.004 0.000 0.960
#> GSM25350 4 0.1398 0.865 0.040 0.004 0.000 0.956
#> GSM25356 2 0.0921 0.895 0.028 0.972 0.000 0.000
#> GSM25357 2 0.1022 0.920 0.000 0.968 0.000 0.032
#> GSM25385 3 0.1821 0.879 0.008 0.032 0.948 0.012
#> GSM25386 3 0.0376 0.900 0.004 0.004 0.992 0.000
#> GSM25399 1 0.0657 0.888 0.984 0.000 0.004 0.012
#> GSM25400 1 0.5546 0.465 0.620 0.356 0.008 0.016
#> GSM48659 2 0.4837 0.451 0.000 0.648 0.004 0.348
#> GSM48660 4 0.3105 0.860 0.004 0.140 0.000 0.856
#> GSM25409 1 0.3215 0.827 0.876 0.032 0.000 0.092
#> GSM25410 3 0.0927 0.893 0.000 0.016 0.976 0.008
#> GSM25426 2 0.0817 0.918 0.000 0.976 0.000 0.024
#> GSM25427 2 0.2924 0.825 0.100 0.884 0.000 0.016
#> GSM25540 3 0.0469 0.899 0.012 0.000 0.988 0.000
#> GSM25541 1 0.3048 0.823 0.876 0.000 0.108 0.016
#> GSM25542 3 0.1004 0.889 0.000 0.004 0.972 0.024
#> GSM25543 3 0.4898 0.221 0.000 0.000 0.584 0.416
#> GSM25479 1 0.0188 0.887 0.996 0.000 0.004 0.000
#> GSM25480 1 0.0376 0.888 0.992 0.000 0.004 0.004
#> GSM25481 2 0.2282 0.911 0.024 0.924 0.000 0.052
#> GSM25482 2 0.2522 0.903 0.016 0.908 0.000 0.076
#> GSM48654 4 0.2224 0.876 0.000 0.032 0.040 0.928
#> GSM48650 4 0.1545 0.888 0.008 0.040 0.000 0.952
#> GSM48651 4 0.2773 0.877 0.004 0.116 0.000 0.880
#> GSM48652 4 0.1978 0.892 0.004 0.068 0.000 0.928
#> GSM48653 4 0.2125 0.891 0.004 0.076 0.000 0.920
#> GSM48662 4 0.2611 0.886 0.008 0.096 0.000 0.896
#> GSM48663 4 0.2831 0.874 0.004 0.120 0.000 0.876
#> GSM25524 1 0.7913 0.344 0.484 0.308 0.192 0.016
#> GSM25525 1 0.4569 0.696 0.760 0.220 0.008 0.012
#> GSM25526 3 0.0672 0.897 0.008 0.000 0.984 0.008
#> GSM25527 1 0.1151 0.884 0.968 0.000 0.024 0.008
#> GSM25528 3 0.4000 0.778 0.144 0.016 0.828 0.012
#> GSM25529 1 0.1262 0.882 0.968 0.008 0.016 0.008
#> GSM25530 3 0.2271 0.868 0.052 0.008 0.928 0.012
#> GSM25531 1 0.5270 0.505 0.660 0.008 0.320 0.012
#> GSM48661 4 0.4454 0.530 0.000 0.000 0.308 0.692
#> GSM25561 3 0.0524 0.898 0.008 0.000 0.988 0.004
#> GSM25562 1 0.1118 0.883 0.964 0.000 0.000 0.036
#> GSM25563 3 0.0188 0.900 0.004 0.000 0.996 0.000
#> GSM25564 1 0.5668 0.195 0.532 0.024 0.000 0.444
#> GSM25565 4 0.3668 0.808 0.004 0.188 0.000 0.808
#> GSM25566 4 0.4829 0.797 0.068 0.156 0.000 0.776
#> GSM25568 4 0.2742 0.889 0.000 0.076 0.024 0.900
#> GSM25569 4 0.2773 0.878 0.000 0.116 0.004 0.880
#> GSM25552 1 0.2032 0.872 0.936 0.028 0.000 0.036
#> GSM25553 1 0.1389 0.878 0.952 0.000 0.000 0.048
