Date: 2019-12-25 22:13:50 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 51941 rows and 84 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] 51941 84
The density distribution for each sample is visualized as in one column in the following heatmap. The clustering is based on the distance which is the Kolmogorov-Smirnov statistic between two distributions.
library(ComplexHeatmap)
densityHeatmap(mat, top_annotation = HeatmapAnnotation(df = get_anno(res_list),
col = get_anno_col(res_list)), ylab = "value", cluster_columns = TRUE, show_column_names = FALSE,
mc.cores = 4)
Folowing table shows the best k (number of partitions) for each combination
of top-value methods and partition methods. Clicking on the method name in
the table goes to the section for a single combination of methods.
The cola vignette explains the definition of the metrics used for determining the best number of partitions.
suggest_best_k(res_list)
| The best k | 1-PAC | Mean silhouette | Concordance | Optional k | ||
|---|---|---|---|---|---|---|
| ATC:skmeans | 3 | 1.000 | 0.983 | 0.993 | ** | 2 |
| CV:NMF | 2 | 0.998 | 0.974 | 0.983 | ** | |
| CV:mclust | 2 | 0.985 | 0.982 | 0.968 | ** | |
| ATC:NMF | 2 | 0.950 | 0.937 | 0.974 | ** | |
| ATC:mclust | 2 | 0.925 | 0.942 | 0.974 | * | |
| ATC:kmeans | 2 | 0.901 | 0.911 | 0.960 | * | |
| ATC:pam | 3 | 0.780 | 0.800 | 0.921 | ||
| CV:kmeans | 2 | 0.543 | 0.927 | 0.924 | ||
| ATC:hclust | 4 | 0.486 | 0.595 | 0.756 | ||
| MAD:kmeans | 2 | 0.360 | 0.844 | 0.863 | ||
| SD:mclust | 3 | 0.265 | 0.795 | 0.819 | ||
| SD:kmeans | 3 | 0.242 | 0.495 | 0.695 | ||
| SD:NMF | 2 | 0.218 | 0.563 | 0.806 | ||
| MAD:NMF | 2 | 0.204 | 0.667 | 0.815 | ||
| SD:pam | 3 | 0.167 | 0.449 | 0.707 | ||
| SD:hclust | 3 | 0.157 | 0.441 | 0.641 | ||
| CV:pam | 2 | 0.136 | 0.623 | 0.809 | ||
| MAD:mclust | 3 | 0.080 | 0.638 | 0.706 | ||
| MAD:hclust | 3 | 0.018 | 0.454 | 0.595 | ||
| MAD:pam | 2 | 0.014 | 0.399 | 0.700 | ||
| CV:hclust | 3 | 0.011 | 0.669 | 0.670 | ||
| SD:skmeans | 2 | 0.007 | 0.307 | 0.653 | ||
| CV:skmeans | 2 | 0.000 | 0.819 | 0.780 | ||
| MAD:skmeans | 2 | 0.000 | 0.637 | 0.713 |
**: 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 0.21811 0.5630 0.806 0.491 0.504 0.504
#> CV:NMF 2 0.99773 0.9739 0.983 0.498 0.504 0.504
#> MAD:NMF 2 0.20415 0.6667 0.815 0.492 0.494 0.494
#> ATC:NMF 2 0.95002 0.9368 0.974 0.479 0.523 0.523
#> SD:skmeans 2 0.00714 0.3068 0.653 0.503 0.494 0.494
#> CV:skmeans 2 0.00000 0.8188 0.780 0.503 0.504 0.504
#> MAD:skmeans 2 0.00000 0.6370 0.713 0.504 0.504 0.504
#> ATC:skmeans 2 1.00000 0.9838 0.994 0.503 0.499 0.499
#> SD:mclust 2 0.29536 0.8665 0.803 0.369 0.504 0.504
#> CV:mclust 2 0.98475 0.9825 0.968 0.477 0.504 0.504
#> MAD:mclust 2 0.08893 0.1185 0.745 0.331 0.953 0.953
#> ATC:mclust 2 0.92535 0.9420 0.974 0.352 0.646 0.646
#> SD:kmeans 2 0.12204 0.3246 0.632 0.449 0.499 0.499
#> CV:kmeans 2 0.54333 0.9272 0.924 0.489 0.504 0.504
#> MAD:kmeans 2 0.35995 0.8439 0.863 0.480 0.523 0.523
#> ATC:kmeans 2 0.90133 0.9109 0.960 0.461 0.535 0.535
#> SD:pam 2 0.05388 0.2387 0.599 0.479 0.512 0.512
#> CV:pam 2 0.13599 0.6228 0.809 0.462 0.523 0.523
#> MAD:pam 2 0.01363 0.3989 0.700 0.455 0.587 0.587
#> ATC:pam 2 0.36157 0.6840 0.859 0.477 0.494 0.494
#> SD:hclust 2 0.06677 0.4900 0.742 0.348 0.826 0.826
#> CV:hclust 2 0.04090 0.0814 0.643 0.348 0.719 0.719
#> MAD:hclust 2 0.02077 0.3228 0.728 0.339 0.845 0.845
#> ATC:hclust 2 0.28205 0.7697 0.827 0.402 0.633 0.633
get_stats(res_list, k = 3)
#> k 1-PAC mean_silhouette concordance area_increased Rand Jaccard
#> SD:NMF 3 0.273288 0.455 0.671 0.287 0.742 0.566
#> CV:NMF 3 0.530996 0.665 0.844 0.217 0.987 0.973
#> MAD:NMF 3 0.235638 0.477 0.719 0.333 0.690 0.455
#> ATC:NMF 3 0.666667 0.749 0.891 0.392 0.677 0.451
#> SD:skmeans 3 0.037975 0.460 0.629 0.335 0.617 0.364
#> CV:skmeans 3 0.000974 0.409 0.593 0.330 0.835 0.673
#> MAD:skmeans 3 0.008114 0.222 0.515 0.331 0.813 0.649
#> ATC:skmeans 3 1.000000 0.983 0.993 0.323 0.752 0.542
#> SD:mclust 3 0.265174 0.795 0.819 0.555 0.884 0.773
#> CV:mclust 3 0.611814 0.627 0.825 0.256 0.950 0.900
#> MAD:mclust 3 0.079844 0.638 0.706 0.618 0.517 0.499
#> ATC:mclust 3 0.519636 0.771 0.854 0.784 0.652 0.483
#> SD:kmeans 3 0.242454 0.495 0.695 0.385 0.807 0.637
#> CV:kmeans 3 0.525154 0.713 0.811 0.302 0.844 0.693
#> MAD:kmeans 3 0.277183 0.506 0.649 0.318 0.937 0.880
#> ATC:kmeans 3 0.789030 0.800 0.917 0.429 0.715 0.504
#> SD:pam 3 0.167478 0.449 0.707 0.350 0.639 0.406
#> CV:pam 3 0.132425 0.345 0.658 0.316 0.902 0.816
#> MAD:pam 3 0.044142 0.419 0.650 0.363 0.672 0.491
#> ATC:pam 3 0.779942 0.800 0.921 0.370 0.663 0.424
#> SD:hclust 3 0.157092 0.441 0.641 0.498 0.629 0.556
#> CV:hclust 3 0.010759 0.669 0.670 0.444 0.518 0.430
#> MAD:hclust 3 0.017527 0.454 0.595 0.578 0.617 0.567
#> ATC:hclust 3 0.365823 0.577 0.748 0.538 0.682 0.514
get_stats(res_list, k = 4)
#> k 1-PAC mean_silhouette concordance area_increased Rand Jaccard
#> SD:NMF 4 0.4106 0.5097 0.700 0.1431 0.783 0.527
#> CV:NMF 4 0.4463 0.5079 0.676 0.1635 0.883 0.763
#> MAD:NMF 4 0.2986 0.3452 0.609 0.1318 0.790 0.470
#> ATC:NMF 4 0.4492 0.4820 0.719 0.1214 0.764 0.419
#> SD:skmeans 4 0.1376 0.4226 0.564 0.1215 0.840 0.562
#> CV:skmeans 4 0.0474 0.3207 0.492 0.1236 0.902 0.747
#> MAD:skmeans 4 0.0409 0.0857 0.376 0.1241 0.761 0.448
#> ATC:skmeans 4 0.7124 0.6298 0.813 0.1128 0.922 0.781
#> SD:mclust 4 0.4492 0.6695 0.741 0.2210 0.865 0.671
#> CV:mclust 4 0.5923 0.7099 0.754 0.1584 0.800 0.573
#> MAD:mclust 4 0.3093 0.5212 0.647 0.2487 0.764 0.523
#> ATC:mclust 4 0.5755 0.5647 0.779 0.1171 0.849 0.611
#> SD:kmeans 4 0.3592 0.4156 0.625 0.1347 0.748 0.452
#> CV:kmeans 4 0.5115 0.5497 0.728 0.1289 0.937 0.831
#> MAD:kmeans 4 0.3457 0.4054 0.609 0.1388 0.743 0.476
#> ATC:kmeans 4 0.6560 0.7293 0.840 0.1326 0.863 0.620
#> SD:pam 4 0.2136 0.3864 0.638 0.0954 0.895 0.714
#> CV:pam 4 0.1870 0.2656 0.612 0.1364 0.863 0.716
#> MAD:pam 4 0.0970 0.3195 0.579 0.1232 0.888 0.728
#> ATC:pam 4 0.6861 0.7290 0.850 0.1462 0.853 0.599
#> SD:hclust 4 0.2301 0.4792 0.638 0.1866 0.869 0.736
#> CV:hclust 4 0.0419 0.4726 0.634 0.2464 0.920 0.836
#> MAD:hclust 4 0.0730 0.4105 0.568 0.1830 0.863 0.752
#> ATC:hclust 4 0.4861 0.5953 0.756 0.1687 0.881 0.681
get_stats(res_list, k = 5)
#> k 1-PAC mean_silhouette concordance area_increased Rand Jaccard
#> SD:NMF 5 0.451 0.307 0.602 0.0825 0.972 0.905
#> CV:NMF 5 0.462 0.359 0.566 0.0898 0.913 0.781
#> MAD:NMF 5 0.351 0.260 0.512 0.0713 0.844 0.489
#> ATC:NMF 5 0.502 0.410 0.662 0.0660 0.838 0.462
#> SD:skmeans 5 0.261 0.302 0.489 0.0651 0.919 0.691
#> CV:skmeans 5 0.163 0.125 0.378 0.0688 0.836 0.563
#> MAD:skmeans 5 0.116 0.123 0.369 0.0656 0.817 0.424
#> ATC:skmeans 5 0.716 0.662 0.816 0.0641 0.867 0.589
#> SD:mclust 5 0.527 0.685 0.733 0.0859 0.816 0.462
#> CV:mclust 5 0.612 0.685 0.781 0.0882 0.953 0.835
#> MAD:mclust 5 0.426 0.467 0.674 0.1177 0.917 0.720
#> ATC:mclust 5 0.630 0.432 0.722 0.0754 0.915 0.725
#> SD:kmeans 5 0.459 0.561 0.651 0.0697 0.871 0.604
#> CV:kmeans 5 0.553 0.487 0.656 0.0746 0.853 0.568
#> MAD:kmeans 5 0.417 0.347 0.591 0.0759 0.818 0.432
#> ATC:kmeans 5 0.622 0.563 0.717 0.0637 0.918 0.698
#> SD:pam 5 0.288 0.383 0.607 0.0482 0.910 0.719
#> CV:pam 5 0.219 0.245 0.577 0.0573 0.882 0.710
#> MAD:pam 5 0.169 0.299 0.553 0.0655 0.947 0.847
#> ATC:pam 5 0.723 0.674 0.816 0.0581 0.921 0.707
#> SD:hclust 5 0.293 0.352 0.598 0.1153 0.764 0.484
#> CV:hclust 5 0.104 0.475 0.615 0.1105 0.894 0.747
#> MAD:hclust 5 0.184 0.360 0.552 0.1025 0.936 0.855
#> ATC:hclust 5 0.549 0.512 0.658 0.0630 0.954 0.839
get_stats(res_list, k = 6)
#> k 1-PAC mean_silhouette concordance area_increased Rand Jaccard
#> SD:NMF 6 0.491 0.306 0.543 0.0504 0.834 0.468
#> CV:NMF 6 0.469 0.176 0.476 0.0631 0.846 0.567
#> MAD:NMF 6 0.420 0.237 0.482 0.0453 0.871 0.497
#> ATC:NMF 6 0.542 0.362 0.623 0.0416 0.826 0.368
#> SD:skmeans 6 0.382 0.193 0.437 0.0411 0.907 0.613
#> CV:skmeans 6 0.291 0.104 0.376 0.0428 0.869 0.534
#> MAD:skmeans 6 0.220 0.095 0.322 0.0419 0.851 0.426
#> ATC:skmeans 6 0.734 0.587 0.753 0.0410 0.949 0.777
#> SD:mclust 6 0.677 0.671 0.796 0.0658 0.934 0.707
#> CV:mclust 6 0.640 0.683 0.761 0.0563 0.953 0.807
#> MAD:mclust 6 0.497 0.463 0.642 0.0578 0.933 0.752
#> ATC:mclust 6 0.674 0.568 0.760 0.0396 0.870 0.576
#> SD:kmeans 6 0.501 0.515 0.648 0.0482 0.982 0.918
#> CV:kmeans 6 0.573 0.506 0.626 0.0405 0.909 0.658
#> MAD:kmeans 6 0.459 0.376 0.563 0.0447 0.841 0.414
#> ATC:kmeans 6 0.651 0.470 0.665 0.0410 0.961 0.823
#> SD:pam 6 0.344 0.387 0.618 0.0303 0.977 0.914
#> CV:pam 6 0.253 0.251 0.544 0.0378 0.925 0.769
#> MAD:pam 6 0.249 0.304 0.541 0.0400 0.933 0.790
#> ATC:pam 6 0.754 0.693 0.812 0.0466 0.921 0.663
#> SD:hclust 6 0.351 0.355 0.595 0.0612 0.836 0.557
#> CV:hclust 6 0.232 0.425 0.593 0.0718 0.909 0.733
#> MAD:hclust 6 0.248 0.378 0.533 0.0616 0.892 0.746
#> ATC:hclust 6 0.592 0.466 0.615 0.0397 0.962 0.853
Following heatmap plots the partition for each combination of methods and the lightness correspond to the silhouette scores for samples in each method. On top the consensus subgroup is inferred from all methods by taking the mean silhouette scores as weight.
collect_stats(res_list, k = 2)
collect_stats(res_list, k = 3)
collect_stats(res_list, k = 4)
collect_stats(res_list, k = 5)
collect_stats(res_list, k = 6)
Collect partitions from all methods:
collect_classes(res_list, k = 2)
collect_classes(res_list, k = 3)
collect_classes(res_list, k = 4)
collect_classes(res_list, k = 5)
collect_classes(res_list, k = 6)
Overlap of top rows from different top-row methods:
top_rows_overlap(res_list, top_n = 1000, method = "euler")
top_rows_overlap(res_list, top_n = 2000, method = "euler")
top_rows_overlap(res_list, top_n = 3000, method = "euler")
top_rows_overlap(res_list, top_n = 4000, method = "euler")
top_rows_overlap(res_list, top_n = 5000, method = "euler")
Also visualize the correspondance of rankings between different top-row methods:
top_rows_overlap(res_list, top_n = 1000, method = "correspondance")
top_rows_overlap(res_list, top_n = 2000, method = "correspondance")
top_rows_overlap(res_list, top_n = 3000, method = "correspondance")
top_rows_overlap(res_list, top_n = 4000, method = "correspondance")
top_rows_overlap(res_list, top_n = 5000, method = "correspondance")
Heatmaps of the top rows:
top_rows_heatmap(res_list, top_n = 1000)
top_rows_heatmap(res_list, top_n = 2000)
top_rows_heatmap(res_list, top_n = 3000)
top_rows_heatmap(res_list, top_n = 4000)
top_rows_heatmap(res_list, top_n = 5000)
Test correlation between subgroups and known annotations. If the known annotation is numeric, one-way ANOVA test is applied, and if the known annotation is discrete, chi-squared contingency table test is applied.
test_to_known_factors(res_list, k = 2)
#> n agent(p) disease.state(p) gender(p) individual(p) k
#> SD:NMF 58 0.96332 0.719 2.06e-13 1.60e-03 2
#> CV:NMF 84 1.00000 1.000 3.81e-19 8.65e-05 2
#> MAD:NMF 70 0.99455 0.473 7.48e-01 4.53e-03 2
#> ATC:NMF 81 0.47143 0.408 8.98e-01 1.73e-03 2
#> SD:skmeans 10 NA NA NA NA 2
#> CV:skmeans 84 1.00000 1.000 3.81e-19 8.65e-05 2
#> MAD:skmeans 72 0.97897 0.477 3.02e-01 7.14e-04 2
#> ATC:skmeans 83 0.90491 0.446 2.60e-01 6.52e-03 2
#> SD:mclust 84 1.00000 1.000 3.81e-19 8.65e-05 2
#> CV:mclust 84 1.00000 1.000 3.81e-19 8.65e-05 2
#> MAD:mclust 35 0.49215 0.545 1.00e+00 2.82e-02 2
#> ATC:mclust 82 1.00000 1.000 6.59e-01 3.27e-04 2
#> SD:kmeans 18 0.61708 0.393 2.05e-04 5.50e-02 2
#> CV:kmeans 84 1.00000 1.000 3.81e-19 8.65e-05 2
#> MAD:kmeans 83 0.88978 0.445 5.30e-01 2.02e-04 2
#> ATC:kmeans 81 0.75321 0.434 1.00e+00 1.90e-03 2
#> SD:pam 0 NA NA NA NA 2
#> CV:pam 66 0.69763 0.274 1.85e-11 6.36e-03 2
#> MAD:pam 43 0.93745 0.407 3.49e-03 2.51e-01 2
#> ATC:pam 70 0.00418 0.821 1.57e-01 7.98e-02 2
#> SD:hclust 51 1.00000 0.701 8.57e-01 1.60e-03 2
#> CV:hclust 10 NA NA NA NA 2
#> MAD:hclust 50 1.00000 1.000 1.00e+00 2.13e-03 2
#> ATC:hclust 78 1.00000 1.000 7.69e-01 2.08e-04 2
test_to_known_factors(res_list, k = 3)
#> n agent(p) disease.state(p) gender(p) individual(p) k
#> SD:NMF 48 1.00000 0.743 3.16e-11 2.52e-03 3
#> CV:NMF 72 1.00000 0.813 1.59e-16 3.40e-04 3
#> MAD:NMF 46 0.30535 0.597 1.70e-02 3.44e-03 3
#> ATC:NMF 72 0.00438 0.191 1.18e-01 2.15e-02 3
#> SD:skmeans 47 0.87270 0.563 3.72e-09 8.15e-05 3
#> CV:skmeans 35 NA NA NA NA 3
#> MAD:skmeans 12 NA NA NA NA 3
#> ATC:skmeans 83 0.03422 0.673 2.33e-01 4.28e-04 3
#> SD:mclust 82 1.00000 0.781 2.19e-16 2.04e-07 3
#> CV:mclust 59 0.98231 0.851 1.54e-13 5.55e-06 3
#> MAD:mclust 75 0.79434 0.215 1.78e-01 1.81e-06 3
#> ATC:mclust 77 0.02381 0.719 6.50e-02 7.03e-04 3
#> SD:kmeans 48 1.00000 0.319 1.52e-08 2.21e-05 3
#> CV:kmeans 76 0.96896 0.337 3.14e-17 5.31e-07 3
#> MAD:kmeans 60 0.77033 1.000 2.50e-01 1.35e-03 3
#> ATC:kmeans 74 0.03055 0.751 2.93e-01 4.45e-04 3
#> SD:pam 47 0.98896 0.111 9.72e-05 1.07e-02 3
#> CV:pam 34 NA NA NA NA 3
#> MAD:pam 33 0.92744 0.132 1.55e-01 6.42e-02 3
#> ATC:pam 73 0.09593 0.845 3.66e-01 3.67e-04 3
#> SD:hclust 45 0.98186 0.785 1.51e-02 5.31e-05 3
#> CV:hclust 78 1.00000 1.000 7.87e-18 2.08e-04 3
#> MAD:hclust 54 1.00000 0.282 1.00e+00 1.52e-03 3
#> ATC:hclust 61 0.55035 0.971 2.48e-02 7.88e-05 3
test_to_known_factors(res_list, k = 4)
#> n agent(p) disease.state(p) gender(p) individual(p) k
#> SD:NMF 54 0.93327 0.0877 1.12e-11 5.00e-06 4
#> CV:NMF 55 0.86881 1.0000 9.66e-13 2.48e-03 4
#> MAD:NMF 24 0.87837 0.4979 6.70e-02 1.07e-02 4
#> ATC:NMF 48 0.13220 0.9297 2.18e-02 3.40e-03 4
#> SD:skmeans 43 0.83930 0.4642 1.28e-06 5.75e-06 4
#> CV:skmeans 25 NA NA NA NA 4
#> MAD:skmeans 0 NA NA NA NA 4
#> ATC:skmeans 61 0.01411 0.5032 6.23e-01 7.01e-03 4
#> SD:mclust 74 1.00000 0.2845 3.15e-14 2.08e-09 4
#> CV:mclust 76 0.98549 0.3658 2.21e-16 1.03e-08 4
#> MAD:mclust 58 0.73138 0.3302 1.06e-01 9.75e-06 4
#> ATC:mclust 55 0.03518 0.7659 1.45e-01 7.51e-03 4
#> SD:kmeans 38 1.00000 0.2479 1.47e-05 3.52e-06 4
#> CV:kmeans 68 0.99367 0.1936 1.14e-14 8.59e-09 4
#> MAD:kmeans 39 0.51650 0.3541 1.29e-04 2.76e-04 4
#> ATC:kmeans 76 0.02065 0.8736 1.24e-01 5.55e-04 4
#> SD:pam 33 1.00000 0.0932 8.53e-01 9.43e-02 4
#> CV:pam 18 NA NA NA NA 4
#> MAD:pam 20 NA NA NA NA 4
#> ATC:pam 73 0.00526 0.7843 6.22e-01 1.29e-03 4
#> SD:hclust 46 1.00000 0.9488 9.85e-04 5.23e-07 4
#> CV:hclust 51 0.87364 0.6102 8.42e-12 4.25e-05 4
#> MAD:hclust 34 0.59787 0.1031 5.66e-02 3.75e-03 4
#> ATC:hclust 63 0.85909 0.5966 2.44e-02 9.49e-07 4
test_to_known_factors(res_list, k = 5)
#> n agent(p) disease.state(p) gender(p) individual(p) k
#> SD:NMF 6 NA NA NA NA 5
#> CV:NMF 30 0.9418 0.0684 3.06e-07 1.95e-03 5
#> MAD:NMF 8 1.0000 0.4142 NA 4.60e-02 5
#> ATC:NMF 39 0.1040 0.8302 7.69e-02 6.48e-03 5
#> SD:skmeans 9 NA NA NA NA 5
#> CV:skmeans 0 NA NA NA NA 5
#> MAD:skmeans 0 NA NA NA NA 5
#> ATC:skmeans 65 0.2622 0.2916 3.99e-02 1.09e-04 5
#> SD:mclust 74 0.9780 0.1151 4.09e-13 2.52e-11 5
#> CV:mclust 74 0.9750 0.1526 3.24e-15 9.67e-11 5
#> MAD:mclust 45 0.9483 0.1123 5.76e-02 1.57e-04 5
#> ATC:mclust 39 0.1608 0.7939 7.53e-01 1.13e-02 5
#> SD:kmeans 59 0.9993 0.2861 9.16e-10 3.34e-10 5
#> CV:kmeans 53 0.9927 0.5650 1.83e-11 5.22e-07 5
#> MAD:kmeans 17 1.0000 0.4916 4.33e-01 3.01e-02 5
#> ATC:kmeans 48 0.0991 0.7454 7.89e-03 1.11e-03 5
#> SD:pam 30 0.9931 0.0481 4.15e-03 2.17e-03 5
#> CV:pam 15 NA NA NA NA 5
#> MAD:pam 16 NA NA NA NA 5
#> ATC:pam 70 0.0162 0.7321 2.19e-02 5.23e-05 5
#> SD:hclust 26 1.0000 0.0734 3.70e-03 6.24e-05 5
#> CV:hclust 49 1.0000 0.4796 1.90e-11 2.83e-03 5
#> MAD:hclust 10 1.0000 0.2586 6.28e-01 4.04e-02 5
#> ATC:hclust 37 0.9706 0.3629 1.37e-01 8.59e-04 5
test_to_known_factors(res_list, k = 6)
#> n agent(p) disease.state(p) gender(p) individual(p) k
#> SD:NMF 11 0.97415 0.1997 4.09e-03 3.75e-02 6
#> CV:NMF 1 NA NA NA NA 6
#> MAD:NMF 0 NA NA NA NA 6
#> ATC:NMF 24 0.10551 0.2737 7.64e-01 3.09e-02 6
#> SD:skmeans 0 NA NA NA NA 6
#> CV:skmeans 0 NA NA NA NA 6
#> MAD:skmeans 0 NA NA NA NA 6
#> ATC:skmeans 54 0.46388 0.3878 1.66e-01 2.92e-06 6
#> SD:mclust 69 0.99664 0.0770 1.75e-11 2.05e-13 6
#> CV:mclust 74 0.99370 0.1706 1.50e-14 1.65e-11 6
#> MAD:mclust 44 0.78737 0.0634 8.21e-02 2.09e-05 6
#> ATC:mclust 55 0.01048 0.2504 1.94e-01 1.28e-03 6
#> SD:kmeans 51 0.94841 0.3169 1.58e-07 3.74e-10 6
#> CV:kmeans 56 0.55056 0.4308 2.01e-11 7.21e-09 6
#> MAD:kmeans 28 0.98688 0.0520 2.82e-02 1.31e-05 6
#> ATC:kmeans 44 0.22061 0.6256 1.15e-01 5.73e-04 6
#> SD:pam 28 0.97732 0.0287 2.88e-02 1.95e-03 6
#> CV:pam 16 NA NA NA NA 6
#> MAD:pam 18 1.00000 0.0890 8.41e-01 1.58e-01 6
#> ATC:pam 71 0.00221 0.6512 8.82e-02 2.48e-04 6
#> SD:hclust 30 0.99743 0.0518 1.69e-04 6.23e-07 6
#> CV:hclust 43 0.90356 0.6541 4.08e-10 6.93e-03 6
#> MAD:hclust 20 0.89571 0.1222 4.99e-04 1.76e-04 6
#> ATC:hclust 42 0.98993 0.2157 6.33e-02 1.56e-07 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 51941 rows and 84 columns.
#> Top rows (1000, 2000, 3000, 4000, 5000) are extracted by 'SD' method.
#> Subgroups are detected by 'hclust' method.
#> Performed in total 1250 partitions by row resampling.
#> Best k for subgroups seems to be 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 0.0668 0.490 0.742 0.3485 0.826 0.826
#> 3 3 0.1571 0.441 0.641 0.4984 0.629 0.556
#> 4 4 0.2301 0.479 0.638 0.1866 0.869 0.736
#> 5 5 0.2931 0.352 0.598 0.1153 0.764 0.484
#> 6 6 0.3512 0.355 0.595 0.0612 0.836 0.557
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
#> GSM1269647 1 0.963 0.3921 0.612 0.388
#> GSM1269655 1 0.653 0.6286 0.832 0.168
#> GSM1269663 1 0.949 0.2857 0.632 0.368
#> GSM1269671 1 0.999 0.0961 0.516 0.484
#> GSM1269679 1 0.706 0.6118 0.808 0.192
#> GSM1269693 1 0.921 0.1854 0.664 0.336
#> GSM1269701 1 0.552 0.6373 0.872 0.128
#> GSM1269709 1 0.714 0.6125 0.804 0.196
#> GSM1269715 1 0.821 0.3346 0.744 0.256
#> GSM1269717 1 0.469 0.5587 0.900 0.100
#> GSM1269721 2 0.949 0.4933 0.368 0.632
#> GSM1269723 1 0.563 0.6347 0.868 0.132
#> GSM1269645 1 0.745 0.5108 0.788 0.212
#> GSM1269653 1 0.939 0.4244 0.644 0.356
#> GSM1269661 1 0.689 0.6290 0.816 0.184
#> GSM1269669 1 0.373 0.6341 0.928 0.072
#> GSM1269677 2 0.861 0.8142 0.284 0.716
#> GSM1269685 1 0.402 0.6325 0.920 0.080
#> GSM1269691 1 0.402 0.6354 0.920 0.080
#> GSM1269699 1 0.998 0.1100 0.528 0.472
#> GSM1269707 1 0.998 0.1100 0.528 0.472
#> GSM1269651 2 0.821 0.8353 0.256 0.744
#> GSM1269659 1 1.000 -0.2042 0.508 0.492
#> GSM1269667 1 0.653 0.6336 0.832 0.168
#> GSM1269675 1 0.978 0.3373 0.588 0.412
#> GSM1269683 1 0.443 0.5932 0.908 0.092
#> GSM1269689 1 0.949 0.4227 0.632 0.368
#> GSM1269697 1 0.969 0.3699 0.604 0.396
#> GSM1269705 1 0.961 0.3856 0.616 0.384
#> GSM1269713 1 0.900 0.4746 0.684 0.316
#> GSM1269719 1 0.827 0.5701 0.740 0.260
#> GSM1269725 1 0.943 0.4256 0.640 0.360
#> GSM1269727 1 0.605 0.6329 0.852 0.148
#> GSM1269649 1 0.714 0.6181 0.804 0.196
#> GSM1269657 1 0.997 -0.1228 0.532 0.468
#> GSM1269665 1 0.730 0.5142 0.796 0.204
#> GSM1269673 1 0.373 0.6325 0.928 0.072
#> GSM1269681 2 0.821 0.8353 0.256 0.744
#> GSM1269687 1 0.358 0.6246 0.932 0.068
#> GSM1269695 1 0.494 0.6429 0.892 0.108
#> GSM1269703 1 0.574 0.6215 0.864 0.136
#> GSM1269711 1 0.653 0.6292 0.832 0.168
#> GSM1269646 1 0.963 0.3921 0.612 0.388
#> GSM1269654 1 0.653 0.6286 0.832 0.168
#> GSM1269662 1 0.949 0.2857 0.632 0.368
#> GSM1269670 1 0.999 0.0961 0.516 0.484
#> GSM1269678 1 0.706 0.6118 0.808 0.192
#> GSM1269692 1 0.913 0.1968 0.672 0.328
#> GSM1269700 1 0.552 0.6373 0.872 0.128
#> GSM1269708 1 0.714 0.6125 0.804 0.196
#> GSM1269714 1 0.456 0.5605 0.904 0.096
#> GSM1269716 1 0.469 0.5587 0.900 0.100
#> GSM1269720 2 0.949 0.4933 0.368 0.632
#> GSM1269722 1 0.563 0.6347 0.868 0.132
#> GSM1269644 1 0.706 0.5362 0.808 0.192
#> GSM1269652 1 0.939 0.4244 0.644 0.356
#> GSM1269660 1 0.689 0.6290 0.816 0.184
#> GSM1269668 1 0.373 0.6341 0.928 0.072
#> GSM1269676 2 0.861 0.8142 0.284 0.716
#> GSM1269684 1 0.402 0.6325 0.920 0.080
#> GSM1269690 1 0.402 0.6354 0.920 0.080
#> GSM1269698 1 0.998 0.1100 0.528 0.472
#> GSM1269706 1 0.998 0.1100 0.528 0.472
#> GSM1269650 2 0.821 0.8353 0.256 0.744
#> GSM1269658 1 1.000 -0.2042 0.508 0.492
#> GSM1269666 1 0.653 0.6336 0.832 0.168
#> GSM1269674 1 0.978 0.3373 0.588 0.412
#> GSM1269682 1 0.443 0.5932 0.908 0.092
#> GSM1269688 1 0.949 0.4227 0.632 0.368
#> GSM1269696 1 0.969 0.3699 0.604 0.396
#> GSM1269704 1 0.961 0.3856 0.616 0.384
#> GSM1269712 1 0.900 0.4746 0.684 0.316
#> GSM1269718 1 0.827 0.5701 0.740 0.260
#> GSM1269724 1 0.943 0.4256 0.640 0.360
#> GSM1269726 1 0.605 0.6329 0.852 0.148
#> GSM1269648 1 0.714 0.6181 0.804 0.196
#> GSM1269656 1 0.997 -0.1228 0.532 0.468
#> GSM1269664 1 0.714 0.5277 0.804 0.196
#> GSM1269672 1 0.373 0.6325 0.928 0.072
#> GSM1269680 2 0.821 0.8353 0.256 0.744
#> GSM1269686 1 0.358 0.6246 0.932 0.068
#> GSM1269694 1 0.494 0.6429 0.892 0.108
#> GSM1269702 1 0.494 0.6192 0.892 0.108
#> GSM1269710 1 0.653 0.6292 0.832 0.168
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM1269647 3 0.662 0.7988 0.436 0.008 0.556
#> GSM1269655 1 0.580 0.4698 0.776 0.040 0.184
#> GSM1269663 1 0.913 -0.1216 0.480 0.372 0.148
#> GSM1269671 3 0.733 0.7648 0.312 0.052 0.636
#> GSM1269679 1 0.603 0.1570 0.712 0.016 0.272
#> GSM1269693 1 0.958 -0.1912 0.480 0.248 0.272
#> GSM1269701 1 0.563 0.3485 0.768 0.024 0.208
#> GSM1269709 1 0.619 0.3416 0.724 0.028 0.248
#> GSM1269715 1 0.805 0.0916 0.568 0.076 0.356
#> GSM1269717 1 0.486 0.5275 0.840 0.044 0.116
#> GSM1269721 2 0.875 0.6545 0.144 0.564 0.292
#> GSM1269723 1 0.568 0.3423 0.764 0.024 0.212
#> GSM1269645 1 0.774 0.4378 0.672 0.204 0.124
#> GSM1269653 1 0.851 -0.6283 0.476 0.092 0.432
#> GSM1269661 1 0.670 0.2962 0.712 0.052 0.236
#> GSM1269669 1 0.368 0.5177 0.876 0.008 0.116
#> GSM1269677 2 0.409 0.7483 0.068 0.880 0.052
#> GSM1269685 1 0.371 0.5396 0.892 0.032 0.076
#> GSM1269691 1 0.401 0.5338 0.876 0.028 0.096
#> GSM1269699 3 0.824 0.7449 0.336 0.092 0.572
#> GSM1269707 3 0.824 0.7449 0.336 0.092 0.572
#> GSM1269651 2 0.444 0.7533 0.052 0.864 0.084
#> GSM1269659 2 0.986 0.5109 0.304 0.416 0.280
#> GSM1269667 1 0.598 0.2648 0.728 0.020 0.252
#> GSM1269675 3 0.681 0.8035 0.404 0.016 0.580
#> GSM1269683 1 0.440 0.5432 0.864 0.044 0.092
#> GSM1269689 3 0.681 0.7376 0.464 0.012 0.524
#> GSM1269697 3 0.623 0.8005 0.436 0.000 0.564
#> GSM1269705 3 0.775 0.7227 0.460 0.048 0.492
#> GSM1269713 1 0.650 -0.6035 0.532 0.004 0.464
#> GSM1269719 1 0.806 0.4286 0.652 0.156 0.192
#> GSM1269725 3 0.652 0.7265 0.484 0.004 0.512
#> GSM1269727 1 0.594 0.4432 0.760 0.036 0.204
#> GSM1269649 1 0.671 0.1713 0.672 0.032 0.296
#> GSM1269657 2 0.865 0.3528 0.360 0.528 0.112
#> GSM1269665 1 0.750 0.4461 0.688 0.200 0.112
#> GSM1269673 1 0.377 0.5366 0.888 0.028 0.084
#> GSM1269681 2 0.418 0.7542 0.052 0.876 0.072
#> GSM1269687 1 0.324 0.5529 0.912 0.032 0.056
#> GSM1269695 1 0.484 0.4551 0.816 0.016 0.168
#> GSM1269703 1 0.556 0.5141 0.808 0.064 0.128
#> GSM1269711 1 0.623 0.2273 0.700 0.020 0.280
#> GSM1269646 3 0.662 0.7988 0.436 0.008 0.556
#> GSM1269654 1 0.580 0.4698 0.776 0.040 0.184
#> GSM1269662 1 0.913 -0.1216 0.480 0.372 0.148
#> GSM1269670 3 0.733 0.7648 0.312 0.052 0.636
#> GSM1269678 1 0.603 0.1570 0.712 0.016 0.272
#> GSM1269692 1 0.945 -0.1344 0.500 0.232 0.268
#> GSM1269700 1 0.563 0.3485 0.768 0.024 0.208
#> GSM1269708 1 0.619 0.3416 0.724 0.028 0.248
#> GSM1269714 1 0.493 0.5266 0.836 0.044 0.120
#> GSM1269716 1 0.486 0.5275 0.840 0.044 0.116
#> GSM1269720 2 0.875 0.6545 0.144 0.564 0.292
#> GSM1269722 1 0.568 0.3423 0.764 0.024 0.212
#> GSM1269644 1 0.738 0.4601 0.700 0.184 0.116
#> GSM1269652 1 0.851 -0.6283 0.476 0.092 0.432
#> GSM1269660 1 0.670 0.2962 0.712 0.052 0.236
#> GSM1269668 1 0.368 0.5177 0.876 0.008 0.116
#> GSM1269676 2 0.409 0.7483 0.068 0.880 0.052
#> GSM1269684 1 0.383 0.5408 0.888 0.036 0.076
#> GSM1269690 1 0.401 0.5338 0.876 0.028 0.096
#> GSM1269698 3 0.824 0.7449 0.336 0.092 0.572
#> GSM1269706 3 0.824 0.7449 0.336 0.092 0.572
#> GSM1269650 2 0.444 0.7533 0.052 0.864 0.084
#> GSM1269658 2 0.986 0.5109 0.304 0.416 0.280
#> GSM1269666 1 0.598 0.2648 0.728 0.020 0.252
#> GSM1269674 3 0.681 0.8035 0.404 0.016 0.580
#> GSM1269682 1 0.440 0.5432 0.864 0.044 0.092
#> GSM1269688 3 0.681 0.7376 0.464 0.012 0.524
#> GSM1269696 3 0.623 0.8005 0.436 0.000 0.564
#> GSM1269704 3 0.775 0.7227 0.460 0.048 0.492
#> GSM1269712 1 0.650 -0.6035 0.532 0.004 0.464
#> GSM1269718 1 0.806 0.4286 0.652 0.156 0.192
#> GSM1269724 3 0.652 0.7265 0.484 0.004 0.512
#> GSM1269726 1 0.594 0.4432 0.760 0.036 0.204
#> GSM1269648 1 0.671 0.1713 0.672 0.032 0.296
#> GSM1269656 2 0.865 0.3528 0.360 0.528 0.112
#> GSM1269664 1 0.730 0.4593 0.704 0.188 0.108
#> GSM1269672 1 0.377 0.5366 0.888 0.028 0.084
#> GSM1269680 2 0.418 0.7542 0.052 0.876 0.072
#> GSM1269686 1 0.324 0.5529 0.912 0.032 0.056
#> GSM1269694 1 0.484 0.4551 0.816 0.016 0.168
#> GSM1269702 1 0.442 0.5373 0.864 0.048 0.088
#> GSM1269710 1 0.623 0.2273 0.700 0.020 0.280
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM1269647 3 0.5037 0.688 0.264 0.008 0.712 0.016
#> GSM1269655 1 0.6668 0.460 0.652 0.020 0.228 0.100
#> GSM1269663 1 0.9185 -0.247 0.400 0.268 0.084 0.248
#> GSM1269671 3 0.4033 0.634 0.116 0.004 0.836 0.044
#> GSM1269679 1 0.6357 0.173 0.544 0.000 0.388 0.068
#> GSM1269693 4 0.6902 0.597 0.276 0.120 0.008 0.596
#> GSM1269701 1 0.6254 0.346 0.624 0.004 0.300 0.072
#> GSM1269709 1 0.6770 0.356 0.584 0.004 0.304 0.108
#> GSM1269715 4 0.4980 0.489 0.304 0.000 0.016 0.680
#> GSM1269717 1 0.5309 0.485 0.700 0.000 0.044 0.256
#> GSM1269721 4 0.6538 0.353 0.036 0.460 0.020 0.484
#> GSM1269723 1 0.6374 0.322 0.592 0.000 0.324 0.084
#> GSM1269645 1 0.7030 0.397 0.680 0.120 0.116 0.084
#> GSM1269653 3 0.7931 0.525 0.288 0.052 0.540 0.120
#> GSM1269661 1 0.6610 0.333 0.616 0.032 0.304 0.048
#> GSM1269669 1 0.4559 0.542 0.792 0.004 0.164 0.040
#> GSM1269677 2 0.3173 0.631 0.016 0.888 0.016 0.080
#> GSM1269685 1 0.3383 0.577 0.872 0.012 0.100 0.016
#> GSM1269691 1 0.3594 0.572 0.860 0.008 0.108 0.024
#> GSM1269699 3 0.6447 0.607 0.140 0.040 0.708 0.112
#> GSM1269707 3 0.6447 0.607 0.140 0.040 0.708 0.112
#> GSM1269651 2 0.2021 0.645 0.000 0.936 0.024 0.040
#> GSM1269659 4 0.7078 0.593 0.120 0.312 0.008 0.560
#> GSM1269667 1 0.6435 0.289 0.572 0.004 0.356 0.068
#> GSM1269675 3 0.4919 0.680 0.200 0.000 0.752 0.048
#> GSM1269683 1 0.5591 0.541 0.732 0.004 0.096 0.168
#> GSM1269689 3 0.5311 0.606 0.328 0.000 0.648 0.024
#> GSM1269697 3 0.4855 0.682 0.268 0.000 0.712 0.020
#> GSM1269705 3 0.6699 0.624 0.304 0.044 0.612 0.040
#> GSM1269713 3 0.5778 0.506 0.356 0.000 0.604 0.040
#> GSM1269719 1 0.8203 0.437 0.560 0.084 0.220 0.136
#> GSM1269725 3 0.5300 0.622 0.308 0.000 0.664 0.028
#> GSM1269727 1 0.6323 0.436 0.628 0.000 0.272 0.100
#> GSM1269649 1 0.6220 0.248 0.600 0.020 0.348 0.032
#> GSM1269657 2 0.7911 0.168 0.336 0.508 0.108 0.048
#> GSM1269665 1 0.6928 0.413 0.688 0.116 0.112 0.084
#> GSM1269673 1 0.3134 0.573 0.880 0.008 0.100 0.012
#> GSM1269681 2 0.0657 0.664 0.000 0.984 0.012 0.004
#> GSM1269687 1 0.3238 0.580 0.880 0.008 0.092 0.020
#> GSM1269695 1 0.4756 0.498 0.756 0.008 0.216 0.020
#> GSM1269703 1 0.4881 0.558 0.792 0.036 0.148 0.024
#> GSM1269711 1 0.5695 0.281 0.624 0.008 0.344 0.024
#> GSM1269646 3 0.5037 0.688 0.264 0.008 0.712 0.016
#> GSM1269654 1 0.6668 0.460 0.652 0.020 0.228 0.100
#> GSM1269662 1 0.9185 -0.247 0.400 0.268 0.084 0.248
#> GSM1269670 3 0.4033 0.634 0.116 0.004 0.836 0.044
#> GSM1269678 1 0.6357 0.173 0.544 0.000 0.388 0.068
#> GSM1269692 4 0.6962 0.582 0.292 0.108 0.012 0.588
#> GSM1269700 1 0.6254 0.346 0.624 0.004 0.300 0.072
#> GSM1269708 1 0.6770 0.356 0.584 0.004 0.304 0.108
#> GSM1269714 1 0.6188 0.463 0.636 0.004 0.072 0.288
#> GSM1269716 1 0.5309 0.485 0.700 0.000 0.044 0.256
#> GSM1269720 4 0.6538 0.353 0.036 0.460 0.020 0.484
#> GSM1269722 1 0.6374 0.322 0.592 0.000 0.324 0.084
#> GSM1269644 1 0.6643 0.454 0.708 0.112 0.104 0.076
#> GSM1269652 3 0.7931 0.525 0.288 0.052 0.540 0.120
#> GSM1269660 1 0.6610 0.333 0.616 0.032 0.304 0.048
#> GSM1269668 1 0.4559 0.542 0.792 0.004 0.164 0.040
#> GSM1269676 2 0.3173 0.631 0.016 0.888 0.016 0.080
#> GSM1269684 1 0.3502 0.577 0.868 0.016 0.100 0.016
#> GSM1269690 1 0.3594 0.572 0.860 0.008 0.108 0.024
#> GSM1269698 3 0.6447 0.607 0.140 0.040 0.708 0.112
#> GSM1269706 3 0.6447 0.607 0.140 0.040 0.708 0.112
#> GSM1269650 2 0.2021 0.645 0.000 0.936 0.024 0.040
#> GSM1269658 4 0.7078 0.593 0.120 0.312 0.008 0.560
#> GSM1269666 1 0.6435 0.289 0.572 0.004 0.356 0.068
#> GSM1269674 3 0.4919 0.680 0.200 0.000 0.752 0.048
#> GSM1269682 1 0.5591 0.541 0.732 0.004 0.096 0.168
#> GSM1269688 3 0.5311 0.606 0.328 0.000 0.648 0.024
#> GSM1269696 3 0.4855 0.682 0.268 0.000 0.712 0.020
#> GSM1269704 3 0.6699 0.624 0.304 0.044 0.612 0.040
#> GSM1269712 3 0.5778 0.506 0.356 0.000 0.604 0.040
#> GSM1269718 1 0.8203 0.437 0.560 0.084 0.220 0.136
#> GSM1269724 3 0.5300 0.622 0.308 0.000 0.664 0.028
#> GSM1269726 1 0.6323 0.436 0.628 0.000 0.272 0.100
#> GSM1269648 1 0.6220 0.248 0.600 0.020 0.348 0.032
#> GSM1269656 2 0.7911 0.168 0.336 0.508 0.108 0.048
#> GSM1269664 1 0.6713 0.445 0.704 0.112 0.100 0.084
#> GSM1269672 1 0.3134 0.573 0.880 0.008 0.100 0.012
#> GSM1269680 2 0.0657 0.664 0.000 0.984 0.012 0.004
#> GSM1269686 1 0.3238 0.580 0.880 0.008 0.092 0.020
#> GSM1269694 1 0.4756 0.498 0.756 0.008 0.216 0.020
#> GSM1269702 1 0.4112 0.576 0.840 0.028 0.112 0.020
#> GSM1269710 1 0.5695 0.281 0.624 0.008 0.344 0.024
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM1269647 3 0.434 0.240 0.048 0.008 0.764 0.000 0.180
#> GSM1269655 3 0.722 0.112 0.372 0.008 0.452 0.124 0.044
#> GSM1269663 1 0.864 -0.223 0.408 0.160 0.028 0.256 0.148
#> GSM1269671 5 0.529 0.555 0.048 0.000 0.464 0.000 0.488
#> GSM1269679 3 0.543 0.445 0.216 0.000 0.676 0.096 0.012
#> GSM1269693 4 0.374 0.571 0.112 0.040 0.012 0.832 0.004
#> GSM1269701 3 0.602 0.313 0.320 0.000 0.576 0.084 0.020
#> GSM1269709 3 0.691 0.265 0.304 0.000 0.516 0.136 0.044
#> GSM1269715 4 0.533 0.467 0.136 0.000 0.028 0.720 0.116
#> GSM1269717 1 0.713 0.176 0.400 0.000 0.224 0.356 0.020
#> GSM1269721 4 0.590 0.307 0.020 0.372 0.020 0.560 0.028
#> GSM1269723 3 0.606 0.372 0.276 0.000 0.596 0.112 0.016
#> GSM1269645 1 0.465 0.440 0.792 0.024 0.028 0.036 0.120
#> GSM1269653 3 0.719 -0.249 0.152 0.016 0.488 0.024 0.320
#> GSM1269661 1 0.704 0.187 0.456 0.008 0.384 0.036 0.116
#> GSM1269669 1 0.528 0.431 0.624 0.000 0.320 0.012 0.044
#> GSM1269677 2 0.464 0.626 0.020 0.772 0.000 0.116 0.092
#> GSM1269685 1 0.439 0.531 0.744 0.000 0.216 0.024 0.016
#> GSM1269691 1 0.498 0.522 0.708 0.000 0.228 0.028 0.036
#> GSM1269699 5 0.588 0.785 0.084 0.008 0.364 0.000 0.544
#> GSM1269707 5 0.589 0.785 0.084 0.008 0.368 0.000 0.540
#> GSM1269651 2 0.165 0.658 0.004 0.944 0.000 0.020 0.032
#> GSM1269659 4 0.495 0.516 0.036 0.224 0.000 0.712 0.028
#> GSM1269667 3 0.606 0.369 0.272 0.004 0.608 0.100 0.016
#> GSM1269675 3 0.516 -0.260 0.048 0.000 0.620 0.004 0.328
#> GSM1269683 1 0.714 0.127 0.448 0.000 0.296 0.232 0.024
#> GSM1269689 3 0.520 0.256 0.152 0.000 0.708 0.008 0.132
#> GSM1269697 3 0.361 0.294 0.040 0.000 0.812 0.000 0.148
#> GSM1269705 3 0.586 0.294 0.084 0.028 0.700 0.024 0.164
#> GSM1269713 3 0.295 0.481 0.100 0.000 0.868 0.004 0.028
#> GSM1269719 1 0.780 0.123 0.456 0.024 0.332 0.108 0.080
#> GSM1269725 3 0.280 0.426 0.060 0.000 0.888 0.008 0.044
#> GSM1269727 3 0.747 0.179 0.356 0.000 0.428 0.140 0.076
#> GSM1269649 1 0.625 0.290 0.504 0.004 0.356 0.000 0.136
#> GSM1269657 2 0.825 0.281 0.296 0.456 0.108 0.056 0.084
#> GSM1269665 1 0.445 0.451 0.808 0.028 0.024 0.036 0.104
#> GSM1269673 1 0.452 0.526 0.732 0.000 0.224 0.012 0.032
#> GSM1269681 2 0.124 0.682 0.008 0.960 0.000 0.004 0.028
#> GSM1269687 1 0.438 0.523 0.728 0.000 0.240 0.012 0.020
#> GSM1269695 1 0.545 0.452 0.632 0.000 0.280 0.004 0.084
#> GSM1269703 1 0.494 0.523 0.716 0.012 0.224 0.008 0.040
#> GSM1269711 1 0.622 0.279 0.496 0.000 0.352 0.000 0.152
#> GSM1269646 3 0.434 0.240 0.048 0.008 0.764 0.000 0.180
#> GSM1269654 3 0.722 0.112 0.372 0.008 0.452 0.124 0.044
#> GSM1269662 1 0.864 -0.223 0.408 0.160 0.028 0.256 0.148
#> GSM1269670 5 0.529 0.555 0.048 0.000 0.464 0.000 0.488
#> GSM1269678 3 0.543 0.445 0.216 0.000 0.676 0.096 0.012
#> GSM1269692 4 0.412 0.565 0.120 0.040 0.024 0.812 0.004
#> GSM1269700 3 0.602 0.313 0.320 0.000 0.576 0.084 0.020
#> GSM1269708 3 0.691 0.265 0.304 0.000 0.516 0.136 0.044
#> GSM1269714 4 0.735 -0.264 0.340 0.000 0.232 0.396 0.032
#> GSM1269716 1 0.713 0.176 0.400 0.000 0.224 0.356 0.020
#> GSM1269720 4 0.590 0.307 0.020 0.372 0.020 0.560 0.028
#> GSM1269722 3 0.606 0.372 0.276 0.000 0.596 0.112 0.016
#> GSM1269644 1 0.476 0.463 0.792 0.024 0.044 0.036 0.104
#> GSM1269652 3 0.719 -0.249 0.152 0.016 0.488 0.024 0.320
#> GSM1269660 1 0.704 0.187 0.456 0.008 0.384 0.036 0.116
#> GSM1269668 1 0.528 0.431 0.624 0.000 0.320 0.012 0.044
#> GSM1269676 2 0.464 0.626 0.020 0.772 0.000 0.116 0.092
#> GSM1269684 1 0.436 0.533 0.748 0.000 0.212 0.024 0.016
#> GSM1269690 1 0.498 0.522 0.708 0.000 0.228 0.028 0.036
#> GSM1269698 5 0.588 0.785 0.084 0.008 0.364 0.000 0.544
#> GSM1269706 5 0.589 0.785 0.084 0.008 0.368 0.000 0.540
#> GSM1269650 2 0.165 0.658 0.004 0.944 0.000 0.020 0.032
#> GSM1269658 4 0.495 0.516 0.036 0.224 0.000 0.712 0.028
#> GSM1269666 3 0.606 0.369 0.272 0.004 0.608 0.100 0.016
#> GSM1269674 3 0.516 -0.260 0.048 0.000 0.620 0.004 0.328
#> GSM1269682 1 0.714 0.127 0.448 0.000 0.296 0.232 0.024
#> GSM1269688 3 0.520 0.256 0.152 0.000 0.708 0.008 0.132
#> GSM1269696 3 0.361 0.294 0.040 0.000 0.812 0.000 0.148
#> GSM1269704 3 0.586 0.294 0.084 0.028 0.700 0.024 0.164
#> GSM1269712 3 0.295 0.481 0.100 0.000 0.868 0.004 0.028
#> GSM1269718 1 0.780 0.123 0.456 0.024 0.332 0.108 0.080
#> GSM1269724 3 0.280 0.426 0.060 0.000 0.888 0.008 0.044
#> GSM1269726 3 0.747 0.179 0.356 0.000 0.428 0.140 0.076
#> GSM1269648 1 0.625 0.290 0.504 0.004 0.356 0.000 0.136
#> GSM1269656 2 0.825 0.281 0.296 0.456 0.108 0.056 0.084
#> GSM1269664 1 0.465 0.460 0.800 0.028 0.036 0.036 0.100
#> GSM1269672 1 0.452 0.526 0.732 0.000 0.224 0.012 0.032
#> GSM1269680 2 0.124 0.682 0.008 0.960 0.000 0.004 0.028
#> GSM1269686 1 0.438 0.523 0.728 0.000 0.240 0.012 0.020
#> GSM1269694 1 0.545 0.452 0.632 0.000 0.280 0.004 0.084
#> GSM1269702 1 0.490 0.532 0.720 0.012 0.224 0.012 0.032
#> GSM1269710 1 0.622 0.279 0.496 0.000 0.352 0.000 0.152
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM1269647 3 0.548 0.4101 0.128 0.076 0.684 0.004 0.108 0.000
#> GSM1269655 1 0.677 0.0996 0.492 0.052 0.312 0.124 0.016 0.004
#> GSM1269663 2 0.661 1.0000 0.200 0.532 0.008 0.212 0.004 0.044
#> GSM1269671 5 0.548 0.3378 0.012 0.092 0.380 0.000 0.516 0.000
#> GSM1269679 3 0.604 0.2957 0.364 0.020 0.504 0.096 0.016 0.000
#> GSM1269693 4 0.315 0.5176 0.080 0.036 0.012 0.860 0.004 0.008
#> GSM1269701 1 0.641 -0.1327 0.448 0.044 0.408 0.076 0.024 0.000
#> GSM1269709 1 0.669 -0.0623 0.440 0.032 0.376 0.124 0.028 0.000
#> GSM1269715 4 0.657 0.2516 0.076 0.172 0.056 0.608 0.088 0.000
#> GSM1269717 1 0.677 0.1957 0.452 0.080 0.112 0.348 0.008 0.000
#> GSM1269721 4 0.572 0.5090 0.016 0.100 0.012 0.624 0.008 0.240
#> GSM1269723 3 0.649 0.1694 0.400 0.032 0.436 0.112 0.020 0.000
#> GSM1269645 1 0.536 -0.0801 0.584 0.340 0.004 0.016 0.044 0.012
#> GSM1269653 5 0.796 0.2920 0.236 0.088 0.304 0.012 0.336 0.024
#> GSM1269661 1 0.700 0.3007 0.552 0.080 0.200 0.020 0.136 0.012
#> GSM1269669 1 0.445 0.4631 0.752 0.056 0.156 0.004 0.032 0.000
#> GSM1269677 6 0.482 0.5198 0.004 0.048 0.000 0.176 0.052 0.720
#> GSM1269685 1 0.201 0.5287 0.920 0.000 0.036 0.012 0.032 0.000
#> GSM1269691 1 0.299 0.5243 0.876 0.016 0.044 0.020 0.044 0.000
#> GSM1269699 5 0.425 0.6650 0.092 0.000 0.164 0.000 0.740 0.004
#> GSM1269707 5 0.428 0.6646 0.092 0.000 0.168 0.000 0.736 0.004
#> GSM1269651 6 0.350 0.5573 0.000 0.184 0.012 0.004 0.012 0.788
#> GSM1269659 4 0.454 0.6158 0.024 0.120 0.000 0.756 0.008 0.092
#> GSM1269667 3 0.661 0.1609 0.404 0.032 0.432 0.096 0.036 0.000
#> GSM1269675 3 0.648 -0.0598 0.044 0.136 0.520 0.012 0.288 0.000
#> GSM1269683 1 0.668 0.2819 0.540 0.068 0.168 0.212 0.012 0.000
#> GSM1269689 3 0.657 0.2921 0.228 0.076 0.548 0.008 0.140 0.000
#> GSM1269697 3 0.458 0.4522 0.132 0.036 0.744 0.000 0.088 0.000
#> GSM1269705 3 0.673 0.4161 0.188 0.112 0.580 0.016 0.096 0.008
#> GSM1269713 3 0.462 0.5087 0.232 0.032 0.704 0.012 0.020 0.000
#> GSM1269719 1 0.741 0.1043 0.404 0.240 0.268 0.072 0.008 0.008
#> GSM1269725 3 0.433 0.5201 0.188 0.016 0.744 0.008 0.044 0.000
#> GSM1269727 1 0.768 0.0737 0.424 0.064 0.296 0.132 0.084 0.000
#> GSM1269649 1 0.617 0.3489 0.580 0.064 0.188 0.000 0.168 0.000
#> GSM1269657 6 0.797 0.2255 0.308 0.056 0.044 0.080 0.080 0.432
#> GSM1269665 1 0.513 -0.0403 0.604 0.332 0.004 0.012 0.036 0.012
#> GSM1269673 1 0.246 0.5253 0.900 0.016 0.040 0.004 0.040 0.000
#> GSM1269681 6 0.026 0.6171 0.000 0.008 0.000 0.000 0.000 0.992
#> GSM1269687 1 0.265 0.5257 0.888 0.020 0.060 0.004 0.028 0.000
#> GSM1269695 1 0.465 0.4623 0.744 0.044 0.116 0.000 0.096 0.000
#> GSM1269703 1 0.429 0.4945 0.784 0.084 0.080 0.000 0.048 0.004
#> GSM1269711 1 0.604 0.3423 0.592 0.056 0.168 0.000 0.184 0.000
#> GSM1269646 3 0.548 0.4101 0.128 0.076 0.684 0.004 0.108 0.000
#> GSM1269654 1 0.677 0.0996 0.492 0.052 0.312 0.124 0.016 0.004
#> GSM1269662 2 0.661 1.0000 0.200 0.532 0.008 0.212 0.004 0.044
#> GSM1269670 5 0.548 0.3378 0.012 0.092 0.380 0.000 0.516 0.000
#> GSM1269678 3 0.612 0.2946 0.364 0.020 0.500 0.096 0.020 0.000
#> GSM1269692 4 0.345 0.4868 0.096 0.036 0.016 0.840 0.004 0.008
#> GSM1269700 1 0.641 -0.1327 0.448 0.044 0.408 0.076 0.024 0.000
#> GSM1269708 1 0.669 -0.0623 0.440 0.032 0.376 0.124 0.028 0.000
#> GSM1269714 1 0.682 0.1569 0.416 0.068 0.108 0.392 0.016 0.000
#> GSM1269716 1 0.677 0.1957 0.452 0.080 0.112 0.348 0.008 0.000
#> GSM1269720 4 0.572 0.5090 0.016 0.100 0.012 0.624 0.008 0.240
#> GSM1269722 3 0.649 0.1694 0.400 0.032 0.436 0.112 0.020 0.000
#> GSM1269644 1 0.525 0.0276 0.616 0.308 0.004 0.016 0.044 0.012
#> GSM1269652 5 0.796 0.2920 0.236 0.088 0.304 0.012 0.336 0.024
#> GSM1269660 1 0.700 0.3007 0.552 0.080 0.200 0.020 0.136 0.012
#> GSM1269668 1 0.445 0.4631 0.752 0.056 0.156 0.004 0.032 0.000
#> GSM1269676 6 0.482 0.5198 0.004 0.048 0.000 0.176 0.052 0.720
#> GSM1269684 1 0.236 0.5292 0.904 0.004 0.048 0.012 0.032 0.000
#> GSM1269690 1 0.299 0.5243 0.876 0.016 0.044 0.020 0.044 0.000
#> GSM1269698 5 0.425 0.6650 0.092 0.000 0.164 0.000 0.740 0.004
#> GSM1269706 5 0.428 0.6646 0.092 0.000 0.168 0.000 0.736 0.004
#> GSM1269650 6 0.350 0.5573 0.000 0.184 0.012 0.004 0.012 0.788
#> GSM1269658 4 0.454 0.6158 0.024 0.120 0.000 0.756 0.008 0.092
#> GSM1269666 3 0.661 0.1609 0.404 0.032 0.432 0.096 0.036 0.000
#> GSM1269674 3 0.648 -0.0598 0.044 0.136 0.520 0.012 0.288 0.000
#> GSM1269682 1 0.668 0.2819 0.540 0.068 0.168 0.212 0.012 0.000
#> GSM1269688 3 0.657 0.2921 0.228 0.076 0.548 0.008 0.140 0.000
#> GSM1269696 3 0.458 0.4522 0.132 0.036 0.744 0.000 0.088 0.000
#> GSM1269704 3 0.673 0.4161 0.188 0.112 0.580 0.016 0.096 0.008
#> GSM1269712 3 0.462 0.5087 0.232 0.032 0.704 0.012 0.020 0.000
#> GSM1269718 1 0.741 0.1043 0.404 0.240 0.268 0.072 0.008 0.008
#> GSM1269724 3 0.433 0.5201 0.188 0.016 0.744 0.008 0.044 0.000
#> GSM1269726 1 0.768 0.0737 0.424 0.064 0.296 0.132 0.084 0.000
#> GSM1269648 1 0.617 0.3489 0.580 0.064 0.188 0.000 0.168 0.000
#> GSM1269656 6 0.797 0.2255 0.308 0.056 0.044 0.080 0.080 0.432
#> GSM1269664 1 0.516 0.0108 0.620 0.312 0.008 0.012 0.036 0.012
#> GSM1269672 1 0.246 0.5253 0.900 0.016 0.040 0.004 0.040 0.000
#> GSM1269680 6 0.026 0.6171 0.000 0.008 0.000 0.000 0.000 0.992
#> GSM1269686 1 0.265 0.5257 0.888 0.020 0.060 0.004 0.028 0.000
#> GSM1269694 1 0.465 0.4623 0.744 0.044 0.116 0.000 0.096 0.000
#> GSM1269702 1 0.353 0.5173 0.844 0.052 0.052 0.004 0.044 0.004
#> GSM1269710 1 0.604 0.3423 0.592 0.056 0.168 0.000 0.184 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 agent(p) disease.state(p) gender(p) individual(p) k
#> SD:hclust 51 1.000 0.7013 0.857290 1.60e-03 2
#> SD:hclust 45 0.982 0.7854 0.015102 5.31e-05 3
#> SD:hclust 46 1.000 0.9488 0.000985 5.23e-07 4
#> SD:hclust 26 1.000 0.0734 0.003703 6.24e-05 5
#> SD:hclust 30 0.997 0.0518 0.000169 6.23e-07 6
If matrix rows can be associated to genes, consider to use functional_enrichment(res,
...) to perform function enrichment for the signature genes. See this vignette for more detailed explanations.
The object with results only for a single top-value method and a single partition method can be extracted as:
res = res_list["SD", "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 51941 rows and 84 columns.
#> Top rows (1000, 2000, 3000, 4000, 5000) are extracted by 'SD' method.
#> Subgroups are detected by 'kmeans' method.
#> Performed in total 1250 partitions by row resampling.
#> Best k for subgroups seems to be 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 0.122 0.325 0.632 0.4486 0.499 0.499
#> 3 3 0.242 0.495 0.695 0.3852 0.807 0.637
#> 4 4 0.359 0.416 0.625 0.1347 0.748 0.452
#> 5 5 0.459 0.561 0.651 0.0697 0.871 0.604
#> 6 6 0.501 0.515 0.648 0.0482 0.982 0.918
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
#> GSM1269647 1 0.327 0.49912 0.940 0.060
#> GSM1269655 2 0.921 0.41361 0.336 0.664
#> GSM1269663 2 0.788 0.35696 0.236 0.764
#> GSM1269671 1 0.358 0.49027 0.932 0.068
#> GSM1269679 1 0.980 0.27062 0.584 0.416
#> GSM1269693 2 0.680 0.43575 0.180 0.820
#> GSM1269701 1 0.985 0.25812 0.572 0.428
#> GSM1269709 1 0.971 0.28960 0.600 0.400
#> GSM1269715 2 0.876 0.44572 0.296 0.704
#> GSM1269717 2 0.850 0.45557 0.276 0.724
#> GSM1269721 1 1.000 -0.06715 0.504 0.496
#> GSM1269723 1 0.980 0.27062 0.584 0.416
#> GSM1269645 2 0.634 0.51956 0.160 0.840
#> GSM1269653 1 0.891 0.34066 0.692 0.308
#> GSM1269661 2 0.958 0.16309 0.380 0.620
#> GSM1269669 2 0.985 0.04563 0.428 0.572
#> GSM1269677 2 0.975 0.10879 0.408 0.592
#> GSM1269685 2 0.653 0.51954 0.168 0.832
#> GSM1269691 2 0.615 0.52200 0.152 0.848
#> GSM1269699 1 0.722 0.34761 0.800 0.200
#> GSM1269707 1 0.767 0.32721 0.776 0.224
#> GSM1269651 1 0.997 -0.04301 0.532 0.468
#> GSM1269659 2 0.980 0.13225 0.416 0.584
#> GSM1269667 1 0.975 0.28192 0.592 0.408
#> GSM1269675 1 0.358 0.49051 0.932 0.068
#> GSM1269683 2 0.891 0.43237 0.308 0.692
#> GSM1269689 1 0.625 0.50610 0.844 0.156
#> GSM1269697 1 0.662 0.49987 0.828 0.172
#> GSM1269705 1 0.402 0.50392 0.920 0.080
#> GSM1269713 1 0.781 0.46327 0.768 0.232
#> GSM1269719 2 0.871 0.44328 0.292 0.708
#> GSM1269725 1 0.808 0.45213 0.752 0.248
#> GSM1269727 1 0.987 0.23760 0.568 0.432
#> GSM1269649 2 0.999 -0.03363 0.480 0.520
#> GSM1269657 2 0.971 0.11518 0.400 0.600
#> GSM1269665 2 0.722 0.50253 0.200 0.800
#> GSM1269673 2 0.706 0.50922 0.192 0.808
#> GSM1269681 2 0.980 0.10087 0.416 0.584
#> GSM1269687 2 0.730 0.50370 0.204 0.796
#> GSM1269695 2 0.992 0.04664 0.448 0.552
#> GSM1269703 2 0.706 0.50719 0.192 0.808
#> GSM1269711 2 0.992 0.03704 0.448 0.552
#> GSM1269646 1 0.358 0.50272 0.932 0.068
#> GSM1269654 2 0.921 0.41361 0.336 0.664
#> GSM1269662 2 0.795 0.35601 0.240 0.760
#> GSM1269670 1 0.358 0.49027 0.932 0.068
#> GSM1269678 1 0.980 0.27062 0.584 0.416
#> GSM1269692 2 0.680 0.43575 0.180 0.820
#> GSM1269700 1 0.985 0.25812 0.572 0.428
#> GSM1269708 1 0.973 0.28412 0.596 0.404
#> GSM1269714 2 0.886 0.43687 0.304 0.696
#> GSM1269716 2 0.850 0.45557 0.276 0.724
#> GSM1269720 1 1.000 -0.06715 0.504 0.496
#> GSM1269722 1 0.980 0.27062 0.584 0.416
#> GSM1269644 2 0.574 0.51607 0.136 0.864
#> GSM1269652 1 0.895 0.33750 0.688 0.312
#> GSM1269660 2 0.952 0.18324 0.372 0.628
#> GSM1269668 2 0.985 0.04563 0.428 0.572
#> GSM1269676 2 0.975 0.10879 0.408 0.592
#> GSM1269684 2 0.625 0.52184 0.156 0.844
#> GSM1269690 2 0.615 0.52200 0.152 0.848
#> GSM1269698 1 0.722 0.34761 0.800 0.200
#> GSM1269706 1 0.767 0.32721 0.776 0.224
#> GSM1269650 1 0.997 -0.04301 0.532 0.468
#> GSM1269658 2 0.980 0.13225 0.416 0.584
#> GSM1269666 1 0.975 0.28192 0.592 0.408
#> GSM1269674 1 0.358 0.49051 0.932 0.068
#> GSM1269682 2 0.891 0.43237 0.308 0.692
#> GSM1269688 1 0.625 0.50610 0.844 0.156
#> GSM1269696 1 0.634 0.50380 0.840 0.160
#> GSM1269704 1 0.402 0.50392 0.920 0.080
#> GSM1269712 1 0.969 0.29612 0.604 0.396
#> GSM1269718 2 0.855 0.45045 0.280 0.720
#> GSM1269724 1 0.808 0.45213 0.752 0.248
#> GSM1269726 1 0.990 0.22620 0.560 0.440
#> GSM1269648 2 0.997 -0.00167 0.468 0.532
#> GSM1269656 2 0.978 0.11507 0.412 0.588
#> GSM1269664 2 0.730 0.49848 0.204 0.796
#> GSM1269672 2 0.706 0.50624 0.192 0.808
#> GSM1269680 2 0.980 0.10087 0.416 0.584
#> GSM1269686 2 0.745 0.49578 0.212 0.788
#> GSM1269694 2 0.992 0.04664 0.448 0.552
#> GSM1269702 2 0.680 0.51697 0.180 0.820
#> GSM1269710 2 0.992 0.03704 0.448 0.552
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM1269647 3 0.580 0.600 0.044 0.176 0.780
#> GSM1269655 1 0.947 0.274 0.456 0.188 0.356
#> GSM1269663 2 0.981 0.120 0.320 0.424 0.256
#> GSM1269671 3 0.730 0.510 0.072 0.252 0.676
#> GSM1269679 3 0.633 0.313 0.332 0.012 0.656
#> GSM1269693 1 0.933 0.311 0.516 0.268 0.216
#> GSM1269701 3 0.697 0.240 0.356 0.028 0.616
#> GSM1269709 3 0.512 0.541 0.188 0.016 0.796
#> GSM1269715 1 0.803 0.456 0.616 0.096 0.288
#> GSM1269717 1 0.823 0.460 0.608 0.112 0.280
#> GSM1269721 2 0.479 0.782 0.112 0.844 0.044
#> GSM1269723 3 0.739 0.212 0.356 0.044 0.600
#> GSM1269645 1 0.264 0.635 0.932 0.048 0.020
#> GSM1269653 3 0.897 0.239 0.412 0.128 0.460
#> GSM1269661 1 0.532 0.520 0.780 0.016 0.204
#> GSM1269669 1 0.558 0.539 0.772 0.024 0.204
#> GSM1269677 2 0.501 0.772 0.180 0.804 0.016
#> GSM1269685 1 0.177 0.643 0.960 0.024 0.016
#> GSM1269691 1 0.175 0.645 0.960 0.028 0.012
#> GSM1269699 3 0.985 0.278 0.288 0.292 0.420
#> GSM1269707 3 0.982 0.246 0.356 0.244 0.400
#> GSM1269651 2 0.510 0.764 0.084 0.836 0.080
#> GSM1269659 2 0.594 0.763 0.140 0.788 0.072
#> GSM1269667 3 0.679 0.302 0.324 0.028 0.648
#> GSM1269675 3 0.634 0.555 0.044 0.220 0.736
#> GSM1269683 1 0.817 0.443 0.600 0.100 0.300
#> GSM1269689 3 0.572 0.614 0.068 0.132 0.800
#> GSM1269697 3 0.473 0.623 0.060 0.088 0.852
#> GSM1269705 3 0.614 0.600 0.060 0.172 0.768
#> GSM1269713 3 0.391 0.589 0.104 0.020 0.876
#> GSM1269719 1 0.966 0.285 0.444 0.224 0.332
#> GSM1269725 3 0.421 0.586 0.120 0.020 0.860
#> GSM1269727 1 0.845 0.251 0.488 0.088 0.424
#> GSM1269649 1 0.740 0.321 0.644 0.060 0.296
#> GSM1269657 2 0.560 0.739 0.228 0.756 0.016
#> GSM1269665 1 0.255 0.643 0.936 0.040 0.024
#> GSM1269673 1 0.134 0.647 0.972 0.016 0.012
#> GSM1269681 2 0.462 0.769 0.136 0.840 0.024
#> GSM1269687 1 0.255 0.638 0.932 0.012 0.056
#> GSM1269695 1 0.659 0.436 0.732 0.060 0.208
#> GSM1269703 1 0.118 0.645 0.976 0.012 0.012
#> GSM1269711 1 0.706 0.313 0.656 0.044 0.300
#> GSM1269646 3 0.569 0.605 0.044 0.168 0.788
#> GSM1269654 1 0.947 0.274 0.456 0.188 0.356
#> GSM1269662 2 0.980 0.116 0.324 0.424 0.252
#> GSM1269670 3 0.730 0.510 0.072 0.252 0.676
#> GSM1269678 3 0.636 0.307 0.336 0.012 0.652
#> GSM1269692 1 0.933 0.311 0.516 0.268 0.216
#> GSM1269700 3 0.697 0.240 0.356 0.028 0.616
#> GSM1269708 3 0.527 0.528 0.200 0.016 0.784
#> GSM1269714 1 0.799 0.453 0.616 0.092 0.292
#> GSM1269716 1 0.823 0.460 0.608 0.112 0.280
#> GSM1269720 2 0.479 0.782 0.112 0.844 0.044
#> GSM1269722 3 0.733 0.242 0.344 0.044 0.612
#> GSM1269644 1 0.223 0.631 0.944 0.044 0.012
#> GSM1269652 3 0.902 0.228 0.416 0.132 0.452
#> GSM1269660 1 0.537 0.519 0.776 0.016 0.208
#> GSM1269668 1 0.558 0.539 0.772 0.024 0.204
#> GSM1269676 2 0.501 0.772 0.180 0.804 0.016
#> GSM1269684 1 0.145 0.646 0.968 0.024 0.008
#> GSM1269690 1 0.175 0.645 0.960 0.028 0.012
#> GSM1269698 3 0.985 0.278 0.288 0.292 0.420
#> GSM1269706 3 0.982 0.246 0.356 0.244 0.400
#> GSM1269650 2 0.510 0.764 0.084 0.836 0.080
#> GSM1269658 2 0.594 0.763 0.140 0.788 0.072
#> GSM1269666 3 0.655 0.301 0.324 0.020 0.656
#> GSM1269674 3 0.634 0.556 0.044 0.220 0.736
#> GSM1269682 1 0.817 0.439 0.600 0.100 0.300
#> GSM1269688 3 0.581 0.615 0.072 0.132 0.796
#> GSM1269696 3 0.473 0.623 0.060 0.088 0.852
#> GSM1269704 3 0.614 0.600 0.060 0.172 0.768
#> GSM1269712 3 0.465 0.530 0.176 0.008 0.816
#> GSM1269718 1 0.958 0.286 0.452 0.208 0.340
#> GSM1269724 3 0.435 0.585 0.128 0.020 0.852
#> GSM1269726 1 0.844 0.260 0.492 0.088 0.420
#> GSM1269648 1 0.731 0.331 0.656 0.060 0.284
#> GSM1269656 2 0.657 0.600 0.348 0.636 0.016
#> GSM1269664 1 0.232 0.644 0.944 0.028 0.028
#> GSM1269672 1 0.101 0.646 0.980 0.012 0.008
#> GSM1269680 2 0.481 0.771 0.148 0.828 0.024
#> GSM1269686 1 0.249 0.637 0.932 0.008 0.060
#> GSM1269694 1 0.659 0.436 0.732 0.060 0.208
#> GSM1269702 1 0.178 0.644 0.960 0.020 0.020
#> GSM1269710 1 0.703 0.321 0.660 0.044 0.296
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM1269647 3 0.6506 -0.26871 0.016 0.444 0.500 0.040
#> GSM1269655 3 0.7960 0.40881 0.292 0.064 0.540 0.104
#> GSM1269663 3 0.9135 -0.00855 0.192 0.088 0.360 0.360
#> GSM1269671 2 0.5596 0.56490 0.016 0.712 0.232 0.040
#> GSM1269679 3 0.3760 0.50259 0.136 0.028 0.836 0.000
#> GSM1269693 1 0.8734 -0.19070 0.392 0.068 0.380 0.160
#> GSM1269701 3 0.4466 0.49600 0.156 0.040 0.800 0.004
#> GSM1269709 3 0.5595 0.37122 0.112 0.148 0.736 0.004
#> GSM1269715 3 0.7967 0.19407 0.412 0.068 0.444 0.076
#> GSM1269717 1 0.8020 -0.21927 0.428 0.068 0.424 0.080
#> GSM1269721 4 0.5738 0.77589 0.072 0.072 0.088 0.768
#> GSM1269723 3 0.3972 0.51985 0.164 0.004 0.816 0.016
#> GSM1269645 1 0.3662 0.68249 0.876 0.048 0.048 0.028
#> GSM1269653 1 0.9272 -0.21019 0.396 0.280 0.228 0.096
#> GSM1269661 1 0.4333 0.60523 0.776 0.008 0.208 0.008
#> GSM1269669 1 0.4682 0.60511 0.760 0.024 0.212 0.004
#> GSM1269677 4 0.4224 0.78694 0.100 0.076 0.000 0.824
#> GSM1269685 1 0.0712 0.70511 0.984 0.004 0.004 0.008
#> GSM1269691 1 0.1042 0.70108 0.972 0.000 0.020 0.008
#> GSM1269699 2 0.7940 0.48455 0.200 0.576 0.056 0.168
#> GSM1269707 2 0.8384 0.42600 0.268 0.508 0.060 0.164
#> GSM1269651 4 0.5324 0.76735 0.036 0.128 0.056 0.780
#> GSM1269659 4 0.5593 0.76385 0.072 0.040 0.120 0.768
#> GSM1269667 3 0.4569 0.50135 0.144 0.052 0.800 0.004
#> GSM1269675 2 0.6572 0.45868 0.020 0.588 0.340 0.052
#> GSM1269683 3 0.7958 0.22187 0.400 0.076 0.456 0.068
#> GSM1269689 2 0.6102 0.43190 0.032 0.540 0.420 0.008
#> GSM1269697 3 0.6310 -0.17380 0.024 0.404 0.548 0.024
#> GSM1269705 3 0.7056 -0.25144 0.036 0.424 0.492 0.048
#> GSM1269713 3 0.5292 0.15984 0.036 0.252 0.708 0.004
#> GSM1269719 3 0.8559 0.32702 0.320 0.068 0.464 0.148
#> GSM1269725 3 0.5514 0.16834 0.040 0.252 0.700 0.008
#> GSM1269727 3 0.7317 0.40314 0.260 0.088 0.604 0.048
#> GSM1269649 1 0.6880 0.46060 0.644 0.216 0.116 0.024
#> GSM1269657 4 0.4756 0.75793 0.144 0.072 0.000 0.784
#> GSM1269665 1 0.3082 0.68702 0.896 0.040 0.056 0.008
#> GSM1269673 1 0.0564 0.70559 0.988 0.004 0.004 0.004
#> GSM1269681 4 0.3399 0.77937 0.040 0.092 0.000 0.868
#> GSM1269687 1 0.2161 0.70126 0.932 0.016 0.048 0.004
#> GSM1269695 1 0.6112 0.47166 0.668 0.256 0.064 0.012
#> GSM1269703 1 0.1369 0.70735 0.964 0.016 0.016 0.004
#> GSM1269711 1 0.6541 0.44802 0.648 0.228 0.116 0.008
#> GSM1269646 3 0.6506 -0.26871 0.016 0.444 0.500 0.040
#> GSM1269654 3 0.7978 0.40512 0.296 0.064 0.536 0.104
#> GSM1269662 3 0.9156 -0.01371 0.188 0.092 0.360 0.360
#> GSM1269670 2 0.5596 0.56490 0.016 0.712 0.232 0.040
#> GSM1269678 3 0.3962 0.51054 0.152 0.028 0.820 0.000
#> GSM1269692 1 0.8734 -0.19070 0.392 0.068 0.380 0.160
#> GSM1269700 3 0.4466 0.49600 0.156 0.040 0.800 0.004
#> GSM1269708 3 0.5595 0.37122 0.112 0.148 0.736 0.004
#> GSM1269714 3 0.7967 0.19407 0.412 0.068 0.444 0.076
#> GSM1269716 1 0.8020 -0.21927 0.428 0.068 0.424 0.080
#> GSM1269720 4 0.5738 0.77589 0.072 0.072 0.088 0.768
#> GSM1269722 3 0.3940 0.51923 0.152 0.004 0.824 0.020
#> GSM1269644 1 0.1877 0.69730 0.948 0.020 0.012 0.020
#> GSM1269652 1 0.9231 -0.19490 0.404 0.284 0.216 0.096
#> GSM1269660 1 0.4574 0.60206 0.768 0.016 0.208 0.008
#> GSM1269668 1 0.4754 0.59744 0.752 0.024 0.220 0.004
#> GSM1269676 4 0.4224 0.78694 0.100 0.076 0.000 0.824
#> GSM1269684 1 0.0672 0.70429 0.984 0.008 0.008 0.000
#> GSM1269690 1 0.1042 0.70108 0.972 0.000 0.020 0.008
#> GSM1269698 2 0.7940 0.48455 0.200 0.576 0.056 0.168
#> GSM1269706 2 0.8384 0.42600 0.268 0.508 0.060 0.164
#> GSM1269650 4 0.5324 0.76735 0.036 0.128 0.056 0.780
#> GSM1269658 4 0.5593 0.76385 0.072 0.040 0.120 0.768
#> GSM1269666 3 0.4232 0.50899 0.144 0.036 0.816 0.004
#> GSM1269674 2 0.6572 0.45868 0.020 0.588 0.340 0.052
#> GSM1269682 3 0.7950 0.22433 0.392 0.076 0.464 0.068
#> GSM1269688 2 0.6102 0.43190 0.032 0.540 0.420 0.008
#> GSM1269696 3 0.6389 -0.17457 0.024 0.400 0.548 0.028
#> GSM1269704 3 0.7056 -0.25144 0.036 0.424 0.492 0.048
#> GSM1269712 3 0.4001 0.38240 0.048 0.108 0.840 0.004
#> GSM1269718 3 0.8479 0.34198 0.316 0.068 0.476 0.140
#> GSM1269724 3 0.5514 0.16834 0.040 0.252 0.700 0.008
#> GSM1269726 3 0.7472 0.36929 0.288 0.088 0.576 0.048
#> GSM1269648 1 0.6558 0.46613 0.660 0.228 0.092 0.020
#> GSM1269656 4 0.6141 0.56364 0.312 0.072 0.000 0.616
#> GSM1269664 1 0.2909 0.68771 0.904 0.036 0.052 0.008
#> GSM1269672 1 0.0524 0.70598 0.988 0.000 0.008 0.004
#> GSM1269680 4 0.3333 0.77980 0.040 0.088 0.000 0.872
#> GSM1269686 1 0.2161 0.70126 0.932 0.016 0.048 0.004
#> GSM1269694 1 0.6112 0.47166 0.668 0.256 0.064 0.012
#> GSM1269702 1 0.1007 0.70632 0.976 0.008 0.008 0.008
#> GSM1269710 1 0.6417 0.45972 0.656 0.232 0.104 0.008
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM1269647 3 0.4877 0.36307 0.004 0.032 0.732 0.028 0.204
#> GSM1269655 4 0.8161 0.61338 0.224 0.048 0.260 0.432 0.036
#> GSM1269663 4 0.7765 0.52461 0.160 0.180 0.072 0.548 0.040
#> GSM1269671 5 0.6050 0.43533 0.000 0.020 0.344 0.080 0.556
#> GSM1269679 3 0.5175 0.47362 0.072 0.000 0.680 0.240 0.008
#> GSM1269693 4 0.5869 0.71222 0.252 0.068 0.040 0.640 0.000
#> GSM1269701 3 0.6665 0.44893 0.092 0.004 0.604 0.228 0.072
#> GSM1269709 3 0.6423 0.53554 0.100 0.000 0.628 0.200 0.072
#> GSM1269715 4 0.6015 0.74440 0.292 0.012 0.096 0.596 0.004
#> GSM1269717 4 0.5971 0.74618 0.296 0.012 0.104 0.588 0.000
#> GSM1269721 2 0.6594 0.67803 0.048 0.620 0.044 0.244 0.044
#> GSM1269723 3 0.6432 0.20486 0.100 0.004 0.528 0.348 0.020
#> GSM1269645 1 0.3355 0.69290 0.848 0.008 0.004 0.116 0.024
#> GSM1269653 1 0.8195 0.00485 0.404 0.052 0.252 0.028 0.264
#> GSM1269661 1 0.5949 0.59879 0.688 0.008 0.168 0.084 0.052
#> GSM1269669 1 0.5550 0.65125 0.720 0.004 0.148 0.064 0.064
#> GSM1269677 2 0.2766 0.73163 0.056 0.892 0.000 0.012 0.040
#> GSM1269685 1 0.1657 0.73455 0.948 0.008 0.008 0.028 0.008
#> GSM1269691 1 0.1924 0.71363 0.924 0.008 0.004 0.064 0.000
#> GSM1269699 5 0.6570 0.53529 0.128 0.148 0.072 0.008 0.644
#> GSM1269707 5 0.7655 0.49511 0.180 0.152 0.092 0.024 0.552
#> GSM1269651 2 0.5574 0.70251 0.020 0.740 0.056 0.112 0.072
#> GSM1269659 2 0.5884 0.67144 0.064 0.648 0.012 0.252 0.024
#> GSM1269667 3 0.5835 0.43865 0.092 0.004 0.640 0.248 0.016
#> GSM1269675 5 0.6917 0.25518 0.004 0.020 0.408 0.148 0.420
#> GSM1269683 4 0.6332 0.72514 0.300 0.000 0.140 0.548 0.012
#> GSM1269689 3 0.6394 0.06664 0.028 0.008 0.552 0.076 0.336
#> GSM1269697 3 0.3458 0.45499 0.004 0.012 0.832 0.012 0.140
#> GSM1269705 3 0.5839 0.33084 0.024 0.032 0.684 0.052 0.208
#> GSM1269713 3 0.2302 0.59170 0.016 0.000 0.916 0.048 0.020
#> GSM1269719 4 0.8283 0.65012 0.272 0.072 0.192 0.432 0.032
#> GSM1269725 3 0.2362 0.58247 0.024 0.000 0.916 0.032 0.028
#> GSM1269727 4 0.7560 0.62274 0.204 0.004 0.204 0.508 0.080
#> GSM1269649 1 0.6409 0.55244 0.600 0.008 0.036 0.088 0.268
#> GSM1269657 2 0.4112 0.69732 0.124 0.804 0.000 0.016 0.056
#> GSM1269665 1 0.3360 0.70021 0.864 0.008 0.020 0.084 0.024
#> GSM1269673 1 0.0671 0.73295 0.980 0.000 0.004 0.016 0.000
#> GSM1269681 2 0.3359 0.71391 0.036 0.868 0.004 0.028 0.064
#> GSM1269687 1 0.2140 0.73176 0.924 0.000 0.040 0.024 0.012
#> GSM1269695 1 0.5779 0.53213 0.616 0.004 0.032 0.044 0.304
#> GSM1269703 1 0.1074 0.73965 0.968 0.000 0.004 0.016 0.012
#> GSM1269711 1 0.6039 0.50802 0.592 0.000 0.056 0.044 0.308
#> GSM1269646 3 0.4877 0.36307 0.004 0.032 0.732 0.028 0.204
#> GSM1269654 4 0.8144 0.61318 0.220 0.048 0.260 0.436 0.036
#> GSM1269662 4 0.7822 0.52947 0.160 0.172 0.076 0.548 0.044
#> GSM1269670 5 0.6050 0.43533 0.000 0.020 0.344 0.080 0.556
#> GSM1269678 3 0.5240 0.45003 0.080 0.000 0.664 0.252 0.004
#> GSM1269692 4 0.5821 0.70999 0.256 0.068 0.036 0.640 0.000
#> GSM1269700 3 0.6665 0.44893 0.092 0.004 0.604 0.228 0.072
#> GSM1269708 3 0.6451 0.53131 0.100 0.000 0.624 0.204 0.072
#> GSM1269714 4 0.6015 0.74537 0.292 0.012 0.096 0.596 0.004
#> GSM1269716 4 0.5971 0.74618 0.296 0.012 0.104 0.588 0.000
#> GSM1269720 2 0.6594 0.67803 0.048 0.620 0.044 0.244 0.044
#> GSM1269722 3 0.6390 0.21538 0.096 0.004 0.532 0.348 0.020
#> GSM1269644 1 0.2054 0.72064 0.916 0.004 0.000 0.072 0.008
#> GSM1269652 1 0.8157 0.03365 0.416 0.052 0.240 0.028 0.264
#> GSM1269660 1 0.6069 0.58470 0.676 0.008 0.176 0.088 0.052
#> GSM1269668 1 0.5589 0.64621 0.716 0.004 0.152 0.064 0.064
#> GSM1269676 2 0.2766 0.73163 0.056 0.892 0.000 0.012 0.040
#> GSM1269684 1 0.1153 0.73528 0.964 0.000 0.008 0.024 0.004
#> GSM1269690 1 0.1857 0.71649 0.928 0.008 0.004 0.060 0.000
#> GSM1269698 5 0.6570 0.53529 0.128 0.148 0.072 0.008 0.644
#> GSM1269706 5 0.7655 0.49511 0.180 0.152 0.092 0.024 0.552
#> GSM1269650 2 0.5574 0.70251 0.020 0.740 0.056 0.112 0.072
#> GSM1269658 2 0.5884 0.67144 0.064 0.648 0.012 0.252 0.024
#> GSM1269666 3 0.5465 0.46060 0.084 0.000 0.660 0.244 0.012
#> GSM1269674 5 0.6917 0.25518 0.004 0.020 0.408 0.148 0.420
#> GSM1269682 4 0.6317 0.72458 0.296 0.000 0.140 0.552 0.012
#> GSM1269688 3 0.6394 0.06664 0.028 0.008 0.552 0.076 0.336
#> GSM1269696 3 0.3559 0.45488 0.004 0.012 0.828 0.016 0.140
#> GSM1269704 3 0.5809 0.33757 0.024 0.032 0.688 0.052 0.204
#> GSM1269712 3 0.3292 0.60991 0.016 0.000 0.836 0.140 0.008
#> GSM1269718 4 0.8290 0.64369 0.284 0.068 0.196 0.420 0.032
#> GSM1269724 3 0.2362 0.58247 0.024 0.000 0.916 0.032 0.028
#> GSM1269726 4 0.7558 0.62898 0.212 0.004 0.196 0.508 0.080
#> GSM1269648 1 0.6151 0.55946 0.624 0.008 0.028 0.084 0.256
#> GSM1269656 2 0.5931 0.46212 0.296 0.612 0.004 0.032 0.056
#> GSM1269664 1 0.3451 0.69959 0.860 0.008 0.024 0.084 0.024
#> GSM1269672 1 0.0898 0.73407 0.972 0.000 0.008 0.020 0.000
#> GSM1269680 2 0.3359 0.71391 0.036 0.868 0.004 0.028 0.064
#> GSM1269686 1 0.2124 0.73316 0.924 0.000 0.044 0.020 0.012
#> GSM1269694 1 0.5779 0.53213 0.616 0.004 0.032 0.044 0.304
#> GSM1269702 1 0.0854 0.73778 0.976 0.000 0.004 0.008 0.012
#> GSM1269710 1 0.5995 0.50816 0.592 0.000 0.052 0.044 0.312
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM1269647 3 0.583 -0.0448 0.004 0.312 0.580 0.024 0.052 0.028
#> GSM1269655 4 0.788 0.4506 0.180 0.044 0.276 0.424 0.032 0.044
#> GSM1269663 4 0.818 0.3851 0.076 0.088 0.060 0.504 0.132 0.140
#> GSM1269671 2 0.434 0.5717 0.004 0.760 0.152 0.008 0.068 0.008
#> GSM1269679 3 0.460 0.4944 0.012 0.020 0.668 0.284 0.016 0.000
#> GSM1269693 4 0.505 0.6293 0.128 0.016 0.016 0.744 0.056 0.040
#> GSM1269701 3 0.687 0.4019 0.052 0.068 0.544 0.244 0.092 0.000
#> GSM1269709 3 0.658 0.4316 0.104 0.084 0.600 0.184 0.020 0.008
#> GSM1269715 4 0.466 0.6740 0.168 0.004 0.064 0.736 0.024 0.004
#> GSM1269717 4 0.461 0.6746 0.172 0.004 0.064 0.736 0.020 0.004
#> GSM1269721 6 0.711 0.6215 0.024 0.060 0.012 0.228 0.152 0.524
#> GSM1269723 3 0.630 0.2306 0.040 0.056 0.480 0.388 0.036 0.000
#> GSM1269645 1 0.415 0.6784 0.800 0.044 0.008 0.100 0.044 0.004
#> GSM1269653 1 0.868 -0.3396 0.348 0.088 0.176 0.012 0.236 0.140
#> GSM1269661 1 0.634 0.5641 0.636 0.020 0.156 0.084 0.092 0.012
#> GSM1269669 1 0.640 0.5615 0.632 0.056 0.144 0.084 0.084 0.000
#> GSM1269677 6 0.189 0.6633 0.016 0.008 0.000 0.020 0.024 0.932
#> GSM1269685 1 0.194 0.7108 0.928 0.012 0.004 0.040 0.012 0.004
#> GSM1269691 1 0.226 0.7040 0.892 0.000 0.004 0.092 0.008 0.004
#> GSM1269699 5 0.796 0.8646 0.120 0.252 0.060 0.000 0.412 0.156
#> GSM1269707 5 0.811 0.8764 0.172 0.180 0.036 0.012 0.424 0.176
#> GSM1269651 6 0.601 0.6543 0.000 0.100 0.024 0.088 0.132 0.656
#> GSM1269659 6 0.621 0.6568 0.016 0.036 0.004 0.220 0.128 0.596
#> GSM1269667 3 0.547 0.4704 0.040 0.024 0.632 0.268 0.036 0.000
#> GSM1269675 2 0.601 0.6037 0.012 0.628 0.224 0.080 0.040 0.016
#> GSM1269683 4 0.501 0.6583 0.196 0.016 0.100 0.684 0.004 0.000
#> GSM1269689 2 0.701 0.4167 0.024 0.384 0.372 0.036 0.184 0.000
#> GSM1269697 3 0.406 0.1670 0.000 0.236 0.728 0.008 0.020 0.008
#> GSM1269705 3 0.655 -0.1557 0.036 0.364 0.500 0.028 0.032 0.040
#> GSM1269713 3 0.273 0.4772 0.000 0.028 0.880 0.064 0.028 0.000
#> GSM1269719 4 0.836 0.5002 0.248 0.052 0.208 0.388 0.064 0.040
#> GSM1269725 3 0.202 0.4653 0.000 0.020 0.920 0.040 0.020 0.000
#> GSM1269727 4 0.657 0.5381 0.128 0.068 0.148 0.608 0.048 0.000
#> GSM1269649 1 0.595 0.5400 0.632 0.116 0.008 0.040 0.196 0.008
#> GSM1269657 6 0.395 0.5641 0.112 0.008 0.004 0.032 0.036 0.808
#> GSM1269665 1 0.384 0.6859 0.820 0.032 0.012 0.084 0.052 0.000
#> GSM1269673 1 0.145 0.7165 0.948 0.008 0.000 0.032 0.008 0.004
#> GSM1269681 6 0.261 0.6567 0.004 0.028 0.004 0.012 0.060 0.892
#> GSM1269687 1 0.310 0.7109 0.868 0.016 0.036 0.060 0.020 0.000
#> GSM1269695 1 0.593 0.4786 0.600 0.124 0.008 0.036 0.232 0.000
#> GSM1269703 1 0.168 0.7204 0.944 0.012 0.008 0.016 0.016 0.004
#> GSM1269711 1 0.602 0.4438 0.588 0.124 0.012 0.032 0.244 0.000
#> GSM1269646 3 0.583 -0.0448 0.004 0.312 0.580 0.024 0.052 0.028
#> GSM1269654 4 0.788 0.4506 0.180 0.044 0.276 0.424 0.032 0.044
#> GSM1269662 4 0.815 0.3870 0.072 0.092 0.060 0.508 0.132 0.136
#> GSM1269670 2 0.434 0.5717 0.004 0.760 0.152 0.008 0.068 0.008
#> GSM1269678 3 0.461 0.4772 0.012 0.016 0.652 0.304 0.016 0.000
#> GSM1269692 4 0.514 0.6316 0.128 0.016 0.020 0.740 0.056 0.040
#> GSM1269700 3 0.687 0.4019 0.052 0.068 0.544 0.244 0.092 0.000
#> GSM1269708 3 0.660 0.4325 0.104 0.084 0.596 0.188 0.020 0.008
#> GSM1269714 4 0.458 0.6736 0.168 0.004 0.064 0.740 0.020 0.004
#> GSM1269716 4 0.461 0.6746 0.172 0.004 0.064 0.736 0.020 0.004
#> GSM1269720 6 0.711 0.6215 0.024 0.060 0.012 0.228 0.152 0.524
#> GSM1269722 3 0.619 0.2453 0.036 0.056 0.488 0.388 0.032 0.000
#> GSM1269644 1 0.331 0.6999 0.856 0.028 0.004 0.072 0.032 0.008
#> GSM1269652 1 0.862 -0.3123 0.368 0.088 0.164 0.012 0.228 0.140
#> GSM1269660 1 0.630 0.5630 0.640 0.020 0.156 0.080 0.092 0.012
#> GSM1269668 1 0.665 0.5323 0.604 0.056 0.160 0.096 0.084 0.000
#> GSM1269676 6 0.189 0.6633 0.016 0.008 0.000 0.020 0.024 0.932
#> GSM1269684 1 0.187 0.7193 0.932 0.008 0.004 0.036 0.016 0.004
#> GSM1269690 1 0.226 0.7040 0.892 0.000 0.004 0.092 0.008 0.004
#> GSM1269698 5 0.796 0.8646 0.120 0.252 0.060 0.000 0.412 0.156
#> GSM1269706 5 0.811 0.8764 0.172 0.180 0.036 0.012 0.424 0.176
#> GSM1269650 6 0.601 0.6543 0.000 0.100 0.024 0.088 0.132 0.656
#> GSM1269658 6 0.621 0.6568 0.016 0.036 0.004 0.220 0.128 0.596
#> GSM1269666 3 0.524 0.4753 0.028 0.020 0.640 0.276 0.036 0.000
#> GSM1269674 2 0.601 0.6037 0.012 0.628 0.224 0.080 0.040 0.016
#> GSM1269682 4 0.494 0.6566 0.180 0.016 0.104 0.696 0.004 0.000
#> GSM1269688 2 0.701 0.4167 0.024 0.384 0.372 0.036 0.184 0.000
#> GSM1269696 3 0.403 0.1739 0.000 0.232 0.732 0.008 0.020 0.008
#> GSM1269704 3 0.649 -0.1516 0.032 0.364 0.504 0.028 0.032 0.040
#> GSM1269712 3 0.330 0.5262 0.000 0.012 0.820 0.140 0.028 0.000
#> GSM1269718 4 0.838 0.4639 0.248 0.052 0.232 0.368 0.064 0.036
#> GSM1269724 3 0.202 0.4653 0.000 0.020 0.920 0.040 0.020 0.000
#> GSM1269726 4 0.657 0.5454 0.136 0.068 0.140 0.608 0.048 0.000
#> GSM1269648 1 0.582 0.5373 0.644 0.112 0.008 0.036 0.192 0.008
#> GSM1269656 6 0.551 0.2444 0.276 0.012 0.008 0.040 0.036 0.628
#> GSM1269664 1 0.371 0.6886 0.828 0.028 0.012 0.080 0.052 0.000
#> GSM1269672 1 0.123 0.7173 0.952 0.000 0.000 0.040 0.004 0.004
#> GSM1269680 6 0.261 0.6567 0.004 0.028 0.004 0.012 0.060 0.892
#> GSM1269686 1 0.303 0.7118 0.872 0.016 0.032 0.060 0.020 0.000
#> GSM1269694 1 0.593 0.4786 0.600 0.124 0.008 0.036 0.232 0.000
#> GSM1269702 1 0.135 0.7146 0.956 0.008 0.004 0.020 0.008 0.004
#> GSM1269710 1 0.602 0.4438 0.588 0.124 0.012 0.032 0.244 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 agent(p) disease.state(p) gender(p) individual(p) k
#> SD:kmeans 18 0.617 0.393 2.05e-04 5.50e-02 2
#> SD:kmeans 48 1.000 0.319 1.52e-08 2.21e-05 3
#> SD:kmeans 38 1.000 0.248 1.47e-05 3.52e-06 4
#> SD:kmeans 59 0.999 0.286 9.16e-10 3.34e-10 5
#> SD:kmeans 51 0.948 0.317 1.58e-07 3.74e-10 6
If matrix rows can be associated to genes, consider to use functional_enrichment(res,
...) to perform function enrichment for the signature genes. See this vignette for more detailed explanations.
The object with results only for a single top-value method and a single partition method can be extracted as:
res = res_list["SD", "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 51941 rows and 84 columns.
#> Top rows (1000, 2000, 3000, 4000, 5000) are extracted by 'SD' method.
#> Subgroups are detected by 'skmeans' method.
#> Performed in total 1250 partitions by row resampling.
#> Best k for subgroups seems to be 2.
#>
#> Following methods can be applied to this 'ConsensusPartition' object:
#> [1] "cola_report" "collect_classes" "collect_plots"
#> [4] "collect_stats" "colnames" "compare_signatures"
#> [7] "consensus_heatmap" "dimension_reduction" "functional_enrichment"
#> [10] "get_anno_col" "get_anno" "get_classes"
#> [13] "get_consensus" "get_matrix" "get_membership"
#> [16] "get_param" "get_signatures" "get_stats"
#> [19] "is_best_k" "is_stable_k" "membership_heatmap"
#> [22] "ncol" "nrow" "plot_ecdf"
#> [25] "rownames" "select_partition_number" "show"
#> [28] "suggest_best_k" "test_to_known_factors"
collect_plots() function collects all the plots made from res for all k (number of partitions)
into one single page to provide an easy and fast comparison between different k.
collect_plots(res)
The plots are:
k and the heatmap of
predicted classes for each k.k.k.k.All the plots in panels can be made by individual functions and they are plotted later in this section.
select_partition_number() produces several plots showing different
statistics for choosing “optimized” k. There are following statistics:
k;k, the area increased is defined as \(A_k - A_{k-1}\).The detailed explanations of these statistics can be found in the cola vignette.
Generally speaking, lower PAC score, higher mean silhouette score or higher
concordance corresponds to better partition. Rand index and Jaccard index
measure how similar the current partition is compared to partition with k-1.
