Date: 2019-12-25 21:09:11 CET, cola version: 1.3.2
Document is loading...
All available functions which can be applied to this res_list object:
res_list
#> A 'ConsensusPartitionList' object with 24 methods.
#> On a matrix with 21168 rows and 86 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] 21168 86
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:kmeans | 2 | 1.000 | 0.993 | 0.993 | ** | |
| ATC:NMF | 2 | 0.999 | 0.962 | 0.984 | ** | |
| ATC:skmeans | 3 | 0.963 | 0.957 | 0.979 | ** | 2 |
| MAD:mclust | 4 | 0.945 | 0.927 | 0.953 | * | |
| ATC:pam | 4 | 0.916 | 0.871 | 0.952 | * | 2 |
| MAD:pam | 2 | 0.904 | 0.946 | 0.975 | * | |
| MAD:NMF | 2 | 0.900 | 0.917 | 0.962 | ||
| SD:mclust | 5 | 0.881 | 0.875 | 0.928 | ||
| CV:pam | 2 | 0.832 | 0.906 | 0.958 | ||
| SD:pam | 2 | 0.812 | 0.926 | 0.965 | ||
| CV:NMF | 3 | 0.801 | 0.862 | 0.939 | ||
| CV:mclust | 5 | 0.769 | 0.782 | 0.879 | ||
| SD:NMF | 3 | 0.765 | 0.860 | 0.934 | ||
| ATC:hclust | 5 | 0.762 | 0.775 | 0.878 | ||
| ATC:mclust | 5 | 0.750 | 0.769 | 0.872 | ||
| CV:skmeans | 3 | 0.712 | 0.833 | 0.920 | ||
| MAD:skmeans | 2 | 0.706 | 0.885 | 0.940 | ||
| SD:skmeans | 3 | 0.687 | 0.854 | 0.922 | ||
| MAD:hclust | 3 | 0.685 | 0.789 | 0.898 | ||
| CV:hclust | 3 | 0.663 | 0.847 | 0.928 | ||
| SD:hclust | 3 | 0.441 | 0.783 | 0.889 | ||
| SD:kmeans | 3 | 0.285 | 0.685 | 0.798 | ||
| CV:kmeans | 3 | 0.277 | 0.656 | 0.785 | ||
| MAD:kmeans | 2 | 0.188 | 0.617 | 0.756 |
**: 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 0.467 0.781 0.888 0.473 0.508 0.508
#> CV:NMF 2 0.345 0.554 0.813 0.468 0.540 0.540
#> MAD:NMF 2 0.900 0.917 0.962 0.496 0.501 0.501
#> ATC:NMF 2 0.999 0.962 0.984 0.468 0.534 0.534
#> SD:skmeans 2 0.402 0.495 0.816 0.502 0.498 0.498
#> CV:skmeans 2 0.325 0.490 0.780 0.499 0.495 0.495
#> MAD:skmeans 2 0.706 0.885 0.940 0.500 0.501 0.501
#> ATC:skmeans 2 1.000 0.978 0.991 0.496 0.504 0.504
#> SD:mclust 2 0.474 0.814 0.906 0.494 0.497 0.497
#> CV:mclust 2 0.331 0.563 0.759 0.472 0.495 0.495
#> MAD:mclust 2 0.698 0.805 0.902 0.488 0.498 0.498
#> ATC:mclust 2 0.276 0.752 0.812 0.446 0.497 0.497
#> SD:kmeans 2 0.163 0.437 0.683 0.383 0.665 0.665
#> CV:kmeans 2 0.154 0.421 0.736 0.382 0.615 0.615
#> MAD:kmeans 2 0.188 0.617 0.756 0.423 0.615 0.615
#> ATC:kmeans 2 1.000 0.993 0.993 0.460 0.540 0.540
#> SD:pam 2 0.812 0.926 0.965 0.489 0.512 0.512
#> CV:pam 2 0.832 0.906 0.958 0.483 0.512 0.512
#> MAD:pam 2 0.904 0.946 0.975 0.488 0.512 0.512
#> ATC:pam 2 0.927 0.936 0.973 0.475 0.521 0.521
#> SD:hclust 2 0.374 0.856 0.881 0.272 0.774 0.774
#> CV:hclust 2 0.535 0.908 0.931 0.252 0.774 0.774
#> MAD:hclust 2 0.426 0.746 0.820 0.318 0.774 0.774
#> ATC:hclust 2 0.667 0.775 0.918 0.336 0.665 0.665
get_stats(res_list, k = 3)
#> k 1-PAC mean_silhouette concordance area_increased Rand Jaccard
#> SD:NMF 3 0.765 0.860 0.934 0.338 0.647 0.423
#> CV:NMF 3 0.801 0.862 0.939 0.401 0.660 0.447
#> MAD:NMF 3 0.786 0.853 0.937 0.332 0.722 0.502
#> ATC:NMF 3 0.723 0.814 0.915 0.362 0.735 0.541
#> SD:skmeans 3 0.687 0.854 0.922 0.330 0.748 0.533
#> CV:skmeans 3 0.712 0.833 0.920 0.343 0.726 0.501
#> MAD:skmeans 3 0.744 0.788 0.905 0.340 0.694 0.460
#> ATC:skmeans 3 0.963 0.957 0.979 0.298 0.808 0.635
#> SD:mclust 3 0.415 0.506 0.690 0.271 0.782 0.587
#> CV:mclust 3 0.324 0.567 0.731 0.330 0.763 0.578
#> MAD:mclust 3 0.636 0.830 0.871 0.341 0.758 0.548
#> ATC:mclust 3 0.421 0.697 0.824 0.282 0.854 0.726
#> SD:kmeans 3 0.285 0.685 0.798 0.436 0.741 0.627
#> CV:kmeans 3 0.277 0.656 0.785 0.440 0.787 0.670
#> MAD:kmeans 3 0.373 0.537 0.738 0.395 0.648 0.469
#> ATC:kmeans 3 0.595 0.628 0.829 0.308 0.810 0.663
#> SD:pam 3 0.894 0.914 0.963 0.238 0.880 0.765
#> CV:pam 3 0.786 0.877 0.944 0.253 0.880 0.765
#> MAD:pam 3 0.780 0.890 0.937 0.267 0.880 0.765
#> ATC:pam 3 0.702 0.790 0.886 0.195 0.904 0.821
#> SD:hclust 3 0.441 0.783 0.889 0.638 0.773 0.708
#> CV:hclust 3 0.663 0.847 0.928 0.744 0.773 0.708
#> MAD:hclust 3 0.685 0.789 0.898 0.893 0.635 0.529
#> ATC:hclust 3 0.378 0.559 0.769 0.470 0.788 0.697
get_stats(res_list, k = 4)
#> k 1-PAC mean_silhouette concordance area_increased Rand Jaccard
#> SD:NMF 4 0.684 0.717 0.867 0.1630 0.749 0.425
#> CV:NMF 4 0.644 0.702 0.848 0.1306 0.768 0.441
#> MAD:NMF 4 0.745 0.782 0.883 0.1214 0.818 0.532
#> ATC:NMF 4 0.766 0.804 0.912 0.1146 0.873 0.673
#> SD:skmeans 4 0.730 0.785 0.862 0.1215 0.842 0.568
#> CV:skmeans 4 0.698 0.786 0.862 0.1186 0.844 0.571
#> MAD:skmeans 4 0.759 0.790 0.870 0.1155 0.851 0.588
#> ATC:skmeans 4 0.722 0.701 0.847 0.1159 0.909 0.757
#> SD:mclust 4 0.819 0.858 0.917 0.1445 0.833 0.581
#> CV:mclust 4 0.666 0.752 0.826 0.1264 0.876 0.693
#> MAD:mclust 4 0.945 0.927 0.953 0.1047 0.947 0.841
#> ATC:mclust 4 0.516 0.746 0.756 0.2217 0.819 0.595
#> SD:kmeans 4 0.343 0.570 0.705 0.2314 0.807 0.602
#> CV:kmeans 4 0.362 0.503 0.671 0.2087 0.795 0.586
#> MAD:kmeans 4 0.427 0.665 0.746 0.1736 0.770 0.486
#> ATC:kmeans 4 0.567 0.509 0.726 0.1519 0.856 0.666
#> SD:pam 4 0.817 0.837 0.922 0.1203 0.904 0.759
#> CV:pam 4 0.724 0.824 0.910 0.1106 0.914 0.786
#> MAD:pam 4 0.713 0.619 0.798 0.1312 0.878 0.702
#> ATC:pam 4 0.916 0.871 0.952 0.1853 0.830 0.647
#> SD:hclust 4 0.578 0.755 0.864 0.1688 0.996 0.992
#> CV:hclust 4 0.670 0.776 0.886 0.1281 0.996 0.992
#> MAD:hclust 4 0.685 0.753 0.872 0.0419 0.993 0.984
#> ATC:hclust 4 0.631 0.661 0.848 0.3137 0.690 0.483
get_stats(res_list, k = 5)
#> k 1-PAC mean_silhouette concordance area_increased Rand Jaccard
#> SD:NMF 5 0.660 0.646 0.794 0.0668 0.887 0.608
#> CV:NMF 5 0.605 0.551 0.741 0.0651 0.903 0.646
#> MAD:NMF 5 0.656 0.568 0.752 0.0674 0.933 0.749
#> ATC:NMF 5 0.610 0.498 0.713 0.0676 0.880 0.650
#> SD:skmeans 5 0.787 0.841 0.878 0.0614 0.945 0.784
#> CV:skmeans 5 0.725 0.764 0.835 0.0606 0.933 0.744
#> MAD:skmeans 5 0.744 0.829 0.871 0.0610 0.950 0.801
#> ATC:skmeans 5 0.728 0.726 0.810 0.0787 0.889 0.647
#> SD:mclust 5 0.881 0.875 0.928 0.0900 0.903 0.675
#> CV:mclust 5 0.769 0.782 0.879 0.1034 0.912 0.706
#> MAD:mclust 5 0.833 0.858 0.902 0.0760 0.943 0.798
#> ATC:mclust 5 0.750 0.769 0.872 0.1136 0.923 0.735
#> SD:kmeans 5 0.519 0.622 0.723 0.0983 0.844 0.543
#> CV:kmeans 5 0.490 0.580 0.701 0.1141 0.749 0.375
#> MAD:kmeans 5 0.497 0.609 0.694 0.0963 0.804 0.464
#> ATC:kmeans 5 0.589 0.386 0.639 0.0903 0.869 0.655
#> SD:pam 5 0.790 0.802 0.886 0.0738 0.970 0.907
#> CV:pam 5 0.787 0.858 0.910 0.0630 0.962 0.886
#> MAD:pam 5 0.845 0.824 0.920 0.1004 0.862 0.594
#> ATC:pam 5 0.831 0.793 0.906 0.1292 0.841 0.567
#> SD:hclust 5 0.810 0.828 0.904 0.1818 0.882 0.785
#> CV:hclust 5 0.729 0.881 0.926 0.1924 0.882 0.785
#> MAD:hclust 5 0.707 0.728 0.841 0.0602 0.958 0.895
#> ATC:hclust 5 0.762 0.775 0.878 0.0830 0.914 0.768
get_stats(res_list, k = 6)
#> k 1-PAC mean_silhouette concordance area_increased Rand Jaccard
#> SD:NMF 6 0.757 0.678 0.828 0.0391 0.880 0.529
#> CV:NMF 6 0.709 0.604 0.742 0.0401 0.892 0.567
#> MAD:NMF 6 0.700 0.664 0.782 0.0401 0.897 0.590
#> ATC:NMF 6 0.704 0.650 0.771 0.0367 0.907 0.703
#> SD:skmeans 6 0.810 0.826 0.819 0.0371 0.969 0.854
#> CV:skmeans 6 0.750 0.748 0.775 0.0391 0.957 0.799
#> MAD:skmeans 6 0.792 0.790 0.829 0.0364 1.000 1.000
#> ATC:skmeans 6 0.714 0.706 0.807 0.0490 0.964 0.834
#> SD:mclust 6 0.816 0.752 0.826 0.0425 0.989 0.951
#> CV:mclust 6 0.812 0.733 0.836 0.0465 0.948 0.762
#> MAD:mclust 6 0.798 0.575 0.750 0.0453 0.963 0.836
#> ATC:mclust 6 0.752 0.690 0.804 0.0390 0.966 0.853
#> SD:kmeans 6 0.638 0.655 0.711 0.0563 0.958 0.812
#> CV:kmeans 6 0.621 0.687 0.699 0.0673 0.894 0.593
#> MAD:kmeans 6 0.638 0.656 0.710 0.0510 0.967 0.858
#> ATC:kmeans 6 0.631 0.467 0.675 0.0574 0.801 0.436
#> SD:pam 6 0.819 0.719 0.839 0.0586 0.888 0.639
#> CV:pam 6 0.735 0.596 0.785 0.0750 0.919 0.747
#> MAD:pam 6 0.828 0.789 0.870 0.0456 0.962 0.839
#> ATC:pam 6 0.871 0.844 0.934 0.0276 0.960 0.838
#> SD:hclust 6 0.768 0.875 0.909 0.0420 0.987 0.970
#> CV:hclust 6 0.755 0.930 0.949 0.0423 0.987 0.970
#> MAD:hclust 6 0.688 0.695 0.832 0.0843 0.879 0.671
#> ATC:hclust 6 0.829 0.692 0.869 0.0335 0.975 0.916
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 individual(p) protocol(p) time(p) other(p) k
#> SD:NMF 78 2.47e-11 0.26886 0.978 0.84278 2
#> CV:NMF 56 4.75e-06 0.00743 0.707 0.71411 2
#> MAD:NMF 83 3.41e-11 0.35873 0.870 0.84711 2
#> ATC:NMF 85 1.53e-06 0.01748 0.896 0.01213 2
#> SD:skmeans 51 2.62e-08 1.00000 0.995 0.36596 2
#> CV:skmeans 51 9.46e-07 0.29041 0.948 0.04874 2
#> MAD:skmeans 86 8.23e-13 0.49189 0.975 0.75145 2
#> ATC:skmeans 85 1.43e-06 0.01342 0.686 0.00765 2
#> SD:mclust 79 1.18e-11 0.61699 0.983 0.87401 2
#> CV:mclust 69 3.73e-11 0.62828 0.997 0.80215 2
#> MAD:mclust 82 1.00e-12 0.92553 0.999 0.69384 2
#> ATC:mclust 77 3.13e-12 0.53563 0.991 0.46354 2
#> SD:kmeans 53 2.92e-08 1.00000 1.000 0.22182 2
#> CV:kmeans 43 NA NA NA NA 2
#> MAD:kmeans 68 3.75e-11 0.98496 0.999 0.67199 2
#> ATC:kmeans 86 3.56e-06 0.01542 0.845 0.01088 2
#> SD:pam 85 1.06e-09 0.05229 0.898 0.90095 2
#> CV:pam 84 8.03e-09 0.07825 0.801 0.96149 2
#> MAD:pam 86 7.09e-10 0.05863 0.889 0.87869 2
#> ATC:pam 83 4.67e-07 0.08055 0.968 0.00607 2
#> SD:hclust 85 3.47e-11 0.36263 0.992 0.00449 2
#> CV:hclust 85 3.47e-11 0.36263 0.992 0.00449 2
#> MAD:hclust 85 3.47e-11 0.36263 0.992 0.00449 2
#> ATC:hclust 73 1.10e-06 0.02978 0.908 0.00196 2
test_to_known_factors(res_list, k = 3)
#> n individual(p) protocol(p) time(p) other(p) k
#> SD:NMF 82 6.82e-18 0.032129 0.978 3.62e-01 3
#> CV:NMF 80 7.06e-17 0.179505 0.996 3.43e-01 3
#> MAD:NMF 79 4.52e-18 0.145302 0.991 3.62e-01 3
#> ATC:NMF 78 9.42e-19 0.029081 0.999 2.11e-01 3
#> SD:skmeans 82 2.64e-21 0.388592 1.000 2.90e-01 3
#> CV:skmeans 81 7.24e-20 0.205557 1.000 2.88e-01 3
#> MAD:skmeans 74 1.38e-20 0.560569 1.000 2.05e-01 3
#> ATC:skmeans 86 9.93e-14 0.000205 0.945 2.69e-01 3
#> SD:mclust 50 3.46e-08 0.700928 0.990 5.25e-01 3
#> CV:mclust 55 4.49e-16 0.605053 0.999 4.07e-01 3
#> MAD:mclust 84 2.12e-25 0.814462 1.000 1.29e-01 3
#> ATC:mclust 81 3.08e-24 0.773545 1.000 2.34e-03 3
#> SD:kmeans 79 1.03e-20 0.366003 1.000 1.08e-01 3
#> CV:kmeans 75 1.36e-19 0.369033 1.000 5.46e-02 3
#> MAD:kmeans 57 6.39e-16 0.361230 1.000 1.41e-01 3
#> ATC:kmeans 69 2.03e-12 0.025029 0.959 7.77e-02 3
#> SD:pam 85 3.34e-17 0.102635 0.991 1.65e-01 3
#> CV:pam 83 7.10e-16 0.061169 0.990 3.10e-01 3
#> MAD:pam 85 3.34e-17 0.102635 0.991 1.65e-01 3
#> ATC:pam 80 1.51e-18 0.170700 0.979 4.03e-05 3
#> SD:hclust 85 6.33e-23 0.364583 1.000 2.04e-02 3
#> CV:hclust 79 9.73e-22 0.490417 0.999 9.16e-03 3
#> MAD:hclust 85 8.69e-23 0.423246 1.000 3.89e-02 3
#> ATC:hclust 55 6.76e-12 0.002925 0.959 1.53e-03 3
test_to_known_factors(res_list, k = 4)
#> n individual(p) protocol(p) time(p) other(p) k
#> SD:NMF 75 1.77e-27 0.381331 1.000 0.175134 4
#> CV:NMF 74 9.33e-26 0.072373 1.000 0.129691 4
#> MAD:NMF 79 7.27e-28 0.087966 1.000 0.299899 4
#> ATC:NMF 78 1.21e-21 0.002520 0.994 0.179939 4
#> SD:skmeans 80 4.69e-33 0.645081 1.000 0.068894 4
#> CV:skmeans 81 1.21e-33 0.709775 1.000 0.085118 4
#> MAD:skmeans 82 3.15e-34 0.766257 1.000 0.087027 4
#> ATC:skmeans 72 5.45e-22 0.011462 1.000 0.041178 4
#> SD:mclust 82 9.05e-35 0.959172 1.000 0.013517 4
#> CV:mclust 83 2.36e-36 0.962514 1.000 0.049829 4
#> MAD:mclust 85 1.64e-37 0.954336 1.000 0.015599 4
#> ATC:mclust 80 4.85e-33 0.728585 1.000 0.030007 4
#> SD:kmeans 59 2.74e-25 0.882252 1.000 0.029224 4
#> CV:kmeans 46 2.15e-15 0.719952 1.000 0.140466 4
#> MAD:kmeans 69 1.15e-29 0.942238 1.000 0.041522 4
#> ATC:kmeans 56 1.54e-14 0.131264 0.981 0.132765 4
#> SD:pam 81 1.17e-28 0.234511 1.000 0.124482 4
#> CV:pam 82 2.41e-23 0.286467 0.997 0.296668 4
#> MAD:pam 56 2.71e-12 0.173312 0.996 0.392364 4
#> ATC:pam 78 9.14e-25 0.018227 0.999 0.005532 4
#> SD:hclust 77 6.78e-23 0.069667 1.000 0.000256 4
#> CV:hclust 77 6.78e-23 0.069667 1.000 0.000256 4
#> MAD:hclust 77 7.95e-23 0.076635 1.000 0.000202 4
#> ATC:hclust 63 1.49e-20 0.000652 0.967 0.000497 4
test_to_known_factors(res_list, k = 5)
#> n individual(p) protocol(p) time(p) other(p) k
#> SD:NMF 69 7.42e-35 0.264111 1.000 0.304464 5
#> CV:NMF 58 9.52e-28 0.606575 0.999 0.360312 5
#> MAD:NMF 60 3.63e-23 0.137910 1.000 0.213719 5
#> ATC:NMF 52 2.36e-16 0.259093 0.686 0.033399 5
#> SD:skmeans 86 2.10e-47 0.911799 1.000 0.020797 5
#> CV:skmeans 82 1.86e-44 0.864930 1.000 0.012907 5
#> MAD:skmeans 85 1.15e-46 0.903932 1.000 0.020566 5
#> ATC:skmeans 73 1.56e-32 0.003642 1.000 0.017190 5
#> SD:mclust 82 1.38e-45 0.952869 1.000 0.023930 5
#> CV:mclust 78 2.53e-41 0.827142 1.000 0.052475 5
#> MAD:mclust 83 1.14e-47 0.955148 1.000 0.014689 5
#> ATC:mclust 78 1.21e-40 0.902794 1.000 0.048179 5
#> SD:kmeans 63 1.38e-27 0.796626 1.000 0.057793 5
#> CV:kmeans 55 2.37e-24 0.985068 1.000 0.028570 5
#> MAD:kmeans 62 3.07e-35 0.976672 1.000 0.017669 5
#> ATC:kmeans 38 2.38e-11 0.002938 0.971 0.014039 5
#> SD:pam 82 4.06e-41 0.325971 1.000 0.105169 5
#> CV:pam 84 6.13e-40 0.309353 1.000 0.128415 5
#> MAD:pam 78 8.88e-36 0.086101 1.000 0.044420 5
#> ATC:pam 79 4.45e-34 0.296669 1.000 0.006147 5
#> SD:hclust 79 1.04e-34 0.183760 1.000 0.001081 5
#> CV:hclust 85 1.26e-36 0.133280 1.000 0.001249 5
#> MAD:hclust 79 1.22e-34 0.181113 1.000 0.000192 5
#> ATC:hclust 77 1.26e-26 0.000909 0.996 0.000028 5
test_to_known_factors(res_list, k = 6)
#> n individual(p) protocol(p) time(p) other(p) k
#> SD:NMF 69 7.17e-48 0.920609 1.000 6.30e-02 6
#> CV:NMF 61 1.68e-42 0.540386 1.000 4.07e-02 6
#> MAD:NMF 72 2.92e-45 0.317302 1.000 1.07e-01 6
#> ATC:NMF 68 5.46e-44 0.605093 1.000 9.63e-02 6
#> SD:skmeans 86 2.76e-59 0.933464 1.000 9.69e-02 6
#> CV:skmeans 82 1.43e-55 0.906327 1.000 5.80e-02 6
#> MAD:skmeans 85 3.27e-49 0.977079 1.000 2.25e-02 6
#> ATC:skmeans 78 1.35e-36 0.007309 1.000 2.29e-02 6
#> SD:mclust 76 2.60e-42 0.909835 1.000 5.77e-02 6
#> CV:mclust 72 4.28e-49 0.894250 1.000 1.93e-01 6
#> MAD:mclust 64 5.53e-35 0.990143 1.000 2.08e-01 6
#> ATC:mclust 73 7.75e-42 0.860840 1.000 5.14e-02 6
#> SD:kmeans 63 6.86e-38 0.982985 1.000 3.71e-02 6
#> CV:kmeans 79 1.80e-57 0.999434 1.000 6.95e-02 6
#> MAD:kmeans 68 3.27e-49 0.856994 1.000 7.99e-02 6
#> ATC:kmeans 44 6.73e-17 0.344957 0.997 4.47e-02 6
#> SD:pam 76 4.13e-50 0.797359 1.000 4.23e-02 6
#> CV:pam 60 1.12e-23 0.361805 1.000 2.83e-01 6
#> MAD:pam 78 1.05e-48 0.186265 1.000 1.00e-01 6
#> ATC:pam 81 4.36e-46 0.364464 1.000 2.09e-02 6
#> SD:hclust 85 2.70e-48 0.130018 1.000 1.10e-03 6
#> CV:hclust 85 2.70e-48 0.130018 1.000 1.10e-03 6
#> MAD:hclust 77 1.54e-55 0.974827 1.000 3.15e-02 6
#> ATC:hclust 74 1.12e-28 0.000281 0.991 5.64e-05 6
The object with results only for a single top-value method and a single partition method can be extracted as:
res = res_list["SD", "hclust"]
# you can also extract it by
# res = res_list["SD:hclust"]
A summary of res and all the functions that can be applied to it:
res
#> A 'ConsensusPartition' object with k = 2, 3, 4, 5, 6.
#> On a matrix with 21168 rows and 86 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 0.374 0.856 0.881 0.272 0.774 0.774
#> 3 3 0.441 0.783 0.889 0.638 0.773 0.708
#> 4 4 0.578 0.755 0.864 0.169 0.996 0.992
#> 5 5 0.810 0.828 0.904 0.182 0.882 0.785
#> 6 6 0.768 0.875 0.909 0.042 0.987 0.970
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
#> GSM614415 2 0.7219 0.783 0.200 0.800
#> GSM614416 2 0.7219 0.783 0.200 0.800
#> GSM614417 2 0.7219 0.783 0.200 0.800
#> GSM614418 2 0.7219 0.783 0.200 0.800
#> GSM614419 2 0.7219 0.783 0.200 0.800
#> GSM614420 2 0.7219 0.783 0.200 0.800
#> GSM614421 2 0.6048 0.771 0.148 0.852
#> GSM614422 2 0.6048 0.771 0.148 0.852
#> GSM614423 2 0.6048 0.771 0.148 0.852
#> GSM614424 2 0.6048 0.771 0.148 0.852
#> GSM614425 2 0.6048 0.771 0.148 0.852
#> GSM614426 2 0.6048 0.771 0.148 0.852
#> GSM614427 2 0.6048 0.771 0.148 0.852
#> GSM614428 2 0.6048 0.771 0.148 0.852
#> GSM614429 2 0.0376 0.898 0.004 0.996
#> GSM614430 2 0.0376 0.898 0.004 0.996
#> GSM614431 2 0.0376 0.898 0.004 0.996
#> GSM614432 2 0.0376 0.898 0.004 0.996
#> GSM614433 2 0.0376 0.898 0.004 0.996
#> GSM614434 2 0.0376 0.898 0.004 0.996
#> GSM614435 2 0.0376 0.898 0.004 0.996
#> GSM614436 2 0.0376 0.898 0.004 0.996
#> GSM614437 1 0.7219 0.952 0.800 0.200
#> GSM614438 1 0.7219 0.952 0.800 0.200
#> GSM614439 1 0.7219 0.952 0.800 0.200
#> GSM614440 1 0.7219 0.952 0.800 0.200
#> GSM614441 1 0.7219 0.952 0.800 0.200
#> GSM614442 1 0.7219 0.952 0.800 0.200
#> GSM614443 1 0.7219 0.952 0.800 0.200
#> GSM614444 1 0.7219 0.952 0.800 0.200
#> GSM614391 2 0.7219 0.783 0.200 0.800
#> GSM614392 2 0.7219 0.783 0.200 0.800
#> GSM614393 2 0.7219 0.783 0.200 0.800
#> GSM614394 2 0.7219 0.783 0.200 0.800
#> GSM614395 1 0.9552 0.275 0.624 0.376
#> GSM614396 2 0.7219 0.783 0.200 0.800
#> GSM614397 2 0.7219 0.783 0.200 0.800
#> GSM614398 2 0.7219 0.783 0.200 0.800
#> GSM614399 2 0.0000 0.900 0.000 1.000
#> GSM614400 2 0.0000 0.900 0.000 1.000
#> GSM614401 2 0.0000 0.900 0.000 1.000
#> GSM614402 2 0.0000 0.900 0.000 1.000
#> GSM614403 2 0.0000 0.900 0.000 1.000
#> GSM614404 2 0.0000 0.900 0.000 1.000
#> GSM614405 2 0.0000 0.900 0.000 1.000
#> GSM614406 2 0.0000 0.900 0.000 1.000
#> GSM614407 2 0.5946 0.827 0.144 0.856
#> GSM614408 2 0.5946 0.827 0.144 0.856
#> GSM614409 2 0.5946 0.827 0.144 0.856
#> GSM614410 2 0.5946 0.827 0.144 0.856
#> GSM614411 2 0.5946 0.827 0.144 0.856
#> GSM614412 2 0.5946 0.827 0.144 0.856
#> GSM614413 2 0.5946 0.827 0.144 0.856
#> GSM614414 2 0.5946 0.827 0.144 0.856
#> GSM614445 2 0.4815 0.821 0.104 0.896
#> GSM614446 2 0.4815 0.821 0.104 0.896
#> GSM614447 2 0.4815 0.821 0.104 0.896
#> GSM614448 2 0.4815 0.821 0.104 0.896
#> GSM614449 2 0.4815 0.821 0.104 0.896
#> GSM614450 2 0.4815 0.821 0.104 0.896
#> GSM614451 1 0.7219 0.952 0.800 0.200
#> GSM614452 1 0.7219 0.952 0.800 0.200
#> GSM614453 2 0.0000 0.900 0.000 1.000
#> GSM614454 2 0.0000 0.900 0.000 1.000
#> GSM614455 2 0.0000 0.900 0.000 1.000
#> GSM614456 2 0.0000 0.900 0.000 1.000
#> GSM614457 2 0.0000 0.900 0.000 1.000
#> GSM614458 2 0.0000 0.900 0.000 1.000
#> GSM614459 2 0.0000 0.900 0.000 1.000
#> GSM614460 2 0.0000 0.900 0.000 1.000
#> GSM614461 2 0.0000 0.900 0.000 1.000
#> GSM614462 2 0.0000 0.900 0.000 1.000
#> GSM614463 2 0.0000 0.900 0.000 1.000
#> GSM614464 2 0.0000 0.900 0.000 1.000
#> GSM614465 2 0.0000 0.900 0.000 1.000
#> GSM614466 2 0.0000 0.900 0.000 1.000
#> GSM614467 2 0.0000 0.900 0.000 1.000
#> GSM614468 2 0.0000 0.900 0.000 1.000
#> GSM614469 2 0.0672 0.898 0.008 0.992
#> GSM614470 2 0.0672 0.898 0.008 0.992
#> GSM614471 2 0.0672 0.898 0.008 0.992
#> GSM614472 2 0.0672 0.898 0.008 0.992
#> GSM614473 2 0.0672 0.898 0.008 0.992
#> GSM614474 2 0.0672 0.898 0.008 0.992
#> GSM614475 2 0.0672 0.898 0.008 0.992
#> GSM614476 2 0.0672 0.898 0.008 0.992
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM614415 1 0.6215 0.5853 0.572 0.428 0.000
#> GSM614416 1 0.6215 0.5853 0.572 0.428 0.000
#> GSM614417 1 0.6215 0.5853 0.572 0.428 0.000
#> GSM614418 1 0.6215 0.5853 0.572 0.428 0.000
#> GSM614419 1 0.6215 0.5853 0.572 0.428 0.000
#> GSM614420 1 0.6215 0.5853 0.572 0.428 0.000
#> GSM614421 2 0.5291 0.6446 0.000 0.732 0.268
#> GSM614422 2 0.5291 0.6446 0.000 0.732 0.268
#> GSM614423 2 0.5291 0.6446 0.000 0.732 0.268
#> GSM614424 2 0.5291 0.6446 0.000 0.732 0.268
#> GSM614425 2 0.5291 0.6446 0.000 0.732 0.268
#> GSM614426 2 0.5291 0.6446 0.000 0.732 0.268
#> GSM614427 2 0.5291 0.6446 0.000 0.732 0.268
#> GSM614428 2 0.5291 0.6446 0.000 0.732 0.268
#> GSM614429 2 0.0237 0.8731 0.000 0.996 0.004
#> GSM614430 2 0.0237 0.8731 0.000 0.996 0.004
#> GSM614431 2 0.0237 0.8731 0.000 0.996 0.004
#> GSM614432 2 0.0237 0.8731 0.000 0.996 0.004
#> GSM614433 2 0.0237 0.8731 0.000 0.996 0.004
#> GSM614434 2 0.0237 0.8731 0.000 0.996 0.004
#> GSM614435 2 0.0237 0.8731 0.000 0.996 0.004
#> GSM614436 2 0.0237 0.8731 0.000 0.996 0.004
#> GSM614437 3 0.0237 0.9768 0.000 0.004 0.996
#> GSM614438 3 0.0237 0.9768 0.000 0.004 0.996
#> GSM614439 3 0.0237 0.9768 0.000 0.004 0.996
#> GSM614440 3 0.0237 0.9768 0.000 0.004 0.996
#> GSM614441 3 0.0237 0.9768 0.000 0.004 0.996
#> GSM614442 3 0.0237 0.9768 0.000 0.004 0.996
#> GSM614443 3 0.0237 0.9768 0.000 0.004 0.996
#> GSM614444 3 0.0237 0.9768 0.000 0.004 0.996
#> GSM614391 1 0.3551 0.7102 0.868 0.132 0.000
#> GSM614392 1 0.3551 0.7102 0.868 0.132 0.000
#> GSM614393 1 0.3551 0.7102 0.868 0.132 0.000
#> GSM614394 1 0.3551 0.7102 0.868 0.132 0.000
#> GSM614395 1 0.5560 -0.0144 0.700 0.000 0.300
#> GSM614396 1 0.3551 0.7102 0.868 0.132 0.000
#> GSM614397 1 0.3551 0.7102 0.868 0.132 0.000
#> GSM614398 1 0.3551 0.7102 0.868 0.132 0.000
#> GSM614399 2 0.0000 0.8740 0.000 1.000 0.000
#> GSM614400 2 0.0000 0.8740 0.000 1.000 0.000
#> GSM614401 2 0.0000 0.8740 0.000 1.000 0.000
#> GSM614402 2 0.0000 0.8740 0.000 1.000 0.000
#> GSM614403 2 0.0000 0.8740 0.000 1.000 0.000
#> GSM614404 2 0.0000 0.8740 0.000 1.000 0.000
#> GSM614405 2 0.0000 0.8740 0.000 1.000 0.000
#> GSM614406 2 0.0000 0.8740 0.000 1.000 0.000
#> GSM614407 2 0.4796 0.6235 0.220 0.780 0.000
#> GSM614408 2 0.4796 0.6235 0.220 0.780 0.000
#> GSM614409 2 0.4796 0.6235 0.220 0.780 0.000
#> GSM614410 2 0.4796 0.6235 0.220 0.780 0.000
#> GSM614411 2 0.4796 0.6235 0.220 0.780 0.000
#> GSM614412 2 0.4796 0.6235 0.220 0.780 0.000
#> GSM614413 2 0.4796 0.6235 0.220 0.780 0.000
#> GSM614414 2 0.4796 0.6235 0.220 0.780 0.000
#> GSM614445 2 0.4750 0.7077 0.000 0.784 0.216
#> GSM614446 2 0.4750 0.7077 0.000 0.784 0.216
#> GSM614447 2 0.4750 0.7077 0.000 0.784 0.216
#> GSM614448 2 0.4750 0.7077 0.000 0.784 0.216
#> GSM614449 2 0.4750 0.7077 0.000 0.784 0.216
#> GSM614450 2 0.4750 0.7077 0.000 0.784 0.216
#> GSM614451 3 0.4277 0.9032 0.132 0.016 0.852
#> GSM614452 3 0.4277 0.9032 0.132 0.016 0.852
#> GSM614453 2 0.0000 0.8740 0.000 1.000 0.000
#> GSM614454 2 0.0000 0.8740 0.000 1.000 0.000
#> GSM614455 2 0.0000 0.8740 0.000 1.000 0.000
#> GSM614456 2 0.0000 0.8740 0.000 1.000 0.000
#> GSM614457 2 0.0000 0.8740 0.000 1.000 0.000
#> GSM614458 2 0.0000 0.8740 0.000 1.000 0.000
#> GSM614459 2 0.0000 0.8740 0.000 1.000 0.000
#> GSM614460 2 0.0000 0.8740 0.000 1.000 0.000
#> GSM614461 2 0.0000 0.8740 0.000 1.000 0.000
#> GSM614462 2 0.0000 0.8740 0.000 1.000 0.000
#> GSM614463 2 0.0000 0.8740 0.000 1.000 0.000
#> GSM614464 2 0.0000 0.8740 0.000 1.000 0.000
#> GSM614465 2 0.0000 0.8740 0.000 1.000 0.000
#> GSM614466 2 0.0000 0.8740 0.000 1.000 0.000
#> GSM614467 2 0.0000 0.8740 0.000 1.000 0.000
#> GSM614468 2 0.0000 0.8740 0.000 1.000 0.000
#> GSM614469 2 0.1031 0.8615 0.024 0.976 0.000
#> GSM614470 2 0.1031 0.8615 0.024 0.976 0.000
#> GSM614471 2 0.1031 0.8615 0.024 0.976 0.000
#> GSM614472 2 0.1031 0.8615 0.024 0.976 0.000
#> GSM614473 2 0.1031 0.8615 0.024 0.976 0.000
#> GSM614474 2 0.1031 0.8615 0.024 0.976 0.000
#> GSM614475 2 0.1031 0.8615 0.024 0.976 0.000
#> GSM614476 2 0.1031 0.8615 0.024 0.976 0.000
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM614415 1 0.6565 0.6249 0.628 0.224 0.148 0.000
#> GSM614416 1 0.6565 0.6249 0.628 0.224 0.148 0.000
#> GSM614417 1 0.6565 0.6249 0.628 0.224 0.148 0.000
#> GSM614418 1 0.6565 0.6249 0.628 0.224 0.148 0.000
#> GSM614419 1 0.6565 0.6249 0.628 0.224 0.148 0.000
#> GSM614420 1 0.6565 0.6249 0.628 0.224 0.148 0.000
#> GSM614421 2 0.5646 0.6736 0.000 0.708 0.204 0.088
#> GSM614422 2 0.5646 0.6736 0.000 0.708 0.204 0.088
#> GSM614423 2 0.5646 0.6736 0.000 0.708 0.204 0.088
#> GSM614424 2 0.5646 0.6736 0.000 0.708 0.204 0.088
#> GSM614425 2 0.5646 0.6736 0.000 0.708 0.204 0.088
#> GSM614426 2 0.5646 0.6736 0.000 0.708 0.204 0.088
#> GSM614427 2 0.5646 0.6736 0.000 0.708 0.204 0.088
#> GSM614428 2 0.5646 0.6736 0.000 0.708 0.204 0.088
#> GSM614429 2 0.0188 0.8579 0.000 0.996 0.004 0.000
#> GSM614430 2 0.0188 0.8579 0.000 0.996 0.004 0.000
#> GSM614431 2 0.0188 0.8579 0.000 0.996 0.004 0.000
#> GSM614432 2 0.0188 0.8579 0.000 0.996 0.004 0.000
#> GSM614433 2 0.0188 0.8579 0.000 0.996 0.004 0.000
#> GSM614434 2 0.0188 0.8579 0.000 0.996 0.004 0.000
#> GSM614435 2 0.0188 0.8579 0.000 0.996 0.004 0.000
#> GSM614436 2 0.0188 0.8579 0.000 0.996 0.004 0.000
#> GSM614437 4 0.0000 1.0000 0.000 0.000 0.000 1.000
#> GSM614438 4 0.0000 1.0000 0.000 0.000 0.000 1.000
#> GSM614439 4 0.0000 1.0000 0.000 0.000 0.000 1.000
#> GSM614440 4 0.0000 1.0000 0.000 0.000 0.000 1.000
#> GSM614441 4 0.0000 1.0000 0.000 0.000 0.000 1.000
#> GSM614442 4 0.0000 1.0000 0.000 0.000 0.000 1.000
#> GSM614443 4 0.0000 1.0000 0.000 0.000 0.000 1.000
#> GSM614444 4 0.0000 1.0000 0.000 0.000 0.000 1.000
#> GSM614391 1 0.0000 0.6624 1.000 0.000 0.000 0.000
#> GSM614392 1 0.0000 0.6624 1.000 0.000 0.000 0.000
#> GSM614393 1 0.0000 0.6624 1.000 0.000 0.000 0.000
#> GSM614394 1 0.0000 0.6624 1.000 0.000 0.000 0.000
#> GSM614395 1 0.4933 0.0228 0.568 0.000 0.432 0.000
#> GSM614396 1 0.0000 0.6624 1.000 0.000 0.000 0.000
#> GSM614397 1 0.0000 0.6624 1.000 0.000 0.000 0.000
#> GSM614398 1 0.0000 0.6624 1.000 0.000 0.000 0.000
#> GSM614399 2 0.0592 0.8567 0.000 0.984 0.016 0.000
#> GSM614400 2 0.0592 0.8567 0.000 0.984 0.016 0.000
#> GSM614401 2 0.0592 0.8567 0.000 0.984 0.016 0.000
#> GSM614402 2 0.0592 0.8567 0.000 0.984 0.016 0.000
#> GSM614403 2 0.0592 0.8567 0.000 0.984 0.016 0.000
#> GSM614404 2 0.0592 0.8567 0.000 0.984 0.016 0.000
#> GSM614405 2 0.0469 0.8574 0.000 0.988 0.012 0.000
#> GSM614406 2 0.0469 0.8574 0.000 0.988 0.012 0.000
#> GSM614407 2 0.7175 0.3236 0.220 0.556 0.224 0.000
#> GSM614408 2 0.7175 0.3236 0.220 0.556 0.224 0.000
#> GSM614409 2 0.7175 0.3236 0.220 0.556 0.224 0.000
#> GSM614410 2 0.7175 0.3236 0.220 0.556 0.224 0.000
#> GSM614411 2 0.7175 0.3236 0.220 0.556 0.224 0.000
#> GSM614412 2 0.7175 0.3236 0.220 0.556 0.224 0.000
#> GSM614413 2 0.7175 0.3236 0.220 0.556 0.224 0.000
#> GSM614414 2 0.7175 0.3236 0.220 0.556 0.224 0.000
#> GSM614445 2 0.4951 0.7190 0.000 0.744 0.212 0.044
#> GSM614446 2 0.4951 0.7190 0.000 0.744 0.212 0.044
#> GSM614447 2 0.4951 0.7190 0.000 0.744 0.212 0.044
#> GSM614448 2 0.4951 0.7190 0.000 0.744 0.212 0.044
#> GSM614449 2 0.4951 0.7190 0.000 0.744 0.212 0.044
#> GSM614450 2 0.4951 0.7190 0.000 0.744 0.212 0.044
#> GSM614451 3 0.3837 1.0000 0.000 0.000 0.776 0.224
#> GSM614452 3 0.3837 1.0000 0.000 0.000 0.776 0.224
#> GSM614453 2 0.0000 0.8582 0.000 1.000 0.000 0.000
#> GSM614454 2 0.0000 0.8582 0.000 1.000 0.000 0.000
#> GSM614455 2 0.0000 0.8582 0.000 1.000 0.000 0.000
#> GSM614456 2 0.0000 0.8582 0.000 1.000 0.000 0.000
#> GSM614457 2 0.0000 0.8582 0.000 1.000 0.000 0.000
#> GSM614458 2 0.0000 0.8582 0.000 1.000 0.000 0.000
#> GSM614459 2 0.0000 0.8582 0.000 1.000 0.000 0.000
#> GSM614460 2 0.0000 0.8582 0.000 1.000 0.000 0.000
#> GSM614461 2 0.0000 0.8582 0.000 1.000 0.000 0.000
#> GSM614462 2 0.0000 0.8582 0.000 1.000 0.000 0.000
#> GSM614463 2 0.0000 0.8582 0.000 1.000 0.000 0.000
#> GSM614464 2 0.0000 0.8582 0.000 1.000 0.000 0.000
#> GSM614465 2 0.0000 0.8582 0.000 1.000 0.000 0.000
#> GSM614466 2 0.0000 0.8582 0.000 1.000 0.000 0.000
#> GSM614467 2 0.0000 0.8582 0.000 1.000 0.000 0.000
#> GSM614468 2 0.0000 0.8582 0.000 1.000 0.000 0.000
#> GSM614469 2 0.1629 0.8452 0.024 0.952 0.024 0.000
#> GSM614470 2 0.1629 0.8452 0.024 0.952 0.024 0.000
#> GSM614471 2 0.1629 0.8452 0.024 0.952 0.024 0.000
#> GSM614472 2 0.1629 0.8452 0.024 0.952 0.024 0.000
#> GSM614473 2 0.1629 0.8452 0.024 0.952 0.024 0.000
#> GSM614474 2 0.1629 0.8452 0.024 0.952 0.024 0.000
#> GSM614475 2 0.1629 0.8452 0.024 0.952 0.024 0.000
#> GSM614476 2 0.1629 0.8452 0.024 0.952 0.024 0.000
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM614415 5 0.4443 0.4316 0.472 0.000 0.004 0.000 0.524
#> GSM614416 5 0.4443 0.4316 0.472 0.000 0.004 0.000 0.524
#> GSM614417 5 0.4443 0.4316 0.472 0.000 0.004 0.000 0.524
#> GSM614418 5 0.4443 0.4316 0.472 0.000 0.004 0.000 0.524
#> GSM614419 5 0.4443 0.4316 0.472 0.000 0.004 0.000 0.524
#> GSM614420 5 0.4443 0.4316 0.472 0.000 0.004 0.000 0.524
#> GSM614421 2 0.4111 0.7008 0.008 0.708 0.280 0.004 0.000
#> GSM614422 2 0.4111 0.7008 0.008 0.708 0.280 0.004 0.000
#> GSM614423 2 0.4111 0.7008 0.008 0.708 0.280 0.004 0.000
#> GSM614424 2 0.4111 0.7008 0.008 0.708 0.280 0.004 0.000
#> GSM614425 2 0.4111 0.7008 0.008 0.708 0.280 0.004 0.000
#> GSM614426 2 0.4111 0.7008 0.008 0.708 0.280 0.004 0.000
#> GSM614427 2 0.4111 0.7008 0.008 0.708 0.280 0.004 0.000
#> GSM614428 2 0.4111 0.7008 0.008 0.708 0.280 0.004 0.000
#> GSM614429 2 0.0162 0.9032 0.000 0.996 0.004 0.000 0.000
#> GSM614430 2 0.0162 0.9032 0.000 0.996 0.004 0.000 0.000
#> GSM614431 2 0.0162 0.9032 0.000 0.996 0.004 0.000 0.000
#> GSM614432 2 0.0162 0.9032 0.000 0.996 0.004 0.000 0.000
#> GSM614433 2 0.0162 0.9032 0.000 0.996 0.004 0.000 0.000
#> GSM614434 2 0.0162 0.9032 0.000 0.996 0.004 0.000 0.000
#> GSM614435 2 0.0162 0.9032 0.000 0.996 0.004 0.000 0.000
#> GSM614436 2 0.0162 0.9032 0.000 0.996 0.004 0.000 0.000
#> GSM614437 4 0.0000 1.0000 0.000 0.000 0.000 1.000 0.000
#> GSM614438 4 0.0000 1.0000 0.000 0.000 0.000 1.000 0.000
#> GSM614439 4 0.0000 1.0000 0.000 0.000 0.000 1.000 0.000
#> GSM614440 4 0.0000 1.0000 0.000 0.000 0.000 1.000 0.000
#> GSM614441 4 0.0000 1.0000 0.000 0.000 0.000 1.000 0.000
#> GSM614442 4 0.0000 1.0000 0.000 0.000 0.000 1.000 0.000
#> GSM614443 4 0.0000 1.0000 0.000 0.000 0.000 1.000 0.000
#> GSM614444 4 0.0000 1.0000 0.000 0.000 0.000 1.000 0.000
#> GSM614391 5 0.0000 0.6670 0.000 0.000 0.000 0.000 1.000
#> GSM614392 5 0.0000 0.6670 0.000 0.000 0.000 0.000 1.000
#> GSM614393 5 0.0000 0.6670 0.000 0.000 0.000 0.000 1.000
#> GSM614394 5 0.0000 0.6670 0.000 0.000 0.000 0.000 1.000
#> GSM614395 5 0.4249 0.0755 0.000 0.000 0.432 0.000 0.568
#> GSM614396 5 0.0000 0.6670 0.000 0.000 0.000 0.000 1.000
#> GSM614397 5 0.0000 0.6670 0.000 0.000 0.000 0.000 1.000
#> GSM614398 5 0.0000 0.6670 0.000 0.000 0.000 0.000 1.000
#> GSM614399 2 0.0798 0.8997 0.016 0.976 0.008 0.000 0.000
#> GSM614400 2 0.0798 0.8997 0.016 0.976 0.008 0.000 0.000
#> GSM614401 2 0.0798 0.8997 0.016 0.976 0.008 0.000 0.000
#> GSM614402 2 0.0798 0.8997 0.016 0.976 0.008 0.000 0.000
#> GSM614403 2 0.0798 0.8997 0.016 0.976 0.008 0.000 0.000
#> GSM614404 2 0.0798 0.8997 0.016 0.976 0.008 0.000 0.000
#> GSM614405 2 0.0693 0.9008 0.012 0.980 0.008 0.000 0.000
#> GSM614406 2 0.0693 0.9008 0.012 0.980 0.008 0.000 0.000
#> GSM614407 1 0.0000 1.0000 1.000 0.000 0.000 0.000 0.000
#> GSM614408 1 0.0000 1.0000 1.000 0.000 0.000 0.000 0.000
#> GSM614409 1 0.0000 1.0000 1.000 0.000 0.000 0.000 0.000
#> GSM614410 1 0.0000 1.0000 1.000 0.000 0.000 0.000 0.000
#> GSM614411 1 0.0000 1.0000 1.000 0.000 0.000 0.000 0.000
#> GSM614412 1 0.0000 1.0000 1.000 0.000 0.000 0.000 0.000
#> GSM614413 1 0.0000 1.0000 1.000 0.000 0.000 0.000 0.000
#> GSM614414 1 0.0000 1.0000 1.000 0.000 0.000 0.000 0.000
#> GSM614445 2 0.4086 0.7443 0.024 0.736 0.240 0.000 0.000
#> GSM614446 2 0.4086 0.7443 0.024 0.736 0.240 0.000 0.000
#> GSM614447 2 0.4086 0.7443 0.024 0.736 0.240 0.000 0.000
#> GSM614448 2 0.4086 0.7443 0.024 0.736 0.240 0.000 0.000
#> GSM614449 2 0.4086 0.7443 0.024 0.736 0.240 0.000 0.000
#> GSM614450 2 0.4086 0.7443 0.024 0.736 0.240 0.000 0.000
#> GSM614451 3 0.0404 1.0000 0.000 0.000 0.988 0.012 0.000
#> GSM614452 3 0.0404 1.0000 0.000 0.000 0.988 0.012 0.000
#> GSM614453 2 0.0000 0.9034 0.000 1.000 0.000 0.000 0.000
#> GSM614454 2 0.0000 0.9034 0.000 1.000 0.000 0.000 0.000
#> GSM614455 2 0.0000 0.9034 0.000 1.000 0.000 0.000 0.000
#> GSM614456 2 0.0000 0.9034 0.000 1.000 0.000 0.000 0.000
#> GSM614457 2 0.0000 0.9034 0.000 1.000 0.000 0.000 0.000
#> GSM614458 2 0.0000 0.9034 0.000 1.000 0.000 0.000 0.000
#> GSM614459 2 0.0000 0.9034 0.000 1.000 0.000 0.000 0.000
#> GSM614460 2 0.0000 0.9034 0.000 1.000 0.000 0.000 0.000
#> GSM614461 2 0.0000 0.9034 0.000 1.000 0.000 0.000 0.000
#> GSM614462 2 0.0000 0.9034 0.000 1.000 0.000 0.000 0.000
#> GSM614463 2 0.0000 0.9034 0.000 1.000 0.000 0.000 0.000
#> GSM614464 2 0.0000 0.9034 0.000 1.000 0.000 0.000 0.000
#> GSM614465 2 0.0000 0.9034 0.000 1.000 0.000 0.000 0.000
#> GSM614466 2 0.0000 0.9034 0.000 1.000 0.000 0.000 0.000
#> GSM614467 2 0.0000 0.9034 0.000 1.000 0.000 0.000 0.000
#> GSM614468 2 0.0000 0.9034 0.000 1.000 0.000 0.000 0.000
#> GSM614469 2 0.2017 0.8653 0.080 0.912 0.008 0.000 0.000
#> GSM614470 2 0.2017 0.8653 0.080 0.912 0.008 0.000 0.000
#> GSM614471 2 0.2017 0.8653 0.080 0.912 0.008 0.000 0.000
#> GSM614472 2 0.2017 0.8653 0.080 0.912 0.008 0.000 0.000
#> GSM614473 2 0.2017 0.8653 0.080 0.912 0.008 0.000 0.000
#> GSM614474 2 0.2017 0.8653 0.080 0.912 0.008 0.000 0.000
#> GSM614475 2 0.2017 0.8653 0.080 0.912 0.008 0.000 0.000
#> GSM614476 2 0.2017 0.8653 0.080 0.912 0.008 0.000 0.000
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM614415 1 0.2350 1.000 0.888 0.000 0.000 0 0.076 0.036
#> GSM614416 1 0.2350 1.000 0.888 0.000 0.000 0 0.076 0.036
#> GSM614417 1 0.2350 1.000 0.888 0.000 0.000 0 0.076 0.036
#> GSM614418 1 0.2350 1.000 0.888 0.000 0.000 0 0.076 0.036
#> GSM614419 1 0.2350 1.000 0.888 0.000 0.000 0 0.076 0.036
#> GSM614420 1 0.2350 1.000 0.888 0.000 0.000 0 0.076 0.036
#> GSM614421 2 0.3758 0.678 0.000 0.668 0.324 0 0.000 0.008
#> GSM614422 2 0.3758 0.678 0.000 0.668 0.324 0 0.000 0.008
#> GSM614423 2 0.3758 0.678 0.000 0.668 0.324 0 0.000 0.008
#> GSM614424 2 0.3758 0.678 0.000 0.668 0.324 0 0.000 0.008
#> GSM614425 2 0.3758 0.678 0.000 0.668 0.324 0 0.000 0.008
#> GSM614426 2 0.3758 0.678 0.000 0.668 0.324 0 0.000 0.008
#> GSM614427 2 0.3758 0.678 0.000 0.668 0.324 0 0.000 0.008
#> GSM614428 2 0.3758 0.678 0.000 0.668 0.324 0 0.000 0.008
#> GSM614429 2 0.0146 0.876 0.000 0.996 0.004 0 0.000 0.000
#> GSM614430 2 0.0146 0.876 0.000 0.996 0.004 0 0.000 0.000
#> GSM614431 2 0.0146 0.876 0.000 0.996 0.004 0 0.000 0.000
#> GSM614432 2 0.0146 0.876 0.000 0.996 0.004 0 0.000 0.000
#> GSM614433 2 0.0146 0.876 0.000 0.996 0.004 0 0.000 0.000
#> GSM614434 2 0.0146 0.876 0.000 0.996 0.004 0 0.000 0.000
#> GSM614435 2 0.0146 0.876 0.000 0.996 0.004 0 0.000 0.000
#> GSM614436 2 0.0146 0.876 0.000 0.996 0.004 0 0.000 0.000
#> GSM614437 4 0.0000 1.000 0.000 0.000 0.000 1 0.000 0.000
#> GSM614438 4 0.0000 1.000 0.000 0.000 0.000 1 0.000 0.000
#> GSM614439 4 0.0000 1.000 0.000 0.000 0.000 1 0.000 0.000
#> GSM614440 4 0.0000 1.000 0.000 0.000 0.000 1 0.000 0.000
#> GSM614441 4 0.0000 1.000 0.000 0.000 0.000 1 0.000 0.000
#> GSM614442 4 0.0000 1.000 0.000 0.000 0.000 1 0.000 0.000
#> GSM614443 4 0.0000 1.000 0.000 0.000 0.000 1 0.000 0.000
#> GSM614444 4 0.0000 1.000 0.000 0.000 0.000 1 0.000 0.000
#> GSM614391 5 0.0000 0.938 0.000 0.000 0.000 0 1.000 0.000
#> GSM614392 5 0.0000 0.938 0.000 0.000 0.000 0 1.000 0.000
#> GSM614393 5 0.0000 0.938 0.000 0.000 0.000 0 1.000 0.000
#> GSM614394 5 0.0000 0.938 0.000 0.000 0.000 0 1.000 0.000
#> GSM614395 5 0.4199 0.335 0.016 0.000 0.416 0 0.568 0.000
#> GSM614396 5 0.0000 0.938 0.000 0.000 0.000 0 1.000 0.000
#> GSM614397 5 0.0000 0.938 0.000 0.000 0.000 0 1.000 0.000
#> GSM614398 5 0.0000 0.938 0.000 0.000 0.000 0 1.000 0.000
#> GSM614399 2 0.2039 0.864 0.076 0.904 0.020 0 0.000 0.000
#> GSM614400 2 0.2039 0.864 0.076 0.904 0.020 0 0.000 0.000
#> GSM614401 2 0.2039 0.864 0.076 0.904 0.020 0 0.000 0.000
#> GSM614402 2 0.2039 0.864 0.076 0.904 0.020 0 0.000 0.000
#> GSM614403 2 0.2039 0.864 0.076 0.904 0.020 0 0.000 0.000
#> GSM614404 2 0.2039 0.864 0.076 0.904 0.020 0 0.000 0.000
#> GSM614405 2 0.1983 0.865 0.072 0.908 0.020 0 0.000 0.000
#> GSM614406 2 0.1983 0.865 0.072 0.908 0.020 0 0.000 0.000
#> GSM614407 6 0.0000 1.000 0.000 0.000 0.000 0 0.000 1.000
#> GSM614408 6 0.0000 1.000 0.000 0.000 0.000 0 0.000 1.000
#> GSM614409 6 0.0000 1.000 0.000 0.000 0.000 0 0.000 1.000
#> GSM614410 6 0.0000 1.000 0.000 0.000 0.000 0 0.000 1.000
#> GSM614411 6 0.0000 1.000 0.000 0.000 0.000 0 0.000 1.000
#> GSM614412 6 0.0000 1.000 0.000 0.000 0.000 0 0.000 1.000
#> GSM614413 6 0.0000 1.000 0.000 0.000 0.000 0 0.000 1.000
#> GSM614414 6 0.0000 1.000 0.000 0.000 0.000 0 0.000 1.000
#> GSM614445 2 0.4436 0.710 0.044 0.676 0.272 0 0.000 0.008
#> GSM614446 2 0.4436 0.710 0.044 0.676 0.272 0 0.000 0.008
#> GSM614447 2 0.4436 0.710 0.044 0.676 0.272 0 0.000 0.008
#> GSM614448 2 0.4436 0.710 0.044 0.676 0.272 0 0.000 0.008
#> GSM614449 2 0.4436 0.710 0.044 0.676 0.272 0 0.000 0.008
#> GSM614450 2 0.4436 0.710 0.044 0.676 0.272 0 0.000 0.008
#> GSM614451 3 0.0937 1.000 0.040 0.000 0.960 0 0.000 0.000
#> GSM614452 3 0.0937 1.000 0.040 0.000 0.960 0 0.000 0.000
#> GSM614453 2 0.0000 0.876 0.000 1.000 0.000 0 0.000 0.000
#> GSM614454 2 0.0000 0.876 0.000 1.000 0.000 0 0.000 0.000
#> GSM614455 2 0.0000 0.876 0.000 1.000 0.000 0 0.000 0.000
#> GSM614456 2 0.0000 0.876 0.000 1.000 0.000 0 0.000 0.000
#> GSM614457 2 0.0000 0.876 0.000 1.000 0.000 0 0.000 0.000
#> GSM614458 2 0.0000 0.876 0.000 1.000 0.000 0 0.000 0.000
#> GSM614459 2 0.0000 0.876 0.000 1.000 0.000 0 0.000 0.000
#> GSM614460 2 0.0000 0.876 0.000 1.000 0.000 0 0.000 0.000
#> GSM614461 2 0.0000 0.876 0.000 1.000 0.000 0 0.000 0.000
#> GSM614462 2 0.0000 0.876 0.000 1.000 0.000 0 0.000 0.000
#> GSM614463 2 0.0000 0.876 0.000 1.000 0.000 0 0.000 0.000
#> GSM614464 2 0.0000 0.876 0.000 1.000 0.000 0 0.000 0.000
#> GSM614465 2 0.0000 0.876 0.000 1.000 0.000 0 0.000 0.000
#> GSM614466 2 0.0000 0.876 0.000 1.000 0.000 0 0.000 0.000
#> GSM614467 2 0.0000 0.876 0.000 1.000 0.000 0 0.000 0.000
#> GSM614468 2 0.0000 0.876 0.000 1.000 0.000 0 0.000 0.000
#> GSM614469 2 0.2939 0.839 0.120 0.848 0.016 0 0.000 0.016
#> GSM614470 2 0.2939 0.839 0.120 0.848 0.016 0 0.000 0.016
#> GSM614471 2 0.2939 0.839 0.120 0.848 0.016 0 0.000 0.016
#> GSM614472 2 0.2939 0.839 0.120 0.848 0.016 0 0.000 0.016
#> GSM614473 2 0.2939 0.839 0.120 0.848 0.016 0 0.000 0.016
#> GSM614474 2 0.2939 0.839 0.120 0.848 0.016 0 0.000 0.016
#> GSM614475 2 0.2939 0.839 0.120 0.848 0.016 0 0.000 0.016
#> GSM614476 2 0.2939 0.839 0.120 0.848 0.016 0 0.000 0.016
Heatmaps for the consensus matrix. It visualizes the probability of two samples to be in a same group.
consensus_heatmap(res, k = 2)
consensus_heatmap(res, k = 3)
consensus_heatmap(res, k = 4)
consensus_heatmap(res, k = 5)
consensus_heatmap(res, k = 6)
Heatmaps for the membership of samples in all partitions to see how consistent they are:
membership_heatmap(res, k = 2)
membership_heatmap(res, k = 3)
membership_heatmap(res, k = 4)
membership_heatmap(res, k = 5)
membership_heatmap(res, k = 6)
As soon as we have had the classes for columns, we can look for signatures which are significantly different between classes which can be candidate marks for certain classes. Following are the heatmaps for signatures.
Signature heatmaps where rows are scaled:
get_signatures(res, k = 2)
get_signatures(res, k = 3)
get_signatures(res, k = 4)
get_signatures(res, k = 5)
get_signatures(res, k = 6)
Signature heatmaps where rows are not scaled:
get_signatures(res, k = 2, scale_rows = FALSE)
get_signatures(res, k = 3, scale_rows = FALSE)
get_signatures(res, k = 4, scale_rows = FALSE)
get_signatures(res, k = 5, scale_rows = FALSE)
get_signatures(res, k = 6, scale_rows = FALSE)
Compare the overlap of signatures from different k:
compare_signatures(res)
get_signature() returns a data frame invisibly. TO get the list of signatures, the function
call should be assigned to a variable explicitly. In following code, if plot argument is set
to FALSE, no heatmap is plotted while only the differential analysis is performed.
# code only for demonstration
tb = get_signature(res, k = ..., plot = FALSE)
An example of the output of tb is:
#> which_row fdr mean_1 mean_2 scaled_mean_1 scaled_mean_2 km
#> 1 38 0.042760348 8.373488 9.131774 -0.5533452 0.5164555 1
#> 2 40 0.018707592 7.106213 8.469186 -0.6173731 0.5762149 1
#> 3 55 0.019134737 10.221463 11.207825 -0.6159697 0.5749050 1
#> 4 59 0.006059896 5.921854 7.869574 -0.6899429 0.6439467 1
#> 5 60 0.018055526 8.928898 10.211722 -0.6204761 0.5791110 1
#> 6 98 0.009384629 15.714769 14.887706 0.6635654 -0.6193277 2
...
The columns in tb are:
which_row: row indices corresponding to the input matrix.fdr: FDR for the differential test. mean_x: The mean value in group x.scaled_mean_x: The mean value in group x after rows are scaled.km: Row groups if k-means clustering is applied to rows.UMAP plot which shows how samples are separated.
dimension_reduction(res, k = 2, method = "UMAP")
dimension_reduction(res, k = 3, method = "UMAP")
dimension_reduction(res, k = 4, method = "UMAP")
dimension_reduction(res, k = 5, method = "UMAP")
dimension_reduction(res, k = 6, method = "UMAP")
Following heatmap shows how subgroups are split when increasing k:
collect_classes(res)
Test correlation between subgroups and known annotations. If the known annotation is numeric, one-way ANOVA test is applied, and if the known annotation is discrete, chi-squared contingency table test is applied.
test_to_known_factors(res)
#> n individual(p) protocol(p) time(p) other(p) k
#> SD:hclust 85 3.47e-11 0.3626 0.992 0.004489 2
#> SD:hclust 85 6.33e-23 0.3646 1.000 0.020415 3
#> SD:hclust 77 6.78e-23 0.0697 1.000 0.000256 4
#> SD:hclust 79 1.04e-34 0.1838 1.000 0.001081 5
#> SD:hclust 85 2.70e-48 0.1300 1.000 0.001105 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 21168 rows and 86 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 0.163 0.437 0.683 0.3825 0.665 0.665
#> 3 3 0.285 0.685 0.798 0.4364 0.741 0.627
#> 4 4 0.343 0.570 0.705 0.2314 0.807 0.602
#> 5 5 0.519 0.622 0.723 0.0983 0.844 0.543
#> 6 6 0.638 0.655 0.711 0.0563 0.958 0.812
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
#> GSM614415 1 0.9922 0.623 0.552 0.448
#> GSM614416 1 0.9922 0.623 0.552 0.448
#> GSM614417 1 0.9922 0.623 0.552 0.448
#> GSM614418 1 0.9922 0.623 0.552 0.448
#> GSM614419 1 0.9358 0.737 0.648 0.352
#> GSM614420 1 0.9358 0.737 0.648 0.352
#> GSM614421 2 0.9522 0.286 0.372 0.628
#> GSM614422 2 0.9552 0.280 0.376 0.624
#> GSM614423 2 0.6438 0.528 0.164 0.836
#> GSM614424 2 0.9522 0.286 0.372 0.628
#> GSM614425 2 0.9522 0.286 0.372 0.628
#> GSM614426 2 0.9522 0.286 0.372 0.628
#> GSM614427 2 0.9580 0.272 0.380 0.620
#> GSM614428 2 0.9909 0.177 0.444 0.556
#> GSM614429 2 0.0672 0.628 0.008 0.992
#> GSM614430 2 0.0938 0.627 0.012 0.988
#> GSM614431 2 0.0672 0.628 0.008 0.992
#> GSM614432 2 0.0672 0.628 0.008 0.992
#> GSM614433 2 0.0672 0.628 0.008 0.992
#> GSM614434 2 0.0672 0.628 0.008 0.992
#> GSM614435 2 0.3114 0.608 0.056 0.944
#> GSM614436 2 0.6973 0.510 0.188 0.812
#> GSM614437 2 0.9909 0.243 0.444 0.556
#> GSM614438 2 0.9977 0.217 0.472 0.528
#> GSM614439 2 0.9977 0.217 0.472 0.528
#> GSM614440 2 0.9977 0.217 0.472 0.528
#> GSM614441 2 0.9977 0.217 0.472 0.528
#> GSM614442 2 0.9977 0.217 0.472 0.528
#> GSM614443 2 0.9963 0.225 0.464 0.536
#> GSM614444 2 0.9977 0.217 0.472 0.528
#> GSM614391 1 0.9248 0.738 0.660 0.340
#> GSM614392 1 0.9522 0.723 0.628 0.372
#> GSM614393 1 0.9522 0.723 0.628 0.372
#> GSM614394 1 0.9248 0.738 0.660 0.340
#> GSM614395 1 0.5294 0.521 0.880 0.120
#> GSM614396 1 0.9248 0.738 0.660 0.340
#> GSM614397 1 0.7674 0.638 0.776 0.224
#> GSM614398 1 0.8499 0.688 0.724 0.276
#> GSM614399 2 0.5059 0.585 0.112 0.888
#> GSM614400 2 0.5059 0.585 0.112 0.888
#> GSM614401 2 0.5059 0.585 0.112 0.888
#> GSM614402 2 0.5059 0.585 0.112 0.888
#> GSM614403 2 0.5178 0.582 0.116 0.884
#> GSM614404 2 0.5059 0.585 0.112 0.888
#> GSM614405 2 0.5059 0.585 0.112 0.888
#> GSM614406 2 0.9323 0.347 0.348 0.652
#> GSM614407 2 0.9977 -0.446 0.472 0.528
#> GSM614408 2 0.9977 -0.446 0.472 0.528
#> GSM614409 2 0.9983 -0.456 0.476 0.524
#> GSM614410 2 0.9977 -0.446 0.472 0.528
#> GSM614411 2 0.9983 -0.456 0.476 0.524
#> GSM614412 2 0.9983 -0.456 0.476 0.524
#> GSM614413 1 0.9522 0.649 0.628 0.372
#> GSM614414 1 0.9522 0.649 0.628 0.372
#> GSM614445 2 0.5408 0.576 0.124 0.876
#> GSM614446 2 0.5408 0.576 0.124 0.876
#> GSM614447 2 0.5408 0.576 0.124 0.876
#> GSM614448 2 0.9209 0.364 0.336 0.664
#> GSM614449 2 0.8267 0.459 0.260 0.740
#> GSM614450 2 0.5737 0.567 0.136 0.864
#> GSM614451 1 0.9909 -0.121 0.556 0.444
#> GSM614452 1 0.9909 -0.121 0.556 0.444
#> GSM614453 2 0.3274 0.606 0.060 0.940
#> GSM614454 2 0.3274 0.606 0.060 0.940
#> GSM614455 2 0.3274 0.606 0.060 0.940
#> GSM614456 2 0.3274 0.606 0.060 0.940
#> GSM614457 2 0.3274 0.606 0.060 0.940
#> GSM614458 2 0.3274 0.606 0.060 0.940
#> GSM614459 2 0.3274 0.606 0.060 0.940
#> GSM614460 2 0.3274 0.606 0.060 0.940
#> GSM614461 2 0.0000 0.629 0.000 1.000
#> GSM614462 2 0.0000 0.629 0.000 1.000
#> GSM614463 2 0.0000 0.629 0.000 1.000
#> GSM614464 2 0.0000 0.629 0.000 1.000
#> GSM614465 2 0.0000 0.629 0.000 1.000
#> GSM614466 2 0.0000 0.629 0.000 1.000
#> GSM614467 2 0.0000 0.629 0.000 1.000
#> GSM614468 2 0.0000 0.629 0.000 1.000
#> GSM614469 2 0.8016 0.367 0.244 0.756
#> GSM614470 2 0.8016 0.367 0.244 0.756
#> GSM614471 2 0.8016 0.367 0.244 0.756
#> GSM614472 2 0.8016 0.367 0.244 0.756
#> GSM614473 2 0.8016 0.367 0.244 0.756
#> GSM614474 2 0.8016 0.367 0.244 0.756
#> GSM614475 2 0.6973 0.486 0.188 0.812
#> GSM614476 2 0.6343 0.539 0.160 0.840
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM614415 1 0.2955 0.777 0.912 0.080 0.008
#> GSM614416 1 0.2955 0.777 0.912 0.080 0.008
#> GSM614417 1 0.2955 0.777 0.912 0.080 0.008
#> GSM614418 1 0.2955 0.777 0.912 0.080 0.008
#> GSM614419 1 0.2492 0.774 0.936 0.048 0.016
#> GSM614420 1 0.2492 0.774 0.936 0.048 0.016
#> GSM614421 2 0.9501 0.406 0.224 0.488 0.288
#> GSM614422 2 0.9501 0.406 0.224 0.488 0.288
#> GSM614423 2 0.6915 0.704 0.140 0.736 0.124
#> GSM614424 2 0.9501 0.406 0.224 0.488 0.288
#> GSM614425 2 0.9501 0.406 0.224 0.488 0.288
#> GSM614426 2 0.9501 0.406 0.224 0.488 0.288
#> GSM614427 2 0.9501 0.406 0.224 0.488 0.288
#> GSM614428 2 0.9519 0.397 0.224 0.484 0.292
#> GSM614429 2 0.0237 0.717 0.000 0.996 0.004
#> GSM614430 2 0.0237 0.717 0.000 0.996 0.004
#> GSM614431 2 0.0237 0.717 0.000 0.996 0.004
#> GSM614432 2 0.0237 0.717 0.000 0.996 0.004
#> GSM614433 2 0.0000 0.718 0.000 1.000 0.000
#> GSM614434 2 0.0237 0.717 0.000 0.996 0.004
#> GSM614435 2 0.0592 0.712 0.000 0.988 0.012
#> GSM614436 2 0.2711 0.652 0.000 0.912 0.088
#> GSM614437 3 0.5588 0.921 0.004 0.276 0.720
#> GSM614438 3 0.5728 0.925 0.008 0.272 0.720
#> GSM614439 3 0.5728 0.925 0.008 0.272 0.720
#> GSM614440 3 0.5728 0.925 0.008 0.272 0.720
#> GSM614441 3 0.5728 0.925 0.008 0.272 0.720
#> GSM614442 3 0.5728 0.925 0.008 0.272 0.720
#> GSM614443 3 0.5588 0.921 0.004 0.276 0.720
#> GSM614444 3 0.5728 0.925 0.008 0.272 0.720
#> GSM614391 1 0.3886 0.749 0.880 0.024 0.096
#> GSM614392 1 0.3886 0.749 0.880 0.024 0.096
#> GSM614393 1 0.3886 0.749 0.880 0.024 0.096
#> GSM614394 1 0.3886 0.749 0.880 0.024 0.096
#> GSM614395 1 0.4978 0.629 0.780 0.004 0.216
#> GSM614396 1 0.3886 0.749 0.880 0.024 0.096
#> GSM614397 1 0.4164 0.712 0.848 0.008 0.144
#> GSM614398 1 0.3989 0.731 0.864 0.012 0.124
#> GSM614399 2 0.6809 0.715 0.156 0.740 0.104
#> GSM614400 2 0.6809 0.715 0.156 0.740 0.104
#> GSM614401 2 0.6809 0.715 0.156 0.740 0.104
#> GSM614402 2 0.6809 0.715 0.156 0.740 0.104
#> GSM614403 2 0.6455 0.722 0.128 0.764 0.108
#> GSM614404 2 0.6809 0.715 0.156 0.740 0.104
#> GSM614405 2 0.6583 0.720 0.136 0.756 0.108
#> GSM614406 2 0.8464 0.528 0.128 0.592 0.280
#> GSM614407 1 0.7446 0.613 0.664 0.260 0.076
#> GSM614408 1 0.7446 0.613 0.664 0.260 0.076
#> GSM614409 1 0.7446 0.613 0.664 0.260 0.076
#> GSM614410 1 0.7446 0.613 0.664 0.260 0.076
#> GSM614411 1 0.7446 0.613 0.664 0.260 0.076
#> GSM614412 1 0.7376 0.622 0.672 0.252 0.076
#> GSM614413 1 0.7605 0.650 0.684 0.192 0.124
#> GSM614414 1 0.7605 0.650 0.684 0.192 0.124
#> GSM614445 2 0.6079 0.720 0.088 0.784 0.128
#> GSM614446 2 0.6079 0.720 0.088 0.784 0.128
#> GSM614447 2 0.6079 0.720 0.088 0.784 0.128
#> GSM614448 2 0.8622 0.505 0.132 0.572 0.296
#> GSM614449 2 0.7935 0.612 0.116 0.648 0.236
#> GSM614450 2 0.6079 0.720 0.088 0.784 0.128
#> GSM614451 3 0.7059 0.670 0.092 0.192 0.716
#> GSM614452 3 0.7059 0.670 0.092 0.192 0.716
#> GSM614453 2 0.2496 0.677 0.004 0.928 0.068
#> GSM614454 2 0.2496 0.677 0.004 0.928 0.068
#> GSM614455 2 0.2496 0.677 0.004 0.928 0.068
#> GSM614456 2 0.2590 0.673 0.004 0.924 0.072
#> GSM614457 2 0.2590 0.673 0.004 0.924 0.072
#> GSM614458 2 0.2496 0.677 0.004 0.928 0.068
#> GSM614459 2 0.2590 0.673 0.004 0.924 0.072
#> GSM614460 2 0.2590 0.673 0.004 0.924 0.072
#> GSM614461 2 0.1015 0.717 0.008 0.980 0.012
#> GSM614462 2 0.1015 0.717 0.008 0.980 0.012
#> GSM614463 2 0.1015 0.717 0.008 0.980 0.012
#> GSM614464 2 0.1015 0.717 0.008 0.980 0.012
#> GSM614465 2 0.1015 0.717 0.008 0.980 0.012
#> GSM614466 2 0.1015 0.717 0.008 0.980 0.012
#> GSM614467 2 0.0829 0.715 0.004 0.984 0.012
#> GSM614468 2 0.1015 0.717 0.008 0.980 0.012
#> GSM614469 2 0.7477 0.592 0.284 0.648 0.068
#> GSM614470 2 0.7477 0.592 0.284 0.648 0.068
#> GSM614471 2 0.7477 0.592 0.284 0.648 0.068
#> GSM614472 2 0.7477 0.592 0.284 0.648 0.068
#> GSM614473 2 0.7477 0.592 0.284 0.648 0.068
#> GSM614474 2 0.7477 0.592 0.284 0.648 0.068
#> GSM614475 2 0.7376 0.633 0.252 0.672 0.076
#> GSM614476 2 0.7413 0.659 0.224 0.684 0.092
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM614415 1 0.2694 0.6873 0.916 0.016 0.044 0.024
#> GSM614416 1 0.2694 0.6873 0.916 0.016 0.044 0.024
#> GSM614417 1 0.2694 0.6873 0.916 0.016 0.044 0.024
#> GSM614418 1 0.2694 0.6873 0.916 0.016 0.044 0.024
#> GSM614419 1 0.3159 0.6875 0.892 0.012 0.068 0.028
#> GSM614420 1 0.3159 0.6875 0.892 0.012 0.068 0.028
#> GSM614421 3 0.8227 0.7064 0.096 0.172 0.572 0.160
#> GSM614422 3 0.8190 0.7068 0.096 0.172 0.576 0.156
#> GSM614423 3 0.6396 0.5724 0.064 0.328 0.600 0.008
#> GSM614424 3 0.8227 0.7064 0.096 0.172 0.572 0.160
#> GSM614425 3 0.8227 0.7064 0.096 0.172 0.572 0.160
#> GSM614426 3 0.8227 0.7064 0.096 0.172 0.572 0.160
#> GSM614427 3 0.8227 0.7064 0.096 0.172 0.572 0.160
#> GSM614428 3 0.8227 0.7064 0.096 0.172 0.572 0.160
#> GSM614429 2 0.0000 0.6606 0.000 1.000 0.000 0.000
#> GSM614430 2 0.0000 0.6606 0.000 1.000 0.000 0.000
#> GSM614431 2 0.0000 0.6606 0.000 1.000 0.000 0.000
#> GSM614432 2 0.0000 0.6606 0.000 1.000 0.000 0.000
#> GSM614433 2 0.0336 0.6589 0.000 0.992 0.008 0.000
#> GSM614434 2 0.0000 0.6606 0.000 1.000 0.000 0.000
#> GSM614435 2 0.0188 0.6596 0.000 0.996 0.000 0.004
#> GSM614436 2 0.1743 0.6168 0.000 0.940 0.004 0.056
#> GSM614437 4 0.3402 0.9899 0.000 0.164 0.004 0.832
#> GSM614438 4 0.3498 0.9966 0.000 0.160 0.008 0.832
#> GSM614439 4 0.3498 0.9966 0.000 0.160 0.008 0.832
#> GSM614440 4 0.3498 0.9966 0.000 0.160 0.008 0.832
#> GSM614441 4 0.3498 0.9966 0.000 0.160 0.008 0.832
#> GSM614442 4 0.3498 0.9966 0.000 0.160 0.008 0.832
#> GSM614443 4 0.3402 0.9899 0.000 0.164 0.004 0.832
#> GSM614444 4 0.3498 0.9966 0.000 0.160 0.008 0.832
#> GSM614391 1 0.4139 0.6732 0.816 0.000 0.144 0.040
#> GSM614392 1 0.4139 0.6732 0.816 0.000 0.144 0.040
#> GSM614393 1 0.4139 0.6732 0.816 0.000 0.144 0.040
#> GSM614394 1 0.4237 0.6691 0.808 0.000 0.152 0.040
#> GSM614395 1 0.5663 0.5243 0.676 0.000 0.264 0.060
#> GSM614396 1 0.4237 0.6691 0.808 0.000 0.152 0.040
#> GSM614397 1 0.4831 0.6189 0.752 0.000 0.208 0.040
#> GSM614398 1 0.4755 0.6271 0.760 0.000 0.200 0.040
#> GSM614399 2 0.8235 0.3374 0.160 0.492 0.304 0.044
#> GSM614400 2 0.8235 0.3374 0.160 0.492 0.304 0.044
#> GSM614401 2 0.8235 0.3374 0.160 0.492 0.304 0.044
#> GSM614402 2 0.8235 0.3374 0.160 0.492 0.304 0.044
#> GSM614403 2 0.7802 0.0354 0.112 0.436 0.420 0.032
#> GSM614404 2 0.8235 0.3374 0.160 0.492 0.304 0.044
#> GSM614405 2 0.8185 0.2268 0.136 0.456 0.364 0.044
#> GSM614406 3 0.7429 0.6287 0.020 0.240 0.580 0.160
#> GSM614407 1 0.8272 0.4368 0.508 0.148 0.288 0.056
#> GSM614408 1 0.8272 0.4368 0.508 0.148 0.288 0.056
#> GSM614409 1 0.8231 0.4361 0.508 0.140 0.296 0.056
#> GSM614410 1 0.8272 0.4368 0.508 0.148 0.288 0.056
#> GSM614411 1 0.8231 0.4361 0.508 0.140 0.296 0.056
#> GSM614412 1 0.8155 0.4406 0.516 0.132 0.296 0.056
#> GSM614413 3 0.7679 -0.3365 0.428 0.048 0.448 0.076
#> GSM614414 3 0.7679 -0.3365 0.428 0.048 0.448 0.076
#> GSM614445 3 0.6782 0.4188 0.060 0.400 0.524 0.016
#> GSM614446 3 0.6612 0.5137 0.056 0.360 0.568 0.016
#> GSM614447 3 0.6746 0.4565 0.060 0.384 0.540 0.016
#> GSM614448 3 0.7521 0.6870 0.040 0.208 0.604 0.148
#> GSM614449 3 0.7170 0.6667 0.036 0.256 0.612 0.096
#> GSM614450 3 0.6555 0.5395 0.056 0.344 0.584 0.016
#> GSM614451 3 0.7588 0.2847 0.040 0.080 0.464 0.416
#> GSM614452 3 0.7588 0.2847 0.040 0.080 0.464 0.416
#> GSM614453 2 0.3082 0.6270 0.000 0.884 0.032 0.084
#> GSM614454 2 0.3082 0.6270 0.000 0.884 0.032 0.084
#> GSM614455 2 0.3082 0.6270 0.000 0.884 0.032 0.084
#> GSM614456 2 0.3082 0.6270 0.000 0.884 0.032 0.084
#> GSM614457 2 0.3082 0.6270 0.000 0.884 0.032 0.084
#> GSM614458 2 0.3013 0.6294 0.000 0.888 0.032 0.080
#> GSM614459 2 0.3082 0.6270 0.000 0.884 0.032 0.084
#> GSM614460 2 0.3082 0.6270 0.000 0.884 0.032 0.084
#> GSM614461 2 0.2408 0.6545 0.004 0.920 0.060 0.016
#> GSM614462 2 0.2408 0.6545 0.004 0.920 0.060 0.016
#> GSM614463 2 0.2408 0.6545 0.004 0.920 0.060 0.016
#> GSM614464 2 0.2408 0.6545 0.004 0.920 0.060 0.016
#> GSM614465 2 0.2408 0.6545 0.004 0.920 0.060 0.016
#> GSM614466 2 0.2408 0.6545 0.004 0.920 0.060 0.016
#> GSM614467 2 0.2408 0.6545 0.004 0.920 0.060 0.016
#> GSM614468 2 0.2408 0.6545 0.004 0.920 0.060 0.016
#> GSM614469 2 0.8367 0.3086 0.328 0.428 0.216 0.028
#> GSM614470 2 0.8367 0.3086 0.328 0.428 0.216 0.028
#> GSM614471 2 0.8367 0.3086 0.328 0.428 0.216 0.028
#> GSM614472 2 0.8367 0.3086 0.328 0.428 0.216 0.028
#> GSM614473 2 0.8367 0.3086 0.328 0.428 0.216 0.028
#> GSM614474 2 0.8367 0.3086 0.328 0.428 0.216 0.028
#> GSM614475 2 0.8390 0.3066 0.320 0.428 0.224 0.028
#> GSM614476 2 0.8450 0.2868 0.280 0.428 0.264 0.028
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM614415 5 0.6100 0.6615 0.276 0.008 0.100 0.012 0.604
#> GSM614416 5 0.6100 0.6615 0.276 0.008 0.100 0.012 0.604
#> GSM614417 5 0.6100 0.6615 0.276 0.008 0.100 0.012 0.604
#> GSM614418 5 0.6100 0.6615 0.276 0.008 0.100 0.012 0.604
#> GSM614419 5 0.5868 0.6750 0.264 0.000 0.100 0.016 0.620
#> GSM614420 5 0.5868 0.6750 0.264 0.000 0.100 0.016 0.620
#> GSM614421 3 0.6270 0.7864 0.064 0.080 0.704 0.100 0.052
#> GSM614422 3 0.6280 0.7845 0.068 0.080 0.704 0.096 0.052
#> GSM614423 3 0.5408 0.7404 0.088 0.148 0.728 0.012 0.024
#> GSM614424 3 0.6270 0.7864 0.064 0.080 0.704 0.100 0.052
#> GSM614425 3 0.6270 0.7864 0.064 0.080 0.704 0.100 0.052
#> GSM614426 3 0.6270 0.7864 0.064 0.080 0.704 0.100 0.052
#> GSM614427 3 0.6270 0.7864 0.064 0.080 0.704 0.100 0.052
#> GSM614428 3 0.6328 0.7845 0.068 0.080 0.700 0.100 0.052
#> GSM614429 2 0.1364 0.7673 0.012 0.952 0.036 0.000 0.000
#> GSM614430 2 0.1364 0.7673 0.012 0.952 0.036 0.000 0.000
#> GSM614431 2 0.1364 0.7673 0.012 0.952 0.036 0.000 0.000
#> GSM614432 2 0.1364 0.7673 0.012 0.952 0.036 0.000 0.000
#> GSM614433 2 0.1364 0.7673 0.012 0.952 0.036 0.000 0.000
#> GSM614434 2 0.1364 0.7673 0.012 0.952 0.036 0.000 0.000
#> GSM614435 2 0.1412 0.7673 0.008 0.952 0.036 0.004 0.000
#> GSM614436 2 0.1756 0.7625 0.008 0.940 0.036 0.016 0.000
#> GSM614437 4 0.1197 1.0000 0.000 0.048 0.000 0.952 0.000
#> GSM614438 4 0.1197 1.0000 0.000 0.048 0.000 0.952 0.000
#> GSM614439 4 0.1197 1.0000 0.000 0.048 0.000 0.952 0.000
#> GSM614440 4 0.1197 1.0000 0.000 0.048 0.000 0.952 0.000
#> GSM614441 4 0.1197 1.0000 0.000 0.048 0.000 0.952 0.000
#> GSM614442 4 0.1197 1.0000 0.000 0.048 0.000 0.952 0.000
#> GSM614443 4 0.1197 1.0000 0.000 0.048 0.000 0.952 0.000
#> GSM614444 4 0.1197 1.0000 0.000 0.048 0.000 0.952 0.000
#> GSM614391 5 0.0451 0.7781 0.004 0.000 0.008 0.000 0.988
#> GSM614392 5 0.0451 0.7781 0.004 0.000 0.008 0.000 0.988
#> GSM614393 5 0.0451 0.7781 0.004 0.000 0.008 0.000 0.988
#> GSM614394 5 0.0451 0.7781 0.004 0.000 0.008 0.000 0.988
#> GSM614395 5 0.2802 0.7060 0.016 0.000 0.100 0.008 0.876
#> GSM614396 5 0.0451 0.7781 0.004 0.000 0.008 0.000 0.988
#> GSM614397 5 0.2110 0.7397 0.016 0.000 0.072 0.000 0.912
#> GSM614398 5 0.2046 0.7427 0.016 0.000 0.068 0.000 0.916
#> GSM614399 1 0.7905 0.1599 0.360 0.360 0.220 0.016 0.044
#> GSM614400 2 0.7905 -0.2336 0.360 0.360 0.220 0.016 0.044
#> GSM614401 2 0.7905 -0.2336 0.360 0.360 0.220 0.016 0.044
#> GSM614402 2 0.7905 -0.2336 0.360 0.360 0.220 0.016 0.044
#> GSM614403 3 0.7736 -0.0658 0.296 0.264 0.396 0.016 0.028
#> GSM614404 1 0.7905 0.1599 0.360 0.360 0.220 0.016 0.044
#> GSM614405 1 0.7857 0.1891 0.356 0.316 0.280 0.016 0.032
#> GSM614406 3 0.7140 0.5690 0.200 0.124 0.580 0.088 0.008
#> GSM614407 1 0.6184 0.2777 0.652 0.060 0.056 0.012 0.220
#> GSM614408 1 0.6184 0.2777 0.652 0.060 0.056 0.012 0.220
#> GSM614409 1 0.6184 0.2777 0.652 0.060 0.056 0.012 0.220
#> GSM614410 1 0.6184 0.2777 0.652 0.060 0.056 0.012 0.220
#> GSM614411 1 0.6184 0.2777 0.652 0.060 0.056 0.012 0.220
#> GSM614412 1 0.6184 0.2777 0.652 0.060 0.056 0.012 0.220
#> GSM614413 1 0.6671 0.1250 0.588 0.020 0.176 0.012 0.204
#> GSM614414 1 0.6671 0.1250 0.588 0.020 0.176 0.012 0.204
#> GSM614445 3 0.4444 0.7059 0.052 0.172 0.764 0.012 0.000
#> GSM614446 3 0.4230 0.7208 0.044 0.164 0.780 0.012 0.000
#> GSM614447 3 0.4407 0.7102 0.052 0.168 0.768 0.012 0.000
#> GSM614448 3 0.4490 0.7710 0.020 0.088 0.808 0.060 0.024
#> GSM614449 3 0.3941 0.7546 0.024 0.132 0.816 0.024 0.004
#> GSM614450 3 0.3962 0.7333 0.036 0.152 0.800 0.012 0.000
#> GSM614451 3 0.5475 0.5131 0.012 0.020 0.636 0.304 0.028
#> GSM614452 3 0.5475 0.5131 0.012 0.020 0.636 0.304 0.028
#> GSM614453 2 0.3448 0.7281 0.052 0.860 0.032 0.056 0.000
#> GSM614454 2 0.3448 0.7281 0.052 0.860 0.032 0.056 0.000
#> GSM614455 2 0.3448 0.7281 0.052 0.860 0.032 0.056 0.000
#> GSM614456 2 0.3448 0.7281 0.052 0.860 0.032 0.056 0.000
#> GSM614457 2 0.3448 0.7281 0.052 0.860 0.032 0.056 0.000
#> GSM614458 2 0.3448 0.7281 0.052 0.860 0.032 0.056 0.000
#> GSM614459 2 0.3448 0.7281 0.052 0.860 0.032 0.056 0.000
#> GSM614460 2 0.3448 0.7281 0.052 0.860 0.032 0.056 0.000
#> GSM614461 2 0.4231 0.7171 0.100 0.796 0.096 0.004 0.004
#> GSM614462 2 0.4231 0.7171 0.100 0.796 0.096 0.004 0.004
#> GSM614463 2 0.4231 0.7171 0.100 0.796 0.096 0.004 0.004
#> GSM614464 2 0.4231 0.7171 0.100 0.796 0.096 0.004 0.004
#> GSM614465 2 0.4231 0.7171 0.100 0.796 0.096 0.004 0.004
#> GSM614466 2 0.4231 0.7171 0.100 0.796 0.096 0.004 0.004
#> GSM614467 2 0.4231 0.7171 0.100 0.796 0.096 0.004 0.004
#> GSM614468 2 0.4231 0.7171 0.100 0.796 0.096 0.004 0.004
#> GSM614469 1 0.8222 0.4804 0.380 0.352 0.124 0.016 0.128
#> GSM614470 1 0.8222 0.4804 0.380 0.352 0.124 0.016 0.128
#> GSM614471 1 0.8222 0.4804 0.380 0.352 0.124 0.016 0.128
#> GSM614472 1 0.8222 0.4804 0.380 0.352 0.124 0.016 0.128
#> GSM614473 1 0.8222 0.4804 0.380 0.352 0.124 0.016 0.128
#> GSM614474 1 0.8222 0.4804 0.380 0.352 0.124 0.016 0.128
#> GSM614475 1 0.8221 0.4744 0.380 0.352 0.132 0.016 0.120
#> GSM614476 1 0.8259 0.4636 0.376 0.348 0.148 0.016 0.112
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM614415 5 0.709 0.48480 0.248 0.004 0.008 0.040 0.360 0.340