#> GSM25578 1 0.0336 0.887 0.992 0.000 0.008 0.000
#> GSM25579 1 0.0657 0.888 0.984 0.000 0.004 0.012
#> GSM25580 1 0.0672 0.887 0.984 0.000 0.008 0.008
#> GSM25581 1 0.0895 0.885 0.976 0.000 0.020 0.004
#> GSM48655 4 0.3142 0.867 0.000 0.132 0.008 0.860
#> GSM48656 4 0.1398 0.863 0.004 0.000 0.040 0.956
#> GSM48657 4 0.0895 0.882 0.004 0.020 0.000 0.976
#> GSM48658 4 0.3791 0.704 0.004 0.000 0.200 0.796
#> GSM25624 4 0.1610 0.862 0.016 0.000 0.032 0.952
#> GSM25625 3 0.0336 0.898 0.000 0.000 0.992 0.008
#> GSM25626 3 0.0336 0.898 0.000 0.000 0.992 0.008
#> GSM25627 4 0.4053 0.662 0.004 0.000 0.228 0.768
#> GSM25628 3 0.0336 0.898 0.000 0.000 0.992 0.008
#> GSM25629 3 0.3539 0.765 0.004 0.000 0.820 0.176
#> GSM25630 3 0.0376 0.900 0.004 0.000 0.992 0.004
#> GSM25631 3 0.6050 0.222 0.044 0.000 0.524 0.432
#> GSM25632 3 0.0376 0.900 0.004 0.000 0.992 0.004
#> GSM25633 1 0.3548 0.830 0.864 0.000 0.068 0.068
#> GSM25634 1 0.4907 0.745 0.764 0.000 0.060 0.176
#> GSM25635 3 0.3143 0.830 0.024 0.000 0.876 0.100
#> GSM25656 3 0.0188 0.899 0.000 0.000 0.996 0.004
#> GSM25657 1 0.1042 0.886 0.972 0.000 0.008 0.020
#> GSM25658 1 0.0859 0.884 0.980 0.004 0.008 0.008
#> GSM25659 1 0.0188 0.888 0.996 0.000 0.000 0.004
#> GSM25660 1 0.0992 0.883 0.976 0.012 0.008 0.004
#> GSM25661 1 0.0592 0.887 0.984 0.000 0.000 0.016
#> GSM25662 2 0.1302 0.919 0.000 0.956 0.000 0.044
#> GSM25663 1 0.7377 0.422 0.552 0.300 0.132 0.016
#> GSM25680 2 0.1211 0.920 0.000 0.960 0.000 0.040
#> GSM25681 2 0.1022 0.920 0.000 0.968 0.000 0.032
#> GSM25682 2 0.1302 0.919 0.000 0.956 0.000 0.044
#> GSM25683 2 0.0817 0.918 0.000 0.976 0.000 0.024
#> GSM25684 2 0.1389 0.918 0.000 0.952 0.000 0.048
#> GSM25685 2 0.1022 0.920 0.000 0.968 0.000 0.032
#> GSM25686 2 0.1716 0.912 0.000 0.936 0.000 0.064
#> GSM25687 2 0.1716 0.912 0.000 0.936 0.000 0.064
#> GSM48664 1 0.1022 0.884 0.968 0.000 0.000 0.032
#> GSM48665 1 0.1389 0.879 0.952 0.000 0.000 0.048
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM25548 4 0.2331 0.86257 0.000 0.020 0.000 0.900 0.080
#> GSM25549 4 0.2969 0.83798 0.000 0.128 0.000 0.852 0.020
#> GSM25550 4 0.2331 0.86626 0.000 0.080 0.000 0.900 0.020
#> GSM25551 4 0.1638 0.85503 0.000 0.004 0.000 0.932 0.064
#> GSM25570 4 0.2580 0.86934 0.000 0.064 0.000 0.892 0.044
#> GSM25571 4 0.2006 0.86137 0.000 0.012 0.000 0.916 0.072
#> GSM25358 4 0.5601 0.68425 0.032 0.028 0.104 0.736 0.100