If they are too similar, we won't accept k is better than k-1.
select_partition_number(res)
The numeric values for all these statistics can be obtained by get_stats().
get_stats(res)
#> k 1-PAC mean_silhouette concordance area_increased Rand Jaccard
#> 2 2 0.00714 0.307 0.653 0.5028 0.494 0.494
#> 3 3 0.03797 0.460 0.629 0.3354 0.617 0.364
#> 4 4 0.13762 0.423 0.564 0.1215 0.840 0.562
#> 5 5 0.26095 0.302 0.489 0.0651 0.919 0.691
#> 6 6 0.38234 0.193 0.437 0.0411 0.907 0.613
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
#> GSM1269647 1 0.605 0.48258 0.852 0.148
#> GSM1269655 1 0.990 0.36344 0.560 0.440
#> GSM1269663 1 1.000 0.24394 0.508 0.492
#> GSM1269671 1 0.767 0.40849 0.776 0.224
#> GSM1269679 1 0.866 0.47043 0.712 0.288
#> GSM1269693 2 0.936 -0.10977 0.352 0.648
#> GSM1269701 1 0.833 0.45312 0.736 0.264
#> GSM1269709 1 0.821 0.52037 0.744 0.256
#> GSM1269715 1 1.000 0.29009 0.504 0.496
#> GSM1269717 2 0.991 -0.23234 0.444 0.556
#> GSM1269721 1 0.999 0.15663 0.516 0.484
#> GSM1269723 1 0.929 0.46677 0.656 0.344
#> GSM1269645 2 0.814 0.47613 0.252 0.748
#> GSM1269653 2 1.000 0.25769 0.496 0.504
#> GSM1269661 2 0.991 0.30419 0.444 0.556
#> GSM1269669 2 0.993 0.24828 0.452 0.548
#> GSM1269677 2 0.802 0.38327 0.244 0.756
#> GSM1269685 2 0.584 0.48670 0.140 0.860
#> GSM1269691 2 0.615 0.47821 0.152 0.848
#> GSM1269699 1 0.990 -0.22809 0.560 0.440
#> GSM1269707 2 0.963 0.35271 0.388 0.612
#> GSM1269651 1 0.999 0.13447 0.516 0.484
#> GSM1269659 2 0.980 -0.05718 0.416 0.584
#> GSM1269667 1 0.866 0.47465 0.712 0.288
#> GSM1269675 1 0.775 0.49491 0.772 0.228
#> GSM1269683 1 0.998 0.30973 0.528 0.472
#> GSM1269689 1 0.506 0.50295 0.888 0.112
#> GSM1269697 1 0.416 0.51976 0.916 0.084
#> GSM1269705 1 0.788 0.48072 0.764 0.236
#> GSM1269713 1 0.518 0.52580 0.884 0.116
#> GSM1269719 1 0.999 0.24107 0.516 0.484
#> GSM1269725 1 0.456 0.53771 0.904 0.096
#> GSM1269727 1 0.939 0.44671 0.644 0.356
#> GSM1269649 1 1.000 -0.28903 0.512 0.488
#> GSM1269657 2 0.781 0.40050 0.232 0.768
#> GSM1269665 2 0.855 0.45117 0.280 0.720
#> GSM1269673 2 0.760 0.48369 0.220 0.780
#> GSM1269681 2 0.913 0.36261 0.328 0.672
#> GSM1269687 2 0.839 0.45596 0.268 0.732
#> GSM1269695 2 0.990 0.35337 0.440 0.560
#> GSM1269703 2 0.767 0.48040 0.224 0.776
#> GSM1269711 2 0.997 0.29732 0.468 0.532
#> GSM1269646 1 0.595 0.49774 0.856 0.144
#> GSM1269654 1 1.000 0.28940 0.512 0.488
#> GSM1269662 2 0.998 -0.22429 0.476 0.524
#> GSM1269670 1 0.625 0.45642 0.844 0.156
#> GSM1269678 1 0.876 0.48782 0.704 0.296
#> GSM1269692 2 0.891 -0.07219 0.308 0.692
#> GSM1269700 1 0.775 0.45178 0.772 0.228
#> GSM1269708 1 0.827 0.52093 0.740 0.260
#> GSM1269714 2 0.998 -0.27100 0.472 0.528
#> GSM1269716 2 0.988 -0.22212 0.436 0.564
#> GSM1269720 2 1.000 -0.15273 0.488 0.512
#> GSM1269722 1 0.913 0.46990 0.672 0.328
#> GSM1269644 2 0.541 0.48603 0.124 0.876
#> GSM1269652 1 1.000 -0.27987 0.508 0.492
#> GSM1269660 2 0.983 0.35183 0.424 0.576
#> GSM1269668 2 0.988 0.27580 0.436 0.564
#> GSM1269676 2 0.788 0.38910 0.236 0.764
#> GSM1269684 2 0.644 0.48415 0.164 0.836
#> GSM1269690 2 0.563 0.46691 0.132 0.868
#> GSM1269698 1 0.998 -0.25975 0.528 0.472
#> GSM1269706 2 0.988 0.32182 0.436 0.564
#> GSM1269650 2 1.000 -0.12898 0.488 0.512
#> GSM1269658 2 0.973 -0.00673 0.404 0.596
#> GSM1269666 1 0.855 0.47648 0.720 0.280
#> GSM1269674 1 0.844 0.42453 0.728 0.272
#> GSM1269682 2 0.999 -0.28116 0.480 0.520
#> GSM1269688 1 0.552 0.50883 0.872 0.128
#> GSM1269696 1 0.311 0.51729 0.944 0.056
#> GSM1269704 1 0.814 0.44764 0.748 0.252
#> GSM1269712 1 0.753 0.51672 0.784 0.216
#> GSM1269718 1 0.999 0.30011 0.520 0.480
#> GSM1269724 1 0.615 0.53794 0.848 0.152
#> GSM1269726 1 0.936 0.43114 0.648 0.352
#> GSM1269648 2 0.993 0.33995 0.452 0.548
#> GSM1269656 2 0.802 0.41262 0.244 0.756
#> GSM1269664 2 0.850 0.43702 0.276 0.724
#> GSM1269672 2 0.788 0.46313 0.236 0.764
#> GSM1269680 2 0.833 0.38327 0.264 0.736
#> GSM1269686 2 0.895 0.39610 0.312 0.688
#> GSM1269694 2 0.966 0.40685 0.392 0.608
#> GSM1269702 2 0.615 0.49234 0.152 0.848
#> GSM1269710 2 0.987 0.35008 0.432 0.568
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM1269647 2 0.861 0.34582 0.364 0.528 0.108
#> GSM1269655 1 0.886 0.40740 0.532 0.332 0.136
#> GSM1269663 2 0.922 -0.10152 0.360 0.480 0.160
#> GSM1269671 2 0.932 0.39285 0.272 0.516 0.212
#> GSM1269679 1 0.422 0.59269 0.868 0.032 0.100
#> GSM1269693 1 0.971 0.31760 0.404 0.376 0.220
#> GSM1269701 1 0.701 0.54451 0.696 0.064 0.240
#> GSM1269709 1 0.802 0.52723 0.632 0.108 0.260
#> GSM1269715 1 0.880 0.49964 0.568 0.156 0.276
#> GSM1269717 1 0.887 0.47934 0.556 0.156 0.288
#> GSM1269721 2 0.543 0.54886 0.088 0.820 0.092
#> GSM1269723 1 0.718 0.58196 0.716 0.168 0.116
#> GSM1269645 3 0.892 0.52805 0.152 0.304 0.544
#> GSM1269653 3 0.958 0.26508 0.212 0.332 0.456
#> GSM1269661 3 0.917 0.51383 0.312 0.172 0.516
#> GSM1269669 3 0.588 0.57182 0.272 0.012 0.716
#> GSM1269677 2 0.313 0.55347 0.008 0.904 0.088
#> GSM1269685 3 0.652 0.66063 0.080 0.168 0.752
#> GSM1269691 3 0.802 0.61586 0.156 0.188 0.656
#> GSM1269699 2 0.902 0.28162 0.148 0.516 0.336
#> GSM1269707 2 0.833 0.23228 0.096 0.564 0.340
#> GSM1269651 2 0.401 0.55787 0.084 0.880 0.036
#> GSM1269659 2 0.593 0.46776 0.124 0.792 0.084
#> GSM1269667 1 0.627 0.58750 0.768 0.076 0.156
#> GSM1269675 2 0.898 0.41838 0.232 0.564 0.204
#> GSM1269683 1 0.836 0.54982 0.616 0.140 0.244
#> GSM1269689 2 0.982 0.12848 0.364 0.392 0.244
#> GSM1269697 1 0.926 0.10921 0.520 0.284 0.196
#> GSM1269705 2 0.918 0.36667 0.324 0.508 0.168
#> GSM1269713 1 0.779 0.42022 0.672 0.192 0.136
#> GSM1269719 2 0.882 -0.12528 0.408 0.476 0.116
#> GSM1269725 1 0.732 0.44230 0.700 0.196 0.104
#> GSM1269727 1 0.880 0.54531 0.580 0.180 0.240
#> GSM1269649 3 0.767 0.59538 0.168 0.148 0.684
#> GSM1269657 2 0.560 0.44378 0.020 0.764 0.216
#> GSM1269665 3 0.850 0.60875 0.220 0.168 0.612
#> GSM1269673 3 0.711 0.66229 0.184 0.100 0.716
#> GSM1269681 2 0.357 0.58009 0.040 0.900 0.060
#> GSM1269687 3 0.852 0.62630 0.204 0.184 0.612
#> GSM1269695 3 0.596 0.64932 0.136 0.076 0.788
#> GSM1269703 3 0.740 0.66151 0.144 0.152 0.704
#> GSM1269711 3 0.740 0.62369 0.180 0.120 0.700
#> GSM1269646 2 0.879 0.36221 0.328 0.540 0.132
#> GSM1269654 1 0.932 0.39005 0.464 0.368 0.168
#> GSM1269662 2 0.918 -0.00996 0.324 0.508 0.168
#> GSM1269670 2 0.938 0.38319 0.276 0.508 0.216
#> GSM1269678 1 0.651 0.59013 0.748 0.072 0.180
#> GSM1269692 1 0.977 0.33994 0.416 0.348 0.236
#> GSM1269700 1 0.573 0.56104 0.752 0.020 0.228
#> GSM1269708 1 0.802 0.54620 0.632 0.108 0.260
#> GSM1269714 1 0.829 0.54271 0.624 0.140 0.236
#> GSM1269716 1 0.860 0.49805 0.580 0.136 0.284
#> GSM1269720 2 0.545 0.54008 0.116 0.816 0.068
#> GSM1269722 1 0.674 0.60105 0.744 0.100 0.156
#> GSM1269644 3 0.808 0.56467 0.096 0.296 0.608
#> GSM1269652 3 0.906 0.49767 0.200 0.248 0.552
#> GSM1269660 3 0.979 0.39122 0.296 0.268 0.436
#> GSM1269668 3 0.686 0.50764 0.356 0.024 0.620
#> GSM1269676 2 0.392 0.54178 0.012 0.868 0.120
#> GSM1269684 3 0.691 0.65927 0.120 0.144 0.736
#> GSM1269690 3 0.770 0.62400 0.140 0.180 0.680
#> GSM1269698 2 0.888 0.37118 0.144 0.540 0.316
#> GSM1269706 3 0.907 -0.03641 0.136 0.428 0.436
#> GSM1269650 2 0.380 0.56418 0.080 0.888 0.032
#> GSM1269658 2 0.633 0.43752 0.144 0.768 0.088
#> GSM1269666 1 0.554 0.59945 0.812 0.072 0.116
#> GSM1269674 2 0.918 0.35123 0.324 0.508 0.168
#> GSM1269682 1 0.877 0.53183 0.584 0.236 0.180
#> GSM1269688 1 0.995 -0.10378 0.380 0.328 0.292
#> GSM1269696 1 0.874 0.06319 0.536 0.340 0.124
#> GSM1269704 2 0.910 0.34544 0.348 0.500 0.152
#> GSM1269712 1 0.661 0.57183 0.752 0.096 0.152
#> GSM1269718 1 0.947 0.30283 0.452 0.360 0.188
#> GSM1269724 1 0.777 0.44312 0.676 0.176 0.148
#> GSM1269726 1 0.902 0.51237 0.528 0.156 0.316
#> GSM1269648 3 0.711 0.63903 0.100 0.184 0.716
#> GSM1269656 2 0.682 0.23596 0.028 0.644 0.328
#> GSM1269664 3 0.829 0.61018 0.256 0.128 0.616
#> GSM1269672 3 0.646 0.66242 0.176 0.072 0.752
#> GSM1269680 2 0.338 0.56431 0.012 0.896 0.092
#> GSM1269686 3 0.772 0.65044 0.220 0.112 0.668
#> GSM1269694 3 0.537 0.65747 0.128 0.056 0.816
#> GSM1269702 3 0.719 0.64099 0.080 0.224 0.696
#> GSM1269710 3 0.691 0.64640 0.180 0.092 0.728
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM1269647 2 0.683 0.53630 0.036 0.656 0.092 0.216
#> GSM1269655 3 0.929 0.22940 0.096 0.252 0.396 0.256
#> GSM1269663 4 0.864 0.18716 0.104 0.112 0.312 0.472
#> GSM1269671 2 0.669 0.51137 0.096 0.680 0.040 0.184
#> GSM1269679 3 0.730 0.24436 0.072 0.360 0.532 0.036
#> GSM1269693 3 0.790 0.30781 0.144 0.028 0.492 0.336
#> GSM1269701 3 0.760 0.36211 0.188 0.264 0.536 0.012
#> GSM1269709 2 0.810 0.22473 0.124 0.528 0.288 0.060
#> GSM1269715 3 0.711 0.51919 0.176 0.064 0.660 0.100
#> GSM1269717 3 0.753 0.53816 0.160 0.084 0.636 0.120
#> GSM1269721 4 0.661 0.56017 0.068 0.100 0.124 0.708
#> GSM1269723 3 0.806 0.32619 0.088 0.324 0.512 0.076
#> GSM1269645 1 0.882 0.52711 0.500 0.104 0.192 0.204
#> GSM1269653 2 0.977 0.05796 0.260 0.320 0.152 0.268
#> GSM1269661 1 0.933 0.36675 0.412 0.232 0.244 0.112
#> GSM1269669 1 0.690 0.53812 0.604 0.112 0.272 0.012
#> GSM1269677 4 0.398 0.59370 0.060 0.052 0.028 0.860
#> GSM1269685 1 0.780 0.59594 0.596 0.060 0.188 0.156
#> GSM1269691 1 0.771 0.54204 0.580 0.036 0.180 0.204
#> GSM1269699 4 0.910 -0.00563 0.248 0.332 0.068 0.352
#> GSM1269707 4 0.879 0.27848 0.228 0.240 0.068 0.464
#> GSM1269651 4 0.625 0.52292 0.040 0.128 0.108 0.724
#> GSM1269659 4 0.445 0.56387 0.044 0.008 0.136 0.812
#> GSM1269667 3 0.801 0.27625 0.168 0.308 0.496 0.028
#> GSM1269675 2 0.881 0.44076 0.132 0.508 0.144 0.216
#> GSM1269683 3 0.754 0.52912 0.140 0.100 0.640 0.120
#> GSM1269689 2 0.835 0.44347 0.204 0.556 0.136 0.104
#> GSM1269697 2 0.580 0.52892 0.080 0.760 0.108 0.052
#> GSM1269705 2 0.828 0.44104 0.088 0.552 0.136 0.224
#> GSM1269713 2 0.694 0.34432 0.056 0.616 0.280 0.048
#> GSM1269719 4 0.962 0.02718 0.144 0.208 0.296 0.352
#> GSM1269725 2 0.596 0.38236 0.028 0.676 0.264 0.032
#> GSM1269727 3 0.868 0.37666 0.152 0.224 0.516 0.108
#> GSM1269649 1 0.790 0.52932 0.584 0.228 0.100 0.088
#> GSM1269657 4 0.605 0.57504 0.116 0.084 0.056 0.744
#> GSM1269665 1 0.818 0.57961 0.584 0.140 0.156 0.120
#> GSM1269673 1 0.718 0.61731 0.664 0.092 0.156 0.088
#> GSM1269681 4 0.460 0.56767 0.044 0.100 0.032 0.824
#> GSM1269687 1 0.842 0.57159 0.544 0.108 0.220 0.128
#> GSM1269695 1 0.677 0.58900 0.684 0.160 0.108 0.048
#> GSM1269703 1 0.763 0.60914 0.632 0.096 0.136 0.136
#> GSM1269711 1 0.721 0.57197 0.648 0.188 0.104 0.060
#> GSM1269646 2 0.776 0.49382 0.080 0.592 0.096 0.232
#> GSM1269654 3 0.896 0.32292 0.092 0.172 0.444 0.292
#> GSM1269662 4 0.930 0.11567 0.132 0.172 0.276 0.420
#> GSM1269670 2 0.703 0.48277 0.096 0.652 0.048 0.204
#> GSM1269678 3 0.777 0.32547 0.104 0.292 0.552 0.052
#> GSM1269692 3 0.767 0.32734 0.136 0.024 0.524 0.316
#> GSM1269700 3 0.763 0.32698 0.148 0.300 0.532 0.020
#> GSM1269708 2 0.821 0.10523 0.112 0.488 0.336 0.064
#> GSM1269714 3 0.610 0.52383 0.160 0.048 0.728 0.064
#> GSM1269716 3 0.691 0.53744 0.140 0.068 0.684 0.108
#> GSM1269720 4 0.624 0.56113 0.040 0.132 0.104 0.724
#> GSM1269722 3 0.754 0.39714 0.084 0.316 0.552 0.048
#> GSM1269644 1 0.801 0.50752 0.512 0.032 0.168 0.288
#> GSM1269652 1 0.963 0.05973 0.336 0.284 0.128 0.252
#> GSM1269660 1 0.978 0.24788 0.356 0.244 0.204 0.196
#> GSM1269668 1 0.692 0.46214 0.552 0.096 0.344 0.008
#> GSM1269676 4 0.468 0.58248 0.100 0.048 0.032 0.820
#> GSM1269684 1 0.786 0.58404 0.576 0.056 0.232 0.136
#> GSM1269690 1 0.766 0.54113 0.584 0.036 0.212 0.168
#> GSM1269698 4 0.846 0.05817 0.196 0.368 0.036 0.400
#> GSM1269706 4 0.911 0.13826 0.276 0.280 0.068 0.376
#> GSM1269650 4 0.505 0.56228 0.028 0.096 0.076 0.800
#> GSM1269658 4 0.466 0.55687 0.040 0.016 0.140 0.804
#> GSM1269666 3 0.720 0.30349 0.068 0.324 0.568 0.040
#> GSM1269674 2 0.865 0.41293 0.080 0.484 0.156 0.280
#> GSM1269682 3 0.738 0.52872 0.160 0.068 0.644 0.128
#> GSM1269688 2 0.869 0.42352 0.216 0.520 0.144 0.120
#> GSM1269696 2 0.569 0.52782 0.056 0.768 0.104 0.072
#> GSM1269704 2 0.786 0.41057 0.076 0.572 0.096 0.256
#> GSM1269712 2 0.814 -0.02370 0.128 0.444 0.384 0.044
#> GSM1269718 4 0.972 -0.12028 0.160 0.208 0.312 0.320
#> GSM1269724 2 0.730 0.35471 0.076 0.588 0.288 0.048
#> GSM1269726 3 0.897 0.45318 0.244 0.132 0.480 0.144
#> GSM1269648 1 0.714 0.56983 0.648 0.204 0.068 0.080
#> GSM1269656 4 0.732 0.52971 0.148 0.096 0.100 0.656
#> GSM1269664 1 0.769 0.56919 0.608 0.124 0.200 0.068
#> GSM1269672 1 0.637 0.62783 0.720 0.064 0.140 0.076
#> GSM1269680 4 0.486 0.57510 0.068 0.076 0.040 0.816
#> GSM1269686 1 0.791 0.58833 0.596 0.120 0.196 0.088
#> GSM1269694 1 0.679 0.60658 0.692 0.140 0.104 0.064
#> GSM1269702 1 0.713 0.59569 0.648 0.048 0.112 0.192
#> GSM1269710 1 0.670 0.61234 0.700 0.140 0.080 0.080
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM1269647 3 0.730 0.4275 0.016 0.204 0.572 0.080 0.128
#> GSM1269655 4 0.886 0.1483 0.076 0.260 0.264 0.344 0.056
#> GSM1269663 2 0.920 0.1803 0.140 0.396 0.116 0.232 0.116
#> GSM1269671 3 0.809 0.2894 0.052 0.120 0.444 0.060 0.324
#> GSM1269679 4 0.764 0.1770 0.076 0.028 0.404 0.412 0.080
#> GSM1269693 4 0.755 0.2415 0.140 0.300 0.032 0.492 0.036
#> GSM1269701 3 0.866 -0.1201 0.176 0.020 0.368 0.288 0.148
#> GSM1269709 3 0.855 0.0461 0.124 0.056 0.404 0.320 0.096
#> GSM1269715 4 0.642 0.4490 0.156 0.076 0.040 0.676 0.052
#> GSM1269717 4 0.672 0.4485 0.132 0.104 0.064 0.660 0.040
#> GSM1269721 2 0.664 0.5376 0.048 0.672 0.072 0.124 0.084
#> GSM1269723 4 0.847 0.2456 0.092 0.056 0.312 0.420 0.120
#> GSM1269645 1 0.877 0.2562 0.420 0.172 0.040 0.148 0.220
#> GSM1269653 5 0.883 0.3325 0.204 0.216 0.196 0.020 0.364
#> GSM1269661 1 0.961 0.0907 0.312 0.096 0.208 0.172 0.212
#> GSM1269669 1 0.769 0.3305 0.524 0.008 0.116 0.168 0.184
#> GSM1269677 2 0.519 0.5241 0.080 0.760 0.024 0.024 0.112
#> GSM1269685 1 0.775 0.3187 0.564 0.136 0.044 0.092 0.164
#> GSM1269691 1 0.769 0.3807 0.556 0.168 0.032 0.160 0.084
#> GSM1269699 5 0.776 0.3632 0.096 0.224 0.136 0.020 0.524
#> GSM1269707 5 0.854 0.3321 0.136 0.264 0.100 0.056 0.444
#> GSM1269651 2 0.604 0.5509 0.036 0.716 0.104 0.080 0.064
#> GSM1269659 2 0.469 0.5761 0.028 0.780 0.016 0.140 0.036
#> GSM1269667 4 0.827 0.1294 0.088 0.040 0.348 0.404 0.120
#> GSM1269675 3 0.866 0.3235 0.040 0.176 0.396 0.104 0.284
#> GSM1269683 4 0.743 0.4654 0.136 0.084 0.084 0.612 0.084
#> GSM1269689 3 0.766 0.3568 0.080 0.040 0.492 0.072 0.316
#> GSM1269697 3 0.481 0.4750 0.040 0.036 0.792 0.032 0.100
#> GSM1269705 3 0.834 0.3712 0.040 0.180 0.480 0.100 0.200
#> GSM1269713 3 0.682 0.3172 0.024 0.032 0.604 0.208 0.132
#> GSM1269719 2 0.912 0.2349 0.108 0.420 0.160 0.196 0.116
#> GSM1269725 3 0.641 0.3511 0.040 0.032 0.672 0.140 0.116
#> GSM1269727 4 0.874 0.3092 0.196 0.048 0.136 0.440 0.180
#> GSM1269649 5 0.883 -0.0126 0.332 0.072 0.164 0.080 0.352
#> GSM1269657 2 0.656 0.4478 0.116 0.660 0.036 0.040 0.148
#> GSM1269665 1 0.830 0.3784 0.524 0.104 0.080 0.132 0.160
#> GSM1269673 1 0.663 0.4136 0.672 0.084 0.044 0.124 0.076
#> GSM1269681 2 0.606 0.4891 0.040 0.688 0.108 0.016 0.148
#> GSM1269687 1 0.850 0.3812 0.512 0.120 0.108 0.136 0.124
#> GSM1269695 1 0.709 0.0164 0.432 0.016 0.080 0.048 0.424
#> GSM1269703 1 0.766 0.3128 0.564 0.072 0.056 0.104 0.204
#> GSM1269711 5 0.787 0.0103 0.344 0.028 0.096 0.088 0.444
#> GSM1269646 3 0.772 0.4041 0.036 0.208 0.544 0.072 0.140
#> GSM1269654 4 0.925 0.2519 0.104 0.236 0.196 0.368 0.096
#> GSM1269662 2 0.882 0.2106 0.104 0.428 0.088 0.256 0.124
#> GSM1269670 3 0.795 0.2722 0.064 0.108 0.404 0.040 0.384
#> GSM1269678 4 0.767 0.2226 0.072 0.032 0.320 0.484 0.092
#> GSM1269692 4 0.792 0.2260 0.148 0.296 0.028 0.464 0.064
#> GSM1269700 4 0.878 0.1258 0.188 0.024 0.288 0.352 0.148
#> GSM1269708 4 0.845 0.1193 0.112 0.032 0.332 0.388 0.136
#> GSM1269714 4 0.634 0.4645 0.152 0.060 0.064 0.684 0.040
#> GSM1269716 4 0.643 0.4797 0.128 0.080 0.072 0.684 0.036
#> GSM1269720 2 0.617 0.5641 0.028 0.700 0.096 0.116 0.060
#> GSM1269722 4 0.810 0.3231 0.104 0.044 0.272 0.484 0.096
#> GSM1269644 1 0.808 0.3173 0.504 0.228 0.036 0.100 0.132
#> GSM1269652 5 0.921 0.2452 0.284 0.160 0.156 0.064 0.336
#> GSM1269660 1 0.983 0.0124 0.268 0.132 0.188 0.180 0.232
#> GSM1269668 1 0.792 0.3209 0.468 0.008 0.116 0.264 0.144
#> GSM1269676 2 0.565 0.5215 0.080 0.732 0.032 0.032 0.124
#> GSM1269684 1 0.748 0.4120 0.580 0.108 0.028 0.172 0.112
#> GSM1269690 1 0.721 0.3950 0.588 0.140 0.012 0.168 0.092
#> GSM1269698 5 0.798 0.3437 0.084 0.216 0.168 0.028 0.504
#> GSM1269706 5 0.824 0.3750 0.136 0.248 0.112 0.032 0.472
#> GSM1269650 2 0.516 0.5707 0.020 0.772 0.080 0.072 0.056
#> GSM1269658 2 0.513 0.5763 0.052 0.756 0.016 0.140 0.036
#> GSM1269666 4 0.801 0.2547 0.100 0.040 0.340 0.440 0.080
#> GSM1269674 3 0.910 0.3321 0.060 0.196 0.384 0.140 0.220
#> GSM1269682 4 0.625 0.4754 0.100 0.076 0.076 0.704 0.044
#> GSM1269688 3 0.813 0.2956 0.088 0.044 0.420 0.096 0.352
#> GSM1269696 3 0.512 0.4770 0.016 0.060 0.772 0.064 0.088
#> GSM1269704 3 0.830 0.3581 0.064 0.188 0.492 0.068 0.188
#> GSM1269712 3 0.843 -0.0322 0.088 0.040 0.400 0.324 0.148
#> GSM1269718 2 0.987 -0.1086 0.184 0.260 0.232 0.180 0.144
#> GSM1269724 3 0.711 0.2837 0.076 0.032 0.608 0.192 0.092
#> GSM1269726 4 0.872 0.2651 0.168 0.096 0.064 0.436 0.236
#> GSM1269648 5 0.728 0.0104 0.424 0.064 0.060 0.028 0.424
#> GSM1269656 2 0.831 0.1756 0.152 0.480 0.056 0.076 0.236
#> GSM1269664 1 0.824 0.3997 0.516 0.080 0.072 0.188 0.144
#> GSM1269672 1 0.622 0.4274 0.692 0.064 0.024 0.132 0.088
#> GSM1269680 2 0.559 0.5103 0.088 0.732 0.052 0.012 0.116
#> GSM1269686 1 0.834 0.3499 0.504 0.056 0.108 0.152 0.180
#> GSM1269694 1 0.709 -0.0162 0.452 0.024 0.072 0.044 0.408
#> GSM1269702 1 0.732 0.3072 0.568 0.132 0.032 0.048 0.220
#> GSM1269710 5 0.738 -0.0359 0.396 0.048 0.056 0.052 0.448
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM1269647 3 0.763 0.18737 0.016 0.192 0.480 0.016 0.144 0.152
#> GSM1269655 4 0.928 0.03067 0.052 0.068 0.228 0.256 0.180 0.216
#> GSM1269663 6 0.871 -0.14855 0.068 0.072 0.060 0.156 0.320 0.324
#> GSM1269671 2 0.746 0.06292 0.028 0.436 0.332 0.020 0.064 0.120
#> GSM1269679 3 0.772 0.08115 0.060 0.068 0.452 0.264 0.148 0.008
#> GSM1269693 4 0.708 0.27577 0.044 0.036 0.032 0.568 0.192 0.128
#> GSM1269701 3 0.889 0.07558 0.140 0.196 0.328 0.204 0.120 0.012
#> GSM1269709 3 0.844 0.09783 0.080 0.116 0.340 0.328 0.116 0.020
#> GSM1269715 4 0.568 0.39636 0.096 0.028 0.052 0.720 0.068 0.036
#> GSM1269717 4 0.551 0.39762 0.080 0.016 0.060 0.732 0.060 0.052
#> GSM1269721 6 0.774 0.25423 0.028 0.096 0.032 0.144 0.204 0.496
#> GSM1269723 4 0.842 0.07684 0.056 0.080 0.336 0.340 0.148 0.040
#> GSM1269645 1 0.907 0.11707 0.312 0.136 0.036 0.108 0.264 0.144
#> GSM1269653 6 0.916 -0.17748 0.136 0.244 0.184 0.032 0.112 0.292
#> GSM1269661 1 0.968 -0.09086 0.244 0.148 0.176 0.112 0.228 0.092
#> GSM1269669 1 0.783 0.38047 0.500 0.132 0.128 0.112 0.124 0.004
#> GSM1269677 6 0.346 0.40852 0.040 0.028 0.012 0.024 0.036 0.860
#> GSM1269685 1 0.764 0.39451 0.508 0.132 0.012 0.164 0.040 0.144
#> GSM1269691 1 0.788 0.35569 0.500 0.056 0.028 0.192 0.132 0.092
#> GSM1269699 2 0.757 0.25459 0.100 0.480 0.080 0.020 0.044 0.276
#> GSM1269707 2 0.831 0.14660 0.128 0.368 0.064 0.020 0.108 0.312
#> GSM1269651 6 0.657 0.29113 0.016 0.060 0.040 0.084 0.180 0.620
#> GSM1269659 6 0.640 0.29403 0.024 0.020 0.008 0.164 0.208 0.576
#> GSM1269667 3 0.865 0.05023 0.072 0.120 0.336 0.224 0.232 0.016
#> GSM1269675 2 0.892 -0.03255 0.056 0.312 0.288 0.056 0.172 0.116
#> GSM1269683 4 0.763 0.35525 0.092 0.076 0.084 0.564 0.132 0.052
#> GSM1269689 3 0.823 0.08998 0.076 0.344 0.368 0.052 0.112 0.048
#> GSM1269697 3 0.527 0.33316 0.032 0.152 0.724 0.020 0.048 0.024
#> GSM1269705 3 0.865 0.15281 0.024 0.220 0.388 0.080 0.144 0.144
#> GSM1269713 3 0.739 0.35737 0.036 0.128 0.576 0.108 0.100 0.052
#> GSM1269719 6 0.909 -0.12946 0.052 0.080 0.104 0.208 0.264 0.292
#> GSM1269725 3 0.526 0.40644 0.024 0.060 0.744 0.076 0.080 0.016
#> GSM1269727 4 0.919 0.19221 0.080 0.192 0.172 0.328 0.176 0.052
#> GSM1269649 2 0.788 -0.16548 0.292 0.432 0.060 0.028 0.140 0.048
#> GSM1269657 6 0.556 0.34681 0.076 0.040 0.028 0.052 0.072 0.732
#> GSM1269665 1 0.839 0.30279 0.436 0.128 0.044 0.104 0.228 0.060
#> GSM1269673 1 0.681 0.45443 0.620 0.064 0.020 0.124 0.116 0.056
#> GSM1269681 6 0.383 0.39475 0.024 0.040 0.032 0.004 0.068 0.832
#> GSM1269687 1 0.812 0.37855 0.496 0.064 0.088 0.168 0.136 0.048
#> GSM1269695 1 0.702 0.20930 0.432 0.396 0.032 0.048 0.068 0.024
#> GSM1269703 1 0.723 0.41510 0.572 0.116 0.020 0.072 0.164 0.056
#> GSM1269711 2 0.705 -0.19225 0.408 0.420 0.048 0.040 0.060 0.024
#> GSM1269646 3 0.851 0.15018 0.060 0.184 0.404 0.024 0.140 0.188
#> GSM1269654 4 0.908 0.15559 0.084 0.072 0.144 0.384 0.156 0.160
#> GSM1269662 5 0.841 -0.08951 0.064 0.068 0.052 0.128 0.380 0.308
#> GSM1269670 2 0.796 0.05809 0.036 0.428 0.292 0.032 0.092 0.120
#> GSM1269678 4 0.810 0.05446 0.096 0.024 0.340 0.356 0.148 0.036
#> GSM1269692 4 0.729 0.20405 0.124 0.020 0.004 0.504 0.208 0.140
#> GSM1269700 3 0.871 0.09641 0.120 0.144 0.372 0.204 0.148 0.012
#> GSM1269708 3 0.830 0.05508 0.128 0.080 0.368 0.312 0.096 0.016
#> GSM1269714 4 0.532 0.39682 0.092 0.016 0.056 0.740 0.068 0.028
#> GSM1269716 4 0.483 0.42465 0.052 0.020 0.044 0.780 0.068 0.036
#> GSM1269720 6 0.718 0.29916 0.020 0.072 0.024 0.120 0.232 0.532
#> GSM1269722 4 0.852 0.12638 0.044 0.140 0.240 0.364 0.188 0.024
#> GSM1269644 1 0.848 0.26902 0.416 0.100 0.012 0.148 0.152 0.172
#> GSM1269652 6 0.944 -0.17709 0.224 0.224 0.136 0.056 0.108 0.252
#> GSM1269660 5 0.980 0.00848 0.152 0.140 0.156 0.136 0.272 0.144
#> GSM1269668 1 0.855 0.28552 0.380 0.108 0.180 0.156 0.172 0.004
#> GSM1269676 6 0.337 0.40663 0.052 0.024 0.016 0.020 0.024 0.864
#> GSM1269684 1 0.720 0.41979 0.556 0.076 0.032 0.208 0.100 0.028
#> GSM1269690 1 0.784 0.38079 0.504 0.072 0.028 0.208 0.084 0.104
#> GSM1269698 2 0.772 0.22055 0.080 0.416 0.108 0.024 0.036 0.336
#> GSM1269706 2 0.830 0.19921 0.120 0.408 0.064 0.032 0.096 0.280
#> GSM1269650 6 0.620 0.31700 0.024 0.040 0.048 0.064 0.164 0.660
#> GSM1269658 6 0.602 0.28441 0.004 0.028 0.000 0.160 0.240 0.568
#> GSM1269666 4 0.853 -0.02391 0.068 0.064 0.308 0.324 0.200 0.036
#> GSM1269674 2 0.899 0.03304 0.040 0.340 0.240 0.084 0.136 0.160
#> GSM1269682 4 0.778 0.34365 0.092 0.024 0.100 0.512 0.200 0.072
#> GSM1269688 2 0.836 -0.13728 0.096 0.364 0.324 0.064 0.120 0.032
#> GSM1269696 3 0.569 0.32849 0.016 0.152 0.688 0.024 0.092 0.028
#> GSM1269704 3 0.856 0.13121 0.036 0.228 0.392 0.052 0.176 0.116
#> GSM1269712 3 0.751 0.21262 0.064 0.088 0.508 0.244 0.080 0.016
#> GSM1269718 6 0.981 -0.22153 0.128 0.116 0.128 0.196 0.212 0.220
#> GSM1269724 3 0.680 0.36758 0.044 0.060 0.612 0.124 0.140 0.020
#> GSM1269726 4 0.913 0.15374 0.136 0.244 0.068 0.296 0.204 0.052
#> GSM1269648 1 0.796 0.13557 0.372 0.352 0.028 0.032 0.104 0.112
#> GSM1269656 6 0.716 0.25404 0.164 0.060 0.052 0.048 0.084 0.592
#> GSM1269664 1 0.821 0.30728 0.452 0.076 0.084 0.096 0.248 0.044
#> GSM1269672 1 0.695 0.44776 0.592 0.084 0.028 0.164 0.108 0.024
#> GSM1269680 6 0.376 0.39536 0.040 0.040 0.028 0.012 0.036 0.844
#> GSM1269686 1 0.774 0.40526 0.548 0.116 0.124 0.100 0.072 0.040
#> GSM1269694 1 0.692 0.19550 0.460 0.372 0.036 0.024 0.076 0.032
#> GSM1269702 1 0.679 0.41170 0.624 0.088 0.020 0.064 0.068 0.136
#> GSM1269710 2 0.698 -0.18226 0.388 0.448 0.040 0.040 0.044 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 agent(p) disease.state(p) gender(p) individual(p) k
#> SD:skmeans 10 NA NA NA NA 2
#> SD:skmeans 47 0.873 0.563 3.72e-09 8.15e-05 3
#> SD:skmeans 43 0.839 0.464 1.28e-06 5.75e-06 4
#> SD:skmeans 9 NA NA NA NA 5
#> SD:skmeans 0 NA NA NA 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["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 51941 rows and 84 columns.
#> Top rows (1000, 2000, 3000, 4000, 5000) are extracted by 'SD' method.
#> Subgroups are detected by 'pam' method.
#> Performed in total 1250 partitions by row resampling.
#> Best k for subgroups seems to be 3.
#>
#> Following methods can be applied to this 'ConsensusPartition' object:
#> [1] "cola_report" "collect_classes" "collect_plots"
#> [4] "collect_stats" "colnames" "compare_signatures"
#> [7] "consensus_heatmap" "dimension_reduction" "functional_enrichment"
#> [10] "get_anno_col" "get_anno" "get_classes"
#> [13] "get_consensus" "get_matrix" "get_membership"
#> [16] "get_param" "get_signatures" "get_stats"
#> [19] "is_best_k" "is_stable_k" "membership_heatmap"
#> [22] "ncol" "nrow" "plot_ecdf"
#> [25] "rownames" "select_partition_number" "show"
#> [28] "suggest_best_k" "test_to_known_factors"
collect_plots() function collects all the plots made from res for all k (number of partitions)
into one single page to provide an easy and fast comparison between different k.
collect_plots(res)
The plots are:
k and the heatmap of
predicted classes for each k.k.k.k.All the plots in panels can be made by individual functions and they are plotted later in this section.
select_partition_number() produces several plots showing different
statistics for choosing “optimized” k. There are following statistics:
k;k, the area increased is defined as \(A_k - A_{k-1}\).The detailed explanations of these statistics can be found in the cola vignette.
Generally speaking, lower PAC score, higher mean silhouette score or higher
concordance corresponds to better partition. Rand index and Jaccard index
measure how similar the current partition is compared to partition with k-1.
If they are too similar, we won't accept k is better than k-1.
select_partition_number(res)
The numeric values for all these statistics can be obtained by get_stats().
get_stats(res)
#> k 1-PAC mean_silhouette concordance area_increased Rand Jaccard
#> 2 2 0.0539 0.239 0.599 0.4794 0.512 0.512
#> 3 3 0.1675 0.449 0.707 0.3495 0.639 0.406
#> 4 4 0.2136 0.386 0.638 0.0954 0.895 0.714
#> 5 5 0.2882 0.383 0.607 0.0482 0.910 0.719
#> 6 6 0.3444 0.387 0.618 0.0303 0.977 0.914
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
#> GSM1269647 2 0.4161 0.43028 0.084 0.916
#> GSM1269655 1 0.9998 -0.28865 0.508 0.492
#> GSM1269663 1 0.9775 -0.17234 0.588 0.412
#> GSM1269671 2 0.8386 0.21106 0.268 0.732
#> GSM1269679 2 0.9815 0.28779 0.420 0.580
#> GSM1269693 1 0.9635 -0.19214 0.612 0.388
#> GSM1269701 1 0.9977 -0.26059 0.528 0.472
#> GSM1269709 2 0.9552 -0.13414 0.376 0.624
#> GSM1269715 2 0.9954 0.30853 0.460 0.540
#> GSM1269717 2 0.9996 -0.00829 0.488 0.512
#> GSM1269721 1 0.8763 -0.10014 0.704 0.296
#> GSM1269723 2 0.9754 0.33136 0.408 0.592
#> GSM1269645 1 0.9983 0.33759 0.524 0.476
#> GSM1269653 1 0.4431 0.25778 0.908 0.092
#> GSM1269661 1 0.9977 0.24535 0.528 0.472
#> GSM1269669 1 0.6531 0.14266 0.832 0.168
#> GSM1269677 1 0.9323 0.38403 0.652 0.348
#> GSM1269685 1 0.9833 0.36759 0.576 0.424
#> GSM1269691 1 0.0672 0.31679 0.992 0.008
#> GSM1269699 1 0.9881 0.36333 0.564 0.436
#> GSM1269707 1 0.7453 0.35850 0.788 0.212
#> GSM1269651 2 0.9393 0.04448 0.356 0.644
#> GSM1269659 1 0.8081 0.34238 0.752 0.248
#> GSM1269667 2 0.8443 0.43980 0.272 0.728
#> GSM1269675 2 0.7139 0.39858 0.196 0.804
#> GSM1269683 2 0.9580 0.38842 0.380 0.620
#> GSM1269689 2 0.9552 0.33888 0.376 0.624
#> GSM1269697 2 0.7745 0.44576 0.228 0.772
#> GSM1269705 2 0.9170 0.30585 0.332 0.668
#> GSM1269713 2 0.9491 0.32489 0.368 0.632
#> GSM1269719 2 0.9881 -0.22440 0.436 0.564
#> GSM1269725 2 0.7376 0.44978 0.208 0.792
#> GSM1269727 2 0.9996 0.24753 0.488 0.512
#> GSM1269649 1 0.6801 0.19370 0.820 0.180
#> GSM1269657 1 0.9988 0.34257 0.520 0.480
#> GSM1269665 1 0.9909 0.35886 0.556 0.444
#> GSM1269673 1 0.5946 0.37078 0.856 0.144
#> GSM1269681 1 0.9988 0.33957 0.520 0.480
#> GSM1269687 1 0.9393 0.37577 0.644 0.356
#> GSM1269695 1 0.9988 0.34383 0.520 0.480
#> GSM1269703 1 0.9988 0.33810 0.520 0.480
#> GSM1269711 1 0.9909 0.30459 0.556 0.444
#> GSM1269646 2 0.6438 0.45782 0.164 0.836
#> GSM1269654 2 0.8955 0.17318 0.312 0.688
#> GSM1269662 1 0.9933 0.13250 0.548 0.452
#> GSM1269670 2 0.7950 0.25788 0.240 0.760
#> GSM1269678 2 0.5519 0.45354 0.128 0.872
#> GSM1269692 1 0.5059 0.22046 0.888 0.112
#> GSM1269700 2 0.8713 0.43148 0.292 0.708
#> GSM1269708 2 0.9608 -0.13103 0.384 0.616
#> GSM1269714 2 0.9552 0.37984 0.376 0.624
#> GSM1269716 2 0.9944 0.22040 0.456 0.544
#> GSM1269720 1 0.9732 -0.18793 0.596 0.404
#> GSM1269722 2 0.9988 0.24680 0.480 0.520
#> GSM1269644 1 0.2423 0.33285 0.960 0.040
#> GSM1269652 1 0.9491 0.26812 0.632 0.368
#> GSM1269660 1 0.9909 0.16918 0.556 0.444
#> GSM1269668 1 0.6531 0.21644 0.832 0.168
#> GSM1269676 1 0.6801 0.37942 0.820 0.180
#> GSM1269684 1 0.9866 0.36350 0.568 0.432
#> GSM1269690 1 0.0376 0.31897 0.996 0.004
#> GSM1269698 1 0.9922 0.34902 0.552 0.448
#> GSM1269706 1 0.9686 0.31554 0.604 0.396
#> GSM1269650 1 0.9393 0.28210 0.644 0.356
#> GSM1269658 1 0.5946 0.20449 0.856 0.144
#> GSM1269666 2 0.9661 0.35729 0.392 0.608
#> GSM1269674 1 0.9833 -0.24025 0.576 0.424
#> GSM1269682 2 0.9977 -0.27980 0.472 0.528
#> GSM1269688 1 0.9944 -0.25190 0.544 0.456
#> GSM1269696 2 0.2603 0.45623 0.044 0.956
#> GSM1269704 2 0.6801 0.42403 0.180 0.820
#> GSM1269712 2 0.6623 0.38250 0.172 0.828
#> GSM1269718 2 0.7299 0.36272 0.204 0.796
#> GSM1269724 2 0.8861 0.41046 0.304 0.696
#> GSM1269726 1 0.9944 -0.24589 0.544 0.456
#> GSM1269648 1 0.9977 0.34561 0.528 0.472
#> GSM1269656 1 0.9775 0.37345 0.588 0.412
#> GSM1269664 1 0.7299 0.36006 0.796 0.204
#> GSM1269672 1 0.4161 0.34353 0.916 0.084
#> GSM1269680 1 0.9996 0.33357 0.512 0.488
#> GSM1269686 1 0.9983 0.34558 0.524 0.476
#> GSM1269694 1 0.9970 0.35073 0.532 0.468
#> GSM1269702 1 0.9977 0.34626 0.528 0.472
#> GSM1269710 1 0.8763 0.36183 0.704 0.296
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM1269647 2 0.6661 0.41121 0.400 0.588 0.012
#> GSM1269655 3 0.7979 0.37970 0.100 0.272 0.628
#> GSM1269663 3 0.9217 0.34522 0.260 0.208 0.532
#> GSM1269671 1 0.4968 0.60770 0.800 0.188 0.012
#> GSM1269679 2 0.5443 0.50425 0.004 0.736 0.260
#> GSM1269693 3 0.3686 0.55358 0.000 0.140 0.860
#> GSM1269701 3 0.6888 0.07267 0.016 0.432 0.552
#> GSM1269709 1 0.7571 0.13095 0.508 0.452 0.040
#> GSM1269715 3 0.9387 0.28234 0.220 0.272 0.508
#> GSM1269717 3 0.9370 -0.10537 0.416 0.168 0.416
#> GSM1269721 3 0.5598 0.54341 0.148 0.052 0.800
#> GSM1269723 2 0.7383 0.50451 0.084 0.680 0.236
#> GSM1269645 1 0.4399 0.65407 0.812 0.000 0.188
#> GSM1269653 3 0.5117 0.60588 0.108 0.060 0.832
#> GSM1269661 1 0.8925 0.44464 0.564 0.180 0.256
#> GSM1269669 3 0.2651 0.59002 0.012 0.060 0.928
#> GSM1269677 1 0.4654 0.55161 0.792 0.000 0.208
#> GSM1269685 1 0.5138 0.62793 0.748 0.000 0.252
#> GSM1269691 3 0.1529 0.60570 0.040 0.000 0.960
#> GSM1269699 1 0.2301 0.67032 0.936 0.004 0.060
#> GSM1269707 1 0.5968 0.18945 0.636 0.000 0.364
#> GSM1269651 1 0.5505 0.63947 0.816 0.088 0.096
#> GSM1269659 3 0.7767 0.23682 0.412 0.052 0.536
#> GSM1269667 2 0.9292 0.37039 0.284 0.516 0.200
#> GSM1269675 2 0.7742 0.39818 0.356 0.584 0.060
#> GSM1269683 2 0.9223 0.27279 0.172 0.504 0.324
#> GSM1269689 2 0.4861 0.60069 0.012 0.808 0.180
#> GSM1269697 2 0.3295 0.64097 0.096 0.896 0.008
#> GSM1269705 1 0.8749 0.34293 0.572 0.276 0.152
#> GSM1269713 2 0.5042 0.61670 0.104 0.836 0.060
#> GSM1269719 1 0.4056 0.66891 0.876 0.092 0.032
#> GSM1269725 2 0.2187 0.64214 0.028 0.948 0.024
#> GSM1269727 2 0.6490 0.37893 0.012 0.628 0.360
#> GSM1269649 3 0.7330 0.57895 0.216 0.092 0.692
#> GSM1269657 1 0.0000 0.66129 1.000 0.000 0.000
#> GSM1269665 1 0.5202 0.64635 0.772 0.008 0.220
#> GSM1269673 3 0.5404 0.46966 0.256 0.004 0.740
#> GSM1269681 1 0.0424 0.66317 0.992 0.000 0.008
#> GSM1269687 1 0.6379 0.47753 0.624 0.008 0.368
#> GSM1269695 1 0.4062 0.66078 0.836 0.000 0.164
#> GSM1269703 1 0.2711 0.67376 0.912 0.000 0.088
#> GSM1269711 1 0.8105 0.44841 0.580 0.084 0.336
#> GSM1269646 2 0.2229 0.63602 0.044 0.944 0.012
#> GSM1269654 1 0.6633 0.57120 0.728 0.212 0.060
#> GSM1269662 1 0.9829 -0.11137 0.400 0.248 0.352
#> GSM1269670 1 0.6252 0.51840 0.708 0.268 0.024
#> GSM1269678 2 0.3481 0.64341 0.044 0.904 0.052
#> GSM1269692 3 0.3148 0.60874 0.048 0.036 0.916
#> GSM1269700 2 0.8459 0.47612 0.232 0.612 0.156
#> GSM1269708 1 0.8342 0.07191 0.464 0.456 0.080
#> GSM1269714 2 0.8309 0.46261 0.188 0.632 0.180
#> GSM1269716 3 0.9547 0.17284 0.320 0.212 0.468
#> GSM1269720 3 0.9133 0.35338 0.296 0.176 0.528
#> GSM1269722 2 0.5618 0.52554 0.008 0.732 0.260
#> GSM1269644 3 0.3482 0.60902 0.128 0.000 0.872
#> GSM1269652 1 0.9783 0.00242 0.436 0.264 0.300
#> GSM1269660 1 0.8304 0.22609 0.504 0.080 0.416
#> GSM1269668 3 0.5307 0.61028 0.124 0.056 0.820
#> GSM1269676 3 0.6192 0.35262 0.420 0.000 0.580
#> GSM1269684 1 0.5254 0.62819 0.736 0.000 0.264
#> GSM1269690 3 0.1753 0.60841 0.048 0.000 0.952
#> GSM1269698 1 0.3752 0.64976 0.884 0.020 0.096
#> GSM1269706 1 0.8230 0.17903 0.564 0.088 0.348
#> GSM1269650 1 0.7095 0.37115 0.660 0.048 0.292
#> GSM1269658 3 0.5263 0.60849 0.116 0.060 0.824
#> GSM1269666 2 0.6970 0.51290 0.048 0.676 0.276
#> GSM1269674 3 0.7841 0.11762 0.056 0.408 0.536
#> GSM1269682 1 0.7726 0.59428 0.676 0.192 0.132
#> GSM1269688 3 0.7466 0.03049 0.036 0.444 0.520
#> GSM1269696 2 0.2096 0.64561 0.052 0.944 0.004
#> GSM1269704 2 0.7392 0.12189 0.468 0.500 0.032
#> GSM1269712 2 0.7138 0.14362 0.436 0.540 0.024
#> GSM1269718 1 0.7325 0.22539 0.576 0.388 0.036
#> GSM1269724 2 0.6677 0.56967 0.088 0.744 0.168
#> GSM1269726 3 0.7698 0.33383 0.072 0.304 0.624
#> GSM1269648 1 0.4002 0.66858 0.840 0.000 0.160
#> GSM1269656 1 0.3619 0.62382 0.864 0.000 0.136
#> GSM1269664 3 0.6521 -0.09641 0.492 0.004 0.504
#> GSM1269672 3 0.4750 0.54513 0.216 0.000 0.784
#> GSM1269680 1 0.0000 0.66129 1.000 0.000 0.000
#> GSM1269686 1 0.4110 0.66183 0.844 0.004 0.152
#> GSM1269694 1 0.3686 0.66639 0.860 0.000 0.140
#> GSM1269702 1 0.3412 0.67002 0.876 0.000 0.124
#> GSM1269710 3 0.6779 0.02768 0.444 0.012 0.544
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM1269647 2 0.625 0.25001 0.276 0.652 0.052 0.020
#> GSM1269655 4 0.835 0.24868 0.076 0.256 0.140 0.528
#> GSM1269663 4 0.822 0.34733 0.244 0.204 0.040 0.512
#> GSM1269671 1 0.511 0.58943 0.772 0.164 0.048 0.016
#> GSM1269679 3 0.794 -0.17230 0.004 0.376 0.380 0.240
#> GSM1269693 4 0.448 0.53212 0.000 0.052 0.152 0.796
#> GSM1269701 2 0.605 0.19003 0.004 0.540 0.036 0.420
#> GSM1269709 3 0.554 0.46022 0.272 0.020 0.688 0.020
#> GSM1269715 4 0.924 0.31965 0.216 0.136 0.196 0.452
#> GSM1269717 1 0.887 0.01977 0.384 0.084 0.152 0.380
#> GSM1269721 4 0.603 0.51897 0.120 0.040 0.100 0.740
#> GSM1269723 2 0.444 0.47510 0.040 0.836 0.040 0.084
#> GSM1269645 1 0.405 0.64361 0.804 0.008 0.008 0.180
#> GSM1269653 4 0.435 0.59242 0.092 0.004 0.080 0.824
#> GSM1269661 1 0.759 0.45286 0.564 0.184 0.020 0.232
#> GSM1269669 4 0.433 0.52835 0.008 0.148 0.032 0.812
#> GSM1269677 1 0.617 0.47592 0.700 0.032 0.060 0.208
#> GSM1269685 1 0.442 0.61441 0.736 0.000 0.008 0.256
#> GSM1269691 4 0.102 0.59350 0.032 0.000 0.000 0.968
#> GSM1269699 1 0.238 0.65658 0.916 0.008 0.004 0.072
#> GSM1269707 1 0.530 0.12770 0.600 0.004 0.008 0.388
#> GSM1269651 1 0.836 0.37659 0.548 0.200 0.168 0.084
#> GSM1269659 4 0.773 0.26365 0.364 0.060 0.072 0.504
#> GSM1269667 2 0.819 0.28915 0.212 0.568 0.100 0.120
#> GSM1269675 3 0.839 0.23611 0.292 0.288 0.400 0.020
#> GSM1269683 2 0.965 0.06428 0.144 0.360 0.264 0.232
#> GSM1269689 2 0.650 0.32665 0.008 0.636 0.260 0.096
#> GSM1269697 3 0.575 0.31902 0.052 0.272 0.672 0.004
#> GSM1269705 1 0.888 0.18904 0.504 0.152 0.204 0.140
#> GSM1269713 3 0.756 0.19695 0.080 0.316 0.552 0.052
#> GSM1269719 1 0.377 0.66204 0.872 0.048 0.040 0.040
#> GSM1269725 2 0.573 0.16107 0.016 0.560 0.416 0.008
#> GSM1269727 2 0.691 0.40937 0.000 0.588 0.172 0.240
#> GSM1269649 4 0.614 0.56392 0.168 0.108 0.016 0.708
#> GSM1269657 1 0.123 0.64184 0.968 0.008 0.004 0.020
#> GSM1269665 1 0.434 0.63099 0.764 0.008 0.004 0.224
#> GSM1269673 4 0.448 0.46693 0.248 0.000 0.012 0.740
#> GSM1269681 1 0.380 0.62451 0.868 0.048 0.060 0.024
#> GSM1269687 1 0.577 0.47051 0.612 0.016 0.016 0.356
#> GSM1269695 1 0.406 0.65262 0.828 0.020 0.012 0.140
#> GSM1269703 1 0.299 0.65977 0.884 0.012 0.004 0.100
#> GSM1269711 1 0.749 0.42989 0.560 0.088 0.044 0.308
#> GSM1269646 2 0.609 0.15111 0.032 0.544 0.416 0.008
#> GSM1269654 1 0.712 0.51665 0.652 0.168 0.136 0.044
#> GSM1269662 1 0.839 -0.11358 0.328 0.328 0.016 0.328
#> GSM1269670 1 0.651 0.50260 0.684 0.160 0.136 0.020
#> GSM1269678 3 0.643 0.13994 0.032 0.356 0.584 0.028
#> GSM1269692 4 0.356 0.58717 0.032 0.020 0.072 0.876
#> GSM1269700 2 0.569 0.43830 0.124 0.760 0.036 0.080
#> GSM1269708 3 0.564 0.46043 0.260 0.016 0.692 0.032
#> GSM1269714 3 0.628 0.39807 0.076 0.116 0.732 0.076
#> GSM1269716 4 0.932 0.20946 0.280 0.168 0.136 0.416
#> GSM1269720 4 0.890 0.35268 0.244 0.112 0.160 0.484
#> GSM1269722 2 0.575 0.45400 0.008 0.732 0.128 0.132
#> GSM1269644 4 0.293 0.60320 0.108 0.000 0.012 0.880
#> GSM1269652 3 0.776 0.26499 0.316 0.000 0.428 0.256
#> GSM1269660 1 0.733 0.19509 0.476 0.100 0.016 0.408
#> GSM1269668 4 0.629 0.52408 0.104 0.168 0.024 0.704
#> GSM1269676 4 0.679 0.35092 0.364 0.020 0.060 0.556
#> GSM1269684 1 0.477 0.62029 0.736 0.008 0.012 0.244
#> GSM1269690 4 0.130 0.59727 0.044 0.000 0.000 0.956
#> GSM1269698 1 0.316 0.64283 0.880 0.004 0.020 0.096
#> GSM1269706 1 0.707 0.15375 0.524 0.008 0.104 0.364
#> GSM1269650 1 0.919 0.12908 0.460 0.148 0.164 0.228
#> GSM1269658 4 0.553 0.57666 0.076 0.084 0.060 0.780
#> GSM1269666 2 0.758 0.35761 0.020 0.560 0.248 0.172
#> GSM1269674 4 0.834 -0.04396 0.044 0.328 0.164 0.464
#> GSM1269682 1 0.742 0.51239 0.636 0.188 0.096 0.080
#> GSM1269688 2 0.768 0.15149 0.028 0.444 0.108 0.420
#> GSM1269696 2 0.569 -0.03142 0.024 0.508 0.468 0.000
#> GSM1269704 3 0.834 0.30539 0.324 0.212 0.436 0.028
#> GSM1269712 3 0.803 0.26272 0.396 0.188 0.400 0.016
#> GSM1269718 1 0.788 0.20449 0.544 0.248 0.176 0.032
#> GSM1269724 3 0.657 0.27928 0.032 0.232 0.664 0.072
#> GSM1269726 4 0.777 0.27148 0.068 0.272 0.092 0.568
#> GSM1269648 1 0.371 0.65854 0.832 0.004 0.012 0.152
#> GSM1269656 1 0.368 0.60332 0.828 0.008 0.004 0.160
#> GSM1269664 4 0.540 -0.06302 0.476 0.012 0.000 0.512
#> GSM1269672 4 0.391 0.54443 0.212 0.000 0.004 0.784
#> GSM1269680 1 0.233 0.63914 0.932 0.024 0.024 0.020
#> GSM1269686 1 0.416 0.65079 0.824 0.024 0.012 0.140
#> GSM1269694 1 0.385 0.65436 0.844 0.020 0.012 0.124
#> GSM1269702 1 0.294 0.65526 0.868 0.000 0.004 0.128
#> GSM1269710 4 0.697 -0.00494 0.428 0.052 0.028 0.492
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM1269647 3 0.659 0.26855 0.176 0.148 0.620 0.004 0.052
#> GSM1269655 4 0.851 0.16392 0.064 0.108 0.196 0.484 0.148
#> GSM1269663 4 0.801 0.33651 0.248 0.060 0.164 0.488 0.040
#> GSM1269671 1 0.544 0.54969 0.720 0.044 0.164 0.004 0.068
#> GSM1269679 3 0.681 0.07900 0.004 0.000 0.384 0.232 0.380
#> GSM1269693 4 0.342 0.52048 0.000 0.004 0.036 0.836 0.124
#> GSM1269701 3 0.545 0.18630 0.008 0.008 0.564 0.388 0.032
#> GSM1269709 5 0.452 0.50040 0.236 0.000 0.004 0.040 0.720
#> GSM1269715 4 0.783 0.33535 0.212 0.016 0.112 0.516 0.144
#> GSM1269717 4 0.803 -0.00598 0.376 0.036 0.084 0.404 0.100
#> GSM1269721 4 0.609 0.46495 0.036 0.224 0.016 0.656 0.068
#> GSM1269723 3 0.335 0.44309 0.028 0.008 0.872 0.056 0.036
#> GSM1269645 1 0.273 0.65458 0.872 0.004 0.000 0.112 0.012
#> GSM1269653 4 0.487 0.56286 0.080 0.044 0.004 0.776 0.096
#> GSM1269661 1 0.665 0.45994 0.588 0.008 0.176 0.204 0.024
#> GSM1269669 4 0.441 0.48166 0.016 0.004 0.180 0.768 0.032
#> GSM1269677 1 0.759 0.02044 0.428 0.372 0.020 0.132 0.048
#> GSM1269685 1 0.289 0.63988 0.824 0.000 0.000 0.176 0.000
#> GSM1269691 4 0.173 0.56851 0.080 0.000 0.000 0.920 0.000
#> GSM1269699 1 0.385 0.62877 0.836 0.084 0.004 0.056 0.020
#> GSM1269707 1 0.619 0.15875 0.528 0.096 0.000 0.360 0.016
#> GSM1269651 2 0.703 0.78397 0.184 0.620 0.064 0.040 0.092
#> GSM1269659 4 0.787 0.22055 0.324 0.212 0.020 0.404 0.040
#> GSM1269667 3 0.746 0.32437 0.188 0.012 0.560 0.120 0.120
#> GSM1269675 5 0.771 0.21313 0.272 0.028 0.276 0.016 0.408
#> GSM1269683 3 0.875 0.07080 0.136 0.020 0.348 0.236 0.260
#> GSM1269689 3 0.551 0.28176 0.004 0.032 0.680 0.052 0.232
#> GSM1269697 5 0.406 0.43590 0.024 0.012 0.188 0.000 0.776
#> GSM1269705 1 0.892 0.06893 0.436 0.100 0.124 0.116 0.224
#> GSM1269713 5 0.630 0.33318 0.020 0.084 0.260 0.020 0.616
#> GSM1269719 1 0.344 0.64252 0.872 0.036 0.040 0.016 0.036
#> GSM1269725 3 0.531 0.06268 0.012 0.020 0.520 0.004 0.444
#> GSM1269727 3 0.619 0.40053 0.000 0.016 0.604 0.220 0.160
#> GSM1269649 4 0.628 0.52087 0.220 0.036 0.092 0.640 0.012
#> GSM1269657 1 0.298 0.59488 0.856 0.128 0.004 0.004 0.008
#> GSM1269665 1 0.289 0.64925 0.844 0.000 0.008 0.148 0.000
#> GSM1269673 4 0.407 0.41807 0.324 0.000 0.000 0.672 0.004
#> GSM1269681 1 0.601 0.35844 0.608 0.296 0.040 0.004 0.052
#> GSM1269687 1 0.427 0.52112 0.696 0.000 0.008 0.288 0.008
#> GSM1269695 1 0.251 0.65576 0.900 0.012 0.008 0.076 0.004
#> GSM1269703 1 0.106 0.64985 0.968 0.008 0.004 0.020 0.000
#> GSM1269711 1 0.673 0.46180 0.576 0.016 0.096 0.276 0.036
#> GSM1269646 3 0.564 0.04758 0.016 0.044 0.520 0.000 0.420
#> GSM1269654 1 0.740 0.37381 0.600 0.120 0.104 0.036 0.140
#> GSM1269662 3 0.860 -0.05073 0.292 0.124 0.320 0.252 0.012
#> GSM1269670 1 0.615 0.49460 0.668 0.032 0.136 0.012 0.152
#> GSM1269678 5 0.591 0.24692 0.028 0.004 0.324 0.052 0.592
#> GSM1269692 4 0.311 0.55723 0.032 0.012 0.016 0.884 0.056
#> GSM1269700 3 0.444 0.43294 0.104 0.004 0.800 0.060 0.032
#> GSM1269708 5 0.471 0.50028 0.216 0.000 0.004 0.060 0.720
#> GSM1269714 5 0.540 0.46014 0.060 0.004 0.092 0.104 0.740
#> GSM1269716 4 0.819 0.18251 0.280 0.044 0.136 0.464 0.076
#> GSM1269720 4 0.859 0.30415 0.184 0.192 0.040 0.452 0.132
#> GSM1269722 3 0.520 0.41607 0.012 0.008 0.732 0.132 0.116
#> GSM1269644 4 0.304 0.57316 0.148 0.004 0.000 0.840 0.008
#> GSM1269652 5 0.778 0.23920 0.192 0.108 0.000 0.232 0.468
#> GSM1269660 1 0.666 0.23268 0.464 0.016 0.100 0.408 0.012
#> GSM1269668 4 0.601 0.45906 0.108 0.004 0.208 0.652 0.028
#> GSM1269676 4 0.789 0.03900 0.184 0.348 0.020 0.400 0.048
#> GSM1269684 1 0.349 0.63836 0.784 0.000 0.004 0.208 0.004
#> GSM1269690 4 0.201 0.57324 0.088 0.000 0.000 0.908 0.004
#> GSM1269698 1 0.427 0.61970 0.812 0.064 0.004 0.092 0.028
#> GSM1269706 1 0.793 0.06537 0.404 0.116 0.008 0.352 0.120
#> GSM1269650 2 0.660 0.78372 0.108 0.680 0.056 0.076 0.080
#> GSM1269658 4 0.640 0.46651 0.044 0.224 0.052 0.644 0.036
#> GSM1269666 3 0.777 0.32528 0.024 0.080 0.524 0.156 0.216
#> GSM1269674 4 0.730 -0.01038 0.044 0.004 0.316 0.472 0.164
#> GSM1269682 1 0.612 0.48488 0.668 0.004 0.180 0.080 0.068
#> GSM1269688 3 0.724 0.16244 0.024 0.032 0.456 0.380 0.108
#> GSM1269696 5 0.516 0.10360 0.024 0.008 0.476 0.000 0.492
#> GSM1269704 5 0.796 0.31383 0.248 0.076 0.180 0.020 0.476
#> GSM1269712 5 0.694 0.25434 0.368 0.004 0.168 0.016 0.444
#> GSM1269718 1 0.751 0.22360 0.516 0.032 0.228 0.028 0.196
#> GSM1269724 5 0.500 0.35951 0.012 0.004 0.212 0.056 0.716
#> GSM1269726 4 0.607 0.28988 0.064 0.004 0.272 0.620 0.040
#> GSM1269648 1 0.245 0.65961 0.896 0.016 0.000 0.084 0.004
#> GSM1269656 1 0.523 0.55414 0.716 0.128 0.004 0.144 0.008
#> GSM1269664 1 0.470 0.10127 0.516 0.004 0.008 0.472 0.000
#> GSM1269672 4 0.379 0.52171 0.248 0.000 0.004 0.744 0.004
#> GSM1269680 1 0.430 0.54362 0.748 0.216 0.004 0.004 0.028
#> GSM1269686 1 0.249 0.65498 0.900 0.000 0.020 0.072 0.008
#> GSM1269694 1 0.189 0.65214 0.936 0.012 0.008 0.040 0.004
#> GSM1269702 1 0.157 0.65152 0.936 0.004 0.000 0.060 0.000
#> GSM1269710 1 0.578 0.01451 0.480 0.008 0.040 0.460 0.012
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM1269647 3 0.6631 0.28562 0.116 0.064 0.584 0.000 0.040 0.196
#> GSM1269655 4 0.7457 0.19044 0.052 0.248 0.140 0.476 0.084 0.000
#> GSM1269663 4 0.7332 0.36656 0.236 0.096 0.152 0.488 0.024 0.004
#> GSM1269671 1 0.6415 0.48062 0.624 0.044 0.176 0.004 0.064 0.088
#> GSM1269679 5 0.6317 -0.06079 0.004 0.004 0.376 0.216 0.396 0.004
#> GSM1269693 4 0.3333 0.53652 0.000 0.012 0.016 0.832 0.124 0.016
#> GSM1269701 3 0.5670 0.17442 0.020 0.012 0.556 0.348 0.060 0.004
#> GSM1269709 5 0.3698 0.48845 0.212 0.000 0.004 0.028 0.756 0.000
#> GSM1269715 4 0.7704 0.31919 0.220 0.032 0.088 0.472 0.172 0.016
#> GSM1269717 4 0.7886 -0.00458 0.360 0.084 0.072 0.380 0.088 0.016
#> GSM1269721 4 0.6584 0.41027 0.016 0.184 0.008 0.588 0.064 0.140
#> GSM1269723 3 0.3518 0.43102 0.020 0.020 0.852 0.056 0.044 0.008
#> GSM1269645 1 0.2666 0.65119 0.864 0.000 0.000 0.112 0.012 0.012
#> GSM1269653 4 0.4284 0.54694 0.072 0.000 0.004 0.784 0.092 0.048
#> GSM1269661 1 0.5845 0.46897 0.600 0.000 0.168 0.204 0.020 0.008
#> GSM1269669 4 0.4744 0.44966 0.020 0.004 0.204 0.712 0.056 0.004
#> GSM1269677 6 0.4310 0.65847 0.204 0.004 0.000 0.072 0.000 0.720
#> GSM1269685 1 0.2527 0.63411 0.832 0.000 0.000 0.168 0.000 0.000
#> GSM1269691 4 0.1387 0.55899 0.068 0.000 0.000 0.932 0.000 0.000
#> GSM1269699 1 0.4341 0.60211 0.780 0.004 0.008 0.060 0.028 0.120
#> GSM1269707 1 0.6361 0.15678 0.480 0.012 0.004 0.344 0.016 0.144
#> GSM1269651 2 0.3193 0.86636 0.076 0.860 0.004 0.008 0.016 0.036
#> GSM1269659 4 0.7877 0.27408 0.272 0.216 0.012 0.396 0.044 0.060
#> GSM1269667 3 0.7107 0.29685 0.192 0.028 0.540 0.100 0.136 0.004
#> GSM1269675 5 0.7649 0.19240 0.264 0.052 0.236 0.008 0.404 0.036
#> GSM1269683 3 0.8264 -0.01012 0.140 0.040 0.320 0.228 0.268 0.004
#> GSM1269689 3 0.5880 0.26177 0.000 0.036 0.648 0.028 0.172 0.116
#> GSM1269697 5 0.3683 0.43748 0.016 0.028 0.172 0.000 0.784 0.000
#> GSM1269705 1 0.8748 0.20268 0.424 0.168 0.096 0.104 0.156 0.052
#> GSM1269713 5 0.6441 0.31806 0.016 0.028 0.236 0.016 0.576 0.128
#> GSM1269719 1 0.3587 0.64012 0.856 0.028 0.036 0.024 0.028 0.028
#> GSM1269725 3 0.5656 0.03254 0.012 0.092 0.472 0.000 0.420 0.004
#> GSM1269727 3 0.6069 0.36747 0.000 0.028 0.588 0.212 0.160 0.012
#> GSM1269649 4 0.6234 0.47734 0.220 0.000 0.088 0.604 0.024 0.064
#> GSM1269657 1 0.2730 0.56382 0.808 0.000 0.000 0.000 0.000 0.192
#> GSM1269665 1 0.2833 0.64348 0.836 0.004 0.012 0.148 0.000 0.000
#> GSM1269673 4 0.3499 0.41559 0.320 0.000 0.000 0.680 0.000 0.000
#> GSM1269681 1 0.5655 -0.00560 0.476 0.120 0.008 0.000 0.000 0.396
#> GSM1269687 1 0.3895 0.52991 0.700 0.000 0.008 0.280 0.012 0.000
#> GSM1269695 1 0.1965 0.65005 0.924 0.000 0.004 0.040 0.008 0.024
#> GSM1269703 1 0.0909 0.64325 0.968 0.000 0.000 0.012 0.000 0.020
#> GSM1269711 1 0.6419 0.44810 0.556 0.004 0.096 0.280 0.032 0.032
#> GSM1269646 3 0.5876 0.01542 0.012 0.064 0.500 0.000 0.392 0.032
#> GSM1269654 1 0.6579 0.43189 0.580 0.248 0.056 0.028 0.072 0.016
#> GSM1269662 3 0.8473 -0.07783 0.248 0.044 0.316 0.240 0.012 0.140
#> GSM1269670 1 0.6447 0.50567 0.644 0.048 0.124 0.012 0.112 0.060
#> GSM1269678 5 0.5235 0.27358 0.016 0.028 0.292 0.028 0.632 0.004
#> GSM1269692 4 0.3010 0.55662 0.024 0.012 0.008 0.876 0.064 0.016
#> GSM1269700 3 0.4142 0.42218 0.076 0.004 0.808 0.044 0.056 0.012
#> GSM1269708 5 0.3853 0.48972 0.196 0.000 0.004 0.044 0.756 0.000
#> GSM1269714 5 0.4548 0.45717 0.036 0.012 0.076 0.088 0.780 0.008
#> GSM1269716 4 0.7936 0.21399 0.268 0.080 0.112 0.448 0.076 0.016
#> GSM1269720 4 0.8481 0.32499 0.132 0.220 0.028 0.420 0.076 0.124
#> GSM1269722 3 0.5244 0.39059 0.008 0.044 0.712 0.112 0.120 0.004
#> GSM1269644 4 0.2757 0.56190 0.136 0.004 0.000 0.848 0.004 0.008
#> GSM1269652 5 0.6998 0.20509 0.148 0.000 0.000 0.208 0.484 0.160
#> GSM1269660 1 0.6284 0.23510 0.460 0.008 0.092 0.404 0.020 0.016
#> GSM1269668 4 0.6090 0.40094 0.108 0.004 0.240 0.592 0.052 0.004
#> GSM1269676 6 0.4382 0.65218 0.080 0.004 0.000 0.200 0.000 0.716
#> GSM1269684 1 0.2871 0.63936 0.804 0.000 0.004 0.192 0.000 0.000
#> GSM1269690 4 0.1588 0.56118 0.072 0.000 0.000 0.924 0.000 0.004
#> GSM1269698 1 0.4355 0.60040 0.776 0.004 0.004 0.092 0.024 0.100
#> GSM1269706 1 0.7528 -0.05265 0.360 0.004 0.008 0.336 0.112 0.180
#> GSM1269650 2 0.3737 0.86274 0.036 0.832 0.004 0.036 0.012 0.080
#> GSM1269658 4 0.6260 0.46450 0.024 0.220 0.040 0.624 0.028 0.064
#> GSM1269666 3 0.7431 0.31475 0.024 0.180 0.480 0.140 0.176 0.000
#> GSM1269674 4 0.7445 0.02621 0.040 0.020 0.316 0.428 0.168 0.028
#> GSM1269682 1 0.5624 0.49984 0.664 0.008 0.176 0.072 0.080 0.000
#> GSM1269688 3 0.7611 0.16063 0.016 0.036 0.436 0.316 0.080 0.116
#> GSM1269696 5 0.4874 0.11977 0.024 0.020 0.472 0.000 0.484 0.000
#> GSM1269704 5 0.8148 0.29501 0.232 0.148 0.128 0.016 0.428 0.048
#> GSM1269712 5 0.5989 0.23434 0.376 0.004 0.148 0.004 0.464 0.004
#> GSM1269718 1 0.7359 0.26498 0.508 0.072 0.212 0.020 0.168 0.020
#> GSM1269724 5 0.4711 0.38220 0.008 0.040 0.188 0.040 0.724 0.000
#> GSM1269726 4 0.5697 0.33506 0.064 0.008 0.264 0.620 0.036 0.008
#> GSM1269648 1 0.2189 0.65347 0.904 0.000 0.000 0.060 0.004 0.032
#> GSM1269656 1 0.4774 0.50366 0.672 0.000 0.000 0.136 0.000 0.192
#> GSM1269664 1 0.4097 0.08743 0.504 0.000 0.008 0.488 0.000 0.000
#> GSM1269672 4 0.3411 0.51793 0.232 0.000 0.008 0.756 0.004 0.000
#> GSM1269680 1 0.3979 0.37976 0.628 0.012 0.000 0.000 0.000 0.360
#> GSM1269686 1 0.2170 0.65060 0.908 0.000 0.016 0.060 0.016 0.000
#> GSM1269694 1 0.1515 0.64524 0.944 0.000 0.000 0.020 0.008 0.028
#> GSM1269702 1 0.1462 0.64635 0.936 0.000 0.000 0.056 0.000 0.008
#> GSM1269710 1 0.5380 0.05186 0.492 0.004 0.028 0.444 0.016 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 agent(p) disease.state(p) gender(p) individual(p) k
#> SD:pam 0 NA NA NA NA 2
#> SD:pam 47 0.989 0.1107 9.72e-05 0.01066 3
#> SD:pam 33 1.000 0.0932 8.53e-01 0.09428 4
#> SD:pam 30 0.993 0.0481 4.15e-03 0.00217 5
#> SD:pam 28 0.977 0.0287 2.88e-02 0.00195 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 51941 rows and 84 columns.
#> Top rows (1000, 2000, 3000, 4000, 5000) are extracted by 'SD' method.
#> Subgroups are detected by 'mclust' method.
#> Performed in total 1250 partitions by row resampling.
#> Best k for subgroups seems to be 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 0.295 0.867 0.803 0.3692 0.504 0.504
#> 3 3 0.265 0.795 0.819 0.5548 0.884 0.773
#> 4 4 0.449 0.669 0.741 0.2210 0.865 0.671
#> 5 5 0.527 0.685 0.733 0.0859 0.816 0.462
#> 6 6 0.677 0.671 0.796 0.0658 0.934 0.707
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
#> GSM1269647 1 0.876 0.916 0.704 0.296
#> GSM1269655 1 0.866 0.915 0.712 0.288
#> GSM1269663 1 0.866 0.917 0.712 0.288
#> GSM1269671 1 0.876 0.916 0.704 0.296
#> GSM1269679 1 0.671 0.837 0.824 0.176
#> GSM1269693 1 0.866 0.914 0.712 0.288
#> GSM1269701 1 0.781 0.887 0.768 0.232
#> GSM1269709 1 0.886 0.916 0.696 0.304
#> GSM1269715 1 0.881 0.916 0.700 0.300
#> GSM1269717 1 0.881 0.916 0.700 0.300
#> GSM1269721 1 0.767 0.861 0.776 0.224
#> GSM1269723 1 0.689 0.844 0.816 0.184
#> GSM1269645 2 0.494 0.881 0.108 0.892
#> GSM1269653 2 0.595 0.881 0.144 0.856
#> GSM1269661 2 0.738 0.828 0.208 0.792
#> GSM1269669 2 0.634 0.871 0.160 0.840
#> GSM1269677 2 0.913 0.755 0.328 0.672
#> GSM1269685 2 0.278 0.862 0.048 0.952
#> GSM1269691 2 0.163 0.847 0.024 0.976
#> GSM1269699 2 0.615 0.878 0.152 0.848
#> GSM1269707 2 0.625 0.875 0.156 0.844
#> GSM1269651 1 0.563 0.744 0.868 0.132
#> GSM1269659 1 0.644 0.712 0.836 0.164
#> GSM1269667 1 0.738 0.867 0.792 0.208
#> GSM1269675 1 0.881 0.916 0.700 0.300
#> GSM1269683 1 0.876 0.916 0.704 0.296
#> GSM1269689 1 0.881 0.916 0.700 0.300
#> GSM1269697 1 0.876 0.916 0.704 0.296
#> GSM1269705 1 0.881 0.916 0.700 0.300
#> GSM1269713 1 0.871 0.917 0.708 0.292
#> GSM1269719 1 0.866 0.917 0.712 0.288
#> GSM1269725 1 0.827 0.903 0.740 0.260
#> GSM1269727 1 0.671 0.835 0.824 0.176
#> GSM1269649 2 0.634 0.866 0.160 0.840
#> GSM1269657 2 0.753 0.859 0.216 0.784
#> GSM1269665 2 0.541 0.875 0.124 0.876
#> GSM1269673 2 0.163 0.847 0.024 0.976
#> GSM1269681 2 0.980 0.660 0.416 0.584
#> GSM1269687 2 0.204 0.853 0.032 0.968
#> GSM1269695 2 0.552 0.883 0.128 0.872
#> GSM1269703 2 0.163 0.847 0.024 0.976
#> GSM1269711 2 0.552 0.883 0.128 0.872
#> GSM1269646 1 0.876 0.916 0.704 0.296
#> GSM1269654 1 0.866 0.917 0.712 0.288
#> GSM1269662 1 0.871 0.916 0.708 0.292
#> GSM1269670 1 0.876 0.916 0.704 0.296
#> GSM1269678 1 0.653 0.828 0.832 0.168
#> GSM1269692 1 0.866 0.914 0.712 0.288
#> GSM1269700 1 0.775 0.884 0.772 0.228
#> GSM1269708 1 0.886 0.916 0.696 0.304
#> GSM1269714 1 0.881 0.916 0.700 0.300
#> GSM1269716 1 0.881 0.916 0.700 0.300
#> GSM1269720 1 0.671 0.804 0.824 0.176
#> GSM1269722 1 0.662 0.830 0.828 0.172
#> GSM1269644 2 0.469 0.882 0.100 0.900
#> GSM1269652 2 0.584 0.882 0.140 0.860
#> GSM1269660 2 0.839 0.715 0.268 0.732
#> GSM1269668 2 0.634 0.871 0.160 0.840
#> GSM1269676 2 0.913 0.755 0.328 0.672
#> GSM1269684 2 0.163 0.847 0.024 0.976
#> GSM1269690 2 0.260 0.862 0.044 0.956
#> GSM1269698 2 0.625 0.875 0.156 0.844
#> GSM1269706 2 0.615 0.878 0.152 0.848
#> GSM1269650 1 0.563 0.744 0.868 0.132
#> GSM1269658 1 0.644 0.712 0.836 0.164
#> GSM1269666 1 0.706 0.850 0.808 0.192
#> GSM1269674 1 0.881 0.916 0.700 0.300
#> GSM1269682 1 0.881 0.916 0.700 0.300
#> GSM1269688 1 0.881 0.916 0.700 0.300
#> GSM1269696 1 0.876 0.916 0.704 0.296
#> GSM1269704 1 0.881 0.916 0.700 0.300
#> GSM1269712 1 0.662 0.833 0.828 0.172
#> GSM1269718 1 0.866 0.917 0.712 0.288
#> GSM1269724 1 0.714 0.858 0.804 0.196
#> GSM1269726 1 0.767 0.880 0.776 0.224
#> GSM1269648 2 0.605 0.876 0.148 0.852
#> GSM1269656 2 0.644 0.872 0.164 0.836
#> GSM1269664 2 0.563 0.879 0.132 0.868
#> GSM1269672 2 0.163 0.847 0.024 0.976
#> GSM1269680 2 0.936 0.740 0.352 0.648
#> GSM1269686 2 0.224 0.857 0.036 0.964
#> GSM1269694 2 0.552 0.883 0.128 0.872
#> GSM1269702 2 0.278 0.864 0.048 0.952
#> GSM1269710 2 0.552 0.883 0.128 0.872
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM1269647 3 0.419 0.823 0.064 0.060 0.876
#> GSM1269655 3 0.300 0.855 0.068 0.016 0.916
#> GSM1269663 3 0.651 0.785 0.072 0.180 0.748
#> GSM1269671 3 0.473 0.810 0.088 0.060 0.852
#> GSM1269679 3 0.249 0.848 0.016 0.048 0.936
#> GSM1269693 3 0.701 0.745 0.080 0.208 0.712
#> GSM1269701 3 0.314 0.853 0.020 0.068 0.912
#> GSM1269709 3 0.425 0.849 0.080 0.048 0.872
#> GSM1269715 3 0.654 0.776 0.084 0.164 0.752
#> GSM1269717 3 0.646 0.779 0.080 0.164 0.756
#> GSM1269721 2 0.725 0.568 0.056 0.656 0.288
#> GSM1269723 3 0.249 0.852 0.016 0.048 0.936
#> GSM1269645 1 0.569 0.771 0.756 0.020 0.224
#> GSM1269653 1 0.463 0.795 0.856 0.056 0.088
#> GSM1269661 1 0.731 0.665 0.616 0.044 0.340
#> GSM1269669 1 0.634 0.810 0.736 0.044 0.220
#> GSM1269677 2 0.639 0.746 0.184 0.752 0.064
#> GSM1269685 1 0.466 0.826 0.852 0.048 0.100
#> GSM1269691 1 0.425 0.826 0.864 0.028 0.108
#> GSM1269699 1 0.541 0.782 0.820 0.076 0.104
#> GSM1269707 1 0.475 0.801 0.852 0.068 0.080
#> GSM1269651 2 0.586 0.780 0.024 0.748 0.228
#> GSM1269659 2 0.437 0.802 0.032 0.860 0.108
#> GSM1269667 3 0.223 0.852 0.012 0.044 0.944
#> GSM1269675 3 0.315 0.847 0.044 0.040 0.916
#> GSM1269683 3 0.594 0.800 0.072 0.140 0.788
#> GSM1269689 3 0.415 0.827 0.080 0.044 0.876
#> GSM1269697 3 0.346 0.837 0.060 0.036 0.904
#> GSM1269705 3 0.274 0.855 0.052 0.020 0.928
#> GSM1269713 3 0.227 0.853 0.040 0.016 0.944
#> GSM1269719 3 0.600 0.810 0.072 0.144 0.784
#> GSM1269725 3 0.244 0.855 0.032 0.028 0.940
#> GSM1269727 3 0.290 0.853 0.016 0.064 0.920
#> GSM1269649 1 0.641 0.770 0.716 0.036 0.248
#> GSM1269657 1 0.826 0.374 0.556 0.356 0.088
#> GSM1269665 1 0.588 0.739 0.728 0.016 0.256
#> GSM1269673 1 0.435 0.831 0.852 0.020 0.128
#> GSM1269681 2 0.635 0.773 0.156 0.764 0.080
#> GSM1269687 1 0.474 0.825 0.828 0.020 0.152
#> GSM1269695 1 0.448 0.777 0.864 0.064 0.072
#> GSM1269703 1 0.421 0.825 0.860 0.020 0.120
#> GSM1269711 1 0.438 0.772 0.868 0.064 0.068
#> GSM1269646 3 0.419 0.823 0.064 0.060 0.876
#> GSM1269654 3 0.353 0.853 0.068 0.032 0.900
#> GSM1269662 3 0.702 0.746 0.072 0.224 0.704
#> GSM1269670 3 0.456 0.815 0.080 0.060 0.860
#> GSM1269678 3 0.270 0.847 0.016 0.056 0.928
#> GSM1269692 3 0.719 0.726 0.080 0.224 0.696
#> GSM1269700 3 0.280 0.854 0.016 0.060 0.924
#> GSM1269708 3 0.446 0.844 0.080 0.056 0.864
#> GSM1269714 3 0.635 0.784 0.080 0.156 0.764
#> GSM1269716 3 0.646 0.779 0.080 0.164 0.756
#> GSM1269720 2 0.626 0.722 0.032 0.724 0.244
#> GSM1269722 3 0.270 0.847 0.016 0.056 0.928
#> GSM1269644 1 0.509 0.825 0.824 0.040 0.136
#> GSM1269652 1 0.447 0.781 0.864 0.060 0.076
#> GSM1269660 1 0.798 0.415 0.500 0.060 0.440
#> GSM1269668 1 0.640 0.803 0.724 0.040 0.236
#> GSM1269676 2 0.639 0.746 0.184 0.752 0.064
#> GSM1269684 1 0.392 0.829 0.872 0.016 0.112
#> GSM1269690 1 0.453 0.826 0.856 0.040 0.104
#> GSM1269698 1 0.533 0.786 0.824 0.076 0.100
#> GSM1269706 1 0.475 0.797 0.852 0.068 0.080
#> GSM1269650 2 0.582 0.783 0.024 0.752 0.224
#> GSM1269658 2 0.437 0.802 0.032 0.860 0.108
#> GSM1269666 3 0.212 0.851 0.012 0.040 0.948
#> GSM1269674 3 0.338 0.856 0.048 0.044 0.908
#> GSM1269682 3 0.591 0.801 0.068 0.144 0.788
#> GSM1269688 3 0.397 0.828 0.088 0.032 0.880
#> GSM1269696 3 0.409 0.826 0.068 0.052 0.880
#> GSM1269704 3 0.298 0.854 0.056 0.024 0.920
#> GSM1269712 3 0.212 0.850 0.012 0.040 0.948
#> GSM1269718 3 0.559 0.817 0.068 0.124 0.808
#> GSM1269724 3 0.175 0.856 0.012 0.028 0.960
#> GSM1269726 3 0.412 0.846 0.024 0.108 0.868
#> GSM1269648 1 0.539 0.817 0.808 0.044 0.148
#> GSM1269656 1 0.618 0.804 0.780 0.104 0.116
#> GSM1269664 1 0.603 0.752 0.732 0.024 0.244
#> GSM1269672 1 0.421 0.830 0.856 0.016 0.128
#> GSM1269680 2 0.628 0.754 0.176 0.760 0.064
#> GSM1269686 1 0.445 0.827 0.836 0.012 0.152
#> GSM1269694 1 0.474 0.786 0.852 0.064 0.084
#> GSM1269702 1 0.445 0.826 0.860 0.040 0.100
#> GSM1269710 1 0.456 0.780 0.860 0.064 0.076
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM1269647 3 0.2673 0.770 0.048 0.016 0.916 0.020
#> GSM1269655 2 0.5512 0.127 0.000 0.496 0.488 0.016
#> GSM1269663 2 0.4004 0.797 0.000 0.812 0.164 0.024
#> GSM1269671 3 0.4217 0.736 0.112 0.024 0.836 0.028
#> GSM1269679 3 0.3545 0.732 0.000 0.164 0.828 0.008
#> GSM1269693 2 0.3172 0.796 0.004 0.872 0.112 0.012
#> GSM1269701 3 0.5345 0.391 0.004 0.404 0.584 0.008
#> GSM1269709 3 0.7086 0.542 0.112 0.308 0.568 0.012
#> GSM1269715 2 0.3249 0.816 0.008 0.852 0.140 0.000
#> GSM1269717 2 0.3052 0.817 0.004 0.860 0.136 0.000
#> GSM1269721 2 0.7269 -0.203 0.000 0.456 0.148 0.396
#> GSM1269723 3 0.4049 0.693 0.000 0.212 0.780 0.008
#> GSM1269645 1 0.7258 0.676 0.544 0.352 0.052 0.052
#> GSM1269653 1 0.2238 0.650 0.920 0.004 0.004 0.072
#> GSM1269661 1 0.9250 0.544 0.448 0.216 0.204 0.132
#> GSM1269669 1 0.6925 0.703 0.660 0.164 0.144 0.032
#> GSM1269677 4 0.1820 0.745 0.036 0.020 0.000 0.944
#> GSM1269685 1 0.6262 0.754 0.636 0.296 0.016 0.052
#> GSM1269691 1 0.6355 0.751 0.628 0.300 0.016 0.056
#> GSM1269699 1 0.2831 0.615 0.876 0.004 0.000 0.120
#> GSM1269707 1 0.2075 0.677 0.936 0.016 0.004 0.044
#> GSM1269651 4 0.6240 0.656 0.000 0.156 0.176 0.668
#> GSM1269659 4 0.4897 0.606 0.000 0.332 0.008 0.660
#> GSM1269667 3 0.3300 0.759 0.000 0.144 0.848 0.008
#> GSM1269675 3 0.3385 0.778 0.048 0.056 0.884 0.012
#> GSM1269683 2 0.3444 0.805 0.000 0.816 0.184 0.000
#> GSM1269689 3 0.3769 0.746 0.096 0.020 0.860 0.024
#> GSM1269697 3 0.3290 0.772 0.068 0.024 0.888 0.020
#> GSM1269705 3 0.3656 0.778 0.040 0.080 0.868 0.012
#> GSM1269713 3 0.1721 0.782 0.012 0.028 0.952 0.008
#> GSM1269719 2 0.3351 0.797 0.000 0.844 0.148 0.008
#> GSM1269725 3 0.1585 0.782 0.004 0.040 0.952 0.004
#> GSM1269727 3 0.5070 0.381 0.000 0.372 0.620 0.008
#> GSM1269649 1 0.5285 0.635 0.760 0.040 0.176 0.024
#> GSM1269657 4 0.6674 0.141 0.316 0.096 0.004 0.584
#> GSM1269665 1 0.7499 0.665 0.532 0.348 0.068 0.052
#> GSM1269673 1 0.6080 0.756 0.660 0.272 0.012 0.056
#> GSM1269681 4 0.2261 0.745 0.036 0.024 0.008 0.932
#> GSM1269687 1 0.6467 0.746 0.612 0.316 0.020 0.052
#> GSM1269695 1 0.0804 0.682 0.980 0.000 0.012 0.008
#> GSM1269703 1 0.6457 0.749 0.624 0.300 0.020 0.056
#> GSM1269711 1 0.0672 0.681 0.984 0.000 0.008 0.008
#> GSM1269646 3 0.2786 0.771 0.048 0.020 0.912 0.020
#> GSM1269654 2 0.5150 0.438 0.000 0.596 0.396 0.008
#> GSM1269662 2 0.4037 0.779 0.000 0.824 0.136 0.040
#> GSM1269670 3 0.3601 0.746 0.100 0.012 0.864 0.024
#> GSM1269678 3 0.3852 0.710 0.000 0.192 0.800 0.008
#> GSM1269692 2 0.3450 0.784 0.004 0.864 0.108 0.024
#> GSM1269700 3 0.5349 0.460 0.008 0.364 0.620 0.008
#> GSM1269708 3 0.6330 0.535 0.056 0.344 0.592 0.008
#> GSM1269714 2 0.3257 0.817 0.004 0.844 0.152 0.000
#> GSM1269716 2 0.3052 0.817 0.004 0.860 0.136 0.000
#> GSM1269720 4 0.7290 0.457 0.000 0.328 0.168 0.504
#> GSM1269722 3 0.4228 0.684 0.000 0.232 0.760 0.008
#> GSM1269644 1 0.6769 0.723 0.588 0.324 0.020 0.068
#> GSM1269652 1 0.1389 0.661 0.952 0.000 0.000 0.048
#> GSM1269660 1 0.9533 0.327 0.340 0.328 0.200 0.132
#> GSM1269668 1 0.7369 0.690 0.624 0.180 0.156 0.040
#> GSM1269676 4 0.1820 0.745 0.036 0.020 0.000 0.944
#> GSM1269684 1 0.6355 0.750 0.628 0.300 0.016 0.056
#> GSM1269690 1 0.6333 0.753 0.632 0.296 0.016 0.056
#> GSM1269698 1 0.2888 0.611 0.872 0.004 0.000 0.124
#> GSM1269706 1 0.2125 0.671 0.932 0.012 0.004 0.052
#> GSM1269650 4 0.6240 0.656 0.000 0.156 0.176 0.668
#> GSM1269658 4 0.4999 0.607 0.000 0.328 0.012 0.660
#> GSM1269666 3 0.3450 0.744 0.000 0.156 0.836 0.008
#> GSM1269674 3 0.3845 0.737 0.016 0.132 0.840 0.012
#> GSM1269682 2 0.3402 0.817 0.000 0.832 0.164 0.004
#> GSM1269688 3 0.4664 0.738 0.116 0.056 0.812 0.016
#> GSM1269696 3 0.3161 0.775 0.056 0.028 0.896 0.020
#> GSM1269704 3 0.3790 0.770 0.040 0.096 0.856 0.008
#> GSM1269712 3 0.3933 0.711 0.000 0.200 0.792 0.008
#> GSM1269718 2 0.3529 0.782 0.000 0.836 0.152 0.012
#> GSM1269724 3 0.1978 0.777 0.000 0.068 0.928 0.004
#> GSM1269726 2 0.5212 0.224 0.000 0.572 0.420 0.008
#> GSM1269648 1 0.2497 0.692 0.924 0.016 0.040 0.020
#> GSM1269656 1 0.7720 0.654 0.512 0.284 0.012 0.192
#> GSM1269664 1 0.7325 0.681 0.540 0.352 0.056 0.052
#> GSM1269672 1 0.6082 0.755 0.652 0.284 0.012 0.052
#> GSM1269680 4 0.1913 0.745 0.040 0.020 0.000 0.940
#> GSM1269686 1 0.6436 0.752 0.628 0.296 0.020 0.056
#> GSM1269694 1 0.0672 0.681 0.984 0.000 0.008 0.008
#> GSM1269702 1 0.6217 0.755 0.644 0.288 0.016 0.052
#> GSM1269710 1 0.0524 0.680 0.988 0.000 0.004 0.008
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM1269647 3 0.157 0.8409 0.000 0.000 0.936 0.060 0.004
#> GSM1269655 4 0.683 0.6642 0.164 0.076 0.148 0.608 0.004
#> GSM1269663 4 0.641 0.6509 0.180 0.120 0.048 0.644 0.008
#> GSM1269671 3 0.186 0.8015 0.000 0.016 0.932 0.004 0.048
#> GSM1269679 4 0.586 0.5310 0.108 0.004 0.276 0.608 0.004
#> GSM1269693 4 0.545 0.5478 0.100 0.148 0.008 0.720 0.024
#> GSM1269701 4 0.587 0.5699 0.144 0.000 0.240 0.612 0.004
#> GSM1269709 4 0.754 0.4788 0.116 0.008 0.264 0.512 0.100
#> GSM1269715 4 0.498 0.5766 0.096 0.112 0.004 0.760 0.028
#> GSM1269717 4 0.502 0.5732 0.096 0.116 0.004 0.756 0.028
#> GSM1269721 2 0.611 0.5828 0.128 0.612 0.012 0.244 0.004
#> GSM1269723 4 0.528 0.5583 0.076 0.004 0.268 0.652 0.000
#> GSM1269645 1 0.351 0.6848 0.828 0.008 0.012 0.144 0.008
#> GSM1269653 5 0.199 0.9342 0.076 0.004 0.004 0.000 0.916
#> GSM1269661 1 0.765 0.4354 0.536 0.032 0.080 0.244 0.108
#> GSM1269669 1 0.711 0.4557 0.556 0.012 0.040 0.220 0.172
#> GSM1269677 2 0.348 0.7839 0.076 0.836 0.000 0.000 0.088
#> GSM1269685 1 0.185 0.7224 0.912 0.000 0.000 0.000 0.088
#> GSM1269691 1 0.218 0.7153 0.896 0.000 0.000 0.004 0.100
#> GSM1269699 5 0.193 0.9308 0.072 0.004 0.004 0.000 0.920
#> GSM1269707 5 0.218 0.9386 0.100 0.004 0.000 0.000 0.896
#> GSM1269651 2 0.260 0.8049 0.008 0.904 0.040 0.044 0.004
#> GSM1269659 2 0.336 0.7613 0.012 0.816 0.000 0.168 0.004
#> GSM1269667 4 0.600 0.4971 0.108 0.004 0.308 0.576 0.004
#> GSM1269675 3 0.199 0.8402 0.004 0.000 0.916 0.076 0.004
#> GSM1269683 4 0.583 0.6425 0.092 0.100 0.064 0.724 0.020
#> GSM1269689 3 0.242 0.8120 0.000 0.016 0.912 0.036 0.036
#> GSM1269697 3 0.265 0.8330 0.000 0.000 0.884 0.084 0.032
#> GSM1269705 3 0.365 0.7836 0.028 0.000 0.808 0.160 0.004
#> GSM1269713 3 0.281 0.7990 0.000 0.000 0.844 0.152 0.004
#> GSM1269719 4 0.647 0.6579 0.212 0.084 0.064 0.632 0.008
#> GSM1269725 3 0.343 0.7583 0.000 0.000 0.776 0.220 0.004
#> GSM1269727 4 0.514 0.6305 0.124 0.004 0.168 0.704 0.000
#> GSM1269649 5 0.689 0.5713 0.120 0.004 0.200 0.080 0.596
#> GSM1269657 1 0.683 -0.0425 0.440 0.396 0.000 0.028 0.136
#> GSM1269665 1 0.367 0.6619 0.808 0.004 0.020 0.164 0.004
#> GSM1269673 1 0.244 0.7000 0.876 0.000 0.000 0.004 0.120
#> GSM1269681 2 0.368 0.7837 0.072 0.828 0.000 0.004 0.096
#> GSM1269687 1 0.301 0.7423 0.876 0.000 0.008 0.052 0.064
#> GSM1269695 5 0.218 0.9381 0.112 0.000 0.000 0.000 0.888
#> GSM1269703 1 0.163 0.7327 0.936 0.000 0.000 0.008 0.056
#> GSM1269711 5 0.213 0.9393 0.108 0.000 0.000 0.000 0.892
#> GSM1269646 3 0.170 0.8413 0.000 0.000 0.928 0.068 0.004
#> GSM1269654 4 0.670 0.6711 0.164 0.080 0.128 0.624 0.004
#> GSM1269662 4 0.674 0.6101 0.160 0.188 0.036 0.604 0.012
#> GSM1269670 3 0.163 0.8041 0.000 0.016 0.944 0.004 0.036
#> GSM1269678 4 0.577 0.5587 0.112 0.004 0.252 0.628 0.004
#> GSM1269692 4 0.545 0.5478 0.100 0.148 0.008 0.720 0.024
#> GSM1269700 4 0.572 0.5703 0.144 0.000 0.240 0.616 0.000
#> GSM1269708 4 0.718 0.5582 0.128 0.008 0.216 0.568 0.080
#> GSM1269714 4 0.498 0.5766 0.096 0.112 0.004 0.760 0.028
#> GSM1269716 4 0.498 0.5766 0.096 0.112 0.004 0.760 0.028
#> GSM1269720 2 0.607 0.6321 0.108 0.644 0.028 0.216 0.004
#> GSM1269722 4 0.519 0.5453 0.068 0.004 0.272 0.656 0.000
#> GSM1269644 1 0.326 0.7165 0.860 0.024 0.004 0.100 0.012
#> GSM1269652 5 0.212 0.9408 0.096 0.004 0.000 0.000 0.900
#> GSM1269660 1 0.798 0.3485 0.508 0.036 0.124 0.240 0.092
#> GSM1269668 1 0.707 0.4573 0.560 0.012 0.040 0.224 0.164
#> GSM1269676 2 0.348 0.7839 0.076 0.836 0.000 0.000 0.088
#> GSM1269684 1 0.236 0.7173 0.892 0.000 0.000 0.012 0.096
#> GSM1269690 1 0.262 0.7131 0.876 0.008 0.000 0.004 0.112
#> GSM1269698 5 0.186 0.9197 0.060 0.008 0.004 0.000 0.928
#> GSM1269706 5 0.212 0.9383 0.096 0.004 0.000 0.000 0.900
#> GSM1269650 2 0.260 0.8049 0.008 0.904 0.040 0.044 0.004
#> GSM1269658 2 0.356 0.7585 0.020 0.808 0.000 0.168 0.004
#> GSM1269666 4 0.587 0.5230 0.108 0.004 0.280 0.604 0.004
#> GSM1269674 3 0.412 0.7340 0.032 0.004 0.776 0.184 0.004
#> GSM1269682 4 0.611 0.6578 0.100 0.112 0.080 0.696 0.012
#> GSM1269688 3 0.433 0.7493 0.004 0.016 0.796 0.060 0.124
#> GSM1269696 3 0.219 0.8356 0.000 0.000 0.904 0.084 0.012
#> GSM1269704 3 0.373 0.7844 0.028 0.004 0.812 0.152 0.004
#> GSM1269712 4 0.520 0.3725 0.040 0.004 0.380 0.576 0.000
#> GSM1269718 4 0.661 0.6711 0.196 0.068 0.100 0.628 0.008
#> GSM1269724 3 0.449 0.4062 0.008 0.000 0.624 0.364 0.004
#> GSM1269726 4 0.603 0.6552 0.168 0.028 0.156 0.648 0.000
#> GSM1269648 5 0.338 0.9089 0.108 0.000 0.032 0.012 0.848
#> GSM1269656 1 0.639 0.3591 0.560 0.288 0.000 0.020 0.132
#> GSM1269664 1 0.420 0.6426 0.784 0.008 0.040 0.164 0.004
#> GSM1269672 1 0.202 0.7196 0.900 0.000 0.000 0.000 0.100
#> GSM1269680 2 0.353 0.7835 0.072 0.832 0.000 0.000 0.096
#> GSM1269686 1 0.308 0.7420 0.872 0.000 0.008 0.060 0.060
#> GSM1269694 5 0.218 0.9381 0.112 0.000 0.000 0.000 0.888
#> GSM1269702 1 0.258 0.6853 0.864 0.000 0.000 0.004 0.132
#> GSM1269710 5 0.213 0.9393 0.108 0.000 0.000 0.000 0.892
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM1269647 2 0.1970 0.8020 0.000 0.900 0.092 0.008 0.000 0.000
#> GSM1269655 3 0.5930 0.2244 0.032 0.028 0.524 0.368 0.000 0.048
#> GSM1269663 4 0.6192 0.2875 0.028 0.004 0.356 0.480 0.000 0.132
#> GSM1269671 2 0.1053 0.7804 0.000 0.964 0.020 0.004 0.012 0.000
#> GSM1269679 3 0.1528 0.7577 0.016 0.028 0.944 0.012 0.000 0.000
#> GSM1269693 4 0.3176 0.6900 0.040 0.000 0.048 0.856 0.000 0.056
#> GSM1269701 3 0.3520 0.7143 0.016 0.084 0.836 0.052 0.012 0.000
#> GSM1269709 3 0.5190 0.6651 0.060 0.052 0.744 0.048 0.092 0.004
#> GSM1269715 4 0.2344 0.7139 0.048 0.000 0.052 0.896 0.000 0.004
#> GSM1269717 4 0.2550 0.7186 0.048 0.004 0.056 0.888 0.000 0.004
#> GSM1269721 6 0.6407 0.5023 0.032 0.024 0.116 0.300 0.000 0.528
#> GSM1269723 3 0.2016 0.7527 0.024 0.016 0.920 0.040 0.000 0.000
#> GSM1269645 1 0.4652 0.6666 0.720 0.008 0.152 0.116 0.004 0.000
#> GSM1269653 5 0.1003 0.9421 0.020 0.000 0.000 0.000 0.964 0.016
#> GSM1269661 1 0.7485 0.3682 0.456 0.012 0.308 0.072 0.108 0.044
#> GSM1269669 1 0.7013 0.4860 0.500 0.008 0.224 0.048 0.204 0.016
#> GSM1269677 6 0.1059 0.7239 0.016 0.000 0.000 0.004 0.016 0.964
#> GSM1269685 1 0.0790 0.7662 0.968 0.000 0.000 0.000 0.032 0.000
#> GSM1269691 1 0.0363 0.7678 0.988 0.000 0.000 0.000 0.012 0.000
#> GSM1269699 5 0.1148 0.9380 0.004 0.016 0.000 0.000 0.960 0.020
#> GSM1269707 5 0.0909 0.9411 0.012 0.000 0.000 0.000 0.968 0.020
#> GSM1269651 6 0.4066 0.7076 0.000 0.040 0.032 0.156 0.000 0.772
#> GSM1269659 6 0.4748 0.5970 0.016 0.004 0.028 0.332 0.000 0.620
#> GSM1269667 3 0.2521 0.7443 0.020 0.056 0.892 0.032 0.000 0.000
#> GSM1269675 2 0.3625 0.7754 0.000 0.768 0.204 0.020 0.004 0.004
#> GSM1269683 4 0.5803 0.2799 0.052 0.020 0.380 0.520 0.000 0.028
#> GSM1269689 2 0.2760 0.7981 0.012 0.868 0.100 0.004 0.016 0.000
#> GSM1269697 2 0.3081 0.8069 0.012 0.824 0.152 0.000 0.012 0.000
#> GSM1269705 2 0.4874 0.6414 0.008 0.624 0.324 0.028 0.004 0.012
#> GSM1269713 2 0.4317 0.6873 0.012 0.660 0.312 0.008 0.004 0.004
#> GSM1269719 3 0.5724 0.2280 0.020 0.012 0.544 0.348 0.000 0.076
#> GSM1269725 2 0.4203 0.6277 0.008 0.608 0.376 0.004 0.000 0.004
#> GSM1269727 3 0.1649 0.7495 0.032 0.000 0.932 0.036 0.000 0.000
#> GSM1269649 5 0.5408 0.6151 0.048 0.104 0.152 0.004 0.688 0.004
#> GSM1269657 6 0.5506 0.3323 0.324 0.004 0.008 0.040 0.036 0.588
#> GSM1269665 1 0.4819 0.6346 0.688 0.004 0.192 0.112 0.000 0.004
#> GSM1269673 1 0.1049 0.7619 0.960 0.000 0.000 0.008 0.032 0.000
#> GSM1269681 6 0.0862 0.7232 0.004 0.000 0.000 0.008 0.016 0.972
#> GSM1269687 1 0.1686 0.7715 0.940 0.000 0.016 0.024 0.016 0.004
#> GSM1269695 5 0.0547 0.9412 0.020 0.000 0.000 0.000 0.980 0.000
#> GSM1269703 1 0.0405 0.7706 0.988 0.000 0.008 0.000 0.004 0.000
#> GSM1269711 5 0.0458 0.9412 0.016 0.000 0.000 0.000 0.984 0.000
#> GSM1269646 2 0.1956 0.8035 0.000 0.908 0.080 0.008 0.004 0.000
#> GSM1269654 3 0.6154 0.1378 0.040 0.032 0.484 0.396 0.000 0.048
#> GSM1269662 4 0.6368 0.3643 0.032 0.000 0.256 0.492 0.000 0.220
#> GSM1269670 2 0.0964 0.7789 0.000 0.968 0.016 0.004 0.012 0.000
#> GSM1269678 3 0.1346 0.7579 0.016 0.024 0.952 0.008 0.000 0.000
#> GSM1269692 4 0.3110 0.6872 0.040 0.000 0.044 0.860 0.000 0.056
#> GSM1269700 3 0.3469 0.7169 0.016 0.080 0.840 0.052 0.012 0.000
#> GSM1269708 3 0.4422 0.6940 0.064 0.032 0.796 0.036 0.068 0.004
#> GSM1269714 4 0.2662 0.7190 0.048 0.004 0.056 0.884 0.000 0.008
#> GSM1269716 4 0.2550 0.7186 0.048 0.004 0.056 0.888 0.000 0.004
#> GSM1269720 6 0.6271 0.5346 0.020 0.028 0.128 0.272 0.000 0.552
#> GSM1269722 3 0.1536 0.7577 0.024 0.020 0.944 0.012 0.000 0.000
#> GSM1269644 1 0.3209 0.7458 0.856 0.000 0.060 0.056 0.004 0.024
#> GSM1269652 5 0.0725 0.9428 0.012 0.000 0.000 0.000 0.976 0.012
#> GSM1269660 1 0.8003 0.2772 0.404 0.028 0.320 0.100 0.104 0.044
#> GSM1269668 1 0.7028 0.4789 0.496 0.008 0.232 0.048 0.200 0.016
#> GSM1269676 6 0.1059 0.7239 0.016 0.000 0.000 0.004 0.016 0.964
#> GSM1269684 1 0.0363 0.7678 0.988 0.000 0.000 0.000 0.012 0.000
#> GSM1269690 1 0.0547 0.7682 0.980 0.000 0.000 0.000 0.020 0.000
#> GSM1269698 5 0.1313 0.9355 0.004 0.016 0.000 0.000 0.952 0.028
#> GSM1269706 5 0.1003 0.9402 0.016 0.000 0.000 0.000 0.964 0.020
#> GSM1269650 6 0.4066 0.7076 0.000 0.040 0.032 0.156 0.000 0.772
#> GSM1269658 6 0.4634 0.5896 0.012 0.004 0.024 0.344 0.000 0.616
#> GSM1269666 3 0.2421 0.7536 0.028 0.040 0.900 0.032 0.000 0.000
#> GSM1269674 2 0.4890 0.6912 0.008 0.660 0.276 0.036 0.004 0.016
#> GSM1269682 4 0.5668 0.4060 0.036 0.032 0.332 0.572 0.000 0.028
#> GSM1269688 2 0.4553 0.7459 0.012 0.744 0.144 0.004 0.092 0.004
#> GSM1269696 2 0.2841 0.8022 0.008 0.852 0.124 0.004 0.012 0.000
#> GSM1269704 2 0.4598 0.6755 0.008 0.652 0.304 0.028 0.004 0.004
#> GSM1269712 3 0.1692 0.7488 0.008 0.048 0.932 0.012 0.000 0.000
#> GSM1269718 3 0.5484 0.3649 0.024 0.012 0.596 0.308 0.000 0.060
#> GSM1269724 3 0.4109 -0.0407 0.008 0.392 0.596 0.000 0.000 0.004
#> GSM1269726 3 0.3514 0.6919 0.032 0.004 0.812 0.140 0.000 0.012
#> GSM1269648 5 0.2024 0.9035 0.028 0.016 0.036 0.000 0.920 0.000
#> GSM1269656 1 0.6099 0.1721 0.492 0.004 0.000 0.048 0.084 0.372
#> GSM1269664 1 0.5124 0.6442 0.692 0.016 0.184 0.096 0.004 0.008
#> GSM1269672 1 0.0458 0.7676 0.984 0.000 0.000 0.000 0.016 0.000
#> GSM1269680 6 0.0748 0.7234 0.004 0.000 0.000 0.004 0.016 0.976
#> GSM1269686 1 0.2044 0.7665 0.920 0.000 0.040 0.028 0.008 0.004
#> GSM1269694 5 0.0632 0.9404 0.024 0.000 0.000 0.000 0.976 0.000
#> GSM1269702 1 0.1340 0.7601 0.948 0.000 0.000 0.008 0.040 0.004
#> GSM1269710 5 0.0632 0.9404 0.024 0.000 0.000 0.000 0.976 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 agent(p) disease.state(p) gender(p) individual(p) k
#> SD:mclust 84 1.000 1.000 3.81e-19 8.65e-05 2
#> SD:mclust 82 1.000 0.781 2.19e-16 2.04e-07 3
#> SD:mclust 74 1.000 0.284 3.15e-14 2.08e-09 4
#> SD:mclust 74 0.978 0.115 4.09e-13 2.52e-11 5
#> SD:mclust 69 0.997 0.077 1.75e-11 2.05e-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["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 51941 rows and 84 columns.
#> Top rows (1000, 2000, 3000, 4000, 5000) are extracted by 'SD' method.
#> Subgroups are detected by 'NMF' method.
#> Performed in total 1250 partitions by row resampling.
#> Best k for subgroups seems to be 2.
#>
#> Following methods can be applied to this 'ConsensusPartition' object:
#> [1] "cola_report" "collect_classes" "collect_plots"
#> [4] "collect_stats" "colnames" "compare_signatures"
#> [7] "consensus_heatmap" "dimension_reduction" "functional_enrichment"
#> [10] "get_anno_col" "get_anno" "get_classes"
#> [13] "get_consensus" "get_matrix" "get_membership"
#> [16] "get_param" "get_signatures" "get_stats"
#> [19] "is_best_k" "is_stable_k" "membership_heatmap"
#> [22] "ncol" "nrow" "plot_ecdf"
#> [25] "rownames" "select_partition_number" "show"
#> [28] "suggest_best_k" "test_to_known_factors"
collect_plots() function collects all the plots made from res for all k (number of partitions)
into one single page to provide an easy and fast comparison between different k.
collect_plots(res)
The plots are:
k and the heatmap of
predicted classes for each k.k.k.k.All the plots in panels can be made by individual functions and they are plotted later in this section.
select_partition_number() produces several plots showing different
statistics for choosing “optimized” k. There are following statistics:
k;k, the area increased is defined as \(A_k - A_{k-1}\).The detailed explanations of these statistics can be found in the cola vignette.
Generally speaking, lower PAC score, higher mean silhouette score or higher
concordance corresponds to better partition. Rand index and Jaccard index
measure how similar the current partition is compared to partition with k-1.
If they are too similar, we won't accept k is better than k-1.
select_partition_number(res)
The numeric values for all these statistics can be obtained by get_stats().
get_stats(res)
#> k 1-PAC mean_silhouette concordance area_increased Rand Jaccard
#> 2 2 0.218 0.563 0.806 0.4909 0.504 0.504
#> 3 3 0.273 0.455 0.671 0.2873 0.742 0.566
#> 4 4 0.411 0.510 0.700 0.1431 0.783 0.527
#> 5 5 0.451 0.307 0.602 0.0825 0.972 0.905
#> 6 6 0.491 0.306 0.543 0.0504 0.834 0.468
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
#> GSM1269647 1 0.5629 0.71552 0.868 0.132
#> GSM1269655 1 0.6712 0.65801 0.824 0.176
#> GSM1269663 1 0.9963 0.11966 0.536 0.464
#> GSM1269671 1 0.9580 0.32162 0.620 0.380
#> GSM1269679 1 0.2603 0.75764 0.956 0.044
#> GSM1269693 1 0.9896 0.21665 0.560 0.440
#> GSM1269701 1 0.3431 0.74791 0.936 0.064
#> GSM1269709 1 0.2043 0.75866 0.968 0.032
#> GSM1269715 1 0.4431 0.74847 0.908 0.092
#> GSM1269717 1 0.4939 0.73886 0.892 0.108
#> GSM1269721 2 0.9522 0.28969 0.372 0.628
#> GSM1269723 1 0.2423 0.74780 0.960 0.040
#> GSM1269645 2 0.4022 0.72047 0.080 0.920
#> GSM1269653 2 0.7950 0.61914 0.240 0.760
#> GSM1269661 1 0.9977 -0.00822 0.528 0.472
#> GSM1269669 1 0.9358 0.29676 0.648 0.352
#> GSM1269677 2 0.2778 0.69876 0.048 0.952
#> GSM1269685 2 0.2603 0.71992 0.044 0.956
#> GSM1269691 2 0.3114 0.71987 0.056 0.944
#> GSM1269699 2 0.5629 0.70399 0.132 0.868
#> GSM1269707 2 0.3431 0.71873 0.064 0.936
#> GSM1269651 2 0.9635 0.25214 0.388 0.612
#> GSM1269659 2 0.9323 0.34338 0.348 0.652
#> GSM1269667 1 0.2043 0.76052 0.968 0.032
#> GSM1269675 1 0.7056 0.66259 0.808 0.192
#> GSM1269683 1 0.3733 0.74958 0.928 0.072
#> GSM1269689 1 0.3274 0.75405 0.940 0.060
#> GSM1269697 1 0.2043 0.75853 0.968 0.032
#> GSM1269705 1 0.5519 0.71988 0.872 0.128
#> GSM1269713 1 0.2603 0.75737 0.956 0.044
#> GSM1269719 1 1.0000 0.01894 0.504 0.496
#> GSM1269725 1 0.2043 0.75864 0.968 0.032
#> GSM1269727 1 0.2236 0.75754 0.964 0.036
#> GSM1269649 1 1.0000 -0.08604 0.500 0.500
#> GSM1269657 2 0.2236 0.70250 0.036 0.964
#> GSM1269665 2 0.9491 0.39771 0.368 0.632
#> GSM1269673 2 0.8499 0.58266 0.276 0.724
#> GSM1269681 2 0.3879 0.69358 0.076 0.924
#> GSM1269687 2 0.9248 0.47228 0.340 0.660
#> GSM1269695 2 0.9087 0.49780 0.324 0.676
#> GSM1269703 2 0.5178 0.70498 0.116 0.884
#> GSM1269711 2 0.9963 0.15933 0.464 0.536
#> GSM1269646 1 0.4562 0.73681 0.904 0.096
#> GSM1269654 1 0.6801 0.65285 0.820 0.180
#> GSM1269662 1 0.9866 0.20128 0.568 0.432
#> GSM1269670 1 0.8081 0.59119 0.752 0.248
#> GSM1269678 1 0.2948 0.75702 0.948 0.052
#> GSM1269692 2 0.9996 -0.07127 0.488 0.512
#> GSM1269700 1 0.3114 0.75202 0.944 0.056
#> GSM1269708 1 0.2603 0.75730 0.956 0.044
#> GSM1269714 1 0.3879 0.75116 0.924 0.076
#> GSM1269716 1 0.5946 0.72075 0.856 0.144
#> GSM1269720 2 0.9661 0.24891 0.392 0.608
#> GSM1269722 1 0.0672 0.75868 0.992 0.008
#> GSM1269644 2 0.2043 0.71545 0.032 0.968
#> GSM1269652 2 0.8144 0.60610 0.252 0.748
#> GSM1269660 1 0.9993 -0.04596 0.516 0.484
#> GSM1269668 1 0.9209 0.33169 0.664 0.336
#> GSM1269676 2 0.2948 0.69973 0.052 0.948
#> GSM1269684 2 0.4815 0.70306 0.104 0.896
#> GSM1269690 2 0.2423 0.71780 0.040 0.960
#> GSM1269698 2 0.4815 0.71433 0.104 0.896
#> GSM1269706 2 0.4022 0.71763 0.080 0.920
#> GSM1269650 2 0.9491 0.29845 0.368 0.632
#> GSM1269658 2 0.9286 0.35106 0.344 0.656
#> GSM1269666 1 0.2236 0.75988 0.964 0.036
#> GSM1269674 1 0.9129 0.46532 0.672 0.328
#> GSM1269682 1 0.5629 0.72435 0.868 0.132
#> GSM1269688 1 0.4161 0.74247 0.916 0.084
#> GSM1269696 1 0.2423 0.75899 0.960 0.040
#> GSM1269704 1 0.5519 0.71580 0.872 0.128
#> GSM1269712 1 0.3114 0.75466 0.944 0.056
#> GSM1269718 1 0.9491 0.35175 0.632 0.368
#> GSM1269724 1 0.2236 0.75848 0.964 0.036
#> GSM1269726 1 0.3114 0.75750 0.944 0.056
#> GSM1269648 2 0.8267 0.59824 0.260 0.740
#> GSM1269656 2 0.1184 0.71552 0.016 0.984
#> GSM1269664 1 0.9922 0.10208 0.552 0.448
#> GSM1269672 2 0.8443 0.58251 0.272 0.728
#> GSM1269680 2 0.3431 0.69602 0.064 0.936
#> GSM1269686 1 0.9933 0.08187 0.548 0.452
#> GSM1269694 2 0.8499 0.57398 0.276 0.724
#> GSM1269702 2 0.2603 0.71888 0.044 0.956
#> GSM1269710 2 0.9710 0.33773 0.400 0.600
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM1269647 1 0.686 0.5428 0.728 0.188 0.084
#> GSM1269655 1 0.405 0.5197 0.848 0.004 0.148
#> GSM1269663 1 0.654 0.3834 0.672 0.024 0.304
#> GSM1269671 1 0.855 0.3685 0.568 0.312 0.120
#> GSM1269679 1 0.468 0.5281 0.840 0.028 0.132
#> GSM1269693 1 0.867 -0.0793 0.484 0.104 0.412
#> GSM1269701 1 0.761 0.2844 0.644 0.076 0.280
#> GSM1269709 1 0.672 0.5563 0.744 0.160 0.096
#> GSM1269715 3 0.832 0.1550 0.428 0.080 0.492
#> GSM1269717 3 0.828 0.1202 0.456 0.076 0.468
#> GSM1269721 1 0.922 -0.0107 0.448 0.152 0.400
#> GSM1269723 1 0.153 0.5724 0.964 0.004 0.032
#> GSM1269645 2 0.651 0.6234 0.044 0.720 0.236
#> GSM1269653 2 0.448 0.6474 0.068 0.864 0.068
#> GSM1269661 2 0.734 0.4891 0.148 0.708 0.144
#> GSM1269669 3 0.952 0.0676 0.188 0.396 0.416
#> GSM1269677 2 0.613 0.4993 0.000 0.600 0.400
#> GSM1269685 2 0.440 0.6620 0.000 0.812 0.188
#> GSM1269691 2 0.562 0.6401 0.012 0.744 0.244
#> GSM1269699 2 0.506 0.6439 0.064 0.836 0.100
#> GSM1269707 2 0.343 0.6746 0.004 0.884 0.112
#> GSM1269651 1 0.780 0.2253 0.552 0.056 0.392
#> GSM1269659 3 0.977 0.1498 0.320 0.248 0.432
#> GSM1269667 1 0.191 0.5829 0.956 0.016 0.028
#> GSM1269675 1 0.686 0.5495 0.732 0.176 0.092
#> GSM1269683 1 0.649 0.3854 0.740 0.060 0.200
#> GSM1269689 1 0.791 0.4849 0.648 0.240 0.112
#> GSM1269697 1 0.741 0.5216 0.692 0.204 0.104
#> GSM1269705 1 0.564 0.5819 0.808 0.112 0.080
#> GSM1269713 1 0.609 0.5702 0.784 0.124 0.092
#> GSM1269719 1 0.775 0.3029 0.624 0.076 0.300
#> GSM1269725 1 0.512 0.5852 0.832 0.108 0.060
#> GSM1269727 1 0.313 0.5499 0.904 0.008 0.088
#> GSM1269649 2 0.698 0.5027 0.132 0.732 0.136
#> GSM1269657 2 0.595 0.5405 0.000 0.640 0.360
#> GSM1269665 2 0.778 0.4779 0.208 0.668 0.124
#> GSM1269673 2 0.269 0.6789 0.032 0.932 0.036
#> GSM1269681 2 0.731 0.4613 0.032 0.552 0.416
#> GSM1269687 2 0.666 0.5906 0.116 0.752 0.132
#> GSM1269695 2 0.457 0.6249 0.068 0.860 0.072
#> GSM1269703 2 0.570 0.6636 0.064 0.800 0.136
#> GSM1269711 2 0.600 0.5604 0.084 0.788 0.128
#> GSM1269646 1 0.654 0.5602 0.752 0.164 0.084
#> GSM1269654 1 0.453 0.5042 0.824 0.008 0.168
#> GSM1269662 1 0.606 0.4243 0.708 0.016 0.276
#> GSM1269670 1 0.801 0.4413 0.624 0.276 0.100
#> GSM1269678 1 0.551 0.4585 0.784 0.028 0.188
#> GSM1269692 3 0.917 0.1406 0.372 0.152 0.476
#> GSM1269700 1 0.704 0.3512 0.688 0.060 0.252
#> GSM1269708 1 0.720 0.5283 0.712 0.180 0.108
#> GSM1269714 1 0.813 -0.1975 0.488 0.068 0.444
#> GSM1269716 1 0.828 -0.2408 0.464 0.076 0.460
#> GSM1269720 1 0.910 0.0439 0.476 0.144 0.380
#> GSM1269722 1 0.368 0.5704 0.892 0.028 0.080
#> GSM1269644 2 0.559 0.6061 0.004 0.720 0.276
#> GSM1269652 2 0.389 0.6504 0.064 0.888 0.048
#> GSM1269660 2 0.815 0.4280 0.240 0.632 0.128
#> GSM1269668 3 0.970 0.0790 0.216 0.388 0.396
#> GSM1269676 2 0.634 0.4945 0.004 0.596 0.400
#> GSM1269684 2 0.614 0.6373 0.040 0.748 0.212
#> GSM1269690 2 0.548 0.6324 0.004 0.732 0.264
#> GSM1269698 2 0.588 0.6348 0.064 0.788 0.148
#> GSM1269706 2 0.437 0.6739 0.032 0.860 0.108
#> GSM1269650 1 0.817 0.1609 0.512 0.072 0.416
#> GSM1269658 3 0.982 0.1443 0.328 0.256 0.416
#> GSM1269666 1 0.165 0.5752 0.960 0.004 0.036
#> GSM1269674 1 0.688 0.5428 0.736 0.108 0.156
#> GSM1269682 1 0.619 0.3968 0.744 0.040 0.216
#> GSM1269688 1 0.832 0.4289 0.600 0.284 0.116
#> GSM1269696 1 0.649 0.5480 0.744 0.192 0.064
#> GSM1269704 1 0.603 0.5718 0.780 0.152 0.068
#> GSM1269712 1 0.459 0.5595 0.856 0.048 0.096
#> GSM1269718 1 0.671 0.4327 0.716 0.056 0.228
#> GSM1269724 1 0.397 0.5788 0.884 0.044 0.072
#> GSM1269726 1 0.462 0.5042 0.836 0.020 0.144
#> GSM1269648 2 0.406 0.6394 0.076 0.880 0.044
#> GSM1269656 2 0.568 0.5811 0.000 0.684 0.316
#> GSM1269664 2 0.912 0.2737 0.236 0.548 0.216
#> GSM1269672 2 0.357 0.6744 0.040 0.900 0.060
#> GSM1269680 2 0.658 0.4796 0.008 0.572 0.420
#> GSM1269686 2 0.849 0.3471 0.148 0.604 0.248
#> GSM1269694 2 0.379 0.6458 0.060 0.892 0.048
#> GSM1269702 2 0.406 0.6673 0.000 0.836 0.164
#> GSM1269710 2 0.518 0.6040 0.084 0.832 0.084
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM1269647 3 0.373 0.7299 0.076 0.060 0.860 0.004
#> GSM1269655 3 0.543 0.5822 0.000 0.216 0.716 0.068
#> GSM1269663 2 0.672 0.1663 0.000 0.524 0.380 0.096
#> GSM1269671 3 0.605 0.4808 0.384 0.028 0.576 0.012
#> GSM1269679 3 0.604 0.6747 0.036 0.052 0.712 0.200
#> GSM1269693 4 0.640 0.0527 0.000 0.464 0.064 0.472
#> GSM1269701 3 0.707 0.4357 0.124 0.004 0.540 0.332
#> GSM1269709 3 0.632 0.6681 0.272 0.052 0.652 0.024
#> GSM1269715 4 0.316 0.5297 0.004 0.096 0.020 0.880
#> GSM1269717 4 0.414 0.5256 0.004 0.144 0.032 0.820
#> GSM1269721 2 0.468 0.4777 0.012 0.808 0.120 0.060
#> GSM1269723 3 0.498 0.6725 0.008 0.136 0.784 0.072
#> GSM1269645 1 0.714 0.5656 0.584 0.252 0.008 0.156
#> GSM1269653 1 0.310 0.6805 0.892 0.044 0.060 0.004
#> GSM1269661 1 0.653 0.6003 0.696 0.056 0.068 0.180
#> GSM1269669 4 0.639 0.0972 0.384 0.004 0.060 0.552
#> GSM1269677 2 0.380 0.4374 0.220 0.780 0.000 0.000
#> GSM1269685 1 0.627 0.6257 0.648 0.240 0.000 0.112
#> GSM1269691 1 0.718 0.5462 0.540 0.284 0.000 0.176
#> GSM1269699 1 0.412 0.6134 0.836 0.044 0.112 0.008
#> GSM1269707 1 0.428 0.6871 0.816 0.144 0.032 0.008
#> GSM1269651 2 0.566 0.3747 0.008 0.632 0.336 0.024
#> GSM1269659 2 0.286 0.5044 0.052 0.904 0.004 0.040
#> GSM1269667 3 0.450 0.6977 0.004 0.068 0.812 0.116
#> GSM1269675 3 0.450 0.7115 0.140 0.044 0.808 0.008
#> GSM1269683 4 0.725 0.0895 0.004 0.148 0.316 0.532
#> GSM1269689 3 0.461 0.6405 0.264 0.000 0.724 0.012
#> GSM1269697 3 0.432 0.6684 0.228 0.000 0.760 0.012
#> GSM1269705 3 0.447 0.7115 0.052 0.116 0.820 0.012
#> GSM1269713 3 0.381 0.7069 0.156 0.000 0.824 0.020
#> GSM1269719 2 0.649 0.2174 0.008 0.560 0.372 0.060
#> GSM1269725 3 0.379 0.7269 0.108 0.008 0.852 0.032
#> GSM1269727 3 0.615 0.6050 0.000 0.144 0.676 0.180
#> GSM1269649 1 0.450 0.5746 0.808 0.008 0.140 0.044
#> GSM1269657 2 0.438 0.3205 0.296 0.704 0.000 0.000
#> GSM1269665 1 0.754 0.4546 0.536 0.104 0.032 0.328
#> GSM1269673 1 0.521 0.6835 0.768 0.144 0.008 0.080
#> GSM1269681 2 0.617 0.3131 0.308 0.628 0.056 0.008
#> GSM1269687 1 0.662 0.5701 0.604 0.124 0.000 0.272
#> GSM1269695 1 0.236 0.6732 0.928 0.008 0.036 0.028
#> GSM1269703 1 0.714 0.5719 0.576 0.176 0.004 0.244
#> GSM1269711 1 0.328 0.6125 0.872 0.004 0.104 0.020
#> GSM1269646 3 0.265 0.7188 0.036 0.040 0.916 0.008
#> GSM1269654 3 0.662 0.4543 0.000 0.272 0.604 0.124
#> GSM1269662 2 0.667 0.2522 0.004 0.540 0.376 0.080
#> GSM1269670 3 0.571 0.5857 0.300 0.024 0.660 0.016
#> GSM1269678 3 0.679 0.5728 0.024 0.076 0.616 0.284
#> GSM1269692 2 0.540 0.1626 0.000 0.628 0.024 0.348
#> GSM1269700 3 0.647 0.5522 0.080 0.008 0.620 0.292
#> GSM1269708 3 0.719 0.6647 0.256 0.076 0.616 0.052
#> GSM1269714 4 0.444 0.5247 0.004 0.140 0.048 0.808
#> GSM1269716 4 0.386 0.5272 0.000 0.144 0.028 0.828
#> GSM1269720 2 0.476 0.4773 0.012 0.796 0.144 0.048
#> GSM1269722 3 0.593 0.6836 0.028 0.108 0.740 0.124
#> GSM1269644 1 0.678 0.4618 0.532 0.376 0.004 0.088
#> GSM1269652 1 0.288 0.6852 0.908 0.028 0.048 0.016
#> GSM1269660 1 0.744 0.4990 0.612 0.052 0.108 0.228
#> GSM1269668 4 0.592 0.2757 0.300 0.004 0.052 0.644
#> GSM1269676 2 0.398 0.4172 0.240 0.760 0.000 0.000
#> GSM1269684 1 0.736 0.5066 0.520 0.204 0.000 0.276
#> GSM1269690 1 0.757 0.4693 0.476 0.300 0.000 0.224
#> GSM1269698 1 0.426 0.6143 0.828 0.048 0.116 0.008
#> GSM1269706 1 0.367 0.6924 0.864 0.092 0.032 0.012
#> GSM1269650 2 0.498 0.4299 0.008 0.708 0.272 0.012
#> GSM1269658 2 0.276 0.5008 0.036 0.908 0.004 0.052
#> GSM1269666 3 0.496 0.6670 0.000 0.116 0.776 0.108
#> GSM1269674 3 0.390 0.6889 0.024 0.120 0.844 0.012
#> GSM1269682 4 0.747 0.2308 0.000 0.228 0.268 0.504
#> GSM1269688 3 0.519 0.5335 0.372 0.000 0.616 0.012
#> GSM1269696 3 0.373 0.6999 0.164 0.004 0.824 0.008
#> GSM1269704 3 0.454 0.7296 0.104 0.072 0.816 0.008
#> GSM1269712 3 0.560 0.7113 0.052 0.056 0.768 0.124
#> GSM1269718 3 0.700 0.3362 0.024 0.360 0.548 0.068
#> GSM1269724 3 0.414 0.7270 0.052 0.020 0.848 0.080
#> GSM1269726 3 0.750 0.3987 0.016 0.128 0.512 0.344
#> GSM1269648 1 0.259 0.6928 0.920 0.036 0.032 0.012
#> GSM1269656 2 0.583 -0.1878 0.440 0.528 0.000 0.032
#> GSM1269664 4 0.728 -0.2215 0.444 0.048 0.048 0.460
#> GSM1269672 1 0.599 0.6529 0.692 0.156 0.000 0.152
#> GSM1269680 2 0.514 0.3393 0.296 0.680 0.024 0.000
#> GSM1269686 4 0.639 -0.1447 0.404 0.068 0.000 0.528
#> GSM1269694 1 0.263 0.6938 0.920 0.036 0.024 0.020
#> GSM1269702 1 0.555 0.6225 0.680 0.268 0.000 0.052
#> GSM1269710 1 0.310 0.6377 0.888 0.008 0.084 0.020
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM1269647 3 0.351 0.3713 0.016 0.020 0.832 0.000 0.132
#> GSM1269655 3 0.650 0.4129 0.008 0.208 0.632 0.084 0.068
#> GSM1269663 2 0.632 0.3744 0.000 0.636 0.200 0.068 0.096
#> GSM1269671 3 0.708 -0.5373 0.184 0.028 0.444 0.000 0.344
#> GSM1269679 3 0.543 0.4259 0.004 0.024 0.692 0.216 0.064
#> GSM1269693 2 0.766 0.0716 0.008 0.448 0.052 0.292 0.200
#> GSM1269701 3 0.707 0.0318 0.040 0.000 0.492 0.300 0.168
#> GSM1269709 3 0.658 -0.0560 0.052 0.028 0.568 0.036 0.316
#> GSM1269715 4 0.708 0.3840 0.012 0.164 0.068 0.592 0.164
#> GSM1269717 4 0.401 0.4967 0.004 0.076 0.040 0.832 0.048
#> GSM1269721 2 0.697 0.3289 0.008 0.516 0.100 0.048 0.328
#> GSM1269723 3 0.561 0.4559 0.000 0.096 0.716 0.072 0.116
#> GSM1269645 1 0.687 0.4077 0.588 0.196 0.000 0.136 0.080
#> GSM1269653 1 0.548 0.4284 0.616 0.008 0.068 0.000 0.308
#> GSM1269661 1 0.575 0.4220 0.664 0.004 0.052 0.236 0.044
#> GSM1269669 4 0.495 0.3022 0.264 0.000 0.016 0.684 0.036
#> GSM1269677 2 0.661 0.3043 0.280 0.508 0.000 0.008 0.204
#> GSM1269685 1 0.641 0.4808 0.624 0.096 0.000 0.068 0.212
#> GSM1269691 1 0.711 0.4332 0.576 0.128 0.000 0.164 0.132
#> GSM1269699 1 0.568 0.3065 0.564 0.004 0.080 0.000 0.352
#> GSM1269707 1 0.563 0.3946 0.504 0.028 0.020 0.004 0.444
#> GSM1269651 2 0.536 0.3841 0.016 0.664 0.256 0.000 0.064
#> GSM1269659 2 0.491 0.4598 0.056 0.736 0.004 0.016 0.188
#> GSM1269667 3 0.575 0.4344 0.028 0.036 0.720 0.144 0.072
#> GSM1269675 3 0.640 0.1609 0.060 0.080 0.620 0.004 0.236
#> GSM1269683 4 0.681 0.3841 0.012 0.112 0.208 0.608 0.060
#> GSM1269689 3 0.610 -0.5373 0.100 0.008 0.560 0.004 0.328
#> GSM1269697 3 0.369 0.2861 0.036 0.000 0.812 0.004 0.148
#> GSM1269705 3 0.391 0.4511 0.004 0.056 0.820 0.008 0.112
#> GSM1269713 3 0.365 0.3014 0.016 0.004 0.820 0.012 0.148
#> GSM1269719 2 0.651 0.3700 0.064 0.620 0.248 0.028 0.040
#> GSM1269725 3 0.286 0.4171 0.004 0.004 0.880 0.024 0.088
#> GSM1269727 3 0.845 0.2113 0.008 0.188 0.404 0.176 0.224
#> GSM1269649 1 0.606 0.4728 0.652 0.012 0.076 0.032 0.228
#> GSM1269657 2 0.655 0.1883 0.368 0.452 0.000 0.004 0.176
#> GSM1269665 1 0.718 0.1357 0.472 0.092 0.008 0.364 0.064
#> GSM1269673 1 0.404 0.5388 0.808 0.028 0.000 0.132 0.032
#> GSM1269681 2 0.680 0.3032 0.316 0.524 0.048 0.000 0.112
#> GSM1269687 1 0.510 0.3788 0.636 0.024 0.000 0.320 0.020
#> GSM1269695 1 0.391 0.5495 0.772 0.000 0.008 0.016 0.204
#> GSM1269703 1 0.553 0.4101 0.652 0.076 0.000 0.256 0.016
#> GSM1269711 1 0.531 0.3705 0.624 0.000 0.064 0.004 0.308
#> GSM1269646 3 0.327 0.4169 0.004 0.036 0.848 0.000 0.112
#> GSM1269654 3 0.663 0.3912 0.004 0.248 0.596 0.080 0.072
#> GSM1269662 2 0.578 0.4303 0.004 0.680 0.200 0.040 0.076
#> GSM1269670 3 0.683 -0.3891 0.164 0.028 0.512 0.000 0.296
#> GSM1269678 3 0.611 0.4006 0.004 0.036 0.640 0.228 0.092
#> GSM1269692 2 0.728 0.2435 0.028 0.548 0.032 0.232 0.160
#> GSM1269700 3 0.650 0.1965 0.028 0.000 0.560 0.284 0.128
#> GSM1269708 3 0.709 -0.0294 0.040 0.052 0.524 0.052 0.332
#> GSM1269714 4 0.762 0.3112 0.000 0.180 0.112 0.500 0.208
#> GSM1269716 4 0.578 0.4639 0.008 0.136 0.068 0.712 0.076
#> GSM1269720 2 0.656 0.3658 0.008 0.548 0.112 0.020 0.312
#> GSM1269722 3 0.662 0.4108 0.016 0.116 0.656 0.088 0.124
#> GSM1269644 1 0.647 0.3957 0.600 0.224 0.000 0.136 0.040
#> GSM1269652 1 0.556 0.4159 0.568 0.004 0.056 0.004 0.368
#> GSM1269660 1 0.755 0.2254 0.516 0.040 0.064 0.296 0.084
#> GSM1269668 4 0.542 0.2858 0.268 0.000 0.024 0.656 0.052
#> GSM1269676 2 0.639 0.2995 0.304 0.500 0.000 0.000 0.196
#> GSM1269684 1 0.681 0.2704 0.488 0.064 0.000 0.368 0.080
#> GSM1269690 1 0.762 0.3521 0.508 0.140 0.000 0.204 0.148
#> GSM1269698 1 0.548 0.3488 0.580 0.004 0.064 0.000 0.352
#> GSM1269706 1 0.564 0.3602 0.488 0.028 0.028 0.000 0.456
#> GSM1269650 2 0.529 0.4116 0.020 0.680 0.240 0.000 0.060
#> GSM1269658 2 0.394 0.4780 0.052 0.812 0.000 0.012 0.124
#> GSM1269666 3 0.538 0.4707 0.004 0.052 0.732 0.144 0.068
#> GSM1269674 3 0.621 0.3151 0.024 0.140 0.632 0.004 0.200
#> GSM1269682 4 0.677 0.3103 0.000 0.200 0.140 0.592 0.068
#> GSM1269688 5 0.675 0.0000 0.156 0.016 0.404 0.000 0.424
#> GSM1269696 3 0.356 0.3578 0.020 0.012 0.832 0.004 0.132
#> GSM1269704 3 0.400 0.4036 0.020 0.028 0.800 0.000 0.152
#> GSM1269712 3 0.481 0.4341 0.004 0.020 0.768 0.104 0.104
#> GSM1269718 3 0.817 0.0403 0.068 0.384 0.392 0.084 0.072
#> GSM1269724 3 0.378 0.4585 0.004 0.012 0.836 0.064 0.084
#> GSM1269726 3 0.896 0.0983 0.024 0.180 0.332 0.236 0.228
#> GSM1269648 1 0.322 0.5767 0.852 0.004 0.036 0.000 0.108
#> GSM1269656 1 0.692 0.1011 0.476 0.280 0.000 0.016 0.228
#> GSM1269664 4 0.617 -0.0491 0.420 0.040 0.012 0.500 0.028
#> GSM1269672 1 0.460 0.5280 0.764 0.028 0.000 0.164 0.044
#> GSM1269680 2 0.590 0.3413 0.304 0.584 0.008 0.000 0.104
#> GSM1269686 4 0.531 -0.1351 0.452 0.012 0.004 0.512 0.020
#> GSM1269694 1 0.348 0.5699 0.812 0.000 0.008 0.012 0.168
#> GSM1269702 1 0.489 0.5358 0.768 0.088 0.000 0.048 0.096
#> GSM1269710 1 0.468 0.4663 0.696 0.000 0.040 0.004 0.260
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM1269647 3 0.434 0.4908 0.028 0.228 0.720 0.016 0.000 0.008
#> GSM1269655 3 0.724 0.1988 0.004 0.244 0.516 0.100 0.048 0.088
#> GSM1269663 4 0.779 0.1948 0.016 0.240 0.116 0.456 0.028 0.144
#> GSM1269671 2 0.685 0.2350 0.308 0.408 0.244 0.032 0.004 0.004
#> GSM1269679 3 0.496 0.5604 0.020 0.040 0.740 0.088 0.112 0.000
#> GSM1269693 4 0.576 0.5199 0.000 0.012 0.052 0.652 0.176 0.108
#> GSM1269701 3 0.702 0.3936 0.120 0.092 0.560 0.052 0.176 0.000
#> GSM1269709 3 0.639 0.4860 0.068 0.132 0.648 0.104 0.020 0.028
#> GSM1269715 4 0.584 0.1511 0.008 0.008 0.076 0.452 0.444 0.012
#> GSM1269717 5 0.554 0.2109 0.000 0.032 0.116 0.156 0.676 0.020
#> GSM1269721 4 0.524 0.4903 0.012 0.044 0.080 0.700 0.000 0.164
#> GSM1269723 3 0.473 0.4976 0.004 0.076 0.700 0.208 0.012 0.000
#> GSM1269645 1 0.851 -0.1033 0.300 0.212 0.000 0.084 0.264 0.140
#> GSM1269653 1 0.724 0.3282 0.520 0.132 0.096 0.012 0.020 0.220
#> GSM1269661 1 0.828 -0.0388 0.372 0.080 0.088 0.012 0.296 0.152
#> GSM1269669 5 0.529 0.3564 0.272 0.024 0.040 0.012 0.644 0.008
#> GSM1269677 6 0.222 0.4955 0.016 0.024 0.000 0.052 0.000 0.908
#> GSM1269685 6 0.741 0.0888 0.336 0.056 0.000 0.064 0.124 0.420
#> GSM1269691 6 0.769 0.0694 0.312 0.020 0.000 0.104 0.220 0.344
#> GSM1269699 1 0.472 0.4931 0.748 0.152 0.040 0.004 0.012 0.044
#> GSM1269707 1 0.697 0.3589 0.564 0.156 0.024 0.088 0.008 0.160
#> GSM1269651 2 0.748 0.0862 0.000 0.384 0.132 0.264 0.004 0.216
#> GSM1269659 4 0.481 0.3919 0.008 0.048 0.000 0.592 0.000 0.352
#> GSM1269667 3 0.632 0.4205 0.036 0.172 0.588 0.024 0.180 0.000
#> GSM1269675 2 0.740 0.1891 0.160 0.376 0.316 0.144 0.000 0.004
#> GSM1269683 5 0.609 0.0089 0.000 0.028 0.172 0.268 0.532 0.000
#> GSM1269689 3 0.706 0.1535 0.232 0.244 0.448 0.068 0.008 0.000
#> GSM1269697 3 0.380 0.5257 0.068 0.136 0.788 0.000 0.008 0.000
#> GSM1269705 3 0.554 0.4642 0.036 0.176 0.660 0.120 0.008 0.000
#> GSM1269713 3 0.359 0.5647 0.044 0.104 0.828 0.008 0.012 0.004
#> GSM1269719 2 0.839 0.1029 0.024 0.352 0.168 0.160 0.028 0.268
#> GSM1269725 3 0.272 0.5779 0.024 0.056 0.888 0.020 0.012 0.000
#> GSM1269727 4 0.723 0.3031 0.020 0.168 0.192 0.512 0.104 0.004
#> GSM1269649 1 0.578 0.3993 0.664 0.176 0.016 0.008 0.096 0.040
#> GSM1269657 6 0.279 0.5214 0.040 0.036 0.000 0.036 0.004 0.884
#> GSM1269665 5 0.733 0.2630 0.248 0.128 0.000 0.020 0.468 0.136
#> GSM1269673 1 0.700 0.0437 0.416 0.056 0.004 0.008 0.348 0.168
#> GSM1269681 6 0.650 0.2384 0.080 0.300 0.008 0.064 0.012 0.536
#> GSM1269687 5 0.663 0.1684 0.376 0.048 0.012 0.004 0.452 0.108
#> GSM1269695 1 0.486 0.4665 0.744 0.112 0.008 0.004 0.096 0.036
#> GSM1269703 5 0.714 0.1427 0.328 0.056 0.000 0.024 0.424 0.168
#> GSM1269711 1 0.412 0.4958 0.816 0.072 0.056 0.024 0.020 0.012
#> GSM1269646 3 0.515 0.4502 0.032 0.244 0.672 0.024 0.004 0.024
#> GSM1269654 3 0.791 0.1235 0.004 0.232 0.456 0.124 0.092 0.092
#> GSM1269662 4 0.792 0.0303 0.004 0.292 0.104 0.376 0.036 0.188
#> GSM1269670 2 0.660 0.2273 0.264 0.428 0.280 0.024 0.000 0.004
#> GSM1269678 3 0.463 0.5461 0.000 0.040 0.744 0.116 0.100 0.000
#> GSM1269692 4 0.584 0.5060 0.000 0.012 0.016 0.604 0.188 0.180
#> GSM1269700 3 0.683 0.4401 0.088 0.076 0.588 0.080 0.168 0.000
#> GSM1269708 3 0.650 0.4604 0.068 0.104 0.616 0.176 0.024 0.012
#> GSM1269714 4 0.618 0.3118 0.000 0.012 0.160 0.504 0.312 0.012
#> GSM1269716 5 0.636 0.0569 0.000 0.024 0.136 0.232 0.572 0.036
#> GSM1269720 4 0.542 0.4846 0.008 0.048 0.076 0.668 0.000 0.200
#> GSM1269722 3 0.559 0.4142 0.012 0.060 0.596 0.300 0.032 0.000
#> GSM1269644 1 0.798 -0.1016 0.292 0.104 0.000 0.036 0.280 0.288
#> GSM1269652 1 0.771 0.2616 0.456 0.168 0.100 0.012 0.028 0.236
#> GSM1269660 5 0.862 0.1122 0.252 0.084 0.092 0.016 0.324 0.232
#> GSM1269668 5 0.538 0.3586 0.280 0.020 0.060 0.016 0.624 0.000
#> GSM1269676 6 0.235 0.5177 0.036 0.028 0.000 0.032 0.000 0.904
#> GSM1269684 5 0.620 0.3283 0.180 0.032 0.000 0.024 0.600 0.164
#> GSM1269690 6 0.747 0.1006 0.192 0.020 0.000 0.084 0.340 0.364
#> GSM1269698 1 0.545 0.4854 0.704 0.140 0.052 0.008 0.012 0.084
#> GSM1269706 1 0.730 0.3414 0.540 0.156 0.044 0.120 0.004 0.136
#> GSM1269650 2 0.761 0.0704 0.000 0.352 0.124 0.240 0.008 0.276
#> GSM1269658 4 0.529 0.3954 0.008 0.072 0.000 0.580 0.008 0.332
#> GSM1269666 3 0.562 0.4891 0.004 0.168 0.656 0.052 0.120 0.000
#> GSM1269674 2 0.713 0.1095 0.088 0.376 0.356 0.176 0.000 0.004
#> GSM1269682 5 0.674 0.0214 0.004 0.104 0.088 0.256 0.536 0.012
#> GSM1269688 3 0.734 0.0132 0.320 0.220 0.344 0.116 0.000 0.000
#> GSM1269696 3 0.409 0.4888 0.052 0.212 0.732 0.004 0.000 0.000
#> GSM1269704 3 0.574 0.4452 0.036 0.180 0.648 0.124 0.004 0.008
#> GSM1269712 3 0.422 0.5760 0.024 0.048 0.804 0.060 0.064 0.000
#> GSM1269718 3 0.866 -0.1528 0.020 0.288 0.304 0.084 0.088 0.216
#> GSM1269724 3 0.398 0.5767 0.016 0.084 0.812 0.060 0.028 0.000
#> GSM1269726 4 0.732 0.3289 0.032 0.132 0.192 0.520 0.120 0.004
#> GSM1269648 1 0.624 0.4210 0.636 0.088 0.020 0.008 0.076 0.172
#> GSM1269656 6 0.437 0.4856 0.112 0.044 0.000 0.036 0.024 0.784
#> GSM1269664 5 0.701 0.3292 0.236 0.096 0.016 0.016 0.540 0.096
#> GSM1269672 1 0.675 0.0625 0.412 0.020 0.000 0.024 0.364 0.180
#> GSM1269680 6 0.585 0.3512 0.060 0.220 0.000 0.072 0.016 0.632
#> GSM1269686 5 0.618 0.3356 0.304 0.032 0.028 0.020 0.572 0.044
#> GSM1269694 1 0.484 0.4596 0.732 0.136 0.000 0.004 0.084 0.044
#> GSM1269702 6 0.668 -0.0440 0.392 0.040 0.000 0.020 0.124 0.424
#> GSM1269710 1 0.345 0.4930 0.844 0.072 0.032 0.008 0.044 0.000
Heatmaps for the consensus matrix. It visualizes the probability of two samples to be in a same group.