#> GSM614416 5 0.709 0.48480 0.248 0.004 0.008 0.040 0.360 0.340
#> GSM614417 5 0.709 0.48480 0.248 0.004 0.008 0.040 0.360 0.340
#> GSM614418 5 0.709 0.48480 0.248 0.004 0.008 0.040 0.360 0.340
#> GSM614419 5 0.717 0.49930 0.248 0.000 0.020 0.040 0.372 0.320
#> GSM614420 5 0.717 0.49930 0.248 0.000 0.020 0.040 0.372 0.320
#> GSM614421 3 0.252 0.82446 0.000 0.012 0.900 0.028 0.044 0.016
#> GSM614422 3 0.252 0.82446 0.000 0.012 0.900 0.028 0.044 0.016
#> GSM614423 3 0.342 0.79205 0.016 0.052 0.856 0.004 0.044 0.028
#> GSM614424 3 0.252 0.82446 0.000 0.012 0.900 0.028 0.044 0.016
#> GSM614425 3 0.252 0.82446 0.000 0.012 0.900 0.028 0.044 0.016
#> GSM614426 3 0.252 0.82446 0.000 0.012 0.900 0.028 0.044 0.016
#> GSM614427 3 0.252 0.82446 0.000 0.012 0.900 0.028 0.044 0.016
#> GSM614428 3 0.259 0.82225 0.000 0.012 0.896 0.028 0.048 0.016
#> GSM614429 2 0.130 0.77991 0.000 0.952 0.032 0.004 0.000 0.012
#> GSM614430 2 0.130 0.77991 0.000 0.952 0.032 0.004 0.000 0.012
#> GSM614431 2 0.130 0.77991 0.000 0.952 0.032 0.004 0.000 0.012
#> GSM614432 2 0.130 0.77991 0.000 0.952 0.032 0.004 0.000 0.012
#> GSM614433 2 0.130 0.77991 0.000 0.952 0.032 0.004 0.000 0.012
#> GSM614434 2 0.130 0.77991 0.000 0.952 0.032 0.004 0.000 0.012
#> GSM614435 2 0.130 0.77991 0.000 0.952 0.032 0.004 0.000 0.012
#> GSM614436 2 0.166 0.77436 0.000 0.936 0.040 0.012 0.000 0.012
#> GSM614437 4 0.279 0.98919 0.008 0.056 0.056 0.876 0.000 0.004
#> GSM614438 4 0.245 0.99641 0.000 0.060 0.056 0.884 0.000 0.000
#> GSM614439 4 0.245 0.99641 0.000 0.060 0.056 0.884 0.000 0.000
#> GSM614440 4 0.245 0.99641 0.000 0.060 0.056 0.884 0.000 0.000
#> GSM614441 4 0.245 0.99641 0.000 0.060 0.056 0.884 0.000 0.000
#> GSM614442 4 0.245 0.99641 0.000 0.060 0.056 0.884 0.000 0.000
#> GSM614443 4 0.279 0.98919 0.008 0.056 0.056 0.876 0.000 0.004
#> GSM614444 4 0.245 0.99641 0.000 0.060 0.056 0.884 0.000 0.000
#> GSM614391 5 0.146 0.69219 0.000 0.000 0.008 0.000 0.936 0.056
#> GSM614392 5 0.146 0.69219 0.000 0.000 0.008 0.000 0.936 0.056
#> GSM614393 5 0.146 0.69219 0.000 0.000 0.008 0.000 0.936 0.056
#> GSM614394 5 0.146 0.69219 0.000 0.000 0.008 0.000 0.936 0.056
#> GSM614395 5 0.354 0.64653 0.060 0.000 0.052 0.020 0.844 0.024
#> GSM614396 5 0.146 0.69219 0.000 0.000 0.008 0.000 0.936 0.056
#> GSM614397 5 0.285 0.66647 0.040 0.000 0.032 0.012 0.884 0.032
#> GSM614398 5 0.275 0.66852 0.040 0.000 0.032 0.008 0.888 0.032
#> GSM614399 1 0.727 0.93636 0.448 0.236 0.092 0.000 0.012 0.212
#> GSM614400 1 0.727 0.93636 0.448 0.236 0.092 0.000 0.012 0.212
#> GSM614401 1 0.727 0.93636 0.448 0.236 0.092 0.000 0.012 0.212
#> GSM614402 1 0.727 0.93636 0.448 0.236 0.092 0.000 0.012 0.212
#> GSM614403 1 0.736 0.73532 0.456 0.164 0.200 0.000 0.008 0.172
#> GSM614404 1 0.727 0.93636 0.448 0.236 0.092 0.000 0.012 0.212
#> GSM614405 1 0.734 0.89106 0.448 0.212 0.128 0.000 0.008 0.204
#> GSM614406 3 0.683 0.09626 0.340 0.088 0.488 0.016 0.024 0.044
#> GSM614407 6 0.323 0.47775 0.000 0.056 0.036 0.000 0.056 0.852
#> GSM614408 6 0.323 0.47775 0.000 0.056 0.036 0.000 0.056 0.852
#> GSM614409 6 0.323 0.47775 0.000 0.056 0.036 0.000 0.056 0.852
#> GSM614410 6 0.323 0.47775 0.000 0.056 0.036 0.000 0.056 0.852
#> GSM614411 6 0.323 0.47775 0.000 0.056 0.036 0.000 0.056 0.852
#> GSM614412 6 0.323 0.47775 0.000 0.056 0.036 0.000 0.056 0.852
#> GSM614413 6 0.508 0.27753 0.008 0.008 0.188 0.024 0.068 0.704
#> GSM614414 6 0.508 0.27753 0.008 0.008 0.188 0.024 0.068 0.704
#> GSM614445 3 0.509 0.72491 0.148 0.072 0.728 0.016 0.012 0.024
#> GSM614446 3 0.478 0.75079 0.140 0.060 0.752 0.016 0.012 0.020
#> GSM614447 3 0.504 0.73004 0.148 0.068 0.732 0.016 0.012 0.024
#> GSM614448 3 0.341 0.79458 0.116 0.008 0.836 0.016 0.012 0.012
#> GSM614449 3 0.389 0.78449 0.128 0.024 0.808 0.016 0.012 0.012
#> GSM614450 3 0.438 0.76867 0.132 0.044 0.780 0.016 0.012 0.016
#> GSM614451 3 0.454 0.73530 0.100 0.004 0.764 0.104 0.016 0.012
#> GSM614452 3 0.454 0.73530 0.100 0.004 0.764 0.104 0.016 0.012
#> GSM614453 2 0.444 0.73241 0.084 0.784 0.004 0.084 0.028 0.016
#> GSM614454 2 0.444 0.73241 0.084 0.784 0.004 0.084 0.028 0.016
#> GSM614455 2 0.444 0.73241 0.084 0.784 0.004 0.084 0.028 0.016
#> GSM614456 2 0.444 0.73241 0.084 0.784 0.004 0.084 0.028 0.016
#> GSM614457 2 0.444 0.73241 0.084 0.784 0.004 0.084 0.028 0.016
#> GSM614458 2 0.444 0.73241 0.084 0.784 0.004 0.084 0.028 0.016
#> GSM614459 2 0.444 0.73241 0.084 0.784 0.004 0.084 0.028 0.016
#> GSM614460 2 0.444 0.73241 0.084 0.784 0.004 0.084 0.028 0.016
#> GSM614461 2 0.435 0.68372 0.200 0.736 0.024 0.004 0.000 0.036
#> GSM614462 2 0.435 0.68372 0.200 0.736 0.024 0.004 0.000 0.036
#> GSM614463 2 0.435 0.68372 0.200 0.736 0.024 0.004 0.000 0.036
#> GSM614464 2 0.435 0.68372 0.200 0.736 0.024 0.004 0.000 0.036
#> GSM614465 2 0.435 0.68372 0.200 0.736 0.024 0.004 0.000 0.036
#> GSM614466 2 0.435 0.68372 0.200 0.736 0.024 0.004 0.000 0.036
#> GSM614467 2 0.435 0.68372 0.200 0.736 0.024 0.004 0.000 0.036
#> GSM614468 2 0.435 0.68372 0.200 0.736 0.024 0.004 0.000 0.036
#> GSM614469 6 0.810 0.00453 0.220 0.280 0.068 0.016 0.048 0.368
#> GSM614470 6 0.810 0.00453 0.220 0.280 0.068 0.016 0.048 0.368
#> GSM614471 6 0.810 0.00453 0.220 0.280 0.068 0.016 0.048 0.368
#> GSM614472 6 0.810 0.00453 0.220 0.280 0.068 0.016 0.048 0.368
#> GSM614473 6 0.810 0.00453 0.220 0.280 0.068 0.016 0.048 0.368
#> GSM614474 6 0.810 0.00453 0.220 0.280 0.068 0.016 0.048 0.368
#> GSM614475 6 0.810 0.00453 0.220 0.280 0.068 0.016 0.048 0.368
#> GSM614476 6 0.824 -0.04818 0.220 0.276 0.084 0.016 0.048 0.356
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 individual(p) protocol(p) time(p) other(p) k
#> SD:kmeans 53 2.92e-08 1.000 1 0.2218 2
#> SD:kmeans 79 1.03e-20 0.366 1 0.1083 3
#> SD:kmeans 59 2.74e-25 0.882 1 0.0292 4
#> SD:kmeans 63 1.38e-27 0.797 1 0.0578 5
#> SD:kmeans 63 6.86e-38 0.983 1 0.0371 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 21168 rows and 86 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 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 0.402 0.495 0.816 0.5017 0.498 0.498
#> 3 3 0.687 0.854 0.922 0.3298 0.748 0.533
#> 4 4 0.730 0.785 0.862 0.1215 0.842 0.568
#> 5 5 0.787 0.841 0.878 0.0614 0.945 0.784
#> 6 6 0.810 0.826 0.819 0.0371 0.969 0.854
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
#> GSM614415 1 0.118 0.71188 0.984 0.016
#> GSM614416 1 0.118 0.71188 0.984 0.016
#> GSM614417 1 0.118 0.71188 0.984 0.016
#> GSM614418 1 0.118 0.71188 0.984 0.016
#> GSM614419 1 0.000 0.71396 1.000 0.000
#> GSM614420 1 0.000 0.71396 1.000 0.000
#> GSM614421 1 0.978 0.16637 0.588 0.412
#> GSM614422 1 0.795 0.45993 0.760 0.240
#> GSM614423 2 0.909 0.38721 0.324 0.676
#> GSM614424 1 0.978 0.16637 0.588 0.412
#> GSM614425 1 0.975 0.17402 0.592 0.408
#> GSM614426 1 0.975 0.17402 0.592 0.408
#> GSM614427 1 0.980 0.15772 0.584 0.416
#> GSM614428 1 0.980 0.15772 0.584 0.416
#> GSM614429 2 0.000 0.72906 0.000 1.000
#> GSM614430 2 0.000 0.72906 0.000 1.000
#> GSM614431 2 0.000 0.72906 0.000 1.000
#> GSM614432 2 0.000 0.72906 0.000 1.000
#> GSM614433 2 0.000 0.72906 0.000 1.000
#> GSM614434 2 0.000 0.72906 0.000 1.000
#> GSM614435 2 0.000 0.72906 0.000 1.000
#> GSM614436 2 0.163 0.71486 0.024 0.976
#> GSM614437 2 0.706 0.57069 0.192 0.808
#> GSM614438 2 0.992 0.18435 0.448 0.552
#> GSM614439 2 0.992 0.18435 0.448 0.552
#> GSM614440 2 0.992 0.18435 0.448 0.552
#> GSM614441 2 0.992 0.18435 0.448 0.552
#> GSM614442 2 0.992 0.18435 0.448 0.552
#> GSM614443 2 0.929 0.36037 0.344 0.656
#> GSM614444 2 0.992 0.18435 0.448 0.552
#> GSM614391 1 0.000 0.71396 1.000 0.000
#> GSM614392 1 0.000 0.71396 1.000 0.000
#> GSM614393 1 0.000 0.71396 1.000 0.000
#> GSM614394 1 0.000 0.71396 1.000 0.000
#> GSM614395 1 0.000 0.71396 1.000 0.000
#> GSM614396 1 0.000 0.71396 1.000 0.000
#> GSM614397 1 0.000 0.71396 1.000 0.000
#> GSM614398 1 0.000 0.71396 1.000 0.000
#> GSM614399 2 0.978 0.15238 0.412 0.588
#> GSM614400 2 0.980 0.14397 0.416 0.584
#> GSM614401 2 0.980 0.14397 0.416 0.584
#> GSM614402 2 0.980 0.14397 0.416 0.584
#> GSM614403 2 0.971 0.17621 0.400 0.600
#> GSM614404 2 0.980 0.14397 0.416 0.584
#> GSM614405 2 0.985 0.13184 0.428 0.572
#> GSM614406 2 0.992 0.18435 0.448 0.552
#> GSM614407 1 0.416 0.67630 0.916 0.084
#> GSM614408 1 0.416 0.67630 0.916 0.084
#> GSM614409 1 0.311 0.69427 0.944 0.056
#> GSM614410 1 0.416 0.67630 0.916 0.084
#> GSM614411 1 0.343 0.68973 0.936 0.064
#> GSM614412 1 0.000 0.71396 1.000 0.000
#> GSM614413 1 0.000 0.71396 1.000 0.000
#> GSM614414 1 0.000 0.71396 1.000 0.000
#> GSM614445 2 0.443 0.66513 0.092 0.908
#> GSM614446 2 0.443 0.66513 0.092 0.908
#> GSM614447 2 0.443 0.66513 0.092 0.908
#> GSM614448 1 0.995 -0.00694 0.540 0.460
#> GSM614449 1 0.999 -0.05718 0.520 0.480
#> GSM614450 2 0.932 0.39498 0.348 0.652
#> GSM614451 2 0.996 0.14853 0.464 0.536
#> GSM614452 2 0.996 0.14853 0.464 0.536
#> GSM614453 2 0.000 0.72906 0.000 1.000
#> GSM614454 2 0.000 0.72906 0.000 1.000
#> GSM614455 2 0.000 0.72906 0.000 1.000
#> GSM614456 2 0.000 0.72906 0.000 1.000
#> GSM614457 2 0.000 0.72906 0.000 1.000
#> GSM614458 2 0.000 0.72906 0.000 1.000
#> GSM614459 2 0.000 0.72906 0.000 1.000
#> GSM614460 2 0.000 0.72906 0.000 1.000
#> GSM614461 2 0.000 0.72906 0.000 1.000
#> GSM614462 2 0.000 0.72906 0.000 1.000
#> GSM614463 2 0.000 0.72906 0.000 1.000
#> GSM614464 2 0.000 0.72906 0.000 1.000
#> GSM614465 2 0.000 0.72906 0.000 1.000
#> GSM614466 2 0.000 0.72906 0.000 1.000
#> GSM614467 2 0.000 0.72906 0.000 1.000
#> GSM614468 2 0.000 0.72906 0.000 1.000
#> GSM614469 1 0.993 0.16940 0.548 0.452
#> GSM614470 1 0.993 0.16940 0.548 0.452
#> GSM614471 1 0.993 0.16940 0.548 0.452
#> GSM614472 1 0.993 0.16940 0.548 0.452
#> GSM614473 1 0.993 0.16940 0.548 0.452
#> GSM614474 1 0.993 0.16940 0.548 0.452
#> GSM614475 1 0.993 0.16940 0.548 0.452
#> GSM614476 1 0.416 0.66861 0.916 0.084
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM614415 1 0.0000 0.927 1.000 0.000 0.000
#> GSM614416 1 0.0000 0.927 1.000 0.000 0.000
#> GSM614417 1 0.0000 0.927 1.000 0.000 0.000
#> GSM614418 1 0.0000 0.927 1.000 0.000 0.000
#> GSM614419 1 0.0000 0.927 1.000 0.000 0.000
#> GSM614420 1 0.0000 0.927 1.000 0.000 0.000
#> GSM614421 3 0.0237 0.882 0.004 0.000 0.996
#> GSM614422 3 0.0237 0.882 0.004 0.000 0.996
#> GSM614423 3 0.2116 0.860 0.012 0.040 0.948
#> GSM614424 3 0.0237 0.882 0.004 0.000 0.996
#> GSM614425 3 0.0237 0.882 0.004 0.000 0.996
#> GSM614426 3 0.0237 0.882 0.004 0.000 0.996
#> GSM614427 3 0.0237 0.882 0.004 0.000 0.996
#> GSM614428 3 0.0237 0.882 0.004 0.000 0.996
#> GSM614429 2 0.0237 0.922 0.000 0.996 0.004
#> GSM614430 2 0.0237 0.922 0.000 0.996 0.004
#> GSM614431 2 0.0237 0.922 0.000 0.996 0.004
#> GSM614432 2 0.0237 0.922 0.000 0.996 0.004
#> GSM614433 2 0.0237 0.922 0.000 0.996 0.004
#> GSM614434 2 0.0237 0.922 0.000 0.996 0.004
#> GSM614435 2 0.0237 0.922 0.000 0.996 0.004
#> GSM614436 3 0.6244 0.382 0.000 0.440 0.560
#> GSM614437 3 0.3941 0.835 0.000 0.156 0.844
#> GSM614438 3 0.3879 0.838 0.000 0.152 0.848
#> GSM614439 3 0.3879 0.838 0.000 0.152 0.848
#> GSM614440 3 0.3879 0.838 0.000 0.152 0.848
#> GSM614441 3 0.3879 0.838 0.000 0.152 0.848
#> GSM614442 3 0.3879 0.838 0.000 0.152 0.848
#> GSM614443 3 0.3879 0.838 0.000 0.152 0.848
#> GSM614444 3 0.3879 0.838 0.000 0.152 0.848
#> GSM614391 1 0.0000 0.927 1.000 0.000 0.000
#> GSM614392 1 0.0000 0.927 1.000 0.000 0.000
#> GSM614393 1 0.0000 0.927 1.000 0.000 0.000
#> GSM614394 1 0.0000 0.927 1.000 0.000 0.000
#> GSM614395 3 0.5810 0.473 0.336 0.000 0.664
#> GSM614396 1 0.0000 0.927 1.000 0.000 0.000
#> GSM614397 1 0.4452 0.737 0.808 0.000 0.192
#> GSM614398 1 0.1643 0.900 0.956 0.000 0.044
#> GSM614399 2 0.5393 0.810 0.044 0.808 0.148
#> GSM614400 2 0.5598 0.805 0.052 0.800 0.148
#> GSM614401 2 0.5598 0.805 0.052 0.800 0.148
#> GSM614402 2 0.5497 0.808 0.048 0.804 0.148
#> GSM614403 2 0.7175 0.480 0.032 0.592 0.376
#> GSM614404 2 0.5598 0.805 0.052 0.800 0.148
#> GSM614405 3 0.7250 0.132 0.032 0.396 0.572
#> GSM614406 3 0.0000 0.882 0.000 0.000 1.000
#> GSM614407 1 0.0000 0.927 1.000 0.000 0.000
#> GSM614408 1 0.0000 0.927 1.000 0.000 0.000
#> GSM614409 1 0.0000 0.927 1.000 0.000 0.000
#> GSM614410 1 0.0000 0.927 1.000 0.000 0.000
#> GSM614411 1 0.0000 0.927 1.000 0.000 0.000
#> GSM614412 1 0.0000 0.927 1.000 0.000 0.000
#> GSM614413 1 0.2165 0.884 0.936 0.000 0.064
#> GSM614414 1 0.1860 0.894 0.948 0.000 0.052
#> GSM614445 2 0.4931 0.759 0.000 0.768 0.232
#> GSM614446 2 0.5560 0.664 0.000 0.700 0.300
#> GSM614447 2 0.5098 0.740 0.000 0.752 0.248
#> GSM614448 3 0.0237 0.881 0.000 0.004 0.996
#> GSM614449 3 0.0237 0.881 0.000 0.004 0.996
#> GSM614450 3 0.1643 0.862 0.000 0.044 0.956
#> GSM614451 3 0.0000 0.882 0.000 0.000 1.000
#> GSM614452 3 0.0000 0.882 0.000 0.000 1.000
#> GSM614453 2 0.0237 0.922 0.000 0.996 0.004
#> GSM614454 2 0.0237 0.922 0.000 0.996 0.004
#> GSM614455 2 0.0237 0.922 0.000 0.996 0.004
#> GSM614456 2 0.0237 0.922 0.000 0.996 0.004
#> GSM614457 2 0.0237 0.922 0.000 0.996 0.004
#> GSM614458 2 0.0237 0.922 0.000 0.996 0.004
#> GSM614459 2 0.0237 0.922 0.000 0.996 0.004
#> GSM614460 2 0.0237 0.922 0.000 0.996 0.004
#> GSM614461 2 0.0592 0.920 0.000 0.988 0.012
#> GSM614462 2 0.0592 0.920 0.000 0.988 0.012
#> GSM614463 2 0.0592 0.920 0.000 0.988 0.012
#> GSM614464 2 0.0592 0.920 0.000 0.988 0.012
#> GSM614465 2 0.0592 0.920 0.000 0.988 0.012
#> GSM614466 2 0.0592 0.920 0.000 0.988 0.012
#> GSM614467 2 0.0592 0.920 0.000 0.988 0.012
#> GSM614468 2 0.0592 0.920 0.000 0.988 0.012
#> GSM614469 1 0.4291 0.816 0.820 0.180 0.000
#> GSM614470 1 0.4291 0.816 0.820 0.180 0.000
#> GSM614471 1 0.4291 0.816 0.820 0.180 0.000
#> GSM614472 1 0.4291 0.816 0.820 0.180 0.000
#> GSM614473 1 0.4291 0.816 0.820 0.180 0.000
#> GSM614474 1 0.4291 0.816 0.820 0.180 0.000
#> GSM614475 1 0.4291 0.816 0.820 0.180 0.000
#> GSM614476 1 0.5167 0.755 0.792 0.016 0.192
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM614415 1 0.0000 0.9411 1.000 0.000 0.000 0.000
#> GSM614416 1 0.0000 0.9411 1.000 0.000 0.000 0.000
#> GSM614417 1 0.0000 0.9411 1.000 0.000 0.000 0.000
#> GSM614418 1 0.0000 0.9411 1.000 0.000 0.000 0.000
#> GSM614419 1 0.0000 0.9411 1.000 0.000 0.000 0.000
#> GSM614420 1 0.0000 0.9411 1.000 0.000 0.000 0.000
#> GSM614421 3 0.2408 0.7901 0.000 0.000 0.896 0.104
#> GSM614422 3 0.2408 0.7901 0.000 0.000 0.896 0.104
#> GSM614423 3 0.5163 0.1525 0.000 0.004 0.516 0.480
#> GSM614424 3 0.2408 0.7901 0.000 0.000 0.896 0.104
#> GSM614425 3 0.2408 0.7901 0.000 0.000 0.896 0.104
#> GSM614426 3 0.2408 0.7901 0.000 0.000 0.896 0.104
#> GSM614427 3 0.2149 0.7930 0.000 0.000 0.912 0.088
#> GSM614428 3 0.1940 0.7940 0.000 0.000 0.924 0.076
#> GSM614429 2 0.1022 0.8923 0.000 0.968 0.000 0.032
#> GSM614430 2 0.1022 0.8923 0.000 0.968 0.000 0.032
#> GSM614431 2 0.1389 0.8915 0.000 0.952 0.000 0.048
#> GSM614432 2 0.1389 0.8915 0.000 0.952 0.000 0.048
#> GSM614433 2 0.1557 0.8891 0.000 0.944 0.000 0.056
#> GSM614434 2 0.1389 0.8915 0.000 0.952 0.000 0.048
#> GSM614435 2 0.0817 0.8913 0.000 0.976 0.000 0.024
#> GSM614436 2 0.3156 0.8112 0.000 0.884 0.068 0.048
#> GSM614437 3 0.4881 0.7304 0.000 0.196 0.756 0.048
#> GSM614438 3 0.4544 0.7550 0.000 0.164 0.788 0.048
#> GSM614439 3 0.4544 0.7550 0.000 0.164 0.788 0.048
#> GSM614440 3 0.4544 0.7550 0.000 0.164 0.788 0.048
#> GSM614441 3 0.4544 0.7550 0.000 0.164 0.788 0.048
#> GSM614442 3 0.4544 0.7550 0.000 0.164 0.788 0.048
#> GSM614443 3 0.4677 0.7464 0.000 0.176 0.776 0.048
#> GSM614444 3 0.4544 0.7550 0.000 0.164 0.788 0.048
#> GSM614391 1 0.0188 0.9408 0.996 0.000 0.000 0.004
#> GSM614392 1 0.0188 0.9408 0.996 0.000 0.000 0.004
#> GSM614393 1 0.0188 0.9408 0.996 0.000 0.000 0.004
#> GSM614394 1 0.0188 0.9408 0.996 0.000 0.000 0.004
#> GSM614395 1 0.5295 0.0326 0.504 0.000 0.488 0.008
#> GSM614396 1 0.0188 0.9408 0.996 0.000 0.000 0.004
#> GSM614397 1 0.2611 0.8524 0.896 0.000 0.096 0.008
#> GSM614398 1 0.1305 0.9185 0.960 0.000 0.036 0.004
#> GSM614399 4 0.2521 0.7334 0.024 0.064 0.000 0.912
#> GSM614400 4 0.2521 0.7334 0.024 0.064 0.000 0.912
#> GSM614401 4 0.2521 0.7334 0.024 0.064 0.000 0.912
#> GSM614402 4 0.2521 0.7334 0.024 0.064 0.000 0.912
#> GSM614403 4 0.2730 0.6579 0.000 0.016 0.088 0.896
#> GSM614404 4 0.2521 0.7334 0.024 0.064 0.000 0.912
#> GSM614405 4 0.1913 0.6848 0.000 0.020 0.040 0.940
#> GSM614406 3 0.4137 0.7435 0.000 0.012 0.780 0.208
#> GSM614407 1 0.1211 0.9232 0.960 0.000 0.000 0.040
#> GSM614408 1 0.1211 0.9232 0.960 0.000 0.000 0.040
#> GSM614409 1 0.1118 0.9261 0.964 0.000 0.000 0.036
#> GSM614410 1 0.1211 0.9232 0.960 0.000 0.000 0.040
#> GSM614411 1 0.1118 0.9261 0.964 0.000 0.000 0.036
#> GSM614412 1 0.1118 0.9261 0.964 0.000 0.000 0.036
#> GSM614413 1 0.1970 0.9007 0.932 0.000 0.060 0.008
#> GSM614414 1 0.1722 0.9105 0.944 0.000 0.048 0.008
#> GSM614445 4 0.7148 0.3739 0.000 0.220 0.220 0.560
#> GSM614446 4 0.7048 0.3048 0.000 0.160 0.284 0.556
#> GSM614447 4 0.7067 0.3629 0.000 0.188 0.244 0.568
#> GSM614448 3 0.3764 0.7109 0.000 0.000 0.784 0.216
#> GSM614449 3 0.3873 0.6995 0.000 0.000 0.772 0.228
#> GSM614450 3 0.5132 0.3008 0.000 0.004 0.548 0.448
#> GSM614451 3 0.1022 0.7938 0.000 0.000 0.968 0.032
#> GSM614452 3 0.0921 0.7944 0.000 0.000 0.972 0.028
#> GSM614453 2 0.0817 0.8843 0.000 0.976 0.024 0.000
#> GSM614454 2 0.0817 0.8843 0.000 0.976 0.024 0.000
#> GSM614455 2 0.0817 0.8843 0.000 0.976 0.024 0.000
#> GSM614456 2 0.0817 0.8843 0.000 0.976 0.024 0.000
#> GSM614457 2 0.0817 0.8843 0.000 0.976 0.024 0.000
#> GSM614458 2 0.0817 0.8843 0.000 0.976 0.024 0.000
#> GSM614459 2 0.0817 0.8843 0.000 0.976 0.024 0.000
#> GSM614460 2 0.0817 0.8843 0.000 0.976 0.024 0.000
#> GSM614461 2 0.3870 0.8151 0.000 0.788 0.004 0.208
#> GSM614462 2 0.3870 0.8151 0.000 0.788 0.004 0.208
#> GSM614463 2 0.3870 0.8151 0.000 0.788 0.004 0.208
#> GSM614464 2 0.3870 0.8151 0.000 0.788 0.004 0.208
#> GSM614465 2 0.3870 0.8151 0.000 0.788 0.004 0.208
#> GSM614466 2 0.3870 0.8151 0.000 0.788 0.004 0.208
#> GSM614467 2 0.3870 0.8151 0.000 0.788 0.004 0.208
#> GSM614468 2 0.3870 0.8151 0.000 0.788 0.004 0.208
#> GSM614469 4 0.5334 0.6958 0.284 0.036 0.000 0.680
#> GSM614470 4 0.5334 0.6958 0.284 0.036 0.000 0.680
#> GSM614471 4 0.5334 0.6958 0.284 0.036 0.000 0.680
#> GSM614472 4 0.5334 0.6958 0.284 0.036 0.000 0.680
#> GSM614473 4 0.5334 0.6958 0.284 0.036 0.000 0.680
#> GSM614474 4 0.5334 0.6958 0.284 0.036 0.000 0.680
#> GSM614475 4 0.5334 0.6958 0.284 0.036 0.000 0.680
#> GSM614476 4 0.5723 0.6724 0.268 0.004 0.052 0.676
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM614415 5 0.0404 0.895 0.012 0.000 0.000 0.000 0.988
#> GSM614416 5 0.0404 0.895 0.012 0.000 0.000 0.000 0.988
#> GSM614417 5 0.0404 0.895 0.012 0.000 0.000 0.000 0.988
#> GSM614418 5 0.0404 0.895 0.012 0.000 0.000 0.000 0.988
#> GSM614419 5 0.0404 0.895 0.012 0.000 0.000 0.000 0.988
#> GSM614420 5 0.0404 0.895 0.012 0.000 0.000 0.000 0.988
#> GSM614421 3 0.2690 0.849 0.000 0.000 0.844 0.156 0.000
#> GSM614422 3 0.2690 0.849 0.000 0.000 0.844 0.156 0.000
#> GSM614423 3 0.3043 0.814 0.080 0.000 0.864 0.056 0.000
#> GSM614424 3 0.2690 0.849 0.000 0.000 0.844 0.156 0.000
#> GSM614425 3 0.2690 0.849 0.000 0.000 0.844 0.156 0.000
#> GSM614426 3 0.2690 0.849 0.000 0.000 0.844 0.156 0.000
#> GSM614427 3 0.2690 0.849 0.000 0.000 0.844 0.156 0.000
#> GSM614428 3 0.2773 0.842 0.000 0.000 0.836 0.164 0.000
#> GSM614429 2 0.0609 0.871 0.000 0.980 0.000 0.020 0.000
#> GSM614430 2 0.0609 0.871 0.000 0.980 0.000 0.020 0.000
#> GSM614431 2 0.0609 0.871 0.000 0.980 0.000 0.020 0.000
#> GSM614432 2 0.0609 0.871 0.000 0.980 0.000 0.020 0.000
#> GSM614433 2 0.0609 0.871 0.000 0.980 0.000 0.020 0.000
#> GSM614434 2 0.0609 0.871 0.000 0.980 0.000 0.020 0.000
#> GSM614435 2 0.0703 0.870 0.000 0.976 0.000 0.024 0.000
#> GSM614436 2 0.3010 0.761 0.000 0.824 0.004 0.172 0.000
#> GSM614437 4 0.1485 0.939 0.000 0.020 0.032 0.948 0.000
#> GSM614438 4 0.1522 0.957 0.000 0.012 0.044 0.944 0.000
#> GSM614439 4 0.1522 0.957 0.000 0.012 0.044 0.944 0.000
#> GSM614440 4 0.1522 0.957 0.000 0.012 0.044 0.944 0.000
#> GSM614441 4 0.1522 0.957 0.000 0.012 0.044 0.944 0.000
#> GSM614442 4 0.1522 0.957 0.000 0.012 0.044 0.944 0.000
#> GSM614443 4 0.1281 0.948 0.000 0.012 0.032 0.956 0.000
#> GSM614444 4 0.1522 0.957 0.000 0.012 0.044 0.944 0.000
#> GSM614391 5 0.0451 0.894 0.004 0.000 0.008 0.000 0.988
#> GSM614392 5 0.0451 0.894 0.004 0.000 0.008 0.000 0.988
#> GSM614393 5 0.0451 0.894 0.004 0.000 0.008 0.000 0.988
#> GSM614394 5 0.0451 0.894 0.004 0.000 0.008 0.000 0.988
#> GSM614395 5 0.4302 0.629 0.000 0.000 0.048 0.208 0.744
#> GSM614396 5 0.0566 0.892 0.004 0.000 0.012 0.000 0.984
#> GSM614397 5 0.1195 0.878 0.000 0.000 0.028 0.012 0.960
#> GSM614398 5 0.1026 0.885 0.004 0.000 0.024 0.004 0.968
#> GSM614399 1 0.3966 0.832 0.808 0.048 0.132 0.012 0.000
#> GSM614400 1 0.3966 0.832 0.808 0.048 0.132 0.012 0.000
#> GSM614401 1 0.3895 0.833 0.812 0.044 0.132 0.012 0.000
#> GSM614402 1 0.3966 0.832 0.808 0.048 0.132 0.012 0.000
#> GSM614403 1 0.4486 0.711 0.712 0.020 0.256 0.012 0.000
#> GSM614404 1 0.3966 0.832 0.808 0.048 0.132 0.012 0.000
#> GSM614405 1 0.4443 0.803 0.776 0.020 0.152 0.052 0.000
#> GSM614406 4 0.5251 0.655 0.128 0.020 0.132 0.720 0.000
#> GSM614407 5 0.4177 0.804 0.200 0.000 0.004 0.036 0.760
#> GSM614408 5 0.4177 0.804 0.200 0.000 0.004 0.036 0.760
#> GSM614409 5 0.4177 0.804 0.200 0.000 0.004 0.036 0.760
#> GSM614410 5 0.4177 0.804 0.200 0.000 0.004 0.036 0.760
#> GSM614411 5 0.4177 0.804 0.200 0.000 0.004 0.036 0.760
#> GSM614412 5 0.4177 0.804 0.200 0.000 0.004 0.036 0.760
#> GSM614413 5 0.4425 0.835 0.132 0.000 0.036 0.044 0.788
#> GSM614414 5 0.4265 0.838 0.132 0.000 0.028 0.044 0.796
#> GSM614445 3 0.2569 0.763 0.068 0.032 0.896 0.004 0.000
#> GSM614446 3 0.2284 0.776 0.056 0.028 0.912 0.004 0.000
#> GSM614447 3 0.2369 0.771 0.056 0.032 0.908 0.004 0.000
#> GSM614448 3 0.1981 0.829 0.016 0.000 0.920 0.064 0.000
#> GSM614449 3 0.1549 0.821 0.016 0.000 0.944 0.040 0.000
#> GSM614450 3 0.1682 0.800 0.044 0.004 0.940 0.012 0.000
#> GSM614451 3 0.4126 0.556 0.000 0.000 0.620 0.380 0.000
#> GSM614452 3 0.4060 0.596 0.000 0.000 0.640 0.360 0.000
#> GSM614453 2 0.3353 0.811 0.008 0.796 0.000 0.196 0.000
#> GSM614454 2 0.3353 0.811 0.008 0.796 0.000 0.196 0.000
#> GSM614455 2 0.3353 0.811 0.008 0.796 0.000 0.196 0.000
#> GSM614456 2 0.3353 0.811 0.008 0.796 0.000 0.196 0.000
#> GSM614457 2 0.3353 0.811 0.008 0.796 0.000 0.196 0.000
#> GSM614458 2 0.3353 0.811 0.008 0.796 0.000 0.196 0.000
#> GSM614459 2 0.3353 0.811 0.008 0.796 0.000 0.196 0.000
#> GSM614460 2 0.3353 0.811 0.008 0.796 0.000 0.196 0.000
#> GSM614461 2 0.2632 0.846 0.072 0.888 0.040 0.000 0.000
#> GSM614462 2 0.2632 0.846 0.072 0.888 0.040 0.000 0.000
#> GSM614463 2 0.2632 0.846 0.072 0.888 0.040 0.000 0.000
#> GSM614464 2 0.2632 0.846 0.072 0.888 0.040 0.000 0.000
#> GSM614465 2 0.2632 0.846 0.072 0.888 0.040 0.000 0.000
#> GSM614466 2 0.2632 0.846 0.072 0.888 0.040 0.000 0.000
#> GSM614467 2 0.2632 0.846 0.072 0.888 0.040 0.000 0.000
#> GSM614468 2 0.2632 0.846 0.072 0.888 0.040 0.000 0.000
#> GSM614469 1 0.1942 0.863 0.920 0.012 0.000 0.000 0.068
#> GSM614470 1 0.1942 0.863 0.920 0.012 0.000 0.000 0.068
#> GSM614471 1 0.1942 0.863 0.920 0.012 0.000 0.000 0.068
#> GSM614472 1 0.1942 0.863 0.920 0.012 0.000 0.000 0.068
#> GSM614473 1 0.1942 0.863 0.920 0.012 0.000 0.000 0.068
#> GSM614474 1 0.1942 0.863 0.920 0.012 0.000 0.000 0.068
#> GSM614475 1 0.1942 0.863 0.920 0.012 0.000 0.000 0.068
#> GSM614476 1 0.2242 0.858 0.920 0.012 0.008 0.008 0.052
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM614415 5 0.2488 0.851 0.004 0.000 0.008 0.000 0.864 0.124
#> GSM614416 5 0.2488 0.851 0.004 0.000 0.008 0.000 0.864 0.124
#> GSM614417 5 0.2488 0.851 0.004 0.000 0.008 0.000 0.864 0.124
#> GSM614418 5 0.2488 0.851 0.004 0.000 0.008 0.000 0.864 0.124
#> GSM614419 5 0.2355 0.859 0.004 0.000 0.008 0.000 0.876 0.112
#> GSM614420 5 0.2355 0.859 0.004 0.000 0.008 0.000 0.876 0.112
#> GSM614421 3 0.1524 0.863 0.000 0.000 0.932 0.060 0.008 0.000
#> GSM614422 3 0.1524 0.863 0.000 0.000 0.932 0.060 0.008 0.000
#> GSM614423 3 0.1570 0.857 0.016 0.000 0.944 0.028 0.008 0.004
#> GSM614424 3 0.1524 0.863 0.000 0.000 0.932 0.060 0.008 0.000
#> GSM614425 3 0.1524 0.863 0.000 0.000 0.932 0.060 0.008 0.000
#> GSM614426 3 0.1524 0.863 0.000 0.000 0.932 0.060 0.008 0.000
#> GSM614427 3 0.1584 0.861 0.000 0.000 0.928 0.064 0.008 0.000
#> GSM614428 3 0.1643 0.859 0.000 0.000 0.924 0.068 0.008 0.000
#> GSM614429 2 0.0363 0.838 0.000 0.988 0.000 0.000 0.000 0.012
#> GSM614430 2 0.0363 0.838 0.000 0.988 0.000 0.000 0.000 0.012
#> GSM614431 2 0.0363 0.838 0.000 0.988 0.000 0.000 0.000 0.012
#> GSM614432 2 0.0363 0.838 0.000 0.988 0.000 0.000 0.000 0.012
#> GSM614433 2 0.0363 0.838 0.000 0.988 0.000 0.000 0.000 0.012
#> GSM614434 2 0.0363 0.838 0.000 0.988 0.000 0.000 0.000 0.012
#> GSM614435 2 0.0363 0.838 0.000 0.988 0.000 0.000 0.000 0.012
#> GSM614436 2 0.2572 0.754 0.000 0.852 0.000 0.136 0.000 0.012
#> GSM614437 4 0.0458 0.953 0.000 0.016 0.000 0.984 0.000 0.000
#> GSM614438 4 0.0508 0.959 0.000 0.012 0.004 0.984 0.000 0.000
#> GSM614439 4 0.0508 0.959 0.000 0.012 0.004 0.984 0.000 0.000
#> GSM614440 4 0.0508 0.959 0.000 0.012 0.004 0.984 0.000 0.000
#> GSM614441 4 0.0508 0.959 0.000 0.012 0.004 0.984 0.000 0.000
#> GSM614442 4 0.0508 0.959 0.000 0.012 0.004 0.984 0.000 0.000
#> GSM614443 4 0.0363 0.957 0.000 0.012 0.000 0.988 0.000 0.000
#> GSM614444 4 0.0508 0.959 0.000 0.012 0.004 0.984 0.000 0.000
#> GSM614391 5 0.0000 0.888 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM614392 5 0.0000 0.888 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM614393 5 0.0000 0.888 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM614394 5 0.0000 0.888 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM614395 5 0.2775 0.691 0.000 0.000 0.040 0.104 0.856 0.000
#> GSM614396 5 0.0000 0.888 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM614397 5 0.0260 0.881 0.000 0.000 0.008 0.000 0.992 0.000
#> GSM614398 5 0.0260 0.881 0.000 0.000 0.008 0.000 0.992 0.000
#> GSM614399 1 0.1152 0.726 0.952 0.000 0.044 0.000 0.000 0.004
#> GSM614400 1 0.1152 0.726 0.952 0.000 0.044 0.000 0.000 0.004
#> GSM614401 1 0.1152 0.726 0.952 0.000 0.044 0.000 0.000 0.004
#> GSM614402 1 0.1152 0.726 0.952 0.000 0.044 0.000 0.000 0.004
#> GSM614403 1 0.1471 0.709 0.932 0.000 0.064 0.000 0.000 0.004
#> GSM614404 1 0.1152 0.726 0.952 0.000 0.044 0.000 0.000 0.004
#> GSM614405 1 0.1296 0.723 0.948 0.000 0.044 0.004 0.000 0.004
#> GSM614406 4 0.4193 0.625 0.272 0.000 0.044 0.684 0.000 0.000
#> GSM614407 6 0.4079 0.978 0.008 0.000 0.004 0.000 0.380 0.608
#> GSM614408 6 0.4079 0.978 0.008 0.000 0.004 0.000 0.380 0.608
#> GSM614409 6 0.4079 0.978 0.008 0.000 0.004 0.000 0.380 0.608
#> GSM614410 6 0.4079 0.978 0.008 0.000 0.004 0.000 0.380 0.608
#> GSM614411 6 0.4079 0.978 0.008 0.000 0.004 0.000 0.380 0.608
#> GSM614412 6 0.4079 0.978 0.008 0.000 0.004 0.000 0.380 0.608
#> GSM614413 6 0.4672 0.926 0.008 0.000 0.032 0.000 0.392 0.568
#> GSM614414 6 0.4481 0.935 0.008 0.000 0.020 0.000 0.400 0.572
#> GSM614445 3 0.4131 0.764 0.180 0.004 0.744 0.000 0.000 0.072
#> GSM614446 3 0.3695 0.786 0.164 0.000 0.776 0.000 0.000 0.060
#> GSM614447 3 0.3819 0.777 0.172 0.000 0.764 0.000 0.000 0.064
#> GSM614448 3 0.3191 0.832 0.096 0.000 0.844 0.016 0.000 0.044
#> GSM614449 3 0.3239 0.831 0.100 0.000 0.840 0.016 0.000 0.044
#> GSM614450 3 0.3088 0.817 0.120 0.000 0.832 0.000 0.000 0.048
#> GSM614451 3 0.3953 0.594 0.000 0.000 0.656 0.328 0.000 0.016
#> GSM614452 3 0.3853 0.633 0.000 0.000 0.680 0.304 0.000 0.016
#> GSM614453 2 0.3475 0.791 0.008 0.816 0.004 0.132 0.000 0.040
#> GSM614454 2 0.3475 0.791 0.008 0.816 0.004 0.132 0.000 0.040
#> GSM614455 2 0.3475 0.791 0.008 0.816 0.004 0.132 0.000 0.040
#> GSM614456 2 0.3475 0.791 0.008 0.816 0.004 0.132 0.000 0.040
#> GSM614457 2 0.3475 0.791 0.008 0.816 0.004 0.132 0.000 0.040
#> GSM614458 2 0.3475 0.791 0.008 0.816 0.004 0.132 0.000 0.040
#> GSM614459 2 0.3475 0.791 0.008 0.816 0.004 0.132 0.000 0.040
#> GSM614460 2 0.3475 0.791 0.008 0.816 0.004 0.132 0.000 0.040
#> GSM614461 2 0.4286 0.783 0.088 0.760 0.020 0.000 0.000 0.132
#> GSM614462 2 0.4286 0.783 0.088 0.760 0.020 0.000 0.000 0.132
#> GSM614463 2 0.4286 0.783 0.088 0.760 0.020 0.000 0.000 0.132
#> GSM614464 2 0.4286 0.783 0.088 0.760 0.020 0.000 0.000 0.132
#> GSM614465 2 0.4286 0.783 0.088 0.760 0.020 0.000 0.000 0.132
#> GSM614466 2 0.4286 0.783 0.088 0.760 0.020 0.000 0.000 0.132
#> GSM614467 2 0.4286 0.783 0.088 0.760 0.020 0.000 0.000 0.132
#> GSM614468 2 0.4286 0.783 0.088 0.760 0.020 0.000 0.000 0.132
#> GSM614469 1 0.4727 0.729 0.580 0.000 0.000 0.012 0.032 0.376
#> GSM614470 1 0.4727 0.729 0.580 0.000 0.000 0.012 0.032 0.376
#> GSM614471 1 0.4727 0.729 0.580 0.000 0.000 0.012 0.032 0.376
#> GSM614472 1 0.4727 0.729 0.580 0.000 0.000 0.012 0.032 0.376
#> GSM614473 1 0.4727 0.729 0.580 0.000 0.000 0.012 0.032 0.376
#> GSM614474 1 0.4727 0.729 0.580 0.000 0.000 0.012 0.032 0.376
#> GSM614475 1 0.4727 0.729 0.580 0.000 0.000 0.012 0.032 0.376
#> GSM614476 1 0.5272 0.720 0.572 0.000 0.024 0.020 0.024 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 individual(p) protocol(p) time(p) other(p) k
#> SD:skmeans 51 2.62e-08 1.000 0.995 0.3660 2
#> SD:skmeans 82 2.64e-21 0.389 1.000 0.2904 3
#> SD:skmeans 80 4.69e-33 0.645 1.000 0.0689 4
#> SD:skmeans 86 2.10e-47 0.912 1.000 0.0208 5
#> SD:skmeans 86 2.76e-59 0.933 1.000 0.0969 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 21168 rows and 86 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 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 0.812 0.926 0.965 0.4889 0.512 0.512
#> 3 3 0.894 0.914 0.963 0.2382 0.880 0.765
#> 4 4 0.817 0.837 0.922 0.1203 0.904 0.759
#> 5 5 0.790 0.802 0.886 0.0738 0.970 0.907
#> 6 6 0.819 0.719 0.839 0.0586 0.888 0.639
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
#> GSM614415 1 0.0000 0.964 1.000 0.000
#> GSM614416 1 0.0000 0.964 1.000 0.000
#> GSM614417 1 0.0000 0.964 1.000 0.000
#> GSM614418 1 0.0000 0.964 1.000 0.000
#> GSM614419 1 0.0000 0.964 1.000 0.000
#> GSM614420 1 0.0000 0.964 1.000 0.000
#> GSM614421 2 0.0000 0.961 0.000 1.000
#> GSM614422 1 0.7815 0.704 0.768 0.232
#> GSM614423 1 0.5629 0.844 0.868 0.132
#> GSM614424 2 0.4939 0.877 0.108 0.892
#> GSM614425 2 0.6801 0.796 0.180 0.820
#> GSM614426 2 0.2043 0.941 0.032 0.968
#> GSM614427 2 0.0000 0.961 0.000 1.000
#> GSM614428 2 0.0000 0.961 0.000 1.000
#> GSM614429 2 0.0000 0.961 0.000 1.000
#> GSM614430 2 0.0000 0.961 0.000 1.000
#> GSM614431 2 0.0000 0.961 0.000 1.000
#> GSM614432 2 0.0000 0.961 0.000 1.000
#> GSM614433 2 0.0000 0.961 0.000 1.000
#> GSM614434 2 0.0000 0.961 0.000 1.000
#> GSM614435 2 0.0000 0.961 0.000 1.000
#> GSM614436 2 0.0000 0.961 0.000 1.000
#> GSM614437 2 0.0000 0.961 0.000 1.000
#> GSM614438 2 0.0000 0.961 0.000 1.000
#> GSM614439 2 0.0000 0.961 0.000 1.000
#> GSM614440 2 0.0000 0.961 0.000 1.000
#> GSM614441 2 0.0000 0.961 0.000 1.000
#> GSM614442 2 0.0000 0.961 0.000 1.000
#> GSM614443 2 0.0000 0.961 0.000 1.000
#> GSM614444 2 0.0000 0.961 0.000 1.000
#> GSM614391 1 0.0000 0.964 1.000 0.000
#> GSM614392 1 0.0000 0.964 1.000 0.000
#> GSM614393 1 0.0000 0.964 1.000 0.000
#> GSM614394 1 0.0000 0.964 1.000 0.000
#> GSM614395 2 0.7528 0.750 0.216 0.784
#> GSM614396 1 0.0000 0.964 1.000 0.000
#> GSM614397 2 0.5842 0.846 0.140 0.860
#> GSM614398 1 0.0672 0.960 0.992 0.008
#> GSM614399 1 0.7674 0.719 0.776 0.224
#> GSM614400 1 0.0376 0.964 0.996 0.004
#> GSM614401 1 0.0376 0.964 0.996 0.004
#> GSM614402 1 0.0672 0.962 0.992 0.008
#> GSM614403 1 0.8016 0.691 0.756 0.244
#> GSM614404 1 0.0376 0.964 0.996 0.004
#> GSM614405 1 0.8144 0.665 0.748 0.252
#> GSM614406 2 0.0000 0.961 0.000 1.000
#> GSM614407 1 0.0376 0.964 0.996 0.004
#> GSM614408 1 0.0000 0.964 1.000 0.000
#> GSM614409 1 0.0000 0.964 1.000 0.000
#> GSM614410 1 0.0000 0.964 1.000 0.000
#> GSM614411 1 0.0376 0.964 0.996 0.004
#> GSM614412 1 0.0376 0.963 0.996 0.004
#> GSM614413 2 0.7056 0.781 0.192 0.808
#> GSM614414 2 0.8443 0.657 0.272 0.728
#> GSM614445 2 0.0938 0.954 0.012 0.988
#> GSM614446 2 0.7219 0.770 0.200 0.800
#> GSM614447 2 0.3431 0.913 0.064 0.936
#> GSM614448 2 0.1843 0.944 0.028 0.972
#> GSM614449 2 0.0000 0.961 0.000 1.000
#> GSM614450 2 0.2423 0.936 0.040 0.960
#> GSM614451 2 0.0000 0.961 0.000 1.000
#> GSM614452 2 0.0000 0.961 0.000 1.000
#> GSM614453 2 0.0000 0.961 0.000 1.000
#> GSM614454 2 0.0000 0.961 0.000 1.000
#> GSM614455 2 0.0000 0.961 0.000 1.000
#> GSM614456 2 0.0000 0.961 0.000 1.000
#> GSM614457 2 0.0000 0.961 0.000 1.000
#> GSM614458 2 0.0000 0.961 0.000 1.000
#> GSM614459 2 0.0000 0.961 0.000 1.000
#> GSM614460 2 0.0000 0.961 0.000 1.000
#> GSM614461 2 0.0000 0.961 0.000 1.000
#> GSM614462 2 0.0000 0.961 0.000 1.000
#> GSM614463 2 0.9358 0.472 0.352 0.648
#> GSM614464 2 0.0000 0.961 0.000 1.000
#> GSM614465 2 0.0000 0.961 0.000 1.000
#> GSM614466 2 0.2603 0.933 0.044 0.956
#> GSM614467 2 0.0000 0.961 0.000 1.000
#> GSM614468 2 0.0000 0.961 0.000 1.000
#> GSM614469 1 0.0376 0.964 0.996 0.004
#> GSM614470 1 0.0376 0.964 0.996 0.004
#> GSM614471 1 0.0376 0.964 0.996 0.004
#> GSM614472 1 0.0376 0.964 0.996 0.004
#> GSM614473 1 0.0376 0.964 0.996 0.004
#> GSM614474 1 0.0376 0.964 0.996 0.004
#> GSM614475 1 0.0376 0.964 0.996 0.004
#> GSM614476 1 0.0376 0.964 0.996 0.004
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM614415 1 0.0237 0.947 0.996 0.000 0.004
#> GSM614416 1 0.0237 0.947 0.996 0.000 0.004
#> GSM614417 1 0.0237 0.947 0.996 0.000 0.004
#> GSM614418 1 0.0237 0.947 0.996 0.000 0.004
#> GSM614419 1 0.0237 0.947 0.996 0.000 0.004
#> GSM614420 1 0.0237 0.947 0.996 0.000 0.004
#> GSM614421 2 0.0237 0.956 0.004 0.996 0.000
#> GSM614422 1 0.5529 0.583 0.704 0.296 0.000
#> GSM614423 1 0.4654 0.715 0.792 0.208 0.000
#> GSM614424 2 0.2448 0.899 0.076 0.924 0.000
#> GSM614425 2 0.2356 0.902 0.072 0.928 0.000
#> GSM614426 2 0.1031 0.944 0.024 0.976 0.000
#> GSM614427 2 0.0237 0.956 0.004 0.996 0.000
#> GSM614428 2 0.0237 0.956 0.004 0.996 0.000
#> GSM614429 2 0.0000 0.957 0.000 1.000 0.000
#> GSM614430 2 0.0000 0.957 0.000 1.000 0.000
#> GSM614431 2 0.0000 0.957 0.000 1.000 0.000
#> GSM614432 2 0.0000 0.957 0.000 1.000 0.000
#> GSM614433 2 0.0000 0.957 0.000 1.000 0.000
#> GSM614434 2 0.0000 0.957 0.000 1.000 0.000
#> GSM614435 2 0.0000 0.957 0.000 1.000 0.000
#> GSM614436 2 0.0000 0.957 0.000 1.000 0.000
#> GSM614437 3 0.0237 0.967 0.000 0.004 0.996
#> GSM614438 3 0.0237 0.967 0.000 0.004 0.996
#> GSM614439 3 0.0237 0.967 0.000 0.004 0.996
#> GSM614440 3 0.0237 0.967 0.000 0.004 0.996
#> GSM614441 3 0.0237 0.967 0.000 0.004 0.996
#> GSM614442 3 0.0237 0.967 0.000 0.004 0.996
#> GSM614443 3 0.0237 0.967 0.000 0.004 0.996
#> GSM614444 3 0.0237 0.967 0.000 0.004 0.996
#> GSM614391 1 0.0237 0.947 0.996 0.000 0.004
#> GSM614392 1 0.0237 0.947 0.996 0.000 0.004
#> GSM614393 1 0.0237 0.947 0.996 0.000 0.004
#> GSM614394 1 0.0237 0.947 0.996 0.000 0.004
#> GSM614395 3 0.7676 0.618 0.216 0.112 0.672
#> GSM614396 1 0.0237 0.947 0.996 0.000 0.004
#> GSM614397 2 0.5763 0.623 0.276 0.716 0.008
#> GSM614398 1 0.0661 0.942 0.988 0.008 0.004
#> GSM614399 1 0.4555 0.724 0.800 0.200 0.000
#> GSM614400 1 0.0000 0.947 1.000 0.000 0.000
#> GSM614401 1 0.0237 0.946 0.996 0.004 0.000
#> GSM614402 1 0.0592 0.939 0.988 0.012 0.000
#> GSM614403 1 0.5948 0.451 0.640 0.360 0.000
#> GSM614404 1 0.0592 0.938 0.988 0.012 0.000
#> GSM614405 1 0.4931 0.671 0.768 0.232 0.000
#> GSM614406 2 0.0424 0.955 0.008 0.992 0.000
#> GSM614407 1 0.0000 0.947 1.000 0.000 0.000
#> GSM614408 1 0.0000 0.947 1.000 0.000 0.000
#> GSM614409 1 0.0000 0.947 1.000 0.000 0.000
#> GSM614410 1 0.0000 0.947 1.000 0.000 0.000
#> GSM614411 1 0.0000 0.947 1.000 0.000 0.000
#> GSM614412 1 0.0237 0.945 0.996 0.004 0.000
#> GSM614413 2 0.4842 0.714 0.224 0.776 0.000
#> GSM614414 2 0.5465 0.616 0.288 0.712 0.000
#> GSM614445 2 0.0237 0.956 0.004 0.996 0.000
#> GSM614446 2 0.2261 0.907 0.068 0.932 0.000
#> GSM614447 2 0.2261 0.903 0.068 0.932 0.000
#> GSM614448 2 0.0592 0.952 0.012 0.988 0.000
#> GSM614449 2 0.0237 0.956 0.004 0.996 0.000
#> GSM614450 2 0.0592 0.952 0.012 0.988 0.000
#> GSM614451 3 0.0661 0.961 0.004 0.008 0.988
#> GSM614452 3 0.1267 0.949 0.004 0.024 0.972
#> GSM614453 2 0.0000 0.957 0.000 1.000 0.000
#> GSM614454 2 0.0000 0.957 0.000 1.000 0.000
#> GSM614455 2 0.0000 0.957 0.000 1.000 0.000
#> GSM614456 2 0.0000 0.957 0.000 1.000 0.000
#> GSM614457 2 0.0000 0.957 0.000 1.000 0.000
#> GSM614458 2 0.0000 0.957 0.000 1.000 0.000
#> GSM614459 2 0.0000 0.957 0.000 1.000 0.000
#> GSM614460 2 0.0000 0.957 0.000 1.000 0.000
#> GSM614461 2 0.0000 0.957 0.000 1.000 0.000
#> GSM614462 2 0.0000 0.957 0.000 1.000 0.000
#> GSM614463 2 0.4842 0.704 0.224 0.776 0.000
#> GSM614464 2 0.0000 0.957 0.000 1.000 0.000
#> GSM614465 2 0.0000 0.957 0.000 1.000 0.000
#> GSM614466 2 0.0592 0.951 0.012 0.988 0.000
#> GSM614467 2 0.0237 0.956 0.004 0.996 0.000
#> GSM614468 2 0.0237 0.956 0.004 0.996 0.000
#> GSM614469 1 0.0000 0.947 1.000 0.000 0.000
#> GSM614470 1 0.0000 0.947 1.000 0.000 0.000
#> GSM614471 1 0.0237 0.946 0.996 0.004 0.000
#> GSM614472 1 0.0000 0.947 1.000 0.000 0.000
#> GSM614473 1 0.0000 0.947 1.000 0.000 0.000
#> GSM614474 1 0.0000 0.947 1.000 0.000 0.000
#> GSM614475 1 0.0000 0.947 1.000 0.000 0.000
#> GSM614476 1 0.0000 0.947 1.000 0.000 0.000
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM614415 1 0.4072 0.699 0.748 0.000 0.252 0.000
#> GSM614416 1 0.4877 0.485 0.592 0.000 0.408 0.000
#> GSM614417 1 0.4713 0.572 0.640 0.000 0.360 0.000
#> GSM614418 1 0.3975 0.706 0.760 0.000 0.240 0.000
#> GSM614419 1 0.3975 0.706 0.760 0.000 0.240 0.000
#> GSM614420 1 0.4040 0.699 0.752 0.000 0.248 0.000
#> GSM614421 2 0.1302 0.940 0.044 0.956 0.000 0.000
#> GSM614422 3 0.6055 0.153 0.044 0.436 0.520 0.000
#> GSM614423 3 0.4224 0.658 0.044 0.144 0.812 0.000
#> GSM614424 2 0.2996 0.894 0.044 0.892 0.064 0.000
#> GSM614425 2 0.2675 0.907 0.044 0.908 0.048 0.000
#> GSM614426 2 0.2111 0.929 0.044 0.932 0.024 0.000
#> GSM614427 2 0.1302 0.940 0.044 0.956 0.000 0.000
#> GSM614428 2 0.1302 0.940 0.044 0.956 0.000 0.000
#> GSM614429 2 0.0000 0.958 0.000 1.000 0.000 0.000
#> GSM614430 2 0.0000 0.958 0.000 1.000 0.000 0.000
#> GSM614431 2 0.0000 0.958 0.000 1.000 0.000 0.000
#> GSM614432 2 0.0000 0.958 0.000 1.000 0.000 0.000
#> GSM614433 2 0.0000 0.958 0.000 1.000 0.000 0.000
#> GSM614434 2 0.0000 0.958 0.000 1.000 0.000 0.000
#> GSM614435 2 0.0000 0.958 0.000 1.000 0.000 0.000
#> GSM614436 2 0.0000 0.958 0.000 1.000 0.000 0.000
#> GSM614437 4 0.0000 0.986 0.000 0.000 0.000 1.000
#> GSM614438 4 0.0000 0.986 0.000 0.000 0.000 1.000
#> GSM614439 4 0.0000 0.986 0.000 0.000 0.000 1.000
#> GSM614440 4 0.0000 0.986 0.000 0.000 0.000 1.000
#> GSM614441 4 0.0000 0.986 0.000 0.000 0.000 1.000
#> GSM614442 4 0.0000 0.986 0.000 0.000 0.000 1.000
#> GSM614443 4 0.0000 0.986 0.000 0.000 0.000 1.000
#> GSM614444 4 0.0000 0.986 0.000 0.000 0.000 1.000
#> GSM614391 1 0.3569 0.691 0.804 0.000 0.196 0.000
#> GSM614392 1 0.4730 0.518 0.636 0.000 0.364 0.000
#> GSM614393 1 0.4564 0.574 0.672 0.000 0.328 0.000
#> GSM614394 1 0.2216 0.688 0.908 0.000 0.092 0.000
#> GSM614395 1 0.5838 -0.103 0.524 0.032 0.000 0.444
#> GSM614396 1 0.1716 0.676 0.936 0.000 0.064 0.000
#> GSM614397 1 0.4855 0.175 0.600 0.400 0.000 0.000
#> GSM614398 1 0.0921 0.648 0.972 0.000 0.028 0.000
#> GSM614399 3 0.2868 0.720 0.000 0.136 0.864 0.000
#> GSM614400 3 0.0000 0.889 0.000 0.000 1.000 0.000
#> GSM614401 3 0.0000 0.889 0.000 0.000 1.000 0.000
#> GSM614402 3 0.0336 0.882 0.000 0.008 0.992 0.000
#> GSM614403 3 0.4746 0.337 0.000 0.368 0.632 0.000
#> GSM614404 3 0.0000 0.889 0.000 0.000 1.000 0.000
#> GSM614405 3 0.3528 0.624 0.000 0.192 0.808 0.000
#> GSM614406 2 0.0592 0.952 0.016 0.984 0.000 0.000
#> GSM614407 3 0.0000 0.889 0.000 0.000 1.000 0.000
#> GSM614408 3 0.0000 0.889 0.000 0.000 1.000 0.000
#> GSM614409 3 0.0000 0.889 0.000 0.000 1.000 0.000
#> GSM614410 3 0.0000 0.889 0.000 0.000 1.000 0.000
#> GSM614411 3 0.0000 0.889 0.000 0.000 1.000 0.000
#> GSM614412 3 0.0469 0.878 0.000 0.012 0.988 0.000
#> GSM614413 2 0.4839 0.710 0.044 0.756 0.200 0.000
#> GSM614414 2 0.5168 0.636 0.040 0.712 0.248 0.000
#> GSM614445 2 0.0336 0.955 0.008 0.992 0.000 0.000
#> GSM614446 2 0.2111 0.923 0.024 0.932 0.044 0.000
#> GSM614447 2 0.2480 0.879 0.008 0.904 0.088 0.000
#> GSM614448 2 0.1635 0.936 0.044 0.948 0.008 0.000
#> GSM614449 2 0.1302 0.940 0.044 0.956 0.000 0.000
#> GSM614450 2 0.1109 0.947 0.028 0.968 0.004 0.000
#> GSM614451 4 0.1489 0.945 0.044 0.004 0.000 0.952
#> GSM614452 4 0.1888 0.932 0.044 0.016 0.000 0.940
#> GSM614453 2 0.0000 0.958 0.000 1.000 0.000 0.000
#> GSM614454 2 0.0000 0.958 0.000 1.000 0.000 0.000
#> GSM614455 2 0.0000 0.958 0.000 1.000 0.000 0.000
#> GSM614456 2 0.0000 0.958 0.000 1.000 0.000 0.000
#> GSM614457 2 0.0000 0.958 0.000 1.000 0.000 0.000
#> GSM614458 2 0.0000 0.958 0.000 1.000 0.000 0.000
#> GSM614459 2 0.0000 0.958 0.000 1.000 0.000 0.000
#> GSM614460 2 0.0000 0.958 0.000 1.000 0.000 0.000
#> GSM614461 2 0.0000 0.958 0.000 1.000 0.000 0.000
#> GSM614462 2 0.0000 0.958 0.000 1.000 0.000 0.000
#> GSM614463 2 0.4250 0.600 0.000 0.724 0.276 0.000
#> GSM614464 2 0.0000 0.958 0.000 1.000 0.000 0.000
#> GSM614465 2 0.0000 0.958 0.000 1.000 0.000 0.000
#> GSM614466 2 0.0336 0.954 0.000 0.992 0.008 0.000
#> GSM614467 2 0.0000 0.958 0.000 1.000 0.000 0.000
#> GSM614468 2 0.0000 0.958 0.000 1.000 0.000 0.000
#> GSM614469 3 0.0000 0.889 0.000 0.000 1.000 0.000
#> GSM614470 3 0.0000 0.889 0.000 0.000 1.000 0.000
#> GSM614471 3 0.0000 0.889 0.000 0.000 1.000 0.000
#> GSM614472 3 0.0000 0.889 0.000 0.000 1.000 0.000
#> GSM614473 3 0.0000 0.889 0.000 0.000 1.000 0.000
#> GSM614474 3 0.0000 0.889 0.000 0.000 1.000 0.000
#> GSM614475 3 0.0000 0.889 0.000 0.000 1.000 0.000
#> GSM614476 3 0.0000 0.889 0.000 0.000 1.000 0.000
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM614415 3 0.0609 0.874 0.020 0.000 0.980 0.000 0.000
#> GSM614416 3 0.2648 0.774 0.152 0.000 0.848 0.000 0.000
#> GSM614417 3 0.2280 0.819 0.120 0.000 0.880 0.000 0.000
#> GSM614418 3 0.0404 0.867 0.012 0.000 0.988 0.000 0.000
#> GSM614419 3 0.0290 0.860 0.008 0.000 0.992 0.000 0.000
#> GSM614420 3 0.1205 0.874 0.040 0.000 0.956 0.000 0.004
#> GSM614421 2 0.4138 0.656 0.000 0.616 0.000 0.000 0.384
#> GSM614422 2 0.6633 0.345 0.220 0.396 0.000 0.000 0.384
#> GSM614423 1 0.5415 0.341 0.552 0.064 0.000 0.000 0.384
#> GSM614424 2 0.5123 0.617 0.044 0.572 0.000 0.000 0.384
#> GSM614425 2 0.4138 0.656 0.000 0.616 0.000 0.000 0.384
#> GSM614426 2 0.4846 0.633 0.028 0.588 0.000 0.000 0.384
#> GSM614427 2 0.4138 0.656 0.000 0.616 0.000 0.000 0.384
#> GSM614428 2 0.4138 0.656 0.000 0.616 0.000 0.000 0.384
#> GSM614429 2 0.0000 0.853 0.000 1.000 0.000 0.000 0.000
#> GSM614430 2 0.0000 0.853 0.000 1.000 0.000 0.000 0.000
#> GSM614431 2 0.0000 0.853 0.000 1.000 0.000 0.000 0.000
#> GSM614432 2 0.0000 0.853 0.000 1.000 0.000 0.000 0.000
#> GSM614433 2 0.0000 0.853 0.000 1.000 0.000 0.000 0.000
#> GSM614434 2 0.0000 0.853 0.000 1.000 0.000 0.000 0.000
#> GSM614435 2 0.0000 0.853 0.000 1.000 0.000 0.000 0.000
#> GSM614436 2 0.0000 0.853 0.000 1.000 0.000 0.000 0.000
#> GSM614437 4 0.0000 0.891 0.000 0.000 0.000 1.000 0.000
#> GSM614438 4 0.0000 0.891 0.000 0.000 0.000 1.000 0.000
#> GSM614439 4 0.0000 0.891 0.000 0.000 0.000 1.000 0.000
#> GSM614440 4 0.0000 0.891 0.000 0.000 0.000 1.000 0.000
#> GSM614441 4 0.0000 0.891 0.000 0.000 0.000 1.000 0.000
#> GSM614442 4 0.0000 0.891 0.000 0.000 0.000 1.000 0.000
#> GSM614443 4 0.0000 0.891 0.000 0.000 0.000 1.000 0.000
#> GSM614444 4 0.0000 0.891 0.000 0.000 0.000 1.000 0.000
#> GSM614391 5 0.4836 0.762 0.032 0.000 0.356 0.000 0.612
#> GSM614392 5 0.5376 0.751 0.080 0.000 0.308 0.000 0.612
#> GSM614393 5 0.5342 0.754 0.076 0.000 0.312 0.000 0.612
#> GSM614394 5 0.4288 0.746 0.004 0.000 0.384 0.000 0.612
#> GSM614395 5 0.2773 0.714 0.000 0.000 0.164 0.000 0.836
#> GSM614396 5 0.4150 0.741 0.000 0.000 0.388 0.000 0.612
#> GSM614397 5 0.2773 0.714 0.000 0.000 0.164 0.000 0.836
#> GSM614398 5 0.3003 0.730 0.000 0.000 0.188 0.000 0.812
#> GSM614399 1 0.2230 0.798 0.884 0.116 0.000 0.000 0.000
#> GSM614400 1 0.0000 0.926 1.000 0.000 0.000 0.000 0.000
#> GSM614401 1 0.0000 0.926 1.000 0.000 0.000 0.000 0.000
#> GSM614402 1 0.0162 0.924 0.996 0.004 0.000 0.000 0.000
#> GSM614403 1 0.4949 0.474 0.656 0.288 0.000 0.000 0.056
#> GSM614404 1 0.0000 0.926 1.000 0.000 0.000 0.000 0.000
#> GSM614405 1 0.2690 0.736 0.844 0.156 0.000 0.000 0.000
#> GSM614406 2 0.2280 0.811 0.000 0.880 0.000 0.000 0.120
#> GSM614407 1 0.0324 0.924 0.992 0.000 0.004 0.000 0.004
#> GSM614408 1 0.0324 0.924 0.992 0.000 0.004 0.000 0.004
#> GSM614409 1 0.0324 0.924 0.992 0.000 0.004 0.000 0.004
#> GSM614410 1 0.0324 0.924 0.992 0.000 0.004 0.000 0.004
#> GSM614411 1 0.0324 0.924 0.992 0.000 0.004 0.000 0.004
#> GSM614412 1 0.1202 0.895 0.960 0.032 0.004 0.000 0.004
#> GSM614413 2 0.5292 0.624 0.048 0.580 0.004 0.000 0.368
#> GSM614414 2 0.5937 0.604 0.120 0.576 0.004 0.000 0.300
#> GSM614445 2 0.1502 0.837 0.004 0.940 0.000 0.000 0.056
#> GSM614446 2 0.3596 0.771 0.016 0.784 0.000 0.000 0.200
#> GSM614447 2 0.3644 0.771 0.096 0.824 0.000 0.000 0.080
#> GSM614448 2 0.4138 0.656 0.000 0.616 0.000 0.000 0.384
#> GSM614449 2 0.4088 0.668 0.000 0.632 0.000 0.000 0.368
#> GSM614450 2 0.3530 0.769 0.012 0.784 0.000 0.000 0.204
#> GSM614451 4 0.4138 0.517 0.000 0.000 0.000 0.616 0.384
#> GSM614452 4 0.4505 0.503 0.000 0.012 0.000 0.604 0.384
#> GSM614453 2 0.0000 0.853 0.000 1.000 0.000 0.000 0.000
#> GSM614454 2 0.0000 0.853 0.000 1.000 0.000 0.000 0.000
#> GSM614455 2 0.0000 0.853 0.000 1.000 0.000 0.000 0.000
#> GSM614456 2 0.0000 0.853 0.000 1.000 0.000 0.000 0.000
#> GSM614457 2 0.0000 0.853 0.000 1.000 0.000 0.000 0.000
#> GSM614458 2 0.0000 0.853 0.000 1.000 0.000 0.000 0.000
#> GSM614459 2 0.0000 0.853 0.000 1.000 0.000 0.000 0.000
#> GSM614460 2 0.0000 0.853 0.000 1.000 0.000 0.000 0.000
#> GSM614461 2 0.0000 0.853 0.000 1.000 0.000 0.000 0.000
#> GSM614462 2 0.0000 0.853 0.000 1.000 0.000 0.000 0.000
#> GSM614463 2 0.3837 0.498 0.308 0.692 0.000 0.000 0.000
#> GSM614464 2 0.0000 0.853 0.000 1.000 0.000 0.000 0.000
#> GSM614465 2 0.0000 0.853 0.000 1.000 0.000 0.000 0.000
#> GSM614466 2 0.0510 0.846 0.016 0.984 0.000 0.000 0.000
#> GSM614467 2 0.0609 0.848 0.000 0.980 0.000 0.000 0.020
#> GSM614468 2 0.0000 0.853 0.000 1.000 0.000 0.000 0.000
#> GSM614469 1 0.0000 0.926 1.000 0.000 0.000 0.000 0.000
#> GSM614470 1 0.0000 0.926 1.000 0.000 0.000 0.000 0.000
#> GSM614471 1 0.0000 0.926 1.000 0.000 0.000 0.000 0.000
#> GSM614472 1 0.0000 0.926 1.000 0.000 0.000 0.000 0.000
#> GSM614473 1 0.0000 0.926 1.000 0.000 0.000 0.000 0.000
#> GSM614474 1 0.0000 0.926 1.000 0.000 0.000 0.000 0.000
#> GSM614475 1 0.0000 0.926 1.000 0.000 0.000 0.000 0.000
#> GSM614476 1 0.0000 0.926 1.000 0.000 0.000 0.000 0.000
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM614415 6 0.1007 0.9828 0.000 0.000 0.000 0.000 0.044 0.956
#> GSM614416 6 0.1176 0.9740 0.024 0.000 0.000 0.000 0.020 0.956
#> GSM614417 6 0.1176 0.9781 0.020 0.000 0.000 0.000 0.024 0.956
#> GSM614418 6 0.1007 0.9828 0.000 0.000 0.000 0.000 0.044 0.956
#> GSM614419 6 0.1007 0.9828 0.000 0.000 0.000 0.000 0.044 0.956
#> GSM614420 6 0.1151 0.9827 0.012 0.000 0.000 0.000 0.032 0.956
#> GSM614421 3 0.0000 0.6528 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM614422 3 0.1765 0.6017 0.096 0.000 0.904 0.000 0.000 0.000
#> GSM614423 3 0.3747 0.1221 0.396 0.000 0.604 0.000 0.000 0.000
#> GSM614424 3 0.0458 0.6514 0.016 0.000 0.984 0.000 0.000 0.000
#> GSM614425 3 0.0000 0.6528 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM614426 3 0.0363 0.6524 0.012 0.000 0.988 0.000 0.000 0.000
#> GSM614427 3 0.0000 0.6528 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM614428 3 0.0000 0.6528 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM614429 2 0.4039 0.8395 0.000 0.568 0.424 0.000 0.000 0.008
#> GSM614430 2 0.4039 0.8397 0.000 0.568 0.424 0.000 0.000 0.008
#> GSM614431 2 0.3804 0.8394 0.000 0.576 0.424 0.000 0.000 0.000
#> GSM614432 2 0.4212 0.8354 0.000 0.560 0.424 0.000 0.000 0.016
#> GSM614433 2 0.4039 0.8397 0.000 0.568 0.424 0.000 0.000 0.008
#> GSM614434 2 0.4039 0.8397 0.000 0.568 0.424 0.000 0.000 0.008
#> GSM614435 2 0.4212 0.8354 0.000 0.560 0.424 0.000 0.000 0.016
#> GSM614436 2 0.4039 0.8395 0.000 0.568 0.424 0.000 0.000 0.008
#> GSM614437 4 0.0000 1.0000 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM614438 4 0.0000 1.0000 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM614439 4 0.0000 1.0000 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM614440 4 0.0000 1.0000 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM614441 4 0.0000 1.0000 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM614442 4 0.0000 1.0000 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM614443 4 0.0000 1.0000 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM614444 4 0.0000 1.0000 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM614391 5 0.0146 0.9942 0.000 0.000 0.000 0.000 0.996 0.004
#> GSM614392 5 0.0146 0.9942 0.000 0.000 0.000 0.000 0.996 0.004
#> GSM614393 5 0.0146 0.9942 0.000 0.000 0.000 0.000 0.996 0.004
#> GSM614394 5 0.0146 0.9942 0.000 0.000 0.000 0.000 0.996 0.004
#> GSM614395 5 0.0363 0.9859 0.000 0.000 0.012 0.000 0.988 0.000
#> GSM614396 5 0.0146 0.9942 0.000 0.000 0.000 0.000 0.996 0.004
#> GSM614397 5 0.0363 0.9859 0.000 0.000 0.012 0.000 0.988 0.000
#> GSM614398 5 0.0146 0.9924 0.000 0.000 0.004 0.000 0.996 0.000
#> GSM614399 1 0.2308 0.7432 0.892 0.040 0.068 0.000 0.000 0.000
#> GSM614400 1 0.0000 0.8210 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM614401 1 0.0000 0.8210 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM614402 1 0.0146 0.8188 0.996 0.000 0.004 0.000 0.000 0.000
#> GSM614403 1 0.4968 0.3499 0.632 0.120 0.248 0.000 0.000 0.000
#> GSM614404 1 0.0000 0.8210 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM614405 1 0.2383 0.7307 0.880 0.024 0.096 0.000 0.000 0.000
#> GSM614406 3 0.4025 -0.5628 0.000 0.416 0.576 0.000 0.000 0.008
#> GSM614407 1 0.4344 0.6043 0.556 0.424 0.000 0.000 0.004 0.016
#> GSM614408 1 0.4344 0.6043 0.556 0.424 0.000 0.000 0.004 0.016
#> GSM614409 1 0.4344 0.6043 0.556 0.424 0.000 0.000 0.004 0.016
#> GSM614410 1 0.4344 0.6043 0.556 0.424 0.000 0.000 0.004 0.016
#> GSM614411 1 0.4344 0.6043 0.556 0.424 0.000 0.000 0.004 0.016
#> GSM614412 1 0.4370 0.5867 0.536 0.444 0.000 0.000 0.004 0.016
#> GSM614413 2 0.4930 -0.3984 0.024 0.484 0.472 0.000 0.004 0.016
#> GSM614414 2 0.5666 -0.3762 0.084 0.464 0.432 0.000 0.004 0.016
#> GSM614445 2 0.4181 0.7626 0.012 0.512 0.476 0.000 0.000 0.000
#> GSM614446 3 0.3940 -0.3628 0.012 0.348 0.640 0.000 0.000 0.000
#> GSM614447 2 0.5191 0.5504 0.088 0.456 0.456 0.000 0.000 0.000
#> GSM614448 3 0.0458 0.6373 0.000 0.016 0.984 0.000 0.000 0.000
#> GSM614449 3 0.0937 0.6051 0.000 0.040 0.960 0.000 0.000 0.000
#> GSM614450 3 0.3766 -0.2048 0.012 0.304 0.684 0.000 0.000 0.000
#> GSM614451 3 0.3804 -0.0890 0.000 0.000 0.576 0.424 0.000 0.000
#> GSM614452 3 0.3789 -0.0638 0.000 0.000 0.584 0.416 0.000 0.000
#> GSM614453 2 0.4039 0.8397 0.000 0.568 0.424 0.000 0.000 0.008
#> GSM614454 2 0.4039 0.8397 0.000 0.568 0.424 0.000 0.000 0.008
#> GSM614455 2 0.4039 0.8397 0.000 0.568 0.424 0.000 0.000 0.008
#> GSM614456 2 0.4039 0.8397 0.000 0.568 0.424 0.000 0.000 0.008
#> GSM614457 2 0.4039 0.8397 0.000 0.568 0.424 0.000 0.000 0.008
#> GSM614458 2 0.4039 0.8397 0.000 0.568 0.424 0.000 0.000 0.008
#> GSM614459 2 0.4039 0.8397 0.000 0.568 0.424 0.000 0.000 0.008
#> GSM614460 2 0.4039 0.8397 0.000 0.568 0.424 0.000 0.000 0.008
#> GSM614461 2 0.4289 0.8338 0.000 0.556 0.424 0.000 0.000 0.020
#> GSM614462 2 0.4289 0.8338 0.000 0.556 0.424 0.000 0.000 0.020
#> GSM614463 2 0.6345 0.3368 0.280 0.456 0.244 0.000 0.000 0.020
#> GSM614464 2 0.4289 0.8338 0.000 0.556 0.424 0.000 0.000 0.020
#> GSM614465 2 0.4289 0.8338 0.000 0.556 0.424 0.000 0.000 0.020
#> GSM614466 2 0.4693 0.8182 0.016 0.540 0.424 0.000 0.000 0.020
#> GSM614467 2 0.4294 0.8311 0.000 0.552 0.428 0.000 0.000 0.020
#> GSM614468 2 0.4289 0.8338 0.000 0.556 0.424 0.000 0.000 0.020
#> GSM614469 1 0.0000 0.8210 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM614470 1 0.0000 0.8210 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM614471 1 0.0000 0.8210 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM614472 1 0.0000 0.8210 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM614473 1 0.0000 0.8210 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM614474 1 0.0000 0.8210 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM614475 1 0.0000 0.8210 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM614476 1 0.0000 0.8210 1.000 0.000 0.000 0.000 0.000 0.000
Heatmaps for the consensus matrix. It visualizes the probability of two samples to be in a same group.
consensus_heatmap(res, k = 2)
consensus_heatmap(res, k = 3)
consensus_heatmap(res, k = 4)
consensus_heatmap(res, k = 5)
consensus_heatmap(res, k = 6)
Heatmaps for the membership of samples in all partitions to see how consistent they are:
membership_heatmap(res, k = 2)
membership_heatmap(res, k = 3)
membership_heatmap(res, k = 4)
membership_heatmap(res, k = 5)
membership_heatmap(res, k = 6)
As soon as we have had the classes for columns, we can look for signatures which are significantly different between classes which can be candidate marks for certain classes. Following are the heatmaps for signatures.
Signature heatmaps where rows are scaled:
get_signatures(res, k = 2)
get_signatures(res, k = 3)
get_signatures(res, k = 4)
get_signatures(res, k = 5)
get_signatures(res, k = 6)
Signature heatmaps where rows are not scaled:
get_signatures(res, k = 2, scale_rows = FALSE)
get_signatures(res, k = 3, scale_rows = FALSE)
get_signatures(res, k = 4, scale_rows = FALSE)
get_signatures(res, k = 5, scale_rows = FALSE)
get_signatures(res, k = 6, scale_rows = FALSE)
Compare the overlap of signatures from different k:
compare_signatures(res)
get_signature() returns a data frame invisibly. TO get the list of signatures, the function
call should be assigned to a variable explicitly. In following code, if plot argument is set
to FALSE, no heatmap is plotted while only the differential analysis is performed.
# code only for demonstration
tb = get_signature(res, k = ..., plot = FALSE)
An example of the output of tb is:
#> which_row fdr mean_1 mean_2 scaled_mean_1 scaled_mean_2 km
#> 1 38 0.042760348 8.373488 9.131774 -0.5533452 0.5164555 1
#> 2 40 0.018707592 7.106213 8.469186 -0.6173731 0.5762149 1
#> 3 55 0.019134737 10.221463 11.207825 -0.6159697 0.5749050 1
#> 4 59 0.006059896 5.921854 7.869574 -0.6899429 0.6439467 1
#> 5 60 0.018055526 8.928898 10.211722 -0.6204761 0.5791110 1
#> 6 98 0.009384629 15.714769 14.887706 0.6635654 -0.6193277 2
...
The columns in tb are:
which_row: row indices corresponding to the input matrix.fdr: FDR for the differential test. mean_x: The mean value in group x.scaled_mean_x: The mean value in group x after rows are scaled.km: Row groups if k-means clustering is applied to rows.UMAP plot which shows how samples are separated.
dimension_reduction(res, k = 2, method = "UMAP")
dimension_reduction(res, k = 3, method = "UMAP")
dimension_reduction(res, k = 4, method = "UMAP")
dimension_reduction(res, k = 5, method = "UMAP")
dimension_reduction(res, k = 6, method = "UMAP")
Following heatmap shows how subgroups are split when increasing k:
collect_classes(res)
Test correlation between subgroups and known annotations. If the known annotation is numeric, one-way ANOVA test is applied, and if the known annotation is discrete, chi-squared contingency table test is applied.
test_to_known_factors(res)
#> n individual(p) protocol(p) time(p) other(p) k
#> SD:pam 85 1.06e-09 0.0523 0.898 0.9010 2
#> SD:pam 85 3.34e-17 0.1026 0.991 0.1653 3
#> SD:pam 81 1.17e-28 0.2345 1.000 0.1245 4
#> SD:pam 82 4.06e-41 0.3260 1.000 0.1052 5
#> SD:pam 76 4.13e-50 0.7974 1.000 0.0423 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 21168 rows and 86 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 5.
#>
#> Following methods can be applied to this 'ConsensusPartition' object:
#> [1] "cola_report" "collect_classes" "collect_plots"
#> [4] "collect_stats" "colnames" "compare_signatures"
#> [7] "consensus_heatmap" "dimension_reduction" "functional_enrichment"
#> [10] "get_anno_col" "get_anno" "get_classes"
#> [13] "get_consensus" "get_matrix" "get_membership"
#> [16] "get_param" "get_signatures" "get_stats"
#> [19] "is_best_k" "is_stable_k" "membership_heatmap"
#> [22] "ncol" "nrow" "plot_ecdf"
#> [25] "rownames" "select_partition_number" "show"
#> [28] "suggest_best_k" "test_to_known_factors"
collect_plots() function collects all the plots made from res for all k (number of partitions)
into one single page to provide an easy and fast comparison between different k.
collect_plots(res)
The plots are:
k and the heatmap of
predicted classes for each k.k.k.k.All the plots in panels can be made by individual functions and they are plotted later in this section.
select_partition_number() produces several plots showing different
statistics for choosing “optimized” k. There are following statistics:
k;k, the area increased is defined as \(A_k - A_{k-1}\).The detailed explanations of these statistics can be found in the cola vignette.
Generally speaking, lower PAC score, higher mean silhouette score or higher
concordance corresponds to better partition. Rand index and Jaccard index
measure how similar the current partition is compared to partition with k-1.
If they are too similar, we won't accept k is better than k-1.
select_partition_number(res)
The numeric values for all these statistics can be obtained by get_stats().
get_stats(res)
#> k 1-PAC mean_silhouette concordance area_increased Rand Jaccard
#> 2 2 0.474 0.814 0.906 0.4941 0.497 0.497
#> 3 3 0.415 0.506 0.690 0.2709 0.782 0.587
#> 4 4 0.819 0.858 0.917 0.1445 0.833 0.581
#> 5 5 0.881 0.875 0.928 0.0900 0.903 0.675
#> 6 6 0.816 0.752 0.826 0.0425 0.989 0.951
suggest_best_k() suggests the best \(k\) based on these statistics. The rules are as follows:
suggest_best_k(res)
#> [1] 5
Following shows the table of the partitions (You need to click the show/hide
code output link to see it). The membership matrix (columns with name p*)
is inferred by
clue::cl_consensus()
function with the SE method. Basically the value in the membership matrix
represents the probability to belong to a certain group. The finall class
label for an item is determined with the group with highest probability it
belongs to.