#> GSM25359 3 0.4185 0.62462 0.008 0.008 0.732 0.248 0.004
#> GSM25360 3 0.0290 0.91393 0.000 0.000 0.992 0.000 0.008
#> GSM25361 3 0.3516 0.78392 0.108 0.052 0.836 0.000 0.004
#> GSM25377 1 0.3403 0.55921 0.820 0.160 0.000 0.008 0.012
#> GSM25378 4 0.4405 0.75048 0.052 0.028 0.000 0.788 0.132
#> GSM25401 1 0.5594 0.44928 0.696 0.192 0.004 0.072 0.036
#> GSM25402 4 0.4699 0.67107 0.216 0.036 0.008 0.732 0.008
#> GSM25349 2 0.2011 0.66426 0.044 0.928 0.000 0.020 0.008
#> GSM25350 2 0.2228 0.66137 0.056 0.916 0.000 0.020 0.008
#> GSM25356 4 0.2720 0.84831 0.020 0.096 0.000 0.880 0.004
#> GSM25357 4 0.1544 0.86824 0.000 0.068 0.000 0.932 0.000
#> GSM25385 3 0.0486 0.91404 0.004 0.000 0.988 0.004 0.004
#> GSM25386 3 0.0727 0.91227 0.004 0.000 0.980 0.004 0.012
#> GSM25399 1 0.2623 0.59640 0.884 0.096 0.004 0.000 0.016
#> GSM25400 1 0.5560 0.14038 0.508 0.024 0.000 0.440 0.028
#> GSM48659 4 0.6114 0.14445 0.000 0.376 0.000 0.492 0.132
#> GSM48660 2 0.3563 0.65417 0.000 0.780 0.000 0.208 0.012
#> GSM25409 2 0.6381 0.13791 0.428 0.436 0.000 0.128 0.008
#> GSM25410 3 0.0727 0.91227 0.004 0.000 0.980 0.004 0.012
#> GSM25426 4 0.2068 0.85771 0.000 0.092 0.000 0.904 0.004
#> GSM25427 4 0.3518 0.80434 0.104 0.048 0.000 0.840 0.008
#> GSM25540 3 0.0693 0.91306 0.008 0.000 0.980 0.000 0.012
#> GSM25541 1 0.4185 0.61766 0.816 0.040 0.072 0.000 0.072
#> GSM25542 3 0.1082 0.90223 0.000 0.008 0.964 0.000 0.028
#> GSM25543 3 0.4394 0.69219 0.016 0.196 0.756 0.000 0.032
#> GSM25479 1 0.4589 0.53250 0.660 0.004 0.000 0.020 0.316
#> GSM25480 1 0.4669 0.53633 0.664 0.008 0.000 0.020 0.308
#> GSM25481 4 0.3360 0.78609 0.012 0.168 0.000 0.816 0.004
#> GSM25482 4 0.2583 0.83233 0.000 0.132 0.000 0.864 0.004
#> GSM48654 2 0.5208 0.15133 0.000 0.544 0.024 0.012 0.420
#> GSM48650 2 0.1992 0.68217 0.000 0.924 0.000 0.044 0.032
#> GSM48651 2 0.3053 0.67639 0.000 0.828 0.000 0.164 0.008
#> GSM48652 2 0.3800 0.65746 0.000 0.812 0.000 0.080 0.108
#> GSM48653 2 0.3090 0.68452 0.000 0.860 0.000 0.088 0.052
#> GSM48662 2 0.2124 0.69162 0.000 0.900 0.000 0.096 0.004
#> GSM48663 2 0.3461 0.67041 0.016 0.812 0.000 0.168 0.004
#> GSM25524 1 0.7588 0.17003 0.388 0.028 0.012 0.352 0.220
#> GSM25525 5 0.6961 -0.01358 0.300 0.020 0.000 0.208 0.472
#> GSM25526 5 0.5884 0.40139 0.012 0.016 0.284 0.064 0.624
#> GSM25527 5 0.4249 0.34493 0.296 0.000 0.000 0.016 0.688
#> GSM25528 3 0.6000 0.16492 0.360 0.008 0.536 0.000 0.096