consensus_heatmap(res, k = 2)
consensus_heatmap(res, k = 3)
consensus_heatmap(res, k = 4)
consensus_heatmap(res, k = 5)
consensus_heatmap(res, k = 6)
Heatmaps for the membership of samples in all partitions to see how consistent they are:
membership_heatmap(res, k = 2)
membership_heatmap(res, k = 3)
membership_heatmap(res, k = 4)
membership_heatmap(res, k = 5)
membership_heatmap(res, k = 6)
As soon as we have had the classes for columns, we can look for signatures which are significantly different between classes which can be candidate marks for certain classes. Following are the heatmaps for signatures.
Signature heatmaps where rows are scaled:
get_signatures(res, k = 2)
get_signatures(res, k = 3)
get_signatures(res, k = 4)
get_signatures(res, k = 5)
get_signatures(res, k = 6)
Signature heatmaps where rows are not scaled:
get_signatures(res, k = 2, scale_rows = FALSE)
get_signatures(res, k = 3, scale_rows = FALSE)
get_signatures(res, k = 4, scale_rows = FALSE)
get_signatures(res, k = 5, scale_rows = FALSE)
get_signatures(res, k = 6, scale_rows = FALSE)
Compare the overlap of signatures from different k:
compare_signatures(res)
get_signature() returns a data frame invisibly. TO get the list of signatures, the function
call should be assigned to a variable explicitly. In following code, if plot argument is set
to FALSE, no heatmap is plotted while only the differential analysis is performed.
# code only for demonstration
tb = get_signature(res, k = ..., plot = FALSE)
An example of the output of tb is:
#> which_row fdr mean_1 mean_2 scaled_mean_1 scaled_mean_2 km
#> 1 38 0.042760348 8.373488 9.131774 -0.5533452 0.5164555 1
#> 2 40 0.018707592 7.106213 8.469186 -0.6173731 0.5762149 1
#> 3 55 0.019134737 10.221463 11.207825 -0.6159697 0.5749050 1
#> 4 59 0.006059896 5.921854 7.869574 -0.6899429 0.6439467 1
#> 5 60 0.018055526 8.928898 10.211722 -0.6204761 0.5791110 1
#> 6 98 0.009384629 15.714769 14.887706 0.6635654 -0.6193277 2
...
The columns in tb are:
which_row: row indices corresponding to the input matrix.fdr: FDR for the differential test. mean_x: The mean value in group x.scaled_mean_x: The mean value in group x after rows are scaled.km: Row groups if k-means clustering is applied to rows.UMAP plot which shows how samples are separated.
dimension_reduction(res, k = 2, method = "UMAP")
dimension_reduction(res, k = 3, method = "UMAP")
dimension_reduction(res, k = 4, method = "UMAP")
dimension_reduction(res, k = 5, method = "UMAP")
dimension_reduction(res, k = 6, method = "UMAP")
Following heatmap shows how subgroups are split when increasing k:
collect_classes(res)
Test correlation between subgroups and known annotations. If the known annotation is numeric, one-way ANOVA test is applied, and if the known annotation is discrete, chi-squared contingency table test is applied.
test_to_known_factors(res)
#> n agent(p) disease.state(p) gender(p) individual(p) k
#> SD:NMF 58 0.963 0.7193 2.06e-13 0.001597 2
#> SD:NMF 48 1.000 0.7432 3.16e-11 0.002524 3
#> SD:NMF 54 0.933 0.0877 1.12e-11 0.000005 4
#> SD:NMF 6 NA NA NA NA 5
#> SD:NMF 11 0.974 0.1997 4.09e-03 0.037520 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 51941 rows and 84 columns.
#> Top rows (1000, 2000, 3000, 4000, 5000) are extracted by 'CV' method.
#> Subgroups are detected by 'hclust' method.
#> Performed in total 1250 partitions by row resampling.
#> Best k for subgroups seems to be 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 0.0409 0.0814 0.643 0.3478 0.719 0.719
#> 3 3 0.0108 0.6686 0.670 0.4445 0.518 0.430
#> 4 4 0.0419 0.4726 0.634 0.2464 0.920 0.836
#> 5 5 0.1039 0.4755 0.615 0.1105 0.894 0.747
#> 6 6 0.2321 0.4251 0.593 0.0718 0.909 0.733
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
#> GSM1269647 1 0.995 -0.457 0.540 0.460
#> GSM1269655 1 0.994 -0.523 0.544 0.456
#> GSM1269663 2 0.990 0.587 0.440 0.560
#> GSM1269671 2 1.000 0.246 0.496 0.504
#> GSM1269679 2 0.998 0.611 0.472 0.528
#> GSM1269693 1 1.000 -0.603 0.500 0.500
#> GSM1269701 1 0.996 -0.525 0.536 0.464
#> GSM1269709 1 1.000 -0.601 0.508 0.492
#> GSM1269715 2 0.722 0.216 0.200 0.800
#> GSM1269717 2 0.999 0.616 0.484 0.516
#> GSM1269721 1 0.991 -0.350 0.556 0.444
#> GSM1269723 1 0.998 -0.561 0.528 0.472
#> GSM1269645 1 0.697 0.413 0.812 0.188
#> GSM1269653 1 0.552 0.445 0.872 0.128
#> GSM1269661 1 0.680 0.396 0.820 0.180
#> GSM1269669 1 0.469 0.451 0.900 0.100
#> GSM1269677 1 0.730 0.392 0.796 0.204
#> GSM1269685 1 0.430 0.458 0.912 0.088
#> GSM1269691 1 0.541 0.446 0.876 0.124
#> GSM1269699 1 0.625 0.418 0.844 0.156
#> GSM1269707 1 0.644 0.423 0.836 0.164
#> GSM1269651 1 0.996 -0.284 0.536 0.464
#> GSM1269659 1 0.994 -0.408 0.544 0.456
#> GSM1269667 1 0.998 -0.565 0.524 0.476
#> GSM1269675 2 0.999 0.378 0.484 0.516
#> GSM1269683 2 1.000 0.569 0.500 0.500
#> GSM1269689 1 0.998 -0.430 0.524 0.476
#> GSM1269697 1 0.998 -0.417 0.524 0.476
#> GSM1269705 1 1.000 -0.392 0.512 0.488
#> GSM1269713 2 1.000 0.603 0.492 0.508
#> GSM1269719 1 0.994 -0.487 0.544 0.456
#> GSM1269725 1 1.000 -0.576 0.500 0.500
#> GSM1269727 2 1.000 0.605 0.488 0.512
#> GSM1269649 1 0.456 0.456 0.904 0.096
#> GSM1269657 1 0.689 0.419 0.816 0.184
#> GSM1269665 1 0.745 0.404 0.788 0.212
#> GSM1269673 1 0.456 0.453 0.904 0.096
#> GSM1269681 1 0.738 0.380 0.792 0.208
#> GSM1269687 1 0.430 0.458 0.912 0.088
#> GSM1269695 1 0.343 0.457 0.936 0.064
#> GSM1269703 1 0.574 0.451 0.864 0.136
#> GSM1269711 1 0.456 0.455 0.904 0.096
#> GSM1269646 1 0.995 -0.454 0.540 0.460
#> GSM1269654 1 0.994 -0.523 0.544 0.456
#> GSM1269662 2 0.990 0.547 0.440 0.560
#> GSM1269670 1 1.000 -0.289 0.504 0.496
#> GSM1269678 1 0.999 -0.584 0.516 0.484
#> GSM1269692 1 0.961 -0.220 0.616 0.384
#> GSM1269700 1 0.998 -0.544 0.528 0.472
#> GSM1269708 1 1.000 -0.601 0.508 0.492
#> GSM1269714 1 0.998 -0.601 0.524 0.476
#> GSM1269716 2 0.999 0.615 0.484 0.516
#> GSM1269720 1 0.991 -0.338 0.556 0.444
#> GSM1269722 1 0.998 -0.576 0.528 0.472
#> GSM1269644 1 0.506 0.448 0.888 0.112
#> GSM1269652 1 0.373 0.461 0.928 0.072
#> GSM1269660 1 0.644 0.421 0.836 0.164
#> GSM1269668 1 0.518 0.450 0.884 0.116
#> GSM1269676 1 0.730 0.392 0.796 0.204
#> GSM1269684 1 0.506 0.441 0.888 0.112
#> GSM1269690 1 0.541 0.446 0.876 0.124
#> GSM1269698 1 0.680 0.414 0.820 0.180
#> GSM1269706 1 0.644 0.423 0.836 0.164
#> GSM1269650 1 0.980 -0.248 0.584 0.416
#> GSM1269658 1 0.994 -0.408 0.544 0.456
#> GSM1269666 1 0.998 -0.562 0.528 0.472
#> GSM1269674 2 1.000 0.379 0.488 0.512
#> GSM1269682 1 1.000 -0.606 0.504 0.496
#> GSM1269688 1 1.000 -0.473 0.508 0.492
#> GSM1269696 1 0.998 -0.447 0.524 0.476
#> GSM1269704 1 0.999 -0.446 0.520 0.480
#> GSM1269712 2 0.998 0.620 0.472 0.528
#> GSM1269718 1 0.981 -0.418 0.580 0.420
#> GSM1269724 1 0.999 -0.542 0.516 0.484
#> GSM1269726 2 1.000 0.601 0.496 0.504
#> GSM1269648 1 0.416 0.457 0.916 0.084
#> GSM1269656 1 0.574 0.443 0.864 0.136
#> GSM1269664 1 0.529 0.450 0.880 0.120
#> GSM1269672 1 0.443 0.453 0.908 0.092
#> GSM1269680 1 0.680 0.404 0.820 0.180
#> GSM1269686 1 0.416 0.457 0.916 0.084
#> GSM1269694 1 0.343 0.457 0.936 0.064
#> GSM1269702 1 0.443 0.457 0.908 0.092
#> GSM1269710 1 0.402 0.453 0.920 0.080
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM1269647 1 0.673 0.685 0.740 0.172 0.088
#> GSM1269655 1 0.471 0.728 0.848 0.108 0.044
#> GSM1269663 1 0.768 0.618 0.680 0.188 0.132
#> GSM1269671 1 0.944 0.447 0.480 0.324 0.196
#> GSM1269679 1 0.388 0.729 0.888 0.068 0.044
#> GSM1269693 1 0.544 0.713 0.812 0.132 0.056
#> GSM1269701 1 0.503 0.719 0.828 0.132 0.040
#> GSM1269709 1 0.417 0.725 0.868 0.104 0.028
#> GSM1269715 3 0.527 0.000 0.200 0.016 0.784
#> GSM1269717 1 0.491 0.699 0.844 0.088 0.068
#> GSM1269721 1 0.856 0.527 0.572 0.304 0.124
#> GSM1269723 1 0.449 0.718 0.856 0.108 0.036
#> GSM1269645 2 0.578 0.670 0.160 0.788 0.052
#> GSM1269653 2 0.667 0.740 0.224 0.720 0.056
#> GSM1269661 2 0.660 0.653 0.428 0.564 0.008
#> GSM1269669 2 0.676 0.761 0.288 0.676 0.036
#> GSM1269677 2 0.789 0.493 0.372 0.564 0.064
#> GSM1269685 2 0.651 0.776 0.300 0.676 0.024
#> GSM1269691 2 0.636 0.739 0.364 0.628 0.008
#> GSM1269699 2 0.771 0.651 0.256 0.652 0.092
#> GSM1269707 2 0.788 0.638 0.268 0.636 0.096
#> GSM1269651 1 0.894 0.534 0.548 0.292 0.160
#> GSM1269659 1 0.837 0.586 0.608 0.260 0.132
#> GSM1269667 1 0.570 0.723 0.796 0.148 0.056
#> GSM1269675 1 0.902 0.541 0.548 0.276 0.176
#> GSM1269683 1 0.572 0.714 0.800 0.132 0.068
#> GSM1269689 1 0.760 0.639 0.668 0.236 0.096
#> GSM1269697 1 0.747 0.645 0.684 0.216 0.100
#> GSM1269705 1 0.844 0.602 0.612 0.236 0.152
#> GSM1269713 1 0.491 0.725 0.844 0.088 0.068
#> GSM1269719 1 0.635 0.684 0.744 0.204 0.052
#> GSM1269725 1 0.448 0.733 0.860 0.096 0.044
#> GSM1269727 1 0.367 0.718 0.896 0.064 0.040
#> GSM1269649 2 0.680 0.768 0.308 0.660 0.032
#> GSM1269657 2 0.751 0.622 0.344 0.604 0.052
#> GSM1269665 2 0.673 0.674 0.260 0.696 0.044
#> GSM1269673 2 0.694 0.769 0.312 0.652 0.036
#> GSM1269681 2 0.742 0.555 0.172 0.700 0.128
#> GSM1269687 2 0.642 0.766 0.324 0.660 0.016
#> GSM1269695 2 0.590 0.778 0.292 0.700 0.008
#> GSM1269703 2 0.640 0.761 0.344 0.644 0.012
#> GSM1269711 2 0.659 0.771 0.280 0.688 0.032
#> GSM1269646 1 0.681 0.683 0.736 0.172 0.092
#> GSM1269654 1 0.471 0.728 0.848 0.108 0.044
#> GSM1269662 1 0.845 0.602 0.616 0.220 0.164
#> GSM1269670 1 0.940 0.443 0.480 0.332 0.188
#> GSM1269678 1 0.448 0.722 0.860 0.096 0.044
#> GSM1269692 1 0.777 0.274 0.592 0.344 0.064
#> GSM1269700 1 0.471 0.722 0.844 0.120 0.036
#> GSM1269708 1 0.441 0.723 0.860 0.104 0.036
#> GSM1269714 1 0.489 0.714 0.844 0.096 0.060
#> GSM1269716 1 0.500 0.695 0.840 0.088 0.072
#> GSM1269720 1 0.865 0.527 0.568 0.300 0.132
#> GSM1269722 1 0.353 0.727 0.892 0.092 0.016
#> GSM1269644 2 0.660 0.762 0.296 0.676 0.028
#> GSM1269652 2 0.703 0.780 0.296 0.660 0.044
#> GSM1269660 2 0.666 0.694 0.400 0.588 0.012
#> GSM1269668 2 0.704 0.756 0.312 0.648 0.040
#> GSM1269676 2 0.789 0.493 0.372 0.564 0.064
#> GSM1269684 2 0.704 0.761 0.348 0.620 0.032
#> GSM1269690 2 0.636 0.739 0.364 0.628 0.008
#> GSM1269698 2 0.797 0.593 0.280 0.624 0.096
#> GSM1269706 2 0.788 0.638 0.268 0.636 0.096
#> GSM1269650 1 0.808 0.579 0.608 0.296 0.096
#> GSM1269658 1 0.834 0.593 0.612 0.256 0.132
#> GSM1269666 1 0.471 0.728 0.844 0.120 0.036
#> GSM1269674 1 0.900 0.542 0.548 0.280 0.172
#> GSM1269682 1 0.514 0.711 0.824 0.132 0.044
#> GSM1269688 1 0.711 0.664 0.696 0.232 0.072
#> GSM1269696 1 0.706 0.668 0.716 0.192 0.092
#> GSM1269704 1 0.787 0.652 0.664 0.200 0.136
#> GSM1269712 1 0.518 0.714 0.832 0.084 0.084
#> GSM1269718 1 0.603 0.671 0.752 0.212 0.036
#> GSM1269724 1 0.572 0.729 0.800 0.132 0.068
#> GSM1269726 1 0.376 0.722 0.892 0.068 0.040
#> GSM1269648 2 0.672 0.769 0.312 0.660 0.028
#> GSM1269656 2 0.709 0.724 0.268 0.676 0.056
#> GSM1269664 2 0.697 0.749 0.356 0.616 0.028
#> GSM1269672 2 0.671 0.768 0.296 0.672 0.032
#> GSM1269680 2 0.794 0.590 0.236 0.648 0.116
#> GSM1269686 2 0.628 0.765 0.324 0.664 0.012
#> GSM1269694 2 0.590 0.778 0.292 0.700 0.008
#> GSM1269702 2 0.610 0.767 0.320 0.672 0.008
#> GSM1269710 2 0.642 0.767 0.260 0.708 0.032
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM1269647 3 0.736 0.1041 0.108 0.312 0.556 0.024
#> GSM1269655 3 0.681 0.4670 0.120 0.156 0.680 0.044
#> GSM1269663 3 0.802 0.1972 0.080 0.248 0.564 0.108
#> GSM1269671 2 0.667 0.6591 0.100 0.652 0.228 0.020
#> GSM1269679 3 0.438 0.5765 0.092 0.044 0.836 0.028
#> GSM1269693 3 0.653 0.4913 0.112 0.116 0.712 0.060
#> GSM1269701 3 0.538 0.5621 0.140 0.072 0.768 0.020
#> GSM1269709 3 0.419 0.5816 0.108 0.044 0.836 0.012
#> GSM1269715 4 0.338 0.0000 0.000 0.012 0.140 0.848
#> GSM1269717 3 0.506 0.5601 0.084 0.040 0.804 0.072
#> GSM1269721 3 0.868 -0.1747 0.188 0.372 0.388 0.052
#> GSM1269723 3 0.496 0.5792 0.116 0.072 0.796 0.016
#> GSM1269645 1 0.626 0.5882 0.716 0.168 0.068 0.048
#> GSM1269653 1 0.633 0.6396 0.712 0.140 0.116 0.032
#> GSM1269661 1 0.653 0.6026 0.620 0.076 0.292 0.012
#> GSM1269669 1 0.521 0.7020 0.768 0.032 0.168 0.032
#> GSM1269677 1 0.882 0.2350 0.440 0.288 0.208 0.064
#> GSM1269685 1 0.544 0.7189 0.756 0.052 0.168 0.024
#> GSM1269691 1 0.595 0.6616 0.672 0.052 0.264 0.012
#> GSM1269699 1 0.790 0.4730 0.536 0.276 0.152 0.036
#> GSM1269707 1 0.798 0.4798 0.540 0.256 0.164 0.040
#> GSM1269651 2 0.821 0.5078 0.132 0.484 0.332 0.052
#> GSM1269659 3 0.911 0.0727 0.192 0.260 0.444 0.104
#> GSM1269667 3 0.657 0.5368 0.144 0.112 0.700 0.044
#> GSM1269675 2 0.808 0.6159 0.128 0.500 0.324 0.048
#> GSM1269683 3 0.588 0.5524 0.124 0.076 0.752 0.048
#> GSM1269689 3 0.792 0.0936 0.168 0.268 0.532 0.032
#> GSM1269697 3 0.719 -0.1275 0.108 0.348 0.532 0.012
#> GSM1269705 2 0.721 0.4584 0.100 0.488 0.400 0.012
#> GSM1269713 3 0.564 0.5702 0.100 0.060 0.772 0.068
#> GSM1269719 3 0.695 0.4284 0.176 0.136 0.656 0.032
#> GSM1269725 3 0.599 0.4978 0.092 0.156 0.728 0.024
#> GSM1269727 3 0.361 0.5845 0.080 0.028 0.872 0.020
#> GSM1269649 1 0.592 0.7039 0.720 0.048 0.196 0.036
#> GSM1269657 1 0.812 0.4756 0.520 0.244 0.200 0.036
#> GSM1269665 1 0.712 0.6106 0.652 0.120 0.180 0.048
#> GSM1269673 1 0.615 0.7119 0.708 0.060 0.196 0.036
#> GSM1269681 1 0.728 0.1943 0.464 0.440 0.048 0.048
#> GSM1269687 1 0.524 0.7110 0.744 0.048 0.200 0.008
#> GSM1269695 1 0.566 0.7202 0.740 0.076 0.168 0.016
#> GSM1269703 1 0.569 0.6985 0.696 0.052 0.244 0.008
#> GSM1269711 1 0.547 0.7140 0.756 0.056 0.164 0.024
#> GSM1269646 3 0.741 0.0933 0.104 0.316 0.552 0.028
#> GSM1269654 3 0.677 0.4725 0.120 0.152 0.684 0.044
#> GSM1269662 3 0.897 -0.0991 0.104 0.324 0.428 0.144
#> GSM1269670 2 0.684 0.6571 0.108 0.644 0.224 0.024
#> GSM1269678 3 0.452 0.5875 0.108 0.044 0.824 0.024
#> GSM1269692 3 0.760 0.2022 0.352 0.096 0.516 0.036
#> GSM1269700 3 0.516 0.5653 0.104 0.088 0.788 0.020
#> GSM1269708 3 0.442 0.5812 0.108 0.044 0.828 0.020
#> GSM1269714 3 0.503 0.5791 0.112 0.052 0.800 0.036
#> GSM1269716 3 0.499 0.5591 0.088 0.044 0.808 0.060
#> GSM1269720 3 0.872 -0.1892 0.184 0.380 0.380 0.056
#> GSM1269722 3 0.455 0.5794 0.108 0.056 0.820 0.016
#> GSM1269644 1 0.625 0.7015 0.708 0.092 0.172 0.028
#> GSM1269652 1 0.607 0.7106 0.728 0.088 0.152 0.032
#> GSM1269660 1 0.669 0.6318 0.620 0.084 0.280 0.016
#> GSM1269668 1 0.587 0.6969 0.716 0.044 0.208 0.032
#> GSM1269676 1 0.882 0.2350 0.440 0.288 0.208 0.064
#> GSM1269684 1 0.584 0.7125 0.716 0.068 0.200 0.016
#> GSM1269690 1 0.595 0.6616 0.672 0.052 0.264 0.012
#> GSM1269698 1 0.804 0.4198 0.496 0.312 0.160 0.032
#> GSM1269706 1 0.798 0.4798 0.540 0.256 0.164 0.040
#> GSM1269650 2 0.798 0.3384 0.132 0.420 0.416 0.032
#> GSM1269658 3 0.910 0.0685 0.188 0.264 0.444 0.104
#> GSM1269666 3 0.601 0.5543 0.116 0.108 0.740 0.036
#> GSM1269674 2 0.808 0.6167 0.128 0.500 0.324 0.048
#> GSM1269682 3 0.529 0.5746 0.120 0.076 0.780 0.024
#> GSM1269688 3 0.763 0.2345 0.164 0.224 0.580 0.032
#> GSM1269696 3 0.720 -0.0793 0.092 0.348 0.540 0.020
#> GSM1269704 3 0.747 -0.3431 0.104 0.412 0.464 0.020
#> GSM1269712 3 0.557 0.5688 0.088 0.052 0.776 0.084
#> GSM1269718 3 0.699 0.4259 0.176 0.132 0.656 0.036
#> GSM1269724 3 0.670 0.4321 0.108 0.196 0.668 0.028
#> GSM1269726 3 0.370 0.5846 0.080 0.032 0.868 0.020
#> GSM1269648 1 0.604 0.7049 0.716 0.056 0.192 0.036
#> GSM1269656 1 0.680 0.6347 0.656 0.180 0.144 0.020
#> GSM1269664 1 0.664 0.6770 0.652 0.064 0.248 0.036
#> GSM1269672 1 0.541 0.7121 0.752 0.044 0.180 0.024
#> GSM1269680 1 0.778 0.2694 0.448 0.416 0.096 0.040
#> GSM1269686 1 0.502 0.7098 0.752 0.044 0.200 0.004
#> GSM1269694 1 0.566 0.7202 0.740 0.076 0.168 0.016
#> GSM1269702 1 0.534 0.7089 0.748 0.064 0.180 0.008
#> GSM1269710 1 0.545 0.7108 0.764 0.052 0.152 0.032
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM1269647 3 0.749 0.1655 0.088 0.060 0.472 0.028 0.352
#> GSM1269655 3 0.720 0.5199 0.092 0.072 0.624 0.060 0.152
#> GSM1269663 3 0.850 -0.0964 0.020 0.260 0.388 0.104 0.228
#> GSM1269671 5 0.532 0.4860 0.044 0.116 0.108 0.000 0.732
#> GSM1269679 3 0.426 0.6462 0.080 0.008 0.816 0.024 0.072
#> GSM1269693 3 0.696 0.4701 0.060 0.168 0.636 0.080 0.056
#> GSM1269701 3 0.552 0.6315 0.144 0.012 0.720 0.024 0.100
#> GSM1269709 3 0.442 0.6454 0.108 0.020 0.800 0.008 0.064
#> GSM1269715 4 0.247 0.0000 0.000 0.012 0.104 0.884 0.000
#> GSM1269717 3 0.508 0.6101 0.076 0.056 0.768 0.092 0.008
#> GSM1269721 2 0.880 0.2870 0.092 0.384 0.240 0.048 0.236
#> GSM1269723 3 0.513 0.6482 0.120 0.028 0.764 0.024 0.064
#> GSM1269645 1 0.688 0.4883 0.608 0.208 0.040 0.028 0.116
#> GSM1269653 1 0.668 0.5527 0.624 0.204 0.072 0.012 0.088
#> GSM1269661 1 0.624 0.5994 0.652 0.068 0.212 0.012 0.056
#> GSM1269669 1 0.483 0.6778 0.772 0.084 0.116 0.012 0.016
#> GSM1269677 2 0.699 0.3489 0.280 0.552 0.112 0.016 0.040
#> GSM1269685 1 0.490 0.6947 0.764 0.104 0.104 0.004 0.024
#> GSM1269691 1 0.578 0.6359 0.664 0.112 0.204 0.004 0.016
#> GSM1269699 1 0.788 0.2588 0.456 0.312 0.104 0.016 0.112
#> GSM1269707 1 0.802 0.2802 0.452 0.308 0.120 0.024 0.096
#> GSM1269651 5 0.828 0.2610 0.068 0.248 0.192 0.040 0.452
#> GSM1269659 2 0.898 0.3542 0.112 0.344 0.336 0.084 0.124
#> GSM1269667 3 0.696 0.6008 0.152 0.040 0.632 0.056 0.120
#> GSM1269675 5 0.575 0.4885 0.056 0.048 0.156 0.024 0.716
#> GSM1269683 3 0.595 0.6121 0.116 0.040 0.720 0.072 0.052
#> GSM1269689 3 0.771 0.1695 0.120 0.060 0.440 0.024 0.356
#> GSM1269697 3 0.691 -0.0235 0.080 0.036 0.436 0.016 0.432
#> GSM1269705 5 0.722 0.4012 0.088 0.096 0.296 0.004 0.516
#> GSM1269713 3 0.615 0.6282 0.104 0.036 0.708 0.084 0.068
#> GSM1269719 3 0.762 0.4165 0.116 0.160 0.580 0.044 0.100
#> GSM1269725 3 0.645 0.5682 0.076 0.064 0.672 0.032 0.156
#> GSM1269727 3 0.417 0.6460 0.076 0.040 0.828 0.012 0.044
#> GSM1269649 1 0.497 0.6893 0.760 0.072 0.136 0.012 0.020
#> GSM1269657 1 0.691 0.0932 0.432 0.408 0.120 0.000 0.040
#> GSM1269665 1 0.727 0.5242 0.576 0.196 0.136 0.016 0.076
#> GSM1269673 1 0.571 0.6898 0.704 0.112 0.148 0.016 0.020
#> GSM1269681 5 0.763 -0.1173 0.284 0.320 0.028 0.008 0.360
#> GSM1269687 1 0.535 0.6837 0.732 0.068 0.148 0.004 0.048
#> GSM1269695 1 0.500 0.7004 0.760 0.092 0.112 0.004 0.032
#> GSM1269703 1 0.532 0.6897 0.716 0.056 0.188 0.004 0.036
#> GSM1269711 1 0.439 0.6961 0.788 0.084 0.112 0.000 0.016
#> GSM1269646 3 0.745 0.1667 0.092 0.060 0.472 0.024 0.352
#> GSM1269654 3 0.716 0.5255 0.092 0.072 0.628 0.060 0.148
#> GSM1269662 2 0.888 -0.0492 0.028 0.336 0.208 0.144 0.284
#> GSM1269670 5 0.543 0.4838 0.048 0.120 0.108 0.000 0.724
#> GSM1269678 3 0.499 0.6545 0.108 0.020 0.776 0.036 0.060
#> GSM1269692 3 0.767 0.1512 0.360 0.060 0.456 0.044 0.080
#> GSM1269700 3 0.539 0.6370 0.116 0.020 0.744 0.028 0.092
#> GSM1269708 3 0.453 0.6449 0.108 0.020 0.796 0.012 0.064
#> GSM1269714 3 0.533 0.6371 0.112 0.044 0.760 0.040 0.044
#> GSM1269716 3 0.486 0.6109 0.080 0.060 0.780 0.076 0.004
#> GSM1269720 2 0.872 0.2937 0.088 0.400 0.216 0.048 0.248
#> GSM1269722 3 0.494 0.6490 0.116 0.024 0.772 0.016 0.072
#> GSM1269644 1 0.604 0.6599 0.692 0.124 0.128 0.016 0.040
#> GSM1269652 1 0.549 0.6661 0.724 0.148 0.088 0.012 0.028
#> GSM1269660 1 0.633 0.6226 0.648 0.076 0.208 0.012 0.056
#> GSM1269668 1 0.551 0.6774 0.720 0.068 0.168 0.020 0.024
#> GSM1269676 2 0.699 0.3489 0.280 0.552 0.112 0.016 0.040
#> GSM1269684 1 0.536 0.6945 0.724 0.096 0.140 0.000 0.040
#> GSM1269690 1 0.578 0.6359 0.664 0.112 0.204 0.004 0.016
#> GSM1269698 1 0.826 0.2267 0.436 0.260 0.112 0.016 0.176
#> GSM1269706 1 0.802 0.2802 0.452 0.308 0.120 0.024 0.096
#> GSM1269650 5 0.847 0.2035 0.064 0.252 0.304 0.032 0.348
#> GSM1269658 2 0.901 0.3509 0.112 0.340 0.336 0.084 0.128
#> GSM1269666 3 0.651 0.6196 0.116 0.036 0.676 0.060 0.112
#> GSM1269674 5 0.577 0.4888 0.056 0.044 0.156 0.028 0.716
#> GSM1269682 3 0.565 0.6316 0.136 0.052 0.732 0.048 0.032
#> GSM1269688 3 0.747 0.3253 0.120 0.052 0.508 0.024 0.296
#> GSM1269696 3 0.678 0.0336 0.056 0.052 0.444 0.012 0.436
#> GSM1269704 5 0.730 0.2061 0.072 0.092 0.384 0.008 0.444
#> GSM1269712 3 0.611 0.6302 0.096 0.044 0.712 0.096 0.052
#> GSM1269718 3 0.747 0.4428 0.116 0.140 0.600 0.048 0.096
#> GSM1269724 3 0.704 0.5028 0.084 0.072 0.616 0.036 0.192
#> GSM1269726 3 0.424 0.6451 0.076 0.040 0.824 0.012 0.048
#> GSM1269648 1 0.514 0.6896 0.748 0.092 0.128 0.008 0.024
#> GSM1269656 1 0.686 0.5317 0.592 0.232 0.088 0.008 0.080
#> GSM1269664 1 0.632 0.6606 0.652 0.088 0.204 0.024 0.032
#> GSM1269672 1 0.484 0.6964 0.768 0.072 0.132 0.012 0.016
#> GSM1269680 2 0.815 -0.0102 0.320 0.328 0.072 0.008 0.272
#> GSM1269686 1 0.495 0.6846 0.752 0.056 0.148 0.000 0.044
#> GSM1269694 1 0.500 0.7004 0.760 0.092 0.112 0.004 0.032
#> GSM1269702 1 0.485 0.6746 0.752 0.116 0.116 0.000 0.016
#> GSM1269710 1 0.441 0.6912 0.800 0.080 0.096 0.008 0.016
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM1269647 3 0.735 -0.07774 0.076 0.072 0.404 0.012 0.388 0.048
#> GSM1269655 3 0.732 0.47715 0.076 0.072 0.588 0.048 0.144 0.072
#> GSM1269663 2 0.825 0.34042 0.012 0.380 0.300 0.080 0.116 0.112
#> GSM1269671 5 0.401 0.30822 0.016 0.068 0.008 0.012 0.812 0.084
#> GSM1269679 3 0.453 0.62663 0.076 0.024 0.784 0.024 0.084 0.008
#> GSM1269693 3 0.690 0.39584 0.032 0.084 0.600 0.076 0.032 0.176
#> GSM1269701 3 0.563 0.61066 0.136 0.016 0.696 0.024 0.100 0.028
#> GSM1269709 3 0.485 0.61217 0.080 0.024 0.768 0.012 0.080 0.036
#> GSM1269715 4 0.201 0.00000 0.000 0.004 0.084 0.904 0.000 0.008
#> GSM1269717 3 0.526 0.57151 0.052 0.060 0.740 0.088 0.004 0.056
#> GSM1269721 6 0.767 -0.06061 0.036 0.068 0.180 0.024 0.204 0.488
#> GSM1269723 3 0.522 0.62547 0.096 0.032 0.748 0.024 0.068 0.032
#> GSM1269645 1 0.744 0.38987 0.544 0.160 0.028 0.032 0.092 0.144
#> GSM1269653 1 0.736 0.40995 0.524 0.132 0.064 0.008 0.060 0.212
#> GSM1269661 1 0.648 0.57148 0.612 0.052 0.196 0.020 0.020 0.100
#> GSM1269669 1 0.481 0.67150 0.768 0.052 0.096 0.012 0.016 0.056
#> GSM1269677 6 0.459 0.40695 0.164 0.008 0.064 0.008 0.012 0.744
#> GSM1269685 1 0.505 0.66721 0.740 0.024 0.080 0.012 0.020 0.124
#> GSM1269691 1 0.590 0.60299 0.632 0.016 0.176 0.004 0.024 0.148
#> GSM1269699 6 0.758 0.27777 0.328 0.092 0.044 0.020 0.076 0.440
#> GSM1269707 6 0.766 0.25429 0.352 0.104 0.060 0.016 0.060 0.408
#> GSM1269651 5 0.773 -0.20989 0.016 0.320 0.100 0.008 0.364 0.192
#> GSM1269659 6 0.827 -0.11262 0.052 0.156 0.264 0.036 0.080 0.412
#> GSM1269667 3 0.718 0.57318 0.132 0.052 0.592 0.064 0.124 0.036
#> GSM1269675 5 0.553 0.31222 0.040 0.160 0.084 0.012 0.692 0.012
#> GSM1269683 3 0.620 0.56752 0.096 0.048 0.688 0.064 0.052 0.052
#> GSM1269689 3 0.774 -0.00852 0.128 0.076 0.368 0.008 0.368 0.052
#> GSM1269697 5 0.697 0.14893 0.072 0.052 0.388 0.008 0.436 0.044
#> GSM1269705 5 0.757 0.25240 0.056 0.092 0.248 0.008 0.488 0.108
#> GSM1269713 3 0.662 0.58686 0.088 0.048 0.656 0.080 0.080 0.048
#> GSM1269719 3 0.773 0.28351 0.096 0.176 0.508 0.016 0.056 0.148
#> GSM1269725 3 0.658 0.51815 0.064 0.048 0.636 0.028 0.160 0.064
#> GSM1269727 3 0.479 0.60309 0.060 0.040 0.784 0.020 0.052 0.044
#> GSM1269649 1 0.531 0.66272 0.732 0.028 0.096 0.012 0.036 0.096
#> GSM1269657 6 0.594 0.22299 0.304 0.016 0.088 0.000 0.028 0.564
#> GSM1269665 1 0.762 0.46005 0.524 0.140 0.116 0.024 0.036 0.160
#> GSM1269673 1 0.598 0.67411 0.680 0.060 0.136 0.016 0.028 0.080
#> GSM1269681 6 0.823 0.21488 0.180 0.164 0.008 0.032 0.268 0.348
#> GSM1269687 1 0.507 0.66777 0.728 0.020 0.116 0.008 0.016 0.112
#> GSM1269695 1 0.500 0.66712 0.720 0.008 0.076 0.004 0.032 0.160
#> GSM1269703 1 0.544 0.67694 0.696 0.032 0.140 0.008 0.012 0.112
#> GSM1269711 1 0.468 0.67099 0.768 0.036 0.080 0.004 0.016 0.096
#> GSM1269646 3 0.739 -0.08445 0.080 0.072 0.400 0.012 0.388 0.048
#> GSM1269654 3 0.729 0.48369 0.076 0.072 0.592 0.048 0.140 0.072
#> GSM1269662 2 0.784 0.31595 0.012 0.504 0.124 0.072 0.144 0.144
#> GSM1269670 5 0.418 0.31213 0.020 0.076 0.016 0.004 0.800 0.084
#> GSM1269678 3 0.502 0.63098 0.076 0.044 0.768 0.032 0.052 0.028
#> GSM1269692 3 0.778 0.10888 0.328 0.052 0.424 0.040 0.044 0.112
#> GSM1269700 3 0.561 0.61418 0.096 0.020 0.712 0.032 0.108 0.032
#> GSM1269708 3 0.496 0.61123 0.084 0.028 0.764 0.016 0.072 0.036
#> GSM1269714 3 0.548 0.59938 0.088 0.036 0.736 0.044 0.032 0.064
#> GSM1269716 3 0.483 0.57379 0.056 0.040 0.764 0.084 0.000 0.056
#> GSM1269720 6 0.763 -0.03700 0.036 0.068 0.156 0.028 0.212 0.500
#> GSM1269722 3 0.507 0.62815 0.096 0.032 0.752 0.012 0.076 0.032
#> GSM1269644 1 0.634 0.62247 0.656 0.076 0.112 0.020 0.028 0.108
#> GSM1269652 1 0.575 0.60822 0.676 0.048 0.072 0.008 0.024 0.172
#> GSM1269660 1 0.637 0.59195 0.616 0.048 0.196 0.012 0.024 0.104
#> GSM1269668 1 0.543 0.66954 0.716 0.044 0.144 0.024 0.016 0.056
#> GSM1269676 6 0.459 0.40695 0.164 0.008 0.064 0.008 0.012 0.744
#> GSM1269684 1 0.512 0.67969 0.728 0.020 0.112 0.008 0.020 0.112
#> GSM1269690 1 0.590 0.60299 0.632 0.016 0.176 0.004 0.024 0.148
#> GSM1269698 6 0.821 0.26458 0.304 0.080 0.080 0.016 0.136 0.384
#> GSM1269706 6 0.766 0.25429 0.352 0.104 0.060 0.016 0.060 0.408
#> GSM1269650 2 0.821 -0.04799 0.012 0.292 0.204 0.012 0.268 0.212
#> GSM1269658 6 0.830 -0.11076 0.052 0.164 0.260 0.036 0.080 0.408
#> GSM1269666 3 0.665 0.58975 0.100 0.064 0.640 0.048 0.120 0.028
#> GSM1269674 5 0.573 0.30954 0.040 0.152 0.092 0.016 0.684 0.016
#> GSM1269682 3 0.580 0.59485 0.104 0.056 0.712 0.040 0.032 0.056
#> GSM1269688 3 0.747 0.17247 0.124 0.052 0.436 0.012 0.328 0.048
#> GSM1269696 5 0.698 0.13499 0.056 0.080 0.372 0.004 0.444 0.044
#> GSM1269704 5 0.747 0.23699 0.060 0.064 0.320 0.008 0.448 0.100
#> GSM1269712 3 0.624 0.59163 0.088 0.036 0.680 0.100 0.048 0.048
#> GSM1269718 3 0.772 0.33556 0.100 0.136 0.540 0.036 0.056 0.132
#> GSM1269724 3 0.705 0.43951 0.064 0.052 0.580 0.024 0.196 0.084
#> GSM1269726 3 0.491 0.60239 0.064 0.040 0.776 0.020 0.052 0.048
#> GSM1269648 1 0.530 0.65887 0.720 0.024 0.088 0.004 0.040 0.124
#> GSM1269656 1 0.681 0.29964 0.504 0.040 0.064 0.004 0.068 0.320
#> GSM1269664 1 0.634 0.64370 0.632 0.052 0.176 0.024 0.016 0.100
#> GSM1269672 1 0.505 0.68994 0.748 0.040 0.116 0.016 0.016 0.064
#> GSM1269680 6 0.816 0.35836 0.236 0.116 0.028 0.020 0.208 0.392
#> GSM1269686 1 0.484 0.67080 0.744 0.020 0.112 0.004 0.016 0.104
#> GSM1269694 1 0.500 0.66712 0.720 0.008 0.076 0.004 0.032 0.160
#> GSM1269702 1 0.501 0.63748 0.704 0.012 0.088 0.000 0.020 0.176
#> GSM1269710 1 0.448 0.66447 0.784 0.040 0.064 0.004 0.016 0.092
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 agent(p) disease.state(p) gender(p) individual(p) k
#> CV:hclust 10 NA NA NA NA 2
#> CV:hclust 78 1.000 1.000 7.87e-18 2.08e-04 3
#> CV:hclust 51 0.874 0.610 8.42e-12 4.25e-05 4
#> CV:hclust 49 1.000 0.480 1.90e-11 2.83e-03 5
#> CV:hclust 43 0.904 0.654 4.08e-10 6.93e-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", "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 51941 rows and 84 columns.
#> Top rows (1000, 2000, 3000, 4000, 5000) are extracted by 'CV' method.
#> Subgroups are detected by 'kmeans' method.
#> Performed in total 1250 partitions by row resampling.
#> Best k for subgroups seems to be 2.
#>
#> Following methods can be applied to this 'ConsensusPartition' object:
#> [1] "cola_report" "collect_classes" "collect_plots"
#> [4] "collect_stats" "colnames" "compare_signatures"
#> [7] "consensus_heatmap" "dimension_reduction" "functional_enrichment"
#> [10] "get_anno_col" "get_anno" "get_classes"
#> [13] "get_consensus" "get_matrix" "get_membership"
#> [16] "get_param" "get_signatures" "get_stats"
#> [19] "is_best_k" "is_stable_k" "membership_heatmap"
#> [22] "ncol" "nrow" "plot_ecdf"
#> [25] "rownames" "select_partition_number" "show"
#> [28] "suggest_best_k" "test_to_known_factors"
collect_plots() function collects all the plots made from res for all k (number of partitions)
into one single page to provide an easy and fast comparison between different k.
collect_plots(res)
The plots are:
k and the heatmap of
predicted classes for each k.k.k.k.All the plots in panels can be made by individual functions and they are plotted later in this section.
select_partition_number() produces several plots showing different
statistics for choosing “optimized” k. There are following statistics:
k;k, the area increased is defined as \(A_k - A_{k-1}\).The detailed explanations of these statistics can be found in the cola vignette.
Generally speaking, lower PAC score, higher mean silhouette score or higher
concordance corresponds to better partition. Rand index and Jaccard index
measure how similar the current partition is compared to partition with k-1.
If they are too similar, we won't accept k is better than k-1.
select_partition_number(res)
The numeric values for all these statistics can be obtained by get_stats().
get_stats(res)
#> k 1-PAC mean_silhouette concordance area_increased Rand Jaccard
#> 2 2 0.543 0.927 0.924 0.4885 0.504 0.504
#> 3 3 0.525 0.713 0.811 0.3018 0.844 0.693
#> 4 4 0.512 0.550 0.728 0.1289 0.937 0.831
#> 5 5 0.553 0.487 0.656 0.0746 0.853 0.568
#> 6 6 0.573 0.506 0.626 0.0405 0.909 0.658
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
#> GSM1269647 1 0.402 0.903 0.920 0.080
#> GSM1269655 1 0.242 0.937 0.960 0.040
#> GSM1269663 1 0.327 0.938 0.940 0.060
#> GSM1269671 1 0.430 0.892 0.912 0.088
#> GSM1269679 1 0.469 0.934 0.900 0.100
#> GSM1269693 1 0.506 0.927 0.888 0.112
#> GSM1269701 1 0.469 0.934 0.900 0.100
#> GSM1269709 1 0.529 0.932 0.880 0.120
#> GSM1269715 1 0.506 0.929 0.888 0.112
#> GSM1269717 1 0.494 0.929 0.892 0.108
#> GSM1269721 1 0.295 0.914 0.948 0.052
#> GSM1269723 1 0.373 0.940 0.928 0.072
#> GSM1269645 2 0.343 0.942 0.064 0.936
#> GSM1269653 2 0.224 0.937 0.036 0.964
#> GSM1269661 2 0.373 0.940 0.072 0.928
#> GSM1269669 2 0.402 0.935 0.080 0.920
#> GSM1269677 2 0.416 0.911 0.084 0.916
#> GSM1269685 2 0.242 0.945 0.040 0.960
#> GSM1269691 2 0.388 0.939 0.076 0.924
#> GSM1269699 2 0.443 0.900 0.092 0.908
#> GSM1269707 2 0.373 0.916 0.072 0.928
#> GSM1269651 1 0.242 0.911 0.960 0.040
#> GSM1269659 1 0.358 0.922 0.932 0.068
#> GSM1269667 1 0.456 0.935 0.904 0.096
#> GSM1269675 1 0.327 0.905 0.940 0.060
#> GSM1269683 1 0.494 0.931 0.892 0.108
#> GSM1269689 1 0.494 0.899 0.892 0.108
#> GSM1269697 1 0.443 0.936 0.908 0.092
#> GSM1269705 1 0.373 0.900 0.928 0.072
#> GSM1269713 1 0.402 0.939 0.920 0.080
#> GSM1269719 1 0.416 0.940 0.916 0.084
#> GSM1269725 1 0.311 0.939 0.944 0.056
#> GSM1269727 1 0.469 0.934 0.900 0.100
#> GSM1269649 2 0.204 0.945 0.032 0.968
#> GSM1269657 2 0.402 0.914 0.080 0.920
#> GSM1269665 2 0.402 0.935 0.080 0.920
#> GSM1269673 2 0.327 0.943 0.060 0.940
#> GSM1269681 2 0.494 0.888 0.108 0.892
#> GSM1269687 2 0.388 0.938 0.076 0.924
#> GSM1269695 2 0.118 0.943 0.016 0.984
#> GSM1269703 2 0.373 0.940 0.072 0.928
#> GSM1269711 2 0.242 0.944 0.040 0.960
#> GSM1269646 1 0.430 0.908 0.912 0.088
#> GSM1269654 1 0.311 0.939 0.944 0.056
#> GSM1269662 1 0.184 0.932 0.972 0.028
#> GSM1269670 1 0.430 0.892 0.912 0.088
#> GSM1269678 1 0.469 0.934 0.900 0.100
#> GSM1269692 1 0.494 0.929 0.892 0.108
#> GSM1269700 1 0.456 0.935 0.904 0.096
#> GSM1269708 1 0.518 0.935 0.884 0.116
#> GSM1269714 1 0.494 0.931 0.892 0.108
#> GSM1269716 1 0.494 0.929 0.892 0.108
#> GSM1269720 1 0.388 0.903 0.924 0.076
#> GSM1269722 1 0.416 0.938 0.916 0.084
#> GSM1269644 2 0.311 0.943 0.056 0.944
#> GSM1269652 2 0.242 0.938 0.040 0.960
#> GSM1269660 2 0.373 0.940 0.072 0.928
#> GSM1269668 2 0.402 0.935 0.080 0.920
#> GSM1269676 2 0.416 0.912 0.084 0.916
#> GSM1269684 2 0.416 0.935 0.084 0.916
#> GSM1269690 2 0.388 0.939 0.076 0.924
#> GSM1269698 2 0.456 0.897 0.096 0.904
#> GSM1269706 2 0.373 0.916 0.072 0.928
#> GSM1269650 1 0.278 0.909 0.952 0.048
#> GSM1269658 1 0.358 0.924 0.932 0.068
#> GSM1269666 1 0.430 0.937 0.912 0.088
#> GSM1269674 1 0.343 0.903 0.936 0.064
#> GSM1269682 1 0.506 0.929 0.888 0.112
#> GSM1269688 1 0.494 0.897 0.892 0.108
#> GSM1269696 1 0.443 0.921 0.908 0.092
#> GSM1269704 1 0.388 0.897 0.924 0.076
#> GSM1269712 1 0.430 0.937 0.912 0.088
#> GSM1269718 1 0.402 0.940 0.920 0.080
#> GSM1269724 1 0.373 0.939 0.928 0.072
#> GSM1269726 1 0.469 0.934 0.900 0.100
#> GSM1269648 2 0.118 0.943 0.016 0.984
#> GSM1269656 2 0.311 0.925 0.056 0.944
#> GSM1269664 2 0.416 0.934 0.084 0.916
#> GSM1269672 2 0.358 0.941 0.068 0.932
#> GSM1269680 2 0.456 0.896 0.096 0.904
#> GSM1269686 2 0.388 0.938 0.076 0.924
#> GSM1269694 2 0.118 0.943 0.016 0.984
#> GSM1269702 2 0.184 0.945 0.028 0.972
#> GSM1269710 2 0.163 0.944 0.024 0.976
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM1269647 1 0.729 -0.3868 0.508 0.028 0.464
#> GSM1269655 1 0.516 0.6266 0.764 0.004 0.232
#> GSM1269663 1 0.525 0.7013 0.792 0.020 0.188
#> GSM1269671 3 0.679 0.7125 0.292 0.036 0.672
#> GSM1269679 1 0.164 0.7813 0.964 0.016 0.020
#> GSM1269693 1 0.541 0.7041 0.812 0.052 0.136
#> GSM1269701 1 0.279 0.7761 0.928 0.028 0.044
#> GSM1269709 1 0.315 0.7625 0.916 0.044 0.040
#> GSM1269715 1 0.516 0.7063 0.820 0.040 0.140
#> GSM1269717 1 0.437 0.7425 0.864 0.040 0.096
#> GSM1269721 3 0.572 0.6972 0.240 0.016 0.744
#> GSM1269723 1 0.364 0.7736 0.892 0.024 0.084
#> GSM1269645 2 0.378 0.8882 0.044 0.892 0.064
#> GSM1269653 2 0.338 0.8859 0.012 0.896 0.092
#> GSM1269661 2 0.408 0.8805 0.072 0.880 0.048
#> GSM1269669 2 0.337 0.8846 0.052 0.908 0.040
#> GSM1269677 2 0.566 0.7766 0.004 0.712 0.284
#> GSM1269685 2 0.145 0.8944 0.008 0.968 0.024
#> GSM1269691 2 0.348 0.8893 0.052 0.904 0.044
#> GSM1269699 2 0.580 0.7701 0.008 0.712 0.280
#> GSM1269707 2 0.497 0.8006 0.000 0.764 0.236
#> GSM1269651 3 0.580 0.7030 0.280 0.008 0.712
#> GSM1269659 3 0.742 0.4282 0.388 0.040 0.572
#> GSM1269667 1 0.355 0.7616 0.896 0.024 0.080
#> GSM1269675 3 0.697 0.6760 0.356 0.028 0.616
#> GSM1269683 1 0.432 0.7491 0.868 0.044 0.088
#> GSM1269689 3 0.812 0.6008 0.396 0.072 0.532
#> GSM1269697 1 0.721 0.0121 0.604 0.036 0.360
#> GSM1269705 3 0.701 0.6869 0.364 0.028 0.608
#> GSM1269713 1 0.383 0.7424 0.880 0.020 0.100
#> GSM1269719 1 0.429 0.7331 0.840 0.008 0.152
#> GSM1269725 1 0.364 0.7531 0.892 0.024 0.084
#> GSM1269727 1 0.231 0.7831 0.944 0.024 0.032
#> GSM1269649 2 0.245 0.8873 0.012 0.936 0.052
#> GSM1269657 2 0.566 0.7768 0.004 0.712 0.284
#> GSM1269665 2 0.386 0.8771 0.072 0.888 0.040
#> GSM1269673 2 0.269 0.8931 0.036 0.932 0.032
#> GSM1269681 3 0.641 -0.2588 0.004 0.420 0.576
#> GSM1269687 2 0.358 0.8912 0.056 0.900 0.044
#> GSM1269695 2 0.304 0.8816 0.008 0.908 0.084
#> GSM1269703 2 0.293 0.8922 0.040 0.924 0.036
#> GSM1269711 2 0.249 0.8854 0.016 0.936 0.048
#> GSM1269646 1 0.737 -0.3363 0.524 0.032 0.444
#> GSM1269654 1 0.397 0.7524 0.860 0.008 0.132
#> GSM1269662 3 0.668 0.2580 0.484 0.008 0.508
#> GSM1269670 3 0.660 0.7139 0.296 0.028 0.676
#> GSM1269678 1 0.255 0.7831 0.936 0.024 0.040
#> GSM1269692 1 0.625 0.6404 0.756 0.056 0.188
#> GSM1269700 1 0.266 0.7743 0.932 0.024 0.044
#> GSM1269708 1 0.348 0.7573 0.904 0.044 0.052
#> GSM1269714 1 0.441 0.7504 0.860 0.036 0.104
#> GSM1269716 1 0.426 0.7497 0.868 0.036 0.096
#> GSM1269720 3 0.608 0.6887 0.216 0.036 0.748
#> GSM1269722 1 0.241 0.7849 0.940 0.020 0.040
#> GSM1269644 2 0.315 0.8932 0.040 0.916 0.044
#> GSM1269652 2 0.265 0.8891 0.012 0.928 0.060
#> GSM1269660 2 0.437 0.8788 0.076 0.868 0.056
#> GSM1269668 2 0.374 0.8780 0.072 0.892 0.036
#> GSM1269676 2 0.562 0.7803 0.004 0.716 0.280
#> GSM1269684 2 0.218 0.8960 0.032 0.948 0.020
#> GSM1269690 2 0.348 0.8893 0.052 0.904 0.044
#> GSM1269698 2 0.586 0.7606 0.008 0.704 0.288
#> GSM1269706 2 0.497 0.8006 0.000 0.764 0.236
#> GSM1269650 3 0.569 0.7035 0.268 0.008 0.724
#> GSM1269658 3 0.746 0.3957 0.400 0.040 0.560
#> GSM1269666 1 0.275 0.7783 0.924 0.012 0.064
#> GSM1269674 3 0.657 0.7071 0.308 0.024 0.668
#> GSM1269682 1 0.404 0.7577 0.880 0.040 0.080
#> GSM1269688 3 0.800 0.6594 0.360 0.072 0.568
#> GSM1269696 1 0.703 -0.1273 0.580 0.024 0.396
#> GSM1269704 3 0.691 0.7034 0.344 0.028 0.628
#> GSM1269712 1 0.205 0.7821 0.952 0.020 0.028
#> GSM1269718 1 0.368 0.7539 0.876 0.008 0.116
#> GSM1269724 1 0.515 0.6305 0.800 0.020 0.180
#> GSM1269726 1 0.343 0.7690 0.904 0.032 0.064
#> GSM1269648 2 0.258 0.8843 0.008 0.928 0.064
#> GSM1269656 2 0.463 0.8416 0.004 0.808 0.188
#> GSM1269664 2 0.355 0.8826 0.064 0.900 0.036
#> GSM1269672 2 0.303 0.8870 0.048 0.920 0.032
#> GSM1269680 2 0.590 0.7382 0.004 0.680 0.316
#> GSM1269686 2 0.324 0.8873 0.056 0.912 0.032
#> GSM1269694 2 0.296 0.8821 0.008 0.912 0.080
#> GSM1269702 2 0.290 0.8938 0.016 0.920 0.064
#> GSM1269710 2 0.188 0.8887 0.004 0.952 0.044
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM1269647 2 0.658 0.4303 0.004 0.540 0.384 0.072
#> GSM1269655 3 0.683 0.5880 0.008 0.236 0.620 0.136
#> GSM1269663 3 0.646 0.6342 0.008 0.120 0.660 0.212
#> GSM1269671 2 0.330 0.6851 0.000 0.876 0.076 0.048
#> GSM1269679 3 0.238 0.7317 0.000 0.052 0.920 0.028
#> GSM1269693 3 0.623 0.6342 0.012 0.068 0.656 0.264
#> GSM1269701 3 0.298 0.7081 0.000 0.084 0.888 0.028
#> GSM1269709 3 0.425 0.7189 0.004 0.104 0.828 0.064
#> GSM1269715 3 0.623 0.6305 0.008 0.072 0.648 0.272
#> GSM1269717 3 0.560 0.6801 0.004 0.060 0.704 0.232
#> GSM1269721 2 0.622 0.6359 0.012 0.688 0.100 0.200
#> GSM1269723 3 0.413 0.7192 0.000 0.108 0.828 0.064
#> GSM1269645 1 0.355 0.6779 0.868 0.020 0.016 0.096
#> GSM1269653 1 0.398 0.6217 0.796 0.012 0.000 0.192
#> GSM1269661 1 0.334 0.6959 0.880 0.008 0.032 0.080
#> GSM1269669 1 0.277 0.6976 0.908 0.008 0.024 0.060
#> GSM1269677 4 0.621 0.3860 0.460 0.052 0.000 0.488
#> GSM1269685 1 0.291 0.6932 0.888 0.020 0.000 0.092
#> GSM1269691 1 0.307 0.7019 0.888 0.012 0.012 0.088
#> GSM1269699 1 0.701 -0.5191 0.452 0.116 0.000 0.432
#> GSM1269707 1 0.630 -0.3582 0.516 0.060 0.000 0.424
#> GSM1269651 2 0.478 0.6815 0.004 0.796 0.088 0.112
#> GSM1269659 2 0.852 0.2843 0.028 0.400 0.268 0.304
#> GSM1269667 3 0.415 0.6600 0.000 0.152 0.812 0.036
#> GSM1269675 2 0.391 0.7041 0.000 0.836 0.120 0.044
#> GSM1269683 3 0.506 0.7064 0.004 0.068 0.768 0.160
#> GSM1269689 2 0.677 0.5771 0.016 0.604 0.296 0.084
#> GSM1269697 3 0.629 -0.1396 0.004 0.436 0.512 0.048
#> GSM1269705 2 0.473 0.6929 0.004 0.780 0.172 0.044
#> GSM1269713 3 0.409 0.6672 0.000 0.140 0.820 0.040
#> GSM1269719 3 0.696 0.6165 0.008 0.180 0.616 0.196
#> GSM1269725 3 0.440 0.6458 0.000 0.152 0.800 0.048
#> GSM1269727 3 0.392 0.7343 0.004 0.044 0.844 0.108
#> GSM1269649 1 0.348 0.6971 0.872 0.032 0.008 0.088
#> GSM1269657 1 0.601 -0.5283 0.484 0.040 0.000 0.476
#> GSM1269665 1 0.363 0.6789 0.860 0.008 0.028 0.104
#> GSM1269673 1 0.201 0.7086 0.932 0.004 0.004 0.060
#> GSM1269681 4 0.749 0.5272 0.192 0.336 0.000 0.472
#> GSM1269687 1 0.298 0.7088 0.896 0.012 0.016 0.076
#> GSM1269695 1 0.375 0.6641 0.840 0.032 0.000 0.128
#> GSM1269703 1 0.276 0.7047 0.908 0.012 0.016 0.064
#> GSM1269711 1 0.409 0.6681 0.828 0.028 0.008 0.136
#> GSM1269646 2 0.669 0.4177 0.004 0.532 0.384 0.080
#> GSM1269654 3 0.625 0.6796 0.008 0.168 0.688 0.136
#> GSM1269662 2 0.745 0.4808 0.004 0.532 0.204 0.260
#> GSM1269670 2 0.353 0.6857 0.000 0.864 0.080 0.056
#> GSM1269678 3 0.266 0.7416 0.000 0.036 0.908 0.056
#> GSM1269692 3 0.727 0.5506 0.024 0.112 0.580 0.284
#> GSM1269700 3 0.308 0.7062 0.000 0.096 0.880 0.024
#> GSM1269708 3 0.370 0.7281 0.004 0.084 0.860 0.052
#> GSM1269714 3 0.578 0.6758 0.008 0.076 0.712 0.204
#> GSM1269716 3 0.522 0.6907 0.000 0.060 0.732 0.208
#> GSM1269720 2 0.602 0.6275 0.016 0.692 0.064 0.228
#> GSM1269722 3 0.354 0.7355 0.000 0.076 0.864 0.060
#> GSM1269644 1 0.238 0.7069 0.916 0.004 0.008 0.072
#> GSM1269652 1 0.365 0.6459 0.832 0.016 0.000 0.152
#> GSM1269660 1 0.391 0.6839 0.844 0.008 0.032 0.116
#> GSM1269668 1 0.276 0.6903 0.904 0.000 0.044 0.052
#> GSM1269676 1 0.608 -0.5173 0.488 0.044 0.000 0.468
#> GSM1269684 1 0.231 0.7089 0.920 0.008 0.004 0.068
#> GSM1269690 1 0.314 0.7013 0.884 0.012 0.012 0.092
#> GSM1269698 1 0.697 -0.5312 0.456 0.112 0.000 0.432
#> GSM1269706 1 0.624 -0.3398 0.520 0.056 0.000 0.424
#> GSM1269650 2 0.517 0.6829 0.004 0.768 0.096 0.132
#> GSM1269658 2 0.853 0.2811 0.028 0.396 0.268 0.308
#> GSM1269666 3 0.459 0.6992 0.000 0.116 0.800 0.084
#> GSM1269674 2 0.324 0.7039 0.000 0.876 0.088 0.036
#> GSM1269682 3 0.510 0.7026 0.004 0.068 0.764 0.164
#> GSM1269688 2 0.635 0.6518 0.020 0.676 0.224 0.080
#> GSM1269696 2 0.631 0.2919 0.004 0.496 0.452 0.048
#> GSM1269704 2 0.485 0.6873 0.004 0.760 0.200 0.036
#> GSM1269712 3 0.314 0.7406 0.000 0.024 0.876 0.100
#> GSM1269718 3 0.619 0.6605 0.020 0.128 0.712 0.140
#> GSM1269724 3 0.665 0.4610 0.004 0.236 0.628 0.132
#> GSM1269726 3 0.462 0.7193 0.004 0.052 0.796 0.148
#> GSM1269648 1 0.358 0.6654 0.852 0.032 0.000 0.116
#> GSM1269656 1 0.534 0.0142 0.624 0.020 0.000 0.356
#> GSM1269664 1 0.366 0.6741 0.864 0.008 0.040 0.088
#> GSM1269672 1 0.233 0.7024 0.924 0.004 0.016 0.056
#> GSM1269680 4 0.728 0.5783 0.380 0.152 0.000 0.468
#> GSM1269686 1 0.298 0.7080 0.896 0.012 0.016 0.076
#> GSM1269694 1 0.380 0.6608 0.836 0.032 0.000 0.132
#> GSM1269702 1 0.303 0.6804 0.868 0.008 0.000 0.124
#> GSM1269710 1 0.318 0.6858 0.876 0.028 0.000 0.096
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM1269647 3 0.604 0.29942 0.004 0.028 0.592 0.064 0.312
#> GSM1269655 4 0.719 -0.02028 0.000 0.040 0.336 0.452 0.172
#> GSM1269663 4 0.605 0.45446 0.000 0.064 0.172 0.668 0.096
#> GSM1269671 5 0.380 0.62147 0.000 0.080 0.108 0.000 0.812
#> GSM1269679 4 0.467 -0.00334 0.000 0.004 0.444 0.544 0.008
#> GSM1269693 4 0.339 0.53966 0.028 0.036 0.052 0.872 0.012
#> GSM1269701 3 0.493 0.22443 0.008 0.016 0.564 0.412 0.000
#> GSM1269709 3 0.637 0.13355 0.012 0.020 0.456 0.448 0.064
#> GSM1269715 4 0.332 0.54485 0.016 0.036 0.052 0.876 0.020
#> GSM1269717 4 0.311 0.56680 0.024 0.024 0.080 0.872 0.000
#> GSM1269721 5 0.642 0.61399 0.004 0.184 0.060 0.112 0.640
#> GSM1269723 4 0.570 0.10405 0.004 0.020 0.392 0.548 0.036
#> GSM1269645 1 0.460 0.68726 0.800 0.064 0.092 0.016 0.028
#> GSM1269653 1 0.527 0.53361 0.676 0.256 0.048 0.004 0.016
#> GSM1269661 1 0.500 0.71100 0.760 0.132 0.044 0.060 0.004
#> GSM1269669 1 0.236 0.73190 0.916 0.024 0.032 0.028 0.000
#> GSM1269677 2 0.526 0.69716 0.260 0.676 0.028 0.004 0.032
#> GSM1269685 1 0.418 0.65363 0.736 0.240 0.008 0.000 0.016
#> GSM1269691 1 0.411 0.70124 0.776 0.184 0.012 0.028 0.000
#> GSM1269699 2 0.591 0.70241 0.208 0.652 0.028 0.000 0.112
#> GSM1269707 2 0.580 0.66416 0.296 0.616 0.044 0.000 0.044
#> GSM1269651 5 0.615 0.61857 0.000 0.112 0.144 0.076 0.668
#> GSM1269659 5 0.845 0.34192 0.008 0.168 0.156 0.324 0.344
#> GSM1269667 3 0.599 0.24865 0.008 0.016 0.520 0.404 0.052
#> GSM1269675 5 0.412 0.60441 0.004 0.020 0.176 0.016 0.784
#> GSM1269683 4 0.290 0.57113 0.020 0.008 0.080 0.884 0.008
#> GSM1269689 3 0.741 -0.03922 0.036 0.080 0.456 0.048 0.380
#> GSM1269697 3 0.626 0.29521 0.004 0.008 0.524 0.108 0.356
#> GSM1269705 5 0.463 0.57163 0.000 0.028 0.224 0.020 0.728
#> GSM1269713 3 0.563 0.26861 0.004 0.012 0.528 0.416 0.040
#> GSM1269719 3 0.723 0.00331 0.008 0.068 0.440 0.396 0.088
#> GSM1269725 3 0.483 0.32338 0.004 0.000 0.616 0.356 0.024
#> GSM1269727 4 0.480 0.43978 0.012 0.016 0.260 0.700 0.012
#> GSM1269649 1 0.496 0.69898 0.756 0.152 0.056 0.008 0.028
#> GSM1269657 2 0.507 0.70238 0.256 0.688 0.020 0.004 0.032
#> GSM1269665 1 0.503 0.69466 0.776 0.084 0.072 0.056 0.012
#> GSM1269673 1 0.301 0.74287 0.876 0.076 0.036 0.012 0.000
#> GSM1269681 2 0.645 0.37062 0.064 0.572 0.068 0.000 0.296
#> GSM1269687 1 0.426 0.71826 0.780 0.172 0.016 0.028 0.004
#> GSM1269695 1 0.473 0.57644 0.700 0.256 0.012 0.000 0.032
#> GSM1269703 1 0.301 0.74110 0.884 0.060 0.040 0.008 0.008
#> GSM1269711 1 0.496 0.64349 0.712 0.228 0.036 0.004 0.020
#> GSM1269646 3 0.602 0.30367 0.004 0.024 0.592 0.068 0.312
#> GSM1269654 4 0.671 0.07284 0.000 0.032 0.352 0.496 0.120
#> GSM1269662 5 0.797 0.36097 0.000 0.088 0.240 0.288 0.384
#> GSM1269670 5 0.391 0.61886 0.000 0.084 0.112 0.000 0.804
#> GSM1269678 4 0.451 0.30200 0.012 0.000 0.340 0.644 0.004
#> GSM1269692 4 0.565 0.47024 0.032 0.052 0.136 0.732 0.048
#> GSM1269700 3 0.501 0.19336 0.008 0.012 0.548 0.428 0.004
#> GSM1269708 4 0.646 -0.18423 0.012 0.028 0.436 0.464 0.060
#> GSM1269714 4 0.190 0.56817 0.024 0.012 0.016 0.940 0.008
#> GSM1269716 4 0.227 0.57136 0.016 0.008 0.064 0.912 0.000
#> GSM1269720 5 0.629 0.61910 0.004 0.188 0.068 0.088 0.652
#> GSM1269722 4 0.560 -0.00762 0.008 0.020 0.452 0.500 0.020
#> GSM1269644 1 0.418 0.72448 0.796 0.152 0.028 0.016 0.008
#> GSM1269652 1 0.486 0.57718 0.696 0.252 0.040 0.000 0.012
#> GSM1269660 1 0.520 0.69355 0.752 0.088 0.108 0.048 0.004
#> GSM1269668 1 0.332 0.70847 0.864 0.020 0.056 0.060 0.000
#> GSM1269676 2 0.525 0.69524 0.272 0.668 0.024 0.004 0.032
#> GSM1269684 1 0.354 0.73140 0.824 0.148 0.016 0.004 0.008
#> GSM1269690 1 0.412 0.70271 0.780 0.176 0.012 0.032 0.000
#> GSM1269698 2 0.571 0.71753 0.184 0.680 0.032 0.000 0.104
#> GSM1269706 2 0.563 0.67388 0.288 0.632 0.036 0.000 0.044
#> GSM1269650 5 0.665 0.60927 0.000 0.136 0.180 0.072 0.612
#> GSM1269658 5 0.845 0.33449 0.008 0.168 0.156 0.328 0.340
#> GSM1269666 3 0.541 0.08034 0.000 0.020 0.512 0.444 0.024
#> GSM1269674 5 0.405 0.63219 0.004 0.020 0.128 0.036 0.812
#> GSM1269682 4 0.243 0.57252 0.020 0.004 0.076 0.900 0.000
#> GSM1269688 5 0.697 0.25613 0.032 0.072 0.356 0.032 0.508
#> GSM1269696 3 0.643 0.22295 0.004 0.032 0.524 0.076 0.364
#> GSM1269704 5 0.485 0.53604 0.000 0.036 0.248 0.016 0.700
#> GSM1269712 4 0.454 0.20472 0.000 0.008 0.380 0.608 0.004
#> GSM1269718 3 0.671 0.19487 0.024 0.060 0.552 0.324 0.040
#> GSM1269724 3 0.574 0.38316 0.004 0.004 0.616 0.280 0.096
#> GSM1269726 4 0.410 0.51356 0.012 0.016 0.196 0.772 0.004
#> GSM1269648 1 0.470 0.60104 0.708 0.248 0.016 0.000 0.028
#> GSM1269656 2 0.504 0.55009 0.376 0.592 0.004 0.004 0.024
#> GSM1269664 1 0.424 0.69799 0.808 0.040 0.116 0.032 0.004
#> GSM1269672 1 0.252 0.74465 0.908 0.036 0.036 0.020 0.000
#> GSM1269680 2 0.632 0.69007 0.180 0.636 0.052 0.000 0.132
#> GSM1269686 1 0.400 0.74781 0.824 0.112 0.024 0.032 0.008
#> GSM1269694 1 0.478 0.57315 0.704 0.248 0.016 0.000 0.032
#> GSM1269702 1 0.379 0.66226 0.744 0.248 0.000 0.004 0.004
#> GSM1269710 1 0.430 0.66585 0.748 0.216 0.016 0.000 0.020
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM1269647 3 0.627 0.3069 0.000 0.240 0.572 0.024 NA 0.028
#> GSM1269655 3 0.766 0.0910 0.004 0.172 0.380 0.264 NA 0.004
#> GSM1269663 4 0.694 0.3904 0.004 0.112 0.256 0.500 NA 0.004
#> GSM1269671 2 0.306 0.6869 0.000 0.868 0.024 0.012 NA 0.068
#> GSM1269679 3 0.393 0.4299 0.004 0.012 0.768 0.188 NA 0.004
#> GSM1269693 4 0.429 0.6027 0.012 0.004 0.200 0.736 NA 0.000
#> GSM1269701 3 0.341 0.5057 0.004 0.016 0.844 0.072 NA 0.004
#> GSM1269709 3 0.511 0.4372 0.004 0.052 0.696 0.204 NA 0.012
#> GSM1269715 4 0.506 0.5886 0.020 0.008 0.156 0.728 NA 0.028
#> GSM1269717 4 0.453 0.6039 0.016 0.004 0.224 0.716 NA 0.008
#> GSM1269721 2 0.761 0.5714 0.000 0.488 0.056 0.176 NA 0.116
#> GSM1269723 3 0.446 0.4144 0.000 0.024 0.720 0.216 NA 0.004
#> GSM1269645 1 0.470 0.6415 0.704 0.008 0.000 0.020 NA 0.048
#> GSM1269653 1 0.535 0.5423 0.604 0.000 0.000 0.004 NA 0.232
#> GSM1269661 1 0.527 0.6702 0.720 0.012 0.004 0.060 NA 0.116
#> GSM1269669 1 0.267 0.7027 0.852 0.000 0.000 0.000 NA 0.020
#> GSM1269677 6 0.623 0.6781 0.216 0.020 0.000 0.072 NA 0.604
#> GSM1269685 1 0.461 0.6577 0.720 0.000 0.000 0.028 NA 0.188
#> GSM1269691 1 0.455 0.6844 0.752 0.000 0.000 0.064 NA 0.128
#> GSM1269699 6 0.484 0.7130 0.144 0.064 0.000 0.004 NA 0.732
#> GSM1269707 6 0.473 0.7201 0.180 0.024 0.004 0.012 NA 0.732
#> GSM1269651 2 0.592 0.6566 0.000 0.624 0.020 0.084 NA 0.048
#> GSM1269659 4 0.817 -0.1736 0.004 0.252 0.072 0.376 NA 0.084
#> GSM1269667 3 0.538 0.4778 0.016 0.068 0.708 0.108 NA 0.000
#> GSM1269675 2 0.554 0.6269 0.000 0.684 0.116 0.040 NA 0.020
#> GSM1269683 4 0.506 0.5298 0.032 0.008 0.292 0.636 NA 0.000
#> GSM1269689 3 0.672 0.2112 0.012 0.272 0.528 0.008 NA 0.056
#> GSM1269697 3 0.548 0.3309 0.000 0.288 0.592 0.008 NA 0.008
#> GSM1269705 2 0.454 0.6593 0.000 0.764 0.128 0.012 NA 0.048
#> GSM1269713 3 0.422 0.5060 0.004 0.052 0.796 0.084 NA 0.004
#> GSM1269719 3 0.750 0.0883 0.008 0.060 0.352 0.280 NA 0.012
#> GSM1269725 3 0.282 0.5259 0.000 0.020 0.876 0.028 NA 0.004
#> GSM1269727 3 0.502 -0.0585 0.000 0.020 0.532 0.412 NA 0.000
#> GSM1269649 1 0.477 0.6515 0.712 0.004 0.000 0.012 NA 0.160
#> GSM1269657 6 0.574 0.6948 0.216 0.008 0.000 0.060 NA 0.636
#> GSM1269665 1 0.543 0.6186 0.680 0.004 0.012 0.036 NA 0.080
#> GSM1269673 1 0.366 0.7024 0.816 0.000 0.000 0.028 NA 0.052
#> GSM1269681 6 0.615 0.3333 0.032 0.276 0.000 0.032 NA 0.576
#> GSM1269687 1 0.418 0.6947 0.776 0.000 0.004 0.024 NA 0.136
#> GSM1269695 1 0.472 0.5793 0.668 0.004 0.000 0.004 NA 0.256
#> GSM1269703 1 0.374 0.7093 0.804 0.000 0.000 0.020 NA 0.056
#> GSM1269711 1 0.522 0.6249 0.660 0.000 0.012 0.008 NA 0.212
#> GSM1269646 3 0.651 0.2461 0.000 0.276 0.524 0.024 NA 0.028
#> GSM1269654 3 0.735 0.0971 0.004 0.100 0.408 0.292 NA 0.004
#> GSM1269662 2 0.791 0.3149 0.004 0.348 0.108 0.272 NA 0.024
#> GSM1269670 2 0.335 0.6837 0.000 0.852 0.024 0.020 NA 0.076
#> GSM1269678 3 0.395 0.2474 0.000 0.000 0.656 0.328 NA 0.000
#> GSM1269692 4 0.586 0.5531 0.016 0.024 0.196 0.640 NA 0.008
#> GSM1269700 3 0.308 0.4962 0.004 0.000 0.852 0.080 NA 0.004
#> GSM1269708 3 0.493 0.4438 0.004 0.044 0.712 0.196 NA 0.012
#> GSM1269714 4 0.405 0.6011 0.020 0.000 0.236 0.728 NA 0.004
#> GSM1269716 4 0.422 0.6005 0.016 0.000 0.232 0.724 NA 0.008
#> GSM1269720 2 0.754 0.5803 0.000 0.492 0.044 0.164 NA 0.136
#> GSM1269722 3 0.366 0.4352 0.000 0.008 0.772 0.192 NA 0.000
#> GSM1269644 1 0.403 0.6955 0.796 0.000 0.000 0.044 NA 0.088
#> GSM1269652 1 0.508 0.5903 0.652 0.000 0.000 0.012 NA 0.228
#> GSM1269660 1 0.534 0.6486 0.696 0.008 0.000 0.064 NA 0.084
#> GSM1269668 1 0.396 0.6856 0.796 0.000 0.048 0.012 NA 0.016
#> GSM1269676 6 0.607 0.6727 0.224 0.012 0.000 0.068 NA 0.608
#> GSM1269684 1 0.435 0.6998 0.760 0.000 0.004 0.024 NA 0.148
#> GSM1269690 1 0.442 0.6834 0.760 0.000 0.000 0.068 NA 0.128
#> GSM1269698 6 0.451 0.7374 0.148 0.076 0.000 0.000 NA 0.744
#> GSM1269706 6 0.459 0.7241 0.180 0.020 0.004 0.012 NA 0.740
#> GSM1269650 2 0.650 0.6476 0.000 0.580 0.040 0.116 NA 0.044
#> GSM1269658 4 0.819 -0.1318 0.004 0.240 0.076 0.380 NA 0.084
#> GSM1269666 3 0.595 0.3749 0.000 0.048 0.608 0.204 NA 0.004
#> GSM1269674 2 0.426 0.6900 0.000 0.796 0.052 0.036 NA 0.024
#> GSM1269682 4 0.461 0.5050 0.028 0.000 0.332 0.624 NA 0.000
#> GSM1269688 3 0.698 -0.1159 0.012 0.384 0.416 0.008 NA 0.064
#> GSM1269696 3 0.608 0.2323 0.000 0.316 0.524 0.016 NA 0.012
#> GSM1269704 2 0.559 0.5660 0.000 0.640 0.224 0.004 NA 0.068
#> GSM1269712 3 0.478 0.3365 0.000 0.008 0.664 0.268 NA 0.008
#> GSM1269718 3 0.722 0.3136 0.040 0.040 0.488 0.164 NA 0.008
#> GSM1269724 3 0.631 0.4598 0.008 0.100 0.616 0.092 NA 0.008
#> GSM1269726 3 0.519 -0.2377 0.012 0.008 0.480 0.460 NA 0.000
#> GSM1269648 1 0.482 0.5865 0.664 0.000 0.000 0.008 NA 0.244
#> GSM1269656 6 0.478 0.5958 0.304 0.016 0.000 0.020 NA 0.644
#> GSM1269664 1 0.487 0.6453 0.720 0.008 0.020 0.024 NA 0.032
#> GSM1269672 1 0.300 0.7147 0.856 0.000 0.000 0.024 NA 0.024
#> GSM1269680 6 0.614 0.7008 0.132 0.120 0.000 0.040 NA 0.648
#> GSM1269686 1 0.393 0.7109 0.800 0.000 0.004 0.020 NA 0.104
#> GSM1269694 1 0.499 0.5339 0.624 0.004 0.000 0.004 NA 0.292
#> GSM1269702 1 0.433 0.6294 0.708 0.000 0.000 0.020 NA 0.240
#> GSM1269710 1 0.467 0.6409 0.704 0.000 0.000 0.012 NA 0.192
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 agent(p) disease.state(p) gender(p) individual(p) k
#> CV:kmeans 84 1.000 1.000 3.81e-19 8.65e-05 2
#> CV:kmeans 76 0.969 0.337 3.14e-17 5.31e-07 3
#> CV:kmeans 68 0.994 0.194 1.14e-14 8.59e-09 4
#> CV:kmeans 53 0.993 0.565 1.83e-11 5.22e-07 5
#> CV:kmeans 56 0.551 0.431 2.01e-11 7.21e-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["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 51941 rows and 84 columns.
#> Top rows (1000, 2000, 3000, 4000, 5000) are extracted by 'CV' method.
#> Subgroups are detected by 'skmeans' method.
#> Performed in total 1250 partitions by row resampling.
#> Best k for subgroups seems to be 2.
#>
#> Following methods can be applied to this 'ConsensusPartition' object:
#> [1] "cola_report" "collect_classes" "collect_plots"
#> [4] "collect_stats" "colnames" "compare_signatures"
#> [7] "consensus_heatmap" "dimension_reduction" "functional_enrichment"
#> [10] "get_anno_col" "get_anno" "get_classes"
#> [13] "get_consensus" "get_matrix" "get_membership"
#> [16] "get_param" "get_signatures" "get_stats"
#> [19] "is_best_k" "is_stable_k" "membership_heatmap"
#> [22] "ncol" "nrow" "plot_ecdf"
#> [25] "rownames" "select_partition_number" "show"
#> [28] "suggest_best_k" "test_to_known_factors"
collect_plots() function collects all the plots made from res for all k (number of partitions)
into one single page to provide an easy and fast comparison between different k.
collect_plots(res)
The plots are:
k and the heatmap of
predicted classes for each k.k.k.k.All the plots in panels can be made by individual functions and they are plotted later in this section.
select_partition_number() produces several plots showing different
statistics for choosing “optimized” k. There are following statistics:
k;k, the area increased is defined as \(A_k - A_{k-1}\).The detailed explanations of these statistics can be found in the cola vignette.
Generally speaking, lower PAC score, higher mean silhouette score or higher
concordance corresponds to better partition. Rand index and Jaccard index
measure how similar the current partition is compared to partition with k-1.
If they are too similar, we won't accept k is better than k-1.
select_partition_number(res)
The numeric values for all these statistics can be obtained by get_stats().
get_stats(res)
#> k 1-PAC mean_silhouette concordance area_increased Rand Jaccard
#> 2 2 0.000000 0.819 0.780 0.5031 0.504 0.504
#> 3 3 0.000974 0.409 0.593 0.3302 0.835 0.673
#> 4 4 0.047387 0.321 0.492 0.1236 0.902 0.747
#> 5 5 0.162610 0.125 0.378 0.0688 0.836 0.563
#> 6 6 0.291139 0.104 0.376 0.0428 0.869 0.534
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
#> GSM1269647 1 0.730 0.824 0.796 0.204
#> GSM1269655 1 0.760 0.837 0.780 0.220
#> GSM1269663 1 0.680 0.837 0.820 0.180
#> GSM1269671 1 0.900 0.766 0.684 0.316
#> GSM1269679 1 0.584 0.825 0.860 0.140
#> GSM1269693 1 0.827 0.802 0.740 0.260
#> GSM1269701 1 0.795 0.816 0.760 0.240
#> GSM1269709 1 0.788 0.819 0.764 0.236
#> GSM1269715 1 0.839 0.808 0.732 0.268
#> GSM1269717 1 0.814 0.817 0.748 0.252
#> GSM1269721 1 0.891 0.758 0.692 0.308
#> GSM1269723 1 0.753 0.838 0.784 0.216
#> GSM1269645 2 0.753 0.827 0.216 0.784
#> GSM1269653 2 0.662 0.852 0.172 0.828
#> GSM1269661 2 0.917 0.717 0.332 0.668
#> GSM1269669 2 0.671 0.850 0.176 0.824
#> GSM1269677 2 0.730 0.832 0.204 0.796
#> GSM1269685 2 0.529 0.854 0.120 0.880
#> GSM1269691 2 0.706 0.847 0.192 0.808
#> GSM1269699 2 0.625 0.843 0.156 0.844
#> GSM1269707 2 0.680 0.834 0.180 0.820
#> GSM1269651 1 0.730 0.824 0.796 0.204
#> GSM1269659 1 0.909 0.748 0.676 0.324
#> GSM1269667 1 0.671 0.826 0.824 0.176
#> GSM1269675 1 0.844 0.812 0.728 0.272
#> GSM1269683 1 0.753 0.830 0.784 0.216
#> GSM1269689 1 0.904 0.767 0.680 0.320
#> GSM1269697 1 0.814 0.827 0.748 0.252
#> GSM1269705 1 0.781 0.824 0.768 0.232
#> GSM1269713 1 0.697 0.841 0.812 0.188
#> GSM1269719 1 0.861 0.799 0.716 0.284
#> GSM1269725 1 0.615 0.831 0.848 0.152
#> GSM1269727 1 0.518 0.823 0.884 0.116
#> GSM1269649 2 0.775 0.823 0.228 0.772
#> GSM1269657 2 0.745 0.830 0.212 0.788
#> GSM1269665 2 0.821 0.799 0.256 0.744
#> GSM1269673 2 0.634 0.851 0.160 0.840
#> GSM1269681 2 0.861 0.774 0.284 0.716
#> GSM1269687 2 0.671 0.855 0.176 0.824
#> GSM1269695 2 0.584 0.852 0.140 0.860
#> GSM1269703 2 0.714 0.846 0.196 0.804
#> GSM1269711 2 0.697 0.849 0.188 0.812
#> GSM1269646 1 0.753 0.830 0.784 0.216
#> GSM1269654 1 0.722 0.838 0.800 0.200
#> GSM1269662 1 0.855 0.802 0.720 0.280
#> GSM1269670 1 0.904 0.736 0.680 0.320
#> GSM1269678 1 0.738 0.826 0.792 0.208
#> GSM1269692 1 0.855 0.801 0.720 0.280
#> GSM1269700 1 0.714 0.836 0.804 0.196
#> GSM1269708 1 0.821 0.824 0.744 0.256
#> GSM1269714 1 0.714 0.834 0.804 0.196
#> GSM1269716 1 0.808 0.814 0.752 0.248
#> GSM1269720 1 0.871 0.786 0.708 0.292
#> GSM1269722 1 0.802 0.827 0.756 0.244
#> GSM1269644 2 0.760 0.844 0.220 0.780
#> GSM1269652 2 0.653 0.855 0.168 0.832
#> GSM1269660 2 0.827 0.793 0.260 0.740
#> GSM1269668 2 0.833 0.782 0.264 0.736
#> GSM1269676 2 0.574 0.854 0.136 0.864
#> GSM1269684 2 0.662 0.846 0.172 0.828
#> GSM1269690 2 0.745 0.822 0.212 0.788
#> GSM1269698 2 0.714 0.833 0.196 0.804
#> GSM1269706 2 0.671 0.846 0.176 0.824
#> GSM1269650 1 0.795 0.816 0.760 0.240
#> GSM1269658 1 0.939 0.664 0.644 0.356
#> GSM1269666 1 0.730 0.833 0.796 0.204
#> GSM1269674 1 0.745 0.826 0.788 0.212
#> GSM1269682 1 0.722 0.835 0.800 0.200
#> GSM1269688 1 0.939 0.713 0.644 0.356
#> GSM1269696 1 0.706 0.834 0.808 0.192
#> GSM1269704 1 0.689 0.819 0.816 0.184
#> GSM1269712 1 0.802 0.824 0.756 0.244
#> GSM1269718 1 0.891 0.774 0.692 0.308
#> GSM1269724 1 0.722 0.834 0.800 0.200
#> GSM1269726 1 0.814 0.823 0.748 0.252
#> GSM1269648 2 0.671 0.857 0.176 0.824
#> GSM1269656 2 0.706 0.847 0.192 0.808
#> GSM1269664 2 0.886 0.738 0.304 0.696
#> GSM1269672 2 0.689 0.849 0.184 0.816
#> GSM1269680 2 0.680 0.836 0.180 0.820
#> GSM1269686 2 0.730 0.838 0.204 0.796
#> GSM1269694 2 0.541 0.853 0.124 0.876
#> GSM1269702 2 0.416 0.840 0.084 0.916
#> GSM1269710 2 0.605 0.859 0.148 0.852
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM1269647 2 0.828 0.2791 0.148 0.628 0.224
#> GSM1269655 2 0.846 -0.0633 0.088 0.476 0.436
#> GSM1269663 3 0.834 0.2381 0.096 0.344 0.560
#> GSM1269671 2 0.854 0.3032 0.172 0.608 0.220
#> GSM1269679 3 0.804 0.2777 0.076 0.352 0.572
#> GSM1269693 3 0.844 0.2826 0.132 0.268 0.600
#> GSM1269701 3 0.903 0.1454 0.136 0.388 0.476
#> GSM1269709 2 0.943 -0.0684 0.176 0.412 0.412
#> GSM1269715 3 0.787 0.3334 0.112 0.236 0.652
#> GSM1269717 3 0.770 0.3705 0.116 0.212 0.672
#> GSM1269721 2 0.909 0.2393 0.168 0.528 0.304
#> GSM1269723 3 0.835 0.2455 0.092 0.360 0.548
#> GSM1269645 1 0.899 0.6381 0.544 0.164 0.292
#> GSM1269653 1 0.710 0.7384 0.724 0.148 0.128
#> GSM1269661 1 0.884 0.6486 0.580 0.208 0.212
#> GSM1269669 1 0.745 0.7327 0.696 0.120 0.184
#> GSM1269677 1 0.807 0.6852 0.648 0.144 0.208
#> GSM1269685 1 0.642 0.7454 0.752 0.068 0.180
#> GSM1269691 1 0.825 0.7104 0.628 0.140 0.232
#> GSM1269699 1 0.703 0.7090 0.724 0.172 0.104
#> GSM1269707 1 0.761 0.7231 0.688 0.164 0.148
#> GSM1269651 2 0.814 0.2770 0.112 0.620 0.268
#> GSM1269659 2 0.947 0.1699 0.188 0.456 0.356
#> GSM1269667 2 0.889 -0.1372 0.120 0.452 0.428
#> GSM1269675 2 0.843 0.2483 0.120 0.588 0.292
#> GSM1269683 3 0.781 0.3434 0.092 0.268 0.640
#> GSM1269689 2 0.939 0.2282 0.224 0.508 0.268
#> GSM1269697 2 0.858 0.0882 0.108 0.536 0.356
#> GSM1269705 2 0.794 0.3064 0.124 0.652 0.224
#> GSM1269713 3 0.857 0.1248 0.096 0.428 0.476
#> GSM1269719 3 0.916 0.0640 0.148 0.388 0.464
#> GSM1269725 2 0.832 -0.0825 0.080 0.496 0.424
#> GSM1269727 3 0.836 0.2288 0.084 0.412 0.504
#> GSM1269649 1 0.835 0.6556 0.628 0.188 0.184
#> GSM1269657 1 0.782 0.7029 0.672 0.176 0.152
#> GSM1269665 1 0.901 0.5744 0.536 0.160 0.304
#> GSM1269673 1 0.713 0.7426 0.716 0.104 0.180
#> GSM1269681 1 0.900 0.4517 0.504 0.356 0.140
#> GSM1269687 1 0.867 0.6638 0.580 0.148 0.272
#> GSM1269695 1 0.586 0.7423 0.796 0.120 0.084
#> GSM1269703 1 0.778 0.7197 0.668 0.124 0.208
#> GSM1269711 1 0.764 0.7124 0.684 0.180 0.136
#> GSM1269646 2 0.800 0.2860 0.128 0.648 0.224
#> GSM1269654 3 0.873 0.1081 0.108 0.424 0.468
#> GSM1269662 2 0.867 0.0546 0.104 0.480 0.416
#> GSM1269670 2 0.770 0.3378 0.180 0.680 0.140
#> GSM1269678 3 0.788 0.2947 0.080 0.308 0.612
#> GSM1269692 3 0.801 0.2689 0.116 0.244 0.640
#> GSM1269700 3 0.849 0.2405 0.108 0.336 0.556
#> GSM1269708 3 0.920 0.1468 0.160 0.352 0.488
#> GSM1269714 3 0.745 0.3576 0.076 0.260 0.664
#> GSM1269716 3 0.744 0.3652 0.108 0.200 0.692
#> GSM1269720 2 0.873 0.2922 0.208 0.592 0.200
#> GSM1269722 3 0.882 0.1870 0.132 0.336 0.532
#> GSM1269644 1 0.804 0.7102 0.648 0.136 0.216
#> GSM1269652 1 0.646 0.7401 0.764 0.116 0.120
#> GSM1269660 1 0.908 0.6015 0.552 0.236 0.212
#> GSM1269668 1 0.839 0.6806 0.612 0.140 0.248
#> GSM1269676 1 0.644 0.7297 0.756 0.168 0.076
#> GSM1269684 1 0.787 0.7153 0.660 0.124 0.216
#> GSM1269690 1 0.727 0.7245 0.684 0.076 0.240
#> GSM1269698 1 0.749 0.7029 0.680 0.224 0.096
#> GSM1269706 1 0.710 0.7235 0.724 0.148 0.128
#> GSM1269650 2 0.891 0.2505 0.164 0.556 0.280
#> GSM1269658 3 0.946 -0.1047 0.180 0.392 0.428
#> GSM1269666 2 0.879 -0.1415 0.112 0.448 0.440
#> GSM1269674 2 0.880 0.1863 0.124 0.520 0.356
#> GSM1269682 3 0.835 0.3272 0.120 0.280 0.600
#> GSM1269688 2 0.924 0.2427 0.232 0.532 0.236
#> GSM1269696 2 0.805 0.1746 0.096 0.612 0.292
#> GSM1269704 2 0.791 0.3239 0.148 0.664 0.188
#> GSM1269712 3 0.818 0.3123 0.092 0.324 0.584
#> GSM1269718 2 0.971 0.0567 0.236 0.440 0.324
#> GSM1269724 2 0.846 0.0457 0.092 0.512 0.396
#> GSM1269726 3 0.869 0.3074 0.164 0.248 0.588
#> GSM1269648 1 0.618 0.7435 0.780 0.100 0.120
#> GSM1269656 1 0.745 0.7234 0.700 0.148 0.152
#> GSM1269664 1 0.943 0.5599 0.504 0.232 0.264
#> GSM1269672 1 0.730 0.7333 0.692 0.088 0.220
#> GSM1269680 1 0.819 0.6787 0.632 0.232 0.136
#> GSM1269686 1 0.794 0.7166 0.644 0.112 0.244
#> GSM1269694 1 0.664 0.7386 0.752 0.140 0.108
#> GSM1269702 1 0.627 0.7471 0.772 0.088 0.140
#> GSM1269710 1 0.638 0.7416 0.768 0.128 0.104
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM1269647 2 0.819 0.23983 0.088 0.572 0.156 0.184
#> GSM1269655 3 0.847 0.08339 0.028 0.340 0.396 0.236
#> GSM1269663 3 0.825 0.22766 0.036 0.244 0.500 0.220
#> GSM1269671 2 0.797 0.28843 0.148 0.604 0.132 0.116
#> GSM1269679 3 0.851 0.22250 0.048 0.260 0.476 0.216
#> GSM1269693 3 0.718 0.31906 0.072 0.128 0.664 0.136
#> GSM1269701 3 0.979 0.10869 0.156 0.272 0.312 0.260
#> GSM1269709 3 0.949 0.13394 0.116 0.280 0.368 0.236
#> GSM1269715 3 0.787 0.29567 0.096 0.132 0.608 0.164
#> GSM1269717 3 0.781 0.30351 0.096 0.120 0.612 0.172
#> GSM1269721 2 0.915 0.15118 0.092 0.424 0.264 0.220
#> GSM1269723 3 0.870 0.14525 0.068 0.316 0.448 0.168
#> GSM1269645 1 0.939 0.39639 0.364 0.196 0.116 0.324
#> GSM1269653 1 0.765 0.58739 0.592 0.080 0.080 0.248
#> GSM1269661 1 0.912 0.48442 0.444 0.112 0.176 0.268
#> GSM1269669 1 0.726 0.57210 0.624 0.040 0.116 0.220
#> GSM1269677 1 0.858 0.52431 0.496 0.160 0.076 0.268
#> GSM1269685 1 0.620 0.61756 0.716 0.040 0.072 0.172
#> GSM1269691 1 0.792 0.58276 0.564 0.064 0.116 0.256
#> GSM1269699 1 0.874 0.46393 0.460 0.188 0.068 0.284
#> GSM1269707 1 0.881 0.46283 0.384 0.164 0.072 0.380
#> GSM1269651 2 0.843 0.24535 0.084 0.540 0.192 0.184
#> GSM1269659 3 0.955 -0.02734 0.132 0.304 0.356 0.208
#> GSM1269667 2 0.845 -0.11793 0.024 0.372 0.356 0.248
#> GSM1269675 2 0.806 0.26474 0.096 0.584 0.196 0.124
#> GSM1269683 3 0.839 0.27633 0.088 0.228 0.540 0.144
#> GSM1269689 2 0.942 0.15081 0.148 0.424 0.208 0.220
#> GSM1269697 2 0.865 0.10776 0.068 0.484 0.256 0.192
#> GSM1269705 2 0.826 0.24966 0.072 0.552 0.200 0.176
#> GSM1269713 3 0.881 0.14070 0.052 0.340 0.388 0.220
#> GSM1269719 2 0.924 0.00858 0.076 0.332 0.316 0.276
#> GSM1269725 3 0.893 0.13355 0.068 0.336 0.396 0.200
#> GSM1269727 3 0.799 0.26669 0.048 0.232 0.556 0.164
#> GSM1269649 1 0.871 0.51289 0.496 0.160 0.092 0.252
#> GSM1269657 1 0.847 0.51646 0.520 0.136 0.088 0.256
#> GSM1269665 1 0.907 0.44515 0.416 0.112 0.148 0.324
#> GSM1269673 1 0.773 0.59414 0.564 0.072 0.080 0.284
#> GSM1269681 2 0.915 -0.22577 0.332 0.348 0.072 0.248
#> GSM1269687 1 0.870 0.54724 0.492 0.104 0.132 0.272
#> GSM1269695 1 0.651 0.60844 0.692 0.076 0.044 0.188
#> GSM1269703 1 0.808 0.56466 0.556 0.072 0.124 0.248
#> GSM1269711 1 0.740 0.56617 0.588 0.076 0.056 0.280
#> GSM1269646 2 0.853 0.16866 0.104 0.536 0.208 0.152
#> GSM1269654 3 0.910 0.15992 0.080 0.288 0.408 0.224
#> GSM1269662 2 0.909 0.04520 0.084 0.372 0.352 0.192
#> GSM1269670 2 0.791 0.28888 0.112 0.608 0.128 0.152
#> GSM1269678 3 0.876 0.25851 0.116 0.192 0.516 0.176
#> GSM1269692 3 0.912 0.17920 0.124 0.228 0.464 0.184
#> GSM1269700 3 0.902 0.16066 0.084 0.312 0.416 0.188
#> GSM1269708 3 0.946 0.17694 0.136 0.252 0.404 0.208
#> GSM1269714 3 0.782 0.31236 0.092 0.132 0.612 0.164
#> GSM1269716 3 0.715 0.33110 0.104 0.100 0.672 0.124
#> GSM1269720 2 0.889 0.20754 0.108 0.492 0.204 0.196
#> GSM1269722 3 0.900 0.07332 0.064 0.328 0.376 0.232
#> GSM1269644 1 0.846 0.57122 0.520 0.108 0.108 0.264
#> GSM1269652 1 0.634 0.60859 0.680 0.060 0.032 0.228
#> GSM1269660 1 0.980 0.23874 0.332 0.188 0.200 0.280
#> GSM1269668 1 0.890 0.47734 0.452 0.092 0.164 0.292
#> GSM1269676 1 0.790 0.54738 0.556 0.128 0.052 0.264
#> GSM1269684 1 0.839 0.54779 0.464 0.072 0.116 0.348
#> GSM1269690 1 0.791 0.56524 0.576 0.060 0.140 0.224
#> GSM1269698 1 0.820 0.49426 0.464 0.192 0.028 0.316
#> GSM1269706 1 0.818 0.53353 0.532 0.164 0.052 0.252
#> GSM1269650 2 0.805 0.24576 0.060 0.568 0.184 0.188
#> GSM1269658 3 0.936 -0.05204 0.104 0.332 0.352 0.212
#> GSM1269666 3 0.906 0.14437 0.080 0.296 0.412 0.212
#> GSM1269674 2 0.816 0.20282 0.076 0.552 0.244 0.128
#> GSM1269682 3 0.812 0.27756 0.096 0.164 0.584 0.156
#> GSM1269688 2 0.926 0.20924 0.140 0.448 0.184 0.228
#> GSM1269696 2 0.799 0.14912 0.036 0.540 0.232 0.192
#> GSM1269704 2 0.880 0.20789 0.100 0.500 0.208 0.192
#> GSM1269712 3 0.849 0.28395 0.080 0.184 0.528 0.208
#> GSM1269718 3 0.952 0.06126 0.108 0.308 0.320 0.264
#> GSM1269724 2 0.883 -0.01938 0.052 0.388 0.332 0.228
#> GSM1269726 3 0.844 0.28936 0.100 0.188 0.548 0.164
#> GSM1269648 1 0.574 0.60764 0.768 0.080 0.068 0.084
#> GSM1269656 1 0.811 0.55027 0.500 0.104 0.064 0.332
#> GSM1269664 1 0.906 0.42189 0.376 0.100 0.156 0.368
#> GSM1269672 1 0.822 0.55972 0.532 0.060 0.152 0.256
#> GSM1269680 1 0.838 0.45920 0.452 0.216 0.032 0.300
#> GSM1269686 1 0.817 0.54991 0.540 0.064 0.140 0.256
#> GSM1269694 1 0.727 0.59378 0.624 0.088 0.056 0.232
#> GSM1269702 1 0.682 0.61244 0.628 0.044 0.056 0.272
#> GSM1269710 1 0.702 0.60226 0.656 0.072 0.068 0.204
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM1269647 5 0.924 0.02698 0.048 0.176 0.280 0.192 0.304
#> GSM1269655 3 0.914 0.14220 0.072 0.088 0.304 0.260 0.276
#> GSM1269663 4 0.881 0.07846 0.072 0.088 0.188 0.424 0.228
#> GSM1269671 5 0.799 0.19377 0.068 0.152 0.112 0.112 0.556
#> GSM1269679 4 0.722 0.12620 0.060 0.048 0.164 0.612 0.116
#> GSM1269693 4 0.896 0.09932 0.092 0.104 0.236 0.420 0.148
#> GSM1269701 4 0.918 0.04460 0.148 0.088 0.260 0.376 0.128
#> GSM1269709 4 0.915 0.05532 0.108 0.124 0.172 0.420 0.176
#> GSM1269715 4 0.869 0.10755 0.152 0.092 0.252 0.436 0.068
#> GSM1269717 4 0.888 0.04016 0.104 0.108 0.300 0.392 0.096
#> GSM1269721 5 0.878 0.14876 0.064 0.192 0.148 0.144 0.452
#> GSM1269723 4 0.887 0.10957 0.064 0.148 0.180 0.440 0.168
#> GSM1269645 1 0.843 0.21475 0.480 0.192 0.176 0.048 0.104
#> GSM1269653 1 0.769 0.15753 0.496 0.308 0.088 0.064 0.044
#> GSM1269661 1 0.925 0.07562 0.348 0.276 0.164 0.080 0.132
#> GSM1269669 1 0.674 0.32545 0.656 0.100 0.144 0.052 0.048
#> GSM1269677 2 0.799 0.19366 0.284 0.476 0.104 0.028 0.108
#> GSM1269685 1 0.761 0.18187 0.516 0.260 0.136 0.068 0.020
#> GSM1269691 1 0.832 0.13018 0.432 0.288 0.172 0.048 0.060
#> GSM1269699 2 0.834 0.21407 0.312 0.424 0.068 0.052 0.144
#> GSM1269707 2 0.717 0.24044 0.232 0.580 0.052 0.032 0.104
#> GSM1269651 5 0.681 0.11864 0.020 0.072 0.172 0.104 0.632
#> GSM1269659 5 0.944 0.07337 0.072 0.272 0.164 0.196 0.296
#> GSM1269667 4 0.846 0.00268 0.052 0.052 0.288 0.396 0.212
#> GSM1269675 5 0.797 0.10560 0.056 0.084 0.260 0.092 0.508
#> GSM1269683 4 0.916 0.05664 0.108 0.076 0.244 0.364 0.208
#> GSM1269689 5 0.896 0.03847 0.080 0.072 0.320 0.196 0.332
#> GSM1269697 5 0.881 -0.02154 0.068 0.060 0.228 0.320 0.324
#> GSM1269705 5 0.793 0.15006 0.052 0.136 0.076 0.200 0.536
#> GSM1269713 4 0.861 0.05384 0.064 0.068 0.228 0.432 0.208
#> GSM1269719 4 0.951 -0.06608 0.100 0.116 0.256 0.280 0.248
#> GSM1269725 4 0.820 -0.01997 0.036 0.072 0.248 0.464 0.180
#> GSM1269727 4 0.806 0.14537 0.084 0.072 0.156 0.544 0.144
#> GSM1269649 1 0.862 0.14341 0.472 0.216 0.128 0.092 0.092
#> GSM1269657 2 0.804 0.29073 0.208 0.516 0.088 0.040 0.148
#> GSM1269665 1 0.910 0.17112 0.424 0.184 0.164 0.132 0.096
#> GSM1269673 1 0.698 0.27256 0.608 0.204 0.096 0.068 0.024
#> GSM1269681 2 0.858 0.14508 0.136 0.388 0.124 0.036 0.316
#> GSM1269687 1 0.889 0.15191 0.420 0.252 0.132 0.116 0.080
#> GSM1269695 1 0.768 0.14366 0.520 0.280 0.080 0.060 0.060
#> GSM1269703 1 0.843 0.17856 0.496 0.196 0.140 0.072 0.096
#> GSM1269711 1 0.838 0.10542 0.428 0.260 0.204 0.068 0.040
#> GSM1269646 5 0.915 0.06874 0.084 0.100 0.268 0.188 0.360
#> GSM1269654 5 0.909 -0.27894 0.060 0.096 0.276 0.268 0.300
#> GSM1269662 5 0.897 0.05546 0.052 0.144 0.252 0.164 0.388
#> GSM1269670 5 0.830 0.19152 0.068 0.180 0.108 0.128 0.516
#> GSM1269678 4 0.828 0.13793 0.064 0.056 0.232 0.476 0.172
#> GSM1269692 4 0.951 0.03835 0.108 0.120 0.272 0.304 0.196
#> GSM1269700 4 0.852 0.04887 0.060 0.056 0.288 0.404 0.192
#> GSM1269708 4 0.892 0.06792 0.084 0.144 0.172 0.448 0.152
#> GSM1269714 4 0.875 0.14094 0.080 0.076 0.260 0.420 0.164
#> GSM1269716 4 0.839 0.08058 0.136 0.072 0.248 0.472 0.072
#> GSM1269720 5 0.825 0.20488 0.052 0.236 0.120 0.100 0.492
#> GSM1269722 4 0.923 0.03780 0.088 0.108 0.216 0.368 0.220
#> GSM1269644 1 0.888 0.11172 0.388 0.296 0.144 0.084 0.088
#> GSM1269652 1 0.810 0.04771 0.408 0.352 0.140 0.036 0.064
#> GSM1269660 1 0.917 0.15204 0.388 0.184 0.228 0.092 0.108
#> GSM1269668 1 0.808 0.28031 0.540 0.140 0.136 0.132 0.052
#> GSM1269676 2 0.800 0.25584 0.236 0.512 0.100 0.040 0.112
#> GSM1269684 1 0.858 0.16317 0.452 0.252 0.128 0.104 0.064
#> GSM1269690 1 0.830 0.21086 0.488 0.228 0.148 0.072 0.064
#> GSM1269698 2 0.808 0.29154 0.188 0.520 0.092 0.044 0.156
#> GSM1269706 2 0.811 0.19990 0.272 0.484 0.076 0.052 0.116
#> GSM1269650 5 0.862 0.14266 0.064 0.188 0.116 0.160 0.472
#> GSM1269658 5 0.957 0.06875 0.088 0.208 0.184 0.204 0.316
#> GSM1269666 3 0.847 0.09437 0.052 0.052 0.400 0.272 0.224
#> GSM1269674 5 0.796 0.12840 0.032 0.084 0.220 0.152 0.512
#> GSM1269682 4 0.873 0.15374 0.088 0.076 0.244 0.436 0.156
#> GSM1269688 5 0.934 0.13028 0.096 0.148 0.204 0.172 0.380
#> GSM1269696 5 0.852 0.00899 0.036 0.084 0.252 0.216 0.412
#> GSM1269704 5 0.842 0.12114 0.048 0.152 0.096 0.240 0.464
#> GSM1269712 4 0.750 0.11471 0.028 0.072 0.220 0.556 0.124
#> GSM1269718 3 0.938 0.02421 0.076 0.160 0.348 0.196 0.220
#> GSM1269724 4 0.866 -0.06101 0.036 0.092 0.256 0.384 0.232
#> GSM1269726 4 0.907 0.14433 0.128 0.112 0.212 0.424 0.124
#> GSM1269648 1 0.788 0.19131 0.532 0.228 0.108 0.052 0.080
#> GSM1269656 2 0.765 0.22802 0.260 0.536 0.076 0.056 0.072
#> GSM1269664 1 0.882 0.24588 0.460 0.172 0.160 0.124 0.084
#> GSM1269672 1 0.759 0.28512 0.564 0.176 0.156 0.068 0.036
#> GSM1269680 2 0.793 0.22183 0.272 0.456 0.096 0.008 0.168
#> GSM1269686 1 0.778 0.25103 0.556 0.196 0.116 0.076 0.056
#> GSM1269694 1 0.811 0.12005 0.480 0.272 0.096 0.040 0.112
#> GSM1269702 1 0.700 0.06455 0.452 0.404 0.096 0.016 0.032
#> GSM1269710 1 0.749 0.20047 0.536 0.276 0.072 0.044 0.072
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM1269647 3 0.843 0.061551 0.036 0.228 0.428 0.068 0.112 0.128
#> GSM1269655 3 0.862 0.011471 0.036 0.224 0.324 0.272 0.096 0.048
#> GSM1269663 4 0.859 0.041431 0.032 0.236 0.252 0.336 0.088 0.056
#> GSM1269671 2 0.794 0.172500 0.052 0.520 0.148 0.052 0.096 0.132
#> GSM1269679 3 0.824 0.027800 0.052 0.112 0.404 0.284 0.120 0.028
#> GSM1269693 4 0.758 0.199038 0.104 0.068 0.076 0.564 0.140 0.048
#> GSM1269701 3 0.874 0.105752 0.172 0.112 0.412 0.168 0.096 0.040
#> GSM1269709 4 0.939 0.019644 0.084 0.144 0.260 0.288 0.132 0.092
#> GSM1269715 4 0.793 0.187037 0.124 0.068 0.096 0.520 0.152 0.040
#> GSM1269717 4 0.768 0.187872 0.076 0.056 0.136 0.544 0.144 0.044
#> GSM1269721 2 0.911 0.217863 0.072 0.368 0.080 0.136 0.148 0.196
#> GSM1269723 3 0.864 0.039921 0.036 0.168 0.340 0.284 0.124 0.048
#> GSM1269645 5 0.885 0.085473 0.280 0.116 0.076 0.052 0.348 0.128
#> GSM1269653 1 0.884 0.057607 0.316 0.076 0.084 0.048 0.184 0.292
#> GSM1269661 6 0.953 -0.080379 0.232 0.080 0.108 0.116 0.212 0.252
#> GSM1269669 1 0.657 0.092749 0.612 0.032 0.044 0.048 0.212 0.052
#> GSM1269677 6 0.659 0.260984 0.144 0.060 0.016 0.040 0.112 0.628
#> GSM1269685 1 0.746 0.157545 0.484 0.008 0.040 0.072 0.148 0.248
#> GSM1269691 1 0.869 0.015852 0.332 0.032 0.072 0.088 0.200 0.276
#> GSM1269699 6 0.744 0.227689 0.224 0.116 0.048 0.012 0.084 0.516
#> GSM1269707 6 0.803 0.176023 0.192 0.128 0.032 0.036 0.140 0.472
#> GSM1269651 2 0.840 0.210265 0.020 0.444 0.152 0.148 0.088 0.148
#> GSM1269659 2 0.946 0.136419 0.052 0.272 0.116 0.204 0.160 0.196
#> GSM1269667 3 0.858 0.013654 0.048 0.144 0.328 0.288 0.168 0.024
#> GSM1269675 2 0.834 0.132343 0.048 0.456 0.168 0.088 0.180 0.060
#> GSM1269683 4 0.894 0.088150 0.088 0.200 0.176 0.324 0.192 0.020
#> GSM1269689 3 0.945 0.038439 0.120 0.256 0.288 0.088 0.132 0.116
#> GSM1269697 3 0.894 0.039194 0.052 0.276 0.312 0.184 0.116 0.060
#> GSM1269705 2 0.774 0.159705 0.048 0.532 0.172 0.112 0.044 0.092
#> GSM1269713 3 0.809 0.124275 0.036 0.128 0.476 0.212 0.084 0.064
#> GSM1269719 4 0.905 0.052995 0.056 0.220 0.168 0.352 0.128 0.076
#> GSM1269725 3 0.762 0.143761 0.040 0.120 0.544 0.160 0.092 0.044
#> GSM1269727 4 0.823 0.044677 0.036 0.200 0.244 0.376 0.132 0.012
#> GSM1269649 1 0.815 0.002979 0.488 0.064 0.052 0.080 0.184 0.132
#> GSM1269657 6 0.708 0.284280 0.144 0.080 0.048 0.040 0.084 0.604
#> GSM1269665 1 0.868 -0.055631 0.328 0.056 0.048 0.124 0.320 0.124
#> GSM1269673 1 0.834 0.009015 0.396 0.032 0.048 0.084 0.240 0.200
#> GSM1269681 6 0.850 0.182866 0.100 0.248 0.076 0.032 0.140 0.404
#> GSM1269687 1 0.856 0.029613 0.360 0.048 0.068 0.064 0.172 0.288
#> GSM1269695 1 0.734 0.142844 0.520 0.080 0.024 0.024 0.120 0.232
#> GSM1269703 1 0.775 0.040080 0.480 0.044 0.040 0.052 0.248 0.136
#> GSM1269711 1 0.877 0.103504 0.420 0.080 0.120 0.056 0.148 0.176
#> GSM1269646 3 0.873 0.071335 0.072 0.240 0.400 0.108 0.116 0.064
#> GSM1269654 4 0.862 0.000796 0.028 0.184 0.296 0.328 0.096 0.068
#> GSM1269662 2 0.891 0.110683 0.052 0.368 0.100 0.236 0.156 0.088
#> GSM1269670 2 0.812 0.171437 0.052 0.500 0.152 0.060 0.092 0.144
#> GSM1269678 4 0.887 0.036324 0.076 0.144 0.280 0.296 0.180 0.024
#> GSM1269692 4 0.908 0.097617 0.136 0.144 0.084 0.328 0.256 0.052
#> GSM1269700 3 0.810 0.095774 0.056 0.148 0.460 0.224 0.076 0.036
#> GSM1269708 4 0.935 0.060197 0.100 0.124 0.228 0.328 0.128 0.092
#> GSM1269714 4 0.708 0.209801 0.048 0.096 0.116 0.608 0.096 0.036
#> GSM1269716 4 0.724 0.211409 0.068 0.116 0.112 0.592 0.084 0.028
#> GSM1269720 2 0.880 0.225926 0.048 0.368 0.120 0.068 0.144 0.252
#> GSM1269722 3 0.891 0.088556 0.036 0.212 0.324 0.188 0.188 0.052
#> GSM1269644 5 0.862 -0.021438 0.276 0.024 0.056 0.100 0.284 0.260
#> GSM1269652 1 0.788 0.152736 0.464 0.068 0.060 0.028 0.120 0.260
#> GSM1269660 5 0.921 0.109311 0.204 0.100 0.076 0.092 0.340 0.188
#> GSM1269668 1 0.793 0.039759 0.460 0.024 0.108 0.048 0.252 0.108
#> GSM1269676 6 0.703 0.255936 0.164 0.052 0.060 0.024 0.108 0.592
#> GSM1269684 1 0.879 0.027337 0.328 0.040 0.052 0.120 0.256 0.204
#> GSM1269690 1 0.861 0.022000 0.340 0.028 0.052 0.120 0.188 0.272
#> GSM1269698 6 0.779 0.297556 0.128 0.152 0.076 0.028 0.088 0.528
#> GSM1269706 6 0.762 0.197998 0.268 0.072 0.040 0.044 0.084 0.492
#> GSM1269650 2 0.880 0.086164 0.048 0.364 0.260 0.124 0.068 0.136
#> GSM1269658 2 0.963 0.134802 0.084 0.268 0.104 0.172 0.188 0.184
#> GSM1269666 3 0.863 0.031756 0.044 0.148 0.368 0.240 0.164 0.036
#> GSM1269674 2 0.776 0.171066 0.056 0.532 0.088 0.184 0.088 0.052
#> GSM1269682 4 0.867 0.132928 0.052 0.164 0.176 0.412 0.144 0.052
#> GSM1269688 2 0.851 0.054029 0.052 0.420 0.240 0.120 0.060 0.108
#> GSM1269696 3 0.806 0.083224 0.048 0.288 0.436 0.108 0.056 0.064
#> GSM1269704 2 0.841 0.093034 0.076 0.432 0.256 0.084 0.076 0.076
#> GSM1269712 4 0.833 0.054539 0.056 0.108 0.304 0.392 0.096 0.044
#> GSM1269718 3 0.939 0.000841 0.108 0.100 0.272 0.208 0.240 0.072
#> GSM1269724 3 0.913 0.044370 0.044 0.216 0.312 0.168 0.188 0.072
#> GSM1269726 4 0.852 0.073939 0.072 0.084 0.228 0.416 0.156 0.044
#> GSM1269648 1 0.749 0.175396 0.532 0.068 0.040 0.048 0.088 0.224
#> GSM1269656 6 0.730 0.229407 0.160 0.104 0.036 0.048 0.076 0.576
#> GSM1269664 5 0.869 0.106343 0.244 0.072 0.092 0.068 0.404 0.120
#> GSM1269672 1 0.793 0.018641 0.448 0.044 0.040 0.048 0.244 0.176
#> GSM1269680 6 0.803 0.266138 0.152 0.140 0.076 0.032 0.096 0.504
#> GSM1269686 1 0.817 0.024363 0.384 0.024 0.068 0.052 0.288 0.184
#> GSM1269694 1 0.738 0.119196 0.516 0.072 0.024 0.032 0.116 0.240
#> GSM1269702 1 0.717 0.042407 0.400 0.032 0.012 0.040 0.124 0.392
#> GSM1269710 1 0.725 0.132528 0.548 0.020 0.064 0.052 0.100 0.216
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 agent(p) disease.state(p) gender(p) individual(p) k
#> CV:skmeans 84 1 1 3.81e-19 8.65e-05 2
#> CV:skmeans 35 NA NA NA NA 3
#> CV:skmeans 25 NA NA NA NA 4
#> CV:skmeans 0 NA NA NA NA 5
#> CV:skmeans 0 NA NA NA 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 51941 rows and 84 columns.