In get_classes() function, the entropy is calculated from the membership
matrix and the silhouette score is calculated from the consensus matrix.
cbind(get_classes(res, k = 2), get_membership(res, k = 2))
#> class entropy silhouette p1 p2
#> GSM614415 1 0.0376 0.8872 0.996 0.004
#> GSM614416 1 0.0376 0.8872 0.996 0.004
#> GSM614417 1 0.0376 0.8872 0.996 0.004
#> GSM614418 1 0.0376 0.8872 0.996 0.004
#> GSM614419 1 0.0376 0.8872 0.996 0.004
#> GSM614420 1 0.0376 0.8872 0.996 0.004
#> GSM614421 2 0.1843 0.8891 0.028 0.972
#> GSM614422 2 0.2778 0.8811 0.048 0.952
#> GSM614423 2 0.2603 0.8831 0.044 0.956
#> GSM614424 2 0.2043 0.8881 0.032 0.968
#> GSM614425 2 0.2423 0.8849 0.040 0.960
#> GSM614426 2 0.2236 0.8866 0.036 0.964
#> GSM614427 2 0.2603 0.8831 0.044 0.956
#> GSM614428 2 0.2043 0.8881 0.032 0.968
#> GSM614429 2 0.1184 0.8910 0.016 0.984
#> GSM614430 2 0.2778 0.8772 0.048 0.952
#> GSM614431 2 0.3274 0.8695 0.060 0.940
#> GSM614432 2 0.4298 0.8458 0.088 0.912
#> GSM614433 2 0.7883 0.6437 0.236 0.764
#> GSM614434 2 0.1184 0.8910 0.016 0.984
#> GSM614435 2 0.1184 0.8910 0.016 0.984
#> GSM614436 2 0.0938 0.8915 0.012 0.988
#> GSM614437 2 0.6048 0.8067 0.148 0.852
#> GSM614438 2 0.6048 0.8067 0.148 0.852
#> GSM614439 2 0.6048 0.8067 0.148 0.852
#> GSM614440 2 0.6048 0.8067 0.148 0.852
#> GSM614441 2 0.6048 0.8067 0.148 0.852
#> GSM614442 2 0.6048 0.8067 0.148 0.852
#> GSM614443 2 0.6048 0.8067 0.148 0.852
#> GSM614444 2 0.6048 0.8067 0.148 0.852
#> GSM614391 1 0.0000 0.8851 1.000 0.000
#> GSM614392 1 0.0000 0.8851 1.000 0.000
#> GSM614393 1 0.0000 0.8851 1.000 0.000
#> GSM614394 1 0.0000 0.8851 1.000 0.000
#> GSM614395 1 0.0000 0.8851 1.000 0.000
#> GSM614396 1 0.0000 0.8851 1.000 0.000
#> GSM614397 1 0.0000 0.8851 1.000 0.000
#> GSM614398 1 0.0000 0.8851 1.000 0.000
#> GSM614399 1 0.6712 0.8338 0.824 0.176
#> GSM614400 1 0.6048 0.8509 0.852 0.148
#> GSM614401 1 0.6712 0.8338 0.824 0.176
#> GSM614402 1 0.8608 0.6969 0.716 0.284
#> GSM614403 2 0.7139 0.7262 0.196 0.804
#> GSM614404 1 0.6623 0.8369 0.828 0.172
#> GSM614405 1 0.6531 0.8395 0.832 0.168
#> GSM614406 2 0.0000 0.8906 0.000 1.000
#> GSM614407 1 0.1184 0.8897 0.984 0.016
#> GSM614408 1 0.1184 0.8897 0.984 0.016
#> GSM614409 1 0.1184 0.8897 0.984 0.016
#> GSM614410 1 0.1184 0.8897 0.984 0.016
#> GSM614411 1 0.1184 0.8897 0.984 0.016
#> GSM614412 1 0.1184 0.8897 0.984 0.016
#> GSM614413 1 0.1633 0.8887 0.976 0.024
#> GSM614414 1 0.1184 0.8897 0.984 0.016
#> GSM614445 2 0.0376 0.8916 0.004 0.996
#> GSM614446 2 0.0376 0.8916 0.004 0.996
#> GSM614447 2 0.0376 0.8916 0.004 0.996
#> GSM614448 2 0.0376 0.8916 0.004 0.996
#> GSM614449 2 0.0376 0.8916 0.004 0.996
#> GSM614450 2 0.0672 0.8919 0.008 0.992
#> GSM614451 2 0.5059 0.8226 0.112 0.888
#> GSM614452 2 0.5519 0.8057 0.128 0.872
#> GSM614453 2 0.0376 0.8914 0.004 0.996
#> GSM614454 2 0.0376 0.8914 0.004 0.996
#> GSM614455 2 0.0376 0.8914 0.004 0.996
#> GSM614456 2 0.0376 0.8914 0.004 0.996
#> GSM614457 2 0.0376 0.8914 0.004 0.996
#> GSM614458 2 0.0000 0.8906 0.000 1.000
#> GSM614459 2 0.0376 0.8914 0.004 0.996
#> GSM614460 2 0.0376 0.8914 0.004 0.996
#> GSM614461 2 0.9954 0.0179 0.460 0.540
#> GSM614462 1 0.9815 0.3949 0.580 0.420
#> GSM614463 1 0.9922 0.3101 0.552 0.448
#> GSM614464 1 0.7602 0.7849 0.780 0.220
#> GSM614465 2 0.9248 0.4343 0.340 0.660
#> GSM614466 2 0.9635 0.2943 0.388 0.612
#> GSM614467 1 0.9922 0.3069 0.552 0.448
#> GSM614468 2 0.9850 0.1512 0.428 0.572
#> GSM614469 1 0.5842 0.8571 0.860 0.140
#> GSM614470 1 0.5842 0.8571 0.860 0.140
#> GSM614471 1 0.5842 0.8571 0.860 0.140
#> GSM614472 1 0.5842 0.8571 0.860 0.140
#> GSM614473 1 0.5842 0.8571 0.860 0.140
#> GSM614474 1 0.5842 0.8571 0.860 0.140
#> GSM614475 1 0.5842 0.8571 0.860 0.140
#> GSM614476 1 0.5842 0.8571 0.860 0.140
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM614415 1 0.1878 0.83812 0.952 0.004 0.044
#> GSM614416 1 0.1878 0.83812 0.952 0.004 0.044
#> GSM614417 1 0.1878 0.83812 0.952 0.004 0.044
#> GSM614418 1 0.1878 0.83812 0.952 0.004 0.044
#> GSM614419 1 0.4228 0.79326 0.844 0.008 0.148
#> GSM614420 1 0.4228 0.79326 0.844 0.008 0.148
#> GSM614421 2 0.2806 0.61890 0.040 0.928 0.032
#> GSM614422 2 0.1643 0.62679 0.044 0.956 0.000
#> GSM614423 2 0.2926 0.61585 0.040 0.924 0.036
#> GSM614424 2 0.1950 0.62729 0.040 0.952 0.008
#> GSM614425 2 0.2269 0.62535 0.040 0.944 0.016
#> GSM614426 2 0.2806 0.61890 0.040 0.928 0.032
#> GSM614427 2 0.2806 0.61890 0.040 0.928 0.032
#> GSM614428 2 0.2918 0.61647 0.044 0.924 0.032
#> GSM614429 2 0.7634 -0.00340 0.044 0.524 0.432
#> GSM614430 2 0.7715 -0.00223 0.048 0.524 0.428
#> GSM614431 2 0.7962 -0.02327 0.060 0.512 0.428
#> GSM614432 2 0.7814 -0.02011 0.052 0.512 0.436
#> GSM614433 2 0.8779 -0.10754 0.112 0.472 0.416
#> GSM614434 2 0.7627 0.00367 0.044 0.528 0.428
#> GSM614435 2 0.7883 -0.02712 0.056 0.516 0.428
#> GSM614436 2 0.7968 0.00157 0.068 0.560 0.372
#> GSM614437 3 0.5180 0.40483 0.032 0.156 0.812
#> GSM614438 3 0.5521 0.39946 0.032 0.180 0.788
#> GSM614439 3 0.5521 0.39946 0.032 0.180 0.788
#> GSM614440 3 0.5521 0.39946 0.032 0.180 0.788
#> GSM614441 3 0.5521 0.39946 0.032 0.180 0.788
#> GSM614442 3 0.5521 0.39946 0.032 0.180 0.788
#> GSM614443 3 0.5180 0.40483 0.032 0.156 0.812
#> GSM614444 3 0.5521 0.39946 0.032 0.180 0.788
#> GSM614391 1 0.2356 0.83240 0.928 0.000 0.072
#> GSM614392 1 0.2066 0.83573 0.940 0.000 0.060
#> GSM614393 1 0.2301 0.83489 0.936 0.004 0.060
#> GSM614394 1 0.4351 0.78170 0.828 0.004 0.168
#> GSM614395 1 0.6758 0.54839 0.620 0.020 0.360
#> GSM614396 1 0.4589 0.77686 0.820 0.008 0.172
#> GSM614397 1 0.5578 0.71011 0.748 0.012 0.240
#> GSM614398 1 0.4645 0.77325 0.816 0.008 0.176
#> GSM614399 1 0.7310 0.50651 0.628 0.324 0.048
#> GSM614400 1 0.7097 0.56938 0.668 0.280 0.052
#> GSM614401 1 0.7727 0.45857 0.600 0.336 0.064
#> GSM614402 2 0.7864 0.23513 0.332 0.596 0.072
#> GSM614403 2 0.4256 0.58664 0.096 0.868 0.036
#> GSM614404 1 0.7514 0.48599 0.616 0.328 0.056
#> GSM614405 1 0.7159 0.29896 0.528 0.448 0.024
#> GSM614406 2 0.4845 0.57235 0.104 0.844 0.052
#> GSM614407 1 0.1182 0.83837 0.976 0.012 0.012
#> GSM614408 1 0.0829 0.83930 0.984 0.012 0.004
#> GSM614409 1 0.0829 0.83984 0.984 0.012 0.004
#> GSM614410 1 0.1182 0.83837 0.976 0.012 0.012
#> GSM614411 1 0.0829 0.83984 0.984 0.012 0.004
#> GSM614412 1 0.0829 0.84039 0.984 0.012 0.004
#> GSM614413 1 0.3670 0.82120 0.888 0.020 0.092
#> GSM614414 1 0.3832 0.81756 0.880 0.020 0.100
#> GSM614445 2 0.1529 0.62822 0.040 0.960 0.000
#> GSM614446 2 0.1529 0.62822 0.040 0.960 0.000
#> GSM614447 2 0.1529 0.62822 0.040 0.960 0.000
#> GSM614448 2 0.1529 0.62822 0.040 0.960 0.000
#> GSM614449 2 0.1529 0.62822 0.040 0.960 0.000
#> GSM614450 2 0.1529 0.62822 0.040 0.960 0.000
#> GSM614451 2 0.3583 0.60008 0.044 0.900 0.056
#> GSM614452 2 0.3583 0.60008 0.044 0.900 0.056
#> GSM614453 3 0.9191 0.25041 0.148 0.420 0.432
#> GSM614454 3 0.9264 0.26706 0.156 0.412 0.432
#> GSM614455 3 0.9264 0.26706 0.156 0.412 0.432
#> GSM614456 3 0.9264 0.26706 0.156 0.412 0.432
#> GSM614457 3 0.9264 0.26706 0.156 0.412 0.432
#> GSM614458 2 0.8124 -0.10397 0.068 0.496 0.436
#> GSM614459 3 0.9264 0.26706 0.156 0.412 0.432
#> GSM614460 3 0.9264 0.26706 0.156 0.412 0.432
#> GSM614461 3 0.9515 0.16895 0.188 0.388 0.424
#> GSM614462 3 0.9823 0.22879 0.260 0.320 0.420
#> GSM614463 3 0.9860 0.22910 0.280 0.304 0.416
#> GSM614464 3 0.9830 0.23045 0.264 0.316 0.420
#> GSM614465 3 0.9264 0.11878 0.156 0.412 0.432
#> GSM614466 3 0.9464 0.13831 0.180 0.408 0.412
#> GSM614467 2 0.9024 -0.13004 0.132 0.448 0.420
#> GSM614468 2 0.9229 -0.16843 0.152 0.428 0.420
#> GSM614469 1 0.3148 0.82278 0.916 0.048 0.036
#> GSM614470 1 0.3148 0.82278 0.916 0.048 0.036
#> GSM614471 1 0.3148 0.82278 0.916 0.048 0.036
#> GSM614472 1 0.3148 0.82278 0.916 0.048 0.036
#> GSM614473 1 0.3148 0.82278 0.916 0.048 0.036
#> GSM614474 1 0.3148 0.82278 0.916 0.048 0.036
#> GSM614475 1 0.3148 0.82278 0.916 0.048 0.036
#> GSM614476 1 0.4539 0.77749 0.836 0.148 0.016
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM614415 1 0.1022 0.879 0.968 0.000 0.000 0.032
#> GSM614416 1 0.0921 0.879 0.972 0.000 0.000 0.028
#> GSM614417 1 0.1022 0.879 0.968 0.000 0.000 0.032
#> GSM614418 1 0.1022 0.879 0.968 0.000 0.000 0.032
#> GSM614419 1 0.1118 0.878 0.964 0.000 0.000 0.036
#> GSM614420 1 0.1118 0.878 0.964 0.000 0.000 0.036
#> GSM614421 3 0.0000 0.930 0.000 0.000 1.000 0.000
#> GSM614422 3 0.0000 0.930 0.000 0.000 1.000 0.000
#> GSM614423 3 0.0000 0.930 0.000 0.000 1.000 0.000
#> GSM614424 3 0.0000 0.930 0.000 0.000 1.000 0.000
#> GSM614425 3 0.0000 0.930 0.000 0.000 1.000 0.000
#> GSM614426 3 0.0000 0.930 0.000 0.000 1.000 0.000
#> GSM614427 3 0.0000 0.930 0.000 0.000 1.000 0.000
#> GSM614428 3 0.0000 0.930 0.000 0.000 1.000 0.000
#> GSM614429 2 0.1004 0.905 0.000 0.972 0.004 0.024
#> GSM614430 2 0.0469 0.906 0.000 0.988 0.000 0.012
#> GSM614431 2 0.0336 0.906 0.000 0.992 0.000 0.008
#> GSM614432 2 0.0524 0.906 0.000 0.988 0.004 0.008
#> GSM614433 2 0.0779 0.893 0.016 0.980 0.004 0.000
#> GSM614434 2 0.0469 0.906 0.000 0.988 0.000 0.012
#> GSM614435 2 0.2589 0.877 0.000 0.884 0.000 0.116
#> GSM614436 2 0.3157 0.866 0.000 0.852 0.004 0.144
#> GSM614437 4 0.1211 0.991 0.000 0.040 0.000 0.960
#> GSM614438 4 0.1209 0.998 0.000 0.032 0.004 0.964
#> GSM614439 4 0.1209 0.998 0.000 0.032 0.004 0.964
#> GSM614440 4 0.1209 0.998 0.000 0.032 0.004 0.964
#> GSM614441 4 0.1209 0.998 0.000 0.032 0.004 0.964
#> GSM614442 4 0.1209 0.998 0.000 0.032 0.004 0.964
#> GSM614443 4 0.1118 0.994 0.000 0.036 0.000 0.964
#> GSM614444 4 0.1209 0.998 0.000 0.032 0.004 0.964
#> GSM614391 1 0.1022 0.879 0.968 0.000 0.000 0.032
#> GSM614392 1 0.1022 0.879 0.968 0.000 0.000 0.032
#> GSM614393 1 0.1118 0.878 0.964 0.000 0.000 0.036
#> GSM614394 1 0.1118 0.878 0.964 0.000 0.000 0.036
#> GSM614395 1 0.1211 0.876 0.960 0.000 0.000 0.040
#> GSM614396 1 0.1118 0.878 0.964 0.000 0.000 0.036
#> GSM614397 1 0.1211 0.876 0.960 0.000 0.000 0.040
#> GSM614398 1 0.1118 0.878 0.964 0.000 0.000 0.036
#> GSM614399 1 0.6725 0.569 0.620 0.136 0.240 0.004
#> GSM614400 1 0.7241 0.355 0.520 0.140 0.336 0.004
#> GSM614401 3 0.6216 0.601 0.188 0.128 0.680 0.004
#> GSM614402 3 0.4961 0.732 0.096 0.116 0.784 0.004
#> GSM614403 3 0.0967 0.915 0.016 0.004 0.976 0.004
#> GSM614404 3 0.7136 0.263 0.328 0.132 0.536 0.004
#> GSM614405 1 0.6369 0.162 0.500 0.052 0.444 0.004
#> GSM614406 1 0.5308 0.215 0.540 0.004 0.452 0.004
#> GSM614407 1 0.1474 0.879 0.948 0.052 0.000 0.000
#> GSM614408 1 0.1661 0.880 0.944 0.052 0.000 0.004
#> GSM614409 1 0.1474 0.879 0.948 0.052 0.000 0.000
#> GSM614410 1 0.1474 0.879 0.948 0.052 0.000 0.000
#> GSM614411 1 0.1474 0.879 0.948 0.052 0.000 0.000
#> GSM614412 1 0.1302 0.880 0.956 0.044 0.000 0.000
#> GSM614413 1 0.0921 0.881 0.972 0.028 0.000 0.000
#> GSM614414 1 0.0921 0.881 0.972 0.028 0.000 0.000
#> GSM614445 3 0.0000 0.930 0.000 0.000 1.000 0.000
#> GSM614446 3 0.0000 0.930 0.000 0.000 1.000 0.000
#> GSM614447 3 0.0000 0.930 0.000 0.000 1.000 0.000
#> GSM614448 3 0.0000 0.930 0.000 0.000 1.000 0.000
#> GSM614449 3 0.0000 0.930 0.000 0.000 1.000 0.000
#> GSM614450 3 0.0000 0.930 0.000 0.000 1.000 0.000
#> GSM614451 3 0.1488 0.902 0.032 0.000 0.956 0.012
#> GSM614452 3 0.1488 0.902 0.032 0.000 0.956 0.012
#> GSM614453 2 0.3356 0.848 0.000 0.824 0.000 0.176
#> GSM614454 2 0.3356 0.848 0.000 0.824 0.000 0.176
#> GSM614455 2 0.3400 0.845 0.000 0.820 0.000 0.180
#> GSM614456 2 0.3400 0.845 0.000 0.820 0.000 0.180
#> GSM614457 2 0.3400 0.845 0.000 0.820 0.000 0.180
#> GSM614458 2 0.2921 0.865 0.000 0.860 0.000 0.140
#> GSM614459 2 0.3400 0.845 0.000 0.820 0.000 0.180
#> GSM614460 2 0.3400 0.845 0.000 0.820 0.000 0.180
#> GSM614461 2 0.0188 0.904 0.000 0.996 0.004 0.000
#> GSM614462 2 0.0376 0.902 0.004 0.992 0.004 0.000
#> GSM614463 2 0.0524 0.900 0.008 0.988 0.004 0.000
#> GSM614464 2 0.0524 0.900 0.008 0.988 0.004 0.000
#> GSM614465 2 0.0188 0.904 0.000 0.996 0.004 0.000
#> GSM614466 2 0.0524 0.900 0.008 0.988 0.004 0.000
#> GSM614467 2 0.0992 0.900 0.012 0.976 0.004 0.008
#> GSM614468 2 0.0376 0.902 0.004 0.992 0.004 0.000
#> GSM614469 1 0.3157 0.842 0.852 0.144 0.004 0.000
#> GSM614470 1 0.3157 0.842 0.852 0.144 0.004 0.000
#> GSM614471 1 0.3157 0.842 0.852 0.144 0.004 0.000
#> GSM614472 1 0.3157 0.842 0.852 0.144 0.004 0.000
#> GSM614473 1 0.3157 0.842 0.852 0.144 0.004 0.000
#> GSM614474 1 0.3157 0.842 0.852 0.144 0.004 0.000
#> GSM614475 1 0.3157 0.842 0.852 0.144 0.004 0.000
#> GSM614476 1 0.4356 0.818 0.812 0.124 0.064 0.000
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM614415 5 0.2179 0.888 0.112 0.000 0.000 0.000 0.888
#> GSM614416 5 0.2648 0.849 0.152 0.000 0.000 0.000 0.848
#> GSM614417 5 0.2852 0.830 0.172 0.000 0.000 0.000 0.828
#> GSM614418 5 0.2929 0.818 0.180 0.000 0.000 0.000 0.820
#> GSM614419 5 0.0609 0.934 0.020 0.000 0.000 0.000 0.980
#> GSM614420 5 0.0609 0.934 0.020 0.000 0.000 0.000 0.980
#> GSM614421 3 0.0000 0.918 0.000 0.000 1.000 0.000 0.000
#> GSM614422 3 0.0000 0.918 0.000 0.000 1.000 0.000 0.000
#> GSM614423 3 0.0000 0.918 0.000 0.000 1.000 0.000 0.000
#> GSM614424 3 0.0000 0.918 0.000 0.000 1.000 0.000 0.000
#> GSM614425 3 0.0000 0.918 0.000 0.000 1.000 0.000 0.000
#> GSM614426 3 0.0000 0.918 0.000 0.000 1.000 0.000 0.000
#> GSM614427 3 0.0000 0.918 0.000 0.000 1.000 0.000 0.000
#> GSM614428 3 0.0000 0.918 0.000 0.000 1.000 0.000 0.000
#> GSM614429 2 0.0162 0.959 0.000 0.996 0.000 0.000 0.004
#> GSM614430 2 0.0000 0.959 0.000 1.000 0.000 0.000 0.000
#> GSM614431 2 0.0000 0.959 0.000 1.000 0.000 0.000 0.000
#> GSM614432 2 0.0000 0.959 0.000 1.000 0.000 0.000 0.000
#> GSM614433 2 0.1121 0.951 0.044 0.956 0.000 0.000 0.000
#> GSM614434 2 0.0000 0.959 0.000 1.000 0.000 0.000 0.000
#> GSM614435 2 0.0613 0.958 0.004 0.984 0.000 0.008 0.004
#> GSM614436 2 0.0613 0.958 0.004 0.984 0.000 0.008 0.004
#> GSM614437 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000
#> GSM614438 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000
#> GSM614439 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000
#> GSM614440 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000
#> GSM614441 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000
#> GSM614442 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000
#> GSM614443 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000
#> GSM614444 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000
#> GSM614391 5 0.1341 0.923 0.056 0.000 0.000 0.000 0.944
#> GSM614392 5 0.1544 0.919 0.068 0.000 0.000 0.000 0.932
#> GSM614393 5 0.0404 0.934 0.012 0.000 0.000 0.000 0.988
#> GSM614394 5 0.0290 0.932 0.008 0.000 0.000 0.000 0.992
#> GSM614395 5 0.0510 0.934 0.016 0.000 0.000 0.000 0.984
#> GSM614396 5 0.0290 0.932 0.008 0.000 0.000 0.000 0.992
#> GSM614397 5 0.0510 0.934 0.016 0.000 0.000 0.000 0.984
#> GSM614398 5 0.0510 0.934 0.016 0.000 0.000 0.000 0.984
#> GSM614399 1 0.3437 0.690 0.808 0.012 0.176 0.004 0.000
#> GSM614400 1 0.4127 0.446 0.680 0.008 0.312 0.000 0.000
#> GSM614401 3 0.3132 0.769 0.172 0.008 0.820 0.000 0.000
#> GSM614402 3 0.1788 0.875 0.056 0.008 0.932 0.004 0.000
#> GSM614403 3 0.0451 0.912 0.008 0.000 0.988 0.004 0.000
#> GSM614404 3 0.4670 0.257 0.440 0.008 0.548 0.004 0.000
#> GSM614405 3 0.5172 0.322 0.380 0.008 0.580 0.000 0.032
#> GSM614406 3 0.5128 0.316 0.380 0.000 0.580 0.004 0.036
#> GSM614407 1 0.2648 0.818 0.848 0.000 0.000 0.000 0.152
#> GSM614408 1 0.2732 0.816 0.840 0.000 0.000 0.000 0.160
#> GSM614409 1 0.3074 0.797 0.804 0.000 0.000 0.000 0.196
#> GSM614410 1 0.3143 0.792 0.796 0.000 0.000 0.000 0.204
#> GSM614411 1 0.2852 0.810 0.828 0.000 0.000 0.000 0.172
#> GSM614412 1 0.3612 0.731 0.732 0.000 0.000 0.000 0.268
#> GSM614413 1 0.4182 0.555 0.600 0.000 0.000 0.000 0.400
#> GSM614414 1 0.4182 0.555 0.600 0.000 0.000 0.000 0.400
#> GSM614445 3 0.0000 0.918 0.000 0.000 1.000 0.000 0.000
#> GSM614446 3 0.0000 0.918 0.000 0.000 1.000 0.000 0.000
#> GSM614447 3 0.0000 0.918 0.000 0.000 1.000 0.000 0.000
#> GSM614448 3 0.0000 0.918 0.000 0.000 1.000 0.000 0.000
#> GSM614449 3 0.0000 0.918 0.000 0.000 1.000 0.000 0.000
#> GSM614450 3 0.0000 0.918 0.000 0.000 1.000 0.000 0.000
#> GSM614451 3 0.0451 0.912 0.004 0.000 0.988 0.000 0.008
#> GSM614452 3 0.0451 0.912 0.004 0.000 0.988 0.000 0.008
#> GSM614453 2 0.1483 0.952 0.028 0.952 0.000 0.012 0.008
#> GSM614454 2 0.1483 0.952 0.028 0.952 0.000 0.012 0.008
#> GSM614455 2 0.1483 0.952 0.028 0.952 0.000 0.012 0.008
#> GSM614456 2 0.1483 0.952 0.028 0.952 0.000 0.012 0.008
#> GSM614457 2 0.1483 0.952 0.028 0.952 0.000 0.012 0.008
#> GSM614458 2 0.0613 0.959 0.004 0.984 0.000 0.004 0.008
#> GSM614459 2 0.1588 0.950 0.028 0.948 0.000 0.016 0.008
#> GSM614460 2 0.1483 0.952 0.028 0.952 0.000 0.012 0.008
#> GSM614461 2 0.1544 0.937 0.068 0.932 0.000 0.000 0.000
#> GSM614462 2 0.1792 0.924 0.084 0.916 0.000 0.000 0.000
#> GSM614463 2 0.1908 0.916 0.092 0.908 0.000 0.000 0.000
#> GSM614464 2 0.1544 0.937 0.068 0.932 0.000 0.000 0.000
#> GSM614465 2 0.1197 0.948 0.048 0.952 0.000 0.000 0.000
#> GSM614466 2 0.1270 0.947 0.052 0.948 0.000 0.000 0.000
#> GSM614467 2 0.1197 0.948 0.048 0.952 0.000 0.000 0.000
#> GSM614468 2 0.1197 0.948 0.048 0.952 0.000 0.000 0.000
#> GSM614469 1 0.0794 0.836 0.972 0.000 0.000 0.000 0.028
#> GSM614470 1 0.0794 0.836 0.972 0.000 0.000 0.000 0.028
#> GSM614471 1 0.0794 0.836 0.972 0.000 0.000 0.000 0.028
#> GSM614472 1 0.0794 0.836 0.972 0.000 0.000 0.000 0.028
#> GSM614473 1 0.0794 0.836 0.972 0.000 0.000 0.000 0.028
#> GSM614474 1 0.0794 0.836 0.972 0.000 0.000 0.000 0.028
#> GSM614475 1 0.0794 0.836 0.972 0.000 0.000 0.000 0.028
#> GSM614476 1 0.4197 0.781 0.776 0.000 0.076 0.000 0.148
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM614415 5 0.3327 0.858 0.088 0.000 0.000 0.000 0.820 NA
#> GSM614416 5 0.3472 0.845 0.100 0.000 0.000 0.000 0.808 NA
#> GSM614417 5 0.3277 0.861 0.084 0.000 0.000 0.000 0.824 NA
#> GSM614418 5 0.3277 0.861 0.084 0.000 0.000 0.000 0.824 NA
#> GSM614419 5 0.0622 0.887 0.012 0.000 0.000 0.000 0.980 NA
#> GSM614420 5 0.0725 0.887 0.012 0.000 0.000 0.000 0.976 NA
#> GSM614421 3 0.0000 0.888 0.000 0.000 1.000 0.000 0.000 NA
#> GSM614422 3 0.0000 0.888 0.000 0.000 1.000 0.000 0.000 NA
#> GSM614423 3 0.0000 0.888 0.000 0.000 1.000 0.000 0.000 NA
#> GSM614424 3 0.0000 0.888 0.000 0.000 1.000 0.000 0.000 NA
#> GSM614425 3 0.0000 0.888 0.000 0.000 1.000 0.000 0.000 NA
#> GSM614426 3 0.0000 0.888 0.000 0.000 1.000 0.000 0.000 NA
#> GSM614427 3 0.0363 0.883 0.000 0.000 0.988 0.000 0.000 NA
#> GSM614428 3 0.0146 0.887 0.000 0.000 0.996 0.000 0.000 NA
#> GSM614429 2 0.1007 0.814 0.000 0.956 0.000 0.000 0.000 NA
#> GSM614430 2 0.0260 0.813 0.000 0.992 0.000 0.000 0.000 NA
#> GSM614431 2 0.0603 0.815 0.004 0.980 0.000 0.000 0.000 NA
#> GSM614432 2 0.0458 0.813 0.000 0.984 0.000 0.000 0.000 NA
#> GSM614433 2 0.1327 0.808 0.000 0.936 0.000 0.000 0.000 NA
#> GSM614434 2 0.0146 0.814 0.000 0.996 0.000 0.000 0.000 NA
#> GSM614435 2 0.2734 0.797 0.008 0.840 0.004 0.000 0.000 NA
#> GSM614436 2 0.3301 0.785 0.008 0.772 0.004 0.000 0.000 NA
#> GSM614437 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000 NA
#> GSM614438 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000 NA
#> GSM614439 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000 NA
#> GSM614440 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000 NA
#> GSM614441 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000 NA
#> GSM614442 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000 NA
#> GSM614443 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000 NA
#> GSM614444 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000 NA
#> GSM614391 5 0.3172 0.865 0.076 0.000 0.000 0.000 0.832 NA
#> GSM614392 5 0.3123 0.866 0.076 0.000 0.000 0.000 0.836 NA
#> GSM614393 5 0.1895 0.885 0.016 0.000 0.000 0.000 0.912 NA
#> GSM614394 5 0.0260 0.885 0.000 0.000 0.000 0.000 0.992 NA
#> GSM614395 5 0.1387 0.856 0.000 0.000 0.000 0.000 0.932 NA
#> GSM614396 5 0.0260 0.885 0.000 0.000 0.000 0.000 0.992 NA
#> GSM614397 5 0.0790 0.879 0.000 0.000 0.000 0.000 0.968 NA
#> GSM614398 5 0.0458 0.884 0.000 0.000 0.000 0.000 0.984 NA
#> GSM614399 1 0.5868 0.473 0.592 0.040 0.140 0.000 0.000 NA
#> GSM614400 1 0.6304 0.309 0.468 0.024 0.200 0.000 0.000 NA
#> GSM614401 3 0.6326 0.222 0.236 0.016 0.432 0.000 0.000 NA
#> GSM614402 3 0.5528 0.497 0.116 0.016 0.580 0.000 0.000 NA
#> GSM614403 3 0.2357 0.805 0.012 0.000 0.872 0.000 0.000 NA
#> GSM614404 1 0.6467 0.168 0.400 0.020 0.256 0.000 0.000 NA
#> GSM614405 3 0.6324 0.193 0.320 0.000 0.480 0.000 0.036 NA
#> GSM614406 3 0.7229 -0.136 0.348 0.000 0.368 0.008 0.080 NA
#> GSM614407 1 0.4954 0.565 0.640 0.000 0.000 0.000 0.232 NA
#> GSM614408 1 0.4989 0.551 0.628 0.000 0.000 0.000 0.252 NA
#> GSM614409 1 0.5420 0.515 0.572 0.000 0.000 0.000 0.256 NA
#> GSM614410 1 0.4947 0.560 0.636 0.000 0.000 0.000 0.244 NA
#> GSM614411 1 0.5258 0.532 0.596 0.000 0.000 0.000 0.252 NA
#> GSM614412 1 0.5514 0.495 0.552 0.000 0.000 0.000 0.272 NA
#> GSM614413 1 0.5534 0.366 0.444 0.000 0.000 0.000 0.424 NA
#> GSM614414 1 0.5534 0.366 0.444 0.000 0.000 0.000 0.424 NA
#> GSM614445 3 0.0000 0.888 0.000 0.000 1.000 0.000 0.000 NA
#> GSM614446 3 0.0000 0.888 0.000 0.000 1.000 0.000 0.000 NA
#> GSM614447 3 0.0000 0.888 0.000 0.000 1.000 0.000 0.000 NA
#> GSM614448 3 0.0000 0.888 0.000 0.000 1.000 0.000 0.000 NA
#> GSM614449 3 0.0000 0.888 0.000 0.000 1.000 0.000 0.000 NA
#> GSM614450 3 0.0000 0.888 0.000 0.000 1.000 0.000 0.000 NA
#> GSM614451 3 0.0405 0.884 0.000 0.000 0.988 0.000 0.004 NA
#> GSM614452 3 0.0405 0.884 0.000 0.000 0.988 0.000 0.004 NA
#> GSM614453 2 0.3789 0.729 0.008 0.660 0.000 0.000 0.000 NA
#> GSM614454 2 0.3741 0.734 0.008 0.672 0.000 0.000 0.000 NA
#> GSM614455 2 0.3819 0.724 0.008 0.652 0.000 0.000 0.000 NA
#> GSM614456 2 0.3847 0.719 0.008 0.644 0.000 0.000 0.000 NA
#> GSM614457 2 0.3847 0.719 0.008 0.644 0.000 0.000 0.000 NA
#> GSM614458 2 0.2402 0.799 0.000 0.856 0.004 0.000 0.000 NA
#> GSM614459 2 0.3861 0.720 0.008 0.640 0.000 0.000 0.000 NA
#> GSM614460 2 0.3847 0.719 0.008 0.644 0.000 0.000 0.000 NA
#> GSM614461 2 0.3259 0.741 0.012 0.772 0.000 0.000 0.000 NA
#> GSM614462 2 0.3841 0.700 0.028 0.716 0.000 0.000 0.000 NA
#> GSM614463 2 0.3933 0.702 0.036 0.716 0.000 0.000 0.000 NA
#> GSM614464 2 0.2848 0.766 0.008 0.816 0.000 0.000 0.000 NA
#> GSM614465 2 0.2743 0.772 0.008 0.828 0.000 0.000 0.000 NA
#> GSM614466 2 0.2871 0.759 0.004 0.804 0.000 0.000 0.000 NA
#> GSM614467 2 0.2340 0.788 0.000 0.852 0.000 0.000 0.000 NA
#> GSM614468 2 0.2048 0.791 0.000 0.880 0.000 0.000 0.000 NA
#> GSM614469 1 0.0405 0.692 0.988 0.000 0.008 0.000 0.000 NA
#> GSM614470 1 0.0260 0.692 0.992 0.000 0.008 0.000 0.000 NA
#> GSM614471 1 0.0405 0.692 0.988 0.000 0.008 0.000 0.000 NA
#> GSM614472 1 0.0260 0.692 0.992 0.000 0.008 0.000 0.000 NA
#> GSM614473 1 0.0622 0.691 0.980 0.000 0.008 0.000 0.000 NA
#> GSM614474 1 0.0405 0.693 0.988 0.000 0.008 0.000 0.000 NA
#> GSM614475 1 0.1265 0.689 0.948 0.000 0.008 0.000 0.000 NA
#> GSM614476 1 0.6404 0.576 0.572 0.000 0.140 0.000 0.116 NA
Heatmaps for the consensus matrix. It visualizes the probability of two samples to be in a same group.
consensus_heatmap(res, k = 2)
consensus_heatmap(res, k = 3)
consensus_heatmap(res, k = 4)
consensus_heatmap(res, k = 5)
consensus_heatmap(res, k = 6)
Heatmaps for the membership of samples in all partitions to see how consistent they are:
membership_heatmap(res, k = 2)
membership_heatmap(res, k = 3)
membership_heatmap(res, k = 4)
membership_heatmap(res, k = 5)
membership_heatmap(res, k = 6)
As soon as we have had the classes for columns, we can look for signatures which are significantly different between classes which can be candidate marks for certain classes. Following are the heatmaps for signatures.
Signature heatmaps where rows are scaled:
get_signatures(res, k = 2)
get_signatures(res, k = 3)
get_signatures(res, k = 4)
get_signatures(res, k = 5)
get_signatures(res, k = 6)
Signature heatmaps where rows are not scaled:
get_signatures(res, k = 2, scale_rows = FALSE)
get_signatures(res, k = 3, scale_rows = FALSE)
get_signatures(res, k = 4, scale_rows = FALSE)
get_signatures(res, k = 5, scale_rows = FALSE)
get_signatures(res, k = 6, scale_rows = FALSE)
Compare the overlap of signatures from different k:
compare_signatures(res)
get_signature() returns a data frame invisibly. TO get the list of signatures, the function
call should be assigned to a variable explicitly. In following code, if plot argument is set
to FALSE, no heatmap is plotted while only the differential analysis is performed.
# code only for demonstration
tb = get_signature(res, k = ..., plot = FALSE)
An example of the output of tb is:
#> which_row fdr mean_1 mean_2 scaled_mean_1 scaled_mean_2 km
#> 1 38 0.042760348 8.373488 9.131774 -0.5533452 0.5164555 1
#> 2 40 0.018707592 7.106213 8.469186 -0.6173731 0.5762149 1
#> 3 55 0.019134737 10.221463 11.207825 -0.6159697 0.5749050 1
#> 4 59 0.006059896 5.921854 7.869574 -0.6899429 0.6439467 1
#> 5 60 0.018055526 8.928898 10.211722 -0.6204761 0.5791110 1
#> 6 98 0.009384629 15.714769 14.887706 0.6635654 -0.6193277 2
...
The columns in tb are:
which_row: row indices corresponding to the input matrix.fdr: FDR for the differential test. mean_x: The mean value in group x.scaled_mean_x: The mean value in group x after rows are scaled.km: Row groups if k-means clustering is applied to rows.UMAP plot which shows how samples are separated.
dimension_reduction(res, k = 2, method = "UMAP")
dimension_reduction(res, k = 3, method = "UMAP")
dimension_reduction(res, k = 4, method = "UMAP")
dimension_reduction(res, k = 5, method = "UMAP")
dimension_reduction(res, k = 6, method = "UMAP")
Following heatmap shows how subgroups are split when increasing k:
collect_classes(res)
Test correlation between subgroups and known annotations. If the known annotation is numeric, one-way ANOVA test is applied, and if the known annotation is discrete, chi-squared contingency table test is applied.
test_to_known_factors(res)
#> n individual(p) protocol(p) time(p) other(p) k
#> SD:mclust 79 1.18e-11 0.617 0.983 0.8740 2
#> SD:mclust 50 3.46e-08 0.701 0.990 0.5247 3
#> SD:mclust 82 9.05e-35 0.959 1.000 0.0135 4
#> SD:mclust 82 1.38e-45 0.953 1.000 0.0239 5
#> SD:mclust 76 2.60e-42 0.910 1.000 0.0577 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 21168 rows and 86 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 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 0.467 0.781 0.888 0.4730 0.508 0.508
#> 3 3 0.765 0.860 0.934 0.3381 0.647 0.423
#> 4 4 0.684 0.717 0.867 0.1630 0.749 0.425
#> 5 5 0.660 0.646 0.794 0.0668 0.887 0.608
#> 6 6 0.757 0.678 0.828 0.0391 0.880 0.529
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
#> GSM614415 1 0.0000 0.902 1.000 0.000
#> GSM614416 1 0.0000 0.902 1.000 0.000
#> GSM614417 1 0.0000 0.902 1.000 0.000
#> GSM614418 1 0.0000 0.902 1.000 0.000
#> GSM614419 1 0.0000 0.902 1.000 0.000
#> GSM614420 1 0.0000 0.902 1.000 0.000
#> GSM614421 2 0.4022 0.835 0.080 0.920
#> GSM614422 1 0.1184 0.895 0.984 0.016
#> GSM614423 1 0.9393 0.294 0.644 0.356
#> GSM614424 2 0.4022 0.837 0.080 0.920
#> GSM614425 2 0.9522 0.459 0.372 0.628
#> GSM614426 2 0.9732 0.364 0.404 0.596
#> GSM614427 2 0.0376 0.837 0.004 0.996
#> GSM614428 2 0.0376 0.837 0.004 0.996
#> GSM614429 2 0.5629 0.834 0.132 0.868
#> GSM614430 2 0.6623 0.824 0.172 0.828
#> GSM614431 2 0.7528 0.806 0.216 0.784
#> GSM614432 2 0.7528 0.806 0.216 0.784
#> GSM614433 2 0.7602 0.804 0.220 0.780
#> GSM614434 2 0.7299 0.813 0.204 0.796
#> GSM614435 2 0.1414 0.841 0.020 0.980
#> GSM614436 2 0.0000 0.837 0.000 1.000
#> GSM614437 2 0.0000 0.837 0.000 1.000
#> GSM614438 2 0.0376 0.837 0.004 0.996
#> GSM614439 2 0.0376 0.837 0.004 0.996
#> GSM614440 2 0.0376 0.837 0.004 0.996
#> GSM614441 2 0.0376 0.837 0.004 0.996
#> GSM614442 2 0.0376 0.837 0.004 0.996
#> GSM614443 2 0.0000 0.837 0.000 1.000
#> GSM614444 2 0.0376 0.837 0.004 0.996
#> GSM614391 1 0.0000 0.902 1.000 0.000
#> GSM614392 1 0.0000 0.902 1.000 0.000
#> GSM614393 1 0.0000 0.902 1.000 0.000
#> GSM614394 1 0.0000 0.902 1.000 0.000
#> GSM614395 1 0.7528 0.668 0.784 0.216
#> GSM614396 1 0.0000 0.902 1.000 0.000
#> GSM614397 1 0.5408 0.779 0.876 0.124
#> GSM614398 1 0.0672 0.897 0.992 0.008
#> GSM614399 2 0.8861 0.716 0.304 0.696
#> GSM614400 1 0.9661 0.173 0.608 0.392
#> GSM614401 1 0.5294 0.785 0.880 0.120
#> GSM614402 1 0.9993 -0.212 0.516 0.484
#> GSM614403 2 0.9323 0.645 0.348 0.652
#> GSM614404 1 0.9996 -0.227 0.512 0.488
#> GSM614405 2 0.8608 0.742 0.284 0.716
#> GSM614406 2 0.0376 0.837 0.004 0.996
#> GSM614407 1 0.0376 0.901 0.996 0.004
#> GSM614408 1 0.0376 0.901 0.996 0.004
#> GSM614409 1 0.0000 0.902 1.000 0.000
#> GSM614410 1 0.0376 0.901 0.996 0.004
#> GSM614411 1 0.0000 0.902 1.000 0.000
#> GSM614412 1 0.0000 0.902 1.000 0.000
#> GSM614413 1 0.0672 0.897 0.992 0.008
#> GSM614414 1 0.0376 0.900 0.996 0.004
#> GSM614445 2 0.9833 0.484 0.424 0.576
#> GSM614446 2 0.9000 0.702 0.316 0.684
#> GSM614447 2 0.9087 0.686 0.324 0.676
#> GSM614448 2 0.1633 0.834 0.024 0.976
#> GSM614449 2 0.2043 0.839 0.032 0.968
#> GSM614450 2 0.9129 0.682 0.328 0.672
#> GSM614451 2 0.0376 0.837 0.004 0.996
#> GSM614452 2 0.0376 0.837 0.004 0.996
#> GSM614453 2 0.7299 0.813 0.204 0.796
#> GSM614454 2 0.6887 0.821 0.184 0.816
#> GSM614455 2 0.6801 0.822 0.180 0.820
#> GSM614456 2 0.0672 0.840 0.008 0.992
#> GSM614457 2 0.0376 0.839 0.004 0.996
#> GSM614458 2 0.3879 0.842 0.076 0.924
#> GSM614459 2 0.0000 0.837 0.000 1.000
#> GSM614460 2 0.2236 0.843 0.036 0.964
#> GSM614461 2 0.7528 0.806 0.216 0.784
#> GSM614462 2 0.7815 0.795 0.232 0.768
#> GSM614463 2 0.8016 0.784 0.244 0.756
#> GSM614464 2 0.7453 0.809 0.212 0.788
#> GSM614465 2 0.7950 0.788 0.240 0.760
#> GSM614466 2 0.7815 0.795 0.232 0.768
#> GSM614467 2 0.3733 0.843 0.072 0.928
#> GSM614468 2 0.7528 0.806 0.216 0.784
#> GSM614469 1 0.0376 0.901 0.996 0.004
#> GSM614470 1 0.0376 0.901 0.996 0.004
#> GSM614471 1 0.0376 0.901 0.996 0.004
#> GSM614472 1 0.0376 0.901 0.996 0.004
#> GSM614473 1 0.0376 0.901 0.996 0.004
#> GSM614474 1 0.0376 0.901 0.996 0.004
#> GSM614475 1 0.5519 0.774 0.872 0.128
#> GSM614476 1 0.8661 0.478 0.712 0.288
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM614415 1 0.0000 0.941075 1.000 0.000 0.000
#> GSM614416 1 0.0000 0.941075 1.000 0.000 0.000
#> GSM614417 1 0.0000 0.941075 1.000 0.000 0.000
#> GSM614418 1 0.0000 0.941075 1.000 0.000 0.000
#> GSM614419 1 0.0237 0.940652 0.996 0.000 0.004
#> GSM614420 1 0.0237 0.940652 0.996 0.000 0.004
#> GSM614421 3 0.7106 0.707511 0.076 0.224 0.700
#> GSM614422 1 0.3682 0.805584 0.876 0.116 0.008
#> GSM614423 2 0.1860 0.911708 0.052 0.948 0.000
#> GSM614424 3 0.6765 0.729817 0.068 0.208 0.724
#> GSM614425 3 0.9601 0.144282 0.392 0.200 0.408
#> GSM614426 1 0.9585 0.000206 0.456 0.212 0.332
#> GSM614427 3 0.4514 0.804008 0.012 0.156 0.832
#> GSM614428 3 0.1015 0.874994 0.012 0.008 0.980
#> GSM614429 2 0.1031 0.926300 0.000 0.976 0.024
#> GSM614430 2 0.0747 0.930067 0.000 0.984 0.016
#> GSM614431 2 0.0000 0.935564 0.000 1.000 0.000
#> GSM614432 2 0.0000 0.935564 0.000 1.000 0.000
#> GSM614433 2 0.0000 0.935564 0.000 1.000 0.000
#> GSM614434 2 0.0237 0.934799 0.000 0.996 0.004
#> GSM614435 2 0.2261 0.896515 0.000 0.932 0.068
#> GSM614436 3 0.3879 0.814569 0.000 0.152 0.848
#> GSM614437 3 0.0424 0.881041 0.000 0.008 0.992
#> GSM614438 3 0.0237 0.882337 0.000 0.004 0.996
#> GSM614439 3 0.0237 0.882337 0.000 0.004 0.996
#> GSM614440 3 0.0237 0.882337 0.000 0.004 0.996
#> GSM614441 3 0.0237 0.882337 0.000 0.004 0.996
#> GSM614442 3 0.0237 0.882337 0.000 0.004 0.996
#> GSM614443 3 0.0237 0.882337 0.000 0.004 0.996
#> GSM614444 3 0.0237 0.882337 0.000 0.004 0.996
#> GSM614391 1 0.0237 0.940652 0.996 0.000 0.004
#> GSM614392 1 0.0000 0.941075 1.000 0.000 0.000
#> GSM614393 1 0.0000 0.941075 1.000 0.000 0.000
#> GSM614394 1 0.0237 0.940652 0.996 0.000 0.004
#> GSM614395 1 0.2878 0.850002 0.904 0.000 0.096
#> GSM614396 1 0.0237 0.940652 0.996 0.000 0.004
#> GSM614397 1 0.0592 0.935170 0.988 0.000 0.012
#> GSM614398 1 0.0237 0.940652 0.996 0.000 0.004
#> GSM614399 2 0.0237 0.935048 0.004 0.996 0.000
#> GSM614400 2 0.0237 0.935048 0.004 0.996 0.000
#> GSM614401 2 0.0424 0.934106 0.008 0.992 0.000
#> GSM614402 2 0.0237 0.935048 0.004 0.996 0.000
#> GSM614403 2 0.0892 0.929667 0.020 0.980 0.000
#> GSM614404 2 0.0237 0.935048 0.004 0.996 0.000
#> GSM614405 2 0.2689 0.907597 0.036 0.932 0.032
#> GSM614406 3 0.0000 0.880785 0.000 0.000 1.000
#> GSM614407 1 0.1031 0.922883 0.976 0.024 0.000
#> GSM614408 1 0.0592 0.933090 0.988 0.012 0.000
#> GSM614409 1 0.0000 0.941075 1.000 0.000 0.000
#> GSM614410 1 0.0592 0.933090 0.988 0.012 0.000
#> GSM614411 1 0.0000 0.941075 1.000 0.000 0.000
#> GSM614412 1 0.0000 0.941075 1.000 0.000 0.000
#> GSM614413 1 0.0237 0.940652 0.996 0.000 0.004
#> GSM614414 1 0.0237 0.940652 0.996 0.000 0.004
#> GSM614445 2 0.0237 0.935048 0.004 0.996 0.000
#> GSM614446 2 0.0000 0.935564 0.000 1.000 0.000
#> GSM614447 2 0.0237 0.935048 0.004 0.996 0.000
#> GSM614448 3 0.5138 0.809667 0.052 0.120 0.828
#> GSM614449 3 0.5706 0.585087 0.000 0.320 0.680
#> GSM614450 2 0.5947 0.724415 0.052 0.776 0.172
#> GSM614451 3 0.0000 0.880785 0.000 0.000 1.000
#> GSM614452 3 0.0000 0.880785 0.000 0.000 1.000
#> GSM614453 2 0.0000 0.935564 0.000 1.000 0.000
#> GSM614454 2 0.0237 0.934799 0.000 0.996 0.004
#> GSM614455 2 0.0237 0.934799 0.000 0.996 0.004
#> GSM614456 2 0.4291 0.779654 0.000 0.820 0.180
#> GSM614457 2 0.4974 0.699599 0.000 0.764 0.236
#> GSM614458 2 0.1031 0.926248 0.000 0.976 0.024
#> GSM614459 2 0.6286 0.157704 0.000 0.536 0.464
#> GSM614460 2 0.3619 0.831808 0.000 0.864 0.136
#> GSM614461 2 0.0000 0.935564 0.000 1.000 0.000
#> GSM614462 2 0.0000 0.935564 0.000 1.000 0.000
#> GSM614463 2 0.0000 0.935564 0.000 1.000 0.000
#> GSM614464 2 0.0237 0.934799 0.000 0.996 0.004
#> GSM614465 2 0.0000 0.935564 0.000 1.000 0.000
#> GSM614466 2 0.0000 0.935564 0.000 1.000 0.000
#> GSM614467 2 0.2356 0.893217 0.000 0.928 0.072
#> GSM614468 2 0.0000 0.935564 0.000 1.000 0.000
#> GSM614469 2 0.3038 0.864630 0.104 0.896 0.000
#> GSM614470 2 0.3551 0.836006 0.132 0.868 0.000
#> GSM614471 2 0.1163 0.925404 0.028 0.972 0.000
#> GSM614472 2 0.1964 0.906771 0.056 0.944 0.000
#> GSM614473 2 0.5327 0.647369 0.272 0.728 0.000
#> GSM614474 2 0.3941 0.808897 0.156 0.844 0.000
#> GSM614475 2 0.1289 0.923537 0.032 0.968 0.000
#> GSM614476 1 0.6126 0.428500 0.644 0.352 0.004
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM614415 1 0.0336 0.9065 0.992 0.000 0.008 0.000
#> GSM614416 1 0.0188 0.9061 0.996 0.000 0.004 0.000
#> GSM614417 1 0.0188 0.9061 0.996 0.000 0.004 0.000
#> GSM614418 1 0.0188 0.9061 0.996 0.000 0.004 0.000
#> GSM614419 1 0.0336 0.9065 0.992 0.000 0.008 0.000
#> GSM614420 1 0.0336 0.9065 0.992 0.000 0.008 0.000
#> GSM614421 3 0.2216 0.8213 0.000 0.092 0.908 0.000
#> GSM614422 3 0.2216 0.8210 0.000 0.092 0.908 0.000
#> GSM614423 3 0.4679 0.5638 0.000 0.352 0.648 0.000
#> GSM614424 3 0.2345 0.8198 0.000 0.100 0.900 0.000
#> GSM614425 3 0.2216 0.8213 0.000 0.092 0.908 0.000
#> GSM614426 3 0.2281 0.8208 0.000 0.096 0.904 0.000
#> GSM614427 3 0.1824 0.8100 0.000 0.060 0.936 0.004
#> GSM614428 3 0.0707 0.7656 0.000 0.000 0.980 0.020
#> GSM614429 2 0.1305 0.8032 0.000 0.960 0.004 0.036
#> GSM614430 2 0.1109 0.8089 0.000 0.968 0.004 0.028
#> GSM614431 2 0.0188 0.8190 0.000 0.996 0.004 0.000
#> GSM614432 2 0.0188 0.8190 0.000 0.996 0.004 0.000
#> GSM614433 2 0.0707 0.8209 0.000 0.980 0.020 0.000
#> GSM614434 2 0.1004 0.8106 0.000 0.972 0.004 0.024
#> GSM614435 2 0.4252 0.5541 0.000 0.744 0.004 0.252
#> GSM614436 4 0.3793 0.8030 0.000 0.112 0.044 0.844
#> GSM614437 4 0.0672 0.8288 0.000 0.008 0.008 0.984
#> GSM614438 4 0.2408 0.8204 0.000 0.000 0.104 0.896
#> GSM614439 4 0.2469 0.8180 0.000 0.000 0.108 0.892
#> GSM614440 4 0.2408 0.8204 0.000 0.000 0.104 0.896
#> GSM614441 4 0.2469 0.8181 0.000 0.000 0.108 0.892
#> GSM614442 4 0.2081 0.8246 0.000 0.000 0.084 0.916
#> GSM614443 4 0.0524 0.8289 0.000 0.004 0.008 0.988
#> GSM614444 4 0.2408 0.8204 0.000 0.000 0.104 0.896
#> GSM614391 1 0.0336 0.9065 0.992 0.000 0.008 0.000
#> GSM614392 1 0.0336 0.9065 0.992 0.000 0.008 0.000
#> GSM614393 1 0.0336 0.9065 0.992 0.000 0.008 0.000
#> GSM614394 1 0.0336 0.9065 0.992 0.000 0.008 0.000
#> GSM614395 3 0.5594 -0.0358 0.460 0.000 0.520 0.020
#> GSM614396 1 0.0336 0.9065 0.992 0.000 0.008 0.000
#> GSM614397 1 0.2469 0.8341 0.892 0.000 0.108 0.000
#> GSM614398 1 0.0921 0.8976 0.972 0.000 0.028 0.000
#> GSM614399 2 0.0921 0.8177 0.000 0.972 0.028 0.000
#> GSM614400 2 0.0592 0.8209 0.000 0.984 0.016 0.000
#> GSM614401 2 0.0817 0.8192 0.000 0.976 0.024 0.000
#> GSM614402 2 0.1637 0.7985 0.000 0.940 0.060 0.000
#> GSM614403 2 0.4972 -0.0872 0.000 0.544 0.456 0.000
#> GSM614404 2 0.0707 0.8203 0.000 0.980 0.020 0.000
#> GSM614405 2 0.4730 0.2707 0.000 0.636 0.364 0.000
#> GSM614406 3 0.3873 0.5482 0.000 0.000 0.772 0.228
#> GSM614407 1 0.0817 0.9020 0.976 0.000 0.024 0.000
#> GSM614408 1 0.0817 0.9020 0.976 0.000 0.024 0.000
#> GSM614409 1 0.1022 0.8997 0.968 0.000 0.032 0.000
#> GSM614410 1 0.0817 0.9020 0.976 0.000 0.024 0.000
#> GSM614411 1 0.0921 0.9010 0.972 0.000 0.028 0.000
#> GSM614412 1 0.1022 0.8997 0.968 0.000 0.032 0.000
#> GSM614413 1 0.4624 0.5143 0.660 0.000 0.340 0.000
#> GSM614414 1 0.2760 0.8280 0.872 0.000 0.128 0.000
#> GSM614445 3 0.4941 0.3719 0.000 0.436 0.564 0.000
#> GSM614446 3 0.4250 0.6752 0.000 0.276 0.724 0.000
#> GSM614447 3 0.4776 0.5144 0.000 0.376 0.624 0.000
#> GSM614448 3 0.1635 0.8025 0.008 0.044 0.948 0.000
#> GSM614449 3 0.2530 0.8148 0.000 0.112 0.888 0.000
#> GSM614450 3 0.3486 0.7649 0.000 0.188 0.812 0.000
#> GSM614451 3 0.2011 0.7248 0.000 0.000 0.920 0.080
#> GSM614452 3 0.1637 0.7406 0.000 0.000 0.940 0.060
#> GSM614453 2 0.4661 0.3555 0.000 0.652 0.000 0.348
#> GSM614454 4 0.4989 0.1570 0.000 0.472 0.000 0.528
#> GSM614455 4 0.4994 0.1315 0.000 0.480 0.000 0.520
#> GSM614456 4 0.2973 0.7857 0.000 0.144 0.000 0.856
#> GSM614457 4 0.2345 0.8129 0.000 0.100 0.000 0.900
#> GSM614458 2 0.5112 0.0911 0.000 0.560 0.004 0.436
#> GSM614459 4 0.1716 0.8231 0.000 0.064 0.000 0.936
#> GSM614460 4 0.2868 0.7934 0.000 0.136 0.000 0.864
#> GSM614461 2 0.0469 0.8205 0.000 0.988 0.012 0.000
#> GSM614462 2 0.0817 0.8199 0.000 0.976 0.024 0.000
#> GSM614463 2 0.0592 0.8208 0.000 0.984 0.016 0.000
#> GSM614464 2 0.1022 0.8168 0.000 0.968 0.032 0.000
#> GSM614465 2 0.1716 0.7968 0.000 0.936 0.064 0.000
#> GSM614466 2 0.0707 0.8205 0.000 0.980 0.020 0.000
#> GSM614467 2 0.4040 0.5494 0.000 0.752 0.248 0.000
#> GSM614468 2 0.1867 0.7902 0.000 0.928 0.072 0.000
#> GSM614469 1 0.4836 0.5180 0.672 0.320 0.008 0.000
#> GSM614470 1 0.4647 0.5807 0.704 0.288 0.008 0.000
#> GSM614471 2 0.4511 0.5639 0.268 0.724 0.008 0.000
#> GSM614472 2 0.5112 0.3214 0.384 0.608 0.008 0.000
#> GSM614473 1 0.3725 0.7438 0.812 0.180 0.008 0.000
#> GSM614474 1 0.5310 0.2808 0.576 0.412 0.012 0.000
#> GSM614475 2 0.3105 0.7194 0.140 0.856 0.004 0.000
#> GSM614476 2 0.7186 0.0616 0.420 0.444 0.136 0.000
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM614415 5 0.3074 0.76816 0.196 0.000 0.000 0.000 0.804
#> GSM614416 5 0.3074 0.76816 0.196 0.000 0.000 0.000 0.804
#> GSM614417 5 0.3074 0.76816 0.196 0.000 0.000 0.000 0.804
#> GSM614418 5 0.3109 0.76306 0.200 0.000 0.000 0.000 0.800
#> GSM614419 5 0.2966 0.77466 0.184 0.000 0.000 0.000 0.816
#> GSM614420 5 0.2966 0.77466 0.184 0.000 0.000 0.000 0.816
#> GSM614421 3 0.1549 0.83171 0.016 0.040 0.944 0.000 0.000
#> GSM614422 3 0.1444 0.83240 0.012 0.040 0.948 0.000 0.000
#> GSM614423 3 0.3961 0.73558 0.028 0.212 0.760 0.000 0.000
#> GSM614424 3 0.1121 0.83400 0.000 0.044 0.956 0.000 0.000
#> GSM614425 3 0.1408 0.83322 0.008 0.044 0.948 0.000 0.000
#> GSM614426 3 0.1282 0.83389 0.000 0.044 0.952 0.000 0.004
#> GSM614427 3 0.1026 0.82697 0.004 0.024 0.968 0.004 0.000
#> GSM614428 3 0.0566 0.80874 0.004 0.000 0.984 0.012 0.000
#> GSM614429 2 0.1012 0.76202 0.012 0.968 0.000 0.020 0.000
#> GSM614430 2 0.0771 0.76575 0.020 0.976 0.000 0.004 0.000
#> GSM614431 2 0.0290 0.76895 0.008 0.992 0.000 0.000 0.000
#> GSM614432 2 0.0798 0.76971 0.016 0.976 0.008 0.000 0.000
#> GSM614433 2 0.0963 0.77204 0.000 0.964 0.036 0.000 0.000
#> GSM614434 2 0.0798 0.76532 0.016 0.976 0.000 0.008 0.000
#> GSM614435 2 0.4104 0.54956 0.032 0.748 0.000 0.220 0.000
#> GSM614436 4 0.5061 0.60017 0.020 0.312 0.024 0.644 0.000
#> GSM614437 4 0.0404 0.83729 0.000 0.012 0.000 0.988 0.000
#> GSM614438 4 0.1341 0.84093 0.000 0.000 0.056 0.944 0.000
#> GSM614439 4 0.1544 0.83489 0.000 0.000 0.068 0.932 0.000
#> GSM614440 4 0.1478 0.83771 0.000 0.000 0.064 0.936 0.000
#> GSM614441 4 0.1478 0.83782 0.000 0.000 0.064 0.936 0.000
#> GSM614442 4 0.1121 0.84173 0.000 0.000 0.044 0.956 0.000
#> GSM614443 4 0.0451 0.83875 0.000 0.008 0.004 0.988 0.000
#> GSM614444 4 0.1341 0.84093 0.000 0.000 0.056 0.944 0.000
#> GSM614391 5 0.0000 0.81231 0.000 0.000 0.000 0.000 1.000
#> GSM614392 5 0.0000 0.81231 0.000 0.000 0.000 0.000 1.000
#> GSM614393 5 0.0000 0.81231 0.000 0.000 0.000 0.000 1.000
#> GSM614394 5 0.0162 0.81131 0.000 0.000 0.004 0.000 0.996
#> GSM614395 5 0.4088 0.44163 0.004 0.000 0.276 0.008 0.712
#> GSM614396 5 0.0162 0.81131 0.000 0.000 0.004 0.000 0.996
#> GSM614397 5 0.1831 0.73490 0.004 0.000 0.076 0.000 0.920
#> GSM614398 5 0.0671 0.79920 0.004 0.000 0.016 0.000 0.980
#> GSM614399 2 0.4058 0.63429 0.236 0.740 0.024 0.000 0.000
#> GSM614400 2 0.4647 0.50531 0.352 0.628 0.016 0.000 0.004
#> GSM614401 2 0.5057 0.38822 0.412 0.556 0.028 0.000 0.004
#> GSM614402 2 0.4924 0.58294 0.272 0.668 0.060 0.000 0.000
#> GSM614403 3 0.6646 0.24745 0.196 0.356 0.444 0.000 0.004
#> GSM614404 2 0.4642 0.53438 0.328 0.648 0.020 0.000 0.004
#> GSM614405 2 0.7109 -0.00176 0.248 0.404 0.332 0.000 0.016
#> GSM614406 3 0.5449 0.53037 0.072 0.004 0.656 0.260 0.008
#> GSM614407 1 0.3561 0.50964 0.740 0.000 0.000 0.000 0.260
#> GSM614408 1 0.3586 0.50081 0.736 0.000 0.000 0.000 0.264
#> GSM614409 1 0.4039 0.51120 0.720 0.004 0.008 0.000 0.268
#> GSM614410 1 0.3612 0.50942 0.732 0.000 0.000 0.000 0.268
#> GSM614411 1 0.3814 0.50798 0.720 0.004 0.000 0.000 0.276
#> GSM614412 1 0.4063 0.50327 0.708 0.000 0.012 0.000 0.280
#> GSM614413 1 0.5032 0.47971 0.692 0.004 0.076 0.000 0.228
#> GSM614414 1 0.4503 0.49457 0.696 0.000 0.036 0.000 0.268
#> GSM614445 3 0.5886 0.55535 0.144 0.272 0.584 0.000 0.000
#> GSM614446 3 0.5027 0.70597 0.112 0.188 0.700 0.000 0.000
#> GSM614447 3 0.5680 0.62219 0.148 0.228 0.624 0.000 0.000
#> GSM614448 3 0.1121 0.82679 0.008 0.016 0.968 0.004 0.004
#> GSM614449 3 0.2527 0.82607 0.020 0.072 0.900 0.004 0.004
#> GSM614450 3 0.4153 0.78619 0.076 0.116 0.800 0.004 0.004
#> GSM614451 3 0.1557 0.78765 0.000 0.000 0.940 0.052 0.008
#> GSM614452 3 0.1408 0.79223 0.000 0.000 0.948 0.044 0.008
#> GSM614453 2 0.3521 0.55948 0.004 0.764 0.000 0.232 0.000
#> GSM614454 2 0.4151 0.35205 0.004 0.652 0.000 0.344 0.000
#> GSM614455 2 0.3932 0.40253 0.000 0.672 0.000 0.328 0.000
#> GSM614456 4 0.3837 0.62984 0.000 0.308 0.000 0.692 0.000
#> GSM614457 4 0.3274 0.74163 0.000 0.220 0.000 0.780 0.000
#> GSM614458 2 0.3707 0.47489 0.000 0.716 0.000 0.284 0.000
#> GSM614459 4 0.2773 0.78418 0.000 0.164 0.000 0.836 0.000
#> GSM614460 4 0.3949 0.65129 0.004 0.300 0.000 0.696 0.000
#> GSM614461 2 0.0693 0.77278 0.008 0.980 0.012 0.000 0.000
#> GSM614462 2 0.1568 0.77082 0.020 0.944 0.036 0.000 0.000
#> GSM614463 2 0.1018 0.77277 0.016 0.968 0.016 0.000 0.000
#> GSM614464 2 0.1981 0.76559 0.028 0.924 0.048 0.000 0.000
#> GSM614465 2 0.2344 0.75788 0.032 0.904 0.064 0.000 0.000
#> GSM614466 2 0.1568 0.77082 0.020 0.944 0.036 0.000 0.000
#> GSM614467 2 0.3210 0.62544 0.000 0.788 0.212 0.000 0.000
#> GSM614468 2 0.1768 0.76119 0.004 0.924 0.072 0.000 0.000
#> GSM614469 1 0.6308 0.25873 0.484 0.164 0.000 0.000 0.352
#> GSM614470 1 0.6186 0.25630 0.512 0.152 0.000 0.000 0.336
#> GSM614471 1 0.6381 0.06309 0.448 0.384 0.000 0.000 0.168
#> GSM614472 1 0.6438 0.30959 0.500 0.280 0.000 0.000 0.220
#> GSM614473 1 0.5687 0.07771 0.484 0.080 0.000 0.000 0.436
#> GSM614474 1 0.6268 0.26108 0.484 0.156 0.000 0.000 0.360
#> GSM614475 2 0.5555 0.45312 0.220 0.640 0.000 0.000 0.140
#> GSM614476 1 0.8081 0.27825 0.412 0.200 0.128 0.000 0.260
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM614415 5 0.4535 0.3342 0.484 0.000 0.000 0.000 0.484 0.032
#> GSM614416 1 0.4535 -0.4278 0.484 0.000 0.000 0.000 0.484 0.032
#> GSM614417 1 0.4535 -0.4278 0.484 0.000 0.000 0.000 0.484 0.032
#> GSM614418 5 0.4535 0.3342 0.484 0.000 0.000 0.000 0.484 0.032
#> GSM614419 5 0.4601 0.3552 0.472 0.000 0.004 0.000 0.496 0.028
#> GSM614420 5 0.4601 0.3552 0.472 0.000 0.004 0.000 0.496 0.028
#> GSM614421 3 0.1151 0.8187 0.000 0.012 0.956 0.000 0.000 0.032
#> GSM614422 3 0.1723 0.8132 0.004 0.012 0.932 0.000 0.004 0.048
#> GSM614423 3 0.3516 0.7588 0.056 0.076 0.832 0.000 0.000 0.036
#> GSM614424 3 0.0909 0.8194 0.000 0.012 0.968 0.000 0.000 0.020
#> GSM614425 3 0.1151 0.8187 0.000 0.012 0.956 0.000 0.000 0.032
#> GSM614426 3 0.0964 0.8200 0.004 0.012 0.968 0.000 0.000 0.016
#> GSM614427 3 0.1067 0.8189 0.000 0.004 0.964 0.004 0.004 0.024
#> GSM614428 3 0.2119 0.8007 0.000 0.000 0.912 0.044 0.008 0.036
#> GSM614429 2 0.1080 0.8188 0.004 0.960 0.032 0.000 0.000 0.004
#> GSM614430 2 0.1155 0.8193 0.004 0.956 0.036 0.000 0.000 0.004
#> GSM614431 2 0.1226 0.8195 0.004 0.952 0.040 0.000 0.000 0.004
#> GSM614432 2 0.1493 0.8182 0.004 0.936 0.056 0.000 0.000 0.004
#> GSM614433 2 0.1753 0.8126 0.004 0.912 0.084 0.000 0.000 0.000
#> GSM614434 2 0.1226 0.8195 0.004 0.952 0.040 0.000 0.000 0.004
#> GSM614435 2 0.2123 0.7979 0.000 0.908 0.008 0.064 0.000 0.020
#> GSM614436 2 0.5020 0.4733 0.004 0.608 0.024 0.328 0.000 0.036
#> GSM614437 4 0.1493 0.9212 0.004 0.056 0.000 0.936 0.000 0.004
#> GSM614438 4 0.0458 0.9687 0.000 0.000 0.016 0.984 0.000 0.000
#> GSM614439 4 0.0547 0.9662 0.000 0.000 0.020 0.980 0.000 0.000
#> GSM614440 4 0.0547 0.9662 0.000 0.000 0.020 0.980 0.000 0.000
#> GSM614441 4 0.0458 0.9687 0.000 0.000 0.016 0.984 0.000 0.000
#> GSM614442 4 0.0551 0.9627 0.000 0.004 0.008 0.984 0.000 0.004
#> GSM614443 4 0.1555 0.9175 0.004 0.060 0.000 0.932 0.000 0.004
#> GSM614444 4 0.0458 0.9687 0.000 0.000 0.016 0.984 0.000 0.000
#> GSM614391 5 0.0458 0.7721 0.000 0.000 0.000 0.000 0.984 0.016
#> GSM614392 5 0.0458 0.7721 0.000 0.000 0.000 0.000 0.984 0.016
#> GSM614393 5 0.0458 0.7721 0.000 0.000 0.000 0.000 0.984 0.016
#> GSM614394 5 0.0458 0.7721 0.000 0.000 0.000 0.000 0.984 0.016
#> GSM614395 5 0.1793 0.7324 0.000 0.000 0.048 0.012 0.928 0.012
#> GSM614396 5 0.0458 0.7721 0.000 0.000 0.000 0.000 0.984 0.016
#> GSM614397 5 0.0725 0.7662 0.000 0.000 0.012 0.000 0.976 0.012
#> GSM614398 5 0.0717 0.7684 0.000 0.000 0.008 0.000 0.976 0.016
#> GSM614399 1 0.4951 0.3707 0.568 0.364 0.064 0.000 0.000 0.004
#> GSM614400 1 0.3900 0.6066 0.760 0.188 0.044 0.000 0.000 0.008
#> GSM614401 1 0.3453 0.6209 0.808 0.144 0.040 0.000 0.000 0.008
#> GSM614402 1 0.4980 0.4988 0.624 0.280 0.092 0.000 0.000 0.004
#> GSM614403 1 0.5493 0.1146 0.520 0.120 0.356 0.000 0.000 0.004
#> GSM614404 1 0.4364 0.5601 0.688 0.256 0.052 0.000 0.000 0.004
#> GSM614405 1 0.5768 0.2819 0.544 0.172 0.276 0.000 0.004 0.004
#> GSM614406 3 0.6426 0.2540 0.268 0.000 0.392 0.324 0.016 0.000
#> GSM614407 6 0.1644 0.9506 0.052 0.004 0.000 0.000 0.012 0.932
#> GSM614408 6 0.1863 0.9415 0.060 0.004 0.000 0.000 0.016 0.920
#> GSM614409 6 0.0820 0.9646 0.012 0.000 0.000 0.000 0.016 0.972
#> GSM614410 6 0.1391 0.9585 0.040 0.000 0.000 0.000 0.016 0.944
#> GSM614411 6 0.1003 0.9646 0.020 0.000 0.000 0.000 0.016 0.964
#> GSM614412 6 0.0748 0.9609 0.004 0.000 0.004 0.000 0.016 0.976
#> GSM614413 6 0.1341 0.9403 0.000 0.000 0.028 0.000 0.024 0.948
#> GSM614414 6 0.1168 0.9489 0.000 0.000 0.016 0.000 0.028 0.956
#> GSM614445 3 0.5534 0.2985 0.360 0.124 0.512 0.000 0.000 0.004
#> GSM614446 3 0.4353 0.6171 0.244 0.056 0.696 0.000 0.000 0.004
#> GSM614447 3 0.4989 0.4652 0.328 0.076 0.592 0.000 0.000 0.004
#> GSM614448 3 0.2100 0.8090 0.024 0.000 0.916 0.048 0.004 0.008
#> GSM614449 3 0.1649 0.8126 0.040 0.000 0.936 0.016 0.000 0.008
#> GSM614450 3 0.3273 0.7243 0.180 0.008 0.800 0.008 0.000 0.004
#> GSM614451 3 0.2520 0.7739 0.000 0.000 0.872 0.108 0.008 0.012
#> GSM614452 3 0.2473 0.7767 0.000 0.000 0.876 0.104 0.008 0.012
#> GSM614453 2 0.1606 0.7989 0.004 0.932 0.000 0.056 0.000 0.008
#> GSM614454 2 0.1843 0.7914 0.004 0.912 0.000 0.080 0.000 0.004
#> GSM614455 2 0.1845 0.7939 0.004 0.916 0.000 0.072 0.000 0.008
#> GSM614456 2 0.3452 0.6203 0.004 0.736 0.000 0.256 0.000 0.004
#> GSM614457 2 0.3930 0.4436 0.004 0.628 0.000 0.364 0.000 0.004
#> GSM614458 2 0.2009 0.7881 0.004 0.904 0.000 0.084 0.000 0.008
#> GSM614459 2 0.4124 0.1511 0.004 0.516 0.000 0.476 0.000 0.004
#> GSM614460 2 0.3646 0.5671 0.004 0.700 0.000 0.292 0.000 0.004
#> GSM614461 2 0.1838 0.8135 0.016 0.916 0.068 0.000 0.000 0.000
#> GSM614462 2 0.2106 0.8060 0.032 0.904 0.064 0.000 0.000 0.000
#> GSM614463 2 0.1984 0.8093 0.032 0.912 0.056 0.000 0.000 0.000
#> GSM614464 2 0.2526 0.7949 0.024 0.876 0.096 0.000 0.004 0.000
#> GSM614465 2 0.2476 0.7970 0.024 0.880 0.092 0.000 0.004 0.000
#> GSM614466 2 0.2176 0.8048 0.024 0.896 0.080 0.000 0.000 0.000
#> GSM614467 2 0.3230 0.6985 0.012 0.776 0.212 0.000 0.000 0.000
#> GSM614468 2 0.2408 0.7989 0.012 0.876 0.108 0.000 0.000 0.004
#> GSM614469 1 0.3915 0.5244 0.792 0.012 0.008 0.000 0.052 0.136
#> GSM614470 1 0.2557 0.5848 0.892 0.012 0.004 0.000 0.036 0.056
#> GSM614471 1 0.2952 0.6073 0.872 0.064 0.008 0.000 0.016 0.040
#> GSM614472 1 0.2364 0.5923 0.904 0.016 0.008 0.000 0.016 0.056
#> GSM614473 1 0.3091 0.5615 0.856 0.012 0.004 0.000 0.044 0.084
#> GSM614474 1 0.4378 0.5079 0.752 0.016 0.008 0.000 0.060 0.164
#> GSM614475 2 0.5676 -0.0626 0.412 0.500 0.040 0.000 0.016 0.032
#> GSM614476 1 0.4256 0.5838 0.780 0.024 0.140 0.004 0.012 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 individual(p) protocol(p) time(p) other(p) k
#> SD:NMF 78 2.47e-11 0.2689 0.978 0.843 2
#> SD:NMF 82 6.82e-18 0.0321 0.978 0.362 3
#> SD:NMF 75 1.77e-27 0.3813 1.000 0.175 4
#> SD:NMF 69 7.42e-35 0.2641 1.000 0.304 5
#> SD:NMF 69 7.17e-48 0.9206 1.000 0.063 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 21168 rows and 86 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 0.535 0.908 0.931 0.2521 0.774 0.774
#> 3 3 0.663 0.847 0.928 0.7440 0.773 0.708
#> 4 4 0.670 0.776 0.886 0.1281 0.996 0.992
#> 5 5 0.729 0.881 0.926 0.1924 0.882 0.785
#> 6 6 0.755 0.930 0.949 0.0423 0.987 0.970
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
#> GSM614415 2 0.6048 0.858 0.148 0.852
#> GSM614416 2 0.6048 0.858 0.148 0.852
#> GSM614417 2 0.6048 0.858 0.148 0.852
#> GSM614418 2 0.6048 0.858 0.148 0.852
#> GSM614419 2 0.6048 0.858 0.148 0.852
#> GSM614420 2 0.6048 0.858 0.148 0.852
#> GSM614421 2 0.2043 0.922 0.032 0.968
#> GSM614422 2 0.2043 0.922 0.032 0.968
#> GSM614423 2 0.2043 0.922 0.032 0.968
#> GSM614424 2 0.2043 0.922 0.032 0.968
#> GSM614425 2 0.2043 0.922 0.032 0.968
#> GSM614426 2 0.2043 0.922 0.032 0.968
#> GSM614427 2 0.2043 0.922 0.032 0.968
#> GSM614428 2 0.2043 0.922 0.032 0.968
#> GSM614429 2 0.0672 0.936 0.008 0.992
#> GSM614430 2 0.0672 0.936 0.008 0.992
#> GSM614431 2 0.0672 0.936 0.008 0.992
#> GSM614432 2 0.0672 0.936 0.008 0.992
#> GSM614433 2 0.0672 0.936 0.008 0.992
#> GSM614434 2 0.0672 0.936 0.008 0.992
#> GSM614435 2 0.0672 0.936 0.008 0.992
#> GSM614436 2 0.0672 0.936 0.008 0.992
#> GSM614437 1 0.6048 0.952 0.852 0.148
#> GSM614438 1 0.6048 0.952 0.852 0.148
#> GSM614439 1 0.6048 0.952 0.852 0.148
#> GSM614440 1 0.6048 0.952 0.852 0.148
#> GSM614441 1 0.6048 0.952 0.852 0.148
#> GSM614442 1 0.6048 0.952 0.852 0.148
#> GSM614443 1 0.6048 0.952 0.852 0.148
#> GSM614444 1 0.6048 0.952 0.852 0.148
#> GSM614391 2 0.6048 0.858 0.148 0.852
#> GSM614392 2 0.6048 0.858 0.148 0.852
#> GSM614393 2 0.6048 0.858 0.148 0.852
#> GSM614394 2 0.6048 0.858 0.148 0.852
#> GSM614395 1 0.9795 0.177 0.584 0.416
#> GSM614396 2 0.6048 0.858 0.148 0.852
#> GSM614397 2 0.6148 0.855 0.152 0.848
#> GSM614398 2 0.6048 0.858 0.148 0.852
#> GSM614399 2 0.0000 0.936 0.000 1.000
#> GSM614400 2 0.0000 0.936 0.000 1.000
#> GSM614401 2 0.0000 0.936 0.000 1.000
#> GSM614402 2 0.0000 0.936 0.000 1.000
#> GSM614403 2 0.0000 0.936 0.000 1.000
#> GSM614404 2 0.0000 0.936 0.000 1.000
#> GSM614405 2 0.0000 0.936 0.000 1.000
#> GSM614406 2 0.0000 0.936 0.000 1.000
#> GSM614407 2 0.5842 0.864 0.140 0.860
#> GSM614408 2 0.5842 0.864 0.140 0.860
#> GSM614409 2 0.5842 0.864 0.140 0.860
#> GSM614410 2 0.5842 0.864 0.140 0.860
#> GSM614411 2 0.5842 0.864 0.140 0.860
#> GSM614412 2 0.5842 0.864 0.140 0.860
#> GSM614413 2 0.5842 0.864 0.140 0.860
#> GSM614414 2 0.5842 0.864 0.140 0.860
#> GSM614445 2 0.2603 0.913 0.044 0.956
#> GSM614446 2 0.2603 0.913 0.044 0.956
#> GSM614447 2 0.2603 0.913 0.044 0.956
#> GSM614448 2 0.2603 0.913 0.044 0.956
#> GSM614449 2 0.2603 0.913 0.044 0.956
#> GSM614450 2 0.2603 0.913 0.044 0.956
#> GSM614451 1 0.6048 0.952 0.852 0.148
#> GSM614452 1 0.6048 0.952 0.852 0.148
#> GSM614453 2 0.0938 0.935 0.012 0.988
#> GSM614454 2 0.0938 0.935 0.012 0.988
#> GSM614455 2 0.0938 0.935 0.012 0.988
#> GSM614456 2 0.0938 0.935 0.012 0.988
#> GSM614457 2 0.0938 0.935 0.012 0.988
#> GSM614458 2 0.0938 0.935 0.012 0.988
#> GSM614459 2 0.0938 0.935 0.012 0.988
#> GSM614460 2 0.0938 0.935 0.012 0.988
#> GSM614461 2 0.0672 0.936 0.008 0.992
#> GSM614462 2 0.0672 0.936 0.008 0.992
#> GSM614463 2 0.0672 0.936 0.008 0.992
#> GSM614464 2 0.0672 0.936 0.008 0.992
#> GSM614465 2 0.0672 0.936 0.008 0.992
#> GSM614466 2 0.0672 0.936 0.008 0.992
#> GSM614467 2 0.0672 0.936 0.008 0.992
#> GSM614468 2 0.0672 0.936 0.008 0.992
#> GSM614469 2 0.0672 0.937 0.008 0.992
#> GSM614470 2 0.0672 0.937 0.008 0.992
#> GSM614471 2 0.0672 0.937 0.008 0.992
#> GSM614472 2 0.0672 0.937 0.008 0.992
#> GSM614473 2 0.0672 0.937 0.008 0.992
#> GSM614474 2 0.0672 0.937 0.008 0.992
#> GSM614475 2 0.0672 0.937 0.008 0.992
#> GSM614476 2 0.0672 0.937 0.008 0.992
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM614415 1 0.6180 0.4959 0.584 0.416 0.000
#> GSM614416 1 0.6180 0.4959 0.584 0.416 0.000
#> GSM614417 1 0.6180 0.4959 0.584 0.416 0.000
#> GSM614418 1 0.6180 0.4959 0.584 0.416 0.000
#> GSM614419 1 0.6180 0.4959 0.584 0.416 0.000
#> GSM614420 1 0.6180 0.4959 0.584 0.416 0.000
#> GSM614421 2 0.2682 0.9003 0.004 0.920 0.076
#> GSM614422 2 0.2682 0.9003 0.004 0.920 0.076
#> GSM614423 2 0.2682 0.9003 0.004 0.920 0.076
#> GSM614424 2 0.2682 0.9003 0.004 0.920 0.076
#> GSM614425 2 0.2682 0.9003 0.004 0.920 0.076
#> GSM614426 2 0.2682 0.9003 0.004 0.920 0.076
#> GSM614427 2 0.2682 0.9003 0.004 0.920 0.076
#> GSM614428 2 0.2682 0.9003 0.004 0.920 0.076
#> GSM614429 2 0.0237 0.9371 0.004 0.996 0.000
#> GSM614430 2 0.0237 0.9371 0.004 0.996 0.000
#> GSM614431 2 0.0237 0.9371 0.004 0.996 0.000
#> GSM614432 2 0.0237 0.9371 0.004 0.996 0.000
#> GSM614433 2 0.0237 0.9371 0.004 0.996 0.000
#> GSM614434 2 0.0237 0.9371 0.004 0.996 0.000
#> GSM614435 2 0.0237 0.9371 0.004 0.996 0.000
#> GSM614436 2 0.0237 0.9371 0.004 0.996 0.000
#> GSM614437 3 0.0237 0.9985 0.000 0.004 0.996
#> GSM614438 3 0.0237 0.9985 0.000 0.004 0.996
#> GSM614439 3 0.0237 0.9985 0.000 0.004 0.996
#> GSM614440 3 0.0237 0.9985 0.000 0.004 0.996
#> GSM614441 3 0.0237 0.9985 0.000 0.004 0.996
#> GSM614442 3 0.0237 0.9985 0.000 0.004 0.996
#> GSM614443 3 0.0237 0.9985 0.000 0.004 0.996
#> GSM614444 3 0.0237 0.9985 0.000 0.004 0.996
#> GSM614391 1 0.0237 0.5944 0.996 0.004 0.000
#> GSM614392 1 0.0237 0.5944 0.996 0.004 0.000
#> GSM614393 1 0.0237 0.5944 0.996 0.004 0.000
#> GSM614394 1 0.0237 0.5944 0.996 0.004 0.000
#> GSM614395 1 0.6451 -0.0762 0.560 0.004 0.436
#> GSM614396 1 0.0237 0.5944 0.996 0.004 0.000
#> GSM614397 1 0.0475 0.5906 0.992 0.004 0.004
#> GSM614398 1 0.0237 0.5944 0.996 0.004 0.000
#> GSM614399 2 0.0237 0.9366 0.004 0.996 0.000
#> GSM614400 2 0.0237 0.9366 0.004 0.996 0.000
#> GSM614401 2 0.0237 0.9366 0.004 0.996 0.000
#> GSM614402 2 0.0237 0.9366 0.004 0.996 0.000
#> GSM614403 2 0.0237 0.9366 0.004 0.996 0.000
#> GSM614404 2 0.0237 0.9366 0.004 0.996 0.000
#> GSM614405 2 0.0237 0.9366 0.004 0.996 0.000
#> GSM614406 2 0.0237 0.9366 0.004 0.996 0.000
#> GSM614407 2 0.4682 0.7405 0.192 0.804 0.004
#> GSM614408 2 0.4682 0.7405 0.192 0.804 0.004
#> GSM614409 2 0.4682 0.7405 0.192 0.804 0.004
#> GSM614410 2 0.4682 0.7405 0.192 0.804 0.004
#> GSM614411 2 0.4682 0.7405 0.192 0.804 0.004
#> GSM614412 2 0.4682 0.7405 0.192 0.804 0.004
#> GSM614413 2 0.4682 0.7405 0.192 0.804 0.004
#> GSM614414 2 0.4682 0.7405 0.192 0.804 0.004
#> GSM614445 2 0.3193 0.8817 0.004 0.896 0.100
#> GSM614446 2 0.3193 0.8817 0.004 0.896 0.100
#> GSM614447 2 0.3193 0.8817 0.004 0.896 0.100
#> GSM614448 2 0.3193 0.8817 0.004 0.896 0.100
#> GSM614449 2 0.3193 0.8817 0.004 0.896 0.100
#> GSM614450 2 0.3193 0.8817 0.004 0.896 0.100
#> GSM614451 3 0.0424 0.9940 0.000 0.008 0.992
#> GSM614452 3 0.0424 0.9940 0.000 0.008 0.992
#> GSM614453 2 0.0475 0.9364 0.004 0.992 0.004
#> GSM614454 2 0.0475 0.9364 0.004 0.992 0.004
#> GSM614455 2 0.0475 0.9364 0.004 0.992 0.004
#> GSM614456 2 0.0475 0.9364 0.004 0.992 0.004
#> GSM614457 2 0.0475 0.9364 0.004 0.992 0.004
#> GSM614458 2 0.0475 0.9364 0.004 0.992 0.004
#> GSM614459 2 0.0475 0.9364 0.004 0.992 0.004
#> GSM614460 2 0.0475 0.9364 0.004 0.992 0.004
#> GSM614461 2 0.0237 0.9371 0.004 0.996 0.000
#> GSM614462 2 0.0237 0.9371 0.004 0.996 0.000
#> GSM614463 2 0.0237 0.9371 0.004 0.996 0.000
#> GSM614464 2 0.0237 0.9371 0.004 0.996 0.000
#> GSM614465 2 0.0237 0.9371 0.004 0.996 0.000
#> GSM614466 2 0.0237 0.9371 0.004 0.996 0.000
#> GSM614467 2 0.0237 0.9371 0.004 0.996 0.000
#> GSM614468 2 0.0237 0.9371 0.004 0.996 0.000
#> GSM614469 2 0.0892 0.9316 0.020 0.980 0.000
#> GSM614470 2 0.0892 0.9316 0.020 0.980 0.000
#> GSM614471 2 0.0892 0.9316 0.020 0.980 0.000
#> GSM614472 2 0.0892 0.9316 0.020 0.980 0.000
#> GSM614473 2 0.0892 0.9316 0.020 0.980 0.000
#> GSM614474 2 0.0892 0.9316 0.020 0.980 0.000
#> GSM614475 2 0.0892 0.9316 0.020 0.980 0.000
#> GSM614476 2 0.0892 0.9316 0.020 0.980 0.000
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM614415 1 0.6536 0.5606 0.580 0.324 0.096 0.000
#> GSM614416 1 0.6536 0.5606 0.580 0.324 0.096 0.000
#> GSM614417 1 0.6536 0.5606 0.580 0.324 0.096 0.000
#> GSM614418 1 0.6536 0.5606 0.580 0.324 0.096 0.000
#> GSM614419 1 0.6536 0.5606 0.580 0.324 0.096 0.000
#> GSM614420 1 0.6536 0.5606 0.580 0.324 0.096 0.000
#> GSM614421 2 0.2586 0.8587 0.000 0.912 0.040 0.048
#> GSM614422 2 0.2586 0.8587 0.000 0.912 0.040 0.048
#> GSM614423 2 0.2586 0.8587 0.000 0.912 0.040 0.048
#> GSM614424 2 0.2586 0.8587 0.000 0.912 0.040 0.048
#> GSM614425 2 0.2586 0.8587 0.000 0.912 0.040 0.048
#> GSM614426 2 0.2586 0.8587 0.000 0.912 0.040 0.048
#> GSM614427 2 0.2586 0.8587 0.000 0.912 0.040 0.048
#> GSM614428 2 0.2586 0.8587 0.000 0.912 0.040 0.048
#> GSM614429 2 0.0188 0.8921 0.000 0.996 0.004 0.000
#> GSM614430 2 0.0188 0.8921 0.000 0.996 0.004 0.000
#> GSM614431 2 0.0188 0.8921 0.000 0.996 0.004 0.000
#> GSM614432 2 0.0188 0.8921 0.000 0.996 0.004 0.000
#> GSM614433 2 0.0188 0.8921 0.000 0.996 0.004 0.000
#> GSM614434 2 0.0188 0.8921 0.000 0.996 0.004 0.000
#> GSM614435 2 0.0188 0.8921 0.000 0.996 0.004 0.000
#> GSM614436 2 0.0188 0.8921 0.000 0.996 0.004 0.000
#> GSM614437 4 0.0000 1.0000 0.000 0.000 0.000 1.000
#> GSM614438 4 0.0000 1.0000 0.000 0.000 0.000 1.000
#> GSM614439 4 0.0000 1.0000 0.000 0.000 0.000 1.000
#> GSM614440 4 0.0000 1.0000 0.000 0.000 0.000 1.000
#> GSM614441 4 0.0000 1.0000 0.000 0.000 0.000 1.000
#> GSM614442 4 0.0000 1.0000 0.000 0.000 0.000 1.000
#> GSM614443 4 0.0000 1.0000 0.000 0.000 0.000 1.000
#> GSM614444 4 0.0000 1.0000 0.000 0.000 0.000 1.000
#> GSM614391 1 0.0000 0.5727 1.000 0.000 0.000 0.000
#> GSM614392 1 0.0000 0.5727 1.000 0.000 0.000 0.000
#> GSM614393 1 0.0000 0.5727 1.000 0.000 0.000 0.000
#> GSM614394 1 0.0000 0.5727 1.000 0.000 0.000 0.000
#> GSM614395 1 0.4941 0.0455 0.564 0.000 0.436 0.000
#> GSM614396 1 0.0000 0.5727 1.000 0.000 0.000 0.000
#> GSM614397 1 0.0188 0.5688 0.996 0.000 0.004 0.000
#> GSM614398 1 0.0000 0.5727 1.000 0.000 0.000 0.000
#> GSM614399 2 0.0188 0.8916 0.000 0.996 0.004 0.000
#> GSM614400 2 0.0188 0.8916 0.000 0.996 0.004 0.000
#> GSM614401 2 0.0188 0.8916 0.000 0.996 0.004 0.000
#> GSM614402 2 0.0188 0.8916 0.000 0.996 0.004 0.000
#> GSM614403 2 0.0188 0.8916 0.000 0.996 0.004 0.000
#> GSM614404 2 0.0188 0.8916 0.000 0.996 0.004 0.000
#> GSM614405 2 0.0188 0.8916 0.000 0.996 0.004 0.000
#> GSM614406 2 0.0188 0.8916 0.000 0.996 0.004 0.000
#> GSM614407 2 0.6204 0.2330 0.052 0.500 0.448 0.000
#> GSM614408 2 0.6204 0.2330 0.052 0.500 0.448 0.000
#> GSM614409 2 0.6204 0.2330 0.052 0.500 0.448 0.000
#> GSM614410 2 0.6204 0.2330 0.052 0.500 0.448 0.000
#> GSM614411 2 0.6204 0.2330 0.052 0.500 0.448 0.000
#> GSM614412 2 0.6204 0.2330 0.052 0.500 0.448 0.000
#> GSM614413 2 0.6204 0.2330 0.052 0.500 0.448 0.000
#> GSM614414 2 0.6204 0.2330 0.052 0.500 0.448 0.000
#> GSM614445 2 0.3004 0.8442 0.000 0.892 0.048 0.060
#> GSM614446 2 0.3004 0.8442 0.000 0.892 0.048 0.060
#> GSM614447 2 0.3004 0.8442 0.000 0.892 0.048 0.060
#> GSM614448 2 0.3004 0.8442 0.000 0.892 0.048 0.060
#> GSM614449 2 0.3004 0.8442 0.000 0.892 0.048 0.060
#> GSM614450 2 0.3004 0.8442 0.000 0.892 0.048 0.060
#> GSM614451 3 0.5132 1.0000 0.000 0.004 0.548 0.448
#> GSM614452 3 0.5132 1.0000 0.000 0.004 0.548 0.448
#> GSM614453 2 0.0376 0.8913 0.000 0.992 0.004 0.004
#> GSM614454 2 0.0376 0.8913 0.000 0.992 0.004 0.004
#> GSM614455 2 0.0376 0.8913 0.000 0.992 0.004 0.004
#> GSM614456 2 0.0376 0.8913 0.000 0.992 0.004 0.004
#> GSM614457 2 0.0376 0.8913 0.000 0.992 0.004 0.004
#> GSM614458 2 0.0376 0.8913 0.000 0.992 0.004 0.004
#> GSM614459 2 0.0376 0.8913 0.000 0.992 0.004 0.004
#> GSM614460 2 0.0376 0.8913 0.000 0.992 0.004 0.004
#> GSM614461 2 0.0188 0.8921 0.000 0.996 0.004 0.000
#> GSM614462 2 0.0188 0.8921 0.000 0.996 0.004 0.000
#> GSM614463 2 0.0188 0.8921 0.000 0.996 0.004 0.000
#> GSM614464 2 0.0188 0.8921 0.000 0.996 0.004 0.000
#> GSM614465 2 0.0188 0.8921 0.000 0.996 0.004 0.000
#> GSM614466 2 0.0188 0.8921 0.000 0.996 0.004 0.000
#> GSM614467 2 0.0188 0.8921 0.000 0.996 0.004 0.000
#> GSM614468 2 0.0188 0.8921 0.000 0.996 0.004 0.000
#> GSM614469 2 0.1624 0.8728 0.020 0.952 0.028 0.000
#> GSM614470 2 0.1624 0.8728 0.020 0.952 0.028 0.000
#> GSM614471 2 0.1624 0.8728 0.020 0.952 0.028 0.000
#> GSM614472 2 0.1624 0.8728 0.020 0.952 0.028 0.000
#> GSM614473 2 0.1624 0.8728 0.020 0.952 0.028 0.000
#> GSM614474 2 0.1624 0.8728 0.020 0.952 0.028 0.000
#> GSM614475 2 0.1624 0.8728 0.020 0.952 0.028 0.000
#> GSM614476 2 0.1624 0.8728 0.020 0.952 0.028 0.000
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM614415 5 0.6545 0.5331 0.360 0.068 0.056 0.000 0.516
#> GSM614416 5 0.6545 0.5331 0.360 0.068 0.056 0.000 0.516
#> GSM614417 5 0.6545 0.5331 0.360 0.068 0.056 0.000 0.516
#> GSM614418 5 0.6545 0.5331 0.360 0.068 0.056 0.000 0.516
#> GSM614419 5 0.6545 0.5331 0.360 0.068 0.056 0.000 0.516
#> GSM614420 5 0.6545 0.5331 0.360 0.068 0.056 0.000 0.516
#> GSM614421 2 0.2293 0.9073 0.016 0.900 0.084 0.000 0.000
#> GSM614422 2 0.2293 0.9073 0.016 0.900 0.084 0.000 0.000
#> GSM614423 2 0.2293 0.9073 0.016 0.900 0.084 0.000 0.000
#> GSM614424 2 0.2293 0.9073 0.016 0.900 0.084 0.000 0.000
#> GSM614425 2 0.2293 0.9073 0.016 0.900 0.084 0.000 0.000
#> GSM614426 2 0.2293 0.9073 0.016 0.900 0.084 0.000 0.000
#> GSM614427 2 0.2293 0.9073 0.016 0.900 0.084 0.000 0.000
#> GSM614428 2 0.2293 0.9073 0.016 0.900 0.084 0.000 0.000
#> GSM614429 2 0.0290 0.9459 0.000 0.992 0.008 0.000 0.000
#> GSM614430 2 0.0290 0.9459 0.000 0.992 0.008 0.000 0.000
#> GSM614431 2 0.0290 0.9459 0.000 0.992 0.008 0.000 0.000
#> GSM614432 2 0.0290 0.9459 0.000 0.992 0.008 0.000 0.000
#> GSM614433 2 0.0290 0.9459 0.000 0.992 0.008 0.000 0.000
#> GSM614434 2 0.0290 0.9459 0.000 0.992 0.008 0.000 0.000
#> GSM614435 2 0.0290 0.9459 0.000 0.992 0.008 0.000 0.000
#> GSM614436 2 0.0290 0.9459 0.000 0.992 0.008 0.000 0.000
#> GSM614437 4 0.0000 1.0000 0.000 0.000 0.000 1.000 0.000
#> GSM614438 4 0.0000 1.0000 0.000 0.000 0.000 1.000 0.000
#> GSM614439 4 0.0000 1.0000 0.000 0.000 0.000 1.000 0.000
#> GSM614440 4 0.0000 1.0000 0.000 0.000 0.000 1.000 0.000
#> GSM614441 4 0.0000 1.0000 0.000 0.000 0.000 1.000 0.000
#> GSM614442 4 0.0000 1.0000 0.000 0.000 0.000 1.000 0.000
#> GSM614443 4 0.0000 1.0000 0.000 0.000 0.000 1.000 0.000
#> GSM614444 4 0.0000 1.0000 0.000 0.000 0.000 1.000 0.000
#> GSM614391 5 0.0000 0.6653 0.000 0.000 0.000 0.000 1.000
#> GSM614392 5 0.0000 0.6653 0.000 0.000 0.000 0.000 1.000
#> GSM614393 5 0.0000 0.6653 0.000 0.000 0.000 0.000 1.000
#> GSM614394 5 0.0000 0.6653 0.000 0.000 0.000 0.000 1.000
#> GSM614395 5 0.4256 0.0914 0.000 0.000 0.436 0.000 0.564
#> GSM614396 5 0.0000 0.6653 0.000 0.000 0.000 0.000 1.000
#> GSM614397 5 0.0162 0.6623 0.000 0.000 0.004 0.000 0.996
#> GSM614398 5 0.0000 0.6653 0.000 0.000 0.000 0.000 1.000
#> GSM614399 2 0.0404 0.9441 0.000 0.988 0.012 0.000 0.000
#> GSM614400 2 0.0404 0.9441 0.000 0.988 0.012 0.000 0.000
#> GSM614401 2 0.0404 0.9441 0.000 0.988 0.012 0.000 0.000
#> GSM614402 2 0.0404 0.9441 0.000 0.988 0.012 0.000 0.000
#> GSM614403 2 0.0404 0.9441 0.000 0.988 0.012 0.000 0.000
#> GSM614404 2 0.0404 0.9441 0.000 0.988 0.012 0.000 0.000
#> GSM614405 2 0.0404 0.9441 0.000 0.988 0.012 0.000 0.000
#> GSM614406 2 0.0404 0.9441 0.000 0.988 0.012 0.000 0.000
#> GSM614407 1 0.0162 1.0000 0.996 0.004 0.000 0.000 0.000
#> GSM614408 1 0.0162 1.0000 0.996 0.004 0.000 0.000 0.000
#> GSM614409 1 0.0162 1.0000 0.996 0.004 0.000 0.000 0.000
#> GSM614410 1 0.0162 1.0000 0.996 0.004 0.000 0.000 0.000
#> GSM614411 1 0.0162 1.0000 0.996 0.004 0.000 0.000 0.000
#> GSM614412 1 0.0162 1.0000 0.996 0.004 0.000 0.000 0.000
#> GSM614413 1 0.0162 1.0000 0.996 0.004 0.000 0.000 0.000
#> GSM614414 1 0.0162 1.0000 0.996 0.004 0.000 0.000 0.000
#> GSM614445 2 0.2513 0.8912 0.008 0.876 0.116 0.000 0.000
#> GSM614446 2 0.2513 0.8912 0.008 0.876 0.116 0.000 0.000
#> GSM614447 2 0.2513 0.8912 0.008 0.876 0.116 0.000 0.000
#> GSM614448 2 0.2513 0.8912 0.008 0.876 0.116 0.000 0.000
#> GSM614449 2 0.2513 0.8912 0.008 0.876 0.116 0.000 0.000
#> GSM614450 2 0.2513 0.8912 0.008 0.876 0.116 0.000 0.000
#> GSM614451 3 0.1671 1.0000 0.000 0.000 0.924 0.076 0.000
#> GSM614452 3 0.1671 1.0000 0.000 0.000 0.924 0.076 0.000
#> GSM614453 2 0.0566 0.9450 0.000 0.984 0.012 0.004 0.000
#> GSM614454 2 0.0566 0.9450 0.000 0.984 0.012 0.004 0.000
#> GSM614455 2 0.0566 0.9450 0.000 0.984 0.012 0.004 0.000
#> GSM614456 2 0.0566 0.9450 0.000 0.984 0.012 0.004 0.000
#> GSM614457 2 0.0566 0.9450 0.000 0.984 0.012 0.004 0.000
#> GSM614458 2 0.0566 0.9450 0.000 0.984 0.012 0.004 0.000
#> GSM614459 2 0.0566 0.9450 0.000 0.984 0.012 0.004 0.000
#> GSM614460 2 0.0566 0.9450 0.000 0.984 0.012 0.004 0.000
#> GSM614461 2 0.0404 0.9456 0.000 0.988 0.012 0.000 0.000
#> GSM614462 2 0.0404 0.9456 0.000 0.988 0.012 0.000 0.000
#> GSM614463 2 0.0404 0.9456 0.000 0.988 0.012 0.000 0.000
#> GSM614464 2 0.0404 0.9456 0.000 0.988 0.012 0.000 0.000
#> GSM614465 2 0.0404 0.9456 0.000 0.988 0.012 0.000 0.000
#> GSM614466 2 0.0404 0.9456 0.000 0.988 0.012 0.000 0.000
#> GSM614467 2 0.0404 0.9456 0.000 0.988 0.012 0.000 0.000
#> GSM614468 2 0.0404 0.9456 0.000 0.988 0.012 0.000 0.000
#> GSM614469 2 0.2747 0.8746 0.088 0.884 0.012 0.000 0.016
#> GSM614470 2 0.2747 0.8746 0.088 0.884 0.012 0.000 0.016
#> GSM614471 2 0.2747 0.8746 0.088 0.884 0.012 0.000 0.016
#> GSM614472 2 0.2747 0.8746 0.088 0.884 0.012 0.000 0.016
#> GSM614473 2 0.2747 0.8746 0.088 0.884 0.012 0.000 0.016
#> GSM614474 2 0.2747 0.8746 0.088 0.884 0.012 0.000 0.016
#> GSM614475 2 0.2747 0.8746 0.088 0.884 0.012 0.000 0.016
#> GSM614476 2 0.2747 0.8746 0.088 0.884 0.012 0.000 0.016
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM614415 1 0.0790 1.000 0.968 0.000 0.000 0.000 0.032 0.000
#> GSM614416 1 0.0790 1.000 0.968 0.000 0.000 0.000 0.032 0.000
#> GSM614417 1 0.0790 1.000 0.968 0.000 0.000 0.000 0.032 0.000
#> GSM614418 1 0.0790 1.000 0.968 0.000 0.000 0.000 0.032 0.000
#> GSM614419 1 0.0790 1.000 0.968 0.000 0.000 0.000 0.032 0.000
#> GSM614420 1 0.0790 1.000 0.968 0.000 0.000 0.000 0.032 0.000
#> GSM614421 2 0.2527 0.894 0.032 0.880 0.084 0.000 0.000 0.004
#> GSM614422 2 0.2527 0.894 0.032 0.880 0.084 0.000 0.000 0.004
#> GSM614423 2 0.2527 0.894 0.032 0.880 0.084 0.000 0.000 0.004
#> GSM614424 2 0.2527 0.894 0.032 0.880 0.084 0.000 0.000 0.004
#> GSM614425 2 0.2527 0.894 0.032 0.880 0.084 0.000 0.000 0.004
#> GSM614426 2 0.2527 0.894 0.032 0.880 0.084 0.000 0.000 0.004
#> GSM614427 2 0.2527 0.894 0.032 0.880 0.084 0.000 0.000 0.004
#> GSM614428 2 0.2527 0.894 0.032 0.880 0.084 0.000 0.000 0.004
#> GSM614429 2 0.0291 0.936 0.004 0.992 0.004 0.000 0.000 0.000
#> GSM614430 2 0.0291 0.936 0.004 0.992 0.004 0.000 0.000 0.000
#> GSM614431 2 0.0291 0.936 0.004 0.992 0.004 0.000 0.000 0.000
#> GSM614432 2 0.0291 0.936 0.004 0.992 0.004 0.000 0.000 0.000
#> GSM614433 2 0.0291 0.936 0.004 0.992 0.004 0.000 0.000 0.000
#> GSM614434 2 0.0291 0.936 0.004 0.992 0.004 0.000 0.000 0.000
#> GSM614435 2 0.0291 0.936 0.004 0.992 0.004 0.000 0.000 0.000
#> GSM614436 2 0.0291 0.936 0.004 0.992 0.004 0.000 0.000 0.000
#> GSM614437 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM614438 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM614439 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM614440 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM614441 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM614442 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM614443 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM614444 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM614391 5 0.0000 0.934 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM614392 5 0.0000 0.934 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM614393 5 0.0000 0.934 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM614394 5 0.0000 0.934 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM614395 5 0.3823 0.244 0.000 0.000 0.436 0.000 0.564 0.000
#> GSM614396 5 0.0000 0.934 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM614397 5 0.0146 0.931 0.000 0.000 0.004 0.000 0.996 0.000
#> GSM614398 5 0.0000 0.934 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM614399 2 0.0972 0.931 0.028 0.964 0.008 0.000 0.000 0.000
#> GSM614400 2 0.0972 0.931 0.028 0.964 0.008 0.000 0.000 0.000
#> GSM614401 2 0.0972 0.931 0.028 0.964 0.008 0.000 0.000 0.000
#> GSM614402 2 0.0972 0.931 0.028 0.964 0.008 0.000 0.000 0.000
#> GSM614403 2 0.0972 0.931 0.028 0.964 0.008 0.000 0.000 0.000
#> GSM614404 2 0.0972 0.931 0.028 0.964 0.008 0.000 0.000 0.000
#> GSM614405 2 0.0972 0.931 0.028 0.964 0.008 0.000 0.000 0.000
#> GSM614406 2 0.0972 0.931 0.028 0.964 0.008 0.000 0.000 0.000
#> GSM614407 6 0.0000 1.000 0.000 0.000 0.000 0.000 0.000 1.000
#> GSM614408 6 0.0000 1.000 0.000 0.000 0.000 0.000 0.000 1.000
#> GSM614409 6 0.0000 1.000 0.000 0.000 0.000 0.000 0.000 1.000
#> GSM614410 6 0.0000 1.000 0.000 0.000 0.000 0.000 0.000 1.000
#> GSM614411 6 0.0000 1.000 0.000 0.000 0.000 0.000 0.000 1.000
#> GSM614412 6 0.0000 1.000 0.000 0.000 0.000 0.000 0.000 1.000
#> GSM614413 6 0.0000 1.000 0.000 0.000 0.000 0.000 0.000 1.000
#> GSM614414 6 0.0000 1.000 0.000 0.000 0.000 0.000 0.000 1.000
#> GSM614445 2 0.2867 0.873 0.040 0.848 0.112 0.000 0.000 0.000
#> GSM614446 2 0.2867 0.873 0.040 0.848 0.112 0.000 0.000 0.000
#> GSM614447 2 0.2867 0.873 0.040 0.848 0.112 0.000 0.000 0.000
#> GSM614448 2 0.2867 0.873 0.040 0.848 0.112 0.000 0.000 0.000
#> GSM614449 2 0.2867 0.873 0.040 0.848 0.112 0.000 0.000 0.000
#> GSM614450 2 0.2867 0.873 0.040 0.848 0.112 0.000 0.000 0.000
#> GSM614451 3 0.0458 1.000 0.000 0.000 0.984 0.016 0.000 0.000
#> GSM614452 3 0.0458 1.000 0.000 0.000 0.984 0.016 0.000 0.000
#> GSM614453 2 0.0665 0.935 0.008 0.980 0.008 0.004 0.000 0.000
#> GSM614454 2 0.0665 0.935 0.008 0.980 0.008 0.004 0.000 0.000
#> GSM614455 2 0.0665 0.935 0.008 0.980 0.008 0.004 0.000 0.000
#> GSM614456 2 0.0665 0.935 0.008 0.980 0.008 0.004 0.000 0.000
#> GSM614457 2 0.0665 0.935 0.008 0.980 0.008 0.004 0.000 0.000
#> GSM614458 2 0.0665 0.935 0.008 0.980 0.008 0.004 0.000 0.000
#> GSM614459 2 0.0665 0.935 0.008 0.980 0.008 0.004 0.000 0.000
#> GSM614460 2 0.0665 0.935 0.008 0.980 0.008 0.004 0.000 0.000
#> GSM614461 2 0.0520 0.935 0.008 0.984 0.008 0.000 0.000 0.000
#> GSM614462 2 0.0520 0.935 0.008 0.984 0.008 0.000 0.000 0.000
#> GSM614463 2 0.0520 0.935 0.008 0.984 0.008 0.000 0.000 0.000
#> GSM614464 2 0.0520 0.935 0.008 0.984 0.008 0.000 0.000 0.000
#> GSM614465 2 0.0520 0.935 0.008 0.984 0.008 0.000 0.000 0.000
#> GSM614466 2 0.0520 0.935 0.008 0.984 0.008 0.000 0.000 0.000
#> GSM614467 2 0.0520 0.935 0.008 0.984 0.008 0.000 0.000 0.000
#> GSM614468 2 0.0520 0.935 0.008 0.984 0.008 0.000 0.000 0.000
#> GSM614469 2 0.2234 0.869 0.124 0.872 0.004 0.000 0.000 0.000
#> GSM614470 2 0.2234 0.869 0.124 0.872 0.004 0.000 0.000 0.000
#> GSM614471 2 0.2234 0.869 0.124 0.872 0.004 0.000 0.000 0.000
#> GSM614472 2 0.2234 0.869 0.124 0.872 0.004 0.000 0.000 0.000
#> GSM614473 2 0.2234 0.869 0.124 0.872 0.004 0.000 0.000 0.000
#> GSM614474 2 0.2234 0.869 0.124 0.872 0.004 0.000 0.000 0.000
#> GSM614475 2 0.2234 0.869 0.124 0.872 0.004 0.000 0.000 0.000
#> GSM614476 2 0.2234 0.869 0.124 0.872 0.004 0.000 0.000 0.000
Heatmaps for the consensus matrix. It visualizes the probability of two samples to be in a same group.