#> GSM25529 1 0.5104 0.50481 0.632 0.016 0.000 0.028 0.324
#> GSM25530 3 0.0898 0.91035 0.020 0.000 0.972 0.000 0.008
#> GSM25531 1 0.5365 0.55480 0.708 0.008 0.112 0.008 0.164
#> GSM48661 5 0.5663 0.14059 0.000 0.412 0.080 0.000 0.508
#> GSM25561 3 0.0324 0.91459 0.004 0.000 0.992 0.004 0.000
#> GSM25562 1 0.2648 0.57393 0.848 0.152 0.000 0.000 0.000
#> GSM25563 3 0.0324 0.91459 0.004 0.000 0.992 0.004 0.000
#> GSM25564 2 0.5827 0.21582 0.408 0.520 0.000 0.052 0.020
#> GSM25565 2 0.5605 0.57595 0.132 0.660 0.000 0.200 0.008
#> GSM25566 2 0.5716 0.56222 0.156 0.660 0.000 0.172 0.012
#> GSM25568 2 0.4706 0.63277 0.000 0.768 0.024 0.080 0.128
#> GSM25569 2 0.3865 0.66720 0.000 0.808 0.000 0.100 0.092
#> GSM25552 1 0.6284 0.14422 0.544 0.316 0.000 0.128 0.012
#> GSM25553 1 0.5129 0.38676 0.672 0.264 0.000 0.052 0.012
#> GSM25578 1 0.3890 0.59231 0.736 0.000 0.000 0.012 0.252
#> GSM25579 1 0.3352 0.62070 0.800 0.004 0.000 0.004 0.192
#> GSM25580 1 0.4151 0.50901 0.652 0.000 0.000 0.004 0.344
#> GSM25581 1 0.4059 0.56171 0.700 0.000 0.004 0.004 0.292
#> GSM48655 2 0.5770 0.32107 0.000 0.532 0.000 0.096 0.372
#> GSM48656 2 0.4557 -0.00899 0.000 0.516 0.008 0.000 0.476
#> GSM48657 2 0.4201 0.19701 0.000 0.592 0.000 0.000 0.408
#> GSM48658 5 0.4677 0.46663 0.000 0.300 0.036 0.000 0.664
#> GSM25624 5 0.3365 0.59693 0.008 0.180 0.004 0.000 0.808
#> GSM25625 3 0.0880 0.90189 0.000 0.000 0.968 0.000 0.032
#> GSM25626 3 0.0510 0.91136 0.000 0.000 0.984 0.000 0.016
#> GSM25627 5 0.4822 0.48241 0.000 0.288 0.048 0.000 0.664
#> GSM25628 3 0.0162 0.91438 0.000 0.000 0.996 0.000 0.004
#> GSM25629 5 0.5128 0.60210 0.008 0.168 0.112 0.000 0.712
#> GSM25630 3 0.0162 0.91474 0.000 0.000 0.996 0.000 0.004
#> GSM25631 5 0.3530 0.64383 0.024 0.104 0.028 0.000 0.844
#> GSM25632 3 0.0609 0.90967 0.000 0.000 0.980 0.000 0.020
#> GSM25633 5 0.4735 0.38058 0.304 0.024 0.008 0.000 0.664
#> GSM25634 5 0.4952 0.52744 0.216 0.068 0.008 0.000 0.708
#> GSM25635 5 0.3105 0.61457 0.064 0.012 0.036 0.008 0.880
#> GSM25656 3 0.0290 0.91440 0.000 0.000 0.992 0.000 0.008
#> GSM25657 1 0.2879 0.63393 0.868 0.032 0.000 0.000 0.100
#> GSM25658 1 0.4573 0.59522 0.728 0.020 0.000 0.024 0.228
#> GSM25659 1 0.3966 0.60576 0.756 0.008 0.000 0.012 0.224
#> GSM25660 1 0.4839 0.53409 0.660 0.012 0.000 0.024 0.304
#> GSM25661 1 0.2344 0.63288 0.904 0.032 0.000 0.000 0.064
#> GSM25662 4 0.2873 0.81994 0.000 0.016 0.000 0.856 0.128