#> Top rows (1000, 2000, 3000, 4000, 5000) are extracted by 'CV' method.
#> Subgroups are detected by 'pam' method.
#> Performed in total 1250 partitions by row resampling.
#> Best k for subgroups seems to be 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 0.136 0.623 0.809 0.4622 0.523 0.523
#> 3 3 0.132 0.345 0.658 0.3164 0.902 0.816
#> 4 4 0.187 0.266 0.612 0.1364 0.863 0.716
#> 5 5 0.219 0.245 0.577 0.0573 0.882 0.710
#> 6 6 0.253 0.251 0.544 0.0378 0.925 0.769
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
#> GSM1269647 1 0.2423 0.7212 0.960 0.040
#> GSM1269655 1 0.8555 0.6922 0.720 0.280
#> GSM1269663 1 0.8443 0.7025 0.728 0.272
#> GSM1269671 2 0.7602 0.6878 0.220 0.780
#> GSM1269679 1 0.9460 0.5752 0.636 0.364
#> GSM1269693 2 0.9866 0.1551 0.432 0.568
#> GSM1269701 1 0.3431 0.7115 0.936 0.064
#> GSM1269709 2 0.7602 0.6829 0.220 0.780
#> GSM1269715 2 0.9988 -0.1187 0.480 0.520
#> GSM1269717 2 0.9580 0.3681 0.380 0.620
#> GSM1269721 1 0.8081 0.7050 0.752 0.248
#> GSM1269723 1 0.6887 0.7424 0.816 0.184
#> GSM1269645 2 0.6712 0.7008 0.176 0.824
#> GSM1269653 2 0.6247 0.6912 0.156 0.844
#> GSM1269661 2 0.4939 0.7536 0.108 0.892
#> GSM1269669 2 0.2603 0.7704 0.044 0.956
#> GSM1269677 2 0.0938 0.7730 0.012 0.988
#> GSM1269685 2 0.1633 0.7744 0.024 0.976
#> GSM1269691 2 0.1633 0.7722 0.024 0.976
#> GSM1269699 2 0.1414 0.7758 0.020 0.980
#> GSM1269707 2 0.2043 0.7735 0.032 0.968
#> GSM1269651 1 0.5408 0.7478 0.876 0.124
#> GSM1269659 2 0.9044 0.5307 0.320 0.680
#> GSM1269667 1 0.9977 0.2085 0.528 0.472
#> GSM1269675 1 0.7453 0.7129 0.788 0.212
#> GSM1269683 1 0.7745 0.7267 0.772 0.228
#> GSM1269689 1 0.4161 0.7195 0.916 0.084
#> GSM1269697 1 0.8555 0.6361 0.720 0.280
#> GSM1269705 1 0.6623 0.7388 0.828 0.172
#> GSM1269713 1 0.4431 0.7382 0.908 0.092
#> GSM1269719 2 0.9754 0.2089 0.408 0.592
#> GSM1269725 1 0.9491 0.5570 0.632 0.368
#> GSM1269727 1 0.1414 0.7130 0.980 0.020
#> GSM1269649 2 0.6531 0.7002 0.168 0.832
#> GSM1269657 2 0.4298 0.7572 0.088 0.912
#> GSM1269665 2 0.6801 0.6865 0.180 0.820
#> GSM1269673 2 0.0376 0.7690 0.004 0.996
#> GSM1269681 2 0.4161 0.7626 0.084 0.916
#> GSM1269687 2 0.0938 0.7738 0.012 0.988
#> GSM1269695 2 0.0938 0.7731 0.012 0.988
#> GSM1269703 2 0.3274 0.7697 0.060 0.940
#> GSM1269711 2 0.9427 0.3753 0.360 0.640
#> GSM1269646 1 0.3584 0.7181 0.932 0.068
#> GSM1269654 2 0.8327 0.5822 0.264 0.736
#> GSM1269662 2 0.9993 -0.0832 0.484 0.516
#> GSM1269670 2 0.9815 0.2032 0.420 0.580
#> GSM1269678 1 0.8267 0.6939 0.740 0.260
#> GSM1269692 2 0.9286 0.3964 0.344 0.656
#> GSM1269700 1 0.9795 0.4083 0.584 0.416
#> GSM1269708 2 0.8813 0.5472 0.300 0.700
#> GSM1269714 2 0.9491 0.3550 0.368 0.632
#> GSM1269716 1 0.9754 0.4486 0.592 0.408
#> GSM1269720 2 0.9661 0.3597 0.392 0.608
#> GSM1269722 1 0.2423 0.7105 0.960 0.040
#> GSM1269644 2 0.8713 0.4601 0.292 0.708
#> GSM1269652 2 0.1633 0.7734 0.024 0.976
#> GSM1269660 2 0.5059 0.7495 0.112 0.888
#> GSM1269668 2 0.7376 0.6460 0.208 0.792
#> GSM1269676 2 0.2603 0.7726 0.044 0.956
#> GSM1269684 2 0.1184 0.7751 0.016 0.984
#> GSM1269690 2 0.0672 0.7685 0.008 0.992
#> GSM1269698 2 0.7299 0.6909 0.204 0.796
#> GSM1269706 2 0.6438 0.6979 0.164 0.836
#> GSM1269650 1 0.9977 0.2116 0.528 0.472
#> GSM1269658 1 0.6801 0.7438 0.820 0.180
#> GSM1269666 1 0.9833 0.3840 0.576 0.424
#> GSM1269674 1 0.3584 0.7354 0.932 0.068
#> GSM1269682 2 0.9552 0.3512 0.376 0.624
#> GSM1269688 1 0.7376 0.6987 0.792 0.208
#> GSM1269696 1 0.9000 0.6512 0.684 0.316
#> GSM1269704 1 0.9427 0.5269 0.640 0.360
#> GSM1269712 2 0.9661 0.3418 0.392 0.608
#> GSM1269718 1 0.5408 0.7525 0.876 0.124
#> GSM1269724 1 0.9460 0.5706 0.636 0.364
#> GSM1269726 1 0.6438 0.7469 0.836 0.164
#> GSM1269648 2 0.0000 0.7686 0.000 1.000
#> GSM1269656 2 0.0000 0.7686 0.000 1.000
#> GSM1269664 2 0.4562 0.7584 0.096 0.904
#> GSM1269672 2 0.1184 0.7722 0.016 0.984
#> GSM1269680 2 0.3114 0.7683 0.056 0.944
#> GSM1269686 2 0.0376 0.7706 0.004 0.996
#> GSM1269694 2 0.2423 0.7676 0.040 0.960
#> GSM1269702 2 0.0000 0.7686 0.000 1.000
#> GSM1269710 2 0.0938 0.7722 0.012 0.988
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM1269647 3 0.4277 0.29169 0.016 0.132 0.852
#> GSM1269655 3 0.9550 0.09622 0.196 0.368 0.436
#> GSM1269663 3 0.8597 0.24428 0.132 0.292 0.576
#> GSM1269671 1 0.7216 0.54441 0.712 0.112 0.176
#> GSM1269679 3 0.8966 0.09492 0.256 0.184 0.560
#> GSM1269693 2 0.9405 0.16509 0.300 0.496 0.204
#> GSM1269701 3 0.7570 0.16617 0.044 0.404 0.552
#> GSM1269709 1 0.8187 0.41181 0.628 0.244 0.128
#> GSM1269715 1 0.9633 -0.13114 0.444 0.216 0.340
#> GSM1269717 1 0.8965 0.28758 0.564 0.240 0.196
#> GSM1269721 2 0.8932 -0.00598 0.124 0.456 0.420
#> GSM1269723 3 0.4087 0.35237 0.068 0.052 0.880
#> GSM1269645 1 0.6981 0.59551 0.732 0.132 0.136
#> GSM1269653 1 0.7660 0.42839 0.612 0.324 0.064
#> GSM1269661 1 0.3583 0.68689 0.900 0.044 0.056
#> GSM1269669 1 0.6143 0.60038 0.720 0.256 0.024
#> GSM1269677 1 0.2998 0.69393 0.916 0.068 0.016
#> GSM1269685 1 0.1753 0.68999 0.952 0.048 0.000
#> GSM1269691 1 0.5493 0.60988 0.756 0.232 0.012
#> GSM1269699 1 0.1525 0.69520 0.964 0.032 0.004
#> GSM1269707 1 0.4413 0.68995 0.860 0.104 0.036
#> GSM1269651 3 0.7641 0.09697 0.044 0.436 0.520
#> GSM1269659 1 0.9284 0.16139 0.512 0.296 0.192
#> GSM1269667 3 0.9968 -0.09475 0.332 0.300 0.368
#> GSM1269675 2 0.8158 -0.00722 0.080 0.556 0.364
#> GSM1269683 3 0.8246 0.25865 0.100 0.312 0.588
#> GSM1269689 3 0.7043 0.05732 0.024 0.400 0.576
#> GSM1269697 3 0.8631 -0.06305 0.100 0.432 0.468
#> GSM1269705 3 0.9095 -0.07560 0.144 0.376 0.480
#> GSM1269713 3 0.7555 0.00147 0.040 0.440 0.520
#> GSM1269719 1 0.9385 0.03081 0.484 0.188 0.328
#> GSM1269725 3 0.8907 0.14718 0.272 0.168 0.560
#> GSM1269727 3 0.5216 0.30040 0.000 0.260 0.740
#> GSM1269649 1 0.7800 0.51387 0.668 0.204 0.128
#> GSM1269657 1 0.3181 0.69042 0.912 0.064 0.024
#> GSM1269665 1 0.5884 0.60533 0.788 0.148 0.064
#> GSM1269673 1 0.3918 0.68107 0.856 0.140 0.004
#> GSM1269681 1 0.4665 0.67412 0.852 0.100 0.048
#> GSM1269687 1 0.2173 0.69422 0.944 0.048 0.008
#> GSM1269695 1 0.2845 0.69644 0.920 0.068 0.012
#> GSM1269703 1 0.2703 0.69284 0.928 0.056 0.016
#> GSM1269711 2 0.9383 0.19684 0.364 0.460 0.176
#> GSM1269646 3 0.6465 0.24263 0.044 0.232 0.724
#> GSM1269654 1 0.7635 0.50107 0.676 0.212 0.112
#> GSM1269662 3 0.8280 -0.02966 0.404 0.080 0.516
#> GSM1269670 1 0.9025 0.18457 0.544 0.172 0.284
#> GSM1269678 3 0.8101 0.28111 0.132 0.228 0.640
#> GSM1269692 1 0.8824 0.29242 0.572 0.168 0.260
#> GSM1269700 3 0.6981 0.21367 0.228 0.068 0.704
#> GSM1269708 1 0.8423 0.38077 0.616 0.228 0.156
#> GSM1269714 1 0.8869 0.03629 0.496 0.124 0.380
#> GSM1269716 3 0.8588 -0.02211 0.344 0.112 0.544
#> GSM1269720 1 0.9502 -0.09080 0.480 0.308 0.212
#> GSM1269722 3 0.4862 0.33179 0.020 0.160 0.820
#> GSM1269644 1 0.9678 0.05384 0.444 0.328 0.228
#> GSM1269652 1 0.3459 0.69725 0.892 0.096 0.012
#> GSM1269660 1 0.5787 0.64926 0.796 0.068 0.136
#> GSM1269668 1 0.7064 0.53135 0.704 0.076 0.220
#> GSM1269676 1 0.3272 0.69841 0.904 0.080 0.016
#> GSM1269684 1 0.4128 0.69105 0.856 0.132 0.012
#> GSM1269690 1 0.5024 0.61579 0.776 0.220 0.004
#> GSM1269698 1 0.7525 0.49045 0.676 0.228 0.096
#> GSM1269706 1 0.7800 0.51486 0.668 0.128 0.204
#> GSM1269650 2 0.9825 0.15286 0.308 0.424 0.268
#> GSM1269658 3 0.7248 0.30555 0.068 0.256 0.676
#> GSM1269666 2 0.9850 0.02930 0.264 0.412 0.324
#> GSM1269674 2 0.6448 -0.00991 0.012 0.636 0.352
#> GSM1269682 1 0.9355 0.14006 0.516 0.232 0.252
#> GSM1269688 3 0.8157 -0.01517 0.072 0.412 0.516
#> GSM1269696 3 0.7259 0.21158 0.248 0.072 0.680
#> GSM1269704 2 0.9425 0.07415 0.176 0.432 0.392
#> GSM1269712 1 0.9837 -0.33126 0.392 0.360 0.248
#> GSM1269718 3 0.7525 0.28642 0.096 0.228 0.676
#> GSM1269724 3 0.9743 0.06543 0.248 0.312 0.440
#> GSM1269726 3 0.7301 0.27337 0.052 0.308 0.640
#> GSM1269648 1 0.0892 0.68609 0.980 0.020 0.000
#> GSM1269656 1 0.1964 0.68820 0.944 0.056 0.000
#> GSM1269664 1 0.6922 0.59843 0.720 0.200 0.080
#> GSM1269672 1 0.5012 0.63948 0.788 0.204 0.008
#> GSM1269680 1 0.2446 0.69056 0.936 0.052 0.012
#> GSM1269686 1 0.2031 0.69322 0.952 0.032 0.016
#> GSM1269694 1 0.3669 0.68342 0.896 0.064 0.040
#> GSM1269702 1 0.0237 0.68484 0.996 0.004 0.000
#> GSM1269710 1 0.3682 0.69045 0.876 0.116 0.008
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM1269647 4 0.5640 -0.045743 0.008 0.012 0.424 0.556
#> GSM1269655 3 0.8274 0.276381 0.116 0.204 0.564 0.116
#> GSM1269663 3 0.8470 0.296805 0.072 0.256 0.512 0.160
#> GSM1269671 1 0.5482 0.448571 0.716 0.012 0.040 0.232
#> GSM1269679 3 0.9244 0.237826 0.172 0.120 0.416 0.292
#> GSM1269693 2 0.8652 -0.119996 0.100 0.440 0.352 0.108
#> GSM1269701 3 0.8564 0.065538 0.028 0.324 0.376 0.272
#> GSM1269709 1 0.8640 0.157897 0.512 0.092 0.164 0.232
#> GSM1269715 3 0.7722 -0.056013 0.400 0.052 0.472 0.076
#> GSM1269717 1 0.7994 0.177647 0.524 0.064 0.312 0.100
#> GSM1269721 4 0.9018 0.040601 0.064 0.316 0.240 0.380
#> GSM1269723 3 0.6637 0.244242 0.040 0.032 0.584 0.344
#> GSM1269645 1 0.7097 0.433912 0.648 0.160 0.156 0.036
#> GSM1269653 1 0.7476 -0.027487 0.416 0.408 0.000 0.176
#> GSM1269661 1 0.3110 0.556177 0.892 0.056 0.048 0.004
#> GSM1269669 2 0.6473 -0.265305 0.472 0.476 0.024 0.028
#> GSM1269677 1 0.3901 0.547957 0.816 0.168 0.004 0.012
#> GSM1269685 1 0.2831 0.545805 0.876 0.120 0.004 0.000
#> GSM1269691 1 0.5404 -0.043348 0.512 0.476 0.012 0.000
#> GSM1269699 1 0.2924 0.561199 0.884 0.100 0.016 0.000
#> GSM1269707 1 0.4766 0.515820 0.800 0.140 0.020 0.040
#> GSM1269651 3 0.7986 -0.000905 0.020 0.176 0.464 0.340
#> GSM1269659 1 0.9613 -0.225729 0.352 0.240 0.276 0.132
#> GSM1269667 3 0.8043 0.132838 0.272 0.076 0.548 0.104
#> GSM1269675 4 0.7435 0.243990 0.044 0.068 0.376 0.512
#> GSM1269683 3 0.7841 0.337000 0.072 0.200 0.596 0.132
#> GSM1269689 4 0.1909 0.458818 0.004 0.008 0.048 0.940
#> GSM1269697 4 0.5650 0.423533 0.040 0.040 0.176 0.744
#> GSM1269705 4 0.6229 0.348161 0.116 0.000 0.228 0.656
#> GSM1269713 4 0.3376 0.469970 0.016 0.008 0.108 0.868
#> GSM1269719 1 0.8848 -0.003413 0.432 0.132 0.336 0.100
#> GSM1269725 3 0.8365 0.366667 0.180 0.096 0.556 0.168
#> GSM1269727 3 0.4295 0.277854 0.000 0.008 0.752 0.240
#> GSM1269649 1 0.8300 0.193369 0.492 0.288 0.044 0.176
#> GSM1269657 1 0.1909 0.550662 0.940 0.048 0.004 0.008
#> GSM1269665 1 0.5394 0.477156 0.748 0.060 0.180 0.012
#> GSM1269673 1 0.4825 0.462767 0.700 0.288 0.008 0.004
#> GSM1269681 1 0.5365 0.535077 0.780 0.092 0.028 0.100
#> GSM1269687 1 0.3679 0.548267 0.840 0.140 0.016 0.004
#> GSM1269695 1 0.4194 0.533017 0.764 0.228 0.008 0.000
#> GSM1269703 1 0.1930 0.553498 0.936 0.056 0.004 0.004
#> GSM1269711 4 0.8699 0.140592 0.188 0.248 0.076 0.488
#> GSM1269646 3 0.7322 0.007954 0.012 0.108 0.460 0.420
#> GSM1269654 1 0.7145 0.331642 0.620 0.092 0.248 0.040
#> GSM1269662 3 0.8342 0.138004 0.360 0.032 0.420 0.188
#> GSM1269670 1 0.8014 0.171846 0.520 0.028 0.244 0.208
#> GSM1269678 3 0.7903 0.351206 0.080 0.124 0.592 0.204
#> GSM1269692 1 0.9191 0.047809 0.452 0.244 0.168 0.136
#> GSM1269700 3 0.7863 0.271187 0.168 0.024 0.516 0.292
#> GSM1269708 1 0.8231 0.104532 0.496 0.084 0.328 0.092
#> GSM1269714 1 0.8668 -0.016038 0.420 0.092 0.116 0.372
#> GSM1269716 1 0.9755 -0.345305 0.292 0.144 0.276 0.288
#> GSM1269720 1 0.8441 -0.005135 0.448 0.036 0.212 0.304
#> GSM1269722 3 0.5369 0.271067 0.012 0.016 0.676 0.296
#> GSM1269644 2 0.7148 0.217477 0.220 0.560 0.220 0.000
#> GSM1269652 1 0.4604 0.544720 0.788 0.168 0.004 0.040
#> GSM1269660 1 0.6140 0.460335 0.724 0.132 0.116 0.028
#> GSM1269668 1 0.7037 0.276003 0.624 0.160 0.200 0.016
#> GSM1269676 1 0.3727 0.547662 0.832 0.152 0.008 0.008
#> GSM1269684 1 0.5010 0.491512 0.700 0.276 0.024 0.000
#> GSM1269690 1 0.4985 -0.020090 0.532 0.468 0.000 0.000
#> GSM1269698 1 0.7475 0.344789 0.608 0.128 0.044 0.220
#> GSM1269706 1 0.7880 0.240752 0.576 0.216 0.156 0.052
#> GSM1269650 2 0.9444 -0.017024 0.228 0.344 0.320 0.108
#> GSM1269658 3 0.7287 0.330100 0.040 0.224 0.620 0.116
#> GSM1269666 3 0.8335 -0.051119 0.192 0.308 0.464 0.036
#> GSM1269674 4 0.7834 0.134824 0.004 0.236 0.312 0.448
#> GSM1269682 1 0.6311 0.061682 0.492 0.048 0.456 0.004
#> GSM1269688 4 0.4033 0.449115 0.028 0.008 0.132 0.832
#> GSM1269696 3 0.8777 0.267404 0.188 0.064 0.432 0.316
#> GSM1269704 4 0.6166 0.416576 0.124 0.036 0.112 0.728
#> GSM1269712 4 0.9129 0.001925 0.312 0.096 0.184 0.408
#> GSM1269718 3 0.6917 0.366464 0.080 0.072 0.676 0.172
#> GSM1269724 3 0.6980 0.290122 0.176 0.124 0.660 0.040
#> GSM1269726 3 0.7937 0.288698 0.032 0.264 0.532 0.172
#> GSM1269648 1 0.2345 0.553140 0.900 0.100 0.000 0.000
#> GSM1269656 1 0.3306 0.549142 0.840 0.156 0.000 0.004
#> GSM1269664 1 0.6878 0.113180 0.524 0.376 0.096 0.004
#> GSM1269672 1 0.5158 0.159163 0.524 0.472 0.004 0.000
#> GSM1269680 1 0.2198 0.549952 0.920 0.072 0.008 0.000
#> GSM1269686 1 0.3046 0.543812 0.884 0.096 0.016 0.004
#> GSM1269694 1 0.3970 0.529439 0.840 0.076 0.000 0.084
#> GSM1269702 1 0.0707 0.546002 0.980 0.020 0.000 0.000
#> GSM1269710 1 0.5130 0.423639 0.644 0.344 0.008 0.004
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM1269647 3 0.574 0.0661 0.008 0.016 0.488 0.032 0.456
#> GSM1269655 3 0.849 -0.1074 0.080 0.120 0.460 0.260 0.080
#> GSM1269663 3 0.867 0.0796 0.056 0.156 0.428 0.260 0.100
#> GSM1269671 1 0.552 0.4114 0.684 0.000 0.032 0.072 0.212
#> GSM1269679 3 0.936 0.1494 0.144 0.068 0.320 0.224 0.244
#> GSM1269693 4 0.869 0.1057 0.040 0.300 0.248 0.336 0.076
#> GSM1269701 2 0.788 -0.2780 0.012 0.388 0.324 0.048 0.228
#> GSM1269709 1 0.814 -0.1050 0.396 0.020 0.080 0.340 0.164
#> GSM1269715 3 0.811 -0.1204 0.300 0.064 0.440 0.164 0.032
#> GSM1269717 1 0.751 0.2470 0.516 0.040 0.300 0.060 0.084
#> GSM1269721 5 0.928 -0.1912 0.044 0.184 0.228 0.260 0.284
#> GSM1269723 3 0.550 0.2747 0.028 0.004 0.668 0.048 0.252
#> GSM1269645 1 0.666 0.3760 0.620 0.184 0.136 0.008 0.052
#> GSM1269653 2 0.721 0.3674 0.364 0.432 0.000 0.044 0.160
#> GSM1269661 1 0.328 0.5306 0.864 0.076 0.048 0.008 0.004
#> GSM1269669 2 0.650 0.2953 0.384 0.512 0.012 0.048 0.044
#> GSM1269677 1 0.424 0.5179 0.800 0.108 0.004 0.080 0.008
#> GSM1269685 1 0.277 0.5172 0.880 0.076 0.000 0.044 0.000
#> GSM1269691 2 0.486 0.4815 0.428 0.552 0.012 0.008 0.000
#> GSM1269699 1 0.337 0.5316 0.852 0.096 0.004 0.044 0.004
#> GSM1269707 1 0.550 0.4231 0.740 0.124 0.020 0.076 0.040
#> GSM1269651 3 0.807 -0.0713 0.016 0.072 0.440 0.236 0.236
#> GSM1269659 4 0.895 0.3032 0.248 0.060 0.220 0.376 0.096
#> GSM1269667 3 0.822 0.0269 0.264 0.076 0.472 0.048 0.140
#> GSM1269675 5 0.778 0.2126 0.048 0.056 0.300 0.104 0.492
#> GSM1269683 3 0.721 0.1170 0.056 0.156 0.616 0.116 0.056
#> GSM1269689 5 0.257 0.4544 0.000 0.012 0.112 0.000 0.876
#> GSM1269697 5 0.622 0.3596 0.028 0.000 0.224 0.132 0.616
#> GSM1269705 5 0.665 0.3346 0.108 0.000 0.240 0.064 0.588
#> GSM1269713 5 0.397 0.4476 0.004 0.008 0.164 0.028 0.796
#> GSM1269719 1 0.830 -0.0567 0.400 0.040 0.204 0.304 0.052
#> GSM1269725 3 0.832 0.1747 0.164 0.044 0.488 0.204 0.100
#> GSM1269727 3 0.584 0.2721 0.000 0.012 0.624 0.112 0.252
#> GSM1269649 1 0.822 -0.1336 0.432 0.316 0.064 0.048 0.140
#> GSM1269657 1 0.157 0.5212 0.944 0.036 0.000 0.020 0.000
#> GSM1269665 1 0.530 0.4941 0.740 0.048 0.152 0.048 0.012
#> GSM1269673 1 0.506 0.3229 0.668 0.276 0.004 0.048 0.004
#> GSM1269681 1 0.483 0.5078 0.780 0.040 0.008 0.064 0.108
#> GSM1269687 1 0.408 0.5112 0.816 0.108 0.016 0.056 0.004
#> GSM1269695 1 0.555 0.3902 0.652 0.220 0.000 0.124 0.004
#> GSM1269703 1 0.234 0.5290 0.912 0.052 0.000 0.028 0.008
#> GSM1269711 5 0.845 0.1603 0.180 0.172 0.044 0.124 0.480
#> GSM1269646 3 0.776 0.1283 0.016 0.036 0.428 0.272 0.248
#> GSM1269654 1 0.689 0.3188 0.592 0.100 0.236 0.056 0.016
#> GSM1269662 3 0.811 0.1014 0.340 0.020 0.404 0.084 0.152
#> GSM1269670 1 0.787 0.2075 0.488 0.024 0.276 0.084 0.128
#> GSM1269678 3 0.707 0.2405 0.060 0.048 0.628 0.152 0.112
#> GSM1269692 1 0.902 0.0466 0.428 0.144 0.160 0.188 0.080
#> GSM1269700 3 0.764 0.2612 0.152 0.056 0.512 0.024 0.256
#> GSM1269708 1 0.803 -0.1141 0.404 0.012 0.228 0.292 0.064
#> GSM1269714 1 0.885 -0.1805 0.340 0.020 0.164 0.240 0.236
#> GSM1269716 3 0.880 0.0945 0.276 0.148 0.352 0.024 0.200
#> GSM1269720 1 0.764 -0.0247 0.428 0.012 0.180 0.044 0.336
#> GSM1269722 3 0.470 0.2991 0.004 0.008 0.736 0.048 0.204
#> GSM1269644 2 0.721 0.2668 0.188 0.560 0.172 0.076 0.004
#> GSM1269652 1 0.479 0.5048 0.772 0.132 0.004 0.056 0.036
#> GSM1269660 1 0.658 0.3497 0.652 0.168 0.100 0.056 0.024
#> GSM1269668 1 0.682 0.1468 0.568 0.208 0.192 0.016 0.016
#> GSM1269676 1 0.359 0.5145 0.828 0.120 0.004 0.048 0.000
#> GSM1269684 1 0.510 0.4158 0.696 0.240 0.016 0.044 0.004
#> GSM1269690 2 0.470 0.4740 0.432 0.552 0.000 0.016 0.000
#> GSM1269698 1 0.729 0.2908 0.588 0.096 0.032 0.080 0.204
#> GSM1269706 1 0.787 0.0867 0.516 0.236 0.144 0.064 0.040
#> GSM1269650 4 0.947 0.2981 0.168 0.168 0.268 0.312 0.084
#> GSM1269658 3 0.639 0.0852 0.016 0.084 0.608 0.264 0.028
#> GSM1269666 3 0.805 -0.1391 0.156 0.332 0.420 0.064 0.028
#> GSM1269674 3 0.829 -0.1151 0.000 0.204 0.332 0.144 0.320
#> GSM1269682 1 0.611 0.1607 0.488 0.040 0.436 0.012 0.024
#> GSM1269688 5 0.335 0.4568 0.020 0.004 0.092 0.024 0.860
#> GSM1269696 3 0.841 0.2614 0.124 0.044 0.472 0.128 0.232
#> GSM1269704 5 0.573 0.4167 0.120 0.020 0.068 0.064 0.728
#> GSM1269712 5 0.907 -0.1262 0.304 0.056 0.152 0.148 0.340
#> GSM1269718 3 0.647 0.2067 0.080 0.044 0.688 0.096 0.092
#> GSM1269724 3 0.772 0.0688 0.176 0.088 0.564 0.128 0.044
#> GSM1269726 3 0.782 0.1946 0.032 0.244 0.520 0.112 0.092
#> GSM1269648 1 0.251 0.5277 0.892 0.080 0.000 0.028 0.000
#> GSM1269656 1 0.359 0.5175 0.828 0.120 0.000 0.048 0.004
#> GSM1269664 2 0.699 0.3144 0.428 0.436 0.076 0.044 0.016
#> GSM1269672 2 0.481 0.3334 0.436 0.548 0.004 0.004 0.008
#> GSM1269680 1 0.246 0.5233 0.904 0.052 0.004 0.040 0.000
#> GSM1269686 1 0.358 0.5081 0.848 0.096 0.012 0.036 0.008
#> GSM1269694 1 0.556 0.4055 0.720 0.116 0.000 0.096 0.068
#> GSM1269702 1 0.117 0.5179 0.960 0.032 0.000 0.008 0.000
#> GSM1269710 1 0.531 0.2450 0.600 0.352 0.008 0.036 0.004
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM1269647 5 0.553 -0.055462 0.012 0.016 0.464 0.036 0.464 0.008
#> GSM1269655 3 0.836 -0.047496 0.060 0.080 0.384 0.312 0.048 0.116
#> GSM1269663 3 0.876 -0.093037 0.052 0.228 0.372 0.196 0.072 0.080
#> GSM1269671 1 0.572 0.421114 0.660 0.020 0.036 0.072 0.204 0.008
#> GSM1269679 4 0.836 0.000557 0.128 0.008 0.284 0.332 0.196 0.052
#> GSM1269693 4 0.757 -0.074752 0.032 0.048 0.208 0.444 0.016 0.252
#> GSM1269701 6 0.765 -0.202387 0.008 0.048 0.280 0.036 0.240 0.388
#> GSM1269709 4 0.661 0.386190 0.340 0.012 0.048 0.484 0.112 0.004
#> GSM1269715 3 0.894 -0.086245 0.148 0.136 0.352 0.244 0.048 0.072
#> GSM1269717 1 0.722 0.214837 0.508 0.024 0.280 0.084 0.076 0.028
#> GSM1269721 2 0.907 0.342242 0.040 0.304 0.176 0.088 0.264 0.128
#> GSM1269723 3 0.556 0.192221 0.020 0.028 0.632 0.052 0.264 0.004
#> GSM1269645 1 0.663 0.378297 0.596 0.024 0.132 0.016 0.048 0.184
#> GSM1269653 6 0.729 0.333255 0.360 0.080 0.000 0.020 0.156 0.384
#> GSM1269661 1 0.336 0.560909 0.840 0.012 0.040 0.004 0.004 0.100
#> GSM1269669 6 0.634 0.293263 0.324 0.052 0.016 0.028 0.036 0.544
#> GSM1269677 1 0.379 0.554170 0.796 0.132 0.004 0.008 0.000 0.060
#> GSM1269685 1 0.294 0.553070 0.856 0.100 0.000 0.012 0.000 0.032
#> GSM1269691 6 0.456 0.498260 0.376 0.020 0.008 0.004 0.000 0.592
#> GSM1269699 1 0.436 0.541839 0.776 0.116 0.004 0.028 0.004 0.072
#> GSM1269707 1 0.569 0.440850 0.696 0.104 0.020 0.016 0.040 0.124
#> GSM1269651 2 0.694 0.338531 0.008 0.404 0.360 0.020 0.188 0.020
#> GSM1269659 2 0.868 0.295903 0.192 0.380 0.176 0.160 0.068 0.024
#> GSM1269667 3 0.737 0.108316 0.268 0.004 0.488 0.072 0.112 0.056
#> GSM1269675 5 0.779 0.092260 0.040 0.056 0.260 0.164 0.452 0.028
#> GSM1269683 3 0.753 0.104294 0.048 0.064 0.568 0.104 0.076 0.140
#> GSM1269689 5 0.198 0.471770 0.000 0.008 0.044 0.012 0.924 0.012
#> GSM1269697 5 0.557 0.363704 0.024 0.000 0.160 0.196 0.620 0.000
#> GSM1269705 5 0.633 0.338156 0.104 0.012 0.176 0.104 0.604 0.000
#> GSM1269713 5 0.328 0.463489 0.004 0.004 0.112 0.048 0.832 0.000
#> GSM1269719 1 0.821 -0.243431 0.352 0.104 0.164 0.320 0.040 0.020
#> GSM1269725 3 0.742 0.083771 0.144 0.020 0.460 0.288 0.072 0.016
#> GSM1269727 3 0.547 0.220311 0.000 0.012 0.632 0.132 0.216 0.008
#> GSM1269649 1 0.866 -0.126198 0.372 0.120 0.048 0.056 0.132 0.272
#> GSM1269657 1 0.146 0.556315 0.944 0.036 0.000 0.004 0.000 0.016
#> GSM1269665 1 0.538 0.518324 0.724 0.040 0.132 0.056 0.012 0.036
#> GSM1269673 1 0.534 0.358055 0.656 0.080 0.008 0.020 0.004 0.232
#> GSM1269681 1 0.489 0.540762 0.764 0.060 0.012 0.024 0.100 0.040
#> GSM1269687 1 0.442 0.541718 0.792 0.096 0.020 0.036 0.008 0.048
#> GSM1269695 1 0.680 0.247847 0.504 0.240 0.004 0.084 0.000 0.168
#> GSM1269703 1 0.229 0.560924 0.900 0.008 0.000 0.016 0.004 0.072
#> GSM1269711 5 0.821 0.177144 0.172 0.124 0.028 0.120 0.476 0.080
#> GSM1269646 3 0.791 0.005585 0.008 0.168 0.408 0.164 0.232 0.020
#> GSM1269654 1 0.659 0.328042 0.576 0.088 0.236 0.016 0.012 0.072
#> GSM1269662 3 0.782 0.082276 0.324 0.040 0.380 0.100 0.152 0.004
#> GSM1269670 1 0.785 0.080532 0.472 0.032 0.212 0.128 0.136 0.020
#> GSM1269678 3 0.736 0.171515 0.056 0.040 0.532 0.220 0.124 0.028
#> GSM1269692 1 0.892 -0.030289 0.384 0.220 0.132 0.124 0.068 0.072
#> GSM1269700 3 0.734 0.211167 0.128 0.012 0.512 0.036 0.236 0.076
#> GSM1269708 4 0.655 0.373255 0.356 0.004 0.176 0.436 0.020 0.008
#> GSM1269714 4 0.760 0.334743 0.300 0.000 0.100 0.356 0.228 0.016
#> GSM1269716 3 0.841 -0.021667 0.260 0.008 0.312 0.040 0.224 0.156
#> GSM1269720 1 0.748 -0.045766 0.400 0.068 0.160 0.020 0.340 0.012
#> GSM1269722 3 0.474 0.223614 0.004 0.024 0.688 0.044 0.240 0.000
#> GSM1269644 6 0.761 0.343714 0.172 0.112 0.144 0.072 0.000 0.500
#> GSM1269652 1 0.489 0.539564 0.760 0.084 0.004 0.040 0.028 0.084
#> GSM1269660 1 0.662 0.349154 0.608 0.068 0.088 0.044 0.008 0.184
#> GSM1269668 1 0.690 0.111914 0.492 0.024 0.188 0.020 0.012 0.264
#> GSM1269676 1 0.325 0.550879 0.844 0.052 0.000 0.020 0.000 0.084
#> GSM1269684 1 0.525 0.456712 0.684 0.084 0.020 0.020 0.000 0.192
#> GSM1269690 6 0.446 0.493962 0.372 0.028 0.000 0.004 0.000 0.596
#> GSM1269698 1 0.732 0.281248 0.556 0.116 0.028 0.068 0.188 0.044
#> GSM1269706 1 0.796 0.065893 0.468 0.080 0.136 0.048 0.032 0.236
#> GSM1269650 2 0.777 0.421339 0.124 0.480 0.240 0.024 0.040 0.092
#> GSM1269658 3 0.653 -0.265746 0.012 0.364 0.496 0.052 0.044 0.032
#> GSM1269666 3 0.781 0.005918 0.136 0.064 0.408 0.060 0.012 0.320
#> GSM1269674 3 0.865 -0.150965 0.000 0.104 0.276 0.200 0.272 0.148
#> GSM1269682 3 0.580 -0.153040 0.456 0.012 0.456 0.016 0.016 0.044
#> GSM1269688 5 0.286 0.459265 0.012 0.012 0.068 0.024 0.880 0.004
#> GSM1269696 3 0.861 0.126881 0.104 0.072 0.388 0.168 0.236 0.032
#> GSM1269704 5 0.586 0.399719 0.112 0.020 0.072 0.116 0.676 0.004
#> GSM1269712 5 0.868 -0.231321 0.272 0.012 0.112 0.208 0.308 0.088
#> GSM1269718 3 0.611 0.109265 0.080 0.116 0.668 0.024 0.100 0.012
#> GSM1269724 3 0.685 0.160480 0.156 0.024 0.572 0.180 0.016 0.052
#> GSM1269726 3 0.742 0.168704 0.016 0.032 0.488 0.168 0.056 0.240
#> GSM1269648 1 0.310 0.568165 0.856 0.060 0.000 0.020 0.000 0.064
#> GSM1269656 1 0.347 0.554660 0.824 0.092 0.000 0.012 0.000 0.072
#> GSM1269664 6 0.675 0.319803 0.368 0.036 0.068 0.048 0.008 0.472
#> GSM1269672 6 0.473 0.321855 0.412 0.020 0.004 0.012 0.000 0.552
#> GSM1269680 1 0.274 0.558125 0.892 0.040 0.008 0.024 0.004 0.032
#> GSM1269686 1 0.424 0.533557 0.796 0.040 0.016 0.028 0.008 0.112
#> GSM1269694 1 0.680 0.271696 0.580 0.176 0.000 0.060 0.060 0.124
#> GSM1269702 1 0.144 0.552295 0.944 0.012 0.000 0.004 0.000 0.040
#> GSM1269710 1 0.497 0.252742 0.580 0.020 0.000 0.040 0.000 0.360
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 agent(p) disease.state(p) gender(p) individual(p) k
#> CV:pam 66 0.698 0.274 1.85e-11 0.00636 2
#> CV:pam 34 NA NA NA NA 3
#> CV:pam 18 NA NA NA NA 4
#> CV:pam 15 NA NA NA NA 5
#> CV:pam 16 NA NA NA 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", "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 51941 rows and 84 columns.
#> Top rows (1000, 2000, 3000, 4000, 5000) are extracted by 'CV' method.
#> Subgroups are detected by 'mclust' method.
#> Performed in total 1250 partitions by row resampling.
#> Best k for subgroups seems to be 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 0.985 0.982 0.968 0.4773 0.504 0.504
#> 3 3 0.612 0.627 0.825 0.2559 0.950 0.900
#> 4 4 0.592 0.710 0.754 0.1584 0.800 0.573
#> 5 5 0.612 0.685 0.781 0.0882 0.953 0.835
#> 6 6 0.640 0.683 0.761 0.0563 0.953 0.807
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
#> GSM1269647 1 0.3114 0.985 0.944 0.056
#> GSM1269655 1 0.3274 0.984 0.940 0.060
#> GSM1269663 1 0.3114 0.985 0.944 0.056
#> GSM1269671 1 0.3733 0.978 0.928 0.072
#> GSM1269679 1 0.1414 0.967 0.980 0.020
#> GSM1269693 1 0.3274 0.984 0.940 0.060
#> GSM1269701 1 0.2236 0.976 0.964 0.036
#> GSM1269709 1 0.3584 0.980 0.932 0.068
#> GSM1269715 1 0.3114 0.985 0.944 0.056
#> GSM1269717 1 0.3114 0.985 0.944 0.056
#> GSM1269721 1 0.3274 0.984 0.940 0.060
#> GSM1269723 1 0.1633 0.970 0.976 0.024
#> GSM1269645 2 0.0376 0.994 0.004 0.996
#> GSM1269653 2 0.0000 0.991 0.000 1.000
#> GSM1269661 2 0.2423 0.964 0.040 0.960
#> GSM1269669 2 0.0376 0.994 0.004 0.996
#> GSM1269677 2 0.0376 0.994 0.004 0.996
#> GSM1269685 2 0.0376 0.994 0.004 0.996
#> GSM1269691 2 0.0376 0.994 0.004 0.996
#> GSM1269699 2 0.0376 0.994 0.004 0.996
#> GSM1269707 2 0.0376 0.994 0.004 0.996
#> GSM1269651 1 0.3274 0.984 0.940 0.060
#> GSM1269659 1 0.3431 0.983 0.936 0.064
#> GSM1269667 1 0.1414 0.967 0.980 0.020
#> GSM1269675 1 0.3114 0.985 0.944 0.056
#> GSM1269683 1 0.2778 0.982 0.952 0.048
#> GSM1269689 1 0.4161 0.967 0.916 0.084
#> GSM1269697 1 0.3274 0.984 0.940 0.060
#> GSM1269705 1 0.3274 0.984 0.940 0.060
#> GSM1269713 1 0.2236 0.977 0.964 0.036
#> GSM1269719 1 0.3274 0.984 0.940 0.060
#> GSM1269725 1 0.2778 0.982 0.952 0.048
#> GSM1269727 1 0.0672 0.957 0.992 0.008
#> GSM1269649 2 0.0376 0.994 0.004 0.996
#> GSM1269657 2 0.0376 0.994 0.004 0.996
#> GSM1269665 2 0.2236 0.968 0.036 0.964
#> GSM1269673 2 0.0376 0.994 0.004 0.996
#> GSM1269681 2 0.1633 0.979 0.024 0.976
#> GSM1269687 2 0.0672 0.991 0.008 0.992
#> GSM1269695 2 0.0000 0.991 0.000 1.000
#> GSM1269703 2 0.0376 0.994 0.004 0.996
#> GSM1269711 2 0.0376 0.994 0.004 0.996
#> GSM1269646 1 0.3114 0.985 0.944 0.056
#> GSM1269654 1 0.3114 0.984 0.944 0.056
#> GSM1269662 1 0.3274 0.984 0.940 0.060
#> GSM1269670 1 0.3584 0.980 0.932 0.068
#> GSM1269678 1 0.0000 0.951 1.000 0.000
#> GSM1269692 1 0.3274 0.984 0.940 0.060
#> GSM1269700 1 0.1843 0.972 0.972 0.028
#> GSM1269708 1 0.3431 0.983 0.936 0.064
#> GSM1269714 1 0.3114 0.985 0.944 0.056
#> GSM1269716 1 0.3114 0.985 0.944 0.056
#> GSM1269720 1 0.3274 0.984 0.940 0.060
#> GSM1269722 1 0.1633 0.970 0.976 0.024
#> GSM1269644 2 0.0376 0.994 0.004 0.996
#> GSM1269652 2 0.0000 0.991 0.000 1.000
#> GSM1269660 2 0.2778 0.956 0.048 0.952
#> GSM1269668 2 0.0672 0.991 0.008 0.992
#> GSM1269676 2 0.0376 0.994 0.004 0.996
#> GSM1269684 2 0.0376 0.994 0.004 0.996
#> GSM1269690 2 0.0376 0.994 0.004 0.996
#> GSM1269698 2 0.0376 0.994 0.004 0.996
#> GSM1269706 2 0.0376 0.994 0.004 0.996
#> GSM1269650 1 0.3274 0.984 0.940 0.060
#> GSM1269658 1 0.4022 0.971 0.920 0.080
#> GSM1269666 1 0.0672 0.958 0.992 0.008
#> GSM1269674 1 0.3114 0.985 0.944 0.056
#> GSM1269682 1 0.3114 0.985 0.944 0.056
#> GSM1269688 1 0.4815 0.947 0.896 0.104
#> GSM1269696 1 0.3114 0.985 0.944 0.056
#> GSM1269704 1 0.3274 0.984 0.940 0.060
#> GSM1269712 1 0.1184 0.965 0.984 0.016
#> GSM1269718 1 0.3274 0.984 0.940 0.060
#> GSM1269724 1 0.1414 0.966 0.980 0.020
#> GSM1269726 1 0.0938 0.962 0.988 0.012
#> GSM1269648 2 0.0000 0.991 0.000 1.000
#> GSM1269656 2 0.0376 0.994 0.004 0.996
#> GSM1269664 2 0.2423 0.965 0.040 0.960
#> GSM1269672 2 0.0376 0.994 0.004 0.996
#> GSM1269680 2 0.0376 0.994 0.004 0.996
#> GSM1269686 2 0.1414 0.983 0.020 0.980
#> GSM1269694 2 0.0000 0.991 0.000 1.000
#> GSM1269702 2 0.0376 0.994 0.004 0.996
#> GSM1269710 2 0.0000 0.991 0.000 1.000
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM1269647 1 0.5881 0.33111 0.728 0.016 0.256
#> GSM1269655 1 0.3530 0.57265 0.900 0.032 0.068
#> GSM1269663 1 0.4324 0.53115 0.860 0.028 0.112
#> GSM1269671 1 0.6859 -0.00116 0.620 0.024 0.356
#> GSM1269679 1 0.3183 0.58317 0.908 0.016 0.076
#> GSM1269693 1 0.6750 0.06736 0.640 0.024 0.336
#> GSM1269701 1 0.4209 0.55717 0.860 0.020 0.120
#> GSM1269709 1 0.5241 0.51182 0.820 0.048 0.132
#> GSM1269715 1 0.6726 0.08863 0.644 0.024 0.332
#> GSM1269717 1 0.6597 0.14487 0.664 0.024 0.312
#> GSM1269721 3 0.7360 0.73562 0.440 0.032 0.528
#> GSM1269723 1 0.2651 0.58635 0.928 0.012 0.060
#> GSM1269645 2 0.1765 0.93767 0.004 0.956 0.040
#> GSM1269653 2 0.1182 0.94344 0.012 0.976 0.012
#> GSM1269661 2 0.4095 0.90005 0.056 0.880 0.064
#> GSM1269669 2 0.1919 0.94407 0.020 0.956 0.024
#> GSM1269677 2 0.4324 0.90451 0.028 0.860 0.112
#> GSM1269685 2 0.0661 0.94293 0.004 0.988 0.008
#> GSM1269691 2 0.1267 0.94051 0.004 0.972 0.024
#> GSM1269699 2 0.4324 0.90633 0.028 0.860 0.112
#> GSM1269707 2 0.3832 0.91308 0.020 0.880 0.100
#> GSM1269651 1 0.7309 -0.32499 0.552 0.032 0.416
#> GSM1269659 3 0.7263 0.79327 0.400 0.032 0.568
#> GSM1269667 1 0.3293 0.57606 0.900 0.012 0.088
#> GSM1269675 1 0.6143 0.30030 0.720 0.024 0.256
#> GSM1269683 1 0.5119 0.49927 0.812 0.028 0.160
#> GSM1269689 1 0.6703 0.25326 0.692 0.040 0.268
#> GSM1269697 1 0.5058 0.49752 0.820 0.032 0.148
#> GSM1269705 1 0.6653 0.18405 0.680 0.032 0.288
#> GSM1269713 1 0.3293 0.58203 0.900 0.012 0.088
#> GSM1269719 1 0.5874 0.43879 0.760 0.032 0.208
#> GSM1269725 1 0.3502 0.58540 0.896 0.020 0.084
#> GSM1269727 1 0.2537 0.57425 0.920 0.000 0.080
#> GSM1269649 2 0.2056 0.94206 0.024 0.952 0.024
#> GSM1269657 2 0.4324 0.90433 0.028 0.860 0.112
#> GSM1269665 2 0.3337 0.92079 0.032 0.908 0.060
#> GSM1269673 2 0.0747 0.94208 0.000 0.984 0.016
#> GSM1269681 2 0.6062 0.82978 0.064 0.776 0.160
#> GSM1269687 2 0.2527 0.93693 0.020 0.936 0.044
#> GSM1269695 2 0.1170 0.94259 0.008 0.976 0.016
#> GSM1269703 2 0.1163 0.93854 0.000 0.972 0.028
#> GSM1269711 2 0.1482 0.94389 0.012 0.968 0.020
#> GSM1269646 1 0.5639 0.37460 0.752 0.016 0.232
#> GSM1269654 1 0.4045 0.55417 0.872 0.024 0.104
#> GSM1269662 1 0.5874 0.37550 0.760 0.032 0.208
#> GSM1269670 1 0.6818 0.01755 0.628 0.024 0.348
#> GSM1269678 1 0.3272 0.56831 0.892 0.004 0.104
#> GSM1269692 1 0.6934 0.00371 0.624 0.028 0.348
#> GSM1269700 1 0.3587 0.58554 0.892 0.020 0.088
#> GSM1269708 1 0.3692 0.56334 0.896 0.048 0.056
#> GSM1269714 1 0.6301 0.28815 0.712 0.028 0.260
#> GSM1269716 1 0.5986 0.34296 0.736 0.024 0.240
#> GSM1269720 3 0.7263 0.78660 0.400 0.032 0.568
#> GSM1269722 1 0.3532 0.57845 0.884 0.008 0.108
#> GSM1269644 2 0.1182 0.94488 0.012 0.976 0.012
#> GSM1269652 2 0.1315 0.94210 0.008 0.972 0.020
#> GSM1269660 2 0.4087 0.90794 0.052 0.880 0.068
#> GSM1269668 2 0.2926 0.93044 0.036 0.924 0.040
#> GSM1269676 2 0.4324 0.90451 0.028 0.860 0.112
#> GSM1269684 2 0.1399 0.93980 0.004 0.968 0.028
#> GSM1269690 2 0.1751 0.94228 0.012 0.960 0.028
#> GSM1269698 2 0.4540 0.89811 0.028 0.848 0.124
#> GSM1269706 2 0.3769 0.91255 0.016 0.880 0.104
#> GSM1269650 1 0.7299 -0.31343 0.556 0.032 0.412
#> GSM1269658 3 0.7366 0.80137 0.400 0.036 0.564
#> GSM1269666 1 0.3038 0.56065 0.896 0.000 0.104
#> GSM1269674 1 0.6224 0.31532 0.728 0.032 0.240
#> GSM1269682 1 0.6143 0.30169 0.720 0.024 0.256
#> GSM1269688 1 0.7159 0.15895 0.660 0.052 0.288
#> GSM1269696 1 0.5269 0.43574 0.784 0.016 0.200
#> GSM1269704 1 0.6507 0.20360 0.688 0.028 0.284
#> GSM1269712 1 0.4228 0.53705 0.844 0.008 0.148
#> GSM1269718 1 0.5581 0.47160 0.788 0.036 0.176
#> GSM1269724 1 0.2682 0.58616 0.920 0.004 0.076
#> GSM1269726 1 0.2448 0.57936 0.924 0.000 0.076
#> GSM1269648 2 0.1170 0.94274 0.008 0.976 0.016
#> GSM1269656 2 0.3310 0.92646 0.028 0.908 0.064
#> GSM1269664 2 0.3896 0.90653 0.052 0.888 0.060
#> GSM1269672 2 0.1411 0.93849 0.000 0.964 0.036
#> GSM1269680 2 0.4324 0.90520 0.028 0.860 0.112
#> GSM1269686 2 0.2903 0.93066 0.028 0.924 0.048
#> GSM1269694 2 0.1315 0.94210 0.008 0.972 0.020
#> GSM1269702 2 0.0661 0.94386 0.008 0.988 0.004
#> GSM1269710 2 0.1315 0.94210 0.008 0.972 0.020
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM1269647 2 0.5239 0.701 0.004 0.676 0.300 0.020
#> GSM1269655 3 0.4397 0.722 0.008 0.120 0.820 0.052
#> GSM1269663 3 0.5390 0.723 0.016 0.124 0.768 0.092
#> GSM1269671 2 0.4747 0.731 0.008 0.764 0.204 0.024
#> GSM1269679 3 0.1721 0.742 0.008 0.028 0.952 0.012
#> GSM1269693 3 0.7362 0.569 0.016 0.144 0.568 0.272
#> GSM1269701 3 0.2927 0.729 0.008 0.068 0.900 0.024
#> GSM1269709 3 0.6141 0.441 0.036 0.252 0.676 0.036
#> GSM1269715 3 0.7451 0.540 0.016 0.136 0.540 0.308
#> GSM1269717 3 0.7474 0.540 0.016 0.140 0.540 0.304
#> GSM1269721 2 0.4665 0.673 0.004 0.804 0.104 0.088
#> GSM1269723 3 0.0992 0.751 0.004 0.012 0.976 0.008
#> GSM1269645 1 0.0804 0.854 0.980 0.012 0.000 0.008
#> GSM1269653 1 0.3498 0.732 0.832 0.008 0.000 0.160
#> GSM1269661 1 0.2499 0.811 0.924 0.032 0.012 0.032
#> GSM1269669 1 0.2174 0.854 0.928 0.020 0.000 0.052
#> GSM1269677 4 0.5716 0.899 0.420 0.028 0.000 0.552
#> GSM1269685 1 0.2611 0.827 0.896 0.008 0.000 0.096
#> GSM1269691 1 0.1191 0.855 0.968 0.004 0.004 0.024
#> GSM1269699 4 0.5917 0.880 0.444 0.036 0.000 0.520
#> GSM1269707 4 0.5780 0.850 0.476 0.028 0.000 0.496
#> GSM1269651 2 0.5169 0.698 0.004 0.760 0.164 0.072
#> GSM1269659 2 0.5537 0.591 0.012 0.752 0.096 0.140
#> GSM1269667 3 0.1674 0.736 0.004 0.032 0.952 0.012
#> GSM1269675 2 0.5237 0.668 0.000 0.628 0.356 0.016
#> GSM1269683 3 0.4931 0.717 0.016 0.056 0.792 0.136
#> GSM1269689 2 0.6090 0.427 0.012 0.512 0.452 0.024
#> GSM1269697 3 0.5421 0.242 0.008 0.328 0.648 0.016
#> GSM1269705 2 0.4719 0.740 0.008 0.752 0.224 0.016
#> GSM1269713 3 0.2923 0.716 0.008 0.080 0.896 0.016
#> GSM1269719 3 0.6521 0.371 0.008 0.328 0.592 0.072
#> GSM1269725 3 0.3564 0.694 0.012 0.112 0.860 0.016
#> GSM1269727 3 0.2131 0.747 0.000 0.036 0.932 0.032
#> GSM1269649 1 0.2409 0.847 0.924 0.032 0.004 0.040
#> GSM1269657 4 0.5673 0.897 0.448 0.024 0.000 0.528
#> GSM1269665 1 0.2107 0.829 0.940 0.020 0.016 0.024
#> GSM1269673 1 0.1854 0.855 0.940 0.012 0.000 0.048
#> GSM1269681 4 0.7087 0.794 0.364 0.080 0.020 0.536
#> GSM1269687 1 0.1139 0.855 0.972 0.008 0.008 0.012
#> GSM1269695 1 0.3196 0.783 0.856 0.008 0.000 0.136
#> GSM1269703 1 0.0779 0.854 0.980 0.004 0.000 0.016
#> GSM1269711 1 0.2987 0.808 0.880 0.016 0.000 0.104
#> GSM1269646 2 0.5851 0.454 0.004 0.516 0.456 0.024
#> GSM1269654 3 0.4736 0.730 0.012 0.104 0.808 0.076
#> GSM1269662 3 0.6774 0.480 0.016 0.296 0.604 0.084
#> GSM1269670 2 0.4652 0.735 0.004 0.756 0.220 0.020
#> GSM1269678 3 0.2245 0.752 0.008 0.020 0.932 0.040
#> GSM1269692 3 0.7439 0.574 0.016 0.188 0.576 0.220
#> GSM1269700 3 0.2197 0.739 0.012 0.028 0.936 0.024
#> GSM1269708 3 0.5016 0.643 0.024 0.152 0.784 0.040
#> GSM1269714 3 0.6742 0.636 0.016 0.116 0.644 0.224
#> GSM1269716 3 0.7060 0.599 0.016 0.120 0.600 0.264
#> GSM1269720 2 0.4417 0.659 0.004 0.820 0.084 0.092
#> GSM1269722 3 0.2521 0.741 0.004 0.060 0.916 0.020
#> GSM1269644 1 0.1863 0.853 0.944 0.012 0.004 0.040
#> GSM1269652 1 0.3498 0.746 0.832 0.008 0.000 0.160
#> GSM1269660 1 0.2807 0.799 0.912 0.040 0.016 0.032
#> GSM1269668 1 0.1958 0.850 0.944 0.020 0.008 0.028
#> GSM1269676 4 0.5731 0.903 0.428 0.028 0.000 0.544
#> GSM1269684 1 0.0967 0.853 0.976 0.004 0.004 0.016
#> GSM1269690 1 0.1191 0.855 0.968 0.004 0.004 0.024
#> GSM1269698 4 0.5996 0.900 0.448 0.040 0.000 0.512
#> GSM1269706 4 0.5771 0.867 0.460 0.028 0.000 0.512
#> GSM1269650 2 0.5276 0.696 0.004 0.756 0.156 0.084
#> GSM1269658 2 0.6745 0.541 0.024 0.668 0.160 0.148
#> GSM1269666 3 0.2207 0.747 0.004 0.024 0.932 0.040
#> GSM1269674 2 0.5426 0.709 0.004 0.656 0.316 0.024
#> GSM1269682 3 0.6498 0.658 0.020 0.128 0.684 0.168
#> GSM1269688 2 0.6429 0.655 0.024 0.600 0.336 0.040
#> GSM1269696 2 0.5607 0.388 0.000 0.496 0.484 0.020
#> GSM1269704 2 0.4747 0.739 0.004 0.736 0.244 0.016
#> GSM1269712 3 0.2164 0.748 0.004 0.004 0.924 0.068
#> GSM1269718 3 0.5340 0.599 0.008 0.220 0.728 0.044
#> GSM1269724 3 0.1796 0.742 0.004 0.032 0.948 0.016
#> GSM1269726 3 0.1943 0.750 0.008 0.032 0.944 0.016
#> GSM1269648 1 0.3196 0.785 0.856 0.008 0.000 0.136
#> GSM1269656 1 0.5548 -0.593 0.588 0.024 0.000 0.388
#> GSM1269664 1 0.2329 0.817 0.932 0.024 0.024 0.020
#> GSM1269672 1 0.1191 0.855 0.968 0.004 0.004 0.024
#> GSM1269680 4 0.6333 0.894 0.416 0.052 0.004 0.528
#> GSM1269686 1 0.1762 0.834 0.952 0.012 0.016 0.020
#> GSM1269694 1 0.3032 0.797 0.868 0.008 0.000 0.124
#> GSM1269702 1 0.2796 0.814 0.892 0.016 0.000 0.092
#> GSM1269710 1 0.3032 0.797 0.868 0.008 0.000 0.124
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM1269647 5 0.3721 0.7254 0.004 0.008 0.148 0.024 0.816
#> GSM1269655 3 0.4236 0.6704 0.004 0.020 0.808 0.108 0.060
#> GSM1269663 3 0.4927 0.5693 0.016 0.012 0.736 0.196 0.040
#> GSM1269671 5 0.2242 0.7389 0.008 0.008 0.052 0.012 0.920
#> GSM1269679 3 0.1074 0.7227 0.004 0.000 0.968 0.012 0.016
#> GSM1269693 4 0.4781 0.7403 0.008 0.012 0.388 0.592 0.000
#> GSM1269701 3 0.1788 0.7231 0.004 0.000 0.932 0.008 0.056
#> GSM1269709 3 0.6165 0.4240 0.048 0.004 0.604 0.056 0.288
#> GSM1269715 4 0.4289 0.7992 0.008 0.012 0.272 0.708 0.000
#> GSM1269717 4 0.4313 0.8019 0.008 0.012 0.276 0.704 0.000
#> GSM1269721 5 0.6124 0.6662 0.008 0.088 0.060 0.168 0.676
#> GSM1269723 3 0.1646 0.7284 0.004 0.000 0.944 0.032 0.020
#> GSM1269645 1 0.1082 0.8384 0.964 0.028 0.000 0.008 0.000
#> GSM1269653 1 0.6195 0.5159 0.564 0.308 0.000 0.112 0.016
#> GSM1269661 1 0.2417 0.8156 0.912 0.040 0.000 0.032 0.016
#> GSM1269669 1 0.2124 0.8347 0.924 0.044 0.000 0.020 0.012
#> GSM1269677 2 0.2612 0.8953 0.124 0.868 0.000 0.000 0.008
#> GSM1269685 1 0.4356 0.7697 0.784 0.140 0.000 0.060 0.016
#> GSM1269691 1 0.1393 0.8347 0.956 0.024 0.000 0.012 0.008
#> GSM1269699 2 0.2921 0.8853 0.124 0.856 0.000 0.000 0.020
#> GSM1269707 2 0.3365 0.8618 0.180 0.808 0.000 0.004 0.008
#> GSM1269651 5 0.6037 0.6864 0.008 0.084 0.088 0.124 0.696
#> GSM1269659 5 0.6793 0.5956 0.004 0.092 0.084 0.224 0.596
#> GSM1269667 3 0.0671 0.7219 0.004 0.000 0.980 0.000 0.016
#> GSM1269675 5 0.4699 0.7024 0.004 0.008 0.204 0.048 0.736
#> GSM1269683 3 0.3750 0.4554 0.012 0.000 0.756 0.232 0.000
#> GSM1269689 5 0.5690 0.5899 0.020 0.012 0.276 0.048 0.644
#> GSM1269697 3 0.5076 0.2753 0.008 0.000 0.592 0.028 0.372
#> GSM1269705 5 0.2906 0.7468 0.004 0.016 0.088 0.012 0.880
#> GSM1269713 3 0.1862 0.7204 0.004 0.000 0.932 0.016 0.048
#> GSM1269719 3 0.6940 0.3886 0.016 0.056 0.596 0.116 0.216
#> GSM1269725 3 0.2608 0.7041 0.004 0.000 0.888 0.020 0.088
#> GSM1269727 3 0.1830 0.7273 0.000 0.000 0.932 0.040 0.028
#> GSM1269649 1 0.2707 0.8284 0.896 0.068 0.004 0.016 0.016
#> GSM1269657 2 0.2989 0.8948 0.132 0.852 0.000 0.008 0.008
#> GSM1269665 1 0.1875 0.8301 0.940 0.028 0.008 0.016 0.008
#> GSM1269673 1 0.2589 0.8307 0.900 0.048 0.000 0.044 0.008
#> GSM1269681 2 0.3180 0.8578 0.068 0.856 0.000 0.000 0.076
#> GSM1269687 1 0.1854 0.8374 0.936 0.036 0.000 0.020 0.008
#> GSM1269695 1 0.5971 0.6544 0.628 0.236 0.000 0.116 0.020
#> GSM1269703 1 0.1492 0.8400 0.948 0.040 0.000 0.008 0.004
#> GSM1269711 1 0.5321 0.7239 0.704 0.184 0.000 0.092 0.020
#> GSM1269646 5 0.5580 0.4738 0.004 0.008 0.356 0.052 0.580
#> GSM1269654 3 0.3637 0.6565 0.008 0.012 0.836 0.120 0.024
#> GSM1269662 3 0.6996 0.2592 0.004 0.056 0.572 0.180 0.188
#> GSM1269670 5 0.2229 0.7377 0.004 0.012 0.052 0.012 0.920
#> GSM1269678 3 0.2304 0.6643 0.000 0.000 0.892 0.100 0.008
#> GSM1269692 4 0.5902 0.5181 0.012 0.020 0.436 0.500 0.032
#> GSM1269700 3 0.0955 0.7264 0.004 0.000 0.968 0.000 0.028
#> GSM1269708 3 0.5349 0.5900 0.048 0.004 0.728 0.060 0.160
#> GSM1269714 3 0.4653 -0.5147 0.012 0.000 0.516 0.472 0.000
#> GSM1269716 4 0.4491 0.7614 0.008 0.004 0.364 0.624 0.000
#> GSM1269720 5 0.5969 0.6725 0.008 0.092 0.056 0.152 0.692
#> GSM1269722 3 0.2522 0.7244 0.000 0.000 0.896 0.052 0.052
#> GSM1269644 1 0.3023 0.8216 0.868 0.096 0.000 0.028 0.008
#> GSM1269652 1 0.6351 0.5241 0.560 0.296 0.000 0.124 0.020
#> GSM1269660 1 0.2514 0.8114 0.912 0.032 0.004 0.032 0.020
#> GSM1269668 1 0.1235 0.8363 0.964 0.016 0.004 0.012 0.004
#> GSM1269676 2 0.2563 0.8971 0.120 0.872 0.000 0.000 0.008
#> GSM1269684 1 0.1153 0.8387 0.964 0.024 0.000 0.004 0.008
#> GSM1269690 1 0.1538 0.8337 0.948 0.036 0.000 0.008 0.008
#> GSM1269698 2 0.3035 0.8959 0.112 0.856 0.000 0.000 0.032
#> GSM1269706 2 0.3320 0.8751 0.164 0.820 0.000 0.004 0.012
#> GSM1269650 5 0.5962 0.6838 0.008 0.084 0.076 0.132 0.700
#> GSM1269658 5 0.7858 0.4827 0.016 0.088 0.156 0.248 0.492
#> GSM1269666 3 0.1205 0.7076 0.000 0.000 0.956 0.040 0.004
#> GSM1269674 5 0.4874 0.7292 0.008 0.016 0.164 0.060 0.752
#> GSM1269682 3 0.4936 -0.0586 0.012 0.004 0.616 0.356 0.012
#> GSM1269688 5 0.5754 0.6753 0.044 0.012 0.196 0.056 0.692
#> GSM1269696 5 0.5502 0.4383 0.004 0.008 0.372 0.044 0.572
#> GSM1269704 5 0.3005 0.7460 0.008 0.028 0.068 0.012 0.884
#> GSM1269712 3 0.2017 0.6902 0.000 0.000 0.912 0.080 0.008
#> GSM1269718 3 0.5158 0.5994 0.020 0.016 0.744 0.064 0.156
#> GSM1269724 3 0.1403 0.7296 0.000 0.000 0.952 0.024 0.024
#> GSM1269726 3 0.1800 0.7219 0.000 0.000 0.932 0.048 0.020
#> GSM1269648 1 0.5949 0.6464 0.624 0.240 0.000 0.120 0.016
#> GSM1269656 2 0.4777 0.5427 0.356 0.620 0.000 0.016 0.008
#> GSM1269664 1 0.1877 0.8242 0.940 0.024 0.004 0.016 0.016
#> GSM1269672 1 0.1074 0.8383 0.968 0.012 0.000 0.016 0.004
#> GSM1269680 2 0.2595 0.8882 0.080 0.888 0.000 0.000 0.032
#> GSM1269686 1 0.1243 0.8316 0.960 0.028 0.000 0.008 0.004
#> GSM1269694 1 0.5815 0.6802 0.652 0.212 0.000 0.116 0.020
#> GSM1269702 1 0.4863 0.7150 0.716 0.212 0.000 0.064 0.008
#> GSM1269710 1 0.5842 0.6758 0.652 0.204 0.000 0.124 0.020
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM1269647 2 0.265 0.7181 0.000 0.892 0.056 0.016 0.020 0.016
#> GSM1269655 3 0.392 0.7082 0.000 0.068 0.808 0.092 0.024 0.008
#> GSM1269663 3 0.430 0.6535 0.004 0.040 0.740 0.200 0.012 0.004
#> GSM1269671 2 0.152 0.7339 0.000 0.948 0.016 0.008 0.008 0.020
#> GSM1269679 3 0.172 0.7343 0.000 0.016 0.932 0.044 0.008 0.000
#> GSM1269693 4 0.399 0.7431 0.004 0.008 0.264 0.712 0.008 0.004
#> GSM1269701 3 0.207 0.7404 0.000 0.044 0.916 0.028 0.012 0.000
#> GSM1269709 3 0.542 0.4580 0.028 0.332 0.592 0.020 0.020 0.008
#> GSM1269715 4 0.284 0.7762 0.004 0.008 0.144 0.840 0.000 0.004
#> GSM1269717 4 0.299 0.7845 0.004 0.008 0.160 0.824 0.000 0.004
#> GSM1269721 2 0.649 0.6457 0.000 0.604 0.040 0.144 0.156 0.056
#> GSM1269723 3 0.193 0.7454 0.000 0.016 0.924 0.048 0.008 0.004
#> GSM1269645 1 0.171 0.8030 0.928 0.000 0.004 0.000 0.056 0.012
#> GSM1269653 5 0.501 0.8534 0.188 0.000 0.000 0.000 0.644 0.168
#> GSM1269661 1 0.168 0.7953 0.944 0.012 0.004 0.008 0.016 0.016
#> GSM1269669 1 0.331 0.7031 0.788 0.004 0.000 0.000 0.192 0.016
#> GSM1269677 6 0.302 0.8451 0.092 0.008 0.000 0.000 0.048 0.852
#> GSM1269685 1 0.508 -0.1936 0.512 0.000 0.000 0.000 0.408 0.080
#> GSM1269691 1 0.215 0.8007 0.896 0.000 0.000 0.000 0.084 0.020
#> GSM1269699 6 0.320 0.7918 0.016 0.012 0.000 0.004 0.140 0.828
#> GSM1269707 6 0.376 0.8279 0.084 0.008 0.000 0.000 0.112 0.796
#> GSM1269651 2 0.642 0.6464 0.000 0.608 0.032 0.144 0.156 0.060
#> GSM1269659 2 0.753 0.5404 0.004 0.488 0.060 0.212 0.172 0.064
#> GSM1269667 3 0.139 0.7348 0.000 0.016 0.948 0.032 0.004 0.000
#> GSM1269675 2 0.364 0.7079 0.000 0.812 0.128 0.020 0.036 0.004
#> GSM1269683 3 0.401 0.4857 0.004 0.008 0.700 0.276 0.012 0.000
#> GSM1269689 2 0.455 0.6467 0.004 0.748 0.168 0.020 0.048 0.012
#> GSM1269697 3 0.472 0.2962 0.004 0.408 0.556 0.016 0.016 0.000
#> GSM1269705 2 0.272 0.7419 0.004 0.888 0.040 0.008 0.052 0.008
#> GSM1269713 3 0.184 0.7396 0.000 0.048 0.924 0.024 0.004 0.000
#> GSM1269719 3 0.682 0.3944 0.000 0.208 0.552 0.136 0.076 0.028
#> GSM1269725 3 0.281 0.7079 0.000 0.104 0.864 0.012 0.016 0.004
#> GSM1269727 3 0.212 0.7382 0.000 0.020 0.908 0.064 0.008 0.000
#> GSM1269649 1 0.439 0.5890 0.736 0.008 0.004 0.004 0.192 0.056
#> GSM1269657 6 0.303 0.8477 0.088 0.008 0.000 0.000 0.052 0.852
#> GSM1269665 1 0.133 0.8007 0.956 0.004 0.004 0.004 0.024 0.008
#> GSM1269673 1 0.340 0.6623 0.768 0.000 0.000 0.000 0.212 0.020
#> GSM1269681 6 0.268 0.8133 0.020 0.080 0.000 0.004 0.016 0.880
#> GSM1269687 1 0.265 0.7831 0.868 0.000 0.004 0.000 0.100 0.028
#> GSM1269695 5 0.477 0.8904 0.256 0.000 0.000 0.000 0.648 0.096
#> GSM1269703 1 0.175 0.8057 0.912 0.000 0.000 0.000 0.084 0.004
#> GSM1269711 5 0.548 0.8060 0.304 0.008 0.000 0.000 0.564 0.124
#> GSM1269646 2 0.439 0.6054 0.000 0.716 0.228 0.020 0.032 0.004
#> GSM1269654 3 0.360 0.7034 0.000 0.036 0.816 0.124 0.020 0.004
#> GSM1269662 3 0.693 0.3464 0.000 0.160 0.556 0.172 0.068 0.044
#> GSM1269670 2 0.132 0.7353 0.000 0.956 0.016 0.004 0.008 0.016
#> GSM1269678 3 0.274 0.7000 0.000 0.008 0.852 0.128 0.012 0.000
#> GSM1269692 4 0.518 0.4879 0.008 0.024 0.360 0.580 0.020 0.008
#> GSM1269700 3 0.160 0.7394 0.000 0.024 0.940 0.028 0.008 0.000
#> GSM1269708 3 0.529 0.5676 0.036 0.240 0.668 0.028 0.020 0.008
#> GSM1269714 4 0.412 0.5731 0.004 0.004 0.384 0.604 0.004 0.000
#> GSM1269716 4 0.338 0.7821 0.004 0.004 0.204 0.780 0.004 0.004
#> GSM1269720 2 0.650 0.6436 0.000 0.600 0.032 0.144 0.160 0.064
#> GSM1269722 3 0.252 0.7430 0.000 0.056 0.888 0.048 0.008 0.000
#> GSM1269644 1 0.422 0.6129 0.724 0.000 0.000 0.000 0.196 0.080
#> GSM1269652 5 0.480 0.8629 0.176 0.000 0.000 0.000 0.672 0.152
#> GSM1269660 1 0.225 0.7794 0.920 0.024 0.016 0.008 0.012 0.020
#> GSM1269668 1 0.190 0.8000 0.924 0.008 0.004 0.000 0.052 0.012
#> GSM1269676 6 0.293 0.8487 0.080 0.008 0.000 0.000 0.052 0.860
#> GSM1269684 1 0.143 0.8076 0.940 0.000 0.000 0.000 0.048 0.012
#> GSM1269690 1 0.207 0.8048 0.904 0.000 0.000 0.000 0.072 0.024
#> GSM1269698 6 0.279 0.8407 0.028 0.016 0.000 0.004 0.076 0.876
#> GSM1269706 6 0.373 0.8270 0.068 0.008 0.000 0.004 0.116 0.804
#> GSM1269650 2 0.650 0.6462 0.000 0.604 0.044 0.144 0.156 0.052
#> GSM1269658 2 0.805 0.4451 0.004 0.420 0.124 0.232 0.164 0.056
#> GSM1269666 3 0.161 0.7335 0.000 0.008 0.932 0.056 0.004 0.000
#> GSM1269674 2 0.408 0.7204 0.000 0.792 0.128 0.032 0.036 0.012
#> GSM1269682 3 0.459 -0.0408 0.004 0.012 0.548 0.424 0.012 0.000
#> GSM1269688 2 0.380 0.7103 0.012 0.828 0.092 0.016 0.036 0.016
#> GSM1269696 2 0.488 0.5537 0.000 0.668 0.264 0.024 0.032 0.012
#> GSM1269704 2 0.241 0.7423 0.004 0.904 0.044 0.004 0.036 0.008
#> GSM1269712 3 0.257 0.7002 0.000 0.008 0.856 0.132 0.004 0.000
#> GSM1269718 3 0.502 0.6229 0.004 0.168 0.716 0.076 0.024 0.012
#> GSM1269724 3 0.157 0.7474 0.000 0.028 0.940 0.028 0.004 0.000
#> GSM1269726 3 0.236 0.7386 0.000 0.016 0.892 0.080 0.012 0.000
#> GSM1269648 5 0.472 0.8976 0.232 0.000 0.000 0.000 0.664 0.104
#> GSM1269656 6 0.510 0.5572 0.228 0.008 0.000 0.000 0.120 0.644
#> GSM1269664 1 0.128 0.7955 0.960 0.008 0.004 0.008 0.012 0.008
#> GSM1269672 1 0.240 0.7866 0.872 0.000 0.000 0.000 0.112 0.016
#> GSM1269680 6 0.217 0.8390 0.016 0.032 0.000 0.004 0.032 0.916
#> GSM1269686 1 0.148 0.8064 0.948 0.004 0.004 0.004 0.032 0.008
#> GSM1269694 5 0.476 0.8824 0.272 0.000 0.000 0.000 0.640 0.088
#> GSM1269702 1 0.559 -0.2526 0.468 0.000 0.000 0.000 0.388 0.144
#> GSM1269710 5 0.462 0.8922 0.244 0.000 0.000 0.000 0.668 0.088
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 agent(p) disease.state(p) gender(p) individual(p) k
#> CV:mclust 84 1.000 1.000 3.81e-19 8.65e-05 2
#> CV:mclust 59 0.982 0.851 1.54e-13 5.55e-06 3
#> CV:mclust 76 0.985 0.366 2.21e-16 1.03e-08 4
#> CV:mclust 74 0.975 0.153 3.24e-15 9.67e-11 5
#> CV:mclust 74 0.994 0.171 1.50e-14 1.65e-11 6
If matrix rows can be associated to genes, consider to use functional_enrichment(res,
...) to perform function enrichment for the signature genes. See this vignette for more detailed explanations.
The object with results only for a single top-value method and a single partition method can be extracted as:
res = res_list["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 51941 rows and 84 columns.
#> Top rows (1000, 2000, 3000, 4000, 5000) are extracted by 'CV' method.
#> Subgroups are detected by 'NMF' method.
#> Performed in total 1250 partitions by row resampling.
#> Best k for subgroups seems to be 2.
#>
#> Following methods can be applied to this 'ConsensusPartition' object:
#> [1] "cola_report" "collect_classes" "collect_plots"
#> [4] "collect_stats" "colnames" "compare_signatures"
#> [7] "consensus_heatmap" "dimension_reduction" "functional_enrichment"
#> [10] "get_anno_col" "get_anno" "get_classes"
#> [13] "get_consensus" "get_matrix" "get_membership"
#> [16] "get_param" "get_signatures" "get_stats"
#> [19] "is_best_k" "is_stable_k" "membership_heatmap"
#> [22] "ncol" "nrow" "plot_ecdf"
#> [25] "rownames" "select_partition_number" "show"
#> [28] "suggest_best_k" "test_to_known_factors"
collect_plots() function collects all the plots made from res for all k (number of partitions)
into one single page to provide an easy and fast comparison between different k.
collect_plots(res)
The plots are:
k and the heatmap of
predicted classes for each k.k.k.k.All the plots in panels can be made by individual functions and they are plotted later in this section.
select_partition_number() produces several plots showing different
statistics for choosing “optimized” k. There are following statistics:
k;k, the area increased is defined as \(A_k - A_{k-1}\).The detailed explanations of these statistics can be found in the cola vignette.
Generally speaking, lower PAC score, higher mean silhouette score or higher
concordance corresponds to better partition. Rand index and Jaccard index
measure how similar the current partition is compared to partition with k-1.
If they are too similar, we won't accept k is better than k-1.
select_partition_number(res)
The numeric values for all these statistics can be obtained by get_stats().
get_stats(res)
#> k 1-PAC mean_silhouette concordance area_increased Rand Jaccard
#> 2 2 0.998 0.974 0.983 0.4976 0.504 0.504
#> 3 3 0.531 0.665 0.844 0.2167 0.987 0.973
#> 4 4 0.446 0.508 0.676 0.1635 0.883 0.763
#> 5 5 0.462 0.359 0.566 0.0898 0.913 0.781
#> 6 6 0.469 0.176 0.476 0.0631 0.846 0.567
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
#> GSM1269647 1 0.1184 0.974 0.984 0.016
#> GSM1269655 1 0.0000 0.977 1.000 0.000
#> GSM1269663 1 0.0376 0.977 0.996 0.004
#> GSM1269671 1 0.4431 0.920 0.908 0.092
#> GSM1269679 1 0.0000 0.977 1.000 0.000
#> GSM1269693 1 0.0672 0.976 0.992 0.008
#> GSM1269701 1 0.0000 0.977 1.000 0.000
#> GSM1269709 1 0.3274 0.948 0.940 0.060
#> GSM1269715 1 0.0938 0.975 0.988 0.012
#> GSM1269717 1 0.0672 0.976 0.992 0.008
#> GSM1269721 1 0.0938 0.975 0.988 0.012
#> GSM1269723 1 0.0000 0.977 1.000 0.000
#> GSM1269645 2 0.0672 0.990 0.008 0.992
#> GSM1269653 2 0.0000 0.993 0.000 1.000
#> GSM1269661 2 0.2948 0.957 0.052 0.948
#> GSM1269669 2 0.0376 0.991 0.004 0.996
#> GSM1269677 2 0.0000 0.993 0.000 1.000
#> GSM1269685 2 0.0000 0.993 0.000 1.000
#> GSM1269691 2 0.0672 0.990 0.008 0.992
#> GSM1269699 2 0.0376 0.991 0.004 0.996
#> GSM1269707 2 0.0000 0.993 0.000 1.000
#> GSM1269651 1 0.0376 0.977 0.996 0.004
#> GSM1269659 1 0.4815 0.905 0.896 0.104
#> GSM1269667 1 0.0000 0.977 1.000 0.000
#> GSM1269675 1 0.0672 0.977 0.992 0.008
#> GSM1269683 1 0.0000 0.977 1.000 0.000
#> GSM1269689 1 0.5629 0.880 0.868 0.132
#> GSM1269697 1 0.0672 0.977 0.992 0.008
#> GSM1269705 1 0.0672 0.977 0.992 0.008
#> GSM1269713 1 0.0000 0.977 1.000 0.000
#> GSM1269719 1 0.0938 0.976 0.988 0.012
#> GSM1269725 1 0.0376 0.977 0.996 0.004
#> GSM1269727 1 0.0000 0.977 1.000 0.000
#> GSM1269649 2 0.0672 0.989 0.008 0.992
#> GSM1269657 2 0.0000 0.993 0.000 1.000
#> GSM1269665 2 0.1414 0.983 0.020 0.980
#> GSM1269673 2 0.0000 0.993 0.000 1.000
#> GSM1269681 2 0.0938 0.987 0.012 0.988
#> GSM1269687 2 0.0376 0.991 0.004 0.996
#> GSM1269695 2 0.0000 0.993 0.000 1.000
#> GSM1269703 2 0.0000 0.993 0.000 1.000
#> GSM1269711 2 0.0000 0.993 0.000 1.000
#> GSM1269646 1 0.0672 0.977 0.992 0.008
#> GSM1269654 1 0.0000 0.977 1.000 0.000
#> GSM1269662 1 0.0000 0.977 1.000 0.000
#> GSM1269670 1 0.2603 0.958 0.956 0.044
#> GSM1269678 1 0.0000 0.977 1.000 0.000
#> GSM1269692 1 0.2778 0.952 0.952 0.048
#> GSM1269700 1 0.0000 0.977 1.000 0.000
#> GSM1269708 1 0.2423 0.960 0.960 0.040
#> GSM1269714 1 0.0376 0.977 0.996 0.004
#> GSM1269716 1 0.0376 0.977 0.996 0.004
#> GSM1269720 1 0.5408 0.886 0.876 0.124
#> GSM1269722 1 0.0000 0.977 1.000 0.000
#> GSM1269644 2 0.0000 0.993 0.000 1.000
#> GSM1269652 2 0.0000 0.993 0.000 1.000
#> GSM1269660 2 0.2043 0.975 0.032 0.968
#> GSM1269668 2 0.2236 0.970 0.036 0.964
#> GSM1269676 2 0.0000 0.993 0.000 1.000
#> GSM1269684 2 0.0672 0.990 0.008 0.992
#> GSM1269690 2 0.0672 0.990 0.008 0.992
#> GSM1269698 2 0.0000 0.993 0.000 1.000
#> GSM1269706 2 0.0000 0.993 0.000 1.000
#> GSM1269650 1 0.0938 0.976 0.988 0.012
#> GSM1269658 1 0.6531 0.832 0.832 0.168
#> GSM1269666 1 0.0000 0.977 1.000 0.000
#> GSM1269674 1 0.0376 0.977 0.996 0.004
#> GSM1269682 1 0.0672 0.976 0.992 0.008
#> GSM1269688 1 0.5629 0.880 0.868 0.132
#> GSM1269696 1 0.0672 0.977 0.992 0.008
#> GSM1269704 1 0.2043 0.965 0.968 0.032
#> GSM1269712 1 0.0376 0.977 0.996 0.004
#> GSM1269718 1 0.0672 0.977 0.992 0.008
#> GSM1269724 1 0.0000 0.977 1.000 0.000
#> GSM1269726 1 0.0000 0.977 1.000 0.000
#> GSM1269648 2 0.0000 0.993 0.000 1.000
#> GSM1269656 2 0.0000 0.993 0.000 1.000
#> GSM1269664 2 0.2778 0.959 0.048 0.952
#> GSM1269672 2 0.0376 0.991 0.004 0.996
#> GSM1269680 2 0.0376 0.991 0.004 0.996
#> GSM1269686 2 0.1414 0.983 0.020 0.980
#> GSM1269694 2 0.0000 0.993 0.000 1.000
#> GSM1269702 2 0.0000 0.993 0.000 1.000
#> GSM1269710 2 0.0000 0.993 0.000 1.000
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM1269647 1 0.4931 0.618 0.768 0.000 0.232
#> GSM1269655 1 0.1163 0.713 0.972 0.000 0.028
#> GSM1269663 1 0.2200 0.698 0.940 0.004 0.056
#> GSM1269671 1 0.7190 0.384 0.608 0.036 0.356
#> GSM1269679 1 0.2261 0.694 0.932 0.000 0.068
#> GSM1269693 1 0.7114 -0.606 0.584 0.028 0.388
#> GSM1269701 1 0.3715 0.658 0.868 0.004 0.128
#> GSM1269709 1 0.3765 0.706 0.888 0.028 0.084
#> GSM1269715 1 0.7286 -0.869 0.508 0.028 0.464
#> GSM1269717 3 0.7283 0.000 0.460 0.028 0.512
#> GSM1269721 1 0.5109 0.618 0.780 0.008 0.212
#> GSM1269723 1 0.0747 0.712 0.984 0.000 0.016
#> GSM1269645 2 0.3295 0.886 0.008 0.896 0.096
#> GSM1269653 2 0.2796 0.889 0.000 0.908 0.092
#> GSM1269661 2 0.4840 0.834 0.016 0.816 0.168
#> GSM1269669 2 0.4121 0.838 0.000 0.832 0.168
#> GSM1269677 2 0.4291 0.853 0.000 0.820 0.180
#> GSM1269685 2 0.1753 0.895 0.000 0.952 0.048
#> GSM1269691 2 0.2356 0.887 0.000 0.928 0.072
#> GSM1269699 2 0.5098 0.804 0.000 0.752 0.248
#> GSM1269707 2 0.4178 0.856 0.000 0.828 0.172
#> GSM1269651 1 0.4750 0.625 0.784 0.000 0.216
#> GSM1269659 1 0.7712 0.335 0.652 0.092 0.256
#> GSM1269667 1 0.2165 0.709 0.936 0.000 0.064
#> GSM1269675 1 0.4346 0.662 0.816 0.000 0.184
#> GSM1269683 1 0.4475 0.544 0.840 0.016 0.144
#> GSM1269689 1 0.7606 0.440 0.664 0.092 0.244
#> GSM1269697 1 0.1860 0.718 0.948 0.000 0.052
#> GSM1269705 1 0.4121 0.678 0.832 0.000 0.168
#> GSM1269713 1 0.2261 0.719 0.932 0.000 0.068
#> GSM1269719 1 0.2301 0.720 0.936 0.004 0.060
#> GSM1269725 1 0.2261 0.718 0.932 0.000 0.068
#> GSM1269727 1 0.1411 0.706 0.964 0.000 0.036
#> GSM1269649 2 0.3755 0.883 0.008 0.872 0.120
#> GSM1269657 2 0.4399 0.847 0.000 0.812 0.188
#> GSM1269665 2 0.5461 0.760 0.008 0.748 0.244
#> GSM1269673 2 0.2165 0.889 0.000 0.936 0.064
#> GSM1269681 2 0.6651 0.687 0.024 0.656 0.320
#> GSM1269687 2 0.1529 0.891 0.000 0.960 0.040
#> GSM1269695 2 0.2066 0.893 0.000 0.940 0.060
#> GSM1269703 2 0.1643 0.892 0.000 0.956 0.044
#> GSM1269711 2 0.2796 0.894 0.000 0.908 0.092
#> GSM1269646 1 0.3619 0.695 0.864 0.000 0.136
#> GSM1269654 1 0.1411 0.710 0.964 0.000 0.036
#> GSM1269662 1 0.2682 0.710 0.920 0.004 0.076
#> GSM1269670 1 0.6773 0.433 0.636 0.024 0.340
#> GSM1269678 1 0.2165 0.691 0.936 0.000 0.064
#> GSM1269692 1 0.7308 -0.121 0.648 0.056 0.296
#> GSM1269700 1 0.1964 0.704 0.944 0.000 0.056
#> GSM1269708 1 0.3042 0.711 0.920 0.040 0.040
#> GSM1269714 1 0.4047 0.554 0.848 0.004 0.148
#> GSM1269716 1 0.5902 -0.252 0.680 0.004 0.316
#> GSM1269720 1 0.8394 0.252 0.576 0.108 0.316
#> GSM1269722 1 0.0747 0.704 0.984 0.000 0.016
#> GSM1269644 2 0.1964 0.893 0.000 0.944 0.056
#> GSM1269652 2 0.2537 0.891 0.000 0.920 0.080
#> GSM1269660 2 0.3539 0.888 0.012 0.888 0.100
#> GSM1269668 2 0.4047 0.856 0.004 0.848 0.148
#> GSM1269676 2 0.3816 0.868 0.000 0.852 0.148
#> GSM1269684 2 0.1860 0.891 0.000 0.948 0.052
#> GSM1269690 2 0.3752 0.863 0.000 0.856 0.144
#> GSM1269698 2 0.5397 0.774 0.000 0.720 0.280
#> GSM1269706 2 0.4178 0.859 0.000 0.828 0.172
#> GSM1269650 1 0.5158 0.602 0.764 0.004 0.232
#> GSM1269658 1 0.8483 0.115 0.600 0.140 0.260
#> GSM1269666 1 0.1529 0.710 0.960 0.000 0.040
#> GSM1269674 1 0.4351 0.670 0.828 0.004 0.168
#> GSM1269682 1 0.3644 0.612 0.872 0.004 0.124
#> GSM1269688 1 0.7739 0.419 0.644 0.088 0.268
#> GSM1269696 1 0.3941 0.681 0.844 0.000 0.156
#> GSM1269704 1 0.4235 0.673 0.824 0.000 0.176
#> GSM1269712 1 0.2537 0.683 0.920 0.000 0.080
#> GSM1269718 1 0.2878 0.711 0.904 0.000 0.096
#> GSM1269724 1 0.1643 0.708 0.956 0.000 0.044
#> GSM1269726 1 0.1964 0.695 0.944 0.000 0.056
#> GSM1269648 2 0.2066 0.893 0.000 0.940 0.060
#> GSM1269656 2 0.2711 0.891 0.000 0.912 0.088
#> GSM1269664 2 0.5848 0.744 0.012 0.720 0.268
#> GSM1269672 2 0.1643 0.890 0.000 0.956 0.044
#> GSM1269680 2 0.5397 0.767 0.000 0.720 0.280
#> GSM1269686 2 0.3038 0.877 0.000 0.896 0.104
#> GSM1269694 2 0.1964 0.894 0.000 0.944 0.056
#> GSM1269702 2 0.1860 0.895 0.000 0.948 0.052
#> GSM1269710 2 0.1860 0.893 0.000 0.948 0.052
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM1269647 3 0.572 0.39747 0.008 0.312 0.648 0.032
#> GSM1269655 3 0.525 0.59683 0.004 0.116 0.764 0.116
#> GSM1269663 3 0.451 0.59833 0.000 0.076 0.804 0.120
#> GSM1269671 2 0.630 -0.00395 0.020 0.484 0.472 0.024
#> GSM1269679 3 0.395 0.62976 0.000 0.064 0.840 0.096
#> GSM1269693 4 0.699 0.41219 0.016 0.076 0.388 0.520
#> GSM1269701 3 0.639 0.52622 0.016 0.128 0.688 0.168
#> GSM1269709 3 0.573 0.53648 0.040 0.176 0.740 0.044
#> GSM1269715 4 0.661 0.48530 0.020 0.056 0.332 0.592
#> GSM1269717 4 0.566 0.48607 0.016 0.024 0.292 0.668
#> GSM1269721 3 0.681 0.02581 0.004 0.352 0.548 0.096
#> GSM1269723 3 0.247 0.62861 0.000 0.028 0.916 0.056
#> GSM1269645 1 0.617 0.69467 0.672 0.136 0.000 0.192
#> GSM1269653 1 0.421 0.76372 0.820 0.124 0.000 0.056
#> GSM1269661 1 0.634 0.64620 0.636 0.076 0.008 0.280
#> GSM1269669 1 0.605 0.61217 0.616 0.064 0.000 0.320
#> GSM1269677 1 0.680 0.54315 0.568 0.308 0.000 0.124
#> GSM1269685 1 0.395 0.77101 0.840 0.064 0.000 0.096
#> GSM1269691 1 0.494 0.74715 0.768 0.072 0.000 0.160
#> GSM1269699 1 0.582 0.64428 0.652 0.288 0.000 0.060
#> GSM1269707 1 0.553 0.70573 0.720 0.192 0.000 0.088
#> GSM1269651 3 0.663 0.06806 0.008 0.376 0.548 0.068
#> GSM1269659 2 0.886 0.26614 0.088 0.396 0.372 0.144
#> GSM1269667 3 0.477 0.61917 0.000 0.076 0.784 0.140
#> GSM1269675 3 0.563 0.37891 0.004 0.304 0.656 0.036
#> GSM1269683 3 0.564 0.30102 0.000 0.040 0.636 0.324
#> GSM1269689 3 0.794 0.03547 0.104 0.364 0.484 0.048
#> GSM1269697 3 0.317 0.62062 0.000 0.116 0.868 0.016
#> GSM1269705 3 0.544 0.39577 0.000 0.288 0.672 0.040
#> GSM1269713 3 0.295 0.62454 0.000 0.088 0.888 0.024
#> GSM1269719 3 0.501 0.56725 0.000 0.172 0.760 0.068
#> GSM1269725 3 0.322 0.61976 0.000 0.120 0.864 0.016
#> GSM1269727 3 0.284 0.63553 0.000 0.028 0.896 0.076
#> GSM1269649 1 0.531 0.72815 0.744 0.164 0.000 0.092
#> GSM1269657 1 0.630 0.61612 0.632 0.268 0.000 0.100
#> GSM1269665 1 0.648 0.43654 0.508 0.060 0.004 0.428
#> GSM1269673 1 0.451 0.75259 0.788 0.044 0.000 0.168
#> GSM1269681 2 0.644 -0.39040 0.456 0.492 0.016 0.036
#> GSM1269687 1 0.431 0.76707 0.812 0.056 0.000 0.132
#> GSM1269695 1 0.236 0.77096 0.920 0.056 0.000 0.024
#> GSM1269703 1 0.413 0.76627 0.824 0.052 0.000 0.124
#> GSM1269711 1 0.538 0.71464 0.728 0.196 0.000 0.076
#> GSM1269646 3 0.499 0.50977 0.000 0.236 0.728 0.036
#> GSM1269654 3 0.407 0.60793 0.000 0.064 0.832 0.104
#> GSM1269662 3 0.614 0.35207 0.000 0.252 0.652 0.096
#> GSM1269670 2 0.671 0.02827 0.028 0.480 0.456 0.036
#> GSM1269678 3 0.396 0.61948 0.000 0.024 0.816 0.160
#> GSM1269692 4 0.791 0.17997 0.012 0.184 0.400 0.404
#> GSM1269700 3 0.327 0.63350 0.000 0.024 0.868 0.108
#> GSM1269708 3 0.517 0.57557 0.040 0.128 0.788 0.044
#> GSM1269714 3 0.561 0.41227 0.000 0.060 0.684 0.256
#> GSM1269716 3 0.600 -0.22854 0.000 0.040 0.504 0.456
#> GSM1269720 2 0.847 0.30684 0.080 0.420 0.392 0.108
#> GSM1269722 3 0.223 0.63294 0.000 0.036 0.928 0.036
#> GSM1269644 1 0.505 0.75722 0.764 0.084 0.000 0.152
#> GSM1269652 1 0.409 0.75423 0.820 0.140 0.000 0.040
#> GSM1269660 1 0.627 0.70548 0.664 0.152 0.000 0.184
#> GSM1269668 1 0.564 0.64025 0.648 0.044 0.000 0.308
#> GSM1269676 1 0.643 0.60370 0.620 0.272 0.000 0.108
#> GSM1269684 1 0.467 0.75941 0.764 0.036 0.000 0.200
#> GSM1269690 1 0.600 0.67511 0.660 0.084 0.000 0.256
#> GSM1269698 1 0.565 0.66114 0.672 0.272 0.000 0.056
#> GSM1269706 1 0.551 0.70056 0.720 0.196 0.000 0.084
#> GSM1269650 3 0.695 -0.12602 0.012 0.428 0.484 0.076
#> GSM1269658 2 0.885 0.23014 0.072 0.412 0.340 0.176
#> GSM1269666 3 0.434 0.62382 0.004 0.048 0.816 0.132
#> GSM1269674 3 0.583 0.26491 0.000 0.332 0.620 0.048
#> GSM1269682 3 0.604 0.33739 0.004 0.060 0.640 0.296
#> GSM1269688 3 0.784 0.06422 0.088 0.348 0.508 0.056
#> GSM1269696 3 0.540 0.39331 0.000 0.328 0.644 0.028
#> GSM1269704 3 0.525 0.46179 0.008 0.252 0.712 0.028
#> GSM1269712 3 0.442 0.59723 0.000 0.040 0.792 0.168
#> GSM1269718 3 0.671 0.49569 0.012 0.184 0.652 0.152
#> GSM1269724 3 0.307 0.63647 0.000 0.044 0.888 0.068
#> GSM1269726 3 0.447 0.59149 0.000 0.036 0.784 0.180
#> GSM1269648 1 0.274 0.77175 0.904 0.060 0.000 0.036
#> GSM1269656 1 0.451 0.75482 0.800 0.136 0.000 0.064
#> GSM1269664 4 0.696 -0.45307 0.448 0.076 0.012 0.464
#> GSM1269672 1 0.405 0.76224 0.820 0.036 0.000 0.144
#> GSM1269680 1 0.588 0.58687 0.620 0.328 0.000 0.052
#> GSM1269686 1 0.497 0.72919 0.752 0.040 0.004 0.204
#> GSM1269694 1 0.277 0.76872 0.900 0.072 0.000 0.028
#> GSM1269702 1 0.389 0.77009 0.844 0.064 0.000 0.092
#> GSM1269710 1 0.340 0.77381 0.868 0.092 0.000 0.040
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM1269647 3 0.554 0.22707 0.004 0.364 0.580 0.036 0.016
#> GSM1269655 3 0.650 0.35635 0.004 0.216 0.612 0.128 0.040
#> GSM1269663 3 0.639 0.23373 0.004 0.104 0.644 0.184 0.064
#> GSM1269671 2 0.654 0.28821 0.044 0.524 0.372 0.024 0.036
#> GSM1269679 3 0.518 0.48887 0.016 0.132 0.732 0.116 0.004
#> GSM1269693 4 0.780 0.65107 0.024 0.068 0.328 0.456 0.124
#> GSM1269701 3 0.726 0.35539 0.036 0.208 0.556 0.176 0.024
#> GSM1269709 3 0.600 0.42004 0.020 0.208 0.672 0.032 0.068
#> GSM1269715 4 0.694 0.70153 0.032 0.044 0.264 0.580 0.080
#> GSM1269717 4 0.583 0.65109 0.036 0.012 0.268 0.644 0.040
#> GSM1269721 3 0.827 -0.18313 0.000 0.288 0.324 0.120 0.268
#> GSM1269723 3 0.354 0.52463 0.000 0.068 0.852 0.056 0.024
#> GSM1269645 1 0.672 0.53898 0.616 0.116 0.000 0.160 0.108
#> GSM1269653 1 0.671 0.47676 0.580 0.136 0.000 0.052 0.232
#> GSM1269661 1 0.668 0.50073 0.572 0.060 0.000 0.264 0.104
#> GSM1269669 1 0.576 0.53495 0.652 0.060 0.004 0.252 0.032
#> GSM1269677 5 0.449 0.25334 0.256 0.016 0.000 0.016 0.712
#> GSM1269685 1 0.566 0.51392 0.652 0.052 0.000 0.040 0.256
#> GSM1269691 1 0.589 0.46912 0.592 0.020 0.000 0.076 0.312
#> GSM1269699 1 0.733 0.37501 0.500 0.200 0.000 0.060 0.240
#> GSM1269707 1 0.700 0.31335 0.452 0.128 0.000 0.044 0.376
#> GSM1269651 2 0.783 0.24956 0.000 0.372 0.356 0.088 0.184
#> GSM1269659 5 0.767 0.19893 0.012 0.160 0.192 0.104 0.532
#> GSM1269667 3 0.645 0.45510 0.024 0.156 0.632 0.172 0.016
#> GSM1269675 3 0.667 0.05309 0.020 0.368 0.516 0.064 0.032
#> GSM1269683 3 0.669 -0.19853 0.032 0.072 0.512 0.368 0.016
#> GSM1269689 2 0.765 0.00183 0.080 0.428 0.392 0.048 0.052
#> GSM1269697 3 0.491 0.43370 0.004 0.200 0.732 0.040 0.024
#> GSM1269705 3 0.600 0.11474 0.000 0.312 0.592 0.044 0.052
#> GSM1269713 3 0.429 0.50783 0.008 0.160 0.784 0.040 0.008
#> GSM1269719 3 0.700 0.30491 0.004 0.188 0.592 0.116 0.100
#> GSM1269725 3 0.418 0.50423 0.000 0.168 0.776 0.052 0.004
#> GSM1269727 3 0.451 0.48391 0.000 0.112 0.776 0.100 0.012
#> GSM1269649 1 0.568 0.58500 0.712 0.144 0.004 0.060 0.080
#> GSM1269657 5 0.547 0.04342 0.356 0.048 0.000 0.012 0.584
#> GSM1269665 1 0.724 0.39423 0.472 0.076 0.004 0.352 0.096
#> GSM1269673 1 0.413 0.61428 0.804 0.016 0.000 0.120 0.060
#> GSM1269681 2 0.777 -0.22566 0.316 0.400 0.012 0.040 0.232
#> GSM1269687 1 0.493 0.60471 0.748 0.028 0.000 0.072 0.152
#> GSM1269695 1 0.344 0.60952 0.856 0.040 0.000 0.024 0.080
#> GSM1269703 1 0.490 0.61140 0.764 0.044 0.000 0.116 0.076
#> GSM1269711 1 0.668 0.50132 0.612 0.196 0.008 0.048 0.136
#> GSM1269646 3 0.543 0.32646 0.008 0.304 0.636 0.036 0.016
#> GSM1269654 3 0.570 0.44318 0.000 0.140 0.684 0.148 0.028
#> GSM1269662 3 0.772 0.06913 0.000 0.260 0.472 0.132 0.136
#> GSM1269670 2 0.655 0.26779 0.032 0.508 0.388 0.020 0.052
#> GSM1269678 3 0.512 0.44086 0.004 0.068 0.712 0.204 0.012
#> GSM1269692 4 0.851 0.51937 0.024 0.100 0.288 0.384 0.204
#> GSM1269700 3 0.477 0.52295 0.012 0.140 0.752 0.096 0.000
#> GSM1269708 3 0.637 0.44491 0.020 0.188 0.660 0.068 0.064
#> GSM1269714 3 0.693 -0.27512 0.004 0.072 0.516 0.332 0.076
#> GSM1269716 4 0.599 0.60421 0.008 0.016 0.404 0.520 0.052
#> GSM1269720 5 0.774 0.06081 0.008 0.248 0.224 0.060 0.460
#> GSM1269722 3 0.365 0.53353 0.000 0.088 0.840 0.056 0.016
#> GSM1269644 1 0.611 0.54383 0.620 0.032 0.000 0.100 0.248
#> GSM1269652 1 0.691 0.46545 0.564 0.160 0.004 0.044 0.228
#> GSM1269660 1 0.660 0.55006 0.632 0.116 0.000 0.124 0.128
#> GSM1269668 1 0.606 0.54187 0.640 0.064 0.008 0.248 0.040
#> GSM1269676 5 0.552 0.09168 0.320 0.048 0.000 0.020 0.612
#> GSM1269684 1 0.655 0.56438 0.604 0.056 0.000 0.120 0.220
#> GSM1269690 1 0.718 0.32040 0.440 0.024 0.000 0.252 0.284
#> GSM1269698 1 0.726 0.33362 0.456 0.180 0.000 0.044 0.320
#> GSM1269706 1 0.718 0.25383 0.428 0.128 0.000 0.056 0.388
#> GSM1269650 2 0.809 0.27745 0.012 0.364 0.340 0.068 0.216
#> GSM1269658 5 0.795 0.13527 0.016 0.148 0.180 0.144 0.512
#> GSM1269666 3 0.523 0.47538 0.004 0.096 0.716 0.172 0.012
#> GSM1269674 3 0.689 -0.10653 0.000 0.356 0.492 0.072 0.080
#> GSM1269682 3 0.711 -0.17196 0.016 0.092 0.520 0.320 0.052
#> GSM1269688 3 0.815 -0.02525 0.092 0.308 0.456 0.048 0.096
#> GSM1269696 3 0.579 0.08520 0.008 0.384 0.552 0.020 0.036
#> GSM1269704 3 0.529 0.23613 0.000 0.316 0.628 0.016 0.040
#> GSM1269712 3 0.557 0.34080 0.012 0.060 0.704 0.192 0.032
#> GSM1269718 3 0.716 0.35587 0.032 0.200 0.576 0.160 0.032
#> GSM1269724 3 0.392 0.52097 0.000 0.096 0.820 0.072 0.012
#> GSM1269726 3 0.561 0.29266 0.004 0.108 0.652 0.232 0.004
#> GSM1269648 1 0.434 0.59986 0.792 0.064 0.000 0.020 0.124
#> GSM1269656 1 0.544 0.37974 0.564 0.048 0.000 0.008 0.380
#> GSM1269664 1 0.684 0.40385 0.484 0.088 0.008 0.380 0.040
#> GSM1269672 1 0.456 0.60817 0.772 0.012 0.000 0.112 0.104
#> GSM1269680 1 0.713 0.26645 0.456 0.252 0.000 0.024 0.268
#> GSM1269686 1 0.505 0.60600 0.748 0.044 0.000 0.140 0.068
#> GSM1269694 1 0.383 0.60819 0.828 0.060 0.000 0.016 0.096
#> GSM1269702 1 0.492 0.48787 0.636 0.008 0.000 0.028 0.328
#> GSM1269710 1 0.434 0.60604 0.800 0.072 0.000 0.028 0.100
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM1269647 3 0.552 0.3653 0.016 0.144 0.712 0.052 0.044 0.032
#> GSM1269655 2 0.747 0.1106 0.008 0.332 0.316 0.276 0.056 0.012
#> GSM1269663 4 0.649 0.1038 0.000 0.104 0.404 0.436 0.024 0.032
#> GSM1269671 2 0.675 0.0861 0.052 0.436 0.412 0.044 0.044 0.012
#> GSM1269679 3 0.602 0.1496 0.008 0.036 0.552 0.324 0.072 0.008
#> GSM1269693 4 0.715 0.3745 0.004 0.120 0.096 0.572 0.136 0.072
#> GSM1269701 3 0.729 0.1879 0.032 0.056 0.484 0.276 0.140 0.012
#> GSM1269709 3 0.680 0.3102 0.056 0.068 0.612 0.136 0.012 0.116
#> GSM1269715 4 0.637 0.3544 0.016 0.036 0.032 0.616 0.208 0.092
#> GSM1269717 4 0.578 0.4137 0.008 0.028 0.048 0.656 0.212 0.048
#> GSM1269721 3 0.830 -0.2071 0.008 0.272 0.296 0.204 0.024 0.196
#> GSM1269723 3 0.570 0.0676 0.000 0.064 0.536 0.364 0.024 0.012
#> GSM1269645 5 0.667 0.3959 0.368 0.096 0.004 0.008 0.456 0.068
#> GSM1269653 1 0.739 0.2994 0.492 0.048 0.052 0.008 0.164 0.236
#> GSM1269661 1 0.697 -0.2701 0.444 0.064 0.000 0.036 0.360 0.096
#> GSM1269669 1 0.566 -0.2884 0.508 0.008 0.012 0.048 0.408 0.016
#> GSM1269677 6 0.349 0.3936 0.152 0.024 0.000 0.008 0.008 0.808
#> GSM1269685 1 0.631 0.3064 0.464 0.008 0.000 0.024 0.144 0.360
#> GSM1269691 1 0.655 0.2826 0.432 0.020 0.000 0.020 0.152 0.376
#> GSM1269699 1 0.755 0.2887 0.516 0.108 0.048 0.012 0.116 0.200
#> GSM1269707 1 0.807 0.2530 0.412 0.108 0.036 0.024 0.140 0.280
#> GSM1269651 2 0.741 0.4023 0.012 0.524 0.208 0.132 0.048 0.076
#> GSM1269659 6 0.723 0.3465 0.004 0.184 0.132 0.132 0.024 0.524
#> GSM1269667 3 0.712 0.1712 0.032 0.088 0.504 0.272 0.100 0.004
#> GSM1269675 3 0.669 -0.0317 0.040 0.336 0.512 0.056 0.032 0.024
#> GSM1269683 4 0.667 0.3874 0.016 0.044 0.184 0.576 0.164 0.016
#> GSM1269689 3 0.685 0.2634 0.092 0.136 0.604 0.028 0.120 0.020
#> GSM1269697 3 0.598 0.3636 0.016 0.140 0.636 0.172 0.028 0.008
#> GSM1269705 3 0.693 -0.1609 0.012 0.340 0.456 0.140 0.020 0.032
#> GSM1269713 3 0.586 0.2839 0.004 0.064 0.616 0.252 0.052 0.012
#> GSM1269719 2 0.833 0.0986 0.028 0.336 0.328 0.176 0.048 0.084
#> GSM1269725 3 0.578 0.3071 0.012 0.052 0.636 0.236 0.056 0.008
#> GSM1269727 4 0.576 0.0142 0.000 0.076 0.440 0.456 0.020 0.008
#> GSM1269649 1 0.592 0.1438 0.672 0.112 0.028 0.008 0.136 0.044
#> GSM1269657 6 0.483 0.2892 0.220 0.032 0.004 0.004 0.040 0.700
#> GSM1269665 5 0.699 0.4919 0.296 0.068 0.004 0.088 0.504 0.040
#> GSM1269673 1 0.624 -0.0432 0.576 0.040 0.004 0.032 0.280 0.068
#> GSM1269681 2 0.827 -0.1999 0.224 0.372 0.048 0.004 0.168 0.184
#> GSM1269687 1 0.627 0.0694 0.576 0.036 0.004 0.020 0.260 0.104
#> GSM1269695 1 0.492 0.2627 0.752 0.060 0.008 0.008 0.096 0.076
#> GSM1269703 1 0.634 -0.1996 0.516 0.032 0.000 0.020 0.324 0.108
#> GSM1269711 1 0.737 0.2223 0.536 0.064 0.080 0.008 0.196 0.116
#> GSM1269646 3 0.619 0.3692 0.036 0.100 0.680 0.072 0.084 0.028
#> GSM1269654 4 0.721 0.0861 0.004 0.200 0.316 0.416 0.040 0.024
#> GSM1269662 2 0.805 0.2498 0.004 0.340 0.288 0.236 0.048 0.084
#> GSM1269670 3 0.613 -0.1696 0.036 0.440 0.452 0.008 0.028 0.036
#> GSM1269678 4 0.626 0.0442 0.004 0.036 0.428 0.444 0.072 0.016
#> GSM1269692 4 0.866 0.0876 0.020 0.192 0.108 0.400 0.144 0.136
#> GSM1269700 3 0.557 0.1238 0.004 0.040 0.552 0.360 0.040 0.004
#> GSM1269708 3 0.724 0.2469 0.036 0.072 0.552 0.224 0.040 0.076
#> GSM1269714 4 0.606 0.3924 0.004 0.088 0.212 0.628 0.040 0.028
#> GSM1269716 4 0.515 0.4487 0.012 0.028 0.088 0.744 0.096 0.032
#> GSM1269720 6 0.741 0.2272 0.008 0.188 0.232 0.068 0.028 0.476
#> GSM1269722 3 0.595 0.1220 0.000 0.056 0.548 0.336 0.036 0.024
#> GSM1269644 1 0.700 -0.1174 0.436 0.076 0.000 0.016 0.344 0.128
#> GSM1269652 1 0.673 0.3395 0.536 0.032 0.036 0.008 0.116 0.272
#> GSM1269660 1 0.727 -0.3836 0.436 0.116 0.004 0.020 0.328 0.096
#> GSM1269668 1 0.607 -0.3095 0.500 0.012 0.016 0.064 0.388 0.020
#> GSM1269676 6 0.421 0.3367 0.200 0.028 0.000 0.004 0.024 0.744
#> GSM1269684 1 0.671 0.1800 0.492 0.024 0.000 0.024 0.252 0.208
#> GSM1269690 1 0.736 0.1116 0.380 0.016 0.000 0.068 0.236 0.300
#> GSM1269698 1 0.791 0.2440 0.420 0.160 0.040 0.004 0.128 0.248
#> GSM1269706 1 0.783 0.2598 0.428 0.100 0.040 0.016 0.120 0.296
#> GSM1269650 2 0.762 0.3872 0.012 0.508 0.204 0.108 0.052 0.116
#> GSM1269658 6 0.816 0.2117 0.016 0.244 0.104 0.136 0.068 0.432
#> GSM1269666 4 0.694 -0.0148 0.008 0.144 0.392 0.396 0.052 0.008
#> GSM1269674 2 0.710 0.2220 0.024 0.436 0.364 0.120 0.020 0.036
#> GSM1269682 4 0.700 0.4045 0.012 0.088 0.152 0.584 0.124 0.040
#> GSM1269688 3 0.687 0.2518 0.088 0.124 0.628 0.040 0.052 0.068
#> GSM1269696 3 0.563 0.2589 0.008 0.252 0.636 0.040 0.052 0.012
#> GSM1269704 3 0.528 0.2862 0.012 0.152 0.700 0.104 0.004 0.028
#> GSM1269712 4 0.633 0.2170 0.012 0.040 0.356 0.516 0.052 0.024
#> GSM1269718 3 0.884 0.0187 0.072 0.212 0.344 0.208 0.136 0.028
#> GSM1269724 3 0.625 0.1385 0.004 0.108 0.532 0.308 0.044 0.004
#> GSM1269726 4 0.673 0.2439 0.020 0.048 0.336 0.496 0.088 0.012
#> GSM1269648 1 0.416 0.3314 0.784 0.028 0.004 0.000 0.068 0.116
#> GSM1269656 1 0.677 0.2345 0.428 0.084 0.004 0.004 0.100 0.380
#> GSM1269664 5 0.678 0.5422 0.336 0.040 0.008 0.076 0.496 0.044
#> GSM1269672 1 0.654 0.0362 0.516 0.004 0.000 0.056 0.260 0.164
#> GSM1269680 1 0.807 0.0125 0.308 0.280 0.020 0.004 0.152 0.236
#> GSM1269686 1 0.586 -0.1913 0.560 0.020 0.004 0.024 0.332 0.060
#> GSM1269694 1 0.456 0.2565 0.776 0.060 0.008 0.004 0.084 0.068
#> GSM1269702 1 0.510 0.2876 0.504 0.008 0.000 0.004 0.048 0.436
#> GSM1269710 1 0.476 0.2703 0.760 0.032 0.012 0.008 0.108 0.080
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 agent(p) disease.state(p) gender(p) individual(p) k
#> CV:NMF 84 1.000 1.0000 3.81e-19 8.65e-05 2
#> CV:NMF 72 1.000 0.8134 1.59e-16 3.40e-04 3
#> CV:NMF 55 0.869 1.0000 9.66e-13 2.48e-03 4
#> CV:NMF 30 0.942 0.0684 3.06e-07 1.95e-03 5
#> CV:NMF 1 NA NA NA 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["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 51941 rows and 84 columns.