consensus_heatmap(res, k = 2)
consensus_heatmap(res, k = 3)
consensus_heatmap(res, k = 4)
consensus_heatmap(res, k = 5)
consensus_heatmap(res, k = 6)
Heatmaps for the membership of samples in all partitions to see how consistent they are:
membership_heatmap(res, k = 2)
membership_heatmap(res, k = 3)
membership_heatmap(res, k = 4)
membership_heatmap(res, k = 5)
membership_heatmap(res, k = 6)
As soon as we have had the classes for columns, we can look for signatures which are significantly different between classes which can be candidate marks for certain classes. Following are the heatmaps for signatures.
Signature heatmaps where rows are scaled:
get_signatures(res, k = 2)
get_signatures(res, k = 3)
get_signatures(res, k = 4)
get_signatures(res, k = 5)
get_signatures(res, k = 6)
Signature heatmaps where rows are not scaled:
get_signatures(res, k = 2, scale_rows = FALSE)
get_signatures(res, k = 3, scale_rows = FALSE)
get_signatures(res, k = 4, scale_rows = FALSE)
get_signatures(res, k = 5, scale_rows = FALSE)
get_signatures(res, k = 6, scale_rows = FALSE)
Compare the overlap of signatures from different k:
compare_signatures(res)
get_signature() returns a data frame invisibly. TO get the list of signatures, the function
call should be assigned to a variable explicitly. In following code, if plot argument is set
to FALSE, no heatmap is plotted while only the differential analysis is performed.
# code only for demonstration
tb = get_signature(res, k = ..., plot = FALSE)
An example of the output of tb is:
#> which_row fdr mean_1 mean_2 scaled_mean_1 scaled_mean_2 km
#> 1 38 0.042760348 8.373488 9.131774 -0.5533452 0.5164555 1
#> 2 40 0.018707592 7.106213 8.469186 -0.6173731 0.5762149 1
#> 3 55 0.019134737 10.221463 11.207825 -0.6159697 0.5749050 1
#> 4 59 0.006059896 5.921854 7.869574 -0.6899429 0.6439467 1
#> 5 60 0.018055526 8.928898 10.211722 -0.6204761 0.5791110 1
#> 6 98 0.009384629 15.714769 14.887706 0.6635654 -0.6193277 2
...
The columns in tb are:
which_row: row indices corresponding to the input matrix.fdr: FDR for the differential test. mean_x: The mean value in group x.scaled_mean_x: The mean value in group x after rows are scaled.km: Row groups if k-means clustering is applied to rows.UMAP plot which shows how samples are separated.
dimension_reduction(res, k = 2, method = "UMAP")
dimension_reduction(res, k = 3, method = "UMAP")
dimension_reduction(res, k = 4, method = "UMAP")
dimension_reduction(res, k = 5, method = "UMAP")
dimension_reduction(res, k = 6, method = "UMAP")
Following heatmap shows how subgroups are split when increasing k:
collect_classes(res)
Test correlation between subgroups and known annotations. If the known annotation is numeric, one-way ANOVA test is applied, and if the known annotation is discrete, chi-squared contingency table test is applied.
test_to_known_factors(res)
#> n individual(p) protocol(p) time(p) other(p) k
#> CV:hclust 85 3.47e-11 0.3626 0.992 0.004489 2
#> CV:hclust 79 9.73e-22 0.4904 0.999 0.009160 3
#> CV:hclust 77 6.78e-23 0.0697 1.000 0.000256 4
#> CV:hclust 85 1.26e-36 0.1333 1.000 0.001249 5
#> CV:hclust 85 2.70e-48 0.1300 1.000 0.001105 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 21168 rows and 86 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 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 0.154 0.421 0.736 0.3824 0.615 0.615
#> 3 3 0.277 0.656 0.785 0.4402 0.787 0.670
#> 4 4 0.362 0.503 0.671 0.2087 0.795 0.586
#> 5 5 0.490 0.580 0.701 0.1141 0.749 0.375
#> 6 6 0.621 0.687 0.699 0.0673 0.894 0.593
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
#> GSM614415 2 0.996 -0.1098 0.464 0.536
#> GSM614416 2 0.996 -0.1098 0.464 0.536
#> GSM614417 2 0.996 -0.1098 0.464 0.536
#> GSM614418 2 0.996 -0.1098 0.464 0.536
#> GSM614419 1 0.987 0.3054 0.568 0.432
#> GSM614420 1 0.987 0.3054 0.568 0.432
#> GSM614421 2 0.871 0.2114 0.292 0.708
#> GSM614422 2 0.876 0.2111 0.296 0.704
#> GSM614423 2 0.260 0.6329 0.044 0.956
#> GSM614424 2 0.871 0.2114 0.292 0.708
#> GSM614425 2 0.871 0.2114 0.292 0.708
#> GSM614426 2 0.871 0.2114 0.292 0.708
#> GSM614427 2 0.876 0.2004 0.296 0.704
#> GSM614428 2 0.939 0.0112 0.356 0.644
#> GSM614429 2 0.469 0.6316 0.100 0.900
#> GSM614430 2 0.469 0.6316 0.100 0.900
#> GSM614431 2 0.416 0.6433 0.084 0.916
#> GSM614432 2 0.416 0.6433 0.084 0.916
#> GSM614433 2 0.416 0.6433 0.084 0.916
#> GSM614434 2 0.416 0.6433 0.084 0.916
#> GSM614435 2 0.518 0.6171 0.116 0.884
#> GSM614436 2 0.886 0.2898 0.304 0.696
#> GSM614437 1 0.993 0.2922 0.548 0.452
#> GSM614438 1 0.992 0.3300 0.552 0.448
#> GSM614439 1 0.992 0.3300 0.552 0.448
#> GSM614440 1 0.992 0.3300 0.552 0.448
#> GSM614441 1 0.992 0.3300 0.552 0.448
#> GSM614442 1 0.992 0.3300 0.552 0.448
#> GSM614443 1 0.990 0.3129 0.560 0.440
#> GSM614444 1 0.992 0.3300 0.552 0.448
#> GSM614391 1 0.980 0.3223 0.584 0.416
#> GSM614392 1 1.000 0.1560 0.504 0.496
#> GSM614393 1 1.000 0.1560 0.504 0.496
#> GSM614394 1 0.980 0.3223 0.584 0.416
#> GSM614395 1 0.680 0.4062 0.820 0.180
#> GSM614396 1 0.980 0.3223 0.584 0.416
#> GSM614397 1 0.827 0.4096 0.740 0.260
#> GSM614398 1 0.886 0.3968 0.696 0.304
#> GSM614399 2 0.260 0.6421 0.044 0.956
#> GSM614400 2 0.260 0.6421 0.044 0.956
#> GSM614401 2 0.260 0.6421 0.044 0.956
#> GSM614402 2 0.260 0.6421 0.044 0.956
#> GSM614403 2 0.118 0.6470 0.016 0.984
#> GSM614404 2 0.260 0.6421 0.044 0.956
#> GSM614405 2 0.295 0.6410 0.052 0.948
#> GSM614406 2 0.921 0.0460 0.336 0.664
#> GSM614407 2 0.936 0.1800 0.352 0.648
#> GSM614408 2 0.936 0.1800 0.352 0.648
#> GSM614409 2 0.936 0.1800 0.352 0.648
#> GSM614410 2 0.936 0.1800 0.352 0.648
#> GSM614411 2 0.936 0.1800 0.352 0.648
#> GSM614412 2 0.939 0.1706 0.356 0.644
#> GSM614413 1 0.983 0.3009 0.576 0.424
#> GSM614414 1 0.985 0.2963 0.572 0.428
#> GSM614445 2 0.204 0.6392 0.032 0.968
#> GSM614446 2 0.204 0.6392 0.032 0.968
#> GSM614447 2 0.204 0.6392 0.032 0.968
#> GSM614448 2 0.895 0.1545 0.312 0.688
#> GSM614449 2 0.855 0.2300 0.280 0.720
#> GSM614450 2 0.260 0.6328 0.044 0.956
#> GSM614451 1 0.997 0.3146 0.532 0.468
#> GSM614452 1 0.997 0.3146 0.532 0.468
#> GSM614453 2 0.563 0.6040 0.132 0.868
#> GSM614454 2 0.563 0.6040 0.132 0.868
#> GSM614455 2 0.563 0.6040 0.132 0.868
#> GSM614456 2 0.563 0.6040 0.132 0.868
#> GSM614457 2 0.563 0.6040 0.132 0.868
#> GSM614458 2 0.563 0.6040 0.132 0.868
#> GSM614459 2 0.563 0.6040 0.132 0.868
#> GSM614460 2 0.563 0.6040 0.132 0.868
#> GSM614461 2 0.402 0.6468 0.080 0.920
#> GSM614462 2 0.402 0.6468 0.080 0.920
#> GSM614463 2 0.402 0.6468 0.080 0.920
#> GSM614464 2 0.402 0.6468 0.080 0.920
#> GSM614465 2 0.402 0.6468 0.080 0.920
#> GSM614466 2 0.402 0.6468 0.080 0.920
#> GSM614467 2 0.402 0.6468 0.080 0.920
#> GSM614468 2 0.402 0.6468 0.080 0.920
#> GSM614469 2 0.595 0.5464 0.144 0.856
#> GSM614470 2 0.595 0.5464 0.144 0.856
#> GSM614471 2 0.595 0.5464 0.144 0.856
#> GSM614472 2 0.595 0.5464 0.144 0.856
#> GSM614473 2 0.595 0.5464 0.144 0.856
#> GSM614474 2 0.595 0.5464 0.144 0.856
#> GSM614475 2 0.456 0.5985 0.096 0.904
#> GSM614476 2 0.373 0.6238 0.072 0.928
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM614415 1 0.525 0.737 0.792 0.188 0.020
#> GSM614416 1 0.525 0.737 0.792 0.188 0.020
#> GSM614417 1 0.525 0.737 0.792 0.188 0.020
#> GSM614418 1 0.525 0.737 0.792 0.188 0.020
#> GSM614419 1 0.486 0.729 0.840 0.116 0.044
#> GSM614420 1 0.486 0.729 0.840 0.116 0.044
#> GSM614421 2 0.945 0.287 0.212 0.492 0.296
#> GSM614422 2 0.945 0.287 0.212 0.492 0.296
#> GSM614423 2 0.579 0.680 0.136 0.796 0.068
#> GSM614424 2 0.945 0.287 0.212 0.492 0.296
#> GSM614425 2 0.945 0.287 0.212 0.492 0.296
#> GSM614426 2 0.945 0.287 0.212 0.492 0.296
#> GSM614427 2 0.945 0.287 0.212 0.492 0.296
#> GSM614428 2 0.950 0.240 0.208 0.476 0.316
#> GSM614429 2 0.207 0.727 0.000 0.940 0.060
#> GSM614430 2 0.207 0.727 0.000 0.940 0.060
#> GSM614431 2 0.207 0.727 0.000 0.940 0.060
#> GSM614432 2 0.207 0.727 0.000 0.940 0.060
#> GSM614433 2 0.196 0.728 0.000 0.944 0.056
#> GSM614434 2 0.207 0.727 0.000 0.940 0.060
#> GSM614435 2 0.207 0.727 0.000 0.940 0.060
#> GSM614436 2 0.435 0.621 0.000 0.816 0.184
#> GSM614437 3 0.412 0.946 0.000 0.168 0.832
#> GSM614438 3 0.412 0.946 0.000 0.168 0.832
#> GSM614439 3 0.412 0.946 0.000 0.168 0.832
#> GSM614440 3 0.412 0.946 0.000 0.168 0.832
#> GSM614441 3 0.412 0.946 0.000 0.168 0.832
#> GSM614442 3 0.412 0.946 0.000 0.168 0.832
#> GSM614443 3 0.412 0.946 0.000 0.168 0.832
#> GSM614444 3 0.412 0.946 0.000 0.168 0.832
#> GSM614391 1 0.426 0.685 0.868 0.036 0.096
#> GSM614392 1 0.442 0.691 0.864 0.048 0.088
#> GSM614393 1 0.442 0.691 0.864 0.048 0.088
#> GSM614394 1 0.426 0.685 0.868 0.036 0.096
#> GSM614395 1 0.618 0.376 0.660 0.008 0.332
#> GSM614396 1 0.426 0.685 0.868 0.036 0.096
#> GSM614397 1 0.445 0.632 0.836 0.012 0.152
#> GSM614398 1 0.441 0.647 0.844 0.016 0.140
#> GSM614399 2 0.560 0.684 0.136 0.804 0.060
#> GSM614400 2 0.563 0.679 0.144 0.800 0.056
#> GSM614401 2 0.563 0.679 0.144 0.800 0.056
#> GSM614402 2 0.557 0.682 0.140 0.804 0.056
#> GSM614403 2 0.552 0.695 0.120 0.812 0.068
#> GSM614404 2 0.563 0.679 0.144 0.800 0.056
#> GSM614405 2 0.613 0.679 0.136 0.780 0.084
#> GSM614406 2 0.859 0.372 0.120 0.560 0.320
#> GSM614407 1 0.739 0.561 0.600 0.356 0.044
#> GSM614408 1 0.739 0.561 0.600 0.356 0.044
#> GSM614409 1 0.739 0.561 0.600 0.356 0.044
#> GSM614410 1 0.739 0.561 0.600 0.356 0.044
#> GSM614411 1 0.739 0.561 0.600 0.356 0.044
#> GSM614412 1 0.739 0.561 0.600 0.356 0.044
#> GSM614413 1 0.828 0.616 0.628 0.224 0.148
#> GSM614414 1 0.828 0.616 0.628 0.224 0.148
#> GSM614445 2 0.514 0.701 0.104 0.832 0.064
#> GSM614446 2 0.541 0.696 0.104 0.820 0.076
#> GSM614447 2 0.541 0.696 0.104 0.820 0.076
#> GSM614448 2 0.877 0.365 0.140 0.556 0.304
#> GSM614449 2 0.868 0.396 0.140 0.572 0.288
#> GSM614450 2 0.566 0.691 0.104 0.808 0.088
#> GSM614451 3 0.649 0.760 0.076 0.172 0.752
#> GSM614452 3 0.649 0.760 0.076 0.172 0.752
#> GSM614453 2 0.384 0.684 0.012 0.872 0.116
#> GSM614454 2 0.384 0.684 0.012 0.872 0.116
#> GSM614455 2 0.384 0.684 0.012 0.872 0.116
#> GSM614456 2 0.384 0.684 0.012 0.872 0.116
#> GSM614457 2 0.384 0.684 0.012 0.872 0.116
#> GSM614458 2 0.384 0.684 0.012 0.872 0.116
#> GSM614459 2 0.384 0.684 0.012 0.872 0.116
#> GSM614460 2 0.384 0.684 0.012 0.872 0.116
#> GSM614461 2 0.210 0.732 0.004 0.944 0.052
#> GSM614462 2 0.210 0.732 0.004 0.944 0.052
#> GSM614463 2 0.210 0.732 0.004 0.944 0.052
#> GSM614464 2 0.210 0.732 0.004 0.944 0.052
#> GSM614465 2 0.210 0.732 0.004 0.944 0.052
#> GSM614466 2 0.210 0.732 0.004 0.944 0.052
#> GSM614467 2 0.210 0.732 0.004 0.944 0.052
#> GSM614468 2 0.210 0.732 0.004 0.944 0.052
#> GSM614469 2 0.569 0.604 0.224 0.756 0.020
#> GSM614470 2 0.569 0.604 0.224 0.756 0.020
#> GSM614471 2 0.569 0.604 0.224 0.756 0.020
#> GSM614472 2 0.569 0.604 0.224 0.756 0.020
#> GSM614473 2 0.569 0.604 0.224 0.756 0.020
#> GSM614474 2 0.569 0.604 0.224 0.756 0.020
#> GSM614475 2 0.563 0.622 0.208 0.768 0.024
#> GSM614476 2 0.619 0.647 0.176 0.764 0.060
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM614415 1 0.6605 0.6090 0.628 0.080 0.276 0.016
#> GSM614416 1 0.6605 0.6090 0.628 0.080 0.276 0.016
#> GSM614417 1 0.6605 0.6090 0.628 0.080 0.276 0.016
#> GSM614418 1 0.6605 0.6090 0.628 0.080 0.276 0.016
#> GSM614419 1 0.5499 0.6548 0.680 0.012 0.284 0.024
#> GSM614420 1 0.5499 0.6548 0.680 0.012 0.284 0.024
#> GSM614421 3 0.9594 0.4544 0.148 0.236 0.384 0.232
#> GSM614422 3 0.9593 0.4568 0.148 0.240 0.384 0.228
#> GSM614423 3 0.8020 0.2734 0.128 0.408 0.428 0.036
#> GSM614424 3 0.9594 0.4544 0.148 0.236 0.384 0.232
#> GSM614425 3 0.9594 0.4544 0.148 0.236 0.384 0.232
#> GSM614426 3 0.9594 0.4544 0.148 0.236 0.384 0.232
#> GSM614427 3 0.9594 0.4544 0.148 0.236 0.384 0.232
#> GSM614428 3 0.9594 0.4480 0.148 0.232 0.384 0.236
#> GSM614429 2 0.0707 0.6720 0.000 0.980 0.000 0.020
#> GSM614430 2 0.0707 0.6720 0.000 0.980 0.000 0.020
#> GSM614431 2 0.0707 0.6720 0.000 0.980 0.000 0.020
#> GSM614432 2 0.0707 0.6720 0.000 0.980 0.000 0.020
#> GSM614433 2 0.0707 0.6720 0.000 0.980 0.000 0.020
#> GSM614434 2 0.0707 0.6720 0.000 0.980 0.000 0.020
#> GSM614435 2 0.0817 0.6704 0.000 0.976 0.000 0.024
#> GSM614436 2 0.3215 0.5916 0.000 0.876 0.032 0.092
#> GSM614437 4 0.2011 0.8884 0.000 0.080 0.000 0.920
#> GSM614438 4 0.2125 0.8946 0.000 0.076 0.004 0.920
#> GSM614439 4 0.2125 0.8946 0.000 0.076 0.004 0.920
#> GSM614440 4 0.2125 0.8946 0.000 0.076 0.004 0.920
#> GSM614441 4 0.2125 0.8946 0.000 0.076 0.004 0.920
#> GSM614442 4 0.2125 0.8946 0.000 0.076 0.004 0.920
#> GSM614443 4 0.2011 0.8884 0.000 0.080 0.000 0.920
#> GSM614444 4 0.2125 0.8946 0.000 0.076 0.004 0.920
#> GSM614391 1 0.0524 0.7499 0.988 0.000 0.004 0.008
#> GSM614392 1 0.0524 0.7498 0.988 0.000 0.008 0.004
#> GSM614393 1 0.0524 0.7498 0.988 0.000 0.008 0.004
#> GSM614394 1 0.0672 0.7489 0.984 0.000 0.008 0.008
#> GSM614395 1 0.4784 0.5491 0.788 0.000 0.112 0.100
#> GSM614396 1 0.0672 0.7489 0.984 0.000 0.008 0.008
#> GSM614397 1 0.2813 0.6700 0.896 0.000 0.080 0.024
#> GSM614398 1 0.2402 0.6867 0.912 0.000 0.076 0.012
#> GSM614399 2 0.6874 0.4560 0.084 0.560 0.344 0.012
#> GSM614400 2 0.6914 0.4564 0.088 0.560 0.340 0.012
#> GSM614401 2 0.6914 0.4564 0.088 0.560 0.340 0.012
#> GSM614402 2 0.6874 0.4560 0.084 0.560 0.344 0.012
#> GSM614403 2 0.6976 0.3794 0.068 0.524 0.388 0.020
#> GSM614404 2 0.6914 0.4564 0.088 0.560 0.340 0.012
#> GSM614405 2 0.7164 0.3929 0.076 0.520 0.380 0.024
#> GSM614406 3 0.8984 0.3770 0.076 0.284 0.428 0.212
#> GSM614407 3 0.8338 -0.0792 0.356 0.188 0.424 0.032
#> GSM614408 3 0.8338 -0.0792 0.356 0.188 0.424 0.032
#> GSM614409 3 0.8314 -0.0794 0.356 0.184 0.428 0.032
#> GSM614410 3 0.8338 -0.0792 0.356 0.188 0.424 0.032
#> GSM614411 3 0.8314 -0.0794 0.356 0.184 0.428 0.032
#> GSM614412 3 0.8252 -0.0982 0.368 0.172 0.428 0.032
#> GSM614413 3 0.7613 -0.1215 0.352 0.048 0.520 0.080
#> GSM614414 3 0.7601 -0.1200 0.348 0.048 0.524 0.080
#> GSM614445 2 0.6964 0.0165 0.052 0.496 0.424 0.028
#> GSM614446 3 0.7134 0.1090 0.052 0.440 0.472 0.036
#> GSM614447 2 0.7126 -0.0248 0.052 0.484 0.428 0.036
#> GSM614448 3 0.9162 0.4295 0.104 0.240 0.440 0.216
#> GSM614449 3 0.9040 0.4323 0.092 0.252 0.448 0.208
#> GSM614450 3 0.7355 0.2242 0.060 0.396 0.500 0.044
#> GSM614451 4 0.6870 0.4522 0.044 0.048 0.308 0.600
#> GSM614452 4 0.6870 0.4522 0.044 0.048 0.308 0.600
#> GSM614453 2 0.3229 0.6319 0.000 0.880 0.048 0.072
#> GSM614454 2 0.3229 0.6319 0.000 0.880 0.048 0.072
#> GSM614455 2 0.3229 0.6319 0.000 0.880 0.048 0.072
#> GSM614456 2 0.3229 0.6319 0.000 0.880 0.048 0.072
#> GSM614457 2 0.3229 0.6319 0.000 0.880 0.048 0.072
#> GSM614458 2 0.3229 0.6319 0.000 0.880 0.048 0.072
#> GSM614459 2 0.3229 0.6319 0.000 0.880 0.048 0.072
#> GSM614460 2 0.3229 0.6319 0.000 0.880 0.048 0.072
#> GSM614461 2 0.2011 0.6678 0.000 0.920 0.080 0.000
#> GSM614462 2 0.2011 0.6678 0.000 0.920 0.080 0.000
#> GSM614463 2 0.2011 0.6678 0.000 0.920 0.080 0.000
#> GSM614464 2 0.2011 0.6678 0.000 0.920 0.080 0.000
#> GSM614465 2 0.2011 0.6678 0.000 0.920 0.080 0.000
#> GSM614466 2 0.2011 0.6678 0.000 0.920 0.080 0.000
#> GSM614467 2 0.2011 0.6678 0.000 0.920 0.080 0.000
#> GSM614468 2 0.2011 0.6678 0.000 0.920 0.080 0.000
#> GSM614469 2 0.7538 0.3520 0.228 0.520 0.248 0.004
#> GSM614470 2 0.7538 0.3520 0.228 0.520 0.248 0.004
#> GSM614471 2 0.7538 0.3520 0.228 0.520 0.248 0.004
#> GSM614472 2 0.7538 0.3520 0.228 0.520 0.248 0.004
#> GSM614473 2 0.7538 0.3520 0.228 0.520 0.248 0.004
#> GSM614474 2 0.7538 0.3520 0.228 0.520 0.248 0.004
#> GSM614475 2 0.7419 0.3760 0.200 0.536 0.260 0.004
#> GSM614476 2 0.7780 0.3158 0.196 0.500 0.292 0.012
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM614415 1 0.701 -0.0997 0.480 0.028 0.084 0.028 0.380
#> GSM614416 1 0.701 -0.0997 0.480 0.028 0.084 0.028 0.380
#> GSM614417 1 0.701 -0.0997 0.480 0.028 0.084 0.028 0.380
#> GSM614418 1 0.701 -0.0997 0.480 0.028 0.084 0.028 0.380
#> GSM614419 1 0.675 -0.1991 0.464 0.004 0.108 0.028 0.396
#> GSM614420 1 0.675 -0.1991 0.464 0.004 0.108 0.028 0.396
#> GSM614421 3 0.724 0.7277 0.076 0.068 0.624 0.144 0.088
#> GSM614422 3 0.726 0.7279 0.080 0.068 0.624 0.140 0.088
#> GSM614423 3 0.656 0.6913 0.108 0.136 0.664 0.024 0.068
#> GSM614424 3 0.724 0.7277 0.076 0.068 0.624 0.144 0.088
#> GSM614425 3 0.724 0.7277 0.076 0.068 0.624 0.144 0.088
#> GSM614426 3 0.724 0.7277 0.076 0.068 0.624 0.144 0.088
#> GSM614427 3 0.724 0.7277 0.076 0.068 0.624 0.144 0.088
#> GSM614428 3 0.729 0.7246 0.076 0.068 0.620 0.144 0.092
#> GSM614429 2 0.292 0.7895 0.028 0.892 0.028 0.048 0.004
#> GSM614430 2 0.292 0.7895 0.028 0.892 0.028 0.048 0.004
#> GSM614431 2 0.292 0.7895 0.028 0.892 0.028 0.048 0.004
#> GSM614432 2 0.292 0.7895 0.028 0.892 0.028 0.048 0.004
#> GSM614433 2 0.292 0.7895 0.028 0.892 0.028 0.048 0.004
#> GSM614434 2 0.292 0.7895 0.028 0.892 0.028 0.048 0.004
#> GSM614435 2 0.292 0.7895 0.028 0.892 0.028 0.048 0.004
#> GSM614436 2 0.336 0.7785 0.028 0.868 0.036 0.064 0.004
#> GSM614437 4 0.141 1.0000 0.000 0.044 0.008 0.948 0.000
#> GSM614438 4 0.141 1.0000 0.000 0.044 0.008 0.948 0.000
#> GSM614439 4 0.141 1.0000 0.000 0.044 0.008 0.948 0.000
#> GSM614440 4 0.141 1.0000 0.000 0.044 0.008 0.948 0.000
#> GSM614441 4 0.141 1.0000 0.000 0.044 0.008 0.948 0.000
#> GSM614442 4 0.141 1.0000 0.000 0.044 0.008 0.948 0.000
#> GSM614443 4 0.141 1.0000 0.000 0.044 0.008 0.948 0.000
#> GSM614444 4 0.141 1.0000 0.000 0.044 0.008 0.948 0.000
#> GSM614391 5 0.293 0.9279 0.164 0.000 0.004 0.000 0.832
#> GSM614392 5 0.297 0.9242 0.168 0.000 0.004 0.000 0.828
#> GSM614393 5 0.297 0.9242 0.168 0.000 0.004 0.000 0.828
#> GSM614394 5 0.289 0.9297 0.160 0.000 0.004 0.000 0.836
#> GSM614395 5 0.400 0.8114 0.060 0.000 0.072 0.040 0.828
#> GSM614396 5 0.289 0.9297 0.160 0.000 0.004 0.000 0.836
#> GSM614397 5 0.332 0.8837 0.100 0.000 0.040 0.008 0.852
#> GSM614398 5 0.290 0.9033 0.108 0.000 0.028 0.000 0.864
#> GSM614399 1 0.825 0.2246 0.320 0.320 0.280 0.016 0.064
#> GSM614400 2 0.825 -0.2842 0.320 0.320 0.280 0.016 0.064
#> GSM614401 1 0.825 0.2246 0.320 0.320 0.280 0.016 0.064
#> GSM614402 1 0.825 0.2246 0.320 0.320 0.280 0.016 0.064
#> GSM614403 3 0.814 -0.1629 0.244 0.284 0.392 0.016 0.064
#> GSM614404 1 0.825 0.2246 0.320 0.320 0.280 0.016 0.064
#> GSM614405 3 0.832 -0.2706 0.292 0.288 0.336 0.020 0.064
#> GSM614406 3 0.777 0.5645 0.088 0.144 0.580 0.108 0.080
#> GSM614407 1 0.258 0.4042 0.904 0.052 0.008 0.004 0.032
#> GSM614408 1 0.258 0.4042 0.904 0.052 0.008 0.004 0.032
#> GSM614409 1 0.267 0.4004 0.900 0.052 0.008 0.004 0.036
#> GSM614410 1 0.258 0.4042 0.904 0.052 0.008 0.004 0.032
#> GSM614411 1 0.267 0.4004 0.900 0.052 0.008 0.004 0.036
#> GSM614412 1 0.259 0.3966 0.904 0.048 0.008 0.004 0.036
#> GSM614413 1 0.550 0.1587 0.704 0.012 0.184 0.016 0.084
#> GSM614414 1 0.550 0.1587 0.704 0.012 0.184 0.016 0.084
#> GSM614445 3 0.509 0.6393 0.068 0.196 0.716 0.020 0.000
#> GSM614446 3 0.466 0.6759 0.056 0.168 0.756 0.020 0.000
#> GSM614447 3 0.509 0.6393 0.068 0.196 0.716 0.020 0.000
#> GSM614448 3 0.566 0.7154 0.032 0.076 0.736 0.116 0.040
#> GSM614449 3 0.524 0.7163 0.032 0.092 0.756 0.104 0.016
#> GSM614450 3 0.464 0.6882 0.056 0.148 0.768 0.028 0.000
#> GSM614451 3 0.603 0.3517 0.008 0.020 0.540 0.380 0.052
#> GSM614452 3 0.603 0.3517 0.008 0.020 0.540 0.380 0.052
#> GSM614453 2 0.461 0.7478 0.012 0.800 0.040 0.092 0.056
#> GSM614454 2 0.461 0.7478 0.012 0.800 0.040 0.092 0.056
#> GSM614455 2 0.461 0.7478 0.012 0.800 0.040 0.092 0.056
#> GSM614456 2 0.461 0.7478 0.012 0.800 0.040 0.092 0.056
#> GSM614457 2 0.461 0.7478 0.012 0.800 0.040 0.092 0.056
#> GSM614458 2 0.461 0.7478 0.012 0.800 0.040 0.092 0.056
#> GSM614459 2 0.461 0.7478 0.012 0.800 0.040 0.092 0.056
#> GSM614460 2 0.461 0.7478 0.012 0.800 0.040 0.092 0.056
#> GSM614461 2 0.471 0.7160 0.056 0.776 0.136 0.008 0.024
#> GSM614462 2 0.471 0.7160 0.056 0.776 0.136 0.008 0.024
#> GSM614463 2 0.471 0.7160 0.056 0.776 0.136 0.008 0.024
#> GSM614464 2 0.471 0.7160 0.056 0.776 0.136 0.008 0.024
#> GSM614465 2 0.471 0.7160 0.056 0.776 0.136 0.008 0.024
#> GSM614466 2 0.471 0.7160 0.056 0.776 0.136 0.008 0.024
#> GSM614467 2 0.464 0.7169 0.052 0.780 0.136 0.008 0.024
#> GSM614468 2 0.464 0.7169 0.052 0.780 0.136 0.008 0.024
#> GSM614469 1 0.736 0.4256 0.480 0.344 0.100 0.016 0.060
#> GSM614470 1 0.736 0.4256 0.480 0.344 0.100 0.016 0.060
#> GSM614471 1 0.736 0.4256 0.480 0.344 0.100 0.016 0.060
#> GSM614472 1 0.736 0.4256 0.480 0.344 0.100 0.016 0.060
#> GSM614473 1 0.736 0.4256 0.480 0.344 0.100 0.016 0.060
#> GSM614474 1 0.736 0.4256 0.480 0.344 0.100 0.016 0.060
#> GSM614475 1 0.735 0.4050 0.472 0.348 0.116 0.016 0.048
#> GSM614476 1 0.781 0.3931 0.448 0.320 0.156 0.024 0.052
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM614415 5 0.775 0.382 0.228 0.012 0.060 0.028 0.368 0.304
#> GSM614416 5 0.775 0.382 0.228 0.012 0.060 0.028 0.368 0.304
#> GSM614417 5 0.775 0.382 0.228 0.012 0.060 0.028 0.368 0.304
#> GSM614418 5 0.775 0.382 0.228 0.012 0.060 0.028 0.368 0.304
#> GSM614419 5 0.766 0.400 0.212 0.008 0.064 0.028 0.388 0.300
#> GSM614420 5 0.766 0.400 0.212 0.008 0.064 0.028 0.388 0.300
#> GSM614421 3 0.438 0.806 0.012 0.028 0.804 0.056 0.064 0.036
#> GSM614422 3 0.438 0.806 0.012 0.028 0.804 0.056 0.064 0.036
#> GSM614423 3 0.489 0.758 0.052 0.060 0.772 0.008 0.056 0.052
#> GSM614424 3 0.438 0.806 0.012 0.028 0.804 0.056 0.064 0.036
#> GSM614425 3 0.438 0.806 0.012 0.028 0.804 0.056 0.064 0.036
#> GSM614426 3 0.438 0.806 0.012 0.028 0.804 0.056 0.064 0.036
#> GSM614427 3 0.438 0.806 0.012 0.028 0.804 0.056 0.064 0.036
#> GSM614428 3 0.445 0.804 0.012 0.028 0.800 0.056 0.064 0.040
#> GSM614429 2 0.325 0.746 0.036 0.856 0.052 0.000 0.004 0.052
#> GSM614430 2 0.325 0.746 0.036 0.856 0.052 0.000 0.004 0.052
#> GSM614431 2 0.325 0.746 0.036 0.856 0.052 0.000 0.004 0.052
#> GSM614432 2 0.325 0.746 0.036 0.856 0.052 0.000 0.004 0.052
#> GSM614433 2 0.325 0.746 0.036 0.856 0.052 0.000 0.004 0.052
#> GSM614434 2 0.325 0.746 0.036 0.856 0.052 0.000 0.004 0.052
#> GSM614435 2 0.325 0.746 0.036 0.856 0.052 0.000 0.004 0.052
#> GSM614436 2 0.331 0.745 0.036 0.852 0.056 0.000 0.004 0.052
#> GSM614437 4 0.170 0.998 0.000 0.024 0.048 0.928 0.000 0.000
#> GSM614438 4 0.170 0.998 0.000 0.024 0.048 0.928 0.000 0.000
#> GSM614439 4 0.184 0.998 0.000 0.024 0.048 0.924 0.000 0.004
#> GSM614440 4 0.184 0.998 0.000 0.024 0.048 0.924 0.000 0.004
#> GSM614441 4 0.184 0.998 0.000 0.024 0.048 0.924 0.000 0.004
#> GSM614442 4 0.170 0.998 0.000 0.024 0.048 0.928 0.000 0.000
#> GSM614443 4 0.170 0.998 0.000 0.024 0.048 0.928 0.000 0.000
#> GSM614444 4 0.184 0.998 0.000 0.024 0.048 0.924 0.000 0.004
#> GSM614391 5 0.135 0.653 0.024 0.000 0.016 0.000 0.952 0.008
#> GSM614392 5 0.135 0.653 0.024 0.000 0.016 0.000 0.952 0.008
#> GSM614393 5 0.135 0.653 0.024 0.000 0.016 0.000 0.952 0.008
#> GSM614394 5 0.117 0.652 0.016 0.000 0.016 0.000 0.960 0.008
#> GSM614395 5 0.361 0.591 0.024 0.000 0.064 0.024 0.840 0.048
#> GSM614396 5 0.132 0.652 0.016 0.000 0.016 0.004 0.956 0.008
#> GSM614397 5 0.271 0.621 0.020 0.000 0.040 0.016 0.892 0.032
#> GSM614398 5 0.256 0.625 0.020 0.000 0.036 0.016 0.900 0.028
#> GSM614399 1 0.377 0.639 0.792 0.152 0.040 0.000 0.012 0.004
#> GSM614400 1 0.377 0.639 0.792 0.152 0.040 0.000 0.012 0.004
#> GSM614401 1 0.377 0.639 0.792 0.152 0.040 0.000 0.012 0.004
#> GSM614402 1 0.377 0.639 0.792 0.152 0.040 0.000 0.012 0.004
#> GSM614403 1 0.445 0.585 0.756 0.120 0.100 0.000 0.020 0.004
#> GSM614404 1 0.377 0.639 0.792 0.152 0.040 0.000 0.012 0.004
#> GSM614405 1 0.408 0.618 0.784 0.128 0.064 0.000 0.020 0.004
#> GSM614406 1 0.658 -0.149 0.492 0.048 0.368 0.028 0.044 0.020
#> GSM614407 6 0.528 0.880 0.192 0.032 0.020 0.000 0.068 0.688
#> GSM614408 6 0.528 0.880 0.192 0.032 0.020 0.000 0.068 0.688
#> GSM614409 6 0.528 0.880 0.192 0.032 0.020 0.000 0.068 0.688
#> GSM614410 6 0.528 0.880 0.192 0.032 0.020 0.000 0.068 0.688
#> GSM614411 6 0.528 0.880 0.192 0.032 0.020 0.000 0.068 0.688
#> GSM614412 6 0.528 0.880 0.192 0.032 0.020 0.000 0.068 0.688
#> GSM614413 6 0.589 0.639 0.060 0.008 0.176 0.008 0.096 0.652
#> GSM614414 6 0.589 0.639 0.060 0.008 0.176 0.008 0.096 0.652
#> GSM614445 3 0.493 0.710 0.168 0.080 0.720 0.012 0.008 0.012
#> GSM614446 3 0.469 0.733 0.152 0.072 0.744 0.012 0.008 0.012
#> GSM614447 3 0.488 0.715 0.168 0.076 0.724 0.012 0.008 0.012
#> GSM614448 3 0.399 0.773 0.116 0.036 0.808 0.016 0.012 0.012
#> GSM614449 3 0.389 0.772 0.116 0.036 0.812 0.016 0.008 0.012
#> GSM614450 3 0.440 0.749 0.144 0.056 0.768 0.012 0.008 0.012
#> GSM614451 3 0.588 0.513 0.048 0.008 0.616 0.264 0.024 0.040
#> GSM614452 3 0.588 0.513 0.048 0.008 0.616 0.264 0.024 0.040
#> GSM614453 2 0.264 0.713 0.008 0.892 0.004 0.032 0.008 0.056
#> GSM614454 2 0.264 0.713 0.008 0.892 0.004 0.032 0.008 0.056
#> GSM614455 2 0.264 0.713 0.008 0.892 0.004 0.032 0.008 0.056
#> GSM614456 2 0.264 0.713 0.008 0.892 0.004 0.032 0.008 0.056
#> GSM614457 2 0.264 0.713 0.008 0.892 0.004 0.032 0.008 0.056
#> GSM614458 2 0.264 0.713 0.008 0.892 0.004 0.032 0.008 0.056
#> GSM614459 2 0.264 0.713 0.008 0.892 0.004 0.032 0.008 0.056
#> GSM614460 2 0.264 0.713 0.008 0.892 0.004 0.032 0.008 0.056
#> GSM614461 2 0.584 0.578 0.276 0.588 0.076 0.004 0.000 0.056
#> GSM614462 2 0.584 0.578 0.276 0.588 0.076 0.004 0.000 0.056
#> GSM614463 2 0.584 0.578 0.276 0.588 0.076 0.004 0.000 0.056
#> GSM614464 2 0.584 0.578 0.276 0.588 0.076 0.004 0.000 0.056
#> GSM614465 2 0.584 0.578 0.276 0.588 0.076 0.004 0.000 0.056
#> GSM614466 2 0.584 0.578 0.276 0.588 0.076 0.004 0.000 0.056
#> GSM614467 2 0.584 0.578 0.276 0.588 0.076 0.004 0.000 0.056
#> GSM614468 2 0.584 0.578 0.276 0.588 0.076 0.004 0.000 0.056
#> GSM614469 1 0.771 0.591 0.460 0.204 0.080 0.004 0.052 0.200
#> GSM614470 1 0.771 0.591 0.460 0.204 0.080 0.004 0.052 0.200
#> GSM614471 1 0.771 0.591 0.460 0.204 0.080 0.004 0.052 0.200
#> GSM614472 1 0.771 0.591 0.460 0.204 0.080 0.004 0.052 0.200
#> GSM614473 1 0.771 0.591 0.460 0.204 0.080 0.004 0.052 0.200
#> GSM614474 1 0.771 0.591 0.460 0.204 0.080 0.004 0.052 0.200
#> GSM614475 1 0.769 0.593 0.464 0.204 0.080 0.004 0.052 0.196
#> GSM614476 1 0.786 0.579 0.456 0.200 0.124 0.008 0.040 0.172
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 individual(p) protocol(p) time(p) other(p) k
#> CV:kmeans 43 NA NA NA NA 2
#> CV:kmeans 75 1.36e-19 0.369 1 0.0546 3
#> CV:kmeans 46 2.15e-15 0.720 1 0.1405 4
#> CV:kmeans 55 2.37e-24 0.985 1 0.0286 5
#> CV:kmeans 79 1.80e-57 0.999 1 0.0695 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 21168 rows and 86 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 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 0.325 0.490 0.780 0.4988 0.495 0.495
#> 3 3 0.712 0.833 0.920 0.3428 0.726 0.501
#> 4 4 0.698 0.786 0.862 0.1186 0.844 0.571
#> 5 5 0.725 0.764 0.835 0.0606 0.933 0.744
#> 6 6 0.750 0.748 0.775 0.0391 0.957 0.799
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
#> GSM614415 1 0.9732 -0.0441 0.596 0.404
#> GSM614416 1 0.9732 -0.0441 0.596 0.404
#> GSM614417 1 0.9732 -0.0441 0.596 0.404
#> GSM614418 1 0.9732 -0.0441 0.596 0.404
#> GSM614419 1 0.0000 0.5695 1.000 0.000
#> GSM614420 1 0.0000 0.5695 1.000 0.000
#> GSM614421 1 0.9087 0.5435 0.676 0.324
#> GSM614422 1 0.6148 0.5660 0.848 0.152
#> GSM614423 2 0.9754 -0.0304 0.408 0.592
#> GSM614424 1 0.9087 0.5435 0.676 0.324
#> GSM614425 1 0.9087 0.5435 0.676 0.324
#> GSM614426 1 0.9087 0.5435 0.676 0.324
#> GSM614427 1 0.9129 0.5417 0.672 0.328
#> GSM614428 1 0.9087 0.5435 0.676 0.324
#> GSM614429 2 0.0000 0.7573 0.000 1.000
#> GSM614430 2 0.0000 0.7573 0.000 1.000
#> GSM614431 2 0.0000 0.7573 0.000 1.000
#> GSM614432 2 0.0000 0.7573 0.000 1.000
#> GSM614433 2 0.0000 0.7573 0.000 1.000
#> GSM614434 2 0.0000 0.7573 0.000 1.000
#> GSM614435 2 0.0000 0.7573 0.000 1.000
#> GSM614436 2 0.9881 -0.2874 0.436 0.564
#> GSM614437 1 0.9933 0.4631 0.548 0.452
#> GSM614438 1 0.9881 0.4832 0.564 0.436
#> GSM614439 1 0.9881 0.4832 0.564 0.436
#> GSM614440 1 0.9881 0.4832 0.564 0.436
#> GSM614441 1 0.9881 0.4832 0.564 0.436
#> GSM614442 1 0.9881 0.4832 0.564 0.436
#> GSM614443 1 0.9896 0.4787 0.560 0.440
#> GSM614444 1 0.9881 0.4832 0.564 0.436
#> GSM614391 1 0.0000 0.5695 1.000 0.000
#> GSM614392 1 0.0376 0.5677 0.996 0.004
#> GSM614393 1 0.0672 0.5657 0.992 0.008
#> GSM614394 1 0.0000 0.5695 1.000 0.000
#> GSM614395 1 0.0000 0.5695 1.000 0.000
#> GSM614396 1 0.0000 0.5695 1.000 0.000
#> GSM614397 1 0.0000 0.5695 1.000 0.000
#> GSM614398 1 0.0000 0.5695 1.000 0.000
#> GSM614399 2 0.8861 0.5187 0.304 0.696
#> GSM614400 2 0.9129 0.4966 0.328 0.672
#> GSM614401 2 0.9129 0.4966 0.328 0.672
#> GSM614402 2 0.9044 0.5043 0.320 0.680
#> GSM614403 2 0.9608 0.3899 0.384 0.616
#> GSM614404 2 0.9129 0.4966 0.328 0.672
#> GSM614405 1 0.8443 0.3705 0.728 0.272
#> GSM614406 1 0.9833 0.4904 0.576 0.424
#> GSM614407 1 1.0000 -0.2617 0.504 0.496
#> GSM614408 1 1.0000 -0.2617 0.504 0.496
#> GSM614409 1 0.9922 -0.1531 0.552 0.448
#> GSM614410 1 1.0000 -0.2617 0.504 0.496
#> GSM614411 1 0.9963 -0.1908 0.536 0.464
#> GSM614412 1 0.8016 0.2989 0.756 0.244
#> GSM614413 1 0.0000 0.5695 1.000 0.000
#> GSM614414 1 0.0000 0.5695 1.000 0.000
#> GSM614445 2 0.2423 0.7315 0.040 0.960
#> GSM614446 2 0.3584 0.7097 0.068 0.932
#> GSM614447 2 0.2423 0.7315 0.040 0.960
#> GSM614448 1 0.9710 0.5066 0.600 0.400
#> GSM614449 1 0.9710 0.5066 0.600 0.400
#> GSM614450 1 0.9944 0.4439 0.544 0.456
#> GSM614451 1 0.9795 0.4975 0.584 0.416
#> GSM614452 1 0.9775 0.5000 0.588 0.412
#> GSM614453 2 0.0000 0.7573 0.000 1.000
#> GSM614454 2 0.0000 0.7573 0.000 1.000
#> GSM614455 2 0.0000 0.7573 0.000 1.000
#> GSM614456 2 0.0000 0.7573 0.000 1.000
#> GSM614457 2 0.0000 0.7573 0.000 1.000
#> GSM614458 2 0.0000 0.7573 0.000 1.000
#> GSM614459 2 0.0000 0.7573 0.000 1.000
#> GSM614460 2 0.0000 0.7573 0.000 1.000
#> GSM614461 2 0.0000 0.7573 0.000 1.000
#> GSM614462 2 0.0000 0.7573 0.000 1.000
#> GSM614463 2 0.0000 0.7573 0.000 1.000
#> GSM614464 2 0.0000 0.7573 0.000 1.000
#> GSM614465 2 0.0000 0.7573 0.000 1.000
#> GSM614466 2 0.0000 0.7573 0.000 1.000
#> GSM614467 2 0.0000 0.7573 0.000 1.000
#> GSM614468 2 0.0000 0.7573 0.000 1.000
#> GSM614469 2 0.9866 0.3761 0.432 0.568
#> GSM614470 2 0.9866 0.3761 0.432 0.568
#> GSM614471 2 0.9850 0.3808 0.428 0.572
#> GSM614472 2 0.9866 0.3761 0.432 0.568
#> GSM614473 2 0.9866 0.3761 0.432 0.568
#> GSM614474 2 0.9866 0.3761 0.432 0.568
#> GSM614475 2 0.9850 0.3808 0.428 0.572
#> GSM614476 1 0.2603 0.5537 0.956 0.044
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM614415 1 0.0000 0.911 1.000 0.000 0.000
#> GSM614416 1 0.0000 0.911 1.000 0.000 0.000
#> GSM614417 1 0.0000 0.911 1.000 0.000 0.000
#> GSM614418 1 0.0000 0.911 1.000 0.000 0.000
#> GSM614419 1 0.0000 0.911 1.000 0.000 0.000
#> GSM614420 1 0.0000 0.911 1.000 0.000 0.000
#> GSM614421 3 0.0000 0.907 0.000 0.000 1.000
#> GSM614422 3 0.0000 0.907 0.000 0.000 1.000
#> GSM614423 3 0.5667 0.745 0.060 0.140 0.800
#> GSM614424 3 0.0000 0.907 0.000 0.000 1.000
#> GSM614425 3 0.0000 0.907 0.000 0.000 1.000
#> GSM614426 3 0.0000 0.907 0.000 0.000 1.000
#> GSM614427 3 0.0000 0.907 0.000 0.000 1.000
#> GSM614428 3 0.0000 0.907 0.000 0.000 1.000
#> GSM614429 2 0.0000 0.915 0.000 1.000 0.000
#> GSM614430 2 0.0000 0.915 0.000 1.000 0.000
#> GSM614431 2 0.0000 0.915 0.000 1.000 0.000
#> GSM614432 2 0.0000 0.915 0.000 1.000 0.000
#> GSM614433 2 0.0000 0.915 0.000 1.000 0.000
#> GSM614434 2 0.0000 0.915 0.000 1.000 0.000
#> GSM614435 2 0.0000 0.915 0.000 1.000 0.000
#> GSM614436 3 0.6225 0.297 0.000 0.432 0.568
#> GSM614437 3 0.2165 0.890 0.000 0.064 0.936
#> GSM614438 3 0.1860 0.898 0.000 0.052 0.948
#> GSM614439 3 0.1860 0.898 0.000 0.052 0.948
#> GSM614440 3 0.1860 0.898 0.000 0.052 0.948
#> GSM614441 3 0.1860 0.898 0.000 0.052 0.948
#> GSM614442 3 0.1860 0.898 0.000 0.052 0.948
#> GSM614443 3 0.1860 0.898 0.000 0.052 0.948
#> GSM614444 3 0.1860 0.898 0.000 0.052 0.948
#> GSM614391 1 0.0000 0.911 1.000 0.000 0.000
#> GSM614392 1 0.0000 0.911 1.000 0.000 0.000
#> GSM614393 1 0.0000 0.911 1.000 0.000 0.000
#> GSM614394 1 0.0000 0.911 1.000 0.000 0.000
#> GSM614395 3 0.5138 0.629 0.252 0.000 0.748
#> GSM614396 1 0.0000 0.911 1.000 0.000 0.000
#> GSM614397 3 0.6305 0.051 0.484 0.000 0.516
#> GSM614398 1 0.4555 0.713 0.800 0.000 0.200
#> GSM614399 2 0.6318 0.745 0.172 0.760 0.068
#> GSM614400 2 0.6192 0.746 0.176 0.764 0.060
#> GSM614401 2 0.6192 0.746 0.176 0.764 0.060
#> GSM614402 2 0.6138 0.750 0.172 0.768 0.060
#> GSM614403 2 0.9162 0.296 0.152 0.480 0.368
#> GSM614404 2 0.6192 0.746 0.176 0.764 0.060
#> GSM614405 3 0.8303 0.505 0.172 0.196 0.632
#> GSM614406 3 0.0000 0.907 0.000 0.000 1.000
#> GSM614407 1 0.0000 0.911 1.000 0.000 0.000
#> GSM614408 1 0.0000 0.911 1.000 0.000 0.000
#> GSM614409 1 0.0000 0.911 1.000 0.000 0.000
#> GSM614410 1 0.0000 0.911 1.000 0.000 0.000
#> GSM614411 1 0.0000 0.911 1.000 0.000 0.000
#> GSM614412 1 0.0000 0.911 1.000 0.000 0.000
#> GSM614413 1 0.5621 0.531 0.692 0.000 0.308
#> GSM614414 1 0.5138 0.635 0.748 0.000 0.252
#> GSM614445 2 0.4796 0.731 0.000 0.780 0.220
#> GSM614446 2 0.6111 0.405 0.000 0.604 0.396
#> GSM614447 2 0.5363 0.650 0.000 0.724 0.276
#> GSM614448 3 0.0000 0.907 0.000 0.000 1.000
#> GSM614449 3 0.0000 0.907 0.000 0.000 1.000
#> GSM614450 3 0.0747 0.901 0.000 0.016 0.984
#> GSM614451 3 0.0000 0.907 0.000 0.000 1.000
#> GSM614452 3 0.0000 0.907 0.000 0.000 1.000
#> GSM614453 2 0.0000 0.915 0.000 1.000 0.000
#> GSM614454 2 0.0000 0.915 0.000 1.000 0.000
#> GSM614455 2 0.0000 0.915 0.000 1.000 0.000
#> GSM614456 2 0.0000 0.915 0.000 1.000 0.000
#> GSM614457 2 0.0000 0.915 0.000 1.000 0.000
#> GSM614458 2 0.0000 0.915 0.000 1.000 0.000
#> GSM614459 2 0.0000 0.915 0.000 1.000 0.000
#> GSM614460 2 0.0000 0.915 0.000 1.000 0.000
#> GSM614461 2 0.0237 0.915 0.000 0.996 0.004
#> GSM614462 2 0.0237 0.915 0.000 0.996 0.004
#> GSM614463 2 0.0237 0.915 0.000 0.996 0.004
#> GSM614464 2 0.0237 0.915 0.000 0.996 0.004
#> GSM614465 2 0.0237 0.915 0.000 0.996 0.004
#> GSM614466 2 0.0237 0.915 0.000 0.996 0.004
#> GSM614467 2 0.0424 0.913 0.000 0.992 0.008
#> GSM614468 2 0.0237 0.915 0.000 0.996 0.004
#> GSM614469 1 0.3619 0.838 0.864 0.136 0.000
#> GSM614470 1 0.3619 0.838 0.864 0.136 0.000
#> GSM614471 1 0.3686 0.834 0.860 0.140 0.000
#> GSM614472 1 0.3619 0.838 0.864 0.136 0.000
#> GSM614473 1 0.3619 0.838 0.864 0.136 0.000
#> GSM614474 1 0.3619 0.838 0.864 0.136 0.000
#> GSM614475 1 0.3686 0.834 0.860 0.140 0.000
#> GSM614476 1 0.6451 0.390 0.608 0.008 0.384
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM614415 1 0.0921 0.924 0.972 0.000 0.000 0.028
#> GSM614416 1 0.0921 0.924 0.972 0.000 0.000 0.028
#> GSM614417 1 0.0921 0.924 0.972 0.000 0.000 0.028
#> GSM614418 1 0.0921 0.924 0.972 0.000 0.000 0.028
#> GSM614419 1 0.0336 0.925 0.992 0.000 0.000 0.008
#> GSM614420 1 0.0469 0.925 0.988 0.000 0.000 0.012
#> GSM614421 3 0.2489 0.822 0.020 0.000 0.912 0.068
#> GSM614422 3 0.2563 0.821 0.020 0.000 0.908 0.072
#> GSM614423 3 0.6429 0.241 0.024 0.028 0.528 0.420
#> GSM614424 3 0.2563 0.821 0.020 0.000 0.908 0.072
#> GSM614425 3 0.2563 0.821 0.020 0.000 0.908 0.072
#> GSM614426 3 0.2563 0.821 0.020 0.000 0.908 0.072
#> GSM614427 3 0.2413 0.823 0.020 0.000 0.916 0.064
#> GSM614428 3 0.2256 0.824 0.020 0.000 0.924 0.056
#> GSM614429 2 0.0817 0.898 0.000 0.976 0.000 0.024
#> GSM614430 2 0.0921 0.897 0.000 0.972 0.000 0.028
#> GSM614431 2 0.1022 0.896 0.000 0.968 0.000 0.032
#> GSM614432 2 0.0921 0.897 0.000 0.972 0.000 0.028
#> GSM614433 2 0.0921 0.897 0.000 0.972 0.000 0.028
#> GSM614434 2 0.0921 0.897 0.000 0.972 0.000 0.028
#> GSM614435 2 0.0469 0.897 0.000 0.988 0.000 0.012
#> GSM614436 2 0.4464 0.644 0.000 0.768 0.208 0.024
#> GSM614437 3 0.4415 0.778 0.000 0.140 0.804 0.056
#> GSM614438 3 0.3858 0.806 0.000 0.100 0.844 0.056
#> GSM614439 3 0.3858 0.806 0.000 0.100 0.844 0.056
#> GSM614440 3 0.3858 0.806 0.000 0.100 0.844 0.056
#> GSM614441 3 0.3858 0.806 0.000 0.100 0.844 0.056
#> GSM614442 3 0.3858 0.806 0.000 0.100 0.844 0.056
#> GSM614443 3 0.4259 0.787 0.000 0.128 0.816 0.056
#> GSM614444 3 0.3858 0.806 0.000 0.100 0.844 0.056
#> GSM614391 1 0.0000 0.924 1.000 0.000 0.000 0.000
#> GSM614392 1 0.0000 0.924 1.000 0.000 0.000 0.000
#> GSM614393 1 0.0000 0.924 1.000 0.000 0.000 0.000
#> GSM614394 1 0.0188 0.922 0.996 0.000 0.004 0.000
#> GSM614395 3 0.4830 0.344 0.392 0.000 0.608 0.000
#> GSM614396 1 0.0188 0.922 0.996 0.000 0.000 0.004
#> GSM614397 1 0.3402 0.773 0.832 0.000 0.164 0.004
#> GSM614398 1 0.2737 0.837 0.888 0.000 0.104 0.008
#> GSM614399 4 0.2965 0.733 0.036 0.072 0.000 0.892
#> GSM614400 4 0.2965 0.733 0.036 0.072 0.000 0.892
#> GSM614401 4 0.2892 0.734 0.036 0.068 0.000 0.896
#> GSM614402 4 0.2965 0.733 0.036 0.072 0.000 0.892
#> GSM614403 4 0.3072 0.665 0.008 0.024 0.076 0.892
#> GSM614404 4 0.2965 0.733 0.036 0.072 0.000 0.892
#> GSM614405 4 0.3474 0.671 0.024 0.012 0.092 0.872
#> GSM614406 3 0.3852 0.758 0.000 0.008 0.800 0.192
#> GSM614407 1 0.2216 0.896 0.908 0.000 0.000 0.092
#> GSM614408 1 0.2281 0.893 0.904 0.000 0.000 0.096
#> GSM614409 1 0.2149 0.898 0.912 0.000 0.000 0.088
#> GSM614410 1 0.2281 0.893 0.904 0.000 0.000 0.096
#> GSM614411 1 0.2149 0.898 0.912 0.000 0.000 0.088
#> GSM614412 1 0.1867 0.907 0.928 0.000 0.000 0.072
#> GSM614413 1 0.4088 0.793 0.820 0.000 0.140 0.040
#> GSM614414 1 0.3876 0.811 0.836 0.000 0.124 0.040
#> GSM614445 4 0.7486 0.311 0.000 0.272 0.228 0.500
#> GSM614446 4 0.7483 0.129 0.000 0.184 0.360 0.456
#> GSM614447 4 0.7433 0.301 0.000 0.216 0.276 0.508
#> GSM614448 3 0.3172 0.771 0.000 0.000 0.840 0.160
#> GSM614449 3 0.3311 0.761 0.000 0.000 0.828 0.172
#> GSM614450 3 0.4406 0.603 0.000 0.000 0.700 0.300
#> GSM614451 3 0.0000 0.827 0.000 0.000 1.000 0.000
#> GSM614452 3 0.0000 0.827 0.000 0.000 1.000 0.000
#> GSM614453 2 0.0188 0.895 0.000 0.996 0.000 0.004
#> GSM614454 2 0.0000 0.894 0.000 1.000 0.000 0.000
#> GSM614455 2 0.0188 0.895 0.000 0.996 0.000 0.004
#> GSM614456 2 0.0000 0.894 0.000 1.000 0.000 0.000
#> GSM614457 2 0.0000 0.894 0.000 1.000 0.000 0.000
#> GSM614458 2 0.0000 0.894 0.000 1.000 0.000 0.000
#> GSM614459 2 0.0188 0.891 0.000 0.996 0.000 0.004
#> GSM614460 2 0.0000 0.894 0.000 1.000 0.000 0.000
#> GSM614461 2 0.3726 0.809 0.000 0.788 0.000 0.212
#> GSM614462 2 0.3726 0.809 0.000 0.788 0.000 0.212
#> GSM614463 2 0.3726 0.809 0.000 0.788 0.000 0.212
#> GSM614464 2 0.3726 0.809 0.000 0.788 0.000 0.212
#> GSM614465 2 0.3726 0.809 0.000 0.788 0.000 0.212
#> GSM614466 2 0.3764 0.804 0.000 0.784 0.000 0.216
#> GSM614467 2 0.3688 0.814 0.000 0.792 0.000 0.208
#> GSM614468 2 0.3726 0.809 0.000 0.788 0.000 0.212
#> GSM614469 4 0.5282 0.669 0.276 0.036 0.000 0.688
#> GSM614470 4 0.5282 0.669 0.276 0.036 0.000 0.688
#> GSM614471 4 0.5282 0.669 0.276 0.036 0.000 0.688
#> GSM614472 4 0.5282 0.669 0.276 0.036 0.000 0.688
#> GSM614473 4 0.5282 0.669 0.276 0.036 0.000 0.688
#> GSM614474 4 0.5282 0.669 0.276 0.036 0.000 0.688
#> GSM614475 4 0.5282 0.669 0.276 0.036 0.000 0.688
#> GSM614476 4 0.6759 0.617 0.128 0.016 0.208 0.648
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM614415 5 0.1408 0.835 0.044 0.000 0.000 0.008 0.948
#> GSM614416 5 0.1408 0.835 0.044 0.000 0.000 0.008 0.948
#> GSM614417 5 0.1408 0.835 0.044 0.000 0.000 0.008 0.948
#> GSM614418 5 0.1408 0.835 0.044 0.000 0.000 0.008 0.948
#> GSM614419 5 0.0865 0.839 0.024 0.000 0.000 0.004 0.972
#> GSM614420 5 0.0865 0.839 0.024 0.000 0.000 0.004 0.972
#> GSM614421 3 0.0290 0.754 0.000 0.000 0.992 0.008 0.000
#> GSM614422 3 0.0510 0.754 0.000 0.000 0.984 0.016 0.000
#> GSM614423 3 0.3170 0.715 0.036 0.012 0.872 0.076 0.004
#> GSM614424 3 0.0162 0.756 0.000 0.000 0.996 0.004 0.000
#> GSM614425 3 0.0290 0.754 0.000 0.000 0.992 0.008 0.000
#> GSM614426 3 0.0162 0.756 0.000 0.000 0.996 0.004 0.000
#> GSM614427 3 0.0290 0.755 0.000 0.000 0.992 0.008 0.000
#> GSM614428 3 0.0703 0.744 0.000 0.000 0.976 0.024 0.000
#> GSM614429 2 0.0324 0.863 0.000 0.992 0.004 0.004 0.000
#> GSM614430 2 0.0324 0.863 0.000 0.992 0.004 0.004 0.000
#> GSM614431 2 0.0162 0.863 0.000 0.996 0.004 0.000 0.000
#> GSM614432 2 0.0324 0.863 0.000 0.992 0.004 0.004 0.000
#> GSM614433 2 0.0162 0.863 0.000 0.996 0.004 0.000 0.000
#> GSM614434 2 0.0162 0.863 0.000 0.996 0.004 0.000 0.000
#> GSM614435 2 0.0486 0.862 0.004 0.988 0.004 0.004 0.000
#> GSM614436 2 0.4806 0.486 0.000 0.688 0.060 0.252 0.000
#> GSM614437 4 0.4431 0.911 0.000 0.052 0.216 0.732 0.000
#> GSM614438 4 0.4378 0.941 0.000 0.036 0.248 0.716 0.000
#> GSM614439 4 0.4378 0.941 0.000 0.036 0.248 0.716 0.000
#> GSM614440 4 0.4378 0.941 0.000 0.036 0.248 0.716 0.000
#> GSM614441 4 0.4378 0.941 0.000 0.036 0.248 0.716 0.000
#> GSM614442 4 0.4378 0.941 0.000 0.036 0.248 0.716 0.000
#> GSM614443 4 0.4424 0.921 0.000 0.048 0.224 0.728 0.000
#> GSM614444 4 0.4378 0.941 0.000 0.036 0.248 0.716 0.000
#> GSM614391 5 0.0798 0.837 0.000 0.000 0.008 0.016 0.976
#> GSM614392 5 0.0671 0.837 0.000 0.000 0.004 0.016 0.980
#> GSM614393 5 0.0671 0.837 0.000 0.000 0.004 0.016 0.980
#> GSM614394 5 0.1211 0.833 0.000 0.000 0.024 0.016 0.960
#> GSM614395 5 0.6212 0.115 0.000 0.000 0.160 0.324 0.516
#> GSM614396 5 0.1300 0.831 0.000 0.000 0.028 0.016 0.956
#> GSM614397 5 0.2974 0.785 0.000 0.000 0.052 0.080 0.868
#> GSM614398 5 0.2012 0.816 0.000 0.000 0.060 0.020 0.920
#> GSM614399 1 0.1282 0.786 0.952 0.044 0.004 0.000 0.000
#> GSM614400 1 0.1205 0.787 0.956 0.040 0.004 0.000 0.000
#> GSM614401 1 0.1205 0.787 0.956 0.040 0.004 0.000 0.000
#> GSM614402 1 0.1282 0.786 0.952 0.044 0.004 0.000 0.000
#> GSM614403 1 0.3397 0.678 0.852 0.020 0.108 0.012 0.008
#> GSM614404 1 0.1282 0.786 0.952 0.044 0.004 0.000 0.000
#> GSM614405 1 0.2672 0.751 0.900 0.012 0.012 0.064 0.012
#> GSM614406 4 0.6276 0.560 0.232 0.004 0.204 0.560 0.000
#> GSM614407 5 0.5202 0.725 0.152 0.000 0.004 0.144 0.700
#> GSM614408 5 0.5048 0.725 0.152 0.000 0.000 0.144 0.704
#> GSM614409 5 0.4959 0.742 0.128 0.000 0.004 0.144 0.724
#> GSM614410 5 0.5202 0.725 0.152 0.000 0.004 0.144 0.700
#> GSM614411 5 0.5043 0.736 0.136 0.000 0.004 0.144 0.716
#> GSM614412 5 0.4991 0.753 0.120 0.000 0.008 0.144 0.728
#> GSM614413 5 0.6135 0.722 0.044 0.000 0.136 0.168 0.652
#> GSM614414 5 0.5620 0.762 0.052 0.000 0.088 0.156 0.704
#> GSM614445 3 0.6237 0.595 0.248 0.088 0.616 0.048 0.000
#> GSM614446 3 0.5620 0.636 0.240 0.040 0.664 0.056 0.000
#> GSM614447 3 0.6034 0.609 0.256 0.056 0.628 0.060 0.000
#> GSM614448 3 0.3657 0.714 0.116 0.000 0.820 0.064 0.000
#> GSM614449 3 0.3521 0.719 0.140 0.000 0.820 0.040 0.000
#> GSM614450 3 0.4204 0.692 0.196 0.000 0.756 0.048 0.000
#> GSM614451 3 0.4434 -0.320 0.004 0.000 0.536 0.460 0.000
#> GSM614452 3 0.4415 -0.267 0.004 0.000 0.552 0.444 0.000
#> GSM614453 2 0.2583 0.833 0.004 0.864 0.000 0.132 0.000
#> GSM614454 2 0.2583 0.833 0.004 0.864 0.000 0.132 0.000
#> GSM614455 2 0.2583 0.833 0.004 0.864 0.000 0.132 0.000
#> GSM614456 2 0.2583 0.833 0.004 0.864 0.000 0.132 0.000
#> GSM614457 2 0.2629 0.830 0.004 0.860 0.000 0.136 0.000
#> GSM614458 2 0.2536 0.835 0.004 0.868 0.000 0.128 0.000
#> GSM614459 2 0.2629 0.830 0.004 0.860 0.000 0.136 0.000
#> GSM614460 2 0.2629 0.830 0.004 0.860 0.000 0.136 0.000
#> GSM614461 2 0.3804 0.813 0.132 0.812 0.004 0.052 0.000
#> GSM614462 2 0.3849 0.810 0.136 0.808 0.004 0.052 0.000
#> GSM614463 2 0.3849 0.810 0.136 0.808 0.004 0.052 0.000
#> GSM614464 2 0.3849 0.810 0.136 0.808 0.004 0.052 0.000
#> GSM614465 2 0.3849 0.810 0.136 0.808 0.004 0.052 0.000
#> GSM614466 2 0.3849 0.810 0.136 0.808 0.004 0.052 0.000
#> GSM614467 2 0.3834 0.816 0.124 0.816 0.008 0.052 0.000
#> GSM614468 2 0.3804 0.813 0.132 0.812 0.004 0.052 0.000
#> GSM614469 1 0.5394 0.810 0.728 0.008 0.024 0.120 0.120
#> GSM614470 1 0.5394 0.810 0.728 0.008 0.024 0.120 0.120
#> GSM614471 1 0.5394 0.810 0.728 0.008 0.024 0.120 0.120
#> GSM614472 1 0.5394 0.810 0.728 0.008 0.024 0.120 0.120
#> GSM614473 1 0.5394 0.810 0.728 0.008 0.024 0.120 0.120
#> GSM614474 1 0.5394 0.810 0.728 0.008 0.024 0.120 0.120
#> GSM614475 1 0.5349 0.810 0.732 0.008 0.024 0.120 0.116
#> GSM614476 1 0.6124 0.761 0.672 0.004 0.112 0.156 0.056
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM614415 5 0.2734 0.752 0.008 0.000 0.004 0.000 0.840 0.148
#> GSM614416 5 0.2695 0.757 0.008 0.000 0.004 0.000 0.844 0.144
#> GSM614417 5 0.2734 0.752 0.008 0.000 0.004 0.000 0.840 0.148
#> GSM614418 5 0.2734 0.752 0.008 0.000 0.004 0.000 0.840 0.148
#> GSM614419 5 0.2488 0.774 0.004 0.000 0.008 0.000 0.864 0.124
#> GSM614420 5 0.2531 0.772 0.004 0.000 0.008 0.000 0.860 0.128
#> GSM614421 3 0.2118 0.845 0.000 0.000 0.888 0.104 0.008 0.000
#> GSM614422 3 0.2118 0.845 0.000 0.000 0.888 0.104 0.008 0.000
#> GSM614423 3 0.2261 0.823 0.020 0.004 0.916 0.036 0.008 0.016
#> GSM614424 3 0.2118 0.845 0.000 0.000 0.888 0.104 0.008 0.000
#> GSM614425 3 0.2118 0.845 0.000 0.000 0.888 0.104 0.008 0.000
#> GSM614426 3 0.2118 0.845 0.000 0.000 0.888 0.104 0.008 0.000
#> GSM614427 3 0.2165 0.843 0.000 0.000 0.884 0.108 0.008 0.000
#> GSM614428 3 0.2302 0.833 0.000 0.000 0.872 0.120 0.008 0.000
#> GSM614429 2 0.0603 0.793 0.000 0.980 0.000 0.004 0.000 0.016
#> GSM614430 2 0.0692 0.793 0.000 0.976 0.000 0.004 0.000 0.020
#> GSM614431 2 0.0837 0.793 0.004 0.972 0.000 0.004 0.000 0.020
#> GSM614432 2 0.0717 0.793 0.000 0.976 0.000 0.008 0.000 0.016
#> GSM614433 2 0.0717 0.793 0.000 0.976 0.000 0.008 0.000 0.016
#> GSM614434 2 0.0717 0.793 0.000 0.976 0.000 0.008 0.000 0.016
#> GSM614435 2 0.0603 0.793 0.000 0.980 0.000 0.004 0.000 0.016
#> GSM614436 2 0.4143 0.486 0.000 0.692 0.016 0.276 0.000 0.016
#> GSM614437 4 0.0837 0.859 0.000 0.020 0.004 0.972 0.000 0.004
#> GSM614438 4 0.0725 0.869 0.000 0.012 0.012 0.976 0.000 0.000
#> GSM614439 4 0.0725 0.869 0.000 0.012 0.012 0.976 0.000 0.000
#> GSM614440 4 0.0725 0.869 0.000 0.012 0.012 0.976 0.000 0.000
#> GSM614441 4 0.0725 0.869 0.000 0.012 0.012 0.976 0.000 0.000
#> GSM614442 4 0.0725 0.869 0.000 0.012 0.012 0.976 0.000 0.000
#> GSM614443 4 0.0748 0.862 0.000 0.016 0.004 0.976 0.000 0.004
#> GSM614444 4 0.0725 0.869 0.000 0.012 0.012 0.976 0.000 0.000
#> GSM614391 5 0.0603 0.811 0.000 0.000 0.016 0.000 0.980 0.004
#> GSM614392 5 0.0603 0.811 0.000 0.000 0.016 0.000 0.980 0.004
#> GSM614393 5 0.0508 0.811 0.000 0.000 0.012 0.000 0.984 0.004
#> GSM614394 5 0.0547 0.808 0.000 0.000 0.020 0.000 0.980 0.000
#> GSM614395 5 0.4414 0.403 0.000 0.000 0.064 0.260 0.676 0.000
#> GSM614396 5 0.0692 0.806 0.000 0.000 0.020 0.000 0.976 0.004
#> GSM614397 5 0.2333 0.723 0.000 0.000 0.040 0.060 0.896 0.004
#> GSM614398 5 0.0858 0.800 0.000 0.000 0.028 0.000 0.968 0.004
#> GSM614399 1 0.0767 0.673 0.976 0.000 0.012 0.008 0.000 0.004
#> GSM614400 1 0.0508 0.673 0.984 0.000 0.012 0.004 0.000 0.000
#> GSM614401 1 0.0363 0.672 0.988 0.000 0.012 0.000 0.000 0.000
#> GSM614402 1 0.0508 0.673 0.984 0.000 0.012 0.004 0.000 0.000
#> GSM614403 1 0.3064 0.570 0.860 0.004 0.092 0.016 0.004 0.024
#> GSM614404 1 0.0508 0.673 0.984 0.000 0.012 0.004 0.000 0.000
#> GSM614405 1 0.3069 0.620 0.868 0.000 0.044 0.056 0.008 0.024
#> GSM614406 4 0.4515 0.554 0.304 0.000 0.056 0.640 0.000 0.000
#> GSM614407 6 0.4626 0.918 0.032 0.000 0.008 0.000 0.372 0.588
#> GSM614408 6 0.4626 0.918 0.032 0.000 0.008 0.000 0.372 0.588
#> GSM614409 6 0.4598 0.918 0.028 0.000 0.008 0.000 0.388 0.576
#> GSM614410 6 0.4626 0.918 0.032 0.000 0.008 0.000 0.372 0.588
#> GSM614411 6 0.4646 0.920 0.032 0.000 0.008 0.000 0.380 0.580
#> GSM614412 6 0.4630 0.906 0.028 0.000 0.008 0.000 0.404 0.560
#> GSM614413 6 0.5644 0.730 0.004 0.000 0.068 0.024 0.432 0.472
#> GSM614414 6 0.4939 0.824 0.004 0.000 0.044 0.004 0.428 0.520
#> GSM614445 3 0.4865 0.672 0.256 0.004 0.656 0.004 0.000 0.080
#> GSM614446 3 0.4011 0.739 0.204 0.000 0.736 0.000 0.000 0.060
#> GSM614447 3 0.4568 0.697 0.236 0.004 0.684 0.000 0.000 0.076
#> GSM614448 3 0.4228 0.802 0.076 0.000 0.776 0.112 0.000 0.036
#> GSM614449 3 0.4199 0.800 0.108 0.000 0.780 0.072 0.000 0.040
#> GSM614450 3 0.3943 0.775 0.156 0.000 0.776 0.016 0.000 0.052
#> GSM614451 4 0.3892 0.397 0.000 0.000 0.352 0.640 0.004 0.004
#> GSM614452 4 0.4049 0.230 0.000 0.000 0.412 0.580 0.004 0.004
#> GSM614453 2 0.3192 0.768 0.012 0.848 0.004 0.092 0.000 0.044
#> GSM614454 2 0.3241 0.766 0.012 0.844 0.004 0.096 0.000 0.044
#> GSM614455 2 0.3241 0.766 0.012 0.844 0.004 0.096 0.000 0.044
#> GSM614456 2 0.3142 0.766 0.008 0.848 0.004 0.096 0.000 0.044
#> GSM614457 2 0.3142 0.766 0.008 0.848 0.004 0.096 0.000 0.044
#> GSM614458 2 0.3142 0.766 0.008 0.848 0.004 0.096 0.000 0.044
#> GSM614459 2 0.3142 0.766 0.008 0.848 0.004 0.096 0.000 0.044
#> GSM614460 2 0.3142 0.766 0.008 0.848 0.004 0.096 0.000 0.044
#> GSM614461 2 0.5771 0.671 0.184 0.636 0.052 0.004 0.000 0.124
#> GSM614462 2 0.5798 0.668 0.188 0.632 0.052 0.004 0.000 0.124
#> GSM614463 2 0.5798 0.668 0.188 0.632 0.052 0.004 0.000 0.124
#> GSM614464 2 0.5798 0.668 0.188 0.632 0.052 0.004 0.000 0.124
#> GSM614465 2 0.5798 0.668 0.188 0.632 0.052 0.004 0.000 0.124
#> GSM614466 2 0.5798 0.668 0.188 0.632 0.052 0.004 0.000 0.124
#> GSM614467 2 0.5771 0.672 0.184 0.636 0.052 0.004 0.000 0.124
#> GSM614468 2 0.5771 0.671 0.184 0.636 0.052 0.004 0.000 0.124
#> GSM614469 1 0.5880 0.666 0.492 0.008 0.020 0.012 0.060 0.408
#> GSM614470 1 0.5880 0.666 0.492 0.008 0.020 0.012 0.060 0.408
#> GSM614471 1 0.5880 0.666 0.492 0.008 0.020 0.012 0.060 0.408
#> GSM614472 1 0.5880 0.666 0.492 0.008 0.020 0.012 0.060 0.408
#> GSM614473 1 0.5880 0.666 0.492 0.008 0.020 0.012 0.060 0.408
#> GSM614474 1 0.5880 0.666 0.492 0.008 0.020 0.012 0.060 0.408
#> GSM614475 1 0.5874 0.662 0.492 0.012 0.020 0.012 0.052 0.412
#> GSM614476 1 0.6644 0.647 0.480 0.012 0.060 0.036 0.040 0.372
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 individual(p) protocol(p) time(p) other(p) k
#> CV:skmeans 51 9.46e-07 0.290 0.948 0.0487 2
#> CV:skmeans 81 7.24e-20 0.206 1.000 0.2878 3
#> CV:skmeans 81 1.21e-33 0.710 1.000 0.0851 4
#> CV:skmeans 82 1.86e-44 0.865 1.000 0.0129 5
#> CV:skmeans 82 1.43e-55 0.906 1.000 0.0580 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 21168 rows and 86 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 0.832 0.906 0.958 0.483 0.512 0.512
#> 3 3 0.786 0.877 0.944 0.253 0.880 0.765
#> 4 4 0.724 0.824 0.910 0.111 0.914 0.786
#> 5 5 0.787 0.858 0.910 0.063 0.962 0.886
#> 6 6 0.735 0.596 0.785 0.075 0.919 0.747
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
#> GSM614415 1 0.0000 0.932 1.000 0.000
#> GSM614416 1 0.0000 0.932 1.000 0.000
#> GSM614417 1 0.0000 0.932 1.000 0.000
#> GSM614418 1 0.0000 0.932 1.000 0.000
#> GSM614419 1 0.0376 0.930 0.996 0.004
#> GSM614420 1 0.3274 0.894 0.940 0.060
#> GSM614421 2 0.0938 0.968 0.012 0.988
#> GSM614422 1 0.9954 0.174 0.540 0.460
#> GSM614423 1 0.8144 0.679 0.748 0.252
#> GSM614424 2 0.3114 0.930 0.056 0.944
#> GSM614425 2 0.7674 0.712 0.224 0.776
#> GSM614426 2 0.0938 0.967 0.012 0.988
#> GSM614427 2 0.0672 0.970 0.008 0.992
#> GSM614428 2 0.0376 0.971 0.004 0.996
#> GSM614429 2 0.0376 0.971 0.004 0.996
#> GSM614430 2 0.0376 0.971 0.004 0.996
#> GSM614431 2 0.0376 0.971 0.004 0.996
#> GSM614432 2 0.0376 0.971 0.004 0.996
#> GSM614433 2 0.0376 0.971 0.004 0.996
#> GSM614434 2 0.0376 0.971 0.004 0.996
#> GSM614435 2 0.0376 0.971 0.004 0.996
#> GSM614436 2 0.0376 0.971 0.004 0.996
#> GSM614437 2 0.0000 0.970 0.000 1.000
#> GSM614438 2 0.0000 0.970 0.000 1.000
#> GSM614439 2 0.0000 0.970 0.000 1.000
#> GSM614440 2 0.0000 0.970 0.000 1.000
#> GSM614441 2 0.0000 0.970 0.000 1.000
#> GSM614442 2 0.0000 0.970 0.000 1.000
#> GSM614443 2 0.0000 0.970 0.000 1.000
#> GSM614444 2 0.0000 0.970 0.000 1.000
#> GSM614391 1 0.0000 0.932 1.000 0.000
#> GSM614392 1 0.0000 0.932 1.000 0.000
#> GSM614393 1 0.0000 0.932 1.000 0.000
#> GSM614394 1 0.0000 0.932 1.000 0.000
#> GSM614395 2 0.1184 0.966 0.016 0.984
#> GSM614396 1 0.0000 0.932 1.000 0.000
#> GSM614397 2 0.1843 0.957 0.028 0.972
#> GSM614398 1 0.9209 0.512 0.664 0.336
#> GSM614399 1 0.9983 0.151 0.524 0.476
#> GSM614400 1 0.1184 0.923 0.984 0.016
#> GSM614401 1 0.0000 0.932 1.000 0.000
#> GSM614402 1 0.3431 0.889 0.936 0.064
#> GSM614403 2 0.8081 0.658 0.248 0.752
#> GSM614404 1 0.0000 0.932 1.000 0.000
#> GSM614405 1 0.8443 0.643 0.728 0.272
#> GSM614406 2 0.0000 0.970 0.000 1.000
#> GSM614407 1 0.0000 0.932 1.000 0.000
#> GSM614408 1 0.0000 0.932 1.000 0.000
#> GSM614409 1 0.0000 0.932 1.000 0.000
#> GSM614410 1 0.0000 0.932 1.000 0.000
#> GSM614411 1 0.0000 0.932 1.000 0.000
#> GSM614412 1 0.5059 0.852 0.888 0.112
#> GSM614413 2 0.7139 0.757 0.196 0.804
#> GSM614414 2 0.2236 0.950 0.036 0.964
#> GSM614445 2 0.0376 0.971 0.004 0.996
#> GSM614446 2 0.8499 0.612 0.276 0.724
#> GSM614447 2 0.1184 0.965 0.016 0.984
#> GSM614448 2 0.0938 0.968 0.012 0.988
#> GSM614449 2 0.0376 0.971 0.004 0.996
#> GSM614450 2 0.2236 0.949 0.036 0.964
#> GSM614451 2 0.0376 0.971 0.004 0.996
#> GSM614452 2 0.0376 0.971 0.004 0.996
#> GSM614453 2 0.0000 0.970 0.000 1.000
#> GSM614454 2 0.0000 0.970 0.000 1.000
#> GSM614455 2 0.0000 0.970 0.000 1.000
#> GSM614456 2 0.0000 0.970 0.000 1.000
#> GSM614457 2 0.0000 0.970 0.000 1.000
#> GSM614458 2 0.0376 0.971 0.004 0.996
#> GSM614459 2 0.0000 0.970 0.000 1.000
#> GSM614460 2 0.0000 0.970 0.000 1.000
#> GSM614461 2 0.0376 0.971 0.004 0.996
#> GSM614462 2 0.0376 0.971 0.004 0.996
#> GSM614463 1 0.6343 0.801 0.840 0.160
#> GSM614464 2 0.0376 0.971 0.004 0.996
#> GSM614465 2 0.1843 0.957 0.028 0.972
#> GSM614466 2 0.5178 0.866 0.116 0.884
#> GSM614467 2 0.0376 0.971 0.004 0.996
#> GSM614468 2 0.0376 0.971 0.004 0.996
#> GSM614469 1 0.0000 0.932 1.000 0.000
#> GSM614470 1 0.0000 0.932 1.000 0.000
#> GSM614471 1 0.0000 0.932 1.000 0.000
#> GSM614472 1 0.0000 0.932 1.000 0.000
#> GSM614473 1 0.0000 0.932 1.000 0.000
#> GSM614474 1 0.0000 0.932 1.000 0.000
#> GSM614475 1 0.0000 0.932 1.000 0.000
#> GSM614476 1 0.0000 0.932 1.000 0.000
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM614415 1 0.0237 0.904 0.996 0.000 0.004
#> GSM614416 1 0.0237 0.904 0.996 0.000 0.004
#> GSM614417 1 0.0237 0.904 0.996 0.000 0.004
#> GSM614418 1 0.0237 0.904 0.996 0.000 0.004
#> GSM614419 1 0.0475 0.903 0.992 0.004 0.004
#> GSM614420 1 0.2384 0.864 0.936 0.056 0.008
#> GSM614421 2 0.0661 0.950 0.008 0.988 0.004
#> GSM614422 1 0.6509 0.117 0.524 0.472 0.004
#> GSM614423 1 0.5517 0.640 0.728 0.268 0.004
#> GSM614424 2 0.1989 0.923 0.048 0.948 0.004
#> GSM614425 2 0.4409 0.788 0.172 0.824 0.004
#> GSM614426 2 0.0983 0.947 0.016 0.980 0.004
#> GSM614427 2 0.1129 0.945 0.020 0.976 0.004
#> GSM614428 2 0.0848 0.949 0.008 0.984 0.008
#> GSM614429 2 0.0000 0.951 0.000 1.000 0.000
#> GSM614430 2 0.0000 0.951 0.000 1.000 0.000
#> GSM614431 2 0.0000 0.951 0.000 1.000 0.000
#> GSM614432 2 0.0000 0.951 0.000 1.000 0.000
#> GSM614433 2 0.0000 0.951 0.000 1.000 0.000
#> GSM614434 2 0.0000 0.951 0.000 1.000 0.000
#> GSM614435 2 0.0000 0.951 0.000 1.000 0.000
#> GSM614436 2 0.0000 0.951 0.000 1.000 0.000
#> GSM614437 3 0.0424 0.946 0.000 0.008 0.992
#> GSM614438 3 0.0424 0.946 0.000 0.008 0.992
#> GSM614439 3 0.0424 0.946 0.000 0.008 0.992
#> GSM614440 3 0.0237 0.943 0.000 0.004 0.996
#> GSM614441 3 0.0424 0.946 0.000 0.008 0.992
#> GSM614442 3 0.0424 0.946 0.000 0.008 0.992
#> GSM614443 3 0.0424 0.946 0.000 0.008 0.992
#> GSM614444 3 0.0424 0.946 0.000 0.008 0.992
#> GSM614391 1 0.0237 0.904 0.996 0.000 0.004
#> GSM614392 1 0.0237 0.904 0.996 0.000 0.004
#> GSM614393 1 0.0237 0.904 0.996 0.000 0.004
#> GSM614394 1 0.0237 0.904 0.996 0.000 0.004
#> GSM614395 3 0.5551 0.751 0.016 0.224 0.760
#> GSM614396 1 0.0237 0.904 0.996 0.000 0.004
#> GSM614397 2 0.2297 0.929 0.036 0.944 0.020
#> GSM614398 1 0.6252 0.488 0.648 0.344 0.008
#> GSM614399 1 0.6291 0.205 0.532 0.468 0.000
#> GSM614400 1 0.1031 0.894 0.976 0.024 0.000
#> GSM614401 1 0.0424 0.904 0.992 0.008 0.000
#> GSM614402 1 0.3267 0.818 0.884 0.116 0.000
#> GSM614403 2 0.5327 0.604 0.272 0.728 0.000
#> GSM614404 1 0.0424 0.904 0.992 0.008 0.000
#> GSM614405 1 0.6301 0.619 0.712 0.260 0.028
#> GSM614406 2 0.3619 0.851 0.000 0.864 0.136
#> GSM614407 1 0.0237 0.905 0.996 0.004 0.000
#> GSM614408 1 0.0000 0.904 1.000 0.000 0.000
#> GSM614409 1 0.0237 0.905 0.996 0.004 0.000
#> GSM614410 1 0.0237 0.905 0.996 0.004 0.000
#> GSM614411 1 0.0424 0.904 0.992 0.008 0.000
#> GSM614412 1 0.4062 0.772 0.836 0.164 0.000
#> GSM614413 2 0.4575 0.772 0.184 0.812 0.004
#> GSM614414 2 0.1647 0.935 0.036 0.960 0.004
#> GSM614445 2 0.0000 0.951 0.000 1.000 0.000
#> GSM614446 2 0.4978 0.722 0.216 0.780 0.004
#> GSM614447 2 0.0424 0.950 0.008 0.992 0.000
#> GSM614448 2 0.0848 0.949 0.008 0.984 0.008
#> GSM614449 2 0.0237 0.950 0.000 0.996 0.004
#> GSM614450 2 0.1129 0.945 0.020 0.976 0.004
#> GSM614451 3 0.3619 0.867 0.000 0.136 0.864
#> GSM614452 3 0.3752 0.860 0.000 0.144 0.856
#> GSM614453 2 0.0237 0.950 0.000 0.996 0.004
#> GSM614454 2 0.0592 0.948 0.000 0.988 0.012
#> GSM614455 2 0.2796 0.891 0.000 0.908 0.092
#> GSM614456 2 0.1163 0.940 0.000 0.972 0.028
#> GSM614457 2 0.0892 0.944 0.000 0.980 0.020
#> GSM614458 2 0.0000 0.951 0.000 1.000 0.000
#> GSM614459 2 0.3482 0.855 0.000 0.872 0.128
#> GSM614460 2 0.0237 0.950 0.000 0.996 0.004
#> GSM614461 2 0.0000 0.951 0.000 1.000 0.000
#> GSM614462 2 0.0237 0.951 0.004 0.996 0.000
#> GSM614463 1 0.5431 0.625 0.716 0.284 0.000
#> GSM614464 2 0.0000 0.951 0.000 1.000 0.000
#> GSM614465 2 0.1031 0.944 0.024 0.976 0.000
#> GSM614466 2 0.2261 0.908 0.068 0.932 0.000
#> GSM614467 2 0.0000 0.951 0.000 1.000 0.000
#> GSM614468 2 0.0000 0.951 0.000 1.000 0.000
#> GSM614469 1 0.0237 0.905 0.996 0.004 0.000
#> GSM614470 1 0.0237 0.905 0.996 0.004 0.000
#> GSM614471 1 0.0237 0.905 0.996 0.004 0.000
#> GSM614472 1 0.0237 0.905 0.996 0.004 0.000
#> GSM614473 1 0.0237 0.905 0.996 0.004 0.000
#> GSM614474 1 0.0237 0.905 0.996 0.004 0.000
#> GSM614475 1 0.0237 0.905 0.996 0.004 0.000
#> GSM614476 1 0.0237 0.905 0.996 0.004 0.000
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM614415 3 0.2469 0.823 0.108 0.000 0.892 0.000
#> GSM614416 3 0.2469 0.823 0.108 0.000 0.892 0.000
#> GSM614417 3 0.2469 0.823 0.108 0.000 0.892 0.000
#> GSM614418 3 0.2408 0.821 0.104 0.000 0.896 0.000
#> GSM614419 3 0.2469 0.823 0.108 0.000 0.892 0.000
#> GSM614420 3 0.2216 0.809 0.092 0.000 0.908 0.000
#> GSM614421 2 0.1890 0.914 0.008 0.936 0.056 0.000