#> GSM25663 1 0.7752 0.08073 0.340 0.016 0.328 0.292 0.024
#> GSM25680 4 0.2193 0.87220 0.000 0.060 0.000 0.912 0.028
#> GSM25681 4 0.1522 0.86647 0.000 0.012 0.000 0.944 0.044
#> GSM25682 4 0.1430 0.87150 0.000 0.052 0.000 0.944 0.004
#> GSM25683 4 0.0963 0.87191 0.000 0.036 0.000 0.964 0.000
#> GSM25684 4 0.2172 0.86141 0.000 0.016 0.000 0.908 0.076
#> GSM25685 4 0.2818 0.81578 0.000 0.012 0.000 0.856 0.132
#> GSM25686 4 0.1544 0.86809 0.000 0.068 0.000 0.932 0.000
#> GSM25687 4 0.1704 0.86820 0.000 0.068 0.000 0.928 0.004
#> GSM48664 1 0.1952 0.60923 0.912 0.084 0.000 0.000 0.004
#> GSM48665 1 0.2423 0.61818 0.896 0.080 0.000 0.000 0.024
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM25548 5 0.4935 0.1527 0.000 0.040 0.000 0.012 0.524 0.424
#> GSM25549 6 0.4280 0.5495 0.000 0.044 0.000 0.012 0.228 0.716
#> GSM25550 6 0.4760 0.2850 0.000 0.040 0.000 0.008 0.376 0.576
#> GSM25551 5 0.2294 0.7092 0.000 0.008 0.000 0.020 0.896 0.076
#> GSM25570 6 0.4845 0.2316 0.000 0.044 0.000 0.008 0.388 0.560
#> GSM25571 5 0.4505 0.4958 0.000 0.032 0.000 0.020 0.676 0.272
#> GSM25358 5 0.3911 0.6353 0.016 0.000 0.040 0.072 0.820 0.052
#> GSM25359 3 0.3192 0.7298 0.000 0.000 0.776 0.004 0.216 0.004
#> GSM25360 3 0.1074 0.9303 0.000 0.000 0.960 0.028 0.000 0.012
#> GSM25361 6 0.4154 0.5102 0.008 0.004 0.244 0.028 0.000 0.716
#> GSM25377 1 0.3342 0.5550 0.836 0.092 0.000 0.008 0.004 0.060
#> GSM25378 5 0.4339 0.6181 0.048 0.008 0.000 0.092 0.784 0.068
#> GSM25401 1 0.3573 0.5292 0.832 0.100 0.012 0.004 0.012 0.040
#> GSM25402 1 0.5209 0.4378 0.696 0.044 0.016 0.028 0.204 0.012
#> GSM25349 2 0.3545 0.7011 0.036 0.824 0.000 0.012 0.012 0.116
#> GSM25350 2 0.5237 0.1998 0.032 0.548 0.000 0.020 0.012 0.388
#> GSM25356 5 0.4223 0.6647 0.124 0.064 0.000 0.008 0.780 0.024
#> GSM25357 5 0.1777 0.7200 0.000 0.044 0.000 0.004 0.928 0.024
#> GSM25385 3 0.2118 0.9060 0.020 0.000 0.920 0.016 0.036 0.008
#> GSM25386 3 0.0436 0.9314 0.004 0.000 0.988 0.004 0.004 0.000
#> GSM25399 1 0.2186 0.5931 0.916 0.016 0.008 0.008 0.004 0.048
#> GSM25400 1 0.4108 0.5576 0.780 0.004 0.012 0.048 0.148 0.008
#> GSM48659 2 0.4500 0.5254 0.000 0.688 0.000 0.028 0.256 0.028
#> GSM48660 2 0.1931 0.7529 0.008 0.916 0.000 0.004 0.068 0.004
#> GSM25409 6 0.6427 0.5270 0.120 0.212 0.000 0.004 0.096 0.568
#> GSM25410 3 0.0964 0.9272 0.004 0.000 0.968 0.016 0.012 0.000