#> Top rows (1000, 2000, 3000, 4000, 5000) are extracted by 'MAD' method.
#> Subgroups are detected by 'hclust' method.
#> Performed in total 1250 partitions by row resampling.
#> Best k for subgroups seems to be 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 0.0208 0.323 0.728 0.3387 0.845 0.845
#> 3 3 0.0175 0.454 0.595 0.5777 0.617 0.567
#> 4 4 0.0730 0.410 0.568 0.1830 0.863 0.752
#> 5 5 0.1835 0.360 0.552 0.1025 0.936 0.855
#> 6 6 0.2481 0.378 0.533 0.0616 0.892 0.746
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
#> GSM1269647 1 0.955 -0.2745 0.624 0.376
#> GSM1269655 1 0.529 0.5996 0.880 0.120
#> GSM1269663 1 0.904 0.2721 0.680 0.320
#> GSM1269671 2 1.000 0.8391 0.492 0.508
#> GSM1269679 1 0.653 0.5932 0.832 0.168
#> GSM1269693 1 0.767 0.5323 0.776 0.224
#> GSM1269701 1 0.653 0.5870 0.832 0.168
#> GSM1269709 1 0.563 0.5958 0.868 0.132
#> GSM1269715 1 0.994 0.2167 0.544 0.456
#> GSM1269717 1 0.921 0.3652 0.664 0.336
#> GSM1269721 1 0.998 -0.6864 0.524 0.476
#> GSM1269723 1 0.634 0.5935 0.840 0.160
#> GSM1269645 1 0.788 0.5406 0.764 0.236
#> GSM1269653 1 0.990 -0.2435 0.560 0.440
#> GSM1269661 1 0.634 0.5828 0.840 0.160
#> GSM1269669 1 0.443 0.6022 0.908 0.092
#> GSM1269677 2 1.000 0.8633 0.492 0.508
#> GSM1269685 1 0.541 0.6052 0.876 0.124
#> GSM1269691 1 0.541 0.6044 0.876 0.124
#> GSM1269699 1 1.000 -0.7004 0.512 0.488
#> GSM1269707 1 0.994 -0.6320 0.544 0.456
#> GSM1269651 1 1.000 -0.8740 0.500 0.500
#> GSM1269659 1 0.997 -0.6772 0.532 0.468
#> GSM1269667 1 0.518 0.6065 0.884 0.116
#> GSM1269675 1 0.971 -0.4150 0.600 0.400
#> GSM1269683 1 0.584 0.5828 0.860 0.140
#> GSM1269689 1 0.971 -0.2103 0.600 0.400
#> GSM1269697 1 0.904 0.1577 0.680 0.320
#> GSM1269705 1 0.949 -0.3364 0.632 0.368
#> GSM1269713 1 0.833 0.4215 0.736 0.264
#> GSM1269719 1 0.760 0.4490 0.780 0.220
#> GSM1269725 1 0.795 0.4237 0.760 0.240
#> GSM1269727 1 0.605 0.5983 0.852 0.148
#> GSM1269649 1 0.653 0.5704 0.832 0.168
#> GSM1269657 1 0.900 0.1472 0.684 0.316
#> GSM1269665 1 0.563 0.6050 0.868 0.132
#> GSM1269673 1 0.506 0.5934 0.888 0.112
#> GSM1269681 2 0.996 0.8802 0.464 0.536
#> GSM1269687 1 0.430 0.6041 0.912 0.088
#> GSM1269695 1 0.443 0.6040 0.908 0.092
#> GSM1269703 1 0.494 0.6106 0.892 0.108
#> GSM1269711 1 0.563 0.5862 0.868 0.132
#> GSM1269646 1 0.955 -0.2745 0.624 0.376
#> GSM1269654 1 0.529 0.5996 0.880 0.120
#> GSM1269662 1 0.913 0.2124 0.672 0.328
#> GSM1269670 2 1.000 0.8391 0.492 0.508
#> GSM1269678 1 0.563 0.6103 0.868 0.132
#> GSM1269692 1 0.767 0.5406 0.776 0.224
#> GSM1269700 1 0.653 0.5870 0.832 0.168
#> GSM1269708 1 0.563 0.5958 0.868 0.132
#> GSM1269714 1 0.644 0.5946 0.836 0.164
#> GSM1269716 1 0.921 0.3652 0.664 0.336
#> GSM1269720 1 0.998 -0.6864 0.524 0.476
#> GSM1269722 1 0.662 0.5943 0.828 0.172
#> GSM1269644 1 0.563 0.6013 0.868 0.132
#> GSM1269652 1 0.925 -0.0752 0.660 0.340
#> GSM1269660 1 0.644 0.5840 0.836 0.164
#> GSM1269668 1 0.443 0.6014 0.908 0.092
#> GSM1269676 2 1.000 0.8633 0.492 0.508
#> GSM1269684 1 0.469 0.6046 0.900 0.100
#> GSM1269690 1 0.541 0.6044 0.876 0.124
#> GSM1269698 1 1.000 -0.7004 0.512 0.488
#> GSM1269706 1 0.994 -0.6320 0.544 0.456
#> GSM1269650 2 1.000 0.8511 0.500 0.500
#> GSM1269658 1 0.997 -0.6772 0.532 0.468
#> GSM1269666 1 0.529 0.6059 0.880 0.120
#> GSM1269674 1 0.971 -0.4150 0.600 0.400
#> GSM1269682 1 0.584 0.5959 0.860 0.140
#> GSM1269688 1 0.971 -0.2103 0.600 0.400
#> GSM1269696 1 0.929 0.0855 0.656 0.344
#> GSM1269704 1 0.949 -0.3364 0.632 0.368
#> GSM1269712 1 0.808 0.4617 0.752 0.248
#> GSM1269718 1 0.760 0.4490 0.780 0.220
#> GSM1269724 1 0.767 0.4656 0.776 0.224
#> GSM1269726 1 0.615 0.5982 0.848 0.152
#> GSM1269648 1 0.653 0.5644 0.832 0.168
#> GSM1269656 1 0.891 0.1777 0.692 0.308
#> GSM1269664 1 0.482 0.6069 0.896 0.104
#> GSM1269672 1 0.506 0.5959 0.888 0.112
#> GSM1269680 2 0.996 0.8802 0.464 0.536
#> GSM1269686 1 0.430 0.6041 0.912 0.088
#> GSM1269694 1 0.443 0.6040 0.908 0.092
#> GSM1269702 1 0.469 0.5961 0.900 0.100
#> GSM1269710 1 0.541 0.5948 0.876 0.124
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM1269647 2 0.843 0.3459 0.384 0.524 0.092
#> GSM1269655 1 0.755 0.6099 0.684 0.204 0.112
#> GSM1269663 1 0.933 0.2105 0.520 0.232 0.248
#> GSM1269671 2 0.806 0.5833 0.236 0.640 0.124
#> GSM1269679 1 0.727 0.5756 0.700 0.204 0.096
#> GSM1269693 1 0.875 0.2417 0.584 0.172 0.244
#> GSM1269701 1 0.762 0.5205 0.684 0.188 0.128
#> GSM1269709 1 0.694 0.5902 0.708 0.224 0.068
#> GSM1269715 3 0.711 0.0000 0.388 0.028 0.584
#> GSM1269717 1 0.780 -0.4926 0.520 0.052 0.428
#> GSM1269721 2 0.792 0.5629 0.248 0.644 0.108
#> GSM1269723 1 0.723 0.5626 0.708 0.188 0.104
#> GSM1269645 1 0.848 0.3690 0.616 0.196 0.188
#> GSM1269653 2 0.890 0.4148 0.320 0.536 0.144
#> GSM1269661 1 0.702 0.5762 0.704 0.224 0.072
#> GSM1269669 1 0.581 0.5857 0.800 0.108 0.092
#> GSM1269677 2 0.869 0.5373 0.176 0.592 0.232
#> GSM1269685 1 0.613 0.5990 0.780 0.136 0.084
#> GSM1269691 1 0.579 0.6062 0.800 0.116 0.084
#> GSM1269699 2 0.618 0.5662 0.236 0.732 0.032
#> GSM1269707 2 0.613 0.5527 0.268 0.712 0.020
#> GSM1269651 2 0.916 0.5329 0.204 0.540 0.256
#> GSM1269659 2 0.935 0.4785 0.256 0.516 0.228
#> GSM1269667 1 0.683 0.6104 0.740 0.148 0.112
#> GSM1269675 2 0.868 0.4699 0.352 0.532 0.116
#> GSM1269683 1 0.721 0.5375 0.716 0.128 0.156
#> GSM1269689 2 0.804 0.3620 0.372 0.556 0.072
#> GSM1269697 1 0.875 -0.0151 0.492 0.396 0.112
#> GSM1269705 2 0.830 0.3646 0.416 0.504 0.080
#> GSM1269713 1 0.833 0.2924 0.564 0.340 0.096
#> GSM1269719 1 0.844 0.3709 0.592 0.284 0.124
#> GSM1269725 1 0.829 0.3194 0.572 0.332 0.096
#> GSM1269727 1 0.769 0.5452 0.680 0.184 0.136
#> GSM1269649 1 0.677 0.5655 0.720 0.216 0.064
#> GSM1269657 2 0.879 0.1509 0.432 0.456 0.112
#> GSM1269665 1 0.611 0.5850 0.784 0.116 0.100
#> GSM1269673 1 0.573 0.6225 0.796 0.144 0.060
#> GSM1269681 2 0.886 0.5313 0.164 0.564 0.272
#> GSM1269687 1 0.542 0.6014 0.820 0.100 0.080
#> GSM1269695 1 0.537 0.6212 0.812 0.140 0.048
#> GSM1269703 1 0.560 0.6311 0.804 0.136 0.060
#> GSM1269711 1 0.640 0.5811 0.744 0.200 0.056
#> GSM1269646 2 0.844 0.3435 0.388 0.520 0.092
#> GSM1269654 1 0.755 0.6099 0.684 0.204 0.112
#> GSM1269662 1 0.947 0.1414 0.496 0.228 0.276
#> GSM1269670 2 0.806 0.5833 0.236 0.640 0.124
#> GSM1269678 1 0.668 0.6135 0.744 0.168 0.088
#> GSM1269692 1 0.859 0.2971 0.604 0.180 0.216
#> GSM1269700 1 0.762 0.5205 0.684 0.188 0.128
#> GSM1269708 1 0.702 0.5901 0.704 0.224 0.072
#> GSM1269714 1 0.760 0.5543 0.688 0.172 0.140
#> GSM1269716 1 0.781 -0.4986 0.516 0.052 0.432
#> GSM1269720 2 0.792 0.5629 0.248 0.644 0.108
#> GSM1269722 1 0.756 0.5631 0.676 0.224 0.100
#> GSM1269644 1 0.591 0.6251 0.788 0.144 0.068
#> GSM1269652 2 0.792 0.2179 0.468 0.476 0.056
#> GSM1269660 1 0.719 0.5762 0.696 0.224 0.080
#> GSM1269668 1 0.594 0.5879 0.792 0.120 0.088
#> GSM1269676 2 0.869 0.5373 0.176 0.592 0.232
#> GSM1269684 1 0.614 0.5768 0.780 0.132 0.088
#> GSM1269690 1 0.579 0.6062 0.800 0.116 0.084
#> GSM1269698 2 0.618 0.5662 0.236 0.732 0.032
#> GSM1269706 2 0.613 0.5527 0.268 0.712 0.020
#> GSM1269650 2 0.916 0.5329 0.204 0.540 0.256
#> GSM1269658 2 0.938 0.4752 0.260 0.512 0.228
#> GSM1269666 1 0.688 0.6093 0.736 0.156 0.108
#> GSM1269674 2 0.868 0.4699 0.352 0.532 0.116
#> GSM1269682 1 0.722 0.5693 0.716 0.152 0.132
#> GSM1269688 2 0.804 0.3620 0.372 0.556 0.072
#> GSM1269696 1 0.882 -0.0736 0.476 0.408 0.116
#> GSM1269704 2 0.830 0.3646 0.416 0.504 0.080
#> GSM1269712 1 0.853 0.3329 0.556 0.332 0.112
#> GSM1269718 1 0.844 0.3709 0.592 0.284 0.124
#> GSM1269724 1 0.806 0.3469 0.588 0.328 0.084
#> GSM1269726 1 0.770 0.5417 0.680 0.180 0.140
#> GSM1269648 1 0.663 0.5686 0.724 0.220 0.056
#> GSM1269656 1 0.879 -0.1449 0.452 0.436 0.112
#> GSM1269664 1 0.608 0.5934 0.784 0.128 0.088
#> GSM1269672 1 0.563 0.6315 0.800 0.144 0.056
#> GSM1269680 2 0.886 0.5313 0.164 0.564 0.272
#> GSM1269686 1 0.541 0.6007 0.820 0.104 0.076
#> GSM1269694 1 0.537 0.6212 0.812 0.140 0.048
#> GSM1269702 1 0.632 0.6251 0.760 0.172 0.068
#> GSM1269710 1 0.604 0.6020 0.772 0.172 0.056
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM1269647 2 0.795 0.4508 0.304 0.520 0.040 0.136
#> GSM1269655 1 0.690 0.5096 0.668 0.188 0.092 0.052
#> GSM1269663 1 0.916 -0.0500 0.396 0.148 0.120 0.336
#> GSM1269671 2 0.818 0.1928 0.080 0.500 0.092 0.328
#> GSM1269679 1 0.720 0.4942 0.640 0.208 0.096 0.056
#> GSM1269693 1 0.858 -0.0708 0.508 0.080 0.244 0.168
#> GSM1269701 1 0.715 0.4400 0.612 0.240 0.124 0.024
#> GSM1269709 1 0.722 0.5301 0.648 0.192 0.092 0.068
#> GSM1269715 3 0.484 0.5899 0.256 0.016 0.724 0.004
#> GSM1269717 3 0.655 0.7152 0.448 0.028 0.496 0.028
#> GSM1269721 4 0.807 0.1963 0.152 0.348 0.032 0.468
#> GSM1269723 1 0.701 0.4889 0.648 0.208 0.104 0.040
#> GSM1269645 1 0.855 0.0588 0.536 0.200 0.160 0.104
#> GSM1269653 2 0.856 0.3878 0.228 0.504 0.068 0.200
#> GSM1269661 1 0.797 0.4569 0.588 0.208 0.100 0.104
#> GSM1269669 1 0.602 0.4966 0.728 0.144 0.104 0.024
#> GSM1269677 4 0.507 0.5422 0.076 0.104 0.024 0.796
#> GSM1269685 1 0.652 0.5023 0.716 0.108 0.104 0.072
#> GSM1269691 1 0.600 0.5134 0.752 0.080 0.080 0.088
#> GSM1269699 2 0.691 0.4384 0.140 0.624 0.012 0.224
#> GSM1269707 2 0.752 0.4413 0.164 0.588 0.028 0.220
#> GSM1269651 4 0.582 0.5167 0.092 0.096 0.052 0.760
#> GSM1269659 4 0.708 0.5014 0.164 0.136 0.044 0.656
#> GSM1269667 1 0.650 0.5238 0.712 0.136 0.092 0.060
#> GSM1269675 2 0.834 0.4189 0.212 0.540 0.072 0.176
#> GSM1269683 1 0.694 0.3823 0.668 0.072 0.188 0.072
#> GSM1269689 2 0.743 0.4947 0.288 0.580 0.052 0.080
#> GSM1269697 1 0.860 -0.1533 0.408 0.392 0.088 0.112
#> GSM1269705 2 0.812 0.4610 0.348 0.472 0.040 0.140
#> GSM1269713 1 0.844 0.1886 0.500 0.296 0.088 0.116
#> GSM1269719 1 0.844 0.2492 0.544 0.156 0.096 0.204
#> GSM1269725 1 0.774 0.2140 0.516 0.348 0.060 0.076
#> GSM1269727 1 0.759 0.4314 0.628 0.160 0.136 0.076
#> GSM1269649 1 0.671 0.4951 0.656 0.236 0.064 0.044
#> GSM1269657 4 0.873 0.0681 0.328 0.220 0.048 0.404
#> GSM1269665 1 0.648 0.4634 0.720 0.104 0.096 0.080
#> GSM1269673 1 0.613 0.5444 0.740 0.116 0.060 0.084
#> GSM1269681 4 0.469 0.5080 0.048 0.084 0.044 0.824
#> GSM1269687 1 0.593 0.5125 0.756 0.092 0.084 0.068
#> GSM1269695 1 0.573 0.5489 0.740 0.172 0.060 0.028
#> GSM1269703 1 0.573 0.5446 0.756 0.132 0.076 0.036
#> GSM1269711 1 0.630 0.5187 0.668 0.248 0.060 0.024
#> GSM1269646 2 0.797 0.4500 0.308 0.516 0.040 0.136
#> GSM1269654 1 0.690 0.5096 0.668 0.188 0.092 0.052
#> GSM1269662 4 0.891 -0.0117 0.380 0.140 0.096 0.384
#> GSM1269670 2 0.818 0.1928 0.080 0.500 0.092 0.328
#> GSM1269678 1 0.682 0.5271 0.692 0.128 0.112 0.068
#> GSM1269692 1 0.842 0.0609 0.540 0.084 0.204 0.172
#> GSM1269700 1 0.715 0.4400 0.612 0.240 0.124 0.024
#> GSM1269708 1 0.727 0.5289 0.644 0.192 0.096 0.068
#> GSM1269714 1 0.766 0.4226 0.620 0.132 0.172 0.076
#> GSM1269716 3 0.663 0.7179 0.444 0.032 0.496 0.028
#> GSM1269720 4 0.807 0.1963 0.152 0.348 0.032 0.468
#> GSM1269722 1 0.737 0.4880 0.628 0.212 0.096 0.064
#> GSM1269644 1 0.638 0.5426 0.724 0.120 0.068 0.088
#> GSM1269652 2 0.815 0.3561 0.368 0.436 0.028 0.168
#> GSM1269660 1 0.807 0.4529 0.580 0.208 0.108 0.104
#> GSM1269668 1 0.601 0.5013 0.728 0.148 0.100 0.024
#> GSM1269676 4 0.507 0.5422 0.076 0.104 0.024 0.796
#> GSM1269684 1 0.624 0.4807 0.736 0.072 0.100 0.092
#> GSM1269690 1 0.600 0.5134 0.752 0.080 0.080 0.088
#> GSM1269698 2 0.691 0.4384 0.140 0.624 0.012 0.224
#> GSM1269706 2 0.752 0.4413 0.164 0.588 0.028 0.220
#> GSM1269650 4 0.582 0.5167 0.092 0.096 0.052 0.760
#> GSM1269658 4 0.712 0.4983 0.168 0.136 0.044 0.652
#> GSM1269666 1 0.651 0.5222 0.712 0.136 0.088 0.064
#> GSM1269674 2 0.834 0.4189 0.212 0.540 0.072 0.176
#> GSM1269682 1 0.699 0.4288 0.672 0.092 0.168 0.068
#> GSM1269688 2 0.743 0.4947 0.288 0.580 0.052 0.080
#> GSM1269696 2 0.870 0.1731 0.384 0.404 0.088 0.124
#> GSM1269704 2 0.812 0.4610 0.348 0.472 0.040 0.140
#> GSM1269712 1 0.860 0.2315 0.500 0.276 0.108 0.116
#> GSM1269718 1 0.844 0.2492 0.544 0.156 0.096 0.204
#> GSM1269724 1 0.742 0.2655 0.544 0.340 0.056 0.060
#> GSM1269726 1 0.763 0.4275 0.624 0.164 0.136 0.076
#> GSM1269648 1 0.642 0.5071 0.676 0.228 0.060 0.036
#> GSM1269656 4 0.872 0.0399 0.348 0.212 0.048 0.392
#> GSM1269664 1 0.619 0.5005 0.740 0.092 0.088 0.080
#> GSM1269672 1 0.607 0.5549 0.740 0.132 0.056 0.072
#> GSM1269680 4 0.469 0.5080 0.048 0.084 0.044 0.824
#> GSM1269686 1 0.580 0.5107 0.764 0.088 0.080 0.068
#> GSM1269694 1 0.573 0.5489 0.740 0.172 0.060 0.028
#> GSM1269702 1 0.682 0.5323 0.696 0.120 0.084 0.100
#> GSM1269710 1 0.622 0.5401 0.680 0.232 0.068 0.020
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM1269647 5 0.786 0.38870 0.264 0.048 0.140 0.048 0.500
#> GSM1269655 1 0.705 0.40420 0.576 0.040 0.224 0.020 0.140
#> GSM1269663 3 0.891 0.39480 0.248 0.256 0.356 0.068 0.072
#> GSM1269671 5 0.811 0.22304 0.044 0.248 0.072 0.148 0.488
#> GSM1269679 1 0.636 0.41723 0.624 0.016 0.224 0.020 0.116
#> GSM1269693 3 0.810 -0.17480 0.376 0.092 0.384 0.124 0.024
#> GSM1269701 1 0.611 0.38416 0.600 0.008 0.272 0.008 0.112
#> GSM1269709 1 0.634 0.44821 0.648 0.024 0.196 0.024 0.108
#> GSM1269715 4 0.582 0.46044 0.164 0.000 0.228 0.608 0.000
#> GSM1269717 4 0.734 0.67950 0.316 0.012 0.284 0.380 0.008
#> GSM1269721 2 0.775 0.25383 0.108 0.436 0.096 0.012 0.348
#> GSM1269723 1 0.644 0.39737 0.604 0.016 0.252 0.020 0.108
#> GSM1269645 3 0.818 0.07414 0.368 0.028 0.380 0.080 0.144
#> GSM1269653 5 0.895 0.36533 0.216 0.096 0.192 0.080 0.416
#> GSM1269661 1 0.719 0.32198 0.564 0.040 0.228 0.024 0.144
#> GSM1269669 1 0.487 0.43897 0.720 0.008 0.220 0.008 0.044
#> GSM1269677 2 0.461 0.59210 0.024 0.804 0.036 0.044 0.092
#> GSM1269685 1 0.633 0.40846 0.680 0.036 0.160 0.052 0.072
#> GSM1269691 1 0.588 0.41778 0.680 0.036 0.212 0.024 0.048
#> GSM1269699 5 0.649 0.44524 0.144 0.152 0.036 0.020 0.648
#> GSM1269707 5 0.719 0.44483 0.172 0.148 0.052 0.032 0.596
#> GSM1269651 2 0.482 0.55128 0.024 0.792 0.072 0.032 0.080
#> GSM1269659 2 0.738 0.52396 0.096 0.616 0.124 0.064 0.100
#> GSM1269667 1 0.552 0.44970 0.676 0.012 0.240 0.016 0.056
#> GSM1269675 5 0.786 0.42088 0.140 0.072 0.108 0.108 0.572
#> GSM1269683 1 0.686 0.31079 0.584 0.036 0.268 0.072 0.040
#> GSM1269689 5 0.746 0.43080 0.268 0.040 0.144 0.028 0.520
#> GSM1269697 1 0.862 -0.09530 0.372 0.048 0.160 0.088 0.332
#> GSM1269705 5 0.794 0.40303 0.296 0.128 0.072 0.032 0.472
#> GSM1269713 1 0.836 0.16074 0.452 0.056 0.216 0.056 0.220
#> GSM1269719 1 0.806 -0.05055 0.420 0.084 0.360 0.044 0.092
#> GSM1269725 1 0.801 0.20649 0.468 0.060 0.172 0.032 0.268
#> GSM1269727 1 0.667 0.30501 0.580 0.032 0.288 0.028 0.072
#> GSM1269649 1 0.625 0.38945 0.632 0.004 0.136 0.028 0.200
#> GSM1269657 2 0.850 0.05185 0.308 0.388 0.124 0.024 0.156
#> GSM1269665 1 0.638 0.29499 0.620 0.012 0.252 0.056 0.060
#> GSM1269673 1 0.534 0.43134 0.692 0.016 0.232 0.012 0.048
#> GSM1269681 2 0.352 0.55536 0.008 0.864 0.032 0.044 0.052
#> GSM1269687 1 0.517 0.43354 0.740 0.028 0.172 0.020 0.040
#> GSM1269695 1 0.510 0.48104 0.744 0.016 0.136 0.008 0.096
#> GSM1269703 1 0.545 0.44873 0.724 0.024 0.168 0.020 0.064
#> GSM1269711 1 0.557 0.45888 0.700 0.004 0.140 0.020 0.136
#> GSM1269646 5 0.789 0.39028 0.264 0.048 0.144 0.048 0.496
#> GSM1269654 1 0.705 0.40420 0.576 0.040 0.224 0.020 0.140
#> GSM1269662 3 0.899 0.33541 0.220 0.284 0.344 0.088 0.064
#> GSM1269670 5 0.811 0.22304 0.044 0.248 0.072 0.148 0.488
#> GSM1269678 1 0.579 0.46684 0.672 0.012 0.220 0.024 0.072
#> GSM1269692 1 0.822 -0.15394 0.444 0.084 0.304 0.128 0.040
#> GSM1269700 1 0.613 0.38001 0.596 0.008 0.276 0.008 0.112
#> GSM1269708 1 0.628 0.44924 0.648 0.024 0.200 0.020 0.108
#> GSM1269714 1 0.672 0.34380 0.572 0.012 0.288 0.072 0.056
#> GSM1269716 4 0.734 0.68207 0.312 0.012 0.288 0.380 0.008
#> GSM1269720 2 0.775 0.25383 0.108 0.436 0.096 0.012 0.348
#> GSM1269722 1 0.629 0.39491 0.608 0.024 0.268 0.012 0.088
#> GSM1269644 1 0.609 0.43740 0.672 0.044 0.200 0.020 0.064
#> GSM1269652 5 0.850 0.26625 0.348 0.092 0.112 0.060 0.388
#> GSM1269660 1 0.715 0.31986 0.560 0.040 0.236 0.020 0.144
#> GSM1269668 1 0.492 0.44641 0.724 0.008 0.208 0.008 0.052
#> GSM1269676 2 0.461 0.59210 0.024 0.804 0.036 0.044 0.092
#> GSM1269684 1 0.592 0.38548 0.688 0.036 0.196 0.048 0.032
#> GSM1269690 1 0.588 0.41778 0.680 0.036 0.212 0.024 0.048
#> GSM1269698 5 0.649 0.44524 0.144 0.152 0.036 0.020 0.648
#> GSM1269706 5 0.719 0.44483 0.172 0.148 0.052 0.032 0.596
#> GSM1269650 2 0.482 0.55128 0.024 0.792 0.072 0.032 0.080
#> GSM1269658 2 0.743 0.51961 0.100 0.612 0.124 0.064 0.100
#> GSM1269666 1 0.547 0.44950 0.672 0.012 0.248 0.012 0.056
#> GSM1269674 5 0.786 0.42088 0.140 0.072 0.108 0.108 0.572
#> GSM1269682 1 0.680 0.36796 0.624 0.036 0.212 0.072 0.056
#> GSM1269688 5 0.746 0.43080 0.268 0.040 0.144 0.028 0.520
#> GSM1269696 1 0.884 -0.15120 0.360 0.056 0.160 0.104 0.320
#> GSM1269704 5 0.794 0.40303 0.296 0.128 0.072 0.032 0.472
#> GSM1269712 1 0.824 0.21022 0.452 0.052 0.244 0.048 0.204
#> GSM1269718 1 0.806 -0.05055 0.420 0.084 0.360 0.044 0.092
#> GSM1269724 1 0.766 0.27968 0.500 0.044 0.164 0.028 0.264
#> GSM1269726 1 0.674 0.29631 0.572 0.032 0.292 0.028 0.076
#> GSM1269648 1 0.564 0.44027 0.680 0.004 0.108 0.016 0.192
#> GSM1269656 2 0.850 0.00403 0.324 0.376 0.128 0.024 0.148
#> GSM1269664 1 0.572 0.38284 0.664 0.004 0.236 0.032 0.064
#> GSM1269672 1 0.530 0.45673 0.712 0.028 0.196 0.004 0.060
#> GSM1269680 2 0.352 0.55536 0.008 0.864 0.032 0.044 0.052
#> GSM1269686 1 0.516 0.43595 0.740 0.024 0.172 0.020 0.044
#> GSM1269694 1 0.510 0.48104 0.744 0.016 0.136 0.008 0.096
#> GSM1269702 1 0.642 0.42560 0.668 0.064 0.164 0.024 0.080
#> GSM1269710 1 0.539 0.46949 0.716 0.004 0.148 0.020 0.112
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM1269647 3 0.739 0.3000 0.172 0.048 0.504 0.012 0.216 0.048
#> GSM1269655 1 0.728 0.4283 0.544 0.068 0.220 0.100 0.044 0.024
#> GSM1269663 2 0.711 0.5021 0.172 0.564 0.068 0.044 0.012 0.140
#> GSM1269671 5 0.363 0.6489 0.008 0.020 0.068 0.020 0.844 0.040
#> GSM1269679 1 0.680 0.3538 0.512 0.092 0.292 0.084 0.016 0.004
#> GSM1269693 1 0.779 -0.1570 0.388 0.172 0.056 0.304 0.000 0.080
#> GSM1269701 1 0.679 0.3962 0.500 0.096 0.280 0.116 0.008 0.000
#> GSM1269709 1 0.685 0.4417 0.592 0.108 0.184 0.064 0.028 0.024
#> GSM1269715 4 0.267 0.4437 0.100 0.012 0.004 0.872 0.012 0.000
#> GSM1269717 4 0.498 0.7246 0.284 0.032 0.020 0.648 0.000 0.016
#> GSM1269721 6 0.772 0.3299 0.080 0.044 0.236 0.028 0.120 0.492
#> GSM1269723 1 0.694 0.4353 0.504 0.128 0.252 0.104 0.012 0.000
#> GSM1269645 2 0.853 0.0314 0.308 0.312 0.140 0.112 0.124 0.004
#> GSM1269653 3 0.663 0.2286 0.092 0.060 0.632 0.016 0.060 0.140
#> GSM1269661 1 0.768 0.3959 0.536 0.156 0.144 0.048 0.072 0.044
#> GSM1269669 1 0.621 0.4916 0.632 0.124 0.144 0.076 0.024 0.000
#> GSM1269677 6 0.209 0.5425 0.016 0.004 0.032 0.000 0.028 0.920
#> GSM1269685 1 0.595 0.4970 0.684 0.088 0.100 0.088 0.008 0.032
#> GSM1269691 1 0.585 0.5033 0.700 0.092 0.088 0.072 0.016 0.032
#> GSM1269699 3 0.757 0.0671 0.072 0.036 0.372 0.004 0.360 0.156
#> GSM1269707 3 0.791 0.1573 0.092 0.040 0.392 0.008 0.296 0.172
#> GSM1269651 6 0.646 0.4604 0.008 0.180 0.048 0.032 0.120 0.612
#> GSM1269659 6 0.513 0.4661 0.092 0.088 0.048 0.016 0.012 0.744
#> GSM1269667 1 0.667 0.4736 0.592 0.100 0.156 0.120 0.032 0.000
#> GSM1269675 5 0.667 0.5820 0.076 0.132 0.236 0.004 0.544 0.008
#> GSM1269683 1 0.673 0.3292 0.568 0.076 0.080 0.240 0.016 0.020
#> GSM1269689 3 0.699 0.2849 0.152 0.072 0.580 0.012 0.144 0.040
#> GSM1269697 3 0.747 0.3337 0.240 0.076 0.484 0.020 0.160 0.020
#> GSM1269705 3 0.836 0.1988 0.216 0.072 0.312 0.008 0.304 0.088
#> GSM1269713 3 0.816 0.1567 0.308 0.084 0.416 0.068 0.056 0.068
#> GSM1269719 1 0.801 0.1278 0.456 0.268 0.104 0.052 0.048 0.072
#> GSM1269725 3 0.698 -0.0148 0.396 0.032 0.436 0.032 0.068 0.036
#> GSM1269727 1 0.726 0.3652 0.504 0.148 0.176 0.156 0.008 0.008
#> GSM1269649 1 0.686 0.3843 0.564 0.096 0.212 0.024 0.096 0.008
#> GSM1269657 6 0.782 0.1244 0.260 0.056 0.192 0.036 0.024 0.432
#> GSM1269665 1 0.662 0.3852 0.604 0.184 0.092 0.068 0.048 0.004
#> GSM1269673 1 0.573 0.5077 0.684 0.152 0.088 0.040 0.028 0.008
#> GSM1269681 6 0.591 0.4733 0.004 0.132 0.032 0.032 0.140 0.660
#> GSM1269687 1 0.543 0.5231 0.724 0.112 0.068 0.060 0.012 0.024
#> GSM1269695 1 0.546 0.5286 0.676 0.088 0.184 0.032 0.020 0.000
#> GSM1269703 1 0.592 0.5158 0.668 0.144 0.112 0.040 0.028 0.008
#> GSM1269711 1 0.552 0.4456 0.608 0.068 0.288 0.020 0.016 0.000
#> GSM1269646 3 0.744 0.2968 0.172 0.052 0.500 0.012 0.216 0.048
#> GSM1269654 1 0.728 0.4283 0.544 0.068 0.220 0.100 0.044 0.024
#> GSM1269662 2 0.624 0.4626 0.124 0.624 0.052 0.012 0.012 0.176
#> GSM1269670 5 0.363 0.6489 0.008 0.020 0.068 0.020 0.844 0.040
#> GSM1269678 1 0.642 0.4797 0.596 0.084 0.212 0.088 0.012 0.008
#> GSM1269692 1 0.779 0.0689 0.448 0.176 0.076 0.220 0.000 0.080
#> GSM1269700 1 0.679 0.3919 0.500 0.100 0.280 0.112 0.008 0.000
#> GSM1269708 1 0.687 0.4444 0.592 0.108 0.180 0.068 0.028 0.024
#> GSM1269714 1 0.676 0.3300 0.532 0.092 0.116 0.248 0.008 0.004
#> GSM1269716 4 0.497 0.7259 0.280 0.032 0.020 0.652 0.000 0.016
#> GSM1269720 6 0.772 0.3299 0.080 0.044 0.236 0.028 0.120 0.492
#> GSM1269722 1 0.721 0.4250 0.492 0.120 0.252 0.116 0.012 0.008
#> GSM1269644 1 0.653 0.5178 0.648 0.116 0.092 0.080 0.024 0.040
#> GSM1269652 3 0.763 0.3474 0.284 0.060 0.452 0.008 0.064 0.132
#> GSM1269660 1 0.771 0.3941 0.532 0.156 0.148 0.048 0.072 0.044
#> GSM1269668 1 0.622 0.4997 0.624 0.124 0.160 0.072 0.020 0.000
#> GSM1269676 6 0.209 0.5425 0.016 0.004 0.032 0.000 0.028 0.920
#> GSM1269684 1 0.593 0.4895 0.684 0.112 0.076 0.092 0.012 0.024
#> GSM1269690 1 0.585 0.5033 0.700 0.092 0.088 0.072 0.016 0.032
#> GSM1269698 3 0.757 0.0671 0.072 0.036 0.372 0.004 0.360 0.156
#> GSM1269706 3 0.791 0.1573 0.092 0.040 0.392 0.008 0.296 0.172
#> GSM1269650 6 0.646 0.4604 0.008 0.180 0.048 0.032 0.120 0.612
#> GSM1269658 6 0.517 0.4621 0.096 0.088 0.048 0.016 0.012 0.740
#> GSM1269666 1 0.667 0.4729 0.592 0.104 0.156 0.116 0.032 0.000
#> GSM1269674 5 0.667 0.5820 0.076 0.132 0.236 0.004 0.544 0.008
#> GSM1269682 1 0.661 0.4024 0.604 0.076 0.088 0.192 0.012 0.028
#> GSM1269688 3 0.699 0.2849 0.152 0.072 0.580 0.012 0.144 0.040
#> GSM1269696 3 0.749 0.3202 0.220 0.080 0.496 0.016 0.160 0.028
#> GSM1269704 3 0.836 0.1988 0.216 0.072 0.312 0.008 0.304 0.088
#> GSM1269712 3 0.839 0.0893 0.332 0.092 0.368 0.088 0.048 0.072
#> GSM1269718 1 0.801 0.1278 0.456 0.268 0.104 0.052 0.048 0.072
#> GSM1269724 1 0.688 0.0342 0.436 0.040 0.404 0.028 0.064 0.028
#> GSM1269726 1 0.728 0.3541 0.500 0.152 0.180 0.152 0.008 0.008
#> GSM1269648 1 0.613 0.4231 0.616 0.080 0.204 0.016 0.084 0.000
#> GSM1269656 6 0.793 0.0895 0.276 0.060 0.184 0.036 0.028 0.416
#> GSM1269664 1 0.652 0.4564 0.620 0.156 0.112 0.052 0.056 0.004
#> GSM1269672 1 0.545 0.5329 0.716 0.120 0.092 0.032 0.028 0.012
#> GSM1269680 6 0.591 0.4733 0.004 0.132 0.032 0.032 0.140 0.660
#> GSM1269686 1 0.540 0.5266 0.724 0.112 0.076 0.056 0.012 0.020
#> GSM1269694 1 0.546 0.5286 0.676 0.088 0.184 0.032 0.020 0.000
#> GSM1269702 1 0.632 0.5145 0.664 0.108 0.104 0.044 0.020 0.060
#> GSM1269710 1 0.554 0.4632 0.620 0.080 0.264 0.024 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 agent(p) disease.state(p) gender(p) individual(p) k
#> MAD:hclust 50 1.000 1.000 1.000000 0.002131 2
#> MAD:hclust 54 1.000 0.282 1.000000 0.001521 3
#> MAD:hclust 34 0.598 0.103 0.056550 0.003749 4
#> MAD:hclust 10 1.000 0.259 0.628299 0.040428 5
#> MAD:hclust 20 0.896 0.122 0.000499 0.000176 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 51941 rows and 84 columns.
#> Top rows (1000, 2000, 3000, 4000, 5000) are extracted by 'MAD' method.
#> Subgroups are detected by 'kmeans' method.
#> Performed in total 1250 partitions by row resampling.
#> Best k for subgroups seems to be 2.
#>
#> Following methods can be applied to this 'ConsensusPartition' object:
#> [1] "cola_report" "collect_classes" "collect_plots"
#> [4] "collect_stats" "colnames" "compare_signatures"
#> [7] "consensus_heatmap" "dimension_reduction" "functional_enrichment"
#> [10] "get_anno_col" "get_anno" "get_classes"
#> [13] "get_consensus" "get_matrix" "get_membership"
#> [16] "get_param" "get_signatures" "get_stats"
#> [19] "is_best_k" "is_stable_k" "membership_heatmap"
#> [22] "ncol" "nrow" "plot_ecdf"
#> [25] "rownames" "select_partition_number" "show"
#> [28] "suggest_best_k" "test_to_known_factors"
collect_plots() function collects all the plots made from res for all k (number of partitions)
into one single page to provide an easy and fast comparison between different k.
collect_plots(res)
The plots are:
k and the heatmap of
predicted classes for each k.k.k.k.All the plots in panels can be made by individual functions and they are plotted later in this section.
select_partition_number() produces several plots showing different
statistics for choosing “optimized” k. There are following statistics:
k;k, the area increased is defined as \(A_k - A_{k-1}\).The detailed explanations of these statistics can be found in the cola vignette.
Generally speaking, lower PAC score, higher mean silhouette score or higher
concordance corresponds to better partition. Rand index and Jaccard index
measure how similar the current partition is compared to partition with k-1.
If they are too similar, we won't accept k is better than k-1.
select_partition_number(res)
The numeric values for all these statistics can be obtained by get_stats().
get_stats(res)
#> k 1-PAC mean_silhouette concordance area_increased Rand Jaccard
#> 2 2 0.360 0.844 0.863 0.4797 0.523 0.523
#> 3 3 0.277 0.506 0.649 0.3182 0.937 0.880
#> 4 4 0.346 0.405 0.609 0.1388 0.743 0.476
#> 5 5 0.417 0.347 0.591 0.0759 0.818 0.432
#> 6 6 0.459 0.376 0.563 0.0447 0.841 0.414
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
#> GSM1269647 2 0.2778 0.871 0.048 0.952
#> GSM1269655 1 0.4939 0.843 0.892 0.108
#> GSM1269663 1 0.5629 0.818 0.868 0.132
#> GSM1269671 2 0.2236 0.876 0.036 0.964
#> GSM1269679 1 0.6531 0.859 0.832 0.168
#> GSM1269693 1 0.2423 0.857 0.960 0.040
#> GSM1269701 1 0.6531 0.859 0.832 0.168
#> GSM1269709 1 0.7883 0.793 0.764 0.236
#> GSM1269715 1 0.0000 0.874 1.000 0.000
#> GSM1269717 1 0.0672 0.871 0.992 0.008
#> GSM1269721 2 0.6801 0.831 0.180 0.820
#> GSM1269723 1 0.6438 0.860 0.836 0.164
#> GSM1269645 1 0.4161 0.880 0.916 0.084
#> GSM1269653 2 0.4690 0.856 0.100 0.900
#> GSM1269661 1 0.7056 0.853 0.808 0.192
#> GSM1269669 1 0.6438 0.860 0.836 0.164
#> GSM1269677 2 0.6801 0.827 0.180 0.820
#> GSM1269685 1 0.3114 0.859 0.944 0.056
#> GSM1269691 1 0.2423 0.863 0.960 0.040
#> GSM1269699 2 0.2423 0.877 0.040 0.960
#> GSM1269707 2 0.4562 0.877 0.096 0.904
#> GSM1269651 2 0.6623 0.828 0.172 0.828
#> GSM1269659 2 0.7883 0.794 0.236 0.764
#> GSM1269667 1 0.6887 0.854 0.816 0.184
#> GSM1269675 2 0.2423 0.875 0.040 0.960
#> GSM1269683 1 0.1414 0.874 0.980 0.020
#> GSM1269689 2 0.4562 0.858 0.096 0.904
#> GSM1269697 2 0.4431 0.853 0.092 0.908
#> GSM1269705 2 0.2423 0.877 0.040 0.960
#> GSM1269713 2 0.6531 0.782 0.168 0.832
#> GSM1269719 1 0.5842 0.817 0.860 0.140
#> GSM1269725 2 0.6343 0.793 0.160 0.840
#> GSM1269727 1 0.6438 0.862 0.836 0.164
#> GSM1269649 1 0.7139 0.849 0.804 0.196
#> GSM1269657 2 0.7056 0.822 0.192 0.808
#> GSM1269665 1 0.2948 0.879 0.948 0.052
#> GSM1269673 1 0.1633 0.873 0.976 0.024
#> GSM1269681 2 0.5842 0.840 0.140 0.860
#> GSM1269687 1 0.1633 0.876 0.976 0.024
#> GSM1269695 1 0.6148 0.870 0.848 0.152
#> GSM1269703 1 0.1184 0.878 0.984 0.016
#> GSM1269711 1 0.7139 0.846 0.804 0.196
#> GSM1269646 2 0.2948 0.871 0.052 0.948
#> GSM1269654 1 0.4690 0.842 0.900 0.100
#> GSM1269662 1 0.7883 0.660 0.764 0.236
#> GSM1269670 2 0.2043 0.874 0.032 0.968
#> GSM1269678 1 0.6343 0.861 0.840 0.160
#> GSM1269692 1 0.2423 0.857 0.960 0.040
#> GSM1269700 1 0.6623 0.859 0.828 0.172
#> GSM1269708 1 0.6973 0.843 0.812 0.188
#> GSM1269714 1 0.0376 0.876 0.996 0.004
#> GSM1269716 1 0.0672 0.871 0.992 0.008
#> GSM1269720 2 0.6887 0.828 0.184 0.816
#> GSM1269722 1 0.6438 0.864 0.836 0.164
#> GSM1269644 1 0.3274 0.860 0.940 0.060
#> GSM1269652 2 0.5408 0.865 0.124 0.876
#> GSM1269660 1 0.7139 0.850 0.804 0.196
#> GSM1269668 1 0.6438 0.860 0.836 0.164
#> GSM1269676 2 0.6623 0.828 0.172 0.828
#> GSM1269684 1 0.1843 0.866 0.972 0.028
#> GSM1269690 1 0.2423 0.863 0.960 0.040
#> GSM1269698 2 0.2043 0.876 0.032 0.968
#> GSM1269706 2 0.4562 0.877 0.096 0.904
#> GSM1269650 2 0.6623 0.828 0.172 0.828
#> GSM1269658 2 0.9608 0.571 0.384 0.616
#> GSM1269666 1 0.6623 0.858 0.828 0.172
#> GSM1269674 2 0.2236 0.876 0.036 0.964
#> GSM1269682 1 0.2778 0.878 0.952 0.048
#> GSM1269688 2 0.4562 0.858 0.096 0.904
#> GSM1269696 2 0.4939 0.842 0.108 0.892
#> GSM1269704 2 0.2236 0.877 0.036 0.964
#> GSM1269712 1 0.7453 0.836 0.788 0.212
#> GSM1269718 1 0.5629 0.823 0.868 0.132
#> GSM1269724 1 0.9833 0.465 0.576 0.424
#> GSM1269726 1 0.6438 0.865 0.836 0.164
#> GSM1269648 1 0.7056 0.851 0.808 0.192
#> GSM1269656 2 0.8081 0.792 0.248 0.752
#> GSM1269664 1 0.3431 0.879 0.936 0.064
#> GSM1269672 1 0.0938 0.875 0.988 0.012
#> GSM1269680 2 0.6343 0.831 0.160 0.840
#> GSM1269686 1 0.0672 0.875 0.992 0.008
#> GSM1269694 1 0.6247 0.869 0.844 0.156
#> GSM1269702 1 0.2603 0.869 0.956 0.044
#> GSM1269710 1 0.6623 0.861 0.828 0.172
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM1269647 2 0.605 0.5581 0.052 0.768 0.180
#> GSM1269655 1 0.651 0.5736 0.688 0.028 0.284
#> GSM1269663 3 0.762 -0.0911 0.368 0.052 0.580
#> GSM1269671 2 0.287 0.5897 0.008 0.916 0.076
#> GSM1269679 1 0.589 0.6204 0.796 0.104 0.100
#> GSM1269693 1 0.553 0.5892 0.704 0.000 0.296
#> GSM1269701 1 0.681 0.6018 0.740 0.104 0.156
#> GSM1269709 1 0.949 0.4006 0.480 0.208 0.312
#> GSM1269715 1 0.452 0.6457 0.816 0.004 0.180
#> GSM1269717 1 0.458 0.6430 0.812 0.004 0.184
#> GSM1269721 2 0.665 0.1222 0.012 0.592 0.396
#> GSM1269723 1 0.741 0.5877 0.692 0.104 0.204
#> GSM1269645 1 0.607 0.6575 0.728 0.024 0.248
#> GSM1269653 2 0.673 0.5577 0.088 0.740 0.172
#> GSM1269661 1 0.720 0.6524 0.704 0.092 0.204
#> GSM1269669 1 0.459 0.6925 0.848 0.032 0.120
#> GSM1269677 2 0.663 0.0115 0.008 0.548 0.444
#> GSM1269685 1 0.650 0.5580 0.596 0.008 0.396
#> GSM1269691 1 0.630 0.5443 0.608 0.004 0.388
#> GSM1269699 2 0.245 0.5918 0.012 0.936 0.052
#> GSM1269707 2 0.511 0.5428 0.036 0.820 0.144
#> GSM1269651 2 0.714 0.1081 0.028 0.576 0.396
#> GSM1269659 3 0.744 0.4057 0.044 0.368 0.588
#> GSM1269667 1 0.579 0.6391 0.796 0.068 0.136
#> GSM1269675 2 0.344 0.5954 0.016 0.896 0.088
#> GSM1269683 1 0.280 0.6774 0.908 0.000 0.092
#> GSM1269689 2 0.684 0.5281 0.076 0.724 0.200
#> GSM1269697 2 0.683 0.5178 0.080 0.728 0.192
#> GSM1269705 2 0.200 0.6073 0.012 0.952 0.036
#> GSM1269713 2 0.939 0.2737 0.272 0.508 0.220
#> GSM1269719 1 0.776 0.4463 0.488 0.048 0.464
#> GSM1269725 2 0.919 0.2868 0.272 0.532 0.196
#> GSM1269727 1 0.517 0.6743 0.828 0.056 0.116
#> GSM1269649 1 0.930 0.4774 0.508 0.192 0.300
#> GSM1269657 3 0.688 0.1922 0.016 0.428 0.556
#> GSM1269665 1 0.502 0.6792 0.796 0.012 0.192
#> GSM1269673 1 0.649 0.5939 0.628 0.012 0.360
#> GSM1269681 2 0.659 0.2086 0.016 0.632 0.352
#> GSM1269687 1 0.552 0.6806 0.728 0.004 0.268
#> GSM1269695 1 0.874 0.5659 0.544 0.128 0.328
#> GSM1269703 1 0.550 0.6607 0.708 0.000 0.292
#> GSM1269711 1 0.919 0.5098 0.480 0.156 0.364
#> GSM1269646 2 0.644 0.5415 0.064 0.748 0.188
#> GSM1269654 1 0.618 0.5866 0.716 0.024 0.260
#> GSM1269662 3 0.800 0.2202 0.304 0.088 0.608
#> GSM1269670 2 0.294 0.5914 0.012 0.916 0.072
#> GSM1269678 1 0.548 0.6494 0.816 0.076 0.108
#> GSM1269692 1 0.586 0.5830 0.656 0.000 0.344
#> GSM1269700 1 0.703 0.5938 0.724 0.104 0.172
#> GSM1269708 1 0.911 0.4843 0.520 0.164 0.316
#> GSM1269714 1 0.478 0.6438 0.796 0.004 0.200
#> GSM1269716 1 0.458 0.6430 0.812 0.004 0.184
#> GSM1269720 2 0.669 0.0737 0.012 0.580 0.408
#> GSM1269722 1 0.670 0.6379 0.744 0.092 0.164
#> GSM1269644 1 0.636 0.5773 0.592 0.004 0.404
#> GSM1269652 2 0.687 0.4774 0.048 0.688 0.264
#> GSM1269660 1 0.798 0.6185 0.632 0.104 0.264
#> GSM1269668 1 0.429 0.6828 0.868 0.040 0.092
#> GSM1269676 2 0.663 0.0115 0.008 0.548 0.444
#> GSM1269684 1 0.550 0.6419 0.708 0.000 0.292
#> GSM1269690 1 0.626 0.5484 0.616 0.004 0.380
#> GSM1269698 2 0.245 0.5887 0.012 0.936 0.052
#> GSM1269706 2 0.541 0.5330 0.040 0.804 0.156
#> GSM1269650 2 0.713 0.1149 0.028 0.580 0.392
#> GSM1269658 3 0.756 0.4549 0.056 0.336 0.608
#> GSM1269666 1 0.538 0.6416 0.820 0.068 0.112
#> GSM1269674 2 0.344 0.5950 0.016 0.896 0.088
#> GSM1269682 1 0.268 0.6816 0.924 0.008 0.068
#> GSM1269688 2 0.670 0.5351 0.076 0.736 0.188
#> GSM1269696 2 0.744 0.4737 0.108 0.692 0.200
#> GSM1269704 2 0.171 0.6060 0.008 0.960 0.032
#> GSM1269712 1 0.815 0.5054 0.644 0.156 0.200
#> GSM1269718 1 0.767 0.4388 0.484 0.044 0.472
#> GSM1269724 1 0.950 0.0902 0.440 0.372 0.188
#> GSM1269726 1 0.531 0.6672 0.820 0.056 0.124
#> GSM1269648 1 0.934 0.5071 0.476 0.176 0.348
#> GSM1269656 3 0.873 0.3790 0.112 0.388 0.500
#> GSM1269664 1 0.506 0.6829 0.800 0.016 0.184
#> GSM1269672 1 0.617 0.6337 0.680 0.012 0.308
#> GSM1269680 2 0.645 0.1557 0.008 0.608 0.384
#> GSM1269686 1 0.525 0.6658 0.736 0.000 0.264
#> GSM1269694 1 0.868 0.5728 0.556 0.128 0.316
#> GSM1269702 1 0.665 0.5820 0.592 0.012 0.396
#> GSM1269710 1 0.789 0.6257 0.624 0.088 0.288
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM1269647 2 0.731 0.55016 0.020 0.548 0.324 0.108
#> GSM1269655 3 0.810 0.10298 0.416 0.064 0.428 0.092
#> GSM1269663 4 0.876 -0.01644 0.356 0.076 0.152 0.416
#> GSM1269671 2 0.400 0.56843 0.012 0.852 0.072 0.064
#> GSM1269679 3 0.510 0.53584 0.152 0.076 0.768 0.004
#> GSM1269693 1 0.691 0.28272 0.576 0.000 0.272 0.152
#> GSM1269701 3 0.521 0.49915 0.220 0.032 0.736 0.012
#> GSM1269709 3 0.868 0.19161 0.332 0.140 0.448 0.080
#> GSM1269715 1 0.643 0.23080 0.568 0.000 0.352 0.080
#> GSM1269717 1 0.639 0.25924 0.588 0.000 0.328 0.084
#> GSM1269721 4 0.706 0.44719 0.076 0.344 0.024 0.556
#> GSM1269723 3 0.554 0.53596 0.160 0.052 0.756 0.032
#> GSM1269645 1 0.720 0.37018 0.604 0.060 0.276 0.060
#> GSM1269653 2 0.856 0.52088 0.052 0.472 0.272 0.204
#> GSM1269661 3 0.751 0.16698 0.412 0.060 0.476 0.052
#> GSM1269669 1 0.539 0.17707 0.564 0.004 0.424 0.008
#> GSM1269677 4 0.516 0.61713 0.044 0.236 0.000 0.720
#> GSM1269685 1 0.437 0.52528 0.820 0.008 0.048 0.124
#> GSM1269691 1 0.443 0.52267 0.808 0.000 0.068 0.124
#> GSM1269699 2 0.469 0.51064 0.004 0.784 0.044 0.168
#> GSM1269707 2 0.806 0.42846 0.100 0.572 0.100 0.228
#> GSM1269651 4 0.518 0.56064 0.008 0.304 0.012 0.676
#> GSM1269659 4 0.511 0.60523 0.132 0.092 0.004 0.772
#> GSM1269667 3 0.570 0.50404 0.204 0.052 0.724 0.020
#> GSM1269675 2 0.490 0.56018 0.036 0.812 0.084 0.068
#> GSM1269683 3 0.607 -0.02975 0.456 0.000 0.500 0.044
#> GSM1269689 2 0.759 0.58009 0.056 0.600 0.232 0.112
#> GSM1269697 2 0.704 0.53395 0.032 0.564 0.340 0.064
#> GSM1269705 2 0.386 0.59978 0.012 0.860 0.064 0.064
#> GSM1269713 3 0.674 -0.08406 0.032 0.332 0.588 0.048
#> GSM1269719 1 0.792 0.37420 0.592 0.072 0.156 0.180
#> GSM1269725 3 0.696 -0.00459 0.036 0.332 0.576 0.056
#> GSM1269727 3 0.628 0.36623 0.308 0.020 0.628 0.044
#> GSM1269649 1 0.767 0.02432 0.444 0.124 0.412 0.020
#> GSM1269657 4 0.576 0.61534 0.128 0.160 0.000 0.712
#> GSM1269665 1 0.599 0.43071 0.676 0.020 0.260 0.044
#> GSM1269673 1 0.258 0.56097 0.912 0.000 0.052 0.036
#> GSM1269681 4 0.509 0.53883 0.000 0.348 0.012 0.640
#> GSM1269687 1 0.496 0.48695 0.732 0.008 0.240 0.020
#> GSM1269695 1 0.657 0.45983 0.684 0.144 0.148 0.024
#> GSM1269703 1 0.537 0.45650 0.704 0.008 0.256 0.032
#> GSM1269711 1 0.756 0.14951 0.516 0.092 0.356 0.036
#> GSM1269646 2 0.759 0.48750 0.032 0.508 0.360 0.100
#> GSM1269654 3 0.782 0.10381 0.424 0.048 0.440 0.088
#> GSM1269662 4 0.833 0.30763 0.288 0.104 0.092 0.516
#> GSM1269670 2 0.380 0.56737 0.008 0.860 0.072 0.060
#> GSM1269678 3 0.480 0.52727 0.204 0.032 0.760 0.004
#> GSM1269692 1 0.639 0.40435 0.652 0.000 0.192 0.156
#> GSM1269700 3 0.553 0.50870 0.212 0.040 0.728 0.020
#> GSM1269708 3 0.847 0.13617 0.392 0.108 0.420 0.080
#> GSM1269714 1 0.638 0.24721 0.580 0.000 0.340 0.080
#> GSM1269716 1 0.638 0.23603 0.580 0.000 0.340 0.080
#> GSM1269720 4 0.675 0.49827 0.076 0.316 0.016 0.592
#> GSM1269722 3 0.629 0.50763 0.240 0.048 0.676 0.036
#> GSM1269644 1 0.378 0.56140 0.860 0.008 0.052 0.080
#> GSM1269652 2 0.964 0.34326 0.196 0.392 0.192 0.220
#> GSM1269660 3 0.771 0.19404 0.380 0.060 0.492 0.068
#> GSM1269668 3 0.542 0.18505 0.440 0.004 0.548 0.008
#> GSM1269676 4 0.516 0.61713 0.044 0.236 0.000 0.720
#> GSM1269684 1 0.343 0.55501 0.868 0.004 0.100 0.028
#> GSM1269690 1 0.493 0.50617 0.776 0.000 0.092 0.132
#> GSM1269698 2 0.487 0.51180 0.008 0.776 0.044 0.172
#> GSM1269706 2 0.806 0.42846 0.100 0.572 0.100 0.228
#> GSM1269650 4 0.520 0.56383 0.008 0.308 0.012 0.672
#> GSM1269658 4 0.540 0.60304 0.144 0.092 0.008 0.756
#> GSM1269666 3 0.487 0.53036 0.168 0.040 0.780 0.012
#> GSM1269674 2 0.450 0.55996 0.020 0.828 0.088 0.064
#> GSM1269682 3 0.647 0.09374 0.420 0.008 0.520 0.052
#> GSM1269688 2 0.748 0.57982 0.052 0.612 0.220 0.116
#> GSM1269696 2 0.716 0.45503 0.020 0.504 0.396 0.080
#> GSM1269704 2 0.414 0.59010 0.008 0.840 0.060 0.092
#> GSM1269712 3 0.520 0.51337 0.116 0.068 0.788 0.028
#> GSM1269718 1 0.817 0.35026 0.564 0.072 0.180 0.184
#> GSM1269724 3 0.697 0.07091 0.052 0.332 0.576 0.040
#> GSM1269726 3 0.650 0.37628 0.308 0.020 0.616 0.056
#> GSM1269648 1 0.703 0.36983 0.620 0.144 0.220 0.016
#> GSM1269656 4 0.834 0.32283 0.388 0.184 0.032 0.396
#> GSM1269664 1 0.624 0.28771 0.596 0.020 0.352 0.032
#> GSM1269672 1 0.289 0.56434 0.896 0.000 0.068 0.036
#> GSM1269680 4 0.504 0.54401 0.000 0.336 0.012 0.652
#> GSM1269686 1 0.394 0.52131 0.800 0.000 0.188 0.012
#> GSM1269694 1 0.657 0.46117 0.684 0.148 0.144 0.024
#> GSM1269702 1 0.390 0.54936 0.856 0.012 0.048 0.084
#> GSM1269710 1 0.662 0.39583 0.632 0.068 0.276 0.024
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM1269647 3 0.627 0.122534 0.024 0.124 0.592 0.000 0.260
#> GSM1269655 4 0.795 0.226439 0.132 0.104 0.308 0.444 0.012
#> GSM1269663 2 0.867 0.219596 0.240 0.384 0.112 0.236 0.028
#> GSM1269671 5 0.242 0.655811 0.044 0.024 0.020 0.000 0.912
#> GSM1269679 3 0.609 0.351732 0.180 0.000 0.588 0.228 0.004
#> GSM1269693 4 0.287 0.413214 0.064 0.036 0.008 0.888 0.004
#> GSM1269701 3 0.693 0.253859 0.324 0.000 0.440 0.224 0.012
#> GSM1269709 4 0.812 0.067987 0.156 0.036 0.372 0.380 0.056
#> GSM1269715 4 0.207 0.430273 0.036 0.004 0.028 0.928 0.004
#> GSM1269717 4 0.172 0.433622 0.028 0.004 0.020 0.944 0.004
#> GSM1269721 2 0.756 0.353521 0.052 0.540 0.112 0.044 0.252
#> GSM1269723 3 0.693 0.319258 0.284 0.000 0.484 0.212 0.020
#> GSM1269645 1 0.532 0.438295 0.704 0.008 0.036 0.216 0.036
#> GSM1269653 3 0.841 -0.169485 0.148 0.244 0.376 0.004 0.228
#> GSM1269661 1 0.737 0.302836 0.488 0.016 0.284 0.184 0.028
#> GSM1269669 1 0.566 0.363234 0.604 0.000 0.116 0.280 0.000
#> GSM1269677 2 0.340 0.640373 0.012 0.856 0.012 0.020 0.100
#> GSM1269685 4 0.565 0.000703 0.332 0.048 0.012 0.600 0.008
#> GSM1269691 4 0.526 0.045445 0.324 0.056 0.004 0.616 0.000
#> GSM1269699 5 0.546 0.634682 0.036 0.152 0.100 0.000 0.712
#> GSM1269707 5 0.816 0.460839 0.080 0.248 0.160 0.036 0.476
#> GSM1269651 2 0.479 0.597305 0.040 0.752 0.040 0.000 0.168
#> GSM1269659 2 0.552 0.625384 0.064 0.744 0.044 0.120 0.028
#> GSM1269667 3 0.738 0.142172 0.300 0.004 0.388 0.288 0.020
#> GSM1269675 5 0.390 0.652630 0.084 0.044 0.040 0.000 0.832
#> GSM1269683 4 0.593 0.222536 0.276 0.008 0.104 0.608 0.004
#> GSM1269689 5 0.811 0.293382 0.164 0.084 0.344 0.016 0.392
#> GSM1269697 3 0.611 0.068672 0.036 0.048 0.552 0.004 0.360
#> GSM1269705 5 0.468 0.666094 0.012 0.072 0.164 0.000 0.752
#> GSM1269713 3 0.559 0.445818 0.080 0.032 0.740 0.036 0.112
#> GSM1269719 1 0.844 0.264665 0.436 0.176 0.176 0.196 0.016
#> GSM1269725 3 0.437 0.466356 0.048 0.020 0.816 0.028 0.088
#> GSM1269727 4 0.718 0.118638 0.332 0.000 0.224 0.420 0.024
#> GSM1269649 1 0.646 0.432213 0.660 0.016 0.140 0.056 0.128
#> GSM1269657 2 0.383 0.644635 0.032 0.848 0.012 0.064 0.044
#> GSM1269665 1 0.581 0.431128 0.636 0.016 0.084 0.260 0.004
#> GSM1269673 1 0.545 0.256309 0.508 0.036 0.012 0.444 0.000
#> GSM1269681 2 0.446 0.580473 0.024 0.756 0.028 0.000 0.192
#> GSM1269687 1 0.609 0.442779 0.588 0.012 0.088 0.304 0.008
#> GSM1269695 1 0.581 0.486201 0.688 0.012 0.020 0.164 0.116
#> GSM1269703 1 0.454 0.491152 0.744 0.016 0.036 0.204 0.000
#> GSM1269711 1 0.752 0.332293 0.548 0.048 0.232 0.132 0.040
#> GSM1269646 3 0.637 0.181973 0.048 0.092 0.596 0.000 0.264
#> GSM1269654 4 0.790 0.256082 0.128 0.108 0.288 0.464 0.012
#> GSM1269662 2 0.845 0.338286 0.256 0.448 0.108 0.148 0.040
#> GSM1269670 5 0.243 0.656209 0.040 0.024 0.024 0.000 0.912
#> GSM1269678 3 0.673 0.218824 0.224 0.004 0.492 0.276 0.004
#> GSM1269692 4 0.380 0.348874 0.120 0.044 0.008 0.824 0.004
#> GSM1269700 3 0.683 0.279145 0.324 0.000 0.460 0.204 0.012
#> GSM1269708 4 0.798 0.151962 0.164 0.032 0.340 0.416 0.048
#> GSM1269714 4 0.144 0.425945 0.032 0.000 0.012 0.952 0.004
#> GSM1269716 4 0.190 0.432593 0.032 0.004 0.024 0.936 0.004
#> GSM1269720 2 0.717 0.452663 0.048 0.592 0.092 0.048 0.220
#> GSM1269722 4 0.702 -0.097502 0.180 0.004 0.400 0.400 0.016
#> GSM1269644 1 0.636 0.268497 0.488 0.064 0.032 0.412 0.004
#> GSM1269652 3 0.971 -0.181229 0.136 0.244 0.288 0.132 0.200
#> GSM1269660 1 0.755 0.280240 0.484 0.028 0.284 0.176 0.028
#> GSM1269668 1 0.657 0.169331 0.468 0.000 0.240 0.292 0.000
#> GSM1269676 2 0.340 0.640373 0.012 0.856 0.012 0.020 0.100
#> GSM1269684 4 0.498 -0.178209 0.412 0.008 0.012 0.564 0.004
#> GSM1269690 4 0.513 0.080709 0.308 0.052 0.004 0.636 0.000
#> GSM1269698 5 0.544 0.627324 0.016 0.160 0.128 0.000 0.696
#> GSM1269706 5 0.816 0.461821 0.076 0.252 0.164 0.036 0.472
#> GSM1269650 2 0.487 0.603032 0.044 0.752 0.032 0.004 0.168
#> GSM1269658 2 0.569 0.624216 0.080 0.732 0.048 0.116 0.024
#> GSM1269666 3 0.714 0.229514 0.208 0.008 0.464 0.304 0.016
#> GSM1269674 5 0.382 0.656070 0.084 0.044 0.036 0.000 0.836
#> GSM1269682 4 0.617 0.109603 0.348 0.004 0.112 0.532 0.004
#> GSM1269688 5 0.810 0.311123 0.164 0.084 0.336 0.016 0.400
#> GSM1269696 3 0.591 0.164061 0.040 0.032 0.564 0.004 0.360
#> GSM1269704 5 0.480 0.660829 0.004 0.084 0.184 0.000 0.728
#> GSM1269712 3 0.581 0.408616 0.104 0.012 0.668 0.204 0.012
#> GSM1269718 1 0.832 0.283817 0.456 0.164 0.196 0.168 0.016
#> GSM1269724 3 0.494 0.470179 0.076 0.016 0.780 0.036 0.092
#> GSM1269726 4 0.717 0.101600 0.336 0.004 0.224 0.420 0.016
#> GSM1269648 1 0.678 0.452050 0.644 0.020 0.104 0.140 0.092
#> GSM1269656 2 0.809 0.238271 0.144 0.444 0.044 0.312 0.056
#> GSM1269664 1 0.627 0.405720 0.616 0.012 0.172 0.192 0.008
#> GSM1269672 1 0.505 0.244708 0.496 0.024 0.004 0.476 0.000
#> GSM1269680 2 0.431 0.582612 0.016 0.760 0.028 0.000 0.196
#> GSM1269686 1 0.566 0.346397 0.520 0.004 0.056 0.416 0.004
#> GSM1269694 1 0.593 0.485842 0.680 0.012 0.024 0.168 0.116
#> GSM1269702 1 0.576 0.252661 0.500 0.036 0.020 0.440 0.004
#> GSM1269710 1 0.595 0.492848 0.696 0.020 0.100 0.152 0.032
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM1269647 5 0.532 0.3702 0.016 0.108 0.116 0.008 0.716 0.036
#> GSM1269655 3 0.825 0.1810 0.152 0.008 0.444 0.140 0.160 0.096
#> GSM1269663 6 0.897 0.1315 0.244 0.048 0.112 0.240 0.060 0.296
#> GSM1269671 2 0.358 0.6526 0.032 0.848 0.004 0.028 0.060 0.028
#> GSM1269679 3 0.477 0.4711 0.052 0.000 0.712 0.048 0.188 0.000
#> GSM1269693 4 0.679 0.7185 0.172 0.008 0.232 0.528 0.012 0.048
#> GSM1269701 3 0.452 0.5319 0.116 0.012 0.752 0.012 0.108 0.000
#> GSM1269709 5 0.842 0.0616 0.236 0.024 0.248 0.156 0.316 0.020
#> GSM1269715 4 0.584 0.8194 0.172 0.000 0.308 0.512 0.008 0.000
#> GSM1269717 4 0.578 0.8248 0.180 0.000 0.304 0.512 0.004 0.000
#> GSM1269721 6 0.753 0.3566 0.028 0.136 0.020 0.108 0.192 0.516
#> GSM1269723 3 0.488 0.4948 0.076 0.020 0.736 0.028 0.140 0.000
#> GSM1269645 1 0.662 0.3395 0.552 0.036 0.240 0.144 0.020 0.008
#> GSM1269653 5 0.790 0.2937 0.080 0.084 0.124 0.076 0.544 0.092
#> GSM1269661 3 0.718 0.1698 0.340 0.008 0.408 0.076 0.164 0.004
#> GSM1269669 3 0.614 0.0400 0.396 0.008 0.468 0.092 0.036 0.000
#> GSM1269677 6 0.348 0.5731 0.016 0.072 0.000 0.020 0.048 0.844
#> GSM1269685 1 0.664 0.0361 0.472 0.004 0.080 0.368 0.052 0.024
#> GSM1269691 1 0.676 -0.0205 0.456 0.008 0.088 0.376 0.020 0.052
#> GSM1269699 2 0.710 0.5226 0.032 0.500 0.004 0.048 0.240 0.176
#> GSM1269707 5 0.853 -0.2189 0.084 0.264 0.016 0.072 0.292 0.272
#> GSM1269651 6 0.626 0.5036 0.012 0.144 0.008 0.124 0.080 0.632
#> GSM1269659 6 0.560 0.5733 0.052 0.016 0.012 0.180 0.056 0.684
#> GSM1269667 3 0.502 0.5376 0.116 0.008 0.716 0.020 0.136 0.004
#> GSM1269675 2 0.392 0.6303 0.008 0.812 0.024 0.024 0.116 0.016
#> GSM1269683 3 0.559 0.0971 0.264 0.004 0.560 0.172 0.000 0.000
#> GSM1269689 5 0.804 0.0548 0.096 0.308 0.112 0.064 0.404 0.016
#> GSM1269697 5 0.614 0.3277 0.016 0.212 0.108 0.024 0.620 0.020
#> GSM1269705 2 0.609 0.5601 0.016 0.564 0.012 0.032 0.312 0.064
#> GSM1269713 5 0.522 0.3177 0.024 0.024 0.352 0.012 0.584 0.004
#> GSM1269719 1 0.831 0.2896 0.484 0.040 0.144 0.124 0.100 0.108
#> GSM1269725 5 0.512 0.3014 0.020 0.020 0.332 0.012 0.608 0.008
#> GSM1269727 3 0.547 0.4339 0.196 0.020 0.676 0.076 0.028 0.004
#> GSM1269649 1 0.716 0.1885 0.496 0.072 0.252 0.024 0.152 0.004
#> GSM1269657 6 0.447 0.5867 0.048 0.032 0.000 0.076 0.056 0.788
#> GSM1269665 1 0.616 0.3634 0.580 0.000 0.228 0.148 0.020 0.024
#> GSM1269673 1 0.517 0.4232 0.684 0.004 0.072 0.208 0.008 0.024
#> GSM1269681 6 0.538 0.4798 0.004 0.184 0.004 0.084 0.044 0.680
#> GSM1269687 1 0.514 0.4758 0.688 0.004 0.192 0.088 0.024 0.004
#> GSM1269695 1 0.483 0.5033 0.756 0.120 0.056 0.020 0.044 0.004
#> GSM1269703 1 0.564 0.4101 0.624 0.000 0.252 0.080 0.016 0.028
#> GSM1269711 1 0.721 0.3159 0.548 0.036 0.164 0.076 0.164 0.012
#> GSM1269646 5 0.557 0.3926 0.024 0.096 0.160 0.016 0.688 0.016
#> GSM1269654 3 0.797 0.1939 0.144 0.008 0.488 0.132 0.132 0.096
#> GSM1269662 6 0.868 0.3024 0.212 0.052 0.076 0.208 0.068 0.384
#> GSM1269670 2 0.358 0.6526 0.032 0.848 0.004 0.028 0.060 0.028
#> GSM1269678 3 0.521 0.5093 0.092 0.000 0.692 0.060 0.156 0.000
#> GSM1269692 4 0.646 0.6215 0.220 0.004 0.168 0.556 0.008 0.044
#> GSM1269700 3 0.474 0.5221 0.124 0.016 0.736 0.012 0.112 0.000
#> GSM1269708 5 0.836 0.0317 0.248 0.020 0.236 0.176 0.304 0.016
#> GSM1269714 4 0.599 0.7959 0.188 0.004 0.284 0.516 0.008 0.000
#> GSM1269716 4 0.589 0.8207 0.180 0.000 0.308 0.504 0.008 0.000
#> GSM1269720 6 0.702 0.4381 0.032 0.124 0.016 0.096 0.148 0.584
#> GSM1269722 3 0.503 0.4626 0.048 0.024 0.756 0.092 0.068 0.012
#> GSM1269644 1 0.605 0.4060 0.624 0.008 0.076 0.224 0.016 0.052
#> GSM1269652 5 0.808 0.2479 0.192 0.056 0.024 0.136 0.472 0.120
#> GSM1269660 3 0.750 0.2093 0.308 0.008 0.396 0.096 0.184 0.008
#> GSM1269668 3 0.570 0.3239 0.272 0.008 0.596 0.100 0.024 0.000
#> GSM1269676 6 0.348 0.5731 0.016 0.072 0.000 0.020 0.048 0.844
#> GSM1269684 1 0.609 0.2703 0.560 0.000 0.132 0.272 0.020 0.016
#> GSM1269690 1 0.690 -0.1041 0.432 0.008 0.104 0.384 0.020 0.052
#> GSM1269698 2 0.704 0.5091 0.028 0.484 0.000 0.048 0.244 0.196
#> GSM1269706 5 0.856 -0.2201 0.088 0.268 0.016 0.072 0.288 0.268
#> GSM1269650 6 0.637 0.5013 0.012 0.140 0.012 0.124 0.084 0.628
#> GSM1269658 6 0.588 0.5668 0.080 0.020 0.020 0.176 0.036 0.668
#> GSM1269666 3 0.453 0.5197 0.052 0.004 0.756 0.036 0.148 0.004
#> GSM1269674 2 0.405 0.6350 0.012 0.812 0.024 0.024 0.104 0.024
#> GSM1269682 3 0.536 0.2751 0.232 0.004 0.616 0.144 0.004 0.000
#> GSM1269688 5 0.822 0.0411 0.100 0.300 0.112 0.060 0.400 0.028
#> GSM1269696 5 0.664 0.3193 0.016 0.228 0.140 0.032 0.564 0.020
#> GSM1269704 2 0.608 0.5008 0.008 0.516 0.008 0.028 0.360 0.080
#> GSM1269712 3 0.607 0.0576 0.056 0.008 0.480 0.060 0.396 0.000
#> GSM1269718 1 0.813 0.2827 0.496 0.032 0.156 0.120 0.100 0.096
#> GSM1269724 5 0.534 0.2835 0.028 0.036 0.332 0.008 0.592 0.004
#> GSM1269726 3 0.583 0.3934 0.192 0.028 0.648 0.104 0.024 0.004
#> GSM1269648 1 0.574 0.4486 0.676 0.072 0.056 0.036 0.160 0.000
#> GSM1269656 6 0.778 0.1294 0.300 0.036 0.008 0.176 0.080 0.400
#> GSM1269664 1 0.643 0.1739 0.512 0.000 0.332 0.076 0.060 0.020
#> GSM1269672 1 0.571 0.3830 0.632 0.000 0.100 0.224 0.024 0.020
#> GSM1269680 6 0.499 0.4928 0.004 0.180 0.000 0.076 0.036 0.704
#> GSM1269686 1 0.553 0.4571 0.660 0.004 0.164 0.144 0.016 0.012
#> GSM1269694 1 0.490 0.5052 0.752 0.116 0.064 0.020 0.044 0.004
#> GSM1269702 1 0.528 0.4524 0.708 0.000 0.040 0.160 0.032 0.060
#> GSM1269710 1 0.590 0.4541 0.684 0.052 0.140 0.028 0.080 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 agent(p) disease.state(p) gender(p) individual(p) k
#> MAD:kmeans 83 0.890 0.445 0.530170 2.02e-04 2
#> MAD:kmeans 60 0.770 1.000 0.249796 1.35e-03 3
#> MAD:kmeans 39 0.517 0.354 0.000129 2.76e-04 4
#> MAD:kmeans 17 1.000 0.492 0.433012 3.01e-02 5
#> MAD:kmeans 28 0.987 0.052 0.028217 1.31e-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["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 51941 rows and 84 columns.
#> Top rows (1000, 2000, 3000, 4000, 5000) are extracted by 'MAD' method.
#> Subgroups are detected by 'skmeans' method.
#> Performed in total 1250 partitions by row resampling.
#> Best k for subgroups seems to be 2.
#>
#> Following methods can be applied to this 'ConsensusPartition' object:
#> [1] "cola_report" "collect_classes" "collect_plots"
#> [4] "collect_stats" "colnames" "compare_signatures"
#> [7] "consensus_heatmap" "dimension_reduction" "functional_enrichment"
#> [10] "get_anno_col" "get_anno" "get_classes"
#> [13] "get_consensus" "get_matrix" "get_membership"
#> [16] "get_param" "get_signatures" "get_stats"
#> [19] "is_best_k" "is_stable_k" "membership_heatmap"
#> [22] "ncol" "nrow" "plot_ecdf"
#> [25] "rownames" "select_partition_number" "show"
#> [28] "suggest_best_k" "test_to_known_factors"
collect_plots() function collects all the plots made from res for all k (number of partitions)
into one single page to provide an easy and fast comparison between different k.
collect_plots(res)
The plots are:
k and the heatmap of
predicted classes for each k.k.k.k.All the plots in panels can be made by individual functions and they are plotted later in this section.
select_partition_number() produces several plots showing different
statistics for choosing “optimized” k. There are following statistics:
k;k, the area increased is defined as \(A_k - A_{k-1}\).The detailed explanations of these statistics can be found in the cola vignette.
Generally speaking, lower PAC score, higher mean silhouette score or higher
concordance corresponds to better partition. Rand index and Jaccard index
measure how similar the current partition is compared to partition with k-1.
If they are too similar, we won't accept k is better than k-1.
select_partition_number(res)
The numeric values for all these statistics can be obtained by get_stats().
get_stats(res)
#> k 1-PAC mean_silhouette concordance area_increased Rand Jaccard
#> 2 2 0.00000 0.6370 0.713 0.5039 0.504 0.504
#> 3 3 0.00811 0.2219 0.515 0.3314 0.813 0.649
#> 4 4 0.04090 0.0857 0.376 0.1241 0.761 0.448
#> 5 5 0.11555 0.1228 0.369 0.0656 0.817 0.424
#> 6 6 0.22038 0.0950 0.322 0.0419 0.851 0.426
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
#> GSM1269647 2 0.760 0.7367 0.220 0.780
#> GSM1269655 1 0.961 0.5539 0.616 0.384
#> GSM1269663 2 0.994 0.2001 0.456 0.544
#> GSM1269671 2 0.605 0.7441 0.148 0.852
#> GSM1269679 1 0.866 0.6822 0.712 0.288
#> GSM1269693 1 0.781 0.7015 0.768 0.232
#> GSM1269701 1 0.827 0.7070 0.740 0.260
#> GSM1269709 1 0.999 0.0176 0.520 0.480
#> GSM1269715 1 0.574 0.7110 0.864 0.136
#> GSM1269717 1 0.615 0.7164 0.848 0.152
#> GSM1269721 2 0.680 0.7472 0.180 0.820
#> GSM1269723 1 0.990 0.4320 0.560 0.440
#> GSM1269645 1 0.855 0.7083 0.720 0.280
#> GSM1269653 2 0.881 0.6787 0.300 0.700
#> GSM1269661 1 0.966 0.5597 0.608 0.392
#> GSM1269669 1 0.443 0.7031 0.908 0.092
#> GSM1269677 2 0.680 0.7241 0.180 0.820
#> GSM1269685 1 0.855 0.6826 0.720 0.280
#> GSM1269691 1 0.839 0.6886 0.732 0.268
#> GSM1269699 2 0.574 0.7435 0.136 0.864
#> GSM1269707 2 0.730 0.7535 0.204 0.796
#> GSM1269651 2 0.730 0.7395 0.204 0.796
#> GSM1269659 2 0.917 0.6192 0.332 0.668
#> GSM1269667 1 0.861 0.6862 0.716 0.284
#> GSM1269675 2 0.753 0.7452 0.216 0.784
#> GSM1269683 1 0.753 0.7277 0.784 0.216
#> GSM1269689 2 0.814 0.7114 0.252 0.748
#> GSM1269697 2 0.839 0.7026 0.268 0.732
#> GSM1269705 2 0.745 0.7517 0.212 0.788
#> GSM1269713 2 0.925 0.5391 0.340 0.660
#> GSM1269719 1 0.995 0.2936 0.540 0.460
#> GSM1269725 2 0.861 0.6825 0.284 0.716
#> GSM1269727 1 0.827 0.7158 0.740 0.260
#> GSM1269649 1 0.975 0.5487 0.592 0.408
#> GSM1269657 2 0.808 0.7239 0.248 0.752
#> GSM1269665 1 0.821 0.7200 0.744 0.256
#> GSM1269673 1 0.760 0.7260 0.780 0.220
#> GSM1269681 2 0.482 0.7442 0.104 0.896
#> GSM1269687 1 0.821 0.7088 0.744 0.256
#> GSM1269695 1 0.939 0.6397 0.644 0.356
#> GSM1269703 1 0.866 0.7040 0.712 0.288
#> GSM1269711 1 0.987 0.3785 0.568 0.432
#> GSM1269646 2 0.775 0.7176 0.228 0.772
#> GSM1269654 1 0.895 0.6737 0.688 0.312
#> GSM1269662 2 0.992 0.2398 0.448 0.552
#> GSM1269670 2 0.634 0.7412 0.160 0.840
#> GSM1269678 1 0.795 0.7211 0.760 0.240
#> GSM1269692 1 0.781 0.7116 0.768 0.232
#> GSM1269700 1 0.855 0.6923 0.720 0.280
#> GSM1269708 1 0.981 0.3826 0.580 0.420
#> GSM1269714 1 0.552 0.7192 0.872 0.128
#> GSM1269716 1 0.482 0.7115 0.896 0.104
#> GSM1269720 2 0.730 0.7471 0.204 0.796
#> GSM1269722 1 0.952 0.5304 0.628 0.372
#> GSM1269644 1 0.943 0.6002 0.640 0.360
#> GSM1269652 2 0.900 0.6686 0.316 0.684
#> GSM1269660 1 0.997 0.3249 0.532 0.468
#> GSM1269668 1 0.494 0.7018 0.892 0.108
#> GSM1269676 2 0.653 0.7330 0.168 0.832
#> GSM1269684 1 0.653 0.7243 0.832 0.168
#> GSM1269690 1 0.767 0.7108 0.776 0.224
#> GSM1269698 2 0.518 0.7470 0.116 0.884
#> GSM1269706 2 0.861 0.6908 0.284 0.716
#> GSM1269650 2 0.671 0.7461 0.176 0.824
#> GSM1269658 2 0.881 0.6612 0.300 0.700
#> GSM1269666 1 0.767 0.7199 0.776 0.224
#> GSM1269674 2 0.839 0.7321 0.268 0.732
#> GSM1269682 1 0.775 0.7261 0.772 0.228
#> GSM1269688 2 0.808 0.7068 0.248 0.752
#> GSM1269696 2 0.808 0.6902 0.248 0.752
#> GSM1269704 2 0.634 0.7549 0.160 0.840
#> GSM1269712 1 0.980 0.4498 0.584 0.416
#> GSM1269718 2 0.990 0.1368 0.440 0.560
#> GSM1269724 2 1.000 0.1744 0.496 0.504
#> GSM1269726 1 0.821 0.7204 0.744 0.256
#> GSM1269648 1 0.969 0.4987 0.604 0.396
#> GSM1269656 2 0.909 0.6151 0.324 0.676
#> GSM1269664 1 0.767 0.7238 0.776 0.224
#> GSM1269672 1 0.760 0.7243 0.780 0.220
#> GSM1269680 2 0.644 0.7441 0.164 0.836
#> GSM1269686 1 0.563 0.7239 0.868 0.132
#> GSM1269694 1 0.925 0.6472 0.660 0.340
#> GSM1269702 1 0.952 0.5893 0.628 0.372
#> GSM1269710 1 0.955 0.5326 0.624 0.376
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM1269647 2 0.868 0.39585 0.120 0.540 0.340
#> GSM1269655 1 0.996 -0.07851 0.368 0.292 0.340
#> GSM1269663 2 0.997 -0.21895 0.340 0.364 0.296
#> GSM1269671 2 0.706 0.52159 0.052 0.672 0.276
#> GSM1269679 3 0.907 -0.04099 0.416 0.136 0.448
#> GSM1269693 1 0.852 0.28345 0.608 0.164 0.228
#> GSM1269701 3 0.896 0.04588 0.376 0.132 0.492
#> GSM1269709 1 0.954 -0.03106 0.420 0.192 0.388
#> GSM1269715 1 0.637 0.30774 0.756 0.068 0.176
#> GSM1269717 1 0.626 0.30693 0.752 0.052 0.196
#> GSM1269721 2 0.753 0.50213 0.084 0.664 0.252
#> GSM1269723 3 0.951 0.14722 0.296 0.220 0.484
#> GSM1269645 1 0.939 0.11105 0.504 0.212 0.284
#> GSM1269653 2 0.894 0.35528 0.160 0.548 0.292
#> GSM1269661 1 0.990 -0.13118 0.372 0.264 0.364
#> GSM1269669 1 0.740 0.19559 0.612 0.048 0.340
#> GSM1269677 2 0.594 0.52579 0.120 0.792 0.088
#> GSM1269685 1 0.858 0.24438 0.596 0.152 0.252
#> GSM1269691 1 0.905 0.23667 0.556 0.224 0.220
#> GSM1269699 2 0.663 0.52866 0.056 0.724 0.220
#> GSM1269707 2 0.819 0.46096 0.128 0.628 0.244
#> GSM1269651 2 0.699 0.52215 0.108 0.728 0.164
#> GSM1269659 2 0.889 0.34201 0.284 0.556 0.160
#> GSM1269667 1 0.911 0.00955 0.448 0.140 0.412
#> GSM1269675 2 0.889 0.39071 0.132 0.516 0.352
#> GSM1269683 1 0.898 0.21337 0.524 0.148 0.328
#> GSM1269689 2 0.867 0.31636 0.108 0.508 0.384
#> GSM1269697 2 0.904 0.30954 0.140 0.488 0.372
#> GSM1269705 2 0.795 0.48742 0.084 0.608 0.308
#> GSM1269713 3 0.928 0.21296 0.180 0.320 0.500
#> GSM1269719 2 0.997 -0.17690 0.300 0.360 0.340
#> GSM1269725 3 0.876 0.03626 0.120 0.360 0.520
#> GSM1269727 1 0.906 0.12960 0.480 0.140 0.380
#> GSM1269649 3 0.958 0.14151 0.340 0.208 0.452
#> GSM1269657 2 0.875 0.37664 0.224 0.588 0.188
#> GSM1269665 1 0.928 0.15039 0.512 0.192 0.296
#> GSM1269673 1 0.835 0.27209 0.604 0.124 0.272
#> GSM1269681 2 0.524 0.54075 0.028 0.804 0.168
#> GSM1269687 1 0.918 0.10276 0.472 0.152 0.376
#> GSM1269695 1 0.983 -0.02169 0.400 0.248 0.352
#> GSM1269703 1 0.961 0.10376 0.428 0.204 0.368
#> GSM1269711 3 0.962 0.14358 0.324 0.220 0.456
#> GSM1269646 2 0.885 0.30365 0.128 0.516 0.356
#> GSM1269654 1 0.990 -0.01505 0.404 0.300 0.296
#> GSM1269662 1 0.999 -0.03754 0.348 0.340 0.312
#> GSM1269670 2 0.698 0.51074 0.064 0.700 0.236
#> GSM1269678 1 0.888 0.09529 0.508 0.128 0.364
#> GSM1269692 1 0.832 0.29034 0.628 0.212 0.160
#> GSM1269700 3 0.891 0.11712 0.344 0.136 0.520
#> GSM1269708 1 0.946 -0.00690 0.412 0.180 0.408
#> GSM1269714 1 0.648 0.29391 0.728 0.048 0.224
#> GSM1269716 1 0.626 0.29405 0.752 0.052 0.196
#> GSM1269720 2 0.781 0.50813 0.144 0.672 0.184
#> GSM1269722 1 0.944 0.05311 0.444 0.180 0.376
#> GSM1269644 1 0.990 0.04859 0.376 0.264 0.360
#> GSM1269652 2 0.975 0.15099 0.232 0.420 0.348
#> GSM1269660 1 0.988 -0.12347 0.376 0.260 0.364
#> GSM1269668 1 0.746 0.15944 0.584 0.044 0.372
#> GSM1269676 2 0.685 0.52102 0.136 0.740 0.124
#> GSM1269684 1 0.857 0.27105 0.592 0.144 0.264
#> GSM1269690 1 0.857 0.26887 0.604 0.168 0.228
#> GSM1269698 2 0.650 0.52731 0.056 0.736 0.208
#> GSM1269706 2 0.948 0.32158 0.216 0.488 0.296
#> GSM1269650 2 0.657 0.51704 0.088 0.752 0.160
#> GSM1269658 2 0.900 0.34034 0.240 0.560 0.200
#> GSM1269666 1 0.849 0.12301 0.548 0.104 0.348
#> GSM1269674 2 0.824 0.46411 0.160 0.636 0.204
#> GSM1269682 1 0.897 0.19941 0.528 0.148 0.324
#> GSM1269688 2 0.871 0.33853 0.112 0.508 0.380
#> GSM1269696 2 0.913 0.18257 0.144 0.460 0.396
#> GSM1269704 2 0.745 0.48745 0.068 0.652 0.280
#> GSM1269712 3 0.944 0.12517 0.324 0.196 0.480
#> GSM1269718 3 0.985 0.05275 0.264 0.324 0.412
#> GSM1269724 3 0.976 0.24759 0.268 0.288 0.444
#> GSM1269726 1 0.885 0.16159 0.484 0.120 0.396
#> GSM1269648 3 0.961 0.12417 0.332 0.216 0.452
#> GSM1269656 2 0.969 0.13432 0.324 0.444 0.232
#> GSM1269664 1 0.903 0.09246 0.476 0.136 0.388
#> GSM1269672 1 0.814 0.27313 0.616 0.108 0.276
#> GSM1269680 2 0.583 0.54381 0.076 0.796 0.128
#> GSM1269686 1 0.857 0.22446 0.556 0.116 0.328
#> GSM1269694 1 0.983 0.01131 0.408 0.252 0.340
#> GSM1269702 1 0.974 0.09484 0.448 0.268 0.284
#> GSM1269710 3 0.932 0.07986 0.368 0.168 0.464
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM1269647 3 0.892 0.13211 0.068 0.200 0.404 0.328
#> GSM1269655 1 0.991 -0.02139 0.300 0.268 0.240 0.192
#> GSM1269663 4 0.984 -0.05495 0.256 0.224 0.188 0.332
#> GSM1269671 3 0.821 -0.02282 0.076 0.088 0.432 0.404
#> GSM1269679 1 0.883 0.01555 0.392 0.364 0.176 0.068
#> GSM1269693 1 0.761 0.18796 0.624 0.176 0.076 0.124
#> GSM1269701 2 0.871 0.06905 0.284 0.468 0.176 0.072
#> GSM1269709 1 0.997 -0.02501 0.276 0.228 0.268 0.228
#> GSM1269715 1 0.639 0.20398 0.704 0.176 0.076 0.044
#> GSM1269717 1 0.716 0.21566 0.668 0.144 0.100 0.088
#> GSM1269721 4 0.844 0.19565 0.152 0.088 0.224 0.536
#> GSM1269723 2 0.940 0.10953 0.244 0.384 0.264 0.108
#> GSM1269645 2 0.922 0.12870 0.264 0.432 0.188 0.116
#> GSM1269653 4 0.945 -0.10808 0.104 0.244 0.324 0.328
#> GSM1269661 2 0.980 0.08608 0.280 0.332 0.204 0.184
#> GSM1269669 1 0.786 0.06401 0.440 0.416 0.104 0.040
#> GSM1269677 4 0.555 0.32094 0.068 0.052 0.104 0.776
#> GSM1269685 1 0.917 0.17081 0.460 0.212 0.200 0.128
#> GSM1269691 1 0.918 0.14207 0.468 0.172 0.156 0.204
#> GSM1269699 4 0.733 0.03253 0.020 0.092 0.412 0.476
#> GSM1269707 4 0.914 0.02620 0.132 0.140 0.300 0.428
#> GSM1269651 4 0.730 0.26892 0.068 0.112 0.172 0.648
#> GSM1269659 4 0.902 0.17248 0.224 0.100 0.212 0.464
#> GSM1269667 2 0.867 0.07693 0.328 0.440 0.168 0.064
#> GSM1269675 4 0.889 -0.03796 0.084 0.156 0.368 0.392
#> GSM1269683 1 0.854 0.09846 0.484 0.304 0.084 0.128
#> GSM1269689 3 0.936 0.13160 0.112 0.200 0.380 0.308
#> GSM1269697 3 0.877 0.10596 0.100 0.136 0.472 0.292
#> GSM1269705 3 0.868 -0.02130 0.072 0.144 0.392 0.392
#> GSM1269713 3 0.943 0.12321 0.116 0.332 0.348 0.204
#> GSM1269719 4 0.977 -0.11575 0.192 0.308 0.184 0.316
#> GSM1269725 3 0.949 0.19764 0.132 0.272 0.388 0.208
#> GSM1269727 2 0.848 -0.00112 0.388 0.416 0.132 0.064
#> GSM1269649 2 0.898 0.15052 0.152 0.424 0.324 0.100
#> GSM1269657 4 0.869 0.16478 0.128 0.112 0.260 0.500
#> GSM1269665 1 0.965 -0.03101 0.340 0.300 0.216 0.144
#> GSM1269673 1 0.916 0.08951 0.436 0.256 0.204 0.104
#> GSM1269681 4 0.631 0.25008 0.024 0.064 0.240 0.672
#> GSM1269687 2 0.957 0.00620 0.320 0.340 0.204 0.136
#> GSM1269695 3 0.986 -0.12912 0.208 0.288 0.308 0.196
#> GSM1269703 2 0.966 0.07271 0.256 0.376 0.192 0.176
#> GSM1269711 3 0.955 -0.12503 0.224 0.304 0.348 0.124
#> GSM1269646 3 0.887 0.17697 0.068 0.236 0.448 0.248
#> GSM1269654 1 0.988 0.01268 0.308 0.256 0.180 0.256
#> GSM1269662 4 0.989 -0.07446 0.236 0.236 0.204 0.324
#> GSM1269670 3 0.747 -0.05008 0.024 0.096 0.464 0.416
#> GSM1269678 1 0.897 -0.00837 0.376 0.364 0.184 0.076
#> GSM1269692 1 0.821 0.18688 0.568 0.168 0.084 0.180
#> GSM1269700 2 0.926 0.12947 0.260 0.404 0.240 0.096
#> GSM1269708 1 0.952 0.07664 0.388 0.204 0.268 0.140
#> GSM1269714 1 0.717 0.21660 0.648 0.196 0.096 0.060
#> GSM1269716 1 0.656 0.20481 0.696 0.176 0.072 0.056
#> GSM1269720 4 0.769 0.23877 0.088 0.116 0.176 0.620
#> GSM1269722 1 0.949 0.00475 0.388 0.288 0.156 0.168
#> GSM1269644 2 0.982 0.01932 0.260 0.328 0.172 0.240
#> GSM1269652 3 0.941 0.07051 0.140 0.168 0.384 0.308
#> GSM1269660 2 0.985 0.11253 0.208 0.332 0.256 0.204
#> GSM1269668 1 0.802 0.06436 0.472 0.372 0.104 0.052
#> GSM1269676 4 0.493 0.31146 0.064 0.032 0.096 0.808
#> GSM1269684 1 0.909 0.12667 0.444 0.284 0.124 0.148
#> GSM1269690 1 0.888 0.18065 0.508 0.168 0.144 0.180
#> GSM1269698 4 0.677 0.11393 0.016 0.064 0.368 0.552
#> GSM1269706 3 0.890 -0.01612 0.140 0.096 0.404 0.360
#> GSM1269650 4 0.757 0.23933 0.056 0.128 0.204 0.612
#> GSM1269658 4 0.807 0.25931 0.184 0.120 0.108 0.588
#> GSM1269666 1 0.894 -0.00736 0.396 0.356 0.164 0.084
#> GSM1269674 4 0.880 0.07220 0.108 0.144 0.260 0.488
#> GSM1269682 1 0.836 0.03754 0.456 0.364 0.084 0.096
#> GSM1269688 3 0.909 0.13153 0.072 0.236 0.388 0.304
#> GSM1269696 3 0.904 0.17745 0.080 0.240 0.432 0.248
#> GSM1269704 4 0.798 0.00631 0.052 0.096 0.420 0.432
#> GSM1269712 2 0.967 0.07455 0.276 0.348 0.232 0.144
#> GSM1269718 2 0.992 0.01818 0.196 0.300 0.244 0.260
#> GSM1269724 3 0.960 0.02093 0.188 0.312 0.348 0.152
#> GSM1269726 2 0.872 0.02793 0.376 0.384 0.184 0.056
#> GSM1269648 3 0.938 -0.07862 0.172 0.344 0.356 0.128
#> GSM1269656 4 0.892 0.11415 0.240 0.104 0.176 0.480
#> GSM1269664 2 0.931 0.08731 0.240 0.440 0.156 0.164
#> GSM1269672 1 0.890 0.09637 0.476 0.256 0.164 0.104
#> GSM1269680 4 0.580 0.28274 0.032 0.080 0.140 0.748
#> GSM1269686 1 0.899 0.04565 0.408 0.348 0.116 0.128
#> GSM1269694 2 0.972 0.10918 0.240 0.316 0.300 0.144
#> GSM1269702 1 0.996 0.01635 0.292 0.232 0.224 0.252
#> GSM1269710 2 0.964 0.04931 0.292 0.324 0.256 0.128
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM1269647 5 0.853 0.21626 0.124 0.276 0.140 0.036 0.424
#> GSM1269655 4 0.987 -0.01611 0.152 0.176 0.204 0.276 0.192
#> GSM1269663 2 0.978 -0.06125 0.200 0.284 0.144 0.228 0.144
#> GSM1269671 5 0.768 0.26199 0.140 0.284 0.048 0.032 0.496
#> GSM1269679 3 0.874 0.12065 0.092 0.056 0.388 0.300 0.164
#> GSM1269693 4 0.811 0.21047 0.176 0.076 0.132 0.536 0.080
#> GSM1269701 3 0.782 0.20243 0.132 0.036 0.556 0.128 0.148
#> GSM1269709 4 0.953 0.00270 0.272 0.084 0.168 0.288 0.188
#> GSM1269715 4 0.588 0.21666 0.088 0.024 0.152 0.704 0.032
#> GSM1269717 4 0.596 0.23335 0.080 0.044 0.132 0.712 0.032
#> GSM1269721 2 0.891 0.11040 0.148 0.452 0.096 0.124 0.180
#> GSM1269723 3 0.920 0.16898 0.104 0.108 0.396 0.196 0.196
#> GSM1269645 3 0.968 0.01747 0.248 0.128 0.280 0.220 0.124
#> GSM1269653 5 0.960 0.16249 0.172 0.256 0.216 0.084 0.272
#> GSM1269661 4 0.963 -0.05854 0.152 0.096 0.252 0.268 0.232
#> GSM1269669 4 0.856 0.02093 0.212 0.032 0.284 0.384 0.088
#> GSM1269677 2 0.607 0.35510 0.144 0.700 0.024 0.060 0.072
#> GSM1269685 4 0.817 0.09803 0.380 0.072 0.080 0.396 0.072
#> GSM1269691 4 0.862 0.11021 0.304 0.140 0.120 0.396 0.040
#> GSM1269699 5 0.750 0.20147 0.112 0.348 0.040 0.032 0.468
#> GSM1269707 2 0.854 -0.06737 0.160 0.400 0.060 0.068 0.312
#> GSM1269651 2 0.742 0.24473 0.088 0.608 0.080 0.072 0.152
#> GSM1269659 2 0.806 0.26598 0.196 0.504 0.052 0.192 0.056
#> GSM1269667 3 0.877 0.17415 0.116 0.040 0.404 0.212 0.228
#> GSM1269675 5 0.847 0.22605 0.184 0.260 0.080 0.044 0.432
#> GSM1269683 4 0.921 0.03008 0.132 0.116 0.264 0.376 0.112
#> GSM1269689 5 0.954 0.14861 0.180 0.232 0.180 0.088 0.320
#> GSM1269697 5 0.910 0.17460 0.176 0.196 0.156 0.068 0.404
#> GSM1269705 5 0.838 0.23027 0.092 0.300 0.096 0.068 0.444
#> GSM1269713 3 0.900 -0.02128 0.168 0.124 0.356 0.056 0.296
#> GSM1269719 1 0.983 0.08214 0.256 0.252 0.164 0.140 0.188
#> GSM1269725 5 0.901 0.10787 0.096 0.160 0.304 0.080 0.360
#> GSM1269727 3 0.913 0.08589 0.160 0.056 0.344 0.284 0.156
#> GSM1269649 3 0.913 -0.03346 0.240 0.040 0.296 0.148 0.276
#> GSM1269657 2 0.744 0.31925 0.212 0.576 0.040 0.080 0.092
#> GSM1269665 4 0.956 0.00669 0.208 0.108 0.268 0.292 0.124
#> GSM1269673 4 0.922 0.05281 0.292 0.120 0.148 0.348 0.092
#> GSM1269681 2 0.610 0.20552 0.052 0.684 0.036 0.044 0.184
#> GSM1269687 1 0.935 0.01013 0.308 0.096 0.300 0.184 0.112
#> GSM1269695 1 0.961 0.15879 0.332 0.148 0.136 0.156 0.228
#> GSM1269703 3 0.939 0.04255 0.240 0.144 0.328 0.212 0.076
#> GSM1269711 1 0.941 0.07509 0.352 0.096 0.204 0.132 0.216
#> GSM1269646 5 0.838 0.17400 0.088 0.252 0.268 0.016 0.376
#> GSM1269654 4 0.955 -0.00627 0.124 0.196 0.264 0.304 0.112
#> GSM1269662 2 0.965 -0.02312 0.220 0.316 0.200 0.144 0.120
#> GSM1269670 5 0.773 0.23392 0.160 0.288 0.036 0.036 0.480
#> GSM1269678 3 0.911 0.05578 0.156 0.068 0.368 0.272 0.136
#> GSM1269692 4 0.810 0.19351 0.232 0.160 0.100 0.484 0.024
#> GSM1269700 3 0.858 0.20457 0.124 0.068 0.480 0.132 0.196
#> GSM1269708 4 0.899 0.05232 0.316 0.044 0.180 0.324 0.136
#> GSM1269714 4 0.672 0.21279 0.120 0.044 0.120 0.660 0.056
#> GSM1269716 4 0.584 0.23417 0.104 0.024 0.108 0.720 0.044
#> GSM1269720 2 0.786 0.22029 0.148 0.560 0.064 0.076 0.152
#> GSM1269722 3 0.955 0.07734 0.156 0.120 0.332 0.244 0.148
#> GSM1269644 1 0.962 0.00702 0.288 0.212 0.148 0.252 0.100
#> GSM1269652 1 0.958 -0.04297 0.304 0.236 0.104 0.136 0.220
#> GSM1269660 3 0.986 0.05619 0.180 0.164 0.288 0.192 0.176
#> GSM1269668 3 0.784 -0.00126 0.116 0.012 0.392 0.384 0.096
#> GSM1269676 2 0.597 0.32360 0.128 0.708 0.020 0.056 0.088
#> GSM1269684 4 0.910 0.05603 0.288 0.096 0.204 0.340 0.072
#> GSM1269690 4 0.799 0.21451 0.264 0.100 0.072 0.504 0.060
#> GSM1269698 5 0.699 0.11113 0.056 0.424 0.056 0.020 0.444
#> GSM1269706 5 0.908 0.12086 0.240 0.244 0.052 0.120 0.344
#> GSM1269650 2 0.712 0.26305 0.124 0.636 0.076 0.060 0.104
#> GSM1269658 2 0.774 0.32289 0.184 0.568 0.088 0.100 0.060
#> GSM1269666 3 0.871 0.10597 0.092 0.056 0.384 0.312 0.156
#> GSM1269674 5 0.926 0.13759 0.112 0.292 0.128 0.120 0.348
#> GSM1269682 4 0.887 0.04712 0.208 0.068 0.304 0.348 0.072
#> GSM1269688 5 0.935 0.18056 0.172 0.212 0.172 0.080 0.364
#> GSM1269696 5 0.894 0.13690 0.100 0.208 0.248 0.060 0.384
#> GSM1269704 5 0.851 0.24390 0.128 0.280 0.116 0.044 0.432
#> GSM1269712 3 0.932 0.12414 0.140 0.068 0.332 0.228 0.232
#> GSM1269718 1 0.983 0.08832 0.264 0.232 0.200 0.132 0.172
#> GSM1269724 3 0.887 0.16479 0.100 0.080 0.400 0.136 0.284
#> GSM1269726 4 0.933 -0.00491 0.176 0.076 0.224 0.360 0.164
#> GSM1269648 1 0.921 0.11143 0.384 0.080 0.132 0.188 0.216
#> GSM1269656 2 0.902 0.11631 0.304 0.360 0.068 0.136 0.132
#> GSM1269664 3 0.933 0.04343 0.204 0.080 0.356 0.220 0.140
#> GSM1269672 4 0.866 0.13347 0.264 0.088 0.148 0.432 0.068
#> GSM1269680 2 0.571 0.24807 0.108 0.708 0.024 0.016 0.144
#> GSM1269686 4 0.912 0.02066 0.276 0.072 0.280 0.288 0.084
#> GSM1269694 1 0.965 0.09075 0.292 0.104 0.152 0.220 0.232
#> GSM1269702 1 0.940 0.11749 0.352 0.236 0.116 0.188 0.108
#> GSM1269710 1 0.961 0.05650 0.316 0.100 0.224 0.184 0.176
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM1269647 2 0.842 0.14897 0.080 0.436 0.104 0.044 0.100 0.236
#> GSM1269655 5 0.992 -0.02811 0.128 0.132 0.180 0.184 0.204 0.172
#> GSM1269663 5 0.974 0.00915 0.160 0.096 0.124 0.184 0.260 0.176
#> GSM1269671 2 0.794 0.22810 0.144 0.496 0.048 0.036 0.092 0.184
#> GSM1269679 3 0.785 0.14265 0.072 0.100 0.512 0.188 0.100 0.028
#> GSM1269693 4 0.776 0.17853 0.068 0.040 0.104 0.512 0.216 0.060
#> GSM1269701 3 0.882 0.11999 0.216 0.104 0.396 0.144 0.088 0.052
#> GSM1269709 3 0.963 -0.01118 0.148 0.100 0.252 0.240 0.160 0.100
#> GSM1269715 4 0.603 0.23324 0.092 0.020 0.136 0.668 0.072 0.012
#> GSM1269717 4 0.656 0.21243 0.068 0.020 0.108 0.644 0.112 0.048
#> GSM1269721 6 0.924 0.00645 0.084 0.228 0.100 0.088 0.168 0.332
#> GSM1269723 3 0.877 0.10418 0.216 0.088 0.380 0.192 0.076 0.048
#> GSM1269645 1 0.953 0.05036 0.268 0.132 0.136 0.156 0.240 0.068
#> GSM1269653 2 0.944 0.12827 0.172 0.272 0.148 0.048 0.136 0.224
#> GSM1269661 3 0.979 0.04326 0.188 0.128 0.212 0.196 0.188 0.088
#> GSM1269669 4 0.868 0.01410 0.244 0.048 0.236 0.288 0.168 0.016
#> GSM1269677 6 0.516 0.29825 0.024 0.096 0.020 0.016 0.108 0.736
#> GSM1269685 4 0.870 -0.00111 0.136 0.044 0.068 0.380 0.256 0.116
#> GSM1269691 5 0.854 0.09402 0.076 0.056 0.088 0.228 0.428 0.124
#> GSM1269699 2 0.774 0.16111 0.192 0.448 0.020 0.028 0.072 0.240
#> GSM1269707 2 0.896 0.04693 0.088 0.316 0.108 0.056 0.128 0.304
#> GSM1269651 6 0.787 0.19283 0.064 0.172 0.100 0.044 0.096 0.524
#> GSM1269659 6 0.800 0.20231 0.036 0.076 0.060 0.120 0.236 0.472
#> GSM1269667 3 0.891 0.11962 0.184 0.124 0.372 0.196 0.080 0.044
#> GSM1269675 2 0.876 0.18952 0.160 0.420 0.088 0.048 0.124 0.160
#> GSM1269683 4 0.900 0.07785 0.100 0.072 0.208 0.372 0.172 0.076
#> GSM1269689 2 0.889 0.15198 0.260 0.368 0.092 0.064 0.096 0.120
#> GSM1269697 2 0.926 0.15326 0.120 0.348 0.172 0.068 0.108 0.184
#> GSM1269705 2 0.844 0.18221 0.116 0.464 0.068 0.060 0.112 0.180
#> GSM1269713 2 0.893 0.02637 0.108 0.328 0.292 0.068 0.064 0.140
#> GSM1269719 1 0.988 -0.00228 0.204 0.104 0.148 0.188 0.180 0.176
#> GSM1269725 3 0.862 -0.01605 0.076 0.276 0.388 0.092 0.056 0.112
#> GSM1269727 4 0.889 0.04315 0.220 0.080 0.212 0.316 0.152 0.020
#> GSM1269649 1 0.914 0.04964 0.292 0.220 0.232 0.072 0.136 0.048
#> GSM1269657 6 0.761 0.26475 0.092 0.084 0.052 0.080 0.124 0.568
#> GSM1269665 1 0.941 0.03097 0.272 0.072 0.156 0.216 0.208 0.076
#> GSM1269673 5 0.860 0.08155 0.252 0.036 0.092 0.184 0.372 0.064
#> GSM1269681 6 0.664 0.15083 0.064 0.244 0.028 0.032 0.048 0.584
#> GSM1269687 1 0.957 0.03997 0.248 0.096 0.228 0.172 0.180 0.076
#> GSM1269695 1 0.819 0.13109 0.508 0.144 0.076 0.112 0.080 0.080
#> GSM1269703 1 0.949 0.06264 0.288 0.072 0.180 0.152 0.208 0.100
#> GSM1269711 1 0.912 0.07070 0.348 0.116 0.136 0.104 0.236 0.060
#> GSM1269646 2 0.869 0.10697 0.112 0.412 0.220 0.052 0.076 0.128
#> GSM1269654 6 0.970 -0.15155 0.084 0.108 0.216 0.200 0.172 0.220
#> GSM1269662 6 0.951 -0.07345 0.160 0.092 0.132 0.096 0.240 0.280
#> GSM1269670 2 0.742 0.24333 0.160 0.528 0.064 0.024 0.036 0.188
#> GSM1269678 3 0.782 0.12087 0.060 0.076 0.496 0.236 0.092 0.040
#> GSM1269692 4 0.781 0.10057 0.060 0.020 0.084 0.456 0.280 0.100
#> GSM1269700 3 0.871 0.11513 0.180 0.112 0.416 0.160 0.092 0.040
#> GSM1269708 4 0.940 0.05342 0.172 0.116 0.148 0.300 0.208 0.056
#> GSM1269714 4 0.716 0.20386 0.064 0.048 0.128 0.592 0.132 0.036
#> GSM1269716 4 0.581 0.23674 0.036 0.020 0.136 0.688 0.096 0.024
#> GSM1269720 6 0.801 0.17889 0.064 0.160 0.052 0.056 0.172 0.496
#> GSM1269722 3 0.941 0.07412 0.092 0.132 0.312 0.164 0.216 0.084
#> GSM1269644 5 0.939 0.08400 0.156 0.084 0.108 0.140 0.340 0.172
#> GSM1269652 6 0.973 -0.07510 0.172 0.216 0.124 0.104 0.132 0.252
#> GSM1269660 3 0.976 0.04389 0.116 0.164 0.268 0.184 0.124 0.144
#> GSM1269668 4 0.818 0.02482 0.168 0.056 0.304 0.360 0.104 0.008
#> GSM1269676 6 0.536 0.30125 0.036 0.096 0.020 0.032 0.080 0.736
#> GSM1269684 4 0.845 0.08070 0.160 0.044 0.108 0.396 0.252 0.040
#> GSM1269690 4 0.791 -0.04868 0.100 0.020 0.024 0.396 0.304 0.156
#> GSM1269698 6 0.791 -0.04927 0.064 0.328 0.064 0.032 0.096 0.416
#> GSM1269706 2 0.892 0.10361 0.128 0.348 0.040 0.072 0.192 0.220
#> GSM1269650 6 0.712 0.20906 0.076 0.152 0.032 0.024 0.140 0.576
#> GSM1269658 6 0.771 0.19824 0.088 0.028 0.060 0.068 0.292 0.464
#> GSM1269666 3 0.856 0.08584 0.084 0.132 0.392 0.272 0.064 0.056
#> GSM1269674 2 0.886 0.16907 0.140 0.380 0.120 0.060 0.068 0.232
#> GSM1269682 4 0.898 0.07901 0.128 0.064 0.228 0.356 0.164 0.060
#> GSM1269688 2 0.881 0.11603 0.232 0.376 0.164 0.044 0.100 0.084
#> GSM1269696 2 0.914 0.06218 0.092 0.328 0.248 0.068 0.096 0.168
#> GSM1269704 2 0.801 0.09407 0.064 0.400 0.120 0.036 0.052 0.328
#> GSM1269712 4 0.926 -0.08134 0.128 0.148 0.276 0.288 0.100 0.060
#> GSM1269718 1 0.961 0.03568 0.248 0.108 0.204 0.072 0.176 0.192
#> GSM1269724 3 0.861 0.13767 0.096 0.244 0.404 0.124 0.092 0.040
#> GSM1269726 4 0.913 0.07666 0.192 0.088 0.160 0.336 0.180 0.044
#> GSM1269648 1 0.877 0.12436 0.424 0.152 0.124 0.084 0.160 0.056
#> GSM1269656 6 0.905 0.11427 0.128 0.148 0.060 0.108 0.168 0.388
#> GSM1269664 3 0.925 -0.01691 0.136 0.084 0.320 0.168 0.228 0.064
#> GSM1269672 5 0.847 0.02341 0.160 0.044 0.104 0.276 0.376 0.040
#> GSM1269680 6 0.554 0.22656 0.052 0.196 0.012 0.016 0.044 0.680
#> GSM1269686 4 0.908 0.01833 0.248 0.040 0.192 0.260 0.204 0.056
#> GSM1269694 1 0.873 0.11110 0.440 0.164 0.072 0.128 0.084 0.112
#> GSM1269702 5 0.934 0.03158 0.216 0.072 0.060 0.168 0.268 0.216
#> GSM1269710 1 0.895 0.10837 0.400 0.084 0.156 0.108 0.180 0.072
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 agent(p) disease.state(p) gender(p) individual(p) k
#> MAD:skmeans 72 0.979 0.477 0.302 0.000714 2
#> MAD:skmeans 12 NA NA NA NA 3
#> MAD:skmeans 0 NA NA NA NA 4
#> MAD:skmeans 0 NA NA NA NA 5
#> MAD:skmeans 0 NA NA NA 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["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 51941 rows and 84 columns.