#> GSM614422 2 0.6252 0.200 0.432 0.512 0.056 0.000
#> GSM614423 1 0.5035 0.597 0.748 0.196 0.056 0.000
#> GSM614424 2 0.2840 0.899 0.044 0.900 0.056 0.000
#> GSM614425 2 0.4465 0.807 0.144 0.800 0.056 0.000
#> GSM614426 2 0.2363 0.910 0.024 0.920 0.056 0.000
#> GSM614427 2 0.2565 0.905 0.032 0.912 0.056 0.000
#> GSM614428 2 0.2076 0.912 0.008 0.932 0.056 0.004
#> GSM614429 2 0.0000 0.929 0.000 1.000 0.000 0.000
#> GSM614430 2 0.0000 0.929 0.000 1.000 0.000 0.000
#> GSM614431 2 0.0188 0.929 0.004 0.996 0.000 0.000
#> GSM614432 2 0.0000 0.929 0.000 1.000 0.000 0.000
#> GSM614433 2 0.0188 0.929 0.004 0.996 0.000 0.000
#> GSM614434 2 0.0336 0.929 0.008 0.992 0.000 0.000
#> GSM614435 2 0.0000 0.929 0.000 1.000 0.000 0.000
#> GSM614436 2 0.0000 0.929 0.000 1.000 0.000 0.000
#> GSM614437 4 0.0000 0.927 0.000 0.000 0.000 1.000
#> GSM614438 4 0.0000 0.927 0.000 0.000 0.000 1.000
#> GSM614439 4 0.0000 0.927 0.000 0.000 0.000 1.000
#> GSM614440 4 0.0000 0.927 0.000 0.000 0.000 1.000
#> GSM614441 4 0.0000 0.927 0.000 0.000 0.000 1.000
#> GSM614442 4 0.0000 0.927 0.000 0.000 0.000 1.000
#> GSM614443 4 0.0000 0.927 0.000 0.000 0.000 1.000
#> GSM614444 4 0.0000 0.927 0.000 0.000 0.000 1.000
#> GSM614391 3 0.4907 0.450 0.420 0.000 0.580 0.000
#> GSM614392 1 0.4008 0.564 0.756 0.000 0.244 0.000
#> GSM614393 1 0.4454 0.421 0.692 0.000 0.308 0.000
#> GSM614394 3 0.4776 0.565 0.376 0.000 0.624 0.000
#> GSM614395 4 0.5747 0.701 0.008 0.140 0.120 0.732
#> GSM614396 3 0.4697 0.588 0.356 0.000 0.644 0.000
#> GSM614397 2 0.4746 0.689 0.004 0.712 0.276 0.008
#> GSM614398 3 0.5751 0.618 0.164 0.124 0.712 0.000
#> GSM614399 1 0.4830 0.370 0.608 0.392 0.000 0.000
#> GSM614400 1 0.0921 0.845 0.972 0.028 0.000 0.000
#> GSM614401 1 0.0469 0.853 0.988 0.012 0.000 0.000
#> GSM614402 1 0.2081 0.795 0.916 0.084 0.000 0.000
#> GSM614403 2 0.5300 0.515 0.308 0.664 0.028 0.000
#> GSM614404 1 0.0707 0.848 0.980 0.020 0.000 0.000
#> GSM614405 1 0.5005 0.501 0.712 0.264 0.004 0.020
#> GSM614406 2 0.3024 0.842 0.000 0.852 0.000 0.148
#> GSM614407 1 0.0000 0.856 1.000 0.000 0.000 0.000
#> GSM614408 1 0.0188 0.853 0.996 0.000 0.004 0.000
#> GSM614409 1 0.0336 0.854 0.992 0.008 0.000 0.000
#> GSM614410 1 0.0000 0.856 1.000 0.000 0.000 0.000
#> GSM614411 1 0.0336 0.855 0.992 0.008 0.000 0.000
#> GSM614412 1 0.3494 0.682 0.824 0.172 0.004 0.000
#> GSM614413 2 0.4234 0.824 0.132 0.816 0.052 0.000
#> GSM614414 2 0.2174 0.913 0.020 0.928 0.052 0.000
#> GSM614445 2 0.0188 0.929 0.000 0.996 0.004 0.000
#> GSM614446 2 0.4332 0.780 0.176 0.792 0.032 0.000
#> GSM614447 2 0.1297 0.926 0.016 0.964 0.020 0.000
#> GSM614448 2 0.1890 0.914 0.008 0.936 0.056 0.000
#> GSM614449 2 0.1743 0.914 0.004 0.940 0.056 0.000
#> GSM614450 2 0.1510 0.923 0.016 0.956 0.028 0.000
#> GSM614451 4 0.3979 0.810 0.004 0.096 0.056 0.844
#> GSM614452 4 0.4102 0.801 0.004 0.104 0.056 0.836
#> GSM614453 2 0.0000 0.929 0.000 1.000 0.000 0.000
#> GSM614454 2 0.0592 0.927 0.000 0.984 0.000 0.016
#> GSM614455 2 0.2216 0.883 0.000 0.908 0.000 0.092
#> GSM614456 2 0.0707 0.925 0.000 0.980 0.000 0.020
#> GSM614457 2 0.0817 0.924 0.000 0.976 0.000 0.024
#> GSM614458 2 0.0000 0.929 0.000 1.000 0.000 0.000
#> GSM614459 2 0.2921 0.848 0.000 0.860 0.000 0.140
#> GSM614460 2 0.0188 0.929 0.000 0.996 0.000 0.004
#> GSM614461 2 0.0000 0.929 0.000 1.000 0.000 0.000
#> GSM614462 2 0.0188 0.929 0.004 0.996 0.000 0.000
#> GSM614463 1 0.4250 0.533 0.724 0.276 0.000 0.000
#> GSM614464 2 0.0000 0.929 0.000 1.000 0.000 0.000
#> GSM614465 2 0.0817 0.923 0.024 0.976 0.000 0.000
#> GSM614466 2 0.1867 0.894 0.072 0.928 0.000 0.000
#> GSM614467 2 0.0000 0.929 0.000 1.000 0.000 0.000
#> GSM614468 2 0.0000 0.929 0.000 1.000 0.000 0.000
#> GSM614469 1 0.0000 0.856 1.000 0.000 0.000 0.000
#> GSM614470 1 0.0000 0.856 1.000 0.000 0.000 0.000
#> GSM614471 1 0.0188 0.854 0.996 0.004 0.000 0.000
#> GSM614472 1 0.0000 0.856 1.000 0.000 0.000 0.000
#> GSM614473 1 0.0000 0.856 1.000 0.000 0.000 0.000
#> GSM614474 1 0.0000 0.856 1.000 0.000 0.000 0.000
#> GSM614475 1 0.0000 0.856 1.000 0.000 0.000 0.000
#> GSM614476 1 0.0000 0.856 1.000 0.000 0.000 0.000
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM614415 3 0.0162 1.000 0.004 0.000 0.996 0.000 0.000
#> GSM614416 3 0.0162 1.000 0.004 0.000 0.996 0.000 0.000
#> GSM614417 3 0.0162 1.000 0.004 0.000 0.996 0.000 0.000
#> GSM614418 3 0.0162 1.000 0.004 0.000 0.996 0.000 0.000
#> GSM614419 3 0.0162 1.000 0.004 0.000 0.996 0.000 0.000
#> GSM614420 3 0.0162 1.000 0.004 0.000 0.996 0.000 0.000
#> GSM614421 2 0.3243 0.841 0.004 0.812 0.004 0.000 0.180
#> GSM614422 2 0.6421 0.433 0.300 0.516 0.004 0.000 0.180
#> GSM614423 1 0.5246 0.597 0.692 0.124 0.004 0.000 0.180
#> GSM614424 2 0.3670 0.835 0.020 0.796 0.004 0.000 0.180
#> GSM614425 2 0.4767 0.790 0.084 0.732 0.004 0.000 0.180
#> GSM614426 2 0.3844 0.832 0.028 0.788 0.004 0.000 0.180
#> GSM614427 2 0.3742 0.834 0.020 0.788 0.004 0.000 0.188
#> GSM614428 2 0.3644 0.836 0.008 0.800 0.004 0.008 0.180
#> GSM614429 2 0.0000 0.904 0.000 1.000 0.000 0.000 0.000
#> GSM614430 2 0.0000 0.904 0.000 1.000 0.000 0.000 0.000
#> GSM614431 2 0.0162 0.904 0.004 0.996 0.000 0.000 0.000
#> GSM614432 2 0.0000 0.904 0.000 1.000 0.000 0.000 0.000
#> GSM614433 2 0.0162 0.904 0.004 0.996 0.000 0.000 0.000
#> GSM614434 2 0.0404 0.905 0.012 0.988 0.000 0.000 0.000
#> GSM614435 2 0.0000 0.904 0.000 1.000 0.000 0.000 0.000
#> GSM614436 2 0.0162 0.904 0.000 0.996 0.000 0.000 0.004
#> GSM614437 4 0.0000 0.942 0.000 0.000 0.000 1.000 0.000
#> GSM614438 4 0.0000 0.942 0.000 0.000 0.000 1.000 0.000
#> GSM614439 4 0.0000 0.942 0.000 0.000 0.000 1.000 0.000
#> GSM614440 4 0.0000 0.942 0.000 0.000 0.000 1.000 0.000
#> GSM614441 4 0.0000 0.942 0.000 0.000 0.000 1.000 0.000
#> GSM614442 4 0.0000 0.942 0.000 0.000 0.000 1.000 0.000
#> GSM614443 4 0.0000 0.942 0.000 0.000 0.000 1.000 0.000
#> GSM614444 4 0.0000 0.942 0.000 0.000 0.000 1.000 0.000
#> GSM614391 5 0.4078 0.854 0.068 0.000 0.148 0.000 0.784
#> GSM614392 5 0.4149 0.860 0.088 0.000 0.128 0.000 0.784
#> GSM614393 5 0.4096 0.853 0.072 0.000 0.144 0.000 0.784
#> GSM614394 5 0.4117 0.860 0.096 0.000 0.116 0.000 0.788
#> GSM614395 5 0.1764 0.781 0.000 0.008 0.000 0.064 0.928
#> GSM614396 5 0.3758 0.864 0.096 0.000 0.088 0.000 0.816
#> GSM614397 5 0.2077 0.810 0.000 0.040 0.040 0.000 0.920
#> GSM614398 5 0.1597 0.818 0.012 0.000 0.048 0.000 0.940
#> GSM614399 1 0.4138 0.476 0.616 0.384 0.000 0.000 0.000
#> GSM614400 1 0.1282 0.866 0.952 0.044 0.000 0.000 0.004
#> GSM614401 1 0.0703 0.876 0.976 0.024 0.000 0.000 0.000
#> GSM614402 1 0.2233 0.816 0.892 0.104 0.000 0.000 0.004
#> GSM614403 2 0.5082 0.584 0.260 0.664 0.000 0.000 0.076
#> GSM614404 1 0.1484 0.860 0.944 0.048 0.000 0.000 0.008
#> GSM614405 1 0.4532 0.599 0.716 0.248 0.000 0.016 0.020
#> GSM614406 2 0.2930 0.828 0.000 0.832 0.000 0.164 0.004
#> GSM614407 1 0.0609 0.878 0.980 0.000 0.000 0.000 0.020
#> GSM614408 1 0.0771 0.877 0.976 0.000 0.004 0.000 0.020
#> GSM614409 1 0.0865 0.877 0.972 0.004 0.000 0.000 0.024
#> GSM614410 1 0.0609 0.878 0.980 0.000 0.000 0.000 0.020
#> GSM614411 1 0.0771 0.878 0.976 0.004 0.000 0.000 0.020
#> GSM614412 1 0.4095 0.669 0.752 0.220 0.004 0.000 0.024
#> GSM614413 2 0.4643 0.795 0.068 0.736 0.004 0.000 0.192
#> GSM614414 2 0.3087 0.861 0.008 0.836 0.004 0.000 0.152
#> GSM614445 2 0.0963 0.904 0.000 0.964 0.000 0.000 0.036
#> GSM614446 2 0.4365 0.811 0.116 0.768 0.000 0.000 0.116
#> GSM614447 2 0.2270 0.894 0.016 0.908 0.004 0.000 0.072
#> GSM614448 2 0.3317 0.841 0.004 0.804 0.004 0.000 0.188
#> GSM614449 2 0.3167 0.846 0.004 0.820 0.004 0.000 0.172
#> GSM614450 2 0.2037 0.895 0.012 0.920 0.004 0.000 0.064
#> GSM614451 4 0.3952 0.751 0.004 0.044 0.004 0.804 0.144
#> GSM614452 4 0.4134 0.736 0.004 0.052 0.004 0.792 0.148
#> GSM614453 2 0.0000 0.904 0.000 1.000 0.000 0.000 0.000
#> GSM614454 2 0.1043 0.899 0.000 0.960 0.000 0.040 0.000
#> GSM614455 2 0.2305 0.871 0.000 0.896 0.000 0.092 0.012
#> GSM614456 2 0.0703 0.903 0.000 0.976 0.000 0.024 0.000
#> GSM614457 2 0.0963 0.899 0.000 0.964 0.000 0.036 0.000
#> GSM614458 2 0.0000 0.904 0.000 1.000 0.000 0.000 0.000
#> GSM614459 2 0.2648 0.839 0.000 0.848 0.000 0.152 0.000
#> GSM614460 2 0.0510 0.904 0.000 0.984 0.000 0.016 0.000
#> GSM614461 2 0.0510 0.902 0.000 0.984 0.000 0.000 0.016
#> GSM614462 2 0.0671 0.903 0.004 0.980 0.000 0.000 0.016
#> GSM614463 1 0.4227 0.592 0.692 0.292 0.000 0.000 0.016
#> GSM614464 2 0.0510 0.902 0.000 0.984 0.000 0.000 0.016
#> GSM614465 2 0.1117 0.899 0.020 0.964 0.000 0.000 0.016
#> GSM614466 2 0.1701 0.887 0.048 0.936 0.000 0.000 0.016
#> GSM614467 2 0.0510 0.902 0.000 0.984 0.000 0.000 0.016
#> GSM614468 2 0.0703 0.904 0.000 0.976 0.000 0.000 0.024
#> GSM614469 1 0.0000 0.881 1.000 0.000 0.000 0.000 0.000
#> GSM614470 1 0.0000 0.881 1.000 0.000 0.000 0.000 0.000
#> GSM614471 1 0.0162 0.880 0.996 0.004 0.000 0.000 0.000
#> GSM614472 1 0.0000 0.881 1.000 0.000 0.000 0.000 0.000
#> GSM614473 1 0.0000 0.881 1.000 0.000 0.000 0.000 0.000
#> GSM614474 1 0.0000 0.881 1.000 0.000 0.000 0.000 0.000
#> GSM614475 1 0.0000 0.881 1.000 0.000 0.000 0.000 0.000
#> GSM614476 1 0.0162 0.880 0.996 0.000 0.004 0.000 0.000
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM614415 3 0.0000 1.0000 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM614416 3 0.0000 1.0000 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM614417 3 0.0000 1.0000 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM614418 3 0.0000 1.0000 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM614419 3 0.0000 1.0000 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM614420 3 0.0000 1.0000 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM614421 2 0.4343 0.5505 0.380 0.592 0.000 0.000 0.028 0.000
#> GSM614422 1 0.5585 -0.4096 0.460 0.444 0.000 0.000 0.028 0.068
#> GSM614423 1 0.4361 0.0867 0.760 0.088 0.000 0.000 0.028 0.124
#> GSM614424 2 0.4697 0.5362 0.392 0.568 0.000 0.000 0.028 0.012
#> GSM614425 2 0.4788 0.5178 0.396 0.560 0.000 0.000 0.028 0.016
#> GSM614426 2 0.4371 0.5385 0.392 0.580 0.000 0.000 0.028 0.000
#> GSM614427 2 0.4445 0.5449 0.396 0.572 0.000 0.000 0.032 0.000
#> GSM614428 2 0.4343 0.5505 0.380 0.592 0.000 0.000 0.028 0.000
#> GSM614429 2 0.0000 0.8029 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM614430 2 0.0146 0.8027 0.000 0.996 0.000 0.000 0.004 0.000
#> GSM614431 2 0.0291 0.8032 0.004 0.992 0.000 0.000 0.004 0.000
#> GSM614432 2 0.0000 0.8029 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM614433 2 0.0291 0.8032 0.004 0.992 0.000 0.000 0.004 0.000
#> GSM614434 2 0.0551 0.8031 0.004 0.984 0.000 0.000 0.004 0.008
#> GSM614435 2 0.0000 0.8029 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM614436 2 0.0260 0.8041 0.008 0.992 0.000 0.000 0.000 0.000
#> GSM614437 4 0.0000 0.9291 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM614438 4 0.0000 0.9291 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM614439 4 0.0000 0.9291 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM614440 4 0.0000 0.9291 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM614441 4 0.0000 0.9291 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM614442 4 0.0000 0.9291 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM614443 4 0.0000 0.9291 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM614444 4 0.0000 0.9291 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM614391 5 0.1226 0.9692 0.004 0.000 0.040 0.000 0.952 0.004
#> GSM614392 5 0.1196 0.9693 0.008 0.000 0.040 0.000 0.952 0.000
#> GSM614393 5 0.1332 0.9651 0.012 0.000 0.028 0.000 0.952 0.008
#> GSM614394 5 0.1226 0.9694 0.004 0.000 0.040 0.000 0.952 0.004
#> GSM614395 5 0.1257 0.9352 0.020 0.000 0.000 0.028 0.952 0.000
#> GSM614396 5 0.1003 0.9711 0.004 0.000 0.028 0.000 0.964 0.004
#> GSM614397 5 0.0603 0.9562 0.000 0.016 0.004 0.000 0.980 0.000
#> GSM614398 5 0.0717 0.9635 0.008 0.000 0.016 0.000 0.976 0.000
#> GSM614399 2 0.6044 -0.4651 0.368 0.380 0.000 0.000 0.000 0.252
#> GSM614400 1 0.4760 -0.0412 0.520 0.040 0.000 0.000 0.004 0.436
#> GSM614401 1 0.4520 -0.0779 0.520 0.032 0.000 0.000 0.000 0.448
#> GSM614402 1 0.5300 0.0348 0.496 0.104 0.000 0.000 0.000 0.400
#> GSM614403 2 0.5322 0.4523 0.232 0.624 0.000 0.000 0.012 0.132
#> GSM614404 1 0.4866 -0.0335 0.516 0.048 0.000 0.000 0.004 0.432
#> GSM614405 6 0.6454 -0.1992 0.340 0.252 0.000 0.012 0.004 0.392
#> GSM614406 2 0.3522 0.7053 0.044 0.784 0.000 0.172 0.000 0.000
#> GSM614407 6 0.0000 0.3930 0.000 0.000 0.000 0.000 0.000 1.000
#> GSM614408 6 0.0000 0.3930 0.000 0.000 0.000 0.000 0.000 1.000
#> GSM614409 6 0.0458 0.3872 0.016 0.000 0.000 0.000 0.000 0.984
#> GSM614410 6 0.0000 0.3930 0.000 0.000 0.000 0.000 0.000 1.000
#> GSM614411 6 0.0405 0.3887 0.004 0.008 0.000 0.000 0.000 0.988
#> GSM614412 6 0.2342 0.3099 0.020 0.088 0.004 0.000 0.000 0.888
#> GSM614413 6 0.5694 0.0538 0.312 0.184 0.000 0.000 0.000 0.504
#> GSM614414 6 0.5788 0.0483 0.276 0.224 0.000 0.000 0.000 0.500
#> GSM614445 2 0.1918 0.7973 0.088 0.904 0.000 0.000 0.008 0.000
#> GSM614446 2 0.4825 0.6341 0.320 0.620 0.000 0.000 0.016 0.044
#> GSM614447 2 0.3171 0.7576 0.204 0.784 0.000 0.000 0.012 0.000
#> GSM614448 2 0.4409 0.5620 0.380 0.588 0.000 0.000 0.032 0.000
#> GSM614449 2 0.4109 0.6122 0.328 0.648 0.000 0.000 0.024 0.000
#> GSM614450 2 0.2373 0.7808 0.104 0.880 0.000 0.000 0.008 0.008
#> GSM614451 4 0.3754 0.7089 0.212 0.016 0.000 0.756 0.016 0.000
#> GSM614452 4 0.4047 0.6712 0.244 0.016 0.000 0.720 0.020 0.000
#> GSM614453 2 0.0858 0.8002 0.028 0.968 0.000 0.000 0.004 0.000
#> GSM614454 2 0.1268 0.7998 0.008 0.952 0.000 0.036 0.004 0.000
#> GSM614455 2 0.2790 0.7798 0.032 0.868 0.000 0.088 0.012 0.000
#> GSM614456 2 0.1053 0.8035 0.012 0.964 0.000 0.020 0.004 0.000
#> GSM614457 2 0.1080 0.8001 0.004 0.960 0.000 0.032 0.004 0.000
#> GSM614458 2 0.0291 0.8025 0.004 0.992 0.000 0.000 0.004 0.000
#> GSM614459 2 0.2624 0.7457 0.004 0.844 0.000 0.148 0.004 0.000
#> GSM614460 2 0.0748 0.8031 0.004 0.976 0.000 0.016 0.004 0.000
#> GSM614461 2 0.2311 0.7686 0.104 0.880 0.000 0.000 0.016 0.000
#> GSM614462 2 0.2565 0.7674 0.104 0.872 0.000 0.000 0.016 0.008
#> GSM614463 1 0.6109 0.1306 0.480 0.320 0.000 0.000 0.016 0.184
#> GSM614464 2 0.2311 0.7686 0.104 0.880 0.000 0.000 0.016 0.000
#> GSM614465 2 0.2748 0.7596 0.120 0.856 0.000 0.000 0.016 0.008
#> GSM614466 2 0.3233 0.7429 0.132 0.828 0.000 0.000 0.016 0.024
#> GSM614467 2 0.2311 0.7686 0.104 0.880 0.000 0.000 0.016 0.000
#> GSM614468 2 0.2450 0.7698 0.116 0.868 0.000 0.000 0.016 0.000
#> GSM614469 6 0.3868 0.0887 0.492 0.000 0.000 0.000 0.000 0.508
#> GSM614470 6 0.3868 0.0887 0.492 0.000 0.000 0.000 0.000 0.508
#> GSM614471 6 0.3997 0.0795 0.488 0.004 0.000 0.000 0.000 0.508
#> GSM614472 6 0.3868 0.0887 0.492 0.000 0.000 0.000 0.000 0.508
#> GSM614473 6 0.3868 0.0887 0.492 0.000 0.000 0.000 0.000 0.508
#> GSM614474 6 0.3868 0.0887 0.492 0.000 0.000 0.000 0.000 0.508
#> GSM614475 6 0.3868 0.0887 0.492 0.000 0.000 0.000 0.000 0.508
#> GSM614476 6 0.3868 0.0805 0.496 0.000 0.000 0.000 0.000 0.504
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 individual(p) protocol(p) time(p) other(p) k
#> CV:pam 84 8.03e-09 0.0783 0.801 0.961 2
#> CV:pam 83 7.10e-16 0.0612 0.990 0.310 3
#> CV:pam 82 2.41e-23 0.2865 0.997 0.297 4
#> CV:pam 84 6.13e-40 0.3094 1.000 0.128 5
#> CV:pam 60 1.12e-23 0.3618 1.000 0.283 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 21168 rows and 86 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 5.
#>
#> Following methods can be applied to this 'ConsensusPartition' object:
#> [1] "cola_report" "collect_classes" "collect_plots"
#> [4] "collect_stats" "colnames" "compare_signatures"
#> [7] "consensus_heatmap" "dimension_reduction" "functional_enrichment"
#> [10] "get_anno_col" "get_anno" "get_classes"
#> [13] "get_consensus" "get_matrix" "get_membership"
#> [16] "get_param" "get_signatures" "get_stats"
#> [19] "is_best_k" "is_stable_k" "membership_heatmap"
#> [22] "ncol" "nrow" "plot_ecdf"
#> [25] "rownames" "select_partition_number" "show"
#> [28] "suggest_best_k" "test_to_known_factors"
collect_plots() function collects all the plots made from res for all k (number of partitions)
into one single page to provide an easy and fast comparison between different k.
collect_plots(res)
The plots are:
k and the heatmap of
predicted classes for each k.k.k.k.All the plots in panels can be made by individual functions and they are plotted later in this section.
select_partition_number() produces several plots showing different
statistics for choosing “optimized” k. There are following statistics:
k;k, the area increased is defined as \(A_k - A_{k-1}\).The detailed explanations of these statistics can be found in the cola vignette.
Generally speaking, lower PAC score, higher mean silhouette score or higher
concordance corresponds to better partition. Rand index and Jaccard index
measure how similar the current partition is compared to partition with k-1.
If they are too similar, we won't accept k is better than k-1.
select_partition_number(res)
The numeric values for all these statistics can be obtained by get_stats().
get_stats(res)
#> k 1-PAC mean_silhouette concordance area_increased Rand Jaccard
#> 2 2 0.331 0.563 0.759 0.4720 0.495 0.495
#> 3 3 0.324 0.567 0.731 0.3298 0.763 0.578
#> 4 4 0.666 0.752 0.826 0.1264 0.876 0.693
#> 5 5 0.769 0.782 0.879 0.1034 0.912 0.706
#> 6 6 0.812 0.733 0.836 0.0465 0.948 0.762
suggest_best_k() suggests the best \(k\) based on these statistics. The rules are as follows:
suggest_best_k(res)
#> [1] 5
Following shows the table of the partitions (You need to click the show/hide
code output link to see it). The membership matrix (columns with name p*)
is inferred by
clue::cl_consensus()
function with the SE method. Basically the value in the membership matrix
represents the probability to belong to a certain group. The finall class
label for an item is determined with the group with highest probability it
belongs to.
In get_classes() function, the entropy is calculated from the membership
matrix and the silhouette score is calculated from the consensus matrix.
cbind(get_classes(res, k = 2), get_membership(res, k = 2))
#> class entropy silhouette p1 p2
#> GSM614415 1 0.9833 0.5756 0.576 0.424
#> GSM614416 1 0.9833 0.5756 0.576 0.424
#> GSM614417 1 0.9833 0.5756 0.576 0.424
#> GSM614418 1 0.9833 0.5756 0.576 0.424
#> GSM614419 1 0.9833 0.5756 0.576 0.424
#> GSM614420 1 0.9833 0.5756 0.576 0.424
#> GSM614421 2 0.9922 0.7951 0.448 0.552
#> GSM614422 2 0.9944 0.7878 0.456 0.544
#> GSM614423 2 0.9944 0.7849 0.456 0.544
#> GSM614424 2 0.9922 0.7951 0.448 0.552
#> GSM614425 2 0.9944 0.7878 0.456 0.544
#> GSM614426 2 0.9922 0.7951 0.448 0.552
#> GSM614427 2 0.9909 0.7948 0.444 0.556
#> GSM614428 2 0.9922 0.7951 0.448 0.552
#> GSM614429 2 0.9866 0.7929 0.432 0.568
#> GSM614430 2 0.9866 0.7929 0.432 0.568
#> GSM614431 2 0.9881 0.7909 0.436 0.564
#> GSM614432 2 0.9881 0.7909 0.436 0.564
#> GSM614433 2 0.9922 0.7799 0.448 0.552
#> GSM614434 2 0.9866 0.7929 0.432 0.568
#> GSM614435 2 0.9866 0.7929 0.432 0.568
#> GSM614436 2 0.9866 0.7929 0.432 0.568
#> GSM614437 2 0.0376 0.4443 0.004 0.996
#> GSM614438 2 0.0376 0.4443 0.004 0.996
#> GSM614439 2 0.0376 0.4443 0.004 0.996
#> GSM614440 2 0.0376 0.4443 0.004 0.996
#> GSM614441 2 0.0376 0.4443 0.004 0.996
#> GSM614442 2 0.0376 0.4443 0.004 0.996
#> GSM614443 2 0.0376 0.4443 0.004 0.996
#> GSM614444 2 0.0376 0.4443 0.004 0.996
#> GSM614391 1 0.9833 0.5756 0.576 0.424
#> GSM614392 1 0.9833 0.5756 0.576 0.424
#> GSM614393 1 0.9833 0.5756 0.576 0.424
#> GSM614394 1 0.9833 0.5756 0.576 0.424
#> GSM614395 1 0.9833 0.5756 0.576 0.424
#> GSM614396 1 0.9833 0.5756 0.576 0.424
#> GSM614397 1 0.9833 0.5756 0.576 0.424
#> GSM614398 1 0.9833 0.5756 0.576 0.424
#> GSM614399 1 0.2043 0.5698 0.968 0.032
#> GSM614400 1 0.0376 0.5989 0.996 0.004
#> GSM614401 1 0.0376 0.5989 0.996 0.004
#> GSM614402 1 0.1414 0.5823 0.980 0.020
#> GSM614403 1 0.5294 0.4248 0.880 0.120
#> GSM614404 1 0.0376 0.5989 0.996 0.004
#> GSM614405 1 0.0672 0.5948 0.992 0.008
#> GSM614406 2 0.9881 0.7959 0.436 0.564
#> GSM614407 1 0.6148 0.6350 0.848 0.152
#> GSM614408 1 0.6247 0.6351 0.844 0.156
#> GSM614409 1 0.6148 0.6350 0.848 0.152
#> GSM614410 1 0.6247 0.6351 0.844 0.156
#> GSM614411 1 0.6247 0.6351 0.844 0.156
#> GSM614412 1 0.6247 0.6351 0.844 0.156
#> GSM614413 1 0.4022 0.6224 0.920 0.080
#> GSM614414 1 0.6247 0.6351 0.844 0.156
#> GSM614445 2 0.9922 0.7951 0.448 0.552
#> GSM614446 2 0.9922 0.7951 0.448 0.552
#> GSM614447 2 0.9909 0.7966 0.444 0.556
#> GSM614448 2 0.9909 0.7966 0.444 0.556
#> GSM614449 2 0.9909 0.7966 0.444 0.556
#> GSM614450 2 0.9922 0.7951 0.448 0.552
#> GSM614451 2 0.9909 0.7962 0.444 0.556
#> GSM614452 2 0.9909 0.7962 0.444 0.556
#> GSM614453 2 0.8661 0.7370 0.288 0.712
#> GSM614454 2 0.8661 0.7370 0.288 0.712
#> GSM614455 2 0.8661 0.7370 0.288 0.712
#> GSM614456 2 0.8661 0.7370 0.288 0.712
#> GSM614457 2 0.8661 0.7370 0.288 0.712
#> GSM614458 2 0.9248 0.7645 0.340 0.660
#> GSM614459 2 0.8661 0.7370 0.288 0.712
#> GSM614460 2 0.8661 0.7370 0.288 0.712
#> GSM614461 1 0.9896 -0.5796 0.560 0.440
#> GSM614462 1 0.8207 0.0152 0.744 0.256
#> GSM614463 1 0.8955 -0.2069 0.688 0.312
#> GSM614464 1 0.6973 0.2566 0.812 0.188
#> GSM614465 1 0.9850 -0.5513 0.572 0.428
#> GSM614466 1 0.9963 -0.6331 0.536 0.464
#> GSM614467 1 0.9686 -0.4705 0.604 0.396
#> GSM614468 1 0.9866 -0.5606 0.568 0.432
#> GSM614469 1 0.0376 0.5989 0.996 0.004
#> GSM614470 1 0.0376 0.5989 0.996 0.004
#> GSM614471 1 0.0376 0.5989 0.996 0.004
#> GSM614472 1 0.0376 0.5989 0.996 0.004
#> GSM614473 1 0.0376 0.5989 0.996 0.004
#> GSM614474 1 0.0376 0.5989 0.996 0.004
#> GSM614475 1 0.0376 0.5989 0.996 0.004
#> GSM614476 1 0.0376 0.5989 0.996 0.004
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM614415 1 0.2301 0.7267 0.936 0.004 0.060
#> GSM614416 1 0.2590 0.7258 0.924 0.004 0.072
#> GSM614417 1 0.2200 0.7286 0.940 0.004 0.056
#> GSM614418 1 0.2400 0.7256 0.932 0.004 0.064
#> GSM614419 1 0.1964 0.7256 0.944 0.000 0.056
#> GSM614420 1 0.1964 0.7256 0.944 0.000 0.056
#> GSM614421 2 0.9062 0.3338 0.136 0.452 0.412
#> GSM614422 2 0.9088 0.3533 0.140 0.464 0.396
#> GSM614423 2 0.9024 0.3246 0.132 0.448 0.420
#> GSM614424 2 0.9049 0.3521 0.136 0.464 0.400
#> GSM614425 2 0.9054 0.3465 0.136 0.460 0.404
#> GSM614426 2 0.9018 0.3371 0.132 0.456 0.412
#> GSM614427 2 0.8968 0.3447 0.128 0.464 0.408
#> GSM614428 3 0.9018 -0.3314 0.132 0.412 0.456
#> GSM614429 2 0.0661 0.5867 0.004 0.988 0.008
#> GSM614430 2 0.0661 0.5866 0.008 0.988 0.004
#> GSM614431 2 0.0747 0.5867 0.016 0.984 0.000
#> GSM614432 2 0.0237 0.5865 0.004 0.996 0.000
#> GSM614433 2 0.1950 0.5798 0.040 0.952 0.008
#> GSM614434 2 0.0475 0.5867 0.004 0.992 0.004
#> GSM614435 2 0.0475 0.5867 0.004 0.992 0.004
#> GSM614436 2 0.1905 0.5862 0.016 0.956 0.028
#> GSM614437 3 0.7276 0.7109 0.104 0.192 0.704
#> GSM614438 3 0.6974 0.7311 0.104 0.168 0.728
#> GSM614439 3 0.6974 0.7311 0.104 0.168 0.728
#> GSM614440 3 0.6974 0.7311 0.104 0.168 0.728
#> GSM614441 3 0.6974 0.7311 0.104 0.168 0.728
#> GSM614442 3 0.6974 0.7311 0.104 0.168 0.728
#> GSM614443 3 0.7228 0.7151 0.104 0.188 0.708
#> GSM614444 3 0.6974 0.7311 0.104 0.168 0.728
#> GSM614391 1 0.2537 0.7141 0.920 0.000 0.080
#> GSM614392 1 0.2448 0.7166 0.924 0.000 0.076
#> GSM614393 1 0.2959 0.6970 0.900 0.000 0.100
#> GSM614394 1 0.3038 0.6954 0.896 0.000 0.104
#> GSM614395 1 0.5706 0.4051 0.680 0.000 0.320
#> GSM614396 1 0.4121 0.6355 0.832 0.000 0.168
#> GSM614397 1 0.5678 0.4095 0.684 0.000 0.316
#> GSM614398 1 0.4062 0.6420 0.836 0.000 0.164
#> GSM614399 1 0.8007 0.7117 0.640 0.244 0.116
#> GSM614400 1 0.7717 0.7333 0.668 0.220 0.112
#> GSM614401 1 0.7782 0.7334 0.668 0.208 0.124
#> GSM614402 1 0.8137 0.7122 0.640 0.220 0.140
#> GSM614403 1 0.9501 0.3768 0.472 0.324 0.204
#> GSM614404 1 0.7762 0.7336 0.668 0.212 0.120
#> GSM614405 1 0.7710 0.7299 0.680 0.176 0.144
#> GSM614406 2 0.8494 0.3730 0.108 0.556 0.336
#> GSM614407 1 0.4779 0.7746 0.840 0.124 0.036
#> GSM614408 1 0.4749 0.7733 0.844 0.116 0.040
#> GSM614409 1 0.4137 0.7720 0.872 0.096 0.032
#> GSM614410 1 0.4677 0.7751 0.840 0.132 0.028
#> GSM614411 1 0.4449 0.7708 0.860 0.100 0.040
#> GSM614412 1 0.4256 0.7709 0.868 0.096 0.036
#> GSM614413 1 0.4807 0.7637 0.848 0.092 0.060
#> GSM614414 1 0.4725 0.7643 0.852 0.088 0.060
#> GSM614445 2 0.8869 0.3802 0.124 0.496 0.380
#> GSM614446 2 0.8848 0.3871 0.124 0.504 0.372
#> GSM614447 2 0.8683 0.4124 0.120 0.540 0.340
#> GSM614448 2 0.9014 0.3455 0.132 0.460 0.408
#> GSM614449 2 0.8991 0.3639 0.132 0.476 0.392
#> GSM614450 2 0.8991 0.3639 0.132 0.476 0.392
#> GSM614451 3 0.8843 -0.0399 0.160 0.276 0.564
#> GSM614452 3 0.8843 -0.0399 0.160 0.276 0.564
#> GSM614453 2 0.5179 0.4928 0.088 0.832 0.080
#> GSM614454 2 0.5179 0.4928 0.088 0.832 0.080
#> GSM614455 2 0.5179 0.4928 0.088 0.832 0.080
#> GSM614456 2 0.5179 0.4928 0.088 0.832 0.080
#> GSM614457 2 0.5179 0.4928 0.088 0.832 0.080
#> GSM614458 2 0.4745 0.5097 0.068 0.852 0.080
#> GSM614459 2 0.5179 0.4928 0.088 0.832 0.080
#> GSM614460 2 0.5179 0.4928 0.088 0.832 0.080
#> GSM614461 2 0.3445 0.5550 0.088 0.896 0.016
#> GSM614462 2 0.6016 0.3567 0.256 0.724 0.020
#> GSM614463 2 0.6161 0.3156 0.272 0.708 0.020
#> GSM614464 2 0.6016 0.3619 0.256 0.724 0.020
#> GSM614465 2 0.5366 0.4481 0.208 0.776 0.016
#> GSM614466 2 0.4418 0.5227 0.132 0.848 0.020
#> GSM614467 2 0.3722 0.5664 0.088 0.888 0.024
#> GSM614468 2 0.4609 0.5244 0.128 0.844 0.028
#> GSM614469 1 0.7568 0.7402 0.680 0.212 0.108
#> GSM614470 1 0.7568 0.7402 0.680 0.212 0.108
#> GSM614471 1 0.7568 0.7402 0.680 0.212 0.108
#> GSM614472 1 0.7568 0.7402 0.680 0.212 0.108
#> GSM614473 1 0.7568 0.7402 0.680 0.212 0.108
#> GSM614474 1 0.7568 0.7402 0.680 0.212 0.108
#> GSM614475 1 0.7610 0.7374 0.676 0.216 0.108
#> GSM614476 1 0.7633 0.7376 0.680 0.200 0.120
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM614415 1 0.0592 0.761 0.984 0.000 0.000 0.016
#> GSM614416 1 0.0779 0.762 0.980 0.004 0.000 0.016
#> GSM614417 1 0.0592 0.761 0.984 0.000 0.000 0.016
#> GSM614418 1 0.0592 0.761 0.984 0.000 0.000 0.016
#> GSM614419 1 0.1406 0.756 0.960 0.000 0.024 0.016
#> GSM614420 1 0.1297 0.757 0.964 0.000 0.020 0.016
#> GSM614421 3 0.1576 0.906 0.000 0.048 0.948 0.004
#> GSM614422 3 0.1743 0.899 0.004 0.056 0.940 0.000
#> GSM614423 3 0.1576 0.906 0.000 0.048 0.948 0.004
#> GSM614424 3 0.1576 0.906 0.000 0.048 0.948 0.004
#> GSM614425 3 0.1389 0.906 0.000 0.048 0.952 0.000
#> GSM614426 3 0.1576 0.906 0.000 0.048 0.948 0.004
#> GSM614427 3 0.1722 0.904 0.000 0.048 0.944 0.008
#> GSM614428 3 0.1635 0.903 0.000 0.044 0.948 0.008
#> GSM614429 2 0.2473 0.754 0.000 0.908 0.012 0.080
#> GSM614430 2 0.2376 0.758 0.000 0.916 0.016 0.068
#> GSM614431 2 0.1297 0.761 0.000 0.964 0.016 0.020
#> GSM614432 2 0.1610 0.762 0.000 0.952 0.016 0.032
#> GSM614433 2 0.0967 0.757 0.004 0.976 0.016 0.004
#> GSM614434 2 0.1706 0.762 0.000 0.948 0.016 0.036
#> GSM614435 2 0.3495 0.727 0.000 0.844 0.016 0.140
#> GSM614436 2 0.4597 0.718 0.004 0.800 0.056 0.140
#> GSM614437 4 0.0592 1.000 0.000 0.016 0.000 0.984
#> GSM614438 4 0.0592 1.000 0.000 0.016 0.000 0.984
#> GSM614439 4 0.0592 1.000 0.000 0.016 0.000 0.984
#> GSM614440 4 0.0592 1.000 0.000 0.016 0.000 0.984
#> GSM614441 4 0.0592 1.000 0.000 0.016 0.000 0.984
#> GSM614442 4 0.0592 1.000 0.000 0.016 0.000 0.984
#> GSM614443 4 0.0592 1.000 0.000 0.016 0.000 0.984
#> GSM614444 4 0.0592 1.000 0.000 0.016 0.000 0.984
#> GSM614391 1 0.1593 0.757 0.956 0.004 0.024 0.016
#> GSM614392 1 0.1593 0.757 0.956 0.004 0.024 0.016
#> GSM614393 1 0.2246 0.746 0.928 0.004 0.052 0.016
#> GSM614394 1 0.2060 0.743 0.932 0.000 0.052 0.016
#> GSM614395 1 0.2142 0.741 0.928 0.000 0.056 0.016
#> GSM614396 1 0.2142 0.741 0.928 0.000 0.056 0.016
#> GSM614397 1 0.2142 0.741 0.928 0.000 0.056 0.016
#> GSM614398 1 0.2142 0.741 0.928 0.000 0.056 0.016
#> GSM614399 1 0.6197 0.692 0.596 0.344 0.056 0.004
#> GSM614400 1 0.5990 0.704 0.608 0.336 0.056 0.000
#> GSM614401 1 0.6552 0.681 0.576 0.328 0.096 0.000
#> GSM614402 1 0.7845 0.416 0.404 0.304 0.292 0.000
#> GSM614403 3 0.7002 0.260 0.268 0.164 0.568 0.000
#> GSM614404 1 0.6170 0.700 0.600 0.332 0.068 0.000
#> GSM614405 1 0.6848 0.689 0.592 0.248 0.160 0.000
#> GSM614406 3 0.8333 -0.131 0.360 0.200 0.412 0.028
#> GSM614407 1 0.4553 0.780 0.780 0.180 0.040 0.000
#> GSM614408 1 0.4423 0.781 0.792 0.168 0.040 0.000
#> GSM614409 1 0.4595 0.781 0.780 0.176 0.044 0.000
#> GSM614410 1 0.4466 0.780 0.784 0.180 0.036 0.000
#> GSM614411 1 0.4589 0.781 0.784 0.168 0.048 0.000
#> GSM614412 1 0.4669 0.782 0.780 0.168 0.052 0.000
#> GSM614413 1 0.5304 0.758 0.748 0.104 0.148 0.000
#> GSM614414 1 0.5266 0.762 0.752 0.108 0.140 0.000
#> GSM614445 3 0.1474 0.905 0.000 0.052 0.948 0.000
#> GSM614446 3 0.1474 0.905 0.000 0.052 0.948 0.000
#> GSM614447 3 0.1824 0.898 0.004 0.060 0.936 0.000
#> GSM614448 3 0.1389 0.906 0.000 0.048 0.952 0.000
#> GSM614449 3 0.1389 0.906 0.000 0.048 0.952 0.000
#> GSM614450 3 0.1389 0.906 0.000 0.048 0.952 0.000
#> GSM614451 3 0.2363 0.818 0.056 0.000 0.920 0.024
#> GSM614452 3 0.2363 0.818 0.056 0.000 0.920 0.024
#> GSM614453 2 0.4761 0.515 0.000 0.628 0.000 0.372
#> GSM614454 2 0.4761 0.515 0.000 0.628 0.000 0.372
#> GSM614455 2 0.4761 0.515 0.000 0.628 0.000 0.372
#> GSM614456 2 0.4746 0.519 0.000 0.632 0.000 0.368
#> GSM614457 2 0.4761 0.515 0.000 0.628 0.000 0.372
#> GSM614458 2 0.4585 0.566 0.000 0.668 0.000 0.332
#> GSM614459 2 0.4761 0.515 0.000 0.628 0.000 0.372
#> GSM614460 2 0.4761 0.515 0.000 0.628 0.000 0.372
#> GSM614461 2 0.2695 0.743 0.056 0.912 0.024 0.008
#> GSM614462 2 0.3653 0.689 0.112 0.856 0.024 0.008
#> GSM614463 2 0.4033 0.631 0.148 0.824 0.020 0.008
#> GSM614464 2 0.3030 0.726 0.076 0.892 0.028 0.004
#> GSM614465 2 0.3374 0.729 0.080 0.880 0.028 0.012
#> GSM614466 2 0.2142 0.741 0.056 0.928 0.016 0.000
#> GSM614467 2 0.3225 0.742 0.060 0.892 0.032 0.016
#> GSM614468 2 0.2814 0.743 0.052 0.908 0.032 0.008
#> GSM614469 1 0.5420 0.713 0.624 0.352 0.024 0.000
#> GSM614470 1 0.5420 0.713 0.624 0.352 0.024 0.000
#> GSM614471 1 0.5436 0.710 0.620 0.356 0.024 0.000
#> GSM614472 1 0.5420 0.713 0.624 0.352 0.024 0.000
#> GSM614473 1 0.5436 0.710 0.620 0.356 0.024 0.000
#> GSM614474 1 0.5436 0.710 0.620 0.356 0.024 0.000
#> GSM614475 1 0.5436 0.710 0.620 0.356 0.024 0.000
#> GSM614476 1 0.6338 0.707 0.600 0.316 0.084 0.000
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM614415 5 0.2424 0.8002 0.132 0.000 0.000 0.000 0.868
#> GSM614416 5 0.2561 0.7890 0.144 0.000 0.000 0.000 0.856
#> GSM614417 5 0.2732 0.7729 0.160 0.000 0.000 0.000 0.840
#> GSM614418 5 0.2732 0.7729 0.160 0.000 0.000 0.000 0.840
#> GSM614419 5 0.0404 0.8613 0.012 0.000 0.000 0.000 0.988
#> GSM614420 5 0.0404 0.8613 0.012 0.000 0.000 0.000 0.988
#> GSM614421 3 0.0510 0.9498 0.000 0.016 0.984 0.000 0.000
#> GSM614422 3 0.0566 0.9461 0.004 0.012 0.984 0.000 0.000
#> GSM614423 3 0.0609 0.9478 0.000 0.020 0.980 0.000 0.000
#> GSM614424 3 0.0510 0.9498 0.000 0.016 0.984 0.000 0.000
#> GSM614425 3 0.0510 0.9498 0.000 0.016 0.984 0.000 0.000
#> GSM614426 3 0.0510 0.9498 0.000 0.016 0.984 0.000 0.000
#> GSM614427 3 0.0898 0.9428 0.008 0.020 0.972 0.000 0.000
#> GSM614428 3 0.0404 0.9476 0.000 0.012 0.988 0.000 0.000
#> GSM614429 2 0.1697 0.8709 0.060 0.932 0.008 0.000 0.000
#> GSM614430 2 0.1697 0.8709 0.060 0.932 0.008 0.000 0.000
#> GSM614431 2 0.1830 0.8710 0.068 0.924 0.008 0.000 0.000
#> GSM614432 2 0.1764 0.8709 0.064 0.928 0.008 0.000 0.000
#> GSM614433 2 0.2017 0.8687 0.080 0.912 0.008 0.000 0.000
#> GSM614434 2 0.1697 0.8709 0.060 0.932 0.008 0.000 0.000
#> GSM614435 2 0.1857 0.8703 0.060 0.928 0.008 0.004 0.000
#> GSM614436 2 0.2172 0.8685 0.060 0.916 0.020 0.004 0.000
#> GSM614437 4 0.0162 0.9968 0.000 0.004 0.000 0.996 0.000
#> GSM614438 4 0.0000 0.9989 0.000 0.000 0.000 1.000 0.000
#> GSM614439 4 0.0000 0.9989 0.000 0.000 0.000 1.000 0.000
#> GSM614440 4 0.0000 0.9989 0.000 0.000 0.000 1.000 0.000
#> GSM614441 4 0.0000 0.9989 0.000 0.000 0.000 1.000 0.000
#> GSM614442 4 0.0000 0.9989 0.000 0.000 0.000 1.000 0.000
#> GSM614443 4 0.0162 0.9968 0.000 0.004 0.000 0.996 0.000
#> GSM614444 4 0.0000 0.9989 0.000 0.000 0.000 1.000 0.000
#> GSM614391 5 0.0880 0.8614 0.032 0.000 0.000 0.000 0.968
#> GSM614392 5 0.0880 0.8614 0.032 0.000 0.000 0.000 0.968
#> GSM614393 5 0.0880 0.8613 0.032 0.000 0.000 0.000 0.968
#> GSM614394 5 0.0566 0.8611 0.012 0.000 0.004 0.000 0.984
#> GSM614395 5 0.1173 0.8521 0.012 0.000 0.020 0.004 0.964
#> GSM614396 5 0.0451 0.8593 0.008 0.000 0.004 0.000 0.988
#> GSM614397 5 0.0912 0.8555 0.012 0.000 0.016 0.000 0.972
#> GSM614398 5 0.0451 0.8593 0.008 0.000 0.004 0.000 0.988
#> GSM614399 1 0.3170 0.7195 0.828 0.008 0.160 0.000 0.004
#> GSM614400 1 0.2462 0.7314 0.880 0.008 0.112 0.000 0.000
#> GSM614401 1 0.3209 0.7043 0.812 0.008 0.180 0.000 0.000
#> GSM614402 1 0.4889 0.1421 0.504 0.016 0.476 0.000 0.004
#> GSM614403 3 0.4592 0.6163 0.224 0.012 0.728 0.000 0.036
#> GSM614404 1 0.2707 0.7271 0.860 0.008 0.132 0.000 0.000
#> GSM614405 1 0.5406 0.4423 0.592 0.008 0.348 0.000 0.052
#> GSM614406 3 0.6047 0.5025 0.260 0.028 0.636 0.016 0.060
#> GSM614407 1 0.4442 0.5189 0.688 0.000 0.028 0.000 0.284
#> GSM614408 1 0.4565 0.4795 0.664 0.000 0.028 0.000 0.308
#> GSM614409 1 0.4763 0.4392 0.632 0.000 0.032 0.000 0.336
#> GSM614410 1 0.4318 0.5168 0.688 0.000 0.020 0.000 0.292
#> GSM614411 1 0.4679 0.4677 0.652 0.000 0.032 0.000 0.316
#> GSM614412 1 0.5059 0.2538 0.548 0.000 0.036 0.000 0.416
#> GSM614413 5 0.5594 -0.0915 0.444 0.004 0.060 0.000 0.492
#> GSM614414 5 0.5486 -0.0805 0.444 0.004 0.052 0.000 0.500
#> GSM614445 3 0.0510 0.9498 0.000 0.016 0.984 0.000 0.000
#> GSM614446 3 0.0510 0.9498 0.000 0.016 0.984 0.000 0.000
#> GSM614447 3 0.0794 0.9412 0.000 0.028 0.972 0.000 0.000
#> GSM614448 3 0.0510 0.9498 0.000 0.016 0.984 0.000 0.000
#> GSM614449 3 0.0510 0.9498 0.000 0.016 0.984 0.000 0.000
#> GSM614450 3 0.0510 0.9498 0.000 0.016 0.984 0.000 0.000
#> GSM614451 3 0.1605 0.9046 0.004 0.000 0.944 0.012 0.040
#> GSM614452 3 0.1605 0.9046 0.004 0.000 0.944 0.012 0.040
#> GSM614453 2 0.2060 0.8279 0.016 0.924 0.000 0.052 0.008
#> GSM614454 2 0.1988 0.8295 0.016 0.928 0.000 0.048 0.008
#> GSM614455 2 0.2060 0.8279 0.016 0.924 0.000 0.052 0.008
#> GSM614456 2 0.2060 0.8279 0.016 0.924 0.000 0.052 0.008
#> GSM614457 2 0.2060 0.8279 0.016 0.924 0.000 0.052 0.008
#> GSM614458 2 0.1498 0.8432 0.024 0.952 0.000 0.016 0.008
#> GSM614459 2 0.2692 0.7992 0.016 0.884 0.000 0.092 0.008
#> GSM614460 2 0.2060 0.8279 0.016 0.924 0.000 0.052 0.008
#> GSM614461 2 0.3783 0.7556 0.252 0.740 0.008 0.000 0.000
#> GSM614462 2 0.4252 0.6394 0.340 0.652 0.008 0.000 0.000
#> GSM614463 2 0.4283 0.6258 0.348 0.644 0.008 0.000 0.000
#> GSM614464 2 0.4183 0.6624 0.324 0.668 0.008 0.000 0.000
#> GSM614465 2 0.3783 0.7547 0.252 0.740 0.008 0.000 0.000
#> GSM614466 2 0.3318 0.8117 0.192 0.800 0.008 0.000 0.000
#> GSM614467 2 0.2953 0.8399 0.144 0.844 0.012 0.000 0.000
#> GSM614468 2 0.3039 0.8363 0.152 0.836 0.012 0.000 0.000
#> GSM614469 1 0.0740 0.7493 0.980 0.004 0.008 0.000 0.008
#> GSM614470 1 0.0740 0.7493 0.980 0.004 0.008 0.000 0.008
#> GSM614471 1 0.0740 0.7493 0.980 0.004 0.008 0.000 0.008
#> GSM614472 1 0.0740 0.7493 0.980 0.004 0.008 0.000 0.008
#> GSM614473 1 0.0740 0.7493 0.980 0.004 0.008 0.000 0.008
#> GSM614474 1 0.0740 0.7493 0.980 0.004 0.008 0.000 0.008
#> GSM614475 1 0.1243 0.7488 0.960 0.004 0.028 0.000 0.008
#> GSM614476 1 0.5180 0.5982 0.664 0.004 0.260 0.000 0.072
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM614415 5 0.1398 0.820727 0.052 0.000 0.000 0.000 0.940 0.008
#> GSM614416 5 0.1745 0.814259 0.056 0.000 0.000 0.000 0.924 0.020
#> GSM614417 5 0.1643 0.809163 0.068 0.000 0.000 0.000 0.924 0.008
#> GSM614418 5 0.1584 0.812583 0.064 0.000 0.000 0.000 0.928 0.008
#> GSM614419 5 0.1908 0.844072 0.004 0.000 0.000 0.000 0.900 0.096
#> GSM614420 5 0.1908 0.844072 0.004 0.000 0.000 0.000 0.900 0.096
#> GSM614421 3 0.0000 0.891011 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM614422 3 0.0146 0.890603 0.000 0.000 0.996 0.000 0.000 0.004
#> GSM614423 3 0.0000 0.891011 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM614424 3 0.0000 0.891011 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM614425 3 0.0146 0.890603 0.000 0.000 0.996 0.000 0.000 0.004
#> GSM614426 3 0.0146 0.890603 0.000 0.000 0.996 0.000 0.000 0.004
#> GSM614427 3 0.0000 0.891011 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM614428 3 0.0146 0.890603 0.000 0.000 0.996 0.000 0.000 0.004
#> GSM614429 2 0.2219 0.769801 0.000 0.864 0.000 0.000 0.000 0.136
#> GSM614430 2 0.1267 0.805988 0.000 0.940 0.000 0.000 0.000 0.060
#> GSM614431 2 0.1327 0.798972 0.000 0.936 0.000 0.000 0.000 0.064
#> GSM614432 2 0.1267 0.801099 0.000 0.940 0.000 0.000 0.000 0.060
#> GSM614433 2 0.1657 0.812038 0.016 0.928 0.000 0.000 0.000 0.056
#> GSM614434 2 0.1444 0.793168 0.000 0.928 0.000 0.000 0.000 0.072
#> GSM614435 2 0.2340 0.765169 0.000 0.852 0.000 0.000 0.000 0.148
#> GSM614436 2 0.3014 0.737034 0.000 0.804 0.012 0.000 0.000 0.184
#> GSM614437 4 0.0146 0.996476 0.000 0.000 0.000 0.996 0.000 0.004
#> GSM614438 4 0.0000 0.998829 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM614439 4 0.0000 0.998829 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM614440 4 0.0000 0.998829 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM614441 4 0.0000 0.998829 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM614442 4 0.0000 0.998829 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM614443 4 0.0146 0.996476 0.000 0.000 0.000 0.996 0.000 0.004
#> GSM614444 4 0.0000 0.998829 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM614391 5 0.0692 0.838057 0.004 0.000 0.000 0.000 0.976 0.020
#> GSM614392 5 0.0806 0.837114 0.008 0.000 0.000 0.000 0.972 0.020
#> GSM614393 5 0.0622 0.841951 0.008 0.000 0.000 0.000 0.980 0.012
#> GSM614394 5 0.1610 0.842907 0.000 0.000 0.000 0.000 0.916 0.084
#> GSM614395 5 0.2378 0.818910 0.000 0.000 0.000 0.000 0.848 0.152
#> GSM614396 5 0.1765 0.839304 0.000 0.000 0.000 0.000 0.904 0.096
#> GSM614397 5 0.2378 0.818910 0.000 0.000 0.000 0.000 0.848 0.152
#> GSM614398 5 0.1765 0.839304 0.000 0.000 0.000 0.000 0.904 0.096
#> GSM614399 1 0.4906 0.484439 0.672 0.056 0.248 0.000 0.008 0.016
#> GSM614400 1 0.3352 0.593249 0.800 0.016 0.172 0.000 0.000 0.012
#> GSM614401 1 0.4319 0.530602 0.728 0.024 0.216 0.000 0.004 0.028
#> GSM614402 3 0.5859 0.080256 0.420 0.032 0.476 0.000 0.012 0.060
#> GSM614403 3 0.4799 0.663603 0.148 0.024 0.736 0.000 0.016 0.076
#> GSM614404 1 0.3658 0.573648 0.776 0.020 0.188 0.000 0.000 0.016
#> GSM614405 3 0.6825 -0.012276 0.376 0.036 0.420 0.004 0.016 0.148
#> GSM614406 3 0.6571 0.471559 0.192 0.052 0.588 0.012 0.016 0.140
#> GSM614407 1 0.4486 0.364307 0.584 0.000 0.004 0.000 0.384 0.028
#> GSM614408 1 0.4467 0.331791 0.564 0.000 0.004 0.000 0.408 0.024
#> GSM614409 1 0.4763 0.252668 0.516 0.000 0.004 0.000 0.440 0.040
#> GSM614410 1 0.4542 0.322538 0.556 0.000 0.004 0.000 0.412 0.028
#> GSM614411 1 0.4636 0.280259 0.532 0.000 0.004 0.000 0.432 0.032
#> GSM614412 1 0.5031 0.172781 0.476 0.000 0.004 0.000 0.460 0.060
#> GSM614413 5 0.6199 0.000758 0.372 0.000 0.008 0.000 0.392 0.228
#> GSM614414 5 0.6185 0.009700 0.368 0.000 0.008 0.000 0.400 0.224
#> GSM614445 3 0.0146 0.889765 0.000 0.000 0.996 0.000 0.000 0.004
#> GSM614446 3 0.0291 0.887784 0.000 0.004 0.992 0.000 0.000 0.004
#> GSM614447 3 0.0820 0.875512 0.000 0.016 0.972 0.000 0.000 0.012
#> GSM614448 3 0.0000 0.891011 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM614449 3 0.0000 0.891011 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM614450 3 0.0000 0.891011 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM614451 3 0.2215 0.841605 0.000 0.000 0.900 0.012 0.012 0.076
#> GSM614452 3 0.2114 0.843530 0.000 0.000 0.904 0.008 0.012 0.076
#> GSM614453 6 0.3578 0.941499 0.000 0.340 0.000 0.000 0.000 0.660
#> GSM614454 6 0.3774 0.925391 0.000 0.408 0.000 0.000 0.000 0.592
#> GSM614455 6 0.3659 0.968509 0.000 0.364 0.000 0.000 0.000 0.636
#> GSM614456 6 0.3706 0.962860 0.000 0.380 0.000 0.000 0.000 0.620
#> GSM614457 6 0.3659 0.971955 0.000 0.364 0.000 0.000 0.000 0.636
#> GSM614458 2 0.3741 0.117382 0.008 0.672 0.000 0.000 0.000 0.320
#> GSM614459 6 0.3659 0.971955 0.000 0.364 0.000 0.000 0.000 0.636
#> GSM614460 6 0.3684 0.968672 0.000 0.372 0.000 0.000 0.000 0.628
#> GSM614461 2 0.2145 0.785525 0.072 0.900 0.000 0.000 0.000 0.028
#> GSM614462 2 0.2956 0.722330 0.120 0.840 0.000 0.000 0.000 0.040
#> GSM614463 2 0.3094 0.702053 0.140 0.824 0.000 0.000 0.000 0.036
#> GSM614464 2 0.2560 0.759918 0.092 0.872 0.000 0.000 0.000 0.036
#> GSM614465 2 0.1789 0.801693 0.044 0.924 0.000 0.000 0.000 0.032
#> GSM614466 2 0.1333 0.812682 0.048 0.944 0.000 0.000 0.000 0.008
#> GSM614467 2 0.2306 0.796804 0.016 0.888 0.000 0.000 0.004 0.092
#> GSM614468 2 0.1616 0.813683 0.020 0.932 0.000 0.000 0.000 0.048
#> GSM614469 1 0.0000 0.692711 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM614470 1 0.0291 0.692122 0.992 0.000 0.000 0.000 0.004 0.004
#> GSM614471 1 0.0000 0.692711 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM614472 1 0.0146 0.692093 0.996 0.004 0.000 0.000 0.000 0.000
#> GSM614473 1 0.0000 0.692711 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM614474 1 0.0146 0.692449 0.996 0.000 0.000 0.000 0.000 0.004
#> GSM614475 1 0.1881 0.685534 0.928 0.008 0.040 0.000 0.004 0.020
#> GSM614476 1 0.7162 0.043162 0.388 0.036 0.380 0.000 0.048 0.148
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 individual(p) protocol(p) time(p) other(p) k
#> CV:mclust 69 3.73e-11 0.628 0.997 0.8021 2
#> CV:mclust 55 4.49e-16 0.605 0.999 0.4067 3
#> CV:mclust 83 2.36e-36 0.963 1.000 0.0498 4
#> CV:mclust 78 2.53e-41 0.827 1.000 0.0525 5
#> CV:mclust 72 4.28e-49 0.894 1.000 0.1928 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 21168 rows and 86 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 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 0.345 0.554 0.813 0.4680 0.540 0.540
#> 3 3 0.801 0.862 0.939 0.4010 0.660 0.447
#> 4 4 0.644 0.702 0.848 0.1306 0.768 0.441
#> 5 5 0.605 0.551 0.741 0.0651 0.903 0.646
#> 6 6 0.709 0.604 0.742 0.0401 0.892 0.567
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
#> GSM614415 1 0.0000 0.6868 1.000 0.000
#> GSM614416 1 0.0000 0.6868 1.000 0.000
#> GSM614417 1 0.0000 0.6868 1.000 0.000
#> GSM614418 1 0.0000 0.6868 1.000 0.000
#> GSM614419 1 0.5629 0.5635 0.868 0.132
#> GSM614420 1 0.1414 0.6748 0.980 0.020
#> GSM614421 2 0.1414 0.7909 0.020 0.980
#> GSM614422 1 0.9996 -0.1138 0.512 0.488
#> GSM614423 1 0.6801 0.6407 0.820 0.180
#> GSM614424 2 0.0000 0.8028 0.000 1.000
#> GSM614425 2 0.6343 0.6411 0.160 0.840
#> GSM614426 2 0.3733 0.7447 0.072 0.928
#> GSM614427 2 0.0000 0.8028 0.000 1.000
#> GSM614428 2 0.0000 0.8028 0.000 1.000
#> GSM614429 1 0.9977 0.3589 0.528 0.472
#> GSM614430 1 0.9944 0.3977 0.544 0.456
#> GSM614431 1 0.9896 0.4283 0.560 0.440
#> GSM614432 1 0.9909 0.4222 0.556 0.444
#> GSM614433 1 0.9909 0.4222 0.556 0.444
#> GSM614434 1 0.9922 0.4155 0.552 0.448
#> GSM614435 2 0.9732 0.0198 0.404 0.596
#> GSM614436 2 0.0000 0.8028 0.000 1.000
#> GSM614437 2 0.0000 0.8028 0.000 1.000
#> GSM614438 2 0.0000 0.8028 0.000 1.000
#> GSM614439 2 0.0000 0.8028 0.000 1.000
#> GSM614440 2 0.0000 0.8028 0.000 1.000
#> GSM614441 2 0.0000 0.8028 0.000 1.000
#> GSM614442 2 0.0000 0.8028 0.000 1.000
#> GSM614443 2 0.0000 0.8028 0.000 1.000
#> GSM614444 2 0.0000 0.8028 0.000 1.000
#> GSM614391 1 0.0000 0.6868 1.000 0.000
#> GSM614392 1 0.0000 0.6868 1.000 0.000
#> GSM614393 1 0.0000 0.6868 1.000 0.000
#> GSM614394 1 0.0672 0.6824 0.992 0.008
#> GSM614395 2 0.9881 0.2161 0.436 0.564
#> GSM614396 1 0.1184 0.6775 0.984 0.016
#> GSM614397 2 0.9909 0.2026 0.444 0.556
#> GSM614398 1 0.9988 -0.1015 0.520 0.480
#> GSM614399 1 0.9552 0.4979 0.624 0.376
#> GSM614400 1 0.6148 0.6520 0.848 0.152
#> GSM614401 1 0.2948 0.6801 0.948 0.052
#> GSM614402 1 0.8207 0.5978 0.744 0.256
#> GSM614403 1 0.9710 0.4732 0.600 0.400
#> GSM614404 1 0.6438 0.6479 0.836 0.164
#> GSM614405 1 0.8955 0.5564 0.688 0.312
#> GSM614406 2 0.0000 0.8028 0.000 1.000
#> GSM614407 1 0.0000 0.6868 1.000 0.000
#> GSM614408 1 0.0000 0.6868 1.000 0.000
#> GSM614409 1 0.0000 0.6868 1.000 0.000
#> GSM614410 1 0.0000 0.6868 1.000 0.000
#> GSM614411 1 0.0000 0.6868 1.000 0.000
#> GSM614412 1 0.0000 0.6868 1.000 0.000
#> GSM614413 1 0.9988 -0.1035 0.520 0.480
#> GSM614414 1 0.9710 0.0778 0.600 0.400
#> GSM614445 1 0.8608 0.5767 0.716 0.284
#> GSM614446 1 0.8555 0.5736 0.720 0.280
#> GSM614447 1 0.9775 0.4601 0.588 0.412
#> GSM614448 2 0.2603 0.7716 0.044 0.956
#> GSM614449 2 0.0376 0.8004 0.004 0.996
#> GSM614450 2 0.6623 0.6642 0.172 0.828
#> GSM614451 2 0.0000 0.8028 0.000 1.000
#> GSM614452 2 0.0000 0.8028 0.000 1.000
#> GSM614453 1 0.9922 0.4155 0.552 0.448
#> GSM614454 1 0.9922 0.4155 0.552 0.448
#> GSM614455 1 0.9922 0.4155 0.552 0.448
#> GSM614456 2 0.9661 0.0666 0.392 0.608
#> GSM614457 2 0.9552 0.1226 0.376 0.624
#> GSM614458 2 0.9922 -0.1471 0.448 0.552
#> GSM614459 2 0.5519 0.6761 0.128 0.872
#> GSM614460 2 0.9954 -0.1880 0.460 0.540
#> GSM614461 1 0.9922 0.4155 0.552 0.448
#> GSM614462 1 0.9909 0.4222 0.556 0.444
#> GSM614463 1 0.9881 0.4335 0.564 0.436
#> GSM614464 1 0.9922 0.4155 0.552 0.448
#> GSM614465 1 0.9896 0.4283 0.560 0.440
#> GSM614466 1 0.9896 0.4283 0.560 0.440
#> GSM614467 2 0.7745 0.5124 0.228 0.772
#> GSM614468 1 0.9922 0.4155 0.552 0.448
#> GSM614469 1 0.0000 0.6868 1.000 0.000
#> GSM614470 1 0.0000 0.6868 1.000 0.000
#> GSM614471 1 0.0000 0.6868 1.000 0.000
#> GSM614472 1 0.0000 0.6868 1.000 0.000
#> GSM614473 1 0.0000 0.6868 1.000 0.000
#> GSM614474 1 0.0000 0.6868 1.000 0.000
#> GSM614475 1 0.2603 0.6815 0.956 0.044
#> GSM614476 1 0.7528 0.6242 0.784 0.216
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM614415 1 0.0000 0.9181 1.000 0.000 0.000
#> GSM614416 1 0.0000 0.9181 1.000 0.000 0.000
#> GSM614417 1 0.0000 0.9181 1.000 0.000 0.000
#> GSM614418 1 0.0000 0.9181 1.000 0.000 0.000
#> GSM614419 1 0.0237 0.9166 0.996 0.000 0.004
#> GSM614420 1 0.0237 0.9166 0.996 0.000 0.004
#> GSM614421 3 0.1031 0.9529 0.024 0.000 0.976
#> GSM614422 1 0.3816 0.7891 0.852 0.000 0.148
#> GSM614423 2 0.4121 0.7921 0.168 0.832 0.000
#> GSM614424 3 0.0592 0.9612 0.012 0.000 0.988
#> GSM614425 3 0.3816 0.8248 0.148 0.000 0.852
#> GSM614426 3 0.2625 0.8995 0.084 0.000 0.916
#> GSM614427 3 0.0000 0.9684 0.000 0.000 1.000
#> GSM614428 3 0.0000 0.9684 0.000 0.000 1.000
#> GSM614429 2 0.0747 0.9189 0.000 0.984 0.016
#> GSM614430 2 0.0424 0.9212 0.000 0.992 0.008
#> GSM614431 2 0.0000 0.9230 0.000 1.000 0.000
#> GSM614432 2 0.0000 0.9230 0.000 1.000 0.000
#> GSM614433 2 0.0000 0.9230 0.000 1.000 0.000
#> GSM614434 2 0.0000 0.9230 0.000 1.000 0.000
#> GSM614435 2 0.2711 0.8745 0.000 0.912 0.088
#> GSM614436 3 0.3192 0.8642 0.000 0.112 0.888
#> GSM614437 3 0.1411 0.9413 0.000 0.036 0.964
#> GSM614438 3 0.0000 0.9684 0.000 0.000 1.000
#> GSM614439 3 0.0000 0.9684 0.000 0.000 1.000
#> GSM614440 3 0.0000 0.9684 0.000 0.000 1.000
#> GSM614441 3 0.0000 0.9684 0.000 0.000 1.000
#> GSM614442 3 0.0000 0.9684 0.000 0.000 1.000
#> GSM614443 3 0.0592 0.9607 0.000 0.012 0.988
#> GSM614444 3 0.0000 0.9684 0.000 0.000 1.000
#> GSM614391 1 0.0000 0.9181 1.000 0.000 0.000
#> GSM614392 1 0.0000 0.9181 1.000 0.000 0.000
#> GSM614393 1 0.0000 0.9181 1.000 0.000 0.000
#> GSM614394 1 0.0000 0.9181 1.000 0.000 0.000
#> GSM614395 1 0.6302 0.0403 0.520 0.000 0.480
#> GSM614396 1 0.0000 0.9181 1.000 0.000 0.000
#> GSM614397 1 0.2448 0.8619 0.924 0.000 0.076
#> GSM614398 1 0.0592 0.9122 0.988 0.000 0.012
#> GSM614399 2 0.0000 0.9230 0.000 1.000 0.000
#> GSM614400 2 0.0237 0.9219 0.004 0.996 0.000
#> GSM614401 2 0.0747 0.9172 0.016 0.984 0.000
#> GSM614402 2 0.0000 0.9230 0.000 1.000 0.000
#> GSM614403 2 0.3091 0.8846 0.016 0.912 0.072
#> GSM614404 2 0.0000 0.9230 0.000 1.000 0.000
#> GSM614405 2 0.9118 0.4303 0.232 0.548 0.220
#> GSM614406 3 0.0000 0.9684 0.000 0.000 1.000
#> GSM614407 1 0.0237 0.9157 0.996 0.004 0.000
#> GSM614408 1 0.0000 0.9181 1.000 0.000 0.000
#> GSM614409 1 0.0000 0.9181 1.000 0.000 0.000
#> GSM614410 1 0.0000 0.9181 1.000 0.000 0.000
#> GSM614411 1 0.0000 0.9181 1.000 0.000 0.000
#> GSM614412 1 0.0000 0.9181 1.000 0.000 0.000
#> GSM614413 1 0.1529 0.8926 0.960 0.000 0.040
#> GSM614414 1 0.0747 0.9097 0.984 0.000 0.016
#> GSM614445 2 0.0592 0.9194 0.012 0.988 0.000
#> GSM614446 2 0.3918 0.8375 0.120 0.868 0.012
#> GSM614447 2 0.1267 0.9151 0.004 0.972 0.024
#> GSM614448 3 0.0000 0.9684 0.000 0.000 1.000
#> GSM614449 3 0.0000 0.9684 0.000 0.000 1.000
#> GSM614450 3 0.5085 0.8332 0.072 0.092 0.836
#> GSM614451 3 0.0000 0.9684 0.000 0.000 1.000
#> GSM614452 3 0.0000 0.9684 0.000 0.000 1.000
#> GSM614453 2 0.0000 0.9230 0.000 1.000 0.000
#> GSM614454 2 0.0000 0.9230 0.000 1.000 0.000
#> GSM614455 2 0.0000 0.9230 0.000 1.000 0.000
#> GSM614456 2 0.2165 0.8935 0.000 0.936 0.064
#> GSM614457 2 0.2066 0.8963 0.000 0.940 0.060
#> GSM614458 2 0.1163 0.9142 0.000 0.972 0.028
#> GSM614459 2 0.4654 0.7453 0.000 0.792 0.208
#> GSM614460 2 0.1643 0.9061 0.000 0.956 0.044
#> GSM614461 2 0.0000 0.9230 0.000 1.000 0.000
#> GSM614462 2 0.0000 0.9230 0.000 1.000 0.000
#> GSM614463 2 0.0000 0.9230 0.000 1.000 0.000
#> GSM614464 2 0.0000 0.9230 0.000 1.000 0.000
#> GSM614465 2 0.0000 0.9230 0.000 1.000 0.000
#> GSM614466 2 0.0000 0.9230 0.000 1.000 0.000
#> GSM614467 2 0.4504 0.7579 0.000 0.804 0.196
#> GSM614468 2 0.0000 0.9230 0.000 1.000 0.000
#> GSM614469 2 0.6140 0.3482 0.404 0.596 0.000
#> GSM614470 2 0.6295 0.1301 0.472 0.528 0.000
#> GSM614471 2 0.2625 0.8735 0.084 0.916 0.000
#> GSM614472 2 0.3412 0.8386 0.124 0.876 0.000
#> GSM614473 1 0.6204 0.2036 0.576 0.424 0.000
#> GSM614474 1 0.6225 0.1795 0.568 0.432 0.000
#> GSM614475 2 0.3941 0.8035 0.156 0.844 0.000
#> GSM614476 1 0.5662 0.7616 0.808 0.100 0.092
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM614415 1 0.0000 0.9166 1.000 0.000 0.000 0.000
#> GSM614416 1 0.0000 0.9166 1.000 0.000 0.000 0.000
#> GSM614417 1 0.0000 0.9166 1.000 0.000 0.000 0.000
#> GSM614418 1 0.0000 0.9166 1.000 0.000 0.000 0.000
#> GSM614419 1 0.0188 0.9172 0.996 0.000 0.004 0.000
#> GSM614420 1 0.0188 0.9172 0.996 0.000 0.004 0.000
#> GSM614421 3 0.2125 0.7660 0.000 0.076 0.920 0.004
#> GSM614422 3 0.2149 0.7659 0.000 0.088 0.912 0.000
#> GSM614423 3 0.4250 0.6419 0.000 0.276 0.724 0.000
#> GSM614424 3 0.2401 0.7658 0.000 0.092 0.904 0.004
#> GSM614425 3 0.1824 0.7626 0.000 0.060 0.936 0.004
#> GSM614426 3 0.1890 0.7609 0.000 0.056 0.936 0.008
#> GSM614427 3 0.1151 0.7449 0.000 0.024 0.968 0.008
#> GSM614428 3 0.1211 0.7076 0.000 0.000 0.960 0.040
#> GSM614429 2 0.2281 0.7412 0.000 0.904 0.000 0.096
#> GSM614430 2 0.1637 0.7735 0.000 0.940 0.000 0.060
#> GSM614431 2 0.0592 0.8021 0.000 0.984 0.000 0.016
#> GSM614432 2 0.0657 0.8064 0.000 0.984 0.004 0.012
#> GSM614433 2 0.2053 0.8108 0.000 0.924 0.072 0.004
#> GSM614434 2 0.1576 0.7859 0.000 0.948 0.004 0.048
#> GSM614435 2 0.3810 0.6330 0.000 0.804 0.008 0.188
#> GSM614436 4 0.5267 0.6403 0.000 0.076 0.184 0.740
#> GSM614437 4 0.0592 0.7539 0.000 0.016 0.000 0.984
#> GSM614438 4 0.2973 0.7351 0.000 0.000 0.144 0.856
#> GSM614439 4 0.3311 0.7161 0.000 0.000 0.172 0.828
#> GSM614440 4 0.3266 0.7197 0.000 0.000 0.168 0.832
#> GSM614441 4 0.3123 0.7288 0.000 0.000 0.156 0.844
#> GSM614442 4 0.2530 0.7445 0.000 0.000 0.112 0.888
#> GSM614443 4 0.0336 0.7538 0.000 0.000 0.008 0.992
#> GSM614444 4 0.3024 0.7334 0.000 0.000 0.148 0.852
#> GSM614391 1 0.0188 0.9172 0.996 0.000 0.004 0.000
#> GSM614392 1 0.0000 0.9166 1.000 0.000 0.000 0.000
#> GSM614393 1 0.0188 0.9172 0.996 0.000 0.004 0.000
#> GSM614394 1 0.0469 0.9159 0.988 0.000 0.012 0.000
#> GSM614395 3 0.7113 0.2789 0.276 0.000 0.552 0.172
#> GSM614396 1 0.0469 0.9159 0.988 0.000 0.012 0.000
#> GSM614397 1 0.4903 0.6487 0.724 0.000 0.248 0.028
#> GSM614398 1 0.1716 0.8864 0.936 0.000 0.064 0.000
#> GSM614399 2 0.2635 0.8095 0.000 0.904 0.076 0.020
#> GSM614400 2 0.2234 0.8134 0.008 0.924 0.064 0.004
#> GSM614401 2 0.2530 0.7942 0.004 0.896 0.100 0.000
#> GSM614402 2 0.3172 0.7350 0.000 0.840 0.160 0.000
#> GSM614403 3 0.4972 0.2831 0.000 0.456 0.544 0.000
#> GSM614404 2 0.2053 0.8107 0.004 0.924 0.072 0.000
#> GSM614405 3 0.5741 0.2934 0.020 0.440 0.536 0.004
#> GSM614406 3 0.4477 0.3895 0.000 0.000 0.688 0.312
#> GSM614407 1 0.1004 0.9117 0.972 0.004 0.024 0.000
#> GSM614408 1 0.0707 0.9141 0.980 0.000 0.020 0.000
#> GSM614409 1 0.1305 0.9069 0.960 0.004 0.036 0.000
#> GSM614410 1 0.0336 0.9165 0.992 0.000 0.008 0.000
#> GSM614411 1 0.1854 0.8974 0.940 0.012 0.048 0.000
#> GSM614412 1 0.2053 0.8868 0.924 0.004 0.072 0.000
#> GSM614413 3 0.3721 0.6166 0.176 0.004 0.816 0.004
#> GSM614414 1 0.4776 0.4561 0.624 0.000 0.376 0.000
#> GSM614445 3 0.4804 0.4689 0.000 0.384 0.616 0.000
#> GSM614446 3 0.4164 0.6536 0.000 0.264 0.736 0.000
#> GSM614447 3 0.4730 0.5073 0.000 0.364 0.636 0.000
#> GSM614448 3 0.0524 0.7365 0.000 0.008 0.988 0.004
#> GSM614449 3 0.2125 0.7659 0.000 0.076 0.920 0.004
#> GSM614450 3 0.3266 0.7293 0.000 0.168 0.832 0.000
#> GSM614451 3 0.3726 0.5471 0.000 0.000 0.788 0.212
#> GSM614452 3 0.3528 0.5707 0.000 0.000 0.808 0.192
#> GSM614453 2 0.4697 0.2779 0.000 0.644 0.000 0.356
#> GSM614454 2 0.4998 -0.1848 0.000 0.512 0.000 0.488
#> GSM614455 4 0.4977 0.2670 0.000 0.460 0.000 0.540
#> GSM614456 4 0.4222 0.6346 0.000 0.272 0.000 0.728
#> GSM614457 4 0.3837 0.6794 0.000 0.224 0.000 0.776
#> GSM614458 4 0.4967 0.2923 0.000 0.452 0.000 0.548
#> GSM614459 4 0.3569 0.6984 0.000 0.196 0.000 0.804
#> GSM614460 4 0.4250 0.6305 0.000 0.276 0.000 0.724
#> GSM614461 2 0.0895 0.8150 0.000 0.976 0.020 0.004
#> GSM614462 2 0.1792 0.8121 0.000 0.932 0.068 0.000
#> GSM614463 2 0.1118 0.8168 0.000 0.964 0.036 0.000
#> GSM614464 2 0.2125 0.8107 0.000 0.920 0.076 0.004
#> GSM614465 2 0.2589 0.7785 0.000 0.884 0.116 0.000
#> GSM614466 2 0.1824 0.8148 0.000 0.936 0.060 0.004
#> GSM614467 2 0.5132 -0.0244 0.000 0.548 0.448 0.004
#> GSM614468 2 0.3791 0.6762 0.000 0.796 0.200 0.004
#> GSM614469 1 0.3972 0.7346 0.788 0.204 0.008 0.000
#> GSM614470 1 0.4248 0.7104 0.768 0.220 0.012 0.000
#> GSM614471 2 0.4606 0.5822 0.264 0.724 0.012 0.000
#> GSM614472 2 0.5306 0.4196 0.348 0.632 0.020 0.000
#> GSM614473 1 0.2546 0.8565 0.900 0.092 0.008 0.000
#> GSM614474 1 0.4399 0.7165 0.768 0.212 0.020 0.000
#> GSM614475 2 0.4036 0.7553 0.088 0.836 0.076 0.000
#> GSM614476 3 0.4903 0.6581 0.028 0.248 0.724 0.000
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM614415 5 0.2471 0.6383 0.136 0.000 0.000 0.000 0.864
#> GSM614416 5 0.2516 0.6361 0.140 0.000 0.000 0.000 0.860
#> GSM614417 5 0.2516 0.6361 0.140 0.000 0.000 0.000 0.860
#> GSM614418 5 0.2516 0.6361 0.140 0.000 0.000 0.000 0.860
#> GSM614419 5 0.2424 0.6400 0.132 0.000 0.000 0.000 0.868
#> GSM614420 5 0.2424 0.6400 0.132 0.000 0.000 0.000 0.868
#> GSM614421 3 0.1444 0.7787 0.040 0.012 0.948 0.000 0.000
#> GSM614422 3 0.2208 0.7677 0.072 0.020 0.908 0.000 0.000
#> GSM614423 3 0.3798 0.7322 0.064 0.128 0.808 0.000 0.000
#> GSM614424 3 0.1310 0.7807 0.020 0.024 0.956 0.000 0.000
#> GSM614425 3 0.1043 0.7768 0.040 0.000 0.960 0.000 0.000
#> GSM614426 3 0.1082 0.7806 0.028 0.008 0.964 0.000 0.000
#> GSM614427 3 0.1116 0.7784 0.028 0.004 0.964 0.004 0.000
#> GSM614428 3 0.2450 0.7471 0.052 0.000 0.900 0.048 0.000
#> GSM614429 2 0.1538 0.6761 0.008 0.948 0.008 0.036 0.000
#> GSM614430 2 0.1356 0.6762 0.012 0.956 0.004 0.028 0.000
#> GSM614431 2 0.1281 0.7113 0.012 0.956 0.032 0.000 0.000
#> GSM614432 2 0.1717 0.7144 0.008 0.936 0.052 0.004 0.000
#> GSM614433 2 0.2358 0.7238 0.008 0.888 0.104 0.000 0.000
#> GSM614434 2 0.1314 0.6933 0.012 0.960 0.016 0.012 0.000
#> GSM614435 2 0.2733 0.5897 0.012 0.872 0.004 0.112 0.000
#> GSM614436 4 0.7022 0.4500 0.024 0.288 0.212 0.476 0.000
#> GSM614437 4 0.1608 0.7657 0.000 0.072 0.000 0.928 0.000
#> GSM614438 4 0.1410 0.7666 0.000 0.000 0.060 0.940 0.000
#> GSM614439 4 0.1544 0.7624 0.000 0.000 0.068 0.932 0.000
#> GSM614440 4 0.1544 0.7624 0.000 0.000 0.068 0.932 0.000
#> GSM614441 4 0.1544 0.7624 0.000 0.000 0.068 0.932 0.000
#> GSM614442 4 0.0955 0.7708 0.000 0.004 0.028 0.968 0.000
#> GSM614443 4 0.1544 0.7663 0.000 0.068 0.000 0.932 0.000
#> GSM614444 4 0.1341 0.7683 0.000 0.000 0.056 0.944 0.000
#> GSM614391 5 0.2329 0.6420 0.124 0.000 0.000 0.000 0.876
#> GSM614392 5 0.2230 0.6441 0.116 0.000 0.000 0.000 0.884
#> GSM614393 5 0.2230 0.6441 0.116 0.000 0.000 0.000 0.884
#> GSM614394 5 0.2629 0.6360 0.136 0.000 0.004 0.000 0.860
#> GSM614395 5 0.7043 0.2366 0.068 0.000 0.308 0.112 0.512
#> GSM614396 5 0.2674 0.6334 0.140 0.000 0.004 0.000 0.856
#> GSM614397 5 0.4968 0.5313 0.140 0.000 0.104 0.016 0.740
#> GSM614398 5 0.3595 0.6036 0.140 0.000 0.044 0.000 0.816
#> GSM614399 2 0.6202 0.5146 0.308 0.584 0.080 0.012 0.016
#> GSM614400 2 0.6903 0.3312 0.368 0.488 0.060 0.004 0.080
#> GSM614401 1 0.7027 -0.2615 0.420 0.420 0.068 0.000 0.092
#> GSM614402 2 0.6690 0.4910 0.288 0.540 0.140 0.000 0.032
#> GSM614403 3 0.6791 0.2925 0.236 0.260 0.492 0.000 0.012
#> GSM614404 2 0.6661 0.4046 0.352 0.512 0.084 0.000 0.052
#> GSM614405 3 0.7331 0.2467 0.328 0.212 0.428 0.004 0.028
#> GSM614406 3 0.5892 0.3194 0.072 0.000 0.524 0.392 0.012
#> GSM614407 1 0.3480 0.4921 0.752 0.000 0.000 0.000 0.248
#> GSM614408 1 0.3636 0.4720 0.728 0.000 0.000 0.000 0.272
#> GSM614409 1 0.3511 0.5166 0.800 0.004 0.012 0.000 0.184
#> GSM614410 1 0.3336 0.4943 0.772 0.000 0.000 0.000 0.228
#> GSM614411 1 0.3652 0.5094 0.784 0.004 0.012 0.000 0.200
#> GSM614412 1 0.3951 0.4932 0.776 0.004 0.028 0.000 0.192
#> GSM614413 1 0.5366 0.3891 0.684 0.008 0.228 0.008 0.072
#> GSM614414 1 0.5054 0.4511 0.732 0.004 0.144 0.008 0.112
#> GSM614445 3 0.5631 0.5459 0.164 0.200 0.636 0.000 0.000
#> GSM614446 3 0.4458 0.6926 0.120 0.120 0.760 0.000 0.000
#> GSM614447 3 0.5163 0.6289 0.156 0.152 0.692 0.000 0.000
#> GSM614448 3 0.2130 0.7539 0.012 0.000 0.908 0.080 0.000
#> GSM614449 3 0.2234 0.7771 0.032 0.012 0.920 0.036 0.000
#> GSM614450 3 0.3126 0.7622 0.076 0.048 0.868 0.008 0.000
#> GSM614451 3 0.3496 0.6434 0.012 0.000 0.788 0.200 0.000