#> GSM25426 5 0.2630 0.7007 0.012 0.088 0.000 0.012 0.880 0.008
#> GSM25427 5 0.4764 0.3433 0.380 0.020 0.000 0.012 0.580 0.008
#> GSM25540 3 0.1788 0.9000 0.004 0.000 0.916 0.004 0.000 0.076
#> GSM25541 6 0.5304 0.4367 0.060 0.000 0.152 0.104 0.000 0.684
#> GSM25542 3 0.0748 0.9296 0.000 0.004 0.976 0.016 0.000 0.004
#> GSM25543 3 0.2638 0.8713 0.020 0.052 0.892 0.008 0.000 0.028
#> GSM25479 1 0.5456 0.2451 0.452 0.000 0.000 0.440 0.004 0.104
#> GSM25480 4 0.5646 -0.3130 0.436 0.000 0.000 0.440 0.008 0.116
#> GSM25481 5 0.6168 0.3782 0.164 0.300 0.000 0.008 0.512 0.016
#> GSM25482 5 0.5292 0.5231 0.064 0.272 0.000 0.008 0.632 0.024
#> GSM48654 2 0.3672 0.5858 0.000 0.712 0.008 0.276 0.000 0.004
#> GSM48650 2 0.1078 0.7603 0.012 0.964 0.000 0.016 0.000 0.008
#> GSM48651 2 0.2317 0.7427 0.020 0.900 0.000 0.000 0.064 0.016
#> GSM48652 2 0.1296 0.7622 0.000 0.952 0.000 0.032 0.012 0.004
#> GSM48653 2 0.1237 0.7650 0.000 0.956 0.000 0.020 0.020 0.004
#> GSM48662 2 0.1269 0.7604 0.012 0.956 0.000 0.000 0.020 0.012
#> GSM48663 2 0.3159 0.7173 0.064 0.856 0.000 0.004 0.060 0.016
#> GSM25524 1 0.6849 0.3290 0.468 0.004 0.004 0.220 0.256 0.048
#> GSM25525 4 0.6462 0.1976 0.196 0.000 0.000 0.552 0.164 0.088
#> GSM25526 4 0.6126 0.4429 0.072 0.000 0.104 0.664 0.096 0.064
#> GSM25527 4 0.3501 0.5274 0.128 0.000 0.004 0.816 0.008 0.044
#> GSM25528 1 0.5562 0.4533 0.616 0.000 0.184 0.184 0.004 0.012
#> GSM25529 1 0.4797 0.3412 0.524 0.000 0.000 0.432 0.008 0.036
#> GSM25530 3 0.4454 0.6568 0.232 0.000 0.708 0.044 0.008 0.008
#> GSM25531 1 0.4009 0.5916 0.764 0.000 0.040 0.180 0.004 0.012
#> GSM48661 2 0.4878 0.1509 0.000 0.480 0.040 0.472 0.000 0.008
#> GSM25561 3 0.0405 0.9314 0.004 0.000 0.988 0.000 0.008 0.000
#> GSM25562 1 0.3505 0.5799 0.824 0.068 0.000 0.016 0.000 0.092
#> GSM25563 3 0.0146 0.9318 0.000 0.000 0.996 0.000 0.004 0.000
#> GSM25564 2 0.6083 0.4712 0.232 0.612 0.000 0.036 0.040 0.080
#> GSM25565 6 0.6083 0.5433 0.036 0.196 0.012 0.004 0.140 0.612
#> GSM25566 6 0.4752 0.6101 0.028 0.120 0.000 0.000 0.128 0.724
#> GSM25568 2 0.2022 0.7579 0.000 0.916 0.000 0.052 0.024 0.008
#> GSM25569 2 0.2024 0.7640 0.000 0.920 0.000 0.028 0.036 0.016
#> GSM25552 6 0.3237 0.6557 0.012 0.036 0.000 0.020 0.076 0.856
#> GSM25553 6 0.3280 0.6562 0.020 0.040 0.000 0.024 0.056 0.860
#> GSM25578 1 0.4756 0.5018 0.628 0.000 0.000 0.304 0.004 0.064
#> GSM25579 6 0.5658 -0.2276 0.316 0.000 0.000 0.176 0.000 0.508