#> Top rows (1000, 2000, 3000, 4000, 5000) are extracted by 'MAD' method.
#> Subgroups are detected by 'pam' method.
#> Performed in total 1250 partitions by row resampling.
#> Best k for subgroups seems to be 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 0.0136 0.399 0.700 0.4552 0.587 0.587
#> 3 3 0.0441 0.419 0.650 0.3633 0.672 0.491
#> 4 4 0.0970 0.320 0.579 0.1232 0.888 0.728
#> 5 5 0.1694 0.299 0.553 0.0655 0.947 0.847
#> 6 6 0.2486 0.304 0.541 0.0400 0.933 0.790
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
#> GSM1269647 1 0.781 0.5635 0.768 0.232
#> GSM1269655 2 0.722 0.5438 0.200 0.800
#> GSM1269663 2 0.992 0.3328 0.448 0.552
#> GSM1269671 1 0.634 0.5955 0.840 0.160
#> GSM1269679 1 1.000 0.0115 0.500 0.500
#> GSM1269693 2 0.456 0.5300 0.096 0.904
#> GSM1269701 2 1.000 -0.0337 0.488 0.512
#> GSM1269709 1 0.552 0.6182 0.872 0.128
#> GSM1269715 2 0.855 0.5274 0.280 0.720
#> GSM1269717 2 0.991 0.2784 0.444 0.556
#> GSM1269721 2 0.644 0.5641 0.164 0.836
#> GSM1269723 2 0.999 0.0757 0.480 0.520
#> GSM1269645 1 0.552 0.6224 0.872 0.128
#> GSM1269653 1 1.000 -0.2645 0.504 0.496
#> GSM1269661 1 1.000 -0.2671 0.512 0.488
#> GSM1269669 2 0.917 0.4465 0.332 0.668
#> GSM1269677 1 0.552 0.6149 0.872 0.128
#> GSM1269685 1 0.706 0.5624 0.808 0.192
#> GSM1269691 2 0.973 0.3984 0.404 0.596
#> GSM1269699 1 0.653 0.5924 0.832 0.168
#> GSM1269707 1 0.260 0.6196 0.956 0.044
#> GSM1269651 2 0.913 0.4694 0.328 0.672
#> GSM1269659 1 0.388 0.6220 0.924 0.076
#> GSM1269667 1 0.939 0.3585 0.644 0.356
#> GSM1269675 1 0.753 0.5749 0.784 0.216
#> GSM1269683 2 0.913 0.5147 0.328 0.672
#> GSM1269689 2 0.966 0.3049 0.392 0.608
#> GSM1269697 2 1.000 -0.0176 0.496 0.504
#> GSM1269705 1 1.000 -0.0924 0.512 0.488
#> GSM1269713 1 1.000 0.0752 0.508 0.492
#> GSM1269719 1 0.714 0.5932 0.804 0.196
#> GSM1269725 1 0.917 0.4033 0.668 0.332
#> GSM1269727 2 0.990 0.1448 0.440 0.560
#> GSM1269649 1 0.943 0.2673 0.640 0.360
#> GSM1269657 1 0.456 0.6218 0.904 0.096
#> GSM1269665 1 0.788 0.4727 0.764 0.236
#> GSM1269673 1 0.983 0.0295 0.576 0.424
#> GSM1269681 1 0.866 0.5139 0.712 0.288
#> GSM1269687 1 0.416 0.6209 0.916 0.084
#> GSM1269695 1 0.242 0.6215 0.960 0.040
#> GSM1269703 1 0.242 0.6149 0.960 0.040
#> GSM1269711 1 0.988 0.0264 0.564 0.436
#> GSM1269646 2 0.936 0.4743 0.352 0.648
#> GSM1269654 1 0.995 0.0860 0.540 0.460
#> GSM1269662 1 0.738 0.5195 0.792 0.208
#> GSM1269670 1 0.605 0.6097 0.852 0.148
#> GSM1269678 2 0.917 0.4638 0.332 0.668
#> GSM1269692 1 0.909 0.4158 0.676 0.324
#> GSM1269700 2 0.891 0.4386 0.308 0.692
#> GSM1269708 1 0.745 0.5807 0.788 0.212
#> GSM1269714 1 0.552 0.6153 0.872 0.128
#> GSM1269716 2 0.855 0.5408 0.280 0.720
#> GSM1269720 1 0.943 0.3271 0.640 0.360
#> GSM1269722 2 0.625 0.5593 0.156 0.844
#> GSM1269644 1 0.994 -0.2056 0.544 0.456
#> GSM1269652 1 0.625 0.6167 0.844 0.156
#> GSM1269660 1 0.781 0.5207 0.768 0.232
#> GSM1269668 1 1.000 -0.1385 0.508 0.492
#> GSM1269676 1 0.993 -0.0883 0.548 0.452
#> GSM1269684 1 0.653 0.6020 0.832 0.168
#> GSM1269690 2 0.997 0.2731 0.468 0.532
#> GSM1269698 1 0.969 0.2227 0.604 0.396
#> GSM1269706 1 0.714 0.4763 0.804 0.196
#> GSM1269650 1 0.993 0.0605 0.548 0.452
#> GSM1269658 1 0.900 0.4039 0.684 0.316
#> GSM1269666 2 0.808 0.5447 0.248 0.752
#> GSM1269674 2 0.978 0.4117 0.412 0.588
#> GSM1269682 1 0.671 0.6141 0.824 0.176
#> GSM1269688 1 0.921 0.4175 0.664 0.336
#> GSM1269696 1 0.949 0.2501 0.632 0.368
#> GSM1269704 1 0.738 0.5875 0.792 0.208
#> GSM1269712 1 0.388 0.6183 0.924 0.076
#> GSM1269718 1 0.999 0.0544 0.516 0.484
#> GSM1269724 1 0.871 0.4797 0.708 0.292
#> GSM1269726 2 0.775 0.5756 0.228 0.772
#> GSM1269648 1 0.416 0.6231 0.916 0.084
#> GSM1269656 1 0.388 0.6215 0.924 0.076
#> GSM1269664 1 0.781 0.5725 0.768 0.232
#> GSM1269672 1 0.973 0.0544 0.596 0.404
#> GSM1269680 1 0.311 0.6165 0.944 0.056
#> GSM1269686 1 0.311 0.6175 0.944 0.056
#> GSM1269694 1 0.563 0.6114 0.868 0.132
#> GSM1269702 1 0.141 0.6102 0.980 0.020
#> GSM1269710 1 0.605 0.5810 0.852 0.148
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM1269647 1 0.818 0.3703 0.592 0.096 0.312
#> GSM1269655 2 0.708 0.4082 0.060 0.684 0.256
#> GSM1269663 2 0.993 -0.0326 0.284 0.384 0.332
#> GSM1269671 1 0.723 0.4425 0.640 0.048 0.312
#> GSM1269679 3 0.970 0.2786 0.360 0.220 0.420
#> GSM1269693 2 0.493 0.4525 0.024 0.820 0.156
#> GSM1269701 3 0.981 0.3469 0.284 0.284 0.432
#> GSM1269709 1 0.520 0.6424 0.820 0.044 0.136
#> GSM1269715 2 0.808 0.4240 0.148 0.648 0.204
#> GSM1269717 2 0.704 0.4703 0.252 0.688 0.060
#> GSM1269721 2 0.628 0.4337 0.064 0.760 0.176
#> GSM1269723 3 0.671 0.5195 0.176 0.084 0.740
#> GSM1269645 1 0.621 0.6120 0.776 0.088 0.136
#> GSM1269653 2 1.000 0.0385 0.328 0.344 0.328
#> GSM1269661 3 0.894 0.3724 0.328 0.144 0.528
#> GSM1269669 2 0.684 0.4489 0.076 0.724 0.200
#> GSM1269677 1 0.454 0.6574 0.836 0.148 0.016
#> GSM1269685 1 0.626 0.5863 0.724 0.244 0.032
#> GSM1269691 2 0.571 0.5171 0.204 0.768 0.028
#> GSM1269699 1 0.655 0.5940 0.756 0.096 0.148
#> GSM1269707 1 0.321 0.6546 0.912 0.028 0.060
#> GSM1269651 2 0.952 -0.0217 0.188 0.424 0.388
#> GSM1269659 1 0.437 0.6581 0.868 0.056 0.076
#> GSM1269667 3 0.857 0.1874 0.428 0.096 0.476
#> GSM1269675 1 0.759 0.3734 0.640 0.072 0.288
#> GSM1269683 2 0.931 0.0774 0.168 0.468 0.364
#> GSM1269689 3 0.880 0.3751 0.276 0.156 0.568
#> GSM1269697 3 0.928 0.3967 0.212 0.264 0.524
#> GSM1269705 1 0.993 -0.1832 0.388 0.328 0.284
#> GSM1269713 3 0.975 0.2280 0.364 0.228 0.408
#> GSM1269719 1 0.715 0.5939 0.720 0.124 0.156
#> GSM1269725 3 0.828 0.4042 0.344 0.092 0.564
#> GSM1269727 3 0.670 0.4866 0.108 0.144 0.748
#> GSM1269649 1 0.930 0.2170 0.500 0.316 0.184
#> GSM1269657 1 0.377 0.6595 0.880 0.104 0.016
#> GSM1269665 1 0.599 0.4390 0.688 0.304 0.008
#> GSM1269673 2 0.721 0.4476 0.360 0.604 0.036
#> GSM1269681 1 0.905 0.3072 0.556 0.216 0.228
#> GSM1269687 1 0.560 0.6423 0.800 0.052 0.148
#> GSM1269695 1 0.429 0.6525 0.864 0.032 0.104
#> GSM1269703 1 0.178 0.6531 0.960 0.020 0.020
#> GSM1269711 1 0.967 0.1213 0.456 0.304 0.240
#> GSM1269646 3 0.823 0.4170 0.144 0.224 0.632
#> GSM1269654 2 0.882 0.2566 0.336 0.532 0.132
#> GSM1269662 1 0.729 0.0583 0.560 0.032 0.408
#> GSM1269670 1 0.608 0.6321 0.784 0.128 0.088
#> GSM1269678 3 0.799 0.3841 0.144 0.200 0.656
#> GSM1269692 1 0.844 0.2711 0.568 0.324 0.108
#> GSM1269700 3 0.485 0.4176 0.036 0.128 0.836
#> GSM1269708 1 0.739 0.6028 0.704 0.156 0.140
#> GSM1269714 1 0.676 0.6199 0.736 0.084 0.180
#> GSM1269716 2 0.734 0.5111 0.152 0.708 0.140
#> GSM1269720 3 0.887 0.4312 0.380 0.124 0.496
#> GSM1269722 3 0.709 0.3652 0.056 0.268 0.676
#> GSM1269644 2 0.743 0.4873 0.328 0.620 0.052
#> GSM1269652 1 0.651 0.6336 0.760 0.104 0.136
#> GSM1269660 1 0.714 0.0452 0.576 0.028 0.396
#> GSM1269668 2 0.997 0.1641 0.328 0.368 0.304
#> GSM1269676 2 0.623 0.4838 0.316 0.672 0.012
#> GSM1269684 1 0.706 0.5424 0.708 0.080 0.212
#> GSM1269690 2 0.541 0.5195 0.212 0.772 0.016
#> GSM1269698 3 0.866 0.2886 0.408 0.104 0.488
#> GSM1269706 1 0.662 0.3775 0.684 0.284 0.032
#> GSM1269650 2 0.805 0.3367 0.356 0.568 0.076
#> GSM1269658 1 0.902 0.1775 0.560 0.216 0.224
#> GSM1269666 3 0.745 0.3801 0.080 0.252 0.668
#> GSM1269674 2 0.942 0.2596 0.228 0.504 0.268
#> GSM1269682 1 0.687 0.5874 0.736 0.104 0.160
#> GSM1269688 1 0.879 0.4218 0.584 0.224 0.192
#> GSM1269696 3 0.550 0.5299 0.248 0.008 0.744
#> GSM1269704 1 0.691 0.5901 0.728 0.092 0.180
#> GSM1269712 1 0.438 0.6604 0.868 0.060 0.072
#> GSM1269718 3 0.989 0.3026 0.336 0.268 0.396
#> GSM1269724 1 0.887 0.2075 0.496 0.124 0.380
#> GSM1269726 2 0.852 -0.0842 0.092 0.464 0.444
#> GSM1269648 1 0.573 0.6154 0.796 0.060 0.144
#> GSM1269656 1 0.255 0.6584 0.936 0.040 0.024
#> GSM1269664 1 0.787 0.5508 0.660 0.216 0.124
#> GSM1269672 2 0.791 0.3411 0.404 0.536 0.060
#> GSM1269680 1 0.243 0.6491 0.940 0.036 0.024
#> GSM1269686 1 0.426 0.6480 0.868 0.036 0.096
#> GSM1269694 1 0.621 0.6406 0.776 0.136 0.088
#> GSM1269702 1 0.127 0.6446 0.972 0.024 0.004
#> GSM1269710 1 0.617 0.6210 0.768 0.168 0.064
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM1269647 1 0.867 0.2331 0.488 0.076 0.260 0.176
#> GSM1269655 2 0.780 0.1434 0.032 0.504 0.124 0.340
#> GSM1269663 2 0.960 -0.0831 0.244 0.352 0.276 0.128
#> GSM1269671 1 0.768 0.4560 0.592 0.048 0.220 0.140
#> GSM1269679 1 0.969 -0.3317 0.304 0.132 0.280 0.284
#> GSM1269693 2 0.463 0.4410 0.012 0.804 0.140 0.044
#> GSM1269701 4 0.954 0.1292 0.228 0.124 0.288 0.360
#> GSM1269709 1 0.589 0.5423 0.716 0.008 0.104 0.172
#> GSM1269715 2 0.801 0.1662 0.132 0.468 0.036 0.364
#> GSM1269717 2 0.755 0.3296 0.188 0.612 0.048 0.152
#> GSM1269721 2 0.496 0.3976 0.032 0.796 0.132 0.040
#> GSM1269723 3 0.548 0.4143 0.160 0.040 0.760 0.040
#> GSM1269645 1 0.553 0.5187 0.724 0.032 0.024 0.220
#> GSM1269653 4 0.998 0.1684 0.260 0.256 0.220 0.264
#> GSM1269661 3 0.793 0.2997 0.324 0.120 0.512 0.044
#> GSM1269669 2 0.725 0.2138 0.068 0.504 0.032 0.396
#> GSM1269677 1 0.503 0.5796 0.784 0.152 0.036 0.028
#> GSM1269685 1 0.643 0.5163 0.672 0.236 0.044 0.048
#> GSM1269691 2 0.344 0.4875 0.152 0.840 0.004 0.004
#> GSM1269699 1 0.736 0.4964 0.640 0.096 0.188 0.076
#> GSM1269707 1 0.348 0.5919 0.884 0.032 0.052 0.032
#> GSM1269651 4 0.937 0.1475 0.108 0.248 0.248 0.396
#> GSM1269659 1 0.438 0.5933 0.840 0.072 0.056 0.032
#> GSM1269667 1 0.846 -0.1325 0.400 0.024 0.300 0.276
#> GSM1269675 1 0.754 0.3715 0.560 0.072 0.308 0.060
#> GSM1269683 4 0.958 0.1384 0.116 0.276 0.284 0.324
#> GSM1269689 4 0.890 0.1968 0.172 0.076 0.360 0.392
#> GSM1269697 3 0.943 0.0265 0.144 0.204 0.420 0.232
#> GSM1269705 3 0.982 -0.0629 0.292 0.256 0.292 0.160
#> GSM1269713 4 0.927 0.2661 0.200 0.128 0.232 0.440
#> GSM1269719 1 0.802 0.3400 0.548 0.080 0.096 0.276
#> GSM1269725 3 0.858 0.2004 0.272 0.044 0.452 0.232
#> GSM1269727 3 0.642 0.3555 0.072 0.060 0.712 0.156
#> GSM1269649 1 0.961 -0.0678 0.388 0.236 0.160 0.216
#> GSM1269657 1 0.414 0.5914 0.848 0.088 0.028 0.036
#> GSM1269665 1 0.575 0.4332 0.652 0.308 0.024 0.016
#> GSM1269673 2 0.658 0.3535 0.332 0.580 0.004 0.084
#> GSM1269681 1 0.929 -0.0381 0.372 0.088 0.260 0.280
#> GSM1269687 1 0.644 0.4168 0.620 0.012 0.068 0.300
#> GSM1269695 1 0.547 0.5711 0.780 0.044 0.088 0.088
#> GSM1269703 1 0.199 0.5853 0.944 0.024 0.020 0.012
#> GSM1269711 1 0.924 -0.1768 0.364 0.208 0.092 0.336
#> GSM1269646 3 0.840 0.3127 0.116 0.156 0.560 0.168
#> GSM1269654 2 0.907 -0.0921 0.236 0.368 0.068 0.328
#> GSM1269662 1 0.768 0.1416 0.504 0.068 0.368 0.060
#> GSM1269670 1 0.645 0.5505 0.720 0.100 0.112 0.068
#> GSM1269678 3 0.855 0.2305 0.120 0.108 0.516 0.256
#> GSM1269692 1 0.723 0.2930 0.524 0.360 0.100 0.016
#> GSM1269700 3 0.503 0.2418 0.012 0.008 0.700 0.280
#> GSM1269708 1 0.670 0.5477 0.688 0.136 0.136 0.040
#> GSM1269714 1 0.676 0.4593 0.632 0.020 0.092 0.256
#> GSM1269716 2 0.618 0.4707 0.104 0.740 0.076 0.080
#> GSM1269720 3 0.826 0.2634 0.344 0.100 0.480 0.076
#> GSM1269722 3 0.572 0.3556 0.028 0.256 0.692 0.024
#> GSM1269644 2 0.682 0.4080 0.284 0.620 0.044 0.052
#> GSM1269652 1 0.628 0.5781 0.732 0.072 0.084 0.112
#> GSM1269660 1 0.731 0.1499 0.528 0.020 0.352 0.100
#> GSM1269668 4 0.911 0.1260 0.236 0.260 0.084 0.420
#> GSM1269676 2 0.604 0.4346 0.220 0.700 0.052 0.028
#> GSM1269684 1 0.625 0.4756 0.692 0.024 0.076 0.208
#> GSM1269690 2 0.390 0.4793 0.132 0.832 0.000 0.036
#> GSM1269698 3 0.853 0.1796 0.296 0.088 0.492 0.124
#> GSM1269706 1 0.587 0.3542 0.656 0.296 0.016 0.032
#> GSM1269650 2 0.902 0.1100 0.248 0.468 0.116 0.168
#> GSM1269658 1 0.760 0.2295 0.528 0.228 0.236 0.008
#> GSM1269666 3 0.805 0.2147 0.048 0.160 0.544 0.248
#> GSM1269674 2 0.932 0.0993 0.168 0.440 0.236 0.156
#> GSM1269682 1 0.717 0.4963 0.656 0.056 0.160 0.128
#> GSM1269688 1 0.883 0.2417 0.480 0.220 0.216 0.084
#> GSM1269696 3 0.599 0.3891 0.188 0.008 0.704 0.100
#> GSM1269704 1 0.694 0.5241 0.668 0.072 0.188 0.072
#> GSM1269712 1 0.445 0.5893 0.812 0.036 0.012 0.140
#> GSM1269718 4 0.968 0.2133 0.180 0.180 0.280 0.360
#> GSM1269724 4 0.651 0.3152 0.228 0.012 0.104 0.656
#> GSM1269726 3 0.798 0.1966 0.072 0.376 0.476 0.076
#> GSM1269648 1 0.565 0.5178 0.736 0.024 0.052 0.188
#> GSM1269656 1 0.347 0.5922 0.884 0.048 0.044 0.024
#> GSM1269664 1 0.828 0.3473 0.532 0.196 0.056 0.216
#> GSM1269672 2 0.699 0.2520 0.348 0.524 0.000 0.128
#> GSM1269680 1 0.356 0.5885 0.880 0.024 0.044 0.052
#> GSM1269686 1 0.554 0.4581 0.696 0.008 0.040 0.256
#> GSM1269694 1 0.643 0.5548 0.704 0.136 0.128 0.032
#> GSM1269702 1 0.125 0.5787 0.968 0.016 0.004 0.012
#> GSM1269710 1 0.557 0.5524 0.752 0.168 0.040 0.040
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM1269647 1 0.8520 0.02295 0.396 0.104 0.208 0.024 0.268
#> GSM1269655 2 0.7729 0.15048 0.016 0.400 0.096 0.396 0.092
#> GSM1269663 4 0.8606 -0.03628 0.220 0.156 0.284 0.332 0.008
#> GSM1269671 1 0.7531 0.41266 0.556 0.096 0.232 0.036 0.080
#> GSM1269679 1 0.9258 -0.16912 0.276 0.272 0.268 0.124 0.060
#> GSM1269693 4 0.4063 0.43589 0.016 0.040 0.120 0.816 0.008
#> GSM1269701 2 0.8492 -0.03059 0.188 0.416 0.272 0.088 0.036
#> GSM1269709 1 0.5995 0.52559 0.688 0.168 0.092 0.020 0.032
#> GSM1269715 2 0.7181 -0.08421 0.116 0.416 0.044 0.416 0.008
#> GSM1269717 4 0.7660 0.13834 0.164 0.200 0.044 0.544 0.048
#> GSM1269721 4 0.5202 0.36065 0.024 0.020 0.116 0.756 0.084
#> GSM1269723 3 0.4871 0.46914 0.108 0.040 0.784 0.048 0.020
#> GSM1269645 1 0.5824 0.49222 0.668 0.236 0.024 0.044 0.028
#> GSM1269653 5 0.9843 0.09238 0.192 0.240 0.132 0.180 0.256
#> GSM1269661 3 0.7013 0.34684 0.304 0.036 0.532 0.112 0.016
#> GSM1269669 4 0.6603 0.02845 0.052 0.440 0.028 0.456 0.024
#> GSM1269677 1 0.5976 0.49280 0.684 0.016 0.024 0.132 0.144
#> GSM1269685 1 0.5657 0.51388 0.672 0.052 0.032 0.236 0.008
#> GSM1269691 4 0.2674 0.47513 0.140 0.000 0.004 0.856 0.000
#> GSM1269699 1 0.6992 0.37956 0.576 0.008 0.128 0.060 0.228
#> GSM1269707 1 0.3572 0.56783 0.860 0.016 0.028 0.024 0.072
#> GSM1269651 2 0.8960 0.17107 0.056 0.412 0.196 0.192 0.144
#> GSM1269659 1 0.5227 0.57018 0.772 0.044 0.056 0.088 0.040
#> GSM1269667 1 0.8279 -0.05426 0.340 0.248 0.332 0.052 0.028
#> GSM1269675 1 0.7284 0.34068 0.524 0.036 0.308 0.044 0.088
#> GSM1269683 2 0.9163 0.19270 0.096 0.348 0.264 0.212 0.080
#> GSM1269689 5 0.8567 0.22133 0.080 0.280 0.220 0.036 0.384
#> GSM1269697 3 0.9681 -0.07303 0.100 0.208 0.296 0.188 0.208
#> GSM1269705 5 0.8958 0.20400 0.180 0.064 0.188 0.144 0.424
#> GSM1269713 5 0.8744 0.17115 0.088 0.304 0.096 0.108 0.404
#> GSM1269719 1 0.7827 0.24673 0.492 0.296 0.092 0.080 0.040
#> GSM1269725 3 0.8671 0.20441 0.196 0.156 0.420 0.028 0.200
#> GSM1269727 3 0.4992 0.44768 0.052 0.136 0.764 0.036 0.012
#> GSM1269649 1 0.9319 0.01697 0.356 0.176 0.156 0.236 0.076
#> GSM1269657 1 0.4953 0.54729 0.780 0.024 0.036 0.052 0.108
#> GSM1269665 1 0.4977 0.41664 0.656 0.032 0.012 0.300 0.000
#> GSM1269673 4 0.5888 0.33057 0.308 0.088 0.004 0.592 0.008
#> GSM1269681 5 0.8718 0.00820 0.216 0.236 0.112 0.036 0.400
#> GSM1269687 1 0.6176 0.42836 0.604 0.296 0.056 0.024 0.020
#> GSM1269695 1 0.5878 0.53353 0.728 0.072 0.104 0.040 0.056
#> GSM1269703 1 0.1278 0.56497 0.960 0.004 0.016 0.020 0.000
#> GSM1269711 1 0.9189 -0.08142 0.324 0.316 0.084 0.164 0.112
#> GSM1269646 3 0.8050 0.29288 0.064 0.112 0.540 0.100 0.184
#> GSM1269654 2 0.8832 0.16796 0.184 0.368 0.056 0.300 0.092
#> GSM1269662 1 0.7323 0.07850 0.428 0.036 0.416 0.080 0.040
#> GSM1269670 1 0.7072 0.48776 0.640 0.056 0.100 0.080 0.124
#> GSM1269678 3 0.8215 0.27031 0.092 0.260 0.488 0.092 0.068
#> GSM1269692 1 0.6527 0.28306 0.500 0.008 0.124 0.360 0.008
#> GSM1269700 3 0.6410 0.20929 0.004 0.244 0.556 0.004 0.192
#> GSM1269708 1 0.6666 0.52532 0.660 0.064 0.112 0.132 0.032
#> GSM1269714 1 0.6762 0.44209 0.576 0.280 0.088 0.028 0.028
#> GSM1269716 4 0.5220 0.45381 0.096 0.096 0.052 0.752 0.004
#> GSM1269720 3 0.7535 0.30468 0.300 0.088 0.500 0.096 0.016
#> GSM1269722 3 0.4860 0.43270 0.032 0.020 0.720 0.224 0.004
#> GSM1269644 4 0.6225 0.40123 0.280 0.040 0.040 0.616 0.024
#> GSM1269652 1 0.6100 0.56102 0.712 0.104 0.068 0.080 0.036
#> GSM1269660 1 0.6519 0.08171 0.480 0.088 0.404 0.020 0.008
#> GSM1269668 2 0.8318 0.09767 0.180 0.456 0.084 0.244 0.036
#> GSM1269676 4 0.6522 0.38830 0.160 0.020 0.020 0.624 0.176
#> GSM1269684 1 0.6005 0.46180 0.656 0.220 0.084 0.032 0.008
#> GSM1269690 4 0.3346 0.46507 0.108 0.036 0.008 0.848 0.000
#> GSM1269698 3 0.7921 0.05745 0.220 0.020 0.416 0.044 0.300
#> GSM1269706 1 0.6171 0.32386 0.600 0.020 0.012 0.292 0.076
#> GSM1269650 4 0.9026 -0.01097 0.192 0.184 0.048 0.392 0.184
#> GSM1269658 1 0.6908 0.26737 0.528 0.012 0.244 0.204 0.012
#> GSM1269666 3 0.7139 0.20155 0.016 0.264 0.544 0.136 0.040
#> GSM1269674 4 0.9256 0.00538 0.116 0.080 0.192 0.340 0.272
#> GSM1269682 1 0.6950 0.46760 0.620 0.112 0.184 0.048 0.036
#> GSM1269688 5 0.8786 0.21633 0.320 0.032 0.144 0.160 0.344
#> GSM1269696 3 0.5615 0.44254 0.152 0.080 0.716 0.008 0.044
#> GSM1269704 1 0.6755 0.46153 0.624 0.036 0.124 0.028 0.188
#> GSM1269712 1 0.4593 0.57169 0.800 0.108 0.028 0.032 0.032
#> GSM1269718 2 0.9221 0.10269 0.148 0.352 0.280 0.140 0.080
#> GSM1269724 2 0.5792 0.04737 0.140 0.716 0.068 0.016 0.060
#> GSM1269726 3 0.7132 0.25467 0.068 0.060 0.488 0.364 0.020
#> GSM1269648 1 0.5159 0.52034 0.736 0.180 0.036 0.032 0.016
#> GSM1269656 1 0.3883 0.56244 0.832 0.008 0.020 0.036 0.104
#> GSM1269664 1 0.8435 0.31208 0.484 0.188 0.096 0.176 0.056
#> GSM1269672 4 0.6204 0.24425 0.336 0.136 0.000 0.524 0.004
#> GSM1269680 1 0.3462 0.56135 0.836 0.028 0.004 0.004 0.128
#> GSM1269686 1 0.5479 0.46702 0.676 0.244 0.052 0.012 0.016
#> GSM1269694 1 0.6405 0.51842 0.672 0.024 0.140 0.112 0.052
#> GSM1269702 1 0.0613 0.55837 0.984 0.000 0.004 0.008 0.004
#> GSM1269710 1 0.5616 0.52968 0.712 0.052 0.024 0.180 0.032
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM1269647 1 0.846 -0.11305 0.344 0.112 0.172 0.016 0.064 0.292
#> GSM1269655 2 0.683 0.29988 0.020 0.532 0.056 0.260 0.124 0.008
#> GSM1269663 4 0.804 -0.06917 0.204 0.024 0.288 0.348 0.128 0.008
#> GSM1269671 1 0.790 0.36218 0.492 0.072 0.204 0.020 0.120 0.092
#> GSM1269679 5 0.859 0.11631 0.244 0.068 0.244 0.096 0.328 0.020
#> GSM1269693 4 0.278 0.50598 0.020 0.004 0.112 0.860 0.004 0.000
#> GSM1269701 5 0.826 0.12642 0.168 0.132 0.252 0.036 0.392 0.020
#> GSM1269709 1 0.588 0.51342 0.672 0.016 0.092 0.040 0.160 0.020
#> GSM1269715 5 0.605 0.24858 0.104 0.004 0.032 0.368 0.492 0.000
#> GSM1269717 4 0.663 0.05394 0.140 0.316 0.028 0.492 0.020 0.004
#> GSM1269721 4 0.559 0.38800 0.016 0.104 0.092 0.700 0.004 0.084
#> GSM1269723 3 0.379 0.44245 0.072 0.028 0.836 0.016 0.036 0.012
#> GSM1269645 1 0.546 0.45458 0.652 0.036 0.024 0.024 0.252 0.012
#> GSM1269653 5 0.932 -0.00862 0.136 0.068 0.092 0.168 0.316 0.220
#> GSM1269661 3 0.634 0.34417 0.300 0.008 0.532 0.116 0.040 0.004
#> GSM1269669 5 0.511 0.16303 0.044 0.012 0.004 0.412 0.528 0.000
#> GSM1269677 1 0.703 0.35904 0.548 0.228 0.012 0.096 0.028 0.088
#> GSM1269685 1 0.539 0.51578 0.668 0.032 0.024 0.232 0.036 0.008
#> GSM1269691 4 0.275 0.53319 0.096 0.028 0.004 0.868 0.004 0.000
#> GSM1269699 1 0.703 0.34589 0.532 0.032 0.096 0.048 0.024 0.268
#> GSM1269707 1 0.363 0.57820 0.848 0.020 0.024 0.028 0.016 0.064
#> GSM1269651 2 0.795 0.29164 0.040 0.480 0.140 0.084 0.216 0.040
#> GSM1269659 1 0.505 0.57627 0.752 0.028 0.044 0.120 0.036 0.020
#> GSM1269667 1 0.780 -0.18250 0.304 0.068 0.304 0.020 0.292 0.012
#> GSM1269675 1 0.737 0.29668 0.480 0.088 0.288 0.016 0.028 0.100
#> GSM1269683 2 0.786 0.32909 0.068 0.484 0.200 0.140 0.096 0.012
#> GSM1269689 6 0.786 0.25617 0.056 0.036 0.152 0.040 0.268 0.448
#> GSM1269697 3 0.971 -0.12815 0.104 0.200 0.248 0.156 0.100 0.192
#> GSM1269705 6 0.812 0.17266 0.160 0.084 0.120 0.116 0.028 0.492
#> GSM1269713 6 0.847 0.24787 0.064 0.152 0.080 0.048 0.240 0.416
#> GSM1269719 1 0.715 0.22465 0.476 0.316 0.080 0.024 0.092 0.012
#> GSM1269725 3 0.866 0.22158 0.204 0.140 0.388 0.020 0.088 0.160
#> GSM1269727 3 0.442 0.41862 0.036 0.056 0.792 0.024 0.088 0.004
#> GSM1269649 1 0.921 -0.06960 0.308 0.072 0.172 0.228 0.164 0.056
#> GSM1269657 1 0.526 0.53207 0.716 0.156 0.012 0.028 0.020 0.068
#> GSM1269665 1 0.562 0.43177 0.628 0.080 0.028 0.248 0.016 0.000
#> GSM1269673 4 0.603 0.34141 0.276 0.072 0.004 0.580 0.064 0.004
#> GSM1269681 2 0.746 0.13191 0.140 0.492 0.080 0.016 0.028 0.244
#> GSM1269687 1 0.633 0.41277 0.608 0.072 0.060 0.028 0.224 0.008
#> GSM1269695 1 0.701 0.47644 0.612 0.096 0.124 0.032 0.084 0.052
#> GSM1269703 1 0.178 0.56694 0.936 0.008 0.024 0.024 0.008 0.000
#> GSM1269711 5 0.890 0.09076 0.292 0.056 0.068 0.172 0.316 0.096
#> GSM1269646 3 0.847 0.19505 0.052 0.088 0.452 0.088 0.120 0.200
#> GSM1269654 2 0.678 0.31901 0.168 0.544 0.016 0.208 0.060 0.004
#> GSM1269662 3 0.758 -0.00711 0.368 0.048 0.408 0.100 0.040 0.036
#> GSM1269670 1 0.717 0.44087 0.556 0.164 0.104 0.004 0.048 0.124
#> GSM1269678 3 0.729 0.08686 0.084 0.020 0.400 0.084 0.392 0.020
#> GSM1269692 1 0.661 0.27962 0.484 0.056 0.112 0.336 0.008 0.004
#> GSM1269700 3 0.684 0.09742 0.008 0.044 0.516 0.012 0.216 0.204
#> GSM1269708 1 0.638 0.52144 0.640 0.024 0.112 0.148 0.056 0.020
#> GSM1269714 1 0.671 0.38758 0.548 0.028 0.092 0.048 0.272 0.012
#> GSM1269716 4 0.441 0.50075 0.076 0.012 0.056 0.784 0.072 0.000
#> GSM1269720 3 0.758 0.32673 0.252 0.068 0.508 0.080 0.056 0.036
#> GSM1269722 3 0.422 0.40465 0.028 0.016 0.748 0.196 0.012 0.000
#> GSM1269644 4 0.585 0.39921 0.252 0.040 0.044 0.620 0.044 0.000
#> GSM1269652 1 0.565 0.56327 0.712 0.012 0.076 0.084 0.088 0.028
#> GSM1269660 1 0.636 0.05696 0.468 0.008 0.388 0.036 0.092 0.008
#> GSM1269668 5 0.598 0.37495 0.144 0.012 0.036 0.196 0.612 0.000
#> GSM1269676 4 0.724 0.22878 0.096 0.212 0.008 0.532 0.036 0.116
#> GSM1269684 1 0.532 0.41858 0.640 0.004 0.068 0.024 0.260 0.004
#> GSM1269690 4 0.262 0.51970 0.080 0.000 0.016 0.880 0.024 0.000
#> GSM1269698 3 0.829 0.07737 0.188 0.060 0.356 0.056 0.032 0.308
#> GSM1269706 1 0.585 0.30575 0.576 0.012 0.008 0.312 0.028 0.064
#> GSM1269650 2 0.814 0.21567 0.116 0.428 0.008 0.248 0.104 0.096
#> GSM1269658 1 0.672 0.22935 0.508 0.044 0.240 0.196 0.008 0.004
#> GSM1269666 3 0.629 0.09926 0.016 0.356 0.512 0.040 0.068 0.008
#> GSM1269674 2 0.971 0.01518 0.080 0.240 0.140 0.212 0.180 0.148
#> GSM1269682 1 0.648 0.38713 0.568 0.028 0.164 0.036 0.204 0.000
#> GSM1269688 6 0.760 0.25960 0.276 0.016 0.084 0.160 0.020 0.444
#> GSM1269696 3 0.499 0.42917 0.136 0.016 0.736 0.004 0.072 0.036
#> GSM1269704 1 0.643 0.46415 0.604 0.056 0.084 0.016 0.020 0.220
#> GSM1269712 1 0.420 0.55569 0.756 0.032 0.016 0.012 0.184 0.000
#> GSM1269718 2 0.900 0.15482 0.128 0.328 0.260 0.068 0.164 0.052
#> GSM1269724 5 0.756 0.16738 0.132 0.200 0.052 0.024 0.524 0.068
#> GSM1269726 3 0.662 0.24875 0.048 0.060 0.504 0.344 0.036 0.008
#> GSM1269648 1 0.497 0.50394 0.708 0.012 0.060 0.016 0.196 0.008
#> GSM1269656 1 0.380 0.57568 0.832 0.032 0.012 0.044 0.008 0.072
#> GSM1269664 1 0.810 0.22525 0.440 0.072 0.072 0.140 0.252 0.024
#> GSM1269672 4 0.553 0.14968 0.316 0.000 0.000 0.528 0.156 0.000
#> GSM1269680 1 0.416 0.57058 0.800 0.044 0.004 0.012 0.036 0.104
#> GSM1269686 1 0.568 0.43612 0.652 0.048 0.044 0.016 0.228 0.012
#> GSM1269694 1 0.705 0.47653 0.596 0.068 0.172 0.076 0.036 0.052
#> GSM1269702 1 0.115 0.55844 0.960 0.004 0.000 0.020 0.016 0.000
#> GSM1269710 1 0.527 0.52733 0.688 0.008 0.020 0.208 0.052 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 agent(p) disease.state(p) gender(p) individual(p) k
#> MAD:pam 43 0.937 0.407 0.00349 0.2506 2
#> MAD:pam 33 0.927 0.132 0.15509 0.0642 3
#> MAD:pam 20 NA NA NA NA 4
#> MAD:pam 16 NA NA NA NA 5
#> MAD:pam 18 1.000 0.089 0.84113 0.1575 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 51941 rows and 84 columns.
#> Top rows (1000, 2000, 3000, 4000, 5000) are extracted by 'MAD' method.
#> Subgroups are detected by 'mclust' method.
#> Performed in total 1250 partitions by row resampling.
#> Best k for subgroups seems to be 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 0.0889 0.118 0.745 0.3306 0.953 0.953
#> 3 3 0.0798 0.638 0.706 0.6185 0.517 0.499
#> 4 4 0.3093 0.521 0.647 0.2487 0.764 0.523
#> 5 5 0.4265 0.467 0.674 0.1177 0.917 0.720
#> 6 6 0.4972 0.463 0.642 0.0578 0.933 0.752
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
#> GSM1269647 1 0.995 -0.8313 0.540 0.460
#> GSM1269655 1 0.416 0.5317 0.916 0.084
#> GSM1269663 1 0.242 0.5082 0.960 0.040
#> GSM1269671 2 1.000 0.8369 0.496 0.504
#> GSM1269679 1 0.871 0.3919 0.708 0.292
#> GSM1269693 1 0.295 0.5047 0.948 0.052
#> GSM1269701 1 0.814 0.4372 0.748 0.252
#> GSM1269709 1 0.443 0.5060 0.908 0.092
#> GSM1269715 1 0.141 0.5285 0.980 0.020
#> GSM1269717 1 0.184 0.5307 0.972 0.028
#> GSM1269721 1 0.936 -0.5565 0.648 0.352
#> GSM1269723 1 0.876 0.3966 0.704 0.296
#> GSM1269645 1 0.697 0.5148 0.812 0.188
#> GSM1269653 1 0.985 -0.6246 0.572 0.428
#> GSM1269661 1 0.827 0.4548 0.740 0.260
#> GSM1269669 1 0.506 0.5289 0.888 0.112
#> GSM1269677 1 0.995 -0.6462 0.540 0.460
#> GSM1269685 1 0.529 0.4827 0.880 0.120
#> GSM1269691 1 0.443 0.4995 0.908 0.092
#> GSM1269699 2 1.000 0.8298 0.488 0.512
#> GSM1269707 1 0.983 -0.6303 0.576 0.424
#> GSM1269651 1 0.985 -0.6415 0.572 0.428
#> GSM1269659 1 0.929 -0.5030 0.656 0.344
#> GSM1269667 1 0.644 0.5008 0.836 0.164
#> GSM1269675 1 0.988 -0.7634 0.564 0.436
#> GSM1269683 1 0.494 0.5227 0.892 0.108
#> GSM1269689 1 0.988 -0.7102 0.564 0.436
#> GSM1269697 1 0.980 -0.6694 0.584 0.416
#> GSM1269705 1 0.988 -0.7911 0.564 0.436
#> GSM1269713 1 0.745 0.2450 0.788 0.212
#> GSM1269719 1 0.278 0.5193 0.952 0.048
#> GSM1269725 1 0.866 -0.1070 0.712 0.288
#> GSM1269727 1 0.855 0.4147 0.720 0.280
#> GSM1269649 1 0.714 0.4980 0.804 0.196
#> GSM1269657 1 0.961 -0.5717 0.616 0.384
#> GSM1269665 1 0.662 0.5084 0.828 0.172
#> GSM1269673 1 0.430 0.5261 0.912 0.088
#> GSM1269681 1 0.991 -0.6631 0.556 0.444
#> GSM1269687 1 0.518 0.5314 0.884 0.116
#> GSM1269695 1 0.469 0.5206 0.900 0.100
#> GSM1269703 1 0.584 0.5273 0.860 0.140
#> GSM1269711 1 0.644 0.4946 0.836 0.164
#> GSM1269646 1 0.963 -0.6044 0.612 0.388
#> GSM1269654 1 0.327 0.5335 0.940 0.060
#> GSM1269662 1 0.358 0.4727 0.932 0.068
#> GSM1269670 1 0.999 -0.8712 0.520 0.480
#> GSM1269678 1 0.904 0.3909 0.680 0.320
#> GSM1269692 1 0.260 0.5113 0.956 0.044
#> GSM1269700 1 0.821 0.4321 0.744 0.256
#> GSM1269708 1 0.373 0.5083 0.928 0.072
#> GSM1269714 1 0.163 0.5299 0.976 0.024
#> GSM1269716 1 0.184 0.5307 0.972 0.028
#> GSM1269720 1 0.955 -0.6032 0.624 0.376
#> GSM1269722 1 0.494 0.5274 0.892 0.108
#> GSM1269644 1 0.416 0.5113 0.916 0.084
#> GSM1269652 1 0.961 -0.4814 0.616 0.384
#> GSM1269660 1 0.808 0.4654 0.752 0.248
#> GSM1269668 1 0.494 0.5285 0.892 0.108
#> GSM1269676 1 0.998 -0.6590 0.524 0.476
#> GSM1269684 1 0.416 0.5227 0.916 0.084
#> GSM1269690 1 0.494 0.4940 0.892 0.108
#> GSM1269698 1 1.000 -0.8790 0.508 0.492
#> GSM1269706 1 0.963 -0.5032 0.612 0.388
#> GSM1269650 1 0.985 -0.6398 0.572 0.428
#> GSM1269658 1 0.821 -0.1498 0.744 0.256
#> GSM1269666 1 0.900 0.3759 0.684 0.316
#> GSM1269674 1 0.966 -0.6499 0.608 0.392
#> GSM1269682 1 0.563 0.5143 0.868 0.132
#> GSM1269688 1 0.985 -0.7009 0.572 0.428
#> GSM1269696 1 0.949 -0.5296 0.632 0.368
#> GSM1269704 1 1.000 -0.8742 0.512 0.488
#> GSM1269712 1 0.833 0.4173 0.736 0.264
#> GSM1269718 1 0.260 0.5272 0.956 0.044
#> GSM1269724 1 0.808 0.4379 0.752 0.248
#> GSM1269726 1 0.821 0.4333 0.744 0.256
#> GSM1269648 1 0.584 0.5084 0.860 0.140
#> GSM1269656 1 0.866 -0.0861 0.712 0.288
#> GSM1269664 1 0.689 0.5034 0.816 0.184
#> GSM1269672 1 0.388 0.5284 0.924 0.076
#> GSM1269680 1 0.997 -0.6559 0.532 0.468
#> GSM1269686 1 0.443 0.5314 0.908 0.092
#> GSM1269694 1 0.469 0.5213 0.900 0.100
#> GSM1269702 1 0.469 0.4980 0.900 0.100
#> GSM1269710 1 0.662 0.5024 0.828 0.172
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM1269647 3 0.429 0.701 0.092 0.040 0.868
#> GSM1269655 1 0.559 0.737 0.812 0.096 0.092
#> GSM1269663 1 0.532 0.705 0.824 0.072 0.104
#> GSM1269671 3 0.471 0.687 0.092 0.056 0.852
#> GSM1269679 1 0.801 0.654 0.640 0.244 0.116
#> GSM1269693 1 0.516 0.700 0.832 0.096 0.072
#> GSM1269701 1 0.810 0.655 0.640 0.228 0.132
#> GSM1269709 1 0.644 0.665 0.720 0.040 0.240
#> GSM1269715 1 0.463 0.705 0.856 0.088 0.056
#> GSM1269717 1 0.477 0.700 0.848 0.100 0.052
#> GSM1269721 1 0.971 -0.490 0.436 0.232 0.332
#> GSM1269723 1 0.801 0.651 0.640 0.244 0.116
#> GSM1269645 1 0.617 0.743 0.776 0.144 0.080
#> GSM1269653 3 0.756 0.612 0.164 0.144 0.692
#> GSM1269661 1 0.712 0.698 0.708 0.204 0.088
#> GSM1269669 1 0.535 0.747 0.824 0.088 0.088
#> GSM1269677 2 0.876 0.793 0.216 0.588 0.196
#> GSM1269685 1 0.549 0.665 0.816 0.104 0.080
#> GSM1269691 1 0.512 0.677 0.832 0.108 0.060
#> GSM1269699 3 0.697 0.616 0.120 0.148 0.732
#> GSM1269707 3 0.879 0.390 0.244 0.176 0.580
#> GSM1269651 2 0.921 0.790 0.224 0.536 0.240
#> GSM1269659 2 0.951 0.691 0.348 0.456 0.196
#> GSM1269667 1 0.766 0.692 0.676 0.208 0.116
#> GSM1269675 3 0.560 0.697 0.136 0.060 0.804
#> GSM1269683 1 0.453 0.738 0.860 0.088 0.052
#> GSM1269689 3 0.395 0.694 0.076 0.040 0.884
#> GSM1269697 3 0.393 0.687 0.092 0.028 0.880
#> GSM1269705 3 0.533 0.705 0.120 0.060 0.820
#> GSM1269713 3 0.718 0.497 0.304 0.048 0.648
#> GSM1269719 1 0.557 0.705 0.812 0.080 0.108
#> GSM1269725 3 0.719 0.496 0.292 0.052 0.656
#> GSM1269727 1 0.746 0.670 0.672 0.244 0.084
#> GSM1269649 1 0.859 0.627 0.604 0.216 0.180
#> GSM1269657 2 0.948 0.702 0.336 0.468 0.196
#> GSM1269665 1 0.543 0.746 0.808 0.144 0.048
#> GSM1269673 1 0.440 0.705 0.864 0.092 0.044
#> GSM1269681 2 0.932 0.753 0.220 0.520 0.260
#> GSM1269687 1 0.489 0.732 0.840 0.112 0.048
#> GSM1269695 1 0.659 0.682 0.744 0.076 0.180
#> GSM1269703 1 0.386 0.748 0.888 0.072 0.040
#> GSM1269711 1 0.752 0.585 0.660 0.080 0.260
#> GSM1269646 3 0.538 0.676 0.188 0.024 0.788
#> GSM1269654 1 0.543 0.735 0.820 0.092 0.088
#> GSM1269662 1 0.576 0.678 0.800 0.076 0.124
#> GSM1269670 3 0.492 0.688 0.108 0.052 0.840
#> GSM1269678 1 0.738 0.675 0.680 0.236 0.084
#> GSM1269692 1 0.500 0.702 0.840 0.092 0.068
#> GSM1269700 1 0.816 0.659 0.636 0.228 0.136
#> GSM1269708 1 0.572 0.730 0.792 0.052 0.156
#> GSM1269714 1 0.437 0.713 0.868 0.076 0.056
#> GSM1269716 1 0.479 0.699 0.848 0.096 0.056
#> GSM1269720 1 0.998 -0.696 0.364 0.328 0.308
#> GSM1269722 1 0.631 0.733 0.772 0.128 0.100
#> GSM1269644 1 0.447 0.716 0.864 0.060 0.076
#> GSM1269652 3 0.891 0.336 0.280 0.164 0.556
#> GSM1269660 1 0.651 0.716 0.748 0.180 0.072
#> GSM1269668 1 0.534 0.741 0.824 0.092 0.084
#> GSM1269676 2 0.876 0.793 0.216 0.588 0.196
#> GSM1269684 1 0.469 0.686 0.852 0.096 0.052
#> GSM1269690 1 0.519 0.673 0.828 0.112 0.060
#> GSM1269698 3 0.720 0.603 0.124 0.160 0.716
#> GSM1269706 3 0.921 0.264 0.296 0.184 0.520
#> GSM1269650 2 0.917 0.785 0.216 0.540 0.244
#> GSM1269658 2 0.938 0.644 0.380 0.448 0.172
#> GSM1269666 1 0.808 0.639 0.628 0.260 0.112
#> GSM1269674 3 0.795 0.204 0.388 0.064 0.548
#> GSM1269682 1 0.474 0.741 0.848 0.104 0.048
#> GSM1269688 3 0.498 0.698 0.136 0.036 0.828
#> GSM1269696 3 0.537 0.703 0.140 0.048 0.812
#> GSM1269704 3 0.401 0.708 0.096 0.028 0.876
#> GSM1269712 1 0.814 0.659 0.624 0.260 0.116
#> GSM1269718 1 0.523 0.725 0.828 0.068 0.104
#> GSM1269724 1 0.904 0.508 0.544 0.176 0.280
#> GSM1269726 1 0.732 0.687 0.696 0.208 0.096
#> GSM1269648 1 0.730 0.650 0.688 0.084 0.228
#> GSM1269656 1 0.811 0.293 0.648 0.160 0.192
#> GSM1269664 1 0.471 0.749 0.844 0.120 0.036
#> GSM1269672 1 0.453 0.703 0.860 0.088 0.052
#> GSM1269680 2 0.912 0.798 0.224 0.548 0.228
#> GSM1269686 1 0.389 0.714 0.884 0.084 0.032
#> GSM1269694 1 0.625 0.715 0.772 0.084 0.144
#> GSM1269702 1 0.482 0.688 0.848 0.088 0.064
#> GSM1269710 1 0.655 0.705 0.756 0.096 0.148
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM1269647 2 0.395 0.7180 0.012 0.832 0.140 0.016
#> GSM1269655 3 0.556 0.6022 0.316 0.024 0.652 0.008
#> GSM1269663 1 0.732 -0.1432 0.456 0.028 0.440 0.076
#> GSM1269671 2 0.397 0.6839 0.016 0.856 0.072 0.056
#> GSM1269679 3 0.499 0.6647 0.148 0.064 0.780 0.008
#> GSM1269693 1 0.595 0.4928 0.724 0.024 0.176 0.076
#> GSM1269701 3 0.482 0.6819 0.160 0.048 0.784 0.008
#> GSM1269709 3 0.891 0.0399 0.312 0.288 0.352 0.048
#> GSM1269715 1 0.525 0.5035 0.752 0.012 0.188 0.048
#> GSM1269717 1 0.507 0.4999 0.764 0.016 0.184 0.036
#> GSM1269721 4 0.896 0.4190 0.316 0.284 0.052 0.348
#> GSM1269723 3 0.464 0.6833 0.164 0.044 0.788 0.004
#> GSM1269645 3 0.703 0.4493 0.356 0.060 0.552 0.032
#> GSM1269653 2 0.731 0.6473 0.116 0.648 0.168 0.068
#> GSM1269661 3 0.506 0.6731 0.272 0.020 0.704 0.004
#> GSM1269669 3 0.604 0.5098 0.416 0.036 0.544 0.004
#> GSM1269677 4 0.329 0.7823 0.080 0.044 0.000 0.876
#> GSM1269685 1 0.241 0.5568 0.928 0.016 0.036 0.020
#> GSM1269691 1 0.211 0.5603 0.932 0.000 0.044 0.024
#> GSM1269699 2 0.649 0.6316 0.052 0.712 0.116 0.120
#> GSM1269707 2 0.848 0.5063 0.172 0.552 0.124 0.152
#> GSM1269651 4 0.552 0.7784 0.068 0.108 0.048 0.776
#> GSM1269659 4 0.615 0.7226 0.304 0.056 0.008 0.632
#> GSM1269667 3 0.450 0.6926 0.192 0.032 0.776 0.000
#> GSM1269675 2 0.400 0.6985 0.020 0.852 0.092 0.036
#> GSM1269683 3 0.548 0.4592 0.412 0.004 0.572 0.012
#> GSM1269689 2 0.342 0.7184 0.024 0.884 0.064 0.028
#> GSM1269697 2 0.354 0.7140 0.004 0.832 0.160 0.004
#> GSM1269705 2 0.471 0.7156 0.008 0.800 0.132 0.060
#> GSM1269713 2 0.690 0.4623 0.132 0.588 0.276 0.004
#> GSM1269719 3 0.740 0.3648 0.372 0.024 0.508 0.096
#> GSM1269725 2 0.680 0.5411 0.076 0.600 0.304 0.020
#> GSM1269727 3 0.454 0.6814 0.216 0.024 0.760 0.000
#> GSM1269649 3 0.715 0.4039 0.292 0.148 0.556 0.004
#> GSM1269657 4 0.639 0.7144 0.300 0.072 0.008 0.620
#> GSM1269665 3 0.530 0.5380 0.408 0.000 0.580 0.012
#> GSM1269673 1 0.361 0.5004 0.800 0.000 0.200 0.000
#> GSM1269681 4 0.589 0.7506 0.068 0.136 0.048 0.748
#> GSM1269687 1 0.531 -0.0539 0.596 0.008 0.392 0.004
#> GSM1269695 1 0.870 0.1992 0.464 0.212 0.264 0.060
#> GSM1269703 1 0.562 -0.1431 0.564 0.012 0.416 0.008
#> GSM1269711 1 0.871 0.0830 0.396 0.364 0.184 0.056
#> GSM1269646 2 0.512 0.6271 0.016 0.700 0.276 0.008
#> GSM1269654 3 0.577 0.5258 0.356 0.016 0.612 0.016
#> GSM1269662 3 0.822 0.1049 0.396 0.044 0.424 0.136
#> GSM1269670 2 0.399 0.6833 0.012 0.852 0.080 0.056
#> GSM1269678 3 0.519 0.6769 0.212 0.040 0.740 0.008
#> GSM1269692 1 0.537 0.5067 0.756 0.012 0.164 0.068
#> GSM1269700 3 0.477 0.6847 0.168 0.048 0.780 0.004
#> GSM1269708 1 0.827 0.0383 0.456 0.148 0.352 0.044
#> GSM1269714 1 0.496 0.4923 0.752 0.016 0.212 0.020
#> GSM1269716 1 0.503 0.5018 0.768 0.016 0.180 0.036
#> GSM1269720 4 0.779 0.6979 0.244 0.168 0.032 0.556
#> GSM1269722 3 0.546 0.6550 0.276 0.036 0.684 0.004
#> GSM1269644 1 0.512 0.4640 0.748 0.016 0.208 0.028
#> GSM1269652 2 0.861 0.5239 0.200 0.532 0.152 0.116
#> GSM1269660 3 0.480 0.6750 0.260 0.020 0.720 0.000
#> GSM1269668 3 0.604 0.5272 0.412 0.036 0.548 0.004
#> GSM1269676 4 0.329 0.7823 0.080 0.044 0.000 0.876
#> GSM1269684 1 0.286 0.5600 0.888 0.008 0.100 0.004
#> GSM1269690 1 0.189 0.5583 0.940 0.000 0.044 0.016
#> GSM1269698 2 0.553 0.6521 0.032 0.760 0.056 0.152
#> GSM1269706 2 0.870 0.4740 0.200 0.524 0.124 0.152
#> GSM1269650 4 0.547 0.7761 0.068 0.104 0.048 0.780
#> GSM1269658 4 0.600 0.7565 0.240 0.052 0.020 0.688
#> GSM1269666 3 0.445 0.6774 0.156 0.048 0.796 0.000
#> GSM1269674 2 0.759 0.4319 0.080 0.556 0.308 0.056
#> GSM1269682 3 0.546 0.5555 0.368 0.004 0.612 0.016
#> GSM1269688 2 0.439 0.7144 0.064 0.840 0.064 0.032
#> GSM1269696 2 0.527 0.6811 0.028 0.740 0.212 0.020
#> GSM1269704 2 0.354 0.7244 0.012 0.868 0.096 0.024
#> GSM1269712 3 0.507 0.6674 0.168 0.056 0.768 0.008
#> GSM1269718 3 0.679 0.4488 0.376 0.016 0.544 0.064
#> GSM1269724 3 0.642 0.5530 0.140 0.168 0.680 0.012
#> GSM1269726 3 0.448 0.6662 0.248 0.012 0.740 0.000
#> GSM1269648 1 0.891 0.0521 0.400 0.256 0.288 0.056
#> GSM1269656 1 0.628 0.1159 0.660 0.088 0.008 0.244
#> GSM1269664 3 0.543 0.5724 0.392 0.004 0.592 0.012
#> GSM1269672 1 0.354 0.5278 0.828 0.008 0.164 0.000
#> GSM1269680 4 0.458 0.7831 0.068 0.076 0.028 0.828
#> GSM1269686 1 0.437 0.3666 0.728 0.004 0.268 0.000
#> GSM1269694 1 0.873 0.2021 0.448 0.220 0.276 0.056
#> GSM1269702 1 0.357 0.5570 0.872 0.012 0.080 0.036
#> GSM1269710 1 0.808 -0.1645 0.424 0.144 0.400 0.032
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM1269647 5 0.2968 0.6660 0.000 0.008 0.092 0.028 0.872
#> GSM1269655 3 0.6049 0.5384 0.188 0.028 0.684 0.052 0.048
#> GSM1269663 3 0.7479 0.1274 0.332 0.076 0.476 0.104 0.012
#> GSM1269671 5 0.3319 0.6035 0.000 0.020 0.000 0.160 0.820
#> GSM1269679 3 0.2922 0.6185 0.016 0.000 0.880 0.024 0.080
#> GSM1269693 1 0.6407 0.4984 0.628 0.040 0.192 0.136 0.004
#> GSM1269701 3 0.2910 0.6324 0.024 0.000 0.888 0.052 0.036
#> GSM1269709 3 0.8299 -0.1479 0.172 0.000 0.368 0.280 0.180
#> GSM1269715 1 0.5484 0.5152 0.684 0.008 0.192 0.112 0.004
#> GSM1269717 1 0.5592 0.4993 0.676 0.008 0.184 0.128 0.004
#> GSM1269721 2 0.7884 0.3883 0.256 0.452 0.008 0.076 0.208
#> GSM1269723 3 0.2751 0.6363 0.020 0.004 0.900 0.032 0.044
#> GSM1269645 3 0.6943 0.3835 0.308 0.004 0.500 0.164 0.024
#> GSM1269653 5 0.6078 0.0416 0.024 0.016 0.036 0.392 0.532
#> GSM1269661 3 0.3502 0.6198 0.144 0.004 0.828 0.012 0.012
#> GSM1269669 3 0.5687 0.4783 0.280 0.000 0.632 0.060 0.028
#> GSM1269677 2 0.0968 0.7728 0.012 0.972 0.000 0.004 0.012
#> GSM1269685 1 0.3611 0.5334 0.848 0.036 0.036 0.080 0.000
#> GSM1269691 1 0.3309 0.5581 0.868 0.032 0.052 0.048 0.000
#> GSM1269699 5 0.5308 0.4177 0.004 0.072 0.004 0.256 0.664
#> GSM1269707 4 0.7283 0.2061 0.064 0.100 0.008 0.448 0.380
#> GSM1269651 2 0.4678 0.7529 0.004 0.768 0.016 0.144 0.068
#> GSM1269659 2 0.4337 0.7254 0.172 0.776 0.004 0.032 0.016
#> GSM1269667 3 0.1777 0.6382 0.012 0.004 0.944 0.020 0.020
#> GSM1269675 5 0.4290 0.5981 0.008 0.008 0.036 0.168 0.780
#> GSM1269683 3 0.5210 0.4138 0.292 0.004 0.648 0.052 0.004
#> GSM1269689 5 0.4159 0.4857 0.004 0.004 0.016 0.232 0.744
#> GSM1269697 5 0.2780 0.6606 0.004 0.004 0.112 0.008 0.872
#> GSM1269705 5 0.3307 0.6663 0.004 0.036 0.072 0.020 0.868
#> GSM1269713 5 0.5128 0.3987 0.032 0.000 0.312 0.016 0.640
#> GSM1269719 3 0.7412 0.2924 0.256 0.120 0.528 0.088 0.008
#> GSM1269725 5 0.4970 0.4728 0.016 0.008 0.292 0.016 0.668
#> GSM1269727 3 0.1854 0.6356 0.020 0.000 0.936 0.036 0.008
#> GSM1269649 3 0.6515 0.4291 0.228 0.000 0.608 0.092 0.072
#> GSM1269657 2 0.4808 0.7161 0.164 0.760 0.012 0.044 0.020
#> GSM1269665 3 0.5537 0.5195 0.248 0.004 0.656 0.084 0.008
#> GSM1269673 1 0.4834 0.4758 0.692 0.004 0.252 0.052 0.000
#> GSM1269681 2 0.5094 0.7266 0.004 0.740 0.016 0.124 0.116
#> GSM1269687 1 0.5083 -0.0601 0.540 0.000 0.428 0.028 0.004
#> GSM1269695 1 0.8119 -0.1970 0.340 0.004 0.244 0.328 0.084
#> GSM1269703 3 0.5754 0.1712 0.460 0.000 0.472 0.056 0.012
#> GSM1269711 4 0.8364 0.2078 0.292 0.000 0.208 0.340 0.160
#> GSM1269646 5 0.3403 0.6340 0.000 0.008 0.160 0.012 0.820
#> GSM1269654 3 0.6533 0.4649 0.228 0.036 0.632 0.064 0.040
#> GSM1269662 3 0.8353 -0.0147 0.308 0.156 0.388 0.136 0.012
#> GSM1269670 5 0.3566 0.6074 0.000 0.024 0.004 0.160 0.812
#> GSM1269678 3 0.4447 0.6026 0.140 0.004 0.784 0.016 0.056
#> GSM1269692 1 0.6344 0.5160 0.640 0.044 0.180 0.132 0.004
#> GSM1269700 3 0.2578 0.6324 0.016 0.000 0.904 0.040 0.040
#> GSM1269708 3 0.8215 0.0370 0.208 0.004 0.408 0.256 0.124
#> GSM1269714 1 0.5487 0.4792 0.644 0.000 0.252 0.100 0.004
#> GSM1269716 1 0.5550 0.5026 0.680 0.008 0.184 0.124 0.004
#> GSM1269720 2 0.5967 0.6958 0.144 0.684 0.008 0.036 0.128
#> GSM1269722 3 0.4163 0.6210 0.112 0.008 0.816 0.032 0.032
#> GSM1269644 1 0.6213 0.4356 0.644 0.056 0.216 0.080 0.004
#> GSM1269652 4 0.7434 0.3206 0.100 0.060 0.020 0.480 0.340
#> GSM1269660 3 0.3682 0.6197 0.140 0.008 0.824 0.016 0.012
#> GSM1269668 3 0.5667 0.4907 0.276 0.000 0.636 0.060 0.028
#> GSM1269676 2 0.1200 0.7723 0.012 0.964 0.000 0.008 0.016
#> GSM1269684 1 0.2908 0.5847 0.868 0.008 0.108 0.016 0.000
#> GSM1269690 1 0.3229 0.5510 0.872 0.032 0.040 0.056 0.000
#> GSM1269698 5 0.4493 0.5672 0.000 0.100 0.008 0.120 0.772
#> GSM1269706 4 0.7621 0.3529 0.112 0.096 0.008 0.472 0.312
#> GSM1269650 2 0.4493 0.7563 0.004 0.780 0.012 0.136 0.068
#> GSM1269658 2 0.4341 0.7482 0.144 0.792 0.012 0.040 0.012
#> GSM1269666 3 0.1682 0.6274 0.000 0.004 0.940 0.012 0.044
#> GSM1269674 5 0.7535 0.3304 0.036 0.048 0.248 0.132 0.536
#> GSM1269682 3 0.5234 0.5458 0.184 0.012 0.712 0.088 0.004
#> GSM1269688 5 0.5143 0.3509 0.016 0.004 0.032 0.284 0.664
#> GSM1269696 5 0.3620 0.6386 0.004 0.016 0.156 0.008 0.816
#> GSM1269704 5 0.2555 0.6631 0.004 0.016 0.048 0.024 0.908
#> GSM1269712 3 0.4195 0.6173 0.072 0.008 0.812 0.012 0.096
#> GSM1269718 3 0.6743 0.4028 0.248 0.064 0.596 0.080 0.012
#> GSM1269724 3 0.5309 0.4953 0.048 0.008 0.684 0.016 0.244
#> GSM1269726 3 0.2976 0.6345 0.064 0.000 0.880 0.044 0.012
#> GSM1269648 1 0.8287 -0.2520 0.296 0.000 0.292 0.292 0.120
#> GSM1269656 1 0.7381 0.1374 0.484 0.324 0.028 0.136 0.028
#> GSM1269664 3 0.5312 0.5364 0.256 0.004 0.668 0.064 0.008
#> GSM1269672 1 0.4462 0.5125 0.740 0.000 0.196 0.064 0.000
#> GSM1269680 2 0.3576 0.7686 0.000 0.840 0.012 0.100 0.048
#> GSM1269686 1 0.4252 0.3854 0.700 0.000 0.280 0.020 0.000
#> GSM1269694 4 0.7965 -0.0843 0.332 0.000 0.252 0.336 0.080
#> GSM1269702 1 0.5780 0.4936 0.708 0.036 0.128 0.116 0.012
#> GSM1269710 3 0.7703 0.1100 0.324 0.000 0.408 0.196 0.072
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM1269647 2 0.280 0.7198 0.024 0.884 0.048 0.000 0.036 0.008
#> GSM1269655 3 0.634 0.5621 0.040 0.032 0.632 0.144 0.140 0.012
#> GSM1269663 3 0.803 0.2902 0.112 0.004 0.396 0.232 0.212 0.044
#> GSM1269671 2 0.433 0.6791 0.072 0.772 0.000 0.000 0.108 0.048
#> GSM1269679 3 0.317 0.5926 0.048 0.052 0.864 0.012 0.024 0.000
#> GSM1269693 4 0.372 0.5158 0.012 0.000 0.076 0.828 0.056 0.028
#> GSM1269701 3 0.279 0.5935 0.072 0.012 0.880 0.016 0.020 0.000
#> GSM1269709 1 0.644 0.3427 0.524 0.088 0.320 0.048 0.012 0.008
#> GSM1269715 4 0.183 0.5487 0.012 0.000 0.064 0.920 0.004 0.000
#> GSM1269717 4 0.197 0.5357 0.004 0.000 0.056 0.916 0.024 0.000
#> GSM1269721 6 0.828 0.1114 0.128 0.168 0.004 0.092 0.188 0.420
#> GSM1269723 3 0.330 0.6149 0.052 0.024 0.864 0.032 0.024 0.004
#> GSM1269645 3 0.689 0.3996 0.352 0.012 0.444 0.060 0.128 0.004
#> GSM1269653 2 0.662 0.2362 0.292 0.428 0.020 0.000 0.252 0.008
#> GSM1269661 3 0.486 0.6118 0.128 0.012 0.748 0.048 0.060 0.004
#> GSM1269669 3 0.539 0.4479 0.256 0.004 0.636 0.076 0.024 0.004
#> GSM1269677 6 0.362 0.5849 0.004 0.016 0.000 0.000 0.236 0.744
#> GSM1269685 4 0.585 0.3876 0.352 0.000 0.012 0.536 0.068 0.032
#> GSM1269691 4 0.542 0.4431 0.328 0.000 0.012 0.580 0.068 0.012
#> GSM1269699 2 0.547 0.5338 0.068 0.596 0.000 0.000 0.296 0.040
#> GSM1269707 1 0.728 -0.0408 0.352 0.276 0.000 0.008 0.296 0.068
#> GSM1269651 6 0.369 0.5432 0.004 0.064 0.004 0.000 0.128 0.800
#> GSM1269659 6 0.553 0.4780 0.032 0.012 0.004 0.048 0.292 0.612
#> GSM1269667 3 0.157 0.6167 0.032 0.008 0.944 0.008 0.008 0.000
#> GSM1269675 2 0.475 0.6822 0.100 0.760 0.028 0.000 0.080 0.032
#> GSM1269683 3 0.576 0.4971 0.060 0.000 0.588 0.276 0.076 0.000
#> GSM1269689 2 0.486 0.5605 0.248 0.672 0.016 0.000 0.060 0.004
#> GSM1269697 2 0.270 0.7180 0.032 0.888 0.056 0.000 0.012 0.012
#> GSM1269705 2 0.285 0.7219 0.004 0.880 0.036 0.000 0.036 0.044
#> GSM1269713 2 0.527 0.4171 0.072 0.604 0.304 0.004 0.016 0.000
#> GSM1269719 3 0.812 0.3753 0.092 0.004 0.440 0.164 0.196 0.104
#> GSM1269725 2 0.506 0.5104 0.060 0.656 0.256 0.004 0.024 0.000
#> GSM1269727 3 0.282 0.6190 0.064 0.004 0.876 0.044 0.012 0.000
#> GSM1269649 3 0.504 0.3593 0.324 0.036 0.612 0.008 0.020 0.000
#> GSM1269657 6 0.517 0.4908 0.032 0.012 0.000 0.032 0.296 0.628
#> GSM1269665 3 0.674 0.5320 0.172 0.004 0.540 0.116 0.168 0.000
#> GSM1269673 4 0.662 0.4140 0.320 0.000 0.156 0.468 0.052 0.004
#> GSM1269681 6 0.406 0.5147 0.000 0.124 0.004 0.000 0.108 0.764
#> GSM1269687 3 0.699 0.0598 0.296 0.000 0.372 0.280 0.048 0.004
#> GSM1269695 1 0.488 0.3340 0.736 0.028 0.148 0.064 0.024 0.000
#> GSM1269703 3 0.683 0.1683 0.296 0.000 0.372 0.288 0.044 0.000
#> GSM1269711 1 0.485 0.4085 0.724 0.064 0.176 0.012 0.020 0.004
#> GSM1269646 2 0.283 0.7048 0.004 0.868 0.092 0.000 0.024 0.012
#> GSM1269654 3 0.671 0.5270 0.028 0.024 0.584 0.176 0.160 0.028
#> GSM1269662 3 0.873 0.1571 0.124 0.004 0.312 0.224 0.216 0.120
#> GSM1269670 2 0.401 0.6790 0.052 0.796 0.000 0.000 0.100 0.052
#> GSM1269678 3 0.437 0.5750 0.156 0.024 0.764 0.036 0.020 0.000
#> GSM1269692 4 0.348 0.5281 0.020 0.000 0.060 0.848 0.044 0.028
#> GSM1269700 3 0.279 0.5956 0.060 0.016 0.884 0.016 0.024 0.000
#> GSM1269708 1 0.654 0.2514 0.488 0.052 0.356 0.084 0.012 0.008
#> GSM1269714 4 0.312 0.5451 0.032 0.000 0.136 0.828 0.000 0.004
#> GSM1269716 4 0.156 0.5378 0.000 0.000 0.056 0.932 0.012 0.000
#> GSM1269720 6 0.655 0.4596 0.056 0.092 0.000 0.048 0.220 0.584
#> GSM1269722 3 0.436 0.6236 0.076 0.020 0.788 0.092 0.020 0.004
#> GSM1269644 4 0.793 0.2119 0.328 0.000 0.164 0.328 0.148 0.032
#> GSM1269652 1 0.734 0.1128 0.420 0.216 0.020 0.004 0.284 0.056
#> GSM1269660 3 0.454 0.6080 0.132 0.008 0.764 0.032 0.060 0.004
#> GSM1269668 3 0.555 0.4192 0.280 0.004 0.608 0.080 0.024 0.004
#> GSM1269676 6 0.376 0.5820 0.008 0.016 0.000 0.000 0.240 0.736
#> GSM1269684 4 0.564 0.5113 0.252 0.000 0.064 0.624 0.052 0.008
#> GSM1269690 4 0.524 0.4639 0.284 0.000 0.008 0.624 0.068 0.016
#> GSM1269698 2 0.449 0.6608 0.028 0.744 0.000 0.000 0.148 0.080
#> GSM1269706 1 0.731 0.1044 0.392 0.200 0.004 0.004 0.312 0.088
#> GSM1269650 6 0.349 0.5442 0.000 0.064 0.000 0.000 0.136 0.800
#> GSM1269658 6 0.565 0.4789 0.016 0.012 0.020 0.052 0.292 0.608
#> GSM1269666 3 0.180 0.6088 0.016 0.020 0.936 0.008 0.020 0.000
#> GSM1269674 2 0.726 0.4734 0.052 0.560 0.180 0.016 0.108 0.084
#> GSM1269682 3 0.634 0.5753 0.128 0.000 0.596 0.132 0.140 0.004
#> GSM1269688 2 0.522 0.4568 0.312 0.600 0.012 0.000 0.072 0.004
#> GSM1269696 2 0.333 0.6890 0.016 0.832 0.124 0.000 0.012 0.016
#> GSM1269704 2 0.200 0.7211 0.004 0.924 0.028 0.000 0.032 0.012
#> GSM1269712 3 0.516 0.6029 0.080 0.060 0.740 0.080 0.040 0.000
#> GSM1269718 3 0.742 0.4696 0.100 0.004 0.516 0.152 0.184 0.044
#> GSM1269724 3 0.529 0.4669 0.084 0.184 0.688 0.008 0.032 0.004
#> GSM1269726 3 0.358 0.6302 0.080 0.000 0.816 0.092 0.012 0.000
#> GSM1269648 1 0.478 0.4114 0.680 0.028 0.256 0.020 0.016 0.000
#> GSM1269656 5 0.824 0.0000 0.272 0.012 0.024 0.132 0.280 0.280
#> GSM1269664 3 0.626 0.5530 0.172 0.004 0.596 0.088 0.140 0.000
#> GSM1269672 4 0.643 0.3781 0.388 0.000 0.128 0.436 0.044 0.004
#> GSM1269680 6 0.194 0.5763 0.000 0.036 0.000 0.004 0.040 0.920
#> GSM1269686 4 0.636 0.3948 0.268 0.000 0.208 0.496 0.024 0.004
#> GSM1269694 1 0.520 0.3259 0.704 0.020 0.176 0.072 0.024 0.004
#> GSM1269702 1 0.653 -0.3185 0.480 0.000 0.084 0.356 0.064 0.016
#> GSM1269710 1 0.543 0.2590 0.580 0.020 0.336 0.048 0.016 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 agent(p) disease.state(p) gender(p) individual(p) k
#> MAD:mclust 35 0.492 0.5448 1.0000 2.82e-02 2
#> MAD:mclust 75 0.794 0.2152 0.1781 1.81e-06 3
#> MAD:mclust 58 0.731 0.3302 0.1060 9.75e-06 4
#> MAD:mclust 45 0.948 0.1123 0.0576 1.57e-04 5
#> MAD:mclust 44 0.787 0.0634 0.0821 2.09e-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["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 51941 rows and 84 columns.
#> Top rows (1000, 2000, 3000, 4000, 5000) are extracted by 'MAD' method.
#> Subgroups are detected by 'NMF' method.
#> Performed in total 1250 partitions by row resampling.
#> Best k for subgroups seems to be 2.
#>
#> Following methods can be applied to this 'ConsensusPartition' object:
#> [1] "cola_report" "collect_classes" "collect_plots"
#> [4] "collect_stats" "colnames" "compare_signatures"
#> [7] "consensus_heatmap" "dimension_reduction" "functional_enrichment"
#> [10] "get_anno_col" "get_anno" "get_classes"
#> [13] "get_consensus" "get_matrix" "get_membership"
#> [16] "get_param" "get_signatures" "get_stats"
#> [19] "is_best_k" "is_stable_k" "membership_heatmap"
#> [22] "ncol" "nrow" "plot_ecdf"
#> [25] "rownames" "select_partition_number" "show"
#> [28] "suggest_best_k" "test_to_known_factors"
collect_plots() function collects all the plots made from res for all k (number of partitions)
into one single page to provide an easy and fast comparison between different k.
collect_plots(res)
The plots are:
k and the heatmap of
predicted classes for each k.k.k.k.All the plots in panels can be made by individual functions and they are plotted later in this section.
select_partition_number() produces several plots showing different
statistics for choosing “optimized” k. There are following statistics:
k;k, the area increased is defined as \(A_k - A_{k-1}\).The detailed explanations of these statistics can be found in the cola vignette.
Generally speaking, lower PAC score, higher mean silhouette score or higher
concordance corresponds to better partition. Rand index and Jaccard index
measure how similar the current partition is compared to partition with k-1.
If they are too similar, we won't accept k is better than k-1.
select_partition_number(res)
The numeric values for all these statistics can be obtained by get_stats().
get_stats(res)
#> k 1-PAC mean_silhouette concordance area_increased Rand Jaccard
#> 2 2 0.204 0.667 0.815 0.4922 0.494 0.494
#> 3 3 0.236 0.477 0.719 0.3333 0.690 0.455
#> 4 4 0.299 0.345 0.609 0.1318 0.790 0.470
#> 5 5 0.351 0.260 0.512 0.0713 0.844 0.489
#> 6 6 0.420 0.237 0.482 0.0453 0.871 0.497
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
#> GSM1269647 2 0.7299 0.728 0.204 0.796
#> GSM1269655 1 0.8909 0.600 0.692 0.308
#> GSM1269663 2 1.0000 -0.104 0.496 0.504
#> GSM1269671 2 0.3431 0.784 0.064 0.936
#> GSM1269679 1 0.3114 0.780 0.944 0.056
#> GSM1269693 1 0.8443 0.627 0.728 0.272
#> GSM1269701 1 0.3114 0.778 0.944 0.056
#> GSM1269709 2 0.9795 0.325 0.416 0.584
#> GSM1269715 1 0.3274 0.771 0.940 0.060
#> GSM1269717 1 0.3114 0.775 0.944 0.056
#> GSM1269721 2 0.2948 0.774 0.052 0.948
#> GSM1269723 1 0.6438 0.734 0.836 0.164
#> GSM1269645 1 0.9044 0.575 0.680 0.320
#> GSM1269653 2 0.7299 0.735 0.204 0.796
#> GSM1269661 1 0.9552 0.366 0.624 0.376
#> GSM1269669 1 0.0672 0.781 0.992 0.008
#> GSM1269677 2 0.2423 0.774 0.040 0.960
#> GSM1269685 1 0.9635 0.452 0.612 0.388
#> GSM1269691 2 0.9552 0.298 0.376 0.624
#> GSM1269699 2 0.3733 0.783 0.072 0.928
#> GSM1269707 2 0.2423 0.786 0.040 0.960
#> GSM1269651 2 0.2603 0.785 0.044 0.956
#> GSM1269659 2 0.3584 0.765 0.068 0.932
#> GSM1269667 1 0.3879 0.775 0.924 0.076
#> GSM1269675 2 0.6801 0.747 0.180 0.820
#> GSM1269683 1 0.3274 0.783 0.940 0.060
#> GSM1269689 2 0.7674 0.713 0.224 0.776
#> GSM1269697 2 0.8081 0.700 0.248 0.752
#> GSM1269705 2 0.6343 0.757 0.160 0.840
#> GSM1269713 2 0.8813 0.619 0.300 0.700
#> GSM1269719 2 0.8608 0.537 0.284 0.716
#> GSM1269725 2 0.9087 0.576 0.324 0.676
#> GSM1269727 1 0.2236 0.781 0.964 0.036
#> GSM1269649 1 0.9522 0.397 0.628 0.372
#> GSM1269657 2 0.3274 0.767 0.060 0.940
#> GSM1269665 1 0.3879 0.789 0.924 0.076
#> GSM1269673 1 0.6247 0.758 0.844 0.156
#> GSM1269681 2 0.2236 0.783 0.036 0.964
#> GSM1269687 1 0.9358 0.501 0.648 0.352
#> GSM1269695 2 0.9661 0.393 0.392 0.608
#> GSM1269703 1 0.8327 0.662 0.736 0.264
#> GSM1269711 1 0.9286 0.474 0.656 0.344
#> GSM1269646 2 0.7815 0.704 0.232 0.768
#> GSM1269654 1 0.9286 0.565 0.656 0.344
#> GSM1269662 2 0.8081 0.662 0.248 0.752
#> GSM1269670 2 0.5842 0.766 0.140 0.860
#> GSM1269678 1 0.1843 0.781 0.972 0.028
#> GSM1269692 1 0.8443 0.622 0.728 0.272
#> GSM1269700 1 0.4022 0.776 0.920 0.080
#> GSM1269708 1 0.8608 0.610 0.716 0.284
#> GSM1269714 1 0.2778 0.775 0.952 0.048
#> GSM1269716 1 0.2778 0.775 0.952 0.048
#> GSM1269720 2 0.2948 0.773 0.052 0.948
#> GSM1269722 1 0.6247 0.751 0.844 0.156
#> GSM1269644 2 0.6712 0.688 0.176 0.824
#> GSM1269652 2 0.3431 0.782 0.064 0.936
#> GSM1269660 1 0.9427 0.452 0.640 0.360
#> GSM1269668 1 0.1633 0.781 0.976 0.024
#> GSM1269676 2 0.2423 0.773 0.040 0.960
#> GSM1269684 1 0.7815 0.673 0.768 0.232
#> GSM1269690 1 0.9580 0.474 0.620 0.380
#> GSM1269698 2 0.3584 0.782 0.068 0.932
#> GSM1269706 2 0.1633 0.782 0.024 0.976
#> GSM1269650 2 0.1843 0.783 0.028 0.972
#> GSM1269658 2 0.3274 0.768 0.060 0.940
#> GSM1269666 1 0.2043 0.780 0.968 0.032
#> GSM1269674 2 0.5178 0.775 0.116 0.884
#> GSM1269682 1 0.2603 0.787 0.956 0.044
#> GSM1269688 2 0.8207 0.688 0.256 0.744
#> GSM1269696 2 0.8909 0.608 0.308 0.692
#> GSM1269704 2 0.5519 0.772 0.128 0.872
#> GSM1269712 1 0.5059 0.771 0.888 0.112
#> GSM1269718 2 0.9580 0.322 0.380 0.620
#> GSM1269724 1 0.6048 0.743 0.852 0.148
#> GSM1269726 1 0.2603 0.786 0.956 0.044
#> GSM1269648 1 0.9970 0.105 0.532 0.468
#> GSM1269656 2 0.3584 0.768 0.068 0.932
#> GSM1269664 1 0.3431 0.788 0.936 0.064
#> GSM1269672 1 0.6148 0.752 0.848 0.152
#> GSM1269680 2 0.0938 0.781 0.012 0.988
#> GSM1269686 1 0.3114 0.774 0.944 0.056
#> GSM1269694 2 0.9393 0.448 0.356 0.644
#> GSM1269702 2 0.7528 0.656 0.216 0.784
#> GSM1269710 1 0.9460 0.411 0.636 0.364
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM1269647 1 0.188 0.6754 0.952 0.044 0.004
#> GSM1269655 1 0.983 0.0132 0.408 0.252 0.340
#> GSM1269663 2 0.944 0.1148 0.180 0.444 0.376
#> GSM1269671 1 0.384 0.6421 0.872 0.116 0.012
#> GSM1269679 3 0.652 0.1695 0.488 0.004 0.508
#> GSM1269693 3 0.608 0.2966 0.000 0.388 0.612
#> GSM1269701 3 0.652 0.1002 0.492 0.004 0.504
#> GSM1269709 1 0.865 0.4438 0.600 0.192 0.208
#> GSM1269715 3 0.141 0.6548 0.000 0.036 0.964
#> GSM1269717 3 0.175 0.6541 0.000 0.048 0.952
#> GSM1269721 2 0.640 0.4795 0.372 0.620 0.008
#> GSM1269723 1 0.641 0.4696 0.700 0.028 0.272
#> GSM1269645 3 0.878 0.1708 0.416 0.112 0.472
#> GSM1269653 1 0.670 0.4715 0.692 0.268 0.040
#> GSM1269661 1 0.806 0.0570 0.532 0.068 0.400
#> GSM1269669 3 0.249 0.6592 0.060 0.008 0.932
#> GSM1269677 2 0.216 0.6914 0.064 0.936 0.000
#> GSM1269685 2 0.709 0.4622 0.056 0.676 0.268
#> GSM1269691 2 0.540 0.6301 0.060 0.816 0.124
#> GSM1269699 1 0.450 0.5780 0.804 0.196 0.000
#> GSM1269707 2 0.724 0.2762 0.432 0.540 0.028
#> GSM1269651 2 0.674 0.2854 0.428 0.560 0.012
#> GSM1269659 2 0.294 0.6925 0.072 0.916 0.012
#> GSM1269667 3 0.652 0.1658 0.484 0.004 0.512
#> GSM1269675 1 0.296 0.6701 0.912 0.080 0.008
#> GSM1269683 3 0.315 0.6625 0.048 0.036 0.916
#> GSM1269689 1 0.257 0.6735 0.936 0.032 0.032
#> GSM1269697 1 0.333 0.6686 0.904 0.076 0.020
#> GSM1269705 1 0.362 0.6521 0.884 0.104 0.012
#> GSM1269713 1 0.158 0.6805 0.964 0.008 0.028
#> GSM1269719 2 0.919 0.3926 0.256 0.536 0.208
#> GSM1269725 1 0.212 0.6772 0.948 0.012 0.040
#> GSM1269727 3 0.619 0.4382 0.364 0.004 0.632
#> GSM1269649 1 0.583 0.5795 0.784 0.052 0.164
#> GSM1269657 2 0.153 0.6898 0.032 0.964 0.004
#> GSM1269665 3 0.618 0.6339 0.156 0.072 0.772
#> GSM1269673 3 0.626 0.5539 0.052 0.196 0.752
#> GSM1269681 1 0.621 0.0922 0.572 0.428 0.000
#> GSM1269687 3 0.909 0.4220 0.176 0.288 0.536
#> GSM1269695 1 0.962 0.2229 0.472 0.252 0.276
#> GSM1269703 3 0.685 0.5796 0.064 0.224 0.712
#> GSM1269711 1 0.834 0.3559 0.592 0.112 0.296
#> GSM1269646 1 0.203 0.6787 0.952 0.032 0.016
#> GSM1269654 3 0.968 0.1584 0.220 0.352 0.428
#> GSM1269662 2 0.771 0.5624 0.196 0.676 0.128
#> GSM1269670 1 0.287 0.6701 0.916 0.076 0.008
#> GSM1269678 3 0.423 0.6351 0.160 0.004 0.836
#> GSM1269692 3 0.648 0.1419 0.004 0.452 0.544
#> GSM1269700 1 0.608 0.3306 0.652 0.004 0.344
#> GSM1269708 3 0.919 0.2435 0.348 0.160 0.492
#> GSM1269714 3 0.212 0.6578 0.012 0.040 0.948
#> GSM1269716 3 0.176 0.6561 0.004 0.040 0.956
#> GSM1269720 2 0.575 0.5579 0.296 0.700 0.004
#> GSM1269722 3 0.738 0.3079 0.404 0.036 0.560
#> GSM1269644 2 0.425 0.6743 0.048 0.872 0.080
#> GSM1269652 2 0.754 0.4590 0.332 0.612 0.056
#> GSM1269660 1 0.861 0.2893 0.568 0.128 0.304
#> GSM1269668 3 0.338 0.6538 0.100 0.008 0.892
#> GSM1269676 2 0.186 0.6911 0.052 0.948 0.000
#> GSM1269684 3 0.623 0.4324 0.012 0.316 0.672
#> GSM1269690 2 0.569 0.5780 0.036 0.780 0.184
#> GSM1269698 1 0.502 0.5165 0.760 0.240 0.000
#> GSM1269706 2 0.716 0.4552 0.332 0.628 0.040
#> GSM1269650 2 0.658 0.3219 0.420 0.572 0.008
#> GSM1269658 2 0.303 0.6846 0.076 0.912 0.012
#> GSM1269666 3 0.599 0.5284 0.304 0.008 0.688
#> GSM1269674 1 0.507 0.5916 0.792 0.196 0.012
#> GSM1269682 3 0.353 0.6581 0.068 0.032 0.900
#> GSM1269688 1 0.379 0.6601 0.892 0.060 0.048
#> GSM1269696 1 0.277 0.6773 0.928 0.048 0.024
#> GSM1269704 1 0.280 0.6573 0.908 0.092 0.000
#> GSM1269712 3 0.808 0.3098 0.408 0.068 0.524
#> GSM1269718 1 0.898 0.1330 0.496 0.368 0.136
#> GSM1269724 1 0.590 0.4273 0.700 0.008 0.292
#> GSM1269726 3 0.599 0.5221 0.304 0.008 0.688
#> GSM1269648 1 0.958 0.2673 0.480 0.252 0.268
#> GSM1269656 2 0.375 0.6777 0.096 0.884 0.020
#> GSM1269664 3 0.706 0.5865 0.236 0.068 0.696
#> GSM1269672 3 0.782 0.3886 0.080 0.300 0.620
#> GSM1269680 2 0.565 0.5333 0.312 0.688 0.000
#> GSM1269686 3 0.393 0.6500 0.028 0.092 0.880
#> GSM1269694 1 0.978 0.1023 0.420 0.336 0.244
#> GSM1269702 2 0.595 0.6408 0.116 0.792 0.092
#> GSM1269710 1 0.936 0.2016 0.484 0.184 0.332
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM1269647 2 0.381 0.5397 0.156 0.824 0.000 0.020
#> GSM1269655 2 0.846 0.2374 0.048 0.488 0.216 0.248
#> GSM1269663 4 0.844 0.3186 0.084 0.148 0.240 0.528
#> GSM1269671 2 0.576 0.4136 0.304 0.644 0.000 0.052
#> GSM1269679 2 0.645 0.2604 0.076 0.544 0.380 0.000
#> GSM1269693 3 0.623 0.0934 0.036 0.008 0.492 0.464
#> GSM1269701 2 0.728 0.3161 0.168 0.508 0.324 0.000
#> GSM1269709 1 0.916 0.0181 0.380 0.356 0.128 0.136
#> GSM1269715 3 0.240 0.6081 0.048 0.000 0.920 0.032
#> GSM1269717 3 0.232 0.6148 0.036 0.004 0.928 0.032
#> GSM1269721 4 0.785 0.2512 0.248 0.240 0.012 0.500
#> GSM1269723 2 0.712 0.5113 0.100 0.648 0.200 0.052
#> GSM1269645 1 0.942 -0.1255 0.340 0.276 0.288 0.096
#> GSM1269653 1 0.683 0.3255 0.552 0.344 0.004 0.100
#> GSM1269661 2 0.894 0.0720 0.236 0.416 0.284 0.064
#> GSM1269669 3 0.460 0.5900 0.132 0.072 0.796 0.000
#> GSM1269677 4 0.334 0.5147 0.128 0.016 0.000 0.856
#> GSM1269685 1 0.750 -0.0911 0.444 0.004 0.156 0.396
#> GSM1269691 4 0.666 0.1386 0.420 0.008 0.064 0.508
#> GSM1269699 1 0.592 0.1242 0.556 0.404 0.000 0.040
#> GSM1269707 1 0.669 0.4631 0.620 0.180 0.000 0.200
#> GSM1269651 4 0.679 0.3914 0.096 0.312 0.008 0.584
#> GSM1269659 4 0.395 0.5067 0.144 0.004 0.024 0.828
#> GSM1269667 2 0.638 0.0377 0.064 0.500 0.436 0.000
#> GSM1269675 2 0.544 0.3590 0.396 0.588 0.004 0.012
#> GSM1269683 3 0.401 0.6236 0.020 0.060 0.856 0.064
#> GSM1269689 2 0.543 0.3872 0.336 0.640 0.004 0.020
#> GSM1269697 2 0.477 0.5395 0.148 0.788 0.004 0.060
#> GSM1269705 2 0.562 0.5131 0.176 0.728 0.004 0.092
#> GSM1269713 2 0.423 0.5552 0.140 0.820 0.032 0.008
#> GSM1269719 4 0.842 0.4187 0.116 0.232 0.112 0.540
#> GSM1269725 2 0.451 0.5571 0.100 0.828 0.036 0.036
#> GSM1269727 3 0.682 0.1937 0.060 0.376 0.544 0.020
#> GSM1269649 2 0.601 0.2951 0.400 0.560 0.036 0.004
#> GSM1269657 4 0.404 0.4899 0.168 0.008 0.012 0.812
#> GSM1269665 3 0.815 0.5134 0.148 0.176 0.580 0.096
#> GSM1269673 3 0.695 0.1197 0.416 0.004 0.484 0.096
#> GSM1269681 4 0.683 0.1942 0.100 0.424 0.000 0.476
#> GSM1269687 1 0.849 -0.1068 0.428 0.076 0.380 0.116
#> GSM1269695 1 0.560 0.4791 0.756 0.144 0.076 0.024
#> GSM1269703 3 0.736 0.4728 0.228 0.056 0.620 0.096
#> GSM1269711 1 0.588 0.3873 0.672 0.248 0.080 0.000
#> GSM1269646 2 0.249 0.5654 0.048 0.920 0.004 0.028
#> GSM1269654 4 0.881 -0.0187 0.040 0.296 0.320 0.344
#> GSM1269662 4 0.725 0.4922 0.116 0.128 0.092 0.664
#> GSM1269670 2 0.468 0.5017 0.232 0.744 0.000 0.024
#> GSM1269678 3 0.571 0.4961 0.064 0.236 0.696 0.004
#> GSM1269692 3 0.677 0.0634 0.080 0.004 0.472 0.444
#> GSM1269700 2 0.629 0.4625 0.092 0.648 0.256 0.004
#> GSM1269708 1 0.982 0.1067 0.316 0.272 0.244 0.168
#> GSM1269714 3 0.357 0.6076 0.080 0.012 0.872 0.036
#> GSM1269716 3 0.203 0.6144 0.028 0.000 0.936 0.036
#> GSM1269720 4 0.632 0.4490 0.168 0.172 0.000 0.660
#> GSM1269722 3 0.768 0.0367 0.072 0.412 0.464 0.052
#> GSM1269644 4 0.669 0.3609 0.320 0.020 0.064 0.596
#> GSM1269652 1 0.609 0.4139 0.688 0.092 0.008 0.212
#> GSM1269660 2 0.895 0.1591 0.188 0.472 0.240 0.100
#> GSM1269668 3 0.517 0.5567 0.168 0.080 0.752 0.000
#> GSM1269676 4 0.395 0.4905 0.172 0.012 0.004 0.812
#> GSM1269684 3 0.719 0.2845 0.292 0.000 0.536 0.172
#> GSM1269690 4 0.728 0.2012 0.348 0.004 0.140 0.508
#> GSM1269698 2 0.681 0.0313 0.404 0.496 0.000 0.100
#> GSM1269706 1 0.641 0.4256 0.656 0.124 0.004 0.216
#> GSM1269650 4 0.681 0.3719 0.104 0.324 0.004 0.568
#> GSM1269658 4 0.308 0.5396 0.064 0.012 0.028 0.896
#> GSM1269666 3 0.576 0.1647 0.012 0.424 0.552 0.012
#> GSM1269674 2 0.704 0.4212 0.196 0.624 0.016 0.164
#> GSM1269682 3 0.478 0.6180 0.040 0.076 0.820 0.064
#> GSM1269688 2 0.613 0.2115 0.420 0.540 0.012 0.028
#> GSM1269696 2 0.377 0.5644 0.080 0.864 0.016 0.040
#> GSM1269704 2 0.433 0.5251 0.176 0.792 0.000 0.032
#> GSM1269712 2 0.766 -0.0888 0.044 0.444 0.432 0.080
#> GSM1269718 2 0.908 -0.0835 0.144 0.392 0.112 0.352
#> GSM1269724 2 0.602 0.5094 0.092 0.696 0.204 0.008
#> GSM1269726 3 0.688 0.3510 0.080 0.308 0.592 0.020
#> GSM1269648 1 0.605 0.4937 0.720 0.160 0.100 0.020
#> GSM1269656 4 0.589 0.2545 0.392 0.012 0.020 0.576
#> GSM1269664 3 0.858 0.4473 0.160 0.228 0.520 0.092
#> GSM1269672 1 0.699 0.2765 0.588 0.016 0.296 0.100
#> GSM1269680 4 0.632 0.4870 0.168 0.172 0.000 0.660
#> GSM1269686 3 0.474 0.5017 0.240 0.008 0.740 0.012
#> GSM1269694 1 0.672 0.4612 0.688 0.148 0.120 0.044
#> GSM1269702 1 0.570 0.1252 0.608 0.000 0.036 0.356
#> GSM1269710 1 0.695 0.4251 0.644 0.196 0.136 0.024
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM1269647 3 0.448 0.4551 0.124 0.084 0.780 0.004 0.008
#> GSM1269655 3 0.865 0.0153 0.064 0.344 0.352 0.176 0.064
#> GSM1269663 2 0.747 0.4436 0.076 0.600 0.056 0.144 0.124
#> GSM1269671 1 0.674 -0.0808 0.492 0.144 0.340 0.000 0.024
#> GSM1269679 3 0.564 0.3660 0.048 0.012 0.624 0.304 0.012
#> GSM1269693 4 0.720 0.0364 0.012 0.308 0.008 0.432 0.240
#> GSM1269701 3 0.699 0.3279 0.188 0.024 0.508 0.276 0.004
#> GSM1269709 3 0.855 0.1896 0.156 0.048 0.432 0.096 0.268
#> GSM1269715 4 0.295 0.5588 0.004 0.020 0.008 0.876 0.092
#> GSM1269717 4 0.317 0.5542 0.000 0.020 0.012 0.856 0.112
#> GSM1269721 3 0.814 -0.1458 0.056 0.308 0.352 0.016 0.268
#> GSM1269723 3 0.730 0.4492 0.112 0.112 0.576 0.188 0.012
#> GSM1269645 1 0.825 0.0630 0.436 0.284 0.084 0.168 0.028
#> GSM1269653 1 0.735 0.2777 0.424 0.024 0.296 0.004 0.252
#> GSM1269661 1 0.915 0.0460 0.324 0.160 0.268 0.204 0.044
#> GSM1269669 4 0.670 0.4665 0.208 0.024 0.100 0.624 0.044
#> GSM1269677 5 0.490 0.0989 0.012 0.396 0.012 0.000 0.580
#> GSM1269685 5 0.545 0.4136 0.136 0.020 0.012 0.108 0.724
#> GSM1269691 5 0.553 0.4765 0.144 0.076 0.000 0.064 0.716
#> GSM1269699 1 0.630 0.2749 0.556 0.016 0.304 0.000 0.124
#> GSM1269707 1 0.700 0.1860 0.436 0.020 0.196 0.000 0.348
#> GSM1269651 2 0.387 0.5352 0.012 0.824 0.088 0.000 0.076
#> GSM1269659 5 0.488 0.0416 0.008 0.412 0.008 0.004 0.568
#> GSM1269667 3 0.803 0.1263 0.176 0.100 0.380 0.340 0.004
#> GSM1269675 1 0.663 -0.0975 0.476 0.120 0.384 0.008 0.012
#> GSM1269683 4 0.456 0.5543 0.048 0.096 0.048 0.800 0.008
#> GSM1269689 3 0.679 0.2620 0.304 0.056 0.556 0.012 0.072
#> GSM1269697 3 0.588 0.4586 0.124 0.096 0.712 0.016 0.052
#> GSM1269705 3 0.621 0.3694 0.204 0.168 0.612 0.004 0.012
#> GSM1269713 3 0.551 0.4580 0.160 0.068 0.724 0.032 0.016
#> GSM1269719 2 0.720 0.4950 0.096 0.636 0.088 0.080 0.100
#> GSM1269725 3 0.298 0.4880 0.012 0.072 0.884 0.024 0.008
#> GSM1269727 4 0.829 0.0255 0.168 0.140 0.276 0.408 0.008
#> GSM1269649 1 0.594 0.1436 0.612 0.052 0.296 0.036 0.004
#> GSM1269657 5 0.462 0.2354 0.020 0.324 0.004 0.000 0.652
#> GSM1269665 4 0.836 0.3477 0.232 0.204 0.080 0.448 0.036
#> GSM1269673 5 0.829 -0.0463 0.264 0.056 0.024 0.320 0.336
#> GSM1269681 2 0.612 0.4064 0.072 0.636 0.232 0.000 0.060
#> GSM1269687 1 0.856 -0.0493 0.392 0.088 0.044 0.312 0.164
#> GSM1269695 1 0.473 0.3770 0.776 0.016 0.036 0.028 0.144
#> GSM1269703 4 0.821 0.2666 0.328 0.132 0.032 0.420 0.088
#> GSM1269711 1 0.769 0.2569 0.472 0.012 0.296 0.068 0.152
#> GSM1269646 3 0.544 0.4554 0.116 0.156 0.704 0.024 0.000
#> GSM1269654 2 0.839 0.1236 0.012 0.376 0.204 0.296 0.112
#> GSM1269662 2 0.692 0.4549 0.092 0.636 0.044 0.060 0.168
#> GSM1269670 1 0.627 -0.1534 0.468 0.152 0.380 0.000 0.000
#> GSM1269678 4 0.653 0.2815 0.092 0.012 0.312 0.560 0.024
#> GSM1269692 4 0.702 0.0385 0.012 0.264 0.000 0.424 0.300
#> GSM1269700 3 0.698 0.4132 0.124 0.084 0.568 0.224 0.000
#> GSM1269708 3 0.911 0.1137 0.128 0.068 0.332 0.156 0.316
#> GSM1269714 4 0.482 0.5442 0.020 0.032 0.036 0.776 0.136
#> GSM1269716 4 0.340 0.5578 0.004 0.020 0.024 0.856 0.096
#> GSM1269720 2 0.699 0.1494 0.024 0.404 0.172 0.000 0.400
#> GSM1269722 3 0.815 0.1526 0.040 0.120 0.436 0.324 0.080
#> GSM1269644 2 0.797 -0.1135 0.240 0.364 0.008 0.060 0.328
#> GSM1269652 5 0.728 -0.1493 0.364 0.016 0.196 0.012 0.412
#> GSM1269660 1 0.915 -0.0174 0.308 0.244 0.252 0.156 0.040
#> GSM1269668 4 0.604 0.4815 0.172 0.008 0.152 0.652 0.016
#> GSM1269676 5 0.499 0.2305 0.028 0.320 0.012 0.000 0.640
#> GSM1269684 4 0.758 0.1637 0.196 0.052 0.004 0.436 0.312
#> GSM1269690 5 0.566 0.4462 0.096 0.072 0.000 0.120 0.712
#> GSM1269698 3 0.754 -0.0264 0.344 0.124 0.436 0.000 0.096
#> GSM1269706 1 0.711 0.1406 0.412 0.028 0.184 0.000 0.376
#> GSM1269650 2 0.449 0.5376 0.032 0.792 0.088 0.000 0.088
#> GSM1269658 2 0.475 0.1190 0.008 0.564 0.000 0.008 0.420
#> GSM1269666 4 0.642 0.0925 0.040 0.064 0.368 0.524 0.004
#> GSM1269674 3 0.761 0.1500 0.304 0.320 0.344 0.020 0.012
#> GSM1269682 4 0.514 0.5387 0.056 0.132 0.056 0.752 0.004
#> GSM1269688 3 0.711 0.2100 0.280 0.044 0.536 0.012 0.128
#> GSM1269696 3 0.634 0.3945 0.224 0.168 0.592 0.012 0.004
#> GSM1269704 3 0.538 0.4048 0.180 0.080 0.708 0.000 0.032
#> GSM1269712 3 0.769 0.0954 0.044 0.104 0.452 0.356 0.044
#> GSM1269718 2 0.830 0.3142 0.172 0.500 0.184 0.088 0.056
#> GSM1269724 3 0.543 0.4690 0.056 0.060 0.740 0.132 0.012
#> GSM1269726 4 0.760 0.3587 0.108 0.120 0.200 0.552 0.020
#> GSM1269648 1 0.668 0.3154 0.592 0.008 0.120 0.040 0.240
#> GSM1269656 5 0.584 0.4375 0.192 0.136 0.012 0.004 0.656
#> GSM1269664 4 0.911 0.2659 0.232 0.188 0.168 0.364 0.048
#> GSM1269672 1 0.717 -0.0215 0.424 0.012 0.008 0.216 0.340
#> GSM1269680 2 0.691 0.4200 0.104 0.580 0.100 0.000 0.216
#> GSM1269686 4 0.680 0.3815 0.236 0.024 0.024 0.588 0.128
#> GSM1269694 1 0.610 0.3679 0.700 0.052 0.048 0.048 0.152
#> GSM1269702 5 0.574 0.1925 0.380 0.020 0.012 0.028 0.560
#> GSM1269710 1 0.588 0.3925 0.700 0.008 0.076 0.064 0.152
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM1269647 3 0.553 0.3548 0.016 0.172 0.652 0.004 0.148 0.008
#> GSM1269655 2 0.765 0.1968 0.048 0.492 0.252 0.112 0.028 0.068
#> GSM1269663 6 0.895 -0.0682 0.076 0.284 0.032 0.140 0.176 0.292
#> GSM1269671 5 0.700 0.4639 0.144 0.168 0.180 0.000 0.504 0.004
#> GSM1269679 3 0.651 0.3362 0.036 0.076 0.584 0.224 0.080 0.000
#> GSM1269693 4 0.746 0.0386 0.040 0.120 0.012 0.416 0.060 0.352
#> GSM1269701 3 0.773 0.2553 0.096 0.056 0.452 0.216 0.180 0.000
#> GSM1269709 3 0.830 0.2595 0.152 0.068 0.480 0.048 0.112 0.140
#> GSM1269715 4 0.401 0.4853 0.056 0.016 0.020 0.828 0.044 0.036
#> GSM1269717 4 0.348 0.4876 0.040 0.036 0.028 0.856 0.004 0.036
#> GSM1269721 6 0.858 0.0320 0.056 0.156 0.292 0.016 0.164 0.316
#> GSM1269723 3 0.764 0.2715 0.016 0.168 0.484 0.124 0.188 0.020
#> GSM1269645 5 0.839 0.0757 0.256 0.200 0.016 0.184 0.320 0.024
#> GSM1269653 3 0.795 -0.0883 0.344 0.044 0.348 0.004 0.136 0.124
#> GSM1269661 2 0.876 0.0535 0.232 0.320 0.128 0.176 0.140 0.004
#> GSM1269669 4 0.693 0.2718 0.284 0.048 0.064 0.512 0.092 0.000
#> GSM1269677 6 0.238 0.4399 0.020 0.080 0.004 0.000 0.004 0.892
#> GSM1269685 6 0.742 0.1116 0.360 0.008 0.056 0.088 0.068 0.420
#> GSM1269691 6 0.599 0.2881 0.324 0.008 0.004 0.084 0.032 0.548
#> GSM1269699 1 0.750 0.0322 0.348 0.068 0.232 0.000 0.328 0.024
#> GSM1269707 1 0.799 0.2597 0.416 0.048 0.228 0.008 0.208 0.092
#> GSM1269651 2 0.572 0.2982 0.000 0.572 0.036 0.000 0.096 0.296
#> GSM1269659 6 0.470 0.4338 0.028 0.100 0.008 0.012 0.088 0.764
#> GSM1269667 4 0.825 0.0774 0.048 0.172 0.196 0.344 0.240 0.000
#> GSM1269675 5 0.658 0.4087 0.120 0.088 0.192 0.000 0.580 0.020
#> GSM1269683 4 0.573 0.4570 0.060 0.124 0.032 0.704 0.064 0.016
#> GSM1269689 3 0.637 0.1812 0.136 0.048 0.540 0.000 0.268 0.008
#> GSM1269697 3 0.643 0.3237 0.044 0.140 0.620 0.004 0.148 0.044
#> GSM1269705 3 0.775 0.0727 0.076 0.244 0.388 0.004 0.256 0.032
#> GSM1269713 3 0.548 0.3550 0.036 0.080 0.688 0.020 0.172 0.004
#> GSM1269719 2 0.826 0.3458 0.120 0.472 0.060 0.068 0.076 0.204
#> GSM1269725 3 0.450 0.4198 0.024 0.132 0.776 0.016 0.036 0.016
#> GSM1269727 4 0.860 0.1228 0.044 0.140 0.216 0.340 0.236 0.024
#> GSM1269649 5 0.729 0.3621 0.236 0.076 0.204 0.016 0.464 0.004
#> GSM1269657 6 0.321 0.4634 0.080 0.048 0.008 0.000 0.012 0.852
#> GSM1269665 4 0.769 0.0620 0.188 0.336 0.020 0.376 0.056 0.024
#> GSM1269673 1 0.644 0.2890 0.612 0.064 0.020 0.216 0.028 0.060
#> GSM1269681 2 0.624 0.3895 0.020 0.608 0.100 0.000 0.068 0.204
#> GSM1269687 1 0.745 0.2576 0.548 0.116 0.040 0.188 0.068 0.040
#> GSM1269695 1 0.501 0.3002 0.676 0.012 0.016 0.028 0.252 0.016
#> GSM1269703 1 0.846 -0.1157 0.340 0.108 0.036 0.336 0.120 0.060
#> GSM1269711 1 0.708 0.0625 0.440 0.016 0.316 0.028 0.184 0.016
#> GSM1269646 3 0.646 0.2623 0.016 0.244 0.524 0.004 0.196 0.016
#> GSM1269654 2 0.851 0.1506 0.024 0.364 0.164 0.260 0.048 0.140
#> GSM1269662 6 0.872 -0.0390 0.068 0.260 0.040 0.076 0.236 0.320
#> GSM1269670 5 0.687 0.4519 0.112 0.228 0.168 0.000 0.492 0.000
#> GSM1269678 4 0.675 0.2261 0.056 0.064 0.320 0.504 0.056 0.000
#> GSM1269692 4 0.729 -0.0588 0.064 0.092 0.000 0.400 0.064 0.380
#> GSM1269700 3 0.764 0.2850 0.056 0.104 0.496 0.176 0.164 0.004
#> GSM1269708 3 0.918 0.1687 0.168 0.056 0.344 0.104 0.120 0.208
#> GSM1269714 4 0.623 0.4430 0.092 0.036 0.048 0.684 0.064 0.076
#> GSM1269716 4 0.387 0.4868 0.036 0.024 0.036 0.836 0.012 0.056
#> GSM1269720 6 0.727 0.2736 0.028 0.116 0.152 0.008 0.160 0.536
#> GSM1269722 3 0.856 0.0600 0.032 0.144 0.344 0.296 0.136 0.048
#> GSM1269644 6 0.844 0.0893 0.288 0.208 0.016 0.092 0.064 0.332
#> GSM1269652 1 0.746 0.2189 0.424 0.012 0.284 0.004 0.108 0.168
#> GSM1269660 2 0.890 0.1164 0.144 0.384 0.140 0.152 0.152 0.028
#> GSM1269668 4 0.600 0.3917 0.204 0.024 0.120 0.620 0.032 0.000
#> GSM1269676 6 0.339 0.4564 0.068 0.064 0.012 0.000 0.012 0.844
#> GSM1269684 1 0.723 0.0192 0.396 0.052 0.008 0.376 0.024 0.144
#> GSM1269690 6 0.658 0.3130 0.260 0.020 0.004 0.140 0.036 0.540
#> GSM1269698 3 0.818 0.0149 0.232 0.192 0.320 0.000 0.224 0.032
#> GSM1269706 1 0.809 0.2443 0.388 0.032 0.248 0.008 0.176 0.148
#> GSM1269650 2 0.618 0.3374 0.012 0.556 0.052 0.000 0.088 0.292
#> GSM1269658 6 0.536 0.3384 0.028 0.188 0.004 0.028 0.060 0.692
#> GSM1269666 4 0.749 0.1488 0.016 0.168 0.276 0.432 0.104 0.004
#> GSM1269674 5 0.724 0.3776 0.040 0.236 0.156 0.016 0.512 0.040
#> GSM1269682 4 0.612 0.4391 0.052 0.128 0.048 0.672 0.088 0.012
#> GSM1269688 3 0.688 0.1908 0.176 0.032 0.496 0.008 0.268 0.020
#> GSM1269696 3 0.671 0.1513 0.052 0.264 0.480 0.000 0.200 0.004
#> GSM1269704 3 0.621 0.2602 0.036 0.104 0.576 0.004 0.264 0.016
#> GSM1269712 3 0.803 0.1587 0.024 0.188 0.428 0.244 0.068 0.048
#> GSM1269718 2 0.910 0.1400 0.140 0.364 0.112 0.052 0.200 0.132
#> GSM1269724 3 0.660 0.3701 0.040 0.184 0.604 0.084 0.084 0.004
#> GSM1269726 4 0.828 0.2461 0.040 0.108 0.176 0.440 0.196 0.040
#> GSM1269648 1 0.530 0.3963 0.732 0.012 0.076 0.024 0.112 0.044
#> GSM1269656 6 0.548 0.3255 0.296 0.028 0.028 0.004 0.028 0.616
#> GSM1269664 4 0.829 0.1048 0.228 0.284 0.064 0.340 0.068 0.016
#> GSM1269672 1 0.554 0.4071 0.668 0.008 0.004 0.192 0.044 0.084
#> GSM1269680 2 0.681 0.1326 0.084 0.444 0.044 0.000 0.048 0.380
#> GSM1269686 4 0.650 0.0971 0.408 0.032 0.040 0.460 0.044 0.016
#> GSM1269694 1 0.618 0.1695 0.552 0.040 0.012 0.052 0.324 0.020
#> GSM1269702 1 0.566 0.0928 0.572 0.024 0.012 0.016 0.036 0.340
#> GSM1269710 1 0.580 0.3607 0.696 0.028 0.060 0.040 0.144 0.032
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 agent(p) disease.state(p) gender(p) individual(p) k
#> MAD:NMF 70 0.995 0.473 0.748 0.00453 2
#> MAD:NMF 46 0.305 0.597 0.017 0.00344 3
#> MAD:NMF 24 0.878 0.498 0.067 0.01072 4
#> MAD:NMF 8 1.000 0.414 NA 0.04601 5
#> MAD:NMF 0 NA NA NA 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["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 51941 rows and 84 columns.