#> GSM614452 3 0.3280 0.6654 0.012 0.000 0.812 0.176 0.000
#> GSM614453 2 0.3048 0.5212 0.004 0.820 0.000 0.176 0.000
#> GSM614454 2 0.3689 0.3821 0.004 0.740 0.000 0.256 0.000
#> GSM614455 2 0.4182 0.1424 0.004 0.644 0.000 0.352 0.000
#> GSM614456 4 0.4430 0.3948 0.004 0.456 0.000 0.540 0.000
#> GSM614457 4 0.4182 0.5805 0.004 0.352 0.000 0.644 0.000
#> GSM614458 2 0.4182 0.1249 0.004 0.644 0.000 0.352 0.000
#> GSM614459 4 0.3949 0.6353 0.004 0.300 0.000 0.696 0.000
#> GSM614460 4 0.4359 0.4923 0.004 0.412 0.000 0.584 0.000
#> GSM614461 2 0.2609 0.7256 0.048 0.896 0.052 0.004 0.000
#> GSM614462 2 0.3857 0.7110 0.084 0.808 0.108 0.000 0.000
#> GSM614463 2 0.3420 0.7190 0.084 0.840 0.076 0.000 0.000
#> GSM614464 2 0.4300 0.6972 0.096 0.772 0.132 0.000 0.000
#> GSM614465 2 0.4599 0.6744 0.100 0.744 0.156 0.000 0.000
#> GSM614466 2 0.3912 0.7108 0.088 0.804 0.108 0.000 0.000
#> GSM614467 2 0.4979 0.0306 0.028 0.492 0.480 0.000 0.000
#> GSM614468 2 0.3876 0.6754 0.032 0.776 0.192 0.000 0.000
#> GSM614469 5 0.5742 -0.0163 0.404 0.088 0.000 0.000 0.508
#> GSM614470 1 0.5840 0.0926 0.488 0.096 0.000 0.000 0.416
#> GSM614471 1 0.6638 0.2436 0.440 0.320 0.000 0.000 0.240
#> GSM614472 1 0.6610 0.2584 0.460 0.260 0.000 0.000 0.280
#> GSM614473 5 0.5371 0.0505 0.420 0.056 0.000 0.000 0.524
#> GSM614474 5 0.5293 -0.0716 0.460 0.048 0.000 0.000 0.492
#> GSM614475 2 0.6555 0.4515 0.272 0.580 0.068 0.000 0.080
#> GSM614476 3 0.6333 0.6023 0.220 0.068 0.644 0.016 0.052
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM614415 5 0.2790 0.53284 0.020 0.000 0.000 0.000 0.840 0.140
#> GSM614416 5 0.2790 0.53284 0.020 0.000 0.000 0.000 0.840 0.140
#> GSM614417 5 0.2790 0.53284 0.020 0.000 0.000 0.000 0.840 0.140
#> GSM614418 5 0.2790 0.53284 0.020 0.000 0.000 0.000 0.840 0.140
#> GSM614419 5 0.2664 0.53326 0.016 0.000 0.000 0.000 0.848 0.136
#> GSM614420 5 0.2664 0.53326 0.016 0.000 0.000 0.000 0.848 0.136
#> GSM614421 3 0.1448 0.80433 0.012 0.024 0.948 0.000 0.000 0.016
#> GSM614422 3 0.2084 0.79153 0.024 0.016 0.916 0.000 0.000 0.044
#> GSM614423 3 0.4422 0.67513 0.096 0.076 0.768 0.000 0.000 0.060
#> GSM614424 3 0.1313 0.80254 0.016 0.028 0.952 0.000 0.000 0.004
#> GSM614425 3 0.1078 0.80627 0.008 0.016 0.964 0.000 0.000 0.012
#> GSM614426 3 0.1251 0.80361 0.012 0.024 0.956 0.000 0.000 0.008
#> GSM614427 3 0.0603 0.80632 0.004 0.016 0.980 0.000 0.000 0.000
#> GSM614428 3 0.0810 0.80226 0.008 0.004 0.976 0.004 0.000 0.008
#> GSM614429 2 0.1088 0.73364 0.000 0.960 0.016 0.024 0.000 0.000
#> GSM614430 2 0.1109 0.73365 0.004 0.964 0.016 0.012 0.000 0.004
#> GSM614431 2 0.1003 0.72949 0.020 0.964 0.016 0.000 0.000 0.000
#> GSM614432 2 0.1492 0.72570 0.024 0.940 0.036 0.000 0.000 0.000
#> GSM614433 2 0.2263 0.71124 0.048 0.896 0.056 0.000 0.000 0.000
#> GSM614434 2 0.0717 0.73164 0.008 0.976 0.016 0.000 0.000 0.000
#> GSM614435 2 0.2036 0.72352 0.000 0.912 0.016 0.064 0.000 0.008
#> GSM614436 2 0.6001 0.38313 0.004 0.552 0.200 0.228 0.000 0.016
#> GSM614437 4 0.1141 0.79482 0.000 0.052 0.000 0.948 0.000 0.000
#> GSM614438 4 0.1141 0.83534 0.000 0.000 0.052 0.948 0.000 0.000
#> GSM614439 4 0.1141 0.83534 0.000 0.000 0.052 0.948 0.000 0.000
#> GSM614440 4 0.1141 0.83534 0.000 0.000 0.052 0.948 0.000 0.000
#> GSM614441 4 0.1141 0.83534 0.000 0.000 0.052 0.948 0.000 0.000
#> GSM614442 4 0.0972 0.82755 0.000 0.008 0.028 0.964 0.000 0.000
#> GSM614443 4 0.0937 0.80232 0.000 0.040 0.000 0.960 0.000 0.000
#> GSM614444 4 0.1141 0.83534 0.000 0.000 0.052 0.948 0.000 0.000
#> GSM614391 5 0.5240 0.49612 0.256 0.000 0.004 0.012 0.632 0.096
#> GSM614392 5 0.5105 0.49713 0.256 0.000 0.000 0.012 0.636 0.096
#> GSM614393 5 0.5105 0.49713 0.256 0.000 0.000 0.012 0.636 0.096
#> GSM614394 5 0.5322 0.49475 0.260 0.000 0.008 0.012 0.628 0.092
#> GSM614395 5 0.6762 0.43050 0.268 0.000 0.116 0.032 0.528 0.056
#> GSM614396 5 0.5578 0.48945 0.260 0.000 0.020 0.012 0.616 0.092
#> GSM614397 5 0.6032 0.47605 0.260 0.000 0.048 0.012 0.588 0.092
#> GSM614398 5 0.5934 0.47778 0.264 0.000 0.040 0.012 0.592 0.092
#> GSM614399 1 0.4346 0.74702 0.692 0.268 0.016 0.016 0.000 0.008
#> GSM614400 1 0.4583 0.77543 0.708 0.208 0.004 0.000 0.072 0.008
#> GSM614401 1 0.4913 0.77254 0.712 0.168 0.012 0.000 0.092 0.016
#> GSM614402 1 0.4289 0.78907 0.720 0.220 0.048 0.000 0.012 0.000
#> GSM614403 1 0.4399 0.74920 0.728 0.112 0.156 0.000 0.000 0.004
#> GSM614404 1 0.4614 0.78167 0.708 0.228 0.024 0.000 0.024 0.016
#> GSM614405 1 0.4859 0.75920 0.744 0.108 0.108 0.008 0.012 0.020
#> GSM614406 4 0.6178 -0.00431 0.348 0.000 0.260 0.388 0.000 0.004
#> GSM614407 6 0.1503 0.90568 0.016 0.000 0.008 0.000 0.032 0.944
#> GSM614408 6 0.1649 0.89912 0.016 0.000 0.008 0.000 0.040 0.936
#> GSM614409 6 0.1232 0.92526 0.016 0.000 0.024 0.000 0.004 0.956
#> GSM614410 6 0.0976 0.92035 0.016 0.000 0.008 0.000 0.008 0.968
#> GSM614411 6 0.0748 0.92315 0.004 0.000 0.016 0.000 0.004 0.976
#> GSM614412 6 0.0790 0.91876 0.000 0.000 0.032 0.000 0.000 0.968
#> GSM614413 6 0.2340 0.81656 0.000 0.000 0.148 0.000 0.000 0.852
#> GSM614414 6 0.1863 0.86689 0.000 0.000 0.104 0.000 0.000 0.896
#> GSM614445 1 0.5094 0.45954 0.568 0.080 0.348 0.000 0.004 0.000
#> GSM614446 3 0.4908 -0.07377 0.464 0.040 0.488 0.000 0.004 0.004
#> GSM614447 1 0.5014 0.43407 0.576 0.060 0.356 0.000 0.004 0.004
#> GSM614448 3 0.2948 0.75945 0.092 0.000 0.848 0.060 0.000 0.000
#> GSM614449 3 0.3042 0.74677 0.128 0.000 0.836 0.032 0.004 0.000
#> GSM614450 3 0.4478 0.53762 0.284 0.016 0.672 0.024 0.004 0.000
#> GSM614451 3 0.3528 0.71419 0.036 0.000 0.800 0.156 0.004 0.004
#> GSM614452 3 0.3196 0.73498 0.036 0.000 0.824 0.136 0.000 0.004
#> GSM614453 2 0.1663 0.71997 0.000 0.912 0.000 0.088 0.000 0.000
#> GSM614454 2 0.2048 0.71031 0.000 0.880 0.000 0.120 0.000 0.000
#> GSM614455 2 0.2854 0.63931 0.000 0.792 0.000 0.208 0.000 0.000
#> GSM614456 2 0.3309 0.54221 0.000 0.720 0.000 0.280 0.000 0.000
#> GSM614457 2 0.3955 0.33637 0.000 0.608 0.000 0.384 0.000 0.008
#> GSM614458 2 0.2402 0.69602 0.000 0.856 0.000 0.140 0.000 0.004
#> GSM614459 4 0.4098 -0.12858 0.000 0.496 0.000 0.496 0.000 0.008
#> GSM614460 2 0.3804 0.43411 0.000 0.656 0.000 0.336 0.000 0.008
#> GSM614461 2 0.2983 0.66842 0.136 0.832 0.032 0.000 0.000 0.000
#> GSM614462 2 0.3555 0.61582 0.184 0.776 0.040 0.000 0.000 0.000
#> GSM614463 2 0.3210 0.64254 0.168 0.804 0.028 0.000 0.000 0.000
#> GSM614464 2 0.3744 0.60561 0.184 0.764 0.052 0.000 0.000 0.000
#> GSM614465 2 0.4059 0.53164 0.228 0.720 0.052 0.000 0.000 0.000
#> GSM614466 2 0.3649 0.60115 0.196 0.764 0.040 0.000 0.000 0.000
#> GSM614467 2 0.4928 0.30251 0.076 0.572 0.352 0.000 0.000 0.000
#> GSM614468 2 0.3566 0.66059 0.104 0.800 0.096 0.000 0.000 0.000
#> GSM614469 5 0.6531 0.29211 0.228 0.024 0.008 0.000 0.476 0.264
#> GSM614470 5 0.6676 0.16763 0.360 0.032 0.008 0.000 0.412 0.188
#> GSM614471 5 0.7622 0.01420 0.320 0.144 0.008 0.000 0.336 0.192
#> GSM614472 5 0.7153 0.09392 0.360 0.076 0.008 0.000 0.372 0.184
#> GSM614473 5 0.6453 0.33375 0.260 0.028 0.008 0.000 0.504 0.200
#> GSM614474 5 0.6488 0.12029 0.200 0.012 0.012 0.000 0.396 0.380
#> GSM614475 2 0.7470 -0.22452 0.320 0.432 0.056 0.000 0.100 0.092
#> GSM614476 3 0.6880 0.24754 0.260 0.032 0.536 0.004 0.088 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 individual(p) protocol(p) time(p) other(p) k
#> CV:NMF 56 4.75e-06 0.00743 0.707 0.7141 2
#> CV:NMF 80 7.06e-17 0.17951 0.996 0.3429 3
#> CV:NMF 74 9.33e-26 0.07237 1.000 0.1297 4
#> CV:NMF 58 9.52e-28 0.60657 0.999 0.3603 5
#> CV:NMF 61 1.68e-42 0.54039 1.000 0.0407 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 21168 rows and 86 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 0.426 0.746 0.820 0.3184 0.774 0.774
#> 3 3 0.685 0.789 0.898 0.8929 0.635 0.529
#> 4 4 0.685 0.753 0.872 0.0419 0.993 0.984
#> 5 5 0.707 0.728 0.841 0.0602 0.958 0.895
#> 6 6 0.688 0.695 0.832 0.0843 0.879 0.671
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
#> GSM614415 2 0.9209 0.679 0.336 0.664
#> GSM614416 2 0.9209 0.679 0.336 0.664
#> GSM614417 2 0.9209 0.679 0.336 0.664
#> GSM614418 2 0.9209 0.679 0.336 0.664
#> GSM614419 2 0.9209 0.679 0.336 0.664
#> GSM614420 2 0.9209 0.679 0.336 0.664
#> GSM614421 2 0.5519 0.641 0.128 0.872
#> GSM614422 2 0.5519 0.641 0.128 0.872
#> GSM614423 2 0.5519 0.641 0.128 0.872
#> GSM614424 2 0.5519 0.641 0.128 0.872
#> GSM614425 2 0.5519 0.641 0.128 0.872
#> GSM614426 2 0.5519 0.641 0.128 0.872
#> GSM614427 2 0.5519 0.641 0.128 0.872
#> GSM614428 2 0.5519 0.641 0.128 0.872
#> GSM614429 2 0.0000 0.780 0.000 1.000
#> GSM614430 2 0.0000 0.780 0.000 1.000
#> GSM614431 2 0.0000 0.780 0.000 1.000
#> GSM614432 2 0.0000 0.780 0.000 1.000
#> GSM614433 2 0.0000 0.780 0.000 1.000
#> GSM614434 2 0.0000 0.780 0.000 1.000
#> GSM614435 2 0.0000 0.780 0.000 1.000
#> GSM614436 2 0.0000 0.780 0.000 1.000
#> GSM614437 1 0.9209 0.942 0.664 0.336
#> GSM614438 1 0.9209 0.942 0.664 0.336
#> GSM614439 1 0.9209 0.942 0.664 0.336
#> GSM614440 1 0.9209 0.942 0.664 0.336
#> GSM614441 1 0.9209 0.942 0.664 0.336
#> GSM614442 1 0.9209 0.942 0.664 0.336
#> GSM614443 1 0.9209 0.942 0.664 0.336
#> GSM614444 1 0.9209 0.942 0.664 0.336
#> GSM614391 2 0.9209 0.679 0.336 0.664
#> GSM614392 2 0.9209 0.679 0.336 0.664
#> GSM614393 2 0.9209 0.679 0.336 0.664
#> GSM614394 2 0.9209 0.679 0.336 0.664
#> GSM614395 1 0.5737 0.391 0.864 0.136
#> GSM614396 2 0.9209 0.679 0.336 0.664
#> GSM614397 2 0.9209 0.679 0.336 0.664
#> GSM614398 2 0.9209 0.679 0.336 0.664
#> GSM614399 2 0.0672 0.780 0.008 0.992
#> GSM614400 2 0.0672 0.780 0.008 0.992
#> GSM614401 2 0.0672 0.780 0.008 0.992
#> GSM614402 2 0.0672 0.780 0.008 0.992
#> GSM614403 2 0.0672 0.780 0.008 0.992
#> GSM614404 2 0.0672 0.780 0.008 0.992
#> GSM614405 2 0.0672 0.780 0.008 0.992
#> GSM614406 2 0.0376 0.780 0.004 0.996
#> GSM614407 2 0.9209 0.679 0.336 0.664
#> GSM614408 2 0.9209 0.679 0.336 0.664
#> GSM614409 2 0.9209 0.679 0.336 0.664
#> GSM614410 2 0.9209 0.679 0.336 0.664
#> GSM614411 2 0.9209 0.679 0.336 0.664
#> GSM614412 2 0.9209 0.679 0.336 0.664
#> GSM614413 2 0.9209 0.679 0.336 0.664
#> GSM614414 2 0.9209 0.679 0.336 0.664
#> GSM614445 2 0.3879 0.711 0.076 0.924
#> GSM614446 2 0.3879 0.711 0.076 0.924
#> GSM614447 2 0.3879 0.711 0.076 0.924
#> GSM614448 2 0.3879 0.711 0.076 0.924
#> GSM614449 2 0.3879 0.711 0.076 0.924
#> GSM614450 2 0.3879 0.711 0.076 0.924
#> GSM614451 1 0.9209 0.942 0.664 0.336
#> GSM614452 1 0.9209 0.942 0.664 0.336
#> GSM614453 2 0.0000 0.780 0.000 1.000
#> GSM614454 2 0.0000 0.780 0.000 1.000
#> GSM614455 2 0.0000 0.780 0.000 1.000
#> GSM614456 2 0.0000 0.780 0.000 1.000
#> GSM614457 2 0.0000 0.780 0.000 1.000
#> GSM614458 2 0.0000 0.780 0.000 1.000
#> GSM614459 2 0.0000 0.780 0.000 1.000
#> GSM614460 2 0.0000 0.780 0.000 1.000
#> GSM614461 2 0.0000 0.780 0.000 1.000
#> GSM614462 2 0.0000 0.780 0.000 1.000
#> GSM614463 2 0.0000 0.780 0.000 1.000
#> GSM614464 2 0.0000 0.780 0.000 1.000
#> GSM614465 2 0.0000 0.780 0.000 1.000
#> GSM614466 2 0.0000 0.780 0.000 1.000
#> GSM614467 2 0.0000 0.780 0.000 1.000
#> GSM614468 2 0.0000 0.780 0.000 1.000
#> GSM614469 2 0.8207 0.716 0.256 0.744
#> GSM614470 2 0.8207 0.716 0.256 0.744
#> GSM614471 2 0.8207 0.716 0.256 0.744
#> GSM614472 2 0.8207 0.716 0.256 0.744
#> GSM614473 2 0.8207 0.716 0.256 0.744
#> GSM614474 2 0.8207 0.716 0.256 0.744
#> GSM614475 2 0.8207 0.716 0.256 0.744
#> GSM614476 2 0.8207 0.716 0.256 0.744
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM614415 1 0.0424 0.8544 0.992 0.008 0.000
#> GSM614416 1 0.0424 0.8544 0.992 0.008 0.000
#> GSM614417 1 0.0424 0.8544 0.992 0.008 0.000
#> GSM614418 1 0.0424 0.8544 0.992 0.008 0.000
#> GSM614419 1 0.0424 0.8544 0.992 0.008 0.000
#> GSM614420 1 0.0424 0.8544 0.992 0.008 0.000
#> GSM614421 2 0.6513 0.5306 0.008 0.592 0.400
#> GSM614422 2 0.6513 0.5306 0.008 0.592 0.400
#> GSM614423 2 0.6513 0.5306 0.008 0.592 0.400
#> GSM614424 2 0.6513 0.5306 0.008 0.592 0.400
#> GSM614425 2 0.6513 0.5306 0.008 0.592 0.400
#> GSM614426 2 0.6513 0.5306 0.008 0.592 0.400
#> GSM614427 2 0.6513 0.5306 0.008 0.592 0.400
#> GSM614428 2 0.6513 0.5306 0.008 0.592 0.400
#> GSM614429 2 0.0000 0.8700 0.000 1.000 0.000
#> GSM614430 2 0.0000 0.8700 0.000 1.000 0.000
#> GSM614431 2 0.0000 0.8700 0.000 1.000 0.000
#> GSM614432 2 0.0000 0.8700 0.000 1.000 0.000
#> GSM614433 2 0.0000 0.8700 0.000 1.000 0.000
#> GSM614434 2 0.0000 0.8700 0.000 1.000 0.000
#> GSM614435 2 0.0000 0.8700 0.000 1.000 0.000
#> GSM614436 2 0.0000 0.8700 0.000 1.000 0.000
#> GSM614437 3 0.0237 0.9445 0.000 0.004 0.996
#> GSM614438 3 0.0237 0.9445 0.000 0.004 0.996
#> GSM614439 3 0.0237 0.9445 0.000 0.004 0.996
#> GSM614440 3 0.0237 0.9445 0.000 0.004 0.996
#> GSM614441 3 0.0237 0.9445 0.000 0.004 0.996
#> GSM614442 3 0.0237 0.9445 0.000 0.004 0.996
#> GSM614443 3 0.0237 0.9445 0.000 0.004 0.996
#> GSM614444 3 0.0237 0.9445 0.000 0.004 0.996
#> GSM614391 1 0.0000 0.8490 1.000 0.000 0.000
#> GSM614392 1 0.0000 0.8490 1.000 0.000 0.000
#> GSM614393 1 0.0000 0.8490 1.000 0.000 0.000
#> GSM614394 1 0.0000 0.8490 1.000 0.000 0.000
#> GSM614395 3 0.6295 0.0283 0.472 0.000 0.528
#> GSM614396 1 0.0000 0.8490 1.000 0.000 0.000
#> GSM614397 1 0.0000 0.8490 1.000 0.000 0.000
#> GSM614398 1 0.0000 0.8490 1.000 0.000 0.000
#> GSM614399 2 0.1031 0.8571 0.024 0.976 0.000
#> GSM614400 2 0.1031 0.8571 0.024 0.976 0.000
#> GSM614401 2 0.1031 0.8571 0.024 0.976 0.000
#> GSM614402 2 0.1031 0.8571 0.024 0.976 0.000
#> GSM614403 2 0.1031 0.8571 0.024 0.976 0.000
#> GSM614404 2 0.1031 0.8571 0.024 0.976 0.000
#> GSM614405 2 0.0424 0.8666 0.008 0.992 0.000
#> GSM614406 2 0.0475 0.8679 0.004 0.992 0.004
#> GSM614407 1 0.0424 0.8544 0.992 0.008 0.000
#> GSM614408 1 0.0424 0.8544 0.992 0.008 0.000
#> GSM614409 1 0.0424 0.8544 0.992 0.008 0.000
#> GSM614410 1 0.0424 0.8544 0.992 0.008 0.000
#> GSM614411 1 0.0424 0.8544 0.992 0.008 0.000
#> GSM614412 1 0.0424 0.8544 0.992 0.008 0.000
#> GSM614413 1 0.0424 0.8544 0.992 0.008 0.000
#> GSM614414 1 0.0424 0.8544 0.992 0.008 0.000
#> GSM614445 2 0.5873 0.6481 0.004 0.684 0.312
#> GSM614446 2 0.5873 0.6481 0.004 0.684 0.312
#> GSM614447 2 0.5873 0.6481 0.004 0.684 0.312
#> GSM614448 2 0.5873 0.6481 0.004 0.684 0.312
#> GSM614449 2 0.5873 0.6481 0.004 0.684 0.312
#> GSM614450 2 0.5873 0.6481 0.004 0.684 0.312
#> GSM614451 3 0.0424 0.9375 0.000 0.008 0.992
#> GSM614452 3 0.0424 0.9375 0.000 0.008 0.992
#> GSM614453 2 0.0000 0.8700 0.000 1.000 0.000
#> GSM614454 2 0.0000 0.8700 0.000 1.000 0.000
#> GSM614455 2 0.0000 0.8700 0.000 1.000 0.000
#> GSM614456 2 0.0000 0.8700 0.000 1.000 0.000
#> GSM614457 2 0.0000 0.8700 0.000 1.000 0.000
#> GSM614458 2 0.0000 0.8700 0.000 1.000 0.000
#> GSM614459 2 0.0000 0.8700 0.000 1.000 0.000
#> GSM614460 2 0.0000 0.8700 0.000 1.000 0.000
#> GSM614461 2 0.0000 0.8700 0.000 1.000 0.000
#> GSM614462 2 0.0000 0.8700 0.000 1.000 0.000
#> GSM614463 2 0.0000 0.8700 0.000 1.000 0.000
#> GSM614464 2 0.0000 0.8700 0.000 1.000 0.000
#> GSM614465 2 0.0000 0.8700 0.000 1.000 0.000
#> GSM614466 2 0.0000 0.8700 0.000 1.000 0.000
#> GSM614467 2 0.0000 0.8700 0.000 1.000 0.000
#> GSM614468 2 0.0000 0.8700 0.000 1.000 0.000
#> GSM614469 1 0.5905 0.5745 0.648 0.352 0.000
#> GSM614470 1 0.5905 0.5745 0.648 0.352 0.000
#> GSM614471 1 0.5905 0.5745 0.648 0.352 0.000
#> GSM614472 1 0.5905 0.5745 0.648 0.352 0.000
#> GSM614473 1 0.5905 0.5745 0.648 0.352 0.000
#> GSM614474 1 0.5905 0.5745 0.648 0.352 0.000
#> GSM614475 1 0.5905 0.5745 0.648 0.352 0.000
#> GSM614476 1 0.5905 0.5745 0.648 0.352 0.000
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM614415 1 0.0469 0.761 0.988 0.000 0.012 0.000
#> GSM614416 1 0.0469 0.761 0.988 0.000 0.012 0.000
#> GSM614417 1 0.0469 0.761 0.988 0.000 0.012 0.000
#> GSM614418 1 0.0469 0.761 0.988 0.000 0.012 0.000
#> GSM614419 1 0.0469 0.761 0.988 0.000 0.012 0.000
#> GSM614420 1 0.0469 0.761 0.988 0.000 0.012 0.000
#> GSM614421 2 0.5183 0.547 0.008 0.584 0.000 0.408
#> GSM614422 2 0.5183 0.547 0.008 0.584 0.000 0.408
#> GSM614423 2 0.5183 0.547 0.008 0.584 0.000 0.408
#> GSM614424 2 0.5183 0.547 0.008 0.584 0.000 0.408
#> GSM614425 2 0.5183 0.547 0.008 0.584 0.000 0.408
#> GSM614426 2 0.5183 0.547 0.008 0.584 0.000 0.408
#> GSM614427 2 0.5183 0.547 0.008 0.584 0.000 0.408
#> GSM614428 2 0.5183 0.547 0.008 0.584 0.000 0.408
#> GSM614429 2 0.0000 0.866 0.000 1.000 0.000 0.000
#> GSM614430 2 0.0000 0.866 0.000 1.000 0.000 0.000
#> GSM614431 2 0.0000 0.866 0.000 1.000 0.000 0.000
#> GSM614432 2 0.0000 0.866 0.000 1.000 0.000 0.000
#> GSM614433 2 0.0000 0.866 0.000 1.000 0.000 0.000
#> GSM614434 2 0.0000 0.866 0.000 1.000 0.000 0.000
#> GSM614435 2 0.0000 0.866 0.000 1.000 0.000 0.000
#> GSM614436 2 0.0000 0.866 0.000 1.000 0.000 0.000
#> GSM614437 4 0.4655 1.000 0.000 0.004 0.312 0.684
#> GSM614438 4 0.4655 1.000 0.000 0.004 0.312 0.684
#> GSM614439 4 0.4655 1.000 0.000 0.004 0.312 0.684
#> GSM614440 4 0.4655 1.000 0.000 0.004 0.312 0.684
#> GSM614441 4 0.4655 1.000 0.000 0.004 0.312 0.684
#> GSM614442 4 0.4655 1.000 0.000 0.004 0.312 0.684
#> GSM614443 4 0.4655 1.000 0.000 0.004 0.312 0.684
#> GSM614444 4 0.4655 1.000 0.000 0.004 0.312 0.684
#> GSM614391 1 0.2647 0.705 0.880 0.000 0.120 0.000
#> GSM614392 1 0.2647 0.705 0.880 0.000 0.120 0.000
#> GSM614393 1 0.2647 0.705 0.880 0.000 0.120 0.000
#> GSM614394 1 0.2647 0.705 0.880 0.000 0.120 0.000
#> GSM614395 3 0.5386 0.162 0.368 0.000 0.612 0.020
#> GSM614396 1 0.2647 0.705 0.880 0.000 0.120 0.000
#> GSM614397 1 0.2647 0.705 0.880 0.000 0.120 0.000
#> GSM614398 1 0.2647 0.705 0.880 0.000 0.120 0.000
#> GSM614399 2 0.0921 0.851 0.028 0.972 0.000 0.000
#> GSM614400 2 0.0921 0.851 0.028 0.972 0.000 0.000
#> GSM614401 2 0.0921 0.851 0.028 0.972 0.000 0.000
#> GSM614402 2 0.0921 0.851 0.028 0.972 0.000 0.000
#> GSM614403 2 0.0921 0.851 0.028 0.972 0.000 0.000
#> GSM614404 2 0.0921 0.851 0.028 0.972 0.000 0.000
#> GSM614405 2 0.0336 0.863 0.008 0.992 0.000 0.000
#> GSM614406 2 0.0376 0.864 0.004 0.992 0.000 0.004
#> GSM614407 1 0.1389 0.758 0.952 0.000 0.048 0.000
#> GSM614408 1 0.1389 0.758 0.952 0.000 0.048 0.000
#> GSM614409 1 0.1389 0.758 0.952 0.000 0.048 0.000
#> GSM614410 1 0.1389 0.758 0.952 0.000 0.048 0.000
#> GSM614411 1 0.1389 0.758 0.952 0.000 0.048 0.000
#> GSM614412 1 0.1474 0.758 0.948 0.000 0.052 0.000
#> GSM614413 1 0.1474 0.758 0.948 0.000 0.052 0.000
#> GSM614414 1 0.1474 0.758 0.948 0.000 0.052 0.000
#> GSM614445 2 0.4836 0.651 0.008 0.672 0.000 0.320
#> GSM614446 2 0.4836 0.651 0.008 0.672 0.000 0.320
#> GSM614447 2 0.4836 0.651 0.008 0.672 0.000 0.320
#> GSM614448 2 0.4836 0.651 0.008 0.672 0.000 0.320
#> GSM614449 2 0.4836 0.651 0.008 0.672 0.000 0.320
#> GSM614450 2 0.4836 0.651 0.008 0.672 0.000 0.320
#> GSM614451 3 0.4999 0.563 0.000 0.000 0.508 0.492
#> GSM614452 3 0.4999 0.563 0.000 0.000 0.508 0.492
#> GSM614453 2 0.0000 0.866 0.000 1.000 0.000 0.000
#> GSM614454 2 0.0000 0.866 0.000 1.000 0.000 0.000
#> GSM614455 2 0.0000 0.866 0.000 1.000 0.000 0.000
#> GSM614456 2 0.0000 0.866 0.000 1.000 0.000 0.000
#> GSM614457 2 0.0000 0.866 0.000 1.000 0.000 0.000
#> GSM614458 2 0.0000 0.866 0.000 1.000 0.000 0.000
#> GSM614459 2 0.0000 0.866 0.000 1.000 0.000 0.000
#> GSM614460 2 0.0000 0.866 0.000 1.000 0.000 0.000
#> GSM614461 2 0.0000 0.866 0.000 1.000 0.000 0.000
#> GSM614462 2 0.0000 0.866 0.000 1.000 0.000 0.000
#> GSM614463 2 0.0000 0.866 0.000 1.000 0.000 0.000
#> GSM614464 2 0.0000 0.866 0.000 1.000 0.000 0.000
#> GSM614465 2 0.0000 0.866 0.000 1.000 0.000 0.000
#> GSM614466 2 0.0000 0.866 0.000 1.000 0.000 0.000
#> GSM614467 2 0.0000 0.866 0.000 1.000 0.000 0.000
#> GSM614468 2 0.0000 0.866 0.000 1.000 0.000 0.000
#> GSM614469 1 0.5075 0.496 0.644 0.344 0.012 0.000
#> GSM614470 1 0.5075 0.496 0.644 0.344 0.012 0.000
#> GSM614471 1 0.5075 0.496 0.644 0.344 0.012 0.000
#> GSM614472 1 0.5075 0.496 0.644 0.344 0.012 0.000
#> GSM614473 1 0.5075 0.496 0.644 0.344 0.012 0.000
#> GSM614474 1 0.5075 0.496 0.644 0.344 0.012 0.000
#> GSM614475 1 0.5075 0.496 0.644 0.344 0.012 0.000
#> GSM614476 1 0.5075 0.496 0.644 0.344 0.012 0.000
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM614415 1 0.5524 0.359 0.516 0.000 0.068 0.000 0.416
#> GSM614416 1 0.5524 0.359 0.516 0.000 0.068 0.000 0.416
#> GSM614417 1 0.5524 0.359 0.516 0.000 0.068 0.000 0.416
#> GSM614418 1 0.5524 0.359 0.516 0.000 0.068 0.000 0.416
#> GSM614419 1 0.5524 0.359 0.516 0.000 0.068 0.000 0.416
#> GSM614420 1 0.5524 0.359 0.516 0.000 0.068 0.000 0.416
#> GSM614421 2 0.5641 0.549 0.004 0.584 0.340 0.068 0.004
#> GSM614422 2 0.5641 0.549 0.004 0.584 0.340 0.068 0.004
#> GSM614423 2 0.5641 0.549 0.004 0.584 0.340 0.068 0.004
#> GSM614424 2 0.5641 0.549 0.004 0.584 0.340 0.068 0.004
#> GSM614425 2 0.5641 0.549 0.004 0.584 0.340 0.068 0.004
#> GSM614426 2 0.5641 0.549 0.004 0.584 0.340 0.068 0.004
#> GSM614427 2 0.5641 0.549 0.004 0.584 0.340 0.068 0.004
#> GSM614428 2 0.5641 0.549 0.004 0.584 0.340 0.068 0.004
#> GSM614429 2 0.0000 0.861 0.000 1.000 0.000 0.000 0.000
#> GSM614430 2 0.0000 0.861 0.000 1.000 0.000 0.000 0.000
#> GSM614431 2 0.0000 0.861 0.000 1.000 0.000 0.000 0.000
#> GSM614432 2 0.0000 0.861 0.000 1.000 0.000 0.000 0.000
#> GSM614433 2 0.0000 0.861 0.000 1.000 0.000 0.000 0.000
#> GSM614434 2 0.0000 0.861 0.000 1.000 0.000 0.000 0.000
#> GSM614435 2 0.0000 0.861 0.000 1.000 0.000 0.000 0.000
#> GSM614436 2 0.0000 0.861 0.000 1.000 0.000 0.000 0.000
#> GSM614437 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000
#> GSM614438 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000
#> GSM614439 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000
#> GSM614440 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000
#> GSM614441 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000
#> GSM614442 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000
#> GSM614443 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000
#> GSM614444 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000
#> GSM614391 5 0.0000 1.000 0.000 0.000 0.000 0.000 1.000
#> GSM614392 5 0.0000 1.000 0.000 0.000 0.000 0.000 1.000
#> GSM614393 5 0.0000 1.000 0.000 0.000 0.000 0.000 1.000
#> GSM614394 5 0.0000 1.000 0.000 0.000 0.000 0.000 1.000
#> GSM614395 3 0.4297 -0.118 0.000 0.000 0.528 0.000 0.472
#> GSM614396 5 0.0000 1.000 0.000 0.000 0.000 0.000 1.000
#> GSM614397 5 0.0000 1.000 0.000 0.000 0.000 0.000 1.000
#> GSM614398 5 0.0000 1.000 0.000 0.000 0.000 0.000 1.000
#> GSM614399 2 0.1300 0.842 0.028 0.956 0.016 0.000 0.000
#> GSM614400 2 0.1300 0.842 0.028 0.956 0.016 0.000 0.000
#> GSM614401 2 0.1300 0.842 0.028 0.956 0.016 0.000 0.000
#> GSM614402 2 0.1300 0.842 0.028 0.956 0.016 0.000 0.000
#> GSM614403 2 0.1300 0.842 0.028 0.956 0.016 0.000 0.000
#> GSM614404 2 0.1300 0.842 0.028 0.956 0.016 0.000 0.000
#> GSM614405 2 0.0798 0.854 0.008 0.976 0.016 0.000 0.000
#> GSM614406 2 0.0771 0.854 0.004 0.976 0.020 0.000 0.000
#> GSM614407 1 0.1121 0.519 0.956 0.000 0.000 0.000 0.044
#> GSM614408 1 0.1121 0.519 0.956 0.000 0.000 0.000 0.044
#> GSM614409 1 0.1121 0.519 0.956 0.000 0.000 0.000 0.044
#> GSM614410 1 0.1121 0.519 0.956 0.000 0.000 0.000 0.044
#> GSM614411 1 0.1121 0.519 0.956 0.000 0.000 0.000 0.044
#> GSM614412 1 0.1270 0.513 0.948 0.000 0.000 0.000 0.052
#> GSM614413 1 0.1341 0.510 0.944 0.000 0.000 0.000 0.056
#> GSM614414 1 0.1341 0.510 0.944 0.000 0.000 0.000 0.056
#> GSM614445 2 0.4235 0.641 0.008 0.656 0.336 0.000 0.000
#> GSM614446 2 0.4235 0.641 0.008 0.656 0.336 0.000 0.000
#> GSM614447 2 0.4235 0.641 0.008 0.656 0.336 0.000 0.000
#> GSM614448 2 0.4235 0.641 0.008 0.656 0.336 0.000 0.000
#> GSM614449 2 0.4235 0.641 0.008 0.656 0.336 0.000 0.000
#> GSM614450 2 0.4235 0.641 0.008 0.656 0.336 0.000 0.000
#> GSM614451 3 0.1792 0.675 0.000 0.000 0.916 0.084 0.000
#> GSM614452 3 0.1792 0.675 0.000 0.000 0.916 0.084 0.000
#> GSM614453 2 0.0000 0.861 0.000 1.000 0.000 0.000 0.000
#> GSM614454 2 0.0000 0.861 0.000 1.000 0.000 0.000 0.000
#> GSM614455 2 0.0000 0.861 0.000 1.000 0.000 0.000 0.000
#> GSM614456 2 0.0000 0.861 0.000 1.000 0.000 0.000 0.000
#> GSM614457 2 0.0000 0.861 0.000 1.000 0.000 0.000 0.000
#> GSM614458 2 0.0000 0.861 0.000 1.000 0.000 0.000 0.000
#> GSM614459 2 0.0000 0.861 0.000 1.000 0.000 0.000 0.000
#> GSM614460 2 0.0000 0.861 0.000 1.000 0.000 0.000 0.000
#> GSM614461 2 0.0000 0.861 0.000 1.000 0.000 0.000 0.000
#> GSM614462 2 0.0000 0.861 0.000 1.000 0.000 0.000 0.000
#> GSM614463 2 0.0000 0.861 0.000 1.000 0.000 0.000 0.000
#> GSM614464 2 0.0000 0.861 0.000 1.000 0.000 0.000 0.000
#> GSM614465 2 0.0000 0.861 0.000 1.000 0.000 0.000 0.000
#> GSM614466 2 0.0000 0.861 0.000 1.000 0.000 0.000 0.000
#> GSM614467 2 0.0000 0.861 0.000 1.000 0.000 0.000 0.000
#> GSM614468 2 0.0000 0.861 0.000 1.000 0.000 0.000 0.000
#> GSM614469 1 0.6712 0.553 0.516 0.344 0.064 0.000 0.076
#> GSM614470 1 0.6712 0.553 0.516 0.344 0.064 0.000 0.076
#> GSM614471 1 0.6712 0.553 0.516 0.344 0.064 0.000 0.076
#> GSM614472 1 0.6712 0.553 0.516 0.344 0.064 0.000 0.076
#> GSM614473 1 0.6712 0.553 0.516 0.344 0.064 0.000 0.076
#> GSM614474 1 0.6712 0.553 0.516 0.344 0.064 0.000 0.076
#> GSM614475 1 0.6712 0.553 0.516 0.344 0.064 0.000 0.076
#> GSM614476 1 0.6712 0.553 0.516 0.344 0.064 0.000 0.076
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM614415 1 0.2854 0.6078 0.792 0.000 0.000 0.00 0.208 0.000
#> GSM614416 1 0.2854 0.6078 0.792 0.000 0.000 0.00 0.208 0.000
#> GSM614417 1 0.2854 0.6078 0.792 0.000 0.000 0.00 0.208 0.000
#> GSM614418 1 0.2854 0.6078 0.792 0.000 0.000 0.00 0.208 0.000
#> GSM614419 1 0.2854 0.6078 0.792 0.000 0.000 0.00 0.208 0.000
#> GSM614420 1 0.2854 0.6078 0.792 0.000 0.000 0.00 0.208 0.000
#> GSM614421 3 0.4596 0.7270 0.000 0.476 0.496 0.02 0.004 0.004
#> GSM614422 3 0.4596 0.7270 0.000 0.476 0.496 0.02 0.004 0.004
#> GSM614423 3 0.4596 0.7270 0.000 0.476 0.496 0.02 0.004 0.004
#> GSM614424 3 0.4596 0.7270 0.000 0.476 0.496 0.02 0.004 0.004
#> GSM614425 3 0.4596 0.7270 0.000 0.476 0.496 0.02 0.004 0.004
#> GSM614426 3 0.4596 0.7270 0.000 0.476 0.496 0.02 0.004 0.004
#> GSM614427 3 0.4596 0.7270 0.000 0.476 0.496 0.02 0.004 0.004
#> GSM614428 3 0.4596 0.7270 0.000 0.476 0.496 0.02 0.004 0.004
#> GSM614429 2 0.0547 0.7717 0.000 0.980 0.020 0.00 0.000 0.000
#> GSM614430 2 0.0547 0.7717 0.000 0.980 0.020 0.00 0.000 0.000
#> GSM614431 2 0.0547 0.7717 0.000 0.980 0.020 0.00 0.000 0.000
#> GSM614432 2 0.0547 0.7717 0.000 0.980 0.020 0.00 0.000 0.000
#> GSM614433 2 0.0547 0.7717 0.000 0.980 0.020 0.00 0.000 0.000
#> GSM614434 2 0.0547 0.7717 0.000 0.980 0.020 0.00 0.000 0.000
#> GSM614435 2 0.0632 0.7685 0.000 0.976 0.024 0.00 0.000 0.000
#> GSM614436 2 0.0632 0.7685 0.000 0.976 0.024 0.00 0.000 0.000
#> GSM614437 4 0.0000 1.0000 0.000 0.000 0.000 1.00 0.000 0.000
#> GSM614438 4 0.0000 1.0000 0.000 0.000 0.000 1.00 0.000 0.000
#> GSM614439 4 0.0000 1.0000 0.000 0.000 0.000 1.00 0.000 0.000
#> GSM614440 4 0.0000 1.0000 0.000 0.000 0.000 1.00 0.000 0.000
#> GSM614441 4 0.0000 1.0000 0.000 0.000 0.000 1.00 0.000 0.000
#> GSM614442 4 0.0000 1.0000 0.000 0.000 0.000 1.00 0.000 0.000
#> GSM614443 4 0.0000 1.0000 0.000 0.000 0.000 1.00 0.000 0.000
#> GSM614444 4 0.0000 1.0000 0.000 0.000 0.000 1.00 0.000 0.000
#> GSM614391 5 0.0000 0.9372 0.000 0.000 0.000 0.00 1.000 0.000
#> GSM614392 5 0.0000 0.9372 0.000 0.000 0.000 0.00 1.000 0.000
#> GSM614393 5 0.0000 0.9372 0.000 0.000 0.000 0.00 1.000 0.000
#> GSM614394 5 0.0000 0.9372 0.000 0.000 0.000 0.00 1.000 0.000
#> GSM614395 5 0.5529 0.4823 0.092 0.000 0.424 0.00 0.472 0.012
#> GSM614396 5 0.0000 0.9372 0.000 0.000 0.000 0.00 1.000 0.000
#> GSM614397 5 0.0000 0.9372 0.000 0.000 0.000 0.00 1.000 0.000
#> GSM614398 5 0.0000 0.9372 0.000 0.000 0.000 0.00 1.000 0.000
#> GSM614399 2 0.3771 0.7002 0.132 0.800 0.036 0.00 0.000 0.032
#> GSM614400 2 0.3771 0.7002 0.132 0.800 0.036 0.00 0.000 0.032
#> GSM614401 2 0.3771 0.7002 0.132 0.800 0.036 0.00 0.000 0.032
#> GSM614402 2 0.3771 0.7002 0.132 0.800 0.036 0.00 0.000 0.032
#> GSM614403 2 0.3771 0.7002 0.132 0.800 0.036 0.00 0.000 0.032
#> GSM614404 2 0.3771 0.7002 0.132 0.800 0.036 0.00 0.000 0.032
#> GSM614405 2 0.3509 0.7027 0.068 0.832 0.068 0.00 0.000 0.032
#> GSM614406 2 0.3508 0.7009 0.064 0.832 0.072 0.00 0.000 0.032
#> GSM614407 6 0.1007 0.9939 0.044 0.000 0.000 0.00 0.000 0.956
#> GSM614408 6 0.1007 0.9939 0.044 0.000 0.000 0.00 0.000 0.956
#> GSM614409 6 0.1007 0.9939 0.044 0.000 0.000 0.00 0.000 0.956
#> GSM614410 6 0.1007 0.9939 0.044 0.000 0.000 0.00 0.000 0.956
#> GSM614411 6 0.1007 0.9939 0.044 0.000 0.000 0.00 0.000 0.956
#> GSM614412 6 0.1124 0.9904 0.036 0.000 0.000 0.00 0.008 0.956
#> GSM614413 6 0.1049 0.9881 0.032 0.000 0.000 0.00 0.008 0.960
#> GSM614414 6 0.1049 0.9881 0.032 0.000 0.000 0.00 0.008 0.960
#> GSM614445 2 0.4533 -0.5027 0.008 0.540 0.432 0.00 0.000 0.020
#> GSM614446 2 0.4533 -0.5027 0.008 0.540 0.432 0.00 0.000 0.020
#> GSM614447 2 0.4533 -0.5027 0.008 0.540 0.432 0.00 0.000 0.020
#> GSM614448 2 0.4533 -0.5027 0.008 0.540 0.432 0.00 0.000 0.020
#> GSM614449 2 0.4533 -0.5027 0.008 0.540 0.432 0.00 0.000 0.020
#> GSM614450 2 0.4533 -0.5027 0.008 0.540 0.432 0.00 0.000 0.020
#> GSM614451 3 0.2704 0.0152 0.100 0.000 0.868 0.02 0.000 0.012
#> GSM614452 3 0.2704 0.0152 0.100 0.000 0.868 0.02 0.000 0.012
#> GSM614453 2 0.1501 0.7660 0.076 0.924 0.000 0.00 0.000 0.000
#> GSM614454 2 0.1501 0.7660 0.076 0.924 0.000 0.00 0.000 0.000
#> GSM614455 2 0.1501 0.7660 0.076 0.924 0.000 0.00 0.000 0.000
#> GSM614456 2 0.1501 0.7660 0.076 0.924 0.000 0.00 0.000 0.000
#> GSM614457 2 0.1501 0.7660 0.076 0.924 0.000 0.00 0.000 0.000
#> GSM614458 2 0.1501 0.7660 0.076 0.924 0.000 0.00 0.000 0.000
#> GSM614459 2 0.1501 0.7660 0.076 0.924 0.000 0.00 0.000 0.000
#> GSM614460 2 0.1501 0.7660 0.076 0.924 0.000 0.00 0.000 0.000
#> GSM614461 2 0.0000 0.7802 0.000 1.000 0.000 0.00 0.000 0.000
#> GSM614462 2 0.0000 0.7802 0.000 1.000 0.000 0.00 0.000 0.000
#> GSM614463 2 0.0000 0.7802 0.000 1.000 0.000 0.00 0.000 0.000
#> GSM614464 2 0.0000 0.7802 0.000 1.000 0.000 0.00 0.000 0.000
#> GSM614465 2 0.0000 0.7802 0.000 1.000 0.000 0.00 0.000 0.000
#> GSM614466 2 0.0000 0.7802 0.000 1.000 0.000 0.00 0.000 0.000
#> GSM614467 2 0.0000 0.7802 0.000 1.000 0.000 0.00 0.000 0.000
#> GSM614468 2 0.0000 0.7802 0.000 1.000 0.000 0.00 0.000 0.000
#> GSM614469 1 0.3221 0.7689 0.736 0.264 0.000 0.00 0.000 0.000
#> GSM614470 1 0.3221 0.7689 0.736 0.264 0.000 0.00 0.000 0.000
#> GSM614471 1 0.3221 0.7689 0.736 0.264 0.000 0.00 0.000 0.000
#> GSM614472 1 0.3221 0.7689 0.736 0.264 0.000 0.00 0.000 0.000
#> GSM614473 1 0.3221 0.7689 0.736 0.264 0.000 0.00 0.000 0.000
#> GSM614474 1 0.3221 0.7689 0.736 0.264 0.000 0.00 0.000 0.000
#> GSM614475 1 0.3221 0.7689 0.736 0.264 0.000 0.00 0.000 0.000
#> GSM614476 1 0.3221 0.7689 0.736 0.264 0.000 0.00 0.000 0.000
Heatmaps for the consensus matrix. It visualizes the probability of two samples to be in a same group.
consensus_heatmap(res, k = 2)
consensus_heatmap(res, k = 3)
consensus_heatmap(res, k = 4)
consensus_heatmap(res, k = 5)
consensus_heatmap(res, k = 6)
Heatmaps for the membership of samples in all partitions to see how consistent they are:
membership_heatmap(res, k = 2)
membership_heatmap(res, k = 3)
membership_heatmap(res, k = 4)
membership_heatmap(res, k = 5)
membership_heatmap(res, k = 6)
As soon as we have had the classes for columns, we can look for signatures which are significantly different between classes which can be candidate marks for certain classes. Following are the heatmaps for signatures.
Signature heatmaps where rows are scaled:
get_signatures(res, k = 2)
get_signatures(res, k = 3)
get_signatures(res, k = 4)
get_signatures(res, k = 5)
get_signatures(res, k = 6)
Signature heatmaps where rows are not scaled:
get_signatures(res, k = 2, scale_rows = FALSE)
get_signatures(res, k = 3, scale_rows = FALSE)
get_signatures(res, k = 4, scale_rows = FALSE)
get_signatures(res, k = 5, scale_rows = FALSE)
get_signatures(res, k = 6, scale_rows = FALSE)
Compare the overlap of signatures from different k:
compare_signatures(res)
get_signature() returns a data frame invisibly. TO get the list of signatures, the function
call should be assigned to a variable explicitly. In following code, if plot argument is set
to FALSE, no heatmap is plotted while only the differential analysis is performed.
# code only for demonstration
tb = get_signature(res, k = ..., plot = FALSE)
An example of the output of tb is:
#> which_row fdr mean_1 mean_2 scaled_mean_1 scaled_mean_2 km
#> 1 38 0.042760348 8.373488 9.131774 -0.5533452 0.5164555 1
#> 2 40 0.018707592 7.106213 8.469186 -0.6173731 0.5762149 1
#> 3 55 0.019134737 10.221463 11.207825 -0.6159697 0.5749050 1
#> 4 59 0.006059896 5.921854 7.869574 -0.6899429 0.6439467 1
#> 5 60 0.018055526 8.928898 10.211722 -0.6204761 0.5791110 1
#> 6 98 0.009384629 15.714769 14.887706 0.6635654 -0.6193277 2
...
The columns in tb are:
which_row: row indices corresponding to the input matrix.fdr: FDR for the differential test. mean_x: The mean value in group x.scaled_mean_x: The mean value in group x after rows are scaled.km: Row groups if k-means clustering is applied to rows.UMAP plot which shows how samples are separated.
dimension_reduction(res, k = 2, method = "UMAP")
dimension_reduction(res, k = 3, method = "UMAP")
dimension_reduction(res, k = 4, method = "UMAP")
dimension_reduction(res, k = 5, method = "UMAP")
dimension_reduction(res, k = 6, method = "UMAP")
Following heatmap shows how subgroups are split when increasing k:
collect_classes(res)
Test correlation between subgroups and known annotations. If the known annotation is numeric, one-way ANOVA test is applied, and if the known annotation is discrete, chi-squared contingency table test is applied.
test_to_known_factors(res)
#> n individual(p) protocol(p) time(p) other(p) k
#> MAD:hclust 85 3.47e-11 0.3626 0.992 0.004489 2
#> MAD:hclust 85 8.69e-23 0.4232 1.000 0.038914 3
#> MAD:hclust 77 7.95e-23 0.0766 1.000 0.000202 4
#> MAD:hclust 79 1.22e-34 0.1811 1.000 0.000192 5
#> MAD:hclust 77 1.54e-55 0.9748 1.000 0.031459 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 21168 rows and 86 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 0.188 0.617 0.756 0.4227 0.615 0.615
#> 3 3 0.373 0.537 0.738 0.3952 0.648 0.469
#> 4 4 0.427 0.665 0.746 0.1736 0.770 0.486
#> 5 5 0.497 0.609 0.694 0.0963 0.804 0.464
#> 6 6 0.638 0.656 0.710 0.0510 0.967 0.858
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
#> GSM614415 1 0.8267 0.8871 0.740 0.260
#> GSM614416 1 0.8267 0.8871 0.740 0.260
#> GSM614417 1 0.8267 0.8871 0.740 0.260
#> GSM614418 1 0.8267 0.8871 0.740 0.260
#> GSM614419 1 0.7219 0.8891 0.800 0.200
#> GSM614420 1 0.7219 0.8891 0.800 0.200
#> GSM614421 2 0.9881 0.4440 0.436 0.564
#> GSM614422 2 0.9881 0.4440 0.436 0.564
#> GSM614423 2 0.8555 0.5196 0.280 0.720
#> GSM614424 2 0.9881 0.4440 0.436 0.564
#> GSM614425 2 0.9881 0.4440 0.436 0.564
#> GSM614426 2 0.9881 0.4440 0.436 0.564
#> GSM614427 2 0.9896 0.4381 0.440 0.560
#> GSM614428 2 0.9909 0.4363 0.444 0.556
#> GSM614429 2 0.0376 0.7029 0.004 0.996
#> GSM614430 2 0.0376 0.7029 0.004 0.996
#> GSM614431 2 0.0376 0.7029 0.004 0.996
#> GSM614432 2 0.0376 0.7029 0.004 0.996
#> GSM614433 2 0.0376 0.7029 0.004 0.996
#> GSM614434 2 0.0376 0.7029 0.004 0.996
#> GSM614435 2 0.0672 0.7026 0.008 0.992
#> GSM614436 2 0.5737 0.6493 0.136 0.864
#> GSM614437 2 0.8555 0.5550 0.280 0.720
#> GSM614438 2 0.9248 0.5276 0.340 0.660
#> GSM614439 2 0.9248 0.5276 0.340 0.660
#> GSM614440 2 0.9248 0.5276 0.340 0.660
#> GSM614441 2 0.9248 0.5276 0.340 0.660
#> GSM614442 2 0.9248 0.5276 0.340 0.660
#> GSM614443 2 0.8861 0.5463 0.304 0.696
#> GSM614444 2 0.9248 0.5276 0.340 0.660
#> GSM614391 1 0.7219 0.8891 0.800 0.200
#> GSM614392 1 0.7815 0.8938 0.768 0.232
#> GSM614393 1 0.7883 0.8936 0.764 0.236
#> GSM614394 1 0.7219 0.8891 0.800 0.200
#> GSM614395 1 0.1633 0.6738 0.976 0.024
#> GSM614396 1 0.7219 0.8891 0.800 0.200
#> GSM614397 1 0.5178 0.7978 0.884 0.116
#> GSM614398 1 0.5519 0.8133 0.872 0.128
#> GSM614399 2 0.7602 0.5771 0.220 0.780
#> GSM614400 2 0.7950 0.5506 0.240 0.760
#> GSM614401 2 0.7950 0.5506 0.240 0.760
#> GSM614402 2 0.7950 0.5506 0.240 0.760
#> GSM614403 2 0.6887 0.6114 0.184 0.816
#> GSM614404 2 0.7950 0.5506 0.240 0.760
#> GSM614405 2 0.7745 0.5671 0.228 0.772
#> GSM614406 2 0.9580 0.5353 0.380 0.620
#> GSM614407 1 0.8608 0.8677 0.716 0.284
#> GSM614408 1 0.8608 0.8677 0.716 0.284
#> GSM614409 1 0.8555 0.8723 0.720 0.280
#> GSM614410 1 0.8608 0.8677 0.716 0.284
#> GSM614411 1 0.8608 0.8677 0.716 0.284
#> GSM614412 1 0.8327 0.8860 0.736 0.264
#> GSM614413 1 0.5842 0.8147 0.860 0.140
#> GSM614414 1 0.5842 0.8147 0.860 0.140
#> GSM614445 2 0.6887 0.6149 0.184 0.816
#> GSM614446 2 0.6887 0.6149 0.184 0.816
#> GSM614447 2 0.6887 0.6149 0.184 0.816
#> GSM614448 2 0.9732 0.4953 0.404 0.596
#> GSM614449 2 0.9661 0.5109 0.392 0.608
#> GSM614450 2 0.8016 0.5846 0.244 0.756
#> GSM614451 2 0.9993 0.4446 0.484 0.516
#> GSM614452 2 0.9993 0.4446 0.484 0.516
#> GSM614453 2 0.0376 0.7029 0.004 0.996
#> GSM614454 2 0.0376 0.7029 0.004 0.996
#> GSM614455 2 0.0376 0.7029 0.004 0.996
#> GSM614456 2 0.0376 0.7029 0.004 0.996
#> GSM614457 2 0.0376 0.7029 0.004 0.996
#> GSM614458 2 0.0376 0.7029 0.004 0.996
#> GSM614459 2 0.1414 0.6971 0.020 0.980
#> GSM614460 2 0.0376 0.7029 0.004 0.996
#> GSM614461 2 0.0000 0.7027 0.000 1.000
#> GSM614462 2 0.0000 0.7027 0.000 1.000
#> GSM614463 2 0.0000 0.7027 0.000 1.000
#> GSM614464 2 0.0000 0.7027 0.000 1.000
#> GSM614465 2 0.0000 0.7027 0.000 1.000
#> GSM614466 2 0.0000 0.7027 0.000 1.000
#> GSM614467 2 0.0672 0.7014 0.008 0.992
#> GSM614468 2 0.0000 0.7027 0.000 1.000
#> GSM614469 2 0.9754 0.1004 0.408 0.592
#> GSM614470 2 0.9754 0.1004 0.408 0.592
#> GSM614471 2 0.9754 0.1004 0.408 0.592
#> GSM614472 2 0.9754 0.1004 0.408 0.592
#> GSM614473 2 0.9754 0.1004 0.408 0.592
#> GSM614474 2 0.9754 0.1004 0.408 0.592
#> GSM614475 2 0.9754 0.1004 0.408 0.592
#> GSM614476 2 0.9977 -0.0954 0.472 0.528
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM614415 1 0.1636 0.710 0.964 0.020 0.016
#> GSM614416 1 0.1636 0.710 0.964 0.020 0.016
#> GSM614417 1 0.1636 0.710 0.964 0.020 0.016
#> GSM614418 1 0.1636 0.710 0.964 0.020 0.016
#> GSM614419 1 0.1774 0.708 0.960 0.016 0.024
#> GSM614420 1 0.1774 0.708 0.960 0.016 0.024
#> GSM614421 3 0.9724 0.427 0.224 0.364 0.412
#> GSM614422 3 0.9724 0.427 0.224 0.364 0.412
#> GSM614423 2 0.8746 0.273 0.184 0.588 0.228
#> GSM614424 3 0.9724 0.427 0.224 0.364 0.412
#> GSM614425 3 0.9724 0.427 0.224 0.364 0.412
#> GSM614426 3 0.9724 0.427 0.224 0.364 0.412
#> GSM614427 3 0.9724 0.427 0.224 0.364 0.412
#> GSM614428 3 0.9724 0.427 0.224 0.364 0.412
#> GSM614429 2 0.0592 0.705 0.000 0.988 0.012
#> GSM614430 2 0.0592 0.705 0.000 0.988 0.012
#> GSM614431 2 0.0592 0.705 0.000 0.988 0.012
#> GSM614432 2 0.0592 0.705 0.000 0.988 0.012
#> GSM614433 2 0.0000 0.707 0.000 1.000 0.000
#> GSM614434 2 0.0592 0.705 0.000 0.988 0.012
#> GSM614435 2 0.1289 0.694 0.000 0.968 0.032
#> GSM614436 2 0.3941 0.480 0.000 0.844 0.156
#> GSM614437 3 0.6686 0.592 0.016 0.372 0.612
#> GSM614438 3 0.6667 0.597 0.016 0.368 0.616
#> GSM614439 3 0.6667 0.597 0.016 0.368 0.616
#> GSM614440 3 0.6667 0.597 0.016 0.368 0.616
#> GSM614441 3 0.6667 0.597 0.016 0.368 0.616
#> GSM614442 3 0.6667 0.597 0.016 0.368 0.616
#> GSM614443 3 0.6686 0.592 0.016 0.372 0.612
#> GSM614444 3 0.6667 0.597 0.016 0.368 0.616
#> GSM614391 1 0.3528 0.687 0.892 0.016 0.092
#> GSM614392 1 0.3528 0.687 0.892 0.016 0.092
#> GSM614393 1 0.3528 0.687 0.892 0.016 0.092
#> GSM614394 1 0.3528 0.687 0.892 0.016 0.092
#> GSM614395 1 0.5650 0.459 0.688 0.000 0.312
#> GSM614396 1 0.3528 0.687 0.892 0.016 0.092
#> GSM614397 1 0.5012 0.592 0.788 0.008 0.204
#> GSM614398 1 0.3965 0.663 0.860 0.008 0.132
#> GSM614399 2 0.8374 0.437 0.240 0.616 0.144
#> GSM614400 2 0.8379 0.421 0.268 0.604 0.128
#> GSM614401 2 0.8379 0.421 0.268 0.604 0.128
#> GSM614402 2 0.8430 0.423 0.260 0.604 0.136
#> GSM614403 2 0.8017 0.463 0.140 0.652 0.208
#> GSM614404 2 0.8379 0.421 0.268 0.604 0.128
#> GSM614405 2 0.8473 0.437 0.208 0.616 0.176
#> GSM614406 2 0.8566 -0.124 0.096 0.480 0.424
#> GSM614407 1 0.5042 0.693 0.836 0.060 0.104
#> GSM614408 1 0.5042 0.693 0.836 0.060 0.104
#> GSM614409 1 0.5042 0.693 0.836 0.060 0.104
#> GSM614410 1 0.5042 0.693 0.836 0.060 0.104
#> GSM614411 1 0.5042 0.693 0.836 0.060 0.104
#> GSM614412 1 0.4945 0.693 0.840 0.056 0.104
#> GSM614413 1 0.5412 0.664 0.796 0.032 0.172
#> GSM614414 1 0.5239 0.672 0.808 0.032 0.160
#> GSM614445 2 0.6662 0.524 0.072 0.736 0.192
#> GSM614446 2 0.6794 0.516 0.076 0.728 0.196
#> GSM614447 2 0.6662 0.524 0.072 0.736 0.192
#> GSM614448 2 0.8602 -0.139 0.100 0.492 0.408
#> GSM614449 2 0.8602 -0.139 0.100 0.492 0.408
#> GSM614450 2 0.7916 0.357 0.100 0.636 0.264
#> GSM614451 3 0.6796 0.568 0.056 0.236 0.708
#> GSM614452 3 0.6796 0.568 0.056 0.236 0.708
#> GSM614453 2 0.1964 0.677 0.000 0.944 0.056
#> GSM614454 2 0.1964 0.677 0.000 0.944 0.056
#> GSM614455 2 0.1964 0.677 0.000 0.944 0.056
#> GSM614456 2 0.2066 0.675 0.000 0.940 0.060
#> GSM614457 2 0.2066 0.675 0.000 0.940 0.060
#> GSM614458 2 0.2066 0.675 0.000 0.940 0.060
#> GSM614459 2 0.2066 0.675 0.000 0.940 0.060
#> GSM614460 2 0.2066 0.675 0.000 0.940 0.060
#> GSM614461 2 0.1129 0.707 0.004 0.976 0.020
#> GSM614462 2 0.1129 0.707 0.004 0.976 0.020
#> GSM614463 2 0.1129 0.707 0.004 0.976 0.020
#> GSM614464 2 0.1129 0.707 0.004 0.976 0.020
#> GSM614465 2 0.1129 0.707 0.004 0.976 0.020
#> GSM614466 2 0.1129 0.707 0.004 0.976 0.020
#> GSM614467 2 0.1267 0.706 0.004 0.972 0.024
#> GSM614468 2 0.1129 0.707 0.004 0.976 0.020
#> GSM614469 1 0.8277 0.145 0.468 0.456 0.076
#> GSM614470 1 0.8277 0.145 0.468 0.456 0.076
#> GSM614471 1 0.8277 0.145 0.468 0.456 0.076
#> GSM614472 1 0.8277 0.145 0.468 0.456 0.076
#> GSM614473 1 0.8277 0.145 0.468 0.456 0.076
#> GSM614474 1 0.8277 0.145 0.468 0.456 0.076
#> GSM614475 1 0.8277 0.145 0.468 0.456 0.076
#> GSM614476 1 0.8744 0.118 0.448 0.444 0.108
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM614415 1 0.264 0.8101 0.916 0.020 0.052 0.012
#> GSM614416 1 0.264 0.8101 0.916 0.020 0.052 0.012
#> GSM614417 1 0.264 0.8101 0.916 0.020 0.052 0.012
#> GSM614418 1 0.264 0.8101 0.916 0.020 0.052 0.012
#> GSM614419 1 0.285 0.8112 0.904 0.008 0.064 0.024
#> GSM614420 1 0.285 0.8112 0.904 0.008 0.064 0.024
#> GSM614421 3 0.525 0.8017 0.032 0.148 0.776 0.044
#> GSM614422 3 0.525 0.8017 0.032 0.148 0.776 0.044
#> GSM614423 3 0.546 0.7630 0.036 0.240 0.712 0.012
#> GSM614424 3 0.525 0.8017 0.032 0.148 0.776 0.044
#> GSM614425 3 0.525 0.8017 0.032 0.148 0.776 0.044
#> GSM614426 3 0.525 0.8017 0.032 0.148 0.776 0.044
#> GSM614427 3 0.525 0.8017 0.032 0.148 0.776 0.044
#> GSM614428 3 0.534 0.7982 0.036 0.148 0.772 0.044
#> GSM614429 2 0.152 0.6665 0.000 0.956 0.024 0.020
#> GSM614430 2 0.152 0.6665 0.000 0.956 0.024 0.020
#> GSM614431 2 0.152 0.6665 0.000 0.956 0.024 0.020
#> GSM614432 2 0.152 0.6665 0.000 0.956 0.024 0.020
#> GSM614433 2 0.162 0.6661 0.000 0.952 0.028 0.020
#> GSM614434 2 0.152 0.6665 0.000 0.956 0.024 0.020
#> GSM614435 2 0.163 0.6651 0.000 0.952 0.024 0.024
#> GSM614436 2 0.304 0.6087 0.000 0.888 0.076 0.036
#> GSM614437 4 0.590 0.9898 0.000 0.160 0.140 0.700
#> GSM614438 4 0.608 0.9966 0.004 0.156 0.144 0.696
#> GSM614439 4 0.608 0.9966 0.004 0.156 0.144 0.696
#> GSM614440 4 0.608 0.9966 0.004 0.156 0.144 0.696
#> GSM614441 4 0.608 0.9966 0.004 0.156 0.144 0.696
#> GSM614442 4 0.608 0.9966 0.004 0.156 0.144 0.696
#> GSM614443 4 0.590 0.9898 0.000 0.160 0.140 0.700
#> GSM614444 4 0.608 0.9966 0.004 0.156 0.144 0.696
#> GSM614391 1 0.456 0.7860 0.816 0.008 0.080 0.096
#> GSM614392 1 0.456 0.7860 0.816 0.008 0.080 0.096
#> GSM614393 1 0.456 0.7860 0.816 0.008 0.080 0.096
#> GSM614394 1 0.462 0.7847 0.812 0.008 0.084 0.096
#> GSM614395 1 0.657 0.5443 0.604 0.000 0.280 0.116
#> GSM614396 1 0.462 0.7847 0.812 0.008 0.084 0.096
#> GSM614397 1 0.593 0.6884 0.700 0.004 0.196 0.100
#> GSM614398 1 0.496 0.7679 0.784 0.004 0.116 0.096
#> GSM614399 2 0.846 0.3911 0.148 0.520 0.252 0.080
#> GSM614400 2 0.856 0.4162 0.180 0.516 0.224 0.080
#> GSM614401 2 0.856 0.4162 0.180 0.516 0.224 0.080
#> GSM614402 2 0.851 0.3964 0.156 0.516 0.248 0.080
#> GSM614403 3 0.770 0.2756 0.056 0.356 0.512 0.076
#> GSM614404 2 0.856 0.4162 0.180 0.516 0.224 0.080
#> GSM614405 2 0.847 0.0819 0.104 0.424 0.388 0.084
#> GSM614406 3 0.543 0.7527 0.004 0.196 0.732 0.068
#> GSM614407 1 0.613 0.7521 0.740 0.056 0.096 0.108
#> GSM614408 1 0.613 0.7521 0.740 0.056 0.096 0.108
#> GSM614409 1 0.619 0.7530 0.736 0.056 0.100 0.108
#> GSM614410 1 0.613 0.7521 0.740 0.056 0.096 0.108
#> GSM614411 1 0.619 0.7530 0.736 0.056 0.100 0.108
#> GSM614412 1 0.620 0.7540 0.732 0.048 0.108 0.112
#> GSM614413 1 0.631 0.7180 0.676 0.008 0.200 0.116
#> GSM614414 1 0.628 0.7219 0.680 0.008 0.196 0.116
#> GSM614445 3 0.547 0.6919 0.020 0.296 0.672 0.012
#> GSM614446 3 0.545 0.6977 0.020 0.292 0.676 0.012
#> GSM614447 3 0.547 0.6919 0.020 0.296 0.672 0.012
#> GSM614448 3 0.446 0.7967 0.012 0.164 0.800 0.024
#> GSM614449 3 0.454 0.7953 0.012 0.172 0.792 0.024
#> GSM614450 3 0.512 0.7521 0.020 0.244 0.724 0.012
#> GSM614451 3 0.560 0.4416 0.008 0.044 0.696 0.252
#> GSM614452 3 0.560 0.4416 0.008 0.044 0.696 0.252
#> GSM614453 2 0.274 0.6497 0.000 0.900 0.024 0.076
#> GSM614454 2 0.281 0.6483 0.000 0.896 0.024 0.080
#> GSM614455 2 0.281 0.6483 0.000 0.896 0.024 0.080
#> GSM614456 2 0.281 0.6483 0.000 0.896 0.024 0.080
#> GSM614457 2 0.281 0.6483 0.000 0.896 0.024 0.080
#> GSM614458 2 0.281 0.6483 0.000 0.896 0.024 0.080
#> GSM614459 2 0.281 0.6483 0.000 0.896 0.024 0.080
#> GSM614460 2 0.281 0.6483 0.000 0.896 0.024 0.080
#> GSM614461 2 0.327 0.6494 0.004 0.880 0.084 0.032
#> GSM614462 2 0.327 0.6494 0.004 0.880 0.084 0.032
#> GSM614463 2 0.327 0.6494 0.004 0.880 0.084 0.032
#> GSM614464 2 0.327 0.6494 0.004 0.880 0.084 0.032
#> GSM614465 2 0.327 0.6494 0.004 0.880 0.084 0.032
#> GSM614466 2 0.327 0.6494 0.004 0.880 0.084 0.032
#> GSM614467 2 0.327 0.6494 0.004 0.880 0.084 0.032
#> GSM614468 2 0.327 0.6494 0.004 0.880 0.084 0.032
#> GSM614469 2 0.838 0.2536 0.404 0.412 0.120 0.064
#> GSM614470 2 0.838 0.2536 0.404 0.412 0.120 0.064
#> GSM614471 2 0.838 0.2536 0.404 0.412 0.120 0.064
#> GSM614472 2 0.838 0.2536 0.404 0.412 0.120 0.064
#> GSM614473 2 0.838 0.2536 0.404 0.412 0.120 0.064
#> GSM614474 2 0.838 0.2536 0.404 0.412 0.120 0.064
#> GSM614475 2 0.838 0.2536 0.404 0.412 0.120 0.064
#> GSM614476 2 0.873 0.2658 0.368 0.400 0.168 0.064
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM614415 1 0.601 -0.107 0.560 0.000 0.028 0.064 0.348
#> GSM614416 1 0.601 -0.107 0.560 0.000 0.028 0.064 0.348
#> GSM614417 1 0.601 -0.107 0.560 0.000 0.028 0.064 0.348
#> GSM614418 1 0.601 -0.107 0.560 0.000 0.028 0.064 0.348
#> GSM614419 1 0.605 -0.149 0.548 0.000 0.028 0.064 0.360
#> GSM614420 1 0.605 -0.149 0.548 0.000 0.028 0.064 0.360
#> GSM614421 3 0.620 0.593 0.056 0.052 0.708 0.096 0.088
#> GSM614422 3 0.620 0.593 0.056 0.052 0.708 0.096 0.088
#> GSM614423 3 0.585 0.612 0.060 0.084 0.732 0.040 0.084
#> GSM614424 3 0.620 0.593 0.056 0.052 0.708 0.096 0.088
#> GSM614425 3 0.620 0.593 0.056 0.052 0.708 0.096 0.088
#> GSM614426 3 0.620 0.593 0.056 0.052 0.708 0.096 0.088
#> GSM614427 3 0.620 0.593 0.056 0.052 0.708 0.096 0.088
#> GSM614428 3 0.620 0.593 0.056 0.052 0.708 0.096 0.088
#> GSM614429 2 0.144 0.859 0.004 0.948 0.044 0.000 0.004
#> GSM614430 2 0.144 0.859 0.004 0.948 0.044 0.000 0.004
#> GSM614431 2 0.144 0.859 0.004 0.948 0.044 0.000 0.004
#> GSM614432 2 0.144 0.859 0.004 0.948 0.044 0.000 0.004
#> GSM614433 2 0.144 0.859 0.004 0.948 0.044 0.000 0.004
#> GSM614434 2 0.144 0.859 0.004 0.948 0.044 0.000 0.004
#> GSM614435 2 0.144 0.859 0.004 0.948 0.044 0.000 0.004
#> GSM614436 2 0.202 0.844 0.008 0.924 0.060 0.004 0.004
#> GSM614437 4 0.382 1.000 0.004 0.140 0.048 0.808 0.000
#> GSM614438 4 0.382 1.000 0.004 0.140 0.048 0.808 0.000
#> GSM614439 4 0.382 1.000 0.004 0.140 0.048 0.808 0.000
#> GSM614440 4 0.382 1.000 0.004 0.140 0.048 0.808 0.000
#> GSM614441 4 0.382 1.000 0.004 0.140 0.048 0.808 0.000
#> GSM614442 4 0.382 1.000 0.004 0.140 0.048 0.808 0.000
#> GSM614443 4 0.382 1.000 0.004 0.140 0.048 0.808 0.000
#> GSM614444 4 0.382 1.000 0.004 0.140 0.048 0.808 0.000
#> GSM614391 5 0.403 0.903 0.316 0.000 0.004 0.000 0.680
#> GSM614392 5 0.395 0.884 0.332 0.000 0.000 0.000 0.668
#> GSM614393 5 0.395 0.884 0.332 0.000 0.000 0.000 0.668
#> GSM614394 5 0.403 0.903 0.316 0.000 0.004 0.000 0.680
#> GSM614395 5 0.525 0.771 0.220 0.000 0.088 0.008 0.684
#> GSM614396 5 0.403 0.903 0.316 0.000 0.004 0.000 0.680
#> GSM614397 5 0.476 0.834 0.240 0.000 0.052 0.004 0.704
#> GSM614398 5 0.418 0.872 0.268 0.000 0.020 0.000 0.712
#> GSM614399 3 0.934 0.110 0.228 0.196 0.332 0.060 0.184
#> GSM614400 3 0.934 0.110 0.228 0.196 0.332 0.060 0.184
#> GSM614401 3 0.934 0.110 0.228 0.196 0.332 0.060 0.184
#> GSM614402 3 0.934 0.110 0.228 0.196 0.332 0.060 0.184
#> GSM614403 3 0.822 0.358 0.124 0.120 0.524 0.060 0.172
#> GSM614404 3 0.934 0.110 0.228 0.196 0.332 0.060 0.184
#> GSM614405 3 0.920 0.157 0.212 0.160 0.376 0.064 0.188
#> GSM614406 3 0.635 0.543 0.024 0.068 0.680 0.084 0.144
#> GSM614407 1 0.138 0.466 0.956 0.020 0.020 0.004 0.000
#> GSM614408 1 0.138 0.466 0.956 0.020 0.020 0.004 0.000
#> GSM614409 1 0.154 0.463 0.952 0.020 0.020 0.004 0.004
#> GSM614410 1 0.138 0.466 0.956 0.020 0.020 0.004 0.000
#> GSM614411 1 0.154 0.463 0.952 0.020 0.020 0.004 0.004
#> GSM614412 1 0.144 0.460 0.956 0.016 0.020 0.004 0.004
#> GSM614413 1 0.379 0.324 0.836 0.004 0.080 0.012 0.068
#> GSM614414 1 0.361 0.337 0.848 0.004 0.068 0.012 0.068
#> GSM614445 3 0.329 0.615 0.012 0.100 0.860 0.008 0.020
#> GSM614446 3 0.329 0.615 0.012 0.100 0.860 0.008 0.020
#> GSM614447 3 0.329 0.615 0.012 0.100 0.860 0.008 0.020
#> GSM614448 3 0.274 0.612 0.008 0.064 0.896 0.024 0.008
#> GSM614449 3 0.252 0.612 0.008 0.064 0.904 0.020 0.004
#> GSM614450 3 0.274 0.620 0.012 0.084 0.888 0.004 0.012
#> GSM614451 3 0.546 0.380 0.012 0.028 0.664 0.268 0.028
#> GSM614452 3 0.546 0.380 0.012 0.028 0.664 0.268 0.028
#> GSM614453 2 0.296 0.826 0.004 0.884 0.008 0.048 0.056
#> GSM614454 2 0.317 0.820 0.004 0.872 0.008 0.060 0.056
#> GSM614455 2 0.317 0.820 0.004 0.872 0.008 0.060 0.056
#> GSM614456 2 0.317 0.820 0.004 0.872 0.008 0.060 0.056
#> GSM614457 2 0.317 0.820 0.004 0.872 0.008 0.060 0.056
#> GSM614458 2 0.317 0.820 0.004 0.872 0.008 0.060 0.056
#> GSM614459 2 0.317 0.820 0.004 0.872 0.008 0.060 0.056
#> GSM614460 2 0.317 0.820 0.004 0.872 0.008 0.060 0.056
#> GSM614461 2 0.496 0.806 0.016 0.780 0.064 0.044 0.096
#> GSM614462 2 0.496 0.806 0.016 0.780 0.064 0.044 0.096
#> GSM614463 2 0.496 0.806 0.016 0.780 0.064 0.044 0.096
#> GSM614464 2 0.496 0.806 0.016 0.780 0.064 0.044 0.096
#> GSM614465 2 0.496 0.806 0.016 0.780 0.064 0.044 0.096
#> GSM614466 2 0.496 0.806 0.016 0.780 0.064 0.044 0.096
#> GSM614467 2 0.496 0.806 0.016 0.780 0.064 0.044 0.096
#> GSM614468 2 0.496 0.806 0.016 0.780 0.064 0.044 0.096
#> GSM614469 1 0.840 0.503 0.492 0.176 0.124 0.048 0.160
#> GSM614470 1 0.840 0.503 0.492 0.176 0.124 0.048 0.160
#> GSM614471 1 0.840 0.503 0.492 0.176 0.124 0.048 0.160
#> GSM614472 1 0.840 0.503 0.492 0.176 0.124 0.048 0.160
#> GSM614473 1 0.840 0.503 0.492 0.176 0.124 0.048 0.160
#> GSM614474 1 0.840 0.503 0.492 0.176 0.124 0.048 0.160
#> GSM614475 1 0.840 0.503 0.492 0.176 0.124 0.048 0.160
#> GSM614476 1 0.859 0.471 0.468 0.172 0.140 0.048 0.172
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM614415 6 0.650 0.1845 0.216 0.000 0.004 0.020 0.356 0.404
#> GSM614416 6 0.650 0.1845 0.216 0.000 0.004 0.020 0.356 0.404
#> GSM614417 6 0.650 0.1845 0.216 0.000 0.004 0.020 0.356 0.404
#> GSM614418 6 0.650 0.1845 0.216 0.000 0.004 0.020 0.356 0.404
#> GSM614419 6 0.652 0.1745 0.220 0.000 0.004 0.020 0.360 0.396
#> GSM614420 6 0.652 0.1745 0.220 0.000 0.004 0.020 0.360 0.396
#> GSM614421 3 0.383 0.7993 0.016 0.012 0.836 0.052 0.032 0.052
#> GSM614422 3 0.383 0.7993 0.016 0.012 0.836 0.052 0.032 0.052
#> GSM614423 3 0.364 0.7715 0.028 0.028 0.844 0.004 0.032 0.064
#> GSM614424 3 0.383 0.7993 0.016 0.012 0.836 0.052 0.032 0.052
#> GSM614425 3 0.383 0.7993 0.016 0.012 0.836 0.052 0.032 0.052
#> GSM614426 3 0.383 0.7993 0.016 0.012 0.836 0.052 0.032 0.052
#> GSM614427 3 0.383 0.7993 0.016 0.012 0.836 0.052 0.032 0.052
#> GSM614428 3 0.383 0.7993 0.016 0.012 0.836 0.052 0.032 0.052
#> GSM614429 2 0.283 0.7590 0.012 0.884 0.044 0.040 0.000 0.020
#> GSM614430 2 0.283 0.7590 0.012 0.884 0.044 0.040 0.000 0.020
#> GSM614431 2 0.283 0.7590 0.012 0.884 0.044 0.040 0.000 0.020
#> GSM614432 2 0.283 0.7590 0.012 0.884 0.044 0.040 0.000 0.020
#> GSM614433 2 0.283 0.7590 0.012 0.884 0.044 0.040 0.000 0.020
#> GSM614434 2 0.283 0.7590 0.012 0.884 0.044 0.040 0.000 0.020
#> GSM614435 2 0.283 0.7590 0.012 0.884 0.044 0.040 0.000 0.020
#> GSM614436 2 0.304 0.7547 0.012 0.872 0.048 0.044 0.000 0.024
#> GSM614437 4 0.129 0.9919 0.004 0.020 0.016 0.956 0.000 0.004
#> GSM614438 4 0.152 0.9916 0.000 0.020 0.016 0.948 0.008 0.008
#> GSM614439 4 0.115 0.9932 0.000 0.020 0.016 0.960 0.000 0.004
#> GSM614440 4 0.152 0.9916 0.000 0.020 0.016 0.948 0.008 0.008
#> GSM614441 4 0.100 0.9932 0.000 0.020 0.016 0.964 0.000 0.000
#> GSM614442 4 0.115 0.9928 0.000 0.020 0.016 0.960 0.000 0.004
#> GSM614443 4 0.129 0.9919 0.004 0.020 0.016 0.956 0.000 0.004
#> GSM614444 4 0.152 0.9916 0.000 0.020 0.016 0.948 0.008 0.008
#> GSM614391 5 0.171 0.9512 0.000 0.000 0.000 0.000 0.908 0.092
#> GSM614392 5 0.171 0.9512 0.000 0.000 0.000 0.000 0.908 0.092
#> GSM614393 5 0.171 0.9512 0.000 0.000 0.000 0.000 0.908 0.092
#> GSM614394 5 0.181 0.9527 0.000 0.000 0.004 0.000 0.908 0.088
#> GSM614395 5 0.256 0.8564 0.024 0.000 0.064 0.008 0.892 0.012
#> GSM614396 5 0.181 0.9527 0.000 0.000 0.004 0.000 0.908 0.088
#> GSM614397 5 0.224 0.9086 0.016 0.000 0.036 0.000 0.908 0.040
#> GSM614398 5 0.204 0.9405 0.008 0.000 0.016 0.000 0.912 0.064
#> GSM614399 1 0.713 0.9452 0.504 0.148 0.144 0.000 0.012 0.192
#> GSM614400 1 0.712 0.9470 0.504 0.148 0.140 0.000 0.012 0.196
#> GSM614401 1 0.712 0.9470 0.504 0.148 0.140 0.000 0.012 0.196
#> GSM614402 1 0.712 0.9470 0.504 0.148 0.140 0.000 0.012 0.196
#> GSM614403 1 0.672 0.7513 0.508 0.116 0.268 0.000 0.004 0.104
#> GSM614404 1 0.712 0.9470 0.504 0.148 0.140 0.000 0.012 0.196
#> GSM614405 1 0.710 0.9222 0.508 0.132 0.164 0.000 0.012 0.184
#> GSM614406 3 0.662 -0.1875 0.420 0.064 0.436 0.016 0.024 0.040
#> GSM614407 6 0.271 0.5046 0.000 0.016 0.012 0.000 0.108 0.864
#> GSM614408 6 0.271 0.5046 0.000 0.016 0.012 0.000 0.108 0.864
#> GSM614409 6 0.271 0.5046 0.000 0.016 0.012 0.000 0.108 0.864
#> GSM614410 6 0.271 0.5046 0.000 0.016 0.012 0.000 0.108 0.864
#> GSM614411 6 0.271 0.5046 0.000 0.016 0.012 0.000 0.108 0.864
#> GSM614412 6 0.290 0.4907 0.000 0.012 0.012 0.004 0.120 0.852
#> GSM614413 6 0.423 0.4241 0.012 0.004 0.064 0.008 0.136 0.776
#> GSM614414 6 0.423 0.4241 0.012 0.004 0.064 0.008 0.136 0.776
#> GSM614445 3 0.367 0.6940 0.172 0.024 0.788 0.008 0.008 0.000
#> GSM614446 3 0.367 0.6940 0.172 0.024 0.788 0.008 0.008 0.000
#> GSM614447 3 0.367 0.6940 0.172 0.024 0.788 0.008 0.008 0.000
#> GSM614448 3 0.263 0.7532 0.104 0.008 0.872 0.008 0.008 0.000
#> GSM614449 3 0.252 0.7524 0.104 0.008 0.876 0.008 0.004 0.000
#> GSM614450 3 0.310 0.7306 0.132 0.020 0.836 0.008 0.004 0.000
#> GSM614451 3 0.438 0.6808 0.056 0.000 0.744 0.172 0.028 0.000
#> GSM614452 3 0.438 0.6808 0.056 0.000 0.744 0.172 0.028 0.000
#> GSM614453 2 0.517 0.7205 0.112 0.732 0.020 0.096 0.032 0.008
#> GSM614454 2 0.521 0.7192 0.112 0.728 0.020 0.100 0.032 0.008
#> GSM614455 2 0.521 0.7192 0.112 0.728 0.020 0.100 0.032 0.008
#> GSM614456 2 0.521 0.7192 0.112 0.728 0.020 0.100 0.032 0.008
#> GSM614457 2 0.521 0.7192 0.112 0.728 0.020 0.100 0.032 0.008
#> GSM614458 2 0.521 0.7192 0.112 0.728 0.020 0.100 0.032 0.008
#> GSM614459 2 0.521 0.7192 0.112 0.728 0.020 0.100 0.032 0.008
#> GSM614460 2 0.521 0.7192 0.112 0.728 0.020 0.100 0.032 0.008
#> GSM614461 2 0.506 0.6404 0.200 0.704 0.044 0.004 0.020 0.028
#> GSM614462 2 0.506 0.6404 0.200 0.704 0.044 0.004 0.020 0.028
#> GSM614463 2 0.506 0.6404 0.200 0.704 0.044 0.004 0.020 0.028
#> GSM614464 2 0.506 0.6404 0.200 0.704 0.044 0.004 0.020 0.028
#> GSM614465 2 0.506 0.6404 0.200 0.704 0.044 0.004 0.020 0.028
#> GSM614466 2 0.506 0.6404 0.200 0.704 0.044 0.004 0.020 0.028
#> GSM614467 2 0.506 0.6404 0.200 0.704 0.044 0.004 0.020 0.028
#> GSM614468 2 0.506 0.6404 0.200 0.704 0.044 0.004 0.020 0.028
#> GSM614469 6 0.704 0.1120 0.244 0.184 0.048 0.004 0.024 0.496
#> GSM614470 6 0.704 0.1120 0.244 0.184 0.048 0.004 0.024 0.496
#> GSM614471 6 0.704 0.1120 0.244 0.184 0.048 0.004 0.024 0.496
#> GSM614472 6 0.704 0.1120 0.244 0.184 0.048 0.004 0.024 0.496
#> GSM614473 6 0.704 0.1120 0.244 0.184 0.048 0.004 0.024 0.496
#> GSM614474 6 0.704 0.1120 0.244 0.184 0.048 0.004 0.024 0.496
#> GSM614475 6 0.704 0.1120 0.244 0.184 0.048 0.004 0.024 0.496
#> GSM614476 6 0.720 0.0734 0.248 0.176 0.064 0.004 0.024 0.484
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 individual(p) protocol(p) time(p) other(p) k
#> MAD:kmeans 68 3.75e-11 0.985 0.999 0.6720 2
#> MAD:kmeans 57 6.39e-16 0.361 1.000 0.1408 3
#> MAD:kmeans 69 1.15e-29 0.942 1.000 0.0415 4
#> MAD:kmeans 62 3.07e-35 0.977 1.000 0.0177 5
#> MAD:kmeans 68 3.27e-49 0.857 1.000 0.0799 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 21168 rows and 86 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 0.706 0.885 0.940 0.4999 0.501 0.501
#> 3 3 0.744 0.788 0.905 0.3397 0.694 0.460
#> 4 4 0.759 0.790 0.870 0.1155 0.851 0.588
#> 5 5 0.744 0.829 0.871 0.0610 0.950 0.801
#> 6 6 0.792 0.790 0.829 0.0364 1.000 1.000
suggest_best_k() suggests the best \(k\) based on these statistics. The rules are as follows:
suggest_best_k(res)
#> [1] 2
Following shows the table of the partitions (You need to click the show/hide
code output link to see it). The membership matrix (columns with name p*)
is inferred by
clue::cl_consensus()
function with the SE method. Basically the value in the membership matrix
represents the probability to belong to a certain group. The finall class
label for an item is determined with the group with highest probability it
belongs to.