#> GSM25580 1 0.5301 0.3104 0.492 0.000 0.000 0.416 0.004 0.088
#> GSM25581 1 0.5304 0.3592 0.516 0.000 0.000 0.388 0.004 0.092
#> GSM48655 2 0.4150 0.6341 0.000 0.724 0.000 0.228 0.036 0.012
#> GSM48656 2 0.4437 0.2842 0.000 0.540 0.004 0.436 0.000 0.020
#> GSM48657 2 0.3468 0.6056 0.000 0.728 0.000 0.264 0.000 0.008
#> GSM48658 4 0.4060 0.3515 0.000 0.296 0.016 0.680 0.000 0.008
#> GSM25624 4 0.3426 0.4916 0.004 0.220 0.000 0.764 0.000 0.012
#> GSM25625 3 0.1802 0.9114 0.000 0.000 0.916 0.072 0.000 0.012
#> GSM25626 3 0.1010 0.9301 0.000 0.000 0.960 0.036 0.000 0.004
#> GSM25627 4 0.4004 0.2807 0.000 0.328 0.012 0.656 0.000 0.004
#> GSM25628 3 0.0820 0.9320 0.000 0.000 0.972 0.016 0.000 0.012
#> GSM25629 4 0.4909 0.5472 0.000 0.128 0.088 0.724 0.000 0.060
#> GSM25630 3 0.0717 0.9323 0.000 0.000 0.976 0.016 0.000 0.008
#> GSM25631 4 0.4448 0.5017 0.008 0.020 0.016 0.684 0.000 0.272
#> GSM25632 3 0.1719 0.9196 0.008 0.000 0.928 0.056 0.000 0.008
#> GSM25633 4 0.3543 0.5459 0.120 0.004 0.008 0.816 0.000 0.052
#> GSM25634 4 0.3594 0.5660 0.104 0.028 0.000 0.820 0.000 0.048
#> GSM25635 4 0.2502 0.5966 0.016 0.020 0.012 0.908 0.012 0.032
#> GSM25656 3 0.0820 0.9320 0.000 0.000 0.972 0.016 0.000 0.012
#> GSM25657 1 0.4535 0.6019 0.704 0.000 0.000 0.148 0.000 0.148
#> GSM25658 1 0.4108 0.5781 0.724 0.000 0.000 0.232 0.012 0.032
#> GSM25659 1 0.6115 0.3415 0.428 0.000 0.000 0.248 0.004 0.320
#> GSM25660 4 0.6243 -0.1240 0.320 0.000 0.000 0.436 0.012 0.232
#> GSM25661 1 0.5007 0.5723 0.648 0.004 0.000 0.124 0.000 0.224
#> GSM25662 5 0.2865 0.6998 0.000 0.012 0.000 0.064 0.868 0.056
#> GSM25663 6 0.3786 0.6359 0.008 0.004 0.028 0.024 0.124 0.812
#> GSM25680 5 0.5061 -0.0190 0.000 0.044 0.004 0.008 0.480 0.464
#> GSM25681 5 0.5081 0.0865 0.000 0.040 0.004 0.012 0.504 0.440
#> GSM25682 5 0.2457 0.7085 0.000 0.036 0.000 0.000 0.880 0.084
#> GSM25683 5 0.1624 0.7202 0.000 0.020 0.000 0.004 0.936 0.040
#> GSM25684 5 0.2628 0.7132 0.000 0.024 0.000 0.024 0.884 0.068
#> GSM25685 5 0.2879 0.6712 0.000 0.008 0.000 0.072 0.864 0.056
#> GSM25686 5 0.2376 0.7158 0.000 0.068 0.000 0.000 0.888 0.044
#> GSM25687 5 0.2801 0.7067 0.000 0.068 0.000 0.000 0.860 0.072
#> GSM48664 1 0.3977 0.5850 0.752 0.016 0.000 0.032 0.000 0.200
#> GSM48665 1 0.4441 0.5875 0.720 0.016 0.000 0.060 0.000 0.204
Heatmaps for the consensus matrix. It visualizes the probability of two samples to be in a same group.