#> Top rows (1000, 2000, 3000, 4000, 5000) are extracted by 'ATC' method.
#> Subgroups are detected by 'hclust' method.
#> Performed in total 1250 partitions by row resampling.
#> Best k for subgroups seems to be 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 0.282 0.770 0.827 0.4015 0.633 0.633
#> 3 3 0.366 0.577 0.748 0.5379 0.682 0.514
#> 4 4 0.486 0.595 0.756 0.1687 0.881 0.681
#> 5 5 0.549 0.512 0.658 0.0630 0.954 0.839
#> 6 6 0.592 0.466 0.615 0.0397 0.962 0.853
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
#> GSM1269647 2 0.0376 0.87436 0.004 0.996
#> GSM1269655 1 0.4022 0.82978 0.920 0.080
#> GSM1269663 1 0.9909 0.41239 0.556 0.444
#> GSM1269671 2 0.8443 0.51011 0.272 0.728
#> GSM1269679 1 0.8909 0.71193 0.692 0.308
#> GSM1269693 1 0.6343 0.84392 0.840 0.160
#> GSM1269701 1 0.5408 0.85015 0.876 0.124
#> GSM1269709 1 0.0376 0.83562 0.996 0.004
#> GSM1269715 1 0.5519 0.84764 0.872 0.128
#> GSM1269717 1 0.7139 0.83524 0.804 0.196
#> GSM1269721 1 0.6048 0.84064 0.852 0.148
#> GSM1269723 1 0.6148 0.84672 0.848 0.152
#> GSM1269645 2 0.0376 0.87436 0.004 0.996
#> GSM1269653 1 0.6438 0.84452 0.836 0.164
#> GSM1269661 2 0.0376 0.87436 0.004 0.996
#> GSM1269669 1 0.6712 0.83359 0.824 0.176
#> GSM1269677 1 0.9970 0.09111 0.532 0.468
#> GSM1269685 1 0.0376 0.83562 0.996 0.004
#> GSM1269691 1 0.0376 0.83562 0.996 0.004
#> GSM1269699 1 0.0376 0.83800 0.996 0.004
#> GSM1269707 1 0.5519 0.84764 0.872 0.128
#> GSM1269651 2 0.0938 0.87231 0.012 0.988
#> GSM1269659 1 0.6048 0.84064 0.852 0.148
#> GSM1269667 2 0.3733 0.84566 0.072 0.928
#> GSM1269675 1 0.8207 0.78239 0.744 0.256
#> GSM1269683 1 0.8955 0.70756 0.688 0.312
#> GSM1269689 1 0.5408 0.84478 0.876 0.124
#> GSM1269697 1 0.5842 0.80062 0.860 0.140
#> GSM1269705 1 0.0376 0.83800 0.996 0.004
#> GSM1269713 1 0.8443 0.76307 0.728 0.272
#> GSM1269719 1 0.9129 0.50371 0.672 0.328
#> GSM1269725 1 0.6048 0.79407 0.852 0.148
#> GSM1269727 1 0.5946 0.84441 0.856 0.144
#> GSM1269649 2 0.3733 0.84566 0.072 0.928
#> GSM1269657 1 0.1414 0.83570 0.980 0.020
#> GSM1269665 2 0.0376 0.87436 0.004 0.996
#> GSM1269673 1 0.6712 0.83359 0.824 0.176
#> GSM1269681 2 0.9996 0.04871 0.488 0.512
#> GSM1269687 1 0.2236 0.83928 0.964 0.036
#> GSM1269695 1 0.0672 0.83890 0.992 0.008
#> GSM1269703 1 0.5946 0.84933 0.856 0.144
#> GSM1269711 1 0.7883 0.79964 0.764 0.236
#> GSM1269646 2 0.0376 0.87436 0.004 0.996
#> GSM1269654 1 0.4022 0.82978 0.920 0.080
#> GSM1269662 2 0.0376 0.87436 0.004 0.996
#> GSM1269670 2 0.0376 0.87436 0.004 0.996
#> GSM1269678 1 0.8909 0.71193 0.692 0.308
#> GSM1269692 1 0.6343 0.84392 0.840 0.160
#> GSM1269700 1 0.5408 0.85015 0.876 0.124
#> GSM1269708 1 0.0376 0.83562 0.996 0.004
#> GSM1269714 1 0.5294 0.85145 0.880 0.120
#> GSM1269716 1 0.7139 0.83524 0.804 0.196
#> GSM1269720 1 0.6048 0.84064 0.852 0.148
#> GSM1269722 1 0.7453 0.81231 0.788 0.212
#> GSM1269644 2 0.0938 0.87231 0.012 0.988
#> GSM1269652 1 0.0376 0.83562 0.996 0.004
#> GSM1269660 2 0.0376 0.87436 0.004 0.996
#> GSM1269668 1 0.4815 0.85553 0.896 0.104
#> GSM1269676 1 0.9970 0.09111 0.532 0.468
#> GSM1269684 1 0.4022 0.83092 0.920 0.080
#> GSM1269690 1 0.0376 0.83562 0.996 0.004
#> GSM1269698 1 0.0376 0.83800 0.996 0.004
#> GSM1269706 1 0.5519 0.84764 0.872 0.128
#> GSM1269650 2 0.0938 0.87231 0.012 0.988
#> GSM1269658 1 0.6247 0.83970 0.844 0.156
#> GSM1269666 2 0.3733 0.84566 0.072 0.928
#> GSM1269674 1 0.8207 0.78239 0.744 0.256
#> GSM1269682 1 0.8955 0.70756 0.688 0.312
#> GSM1269688 1 0.5408 0.84478 0.876 0.124
#> GSM1269696 1 0.5842 0.80062 0.860 0.140
#> GSM1269704 1 0.0376 0.83800 0.996 0.004
#> GSM1269712 2 0.9850 -0.00524 0.428 0.572
#> GSM1269718 1 0.9129 0.50371 0.672 0.328
#> GSM1269724 1 0.6048 0.79407 0.852 0.148
#> GSM1269726 1 0.5946 0.84441 0.856 0.144
#> GSM1269648 2 0.3733 0.84566 0.072 0.928
#> GSM1269656 1 0.1414 0.83570 0.980 0.020
#> GSM1269664 2 0.0376 0.87436 0.004 0.996
#> GSM1269672 1 0.4815 0.85553 0.896 0.104
#> GSM1269680 2 0.9998 0.03126 0.492 0.508
#> GSM1269686 1 0.2236 0.83928 0.964 0.036
#> GSM1269694 1 0.0672 0.83890 0.992 0.008
#> GSM1269702 1 0.0000 0.83705 1.000 0.000
#> GSM1269710 1 0.7883 0.79964 0.764 0.236
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM1269647 2 0.0000 0.920 0.000 1.000 0.000
#> GSM1269655 1 0.3755 0.659 0.872 0.008 0.120
#> GSM1269663 3 0.8734 0.507 0.168 0.248 0.584
#> GSM1269671 2 0.8028 0.452 0.176 0.656 0.168
#> GSM1269679 3 0.8892 0.357 0.436 0.120 0.444
#> GSM1269693 3 0.4291 0.680 0.180 0.000 0.820
#> GSM1269701 1 0.6505 -0.123 0.528 0.004 0.468
#> GSM1269709 1 0.4974 0.643 0.764 0.000 0.236
#> GSM1269715 3 0.3619 0.700 0.136 0.000 0.864
#> GSM1269717 1 0.6308 -0.238 0.508 0.000 0.492
#> GSM1269721 3 0.1860 0.675 0.052 0.000 0.948
#> GSM1269723 3 0.6099 0.607 0.228 0.032 0.740
#> GSM1269645 2 0.0000 0.920 0.000 1.000 0.000
#> GSM1269653 3 0.6633 0.389 0.444 0.008 0.548
#> GSM1269661 2 0.0000 0.920 0.000 1.000 0.000
#> GSM1269669 1 0.7481 0.336 0.596 0.048 0.356
#> GSM1269677 1 0.8089 0.321 0.600 0.308 0.092
#> GSM1269685 1 0.4974 0.643 0.764 0.000 0.236
#> GSM1269691 1 0.4974 0.643 0.764 0.000 0.236
#> GSM1269699 1 0.3816 0.668 0.852 0.000 0.148
#> GSM1269707 3 0.3619 0.700 0.136 0.000 0.864
#> GSM1269651 2 0.0424 0.919 0.008 0.992 0.000
#> GSM1269659 3 0.2711 0.667 0.088 0.000 0.912
#> GSM1269667 2 0.2261 0.885 0.068 0.932 0.000
#> GSM1269675 3 0.7112 0.637 0.260 0.060 0.680
#> GSM1269683 3 0.8884 0.384 0.420 0.120 0.460
#> GSM1269689 3 0.2796 0.680 0.092 0.000 0.908
#> GSM1269697 1 0.3587 0.600 0.892 0.020 0.088
#> GSM1269705 1 0.3816 0.668 0.852 0.000 0.148
#> GSM1269713 1 0.8304 -0.174 0.504 0.080 0.416
#> GSM1269719 1 0.6746 0.445 0.732 0.192 0.076
#> GSM1269725 1 0.3415 0.596 0.900 0.020 0.080
#> GSM1269727 3 0.3532 0.698 0.108 0.008 0.884
#> GSM1269649 2 0.2261 0.885 0.068 0.932 0.000
#> GSM1269657 1 0.4605 0.658 0.796 0.000 0.204
#> GSM1269665 2 0.0000 0.920 0.000 1.000 0.000
#> GSM1269673 1 0.7481 0.336 0.596 0.048 0.356
#> GSM1269681 1 0.8314 0.235 0.556 0.352 0.092
#> GSM1269687 1 0.4861 0.656 0.800 0.008 0.192
#> GSM1269695 1 0.3941 0.659 0.844 0.000 0.156
#> GSM1269703 3 0.6460 0.356 0.440 0.004 0.556
#> GSM1269711 3 0.6546 0.673 0.240 0.044 0.716
#> GSM1269646 2 0.0000 0.920 0.000 1.000 0.000
#> GSM1269654 1 0.3755 0.659 0.872 0.008 0.120
#> GSM1269662 2 0.0000 0.920 0.000 1.000 0.000
#> GSM1269670 2 0.0237 0.919 0.004 0.996 0.000
#> GSM1269678 3 0.8892 0.357 0.436 0.120 0.444
#> GSM1269692 3 0.4291 0.680 0.180 0.000 0.820
#> GSM1269700 1 0.6505 -0.123 0.528 0.004 0.468
#> GSM1269708 1 0.4974 0.643 0.764 0.000 0.236
#> GSM1269714 3 0.6267 0.220 0.452 0.000 0.548
#> GSM1269716 1 0.6308 -0.238 0.508 0.000 0.492
#> GSM1269720 3 0.1860 0.675 0.052 0.000 0.948
#> GSM1269722 3 0.8131 0.304 0.376 0.076 0.548
#> GSM1269644 2 0.0424 0.919 0.008 0.992 0.000
#> GSM1269652 1 0.4974 0.643 0.764 0.000 0.236
#> GSM1269660 2 0.0000 0.920 0.000 1.000 0.000
#> GSM1269668 1 0.6527 0.503 0.660 0.020 0.320
#> GSM1269676 1 0.8089 0.321 0.600 0.308 0.092
#> GSM1269684 1 0.5450 0.610 0.760 0.012 0.228
#> GSM1269690 1 0.4974 0.643 0.764 0.000 0.236
#> GSM1269698 1 0.3816 0.668 0.852 0.000 0.148
#> GSM1269706 3 0.3619 0.700 0.136 0.000 0.864
#> GSM1269650 2 0.0424 0.919 0.008 0.992 0.000
#> GSM1269658 3 0.2866 0.687 0.076 0.008 0.916
#> GSM1269666 2 0.2261 0.885 0.068 0.932 0.000
#> GSM1269674 3 0.7112 0.637 0.260 0.060 0.680
#> GSM1269682 3 0.8884 0.384 0.420 0.120 0.460
#> GSM1269688 3 0.4555 0.640 0.200 0.000 0.800
#> GSM1269696 1 0.3587 0.600 0.892 0.020 0.088
#> GSM1269704 1 0.3816 0.668 0.852 0.000 0.148
#> GSM1269712 2 0.9666 -0.129 0.316 0.452 0.232
#> GSM1269718 1 0.6746 0.445 0.732 0.192 0.076
#> GSM1269724 1 0.3415 0.596 0.900 0.020 0.080
#> GSM1269726 3 0.3532 0.698 0.108 0.008 0.884
#> GSM1269648 2 0.2261 0.885 0.068 0.932 0.000
#> GSM1269656 1 0.4605 0.658 0.796 0.000 0.204
#> GSM1269664 2 0.0000 0.920 0.000 1.000 0.000
#> GSM1269672 1 0.6527 0.503 0.660 0.020 0.320
#> GSM1269680 1 0.8297 0.246 0.560 0.348 0.092
#> GSM1269686 1 0.4861 0.656 0.800 0.008 0.192
#> GSM1269694 1 0.3941 0.659 0.844 0.000 0.156
#> GSM1269702 1 0.4555 0.657 0.800 0.000 0.200
#> GSM1269710 3 0.6546 0.673 0.240 0.044 0.716
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM1269647 2 0.1389 0.8810 0.000 0.952 0.000 0.048
#> GSM1269655 1 0.5108 0.5689 0.672 0.000 0.020 0.308
#> GSM1269663 3 0.7480 0.4767 0.028 0.192 0.596 0.184
#> GSM1269671 2 0.6751 0.2467 0.004 0.624 0.152 0.220
#> GSM1269679 3 0.7819 0.3165 0.048 0.088 0.452 0.412
#> GSM1269693 3 0.5836 0.6417 0.112 0.000 0.700 0.188
#> GSM1269701 1 0.7823 -0.0129 0.376 0.000 0.368 0.256
#> GSM1269709 1 0.0000 0.7108 1.000 0.000 0.000 0.000
#> GSM1269715 3 0.2644 0.6841 0.032 0.000 0.908 0.060
#> GSM1269717 3 0.6826 0.3193 0.100 0.000 0.484 0.416
#> GSM1269721 3 0.3081 0.6575 0.048 0.000 0.888 0.064
#> GSM1269723 3 0.3925 0.6255 0.000 0.016 0.808 0.176
#> GSM1269645 2 0.0000 0.8850 0.000 1.000 0.000 0.000
#> GSM1269653 3 0.7272 0.4514 0.156 0.004 0.536 0.304
#> GSM1269661 2 0.0000 0.8850 0.000 1.000 0.000 0.000
#> GSM1269669 1 0.8447 0.1907 0.460 0.036 0.264 0.240
#> GSM1269677 4 0.4719 0.6281 0.008 0.224 0.016 0.752
#> GSM1269685 1 0.0000 0.7108 1.000 0.000 0.000 0.000
#> GSM1269691 1 0.0000 0.7108 1.000 0.000 0.000 0.000
#> GSM1269699 1 0.3681 0.7038 0.816 0.000 0.008 0.176
#> GSM1269707 3 0.2644 0.6841 0.032 0.000 0.908 0.060
#> GSM1269651 2 0.1557 0.8800 0.000 0.944 0.000 0.056
#> GSM1269659 3 0.4389 0.6339 0.116 0.000 0.812 0.072
#> GSM1269667 2 0.2704 0.8533 0.000 0.876 0.000 0.124
#> GSM1269675 3 0.6138 0.6089 0.032 0.040 0.676 0.252
#> GSM1269683 3 0.7179 0.3642 0.016 0.088 0.492 0.404
#> GSM1269689 3 0.0707 0.6554 0.000 0.000 0.980 0.020
#> GSM1269697 4 0.3958 0.6077 0.112 0.000 0.052 0.836
#> GSM1269705 1 0.3681 0.7038 0.816 0.000 0.008 0.176
#> GSM1269713 4 0.6796 -0.2859 0.012 0.064 0.448 0.476
#> GSM1269719 4 0.7229 0.4274 0.276 0.120 0.020 0.584
#> GSM1269725 4 0.3840 0.6141 0.104 0.000 0.052 0.844
#> GSM1269727 3 0.1489 0.6748 0.000 0.004 0.952 0.044
#> GSM1269649 2 0.2704 0.8533 0.000 0.876 0.000 0.124
#> GSM1269657 1 0.2011 0.7171 0.920 0.000 0.000 0.080
#> GSM1269665 2 0.0000 0.8850 0.000 1.000 0.000 0.000
#> GSM1269673 1 0.8447 0.1907 0.460 0.036 0.264 0.240
#> GSM1269681 4 0.5065 0.5696 0.008 0.268 0.016 0.708
#> GSM1269687 1 0.5417 0.6421 0.704 0.000 0.056 0.240
#> GSM1269695 1 0.4900 0.6513 0.732 0.000 0.032 0.236
#> GSM1269703 3 0.7269 0.4233 0.180 0.000 0.524 0.296
#> GSM1269711 3 0.4894 0.6471 0.008 0.024 0.748 0.220
#> GSM1269646 2 0.0592 0.8853 0.000 0.984 0.000 0.016
#> GSM1269654 1 0.5108 0.5689 0.672 0.000 0.020 0.308
#> GSM1269662 2 0.0000 0.8850 0.000 1.000 0.000 0.000
#> GSM1269670 2 0.0817 0.8763 0.000 0.976 0.000 0.024
#> GSM1269678 3 0.7819 0.3165 0.048 0.088 0.452 0.412
#> GSM1269692 3 0.5836 0.6417 0.112 0.000 0.700 0.188
#> GSM1269700 1 0.7823 -0.0129 0.376 0.000 0.368 0.256
#> GSM1269708 1 0.0000 0.7108 1.000 0.000 0.000 0.000
#> GSM1269714 3 0.6968 0.2572 0.392 0.000 0.492 0.116
#> GSM1269716 3 0.6826 0.3193 0.100 0.000 0.484 0.416
#> GSM1269720 3 0.3081 0.6575 0.048 0.000 0.888 0.064
#> GSM1269722 3 0.6308 0.3860 0.004 0.060 0.580 0.356
#> GSM1269644 2 0.1716 0.8769 0.000 0.936 0.000 0.064
#> GSM1269652 1 0.0188 0.7122 0.996 0.000 0.000 0.004
#> GSM1269660 2 0.0000 0.8850 0.000 1.000 0.000 0.000
#> GSM1269668 1 0.6717 0.5294 0.652 0.012 0.164 0.172
#> GSM1269676 4 0.4719 0.6281 0.008 0.224 0.016 0.752
#> GSM1269684 1 0.6773 0.4468 0.588 0.000 0.136 0.276
#> GSM1269690 1 0.0000 0.7108 1.000 0.000 0.000 0.000
#> GSM1269698 1 0.3681 0.7038 0.816 0.000 0.008 0.176
#> GSM1269706 3 0.2644 0.6841 0.032 0.000 0.908 0.060
#> GSM1269650 2 0.1557 0.8800 0.000 0.944 0.000 0.056
#> GSM1269658 3 0.4017 0.6669 0.044 0.000 0.828 0.128
#> GSM1269666 2 0.2760 0.8524 0.000 0.872 0.000 0.128
#> GSM1269674 3 0.6138 0.6089 0.032 0.040 0.676 0.252
#> GSM1269682 3 0.7179 0.3642 0.016 0.088 0.492 0.404
#> GSM1269688 3 0.4035 0.6147 0.176 0.000 0.804 0.020
#> GSM1269696 4 0.3958 0.6077 0.112 0.000 0.052 0.836
#> GSM1269704 1 0.3681 0.7038 0.816 0.000 0.008 0.176
#> GSM1269712 2 0.7773 -0.1807 0.000 0.432 0.284 0.284
#> GSM1269718 4 0.7229 0.4274 0.276 0.120 0.020 0.584
#> GSM1269724 4 0.3840 0.6141 0.104 0.000 0.052 0.844
#> GSM1269726 3 0.1489 0.6748 0.000 0.004 0.952 0.044
#> GSM1269648 2 0.2760 0.8524 0.000 0.872 0.000 0.128
#> GSM1269656 1 0.2011 0.7171 0.920 0.000 0.000 0.080
#> GSM1269664 2 0.0000 0.8850 0.000 1.000 0.000 0.000
#> GSM1269672 1 0.6717 0.5294 0.652 0.012 0.164 0.172
#> GSM1269680 4 0.5037 0.5770 0.008 0.264 0.016 0.712
#> GSM1269686 1 0.5417 0.6421 0.704 0.000 0.056 0.240
#> GSM1269694 1 0.4900 0.6513 0.732 0.000 0.032 0.236
#> GSM1269702 1 0.1557 0.7183 0.944 0.000 0.000 0.056
#> GSM1269710 3 0.4894 0.6471 0.008 0.024 0.748 0.220
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM1269647 2 0.1568 0.8554 0.000 0.944 0.020 0.000 0.036
#> GSM1269655 1 0.5273 0.5512 0.636 0.000 0.304 0.012 0.048
#> GSM1269663 4 0.7978 0.2513 0.024 0.152 0.116 0.516 0.192
#> GSM1269671 2 0.6860 0.3403 0.000 0.600 0.148 0.156 0.096
#> GSM1269679 4 0.7670 0.3792 0.032 0.020 0.300 0.444 0.204
#> GSM1269693 4 0.6333 0.3172 0.072 0.000 0.156 0.648 0.124
#> GSM1269701 3 0.8202 -0.0546 0.312 0.000 0.328 0.248 0.112
#> GSM1269709 1 0.0771 0.7058 0.976 0.000 0.000 0.004 0.020
#> GSM1269715 4 0.1806 0.4049 0.016 0.000 0.028 0.940 0.016
#> GSM1269717 4 0.7041 0.3065 0.088 0.000 0.344 0.488 0.080
#> GSM1269721 4 0.3543 0.2699 0.024 0.000 0.012 0.828 0.136
#> GSM1269723 4 0.3966 0.4204 0.000 0.012 0.132 0.808 0.048
#> GSM1269645 2 0.1270 0.8534 0.000 0.948 0.000 0.000 0.052
#> GSM1269653 4 0.7430 0.2501 0.080 0.000 0.340 0.448 0.132
#> GSM1269661 2 0.1270 0.8534 0.000 0.948 0.000 0.000 0.052
#> GSM1269669 1 0.7948 0.1405 0.444 0.012 0.228 0.248 0.068
#> GSM1269677 3 0.6583 0.4522 0.004 0.124 0.532 0.020 0.320
#> GSM1269685 1 0.0865 0.7046 0.972 0.000 0.000 0.004 0.024
#> GSM1269691 1 0.0771 0.7058 0.976 0.000 0.000 0.004 0.020
#> GSM1269699 1 0.3741 0.6711 0.732 0.000 0.264 0.000 0.004
#> GSM1269707 4 0.1806 0.4049 0.016 0.000 0.028 0.940 0.016
#> GSM1269651 2 0.1818 0.8541 0.000 0.932 0.024 0.000 0.044
#> GSM1269659 4 0.5143 0.0860 0.084 0.000 0.008 0.696 0.212
#> GSM1269667 2 0.3105 0.8329 0.000 0.864 0.088 0.004 0.044
#> GSM1269675 4 0.6619 0.4308 0.024 0.008 0.156 0.588 0.224
#> GSM1269683 4 0.7162 0.4234 0.012 0.020 0.268 0.500 0.200
#> GSM1269689 5 0.4192 0.6591 0.000 0.000 0.000 0.404 0.596
#> GSM1269697 3 0.1885 0.4851 0.020 0.000 0.932 0.044 0.004
#> GSM1269705 1 0.3741 0.6711 0.732 0.000 0.264 0.000 0.004
#> GSM1269713 4 0.6752 0.2863 0.008 0.020 0.408 0.452 0.112
#> GSM1269719 3 0.7927 0.3551 0.256 0.036 0.416 0.024 0.268
#> GSM1269725 3 0.2100 0.4926 0.016 0.000 0.924 0.048 0.012
#> GSM1269727 4 0.2522 0.3701 0.000 0.000 0.012 0.880 0.108
#> GSM1269649 2 0.3105 0.8329 0.000 0.864 0.088 0.004 0.044
#> GSM1269657 1 0.2270 0.7131 0.908 0.000 0.072 0.004 0.016
#> GSM1269665 2 0.1270 0.8534 0.000 0.948 0.000 0.000 0.052
#> GSM1269673 1 0.7948 0.1405 0.444 0.012 0.228 0.248 0.068
#> GSM1269681 3 0.6810 0.4376 0.004 0.168 0.504 0.016 0.308
#> GSM1269687 1 0.5447 0.6258 0.664 0.000 0.256 0.048 0.032
#> GSM1269695 1 0.4491 0.5981 0.648 0.000 0.336 0.012 0.004
#> GSM1269703 4 0.7608 0.2191 0.104 0.000 0.344 0.428 0.124
#> GSM1269711 4 0.5116 0.4505 0.000 0.000 0.120 0.692 0.188
#> GSM1269646 2 0.0693 0.8592 0.000 0.980 0.012 0.000 0.008
#> GSM1269654 1 0.5273 0.5512 0.636 0.000 0.304 0.012 0.048
#> GSM1269662 2 0.1270 0.8534 0.000 0.948 0.000 0.000 0.052
#> GSM1269670 2 0.1710 0.8504 0.000 0.940 0.016 0.004 0.040
#> GSM1269678 4 0.7670 0.3792 0.032 0.020 0.300 0.444 0.204
#> GSM1269692 4 0.6333 0.3172 0.072 0.000 0.156 0.648 0.124
#> GSM1269700 3 0.8202 -0.0546 0.312 0.000 0.328 0.248 0.112
#> GSM1269708 1 0.0771 0.7058 0.976 0.000 0.000 0.004 0.020
#> GSM1269714 4 0.6206 0.1033 0.392 0.000 0.084 0.504 0.020
#> GSM1269716 4 0.7041 0.3065 0.088 0.000 0.344 0.488 0.080
#> GSM1269720 4 0.3543 0.2699 0.024 0.000 0.012 0.828 0.136
#> GSM1269722 4 0.6280 0.3573 0.000 0.020 0.288 0.572 0.120
#> GSM1269644 2 0.1981 0.8514 0.000 0.924 0.028 0.000 0.048
#> GSM1269652 1 0.0486 0.7109 0.988 0.000 0.004 0.004 0.004
#> GSM1269660 2 0.1270 0.8534 0.000 0.948 0.000 0.000 0.052
#> GSM1269668 1 0.6408 0.4905 0.632 0.008 0.172 0.156 0.032
#> GSM1269676 3 0.6583 0.4522 0.004 0.124 0.532 0.020 0.320
#> GSM1269684 1 0.7147 0.3868 0.564 0.000 0.188 0.148 0.100
#> GSM1269690 1 0.0771 0.7058 0.976 0.000 0.000 0.004 0.020
#> GSM1269698 1 0.3741 0.6711 0.732 0.000 0.264 0.000 0.004
#> GSM1269706 4 0.1806 0.4049 0.016 0.000 0.028 0.940 0.016
#> GSM1269650 2 0.1818 0.8541 0.000 0.932 0.024 0.000 0.044
#> GSM1269658 4 0.4688 0.2569 0.024 0.000 0.024 0.720 0.232
#> GSM1269666 2 0.3178 0.8318 0.000 0.860 0.088 0.004 0.048
#> GSM1269674 4 0.6619 0.4308 0.024 0.008 0.156 0.588 0.224
#> GSM1269682 4 0.7162 0.4234 0.012 0.020 0.268 0.500 0.200
#> GSM1269688 5 0.5770 0.6934 0.140 0.000 0.000 0.256 0.604
#> GSM1269696 3 0.1885 0.4851 0.020 0.000 0.932 0.044 0.004
#> GSM1269704 1 0.3741 0.6711 0.732 0.000 0.264 0.000 0.004
#> GSM1269712 2 0.8241 -0.1857 0.000 0.360 0.232 0.280 0.128
#> GSM1269718 3 0.7927 0.3551 0.256 0.036 0.416 0.024 0.268
#> GSM1269724 3 0.2100 0.4926 0.016 0.000 0.924 0.048 0.012
#> GSM1269726 4 0.2522 0.3701 0.000 0.000 0.012 0.880 0.108
#> GSM1269648 2 0.3178 0.8318 0.000 0.860 0.088 0.004 0.048
#> GSM1269656 1 0.2270 0.7131 0.908 0.000 0.072 0.004 0.016
#> GSM1269664 2 0.1270 0.8534 0.000 0.948 0.000 0.000 0.052
#> GSM1269672 1 0.6440 0.4898 0.628 0.008 0.176 0.156 0.032
#> GSM1269680 3 0.6782 0.4399 0.004 0.164 0.508 0.016 0.308
#> GSM1269686 1 0.5447 0.6258 0.664 0.000 0.256 0.048 0.032
#> GSM1269694 1 0.4491 0.5981 0.648 0.000 0.336 0.012 0.004
#> GSM1269702 1 0.2352 0.7155 0.896 0.000 0.092 0.004 0.008
#> GSM1269710 4 0.5116 0.4505 0.000 0.000 0.120 0.692 0.188
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM1269647 2 0.2937 0.7877 0.000 0.848 0.000 0.000 0.056 0.096
#> GSM1269655 1 0.5573 0.4961 0.536 0.000 0.288 0.000 0.000 0.176
#> GSM1269663 4 0.7823 0.2669 0.004 0.068 0.108 0.464 0.128 0.228
#> GSM1269671 2 0.5781 0.3415 0.000 0.588 0.028 0.148 0.000 0.236
#> GSM1269679 6 0.6547 -0.1508 0.024 0.008 0.140 0.408 0.008 0.412
#> GSM1269693 4 0.6159 0.3209 0.060 0.000 0.352 0.496 0.000 0.092
#> GSM1269701 3 0.7072 0.2824 0.268 0.000 0.424 0.216 0.092 0.000
#> GSM1269709 1 0.0000 0.6492 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM1269715 4 0.1917 0.5238 0.016 0.000 0.036 0.928 0.016 0.004
#> GSM1269717 4 0.6769 0.1875 0.068 0.000 0.220 0.472 0.000 0.240
#> GSM1269721 4 0.5230 0.3341 0.024 0.000 0.192 0.692 0.028 0.064
#> GSM1269723 4 0.4539 0.4748 0.000 0.008 0.048 0.768 0.076 0.100
#> GSM1269645 2 0.1950 0.7736 0.000 0.912 0.064 0.000 0.024 0.000
#> GSM1269653 4 0.6929 -0.0897 0.044 0.000 0.384 0.428 0.092 0.052
#> GSM1269661 2 0.1890 0.7748 0.000 0.916 0.060 0.000 0.024 0.000
#> GSM1269669 1 0.7538 0.0466 0.420 0.008 0.196 0.192 0.000 0.184
#> GSM1269677 6 0.1152 0.5480 0.000 0.044 0.000 0.004 0.000 0.952
#> GSM1269685 1 0.0146 0.6479 0.996 0.000 0.000 0.000 0.004 0.000
#> GSM1269691 1 0.0000 0.6492 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM1269699 1 0.3659 0.5849 0.636 0.000 0.364 0.000 0.000 0.000
#> GSM1269707 4 0.1917 0.5238 0.016 0.000 0.036 0.928 0.016 0.004
#> GSM1269651 2 0.3211 0.7836 0.000 0.824 0.000 0.000 0.056 0.120
#> GSM1269659 4 0.7280 0.1801 0.084 0.000 0.220 0.520 0.108 0.068
#> GSM1269667 2 0.4279 0.7660 0.000 0.756 0.024 0.000 0.064 0.156
#> GSM1269675 4 0.6617 0.3321 0.004 0.004 0.128 0.528 0.072 0.264
#> GSM1269683 4 0.6002 0.1050 0.004 0.008 0.108 0.472 0.012 0.396
#> GSM1269689 5 0.1663 0.7704 0.000 0.000 0.000 0.088 0.912 0.000
#> GSM1269697 3 0.4184 0.5007 0.000 0.000 0.576 0.016 0.000 0.408
#> GSM1269705 1 0.3659 0.5849 0.636 0.000 0.364 0.000 0.000 0.000
#> GSM1269713 4 0.6164 0.1393 0.000 0.008 0.204 0.468 0.004 0.316
#> GSM1269719 6 0.5147 0.3089 0.176 0.000 0.180 0.004 0.000 0.640
#> GSM1269725 3 0.4294 0.4810 0.000 0.000 0.552 0.020 0.000 0.428
#> GSM1269727 4 0.2653 0.4914 0.000 0.000 0.012 0.844 0.144 0.000
#> GSM1269649 2 0.4223 0.7655 0.000 0.760 0.024 0.000 0.060 0.156
#> GSM1269657 1 0.2706 0.6590 0.860 0.000 0.104 0.000 0.000 0.036
#> GSM1269665 2 0.1950 0.7736 0.000 0.912 0.064 0.000 0.024 0.000
#> GSM1269673 1 0.7538 0.0466 0.420 0.008 0.196 0.192 0.000 0.184
#> GSM1269681 6 0.1610 0.5356 0.000 0.084 0.000 0.000 0.000 0.916
#> GSM1269687 1 0.5370 0.5484 0.564 0.000 0.324 0.008 0.000 0.104
#> GSM1269695 1 0.3833 0.4790 0.556 0.000 0.444 0.000 0.000 0.000
#> GSM1269703 3 0.6827 0.0350 0.068 0.000 0.412 0.404 0.092 0.024
#> GSM1269711 4 0.4925 0.4447 0.004 0.000 0.024 0.692 0.072 0.208
#> GSM1269646 2 0.1655 0.7920 0.000 0.932 0.008 0.000 0.008 0.052
#> GSM1269654 1 0.5573 0.4961 0.536 0.000 0.288 0.000 0.000 0.176
#> GSM1269662 2 0.1950 0.7736 0.000 0.912 0.064 0.000 0.024 0.000
#> GSM1269670 2 0.1636 0.7781 0.000 0.936 0.024 0.004 0.000 0.036
#> GSM1269678 6 0.6547 -0.1508 0.024 0.008 0.140 0.408 0.008 0.412
#> GSM1269692 4 0.6159 0.3209 0.060 0.000 0.352 0.496 0.000 0.092
#> GSM1269700 3 0.7072 0.2824 0.268 0.000 0.424 0.216 0.092 0.000
#> GSM1269708 1 0.0000 0.6492 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM1269714 4 0.6299 0.1210 0.364 0.000 0.112 0.480 0.012 0.032
#> GSM1269716 4 0.6769 0.1875 0.068 0.000 0.220 0.472 0.000 0.240
#> GSM1269720 4 0.5230 0.3341 0.024 0.000 0.192 0.692 0.028 0.064
#> GSM1269722 4 0.6084 0.2618 0.000 0.008 0.124 0.568 0.036 0.264
#> GSM1269644 2 0.3295 0.7806 0.000 0.816 0.000 0.000 0.056 0.128
#> GSM1269652 1 0.0547 0.6530 0.980 0.000 0.020 0.000 0.000 0.000
#> GSM1269660 2 0.1890 0.7748 0.000 0.916 0.060 0.000 0.024 0.000
#> GSM1269668 1 0.6164 0.3667 0.600 0.008 0.208 0.116 0.000 0.068
#> GSM1269676 6 0.1152 0.5480 0.000 0.044 0.000 0.004 0.000 0.952
#> GSM1269684 1 0.7183 0.3693 0.484 0.000 0.180 0.112 0.012 0.212
#> GSM1269690 1 0.0000 0.6492 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM1269698 1 0.3659 0.5849 0.636 0.000 0.364 0.000 0.000 0.000
#> GSM1269706 4 0.1917 0.5238 0.016 0.000 0.036 0.928 0.016 0.004
#> GSM1269650 2 0.3211 0.7836 0.000 0.824 0.000 0.000 0.056 0.120
#> GSM1269658 4 0.6805 0.2906 0.024 0.000 0.224 0.540 0.068 0.144
#> GSM1269666 2 0.4122 0.7628 0.000 0.764 0.020 0.000 0.056 0.160
#> GSM1269674 4 0.6617 0.3321 0.004 0.004 0.128 0.528 0.072 0.264
#> GSM1269682 4 0.6002 0.1050 0.004 0.008 0.108 0.472 0.012 0.396
#> GSM1269688 5 0.2859 0.7647 0.156 0.000 0.000 0.016 0.828 0.000
#> GSM1269696 3 0.4184 0.5007 0.000 0.000 0.576 0.016 0.000 0.408
#> GSM1269704 1 0.3659 0.5849 0.636 0.000 0.364 0.000 0.000 0.000
#> GSM1269712 2 0.7353 -0.2604 0.000 0.356 0.092 0.292 0.004 0.256
#> GSM1269718 6 0.5147 0.3089 0.176 0.000 0.180 0.004 0.000 0.640
#> GSM1269724 3 0.4294 0.4810 0.000 0.000 0.552 0.020 0.000 0.428
#> GSM1269726 4 0.2653 0.4914 0.000 0.000 0.012 0.844 0.144 0.000
#> GSM1269648 2 0.4122 0.7628 0.000 0.764 0.020 0.000 0.056 0.160
#> GSM1269656 1 0.2706 0.6590 0.860 0.000 0.104 0.000 0.000 0.036
#> GSM1269664 2 0.1950 0.7736 0.000 0.912 0.064 0.000 0.024 0.000
#> GSM1269672 1 0.6187 0.3656 0.596 0.008 0.212 0.116 0.000 0.068
#> GSM1269680 6 0.1556 0.5366 0.000 0.080 0.000 0.000 0.000 0.920
#> GSM1269686 1 0.5370 0.5484 0.564 0.000 0.324 0.008 0.000 0.104
#> GSM1269694 1 0.3833 0.4790 0.556 0.000 0.444 0.000 0.000 0.000
#> GSM1269702 1 0.2697 0.6546 0.812 0.000 0.188 0.000 0.000 0.000
#> GSM1269710 4 0.4925 0.4447 0.004 0.000 0.024 0.692 0.072 0.208
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 agent(p) disease.state(p) gender(p) individual(p) k
#> ATC:hclust 78 1.000 1.000 0.7695 2.08e-04 2
#> ATC:hclust 61 0.550 0.971 0.0248 7.88e-05 3
#> ATC:hclust 63 0.859 0.597 0.0244 9.49e-07 4
#> ATC:hclust 37 0.971 0.363 0.1370 8.59e-04 5
#> ATC:hclust 42 0.990 0.216 0.0633 1.56e-07 6
If matrix rows can be associated to genes, consider to use functional_enrichment(res,
...) to perform function enrichment for the signature genes. See this vignette for more detailed explanations.
The object with results only for a single top-value method and a single partition method can be extracted as:
res = res_list["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 51941 rows and 84 columns.
#> Top rows (1000, 2000, 3000, 4000, 5000) are extracted by 'ATC' method.
#> Subgroups are detected by 'kmeans' method.
#> Performed in total 1250 partitions by row resampling.
#> Best k for subgroups seems to be 2.
#>
#> Following methods can be applied to this 'ConsensusPartition' object:
#> [1] "cola_report" "collect_classes" "collect_plots"
#> [4] "collect_stats" "colnames" "compare_signatures"
#> [7] "consensus_heatmap" "dimension_reduction" "functional_enrichment"
#> [10] "get_anno_col" "get_anno" "get_classes"
#> [13] "get_consensus" "get_matrix" "get_membership"
#> [16] "get_param" "get_signatures" "get_stats"
#> [19] "is_best_k" "is_stable_k" "membership_heatmap"
#> [22] "ncol" "nrow" "plot_ecdf"
#> [25] "rownames" "select_partition_number" "show"
#> [28] "suggest_best_k" "test_to_known_factors"
collect_plots() function collects all the plots made from res for all k (number of partitions)
into one single page to provide an easy and fast comparison between different k.
collect_plots(res)
The plots are:
k and the heatmap of
predicted classes for each k.k.k.k.All the plots in panels can be made by individual functions and they are plotted later in this section.
select_partition_number() produces several plots showing different
statistics for choosing “optimized” k. There are following statistics:
k;k, the area increased is defined as \(A_k - A_{k-1}\).The detailed explanations of these statistics can be found in the cola vignette.
Generally speaking, lower PAC score, higher mean silhouette score or higher
concordance corresponds to better partition. Rand index and Jaccard index
measure how similar the current partition is compared to partition with k-1.
If they are too similar, we won't accept k is better than k-1.
select_partition_number(res)
The numeric values for all these statistics can be obtained by get_stats().
get_stats(res)
#> k 1-PAC mean_silhouette concordance area_increased Rand Jaccard
#> 2 2 0.901 0.911 0.960 0.4615 0.535 0.535
#> 3 3 0.789 0.800 0.917 0.4295 0.715 0.504
#> 4 4 0.656 0.729 0.840 0.1326 0.863 0.620
#> 5 5 0.622 0.563 0.717 0.0637 0.918 0.698
#> 6 6 0.651 0.470 0.665 0.0410 0.961 0.823
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
#> GSM1269647 2 0.0000 0.945 0.000 1.000
#> GSM1269655 1 0.0672 0.964 0.992 0.008
#> GSM1269663 2 0.0672 0.940 0.008 0.992
#> GSM1269671 2 0.0000 0.945 0.000 1.000
#> GSM1269679 1 0.2778 0.941 0.952 0.048
#> GSM1269693 1 0.2423 0.947 0.960 0.040
#> GSM1269701 1 0.0000 0.962 1.000 0.000
#> GSM1269709 1 0.0000 0.962 1.000 0.000
#> GSM1269715 1 0.2423 0.947 0.960 0.040
#> GSM1269717 1 0.3114 0.940 0.944 0.056
#> GSM1269721 1 0.2423 0.947 0.960 0.040
#> GSM1269723 2 0.0672 0.940 0.008 0.992
#> GSM1269645 2 0.0672 0.940 0.008 0.992
#> GSM1269653 1 0.0000 0.962 1.000 0.000
#> GSM1269661 2 0.0000 0.945 0.000 1.000
#> GSM1269669 1 0.0376 0.962 0.996 0.004
#> GSM1269677 2 0.0000 0.945 0.000 1.000
#> GSM1269685 1 0.0000 0.962 1.000 0.000
#> GSM1269691 1 0.0000 0.962 1.000 0.000
#> GSM1269699 1 0.0672 0.964 0.992 0.008
#> GSM1269707 1 0.0000 0.962 1.000 0.000
#> GSM1269651 2 0.0000 0.945 0.000 1.000
#> GSM1269659 1 0.2423 0.947 0.960 0.040
#> GSM1269667 2 0.0000 0.945 0.000 1.000
#> GSM1269675 1 0.2423 0.947 0.960 0.040
#> GSM1269683 1 0.9954 0.123 0.540 0.460
#> GSM1269689 1 0.2423 0.947 0.960 0.040
#> GSM1269697 1 0.0672 0.964 0.992 0.008
#> GSM1269705 1 0.0672 0.964 0.992 0.008
#> GSM1269713 1 0.0376 0.962 0.996 0.004
#> GSM1269719 2 0.7453 0.726 0.212 0.788
#> GSM1269725 1 0.0672 0.964 0.992 0.008
#> GSM1269727 1 0.2948 0.937 0.948 0.052
#> GSM1269649 2 0.0000 0.945 0.000 1.000
#> GSM1269657 1 0.0672 0.964 0.992 0.008
#> GSM1269665 2 0.0000 0.945 0.000 1.000
#> GSM1269673 1 0.2423 0.947 0.960 0.040
#> GSM1269681 2 0.0000 0.945 0.000 1.000
#> GSM1269687 1 0.0938 0.963 0.988 0.012
#> GSM1269695 1 0.0000 0.962 1.000 0.000
#> GSM1269703 1 0.2423 0.947 0.960 0.040
#> GSM1269711 1 0.2423 0.947 0.960 0.040
#> GSM1269646 2 0.0000 0.945 0.000 1.000
#> GSM1269654 1 0.0672 0.964 0.992 0.008
#> GSM1269662 2 0.0000 0.945 0.000 1.000
#> GSM1269670 2 0.0000 0.945 0.000 1.000
#> GSM1269678 1 0.0672 0.964 0.992 0.008
#> GSM1269692 1 0.0672 0.964 0.992 0.008
#> GSM1269700 1 0.0000 0.962 1.000 0.000
#> GSM1269708 1 0.0672 0.964 0.992 0.008
#> GSM1269714 1 0.0000 0.962 1.000 0.000
#> GSM1269716 1 0.0672 0.964 0.992 0.008
#> GSM1269720 1 0.8327 0.646 0.736 0.264
#> GSM1269722 1 0.5294 0.873 0.880 0.120
#> GSM1269644 2 0.0000 0.945 0.000 1.000
#> GSM1269652 1 0.0672 0.964 0.992 0.008
#> GSM1269660 2 0.0000 0.945 0.000 1.000
#> GSM1269668 1 0.0672 0.964 0.992 0.008
#> GSM1269676 2 0.9933 0.168 0.452 0.548
#> GSM1269684 1 0.0000 0.962 1.000 0.000
#> GSM1269690 1 0.0672 0.964 0.992 0.008
#> GSM1269698 1 0.0672 0.964 0.992 0.008
#> GSM1269706 1 0.0000 0.962 1.000 0.000
#> GSM1269650 2 0.0000 0.945 0.000 1.000
#> GSM1269658 1 0.8555 0.615 0.720 0.280
#> GSM1269666 2 0.0000 0.945 0.000 1.000
#> GSM1269674 1 0.2423 0.947 0.960 0.040
#> GSM1269682 2 0.0000 0.945 0.000 1.000
#> GSM1269688 1 0.0000 0.962 1.000 0.000
#> GSM1269696 1 0.0672 0.964 0.992 0.008
#> GSM1269704 1 0.0672 0.964 0.992 0.008
#> GSM1269712 2 0.0000 0.945 0.000 1.000
#> GSM1269718 2 0.7453 0.726 0.212 0.788
#> GSM1269724 2 0.9833 0.315 0.424 0.576
#> GSM1269726 2 0.4161 0.884 0.084 0.916
#> GSM1269648 2 0.0000 0.945 0.000 1.000
#> GSM1269656 1 0.0672 0.964 0.992 0.008
#> GSM1269664 2 0.0000 0.945 0.000 1.000
#> GSM1269672 1 0.0672 0.964 0.992 0.008
#> GSM1269680 2 0.0000 0.945 0.000 1.000
#> GSM1269686 1 0.0672 0.964 0.992 0.008
#> GSM1269694 1 0.0672 0.964 0.992 0.008
#> GSM1269702 1 0.0672 0.964 0.992 0.008
#> GSM1269710 2 0.5059 0.857 0.112 0.888
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM1269647 2 0.0592 0.8820 0.000 0.988 0.012
#> GSM1269655 1 0.0237 0.9111 0.996 0.000 0.004
#> GSM1269663 2 0.0237 0.8814 0.000 0.996 0.004
#> GSM1269671 2 0.6225 0.2859 0.000 0.568 0.432
#> GSM1269679 3 0.0829 0.9246 0.012 0.004 0.984
#> GSM1269693 3 0.0592 0.9242 0.012 0.000 0.988
#> GSM1269701 1 0.3619 0.8091 0.864 0.000 0.136
#> GSM1269709 1 0.0000 0.9121 1.000 0.000 0.000
#> GSM1269715 3 0.0592 0.9242 0.012 0.000 0.988
#> GSM1269717 3 0.0592 0.9242 0.012 0.000 0.988
#> GSM1269721 3 0.0592 0.9242 0.012 0.000 0.988
#> GSM1269723 3 0.0424 0.9186 0.000 0.008 0.992
#> GSM1269645 2 0.0237 0.8820 0.000 0.996 0.004
#> GSM1269653 3 0.0747 0.9235 0.016 0.000 0.984
#> GSM1269661 2 0.0424 0.8818 0.000 0.992 0.008
#> GSM1269669 3 0.4931 0.7058 0.212 0.004 0.784
#> GSM1269677 2 0.6309 0.0929 0.000 0.504 0.496
#> GSM1269685 1 0.0000 0.9121 1.000 0.000 0.000
#> GSM1269691 1 0.0000 0.9121 1.000 0.000 0.000
#> GSM1269699 1 0.0000 0.9121 1.000 0.000 0.000
#> GSM1269707 3 0.0592 0.9242 0.012 0.000 0.988
#> GSM1269651 2 0.0237 0.8814 0.000 0.996 0.004
#> GSM1269659 3 0.1163 0.9149 0.028 0.000 0.972
#> GSM1269667 2 0.0424 0.8818 0.000 0.992 0.008
#> GSM1269675 3 0.0829 0.9246 0.012 0.004 0.984
#> GSM1269683 3 0.0424 0.9185 0.000 0.008 0.992
#> GSM1269689 3 0.0829 0.9246 0.012 0.004 0.984
#> GSM1269697 1 0.0424 0.9086 0.992 0.000 0.008
#> GSM1269705 1 0.0000 0.9121 1.000 0.000 0.000
#> GSM1269713 3 0.0829 0.9246 0.012 0.004 0.984
#> GSM1269719 2 0.8382 0.1700 0.084 0.492 0.424
#> GSM1269725 1 0.1163 0.8990 0.972 0.000 0.028
#> GSM1269727 3 0.0661 0.9233 0.008 0.004 0.988
#> GSM1269649 2 0.0592 0.8820 0.000 0.988 0.012
#> GSM1269657 1 0.0237 0.9111 0.996 0.000 0.004
#> GSM1269665 2 0.0237 0.8820 0.000 0.996 0.004
#> GSM1269673 3 0.3030 0.8551 0.092 0.004 0.904
#> GSM1269681 2 0.0000 0.8817 0.000 1.000 0.000
#> GSM1269687 1 0.6505 0.1544 0.528 0.004 0.468
#> GSM1269695 1 0.0000 0.9121 1.000 0.000 0.000
#> GSM1269703 3 0.0829 0.9246 0.012 0.004 0.984
#> GSM1269711 3 0.0829 0.9246 0.012 0.004 0.984
#> GSM1269646 2 0.0424 0.8818 0.000 0.992 0.008
#> GSM1269654 1 0.0237 0.9111 0.996 0.000 0.004
#> GSM1269662 2 0.0237 0.8820 0.000 0.996 0.004
#> GSM1269670 2 0.0424 0.8818 0.000 0.992 0.008
#> GSM1269678 1 0.4842 0.6903 0.776 0.000 0.224
#> GSM1269692 1 0.1163 0.8993 0.972 0.000 0.028
#> GSM1269700 1 0.4702 0.7349 0.788 0.000 0.212
#> GSM1269708 1 0.0000 0.9121 1.000 0.000 0.000
#> GSM1269714 1 0.1163 0.8993 0.972 0.000 0.028
#> GSM1269716 1 0.6140 0.3349 0.596 0.000 0.404
#> GSM1269720 3 0.0475 0.9219 0.004 0.004 0.992
#> GSM1269722 3 0.0237 0.9220 0.004 0.000 0.996
#> GSM1269644 2 0.0237 0.8814 0.000 0.996 0.004
#> GSM1269652 1 0.0000 0.9121 1.000 0.000 0.000
#> GSM1269660 2 0.0424 0.8818 0.000 0.992 0.008
#> GSM1269668 1 0.0000 0.9121 1.000 0.000 0.000
#> GSM1269676 3 0.9773 0.0439 0.240 0.340 0.420
#> GSM1269684 1 0.5733 0.5208 0.676 0.000 0.324
#> GSM1269690 1 0.0000 0.9121 1.000 0.000 0.000
#> GSM1269698 1 0.0000 0.9121 1.000 0.000 0.000
#> GSM1269706 3 0.2066 0.8889 0.060 0.000 0.940
#> GSM1269650 2 0.0237 0.8814 0.000 0.996 0.004
#> GSM1269658 3 0.0848 0.9174 0.008 0.008 0.984
#> GSM1269666 2 0.0424 0.8816 0.000 0.992 0.008
#> GSM1269674 3 0.4733 0.7326 0.196 0.004 0.800
#> GSM1269682 3 0.6305 -0.1006 0.000 0.484 0.516
#> GSM1269688 1 0.3482 0.8150 0.872 0.000 0.128
#> GSM1269696 1 0.2796 0.8527 0.908 0.000 0.092
#> GSM1269704 1 0.0000 0.9121 1.000 0.000 0.000
#> GSM1269712 2 0.6291 0.1851 0.000 0.532 0.468
#> GSM1269718 2 0.8701 0.1941 0.108 0.492 0.400
#> GSM1269724 1 0.8097 0.2610 0.540 0.072 0.388
#> GSM1269726 3 0.0424 0.9186 0.000 0.008 0.992
#> GSM1269648 2 0.0424 0.8816 0.000 0.992 0.008
#> GSM1269656 1 0.0237 0.9111 0.996 0.000 0.004
#> GSM1269664 2 0.0237 0.8820 0.000 0.996 0.004
#> GSM1269672 1 0.0000 0.9121 1.000 0.000 0.000
#> GSM1269680 2 0.0424 0.8792 0.000 0.992 0.008
#> GSM1269686 1 0.0000 0.9121 1.000 0.000 0.000
#> GSM1269694 1 0.0000 0.9121 1.000 0.000 0.000
#> GSM1269702 1 0.0000 0.9121 1.000 0.000 0.000
#> GSM1269710 3 0.0424 0.9186 0.000 0.008 0.992
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM1269647 2 0.1716 0.96192 0.000 0.936 0.000 0.064
#> GSM1269655 1 0.3311 0.76904 0.828 0.000 0.000 0.172
#> GSM1269663 2 0.1792 0.96007 0.000 0.932 0.000 0.068
#> GSM1269671 4 0.7034 0.51108 0.000 0.220 0.204 0.576
#> GSM1269679 3 0.4222 0.58742 0.000 0.000 0.728 0.272
#> GSM1269693 3 0.2589 0.74551 0.000 0.000 0.884 0.116
#> GSM1269701 1 0.5174 0.69656 0.756 0.000 0.152 0.092
#> GSM1269709 1 0.1743 0.84095 0.940 0.000 0.004 0.056
#> GSM1269715 3 0.2011 0.75763 0.000 0.000 0.920 0.080
#> GSM1269717 4 0.4134 0.55562 0.000 0.000 0.260 0.740
#> GSM1269721 3 0.2647 0.74451 0.000 0.000 0.880 0.120
#> GSM1269723 3 0.2408 0.72945 0.000 0.000 0.896 0.104
#> GSM1269645 2 0.0188 0.96338 0.000 0.996 0.000 0.004
#> GSM1269653 3 0.3486 0.68964 0.000 0.000 0.812 0.188
#> GSM1269661 2 0.0469 0.96292 0.000 0.988 0.000 0.012
#> GSM1269669 3 0.6317 0.50836 0.116 0.000 0.644 0.240
#> GSM1269677 4 0.4352 0.66481 0.000 0.080 0.104 0.816
#> GSM1269685 1 0.1118 0.84962 0.964 0.000 0.000 0.036
#> GSM1269691 1 0.1022 0.84939 0.968 0.000 0.000 0.032
#> GSM1269699 1 0.0921 0.85158 0.972 0.000 0.000 0.028
#> GSM1269707 3 0.2081 0.75762 0.000 0.000 0.916 0.084
#> GSM1269651 2 0.1637 0.96054 0.000 0.940 0.000 0.060
#> GSM1269659 3 0.2868 0.73876 0.000 0.000 0.864 0.136
#> GSM1269667 2 0.1716 0.96120 0.000 0.936 0.000 0.064
#> GSM1269675 3 0.1022 0.76416 0.000 0.000 0.968 0.032
#> GSM1269683 3 0.0336 0.76671 0.000 0.000 0.992 0.008
#> GSM1269689 3 0.1022 0.76352 0.000 0.000 0.968 0.032
#> GSM1269697 1 0.2530 0.81658 0.896 0.000 0.004 0.100
#> GSM1269705 1 0.0921 0.85158 0.972 0.000 0.000 0.028
#> GSM1269713 3 0.4994 -0.08395 0.000 0.000 0.520 0.480
#> GSM1269719 4 0.5689 0.70519 0.040 0.108 0.088 0.764
#> GSM1269725 4 0.5496 0.54281 0.312 0.000 0.036 0.652
#> GSM1269727 3 0.0188 0.76625 0.000 0.000 0.996 0.004
#> GSM1269649 2 0.1792 0.96044 0.000 0.932 0.000 0.068
#> GSM1269657 1 0.4040 0.70987 0.752 0.000 0.000 0.248
#> GSM1269665 2 0.0336 0.96278 0.000 0.992 0.000 0.008
#> GSM1269673 3 0.6179 0.51695 0.076 0.004 0.644 0.276
#> GSM1269681 2 0.1637 0.93022 0.000 0.940 0.000 0.060
#> GSM1269687 4 0.5672 0.61399 0.100 0.000 0.188 0.712
#> GSM1269695 1 0.1557 0.84522 0.944 0.000 0.000 0.056
#> GSM1269703 3 0.3356 0.68514 0.000 0.000 0.824 0.176
#> GSM1269711 3 0.1557 0.76009 0.000 0.000 0.944 0.056
#> GSM1269646 2 0.0469 0.96292 0.000 0.988 0.000 0.012
#> GSM1269654 1 0.4877 0.35183 0.592 0.000 0.000 0.408
#> GSM1269662 2 0.0336 0.96278 0.000 0.992 0.000 0.008
#> GSM1269670 2 0.0469 0.96292 0.000 0.988 0.000 0.012
#> GSM1269678 4 0.5219 0.64492 0.244 0.000 0.044 0.712
#> GSM1269692 1 0.4382 0.66225 0.704 0.000 0.000 0.296
#> GSM1269700 1 0.6469 0.53089 0.644 0.000 0.192 0.164
#> GSM1269708 1 0.0817 0.85262 0.976 0.000 0.000 0.024
#> GSM1269714 1 0.4250 0.68356 0.724 0.000 0.000 0.276
#> GSM1269716 4 0.4322 0.67258 0.152 0.000 0.044 0.804
#> GSM1269720 3 0.2647 0.74303 0.000 0.000 0.880 0.120
#> GSM1269722 4 0.4843 0.38672 0.000 0.000 0.396 0.604
#> GSM1269644 2 0.1940 0.95745 0.000 0.924 0.000 0.076
#> GSM1269652 1 0.0817 0.85262 0.976 0.000 0.000 0.024
#> GSM1269660 2 0.0469 0.96292 0.000 0.988 0.000 0.012
#> GSM1269668 1 0.1118 0.85034 0.964 0.000 0.000 0.036
#> GSM1269676 4 0.4319 0.68189 0.096 0.020 0.048 0.836
#> GSM1269684 1 0.6867 0.32870 0.508 0.000 0.108 0.384
#> GSM1269690 1 0.0469 0.85370 0.988 0.000 0.000 0.012
#> GSM1269698 1 0.1389 0.85060 0.952 0.000 0.000 0.048
#> GSM1269706 4 0.5853 0.00372 0.032 0.000 0.460 0.508
#> GSM1269650 2 0.1637 0.96054 0.000 0.940 0.000 0.060
#> GSM1269658 3 0.3831 0.67763 0.000 0.004 0.792 0.204
#> GSM1269666 2 0.1940 0.95910 0.000 0.924 0.000 0.076
#> GSM1269674 3 0.6759 0.36016 0.108 0.000 0.548 0.344
#> GSM1269682 4 0.6198 0.62942 0.000 0.116 0.224 0.660
#> GSM1269688 1 0.4415 0.73211 0.804 0.000 0.140 0.056
#> GSM1269696 4 0.5557 0.55372 0.308 0.000 0.040 0.652
#> GSM1269704 1 0.1389 0.85060 0.952 0.000 0.000 0.048
#> GSM1269712 4 0.6586 0.61925 0.000 0.216 0.156 0.628
#> GSM1269718 4 0.5795 0.70668 0.052 0.112 0.076 0.760
#> GSM1269724 4 0.5644 0.69160 0.144 0.012 0.100 0.744
#> GSM1269726 3 0.2216 0.73509 0.000 0.000 0.908 0.092
#> GSM1269648 2 0.1940 0.95910 0.000 0.924 0.000 0.076
#> GSM1269656 1 0.3356 0.76609 0.824 0.000 0.000 0.176
#> GSM1269664 2 0.0336 0.96278 0.000 0.992 0.000 0.008
#> GSM1269672 1 0.1637 0.84654 0.940 0.000 0.000 0.060
#> GSM1269680 4 0.4914 0.47674 0.012 0.312 0.000 0.676
#> GSM1269686 1 0.1716 0.84046 0.936 0.000 0.000 0.064
#> GSM1269694 1 0.0921 0.85158 0.972 0.000 0.000 0.028
#> GSM1269702 1 0.0707 0.85283 0.980 0.000 0.000 0.020
#> GSM1269710 3 0.4994 -0.02826 0.000 0.000 0.520 0.480
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM1269647 2 0.3724 0.8648 0.204 0.776 0.020 0.000 0.000
#> GSM1269655 5 0.4797 0.5662 0.044 0.000 0.296 0.000 0.660
#> GSM1269663 2 0.4180 0.8521 0.220 0.744 0.036 0.000 0.000
#> GSM1269671 3 0.7144 0.2714 0.300 0.132 0.504 0.064 0.000
#> GSM1269679 1 0.6483 0.0945 0.452 0.000 0.192 0.356 0.000
#> GSM1269693 4 0.1893 0.6530 0.024 0.000 0.048 0.928 0.000
#> GSM1269701 1 0.5427 -0.1798 0.480 0.000 0.020 0.024 0.476
#> GSM1269709 5 0.2719 0.6683 0.144 0.000 0.000 0.004 0.852
#> GSM1269715 4 0.0963 0.6746 0.036 0.000 0.000 0.964 0.000
#> GSM1269717 3 0.4498 0.4590 0.032 0.000 0.688 0.280 0.000
#> GSM1269721 4 0.1981 0.6519 0.028 0.000 0.048 0.924 0.000
#> GSM1269723 4 0.5352 0.5334 0.160 0.004 0.152 0.684 0.000
#> GSM1269645 2 0.0324 0.8728 0.004 0.992 0.004 0.000 0.000
#> GSM1269653 4 0.5723 0.2753 0.392 0.000 0.088 0.520 0.000
#> GSM1269661 2 0.0566 0.8724 0.004 0.984 0.012 0.000 0.000
#> GSM1269669 1 0.7747 0.4782 0.476 0.000 0.172 0.236 0.116
#> GSM1269677 3 0.4406 0.4535 0.060 0.024 0.788 0.128 0.000
#> GSM1269685 5 0.2352 0.7041 0.092 0.000 0.008 0.004 0.896
#> GSM1269691 5 0.2011 0.7096 0.088 0.000 0.000 0.004 0.908
#> GSM1269699 5 0.3236 0.6958 0.152 0.000 0.020 0.000 0.828
#> GSM1269707 4 0.1661 0.6717 0.036 0.000 0.024 0.940 0.000
#> GSM1269651 2 0.4203 0.8545 0.188 0.760 0.052 0.000 0.000
#> GSM1269659 4 0.3254 0.6181 0.060 0.000 0.052 0.868 0.020
#> GSM1269667 2 0.3456 0.8691 0.184 0.800 0.016 0.000 0.000
#> GSM1269675 4 0.4787 0.4465 0.324 0.000 0.036 0.640 0.000
#> GSM1269683 4 0.3106 0.6508 0.140 0.000 0.020 0.840 0.000
#> GSM1269689 4 0.3861 0.5552 0.264 0.000 0.008 0.728 0.000
#> GSM1269697 5 0.5421 0.4893 0.276 0.000 0.096 0.000 0.628
#> GSM1269705 5 0.3278 0.6950 0.156 0.000 0.020 0.000 0.824
#> GSM1269713 3 0.6685 0.1229 0.236 0.000 0.388 0.376 0.000
#> GSM1269719 3 0.2966 0.4994 0.056 0.032 0.888 0.004 0.020
#> GSM1269725 3 0.6389 0.3285 0.284 0.000 0.528 0.004 0.184
#> GSM1269727 4 0.2864 0.6532 0.136 0.000 0.012 0.852 0.000
#> GSM1269649 2 0.3929 0.8611 0.208 0.764 0.028 0.000 0.000
#> GSM1269657 5 0.5669 0.4859 0.040 0.000 0.312 0.036 0.612
#> GSM1269665 2 0.0324 0.8728 0.004 0.992 0.004 0.000 0.000
#> GSM1269673 1 0.7534 0.4528 0.488 0.000 0.180 0.248 0.084
#> GSM1269681 2 0.3182 0.7803 0.032 0.844 0.124 0.000 0.000
#> GSM1269687 1 0.6207 0.2115 0.480 0.000 0.424 0.028 0.068
#> GSM1269695 5 0.3724 0.6763 0.204 0.000 0.020 0.000 0.776
#> GSM1269703 4 0.5976 0.2574 0.376 0.000 0.116 0.508 0.000
#> GSM1269711 4 0.4585 0.4332 0.352 0.000 0.020 0.628 0.000
#> GSM1269646 2 0.0693 0.8746 0.012 0.980 0.008 0.000 0.000
#> GSM1269654 5 0.5350 0.2792 0.052 0.000 0.460 0.000 0.488
#> GSM1269662 2 0.0324 0.8728 0.004 0.992 0.004 0.000 0.000
#> GSM1269670 2 0.0865 0.8674 0.004 0.972 0.024 0.000 0.000
#> GSM1269678 3 0.5309 0.3671 0.160 0.000 0.676 0.000 0.164
#> GSM1269692 5 0.7047 0.3647 0.048 0.000 0.308 0.144 0.500
#> GSM1269700 1 0.6232 0.3106 0.596 0.000 0.092 0.036 0.276
#> GSM1269708 5 0.0579 0.7351 0.008 0.000 0.008 0.000 0.984
#> GSM1269714 5 0.6510 0.4185 0.032 0.000 0.288 0.120 0.560
#> GSM1269716 3 0.4806 0.4630 0.044 0.000 0.772 0.076 0.108
#> GSM1269720 4 0.1915 0.6590 0.032 0.000 0.040 0.928 0.000
#> GSM1269722 3 0.6436 0.3067 0.164 0.004 0.488 0.344 0.000
#> GSM1269644 2 0.4762 0.8318 0.236 0.700 0.064 0.000 0.000
#> GSM1269652 5 0.0579 0.7351 0.008 0.000 0.008 0.000 0.984
#> GSM1269660 2 0.0566 0.8724 0.004 0.984 0.012 0.000 0.000
#> GSM1269668 5 0.2824 0.7003 0.116 0.000 0.020 0.000 0.864
#> GSM1269676 3 0.4815 0.4360 0.068 0.012 0.780 0.112 0.028
#> GSM1269684 3 0.8133 -0.0738 0.136 0.000 0.372 0.176 0.316
#> GSM1269690 5 0.0703 0.7333 0.024 0.000 0.000 0.000 0.976
#> GSM1269698 5 0.3615 0.6946 0.156 0.000 0.036 0.000 0.808
#> GSM1269706 4 0.5415 0.2268 0.028 0.000 0.340 0.604 0.028
#> GSM1269650 2 0.4203 0.8545 0.188 0.760 0.052 0.000 0.000
#> GSM1269658 4 0.4946 0.3874 0.056 0.004 0.260 0.680 0.000
#> GSM1269666 2 0.4424 0.8479 0.224 0.728 0.048 0.000 0.000
#> GSM1269674 1 0.7789 0.4430 0.432 0.000 0.296 0.176 0.096
#> GSM1269682 3 0.6027 0.4354 0.116 0.032 0.644 0.208 0.000
#> GSM1269688 5 0.4404 0.4188 0.264 0.000 0.000 0.032 0.704
#> GSM1269696 3 0.6554 0.3345 0.296 0.004 0.528 0.008 0.164
#> GSM1269704 5 0.3536 0.6961 0.156 0.000 0.032 0.000 0.812
#> GSM1269712 3 0.7667 0.4009 0.164 0.192 0.504 0.140 0.000
#> GSM1269718 3 0.2675 0.5044 0.040 0.032 0.904 0.004 0.020
#> GSM1269724 3 0.6206 0.4475 0.240 0.008 0.632 0.036 0.084
#> GSM1269726 4 0.4961 0.5714 0.140 0.004 0.132 0.724 0.000
#> GSM1269648 2 0.4666 0.8383 0.240 0.704 0.056 0.000 0.000
#> GSM1269656 5 0.4219 0.5853 0.024 0.000 0.260 0.000 0.716
#> GSM1269664 2 0.0324 0.8728 0.004 0.992 0.004 0.000 0.000
#> GSM1269672 5 0.3239 0.6947 0.068 0.000 0.080 0.000 0.852
#> GSM1269680 3 0.5434 0.3487 0.156 0.152 0.684 0.000 0.008
#> GSM1269686 5 0.2645 0.7262 0.044 0.000 0.068 0.000 0.888
#> GSM1269694 5 0.3061 0.7028 0.136 0.000 0.020 0.000 0.844
#> GSM1269702 5 0.1082 0.7369 0.028 0.000 0.008 0.000 0.964
#> GSM1269710 3 0.6890 0.1022 0.264 0.004 0.380 0.352 0.000
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM1269647 2 0.4386 0.7702 0.000 0.652 0.000 0.000 0.048 0.300
#> GSM1269655 1 0.6662 0.2718 0.468 0.000 0.144 0.000 0.076 0.312
#> GSM1269663 2 0.4939 0.7602 0.000 0.612 0.000 0.000 0.096 0.292
#> GSM1269671 3 0.5832 0.2250 0.000 0.052 0.504 0.024 0.396 0.024
#> GSM1269679 5 0.4505 0.5496 0.000 0.000 0.120 0.136 0.732 0.012
#> GSM1269693 4 0.1693 0.6307 0.000 0.000 0.004 0.932 0.020 0.044
#> GSM1269701 5 0.6483 0.0849 0.352 0.000 0.088 0.012 0.484 0.064
#> GSM1269709 1 0.4346 0.5702 0.740 0.000 0.008 0.020 0.196 0.036
#> GSM1269715 4 0.2249 0.6279 0.000 0.000 0.032 0.900 0.064 0.004
#> GSM1269717 3 0.6244 0.1995 0.000 0.000 0.576 0.176 0.072 0.176
#> GSM1269721 4 0.1713 0.6262 0.000 0.000 0.000 0.928 0.028 0.044
#> GSM1269723 4 0.5939 0.1728 0.000 0.000 0.404 0.444 0.136 0.016
#> GSM1269645 2 0.0717 0.7769 0.000 0.976 0.000 0.000 0.008 0.016
#> GSM1269653 5 0.5867 0.4163 0.000 0.000 0.168 0.272 0.544 0.016
#> GSM1269661 2 0.0146 0.7798 0.000 0.996 0.000 0.000 0.004 0.000
#> GSM1269669 5 0.2941 0.5890 0.024 0.000 0.012 0.092 0.864 0.008
#> GSM1269677 6 0.6722 0.4784 0.000 0.008 0.360 0.084 0.100 0.448
#> GSM1269685 1 0.3618 0.6355 0.820 0.000 0.004 0.024 0.112 0.040
#> GSM1269691 1 0.2290 0.6544 0.892 0.000 0.004 0.000 0.084 0.020
#> GSM1269699 1 0.3895 0.6437 0.804 0.000 0.096 0.000 0.040 0.060
#> GSM1269707 4 0.2344 0.6270 0.000 0.000 0.028 0.896 0.068 0.008
#> GSM1269651 2 0.3756 0.7481 0.000 0.600 0.000 0.000 0.000 0.400
#> GSM1269659 4 0.2993 0.5950 0.008 0.000 0.012 0.868 0.048 0.064
#> GSM1269667 2 0.4145 0.7805 0.000 0.700 0.000 0.000 0.048 0.252
#> GSM1269675 5 0.4497 0.3226 0.000 0.000 0.020 0.368 0.600 0.012
#> GSM1269683 4 0.4998 0.4645 0.000 0.000 0.112 0.676 0.196 0.016
#> GSM1269689 4 0.5174 -0.0475 0.000 0.000 0.052 0.512 0.420 0.016
#> GSM1269697 1 0.6278 0.4189 0.536 0.000 0.284 0.000 0.084 0.096
#> GSM1269705 1 0.4180 0.6423 0.784 0.000 0.096 0.000 0.044 0.076
#> GSM1269713 3 0.5577 0.3129 0.000 0.000 0.604 0.212 0.168 0.016
#> GSM1269719 3 0.5500 -0.1556 0.008 0.008 0.580 0.000 0.100 0.304
#> GSM1269725 3 0.5691 0.3310 0.120 0.000 0.664 0.004 0.084 0.128
#> GSM1269727 4 0.4681 0.4948 0.000 0.000 0.104 0.708 0.176 0.012
#> GSM1269649 2 0.4498 0.7687 0.000 0.644 0.000 0.000 0.056 0.300
#> GSM1269657 1 0.7151 0.2127 0.480 0.000 0.096 0.032 0.104 0.288
#> GSM1269665 2 0.0717 0.7769 0.000 0.976 0.000 0.000 0.008 0.016
#> GSM1269673 5 0.3157 0.5862 0.016 0.000 0.020 0.080 0.860 0.024
#> GSM1269681 2 0.5403 0.4146 0.000 0.648 0.132 0.004 0.020 0.196
#> GSM1269687 5 0.4289 0.4402 0.016 0.000 0.156 0.012 0.764 0.052
#> GSM1269695 1 0.4529 0.6369 0.760 0.000 0.096 0.000 0.080 0.064
#> GSM1269703 5 0.6432 0.2882 0.000 0.000 0.260 0.268 0.448 0.024
#> GSM1269711 5 0.4699 0.3288 0.000 0.000 0.036 0.376 0.580 0.008
#> GSM1269646 2 0.0363 0.7807 0.000 0.988 0.000 0.000 0.000 0.012
#> GSM1269654 1 0.7105 -0.0187 0.364 0.000 0.204 0.000 0.088 0.344
#> GSM1269662 2 0.0717 0.7769 0.000 0.976 0.000 0.000 0.008 0.016
#> GSM1269670 2 0.1285 0.7458 0.000 0.944 0.052 0.000 0.000 0.004
#> GSM1269678 3 0.7056 -0.0461 0.096 0.000 0.392 0.000 0.336 0.176
#> GSM1269692 1 0.8234 -0.1923 0.328 0.000 0.100 0.180 0.084 0.308
#> GSM1269700 5 0.6550 0.3591 0.176 0.000 0.188 0.012 0.560 0.064
#> GSM1269708 1 0.2291 0.6624 0.904 0.000 0.012 0.000 0.040 0.044
#> GSM1269714 1 0.8019 0.1038 0.424 0.000 0.096 0.140 0.100 0.240
#> GSM1269716 3 0.7040 -0.2704 0.072 0.000 0.484 0.076 0.056 0.312
#> GSM1269720 4 0.1863 0.6356 0.000 0.000 0.036 0.920 0.000 0.044
#> GSM1269722 3 0.4275 0.4287 0.000 0.000 0.728 0.192 0.076 0.004
#> GSM1269644 2 0.4970 0.7424 0.000 0.580 0.000 0.000 0.084 0.336
#> GSM1269652 1 0.2122 0.6625 0.912 0.000 0.008 0.000 0.040 0.040
#> GSM1269660 2 0.0000 0.7790 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM1269668 1 0.4374 0.5773 0.680 0.000 0.008 0.000 0.272 0.040
#> GSM1269676 6 0.6929 0.5534 0.020 0.004 0.332 0.076 0.096 0.472
#> GSM1269684 6 0.8624 0.2642 0.196 0.000 0.104 0.144 0.268 0.288
#> GSM1269690 1 0.1390 0.6645 0.948 0.000 0.004 0.000 0.032 0.016
#> GSM1269698 1 0.4335 0.6384 0.768 0.000 0.108 0.000 0.036 0.088
#> GSM1269706 4 0.6582 0.1679 0.016 0.000 0.172 0.560 0.060 0.192
#> GSM1269650 2 0.3756 0.7481 0.000 0.600 0.000 0.000 0.000 0.400
#> GSM1269658 4 0.4860 0.4443 0.000 0.000 0.040 0.720 0.100 0.140
#> GSM1269666 2 0.4828 0.7518 0.000 0.604 0.000 0.000 0.076 0.320
#> GSM1269674 5 0.3963 0.5094 0.012 0.000 0.056 0.052 0.816 0.064
#> GSM1269682 3 0.6226 0.3385 0.000 0.020 0.624 0.072 0.152 0.132
#> GSM1269688 1 0.5357 0.2724 0.572 0.000 0.012 0.028 0.352 0.036
#> GSM1269696 3 0.5644 0.3316 0.124 0.000 0.668 0.004 0.080 0.124
#> GSM1269704 1 0.4286 0.6393 0.772 0.000 0.108 0.000 0.036 0.084
#> GSM1269712 3 0.5031 0.4320 0.000 0.120 0.732 0.068 0.068 0.012
#> GSM1269718 3 0.5419 -0.1458 0.008 0.008 0.588 0.000 0.092 0.304
#> GSM1269724 3 0.3552 0.4072 0.040 0.000 0.840 0.008 0.060 0.052
#> GSM1269726 4 0.5875 0.2475 0.000 0.000 0.380 0.480 0.120 0.020
#> GSM1269648 2 0.5147 0.7280 0.000 0.568 0.000 0.000 0.104 0.328
#> GSM1269656 1 0.6166 0.3294 0.552 0.000 0.092 0.000 0.080 0.276
#> GSM1269664 2 0.0622 0.7779 0.000 0.980 0.000 0.000 0.008 0.012
#> GSM1269672 1 0.5088 0.5726 0.676 0.000 0.028 0.000 0.200 0.096
#> GSM1269680 6 0.6081 0.4752 0.008 0.052 0.280 0.000 0.092 0.568
#> GSM1269686 1 0.5137 0.6085 0.708 0.000 0.080 0.000 0.092 0.120
#> GSM1269694 1 0.3720 0.6468 0.816 0.000 0.092 0.000 0.036 0.056
#> GSM1269702 1 0.0436 0.6703 0.988 0.000 0.004 0.000 0.004 0.004
#> GSM1269710 3 0.5679 0.2546 0.000 0.000 0.568 0.208 0.216 0.008
Heatmaps for the consensus matrix. It visualizes the probability of two samples to be in a same group.
consensus_heatmap(res, k = 2)
consensus_heatmap(res, k = 3)
consensus_heatmap(res, k = 4)
consensus_heatmap(res, k = 5)
consensus_heatmap(res, k = 6)
Heatmaps for the membership of samples in all partitions to see how consistent they are:
membership_heatmap(res, k = 2)
membership_heatmap(res, k = 3)
membership_heatmap(res, k = 4)
membership_heatmap(res, k = 5)
membership_heatmap(res, k = 6)
As soon as we have had the classes for columns, we can look for signatures which are significantly different between classes which can be candidate marks for certain classes. Following are the heatmaps for signatures.
Signature heatmaps where rows are scaled:
get_signatures(res, k = 2)
get_signatures(res, k = 3)
get_signatures(res, k = 4)
get_signatures(res, k = 5)
get_signatures(res, k = 6)
Signature heatmaps where rows are not scaled:
get_signatures(res, k = 2, scale_rows = FALSE)
get_signatures(res, k = 3, scale_rows = FALSE)
get_signatures(res, k = 4, scale_rows = FALSE)
get_signatures(res, k = 5, scale_rows = FALSE)
get_signatures(res, k = 6, scale_rows = FALSE)
Compare the overlap of signatures from different k:
compare_signatures(res)
get_signature() returns a data frame invisibly. TO get the list of signatures, the function
call should be assigned to a variable explicitly. In following code, if plot argument is set
to FALSE, no heatmap is plotted while only the differential analysis is performed.
# code only for demonstration
tb = get_signature(res, k = ..., plot = FALSE)
An example of the output of tb is:
#> which_row fdr mean_1 mean_2 scaled_mean_1 scaled_mean_2 km
#> 1 38 0.042760348 8.373488 9.131774 -0.5533452 0.5164555 1
#> 2 40 0.018707592 7.106213 8.469186 -0.6173731 0.5762149 1
#> 3 55 0.019134737 10.221463 11.207825 -0.6159697 0.5749050 1
#> 4 59 0.006059896 5.921854 7.869574 -0.6899429 0.6439467 1
#> 5 60 0.018055526 8.928898 10.211722 -0.6204761 0.5791110 1
#> 6 98 0.009384629 15.714769 14.887706 0.6635654 -0.6193277 2
...
The columns in tb are:
which_row: row indices corresponding to the input matrix.fdr: FDR for the differential test. mean_x: The mean value in group x.scaled_mean_x: The mean value in group x after rows are scaled.km: Row groups if k-means clustering is applied to rows.UMAP plot which shows how samples are separated.
dimension_reduction(res, k = 2, method = "UMAP")
dimension_reduction(res, k = 3, method = "UMAP")
dimension_reduction(res, k = 4, method = "UMAP")
dimension_reduction(res, k = 5, method = "UMAP")
dimension_reduction(res, k = 6, method = "UMAP")
Following heatmap shows how subgroups are split when increasing k:
collect_classes(res)
Test correlation between subgroups and known annotations. If the known annotation is numeric, one-way ANOVA test is applied, and if the known annotation is discrete, chi-squared contingency table test is applied.
test_to_known_factors(res)
#> n agent(p) disease.state(p) gender(p) individual(p) k
#> ATC:kmeans 81 0.7532 0.434 1.00000 0.001896 2
#> ATC:kmeans 74 0.0305 0.751 0.29281 0.000445 3
#> ATC:kmeans 76 0.0206 0.874 0.12393 0.000555 4
#> ATC:kmeans 48 0.0991 0.745 0.00789 0.001111 5
#> ATC:kmeans 44 0.2206 0.626 0.11514 0.000573 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 51941 rows and 84 columns.
#> Top rows (1000, 2000, 3000, 4000, 5000) are extracted by 'ATC' method.
#> Subgroups are detected by 'skmeans' method.
#> Performed in total 1250 partitions by row resampling.
#> Best k for subgroups seems to be 3.
#>
#> Following methods can be applied to this 'ConsensusPartition' object:
#> [1] "cola_report" "collect_classes" "collect_plots"
#> [4] "collect_stats" "colnames" "compare_signatures"
#> [7] "consensus_heatmap" "dimension_reduction" "functional_enrichment"
#> [10] "get_anno_col" "get_anno" "get_classes"
#> [13] "get_consensus" "get_matrix" "get_membership"
#> [16] "get_param" "get_signatures" "get_stats"
#> [19] "is_best_k" "is_stable_k" "membership_heatmap"
#> [22] "ncol" "nrow" "plot_ecdf"
#> [25] "rownames" "select_partition_number" "show"
#> [28] "suggest_best_k" "test_to_known_factors"
collect_plots() function collects all the plots made from res for all k (number of partitions)
into one single page to provide an easy and fast comparison between different k.
collect_plots(res)
The plots are:
k and the heatmap of
predicted classes for each k.k.k.k.All the plots in panels can be made by individual functions and they are plotted later in this section.
select_partition_number() produces several plots showing different
statistics for choosing “optimized” k. There are following statistics:
k;k, the area increased is defined as \(A_k - A_{k-1}\).The detailed explanations of these statistics can be found in the cola vignette.
Generally speaking, lower PAC score, higher mean silhouette score or higher
concordance corresponds to better partition. Rand index and Jaccard index
measure how similar the current partition is compared to partition with k-1.
If they are too similar, we won't accept k is better than k-1.
select_partition_number(res)
The numeric values for all these statistics can be obtained by get_stats().
get_stats(res)
#> k 1-PAC mean_silhouette concordance area_increased Rand Jaccard
#> 2 2 1.000 0.984 0.994 0.5028 0.499 0.499
#> 3 3 1.000 0.983 0.993 0.3234 0.752 0.542
#> 4 4 0.712 0.630 0.813 0.1128 0.922 0.781
#> 5 5 0.716 0.662 0.816 0.0641 0.867 0.589
#> 6 6 0.734 0.587 0.753 0.0410 0.949 0.777
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
#> GSM1269647 2 0.0000 1.000 0.000 1.000
#> GSM1269655 1 0.0000 0.990 1.000 0.000
#> GSM1269663 2 0.0000 1.000 0.000 1.000
#> GSM1269671 2 0.0000 1.000 0.000 1.000
#> GSM1269679 2 0.0000 1.000 0.000 1.000
#> GSM1269693 1 0.0000 0.990 1.000 0.000
#> GSM1269701 1 0.0000 0.990 1.000 0.000
#> GSM1269709 1 0.0000 0.990 1.000 0.000
#> GSM1269715 1 0.0000 0.990 1.000 0.000
#> GSM1269717 2 0.0000 1.000 0.000 1.000
#> GSM1269721 1 0.0000 0.990 1.000 0.000
#> GSM1269723 2 0.0000 1.000 0.000 1.000
#> GSM1269645 2 0.0000 1.000 0.000 1.000
#> GSM1269653 1 0.0000 0.990 1.000 0.000
#> GSM1269661 2 0.0000 1.000 0.000 1.000
#> GSM1269669 1 0.0000 0.990 1.000 0.000
#> GSM1269677 2 0.0000 1.000 0.000 1.000
#> GSM1269685 1 0.0000 0.990 1.000 0.000
#> GSM1269691 1 0.0000 0.990 1.000 0.000
#> GSM1269699 1 0.0000 0.990 1.000 0.000
#> GSM1269707 1 0.0000 0.990 1.000 0.000
#> GSM1269651 2 0.0000 1.000 0.000 1.000
#> GSM1269659 1 0.0000 0.990 1.000 0.000
#> GSM1269667 2 0.0000 1.000 0.000 1.000
#> GSM1269675 2 0.0672 0.992 0.008 0.992
#> GSM1269683 2 0.0000 1.000 0.000 1.000
#> GSM1269689 1 0.0000 0.990 1.000 0.000
#> GSM1269697 1 0.0000 0.990 1.000 0.000
#> GSM1269705 1 0.0000 0.990 1.000 0.000
#> GSM1269713 1 0.0000 0.990 1.000 0.000
#> GSM1269719 2 0.0000 1.000 0.000 1.000
#> GSM1269725 1 0.0000 0.990 1.000 0.000
#> GSM1269727 2 0.0000 1.000 0.000 1.000
#> GSM1269649 2 0.0000 1.000 0.000 1.000
#> GSM1269657 1 0.0000 0.990 1.000 0.000
#> GSM1269665 2 0.0000 1.000 0.000 1.000
#> GSM1269673 1 0.9963 0.134 0.536 0.464
#> GSM1269681 2 0.0000 1.000 0.000 1.000
#> GSM1269687 1 0.0000 0.990 1.000 0.000
#> GSM1269695 1 0.0000 0.990 1.000 0.000
#> GSM1269703 1 0.0000 0.990 1.000 0.000
#> GSM1269711 1 0.0000 0.990 1.000 0.000
#> GSM1269646 2 0.0000 1.000 0.000 1.000
#> GSM1269654 1 0.0000 0.990 1.000 0.000
#> GSM1269662 2 0.0000 1.000 0.000 1.000
#> GSM1269670 2 0.0000 1.000 0.000 1.000
#> GSM1269678 1 0.0000 0.990 1.000 0.000
#> GSM1269692 1 0.0000 0.990 1.000 0.000
#> GSM1269700 1 0.0000 0.990 1.000 0.000
#> GSM1269708 1 0.0000 0.990 1.000 0.000
#> GSM1269714 1 0.0000 0.990 1.000 0.000
#> GSM1269716 1 0.0000 0.990 1.000 0.000
#> GSM1269720 2 0.0000 1.000 0.000 1.000
#> GSM1269722 2 0.0000 1.000 0.000 1.000
#> GSM1269644 2 0.0000 1.000 0.000 1.000
#> GSM1269652 1 0.0000 0.990 1.000 0.000
#> GSM1269660 2 0.0000 1.000 0.000 1.000
#> GSM1269668 1 0.0000 0.990 1.000 0.000
#> GSM1269676 2 0.0000 1.000 0.000 1.000
#> GSM1269684 1 0.0000 0.990 1.000 0.000
#> GSM1269690 1 0.0000 0.990 1.000 0.000
#> GSM1269698 1 0.0000 0.990 1.000 0.000
#> GSM1269706 1 0.0000 0.990 1.000 0.000
#> GSM1269650 2 0.0000 1.000 0.000 1.000
#> GSM1269658 2 0.0000 1.000 0.000 1.000
#> GSM1269666 2 0.0000 1.000 0.000 1.000
#> GSM1269674 1 0.0376 0.986 0.996 0.004
#> GSM1269682 2 0.0000 1.000 0.000 1.000
#> GSM1269688 1 0.0000 0.990 1.000 0.000
#> GSM1269696 1 0.0000 0.990 1.000 0.000
#> GSM1269704 1 0.0000 0.990 1.000 0.000
#> GSM1269712 2 0.0000 1.000 0.000 1.000
#> GSM1269718 2 0.0000 1.000 0.000 1.000
#> GSM1269724 2 0.0000 1.000 0.000 1.000
#> GSM1269726 2 0.0000 1.000 0.000 1.000
#> GSM1269648 2 0.0000 1.000 0.000 1.000
#> GSM1269656 1 0.0000 0.990 1.000 0.000
#> GSM1269664 2 0.0000 1.000 0.000 1.000
#> GSM1269672 1 0.0000 0.990 1.000 0.000
#> GSM1269680 2 0.0000 1.000 0.000 1.000
#> GSM1269686 1 0.0000 0.990 1.000 0.000
#> GSM1269694 1 0.0000 0.990 1.000 0.000
#> GSM1269702 1 0.0000 0.990 1.000 0.000
#> GSM1269710 2 0.0000 1.000 0.000 1.000
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM1269647 2 0.0000 0.984 0.000 1.000 0.000
#> GSM1269655 1 0.0000 0.996 1.000 0.000 0.000
#> GSM1269663 2 0.0000 0.984 0.000 1.000 0.000
#> GSM1269671 2 0.0000 0.984 0.000 1.000 0.000
#> GSM1269679 3 0.0000 1.000 0.000 0.000 1.000
#> GSM1269693 3 0.0000 1.000 0.000 0.000 1.000
#> GSM1269701 1 0.0000 0.996 1.000 0.000 0.000
#> GSM1269709 1 0.0000 0.996 1.000 0.000 0.000
#> GSM1269715 3 0.0000 1.000 0.000 0.000 1.000
#> GSM1269717 3 0.0000 1.000 0.000 0.000 1.000
#> GSM1269721 3 0.0000 1.000 0.000 0.000 1.000
#> GSM1269723 3 0.0000 1.000 0.000 0.000 1.000
#> GSM1269645 2 0.0000 0.984 0.000 1.000 0.000
#> GSM1269653 3 0.0000 1.000 0.000 0.000 1.000
#> GSM1269661 2 0.0000 0.984 0.000 1.000 0.000
#> GSM1269669 1 0.2537 0.915 0.920 0.000 0.080
#> GSM1269677 2 0.0000 0.984 0.000 1.000 0.000
#> GSM1269685 1 0.0000 0.996 1.000 0.000 0.000
#> GSM1269691 1 0.0000 0.996 1.000 0.000 0.000
#> GSM1269699 1 0.0000 0.996 1.000 0.000 0.000
#> GSM1269707 3 0.0000 1.000 0.000 0.000 1.000
#> GSM1269651 2 0.0000 0.984 0.000 1.000 0.000
#> GSM1269659 3 0.0000 1.000 0.000 0.000 1.000
#> GSM1269667 2 0.0000 0.984 0.000 1.000 0.000
#> GSM1269675 3 0.0000 1.000 0.000 0.000 1.000
#> GSM1269683 3 0.0000 1.000 0.000 0.000 1.000
#> GSM1269689 3 0.0000 1.000 0.000 0.000 1.000
#> GSM1269697 1 0.0000 0.996 1.000 0.000 0.000
#> GSM1269705 1 0.0000 0.996 1.000 0.000 0.000
#> GSM1269713 3 0.0000 1.000 0.000 0.000 1.000
#> GSM1269719 2 0.0000 0.984 0.000 1.000 0.000
#> GSM1269725 1 0.0000 0.996 1.000 0.000 0.000
#> GSM1269727 3 0.0000 1.000 0.000 0.000 1.000
#> GSM1269649 2 0.0000 0.984 0.000 1.000 0.000
#> GSM1269657 1 0.0000 0.996 1.000 0.000 0.000
#> GSM1269665 2 0.0000 0.984 0.000 1.000 0.000
#> GSM1269673 2 0.7890 0.244 0.060 0.544 0.396
#> GSM1269681 2 0.0000 0.984 0.000 1.000 0.000
#> GSM1269687 1 0.0000 0.996 1.000 0.000 0.000
#> GSM1269695 1 0.0000 0.996 1.000 0.000 0.000
#> GSM1269703 3 0.0000 1.000 0.000 0.000 1.000
#> GSM1269711 3 0.0000 1.000 0.000 0.000 1.000
#> GSM1269646 2 0.0000 0.984 0.000 1.000 0.000
#> GSM1269654 1 0.0000 0.996 1.000 0.000 0.000
#> GSM1269662 2 0.0000 0.984 0.000 1.000 0.000
#> GSM1269670 2 0.0000 0.984 0.000 1.000 0.000
#> GSM1269678 1 0.0000 0.996 1.000 0.000 0.000
#> GSM1269692 1 0.0000 0.996 1.000 0.000 0.000
#> GSM1269700 1 0.0000 0.996 1.000 0.000 0.000
#> GSM1269708 1 0.0000 0.996 1.000 0.000 0.000
#> GSM1269714 1 0.0000 0.996 1.000 0.000 0.000
#> GSM1269716 1 0.0000 0.996 1.000 0.000 0.000
#> GSM1269720 3 0.0000 1.000 0.000 0.000 1.000
#> GSM1269722 3 0.0000 1.000 0.000 0.000 1.000
#> GSM1269644 2 0.0000 0.984 0.000 1.000 0.000
#> GSM1269652 1 0.0000 0.996 1.000 0.000 0.000
#> GSM1269660 2 0.0000 0.984 0.000 1.000 0.000
#> GSM1269668 1 0.0000 0.996 1.000 0.000 0.000
#> GSM1269676 2 0.0000 0.984 0.000 1.000 0.000
#> GSM1269684 1 0.0000 0.996 1.000 0.000 0.000
#> GSM1269690 1 0.0000 0.996 1.000 0.000 0.000
#> GSM1269698 1 0.0000 0.996 1.000 0.000 0.000
#> GSM1269706 3 0.0237 0.996 0.004 0.000 0.996
#> GSM1269650 2 0.0000 0.984 0.000 1.000 0.000
#> GSM1269658 3 0.0000 1.000 0.000 0.000 1.000
#> GSM1269666 2 0.0000 0.984 0.000 1.000 0.000
#> GSM1269674 1 0.1753 0.950 0.952 0.000 0.048
#> GSM1269682 2 0.0000 0.984 0.000 1.000 0.000
#> GSM1269688 1 0.0000 0.996 1.000 0.000 0.000
#> GSM1269696 1 0.0000 0.996 1.000 0.000 0.000
#> GSM1269704 1 0.0000 0.996 1.000 0.000 0.000
#> GSM1269712 2 0.0000 0.984 0.000 1.000 0.000
#> GSM1269718 2 0.0000 0.984 0.000 1.000 0.000
#> GSM1269724 2 0.0000 0.984 0.000 1.000 0.000
#> GSM1269726 3 0.0000 1.000 0.000 0.000 1.000
#> GSM1269648 2 0.0000 0.984 0.000 1.000 0.000
#> GSM1269656 1 0.0000 0.996 1.000 0.000 0.000
#> GSM1269664 2 0.0000 0.984 0.000 1.000 0.000
#> GSM1269672 1 0.0000 0.996 1.000 0.000 0.000
#> GSM1269680 2 0.0000 0.984 0.000 1.000 0.000
#> GSM1269686 1 0.0000 0.996 1.000 0.000 0.000
#> GSM1269694 1 0.0000 0.996 1.000 0.000 0.000
#> GSM1269702 1 0.0000 0.996 1.000 0.000 0.000
#> GSM1269710 3 0.0000 1.000 0.000 0.000 1.000
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM1269647 2 0.0000 0.90355 0.000 1.000 0.000 0.000
#> GSM1269655 1 0.3400 0.72658 0.820 0.000 0.000 0.180
#> GSM1269663 2 0.0336 0.89795 0.000 0.992 0.000 0.008
#> GSM1269671 2 0.0188 0.90071 0.000 0.996 0.004 0.000
#> GSM1269679 3 0.4050 0.49236 0.000 0.024 0.808 0.168
#> GSM1269693 3 0.4193 0.48216 0.000 0.000 0.732 0.268
#> GSM1269701 1 0.6273 0.56632 0.644 0.000 0.248 0.108
#> GSM1269709 1 0.2816 0.78865 0.900 0.000 0.036 0.064
#> GSM1269715 3 0.3975 0.50671 0.000 0.000 0.760 0.240
#> GSM1269717 4 0.4304 0.43043 0.000 0.000 0.284 0.716
#> GSM1269721 3 0.4164 0.48650 0.000 0.000 0.736 0.264
#> GSM1269723 3 0.3024 0.57157 0.000 0.000 0.852 0.148
#> GSM1269645 2 0.0000 0.90355 0.000 1.000 0.000 0.000
#> GSM1269653 3 0.1488 0.58721 0.012 0.000 0.956 0.032
#> GSM1269661 2 0.0000 0.90355 0.000 1.000 0.000 0.000
#> GSM1269669 3 0.7469 -0.07659 0.392 0.000 0.432 0.176
#> GSM1269677 2 0.4454 0.51835 0.000 0.692 0.000 0.308
#> GSM1269685 1 0.2921 0.78325 0.860 0.000 0.000 0.140
#> GSM1269691 1 0.1557 0.80362 0.944 0.000 0.000 0.056
#> GSM1269699 1 0.0592 0.81414 0.984 0.000 0.000 0.016
#> GSM1269707 3 0.3975 0.50671 0.000 0.000 0.760 0.240
#> GSM1269651 2 0.0000 0.90355 0.000 1.000 0.000 0.000
#> GSM1269659 3 0.4746 0.35261 0.000 0.000 0.632 0.368
#> GSM1269667 2 0.0000 0.90355 0.000 1.000 0.000 0.000
#> GSM1269675 3 0.2868 0.53746 0.000 0.000 0.864 0.136
#> GSM1269683 3 0.2530 0.58898 0.000 0.000 0.888 0.112
#> GSM1269689 3 0.1022 0.58911 0.000 0.000 0.968 0.032
#> GSM1269697 1 0.1677 0.80767 0.948 0.000 0.012 0.040
#> GSM1269705 1 0.0592 0.81414 0.984 0.000 0.000 0.016
#> GSM1269713 3 0.4972 -0.08587 0.000 0.000 0.544 0.456
#> GSM1269719 2 0.4713 0.48878 0.000 0.640 0.000 0.360
#> GSM1269725 1 0.5300 0.25707 0.580 0.000 0.012 0.408
#> GSM1269727 3 0.2408 0.58982 0.000 0.000 0.896 0.104
#> GSM1269649 2 0.0000 0.90355 0.000 1.000 0.000 0.000
#> GSM1269657 1 0.3569 0.71685 0.804 0.000 0.000 0.196
#> GSM1269665 2 0.0000 0.90355 0.000 1.000 0.000 0.000
#> GSM1269673 3 0.9173 0.09172 0.112 0.284 0.428 0.176
#> GSM1269681 2 0.0000 0.90355 0.000 1.000 0.000 0.000
#> GSM1269687 1 0.6903 0.51563 0.592 0.000 0.224 0.184
#> GSM1269695 1 0.1792 0.80376 0.932 0.000 0.000 0.068
#> GSM1269703 3 0.0921 0.59075 0.000 0.000 0.972 0.028
#> GSM1269711 3 0.1389 0.58342 0.000 0.000 0.952 0.048
#> GSM1269646 2 0.0000 0.90355 0.000 1.000 0.000 0.000
#> GSM1269654 1 0.3649 0.71023 0.796 0.000 0.000 0.204
#> GSM1269662 2 0.0000 0.90355 0.000 1.000 0.000 0.000
#> GSM1269670 2 0.0000 0.90355 0.000 1.000 0.000 0.000
#> GSM1269678 1 0.2760 0.77445 0.872 0.000 0.000 0.128
#> GSM1269692 1 0.4564 0.56372 0.672 0.000 0.000 0.328
#> GSM1269700 1 0.6346 0.56345 0.640 0.000 0.244 0.116
#> GSM1269708 1 0.0469 0.81403 0.988 0.000 0.000 0.012
#> GSM1269714 1 0.4585 0.55786 0.668 0.000 0.000 0.332
#> GSM1269716 4 0.3975 0.40137 0.240 0.000 0.000 0.760
#> GSM1269720 3 0.4193 0.48326 0.000 0.000 0.732 0.268
#> GSM1269722 4 0.4830 0.25464 0.000 0.000 0.392 0.608
#> GSM1269644 2 0.0000 0.90355 0.000 1.000 0.000 0.000
#> GSM1269652 1 0.0469 0.81403 0.988 0.000 0.000 0.012
#> GSM1269660 2 0.0000 0.90355 0.000 1.000 0.000 0.000
#> GSM1269668 1 0.2704 0.77756 0.876 0.000 0.000 0.124
#> GSM1269676 4 0.5524 0.33229 0.048 0.276 0.000 0.676
#> GSM1269684 1 0.5147 0.44559 0.536 0.000 0.004 0.460
#> GSM1269690 1 0.0188 0.81457 0.996 0.000 0.000 0.004
#> GSM1269698 1 0.0817 0.81287 0.976 0.000 0.000 0.024
#> GSM1269706 4 0.6052 0.16451 0.048 0.000 0.396 0.556
#> GSM1269650 2 0.0000 0.90355 0.000 1.000 0.000 0.000
#> GSM1269658 3 0.4972 0.21268 0.000 0.000 0.544 0.456
#> GSM1269666 2 0.0000 0.90355 0.000 1.000 0.000 0.000
#> GSM1269674 3 0.7862 -0.03960 0.332 0.000 0.388 0.280
#> GSM1269682 2 0.0336 0.89754 0.000 0.992 0.000 0.008
#> GSM1269688 1 0.5995 0.58895 0.672 0.000 0.232 0.096
#> GSM1269696 1 0.5183 0.26220 0.584 0.000 0.008 0.408
#> GSM1269704 1 0.0817 0.81287 0.976 0.000 0.000 0.024
#> GSM1269712 2 0.4790 0.43327 0.000 0.620 0.000 0.380
#> GSM1269718 2 0.4941 0.33830 0.000 0.564 0.000 0.436
#> GSM1269724 2 0.7476 -0.00149 0.152 0.432 0.004 0.412
#> GSM1269726 3 0.3172 0.56146 0.000 0.000 0.840 0.160
#> GSM1269648 2 0.0000 0.90355 0.000 1.000 0.000 0.000
#> GSM1269656 1 0.3444 0.72564 0.816 0.000 0.000 0.184
#> GSM1269664 2 0.0000 0.90355 0.000 1.000 0.000 0.000
#> GSM1269672 1 0.1118 0.81299 0.964 0.000 0.000 0.036
#> GSM1269680 2 0.2647 0.80125 0.000 0.880 0.000 0.120
#> GSM1269686 1 0.0469 0.81396 0.988 0.000 0.000 0.012
#> GSM1269694 1 0.0592 0.81414 0.984 0.000 0.000 0.016
#> GSM1269702 1 0.0336 0.81402 0.992 0.000 0.000 0.008
#> GSM1269710 3 0.4331 0.38088 0.000 0.000 0.712 0.288
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM1269647 2 0.0290 0.921 0.008 0.992 0.000 0.000 0.000
#> GSM1269655 5 0.3037 0.786 0.032 0.000 0.100 0.004 0.864
#> GSM1269663 2 0.0609 0.917 0.020 0.980 0.000 0.000 0.000
#> GSM1269671 2 0.1364 0.898 0.012 0.952 0.036 0.000 0.000
#> GSM1269679 1 0.2967 0.603 0.868 0.016 0.012 0.104 0.000
#> GSM1269693 4 0.0880 0.744 0.032 0.000 0.000 0.968 0.000
#> GSM1269701 1 0.5052 0.433 0.612 0.000 0.048 0.000 0.340
#> GSM1269709 5 0.3074 0.676 0.196 0.000 0.000 0.000 0.804
#> GSM1269715 4 0.1701 0.744 0.048 0.000 0.016 0.936 0.000
#> GSM1269717 4 0.4584 0.299 0.028 0.000 0.312 0.660 0.000
#> GSM1269721 4 0.0880 0.744 0.032 0.000 0.000 0.968 0.000
#> GSM1269723 4 0.5210 0.518 0.084 0.000 0.264 0.652 0.000
#> GSM1269645 2 0.0000 0.922 0.000 1.000 0.000 0.000 0.000
#> GSM1269653 1 0.6117 0.249 0.516 0.000 0.064 0.392 0.028
#> GSM1269661 2 0.0000 0.922 0.000 1.000 0.000 0.000 0.000
#> GSM1269669 1 0.2482 0.643 0.892 0.000 0.000 0.024 0.084
#> GSM1269677 2 0.7174 0.175 0.040 0.492 0.208 0.260 0.000
#> GSM1269685 5 0.2142 0.805 0.028 0.000 0.048 0.004 0.920
#> GSM1269691 5 0.0963 0.812 0.036 0.000 0.000 0.000 0.964
#> GSM1269699 5 0.2520 0.802 0.048 0.000 0.056 0.000 0.896
#> GSM1269707 4 0.1408 0.746 0.044 0.000 0.008 0.948 0.000
#> GSM1269651 2 0.0000 0.922 0.000 1.000 0.000 0.000 0.000
#> GSM1269659 4 0.2592 0.702 0.052 0.000 0.056 0.892 0.000
#> GSM1269667 2 0.0510 0.919 0.016 0.984 0.000 0.000 0.000
#> GSM1269675 1 0.3957 0.480 0.712 0.000 0.008 0.280 0.000
#> GSM1269683 4 0.4159 0.630 0.156 0.000 0.068 0.776 0.000
#> GSM1269689 1 0.4736 0.287 0.576 0.000 0.020 0.404 0.000
#> GSM1269697 5 0.4780 0.565 0.048 0.000 0.280 0.000 0.672
#> GSM1269705 5 0.2520 0.802 0.048 0.000 0.056 0.000 0.896
#> GSM1269713 3 0.4681 0.407 0.052 0.000 0.696 0.252 0.000
#> GSM1269719 2 0.5040 0.106 0.024 0.516 0.456 0.004 0.000
#> GSM1269725 3 0.4210 0.517 0.036 0.000 0.740 0.000 0.224
#> GSM1269727 4 0.4676 0.553 0.208 0.000 0.072 0.720 0.000
#> GSM1269649 2 0.0510 0.919 0.016 0.984 0.000 0.000 0.000
#> GSM1269657 5 0.3870 0.758 0.032 0.000 0.088 0.048 0.832
#> GSM1269665 2 0.0000 0.922 0.000 1.000 0.000 0.000 0.000
#> GSM1269673 1 0.3151 0.620 0.876 0.056 0.000 0.032 0.036
#> GSM1269681 2 0.1444 0.890 0.012 0.948 0.040 0.000 0.000
#> GSM1269687 1 0.2612 0.613 0.868 0.000 0.008 0.000 0.124
#> GSM1269695 5 0.2790 0.796 0.068 0.000 0.052 0.000 0.880
#> GSM1269703 4 0.6454 0.147 0.340 0.000 0.168 0.488 0.004
#> GSM1269711 1 0.4686 0.329 0.596 0.000 0.020 0.384 0.000
#> GSM1269646 2 0.0000 0.922 0.000 1.000 0.000 0.000 0.000
#> GSM1269654 5 0.3895 0.754 0.032 0.000 0.164 0.008 0.796
#> GSM1269662 2 0.0000 0.922 0.000 1.000 0.000 0.000 0.000
#> GSM1269670 2 0.0404 0.918 0.000 0.988 0.012 0.000 0.000
#> GSM1269678 5 0.5680 0.509 0.228 0.000 0.148 0.000 0.624
#> GSM1269692 5 0.6288 0.483 0.032 0.000 0.100 0.284 0.584
#> GSM1269700 1 0.5644 0.436 0.584 0.000 0.100 0.000 0.316
#> GSM1269708 5 0.0162 0.817 0.004 0.000 0.000 0.000 0.996
#> GSM1269714 5 0.5952 0.532 0.036 0.000 0.076 0.260 0.628
#> GSM1269716 3 0.6100 0.357 0.032 0.000 0.612 0.264 0.092
#> GSM1269720 4 0.0324 0.741 0.004 0.000 0.004 0.992 0.000
#> GSM1269722 3 0.4167 0.440 0.024 0.000 0.724 0.252 0.000
#> GSM1269644 2 0.0404 0.920 0.012 0.988 0.000 0.000 0.000
#> GSM1269652 5 0.0162 0.817 0.004 0.000 0.000 0.000 0.996
#> GSM1269660 2 0.0000 0.922 0.000 1.000 0.000 0.000 0.000
#> GSM1269668 5 0.3424 0.654 0.240 0.000 0.000 0.000 0.760
#> GSM1269676 3 0.8192 0.257 0.048 0.140 0.460 0.284 0.068
#> GSM1269684 5 0.7487 0.402 0.156 0.000 0.100 0.236 0.508
#> GSM1269690 5 0.0290 0.817 0.008 0.000 0.000 0.000 0.992
#> GSM1269698 5 0.2291 0.804 0.036 0.000 0.056 0.000 0.908
#> GSM1269706 4 0.3544 0.672 0.048 0.000 0.048 0.856 0.048
#> GSM1269650 2 0.0000 0.922 0.000 1.000 0.000 0.000 0.000
#> GSM1269658 4 0.4072 0.607 0.108 0.000 0.100 0.792 0.000
#> GSM1269666 2 0.0404 0.920 0.012 0.988 0.000 0.000 0.000
#> GSM1269674 1 0.3410 0.622 0.840 0.000 0.000 0.068 0.092
#> GSM1269682 2 0.3449 0.790 0.004 0.844 0.088 0.064 0.000
#> GSM1269688 1 0.4415 0.306 0.552 0.000 0.000 0.004 0.444
#> GSM1269696 3 0.3847 0.554 0.036 0.000 0.784 0.000 0.180
#> GSM1269704 5 0.2291 0.804 0.036 0.000 0.056 0.000 0.908
#> GSM1269712 3 0.4297 0.548 0.008 0.200 0.756 0.036 0.000
#> GSM1269718 3 0.4637 0.442 0.028 0.292 0.676 0.004 0.000
#> GSM1269724 3 0.2878 0.599 0.012 0.024 0.880 0.000 0.084
#> GSM1269726 4 0.5181 0.514 0.080 0.000 0.268 0.652 0.000
#> GSM1269648 2 0.0609 0.917 0.020 0.980 0.000 0.000 0.000
#> GSM1269656 5 0.2871 0.784 0.032 0.000 0.088 0.004 0.876
#> GSM1269664 2 0.0000 0.922 0.000 1.000 0.000 0.000 0.000
#> GSM1269672 5 0.2377 0.754 0.128 0.000 0.000 0.000 0.872
#> GSM1269680 2 0.4149 0.700 0.040 0.768 0.188 0.004 0.000
#> GSM1269686 5 0.0898 0.818 0.020 0.000 0.008 0.000 0.972
#> GSM1269694 5 0.2450 0.804 0.048 0.000 0.052 0.000 0.900
#> GSM1269702 5 0.0451 0.818 0.004 0.000 0.008 0.000 0.988
#> GSM1269710 3 0.6082 0.134 0.108 0.008 0.540 0.344 0.000
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM1269647 2 0.0914 0.9226 0.000 0.968 0.000 0.016 0.016 0.000
#> GSM1269655 1 0.3907 0.6589 0.764 0.000 0.000 0.084 0.000 0.152
#> GSM1269663 2 0.1464 0.9161 0.000 0.944 0.000 0.016 0.036 0.004
#> GSM1269671 2 0.2240 0.8910 0.000 0.916 0.024 0.012 0.020 0.028
#> GSM1269679 5 0.1728 0.6261 0.000 0.004 0.064 0.008 0.924 0.000
#> GSM1269693 4 0.4041 0.8051 0.000 0.000 0.408 0.584 0.004 0.004
#> GSM1269701 5 0.6306 0.4289 0.280 0.000 0.004 0.116 0.540 0.060
#> GSM1269709 1 0.3673 0.6239 0.764 0.000 0.000 0.024 0.204 0.008
#> GSM1269715 3 0.4177 -0.6958 0.000 0.000 0.520 0.468 0.012 0.000
#> GSM1269717 3 0.5526 -0.0568 0.000 0.000 0.524 0.324 0.000 0.152
#> GSM1269721 4 0.3923 0.8023 0.000 0.000 0.416 0.580 0.004 0.000
#> GSM1269723 3 0.1148 0.4379 0.000 0.016 0.960 0.000 0.020 0.004
#> GSM1269645 2 0.0508 0.9252 0.000 0.984 0.004 0.000 0.000 0.012
#> GSM1269653 5 0.6947 0.2740 0.016 0.000 0.304 0.220 0.424 0.036
#> GSM1269661 2 0.0260 0.9259 0.000 0.992 0.000 0.000 0.000 0.008
#> GSM1269669 5 0.0767 0.6419 0.012 0.000 0.004 0.008 0.976 0.000
#> GSM1269677 6 0.5543 0.4568 0.000 0.156 0.020 0.208 0.000 0.616
#> GSM1269685 1 0.1713 0.7535 0.928 0.000 0.000 0.028 0.044 0.000
#> GSM1269691 1 0.1726 0.7565 0.932 0.000 0.000 0.012 0.044 0.012
#> GSM1269699 1 0.3125 0.7314 0.852 0.000 0.000 0.076 0.016 0.056
#> GSM1269707 4 0.4179 0.7294 0.000 0.000 0.472 0.516 0.012 0.000
#> GSM1269651 2 0.1390 0.9162 0.000 0.948 0.000 0.016 0.004 0.032
#> GSM1269659 4 0.4241 0.7777 0.004 0.000 0.332 0.644 0.016 0.004
#> GSM1269667 2 0.1003 0.9219 0.000 0.964 0.000 0.016 0.020 0.000
#> GSM1269675 5 0.4387 0.5487 0.000 0.000 0.128 0.152 0.720 0.000
#> GSM1269683 3 0.3829 0.0611 0.000 0.000 0.760 0.180 0.060 0.000
#> GSM1269689 5 0.5682 0.3381 0.000 0.000 0.316 0.180 0.504 0.000
#> GSM1269697 1 0.6321 0.3844 0.548 0.000 0.020 0.236 0.020 0.176
#> GSM1269705 1 0.3211 0.7304 0.848 0.000 0.000 0.076 0.020 0.056
#> GSM1269713 3 0.5167 0.3665 0.000 0.000 0.636 0.196 0.004 0.164
#> GSM1269719 6 0.2963 0.5132 0.000 0.152 0.016 0.000 0.004 0.828
#> GSM1269725 6 0.7506 0.2328 0.176 0.000 0.168 0.260 0.004 0.392
#> GSM1269727 3 0.4039 0.0880 0.000 0.000 0.752 0.156 0.092 0.000
#> GSM1269649 2 0.1088 0.9209 0.000 0.960 0.000 0.016 0.024 0.000
#> GSM1269657 1 0.4322 0.6452 0.736 0.000 0.000 0.152 0.004 0.108
#> GSM1269665 2 0.0508 0.9252 0.000 0.984 0.004 0.000 0.000 0.012
#> GSM1269673 5 0.0912 0.6394 0.008 0.004 0.004 0.012 0.972 0.000
#> GSM1269681 2 0.3690 0.5453 0.000 0.684 0.008 0.000 0.000 0.308
#> GSM1269687 5 0.1708 0.6276 0.040 0.000 0.000 0.004 0.932 0.024
#> GSM1269695 1 0.3628 0.7244 0.824 0.000 0.000 0.084 0.040 0.052
#> GSM1269703 3 0.6517 0.2625 0.008 0.000 0.556 0.160 0.208 0.068
#> GSM1269711 5 0.5529 0.4058 0.000 0.000 0.276 0.176 0.548 0.000
#> GSM1269646 2 0.0260 0.9259 0.000 0.992 0.000 0.000 0.000 0.008
#> GSM1269654 1 0.4831 0.5147 0.636 0.000 0.000 0.096 0.000 0.268
#> GSM1269662 2 0.0508 0.9252 0.000 0.984 0.004 0.000 0.000 0.012
#> GSM1269670 2 0.0820 0.9193 0.000 0.972 0.012 0.000 0.000 0.016
#> GSM1269678 1 0.6020 0.4397 0.588 0.000 0.016 0.036 0.260 0.100
#> GSM1269692 1 0.5451 0.2509 0.448 0.000 0.000 0.432 0.000 0.120
#> GSM1269700 5 0.6931 0.4062 0.260 0.000 0.008 0.168 0.484 0.080
#> GSM1269708 1 0.0862 0.7602 0.972 0.000 0.000 0.004 0.016 0.008
#> GSM1269714 1 0.4680 0.4234 0.564 0.000 0.000 0.396 0.008 0.032
#> GSM1269716 6 0.5947 0.3162 0.072 0.000 0.052 0.420 0.000 0.456
#> GSM1269720 4 0.3868 0.7268 0.000 0.000 0.492 0.508 0.000 0.000
#> GSM1269722 3 0.4825 0.3901 0.000 0.000 0.668 0.180 0.000 0.152
#> GSM1269644 2 0.1458 0.9202 0.000 0.948 0.000 0.016 0.016 0.020
#> GSM1269652 1 0.0748 0.7595 0.976 0.000 0.000 0.004 0.016 0.004
#> GSM1269660 2 0.0405 0.9249 0.000 0.988 0.004 0.000 0.000 0.008
#> GSM1269668 1 0.3309 0.5900 0.720 0.000 0.000 0.000 0.280 0.000
#> GSM1269676 6 0.3869 0.4655 0.016 0.004 0.008 0.236 0.000 0.736
#> GSM1269684 1 0.7050 0.1507 0.384 0.000 0.000 0.356 0.112 0.148
#> GSM1269690 1 0.0603 0.7595 0.980 0.000 0.000 0.004 0.016 0.000
#> GSM1269698 1 0.2817 0.7383 0.868 0.000 0.000 0.072 0.008 0.052
#> GSM1269706 4 0.5315 0.6929 0.044 0.000 0.324 0.588 0.000 0.044
#> GSM1269650 2 0.1536 0.9117 0.000 0.940 0.000 0.016 0.004 0.040
#> GSM1269658 4 0.5920 0.5883 0.000 0.000 0.220 0.604 0.068 0.108
#> GSM1269666 2 0.1173 0.9218 0.000 0.960 0.000 0.016 0.016 0.008
#> GSM1269674 5 0.1995 0.6261 0.036 0.000 0.000 0.052 0.912 0.000
#> GSM1269682 2 0.5786 0.2219 0.000 0.524 0.272 0.000 0.004 0.200
#> GSM1269688 5 0.5457 0.2643 0.408 0.000 0.008 0.072 0.504 0.008
#> GSM1269696 6 0.7131 0.2478 0.100 0.000 0.192 0.256 0.004 0.448
#> GSM1269704 1 0.2817 0.7383 0.868 0.000 0.000 0.072 0.008 0.052
#> GSM1269712 3 0.6605 0.1589 0.000 0.088 0.528 0.176 0.000 0.208
#> GSM1269718 6 0.1984 0.5194 0.000 0.056 0.032 0.000 0.000 0.912
#> GSM1269724 6 0.6372 0.1808 0.016 0.000 0.280 0.236 0.004 0.464
#> GSM1269726 3 0.0508 0.4311 0.000 0.000 0.984 0.000 0.012 0.004
#> GSM1269648 2 0.1346 0.9200 0.000 0.952 0.000 0.016 0.024 0.008
#> GSM1269656 1 0.3751 0.6802 0.792 0.000 0.000 0.096 0.004 0.108
#> GSM1269664 2 0.0508 0.9252 0.000 0.984 0.004 0.000 0.000 0.012
#> GSM1269672 1 0.2595 0.6936 0.836 0.000 0.000 0.000 0.160 0.004
#> GSM1269680 6 0.4928 0.4409 0.000 0.260 0.000 0.096 0.004 0.640
#> GSM1269686 1 0.0665 0.7625 0.980 0.000 0.000 0.008 0.004 0.008
#> GSM1269694 1 0.2971 0.7350 0.860 0.000 0.000 0.076 0.012 0.052
#> GSM1269702 1 0.0291 0.7620 0.992 0.000 0.000 0.000 0.004 0.004
#> GSM1269710 3 0.4637 0.4896 0.000 0.024 0.752 0.136 0.016 0.072
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 agent(p) disease.state(p) gender(p) individual(p) k
#> ATC:skmeans 83 0.9049 0.446 0.2601 6.52e-03 2
#> ATC:skmeans 83 0.0342 0.673 0.2329 4.28e-04 3
#> ATC:skmeans 61 0.0141 0.503 0.6235 7.01e-03 4
#> ATC:skmeans 65 0.2622 0.292 0.0399 1.09e-04 5
#> ATC:skmeans 54 0.4639 0.388 0.1665 2.92e-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["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 51941 rows and 84 columns.