In get_classes() function, the entropy is calculated from the membership
matrix and the silhouette score is calculated from the consensus matrix.
cbind(get_classes(res, k = 2), get_membership(res, k = 2))
#> class entropy silhouette p1 p2
#> GSM614415 1 0.0000 0.940 1.000 0.000
#> GSM614416 1 0.0000 0.940 1.000 0.000
#> GSM614417 1 0.0000 0.940 1.000 0.000
#> GSM614418 1 0.0000 0.940 1.000 0.000
#> GSM614419 1 0.0000 0.940 1.000 0.000
#> GSM614420 1 0.0000 0.940 1.000 0.000
#> GSM614421 2 0.9358 0.566 0.352 0.648
#> GSM614422 1 0.4815 0.847 0.896 0.104
#> GSM614423 2 0.9491 0.534 0.368 0.632
#> GSM614424 2 0.9358 0.566 0.352 0.648
#> GSM614425 2 0.9358 0.566 0.352 0.648
#> GSM614426 2 0.9358 0.566 0.352 0.648
#> GSM614427 2 0.9358 0.566 0.352 0.648
#> GSM614428 2 0.9358 0.566 0.352 0.648
#> GSM614429 2 0.0000 0.931 0.000 1.000
#> GSM614430 2 0.0000 0.931 0.000 1.000
#> GSM614431 2 0.0000 0.931 0.000 1.000
#> GSM614432 2 0.0000 0.931 0.000 1.000
#> GSM614433 2 0.0000 0.931 0.000 1.000
#> GSM614434 2 0.0000 0.931 0.000 1.000
#> GSM614435 2 0.0000 0.931 0.000 1.000
#> GSM614436 2 0.0000 0.931 0.000 1.000
#> GSM614437 2 0.0000 0.931 0.000 1.000
#> GSM614438 2 0.2043 0.923 0.032 0.968
#> GSM614439 2 0.2043 0.923 0.032 0.968
#> GSM614440 2 0.2043 0.923 0.032 0.968
#> GSM614441 2 0.2043 0.923 0.032 0.968
#> GSM614442 2 0.2043 0.923 0.032 0.968
#> GSM614443 2 0.0938 0.929 0.012 0.988
#> GSM614444 2 0.2043 0.923 0.032 0.968
#> GSM614391 1 0.0000 0.940 1.000 0.000
#> GSM614392 1 0.0000 0.940 1.000 0.000
#> GSM614393 1 0.0000 0.940 1.000 0.000
#> GSM614394 1 0.0000 0.940 1.000 0.000
#> GSM614395 1 0.0000 0.940 1.000 0.000
#> GSM614396 1 0.0000 0.940 1.000 0.000
#> GSM614397 1 0.0000 0.940 1.000 0.000
#> GSM614398 1 0.0000 0.940 1.000 0.000
#> GSM614399 1 0.7883 0.750 0.764 0.236
#> GSM614400 1 0.7815 0.755 0.768 0.232
#> GSM614401 1 0.7815 0.755 0.768 0.232
#> GSM614402 1 0.7815 0.755 0.768 0.232
#> GSM614403 1 0.7745 0.758 0.772 0.228
#> GSM614404 1 0.7815 0.755 0.768 0.232
#> GSM614405 1 0.7815 0.755 0.768 0.232
#> GSM614406 2 0.3114 0.912 0.056 0.944
#> GSM614407 1 0.0000 0.940 1.000 0.000
#> GSM614408 1 0.0000 0.940 1.000 0.000
#> GSM614409 1 0.0000 0.940 1.000 0.000
#> GSM614410 1 0.0000 0.940 1.000 0.000
#> GSM614411 1 0.0000 0.940 1.000 0.000
#> GSM614412 1 0.0000 0.940 1.000 0.000
#> GSM614413 1 0.0000 0.940 1.000 0.000
#> GSM614414 1 0.0000 0.940 1.000 0.000
#> GSM614445 2 0.2043 0.921 0.032 0.968
#> GSM614446 2 0.2236 0.920 0.036 0.964
#> GSM614447 2 0.2043 0.921 0.032 0.968
#> GSM614448 2 0.3431 0.907 0.064 0.936
#> GSM614449 2 0.3431 0.907 0.064 0.936
#> GSM614450 2 0.3274 0.910 0.060 0.940
#> GSM614451 2 0.2603 0.916 0.044 0.956
#> GSM614452 2 0.2603 0.916 0.044 0.956
#> GSM614453 2 0.0000 0.931 0.000 1.000
#> GSM614454 2 0.0000 0.931 0.000 1.000
#> GSM614455 2 0.0000 0.931 0.000 1.000
#> GSM614456 2 0.0000 0.931 0.000 1.000
#> GSM614457 2 0.0000 0.931 0.000 1.000
#> GSM614458 2 0.0000 0.931 0.000 1.000
#> GSM614459 2 0.0000 0.931 0.000 1.000
#> GSM614460 2 0.0000 0.931 0.000 1.000
#> GSM614461 2 0.0000 0.931 0.000 1.000
#> GSM614462 2 0.0000 0.931 0.000 1.000
#> GSM614463 2 0.0000 0.931 0.000 1.000
#> GSM614464 2 0.0000 0.931 0.000 1.000
#> GSM614465 2 0.0000 0.931 0.000 1.000
#> GSM614466 2 0.0000 0.931 0.000 1.000
#> GSM614467 2 0.0000 0.931 0.000 1.000
#> GSM614468 2 0.0000 0.931 0.000 1.000
#> GSM614469 1 0.2603 0.925 0.956 0.044
#> GSM614470 1 0.2603 0.925 0.956 0.044
#> GSM614471 1 0.2603 0.925 0.956 0.044
#> GSM614472 1 0.2603 0.925 0.956 0.044
#> GSM614473 1 0.2603 0.925 0.956 0.044
#> GSM614474 1 0.2603 0.925 0.956 0.044
#> GSM614475 1 0.2603 0.925 0.956 0.044
#> GSM614476 1 0.1414 0.934 0.980 0.020
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM614415 1 0.0000 0.975 1.000 0.000 0.000
#> GSM614416 1 0.0000 0.975 1.000 0.000 0.000
#> GSM614417 1 0.0000 0.975 1.000 0.000 0.000
#> GSM614418 1 0.0000 0.975 1.000 0.000 0.000
#> GSM614419 1 0.0000 0.975 1.000 0.000 0.000
#> GSM614420 1 0.0000 0.975 1.000 0.000 0.000
#> GSM614421 3 0.0000 0.834 0.000 0.000 1.000
#> GSM614422 3 0.0000 0.834 0.000 0.000 1.000
#> GSM614423 3 0.0424 0.830 0.008 0.000 0.992
#> GSM614424 3 0.0000 0.834 0.000 0.000 1.000
#> GSM614425 3 0.0000 0.834 0.000 0.000 1.000
#> GSM614426 3 0.0000 0.834 0.000 0.000 1.000
#> GSM614427 3 0.0000 0.834 0.000 0.000 1.000
#> GSM614428 3 0.0000 0.834 0.000 0.000 1.000
#> GSM614429 2 0.0000 0.852 0.000 1.000 0.000
#> GSM614430 2 0.0000 0.852 0.000 1.000 0.000
#> GSM614431 2 0.0000 0.852 0.000 1.000 0.000
#> GSM614432 2 0.0000 0.852 0.000 1.000 0.000
#> GSM614433 2 0.0000 0.852 0.000 1.000 0.000
#> GSM614434 2 0.0000 0.852 0.000 1.000 0.000
#> GSM614435 2 0.0000 0.852 0.000 1.000 0.000
#> GSM614436 2 0.6062 0.143 0.000 0.616 0.384
#> GSM614437 3 0.5465 0.643 0.000 0.288 0.712
#> GSM614438 3 0.4605 0.743 0.000 0.204 0.796
#> GSM614439 3 0.4605 0.743 0.000 0.204 0.796
#> GSM614440 3 0.4605 0.743 0.000 0.204 0.796
#> GSM614441 3 0.4605 0.743 0.000 0.204 0.796
#> GSM614442 3 0.4605 0.743 0.000 0.204 0.796
#> GSM614443 3 0.5397 0.654 0.000 0.280 0.720
#> GSM614444 3 0.4605 0.743 0.000 0.204 0.796
#> GSM614391 1 0.0000 0.975 1.000 0.000 0.000
#> GSM614392 1 0.0000 0.975 1.000 0.000 0.000
#> GSM614393 1 0.0000 0.975 1.000 0.000 0.000
#> GSM614394 1 0.0000 0.975 1.000 0.000 0.000
#> GSM614395 3 0.4654 0.680 0.208 0.000 0.792
#> GSM614396 1 0.0000 0.975 1.000 0.000 0.000
#> GSM614397 3 0.6252 0.199 0.444 0.000 0.556
#> GSM614398 1 0.2356 0.909 0.928 0.000 0.072
#> GSM614399 2 0.8825 0.488 0.288 0.560 0.152
#> GSM614400 2 0.8921 0.397 0.348 0.516 0.136
#> GSM614401 2 0.8935 0.389 0.352 0.512 0.136
#> GSM614402 2 0.8971 0.414 0.336 0.520 0.144
#> GSM614403 2 0.9383 0.227 0.172 0.444 0.384
#> GSM614404 2 0.8955 0.402 0.344 0.516 0.140
#> GSM614405 3 0.9773 -0.156 0.232 0.372 0.396
#> GSM614406 3 0.0000 0.834 0.000 0.000 1.000
#> GSM614407 1 0.0000 0.975 1.000 0.000 0.000
#> GSM614408 1 0.0000 0.975 1.000 0.000 0.000
#> GSM614409 1 0.0000 0.975 1.000 0.000 0.000
#> GSM614410 1 0.0000 0.975 1.000 0.000 0.000
#> GSM614411 1 0.0000 0.975 1.000 0.000 0.000
#> GSM614412 1 0.0000 0.975 1.000 0.000 0.000
#> GSM614413 1 0.4887 0.687 0.772 0.000 0.228
#> GSM614414 1 0.2878 0.884 0.904 0.000 0.096
#> GSM614445 2 0.5835 0.497 0.000 0.660 0.340
#> GSM614446 3 0.6302 -0.105 0.000 0.480 0.520
#> GSM614447 2 0.6079 0.410 0.000 0.612 0.388
#> GSM614448 3 0.0000 0.834 0.000 0.000 1.000
#> GSM614449 3 0.0000 0.834 0.000 0.000 1.000
#> GSM614450 3 0.0424 0.830 0.000 0.008 0.992
#> GSM614451 3 0.0000 0.834 0.000 0.000 1.000
#> GSM614452 3 0.0000 0.834 0.000 0.000 1.000
#> GSM614453 2 0.0000 0.852 0.000 1.000 0.000
#> GSM614454 2 0.0000 0.852 0.000 1.000 0.000
#> GSM614455 2 0.0000 0.852 0.000 1.000 0.000
#> GSM614456 2 0.0000 0.852 0.000 1.000 0.000
#> GSM614457 2 0.0000 0.852 0.000 1.000 0.000
#> GSM614458 2 0.0000 0.852 0.000 1.000 0.000
#> GSM614459 2 0.0000 0.852 0.000 1.000 0.000
#> GSM614460 2 0.0000 0.852 0.000 1.000 0.000
#> GSM614461 2 0.0000 0.852 0.000 1.000 0.000
#> GSM614462 2 0.0000 0.852 0.000 1.000 0.000
#> GSM614463 2 0.0000 0.852 0.000 1.000 0.000
#> GSM614464 2 0.0000 0.852 0.000 1.000 0.000
#> GSM614465 2 0.0000 0.852 0.000 1.000 0.000
#> GSM614466 2 0.0000 0.852 0.000 1.000 0.000
#> GSM614467 2 0.0000 0.852 0.000 1.000 0.000
#> GSM614468 2 0.0000 0.852 0.000 1.000 0.000
#> GSM614469 1 0.1031 0.964 0.976 0.024 0.000
#> GSM614470 1 0.1031 0.964 0.976 0.024 0.000
#> GSM614471 1 0.1031 0.964 0.976 0.024 0.000
#> GSM614472 1 0.1031 0.964 0.976 0.024 0.000
#> GSM614473 1 0.1031 0.964 0.976 0.024 0.000
#> GSM614474 1 0.1031 0.964 0.976 0.024 0.000
#> GSM614475 1 0.1031 0.964 0.976 0.024 0.000
#> GSM614476 1 0.1411 0.952 0.964 0.000 0.036
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM614415 1 0.0000 0.9414 1.000 0.000 0.000 0.000
#> GSM614416 1 0.0000 0.9414 1.000 0.000 0.000 0.000
#> GSM614417 1 0.0000 0.9414 1.000 0.000 0.000 0.000
#> GSM614418 1 0.0000 0.9414 1.000 0.000 0.000 0.000
#> GSM614419 1 0.0188 0.9406 0.996 0.000 0.004 0.000
#> GSM614420 1 0.0188 0.9406 0.996 0.000 0.004 0.000
#> GSM614421 3 0.1637 0.7884 0.000 0.000 0.940 0.060
#> GSM614422 3 0.1637 0.7884 0.000 0.000 0.940 0.060
#> GSM614423 3 0.2281 0.7724 0.000 0.000 0.904 0.096
#> GSM614424 3 0.1637 0.7884 0.000 0.000 0.940 0.060
#> GSM614425 3 0.1637 0.7884 0.000 0.000 0.940 0.060
#> GSM614426 3 0.1637 0.7884 0.000 0.000 0.940 0.060
#> GSM614427 3 0.1118 0.7903 0.000 0.000 0.964 0.036
#> GSM614428 3 0.1118 0.7903 0.000 0.000 0.964 0.036
#> GSM614429 2 0.0592 0.9436 0.000 0.984 0.000 0.016
#> GSM614430 2 0.0592 0.9436 0.000 0.984 0.000 0.016
#> GSM614431 2 0.0592 0.9436 0.000 0.984 0.000 0.016
#> GSM614432 2 0.0707 0.9429 0.000 0.980 0.000 0.020
#> GSM614433 2 0.0817 0.9417 0.000 0.976 0.000 0.024
#> GSM614434 2 0.0592 0.9436 0.000 0.984 0.000 0.016
#> GSM614435 2 0.0188 0.9423 0.000 0.996 0.000 0.004
#> GSM614436 2 0.2021 0.8912 0.000 0.936 0.024 0.040
#> GSM614437 3 0.5998 0.6765 0.000 0.248 0.664 0.088
#> GSM614438 3 0.5496 0.7300 0.000 0.188 0.724 0.088
#> GSM614439 3 0.5496 0.7300 0.000 0.188 0.724 0.088
#> GSM614440 3 0.5496 0.7300 0.000 0.188 0.724 0.088
#> GSM614441 3 0.5496 0.7300 0.000 0.188 0.724 0.088
#> GSM614442 3 0.5496 0.7300 0.000 0.188 0.724 0.088
#> GSM614443 3 0.5880 0.6918 0.000 0.232 0.680 0.088
#> GSM614444 3 0.5496 0.7300 0.000 0.188 0.724 0.088
#> GSM614391 1 0.0000 0.9414 1.000 0.000 0.000 0.000
#> GSM614392 1 0.0000 0.9414 1.000 0.000 0.000 0.000
#> GSM614393 1 0.0000 0.9414 1.000 0.000 0.000 0.000
#> GSM614394 1 0.0188 0.9406 0.996 0.000 0.004 0.000
#> GSM614395 1 0.5731 0.2156 0.544 0.000 0.428 0.028
#> GSM614396 1 0.0188 0.9406 0.996 0.000 0.004 0.000
#> GSM614397 1 0.4574 0.6515 0.756 0.000 0.220 0.024
#> GSM614398 1 0.1004 0.9219 0.972 0.000 0.024 0.004
#> GSM614399 4 0.2174 0.6796 0.020 0.052 0.000 0.928
#> GSM614400 4 0.2408 0.6924 0.036 0.044 0.000 0.920
#> GSM614401 4 0.2408 0.6924 0.036 0.044 0.000 0.920
#> GSM614402 4 0.2408 0.6924 0.036 0.044 0.000 0.920
#> GSM614403 4 0.2966 0.6377 0.008 0.020 0.076 0.896
#> GSM614404 4 0.2408 0.6924 0.036 0.044 0.000 0.920
#> GSM614405 4 0.1953 0.6583 0.012 0.012 0.032 0.944
#> GSM614406 3 0.4158 0.7347 0.000 0.008 0.768 0.224
#> GSM614407 1 0.0188 0.9410 0.996 0.000 0.000 0.004
#> GSM614408 1 0.0188 0.9410 0.996 0.000 0.000 0.004
#> GSM614409 1 0.0188 0.9410 0.996 0.000 0.000 0.004
#> GSM614410 1 0.0188 0.9410 0.996 0.000 0.000 0.004
#> GSM614411 1 0.0188 0.9410 0.996 0.000 0.000 0.004
#> GSM614412 1 0.0188 0.9410 0.996 0.000 0.000 0.004
#> GSM614413 1 0.2676 0.8423 0.896 0.000 0.092 0.012
#> GSM614414 1 0.1356 0.9132 0.960 0.000 0.032 0.008
#> GSM614445 4 0.7090 0.0254 0.000 0.132 0.372 0.496
#> GSM614446 3 0.6471 0.2297 0.000 0.072 0.512 0.416
#> GSM614447 4 0.6921 -0.1039 0.000 0.108 0.424 0.468
#> GSM614448 3 0.3649 0.6897 0.000 0.000 0.796 0.204
#> GSM614449 3 0.3764 0.6786 0.000 0.000 0.784 0.216
#> GSM614450 3 0.4624 0.5064 0.000 0.000 0.660 0.340
#> GSM614451 3 0.0921 0.7857 0.000 0.000 0.972 0.028
#> GSM614452 3 0.0707 0.7868 0.000 0.000 0.980 0.020
#> GSM614453 2 0.0188 0.9423 0.000 0.996 0.000 0.004
#> GSM614454 2 0.0188 0.9423 0.000 0.996 0.000 0.004
#> GSM614455 2 0.0188 0.9423 0.000 0.996 0.000 0.004
#> GSM614456 2 0.0188 0.9392 0.000 0.996 0.000 0.004
#> GSM614457 2 0.0336 0.9373 0.000 0.992 0.000 0.008
#> GSM614458 2 0.0188 0.9392 0.000 0.996 0.000 0.004
#> GSM614459 2 0.0336 0.9373 0.000 0.992 0.000 0.008
#> GSM614460 2 0.0336 0.9373 0.000 0.992 0.000 0.008
#> GSM614461 2 0.2999 0.8967 0.000 0.864 0.004 0.132
#> GSM614462 2 0.2999 0.8967 0.000 0.864 0.004 0.132
#> GSM614463 2 0.2999 0.8967 0.000 0.864 0.004 0.132
#> GSM614464 2 0.2999 0.8967 0.000 0.864 0.004 0.132
#> GSM614465 2 0.2999 0.8967 0.000 0.864 0.004 0.132
#> GSM614466 2 0.2999 0.8967 0.000 0.864 0.004 0.132
#> GSM614467 2 0.2944 0.9007 0.000 0.868 0.004 0.128
#> GSM614468 2 0.2999 0.8967 0.000 0.864 0.004 0.132
#> GSM614469 4 0.5070 0.6317 0.372 0.008 0.000 0.620
#> GSM614470 4 0.5070 0.6317 0.372 0.008 0.000 0.620
#> GSM614471 4 0.5070 0.6317 0.372 0.008 0.000 0.620
#> GSM614472 4 0.5070 0.6317 0.372 0.008 0.000 0.620
#> GSM614473 4 0.5070 0.6317 0.372 0.008 0.000 0.620
#> GSM614474 4 0.5070 0.6317 0.372 0.008 0.000 0.620
#> GSM614475 4 0.5070 0.6317 0.372 0.008 0.000 0.620
#> GSM614476 4 0.5481 0.6324 0.348 0.004 0.020 0.628
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM614415 5 0.0963 0.907 0.036 0.000 0.000 0.000 0.964
#> GSM614416 5 0.0963 0.907 0.036 0.000 0.000 0.000 0.964
#> GSM614417 5 0.0963 0.907 0.036 0.000 0.000 0.000 0.964
#> GSM614418 5 0.0963 0.907 0.036 0.000 0.000 0.000 0.964
#> GSM614419 5 0.0880 0.908 0.032 0.000 0.000 0.000 0.968
#> GSM614420 5 0.0794 0.908 0.028 0.000 0.000 0.000 0.972
#> GSM614421 3 0.3086 0.805 0.000 0.000 0.816 0.180 0.004
#> GSM614422 3 0.3086 0.805 0.000 0.000 0.816 0.180 0.004
#> GSM614423 3 0.3693 0.788 0.044 0.000 0.824 0.124 0.008
#> GSM614424 3 0.3086 0.805 0.000 0.000 0.816 0.180 0.004
#> GSM614425 3 0.3086 0.805 0.000 0.000 0.816 0.180 0.004
#> GSM614426 3 0.3086 0.805 0.000 0.000 0.816 0.180 0.004
#> GSM614427 3 0.3123 0.803 0.000 0.000 0.812 0.184 0.004
#> GSM614428 3 0.3123 0.803 0.000 0.000 0.812 0.184 0.004
#> GSM614429 2 0.0162 0.885 0.000 0.996 0.000 0.004 0.000
#> GSM614430 2 0.0162 0.885 0.000 0.996 0.000 0.004 0.000
#> GSM614431 2 0.0162 0.885 0.000 0.996 0.000 0.004 0.000
#> GSM614432 2 0.0162 0.885 0.000 0.996 0.000 0.004 0.000
#> GSM614433 2 0.0000 0.884 0.000 1.000 0.000 0.000 0.000
#> GSM614434 2 0.0162 0.885 0.000 0.996 0.000 0.004 0.000
#> GSM614435 2 0.0290 0.884 0.000 0.992 0.000 0.008 0.000
#> GSM614436 2 0.2127 0.834 0.000 0.892 0.000 0.108 0.000
#> GSM614437 4 0.1357 0.925 0.000 0.048 0.004 0.948 0.000
#> GSM614438 4 0.1106 0.948 0.000 0.024 0.012 0.964 0.000
#> GSM614439 4 0.1106 0.948 0.000 0.024 0.012 0.964 0.000
#> GSM614440 4 0.1106 0.948 0.000 0.024 0.012 0.964 0.000
#> GSM614441 4 0.1106 0.948 0.000 0.024 0.012 0.964 0.000
#> GSM614442 4 0.1106 0.948 0.000 0.024 0.012 0.964 0.000
#> GSM614443 4 0.1205 0.934 0.000 0.040 0.004 0.956 0.000
#> GSM614444 4 0.1106 0.948 0.000 0.024 0.012 0.964 0.000
#> GSM614391 5 0.1442 0.903 0.012 0.000 0.032 0.004 0.952
#> GSM614392 5 0.1173 0.906 0.012 0.000 0.020 0.004 0.964
#> GSM614393 5 0.1074 0.907 0.012 0.000 0.016 0.004 0.968
#> GSM614394 5 0.1682 0.899 0.012 0.000 0.044 0.004 0.940
#> GSM614395 5 0.5420 0.577 0.004 0.000 0.112 0.220 0.664
#> GSM614396 5 0.1605 0.900 0.012 0.000 0.040 0.004 0.944
#> GSM614397 5 0.3234 0.837 0.004 0.000 0.092 0.048 0.856
#> GSM614398 5 0.2349 0.875 0.012 0.000 0.084 0.004 0.900
#> GSM614399 1 0.3827 0.756 0.812 0.020 0.144 0.024 0.000
#> GSM614400 1 0.3732 0.757 0.816 0.016 0.144 0.024 0.000
#> GSM614401 1 0.3732 0.757 0.816 0.016 0.144 0.024 0.000
#> GSM614402 1 0.3827 0.756 0.812 0.020 0.144 0.024 0.000
#> GSM614403 1 0.3989 0.726 0.784 0.012 0.180 0.024 0.000
#> GSM614404 1 0.3827 0.756 0.812 0.020 0.144 0.024 0.000
#> GSM614405 1 0.3887 0.747 0.804 0.004 0.152 0.036 0.004
#> GSM614406 4 0.5083 0.561 0.140 0.000 0.160 0.700 0.000
#> GSM614407 5 0.2699 0.876 0.100 0.000 0.008 0.012 0.880
#> GSM614408 5 0.2805 0.872 0.108 0.000 0.008 0.012 0.872
#> GSM614409 5 0.2589 0.880 0.092 0.000 0.008 0.012 0.888
#> GSM614410 5 0.2699 0.876 0.100 0.000 0.008 0.012 0.880
#> GSM614411 5 0.2645 0.879 0.096 0.000 0.008 0.012 0.884
#> GSM614412 5 0.2414 0.884 0.080 0.000 0.008 0.012 0.900
#> GSM614413 5 0.3871 0.851 0.040 0.000 0.112 0.024 0.824
#> GSM614414 5 0.3426 0.877 0.052 0.000 0.084 0.012 0.852
#> GSM614445 3 0.3190 0.699 0.140 0.012 0.840 0.008 0.000
#> GSM614446 3 0.2911 0.711 0.136 0.004 0.852 0.008 0.000
#> GSM614447 3 0.2911 0.706 0.136 0.008 0.852 0.004 0.000
#> GSM614448 3 0.3862 0.758 0.104 0.000 0.808 0.088 0.000
#> GSM614449 3 0.3670 0.754 0.112 0.000 0.820 0.068 0.000
#> GSM614450 3 0.3165 0.737 0.116 0.000 0.848 0.036 0.000
#> GSM614451 3 0.4268 0.456 0.000 0.000 0.556 0.444 0.000
#> GSM614452 3 0.4201 0.535 0.000 0.000 0.592 0.408 0.000
#> GSM614453 2 0.2424 0.849 0.000 0.868 0.000 0.132 0.000
#> GSM614454 2 0.2471 0.847 0.000 0.864 0.000 0.136 0.000
#> GSM614455 2 0.2471 0.847 0.000 0.864 0.000 0.136 0.000
#> GSM614456 2 0.2516 0.846 0.000 0.860 0.000 0.140 0.000
#> GSM614457 2 0.2516 0.846 0.000 0.860 0.000 0.140 0.000
#> GSM614458 2 0.2516 0.846 0.000 0.860 0.000 0.140 0.000
#> GSM614459 2 0.2516 0.846 0.000 0.860 0.000 0.140 0.000
#> GSM614460 2 0.2516 0.846 0.000 0.860 0.000 0.140 0.000
#> GSM614461 2 0.3485 0.850 0.060 0.852 0.072 0.016 0.000
#> GSM614462 2 0.3485 0.850 0.060 0.852 0.072 0.016 0.000
#> GSM614463 2 0.3485 0.850 0.060 0.852 0.072 0.016 0.000
#> GSM614464 2 0.3485 0.850 0.060 0.852 0.072 0.016 0.000
#> GSM614465 2 0.3485 0.850 0.060 0.852 0.072 0.016 0.000
#> GSM614466 2 0.3485 0.850 0.060 0.852 0.072 0.016 0.000
#> GSM614467 2 0.3485 0.850 0.060 0.852 0.072 0.016 0.000
#> GSM614468 2 0.3485 0.850 0.060 0.852 0.072 0.016 0.000
#> GSM614469 1 0.3203 0.795 0.820 0.000 0.012 0.000 0.168
#> GSM614470 1 0.3203 0.795 0.820 0.000 0.012 0.000 0.168
#> GSM614471 1 0.3203 0.795 0.820 0.000 0.012 0.000 0.168
#> GSM614472 1 0.3203 0.795 0.820 0.000 0.012 0.000 0.168
#> GSM614473 1 0.3203 0.795 0.820 0.000 0.012 0.000 0.168
#> GSM614474 1 0.3203 0.795 0.820 0.000 0.012 0.000 0.168
#> GSM614475 1 0.3203 0.795 0.820 0.000 0.012 0.000 0.168
#> GSM614476 1 0.3799 0.787 0.812 0.000 0.032 0.012 0.144
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM614415 5 0.3045 0.801 0.060 0.000 0.000 0.000 0.840 NA
#> GSM614416 5 0.3045 0.801 0.060 0.000 0.000 0.000 0.840 NA
#> GSM614417 5 0.3045 0.801 0.060 0.000 0.000 0.000 0.840 NA
#> GSM614418 5 0.3045 0.801 0.060 0.000 0.000 0.000 0.840 NA
#> GSM614419 5 0.2860 0.804 0.048 0.000 0.000 0.000 0.852 NA
#> GSM614420 5 0.2860 0.804 0.048 0.000 0.000 0.000 0.852 NA
#> GSM614421 3 0.1477 0.865 0.004 0.000 0.940 0.048 0.008 NA
#> GSM614422 3 0.1477 0.865 0.004 0.000 0.940 0.048 0.008 NA
#> GSM614423 3 0.1592 0.860 0.020 0.000 0.940 0.032 0.008 NA
#> GSM614424 3 0.1477 0.865 0.004 0.000 0.940 0.048 0.008 NA
#> GSM614425 3 0.1477 0.865 0.004 0.000 0.940 0.048 0.008 NA
#> GSM614426 3 0.1477 0.865 0.004 0.000 0.940 0.048 0.008 NA
#> GSM614427 3 0.1477 0.865 0.004 0.000 0.940 0.048 0.008 NA
#> GSM614428 3 0.1477 0.865 0.004 0.000 0.940 0.048 0.008 NA
#> GSM614429 2 0.0665 0.838 0.000 0.980 0.004 0.008 0.000 NA
#> GSM614430 2 0.0665 0.838 0.000 0.980 0.004 0.008 0.000 NA
#> GSM614431 2 0.0551 0.838 0.000 0.984 0.004 0.004 0.000 NA
#> GSM614432 2 0.0551 0.838 0.000 0.984 0.004 0.004 0.000 NA
#> GSM614433 2 0.0551 0.838 0.000 0.984 0.004 0.004 0.000 NA
#> GSM614434 2 0.0665 0.838 0.000 0.980 0.004 0.008 0.000 NA
#> GSM614435 2 0.0767 0.838 0.000 0.976 0.004 0.012 0.000 NA
#> GSM614436 2 0.2400 0.782 0.000 0.872 0.004 0.116 0.000 NA
#> GSM614437 4 0.0603 0.935 0.000 0.016 0.004 0.980 0.000 NA
#> GSM614438 4 0.0622 0.946 0.000 0.008 0.012 0.980 0.000 NA
#> GSM614439 4 0.0622 0.946 0.000 0.008 0.012 0.980 0.000 NA
#> GSM614440 4 0.0622 0.946 0.000 0.008 0.012 0.980 0.000 NA
#> GSM614441 4 0.0622 0.946 0.000 0.008 0.012 0.980 0.000 NA
#> GSM614442 4 0.0622 0.946 0.000 0.008 0.012 0.980 0.000 NA
#> GSM614443 4 0.0622 0.943 0.000 0.012 0.008 0.980 0.000 NA
#> GSM614444 4 0.0622 0.946 0.000 0.008 0.012 0.980 0.000 NA
#> GSM614391 5 0.0146 0.800 0.004 0.000 0.000 0.000 0.996 NA
#> GSM614392 5 0.0146 0.800 0.004 0.000 0.000 0.000 0.996 NA
#> GSM614393 5 0.0146 0.800 0.004 0.000 0.000 0.000 0.996 NA
#> GSM614394 5 0.0000 0.800 0.000 0.000 0.000 0.000 1.000 NA
#> GSM614395 5 0.2434 0.725 0.000 0.000 0.036 0.064 0.892 NA
#> GSM614396 5 0.0000 0.800 0.000 0.000 0.000 0.000 1.000 NA
#> GSM614397 5 0.0405 0.796 0.000 0.000 0.008 0.004 0.988 NA
#> GSM614398 5 0.0260 0.797 0.000 0.000 0.008 0.000 0.992 NA
#> GSM614399 1 0.4906 0.705 0.544 0.004 0.044 0.004 0.000 NA
#> GSM614400 1 0.4906 0.705 0.544 0.004 0.044 0.004 0.000 NA
#> GSM614401 1 0.4906 0.705 0.544 0.004 0.044 0.004 0.000 NA
#> GSM614402 1 0.4906 0.705 0.544 0.004 0.044 0.004 0.000 NA
#> GSM614403 1 0.5219 0.680 0.512 0.004 0.068 0.004 0.000 NA
#> GSM614404 1 0.4906 0.705 0.544 0.004 0.044 0.004 0.000 NA
#> GSM614405 1 0.5082 0.702 0.544 0.004 0.044 0.012 0.000 NA
#> GSM614406 4 0.5988 0.457 0.040 0.000 0.104 0.556 0.004 NA
#> GSM614407 5 0.5631 0.699 0.128 0.000 0.000 0.008 0.520 NA
#> GSM614408 5 0.5692 0.695 0.136 0.000 0.000 0.008 0.512 NA
#> GSM614409 5 0.5631 0.699 0.128 0.000 0.000 0.008 0.520 NA
#> GSM614410 5 0.5631 0.699 0.128 0.000 0.000 0.008 0.520 NA
#> GSM614411 5 0.5631 0.699 0.128 0.000 0.000 0.008 0.520 NA
#> GSM614412 5 0.5389 0.705 0.100 0.000 0.000 0.008 0.548 NA
#> GSM614413 5 0.5546 0.711 0.072 0.000 0.020 0.008 0.568 NA
#> GSM614414 5 0.5295 0.714 0.080 0.000 0.004 0.008 0.572 NA
#> GSM614445 3 0.4030 0.747 0.024 0.008 0.728 0.004 0.000 NA
#> GSM614446 3 0.3417 0.796 0.016 0.004 0.788 0.004 0.000 NA
#> GSM614447 3 0.3872 0.768 0.016 0.008 0.748 0.008 0.000 NA
#> GSM614448 3 0.2742 0.832 0.008 0.004 0.856 0.008 0.000 NA
#> GSM614449 3 0.3157 0.827 0.016 0.004 0.832 0.012 0.000 NA
#> GSM614450 3 0.3178 0.813 0.016 0.004 0.816 0.004 0.000 NA
#> GSM614451 3 0.4265 0.624 0.000 0.000 0.660 0.300 0.000 NA
#> GSM614452 3 0.3938 0.725 0.000 0.000 0.728 0.228 0.000 NA
#> GSM614453 2 0.3481 0.791 0.000 0.804 0.000 0.124 0.000 NA
#> GSM614454 2 0.3522 0.789 0.000 0.800 0.000 0.128 0.000 NA
#> GSM614455 2 0.3563 0.787 0.000 0.796 0.000 0.132 0.000 NA
#> GSM614456 2 0.3563 0.787 0.000 0.796 0.000 0.132 0.000 NA
#> GSM614457 2 0.3563 0.787 0.000 0.796 0.000 0.132 0.000 NA
#> GSM614458 2 0.3522 0.789 0.000 0.800 0.000 0.128 0.000 NA
#> GSM614459 2 0.3602 0.784 0.000 0.792 0.000 0.136 0.000 NA
#> GSM614460 2 0.3522 0.789 0.000 0.800 0.000 0.128 0.000 NA
#> GSM614461 2 0.3788 0.786 0.024 0.772 0.012 0.004 0.000 NA
#> GSM614462 2 0.3788 0.786 0.024 0.772 0.012 0.004 0.000 NA
#> GSM614463 2 0.3788 0.786 0.024 0.772 0.012 0.004 0.000 NA
#> GSM614464 2 0.3788 0.786 0.024 0.772 0.012 0.004 0.000 NA
#> GSM614465 2 0.3788 0.786 0.024 0.772 0.012 0.004 0.000 NA
#> GSM614466 2 0.3788 0.786 0.024 0.772 0.012 0.004 0.000 NA
#> GSM614467 2 0.3788 0.786 0.024 0.772 0.012 0.004 0.000 NA
#> GSM614468 2 0.3788 0.786 0.024 0.772 0.012 0.004 0.000 NA
#> GSM614469 1 0.0937 0.753 0.960 0.000 0.000 0.000 0.040 NA
#> GSM614470 1 0.0937 0.753 0.960 0.000 0.000 0.000 0.040 NA
#> GSM614471 1 0.0937 0.753 0.960 0.000 0.000 0.000 0.040 NA
#> GSM614472 1 0.0937 0.753 0.960 0.000 0.000 0.000 0.040 NA
#> GSM614473 1 0.0937 0.753 0.960 0.000 0.000 0.000 0.040 NA
#> GSM614474 1 0.0937 0.753 0.960 0.000 0.000 0.000 0.040 NA
#> GSM614475 1 0.0937 0.753 0.960 0.000 0.000 0.000 0.040 NA
#> GSM614476 1 0.1149 0.746 0.960 0.000 0.008 0.008 0.024 NA
Heatmaps for the consensus matrix. It visualizes the probability of two samples to be in a same group.
consensus_heatmap(res, k = 2)
consensus_heatmap(res, k = 3)
consensus_heatmap(res, k = 4)
consensus_heatmap(res, k = 5)
consensus_heatmap(res, k = 6)
Heatmaps for the membership of samples in all partitions to see how consistent they are:
membership_heatmap(res, k = 2)
membership_heatmap(res, k = 3)
membership_heatmap(res, k = 4)
membership_heatmap(res, k = 5)
membership_heatmap(res, k = 6)
As soon as we have had the classes for columns, we can look for signatures which are significantly different between classes which can be candidate marks for certain classes. Following are the heatmaps for signatures.
Signature heatmaps where rows are scaled:
get_signatures(res, k = 2)
get_signatures(res, k = 3)
get_signatures(res, k = 4)
get_signatures(res, k = 5)
get_signatures(res, k = 6)
Signature heatmaps where rows are not scaled:
get_signatures(res, k = 2, scale_rows = FALSE)
get_signatures(res, k = 3, scale_rows = FALSE)
get_signatures(res, k = 4, scale_rows = FALSE)
get_signatures(res, k = 5, scale_rows = FALSE)
get_signatures(res, k = 6, scale_rows = FALSE)
Compare the overlap of signatures from different k:
compare_signatures(res)
get_signature() returns a data frame invisibly. TO get the list of signatures, the function
call should be assigned to a variable explicitly. In following code, if plot argument is set
to FALSE, no heatmap is plotted while only the differential analysis is performed.
# code only for demonstration
tb = get_signature(res, k = ..., plot = FALSE)
An example of the output of tb is:
#> which_row fdr mean_1 mean_2 scaled_mean_1 scaled_mean_2 km
#> 1 38 0.042760348 8.373488 9.131774 -0.5533452 0.5164555 1
#> 2 40 0.018707592 7.106213 8.469186 -0.6173731 0.5762149 1
#> 3 55 0.019134737 10.221463 11.207825 -0.6159697 0.5749050 1
#> 4 59 0.006059896 5.921854 7.869574 -0.6899429 0.6439467 1
#> 5 60 0.018055526 8.928898 10.211722 -0.6204761 0.5791110 1
#> 6 98 0.009384629 15.714769 14.887706 0.6635654 -0.6193277 2
...
The columns in tb are:
which_row: row indices corresponding to the input matrix.fdr: FDR for the differential test. mean_x: The mean value in group x.scaled_mean_x: The mean value in group x after rows are scaled.km: Row groups if k-means clustering is applied to rows.UMAP plot which shows how samples are separated.
dimension_reduction(res, k = 2, method = "UMAP")
dimension_reduction(res, k = 3, method = "UMAP")
dimension_reduction(res, k = 4, method = "UMAP")
dimension_reduction(res, k = 5, method = "UMAP")
dimension_reduction(res, k = 6, method = "UMAP")
Following heatmap shows how subgroups are split when increasing k:
collect_classes(res)
Test correlation between subgroups and known annotations. If the known annotation is numeric, one-way ANOVA test is applied, and if the known annotation is discrete, chi-squared contingency table test is applied.
test_to_known_factors(res)
#> n individual(p) protocol(p) time(p) other(p) k
#> MAD:skmeans 86 8.23e-13 0.492 0.975 0.7514 2
#> MAD:skmeans 74 1.38e-20 0.561 1.000 0.2053 3
#> MAD:skmeans 82 3.15e-34 0.766 1.000 0.0870 4
#> MAD:skmeans 85 1.15e-46 0.904 1.000 0.0206 5
#> MAD:skmeans 85 3.27e-49 0.977 1.000 0.0225 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 21168 rows and 86 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 0.904 0.946 0.975 0.4881 0.512 0.512
#> 3 3 0.780 0.890 0.937 0.2669 0.880 0.765
#> 4 4 0.713 0.619 0.798 0.1312 0.878 0.702
#> 5 5 0.845 0.824 0.920 0.1004 0.862 0.594
#> 6 6 0.828 0.789 0.870 0.0456 0.962 0.839
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
#> GSM614415 1 0.0000 0.967 1.000 0.000
#> GSM614416 1 0.0000 0.967 1.000 0.000
#> GSM614417 1 0.0000 0.967 1.000 0.000
#> GSM614418 1 0.0000 0.967 1.000 0.000
#> GSM614419 1 0.0000 0.967 1.000 0.000
#> GSM614420 1 0.0000 0.967 1.000 0.000
#> GSM614421 2 0.0376 0.976 0.004 0.996
#> GSM614422 1 0.0376 0.964 0.996 0.004
#> GSM614423 1 0.8443 0.652 0.728 0.272
#> GSM614424 2 0.7376 0.747 0.208 0.792
#> GSM614425 2 0.0376 0.976 0.004 0.996
#> GSM614426 2 0.1633 0.961 0.024 0.976
#> GSM614427 2 0.0000 0.978 0.000 1.000
#> GSM614428 2 0.0000 0.978 0.000 1.000
#> GSM614429 2 0.0000 0.978 0.000 1.000
#> GSM614430 2 0.0000 0.978 0.000 1.000
#> GSM614431 2 0.0000 0.978 0.000 1.000
#> GSM614432 2 0.0000 0.978 0.000 1.000
#> GSM614433 2 0.0000 0.978 0.000 1.000
#> GSM614434 2 0.0000 0.978 0.000 1.000
#> GSM614435 2 0.0000 0.978 0.000 1.000
#> GSM614436 2 0.0000 0.978 0.000 1.000
#> GSM614437 2 0.0000 0.978 0.000 1.000
#> GSM614438 2 0.0000 0.978 0.000 1.000
#> GSM614439 2 0.0000 0.978 0.000 1.000
#> GSM614440 2 0.0000 0.978 0.000 1.000
#> GSM614441 2 0.0000 0.978 0.000 1.000
#> GSM614442 2 0.0000 0.978 0.000 1.000
#> GSM614443 2 0.0000 0.978 0.000 1.000
#> GSM614444 2 0.0000 0.978 0.000 1.000
#> GSM614391 1 0.0000 0.967 1.000 0.000
#> GSM614392 1 0.0000 0.967 1.000 0.000
#> GSM614393 1 0.0000 0.967 1.000 0.000
#> GSM614394 1 0.0000 0.967 1.000 0.000
#> GSM614395 2 0.5059 0.878 0.112 0.888
#> GSM614396 1 0.0000 0.967 1.000 0.000
#> GSM614397 2 0.2043 0.956 0.032 0.968
#> GSM614398 1 0.0000 0.967 1.000 0.000
#> GSM614399 1 0.8386 0.659 0.732 0.268
#> GSM614400 1 0.0000 0.967 1.000 0.000
#> GSM614401 1 0.0000 0.967 1.000 0.000
#> GSM614402 1 0.0938 0.958 0.988 0.012
#> GSM614403 1 0.8955 0.576 0.688 0.312
#> GSM614404 1 0.0000 0.967 1.000 0.000
#> GSM614405 1 0.7745 0.708 0.772 0.228
#> GSM614406 2 0.0000 0.978 0.000 1.000
#> GSM614407 1 0.0000 0.967 1.000 0.000
#> GSM614408 1 0.0000 0.967 1.000 0.000
#> GSM614409 1 0.0000 0.967 1.000 0.000
#> GSM614410 1 0.0000 0.967 1.000 0.000
#> GSM614411 1 0.0000 0.967 1.000 0.000
#> GSM614412 1 0.0376 0.964 0.996 0.004
#> GSM614413 2 0.5519 0.859 0.128 0.872
#> GSM614414 2 0.8016 0.687 0.244 0.756
#> GSM614445 2 0.0000 0.978 0.000 1.000
#> GSM614446 2 0.6887 0.785 0.184 0.816
#> GSM614447 2 0.2236 0.951 0.036 0.964
#> GSM614448 2 0.0376 0.976 0.004 0.996
#> GSM614449 2 0.0000 0.978 0.000 1.000
#> GSM614450 2 0.3114 0.934 0.056 0.944
#> GSM614451 2 0.0000 0.978 0.000 1.000
#> GSM614452 2 0.0000 0.978 0.000 1.000
#> GSM614453 2 0.0000 0.978 0.000 1.000
#> GSM614454 2 0.0000 0.978 0.000 1.000
#> GSM614455 2 0.0000 0.978 0.000 1.000
#> GSM614456 2 0.0000 0.978 0.000 1.000
#> GSM614457 2 0.0000 0.978 0.000 1.000
#> GSM614458 2 0.0000 0.978 0.000 1.000
#> GSM614459 2 0.0000 0.978 0.000 1.000
#> GSM614460 2 0.0000 0.978 0.000 1.000
#> GSM614461 2 0.0000 0.978 0.000 1.000
#> GSM614462 2 0.0000 0.978 0.000 1.000
#> GSM614463 2 0.1633 0.961 0.024 0.976
#> GSM614464 2 0.0000 0.978 0.000 1.000
#> GSM614465 2 0.0000 0.978 0.000 1.000
#> GSM614466 2 0.0000 0.978 0.000 1.000
#> GSM614467 2 0.0000 0.978 0.000 1.000
#> GSM614468 2 0.0000 0.978 0.000 1.000
#> GSM614469 1 0.0000 0.967 1.000 0.000
#> GSM614470 1 0.0000 0.967 1.000 0.000
#> GSM614471 1 0.0000 0.967 1.000 0.000
#> GSM614472 1 0.0000 0.967 1.000 0.000
#> GSM614473 1 0.0000 0.967 1.000 0.000
#> GSM614474 1 0.0000 0.967 1.000 0.000
#> GSM614475 1 0.0000 0.967 1.000 0.000
#> GSM614476 1 0.0000 0.967 1.000 0.000
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM614415 1 0.0000 0.940 1.000 0.000 0.000
#> GSM614416 1 0.0000 0.940 1.000 0.000 0.000
#> GSM614417 1 0.0000 0.940 1.000 0.000 0.000
#> GSM614418 1 0.0000 0.940 1.000 0.000 0.000
#> GSM614419 1 0.0237 0.938 0.996 0.000 0.004
#> GSM614420 1 0.0000 0.940 1.000 0.000 0.000
#> GSM614421 2 0.3816 0.867 0.000 0.852 0.148
#> GSM614422 1 0.5393 0.773 0.808 0.044 0.148
#> GSM614423 1 0.7766 0.596 0.676 0.176 0.148
#> GSM614424 2 0.7862 0.650 0.184 0.668 0.148
#> GSM614425 2 0.3816 0.867 0.000 0.852 0.148
#> GSM614426 2 0.4679 0.857 0.020 0.832 0.148
#> GSM614427 2 0.3816 0.867 0.000 0.852 0.148
#> GSM614428 2 0.3816 0.867 0.000 0.852 0.148
#> GSM614429 2 0.0000 0.921 0.000 1.000 0.000
#> GSM614430 2 0.0000 0.921 0.000 1.000 0.000
#> GSM614431 2 0.0000 0.921 0.000 1.000 0.000
#> GSM614432 2 0.0000 0.921 0.000 1.000 0.000
#> GSM614433 2 0.0000 0.921 0.000 1.000 0.000
#> GSM614434 2 0.0000 0.921 0.000 1.000 0.000
#> GSM614435 2 0.0000 0.921 0.000 1.000 0.000
#> GSM614436 2 0.0000 0.921 0.000 1.000 0.000
#> GSM614437 3 0.3816 0.890 0.000 0.148 0.852
#> GSM614438 3 0.2448 0.933 0.000 0.076 0.924
#> GSM614439 3 0.2448 0.933 0.000 0.076 0.924
#> GSM614440 3 0.0892 0.925 0.000 0.020 0.980
#> GSM614441 3 0.1411 0.930 0.000 0.036 0.964
#> GSM614442 3 0.3192 0.920 0.000 0.112 0.888
#> GSM614443 3 0.3816 0.890 0.000 0.148 0.852
#> GSM614444 3 0.3192 0.920 0.000 0.112 0.888
#> GSM614391 1 0.0747 0.931 0.984 0.000 0.016
#> GSM614392 1 0.0000 0.940 1.000 0.000 0.000
#> GSM614393 1 0.0000 0.940 1.000 0.000 0.000
#> GSM614394 1 0.0592 0.933 0.988 0.000 0.012
#> GSM614395 3 0.0237 0.913 0.004 0.000 0.996
#> GSM614396 1 0.3551 0.831 0.868 0.000 0.132
#> GSM614397 2 0.5315 0.807 0.012 0.772 0.216
#> GSM614398 1 0.3816 0.814 0.852 0.000 0.148
#> GSM614399 1 0.5216 0.650 0.740 0.260 0.000
#> GSM614400 1 0.0000 0.940 1.000 0.000 0.000
#> GSM614401 1 0.0000 0.940 1.000 0.000 0.000
#> GSM614402 1 0.0747 0.929 0.984 0.016 0.000
#> GSM614403 1 0.8392 0.495 0.616 0.236 0.148
#> GSM614404 1 0.0000 0.940 1.000 0.000 0.000
#> GSM614405 1 0.6940 0.612 0.708 0.224 0.068
#> GSM614406 2 0.2625 0.896 0.000 0.916 0.084
#> GSM614407 1 0.0000 0.940 1.000 0.000 0.000
#> GSM614408 1 0.0000 0.940 1.000 0.000 0.000
#> GSM614409 1 0.0000 0.940 1.000 0.000 0.000
#> GSM614410 1 0.0000 0.940 1.000 0.000 0.000
#> GSM614411 1 0.0000 0.940 1.000 0.000 0.000
#> GSM614412 1 0.1170 0.926 0.976 0.008 0.016
#> GSM614413 2 0.5497 0.837 0.048 0.804 0.148
#> GSM614414 2 0.7954 0.639 0.192 0.660 0.148
#> GSM614445 2 0.3619 0.874 0.000 0.864 0.136
#> GSM614446 2 0.5393 0.839 0.044 0.808 0.148
#> GSM614447 2 0.4982 0.855 0.036 0.828 0.136
#> GSM614448 2 0.3816 0.867 0.000 0.852 0.148
#> GSM614449 2 0.3816 0.867 0.000 0.852 0.148
#> GSM614450 2 0.4164 0.867 0.008 0.848 0.144
#> GSM614451 3 0.0000 0.914 0.000 0.000 1.000
#> GSM614452 3 0.0000 0.914 0.000 0.000 1.000
#> GSM614453 2 0.0000 0.921 0.000 1.000 0.000
#> GSM614454 2 0.0000 0.921 0.000 1.000 0.000
#> GSM614455 2 0.0000 0.921 0.000 1.000 0.000
#> GSM614456 2 0.0000 0.921 0.000 1.000 0.000
#> GSM614457 2 0.0000 0.921 0.000 1.000 0.000
#> GSM614458 2 0.0000 0.921 0.000 1.000 0.000
#> GSM614459 2 0.0000 0.921 0.000 1.000 0.000
#> GSM614460 2 0.0000 0.921 0.000 1.000 0.000
#> GSM614461 2 0.0000 0.921 0.000 1.000 0.000
#> GSM614462 2 0.0000 0.921 0.000 1.000 0.000
#> GSM614463 2 0.0747 0.910 0.016 0.984 0.000
#> GSM614464 2 0.0000 0.921 0.000 1.000 0.000
#> GSM614465 2 0.0000 0.921 0.000 1.000 0.000
#> GSM614466 2 0.0000 0.921 0.000 1.000 0.000
#> GSM614467 2 0.0000 0.921 0.000 1.000 0.000
#> GSM614468 2 0.0000 0.921 0.000 1.000 0.000
#> GSM614469 1 0.0000 0.940 1.000 0.000 0.000
#> GSM614470 1 0.0000 0.940 1.000 0.000 0.000
#> GSM614471 1 0.0000 0.940 1.000 0.000 0.000
#> GSM614472 1 0.0000 0.940 1.000 0.000 0.000
#> GSM614473 1 0.0000 0.940 1.000 0.000 0.000
#> GSM614474 1 0.0000 0.940 1.000 0.000 0.000
#> GSM614475 1 0.0000 0.940 1.000 0.000 0.000
#> GSM614476 1 0.0000 0.940 1.000 0.000 0.000
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM614415 1 0.0000 0.811 1.000 0.000 0.000 0.000
#> GSM614416 1 0.0000 0.811 1.000 0.000 0.000 0.000
#> GSM614417 1 0.0000 0.811 1.000 0.000 0.000 0.000
#> GSM614418 1 0.0000 0.811 1.000 0.000 0.000 0.000
#> GSM614419 1 0.0921 0.795 0.972 0.000 0.028 0.000
#> GSM614420 1 0.0592 0.805 0.984 0.000 0.016 0.000
#> GSM614421 2 0.4989 0.557 0.000 0.528 0.472 0.000
#> GSM614422 3 0.0921 0.146 0.000 0.028 0.972 0.000
#> GSM614423 3 0.1022 0.147 0.000 0.032 0.968 0.000
#> GSM614424 3 0.4925 -0.454 0.000 0.428 0.572 0.000
#> GSM614425 2 0.4989 0.557 0.000 0.528 0.472 0.000
#> GSM614426 2 0.4998 0.539 0.000 0.512 0.488 0.000
#> GSM614427 2 0.4989 0.557 0.000 0.528 0.472 0.000
#> GSM614428 2 0.4989 0.557 0.000 0.528 0.472 0.000
#> GSM614429 2 0.0000 0.862 0.000 1.000 0.000 0.000
#> GSM614430 2 0.0000 0.862 0.000 1.000 0.000 0.000
#> GSM614431 2 0.0000 0.862 0.000 1.000 0.000 0.000
#> GSM614432 2 0.0000 0.862 0.000 1.000 0.000 0.000
#> GSM614433 2 0.0336 0.857 0.008 0.992 0.000 0.000
#> GSM614434 2 0.0000 0.862 0.000 1.000 0.000 0.000
#> GSM614435 2 0.0000 0.862 0.000 1.000 0.000 0.000
#> GSM614436 2 0.0000 0.862 0.000 1.000 0.000 0.000
#> GSM614437 4 0.0000 0.886 0.000 0.000 0.000 1.000
#> GSM614438 4 0.0000 0.886 0.000 0.000 0.000 1.000
#> GSM614439 4 0.0000 0.886 0.000 0.000 0.000 1.000
#> GSM614440 4 0.0000 0.886 0.000 0.000 0.000 1.000
#> GSM614441 4 0.0000 0.886 0.000 0.000 0.000 1.000
#> GSM614442 4 0.0000 0.886 0.000 0.000 0.000 1.000
#> GSM614443 4 0.0000 0.886 0.000 0.000 0.000 1.000
#> GSM614444 4 0.0000 0.886 0.000 0.000 0.000 1.000
#> GSM614391 1 0.2081 0.777 0.916 0.000 0.084 0.000
#> GSM614392 1 0.4585 0.080 0.668 0.000 0.332 0.000
#> GSM614393 1 0.4331 0.240 0.712 0.000 0.288 0.000
#> GSM614394 1 0.2081 0.781 0.916 0.000 0.084 0.000
#> GSM614395 4 0.5040 0.650 0.008 0.000 0.364 0.628
#> GSM614396 1 0.4543 0.461 0.676 0.000 0.324 0.000
#> GSM614397 3 0.7669 -0.328 0.228 0.328 0.444 0.000
#> GSM614398 3 0.4996 -0.267 0.484 0.000 0.516 0.000
#> GSM614399 3 0.7464 0.245 0.296 0.208 0.496 0.000
#> GSM614400 3 0.4989 0.489 0.472 0.000 0.528 0.000
#> GSM614401 3 0.4989 0.489 0.472 0.000 0.528 0.000
#> GSM614402 3 0.5396 0.474 0.464 0.012 0.524 0.000
#> GSM614403 3 0.6078 0.150 0.152 0.164 0.684 0.000
#> GSM614404 3 0.4989 0.489 0.472 0.000 0.528 0.000
#> GSM614405 3 0.7564 0.155 0.328 0.208 0.464 0.000
#> GSM614406 2 0.4284 0.745 0.000 0.764 0.224 0.012
#> GSM614407 3 0.4992 0.482 0.476 0.000 0.524 0.000
#> GSM614408 3 0.4992 0.483 0.476 0.000 0.524 0.000
#> GSM614409 3 0.4989 0.482 0.472 0.000 0.528 0.000
#> GSM614410 3 0.4989 0.489 0.472 0.000 0.528 0.000
#> GSM614411 3 0.4989 0.489 0.472 0.000 0.528 0.000
#> GSM614412 3 0.5378 0.332 0.448 0.012 0.540 0.000
#> GSM614413 2 0.5220 0.598 0.008 0.568 0.424 0.000
#> GSM614414 3 0.6042 -0.389 0.048 0.392 0.560 0.000
#> GSM614445 2 0.2868 0.800 0.000 0.864 0.136 0.000
#> GSM614446 2 0.4406 0.705 0.000 0.700 0.300 0.000
#> GSM614447 2 0.4123 0.754 0.008 0.772 0.220 0.000
#> GSM614448 2 0.4981 0.564 0.000 0.536 0.464 0.000
#> GSM614449 2 0.4804 0.636 0.000 0.616 0.384 0.000
#> GSM614450 2 0.4283 0.731 0.004 0.740 0.256 0.000
#> GSM614451 4 0.4454 0.701 0.000 0.000 0.308 0.692
#> GSM614452 4 0.4830 0.631 0.000 0.000 0.392 0.608
#> GSM614453 2 0.0000 0.862 0.000 1.000 0.000 0.000
#> GSM614454 2 0.0000 0.862 0.000 1.000 0.000 0.000
#> GSM614455 2 0.0000 0.862 0.000 1.000 0.000 0.000
#> GSM614456 2 0.0000 0.862 0.000 1.000 0.000 0.000
#> GSM614457 2 0.0000 0.862 0.000 1.000 0.000 0.000
#> GSM614458 2 0.0000 0.862 0.000 1.000 0.000 0.000
#> GSM614459 2 0.0000 0.862 0.000 1.000 0.000 0.000
#> GSM614460 2 0.0000 0.862 0.000 1.000 0.000 0.000
#> GSM614461 2 0.0000 0.862 0.000 1.000 0.000 0.000
#> GSM614462 2 0.0000 0.862 0.000 1.000 0.000 0.000
#> GSM614463 2 0.0469 0.852 0.000 0.988 0.012 0.000
#> GSM614464 2 0.0000 0.862 0.000 1.000 0.000 0.000
#> GSM614465 2 0.0000 0.862 0.000 1.000 0.000 0.000
#> GSM614466 2 0.0000 0.862 0.000 1.000 0.000 0.000
#> GSM614467 2 0.0000 0.862 0.000 1.000 0.000 0.000
#> GSM614468 2 0.0000 0.862 0.000 1.000 0.000 0.000
#> GSM614469 3 0.4989 0.489 0.472 0.000 0.528 0.000
#> GSM614470 3 0.4989 0.489 0.472 0.000 0.528 0.000
#> GSM614471 3 0.4989 0.489 0.472 0.000 0.528 0.000
#> GSM614472 3 0.4989 0.489 0.472 0.000 0.528 0.000
#> GSM614473 3 0.4989 0.489 0.472 0.000 0.528 0.000
#> GSM614474 3 0.4989 0.489 0.472 0.000 0.528 0.000
#> GSM614475 3 0.4989 0.489 0.472 0.000 0.528 0.000
#> GSM614476 3 0.4989 0.489 0.472 0.000 0.528 0.000
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM614415 5 0.0290 0.8791 0.008 0.000 0.000 0.000 0.992
#> GSM614416 5 0.0404 0.8823 0.012 0.000 0.000 0.000 0.988
#> GSM614417 5 0.0404 0.8823 0.012 0.000 0.000 0.000 0.988
#> GSM614418 5 0.0404 0.8823 0.012 0.000 0.000 0.000 0.988
#> GSM614419 5 0.0404 0.8823 0.012 0.000 0.000 0.000 0.988
#> GSM614420 5 0.0404 0.8823 0.012 0.000 0.000 0.000 0.988
#> GSM614421 3 0.0880 0.8649 0.000 0.032 0.968 0.000 0.000
#> GSM614422 3 0.1018 0.8574 0.016 0.016 0.968 0.000 0.000
#> GSM614423 3 0.0880 0.8452 0.032 0.000 0.968 0.000 0.000
#> GSM614424 3 0.0880 0.8649 0.000 0.032 0.968 0.000 0.000
#> GSM614425 3 0.0880 0.8649 0.000 0.032 0.968 0.000 0.000
#> GSM614426 3 0.0955 0.8636 0.004 0.028 0.968 0.000 0.000
#> GSM614427 3 0.0880 0.8649 0.000 0.032 0.968 0.000 0.000
#> GSM614428 3 0.0880 0.8649 0.000 0.032 0.968 0.000 0.000
#> GSM614429 2 0.0000 0.9327 0.000 1.000 0.000 0.000 0.000
#> GSM614430 2 0.0000 0.9327 0.000 1.000 0.000 0.000 0.000
#> GSM614431 2 0.0000 0.9327 0.000 1.000 0.000 0.000 0.000
#> GSM614432 2 0.0000 0.9327 0.000 1.000 0.000 0.000 0.000
#> GSM614433 2 0.0000 0.9327 0.000 1.000 0.000 0.000 0.000
#> GSM614434 2 0.0000 0.9327 0.000 1.000 0.000 0.000 0.000
#> GSM614435 2 0.0000 0.9327 0.000 1.000 0.000 0.000 0.000
#> GSM614436 2 0.0000 0.9327 0.000 1.000 0.000 0.000 0.000
#> GSM614437 4 0.0000 1.0000 0.000 0.000 0.000 1.000 0.000
#> GSM614438 4 0.0000 1.0000 0.000 0.000 0.000 1.000 0.000
#> GSM614439 4 0.0000 1.0000 0.000 0.000 0.000 1.000 0.000
#> GSM614440 4 0.0000 1.0000 0.000 0.000 0.000 1.000 0.000
#> GSM614441 4 0.0000 1.0000 0.000 0.000 0.000 1.000 0.000
#> GSM614442 4 0.0000 1.0000 0.000 0.000 0.000 1.000 0.000
#> GSM614443 4 0.0000 1.0000 0.000 0.000 0.000 1.000 0.000
#> GSM614444 4 0.0000 1.0000 0.000 0.000 0.000 1.000 0.000
#> GSM614391 5 0.2806 0.8054 0.152 0.000 0.004 0.000 0.844
#> GSM614392 1 0.4171 0.3047 0.604 0.000 0.000 0.000 0.396
#> GSM614393 5 0.3999 0.4888 0.344 0.000 0.000 0.000 0.656
#> GSM614394 5 0.2536 0.8252 0.128 0.000 0.004 0.000 0.868
#> GSM614395 3 0.4206 0.6210 0.000 0.000 0.696 0.288 0.016
#> GSM614396 5 0.4605 0.6798 0.076 0.000 0.192 0.000 0.732
#> GSM614397 3 0.3639 0.7881 0.000 0.076 0.824 0.000 0.100
#> GSM614398 3 0.3480 0.6247 0.000 0.000 0.752 0.000 0.248
#> GSM614399 1 0.2891 0.7120 0.824 0.176 0.000 0.000 0.000
#> GSM614400 1 0.0000 0.8903 1.000 0.000 0.000 0.000 0.000
#> GSM614401 1 0.0000 0.8903 1.000 0.000 0.000 0.000 0.000
#> GSM614402 1 0.0609 0.8795 0.980 0.020 0.000 0.000 0.000
#> GSM614403 1 0.6012 0.1193 0.484 0.116 0.400 0.000 0.000
#> GSM614404 1 0.0000 0.8903 1.000 0.000 0.000 0.000 0.000
#> GSM614405 1 0.5295 0.5471 0.672 0.200 0.128 0.000 0.000
#> GSM614406 2 0.4824 -0.0174 0.000 0.512 0.468 0.020 0.000
#> GSM614407 1 0.1485 0.8734 0.948 0.000 0.032 0.000 0.020
#> GSM614408 1 0.1668 0.8687 0.940 0.000 0.032 0.000 0.028
#> GSM614409 1 0.2370 0.8438 0.904 0.000 0.040 0.000 0.056
#> GSM614410 1 0.0880 0.8795 0.968 0.000 0.032 0.000 0.000
#> GSM614411 1 0.1041 0.8788 0.964 0.000 0.032 0.000 0.004
#> GSM614412 1 0.5571 0.5820 0.668 0.008 0.148 0.000 0.176
#> GSM614413 3 0.2329 0.7803 0.000 0.124 0.876 0.000 0.000
#> GSM614414 3 0.1251 0.8461 0.008 0.036 0.956 0.000 0.000
#> GSM614445 2 0.2516 0.7926 0.000 0.860 0.140 0.000 0.000
#> GSM614446 2 0.4464 0.3059 0.008 0.584 0.408 0.000 0.000
#> GSM614447 2 0.4360 0.5639 0.024 0.692 0.284 0.000 0.000
#> GSM614448 3 0.1544 0.8485 0.000 0.068 0.932 0.000 0.000
#> GSM614449 3 0.3966 0.4858 0.000 0.336 0.664 0.000 0.000
#> GSM614450 2 0.4182 0.4413 0.004 0.644 0.352 0.000 0.000
#> GSM614451 3 0.4088 0.4938 0.000 0.000 0.632 0.368 0.000
#> GSM614452 3 0.3534 0.6724 0.000 0.000 0.744 0.256 0.000
#> GSM614453 2 0.0000 0.9327 0.000 1.000 0.000 0.000 0.000
#> GSM614454 2 0.0000 0.9327 0.000 1.000 0.000 0.000 0.000
#> GSM614455 2 0.0000 0.9327 0.000 1.000 0.000 0.000 0.000
#> GSM614456 2 0.0000 0.9327 0.000 1.000 0.000 0.000 0.000
#> GSM614457 2 0.0000 0.9327 0.000 1.000 0.000 0.000 0.000
#> GSM614458 2 0.0000 0.9327 0.000 1.000 0.000 0.000 0.000
#> GSM614459 2 0.0000 0.9327 0.000 1.000 0.000 0.000 0.000
#> GSM614460 2 0.0000 0.9327 0.000 1.000 0.000 0.000 0.000
#> GSM614461 2 0.0000 0.9327 0.000 1.000 0.000 0.000 0.000
#> GSM614462 2 0.0000 0.9327 0.000 1.000 0.000 0.000 0.000
#> GSM614463 2 0.0404 0.9219 0.012 0.988 0.000 0.000 0.000
#> GSM614464 2 0.0000 0.9327 0.000 1.000 0.000 0.000 0.000
#> GSM614465 2 0.0000 0.9327 0.000 1.000 0.000 0.000 0.000
#> GSM614466 2 0.0000 0.9327 0.000 1.000 0.000 0.000 0.000
#> GSM614467 2 0.0000 0.9327 0.000 1.000 0.000 0.000 0.000
#> GSM614468 2 0.0000 0.9327 0.000 1.000 0.000 0.000 0.000
#> GSM614469 1 0.0000 0.8903 1.000 0.000 0.000 0.000 0.000
#> GSM614470 1 0.0000 0.8903 1.000 0.000 0.000 0.000 0.000
#> GSM614471 1 0.0000 0.8903 1.000 0.000 0.000 0.000 0.000
#> GSM614472 1 0.0000 0.8903 1.000 0.000 0.000 0.000 0.000
#> GSM614473 1 0.0000 0.8903 1.000 0.000 0.000 0.000 0.000
#> GSM614474 1 0.0000 0.8903 1.000 0.000 0.000 0.000 0.000
#> GSM614475 1 0.0000 0.8903 1.000 0.000 0.000 0.000 0.000
#> GSM614476 1 0.0000 0.8903 1.000 0.000 0.000 0.000 0.000
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM614415 5 0.0000 0.876 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM614416 5 0.0000 0.876 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM614417 5 0.0000 0.876 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM614418 5 0.0000 0.876 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM614419 5 0.0000 0.876 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM614420 5 0.0000 0.876 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM614421 3 0.0000 0.872 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM614422 3 0.0000 0.872 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM614423 3 0.0146 0.869 0.004 0.000 0.996 0.000 0.000 0.000
#> GSM614424 3 0.0000 0.872 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM614425 3 0.0000 0.872 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM614426 3 0.0000 0.872 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM614427 3 0.0000 0.872 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM614428 3 0.0000 0.872 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM614429 2 0.0291 0.904 0.000 0.992 0.004 0.000 0.000 0.004
#> GSM614430 2 0.0405 0.904 0.000 0.988 0.004 0.000 0.000 0.008
#> GSM614431 2 0.0146 0.905 0.000 0.996 0.004 0.000 0.000 0.000
#> GSM614432 2 0.0146 0.905 0.000 0.996 0.004 0.000 0.000 0.000
#> GSM614433 2 0.0146 0.905 0.000 0.996 0.004 0.000 0.000 0.000
#> GSM614434 2 0.0291 0.904 0.000 0.992 0.004 0.000 0.000 0.004
#> GSM614435 2 0.0146 0.905 0.000 0.996 0.004 0.000 0.000 0.000
#> GSM614436 2 0.0146 0.905 0.000 0.996 0.004 0.000 0.000 0.000
#> GSM614437 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM614438 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM614439 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM614440 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM614441 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM614442 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM614443 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM614444 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM614391 5 0.4178 0.778 0.092 0.000 0.004 0.000 0.748 0.156
#> GSM614392 1 0.5537 -0.123 0.476 0.000 0.000 0.000 0.388 0.136
#> GSM614393 5 0.4871 0.673 0.212 0.000 0.000 0.000 0.656 0.132
#> GSM614394 5 0.4002 0.789 0.096 0.000 0.004 0.000 0.768 0.132
#> GSM614395 3 0.4493 0.722 0.000 0.000 0.720 0.144 0.004 0.132
#> GSM614396 5 0.5719 0.570 0.032 0.000 0.236 0.000 0.600 0.132
#> GSM614397 3 0.4273 0.740 0.000 0.036 0.768 0.000 0.064 0.132
#> GSM614398 3 0.4845 0.531 0.000 0.000 0.660 0.000 0.208 0.132
#> GSM614399 1 0.4442 0.644 0.712 0.120 0.000 0.000 0.000 0.168
#> GSM614400 1 0.2416 0.750 0.844 0.000 0.000 0.000 0.000 0.156
#> GSM614401 1 0.2416 0.750 0.844 0.000 0.000 0.000 0.000 0.156
#> GSM614402 1 0.2737 0.745 0.832 0.004 0.004 0.000 0.000 0.160
#> GSM614403 1 0.7132 0.269 0.416 0.116 0.296 0.000 0.000 0.172
#> GSM614404 1 0.2454 0.749 0.840 0.000 0.000 0.000 0.000 0.160
#> GSM614405 1 0.6084 0.535 0.612 0.116 0.116 0.000 0.000 0.156
#> GSM614406 2 0.5978 0.345 0.000 0.536 0.284 0.024 0.000 0.156
#> GSM614407 6 0.3782 0.793 0.360 0.000 0.000 0.000 0.004 0.636
#> GSM614408 6 0.3782 0.793 0.360 0.000 0.000 0.000 0.004 0.636
#> GSM614409 6 0.3905 0.794 0.356 0.000 0.004 0.000 0.004 0.636
#> GSM614410 6 0.3659 0.792 0.364 0.000 0.000 0.000 0.000 0.636
#> GSM614411 6 0.3659 0.792 0.364 0.000 0.000 0.000 0.000 0.636
#> GSM614412 6 0.5104 0.754 0.260 0.000 0.044 0.000 0.048 0.648
#> GSM614413 6 0.3874 0.439 0.000 0.008 0.356 0.000 0.000 0.636
#> GSM614414 6 0.3659 0.431 0.000 0.000 0.364 0.000 0.000 0.636
#> GSM614445 2 0.4687 0.668 0.000 0.684 0.136 0.000 0.000 0.180
#> GSM614446 2 0.5708 0.284 0.004 0.488 0.360 0.000 0.000 0.148
#> GSM614447 2 0.5887 0.482 0.024 0.564 0.248 0.000 0.000 0.164
#> GSM614448 3 0.0458 0.862 0.000 0.016 0.984 0.000 0.000 0.000
#> GSM614449 3 0.3161 0.634 0.000 0.216 0.776 0.000 0.000 0.008
#> GSM614450 2 0.4266 0.474 0.004 0.620 0.356 0.000 0.000 0.020
#> GSM614451 3 0.3619 0.582 0.000 0.004 0.680 0.316 0.000 0.000
#> GSM614452 3 0.2597 0.770 0.000 0.000 0.824 0.176 0.000 0.000
#> GSM614453 2 0.1349 0.903 0.000 0.940 0.004 0.000 0.000 0.056
#> GSM614454 2 0.0603 0.905 0.000 0.980 0.004 0.000 0.000 0.016
#> GSM614455 2 0.1285 0.903 0.000 0.944 0.004 0.000 0.000 0.052
#> GSM614456 2 0.1010 0.905 0.000 0.960 0.004 0.000 0.000 0.036
#> GSM614457 2 0.0865 0.904 0.000 0.964 0.000 0.000 0.000 0.036
#> GSM614458 2 0.0405 0.904 0.000 0.988 0.004 0.000 0.000 0.008
#> GSM614459 2 0.0363 0.905 0.000 0.988 0.000 0.000 0.000 0.012
#> GSM614460 2 0.0405 0.904 0.000 0.988 0.004 0.000 0.000 0.008
#> GSM614461 2 0.1387 0.898 0.000 0.932 0.000 0.000 0.000 0.068
#> GSM614462 2 0.1387 0.898 0.000 0.932 0.000 0.000 0.000 0.068
#> GSM614463 2 0.1913 0.887 0.012 0.908 0.000 0.000 0.000 0.080
#> GSM614464 2 0.1387 0.898 0.000 0.932 0.000 0.000 0.000 0.068
#> GSM614465 2 0.1387 0.898 0.000 0.932 0.000 0.000 0.000 0.068
#> GSM614466 2 0.1387 0.898 0.000 0.932 0.000 0.000 0.000 0.068
#> GSM614467 2 0.1387 0.898 0.000 0.932 0.000 0.000 0.000 0.068
#> GSM614468 2 0.1387 0.898 0.000 0.932 0.000 0.000 0.000 0.068
#> GSM614469 1 0.0000 0.777 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM614470 1 0.0000 0.777 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM614471 1 0.0000 0.777 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM614472 1 0.0000 0.777 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM614473 1 0.0000 0.777 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM614474 1 0.0000 0.777 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM614475 1 0.0000 0.777 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM614476 1 0.0000 0.777 1.000 0.000 0.000 0.000 0.000 0.000
Heatmaps for the consensus matrix. It visualizes the probability of two samples to be in a same group.
consensus_heatmap(res, k = 2)
consensus_heatmap(res, k = 3)
consensus_heatmap(res, k = 4)
consensus_heatmap(res, k = 5)
consensus_heatmap(res, k = 6)
Heatmaps for the membership of samples in all partitions to see how consistent they are:
membership_heatmap(res, k = 2)
membership_heatmap(res, k = 3)
membership_heatmap(res, k = 4)
membership_heatmap(res, k = 5)
membership_heatmap(res, k = 6)
As soon as we have had the classes for columns, we can look for signatures which are significantly different between classes which can be candidate marks for certain classes. Following are the heatmaps for signatures.
Signature heatmaps where rows are scaled:
get_signatures(res, k = 2)
get_signatures(res, k = 3)
get_signatures(res, k = 4)
get_signatures(res, k = 5)
get_signatures(res, k = 6)
Signature heatmaps where rows are not scaled:
get_signatures(res, k = 2, scale_rows = FALSE)
get_signatures(res, k = 3, scale_rows = FALSE)
get_signatures(res, k = 4, scale_rows = FALSE)
get_signatures(res, k = 5, scale_rows = FALSE)
get_signatures(res, k = 6, scale_rows = FALSE)
Compare the overlap of signatures from different k:
compare_signatures(res)
get_signature() returns a data frame invisibly. TO get the list of signatures, the function
call should be assigned to a variable explicitly. In following code, if plot argument is set
to FALSE, no heatmap is plotted while only the differential analysis is performed.
# code only for demonstration
tb = get_signature(res, k = ..., plot = FALSE)
An example of the output of tb is:
#> which_row fdr mean_1 mean_2 scaled_mean_1 scaled_mean_2 km
#> 1 38 0.042760348 8.373488 9.131774 -0.5533452 0.5164555 1
#> 2 40 0.018707592 7.106213 8.469186 -0.6173731 0.5762149 1
#> 3 55 0.019134737 10.221463 11.207825 -0.6159697 0.5749050 1
#> 4 59 0.006059896 5.921854 7.869574 -0.6899429 0.6439467 1
#> 5 60 0.018055526 8.928898 10.211722 -0.6204761 0.5791110 1
#> 6 98 0.009384629 15.714769 14.887706 0.6635654 -0.6193277 2
...
The columns in tb are:
which_row: row indices corresponding to the input matrix.fdr: FDR for the differential test. mean_x: The mean value in group x.scaled_mean_x: The mean value in group x after rows are scaled.km: Row groups if k-means clustering is applied to rows.UMAP plot which shows how samples are separated.
dimension_reduction(res, k = 2, method = "UMAP")
dimension_reduction(res, k = 3, method = "UMAP")
dimension_reduction(res, k = 4, method = "UMAP")
dimension_reduction(res, k = 5, method = "UMAP")
dimension_reduction(res, k = 6, method = "UMAP")
Following heatmap shows how subgroups are split when increasing k:
collect_classes(res)
Test correlation between subgroups and known annotations. If the known annotation is numeric, one-way ANOVA test is applied, and if the known annotation is discrete, chi-squared contingency table test is applied.
test_to_known_factors(res)
#> n individual(p) protocol(p) time(p) other(p) k
#> MAD:pam 86 7.09e-10 0.0586 0.889 0.8787 2
#> MAD:pam 85 3.34e-17 0.1026 0.991 0.1653 3
#> MAD:pam 56 2.71e-12 0.1733 0.996 0.3924 4
#> MAD:pam 78 8.88e-36 0.0861 1.000 0.0444 5
#> MAD:pam 78 1.05e-48 0.1863 1.000 0.1000 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 21168 rows and 86 columns.
#> Top rows (1000, 2000, 3000, 4000, 5000) are extracted by 'MAD' method.
#> Subgroups are detected by 'mclust' method.
#> Performed in total 1250 partitions by row resampling.
#> Best k for subgroups seems to be 4.
#>
#> Following methods can be applied to this 'ConsensusPartition' object:
#> [1] "cola_report" "collect_classes" "collect_plots"
#> [4] "collect_stats" "colnames" "compare_signatures"
#> [7] "consensus_heatmap" "dimension_reduction" "functional_enrichment"
#> [10] "get_anno_col" "get_anno" "get_classes"
#> [13] "get_consensus" "get_matrix" "get_membership"
#> [16] "get_param" "get_signatures" "get_stats"
#> [19] "is_best_k" "is_stable_k" "membership_heatmap"
#> [22] "ncol" "nrow" "plot_ecdf"
#> [25] "rownames" "select_partition_number" "show"
#> [28] "suggest_best_k" "test_to_known_factors"
collect_plots() function collects all the plots made from res for all k (number of partitions)
into one single page to provide an easy and fast comparison between different k.
collect_plots(res)
The plots are:
k and the heatmap of
predicted classes for each k.k.k.k.All the plots in panels can be made by individual functions and they are plotted later in this section.
select_partition_number() produces several plots showing different
statistics for choosing “optimized” k. There are following statistics:
k;k, the area increased is defined as \(A_k - A_{k-1}\).The detailed explanations of these statistics can be found in the cola vignette.
Generally speaking, lower PAC score, higher mean silhouette score or higher
concordance corresponds to better partition. Rand index and Jaccard index
measure how similar the current partition is compared to partition with k-1.