consensus_heatmap(res, k = 2)
consensus_heatmap(res, k = 3)
consensus_heatmap(res, k = 4)
consensus_heatmap(res, k = 5)
consensus_heatmap(res, k = 6)
Heatmaps for the membership of samples in all partitions to see how consistent they are:
membership_heatmap(res, k = 2)
membership_heatmap(res, k = 3)
membership_heatmap(res, k = 4)
membership_heatmap(res, k = 5)
membership_heatmap(res, k = 6)
As soon as we have had the classes for columns, we can look for signatures which are significantly different between classes which can be candidate marks for certain classes. Following are the heatmaps for signatures.
Signature heatmaps where rows are scaled:
get_signatures(res, k = 2)
get_signatures(res, k = 3)
get_signatures(res, k = 4)
get_signatures(res, k = 5)
get_signatures(res, k = 6)
Signature heatmaps where rows are not scaled:
get_signatures(res, k = 2, scale_rows = FALSE)
get_signatures(res, k = 3, scale_rows = FALSE)
get_signatures(res, k = 4, scale_rows = FALSE)
get_signatures(res, k = 5, scale_rows = FALSE)
get_signatures(res, k = 6, scale_rows = FALSE)
Compare the overlap of signatures from different k:
compare_signatures(res)
get_signature() returns a data frame invisibly. TO get the list of signatures, the function
call should be assigned to a variable explicitly. In following code, if plot argument is set
to FALSE, no heatmap is plotted while only the differential analysis is performed.
# code only for demonstration
tb = get_signature(res, k = ..., plot = FALSE)
An example of the output of tb is:
#> which_row fdr mean_1 mean_2 scaled_mean_1 scaled_mean_2 km
#> 1 38 0.042760348 8.373488 9.131774 -0.5533452 0.5164555 1
#> 2 40 0.018707592 7.106213 8.469186 -0.6173731 0.5762149 1
#> 3 55 0.019134737 10.221463 11.207825 -0.6159697 0.5749050 1
#> 4 59 0.006059896 5.921854 7.869574 -0.6899429 0.6439467 1
#> 5 60 0.018055526 8.928898 10.211722 -0.6204761 0.5791110 1
#> 6 98 0.009384629 15.714769 14.887706 0.6635654 -0.6193277 2
...
The columns in tb are:
which_row: row indices corresponding to the input matrix.fdr: FDR for the differential test. mean_x: The mean value in group x.scaled_mean_x: The mean value in group x after rows are scaled.km: Row groups if k-means clustering is applied to rows.UMAP plot which shows how samples are separated.
dimension_reduction(res, k = 2, method = "UMAP")
dimension_reduction(res, k = 3, method = "UMAP")
dimension_reduction(res, k = 4, method = "UMAP")
dimension_reduction(res, k = 5, method = "UMAP")
dimension_reduction(res, k = 6, method = "UMAP")
Following heatmap shows how subgroups are split when increasing k:
collect_classes(res)
Test correlation between subgroups and known annotations. If the known annotation is numeric, one-way ANOVA test is applied, and if the known annotation is discrete, chi-squared contingency table test is applied.
test_to_known_factors(res)
#> n genotype/variation(p) k
#> ATC:NMF 93 7.87e-05 2
#> ATC:NMF 97 4.24e-06 3
#> ATC:NMF 91 2.02e-10 4
#> ATC:NMF 79 1.26e-11 5
#> ATC:NMF 71 5.36e-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.
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