#> Top rows (1000, 2000, 3000, 4000, 5000) are extracted by 'ATC' method.
#> Subgroups are detected by 'pam' method.
#> Performed in total 1250 partitions by row resampling.
#> Best k for subgroups seems to be 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 0.362 0.684 0.859 0.4775 0.494 0.494
#> 3 3 0.780 0.800 0.921 0.3696 0.663 0.424
#> 4 4 0.686 0.729 0.850 0.1462 0.853 0.599
#> 5 5 0.723 0.674 0.816 0.0581 0.921 0.707
#> 6 6 0.754 0.693 0.812 0.0466 0.921 0.663
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
#> GSM1269647 2 0.8555 0.46063 0.280 0.720
#> GSM1269655 1 0.0000 0.84164 1.000 0.000
#> GSM1269663 2 0.0000 0.77281 0.000 1.000
#> GSM1269671 2 0.0000 0.77281 0.000 1.000
#> GSM1269679 2 0.7674 0.73346 0.224 0.776
#> GSM1269693 2 0.8267 0.71529 0.260 0.740
#> GSM1269701 1 0.8443 0.47433 0.728 0.272
#> GSM1269709 1 0.0000 0.84164 1.000 0.000
#> GSM1269715 2 0.8267 0.71529 0.260 0.740
#> GSM1269717 2 0.8267 0.71529 0.260 0.740
#> GSM1269721 2 0.8267 0.71529 0.260 0.740
#> GSM1269723 2 0.0000 0.77281 0.000 1.000
#> GSM1269645 2 0.0000 0.77281 0.000 1.000
#> GSM1269653 2 0.9944 0.35534 0.456 0.544
#> GSM1269661 2 0.0000 0.77281 0.000 1.000
#> GSM1269669 1 0.9954 -0.05782 0.540 0.460
#> GSM1269677 2 0.8016 0.72454 0.244 0.756
#> GSM1269685 1 0.0000 0.84164 1.000 0.000
#> GSM1269691 1 0.0000 0.84164 1.000 0.000
#> GSM1269699 1 0.0000 0.84164 1.000 0.000
#> GSM1269707 2 0.9954 0.34545 0.460 0.540
#> GSM1269651 2 0.0000 0.77281 0.000 1.000
#> GSM1269659 2 0.8713 0.67678 0.292 0.708
#> GSM1269667 2 0.3274 0.74210 0.060 0.940
#> GSM1269675 2 0.8267 0.71529 0.260 0.740
#> GSM1269683 2 0.6973 0.74560 0.188 0.812
#> GSM1269689 2 0.8267 0.71529 0.260 0.740
#> GSM1269697 1 0.0000 0.84164 1.000 0.000
#> GSM1269705 1 0.0000 0.84164 1.000 0.000
#> GSM1269713 2 0.8267 0.71529 0.260 0.740
#> GSM1269719 2 0.8499 0.63817 0.276 0.724
#> GSM1269725 1 0.0000 0.84164 1.000 0.000
#> GSM1269727 2 0.8267 0.71529 0.260 0.740
#> GSM1269649 2 0.0000 0.77281 0.000 1.000
#> GSM1269657 1 0.0376 0.84001 0.996 0.004
#> GSM1269665 2 0.0000 0.77281 0.000 1.000
#> GSM1269673 1 0.9909 0.00728 0.556 0.444
#> GSM1269681 2 0.0000 0.77281 0.000 1.000
#> GSM1269687 1 0.5946 0.73112 0.856 0.144
#> GSM1269695 1 0.0000 0.84164 1.000 0.000
#> GSM1269703 2 0.8555 0.69655 0.280 0.720
#> GSM1269711 2 0.8499 0.70010 0.276 0.724
#> GSM1269646 2 0.8555 0.46063 0.280 0.720
#> GSM1269654 1 0.0376 0.83987 0.996 0.004
#> GSM1269662 2 0.0000 0.77281 0.000 1.000
#> GSM1269670 2 0.0000 0.77281 0.000 1.000
#> GSM1269678 1 0.7528 0.64728 0.784 0.216
#> GSM1269692 1 0.1414 0.83148 0.980 0.020
#> GSM1269700 1 0.5629 0.71597 0.868 0.132
#> GSM1269708 1 0.0000 0.84164 1.000 0.000
#> GSM1269714 1 0.1184 0.83328 0.984 0.016
#> GSM1269716 1 0.5737 0.74224 0.864 0.136
#> GSM1269720 2 0.8267 0.71529 0.260 0.740
#> GSM1269722 2 0.8267 0.71529 0.260 0.740
#> GSM1269644 2 0.9686 0.17069 0.396 0.604
#> GSM1269652 1 0.0000 0.84164 1.000 0.000
#> GSM1269660 2 0.0000 0.77281 0.000 1.000
#> GSM1269668 1 0.0000 0.84164 1.000 0.000
#> GSM1269676 1 0.8081 0.59628 0.752 0.248
#> GSM1269684 1 0.9881 0.09671 0.564 0.436
#> GSM1269690 1 0.0000 0.84164 1.000 0.000
#> GSM1269698 1 0.0000 0.84164 1.000 0.000
#> GSM1269706 1 0.8763 0.43182 0.704 0.296
#> GSM1269650 2 0.8909 0.40416 0.308 0.692
#> GSM1269658 2 0.8267 0.71529 0.260 0.740
#> GSM1269666 1 0.9983 0.21796 0.524 0.476
#> GSM1269674 1 0.9954 -0.04507 0.540 0.460
#> GSM1269682 2 0.0000 0.77281 0.000 1.000
#> GSM1269688 1 0.0000 0.84164 1.000 0.000
#> GSM1269696 1 0.0000 0.84164 1.000 0.000
#> GSM1269704 1 0.0000 0.84164 1.000 0.000
#> GSM1269712 2 0.0000 0.77281 0.000 1.000
#> GSM1269718 1 0.7528 0.64728 0.784 0.216
#> GSM1269724 1 0.7674 0.63967 0.776 0.224
#> GSM1269726 2 0.0376 0.77298 0.004 0.996
#> GSM1269648 1 0.9963 0.24197 0.536 0.464
#> GSM1269656 1 0.0000 0.84164 1.000 0.000
#> GSM1269664 2 0.0000 0.77281 0.000 1.000
#> GSM1269672 1 0.0000 0.84164 1.000 0.000
#> GSM1269680 1 0.7528 0.64728 0.784 0.216
#> GSM1269686 1 0.0000 0.84164 1.000 0.000
#> GSM1269694 1 0.0000 0.84164 1.000 0.000
#> GSM1269702 1 0.0000 0.84164 1.000 0.000
#> GSM1269710 2 0.2778 0.77129 0.048 0.952
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM1269647 2 0.0000 0.9404 0.000 1.000 0.000
#> GSM1269655 1 0.0000 0.9307 1.000 0.000 0.000
#> GSM1269663 2 0.0000 0.9404 0.000 1.000 0.000
#> GSM1269671 2 0.4555 0.7039 0.000 0.800 0.200
#> GSM1269679 3 0.0000 0.8699 0.000 0.000 1.000
#> GSM1269693 3 0.0000 0.8699 0.000 0.000 1.000
#> GSM1269701 1 0.0424 0.9246 0.992 0.000 0.008
#> GSM1269709 1 0.0000 0.9307 1.000 0.000 0.000
#> GSM1269715 3 0.0000 0.8699 0.000 0.000 1.000
#> GSM1269717 3 0.0000 0.8699 0.000 0.000 1.000
#> GSM1269721 3 0.0000 0.8699 0.000 0.000 1.000
#> GSM1269723 3 0.0000 0.8699 0.000 0.000 1.000
#> GSM1269645 2 0.0000 0.9404 0.000 1.000 0.000
#> GSM1269653 3 0.0000 0.8699 0.000 0.000 1.000
#> GSM1269661 2 0.0000 0.9404 0.000 1.000 0.000
#> GSM1269669 3 0.0237 0.8671 0.004 0.000 0.996
#> GSM1269677 3 0.0000 0.8699 0.000 0.000 1.000
#> GSM1269685 1 0.0000 0.9307 1.000 0.000 0.000
#> GSM1269691 1 0.0000 0.9307 1.000 0.000 0.000
#> GSM1269699 1 0.0000 0.9307 1.000 0.000 0.000
#> GSM1269707 3 0.0000 0.8699 0.000 0.000 1.000
#> GSM1269651 2 0.0000 0.9404 0.000 1.000 0.000
#> GSM1269659 3 0.0000 0.8699 0.000 0.000 1.000
#> GSM1269667 2 0.0000 0.9404 0.000 1.000 0.000
#> GSM1269675 3 0.0000 0.8699 0.000 0.000 1.000
#> GSM1269683 3 0.0000 0.8699 0.000 0.000 1.000
#> GSM1269689 3 0.0000 0.8699 0.000 0.000 1.000
#> GSM1269697 1 0.0000 0.9307 1.000 0.000 0.000
#> GSM1269705 1 0.0000 0.9307 1.000 0.000 0.000
#> GSM1269713 3 0.0000 0.8699 0.000 0.000 1.000
#> GSM1269719 2 0.6062 0.3266 0.000 0.616 0.384
#> GSM1269725 1 0.2356 0.8669 0.928 0.000 0.072
#> GSM1269727 3 0.0000 0.8699 0.000 0.000 1.000
#> GSM1269649 2 0.0000 0.9404 0.000 1.000 0.000
#> GSM1269657 1 0.5968 0.3628 0.636 0.000 0.364
#> GSM1269665 2 0.0000 0.9404 0.000 1.000 0.000
#> GSM1269673 3 0.4473 0.6839 0.008 0.164 0.828
#> GSM1269681 2 0.0000 0.9404 0.000 1.000 0.000
#> GSM1269687 3 0.6280 0.1998 0.460 0.000 0.540
#> GSM1269695 1 0.0000 0.9307 1.000 0.000 0.000
#> GSM1269703 3 0.0000 0.8699 0.000 0.000 1.000
#> GSM1269711 3 0.0000 0.8699 0.000 0.000 1.000
#> GSM1269646 2 0.0000 0.9404 0.000 1.000 0.000
#> GSM1269654 1 0.5529 0.5200 0.704 0.000 0.296
#> GSM1269662 2 0.0000 0.9404 0.000 1.000 0.000
#> GSM1269670 2 0.0000 0.9404 0.000 1.000 0.000
#> GSM1269678 3 0.6168 0.3348 0.412 0.000 0.588
#> GSM1269692 1 0.4796 0.6692 0.780 0.000 0.220
#> GSM1269700 1 0.0000 0.9307 1.000 0.000 0.000
#> GSM1269708 1 0.0000 0.9307 1.000 0.000 0.000
#> GSM1269714 3 0.6280 0.2049 0.460 0.000 0.540
#> GSM1269716 3 0.6244 0.2589 0.440 0.000 0.560
#> GSM1269720 3 0.0000 0.8699 0.000 0.000 1.000
#> GSM1269722 3 0.0000 0.8699 0.000 0.000 1.000
#> GSM1269644 2 0.0000 0.9404 0.000 1.000 0.000
#> GSM1269652 1 0.0000 0.9307 1.000 0.000 0.000
#> GSM1269660 2 0.0000 0.9404 0.000 1.000 0.000
#> GSM1269668 1 0.0000 0.9307 1.000 0.000 0.000
#> GSM1269676 3 0.5216 0.6096 0.260 0.000 0.740
#> GSM1269684 3 0.6168 0.3460 0.412 0.000 0.588
#> GSM1269690 1 0.0000 0.9307 1.000 0.000 0.000
#> GSM1269698 1 0.0000 0.9307 1.000 0.000 0.000
#> GSM1269706 3 0.6192 0.3125 0.420 0.000 0.580
#> GSM1269650 2 0.0000 0.9404 0.000 1.000 0.000
#> GSM1269658 3 0.0000 0.8699 0.000 0.000 1.000
#> GSM1269666 2 0.0000 0.9404 0.000 1.000 0.000
#> GSM1269674 3 0.0000 0.8699 0.000 0.000 1.000
#> GSM1269682 3 0.0000 0.8699 0.000 0.000 1.000
#> GSM1269688 1 0.0000 0.9307 1.000 0.000 0.000
#> GSM1269696 1 0.2959 0.8378 0.900 0.000 0.100
#> GSM1269704 1 0.0000 0.9307 1.000 0.000 0.000
#> GSM1269712 3 0.5098 0.5978 0.000 0.248 0.752
#> GSM1269718 3 0.9959 0.0654 0.340 0.292 0.368
#> GSM1269724 1 0.9490 0.0399 0.444 0.188 0.368
#> GSM1269726 3 0.0000 0.8699 0.000 0.000 1.000
#> GSM1269648 2 0.0000 0.9404 0.000 1.000 0.000
#> GSM1269656 1 0.0747 0.9190 0.984 0.000 0.016
#> GSM1269664 2 0.0000 0.9404 0.000 1.000 0.000
#> GSM1269672 1 0.0000 0.9307 1.000 0.000 0.000
#> GSM1269680 2 0.9574 0.1496 0.312 0.468 0.220
#> GSM1269686 1 0.0000 0.9307 1.000 0.000 0.000
#> GSM1269694 1 0.0000 0.9307 1.000 0.000 0.000
#> GSM1269702 1 0.0000 0.9307 1.000 0.000 0.000
#> GSM1269710 3 0.0000 0.8699 0.000 0.000 1.000
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM1269647 2 0.0000 0.950 0.000 1.000 0.000 0.000
#> GSM1269655 1 0.5000 -0.180 0.504 0.000 0.000 0.496
#> GSM1269663 2 0.0000 0.950 0.000 1.000 0.000 0.000
#> GSM1269671 2 0.4590 0.760 0.000 0.792 0.060 0.148
#> GSM1269679 3 0.3486 0.782 0.000 0.000 0.812 0.188
#> GSM1269693 3 0.0707 0.802 0.000 0.000 0.980 0.020
#> GSM1269701 1 0.0000 0.880 1.000 0.000 0.000 0.000
#> GSM1269709 1 0.2101 0.826 0.928 0.000 0.060 0.012
#> GSM1269715 3 0.4134 0.664 0.000 0.000 0.740 0.260
#> GSM1269717 3 0.4661 0.659 0.000 0.000 0.652 0.348
#> GSM1269721 3 0.0707 0.800 0.000 0.000 0.980 0.020
#> GSM1269723 3 0.2469 0.813 0.000 0.000 0.892 0.108
#> GSM1269645 2 0.0000 0.950 0.000 1.000 0.000 0.000
#> GSM1269653 3 0.4164 0.660 0.000 0.000 0.736 0.264
#> GSM1269661 2 0.0000 0.950 0.000 1.000 0.000 0.000
#> GSM1269669 3 0.4163 0.766 0.020 0.000 0.792 0.188
#> GSM1269677 3 0.4624 0.668 0.000 0.000 0.660 0.340
#> GSM1269685 1 0.3400 0.693 0.820 0.000 0.000 0.180
#> GSM1269691 1 0.0000 0.880 1.000 0.000 0.000 0.000
#> GSM1269699 1 0.0000 0.880 1.000 0.000 0.000 0.000
#> GSM1269707 3 0.4193 0.658 0.000 0.000 0.732 0.268
#> GSM1269651 2 0.0000 0.950 0.000 1.000 0.000 0.000
#> GSM1269659 3 0.2647 0.764 0.000 0.000 0.880 0.120
#> GSM1269667 2 0.0000 0.950 0.000 1.000 0.000 0.000
#> GSM1269675 3 0.2345 0.768 0.000 0.000 0.900 0.100
#> GSM1269683 3 0.2281 0.813 0.000 0.000 0.904 0.096
#> GSM1269689 3 0.0188 0.800 0.000 0.000 0.996 0.004
#> GSM1269697 1 0.0000 0.880 1.000 0.000 0.000 0.000
#> GSM1269705 1 0.0000 0.880 1.000 0.000 0.000 0.000
#> GSM1269713 3 0.4522 0.661 0.000 0.000 0.680 0.320
#> GSM1269719 2 0.6919 0.230 0.000 0.528 0.120 0.352
#> GSM1269725 1 0.2699 0.810 0.904 0.000 0.028 0.068
#> GSM1269727 3 0.0000 0.802 0.000 0.000 1.000 0.000
#> GSM1269649 2 0.0000 0.950 0.000 1.000 0.000 0.000
#> GSM1269657 4 0.1452 0.681 0.036 0.000 0.008 0.956
#> GSM1269665 2 0.0000 0.950 0.000 1.000 0.000 0.000
#> GSM1269673 3 0.4267 0.762 0.024 0.000 0.788 0.188
#> GSM1269681 2 0.0921 0.930 0.000 0.972 0.000 0.028
#> GSM1269687 4 0.4776 0.490 0.060 0.000 0.164 0.776
#> GSM1269695 1 0.0000 0.880 1.000 0.000 0.000 0.000
#> GSM1269703 3 0.4642 0.739 0.020 0.000 0.740 0.240
#> GSM1269711 3 0.2469 0.761 0.000 0.000 0.892 0.108
#> GSM1269646 2 0.0000 0.950 0.000 1.000 0.000 0.000
#> GSM1269654 4 0.4215 0.688 0.104 0.000 0.072 0.824
#> GSM1269662 2 0.0000 0.950 0.000 1.000 0.000 0.000
#> GSM1269670 2 0.0000 0.950 0.000 1.000 0.000 0.000
#> GSM1269678 4 0.2271 0.651 0.008 0.000 0.076 0.916
#> GSM1269692 4 0.4776 0.567 0.272 0.000 0.016 0.712
#> GSM1269700 1 0.0188 0.878 0.996 0.000 0.000 0.004
#> GSM1269708 4 0.4977 0.207 0.460 0.000 0.000 0.540
#> GSM1269714 4 0.3681 0.664 0.008 0.000 0.176 0.816
#> GSM1269716 4 0.2760 0.662 0.000 0.000 0.128 0.872
#> GSM1269720 3 0.1716 0.812 0.000 0.000 0.936 0.064
#> GSM1269722 3 0.4661 0.659 0.000 0.000 0.652 0.348
#> GSM1269644 2 0.2281 0.886 0.000 0.904 0.000 0.096
#> GSM1269652 4 0.4981 0.203 0.464 0.000 0.000 0.536
#> GSM1269660 2 0.0000 0.950 0.000 1.000 0.000 0.000
#> GSM1269668 4 0.4955 0.376 0.344 0.000 0.008 0.648
#> GSM1269676 4 0.4222 0.365 0.000 0.000 0.272 0.728
#> GSM1269684 4 0.3833 0.694 0.072 0.000 0.080 0.848
#> GSM1269690 1 0.1940 0.827 0.924 0.000 0.000 0.076
#> GSM1269698 1 0.3400 0.685 0.820 0.000 0.000 0.180
#> GSM1269706 4 0.5339 0.561 0.040 0.000 0.272 0.688
#> GSM1269650 2 0.0000 0.950 0.000 1.000 0.000 0.000
#> GSM1269658 3 0.3024 0.806 0.000 0.000 0.852 0.148
#> GSM1269666 2 0.0000 0.950 0.000 1.000 0.000 0.000
#> GSM1269674 3 0.3726 0.772 0.000 0.000 0.788 0.212
#> GSM1269682 3 0.2469 0.813 0.000 0.000 0.892 0.108
#> GSM1269688 1 0.5168 0.525 0.712 0.000 0.040 0.248
#> GSM1269696 1 0.3090 0.776 0.888 0.000 0.056 0.056
#> GSM1269704 1 0.0000 0.880 1.000 0.000 0.000 0.000
#> GSM1269712 3 0.7740 0.313 0.000 0.236 0.416 0.348
#> GSM1269718 4 0.2408 0.655 0.000 0.000 0.104 0.896
#> GSM1269724 4 0.6598 0.583 0.124 0.064 0.104 0.708
#> GSM1269726 3 0.2469 0.813 0.000 0.000 0.892 0.108
#> GSM1269648 2 0.2345 0.877 0.000 0.900 0.000 0.100
#> GSM1269656 4 0.4898 0.314 0.416 0.000 0.000 0.584
#> GSM1269664 2 0.0000 0.950 0.000 1.000 0.000 0.000
#> GSM1269672 4 0.4837 0.372 0.348 0.000 0.004 0.648
#> GSM1269680 4 0.4507 0.631 0.000 0.168 0.044 0.788
#> GSM1269686 4 0.4967 0.234 0.452 0.000 0.000 0.548
#> GSM1269694 1 0.0000 0.880 1.000 0.000 0.000 0.000
#> GSM1269702 1 0.0336 0.877 0.992 0.000 0.000 0.008
#> GSM1269710 3 0.2408 0.813 0.000 0.000 0.896 0.104
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM1269647 2 0.0000 0.945810 0.000 1.000 0.000 0.000 0.000
#> GSM1269655 5 0.4350 0.098499 0.408 0.000 0.004 0.000 0.588
#> GSM1269663 2 0.0000 0.945810 0.000 1.000 0.000 0.000 0.000
#> GSM1269671 2 0.4688 0.709395 0.128 0.764 0.092 0.016 0.000
#> GSM1269679 4 0.4421 0.714974 0.184 0.000 0.068 0.748 0.000
#> GSM1269693 4 0.0404 0.731880 0.000 0.000 0.012 0.988 0.000
#> GSM1269701 5 0.0000 0.880386 0.000 0.000 0.000 0.000 1.000
#> GSM1269709 5 0.2520 0.800485 0.048 0.000 0.000 0.056 0.896
#> GSM1269715 4 0.3710 0.666833 0.048 0.000 0.144 0.808 0.000
#> GSM1269717 4 0.4171 0.601984 0.000 0.000 0.396 0.604 0.000
#> GSM1269721 4 0.1121 0.715842 0.044 0.000 0.000 0.956 0.000
#> GSM1269723 4 0.3534 0.699073 0.000 0.000 0.256 0.744 0.000
#> GSM1269645 2 0.0000 0.945810 0.000 1.000 0.000 0.000 0.000
#> GSM1269653 4 0.4404 0.667565 0.088 0.000 0.152 0.760 0.000
#> GSM1269661 2 0.0000 0.945810 0.000 1.000 0.000 0.000 0.000
#> GSM1269669 4 0.4256 0.561583 0.436 0.000 0.000 0.564 0.000
#> GSM1269677 4 0.4182 0.598627 0.000 0.000 0.400 0.600 0.000
#> GSM1269685 5 0.4297 -0.091347 0.472 0.000 0.000 0.000 0.528
#> GSM1269691 5 0.0000 0.880386 0.000 0.000 0.000 0.000 1.000
#> GSM1269699 5 0.0000 0.880386 0.000 0.000 0.000 0.000 1.000
#> GSM1269707 4 0.3953 0.659687 0.060 0.000 0.148 0.792 0.000
#> GSM1269651 2 0.0000 0.945810 0.000 1.000 0.000 0.000 0.000
#> GSM1269659 4 0.1410 0.715156 0.060 0.000 0.000 0.940 0.000
#> GSM1269667 2 0.0000 0.945810 0.000 1.000 0.000 0.000 0.000
#> GSM1269675 4 0.3143 0.711222 0.204 0.000 0.000 0.796 0.000
#> GSM1269683 4 0.4180 0.710743 0.036 0.000 0.220 0.744 0.000
#> GSM1269689 4 0.1341 0.730422 0.056 0.000 0.000 0.944 0.000
#> GSM1269697 5 0.0162 0.878308 0.004 0.000 0.000 0.000 0.996
#> GSM1269705 5 0.0000 0.880386 0.000 0.000 0.000 0.000 1.000
#> GSM1269713 3 0.3003 0.560729 0.044 0.000 0.864 0.092 0.000
#> GSM1269719 3 0.4448 -0.031532 0.000 0.480 0.516 0.004 0.000
#> GSM1269725 3 0.5249 0.000675 0.024 0.000 0.508 0.012 0.456
#> GSM1269727 4 0.0290 0.731295 0.000 0.000 0.008 0.992 0.000
#> GSM1269649 2 0.0000 0.945810 0.000 1.000 0.000 0.000 0.000
#> GSM1269657 1 0.1830 0.588400 0.932 0.000 0.052 0.004 0.012
#> GSM1269665 2 0.0000 0.945810 0.000 1.000 0.000 0.000 0.000
#> GSM1269673 4 0.4273 0.548660 0.448 0.000 0.000 0.552 0.000
#> GSM1269681 2 0.0404 0.937371 0.000 0.988 0.012 0.000 0.000
#> GSM1269687 4 0.7835 0.429466 0.280 0.000 0.240 0.404 0.076
#> GSM1269695 5 0.0000 0.880386 0.000 0.000 0.000 0.000 1.000
#> GSM1269703 4 0.3895 0.668501 0.000 0.000 0.320 0.680 0.000
#> GSM1269711 4 0.2358 0.728517 0.104 0.000 0.008 0.888 0.000
#> GSM1269646 2 0.0000 0.945810 0.000 1.000 0.000 0.000 0.000
#> GSM1269654 1 0.5635 0.480967 0.496 0.000 0.428 0.000 0.076
#> GSM1269662 2 0.0000 0.945810 0.000 1.000 0.000 0.000 0.000
#> GSM1269670 2 0.0880 0.921500 0.000 0.968 0.032 0.000 0.000
#> GSM1269678 3 0.4201 0.275800 0.408 0.000 0.592 0.000 0.000
#> GSM1269692 1 0.6191 0.581501 0.552 0.000 0.244 0.000 0.204
#> GSM1269700 5 0.0162 0.877850 0.004 0.000 0.000 0.000 0.996
#> GSM1269708 1 0.4227 0.292670 0.580 0.000 0.000 0.000 0.420
#> GSM1269714 1 0.5752 0.483060 0.620 0.000 0.172 0.208 0.000
#> GSM1269716 1 0.4830 0.392502 0.492 0.000 0.488 0.020 0.000
#> GSM1269720 4 0.1965 0.741345 0.000 0.000 0.096 0.904 0.000
#> GSM1269722 3 0.0579 0.630945 0.008 0.000 0.984 0.008 0.000
#> GSM1269644 2 0.3210 0.743312 0.212 0.788 0.000 0.000 0.000
#> GSM1269652 1 0.3586 0.575625 0.736 0.000 0.000 0.000 0.264
#> GSM1269660 2 0.0000 0.945810 0.000 1.000 0.000 0.000 0.000
#> GSM1269668 1 0.1478 0.590556 0.936 0.000 0.000 0.000 0.064
#> GSM1269676 1 0.4307 0.392772 0.504 0.000 0.496 0.000 0.000
#> GSM1269684 1 0.6857 0.477239 0.524 0.000 0.316 0.096 0.064
#> GSM1269690 5 0.2561 0.733386 0.144 0.000 0.000 0.000 0.856
#> GSM1269698 5 0.0000 0.880386 0.000 0.000 0.000 0.000 1.000
#> GSM1269706 4 0.7643 0.110040 0.272 0.000 0.176 0.464 0.088
#> GSM1269650 2 0.1043 0.915847 0.000 0.960 0.040 0.000 0.000
#> GSM1269658 4 0.4164 0.729572 0.096 0.000 0.120 0.784 0.000
#> GSM1269666 2 0.0000 0.945810 0.000 1.000 0.000 0.000 0.000
#> GSM1269674 4 0.4278 0.544905 0.452 0.000 0.000 0.548 0.000
#> GSM1269682 4 0.3689 0.699798 0.004 0.000 0.256 0.740 0.000
#> GSM1269688 5 0.4002 0.686501 0.084 0.000 0.000 0.120 0.796
#> GSM1269696 3 0.3949 0.354146 0.000 0.000 0.668 0.000 0.332
#> GSM1269704 5 0.0000 0.880386 0.000 0.000 0.000 0.000 1.000
#> GSM1269712 3 0.0000 0.634294 0.000 0.000 1.000 0.000 0.000
#> GSM1269718 3 0.0000 0.634294 0.000 0.000 1.000 0.000 0.000
#> GSM1269724 3 0.0000 0.634294 0.000 0.000 1.000 0.000 0.000
#> GSM1269726 4 0.3730 0.679979 0.000 0.000 0.288 0.712 0.000
#> GSM1269648 2 0.4088 0.528143 0.368 0.632 0.000 0.000 0.000
#> GSM1269656 1 0.3480 0.592635 0.752 0.000 0.000 0.000 0.248
#> GSM1269664 2 0.0000 0.945810 0.000 1.000 0.000 0.000 0.000
#> GSM1269672 1 0.1851 0.602096 0.912 0.000 0.000 0.000 0.088
#> GSM1269680 1 0.5137 0.525091 0.676 0.096 0.228 0.000 0.000
#> GSM1269686 1 0.3534 0.588625 0.744 0.000 0.000 0.000 0.256
#> GSM1269694 5 0.0000 0.880386 0.000 0.000 0.000 0.000 1.000
#> GSM1269702 5 0.1043 0.850472 0.040 0.000 0.000 0.000 0.960
#> GSM1269710 4 0.4455 0.693398 0.036 0.000 0.260 0.704 0.000
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM1269647 2 0.0000 0.9146 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM1269655 6 0.4184 0.3472 0.484 0.000 0.012 0.000 0.000 0.504
#> GSM1269663 2 0.0363 0.9088 0.000 0.988 0.000 0.000 0.012 0.000
#> GSM1269671 2 0.4593 0.6168 0.000 0.680 0.048 0.000 0.256 0.016
#> GSM1269679 5 0.1232 0.6356 0.000 0.000 0.016 0.004 0.956 0.024
#> GSM1269693 5 0.3857 0.4261 0.000 0.000 0.000 0.468 0.532 0.000
#> GSM1269701 1 0.0000 0.8967 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM1269709 1 0.5915 0.3085 0.556 0.000 0.000 0.292 0.040 0.112
#> GSM1269715 4 0.1053 0.7860 0.000 0.000 0.012 0.964 0.020 0.004
#> GSM1269717 5 0.5616 0.6141 0.000 0.000 0.188 0.196 0.600 0.016
#> GSM1269721 4 0.0790 0.7817 0.000 0.000 0.000 0.968 0.032 0.000
#> GSM1269723 5 0.4109 0.5124 0.000 0.000 0.412 0.012 0.576 0.000
#> GSM1269645 2 0.0000 0.9146 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM1269653 4 0.3376 0.6938 0.000 0.000 0.016 0.764 0.220 0.000
#> GSM1269661 2 0.0000 0.9146 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM1269669 5 0.2520 0.5519 0.000 0.000 0.000 0.004 0.844 0.152
#> GSM1269677 5 0.7158 0.5129 0.000 0.180 0.148 0.092 0.536 0.044
#> GSM1269685 6 0.3899 0.4523 0.404 0.000 0.000 0.004 0.000 0.592
#> GSM1269691 1 0.0632 0.8873 0.976 0.000 0.000 0.000 0.000 0.024
#> GSM1269699 1 0.0000 0.8967 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM1269707 4 0.1141 0.7862 0.000 0.000 0.000 0.948 0.052 0.000
#> GSM1269651 2 0.0000 0.9146 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM1269659 4 0.0790 0.7817 0.000 0.000 0.000 0.968 0.032 0.000
#> GSM1269667 2 0.0000 0.9146 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM1269675 5 0.0405 0.6369 0.000 0.000 0.000 0.004 0.988 0.008
#> GSM1269683 5 0.3388 0.6456 0.000 0.000 0.172 0.036 0.792 0.000
#> GSM1269689 5 0.3409 0.4867 0.000 0.000 0.000 0.300 0.700 0.000
#> GSM1269697 1 0.0790 0.8785 0.968 0.000 0.000 0.000 0.000 0.032
#> GSM1269705 1 0.0000 0.8967 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM1269713 3 0.3515 0.4319 0.000 0.000 0.676 0.324 0.000 0.000
#> GSM1269719 2 0.6053 0.3320 0.000 0.552 0.280 0.000 0.120 0.048
#> GSM1269725 1 0.5662 0.0491 0.440 0.000 0.424 0.004 0.000 0.132
#> GSM1269727 5 0.4467 0.5373 0.000 0.000 0.048 0.320 0.632 0.000
#> GSM1269649 2 0.0146 0.9130 0.000 0.996 0.000 0.000 0.004 0.000
#> GSM1269657 6 0.1845 0.7061 0.008 0.000 0.000 0.004 0.072 0.916
#> GSM1269665 2 0.0000 0.9146 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM1269673 5 0.2738 0.5426 0.000 0.000 0.000 0.004 0.820 0.176
#> GSM1269681 2 0.0622 0.9047 0.000 0.980 0.000 0.012 0.008 0.000
#> GSM1269687 5 0.5062 0.4374 0.076 0.000 0.128 0.004 0.720 0.072
#> GSM1269695 1 0.0000 0.8967 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM1269703 5 0.5571 0.6343 0.076 0.000 0.128 0.132 0.664 0.000
#> GSM1269711 4 0.3619 0.6128 0.000 0.000 0.004 0.680 0.316 0.000
#> GSM1269646 2 0.0000 0.9146 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM1269654 6 0.5053 0.6905 0.136 0.000 0.184 0.012 0.000 0.668
#> GSM1269662 2 0.0000 0.9146 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM1269670 2 0.3221 0.6105 0.000 0.736 0.264 0.000 0.000 0.000
#> GSM1269678 3 0.6085 0.1864 0.000 0.000 0.412 0.004 0.360 0.224
#> GSM1269692 6 0.5425 0.6378 0.272 0.000 0.016 0.112 0.000 0.600
#> GSM1269700 1 0.0725 0.8852 0.976 0.000 0.000 0.000 0.012 0.012
#> GSM1269708 6 0.3290 0.7003 0.208 0.000 0.000 0.016 0.000 0.776
#> GSM1269714 4 0.4147 0.2913 0.000 0.000 0.012 0.552 0.000 0.436
#> GSM1269716 6 0.4079 0.5739 0.000 0.000 0.288 0.032 0.000 0.680
#> GSM1269720 5 0.5087 0.5573 0.000 0.000 0.092 0.348 0.560 0.000
#> GSM1269722 3 0.0622 0.7751 0.000 0.000 0.980 0.012 0.008 0.000
#> GSM1269644 2 0.3186 0.7933 0.000 0.836 0.000 0.004 0.100 0.060
#> GSM1269652 6 0.1701 0.7390 0.072 0.000 0.000 0.008 0.000 0.920
#> GSM1269660 2 0.0000 0.9146 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM1269668 6 0.2913 0.6613 0.004 0.000 0.000 0.004 0.180 0.812
#> GSM1269676 6 0.4436 0.5750 0.000 0.000 0.272 0.044 0.008 0.676
#> GSM1269684 6 0.5617 0.7123 0.104 0.000 0.052 0.044 0.100 0.700
#> GSM1269690 1 0.1814 0.8131 0.900 0.000 0.000 0.000 0.000 0.100
#> GSM1269698 1 0.0260 0.8935 0.992 0.000 0.000 0.000 0.000 0.008
#> GSM1269706 4 0.2789 0.7582 0.020 0.000 0.016 0.872 0.004 0.088
#> GSM1269650 2 0.0146 0.9129 0.000 0.996 0.000 0.004 0.000 0.000
#> GSM1269658 5 0.5209 0.6226 0.000 0.000 0.048 0.172 0.684 0.096
#> GSM1269666 2 0.0000 0.9146 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM1269674 5 0.2697 0.5401 0.000 0.000 0.000 0.000 0.812 0.188
#> GSM1269682 5 0.4161 0.5416 0.000 0.000 0.376 0.012 0.608 0.004
#> GSM1269688 4 0.5316 0.6118 0.172 0.000 0.000 0.672 0.044 0.112
#> GSM1269696 3 0.2258 0.7576 0.060 0.000 0.896 0.000 0.000 0.044
#> GSM1269704 1 0.0000 0.8967 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM1269712 3 0.0000 0.7814 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM1269718 3 0.1075 0.7868 0.000 0.000 0.952 0.000 0.000 0.048
#> GSM1269724 3 0.1007 0.7865 0.000 0.000 0.956 0.000 0.000 0.044
#> GSM1269726 5 0.4157 0.4647 0.000 0.000 0.444 0.012 0.544 0.000
#> GSM1269648 2 0.4765 0.6088 0.000 0.676 0.000 0.000 0.152 0.172
#> GSM1269656 6 0.1387 0.7396 0.068 0.000 0.000 0.000 0.000 0.932
#> GSM1269664 2 0.0000 0.9146 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM1269672 6 0.2805 0.6750 0.012 0.000 0.000 0.000 0.160 0.828
#> GSM1269680 6 0.4610 0.6246 0.000 0.056 0.200 0.012 0.012 0.720
#> GSM1269686 6 0.2631 0.7400 0.180 0.000 0.000 0.000 0.000 0.820
#> GSM1269694 1 0.0000 0.8967 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM1269702 1 0.1075 0.8718 0.952 0.000 0.000 0.000 0.000 0.048
#> GSM1269710 5 0.4010 0.4986 0.000 0.000 0.408 0.008 0.584 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 agent(p) disease.state(p) gender(p) individual(p) k
#> ATC:pam 70 0.00418 0.821 0.1569 7.98e-02 2
#> ATC:pam 73 0.09593 0.845 0.3656 3.67e-04 3
#> ATC:pam 73 0.00526 0.784 0.6219 1.29e-03 4
#> ATC:pam 70 0.01617 0.732 0.0219 5.23e-05 5
#> ATC:pam 71 0.00221 0.651 0.0882 2.48e-04 6
If matrix rows can be associated to genes, consider to use functional_enrichment(res,
...) to perform function enrichment for the signature genes. See this vignette for more detailed explanations.
The object with results only for a single top-value method and a single partition method can be extracted as:
res = res_list["ATC", "mclust"]
# you can also extract it by
# res = res_list["ATC:mclust"]
A summary of res and all the functions that can be applied to it:
res
#> A 'ConsensusPartition' object with k = 2, 3, 4, 5, 6.
#> On a matrix with 51941 rows and 84 columns.
#> Top rows (1000, 2000, 3000, 4000, 5000) are extracted by 'ATC' method.
#> Subgroups are detected by 'mclust' method.
#> Performed in total 1250 partitions by row resampling.
#> Best k for subgroups seems to be 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 0.925 0.942 0.974 0.3522 0.646 0.646
#> 3 3 0.520 0.771 0.854 0.7843 0.652 0.483
#> 4 4 0.575 0.565 0.779 0.1171 0.849 0.611
#> 5 5 0.630 0.432 0.722 0.0754 0.915 0.725
#> 6 6 0.674 0.568 0.760 0.0396 0.870 0.576
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
#> GSM1269647 2 0.0376 0.93795 0.004 0.996
#> GSM1269655 1 0.0000 0.98281 1.000 0.000
#> GSM1269663 2 0.4939 0.88554 0.108 0.892
#> GSM1269671 1 0.0000 0.98281 1.000 0.000
#> GSM1269679 1 0.0000 0.98281 1.000 0.000
#> GSM1269693 1 0.0376 0.98058 0.996 0.004
#> GSM1269701 1 0.0000 0.98281 1.000 0.000
#> GSM1269709 1 0.0000 0.98281 1.000 0.000
#> GSM1269715 1 0.0376 0.98058 0.996 0.004
#> GSM1269717 1 0.0000 0.98281 1.000 0.000
#> GSM1269721 1 0.0376 0.98058 0.996 0.004
#> GSM1269723 1 0.7453 0.71333 0.788 0.212
#> GSM1269645 2 0.0000 0.93517 0.000 1.000
#> GSM1269653 1 0.0000 0.98281 1.000 0.000
#> GSM1269661 2 0.0376 0.93795 0.004 0.996
#> GSM1269669 1 0.0000 0.98281 1.000 0.000
#> GSM1269677 1 0.6973 0.74868 0.812 0.188
#> GSM1269685 1 0.0000 0.98281 1.000 0.000
#> GSM1269691 1 0.0000 0.98281 1.000 0.000
#> GSM1269699 1 0.0000 0.98281 1.000 0.000
#> GSM1269707 1 0.0376 0.98058 0.996 0.004
#> GSM1269651 2 0.0376 0.93795 0.004 0.996
#> GSM1269659 1 0.0376 0.98058 0.996 0.004
#> GSM1269667 2 0.0376 0.93795 0.004 0.996
#> GSM1269675 1 0.0376 0.98058 0.996 0.004
#> GSM1269683 1 0.0376 0.98058 0.996 0.004
#> GSM1269689 1 0.0376 0.98058 0.996 0.004
#> GSM1269697 1 0.0000 0.98281 1.000 0.000
#> GSM1269705 1 0.0000 0.98281 1.000 0.000
#> GSM1269713 1 0.0000 0.98281 1.000 0.000
#> GSM1269719 1 0.0000 0.98281 1.000 0.000
#> GSM1269725 1 0.0000 0.98281 1.000 0.000
#> GSM1269727 1 0.0376 0.98058 0.996 0.004
#> GSM1269649 2 0.0376 0.93795 0.004 0.996
#> GSM1269657 1 0.0000 0.98281 1.000 0.000
#> GSM1269665 2 0.0376 0.93795 0.004 0.996
#> GSM1269673 1 0.0000 0.98281 1.000 0.000
#> GSM1269681 2 0.4562 0.89231 0.096 0.904
#> GSM1269687 1 0.0000 0.98281 1.000 0.000
#> GSM1269695 1 0.0000 0.98281 1.000 0.000
#> GSM1269703 1 0.0000 0.98281 1.000 0.000
#> GSM1269711 1 0.0376 0.98058 0.996 0.004
#> GSM1269646 2 0.0376 0.93795 0.004 0.996
#> GSM1269654 1 0.0000 0.98281 1.000 0.000
#> GSM1269662 2 0.0376 0.93795 0.004 0.996
#> GSM1269670 2 0.4815 0.88596 0.104 0.896
#> GSM1269678 1 0.0000 0.98281 1.000 0.000
#> GSM1269692 1 0.0000 0.98281 1.000 0.000
#> GSM1269700 1 0.0000 0.98281 1.000 0.000
#> GSM1269708 1 0.0000 0.98281 1.000 0.000
#> GSM1269714 1 0.0000 0.98281 1.000 0.000
#> GSM1269716 1 0.0000 0.98281 1.000 0.000
#> GSM1269720 1 0.0376 0.98058 0.996 0.004
#> GSM1269722 1 0.0000 0.98281 1.000 0.000
#> GSM1269644 2 0.4939 0.88554 0.108 0.892
#> GSM1269652 1 0.0000 0.98281 1.000 0.000
#> GSM1269660 2 0.0376 0.93795 0.004 0.996
#> GSM1269668 1 0.0000 0.98281 1.000 0.000
#> GSM1269676 1 0.0000 0.98281 1.000 0.000
#> GSM1269684 1 0.0000 0.98281 1.000 0.000
#> GSM1269690 1 0.0000 0.98281 1.000 0.000
#> GSM1269698 1 0.0000 0.98281 1.000 0.000
#> GSM1269706 1 0.0000 0.98281 1.000 0.000
#> GSM1269650 2 0.0376 0.93795 0.004 0.996
#> GSM1269658 1 0.0376 0.98058 0.996 0.004
#> GSM1269666 2 0.4431 0.89697 0.092 0.908
#> GSM1269674 1 0.0000 0.98281 1.000 0.000
#> GSM1269682 1 0.4161 0.89294 0.916 0.084
#> GSM1269688 1 0.0376 0.98058 0.996 0.004
#> GSM1269696 1 0.0000 0.98281 1.000 0.000
#> GSM1269704 1 0.0000 0.98281 1.000 0.000
#> GSM1269712 2 0.9732 0.35968 0.404 0.596
#> GSM1269718 1 0.0000 0.98281 1.000 0.000
#> GSM1269724 1 0.0000 0.98281 1.000 0.000
#> GSM1269726 1 0.0376 0.98058 0.996 0.004
#> GSM1269648 1 0.9983 -0.00881 0.524 0.476
#> GSM1269656 1 0.0000 0.98281 1.000 0.000
#> GSM1269664 2 0.0376 0.93795 0.004 0.996
#> GSM1269672 1 0.0000 0.98281 1.000 0.000
#> GSM1269680 2 0.6887 0.80639 0.184 0.816
#> GSM1269686 1 0.0000 0.98281 1.000 0.000
#> GSM1269694 1 0.0000 0.98281 1.000 0.000
#> GSM1269702 1 0.0000 0.98281 1.000 0.000
#> GSM1269710 1 0.0000 0.98281 1.000 0.000
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM1269647 2 0.0000 0.8837 0.000 1.000 0.000
#> GSM1269655 1 0.0000 0.8790 1.000 0.000 0.000
#> GSM1269663 2 0.0000 0.8837 0.000 1.000 0.000
#> GSM1269671 2 0.8386 0.4582 0.112 0.584 0.304
#> GSM1269679 3 0.3340 0.7885 0.120 0.000 0.880
#> GSM1269693 3 0.3482 0.8109 0.128 0.000 0.872
#> GSM1269701 1 0.3192 0.7758 0.888 0.000 0.112
#> GSM1269709 1 0.3192 0.7758 0.888 0.000 0.112
#> GSM1269715 3 0.3482 0.8109 0.128 0.000 0.872
#> GSM1269717 3 0.5859 0.6960 0.344 0.000 0.656
#> GSM1269721 3 0.3482 0.8109 0.128 0.000 0.872
#> GSM1269723 3 0.3192 0.7913 0.112 0.000 0.888
#> GSM1269645 2 0.0000 0.8837 0.000 1.000 0.000
#> GSM1269653 1 0.6180 -0.0638 0.584 0.000 0.416
#> GSM1269661 2 0.0000 0.8837 0.000 1.000 0.000
#> GSM1269669 1 0.5859 0.5139 0.656 0.000 0.344
#> GSM1269677 2 0.7909 0.5599 0.112 0.648 0.240
#> GSM1269685 1 0.3192 0.7758 0.888 0.000 0.112
#> GSM1269691 1 0.0000 0.8790 1.000 0.000 0.000
#> GSM1269699 1 0.0000 0.8790 1.000 0.000 0.000
#> GSM1269707 3 0.3482 0.8109 0.128 0.000 0.872
#> GSM1269651 2 0.0000 0.8837 0.000 1.000 0.000
#> GSM1269659 3 0.3752 0.8003 0.144 0.000 0.856
#> GSM1269667 2 0.0000 0.8837 0.000 1.000 0.000
#> GSM1269675 3 0.3192 0.7913 0.112 0.000 0.888
#> GSM1269683 3 0.3192 0.7913 0.112 0.000 0.888
#> GSM1269689 3 0.3482 0.8109 0.128 0.000 0.872
#> GSM1269697 1 0.0000 0.8790 1.000 0.000 0.000
#> GSM1269705 1 0.0000 0.8790 1.000 0.000 0.000
#> GSM1269713 3 0.6299 0.4248 0.476 0.000 0.524
#> GSM1269719 1 0.9496 0.2350 0.492 0.276 0.232
#> GSM1269725 1 0.0237 0.8776 0.996 0.000 0.004
#> GSM1269727 3 0.3482 0.8109 0.128 0.000 0.872
#> GSM1269649 2 0.0000 0.8837 0.000 1.000 0.000
#> GSM1269657 1 0.0747 0.8726 0.984 0.000 0.016
#> GSM1269665 2 0.0000 0.8837 0.000 1.000 0.000
#> GSM1269673 3 0.6260 0.1908 0.448 0.000 0.552
#> GSM1269681 2 0.2959 0.8359 0.000 0.900 0.100
#> GSM1269687 1 0.4654 0.7353 0.792 0.000 0.208
#> GSM1269695 1 0.0000 0.8790 1.000 0.000 0.000
#> GSM1269703 3 0.6192 0.5668 0.420 0.000 0.580
#> GSM1269711 3 0.5016 0.8006 0.240 0.000 0.760
#> GSM1269646 2 0.0000 0.8837 0.000 1.000 0.000
#> GSM1269654 1 0.0237 0.8777 0.996 0.000 0.004
#> GSM1269662 2 0.0000 0.8837 0.000 1.000 0.000
#> GSM1269670 2 0.2959 0.8359 0.000 0.900 0.100
#> GSM1269678 1 0.4555 0.7432 0.800 0.000 0.200
#> GSM1269692 1 0.0000 0.8790 1.000 0.000 0.000
#> GSM1269700 1 0.2448 0.8322 0.924 0.000 0.076
#> GSM1269708 1 0.0000 0.8790 1.000 0.000 0.000
#> GSM1269714 1 0.0000 0.8790 1.000 0.000 0.000
#> GSM1269716 1 0.2066 0.8449 0.940 0.000 0.060
#> GSM1269720 3 0.3686 0.8191 0.140 0.000 0.860
#> GSM1269722 3 0.5621 0.7474 0.308 0.000 0.692
#> GSM1269644 2 0.0000 0.8837 0.000 1.000 0.000
#> GSM1269652 1 0.0000 0.8790 1.000 0.000 0.000
#> GSM1269660 2 0.0000 0.8837 0.000 1.000 0.000
#> GSM1269668 1 0.3482 0.7850 0.872 0.000 0.128
#> GSM1269676 1 0.5012 0.7326 0.788 0.008 0.204
#> GSM1269684 1 0.4178 0.7651 0.828 0.000 0.172
#> GSM1269690 1 0.0000 0.8790 1.000 0.000 0.000
#> GSM1269698 1 0.0000 0.8790 1.000 0.000 0.000
#> GSM1269706 1 0.3482 0.7879 0.872 0.000 0.128
#> GSM1269650 2 0.0000 0.8837 0.000 1.000 0.000
#> GSM1269658 3 0.3192 0.7913 0.112 0.000 0.888
#> GSM1269666 2 0.0000 0.8837 0.000 1.000 0.000
#> GSM1269674 1 0.5678 0.5745 0.684 0.000 0.316
#> GSM1269682 2 0.8496 0.4185 0.112 0.564 0.324
#> GSM1269688 1 0.3192 0.7758 0.888 0.000 0.112
#> GSM1269696 1 0.2448 0.8322 0.924 0.000 0.076
#> GSM1269704 1 0.0000 0.8790 1.000 0.000 0.000
#> GSM1269712 2 0.8047 0.5378 0.112 0.632 0.256
#> GSM1269718 2 0.9627 0.1473 0.364 0.428 0.208
#> GSM1269724 1 0.2537 0.8286 0.920 0.000 0.080
#> GSM1269726 3 0.4887 0.8051 0.228 0.000 0.772
#> GSM1269648 2 0.1860 0.8440 0.052 0.948 0.000
#> GSM1269656 1 0.0000 0.8790 1.000 0.000 0.000
#> GSM1269664 2 0.0000 0.8837 0.000 1.000 0.000
#> GSM1269672 1 0.3482 0.7850 0.872 0.000 0.128
#> GSM1269680 2 0.2959 0.8359 0.000 0.900 0.100
#> GSM1269686 1 0.0000 0.8790 1.000 0.000 0.000
#> GSM1269694 1 0.0000 0.8790 1.000 0.000 0.000
#> GSM1269702 1 0.0000 0.8790 1.000 0.000 0.000
#> GSM1269710 3 0.3192 0.7913 0.112 0.000 0.888
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM1269647 2 0.0336 0.880135 0.000 0.992 0.000 0.008
#> GSM1269655 1 0.0707 0.820685 0.980 0.000 0.020 0.000
#> GSM1269663 2 0.1716 0.838146 0.000 0.936 0.064 0.000
#> GSM1269671 3 0.4535 0.296384 0.000 0.292 0.704 0.004
#> GSM1269679 3 0.3764 0.166248 0.000 0.000 0.784 0.216
#> GSM1269693 4 0.3311 0.717784 0.000 0.000 0.172 0.828
#> GSM1269701 1 0.6931 0.448868 0.588 0.000 0.184 0.228
#> GSM1269709 1 0.5464 0.614781 0.708 0.000 0.064 0.228
#> GSM1269715 4 0.4477 0.752229 0.000 0.000 0.312 0.688
#> GSM1269717 3 0.6917 0.039623 0.288 0.000 0.568 0.144
#> GSM1269721 4 0.1557 0.651172 0.000 0.000 0.056 0.944
#> GSM1269723 3 0.4222 0.048186 0.000 0.000 0.728 0.272
#> GSM1269645 2 0.0000 0.880583 0.000 1.000 0.000 0.000
#> GSM1269653 3 0.7282 0.032577 0.416 0.000 0.436 0.148
#> GSM1269661 2 0.0336 0.880135 0.000 0.992 0.000 0.008
#> GSM1269669 3 0.5610 0.364829 0.176 0.000 0.720 0.104
#> GSM1269677 3 0.4855 -0.107753 0.000 0.400 0.600 0.000
#> GSM1269685 1 0.5394 0.615761 0.712 0.000 0.060 0.228
#> GSM1269691 1 0.1118 0.821024 0.964 0.000 0.036 0.000
#> GSM1269699 1 0.0000 0.817646 1.000 0.000 0.000 0.000
#> GSM1269707 4 0.4477 0.752229 0.000 0.000 0.312 0.688
#> GSM1269651 2 0.0336 0.879248 0.000 0.992 0.008 0.000
#> GSM1269659 4 0.1557 0.651172 0.000 0.000 0.056 0.944
#> GSM1269667 2 0.0336 0.880135 0.000 0.992 0.000 0.008
#> GSM1269675 3 0.4406 -0.000840 0.000 0.000 0.700 0.300
#> GSM1269683 3 0.4406 -0.000840 0.000 0.000 0.700 0.300
#> GSM1269689 4 0.4477 0.752229 0.000 0.000 0.312 0.688
#> GSM1269697 1 0.2589 0.792068 0.884 0.000 0.116 0.000
#> GSM1269705 1 0.0336 0.817726 0.992 0.000 0.008 0.000
#> GSM1269713 3 0.7448 0.023352 0.400 0.000 0.428 0.172
#> GSM1269719 3 0.6975 0.386150 0.200 0.216 0.584 0.000
#> GSM1269725 1 0.3649 0.709453 0.796 0.000 0.204 0.000
#> GSM1269727 4 0.4477 0.752229 0.000 0.000 0.312 0.688
#> GSM1269649 2 0.0336 0.880135 0.000 0.992 0.000 0.008
#> GSM1269657 1 0.3569 0.695287 0.804 0.000 0.196 0.000
#> GSM1269665 2 0.0000 0.880583 0.000 1.000 0.000 0.000
#> GSM1269673 3 0.5314 0.325009 0.108 0.000 0.748 0.144
#> GSM1269681 2 0.3219 0.758163 0.000 0.836 0.164 0.000
#> GSM1269687 3 0.4888 0.127503 0.412 0.000 0.588 0.000
#> GSM1269695 1 0.1716 0.815128 0.936 0.000 0.064 0.000
#> GSM1269703 3 0.7825 -0.178400 0.284 0.000 0.412 0.304
#> GSM1269711 4 0.7648 0.360271 0.208 0.000 0.392 0.400
#> GSM1269646 2 0.0000 0.880583 0.000 1.000 0.000 0.000
#> GSM1269654 1 0.1716 0.816187 0.936 0.000 0.064 0.000
#> GSM1269662 2 0.0000 0.880583 0.000 1.000 0.000 0.000
#> GSM1269670 2 0.3498 0.761272 0.000 0.832 0.160 0.008
#> GSM1269678 3 0.4955 0.055621 0.444 0.000 0.556 0.000
#> GSM1269692 1 0.1637 0.816606 0.940 0.000 0.060 0.000
#> GSM1269700 1 0.3569 0.714610 0.804 0.000 0.196 0.000
#> GSM1269708 1 0.0000 0.817646 1.000 0.000 0.000 0.000
#> GSM1269714 1 0.1637 0.816606 0.940 0.000 0.060 0.000
#> GSM1269716 1 0.2149 0.809915 0.912 0.000 0.088 0.000
#> GSM1269720 4 0.4908 0.557092 0.016 0.000 0.292 0.692
#> GSM1269722 3 0.7617 -0.017806 0.372 0.000 0.424 0.204
#> GSM1269644 2 0.3266 0.784609 0.000 0.832 0.168 0.000
#> GSM1269652 1 0.0188 0.816996 0.996 0.000 0.004 0.000
#> GSM1269660 2 0.0000 0.880583 0.000 1.000 0.000 0.000
#> GSM1269668 1 0.4843 0.165953 0.604 0.000 0.396 0.000
#> GSM1269676 3 0.5400 0.199838 0.372 0.020 0.608 0.000
#> GSM1269684 1 0.4967 0.123632 0.548 0.000 0.452 0.000
#> GSM1269690 1 0.0188 0.816996 0.996 0.000 0.004 0.000
#> GSM1269698 1 0.0000 0.817646 1.000 0.000 0.000 0.000
#> GSM1269706 1 0.2412 0.806863 0.908 0.000 0.084 0.008
#> GSM1269650 2 0.0817 0.873711 0.000 0.976 0.024 0.000
#> GSM1269658 3 0.4996 0.000766 0.000 0.000 0.516 0.484
#> GSM1269666 2 0.0469 0.877752 0.000 0.988 0.012 0.000
#> GSM1269674 3 0.5576 0.067363 0.444 0.000 0.536 0.020
#> GSM1269682 2 0.5296 0.148635 0.000 0.500 0.492 0.008
#> GSM1269688 1 0.6124 0.532164 0.640 0.000 0.084 0.276
#> GSM1269696 1 0.3726 0.698511 0.788 0.000 0.212 0.000
#> GSM1269704 1 0.0000 0.817646 1.000 0.000 0.000 0.000
#> GSM1269712 2 0.5161 0.203873 0.000 0.520 0.476 0.004
#> GSM1269718 3 0.6411 0.235991 0.092 0.308 0.600 0.000
#> GSM1269724 1 0.3801 0.689882 0.780 0.000 0.220 0.000
#> GSM1269726 4 0.7251 0.452931 0.144 0.000 0.416 0.440
#> GSM1269648 2 0.3754 0.764324 0.064 0.852 0.084 0.000
#> GSM1269656 1 0.1637 0.817928 0.940 0.000 0.060 0.000
#> GSM1269664 2 0.0000 0.880583 0.000 1.000 0.000 0.000
#> GSM1269672 1 0.4843 0.167792 0.604 0.000 0.396 0.000
#> GSM1269680 2 0.4697 0.594575 0.000 0.644 0.356 0.000
#> GSM1269686 1 0.0188 0.817945 0.996 0.000 0.004 0.000
#> GSM1269694 1 0.0000 0.817646 1.000 0.000 0.000 0.000
#> GSM1269702 1 0.0188 0.816996 0.996 0.000 0.004 0.000
#> GSM1269710 3 0.3074 0.201932 0.000 0.000 0.848 0.152
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM1269647 2 0.0290 0.7907 0.000 0.992 0.008 0.000 0.000
#> GSM1269655 5 0.0290 0.8120 0.000 0.000 0.008 0.000 0.992
#> GSM1269663 2 0.1908 0.7610 0.000 0.908 0.092 0.000 0.000
#> GSM1269671 1 0.5335 0.0502 0.548 0.032 0.012 0.408 0.000
#> GSM1269679 1 0.4192 0.1506 0.596 0.000 0.000 0.404 0.000
#> GSM1269693 4 0.6473 -0.0573 0.188 0.000 0.364 0.448 0.000
#> GSM1269701 1 0.6866 0.1342 0.380 0.000 0.252 0.004 0.364
#> GSM1269709 5 0.3452 0.6147 0.000 0.000 0.244 0.000 0.756
#> GSM1269715 1 0.6682 0.2365 0.396 0.000 0.368 0.236 0.000
#> GSM1269717 1 0.1915 0.3703 0.928 0.000 0.000 0.040 0.032
#> GSM1269721 4 0.5221 0.0918 0.048 0.000 0.400 0.552 0.000
#> GSM1269723 1 0.3814 0.2853 0.720 0.000 0.004 0.276 0.000
#> GSM1269645 2 0.0162 0.7910 0.000 0.996 0.004 0.000 0.000
#> GSM1269653 1 0.4182 0.2986 0.600 0.000 0.000 0.000 0.400
#> GSM1269661 2 0.0000 0.7911 0.000 1.000 0.000 0.000 0.000
#> GSM1269669 1 0.4902 0.1136 0.564 0.000 0.000 0.408 0.028
#> GSM1269677 4 0.6726 -0.6518 0.004 0.208 0.388 0.400 0.000
#> GSM1269685 5 0.3452 0.6115 0.000 0.000 0.244 0.000 0.756
#> GSM1269691 5 0.0510 0.8115 0.000 0.000 0.016 0.000 0.984
#> GSM1269699 5 0.0794 0.8110 0.000 0.000 0.028 0.000 0.972
#> GSM1269707 1 0.6682 0.2365 0.396 0.000 0.368 0.236 0.000
#> GSM1269651 2 0.2605 0.7310 0.000 0.852 0.148 0.000 0.000
#> GSM1269659 4 0.5229 0.0928 0.048 0.000 0.404 0.548 0.000
#> GSM1269667 2 0.0162 0.7910 0.000 0.996 0.004 0.000 0.000
#> GSM1269675 4 0.4211 -0.1448 0.360 0.000 0.004 0.636 0.000
#> GSM1269683 4 0.4341 -0.1733 0.404 0.000 0.004 0.592 0.000
#> GSM1269689 1 0.6680 0.2362 0.400 0.000 0.364 0.236 0.000
#> GSM1269697 5 0.5806 0.4952 0.212 0.000 0.144 0.008 0.636
#> GSM1269705 5 0.0794 0.8110 0.000 0.000 0.028 0.000 0.972
#> GSM1269713 1 0.5043 0.3173 0.600 0.000 0.044 0.000 0.356
#> GSM1269719 4 0.9016 -0.5477 0.268 0.144 0.156 0.376 0.056
#> GSM1269725 5 0.6029 0.4498 0.236 0.000 0.152 0.008 0.604
#> GSM1269727 1 0.6680 0.2362 0.400 0.000 0.364 0.236 0.000
#> GSM1269649 2 0.0290 0.7907 0.000 0.992 0.008 0.000 0.000
#> GSM1269657 5 0.3052 0.7366 0.036 0.000 0.016 0.072 0.876
#> GSM1269665 2 0.0162 0.7910 0.000 0.996 0.004 0.000 0.000
#> GSM1269673 1 0.5175 0.0824 0.548 0.000 0.000 0.408 0.044
#> GSM1269681 2 0.3010 0.6324 0.004 0.824 0.000 0.172 0.000
#> GSM1269687 4 0.7452 -0.1592 0.240 0.000 0.040 0.412 0.308
#> GSM1269695 5 0.0510 0.8125 0.000 0.000 0.016 0.000 0.984
#> GSM1269703 1 0.3109 0.3734 0.800 0.000 0.000 0.000 0.200
#> GSM1269711 1 0.3565 0.3488 0.816 0.000 0.000 0.144 0.040
#> GSM1269646 2 0.0000 0.7911 0.000 1.000 0.000 0.000 0.000
#> GSM1269654 5 0.0798 0.8094 0.000 0.000 0.016 0.008 0.976
#> GSM1269662 2 0.0162 0.7910 0.000 0.996 0.004 0.000 0.000
#> GSM1269670 2 0.3010 0.6324 0.004 0.824 0.000 0.172 0.000
#> GSM1269678 4 0.7478 -0.1626 0.264 0.000 0.036 0.380 0.320
#> GSM1269692 5 0.0510 0.8109 0.000 0.000 0.016 0.000 0.984
#> GSM1269700 5 0.5468 0.3674 0.332 0.000 0.060 0.008 0.600
#> GSM1269708 5 0.0290 0.8132 0.000 0.000 0.008 0.000 0.992
#> GSM1269714 5 0.0510 0.8109 0.000 0.000 0.016 0.000 0.984
#> GSM1269716 5 0.2054 0.7857 0.052 0.000 0.028 0.000 0.920
#> GSM1269720 4 0.5257 -0.0863 0.452 0.000 0.020 0.512 0.016
#> GSM1269722 1 0.5394 0.3300 0.604 0.000 0.064 0.004 0.328
#> GSM1269644 2 0.4161 0.4612 0.000 0.608 0.392 0.000 0.000
#> GSM1269652 5 0.0000 0.8131 0.000 0.000 0.000 0.000 1.000
#> GSM1269660 2 0.0162 0.7910 0.000 0.996 0.004 0.000 0.000
#> GSM1269668 5 0.4649 0.1662 0.000 0.000 0.016 0.404 0.580
#> GSM1269676 3 0.7735 0.5316 0.072 0.008 0.404 0.368 0.148
#> GSM1269684 5 0.5144 0.1615 0.020 0.000 0.016 0.384 0.580
#> GSM1269690 5 0.0290 0.8120 0.000 0.000 0.008 0.000 0.992
#> GSM1269698 5 0.0880 0.8102 0.000 0.000 0.032 0.000 0.968
#> GSM1269706 5 0.1725 0.7943 0.044 0.000 0.020 0.000 0.936
#> GSM1269650 2 0.4088 0.4960 0.000 0.632 0.368 0.000 0.000
#> GSM1269658 4 0.1768 -0.1450 0.004 0.000 0.072 0.924 0.000
#> GSM1269666 2 0.2471 0.7382 0.000 0.864 0.136 0.000 0.000
#> GSM1269674 4 0.7013 -0.1577 0.244 0.000 0.012 0.412 0.332
#> GSM1269682 2 0.6333 -0.2432 0.124 0.468 0.008 0.400 0.000
#> GSM1269688 5 0.5080 0.4010 0.048 0.000 0.348 0.000 0.604
#> GSM1269696 5 0.6352 0.3941 0.268 0.000 0.172 0.008 0.552
#> GSM1269704 5 0.0794 0.8114 0.000 0.000 0.028 0.000 0.972
#> GSM1269712 2 0.6385 -0.0427 0.252 0.516 0.000 0.232 0.000
#> GSM1269718 3 0.8382 0.5465 0.180 0.152 0.340 0.324 0.004
#> GSM1269724 5 0.6534 0.3744 0.276 0.004 0.172 0.008 0.540
#> GSM1269726 1 0.4712 0.2904 0.716 0.000 0.028 0.236 0.020
#> GSM1269648 2 0.6732 0.4204 0.108 0.624 0.164 0.004 0.100
#> GSM1269656 5 0.0566 0.8111 0.000 0.000 0.012 0.004 0.984
#> GSM1269664 2 0.0162 0.7910 0.000 0.996 0.004 0.000 0.000
#> GSM1269672 5 0.4789 0.1627 0.004 0.000 0.016 0.400 0.580
#> GSM1269680 2 0.6550 0.0498 0.004 0.436 0.388 0.172 0.000
#> GSM1269686 5 0.0510 0.8129 0.000 0.000 0.016 0.000 0.984
#> GSM1269694 5 0.0290 0.8132 0.000 0.000 0.008 0.000 0.992
#> GSM1269702 5 0.0162 0.8134 0.000 0.000 0.004 0.000 0.996
#> GSM1269710 1 0.2424 0.3365 0.868 0.000 0.000 0.132 0.000
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM1269647 2 0.3986 0.7079 0.000 0.532 0.000 0.000 0.004 0.464
#> GSM1269655 1 0.0000 0.8379 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM1269663 6 0.3314 -0.0598 0.000 0.256 0.004 0.000 0.000 0.740
#> GSM1269671 3 0.5372 0.5294 0.000 0.388 0.520 0.012 0.080 0.000
#> GSM1269679 3 0.5823 0.5478 0.000 0.336 0.540 0.068 0.056 0.000
#> GSM1269693 3 0.3810 -0.6062 0.000 0.000 0.572 0.428 0.000 0.000
#> GSM1269701 3 0.4977 0.2685 0.372 0.000 0.552 0.000 0.076 0.000
#> GSM1269709 1 0.3428 0.6806 0.796 0.004 0.176 0.016 0.008 0.000
#> GSM1269715 3 0.1078 0.4017 0.000 0.016 0.964 0.012 0.008 0.000
#> GSM1269717 3 0.6917 0.5728 0.068 0.136 0.584 0.128 0.084 0.000
#> GSM1269721 4 0.3899 0.6649 0.000 0.000 0.404 0.592 0.004 0.000
#> GSM1269723 3 0.5688 0.5355 0.000 0.224 0.608 0.136 0.032 0.000
#> GSM1269645 2 0.4724 0.7624 0.000 0.588 0.000 0.028 0.016 0.368
#> GSM1269653 3 0.4815 0.3963 0.384 0.000 0.556 0.000 0.060 0.000
#> GSM1269661 2 0.4159 0.7644 0.000 0.588 0.000 0.016 0.000 0.396
#> GSM1269669 3 0.6459 0.5306 0.000 0.192 0.532 0.212 0.064 0.000
#> GSM1269677 6 0.4397 0.4674 0.000 0.376 0.000 0.024 0.004 0.596
#> GSM1269685 1 0.3428 0.6806 0.796 0.004 0.176 0.016 0.008 0.000
#> GSM1269691 1 0.1476 0.8157 0.948 0.004 0.028 0.012 0.008 0.000
#> GSM1269699 1 0.0146 0.8380 0.996 0.000 0.000 0.000 0.004 0.000
#> GSM1269707 3 0.0767 0.4186 0.004 0.000 0.976 0.012 0.008 0.000
#> GSM1269651 6 0.0865 0.4847 0.000 0.036 0.000 0.000 0.000 0.964
#> GSM1269659 4 0.3774 0.6664 0.000 0.000 0.408 0.592 0.000 0.000
#> GSM1269667 2 0.3944 0.7451 0.000 0.568 0.000 0.000 0.004 0.428
#> GSM1269675 3 0.5327 0.5201 0.000 0.164 0.588 0.248 0.000 0.000
#> GSM1269683 3 0.5306 0.5563 0.000 0.268 0.596 0.132 0.004 0.000
#> GSM1269689 3 0.0820 0.4010 0.000 0.016 0.972 0.012 0.000 0.000
#> GSM1269697 5 0.3198 0.7431 0.260 0.000 0.000 0.000 0.740 0.000
#> GSM1269705 1 0.0291 0.8374 0.992 0.004 0.000 0.000 0.004 0.000
#> GSM1269713 3 0.5524 0.4281 0.204 0.000 0.560 0.000 0.236 0.000
#> GSM1269719 6 0.6669 0.4001 0.024 0.360 0.028 0.036 0.060 0.492
#> GSM1269725 5 0.3163 0.7650 0.232 0.000 0.004 0.000 0.764 0.000
#> GSM1269727 3 0.0820 0.4010 0.000 0.016 0.972 0.012 0.000 0.000
#> GSM1269649 2 0.3971 0.7274 0.000 0.548 0.000 0.000 0.004 0.448
#> GSM1269657 1 0.1780 0.8028 0.924 0.048 0.000 0.028 0.000 0.000
#> GSM1269665 2 0.4724 0.7624 0.000 0.588 0.000 0.028 0.016 0.368
#> GSM1269673 3 0.6543 0.5309 0.004 0.192 0.532 0.212 0.060 0.000
#> GSM1269681 2 0.3240 0.5651 0.000 0.752 0.000 0.000 0.004 0.244
#> GSM1269687 3 0.6278 0.5161 0.044 0.384 0.480 0.020 0.072 0.000
#> GSM1269695 1 0.0000 0.8379 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM1269703 3 0.5452 0.5382 0.160 0.008 0.684 0.076 0.072 0.000
#> GSM1269711 3 0.4542 0.5198 0.088 0.000 0.748 0.128 0.036 0.000
#> GSM1269646 2 0.4228 0.7656 0.000 0.588 0.000 0.020 0.000 0.392
#> GSM1269654 1 0.1141 0.8150 0.948 0.052 0.000 0.000 0.000 0.000
#> GSM1269662 2 0.4724 0.7624 0.000 0.588 0.000 0.028 0.016 0.368
#> GSM1269670 2 0.3217 0.5824 0.000 0.768 0.000 0.000 0.008 0.224
#> GSM1269678 1 0.6471 0.2303 0.480 0.364 0.028 0.036 0.092 0.000
#> GSM1269692 1 0.0146 0.8377 0.996 0.000 0.000 0.004 0.000 0.000
#> GSM1269700 3 0.5856 0.1751 0.404 0.000 0.404 0.000 0.192 0.000
#> GSM1269708 1 0.0260 0.8362 0.992 0.000 0.000 0.000 0.008 0.000
#> GSM1269714 1 0.0146 0.8377 0.996 0.000 0.000 0.004 0.000 0.000
#> GSM1269716 1 0.3156 0.6556 0.800 0.000 0.020 0.000 0.180 0.000
#> GSM1269720 4 0.5998 0.4406 0.032 0.060 0.332 0.548 0.028 0.000
#> GSM1269722 3 0.5605 0.4397 0.180 0.000 0.560 0.004 0.256 0.000
#> GSM1269644 6 0.0000 0.5158 0.000 0.000 0.000 0.000 0.000 1.000
#> GSM1269652 1 0.0146 0.8380 0.996 0.000 0.000 0.000 0.004 0.000
#> GSM1269660 2 0.4228 0.7656 0.000 0.588 0.000 0.020 0.000 0.392
#> GSM1269668 1 0.5303 0.4504 0.600 0.204 0.000 0.196 0.000 0.000
#> GSM1269676 6 0.5069 0.4581 0.000 0.360 0.016 0.036 0.008 0.580
#> GSM1269684 1 0.5350 0.4450 0.592 0.196 0.000 0.212 0.000 0.000
#> GSM1269690 1 0.0000 0.8379 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM1269698 1 0.2320 0.7314 0.864 0.004 0.000 0.000 0.132 0.000
#> GSM1269706 1 0.2629 0.7441 0.872 0.000 0.068 0.000 0.060 0.000
#> GSM1269650 6 0.0000 0.5158 0.000 0.000 0.000 0.000 0.000 1.000
#> GSM1269658 4 0.2703 0.3874 0.000 0.172 0.004 0.824 0.000 0.000
#> GSM1269666 6 0.2772 0.2149 0.000 0.180 0.000 0.000 0.004 0.816
#> GSM1269674 3 0.8087 0.4107 0.148 0.192 0.396 0.212 0.052 0.000
#> GSM1269682 2 0.5052 -0.3424 0.000 0.628 0.304 0.012 0.036 0.020
#> GSM1269688 1 0.4529 0.3577 0.576 0.004 0.396 0.016 0.008 0.000
#> GSM1269696 5 0.1219 0.7520 0.048 0.000 0.004 0.000 0.948 0.000
#> GSM1269704 1 0.0291 0.8374 0.992 0.004 0.000 0.000 0.004 0.000
#> GSM1269712 2 0.3704 0.1677 0.000 0.820 0.080 0.044 0.056 0.000
#> GSM1269718 6 0.6498 0.4236 0.024 0.344 0.032 0.040 0.040 0.520
#> GSM1269724 5 0.1152 0.7470 0.044 0.000 0.004 0.000 0.952 0.000
#> GSM1269726 3 0.4714 0.4975 0.032 0.060 0.764 0.108 0.036 0.000
#> GSM1269648 6 0.3354 0.3945 0.008 0.016 0.000 0.000 0.184 0.792
#> GSM1269656 1 0.0146 0.8378 0.996 0.004 0.000 0.000 0.000 0.000
#> GSM1269664 2 0.4228 0.7656 0.000 0.588 0.000 0.020 0.000 0.392
#> GSM1269672 1 0.4432 0.4313 0.600 0.364 0.000 0.036 0.000 0.000
#> GSM1269680 6 0.2946 0.5424 0.000 0.184 0.000 0.004 0.004 0.808
#> GSM1269686 1 0.0146 0.8380 0.996 0.000 0.000 0.000 0.004 0.000
#> GSM1269694 1 0.0146 0.8380 0.996 0.000 0.000 0.000 0.004 0.000
#> GSM1269702 1 0.0291 0.8374 0.992 0.004 0.000 0.000 0.004 0.000
#> GSM1269710 3 0.6052 0.5577 0.000 0.268 0.568 0.084 0.080 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 agent(p) disease.state(p) gender(p) individual(p) k
#> ATC:mclust 82 1.0000 1.000 0.659 0.000327 2
#> ATC:mclust 77 0.0238 0.719 0.065 0.000703 3
#> ATC:mclust 55 0.0352 0.766 0.145 0.007509 4
#> ATC:mclust 39 0.1608 0.794 0.753 0.011297 5
#> ATC:mclust 55 0.0105 0.250 0.194 0.001276 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 51941 rows and 84 columns.
#> Top rows (1000, 2000, 3000, 4000, 5000) are extracted by 'ATC' method.
#> Subgroups are detected by 'NMF' method.
#> Performed in total 1250 partitions by row resampling.
#> Best k for subgroups seems to be 2.
#>
#> Following methods can be applied to this 'ConsensusPartition' object:
#> [1] "cola_report" "collect_classes" "collect_plots"
#> [4] "collect_stats" "colnames" "compare_signatures"
#> [7] "consensus_heatmap" "dimension_reduction" "functional_enrichment"
#> [10] "get_anno_col" "get_anno" "get_classes"
#> [13] "get_consensus" "get_matrix" "get_membership"
#> [16] "get_param" "get_signatures" "get_stats"
#> [19] "is_best_k" "is_stable_k" "membership_heatmap"
#> [22] "ncol" "nrow" "plot_ecdf"
#> [25] "rownames" "select_partition_number" "show"
#> [28] "suggest_best_k" "test_to_known_factors"
collect_plots() function collects all the plots made from res for all k (number of partitions)
into one single page to provide an easy and fast comparison between different k.
collect_plots(res)
The plots are:
k and the heatmap of
predicted classes for each k.k.k.k.All the plots in panels can be made by individual functions and they are plotted later in this section.
select_partition_number() produces several plots showing different
statistics for choosing “optimized” k. There are following statistics:
k;k, the area increased is defined as \(A_k - A_{k-1}\).The detailed explanations of these statistics can be found in the cola vignette.
Generally speaking, lower PAC score, higher mean silhouette score or higher
concordance corresponds to better partition. Rand index and Jaccard index
measure how similar the current partition is compared to partition with k-1.
If they are too similar, we won't accept k is better than k-1.
select_partition_number(res)
The numeric values for all these statistics can be obtained by get_stats().
get_stats(res)
#> k 1-PAC mean_silhouette concordance area_increased Rand Jaccard
#> 2 2 0.950 0.937 0.974 0.4792 0.523 0.523
#> 3 3 0.667 0.749 0.891 0.3920 0.677 0.451
#> 4 4 0.449 0.482 0.719 0.1214 0.764 0.419
#> 5 5 0.502 0.410 0.662 0.0660 0.838 0.462
#> 6 6 0.542 0.362 0.623 0.0416 0.826 0.368
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
#> GSM1269647 2 0.0000 0.974 0.000 1.000
#> GSM1269655 1 0.0000 0.971 1.000 0.000
#> GSM1269663 2 0.0000 0.974 0.000 1.000
#> GSM1269671 2 0.0000 0.974 0.000 1.000
#> GSM1269679 1 0.7745 0.708 0.772 0.228
#> GSM1269693 1 0.0000 0.971 1.000 0.000
#> GSM1269701 1 0.0000 0.971 1.000 0.000
#> GSM1269709 1 0.0000 0.971 1.000 0.000
#> GSM1269715 1 0.0000 0.971 1.000 0.000
#> GSM1269717 1 0.5629 0.841 0.868 0.132
#> GSM1269721 1 0.0000 0.971 1.000 0.000
#> GSM1269723 2 0.0000 0.974 0.000 1.000
#> GSM1269645 2 0.0000 0.974 0.000 1.000
#> GSM1269653 1 0.0000 0.971 1.000 0.000
#> GSM1269661 2 0.0000 0.974 0.000 1.000
#> GSM1269669 1 0.0000 0.971 1.000 0.000
#> GSM1269677 2 0.0000 0.974 0.000 1.000
#> GSM1269685 1 0.0000 0.971 1.000 0.000
#> GSM1269691 1 0.0000 0.971 1.000 0.000
#> GSM1269699 1 0.0000 0.971 1.000 0.000
#> GSM1269707 1 0.0000 0.971 1.000 0.000
#> GSM1269651 2 0.0000 0.974 0.000 1.000
#> GSM1269659 1 0.0000 0.971 1.000 0.000
#> GSM1269667 2 0.0000 0.974 0.000 1.000
#> GSM1269675 1 0.2603 0.934 0.956 0.044
#> GSM1269683 2 0.3114 0.928 0.056 0.944
#> GSM1269689 1 0.0000 0.971 1.000 0.000
#> GSM1269697 1 0.0000 0.971 1.000 0.000
#> GSM1269705 1 0.0000 0.971 1.000 0.000
#> GSM1269713 1 0.0000 0.971 1.000 0.000
#> GSM1269719 2 0.2236 0.946 0.036 0.964
#> GSM1269725 1 0.0000 0.971 1.000 0.000
#> GSM1269727 1 0.6438 0.801 0.836 0.164
#> GSM1269649 2 0.0000 0.974 0.000 1.000
#> GSM1269657 1 0.0000 0.971 1.000 0.000
#> GSM1269665 2 0.0000 0.974 0.000 1.000
#> GSM1269673 1 0.2423 0.938 0.960 0.040
#> GSM1269681 2 0.0000 0.974 0.000 1.000
#> GSM1269687 1 0.0376 0.968 0.996 0.004
#> GSM1269695 1 0.0000 0.971 1.000 0.000
#> GSM1269703 1 0.0000 0.971 1.000 0.000
#> GSM1269711 1 0.0000 0.971 1.000 0.000
#> GSM1269646 2 0.0000 0.974 0.000 1.000
#> GSM1269654 1 0.0000 0.971 1.000 0.000
#> GSM1269662 2 0.0000 0.974 0.000 1.000
#> GSM1269670 2 0.0000 0.974 0.000 1.000
#> GSM1269678 1 0.0000 0.971 1.000 0.000
#> GSM1269692 1 0.0000 0.971 1.000 0.000
#> GSM1269700 1 0.0000 0.971 1.000 0.000
#> GSM1269708 1 0.0000 0.971 1.000 0.000
#> GSM1269714 1 0.0000 0.971 1.000 0.000
#> GSM1269716 1 0.0000 0.971 1.000 0.000
#> GSM1269720 2 0.9881 0.196 0.436 0.564
#> GSM1269722 1 0.9775 0.310 0.588 0.412
#> GSM1269644 2 0.0000 0.974 0.000 1.000
#> GSM1269652 1 0.0000 0.971 1.000 0.000
#> GSM1269660 2 0.0000 0.974 0.000 1.000
#> GSM1269668 1 0.0000 0.971 1.000 0.000
#> GSM1269676 2 0.5946 0.825 0.144 0.856
#> GSM1269684 1 0.0000 0.971 1.000 0.000
#> GSM1269690 1 0.0000 0.971 1.000 0.000
#> GSM1269698 1 0.0000 0.971 1.000 0.000
#> GSM1269706 1 0.0000 0.971 1.000 0.000
#> GSM1269650 2 0.0000 0.974 0.000 1.000
#> GSM1269658 1 0.9580 0.398 0.620 0.380
#> GSM1269666 2 0.0000 0.974 0.000 1.000
#> GSM1269674 1 0.0000 0.971 1.000 0.000
#> GSM1269682 2 0.0000 0.974 0.000 1.000
#> GSM1269688 1 0.0000 0.971 1.000 0.000
#> GSM1269696 1 0.0938 0.962 0.988 0.012
#> GSM1269704 1 0.0000 0.971 1.000 0.000
#> GSM1269712 2 0.0000 0.974 0.000 1.000
#> GSM1269718 2 0.1184 0.962 0.016 0.984
#> GSM1269724 2 0.4161 0.899 0.084 0.916
#> GSM1269726 2 0.0000 0.974 0.000 1.000
#> GSM1269648 2 0.0000 0.974 0.000 1.000
#> GSM1269656 1 0.0000 0.971 1.000 0.000
#> GSM1269664 2 0.0000 0.974 0.000 1.000
#> GSM1269672 1 0.0000 0.971 1.000 0.000
#> GSM1269680 2 0.0000 0.974 0.000 1.000
#> GSM1269686 1 0.0000 0.971 1.000 0.000
#> GSM1269694 1 0.0000 0.971 1.000 0.000
#> GSM1269702 1 0.0000 0.971 1.000 0.000
#> GSM1269710 2 0.0000 0.974 0.000 1.000
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM1269647 2 0.3038 0.8362 0.104 0.896 0.000
#> GSM1269655 1 0.1753 0.8101 0.952 0.000 0.048
#> GSM1269663 2 0.0424 0.9195 0.000 0.992 0.008
#> GSM1269671 2 0.0237 0.9196 0.004 0.996 0.000
#> GSM1269679 3 0.1411 0.8723 0.000 0.036 0.964
#> GSM1269693 3 0.0424 0.8884 0.008 0.000 0.992
#> GSM1269701 3 0.1529 0.8745 0.040 0.000 0.960
#> GSM1269709 3 0.1529 0.8744 0.040 0.000 0.960
#> GSM1269715 3 0.0237 0.8852 0.000 0.004 0.996
#> GSM1269717 3 0.1170 0.8854 0.008 0.016 0.976
#> GSM1269721 3 0.0000 0.8867 0.000 0.000 1.000
#> GSM1269723 2 0.4750 0.7159 0.000 0.784 0.216
#> GSM1269645 2 0.0424 0.9195 0.000 0.992 0.008
#> GSM1269653 3 0.0424 0.8884 0.008 0.000 0.992
#> GSM1269661 2 0.0237 0.9196 0.004 0.996 0.000
#> GSM1269669 3 0.5363 0.6078 0.276 0.000 0.724
#> GSM1269677 2 0.0424 0.9195 0.000 0.992 0.008
#> GSM1269685 3 0.3816 0.7857 0.148 0.000 0.852
#> GSM1269691 1 0.6168 0.2791 0.588 0.000 0.412
#> GSM1269699 1 0.1860 0.8072 0.948 0.000 0.052
#> GSM1269707 3 0.0424 0.8884 0.008 0.000 0.992
#> GSM1269651 2 0.0424 0.9186 0.008 0.992 0.000
#> GSM1269659 3 0.0424 0.8884 0.008 0.000 0.992
#> GSM1269667 2 0.0747 0.9150 0.016 0.984 0.000
#> GSM1269675 3 0.0424 0.8843 0.000 0.008 0.992
#> GSM1269683 3 0.4702 0.6784 0.000 0.212 0.788
#> GSM1269689 3 0.0000 0.8867 0.000 0.000 1.000
#> GSM1269697 1 0.1289 0.8177 0.968 0.000 0.032
#> GSM1269705 1 0.0237 0.8248 0.996 0.000 0.004
#> GSM1269713 3 0.0424 0.8884 0.008 0.000 0.992
#> GSM1269719 2 0.6260 0.1217 0.448 0.552 0.000
#> GSM1269725 1 0.4887 0.6411 0.772 0.000 0.228
#> GSM1269727 3 0.1163 0.8750 0.000 0.028 0.972
#> GSM1269649 2 0.0747 0.9150 0.016 0.984 0.000
#> GSM1269657 1 0.6126 0.3142 0.600 0.000 0.400
#> GSM1269665 2 0.0424 0.9195 0.000 0.992 0.008
#> GSM1269673 1 0.6617 0.2146 0.556 0.008 0.436
#> GSM1269681 2 0.0424 0.9195 0.000 0.992 0.008
#> GSM1269687 1 0.0237 0.8248 0.996 0.000 0.004
#> GSM1269695 1 0.6309 -0.0205 0.500 0.000 0.500
#> GSM1269703 3 0.0424 0.8884 0.008 0.000 0.992
#> GSM1269711 3 0.0424 0.8884 0.008 0.000 0.992
#> GSM1269646 2 0.0747 0.9144 0.016 0.984 0.000
#> GSM1269654 1 0.0237 0.8248 0.996 0.000 0.004
#> GSM1269662 2 0.0424 0.9195 0.000 0.992 0.008
#> GSM1269670 2 0.0237 0.9196 0.004 0.996 0.000
#> GSM1269678 1 0.0237 0.8229 0.996 0.004 0.000
#> GSM1269692 3 0.6045 0.3795 0.380 0.000 0.620
#> GSM1269700 3 0.5363 0.6102 0.276 0.000 0.724
#> GSM1269708 1 0.6062 0.3596 0.616 0.000 0.384
#> GSM1269714 3 0.3551 0.8015 0.132 0.000 0.868
#> GSM1269716 3 0.6008 0.4058 0.372 0.000 0.628
#> GSM1269720 3 0.2878 0.8226 0.000 0.096 0.904
#> GSM1269722 3 0.5327 0.5845 0.000 0.272 0.728
#> GSM1269644 1 0.6204 0.2231 0.576 0.424 0.000
#> GSM1269652 1 0.0424 0.8242 0.992 0.000 0.008
#> GSM1269660 2 0.0237 0.9196 0.004 0.996 0.000
#> GSM1269668 1 0.0237 0.8248 0.996 0.000 0.004
#> GSM1269676 1 0.3619 0.7406 0.864 0.136 0.000
#> GSM1269684 3 0.5363 0.6138 0.276 0.000 0.724
#> GSM1269690 1 0.0747 0.8221 0.984 0.000 0.016
#> GSM1269698 1 0.0237 0.8248 0.996 0.000 0.004
#> GSM1269706 3 0.0424 0.8884 0.008 0.000 0.992
#> GSM1269650 2 0.2959 0.8405 0.100 0.900 0.000
#> GSM1269658 3 0.0747 0.8812 0.000 0.016 0.984
#> GSM1269666 1 0.5363 0.5548 0.724 0.276 0.000
#> GSM1269674 1 0.6307 0.0280 0.512 0.000 0.488
#> GSM1269682 2 0.0592 0.9178 0.000 0.988 0.012
#> GSM1269688 3 0.0592 0.8872 0.012 0.000 0.988
#> GSM1269696 1 0.1950 0.8141 0.952 0.008 0.040
#> GSM1269704 1 0.0237 0.8229 0.996 0.004 0.000
#> GSM1269712 2 0.0424 0.9193 0.000 0.992 0.008
#> GSM1269718 1 0.5810 0.4701 0.664 0.336 0.000
#> GSM1269724 1 0.6341 0.5031 0.672 0.312 0.016
#> GSM1269726 2 0.6192 0.2957 0.000 0.580 0.420
#> GSM1269648 1 0.3192 0.7565 0.888 0.112 0.000
#> GSM1269656 1 0.0237 0.8248 0.996 0.000 0.004
#> GSM1269664 2 0.0237 0.9196 0.004 0.996 0.000
#> GSM1269672 1 0.0424 0.8214 0.992 0.008 0.000
#> GSM1269680 1 0.2796 0.7723 0.908 0.092 0.000
#> GSM1269686 1 0.0237 0.8248 0.996 0.000 0.004
#> GSM1269694 1 0.1411 0.8157 0.964 0.000 0.036
#> GSM1269702 1 0.0237 0.8248 0.996 0.000 0.004
#> GSM1269710 2 0.4178 0.7746 0.000 0.828 0.172
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM1269647 2 0.3378 0.75596 0.060 0.884 0.012 0.044
#> GSM1269655 1 0.5168 -0.01957 0.500 0.000 0.004 0.496
#> GSM1269663 2 0.6295 0.61317 0.000 0.616 0.296 0.088
#> GSM1269671 2 0.4185 0.74454 0.044 0.852 0.060 0.044
#> GSM1269679 3 0.6623 0.58096 0.264 0.048 0.644 0.044
#> GSM1269693 3 0.6065 0.68183 0.176 0.000 0.684 0.140
#> GSM1269701 1 0.5530 0.07774 0.632 0.000 0.336 0.032
#> GSM1269709 1 0.5768 -0.27454 0.516 0.000 0.456 0.028
#> GSM1269715 3 0.4720 0.67270 0.264 0.000 0.720 0.016
#> GSM1269717 3 0.7675 0.51989 0.344 0.072 0.524 0.060
#> GSM1269721 3 0.4300 0.70927 0.088 0.000 0.820 0.092
#> GSM1269723 3 0.4410 0.56209 0.012 0.144 0.812 0.032
#> GSM1269645 2 0.4509 0.66636 0.000 0.708 0.288 0.004
#> GSM1269653 1 0.5407 -0.33882 0.504 0.000 0.484 0.012
#> GSM1269661 2 0.1411 0.77986 0.000 0.960 0.020 0.020
#> GSM1269669 4 0.7822 0.00489 0.364 0.000 0.256 0.380
#> GSM1269677 2 0.7431 0.44493 0.004 0.532 0.268 0.196
#> GSM1269685 4 0.5184 0.49231 0.056 0.000 0.212 0.732
#> GSM1269691 4 0.5478 0.45175 0.248 0.000 0.056 0.696
#> GSM1269699 1 0.1302 0.56596 0.956 0.000 0.000 0.044
#> GSM1269707 3 0.4983 0.66774 0.272 0.000 0.704 0.024
#> GSM1269651 2 0.4758 0.73459 0.000 0.780 0.156 0.064
#> GSM1269659 3 0.4595 0.64600 0.040 0.000 0.776 0.184
#> GSM1269667 2 0.3255 0.76364 0.048 0.892 0.016 0.044
#> GSM1269675 3 0.1182 0.69795 0.016 0.000 0.968 0.016
#> GSM1269683 3 0.2124 0.66146 0.000 0.068 0.924 0.008
#> GSM1269689 3 0.4647 0.65387 0.288 0.000 0.704 0.008
#> GSM1269697 1 0.2644 0.57170 0.916 0.044 0.032 0.008
#> GSM1269705 1 0.5137 0.39509 0.716 0.040 0.000 0.244
#> GSM1269713 1 0.5358 0.37234 0.736 0.020 0.212 0.032
#> GSM1269719 2 0.5619 0.51952 0.268 0.676 0.000 0.056
#> GSM1269725 1 0.4456 0.53805 0.804 0.148 0.044 0.004
#> GSM1269727 3 0.1940 0.71326 0.076 0.000 0.924 0.000
#> GSM1269649 2 0.5085 0.75435 0.036 0.796 0.116 0.052
#> GSM1269657 4 0.3616 0.55303 0.036 0.000 0.112 0.852
#> GSM1269665 2 0.2773 0.76741 0.000 0.880 0.116 0.004
#> GSM1269673 4 0.6505 0.28884 0.064 0.008 0.356 0.572
#> GSM1269681 2 0.3581 0.75939 0.000 0.852 0.116 0.032
#> GSM1269687 1 0.6586 0.02459 0.500 0.080 0.000 0.420
#> GSM1269695 1 0.3009 0.55126 0.892 0.000 0.056 0.052
#> GSM1269703 1 0.4682 0.41361 0.764 0.008 0.208 0.020
#> GSM1269711 3 0.4993 0.66710 0.244 0.008 0.728 0.020
#> GSM1269646 2 0.1624 0.77261 0.028 0.952 0.000 0.020
#> GSM1269654 4 0.5987 0.03812 0.440 0.040 0.000 0.520
#> GSM1269662 2 0.4262 0.70393 0.000 0.756 0.236 0.008
#> GSM1269670 2 0.2368 0.77197 0.032 0.928 0.008 0.032
#> GSM1269678 1 0.6240 0.39393 0.664 0.200 0.000 0.136
#> GSM1269692 4 0.7081 0.19825 0.352 0.000 0.136 0.512
#> GSM1269700 1 0.3198 0.54793 0.884 0.004 0.080 0.032
#> GSM1269708 1 0.2413 0.56053 0.916 0.000 0.020 0.064
#> GSM1269714 3 0.7488 0.29461 0.180 0.000 0.436 0.384
#> GSM1269716 1 0.5961 0.29578 0.656 0.008 0.052 0.284
#> GSM1269720 3 0.3529 0.65499 0.012 0.068 0.876 0.044
#> GSM1269722 1 0.7189 0.26403 0.604 0.136 0.240 0.020
#> GSM1269644 4 0.5649 0.41107 0.044 0.280 0.004 0.672
#> GSM1269652 4 0.4730 0.29473 0.364 0.000 0.000 0.636
#> GSM1269660 2 0.0376 0.77782 0.000 0.992 0.004 0.004
#> GSM1269668 4 0.4741 0.36791 0.328 0.004 0.000 0.668
#> GSM1269676 4 0.7538 0.23800 0.112 0.352 0.024 0.512
#> GSM1269684 4 0.4524 0.50027 0.028 0.000 0.204 0.768
#> GSM1269690 4 0.2973 0.52780 0.144 0.000 0.000 0.856
#> GSM1269698 1 0.3320 0.55034 0.876 0.068 0.000 0.056
#> GSM1269706 3 0.6773 0.63124 0.284 0.000 0.584 0.132
#> GSM1269650 2 0.3113 0.74279 0.004 0.876 0.012 0.108
#> GSM1269658 3 0.5827 0.40341 0.000 0.052 0.632 0.316
#> GSM1269666 2 0.5006 0.65654 0.104 0.772 0.000 0.124
#> GSM1269674 4 0.4542 0.49454 0.020 0.004 0.208 0.768
#> GSM1269682 2 0.4238 0.73711 0.000 0.796 0.176 0.028
#> GSM1269688 3 0.5731 0.43538 0.428 0.000 0.544 0.028
#> GSM1269696 1 0.4579 0.41818 0.720 0.272 0.004 0.004
#> GSM1269704 1 0.5966 0.32363 0.648 0.072 0.000 0.280
#> GSM1269712 2 0.5199 0.67501 0.152 0.776 0.044 0.028
#> GSM1269718 2 0.6340 0.17439 0.408 0.528 0.000 0.064
#> GSM1269724 1 0.5460 0.38605 0.660 0.312 0.016 0.012
#> GSM1269726 3 0.5125 0.66952 0.108 0.076 0.792 0.024
#> GSM1269648 4 0.7158 0.25747 0.148 0.340 0.000 0.512
#> GSM1269656 4 0.2281 0.54162 0.096 0.000 0.000 0.904
#> GSM1269664 2 0.1284 0.77449 0.000 0.964 0.012 0.024
#> GSM1269672 4 0.5277 0.38675 0.304 0.028 0.000 0.668
#> GSM1269680 4 0.6558 -0.02946 0.076 0.452 0.000 0.472
#> GSM1269686 1 0.5668 0.31743 0.652 0.048 0.000 0.300
#> GSM1269694 1 0.3494 0.50336 0.824 0.000 0.004 0.172
#> GSM1269702 1 0.4994 0.00984 0.520 0.000 0.000 0.480
#> GSM1269710 2 0.7699 0.40418 0.144 0.568 0.252 0.036
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM1269647 2 0.4904 0.3778 0.368 0.604 0.008 0.020 0.000
#> GSM1269655 5 0.4567 0.3213 0.012 0.000 0.356 0.004 0.628
#> GSM1269663 2 0.6975 0.2838 0.256 0.440 0.000 0.292 0.012
#> GSM1269671 1 0.6843 -0.0138 0.464 0.392 0.080 0.064 0.000
#> GSM1269679 1 0.5755 0.1491 0.588 0.032 0.044 0.336 0.000
#> GSM1269693 4 0.4673 0.4764 0.004 0.000 0.052 0.716 0.228
#> GSM1269701 1 0.6595 0.2032 0.500 0.000 0.304 0.188 0.008
#> GSM1269709 4 0.7029 0.0662 0.348 0.000 0.284 0.360 0.008
#> GSM1269715 4 0.4449 0.5728 0.004 0.000 0.288 0.688 0.020
#> GSM1269717 3 0.7593 -0.0325 0.004 0.052 0.424 0.328 0.192
#> GSM1269721 4 0.4149 0.5272 0.004 0.000 0.040 0.768 0.188
#> GSM1269723 4 0.4664 0.5598 0.100 0.064 0.052 0.784 0.000
#> GSM1269645 2 0.6162 0.4190 0.160 0.532 0.000 0.308 0.000
#> GSM1269653 3 0.5812 -0.3211 0.092 0.000 0.476 0.432 0.000
#> GSM1269661 2 0.3355 0.6507 0.132 0.832 0.000 0.036 0.000
#> GSM1269669 1 0.2734 0.5214 0.888 0.000 0.028 0.076 0.008
#> GSM1269677 5 0.6626 0.3048 0.004 0.276 0.000 0.232 0.488
#> GSM1269685 5 0.5305 0.5266 0.160 0.000 0.016 0.116 0.708
#> GSM1269691 1 0.5804 0.0186 0.476 0.000 0.060 0.012 0.452
#> GSM1269699 3 0.4296 0.5425 0.204 0.000 0.756 0.016 0.024
#> GSM1269707 4 0.4842 0.5722 0.004 0.000 0.264 0.684 0.048
#> GSM1269651 2 0.5516 0.6229 0.048 0.712 0.000 0.088 0.152
#> GSM1269659 4 0.4491 0.2463 0.004 0.000 0.008 0.624 0.364
#> GSM1269667 1 0.5628 -0.0840 0.508 0.424 0.004 0.064 0.000
#> GSM1269675 4 0.3706 0.4719 0.236 0.004 0.000 0.756 0.004
#> GSM1269683 4 0.1913 0.5992 0.024 0.020 0.000 0.936 0.020
#> GSM1269689 4 0.5692 0.5426 0.168 0.000 0.204 0.628 0.000
#> GSM1269697 3 0.3730 0.5583 0.168 0.028 0.800 0.000 0.004
#> GSM1269705 1 0.5293 -0.1544 0.492 0.000 0.460 0.000 0.048
#> GSM1269713 3 0.3463 0.4752 0.008 0.016 0.820 0.156 0.000
#> GSM1269719 2 0.4885 0.3011 0.000 0.668 0.276 0.000 0.056
#> GSM1269725 3 0.2664 0.6082 0.020 0.092 0.884 0.004 0.000
#> GSM1269727 4 0.2766 0.5931 0.084 0.008 0.024 0.884 0.000
#> GSM1269649 1 0.5637 0.1877 0.604 0.284 0.000 0.112 0.000
#> GSM1269657 5 0.2758 0.6201 0.024 0.000 0.012 0.076 0.888
#> GSM1269665 2 0.4269 0.6401 0.108 0.776 0.000 0.116 0.000
#> GSM1269673 1 0.2935 0.4930 0.860 0.004 0.000 0.120 0.016
#> GSM1269681 2 0.2820 0.6791 0.004 0.884 0.000 0.056 0.056
#> GSM1269687 1 0.3963 0.4590 0.800 0.004 0.152 0.004 0.040
#> GSM1269695 3 0.5522 0.3810 0.312 0.000 0.620 0.028 0.040
#> GSM1269703 3 0.4453 0.4003 0.060 0.004 0.752 0.184 0.000
#> GSM1269711 4 0.5360 0.2785 0.384 0.000 0.060 0.556 0.000
#> GSM1269646 2 0.1179 0.6856 0.016 0.964 0.016 0.000 0.004
#> GSM1269654 5 0.5476 0.3608 0.004 0.048 0.316 0.012 0.620
#> GSM1269662 2 0.4832 0.6087 0.064 0.720 0.000 0.208 0.008
#> GSM1269670 2 0.1386 0.6821 0.016 0.952 0.032 0.000 0.000
#> GSM1269678 3 0.6371 0.5261 0.120 0.212 0.620 0.000 0.048
#> GSM1269692 5 0.4393 0.5614 0.004 0.000 0.052 0.192 0.752
#> GSM1269700 3 0.4810 0.3827 0.296 0.004 0.668 0.028 0.004
#> GSM1269708 3 0.2529 0.6005 0.036 0.000 0.908 0.024 0.032
#> GSM1269714 5 0.5124 0.4018 0.004 0.000 0.048 0.320 0.628
#> GSM1269716 3 0.6928 0.2988 0.004 0.076 0.584 0.116 0.220
#> GSM1269720 4 0.3742 0.5162 0.004 0.020 0.000 0.788 0.188
#> GSM1269722 3 0.4862 0.5564 0.004 0.180 0.740 0.064 0.012
#> GSM1269644 5 0.6059 0.2287 0.184 0.244 0.000 0.000 0.572
#> GSM1269652 5 0.6407 0.2273 0.244 0.000 0.244 0.000 0.512
#> GSM1269660 2 0.0451 0.6880 0.000 0.988 0.008 0.000 0.004
#> GSM1269668 1 0.4490 0.4385 0.724 0.000 0.052 0.000 0.224
#> GSM1269676 5 0.4595 0.5871 0.004 0.084 0.020 0.108 0.784
#> GSM1269684 5 0.3006 0.5916 0.004 0.000 0.004 0.156 0.836
#> GSM1269690 5 0.3731 0.5192 0.160 0.000 0.040 0.000 0.800
#> GSM1269698 3 0.2777 0.6169 0.036 0.028 0.896 0.000 0.040
#> GSM1269706 4 0.6568 0.3125 0.004 0.000 0.276 0.500 0.220
#> GSM1269650 2 0.3132 0.6340 0.000 0.820 0.000 0.008 0.172
#> GSM1269658 5 0.4759 0.4032 0.004 0.024 0.000 0.336 0.636
#> GSM1269666 2 0.6229 0.1548 0.380 0.512 0.020 0.000 0.088
#> GSM1269674 5 0.4690 0.5247 0.140 0.004 0.000 0.108 0.748
#> GSM1269682 2 0.5251 0.5499 0.004 0.692 0.004 0.208 0.092
#> GSM1269688 4 0.6902 0.2049 0.280 0.000 0.324 0.392 0.004
#> GSM1269696 3 0.3992 0.5449 0.004 0.280 0.712 0.000 0.004
#> GSM1269704 3 0.6313 0.4583 0.240 0.032 0.604 0.000 0.124
#> GSM1269712 2 0.4552 0.1969 0.004 0.632 0.352 0.000 0.012
#> GSM1269718 3 0.5350 0.2233 0.000 0.460 0.488 0.000 0.052
#> GSM1269724 3 0.4504 0.3322 0.000 0.428 0.564 0.000 0.008
#> GSM1269726 4 0.4930 0.6131 0.020 0.068 0.160 0.748 0.004
#> GSM1269648 1 0.5592 0.4549 0.664 0.140 0.008 0.000 0.188
#> GSM1269656 5 0.1281 0.6010 0.032 0.000 0.012 0.000 0.956
#> GSM1269664 2 0.0613 0.6893 0.004 0.984 0.004 0.000 0.008
#> GSM1269672 1 0.5030 0.2817 0.604 0.000 0.044 0.000 0.352
#> GSM1269680 5 0.4299 0.3973 0.004 0.316 0.008 0.000 0.672
#> GSM1269686 3 0.5742 0.5131 0.168 0.016 0.664 0.000 0.152
#> GSM1269694 3 0.4928 0.5587 0.132 0.000 0.740 0.012 0.116
#> GSM1269702 5 0.6587 -0.0931 0.208 0.000 0.388 0.000 0.404
#> GSM1269710 4 0.8154 0.1283 0.128 0.304 0.196 0.372 0.000
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM1269647 2 0.5507 0.3404 0.220 0.632 0.016 0.000 0.124 0.008
#> GSM1269655 6 0.5966 0.3057 0.048 0.004 0.312 0.040 0.024 0.572
#> GSM1269663 5 0.5931 0.2241 0.012 0.336 0.004 0.008 0.528 0.112
#> GSM1269671 2 0.6830 -0.0625 0.076 0.468 0.108 0.016 0.332 0.000
#> GSM1269679 5 0.7066 0.2116 0.164 0.064 0.100 0.108 0.564 0.000
#> GSM1269693 6 0.5542 0.0304 0.000 0.000 0.028 0.448 0.064 0.460
#> GSM1269701 3 0.7059 0.2824 0.080 0.000 0.484 0.144 0.272 0.020
#> GSM1269709 4 0.6766 0.2641 0.344 0.000 0.132 0.456 0.052 0.016
#> GSM1269715 4 0.1700 0.5924 0.000 0.000 0.012 0.936 0.024 0.028
#> GSM1269717 4 0.6260 0.4752 0.016 0.056 0.072 0.672 0.080 0.104
#> GSM1269721 4 0.4836 0.4380 0.000 0.000 0.020 0.688 0.080 0.212
#> GSM1269723 4 0.5553 0.4327 0.004 0.092 0.012 0.632 0.244 0.016
#> GSM1269645 5 0.5535 0.0404 0.000 0.436 0.000 0.012 0.460 0.092
#> GSM1269653 4 0.4085 0.5582 0.004 0.000 0.192 0.748 0.052 0.004
#> GSM1269661 2 0.3983 0.4009 0.000 0.776 0.044 0.000 0.156 0.024
#> GSM1269669 1 0.6064 0.0385 0.424 0.000 0.124 0.028 0.424 0.000
#> GSM1269677 6 0.4808 0.4488 0.000 0.176 0.008 0.048 0.044 0.724
#> GSM1269685 6 0.6173 0.3122 0.368 0.000 0.024 0.080 0.028 0.500
#> GSM1269691 3 0.7518 -0.0797 0.300 0.000 0.308 0.016 0.076 0.300
#> GSM1269699 3 0.2508 0.6493 0.024 0.000 0.904 0.024 0.024 0.024
#> GSM1269707 4 0.1785 0.5845 0.000 0.000 0.008 0.928 0.016 0.048
#> GSM1269651 2 0.5905 0.3650 0.028 0.612 0.016 0.004 0.096 0.244
#> GSM1269659 6 0.5222 0.1222 0.008 0.000 0.012 0.424 0.044 0.512
#> GSM1269667 2 0.6459 -0.0761 0.072 0.468 0.112 0.000 0.348 0.000
#> GSM1269675 5 0.6249 0.1563 0.004 0.008 0.092 0.224 0.596 0.076
#> GSM1269683 4 0.5799 0.1960 0.000 0.016 0.000 0.460 0.408 0.116
#> GSM1269689 4 0.6096 0.3123 0.008 0.000 0.176 0.500 0.308 0.008
#> GSM1269697 3 0.2862 0.6373 0.020 0.012 0.880 0.060 0.028 0.000
#> GSM1269705 3 0.3400 0.6170 0.108 0.008 0.836 0.004 0.036 0.008
#> GSM1269713 4 0.4839 0.5357 0.008 0.012 0.196 0.708 0.072 0.004
#> GSM1269719 3 0.6149 0.3814 0.000 0.228 0.544 0.004 0.024 0.200
#> GSM1269725 3 0.7367 0.3518 0.068 0.108 0.532 0.208 0.076 0.008
#> GSM1269727 4 0.4821 0.4614 0.004 0.000 0.020 0.644 0.296 0.036
#> GSM1269649 5 0.6656 0.1212 0.256 0.332 0.016 0.004 0.388 0.004
#> GSM1269657 6 0.6001 0.4557 0.236 0.000 0.004 0.208 0.012 0.540
#> GSM1269665 2 0.4906 0.2523 0.000 0.648 0.004 0.004 0.264 0.080
#> GSM1269673 1 0.5925 0.1136 0.492 0.016 0.056 0.016 0.408 0.012
#> GSM1269681 2 0.4492 0.4148 0.000 0.700 0.000 0.004 0.080 0.216
#> GSM1269687 3 0.6316 0.1531 0.256 0.008 0.468 0.000 0.260 0.008
#> GSM1269695 3 0.4484 0.6166 0.016 0.000 0.780 0.052 0.064 0.088
#> GSM1269703 3 0.4097 0.6212 0.000 0.012 0.800 0.068 0.092 0.028
#> GSM1269711 4 0.6424 0.2975 0.060 0.004 0.096 0.500 0.336 0.004
#> GSM1269646 2 0.3111 0.4962 0.060 0.868 0.012 0.004 0.044 0.012
#> GSM1269654 6 0.6632 0.2559 0.036 0.028 0.324 0.044 0.040 0.528
#> GSM1269662 2 0.5050 0.2388 0.000 0.628 0.000 0.004 0.260 0.108
#> GSM1269670 2 0.2770 0.5022 0.024 0.888 0.036 0.004 0.044 0.004
#> GSM1269678 1 0.8522 0.1377 0.360 0.264 0.164 0.084 0.116 0.012
#> GSM1269692 6 0.3725 0.5661 0.008 0.000 0.140 0.060 0.000 0.792
#> GSM1269700 3 0.3742 0.6110 0.008 0.000 0.796 0.076 0.120 0.000
#> GSM1269708 4 0.7198 0.1449 0.352 0.008 0.088 0.444 0.072 0.036
#> GSM1269714 4 0.5298 0.4303 0.068 0.000 0.016 0.688 0.040 0.188
#> GSM1269716 4 0.8514 0.1950 0.040 0.068 0.236 0.416 0.128 0.112
#> GSM1269720 4 0.4871 0.4457 0.000 0.008 0.000 0.676 0.112 0.204
#> GSM1269722 4 0.6818 0.3874 0.024 0.096 0.208 0.564 0.104 0.004
#> GSM1269644 1 0.5881 0.3469 0.616 0.148 0.012 0.000 0.028 0.196
#> GSM1269652 1 0.3637 0.4738 0.816 0.000 0.020 0.020 0.016 0.128
#> GSM1269660 2 0.0837 0.5048 0.000 0.972 0.004 0.000 0.020 0.004
#> GSM1269668 1 0.2224 0.5423 0.904 0.000 0.020 0.000 0.012 0.064
#> GSM1269676 6 0.3999 0.5562 0.016 0.076 0.040 0.028 0.016 0.824
#> GSM1269684 6 0.3466 0.5642 0.100 0.000 0.020 0.044 0.004 0.832
#> GSM1269690 6 0.5619 0.3925 0.248 0.000 0.188 0.000 0.004 0.560
#> GSM1269698 3 0.3787 0.6244 0.020 0.016 0.836 0.040 0.016 0.072
#> GSM1269706 4 0.4365 0.5358 0.012 0.000 0.044 0.784 0.064 0.096
#> GSM1269650 2 0.5279 0.4107 0.028 0.660 0.016 0.004 0.044 0.248
#> GSM1269658 6 0.3892 0.5317 0.000 0.012 0.000 0.120 0.080 0.788
#> GSM1269666 2 0.5936 0.0519 0.420 0.460 0.088 0.000 0.024 0.008
#> GSM1269674 6 0.7074 0.3942 0.176 0.000 0.056 0.064 0.160 0.544
#> GSM1269682 2 0.5753 0.3313 0.000 0.640 0.000 0.108 0.080 0.172
#> GSM1269688 4 0.7153 0.3694 0.144 0.000 0.216 0.500 0.124 0.016
#> GSM1269696 3 0.4799 0.5782 0.012 0.120 0.764 0.044 0.036 0.024
#> GSM1269704 3 0.3952 0.5994 0.092 0.016 0.808 0.008 0.004 0.072
#> GSM1269712 2 0.6545 0.3461 0.036 0.624 0.144 0.076 0.112 0.008
#> GSM1269718 3 0.7031 0.0887 0.000 0.372 0.412 0.020 0.072 0.124
#> GSM1269724 2 0.7957 -0.1098 0.052 0.368 0.356 0.096 0.112 0.016
#> GSM1269726 4 0.3092 0.5939 0.000 0.036 0.024 0.868 0.060 0.012
#> GSM1269648 1 0.3721 0.4938 0.816 0.116 0.020 0.000 0.036 0.012
#> GSM1269656 6 0.5918 0.2483 0.416 0.000 0.016 0.064 0.028 0.476
#> GSM1269664 2 0.2182 0.4982 0.000 0.900 0.000 0.004 0.020 0.076
#> GSM1269672 1 0.2903 0.5415 0.864 0.004 0.052 0.000 0.004 0.076
#> GSM1269680 6 0.5925 0.3378 0.084 0.244 0.012 0.000 0.052 0.608
#> GSM1269686 1 0.7316 0.2059 0.492 0.020 0.292 0.064 0.092 0.040
#> GSM1269694 3 0.3358 0.6244 0.016 0.000 0.832 0.024 0.008 0.120
#> GSM1269702 3 0.5246 0.4255 0.164 0.000 0.604 0.000 0.000 0.232
#> GSM1269710 4 0.6957 0.2792 0.036 0.268 0.040 0.508 0.144 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 agent(p) disease.state(p) gender(p) individual(p) k
#> ATC:NMF 81 0.47143 0.408 0.8983 0.00173 2
#> ATC:NMF 72 0.00438 0.191 0.1182 0.02146 3
#> ATC:NMF 48 0.13220 0.930 0.0218 0.00340 4
#> ATC:NMF 39 0.10398 0.830 0.0769 0.00648 5
#> ATC:NMF 24 0.10551 0.274 0.7640 0.03094 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