If they are too similar, we won't accept k is better than k-1.
select_partition_number(res)
The numeric values for all these statistics can be obtained by get_stats().
get_stats(res)
#> k 1-PAC mean_silhouette concordance area_increased Rand Jaccard
#> 2 2 0.698 0.805 0.902 0.4878 0.498 0.498
#> 3 3 0.636 0.830 0.871 0.3408 0.758 0.548
#> 4 4 0.945 0.927 0.953 0.1047 0.947 0.841
#> 5 5 0.833 0.858 0.902 0.0760 0.943 0.798
#> 6 6 0.798 0.575 0.750 0.0453 0.963 0.836
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
#> GSM614415 1 0.0000 0.948 1.000 0.000
#> GSM614416 1 0.0000 0.948 1.000 0.000
#> GSM614417 1 0.0000 0.948 1.000 0.000
#> GSM614418 1 0.0000 0.948 1.000 0.000
#> GSM614419 1 0.0000 0.948 1.000 0.000
#> GSM614420 1 0.0000 0.948 1.000 0.000
#> GSM614421 2 0.9635 0.504 0.388 0.612
#> GSM614422 2 0.9710 0.477 0.400 0.600
#> GSM614423 2 0.9635 0.504 0.388 0.612
#> GSM614424 2 0.9635 0.504 0.388 0.612
#> GSM614425 2 0.9635 0.504 0.388 0.612
#> GSM614426 2 0.9635 0.504 0.388 0.612
#> GSM614427 2 0.9635 0.504 0.388 0.612
#> GSM614428 2 0.9661 0.495 0.392 0.608
#> GSM614429 2 0.0938 0.834 0.012 0.988
#> GSM614430 2 0.0938 0.834 0.012 0.988
#> GSM614431 2 0.0938 0.834 0.012 0.988
#> GSM614432 2 0.0938 0.834 0.012 0.988
#> GSM614433 2 0.0938 0.834 0.012 0.988
#> GSM614434 2 0.0938 0.834 0.012 0.988
#> GSM614435 2 0.0938 0.834 0.012 0.988
#> GSM614436 2 0.0938 0.834 0.012 0.988
#> GSM614437 2 0.3114 0.817 0.056 0.944
#> GSM614438 2 0.3114 0.817 0.056 0.944
#> GSM614439 2 0.3114 0.817 0.056 0.944
#> GSM614440 2 0.3114 0.817 0.056 0.944
#> GSM614441 2 0.3114 0.817 0.056 0.944
#> GSM614442 2 0.3114 0.817 0.056 0.944
#> GSM614443 2 0.3114 0.817 0.056 0.944
#> GSM614444 2 0.3114 0.817 0.056 0.944
#> GSM614391 1 0.0000 0.948 1.000 0.000
#> GSM614392 1 0.0000 0.948 1.000 0.000
#> GSM614393 1 0.0000 0.948 1.000 0.000
#> GSM614394 1 0.0000 0.948 1.000 0.000
#> GSM614395 1 0.0000 0.948 1.000 0.000
#> GSM614396 1 0.0000 0.948 1.000 0.000
#> GSM614397 1 0.0000 0.948 1.000 0.000
#> GSM614398 1 0.0000 0.948 1.000 0.000
#> GSM614399 1 0.3584 0.913 0.932 0.068
#> GSM614400 1 0.3274 0.919 0.940 0.060
#> GSM614401 1 0.3274 0.919 0.940 0.060
#> GSM614402 1 0.3879 0.904 0.924 0.076
#> GSM614403 1 0.8713 0.536 0.708 0.292
#> GSM614404 1 0.3274 0.919 0.940 0.060
#> GSM614405 1 0.3431 0.916 0.936 0.064
#> GSM614406 2 0.9323 0.560 0.348 0.652
#> GSM614407 1 0.0000 0.948 1.000 0.000
#> GSM614408 1 0.0000 0.948 1.000 0.000
#> GSM614409 1 0.0000 0.948 1.000 0.000
#> GSM614410 1 0.0000 0.948 1.000 0.000
#> GSM614411 1 0.0000 0.948 1.000 0.000
#> GSM614412 1 0.0000 0.948 1.000 0.000
#> GSM614413 1 0.0000 0.948 1.000 0.000
#> GSM614414 1 0.0000 0.948 1.000 0.000
#> GSM614445 2 0.9608 0.511 0.384 0.616
#> GSM614446 2 0.9608 0.511 0.384 0.616
#> GSM614447 2 0.9608 0.511 0.384 0.616
#> GSM614448 2 0.9608 0.511 0.384 0.616
#> GSM614449 2 0.9608 0.511 0.384 0.616
#> GSM614450 2 0.9608 0.511 0.384 0.616
#> GSM614451 1 0.9580 0.347 0.620 0.380
#> GSM614452 1 0.9393 0.420 0.644 0.356
#> GSM614453 2 0.0672 0.833 0.008 0.992
#> GSM614454 2 0.0672 0.833 0.008 0.992
#> GSM614455 2 0.0672 0.833 0.008 0.992
#> GSM614456 2 0.0672 0.833 0.008 0.992
#> GSM614457 2 0.0672 0.833 0.008 0.992
#> GSM614458 2 0.0672 0.833 0.008 0.992
#> GSM614459 2 0.0672 0.833 0.008 0.992
#> GSM614460 2 0.0672 0.833 0.008 0.992
#> GSM614461 2 0.0938 0.834 0.012 0.988
#> GSM614462 2 0.2423 0.825 0.040 0.960
#> GSM614463 2 0.2236 0.827 0.036 0.964
#> GSM614464 2 0.4161 0.795 0.084 0.916
#> GSM614465 2 0.1414 0.832 0.020 0.980
#> GSM614466 2 0.1184 0.833 0.016 0.984
#> GSM614467 2 0.2423 0.825 0.040 0.960
#> GSM614468 2 0.2043 0.829 0.032 0.968
#> GSM614469 1 0.2043 0.939 0.968 0.032
#> GSM614470 1 0.2043 0.939 0.968 0.032
#> GSM614471 1 0.2043 0.939 0.968 0.032
#> GSM614472 1 0.2043 0.939 0.968 0.032
#> GSM614473 1 0.2043 0.939 0.968 0.032
#> GSM614474 1 0.2043 0.939 0.968 0.032
#> GSM614475 1 0.2043 0.939 0.968 0.032
#> GSM614476 1 0.2043 0.939 0.968 0.032
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM614415 1 0.0000 0.914 1.000 0.000 0.000
#> GSM614416 1 0.0237 0.915 0.996 0.004 0.000
#> GSM614417 1 0.0237 0.915 0.996 0.004 0.000
#> GSM614418 1 0.0237 0.915 0.996 0.004 0.000
#> GSM614419 1 0.1129 0.905 0.976 0.004 0.020
#> GSM614420 1 0.0661 0.910 0.988 0.004 0.008
#> GSM614421 3 0.4139 0.866 0.016 0.124 0.860
#> GSM614422 3 0.4485 0.866 0.020 0.136 0.844
#> GSM614423 3 0.4136 0.864 0.020 0.116 0.864
#> GSM614424 3 0.4345 0.865 0.016 0.136 0.848
#> GSM614425 3 0.4277 0.866 0.016 0.132 0.852
#> GSM614426 3 0.4209 0.866 0.016 0.128 0.856
#> GSM614427 3 0.4139 0.866 0.016 0.124 0.860
#> GSM614428 3 0.4139 0.866 0.016 0.124 0.860
#> GSM614429 2 0.2165 0.876 0.000 0.936 0.064
#> GSM614430 2 0.2165 0.876 0.000 0.936 0.064
#> GSM614431 2 0.2165 0.876 0.000 0.936 0.064
#> GSM614432 2 0.2261 0.876 0.000 0.932 0.068
#> GSM614433 2 0.3043 0.869 0.008 0.908 0.084
#> GSM614434 2 0.2165 0.876 0.000 0.936 0.064
#> GSM614435 2 0.2066 0.877 0.000 0.940 0.060
#> GSM614436 2 0.2356 0.877 0.000 0.928 0.072
#> GSM614437 2 0.6585 0.736 0.044 0.712 0.244
#> GSM614438 2 0.7233 0.709 0.064 0.672 0.264
#> GSM614439 2 0.7233 0.709 0.064 0.672 0.264
#> GSM614440 2 0.7233 0.709 0.064 0.672 0.264
#> GSM614441 2 0.7233 0.709 0.064 0.672 0.264
#> GSM614442 2 0.7233 0.709 0.064 0.672 0.264
#> GSM614443 2 0.6685 0.733 0.048 0.708 0.244
#> GSM614444 2 0.7233 0.709 0.064 0.672 0.264
#> GSM614391 1 0.0237 0.915 0.996 0.004 0.000
#> GSM614392 1 0.0237 0.915 0.996 0.004 0.000
#> GSM614393 1 0.0000 0.914 1.000 0.000 0.000
#> GSM614394 1 0.1129 0.905 0.976 0.004 0.020
#> GSM614395 1 0.2400 0.874 0.932 0.004 0.064
#> GSM614396 1 0.1267 0.903 0.972 0.004 0.024
#> GSM614397 1 0.1989 0.887 0.948 0.004 0.048
#> GSM614398 1 0.1267 0.903 0.972 0.004 0.024
#> GSM614399 3 0.7263 0.412 0.372 0.036 0.592
#> GSM614400 3 0.7128 0.480 0.344 0.036 0.620
#> GSM614401 3 0.6908 0.553 0.308 0.036 0.656
#> GSM614402 3 0.6895 0.700 0.212 0.072 0.716
#> GSM614403 3 0.5471 0.847 0.060 0.128 0.812
#> GSM614404 3 0.6935 0.546 0.312 0.036 0.652
#> GSM614405 3 0.6543 0.506 0.344 0.016 0.640
#> GSM614406 3 0.6046 0.835 0.080 0.136 0.784
#> GSM614407 1 0.1315 0.914 0.972 0.008 0.020
#> GSM614408 1 0.1315 0.914 0.972 0.008 0.020
#> GSM614409 1 0.1315 0.914 0.972 0.008 0.020
#> GSM614410 1 0.1315 0.914 0.972 0.008 0.020
#> GSM614411 1 0.1315 0.914 0.972 0.008 0.020
#> GSM614412 1 0.1315 0.914 0.972 0.008 0.020
#> GSM614413 1 0.1751 0.914 0.960 0.012 0.028
#> GSM614414 1 0.1999 0.913 0.952 0.012 0.036
#> GSM614445 3 0.4663 0.855 0.016 0.156 0.828
#> GSM614446 3 0.4663 0.855 0.016 0.156 0.828
#> GSM614447 3 0.4723 0.852 0.016 0.160 0.824
#> GSM614448 3 0.4539 0.860 0.016 0.148 0.836
#> GSM614449 3 0.4602 0.858 0.016 0.152 0.832
#> GSM614450 3 0.4602 0.858 0.016 0.152 0.832
#> GSM614451 3 0.3769 0.859 0.016 0.104 0.880
#> GSM614452 3 0.3769 0.859 0.016 0.104 0.880
#> GSM614453 2 0.0237 0.875 0.004 0.996 0.000
#> GSM614454 2 0.0237 0.875 0.004 0.996 0.000
#> GSM614455 2 0.0237 0.875 0.004 0.996 0.000
#> GSM614456 2 0.0237 0.875 0.004 0.996 0.000
#> GSM614457 2 0.0237 0.875 0.004 0.996 0.000
#> GSM614458 2 0.0237 0.876 0.000 0.996 0.004
#> GSM614459 2 0.0237 0.875 0.004 0.996 0.000
#> GSM614460 2 0.0237 0.875 0.004 0.996 0.000
#> GSM614461 2 0.2448 0.873 0.000 0.924 0.076
#> GSM614462 2 0.3850 0.869 0.028 0.884 0.088
#> GSM614463 2 0.3973 0.868 0.032 0.880 0.088
#> GSM614464 2 0.3722 0.870 0.024 0.888 0.088
#> GSM614465 2 0.3850 0.869 0.028 0.884 0.088
#> GSM614466 2 0.3502 0.872 0.020 0.896 0.084
#> GSM614467 2 0.2711 0.867 0.000 0.912 0.088
#> GSM614468 2 0.2711 0.867 0.000 0.912 0.088
#> GSM614469 1 0.5235 0.803 0.812 0.036 0.152
#> GSM614470 1 0.5235 0.803 0.812 0.036 0.152
#> GSM614471 1 0.5295 0.799 0.808 0.036 0.156
#> GSM614472 1 0.5295 0.799 0.808 0.036 0.156
#> GSM614473 1 0.5235 0.803 0.812 0.036 0.152
#> GSM614474 1 0.5295 0.799 0.808 0.036 0.156
#> GSM614475 1 0.6295 0.676 0.728 0.036 0.236
#> GSM614476 1 0.6762 0.570 0.676 0.036 0.288
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM614415 1 0.0000 0.969 1.000 0.000 0.000 0.000
#> GSM614416 1 0.0469 0.966 0.988 0.000 0.000 0.012
#> GSM614417 1 0.0000 0.969 1.000 0.000 0.000 0.000
#> GSM614418 1 0.0000 0.969 1.000 0.000 0.000 0.000
#> GSM614419 1 0.0000 0.969 1.000 0.000 0.000 0.000
#> GSM614420 1 0.0000 0.969 1.000 0.000 0.000 0.000
#> GSM614421 3 0.0000 0.906 0.000 0.000 1.000 0.000
#> GSM614422 3 0.0188 0.904 0.000 0.000 0.996 0.004
#> GSM614423 3 0.0000 0.906 0.000 0.000 1.000 0.000
#> GSM614424 3 0.0000 0.906 0.000 0.000 1.000 0.000
#> GSM614425 3 0.0000 0.906 0.000 0.000 1.000 0.000
#> GSM614426 3 0.0000 0.906 0.000 0.000 1.000 0.000
#> GSM614427 3 0.0188 0.905 0.000 0.000 0.996 0.004
#> GSM614428 3 0.0000 0.906 0.000 0.000 1.000 0.000
#> GSM614429 2 0.1489 0.962 0.000 0.952 0.004 0.044
#> GSM614430 2 0.1398 0.963 0.000 0.956 0.004 0.040
#> GSM614431 2 0.0188 0.959 0.000 0.996 0.004 0.000
#> GSM614432 2 0.0376 0.957 0.000 0.992 0.004 0.004
#> GSM614433 2 0.1004 0.942 0.000 0.972 0.004 0.024
#> GSM614434 2 0.0524 0.962 0.000 0.988 0.004 0.008
#> GSM614435 2 0.1743 0.959 0.000 0.940 0.004 0.056
#> GSM614436 2 0.1970 0.955 0.000 0.932 0.008 0.060
#> GSM614437 4 0.0817 1.000 0.000 0.024 0.000 0.976
#> GSM614438 4 0.0817 1.000 0.000 0.024 0.000 0.976
#> GSM614439 4 0.0817 1.000 0.000 0.024 0.000 0.976
#> GSM614440 4 0.0817 1.000 0.000 0.024 0.000 0.976
#> GSM614441 4 0.0817 1.000 0.000 0.024 0.000 0.976
#> GSM614442 4 0.0817 1.000 0.000 0.024 0.000 0.976
#> GSM614443 4 0.0817 1.000 0.000 0.024 0.000 0.976
#> GSM614444 4 0.0817 1.000 0.000 0.024 0.000 0.976
#> GSM614391 1 0.0000 0.969 1.000 0.000 0.000 0.000
#> GSM614392 1 0.0000 0.969 1.000 0.000 0.000 0.000
#> GSM614393 1 0.0000 0.969 1.000 0.000 0.000 0.000
#> GSM614394 1 0.0000 0.969 1.000 0.000 0.000 0.000
#> GSM614395 1 0.0000 0.969 1.000 0.000 0.000 0.000
#> GSM614396 1 0.0000 0.969 1.000 0.000 0.000 0.000
#> GSM614397 1 0.0000 0.969 1.000 0.000 0.000 0.000
#> GSM614398 1 0.0000 0.969 1.000 0.000 0.000 0.000
#> GSM614399 3 0.7898 0.390 0.340 0.072 0.512 0.076
#> GSM614400 3 0.6621 0.702 0.144 0.076 0.704 0.076
#> GSM614401 3 0.5183 0.795 0.068 0.076 0.800 0.056
#> GSM614402 3 0.3888 0.834 0.016 0.072 0.860 0.052
#> GSM614403 3 0.1847 0.877 0.004 0.004 0.940 0.052
#> GSM614404 3 0.5542 0.779 0.076 0.076 0.780 0.068
#> GSM614405 3 0.5658 0.723 0.172 0.028 0.744 0.056
#> GSM614406 3 0.5840 0.614 0.264 0.004 0.672 0.060
#> GSM614407 1 0.0707 0.963 0.980 0.000 0.000 0.020
#> GSM614408 1 0.0779 0.963 0.980 0.004 0.000 0.016
#> GSM614409 1 0.0000 0.969 1.000 0.000 0.000 0.000
#> GSM614410 1 0.0336 0.967 0.992 0.000 0.000 0.008
#> GSM614411 1 0.0188 0.968 0.996 0.000 0.000 0.004
#> GSM614412 1 0.0000 0.969 1.000 0.000 0.000 0.000
#> GSM614413 1 0.0000 0.969 1.000 0.000 0.000 0.000
#> GSM614414 1 0.0000 0.969 1.000 0.000 0.000 0.000
#> GSM614445 3 0.0000 0.906 0.000 0.000 1.000 0.000
#> GSM614446 3 0.0000 0.906 0.000 0.000 1.000 0.000
#> GSM614447 3 0.0000 0.906 0.000 0.000 1.000 0.000
#> GSM614448 3 0.0000 0.906 0.000 0.000 1.000 0.000
#> GSM614449 3 0.0000 0.906 0.000 0.000 1.000 0.000
#> GSM614450 3 0.0000 0.906 0.000 0.000 1.000 0.000
#> GSM614451 3 0.0524 0.901 0.000 0.008 0.988 0.004
#> GSM614452 3 0.0524 0.901 0.000 0.008 0.988 0.004
#> GSM614453 2 0.1902 0.957 0.000 0.932 0.004 0.064
#> GSM614454 2 0.1716 0.956 0.000 0.936 0.000 0.064
#> GSM614455 2 0.1978 0.954 0.000 0.928 0.004 0.068
#> GSM614456 2 0.1118 0.963 0.000 0.964 0.000 0.036
#> GSM614457 2 0.1867 0.951 0.000 0.928 0.000 0.072
#> GSM614458 2 0.1474 0.961 0.000 0.948 0.000 0.052
#> GSM614459 2 0.1867 0.951 0.000 0.928 0.000 0.072
#> GSM614460 2 0.1867 0.951 0.000 0.928 0.000 0.072
#> GSM614461 2 0.0188 0.959 0.000 0.996 0.004 0.000
#> GSM614462 2 0.0188 0.959 0.000 0.996 0.004 0.000
#> GSM614463 2 0.0376 0.961 0.000 0.992 0.004 0.004
#> GSM614464 2 0.0779 0.949 0.000 0.980 0.004 0.016
#> GSM614465 2 0.0779 0.962 0.000 0.980 0.004 0.016
#> GSM614466 2 0.0895 0.946 0.000 0.976 0.004 0.020
#> GSM614467 2 0.1807 0.961 0.000 0.940 0.008 0.052
#> GSM614468 2 0.0657 0.952 0.000 0.984 0.004 0.012
#> GSM614469 1 0.3166 0.918 0.896 0.056 0.024 0.024
#> GSM614470 1 0.3166 0.918 0.896 0.056 0.024 0.024
#> GSM614471 1 0.3321 0.911 0.888 0.064 0.024 0.024
#> GSM614472 1 0.3166 0.918 0.896 0.056 0.024 0.024
#> GSM614473 1 0.3166 0.918 0.896 0.056 0.024 0.024
#> GSM614474 1 0.3321 0.911 0.888 0.064 0.024 0.024
#> GSM614475 1 0.3444 0.908 0.884 0.060 0.032 0.024
#> GSM614476 1 0.2945 0.917 0.904 0.024 0.056 0.016
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM614415 5 0.1270 0.884 0.052 0.000 0.000 0.000 0.948
#> GSM614416 5 0.2516 0.853 0.140 0.000 0.000 0.000 0.860
#> GSM614417 5 0.1121 0.884 0.044 0.000 0.000 0.000 0.956
#> GSM614418 5 0.1608 0.883 0.072 0.000 0.000 0.000 0.928
#> GSM614419 5 0.0000 0.878 0.000 0.000 0.000 0.000 1.000
#> GSM614420 5 0.0000 0.878 0.000 0.000 0.000 0.000 1.000
#> GSM614421 3 0.0000 0.875 0.000 0.000 1.000 0.000 0.000
#> GSM614422 3 0.0162 0.874 0.004 0.000 0.996 0.000 0.000
#> GSM614423 3 0.0693 0.871 0.008 0.000 0.980 0.012 0.000
#> GSM614424 3 0.0000 0.875 0.000 0.000 1.000 0.000 0.000
#> GSM614425 3 0.0000 0.875 0.000 0.000 1.000 0.000 0.000
#> GSM614426 3 0.0162 0.875 0.000 0.000 0.996 0.004 0.000
#> GSM614427 3 0.0671 0.870 0.004 0.000 0.980 0.016 0.000
#> GSM614428 3 0.0000 0.875 0.000 0.000 1.000 0.000 0.000
#> GSM614429 2 0.0000 0.955 0.000 1.000 0.000 0.000 0.000
#> GSM614430 2 0.0000 0.955 0.000 1.000 0.000 0.000 0.000
#> GSM614431 2 0.0000 0.955 0.000 1.000 0.000 0.000 0.000
#> GSM614432 2 0.0000 0.955 0.000 1.000 0.000 0.000 0.000
#> GSM614433 2 0.0162 0.954 0.004 0.996 0.000 0.000 0.000
#> GSM614434 2 0.0000 0.955 0.000 1.000 0.000 0.000 0.000
#> GSM614435 2 0.0162 0.954 0.000 0.996 0.000 0.004 0.000
#> GSM614436 2 0.0324 0.954 0.004 0.992 0.000 0.004 0.000
#> GSM614437 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000
#> GSM614438 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000
#> GSM614439 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000
#> GSM614440 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000
#> GSM614441 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000
#> GSM614442 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000
#> GSM614443 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000
#> GSM614444 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000
#> GSM614391 5 0.1908 0.877 0.092 0.000 0.000 0.000 0.908
#> GSM614392 5 0.1732 0.880 0.080 0.000 0.000 0.000 0.920
#> GSM614393 5 0.0510 0.880 0.016 0.000 0.000 0.000 0.984
#> GSM614394 5 0.0000 0.878 0.000 0.000 0.000 0.000 1.000
#> GSM614395 5 0.0794 0.882 0.028 0.000 0.000 0.000 0.972
#> GSM614396 5 0.0000 0.878 0.000 0.000 0.000 0.000 1.000
#> GSM614397 5 0.0794 0.882 0.028 0.000 0.000 0.000 0.972
#> GSM614398 5 0.0162 0.879 0.004 0.000 0.000 0.000 0.996
#> GSM614399 1 0.5446 -0.162 0.484 0.004 0.472 0.008 0.032
#> GSM614400 3 0.5226 0.342 0.404 0.004 0.560 0.008 0.024
#> GSM614401 3 0.4876 0.552 0.320 0.004 0.648 0.008 0.020
#> GSM614402 3 0.3983 0.702 0.220 0.004 0.760 0.008 0.008
#> GSM614403 3 0.3844 0.736 0.180 0.000 0.788 0.028 0.004
#> GSM614404 3 0.5041 0.461 0.364 0.004 0.604 0.008 0.020
#> GSM614405 3 0.5004 0.549 0.312 0.004 0.648 0.008 0.028
#> GSM614406 3 0.5354 0.572 0.276 0.000 0.656 0.036 0.032
#> GSM614407 5 0.3612 0.762 0.268 0.000 0.000 0.000 0.732
#> GSM614408 5 0.3612 0.762 0.268 0.000 0.000 0.000 0.732
#> GSM614409 5 0.3561 0.773 0.260 0.000 0.000 0.000 0.740
#> GSM614410 5 0.3561 0.772 0.260 0.000 0.000 0.000 0.740
#> GSM614411 5 0.3586 0.774 0.264 0.000 0.000 0.000 0.736
#> GSM614412 5 0.2773 0.841 0.164 0.000 0.000 0.000 0.836
#> GSM614413 5 0.2471 0.847 0.136 0.000 0.000 0.000 0.864
#> GSM614414 5 0.2329 0.853 0.124 0.000 0.000 0.000 0.876
#> GSM614445 3 0.0566 0.872 0.004 0.012 0.984 0.000 0.000
#> GSM614446 3 0.0566 0.872 0.004 0.012 0.984 0.000 0.000
#> GSM614447 3 0.0566 0.872 0.004 0.012 0.984 0.000 0.000
#> GSM614448 3 0.0290 0.873 0.000 0.008 0.992 0.000 0.000
#> GSM614449 3 0.0290 0.873 0.000 0.008 0.992 0.000 0.000
#> GSM614450 3 0.0290 0.873 0.000 0.008 0.992 0.000 0.000
#> GSM614451 3 0.0162 0.875 0.000 0.000 0.996 0.004 0.000
#> GSM614452 3 0.0162 0.875 0.000 0.000 0.996 0.004 0.000
#> GSM614453 2 0.2771 0.909 0.128 0.860 0.000 0.012 0.000
#> GSM614454 2 0.2771 0.909 0.128 0.860 0.000 0.012 0.000
#> GSM614455 2 0.2771 0.909 0.128 0.860 0.000 0.012 0.000
#> GSM614456 2 0.2723 0.911 0.124 0.864 0.000 0.012 0.000
#> GSM614457 2 0.2771 0.909 0.128 0.860 0.000 0.012 0.000
#> GSM614458 2 0.2249 0.923 0.096 0.896 0.000 0.008 0.000
#> GSM614459 2 0.2771 0.909 0.128 0.860 0.000 0.012 0.000
#> GSM614460 2 0.2771 0.909 0.128 0.860 0.000 0.012 0.000
#> GSM614461 2 0.0162 0.954 0.004 0.996 0.000 0.000 0.000
#> GSM614462 2 0.0609 0.946 0.020 0.980 0.000 0.000 0.000
#> GSM614463 2 0.0609 0.946 0.020 0.980 0.000 0.000 0.000
#> GSM614464 2 0.0404 0.950 0.012 0.988 0.000 0.000 0.000
#> GSM614465 2 0.0000 0.955 0.000 1.000 0.000 0.000 0.000
#> GSM614466 2 0.0000 0.955 0.000 1.000 0.000 0.000 0.000
#> GSM614467 2 0.0000 0.955 0.000 1.000 0.000 0.000 0.000
#> GSM614468 2 0.0000 0.955 0.000 1.000 0.000 0.000 0.000
#> GSM614469 1 0.2377 0.874 0.872 0.000 0.000 0.000 0.128
#> GSM614470 1 0.2377 0.874 0.872 0.000 0.000 0.000 0.128
#> GSM614471 1 0.2377 0.874 0.872 0.000 0.000 0.000 0.128
#> GSM614472 1 0.2377 0.874 0.872 0.000 0.000 0.000 0.128
#> GSM614473 1 0.2377 0.874 0.872 0.000 0.000 0.000 0.128
#> GSM614474 1 0.2377 0.874 0.872 0.000 0.000 0.000 0.128
#> GSM614475 1 0.4049 0.827 0.792 0.000 0.084 0.000 0.124
#> GSM614476 1 0.4588 0.795 0.748 0.000 0.116 0.000 0.136
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM614415 5 0.3073 0.7199 0.204 0.000 0.000 0.000 0.788 0.008
#> GSM614416 5 0.3323 0.7006 0.240 0.000 0.000 0.000 0.752 0.008
#> GSM614417 5 0.3073 0.7197 0.204 0.000 0.000 0.000 0.788 0.008
#> GSM614418 5 0.3103 0.7179 0.208 0.000 0.000 0.000 0.784 0.008
#> GSM614419 5 0.2300 0.7084 0.000 0.000 0.000 0.000 0.856 0.144
#> GSM614420 5 0.2260 0.7097 0.000 0.000 0.000 0.000 0.860 0.140
#> GSM614421 3 0.0000 0.8154 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM614422 3 0.0000 0.8154 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM614423 3 0.0858 0.8087 0.004 0.000 0.968 0.000 0.000 0.028
#> GSM614424 3 0.0000 0.8154 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM614425 3 0.0000 0.8154 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM614426 3 0.0000 0.8154 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM614427 3 0.0713 0.8072 0.000 0.000 0.972 0.000 0.000 0.028
#> GSM614428 3 0.0146 0.8154 0.000 0.000 0.996 0.000 0.000 0.004
#> GSM614429 2 0.3862 -0.7756 0.000 0.524 0.000 0.000 0.000 0.476
#> GSM614430 2 0.3864 -0.7860 0.000 0.520 0.000 0.000 0.000 0.480
#> GSM614431 2 0.3864 -0.7859 0.000 0.520 0.000 0.000 0.000 0.480
#> GSM614432 2 0.3864 -0.7859 0.000 0.520 0.000 0.000 0.000 0.480
#> GSM614433 6 0.3868 0.8186 0.000 0.492 0.000 0.000 0.000 0.508
#> GSM614434 2 0.3862 -0.7794 0.000 0.524 0.000 0.000 0.000 0.476
#> GSM614435 2 0.3833 -0.6890 0.000 0.556 0.000 0.000 0.000 0.444
#> GSM614436 2 0.3999 -0.7532 0.004 0.500 0.000 0.000 0.000 0.496
#> GSM614437 4 0.0000 1.0000 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM614438 4 0.0000 1.0000 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM614439 4 0.0000 1.0000 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM614440 4 0.0000 1.0000 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM614441 4 0.0000 1.0000 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM614442 4 0.0000 1.0000 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM614443 4 0.0000 1.0000 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM614444 4 0.0000 1.0000 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM614391 5 0.2994 0.7236 0.208 0.000 0.000 0.000 0.788 0.004
#> GSM614392 5 0.2883 0.7192 0.212 0.000 0.000 0.000 0.788 0.000
#> GSM614393 5 0.2106 0.7352 0.064 0.000 0.000 0.000 0.904 0.032
#> GSM614394 5 0.2300 0.7084 0.000 0.000 0.000 0.000 0.856 0.144
#> GSM614395 5 0.3539 0.7011 0.024 0.000 0.000 0.000 0.756 0.220
#> GSM614396 5 0.2300 0.7084 0.000 0.000 0.000 0.000 0.856 0.144
#> GSM614397 5 0.3614 0.7004 0.028 0.000 0.000 0.000 0.752 0.220
#> GSM614398 5 0.2300 0.7084 0.000 0.000 0.000 0.000 0.856 0.144
#> GSM614399 1 0.6199 -0.1332 0.408 0.008 0.388 0.000 0.004 0.192
#> GSM614400 3 0.6107 0.0687 0.380 0.004 0.388 0.000 0.000 0.228
#> GSM614401 3 0.6214 0.1563 0.348 0.004 0.416 0.000 0.004 0.228
#> GSM614402 3 0.5768 0.3011 0.316 0.000 0.488 0.000 0.000 0.196
#> GSM614403 3 0.4843 0.5532 0.192 0.000 0.664 0.000 0.000 0.144
#> GSM614404 3 0.6226 0.1143 0.364 0.004 0.400 0.000 0.004 0.228
#> GSM614405 3 0.5983 0.3097 0.316 0.000 0.504 0.000 0.016 0.164
#> GSM614406 3 0.6056 0.3204 0.292 0.000 0.504 0.000 0.016 0.188
#> GSM614407 5 0.5015 0.6176 0.352 0.000 0.000 0.000 0.564 0.084
#> GSM614408 5 0.4881 0.6361 0.336 0.000 0.000 0.000 0.588 0.076
#> GSM614409 5 0.4993 0.6259 0.344 0.000 0.000 0.000 0.572 0.084
#> GSM614410 5 0.4972 0.6194 0.352 0.000 0.000 0.000 0.568 0.080
#> GSM614411 5 0.4913 0.6368 0.332 0.000 0.000 0.000 0.588 0.080
#> GSM614412 5 0.4904 0.6876 0.236 0.000 0.000 0.000 0.644 0.120
#> GSM614413 5 0.4650 0.6739 0.104 0.000 0.000 0.000 0.676 0.220
#> GSM614414 5 0.4650 0.6739 0.104 0.000 0.000 0.000 0.676 0.220
#> GSM614445 3 0.0508 0.8143 0.004 0.000 0.984 0.000 0.000 0.012
#> GSM614446 3 0.0508 0.8143 0.004 0.000 0.984 0.000 0.000 0.012
#> GSM614447 3 0.0508 0.8143 0.004 0.000 0.984 0.000 0.000 0.012
#> GSM614448 3 0.0260 0.8151 0.000 0.000 0.992 0.000 0.000 0.008
#> GSM614449 3 0.0363 0.8148 0.000 0.000 0.988 0.000 0.000 0.012
#> GSM614450 3 0.0363 0.8148 0.000 0.000 0.988 0.000 0.000 0.012
#> GSM614451 3 0.0146 0.8148 0.000 0.000 0.996 0.004 0.000 0.000
#> GSM614452 3 0.0146 0.8148 0.000 0.000 0.996 0.004 0.000 0.000
#> GSM614453 2 0.0260 0.4652 0.000 0.992 0.000 0.000 0.000 0.008
#> GSM614454 2 0.0363 0.4628 0.000 0.988 0.000 0.000 0.000 0.012
#> GSM614455 2 0.0260 0.4652 0.000 0.992 0.000 0.000 0.000 0.008
#> GSM614456 2 0.0000 0.4657 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM614457 2 0.0000 0.4657 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM614458 2 0.3531 -0.2806 0.000 0.672 0.000 0.000 0.000 0.328
#> GSM614459 2 0.0000 0.4657 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM614460 2 0.0000 0.4657 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM614461 6 0.3838 0.8755 0.000 0.448 0.000 0.000 0.000 0.552
#> GSM614462 6 0.3810 0.8439 0.000 0.428 0.000 0.000 0.000 0.572
#> GSM614463 6 0.3774 0.7980 0.000 0.408 0.000 0.000 0.000 0.592
#> GSM614464 6 0.3854 0.8719 0.000 0.464 0.000 0.000 0.000 0.536
#> GSM614465 6 0.3838 0.8756 0.000 0.448 0.000 0.000 0.000 0.552
#> GSM614466 6 0.3867 0.8499 0.000 0.488 0.000 0.000 0.000 0.512
#> GSM614467 6 0.3869 0.7867 0.000 0.500 0.000 0.000 0.000 0.500
#> GSM614468 6 0.3868 0.8208 0.000 0.496 0.000 0.000 0.000 0.504
#> GSM614469 1 0.0000 0.8434 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM614470 1 0.0363 0.8329 0.988 0.000 0.000 0.000 0.012 0.000
#> GSM614471 1 0.0000 0.8434 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM614472 1 0.0000 0.8434 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM614473 1 0.0146 0.8408 0.996 0.000 0.000 0.000 0.004 0.000
#> GSM614474 1 0.0000 0.8434 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM614475 1 0.2393 0.7853 0.884 0.000 0.092 0.000 0.004 0.020
#> GSM614476 1 0.4887 0.6202 0.700 0.000 0.192 0.000 0.036 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 individual(p) protocol(p) time(p) other(p) k
#> MAD:mclust 82 1.00e-12 0.926 0.999 0.6938 2
#> MAD:mclust 84 2.12e-25 0.814 1.000 0.1294 3
#> MAD:mclust 85 1.64e-37 0.954 1.000 0.0156 4
#> MAD:mclust 83 1.14e-47 0.955 1.000 0.0147 5
#> MAD:mclust 64 5.53e-35 0.990 1.000 0.2076 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 21168 rows and 86 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 0.900 0.917 0.962 0.4961 0.501 0.501
#> 3 3 0.786 0.853 0.937 0.3323 0.722 0.502
#> 4 4 0.745 0.782 0.883 0.1214 0.818 0.532
#> 5 5 0.656 0.568 0.752 0.0674 0.933 0.749
#> 6 6 0.700 0.664 0.782 0.0401 0.897 0.590
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
#> GSM614415 1 0.0000 0.954 1.000 0.000
#> GSM614416 1 0.0000 0.954 1.000 0.000
#> GSM614417 1 0.0000 0.954 1.000 0.000
#> GSM614418 1 0.0000 0.954 1.000 0.000
#> GSM614419 1 0.0000 0.954 1.000 0.000
#> GSM614420 1 0.0000 0.954 1.000 0.000
#> GSM614421 2 0.6973 0.766 0.188 0.812
#> GSM614422 1 0.0376 0.952 0.996 0.004
#> GSM614423 1 0.9661 0.356 0.608 0.392
#> GSM614424 2 0.6148 0.818 0.152 0.848
#> GSM614425 1 0.9866 0.263 0.568 0.432
#> GSM614426 1 0.7674 0.725 0.776 0.224
#> GSM614427 2 0.1184 0.957 0.016 0.984
#> GSM614428 2 0.1843 0.950 0.028 0.972
#> GSM614429 2 0.0000 0.963 0.000 1.000
#> GSM614430 2 0.0000 0.963 0.000 1.000
#> GSM614431 2 0.1184 0.962 0.016 0.984
#> GSM614432 2 0.1184 0.962 0.016 0.984
#> GSM614433 2 0.1184 0.962 0.016 0.984
#> GSM614434 2 0.1184 0.962 0.016 0.984
#> GSM614435 2 0.0000 0.963 0.000 1.000
#> GSM614436 2 0.0000 0.963 0.000 1.000
#> GSM614437 2 0.0000 0.963 0.000 1.000
#> GSM614438 2 0.0000 0.963 0.000 1.000
#> GSM614439 2 0.0000 0.963 0.000 1.000
#> GSM614440 2 0.0000 0.963 0.000 1.000
#> GSM614441 2 0.0000 0.963 0.000 1.000
#> GSM614442 2 0.0000 0.963 0.000 1.000
#> GSM614443 2 0.0000 0.963 0.000 1.000
#> GSM614444 2 0.0000 0.963 0.000 1.000
#> GSM614391 1 0.0000 0.954 1.000 0.000
#> GSM614392 1 0.0000 0.954 1.000 0.000
#> GSM614393 1 0.0000 0.954 1.000 0.000
#> GSM614394 1 0.0000 0.954 1.000 0.000
#> GSM614395 1 0.1414 0.942 0.980 0.020
#> GSM614396 1 0.0000 0.954 1.000 0.000
#> GSM614397 1 0.1184 0.944 0.984 0.016
#> GSM614398 1 0.0376 0.952 0.996 0.004
#> GSM614399 2 0.3584 0.924 0.068 0.932
#> GSM614400 1 0.3584 0.901 0.932 0.068
#> GSM614401 1 0.0000 0.954 1.000 0.000
#> GSM614402 1 0.7056 0.760 0.808 0.192
#> GSM614403 2 0.9248 0.503 0.340 0.660
#> GSM614404 1 0.7376 0.739 0.792 0.208
#> GSM614405 2 0.9552 0.412 0.376 0.624
#> GSM614406 2 0.0000 0.963 0.000 1.000
#> GSM614407 1 0.0000 0.954 1.000 0.000
#> GSM614408 1 0.0000 0.954 1.000 0.000
#> GSM614409 1 0.0000 0.954 1.000 0.000
#> GSM614410 1 0.0000 0.954 1.000 0.000
#> GSM614411 1 0.0000 0.954 1.000 0.000
#> GSM614412 1 0.0000 0.954 1.000 0.000
#> GSM614413 1 0.0672 0.950 0.992 0.008
#> GSM614414 1 0.0376 0.952 0.996 0.004
#> GSM614445 2 0.4562 0.897 0.096 0.904
#> GSM614446 2 0.2236 0.951 0.036 0.964
#> GSM614447 2 0.2423 0.948 0.040 0.960
#> GSM614448 2 0.1184 0.959 0.016 0.984
#> GSM614449 2 0.0376 0.963 0.004 0.996
#> GSM614450 2 0.3584 0.924 0.068 0.932
#> GSM614451 2 0.0000 0.963 0.000 1.000
#> GSM614452 2 0.0000 0.963 0.000 1.000
#> GSM614453 2 0.1184 0.962 0.016 0.984
#> GSM614454 2 0.1184 0.962 0.016 0.984
#> GSM614455 2 0.1184 0.962 0.016 0.984
#> GSM614456 2 0.0000 0.963 0.000 1.000
#> GSM614457 2 0.0000 0.963 0.000 1.000
#> GSM614458 2 0.0000 0.963 0.000 1.000
#> GSM614459 2 0.0000 0.963 0.000 1.000
#> GSM614460 2 0.0000 0.963 0.000 1.000
#> GSM614461 2 0.1184 0.962 0.016 0.984
#> GSM614462 2 0.1414 0.960 0.020 0.980
#> GSM614463 2 0.1633 0.958 0.024 0.976
#> GSM614464 2 0.1184 0.962 0.016 0.984
#> GSM614465 2 0.1414 0.960 0.020 0.980
#> GSM614466 2 0.1414 0.960 0.020 0.980
#> GSM614467 2 0.0000 0.963 0.000 1.000
#> GSM614468 2 0.1184 0.962 0.016 0.984
#> GSM614469 1 0.0000 0.954 1.000 0.000
#> GSM614470 1 0.0000 0.954 1.000 0.000
#> GSM614471 1 0.0000 0.954 1.000 0.000
#> GSM614472 1 0.0000 0.954 1.000 0.000
#> GSM614473 1 0.0000 0.954 1.000 0.000
#> GSM614474 1 0.0000 0.954 1.000 0.000
#> GSM614475 1 0.0376 0.952 0.996 0.004
#> GSM614476 1 0.2043 0.932 0.968 0.032
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM614415 1 0.0000 0.8905 1.000 0.000 0.000
#> GSM614416 1 0.0000 0.8905 1.000 0.000 0.000
#> GSM614417 1 0.0000 0.8905 1.000 0.000 0.000
#> GSM614418 1 0.0000 0.8905 1.000 0.000 0.000
#> GSM614419 1 0.0000 0.8905 1.000 0.000 0.000
#> GSM614420 1 0.0000 0.8905 1.000 0.000 0.000
#> GSM614421 3 0.1163 0.9427 0.028 0.000 0.972
#> GSM614422 1 0.3412 0.7947 0.876 0.000 0.124
#> GSM614423 1 0.6169 0.4664 0.636 0.360 0.004
#> GSM614424 3 0.1031 0.9458 0.024 0.000 0.976
#> GSM614425 3 0.3482 0.8522 0.128 0.000 0.872
#> GSM614426 3 0.4346 0.7827 0.184 0.000 0.816
#> GSM614427 3 0.0424 0.9541 0.008 0.000 0.992
#> GSM614428 3 0.0237 0.9560 0.004 0.000 0.996
#> GSM614429 2 0.0592 0.9440 0.000 0.988 0.012
#> GSM614430 2 0.0592 0.9440 0.000 0.988 0.012
#> GSM614431 2 0.0000 0.9468 0.000 1.000 0.000
#> GSM614432 2 0.0000 0.9468 0.000 1.000 0.000
#> GSM614433 2 0.0000 0.9468 0.000 1.000 0.000
#> GSM614434 2 0.0237 0.9466 0.000 0.996 0.004
#> GSM614435 2 0.2625 0.8836 0.000 0.916 0.084
#> GSM614436 3 0.3482 0.8358 0.000 0.128 0.872
#> GSM614437 3 0.1031 0.9411 0.000 0.024 0.976
#> GSM614438 3 0.0000 0.9576 0.000 0.000 1.000
#> GSM614439 3 0.0000 0.9576 0.000 0.000 1.000
#> GSM614440 3 0.0000 0.9576 0.000 0.000 1.000
#> GSM614441 3 0.0000 0.9576 0.000 0.000 1.000
#> GSM614442 3 0.0000 0.9576 0.000 0.000 1.000
#> GSM614443 3 0.0424 0.9530 0.000 0.008 0.992
#> GSM614444 3 0.0000 0.9576 0.000 0.000 1.000
#> GSM614391 1 0.0000 0.8905 1.000 0.000 0.000
#> GSM614392 1 0.0000 0.8905 1.000 0.000 0.000
#> GSM614393 1 0.0000 0.8905 1.000 0.000 0.000
#> GSM614394 1 0.0000 0.8905 1.000 0.000 0.000
#> GSM614395 1 0.6225 0.1845 0.568 0.000 0.432
#> GSM614396 1 0.0000 0.8905 1.000 0.000 0.000
#> GSM614397 1 0.2711 0.8281 0.912 0.000 0.088
#> GSM614398 1 0.0747 0.8821 0.984 0.000 0.016
#> GSM614399 2 0.0000 0.9468 0.000 1.000 0.000
#> GSM614400 2 0.0000 0.9468 0.000 1.000 0.000
#> GSM614401 2 0.3038 0.8481 0.104 0.896 0.000
#> GSM614402 2 0.0424 0.9431 0.008 0.992 0.000
#> GSM614403 2 0.5835 0.4212 0.340 0.660 0.000
#> GSM614404 2 0.0237 0.9452 0.004 0.996 0.000
#> GSM614405 1 0.6318 0.4793 0.636 0.356 0.008
#> GSM614406 3 0.0000 0.9576 0.000 0.000 1.000
#> GSM614407 1 0.0000 0.8905 1.000 0.000 0.000
#> GSM614408 1 0.0000 0.8905 1.000 0.000 0.000
#> GSM614409 1 0.0000 0.8905 1.000 0.000 0.000
#> GSM614410 1 0.0000 0.8905 1.000 0.000 0.000
#> GSM614411 1 0.0000 0.8905 1.000 0.000 0.000
#> GSM614412 1 0.0000 0.8905 1.000 0.000 0.000
#> GSM614413 1 0.1643 0.8643 0.956 0.000 0.044
#> GSM614414 1 0.0424 0.8865 0.992 0.000 0.008
#> GSM614445 2 0.0747 0.9380 0.016 0.984 0.000
#> GSM614446 2 0.1877 0.9232 0.012 0.956 0.032
#> GSM614447 2 0.0000 0.9468 0.000 1.000 0.000
#> GSM614448 3 0.0000 0.9576 0.000 0.000 1.000
#> GSM614449 3 0.0000 0.9576 0.000 0.000 1.000
#> GSM614450 3 0.6565 0.6519 0.232 0.048 0.720
#> GSM614451 3 0.0000 0.9576 0.000 0.000 1.000
#> GSM614452 3 0.0000 0.9576 0.000 0.000 1.000
#> GSM614453 2 0.0000 0.9468 0.000 1.000 0.000
#> GSM614454 2 0.0237 0.9466 0.000 0.996 0.004
#> GSM614455 2 0.0237 0.9466 0.000 0.996 0.004
#> GSM614456 2 0.0892 0.9404 0.000 0.980 0.020
#> GSM614457 2 0.0892 0.9398 0.000 0.980 0.020
#> GSM614458 2 0.0592 0.9440 0.000 0.988 0.012
#> GSM614459 2 0.4504 0.7548 0.000 0.804 0.196
#> GSM614460 2 0.0892 0.9405 0.000 0.980 0.020
#> GSM614461 2 0.0000 0.9468 0.000 1.000 0.000
#> GSM614462 2 0.0000 0.9468 0.000 1.000 0.000
#> GSM614463 2 0.0000 0.9468 0.000 1.000 0.000
#> GSM614464 2 0.0237 0.9466 0.000 0.996 0.004
#> GSM614465 2 0.0000 0.9468 0.000 1.000 0.000
#> GSM614466 2 0.0000 0.9468 0.000 1.000 0.000
#> GSM614467 2 0.4346 0.7720 0.000 0.816 0.184
#> GSM614468 2 0.0237 0.9466 0.000 0.996 0.004
#> GSM614469 1 0.4654 0.7308 0.792 0.208 0.000
#> GSM614470 1 0.4002 0.7856 0.840 0.160 0.000
#> GSM614471 2 0.6295 -0.0197 0.472 0.528 0.000
#> GSM614472 1 0.6307 0.1193 0.512 0.488 0.000
#> GSM614473 1 0.3267 0.8239 0.884 0.116 0.000
#> GSM614474 1 0.3412 0.8189 0.876 0.124 0.000
#> GSM614475 1 0.6305 0.1380 0.516 0.484 0.000
#> GSM614476 1 0.0000 0.8905 1.000 0.000 0.000
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM614415 1 0.0000 0.93927 1.000 0.000 0.000 0.000
#> GSM614416 1 0.0000 0.93927 1.000 0.000 0.000 0.000
#> GSM614417 1 0.0000 0.93927 1.000 0.000 0.000 0.000
#> GSM614418 1 0.0000 0.93927 1.000 0.000 0.000 0.000
#> GSM614419 1 0.0188 0.93830 0.996 0.000 0.004 0.000
#> GSM614420 1 0.0188 0.93830 0.996 0.000 0.004 0.000
#> GSM614421 3 0.2021 0.84430 0.000 0.056 0.932 0.012
#> GSM614422 3 0.2124 0.84714 0.008 0.068 0.924 0.000
#> GSM614423 3 0.3873 0.76862 0.000 0.228 0.772 0.000
#> GSM614424 3 0.2197 0.84686 0.000 0.080 0.916 0.004
#> GSM614425 3 0.1890 0.84545 0.000 0.056 0.936 0.008
#> GSM614426 3 0.1824 0.84646 0.000 0.060 0.936 0.004
#> GSM614427 3 0.1584 0.83426 0.000 0.036 0.952 0.012
#> GSM614428 3 0.0927 0.81232 0.000 0.008 0.976 0.016
#> GSM614429 2 0.2345 0.77909 0.000 0.900 0.000 0.100
#> GSM614430 2 0.2334 0.79167 0.000 0.908 0.004 0.088
#> GSM614431 2 0.1545 0.82538 0.000 0.952 0.008 0.040
#> GSM614432 2 0.1411 0.83908 0.000 0.960 0.020 0.020
#> GSM614433 2 0.1489 0.84289 0.000 0.952 0.044 0.004
#> GSM614434 2 0.1398 0.82335 0.000 0.956 0.004 0.040
#> GSM614435 4 0.5168 -0.02432 0.000 0.492 0.004 0.504
#> GSM614436 4 0.3198 0.83025 0.000 0.040 0.080 0.880
#> GSM614437 4 0.1059 0.82544 0.000 0.012 0.016 0.972
#> GSM614438 4 0.2973 0.82101 0.000 0.000 0.144 0.856
#> GSM614439 4 0.3074 0.81505 0.000 0.000 0.152 0.848
#> GSM614440 4 0.2921 0.82337 0.000 0.000 0.140 0.860
#> GSM614441 4 0.3074 0.81505 0.000 0.000 0.152 0.848
#> GSM614442 4 0.2647 0.82877 0.000 0.000 0.120 0.880
#> GSM614443 4 0.1209 0.83004 0.000 0.004 0.032 0.964
#> GSM614444 4 0.2921 0.82337 0.000 0.000 0.140 0.860
#> GSM614391 1 0.0000 0.93927 1.000 0.000 0.000 0.000
#> GSM614392 1 0.0000 0.93927 1.000 0.000 0.000 0.000
#> GSM614393 1 0.0000 0.93927 1.000 0.000 0.000 0.000
#> GSM614394 1 0.0336 0.93683 0.992 0.000 0.008 0.000
#> GSM614395 3 0.6138 0.44563 0.260 0.000 0.648 0.092
#> GSM614396 1 0.0336 0.93683 0.992 0.000 0.008 0.000
#> GSM614397 1 0.4917 0.52336 0.656 0.000 0.336 0.008
#> GSM614398 1 0.1302 0.91613 0.956 0.000 0.044 0.000
#> GSM614399 2 0.1629 0.83727 0.000 0.952 0.024 0.024
#> GSM614400 2 0.1706 0.83899 0.016 0.948 0.036 0.000
#> GSM614401 2 0.2670 0.81598 0.040 0.908 0.052 0.000
#> GSM614402 2 0.2408 0.80078 0.000 0.896 0.104 0.000
#> GSM614403 3 0.4431 0.68033 0.000 0.304 0.696 0.000
#> GSM614404 2 0.1398 0.84165 0.004 0.956 0.040 0.000
#> GSM614405 3 0.5677 0.58984 0.040 0.332 0.628 0.000
#> GSM614406 3 0.2999 0.71115 0.000 0.004 0.864 0.132
#> GSM614407 1 0.0000 0.93927 1.000 0.000 0.000 0.000
#> GSM614408 1 0.0000 0.93927 1.000 0.000 0.000 0.000
#> GSM614409 1 0.0000 0.93927 1.000 0.000 0.000 0.000
#> GSM614410 1 0.0000 0.93927 1.000 0.000 0.000 0.000
#> GSM614411 1 0.0000 0.93927 1.000 0.000 0.000 0.000
#> GSM614412 1 0.0000 0.93927 1.000 0.000 0.000 0.000
#> GSM614413 1 0.4889 0.47186 0.636 0.000 0.360 0.004
#> GSM614414 1 0.1637 0.90478 0.940 0.000 0.060 0.000
#> GSM614445 3 0.4382 0.69156 0.000 0.296 0.704 0.000
#> GSM614446 3 0.3907 0.76528 0.000 0.232 0.768 0.000
#> GSM614447 3 0.4072 0.74543 0.000 0.252 0.748 0.000
#> GSM614448 3 0.1576 0.84384 0.000 0.048 0.948 0.004
#> GSM614449 3 0.2714 0.83741 0.000 0.112 0.884 0.004
#> GSM614450 3 0.3024 0.81948 0.000 0.148 0.852 0.000
#> GSM614451 3 0.2149 0.74891 0.000 0.000 0.912 0.088
#> GSM614452 3 0.1940 0.75902 0.000 0.000 0.924 0.076
#> GSM614453 2 0.4134 0.57559 0.000 0.740 0.000 0.260
#> GSM614454 2 0.4941 0.19959 0.000 0.564 0.000 0.436
#> GSM614455 2 0.4941 0.20207 0.000 0.564 0.000 0.436
#> GSM614456 4 0.3610 0.70259 0.000 0.200 0.000 0.800
#> GSM614457 4 0.3266 0.73638 0.000 0.168 0.000 0.832
#> GSM614458 2 0.4999 0.00719 0.000 0.508 0.000 0.492
#> GSM614459 4 0.2216 0.78803 0.000 0.092 0.000 0.908
#> GSM614460 4 0.3610 0.70471 0.000 0.200 0.000 0.800
#> GSM614461 2 0.0895 0.84207 0.000 0.976 0.020 0.004
#> GSM614462 2 0.1211 0.84241 0.000 0.960 0.040 0.000
#> GSM614463 2 0.0921 0.84294 0.000 0.972 0.028 0.000
#> GSM614464 2 0.1792 0.82873 0.000 0.932 0.068 0.000
#> GSM614465 2 0.2281 0.80809 0.000 0.904 0.096 0.000
#> GSM614466 2 0.1389 0.83971 0.000 0.952 0.048 0.000
#> GSM614467 2 0.4790 0.21886 0.000 0.620 0.380 0.000
#> GSM614468 2 0.2647 0.78166 0.000 0.880 0.120 0.000
#> GSM614469 1 0.0779 0.93166 0.980 0.016 0.004 0.000
#> GSM614470 1 0.0779 0.93166 0.980 0.016 0.004 0.000
#> GSM614471 1 0.3892 0.75268 0.800 0.192 0.004 0.004
#> GSM614472 1 0.2888 0.83688 0.872 0.124 0.004 0.000
#> GSM614473 1 0.0469 0.93448 0.988 0.012 0.000 0.000
#> GSM614474 1 0.0779 0.93125 0.980 0.016 0.004 0.000
#> GSM614475 1 0.4428 0.62266 0.720 0.276 0.004 0.000
#> GSM614476 1 0.2197 0.88541 0.916 0.004 0.080 0.000
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM614415 5 0.0609 0.76453 0.020 0.000 0.000 0.000 0.980
#> GSM614416 5 0.0609 0.76453 0.020 0.000 0.000 0.000 0.980
#> GSM614417 5 0.0510 0.76480 0.016 0.000 0.000 0.000 0.984
#> GSM614418 5 0.0510 0.76480 0.016 0.000 0.000 0.000 0.984
#> GSM614419 5 0.0609 0.76548 0.020 0.000 0.000 0.000 0.980
#> GSM614420 5 0.0404 0.76547 0.012 0.000 0.000 0.000 0.988
#> GSM614421 3 0.1059 0.74628 0.020 0.008 0.968 0.004 0.000
#> GSM614422 3 0.1267 0.74530 0.024 0.012 0.960 0.004 0.000
#> GSM614423 3 0.3527 0.70366 0.056 0.116 0.828 0.000 0.000
#> GSM614424 3 0.1202 0.74998 0.004 0.032 0.960 0.004 0.000
#> GSM614425 3 0.0854 0.74853 0.012 0.008 0.976 0.004 0.000
#> GSM614426 3 0.0960 0.75109 0.016 0.008 0.972 0.004 0.000
#> GSM614427 3 0.0740 0.74938 0.008 0.008 0.980 0.004 0.000
#> GSM614428 3 0.1372 0.73722 0.024 0.004 0.956 0.016 0.000
#> GSM614429 2 0.2793 0.61103 0.036 0.876 0.000 0.088 0.000
#> GSM614430 2 0.3096 0.62122 0.084 0.868 0.008 0.040 0.000
#> GSM614431 2 0.1725 0.65320 0.044 0.936 0.000 0.020 0.000
#> GSM614432 2 0.1901 0.65469 0.056 0.928 0.004 0.012 0.000
#> GSM614433 2 0.1818 0.68300 0.024 0.932 0.044 0.000 0.000
#> GSM614434 2 0.2426 0.64013 0.064 0.900 0.000 0.036 0.000
#> GSM614435 2 0.5755 0.30090 0.100 0.640 0.016 0.244 0.000
#> GSM614436 4 0.6819 0.39996 0.080 0.348 0.068 0.504 0.000
#> GSM614437 4 0.0671 0.78898 0.004 0.016 0.000 0.980 0.000
#> GSM614438 4 0.1544 0.79099 0.000 0.000 0.068 0.932 0.000
#> GSM614439 4 0.1478 0.79393 0.000 0.000 0.064 0.936 0.000
#> GSM614440 4 0.1410 0.79566 0.000 0.000 0.060 0.940 0.000
#> GSM614441 4 0.1478 0.79393 0.000 0.000 0.064 0.936 0.000
#> GSM614442 4 0.1121 0.79791 0.000 0.000 0.044 0.956 0.000
#> GSM614443 4 0.0566 0.78977 0.004 0.012 0.000 0.984 0.000
#> GSM614444 4 0.1270 0.79757 0.000 0.000 0.052 0.948 0.000
#> GSM614391 5 0.3304 0.68280 0.168 0.000 0.016 0.000 0.816
#> GSM614392 5 0.2929 0.70117 0.152 0.000 0.008 0.000 0.840
#> GSM614393 5 0.2674 0.70851 0.140 0.000 0.004 0.000 0.856
#> GSM614394 5 0.3574 0.67248 0.168 0.000 0.028 0.000 0.804
#> GSM614395 3 0.7213 0.22173 0.112 0.000 0.540 0.108 0.240
#> GSM614396 5 0.3495 0.68071 0.160 0.000 0.028 0.000 0.812
#> GSM614397 5 0.6914 0.00292 0.172 0.000 0.388 0.020 0.420
#> GSM614398 5 0.5101 0.54608 0.184 0.000 0.108 0.004 0.704
#> GSM614399 2 0.6506 0.45017 0.364 0.508 0.032 0.000 0.096
#> GSM614400 2 0.7112 0.31422 0.364 0.412 0.024 0.000 0.200
#> GSM614401 1 0.7360 -0.29429 0.376 0.336 0.028 0.000 0.260
#> GSM614402 2 0.7033 0.41306 0.364 0.472 0.076 0.000 0.088
#> GSM614403 3 0.7310 0.25730 0.360 0.236 0.376 0.000 0.028
#> GSM614404 2 0.6698 0.42533 0.364 0.488 0.032 0.000 0.116
#> GSM614405 1 0.7833 -0.39102 0.376 0.216 0.332 0.000 0.076
#> GSM614406 3 0.5869 0.62155 0.164 0.000 0.632 0.196 0.008
#> GSM614407 1 0.4517 0.38060 0.556 0.008 0.000 0.000 0.436
#> GSM614408 5 0.4449 -0.29095 0.484 0.004 0.000 0.000 0.512
#> GSM614409 1 0.4759 0.45477 0.600 0.012 0.008 0.000 0.380
#> GSM614410 1 0.4528 0.36247 0.548 0.008 0.000 0.000 0.444
#> GSM614411 1 0.5000 0.46732 0.604 0.016 0.016 0.000 0.364
#> GSM614412 1 0.5419 0.47788 0.608 0.016 0.044 0.000 0.332
#> GSM614413 1 0.6170 0.42681 0.620 0.016 0.220 0.004 0.140
#> GSM614414 1 0.5916 0.46601 0.612 0.008 0.132 0.000 0.248
#> GSM614445 3 0.6497 0.39794 0.312 0.212 0.476 0.000 0.000
#> GSM614446 3 0.5796 0.56496 0.284 0.128 0.588 0.000 0.000
#> GSM614447 3 0.6358 0.44300 0.328 0.180 0.492 0.000 0.000
#> GSM614448 3 0.2616 0.74594 0.100 0.000 0.880 0.020 0.000
#> GSM614449 3 0.3961 0.71469 0.160 0.044 0.792 0.004 0.000
#> GSM614450 3 0.5030 0.64924 0.264 0.060 0.672 0.000 0.004
#> GSM614451 3 0.3060 0.70078 0.024 0.000 0.848 0.128 0.000
#> GSM614452 3 0.2915 0.70887 0.024 0.000 0.860 0.116 0.000
#> GSM614453 2 0.3388 0.51293 0.008 0.792 0.000 0.200 0.000
#> GSM614454 2 0.3999 0.27691 0.000 0.656 0.000 0.344 0.000
#> GSM614455 2 0.4088 0.22876 0.000 0.632 0.000 0.368 0.000
#> GSM614456 4 0.3932 0.57854 0.000 0.328 0.000 0.672 0.000
#> GSM614457 4 0.3895 0.59464 0.000 0.320 0.000 0.680 0.000
#> GSM614458 2 0.4331 0.08113 0.004 0.596 0.000 0.400 0.000
#> GSM614459 4 0.3612 0.64930 0.000 0.268 0.000 0.732 0.000
#> GSM614460 4 0.4126 0.50445 0.000 0.380 0.000 0.620 0.000
#> GSM614461 2 0.2131 0.68298 0.056 0.920 0.016 0.008 0.000
#> GSM614462 2 0.3835 0.67355 0.156 0.796 0.048 0.000 0.000
#> GSM614463 2 0.3370 0.68022 0.148 0.824 0.028 0.000 0.000
#> GSM614464 2 0.4627 0.63302 0.188 0.732 0.080 0.000 0.000
#> GSM614465 2 0.4810 0.61906 0.204 0.712 0.084 0.000 0.000
#> GSM614466 2 0.4152 0.65960 0.168 0.772 0.060 0.000 0.000
#> GSM614467 2 0.5114 0.34980 0.052 0.608 0.340 0.000 0.000
#> GSM614468 2 0.3339 0.67025 0.048 0.840 0.112 0.000 0.000
#> GSM614469 5 0.1484 0.75302 0.048 0.008 0.000 0.000 0.944
#> GSM614470 5 0.1894 0.73652 0.072 0.008 0.000 0.000 0.920
#> GSM614471 5 0.3471 0.66054 0.072 0.092 0.000 0.000 0.836
#> GSM614472 5 0.3323 0.66877 0.100 0.056 0.000 0.000 0.844
#> GSM614473 5 0.1502 0.75230 0.056 0.004 0.000 0.000 0.940
#> GSM614474 5 0.1710 0.75654 0.040 0.016 0.004 0.000 0.940
#> GSM614475 5 0.5734 0.32520 0.072 0.308 0.016 0.000 0.604
#> GSM614476 5 0.4353 0.65694 0.096 0.008 0.100 0.004 0.792
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM614415 5 0.1738 0.696 0.016 0.000 0.000 0.004 0.928 0.052
#> GSM614416 5 0.1826 0.696 0.020 0.000 0.000 0.004 0.924 0.052
#> GSM614417 5 0.1644 0.696 0.012 0.000 0.000 0.004 0.932 0.052
#> GSM614418 5 0.1644 0.696 0.012 0.000 0.000 0.004 0.932 0.052
#> GSM614419 5 0.1909 0.698 0.024 0.000 0.000 0.004 0.920 0.052
#> GSM614420 5 0.1826 0.698 0.020 0.000 0.000 0.004 0.924 0.052
#> GSM614421 3 0.1138 0.777 0.012 0.000 0.960 0.000 0.004 0.024
#> GSM614422 3 0.1332 0.763 0.012 0.000 0.952 0.000 0.008 0.028
#> GSM614423 3 0.3452 0.712 0.068 0.032 0.848 0.000 0.016 0.036
#> GSM614424 3 0.1528 0.763 0.048 0.000 0.936 0.000 0.000 0.016
#> GSM614425 3 0.1148 0.776 0.016 0.000 0.960 0.000 0.004 0.020
#> GSM614426 3 0.1148 0.776 0.016 0.000 0.960 0.000 0.004 0.020
#> GSM614427 3 0.1148 0.776 0.016 0.000 0.960 0.004 0.000 0.020
#> GSM614428 3 0.1871 0.768 0.016 0.000 0.928 0.024 0.000 0.032
#> GSM614429 2 0.1363 0.749 0.004 0.952 0.028 0.012 0.000 0.004
#> GSM614430 2 0.1659 0.748 0.008 0.940 0.028 0.004 0.000 0.020
#> GSM614431 2 0.1930 0.744 0.028 0.924 0.036 0.000 0.000 0.012
#> GSM614432 2 0.2384 0.738 0.040 0.896 0.056 0.000 0.000 0.008
#> GSM614433 2 0.3325 0.705 0.096 0.820 0.084 0.000 0.000 0.000
#> GSM614434 2 0.1820 0.745 0.016 0.928 0.044 0.000 0.000 0.012
#> GSM614435 2 0.2893 0.731 0.000 0.872 0.028 0.056 0.000 0.044
#> GSM614436 2 0.5391 0.586 0.000 0.668 0.080 0.184 0.000 0.068
#> GSM614437 4 0.2243 0.880 0.000 0.112 0.000 0.880 0.004 0.004
#> GSM614438 4 0.0508 0.952 0.000 0.004 0.012 0.984 0.000 0.000
#> GSM614439 4 0.0508 0.952 0.000 0.004 0.012 0.984 0.000 0.000
#> GSM614440 4 0.0622 0.954 0.000 0.012 0.008 0.980 0.000 0.000
#> GSM614441 4 0.0508 0.952 0.000 0.004 0.012 0.984 0.000 0.000
#> GSM614442 4 0.0692 0.949 0.000 0.020 0.004 0.976 0.000 0.000
#> GSM614443 4 0.2149 0.888 0.000 0.104 0.000 0.888 0.004 0.004
#> GSM614444 4 0.0725 0.954 0.000 0.012 0.012 0.976 0.000 0.000
#> GSM614391 5 0.6295 0.579 0.176 0.000 0.072 0.000 0.568 0.184
#> GSM614392 5 0.6046 0.592 0.176 0.000 0.056 0.000 0.592 0.176
#> GSM614393 5 0.5824 0.600 0.172 0.000 0.044 0.000 0.612 0.172
#> GSM614394 5 0.6587 0.567 0.176 0.000 0.100 0.000 0.540 0.184
#> GSM614395 3 0.8644 -0.134 0.200 0.000 0.316 0.120 0.228 0.136
#> GSM614396 5 0.6624 0.563 0.176 0.000 0.104 0.000 0.536 0.184
#> GSM614397 5 0.8218 0.310 0.192 0.000 0.260 0.040 0.324 0.184
#> GSM614398 5 0.7277 0.474 0.176 0.000 0.188 0.000 0.436 0.200
#> GSM614399 1 0.4846 0.672 0.724 0.168 0.020 0.008 0.076 0.004
#> GSM614400 1 0.4507 0.689 0.736 0.116 0.004 0.008 0.136 0.000
#> GSM614401 1 0.4398 0.674 0.736 0.076 0.008 0.004 0.176 0.000
#> GSM614402 1 0.4440 0.700 0.756 0.144 0.028 0.004 0.068 0.000
#> GSM614403 1 0.3648 0.691 0.788 0.028 0.168 0.000 0.016 0.000
#> GSM614404 1 0.4496 0.696 0.744 0.128 0.008 0.008 0.112 0.000
#> GSM614405 1 0.4783 0.688 0.760 0.044 0.128 0.008 0.036 0.024
#> GSM614406 1 0.6509 0.142 0.376 0.000 0.232 0.372 0.008 0.012
#> GSM614407 6 0.2234 0.930 0.000 0.004 0.000 0.000 0.124 0.872
#> GSM614408 6 0.2624 0.909 0.004 0.004 0.000 0.000 0.148 0.844
#> GSM614409 6 0.2153 0.943 0.004 0.004 0.008 0.000 0.084 0.900
#> GSM614410 6 0.2191 0.933 0.000 0.004 0.000 0.000 0.120 0.876
#> GSM614411 6 0.2062 0.943 0.000 0.004 0.008 0.000 0.088 0.900
#> GSM614412 6 0.1829 0.936 0.000 0.004 0.012 0.000 0.064 0.920
#> GSM614413 6 0.1946 0.870 0.000 0.012 0.072 0.004 0.000 0.912
#> GSM614414 6 0.2036 0.910 0.000 0.008 0.048 0.000 0.028 0.916
#> GSM614445 1 0.4456 0.610 0.672 0.044 0.276 0.000 0.000 0.008
#> GSM614446 1 0.4585 0.517 0.624 0.028 0.336 0.004 0.000 0.008
#> GSM614447 1 0.4283 0.609 0.680 0.032 0.280 0.000 0.000 0.008
#> GSM614448 3 0.5022 0.498 0.232 0.004 0.664 0.088 0.000 0.012
#> GSM614449 3 0.4830 0.316 0.324 0.004 0.620 0.040 0.000 0.012
#> GSM614450 1 0.4932 0.254 0.516 0.004 0.436 0.036 0.000 0.008
#> GSM614451 3 0.4570 0.615 0.056 0.000 0.696 0.232 0.000 0.016
#> GSM614452 3 0.4469 0.634 0.060 0.000 0.716 0.208 0.000 0.016
#> GSM614453 2 0.1523 0.746 0.008 0.940 0.000 0.044 0.000 0.008
#> GSM614454 2 0.2101 0.739 0.008 0.908 0.000 0.072 0.004 0.008
#> GSM614455 2 0.1781 0.744 0.008 0.924 0.000 0.060 0.000 0.008
#> GSM614456 2 0.3323 0.604 0.000 0.752 0.000 0.240 0.000 0.008
#> GSM614457 2 0.3468 0.570 0.000 0.728 0.000 0.264 0.000 0.008
#> GSM614458 2 0.2001 0.731 0.004 0.900 0.000 0.092 0.000 0.004
#> GSM614459 2 0.3934 0.363 0.000 0.616 0.000 0.376 0.000 0.008
#> GSM614460 2 0.3161 0.624 0.000 0.776 0.000 0.216 0.000 0.008
#> GSM614461 2 0.3431 0.632 0.228 0.756 0.016 0.000 0.000 0.000
#> GSM614462 2 0.4105 0.475 0.348 0.632 0.020 0.000 0.000 0.000
#> GSM614463 2 0.3852 0.520 0.324 0.664 0.012 0.000 0.000 0.000
#> GSM614464 2 0.4666 0.358 0.388 0.564 0.048 0.000 0.000 0.000
#> GSM614465 2 0.4756 0.300 0.408 0.540 0.052 0.000 0.000 0.000
#> GSM614466 2 0.4206 0.456 0.356 0.620 0.024 0.000 0.000 0.000
#> GSM614467 2 0.5700 0.328 0.132 0.532 0.324 0.000 0.000 0.012
#> GSM614468 2 0.3822 0.671 0.128 0.776 0.096 0.000 0.000 0.000
#> GSM614469 5 0.2572 0.684 0.064 0.008 0.016 0.000 0.892 0.020
#> GSM614470 5 0.3178 0.671 0.104 0.008 0.016 0.000 0.848 0.024
#> GSM614471 5 0.4284 0.648 0.112 0.064 0.016 0.000 0.784 0.024
#> GSM614472 5 0.3552 0.651 0.128 0.020 0.016 0.000 0.820 0.016
#> GSM614473 5 0.2661 0.682 0.096 0.000 0.016 0.000 0.872 0.016
#> GSM614474 5 0.3478 0.680 0.080 0.024 0.020 0.000 0.844 0.032
#> GSM614475 5 0.7688 0.210 0.120 0.352 0.124 0.000 0.364 0.040
#> GSM614476 5 0.6682 0.533 0.108 0.036 0.224 0.008 0.576 0.048
Heatmaps for the consensus matrix. It visualizes the probability of two samples to be in a same group.
consensus_heatmap(res, k = 2)
consensus_heatmap(res, k = 3)
consensus_heatmap(res, k = 4)
consensus_heatmap(res, k = 5)
consensus_heatmap(res, k = 6)
Heatmaps for the membership of samples in all partitions to see how consistent they are:
membership_heatmap(res, k = 2)
membership_heatmap(res, k = 3)
membership_heatmap(res, k = 4)
membership_heatmap(res, k = 5)
membership_heatmap(res, k = 6)
As soon as we have had the classes for columns, we can look for signatures which are significantly different between classes which can be candidate marks for certain classes. Following are the heatmaps for signatures.
Signature heatmaps where rows are scaled:
get_signatures(res, k = 2)
get_signatures(res, k = 3)
get_signatures(res, k = 4)
get_signatures(res, k = 5)
get_signatures(res, k = 6)
Signature heatmaps where rows are not scaled:
get_signatures(res, k = 2, scale_rows = FALSE)
get_signatures(res, k = 3, scale_rows = FALSE)
get_signatures(res, k = 4, scale_rows = FALSE)
get_signatures(res, k = 5, scale_rows = FALSE)
get_signatures(res, k = 6, scale_rows = FALSE)
Compare the overlap of signatures from different k:
compare_signatures(res)
get_signature() returns a data frame invisibly. TO get the list of signatures, the function
call should be assigned to a variable explicitly. In following code, if plot argument is set
to FALSE, no heatmap is plotted while only the differential analysis is performed.
# code only for demonstration
tb = get_signature(res, k = ..., plot = FALSE)
An example of the output of tb is:
#> which_row fdr mean_1 mean_2 scaled_mean_1 scaled_mean_2 km
#> 1 38 0.042760348 8.373488 9.131774 -0.5533452 0.5164555 1
#> 2 40 0.018707592 7.106213 8.469186 -0.6173731 0.5762149 1
#> 3 55 0.019134737 10.221463 11.207825 -0.6159697 0.5749050 1
#> 4 59 0.006059896 5.921854 7.869574 -0.6899429 0.6439467 1
#> 5 60 0.018055526 8.928898 10.211722 -0.6204761 0.5791110 1
#> 6 98 0.009384629 15.714769 14.887706 0.6635654 -0.6193277 2
...
The columns in tb are:
which_row: row indices corresponding to the input matrix.fdr: FDR for the differential test. mean_x: The mean value in group x.scaled_mean_x: The mean value in group x after rows are scaled.km: Row groups if k-means clustering is applied to rows.UMAP plot which shows how samples are separated.
dimension_reduction(res, k = 2, method = "UMAP")
dimension_reduction(res, k = 3, method = "UMAP")
dimension_reduction(res, k = 4, method = "UMAP")
dimension_reduction(res, k = 5, method = "UMAP")
dimension_reduction(res, k = 6, method = "UMAP")
Following heatmap shows how subgroups are split when increasing k:
collect_classes(res)
Test correlation between subgroups and known annotations. If the known annotation is numeric, one-way ANOVA test is applied, and if the known annotation is discrete, chi-squared contingency table test is applied.
test_to_known_factors(res)
#> n individual(p) protocol(p) time(p) other(p) k
#> MAD:NMF 83 3.41e-11 0.359 0.870 0.847 2
#> MAD:NMF 79 4.52e-18 0.145 0.991 0.362 3
#> MAD:NMF 79 7.27e-28 0.088 1.000 0.300 4
#> MAD:NMF 60 3.63e-23 0.138 1.000 0.214 5
#> MAD:NMF 72 2.92e-45 0.317 1.000 0.107 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 21168 rows and 86 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 5.
#>
#> Following methods can be applied to this 'ConsensusPartition' object:
#> [1] "cola_report" "collect_classes" "collect_plots"
#> [4] "collect_stats" "colnames" "compare_signatures"
#> [7] "consensus_heatmap" "dimension_reduction" "functional_enrichment"
#> [10] "get_anno_col" "get_anno" "get_classes"
#> [13] "get_consensus" "get_matrix" "get_membership"
#> [16] "get_param" "get_signatures" "get_stats"
#> [19] "is_best_k" "is_stable_k" "membership_heatmap"
#> [22] "ncol" "nrow" "plot_ecdf"
#> [25] "rownames" "select_partition_number" "show"
#> [28] "suggest_best_k" "test_to_known_factors"
collect_plots() function collects all the plots made from res for all k (number of partitions)
into one single page to provide an easy and fast comparison between different k.
collect_plots(res)
The plots are:
k and the heatmap of
predicted classes for each k.k.k.k.All the plots in panels can be made by individual functions and they are plotted later in this section.
select_partition_number() produces several plots showing different
statistics for choosing “optimized” k. There are following statistics:
k;k, the area increased is defined as \(A_k - A_{k-1}\).The detailed explanations of these statistics can be found in the cola vignette.
Generally speaking, lower PAC score, higher mean silhouette score or higher
concordance corresponds to better partition. Rand index and Jaccard index
measure how similar the current partition is compared to partition with k-1.
If they are too similar, we won't accept k is better than k-1.
select_partition_number(res)
The numeric values for all these statistics can be obtained by get_stats().
get_stats(res)
#> k 1-PAC mean_silhouette concordance area_increased Rand Jaccard
#> 2 2 0.667 0.775 0.918 0.3361 0.665 0.665
#> 3 3 0.378 0.559 0.769 0.4704 0.788 0.697
#> 4 4 0.631 0.661 0.848 0.3137 0.690 0.483
#> 5 5 0.762 0.775 0.878 0.0830 0.914 0.768
#> 6 6 0.829 0.692 0.869 0.0335 0.975 0.916
suggest_best_k() suggests the best \(k\) based on these statistics. The rules are as follows:
suggest_best_k(res)
#> [1] 5
Following shows the table of the partitions (You need to click the show/hide
code output link to see it). The membership matrix (columns with name p*)
is inferred by
clue::cl_consensus()
function with the SE method. Basically the value in the membership matrix
represents the probability to belong to a certain group. The finall class
label for an item is determined with the group with highest probability it
belongs to.
In get_classes() function, the entropy is calculated from the membership
matrix and the silhouette score is calculated from the consensus matrix.
cbind(get_classes(res, k = 2), get_membership(res, k = 2))
#> class entropy silhouette p1 p2
#> GSM614415 2 0.000 0.924 0.000 1.000
#> GSM614416 2 0.000 0.924 0.000 1.000
#> GSM614417 2 0.000 0.924 0.000 1.000
#> GSM614418 2 0.000 0.924 0.000 1.000
#> GSM614419 2 0.000 0.924 0.000 1.000
#> GSM614420 2 0.000 0.924 0.000 1.000
#> GSM614421 2 0.997 -0.079 0.468 0.532
#> GSM614422 2 0.997 -0.079 0.468 0.532
#> GSM614423 2 0.552 0.782 0.128 0.872
#> GSM614424 2 0.997 -0.079 0.468 0.532
#> GSM614425 2 0.997 -0.079 0.468 0.532
#> GSM614426 2 0.997 -0.079 0.468 0.532
#> GSM614427 1 0.991 0.325 0.556 0.444
#> GSM614428 1 0.946 0.489 0.636 0.364
#> GSM614429 2 0.000 0.924 0.000 1.000
#> GSM614430 2 0.000 0.924 0.000 1.000
#> GSM614431 2 0.000 0.924 0.000 1.000
#> GSM614432 2 0.000 0.924 0.000 1.000
#> GSM614433 2 0.000 0.924 0.000 1.000
#> GSM614434 2 0.000 0.924 0.000 1.000
#> GSM614435 2 0.000 0.924 0.000 1.000
#> GSM614436 2 0.662 0.727 0.172 0.828
#> GSM614437 1 0.998 0.221 0.524 0.476
#> GSM614438 1 0.000 0.796 1.000 0.000
#> GSM614439 1 0.000 0.796 1.000 0.000
#> GSM614440 1 0.000 0.796 1.000 0.000
#> GSM614441 1 0.000 0.796 1.000 0.000
#> GSM614442 1 0.000 0.796 1.000 0.000
#> GSM614443 1 0.998 0.221 0.524 0.476
#> GSM614444 1 0.000 0.796 1.000 0.000
#> GSM614391 2 0.000 0.924 0.000 1.000
#> GSM614392 2 0.000 0.924 0.000 1.000
#> GSM614393 2 0.000 0.924 0.000 1.000
#> GSM614394 2 0.000 0.924 0.000 1.000
#> GSM614395 1 0.000 0.796 1.000 0.000
#> GSM614396 2 0.000 0.924 0.000 1.000
#> GSM614397 1 0.000 0.796 1.000 0.000
#> GSM614398 1 0.000 0.796 1.000 0.000
#> GSM614399 2 0.000 0.924 0.000 1.000
#> GSM614400 2 0.000 0.924 0.000 1.000
#> GSM614401 2 0.000 0.924 0.000 1.000
#> GSM614402 2 0.000 0.924 0.000 1.000
#> GSM614403 2 0.767 0.643 0.224 0.776
#> GSM614404 2 0.000 0.924 0.000 1.000
#> GSM614405 2 0.943 0.319 0.360 0.640
#> GSM614406 1 0.999 0.222 0.520 0.480
#> GSM614407 2 0.000 0.924 0.000 1.000
#> GSM614408 2 0.000 0.924 0.000 1.000
#> GSM614409 2 0.000 0.924 0.000 1.000
#> GSM614410 2 0.000 0.924 0.000 1.000
#> GSM614411 2 0.000 0.924 0.000 1.000
#> GSM614412 2 0.000 0.924 0.000 1.000
#> GSM614413 1 0.975 0.417 0.592 0.408
#> GSM614414 1 0.975 0.417 0.592 0.408
#> GSM614445 2 0.000 0.924 0.000 1.000
#> GSM614446 2 0.000 0.924 0.000 1.000
#> GSM614447 2 0.000 0.924 0.000 1.000
#> GSM614448 2 0.745 0.661 0.212 0.788
#> GSM614449 2 0.738 0.668 0.208 0.792
#> GSM614450 2 0.738 0.668 0.208 0.792
#> GSM614451 1 0.000 0.796 1.000 0.000
#> GSM614452 1 0.000 0.796 1.000 0.000
#> GSM614453 2 0.000 0.924 0.000 1.000
#> GSM614454 2 0.000 0.924 0.000 1.000
#> GSM614455 2 0.000 0.924 0.000 1.000
#> GSM614456 2 0.000 0.924 0.000 1.000
#> GSM614457 2 0.000 0.924 0.000 1.000
#> GSM614458 2 0.000 0.924 0.000 1.000
#> GSM614459 2 0.000 0.924 0.000 1.000
#> GSM614460 2 0.000 0.924 0.000 1.000
#> GSM614461 2 0.000 0.924 0.000 1.000
#> GSM614462 2 0.000 0.924 0.000 1.000
#> GSM614463 2 0.000 0.924 0.000 1.000
#> GSM614464 2 0.000 0.924 0.000 1.000
#> GSM614465 2 0.000 0.924 0.000 1.000
#> GSM614466 2 0.000 0.924 0.000 1.000
#> GSM614467 2 0.000 0.924 0.000 1.000
#> GSM614468 2 0.000 0.924 0.000 1.000
#> GSM614469 2 0.000 0.924 0.000 1.000
#> GSM614470 2 0.000 0.924 0.000 1.000
#> GSM614471 2 0.000 0.924 0.000 1.000
#> GSM614472 2 0.000 0.924 0.000 1.000
#> GSM614473 2 0.000 0.924 0.000 1.000
#> GSM614474 2 0.000 0.924 0.000 1.000
#> GSM614475 2 0.000 0.924 0.000 1.000
#> GSM614476 2 0.662 0.727 0.172 0.828
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM614415 2 0.6180 0.4792 0.000 0.584 0.416
#> GSM614416 2 0.6180 0.4792 0.000 0.584 0.416
#> GSM614417 2 0.6180 0.4792 0.000 0.584 0.416
#> GSM614418 2 0.6180 0.4792 0.000 0.584 0.416
#> GSM614419 2 0.6260 0.4332 0.000 0.552 0.448
#> GSM614420 2 0.6260 0.4332 0.000 0.552 0.448
#> GSM614421 3 0.9640 0.5557 0.252 0.280 0.468
#> GSM614422 3 0.9640 0.5557 0.252 0.280 0.468
#> GSM614423 2 0.6260 0.2854 0.000 0.552 0.448
#> GSM614424 3 0.9640 0.5557 0.252 0.280 0.468
#> GSM614425 3 0.9640 0.5557 0.252 0.280 0.468
#> GSM614426 3 0.9640 0.5557 0.252 0.280 0.468
#> GSM614427 3 0.9379 0.5144 0.288 0.208 0.504
#> GSM614428 3 0.9601 0.4388 0.364 0.204 0.432
#> GSM614429 2 0.0000 0.7826 0.000 1.000 0.000
#> GSM614430 2 0.0000 0.7826 0.000 1.000 0.000
#> GSM614431 2 0.0000 0.7826 0.000 1.000 0.000
#> GSM614432 2 0.0000 0.7826 0.000 1.000 0.000
#> GSM614433 2 0.0000 0.7826 0.000 1.000 0.000
#> GSM614434 2 0.0000 0.7826 0.000 1.000 0.000
#> GSM614435 2 0.0000 0.7826 0.000 1.000 0.000
#> GSM614436 2 0.6299 0.1809 0.000 0.524 0.476
#> GSM614437 3 0.4062 0.3525 0.000 0.164 0.836
#> GSM614438 3 0.6274 -0.2681 0.456 0.000 0.544
#> GSM614439 3 0.6274 -0.2681 0.456 0.000 0.544
#> GSM614440 3 0.6274 -0.2681 0.456 0.000 0.544
#> GSM614441 3 0.6274 -0.2681 0.456 0.000 0.544
#> GSM614442 3 0.6274 -0.2681 0.456 0.000 0.544
#> GSM614443 3 0.4062 0.3525 0.000 0.164 0.836
#> GSM614444 3 0.6274 -0.2681 0.456 0.000 0.544
#> GSM614391 2 0.6267 0.4265 0.000 0.548 0.452
#> GSM614392 2 0.6267 0.4265 0.000 0.548 0.452
#> GSM614393 2 0.6267 0.4265 0.000 0.548 0.452
#> GSM614394 2 0.6267 0.4265 0.000 0.548 0.452
#> GSM614395 1 0.0237 0.9891 0.996 0.000 0.004
#> GSM614396 2 0.6267 0.4265 0.000 0.548 0.452
#> GSM614397 1 0.0237 0.9891 0.996 0.000 0.004
#> GSM614398 1 0.0237 0.9891 0.996 0.000 0.004
#> GSM614399 2 0.0000 0.7826 0.000 1.000 0.000
#> GSM614400 2 0.0000 0.7826 0.000 1.000 0.000
#> GSM614401 2 0.0000 0.7826 0.000 1.000 0.000
#> GSM614402 2 0.0000 0.7826 0.000 1.000 0.000
#> GSM614403 3 0.7286 -0.0232 0.028 0.464 0.508
#> GSM614404 2 0.0000 0.7826 0.000 1.000 0.000
#> GSM614405 3 0.6211 0.4921 0.036 0.228 0.736
#> GSM614406 3 0.8950 0.5296 0.216 0.216 0.568
#> GSM614407 2 0.6045 0.5254 0.000 0.620 0.380
#> GSM614408 2 0.6045 0.5254 0.000 0.620 0.380
#> GSM614409 2 0.6045 0.5254 0.000 0.620 0.380
#> GSM614410 2 0.6045 0.5254 0.000 0.620 0.380
#> GSM614411 2 0.6045 0.5254 0.000 0.620 0.380
#> GSM614412 2 0.6140 0.4975 0.000 0.596 0.404
#> GSM614413 3 0.9531 0.4824 0.324 0.208 0.468
#> GSM614414 3 0.9531 0.4824 0.324 0.208 0.468
#> GSM614445 2 0.4555 0.6739 0.000 0.800 0.200
#> GSM614446 2 0.4555 0.6739 0.000 0.800 0.200
#> GSM614447 2 0.4555 0.6739 0.000 0.800 0.200
#> GSM614448 3 0.7263 0.1309 0.032 0.400 0.568
#> GSM614449 3 0.7366 0.0273 0.032 0.444 0.524
#> GSM614450 3 0.7366 0.0273 0.032 0.444 0.524
#> GSM614451 1 0.0892 0.9835 0.980 0.000 0.020
#> GSM614452 1 0.0892 0.9835 0.980 0.000 0.020
#> GSM614453 2 0.0000 0.7826 0.000 1.000 0.000
#> GSM614454 2 0.0000 0.7826 0.000 1.000 0.000
#> GSM614455 2 0.0000 0.7826 0.000 1.000 0.000
#> GSM614456 2 0.0000 0.7826 0.000 1.000 0.000
#> GSM614457 2 0.0000 0.7826 0.000 1.000 0.000
#> GSM614458 2 0.0000 0.7826 0.000 1.000 0.000
#> GSM614459 2 0.0000 0.7826 0.000 1.000 0.000
#> GSM614460 2 0.0000 0.7826 0.000 1.000 0.000
#> GSM614461 2 0.0000 0.7826 0.000 1.000 0.000
#> GSM614462 2 0.0000 0.7826 0.000 1.000 0.000
#> GSM614463 2 0.0000 0.7826 0.000 1.000 0.000
#> GSM614464 2 0.0000 0.7826 0.000 1.000 0.000
#> GSM614465 2 0.0000 0.7826 0.000 1.000 0.000
#> GSM614466 2 0.0000 0.7826 0.000 1.000 0.000
#> GSM614467 2 0.3551 0.7181 0.000 0.868 0.132
#> GSM614468 2 0.3551 0.7181 0.000 0.868 0.132
#> GSM614469 2 0.0000 0.7826 0.000 1.000 0.000
#> GSM614470 2 0.0000 0.7826 0.000 1.000 0.000
#> GSM614471 2 0.0000 0.7826 0.000 1.000 0.000
#> GSM614472 2 0.0000 0.7826 0.000 1.000 0.000
#> GSM614473 2 0.0000 0.7826 0.000 1.000 0.000
#> GSM614474 2 0.0000 0.7826 0.000 1.000 0.000
#> GSM614475 2 0.0000 0.7826 0.000 1.000 0.000
#> GSM614476 2 0.6062 0.3657 0.000 0.616 0.384
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM614415 1 0.1118 0.6896 0.964 0.036 0.000 0.000
#> GSM614416 1 0.1118 0.6896 0.964 0.036 0.000 0.000
#> GSM614417 1 0.1118 0.6896 0.964 0.036 0.000 0.000
#> GSM614418 1 0.1118 0.6896 0.964 0.036 0.000 0.000
#> GSM614419 1 0.0188 0.6876 0.996 0.004 0.000 0.000
#> GSM614420 1 0.0188 0.6876 0.996 0.004 0.000 0.000
#> GSM614421 1 0.7894 0.3733 0.496 0.016 0.200 0.288
#> GSM614422 1 0.7894 0.3733 0.496 0.016 0.200 0.288
#> GSM614423 1 0.6858 0.3966 0.588 0.284 0.004 0.124
#> GSM614424 1 0.7894 0.3733 0.496 0.016 0.200 0.288
#> GSM614425 1 0.7894 0.3733 0.496 0.016 0.200 0.288
#> GSM614426 1 0.7894 0.3733 0.496 0.016 0.200 0.288
#> GSM614427 1 0.7717 0.2175 0.424 0.000 0.232 0.344
#> GSM614428 4 0.7810 -0.1591 0.364 0.000 0.252 0.384
#> GSM614429 2 0.0000 0.9092 0.000 1.000 0.000 0.000
#> GSM614430 2 0.0000 0.9092 0.000 1.000 0.000 0.000
#> GSM614431 2 0.0000 0.9092 0.000 1.000 0.000 0.000
#> GSM614432 2 0.0000 0.9092 0.000 1.000 0.000 0.000
#> GSM614433 2 0.0000 0.9092 0.000 1.000 0.000 0.000
#> GSM614434 2 0.0000 0.9092 0.000 1.000 0.000 0.000
#> GSM614435 2 0.0000 0.9092 0.000 1.000 0.000 0.000
#> GSM614436 2 0.7465 0.2097 0.288 0.520 0.004 0.188
#> GSM614437 4 0.7274 0.2623 0.296 0.160 0.004 0.540
#> GSM614438 4 0.0592 0.6575 0.000 0.000 0.016 0.984
#> GSM614439 4 0.0592 0.6575 0.000 0.000 0.016 0.984
#> GSM614440 4 0.0592 0.6575 0.000 0.000 0.016 0.984
#> GSM614441 4 0.0592 0.6575 0.000 0.000 0.016 0.984
#> GSM614442 4 0.0592 0.6575 0.000 0.000 0.016 0.984
#> GSM614443 4 0.7274 0.2623 0.296 0.160 0.004 0.540
#> GSM614444 4 0.0592 0.6575 0.000 0.000 0.016 0.984
#> GSM614391 1 0.0000 0.6863 1.000 0.000 0.000 0.000
#> GSM614392 1 0.0000 0.6863 1.000 0.000 0.000 0.000
#> GSM614393 1 0.0000 0.6863 1.000 0.000 0.000 0.000
#> GSM614394 1 0.0000 0.6863 1.000 0.000 0.000 0.000
#> GSM614395 3 0.1474 0.8296 0.000 0.000 0.948 0.052
#> GSM614396 1 0.0000 0.6863 1.000 0.000 0.000 0.000
#> GSM614397 3 0.0469 0.8263 0.000 0.000 0.988 0.012
#> GSM614398 3 0.0469 0.8263 0.000 0.000 0.988 0.012
#> GSM614399 2 0.0000 0.9092 0.000 1.000 0.000 0.000
#> GSM614400 2 0.0000 0.9092 0.000 1.000 0.000 0.000
#> GSM614401 2 0.0000 0.9092 0.000 1.000 0.000 0.000
#> GSM614402 2 0.0000 0.9092 0.000 1.000 0.000 0.000
#> GSM614403 2 0.8092 0.0387 0.292 0.460 0.016 0.232
#> GSM614404 2 0.0000 0.9092 0.000 1.000 0.000 0.000
#> GSM614405 1 0.6164 0.3959 0.596 0.020 0.028 0.356
#> GSM614406 1 0.7258 0.1855 0.448 0.008 0.112 0.432
#> GSM614407 1 0.2530 0.6617 0.888 0.112 0.000 0.000
#> GSM614408 1 0.2530 0.6617 0.888 0.112 0.000 0.000
#> GSM614409 1 0.2530 0.6617 0.888 0.112 0.000 0.000
#> GSM614410 1 0.2530 0.6617 0.888 0.112 0.000 0.000
#> GSM614411 1 0.2530 0.6617 0.888 0.112 0.000 0.000
#> GSM614412 1 0.2149 0.6711 0.912 0.088 0.000 0.000
#> GSM614413 1 0.7836 0.1439 0.400 0.000 0.272 0.328
#> GSM614414 1 0.7836 0.1439 0.400 0.000 0.272 0.328
#> GSM614445 2 0.4998 0.0152 0.488 0.512 0.000 0.000
#> GSM614446 2 0.4998 0.0152 0.488 0.512 0.000 0.000
#> GSM614447 2 0.4998 0.0152 0.488 0.512 0.000 0.000
#> GSM614448 1 0.6732 0.4919 0.652 0.112 0.020 0.216
#> GSM614449 1 0.7195 0.4497 0.612 0.156 0.020 0.212
#> GSM614450 1 0.7195 0.4497 0.612 0.156 0.020 0.212
#> GSM614451 3 0.4564 0.7220 0.000 0.000 0.672 0.328
#> GSM614452 3 0.4564 0.7220 0.000 0.000 0.672 0.328
#> GSM614453 2 0.0000 0.9092 0.000 1.000 0.000 0.000
#> GSM614454 2 0.0000 0.9092 0.000 1.000 0.000 0.000
#> GSM614455 2 0.0000 0.9092 0.000 1.000 0.000 0.000
#> GSM614456 2 0.0000 0.9092 0.000 1.000 0.000 0.000
#> GSM614457 2 0.0000 0.9092 0.000 1.000 0.000 0.000
#> GSM614458 2 0.0000 0.9092 0.000 1.000 0.000 0.000
#> GSM614459 2 0.0000 0.9092 0.000 1.000 0.000 0.000
#> GSM614460 2 0.0000 0.9092 0.000 1.000 0.000 0.000
#> GSM614461 2 0.0000 0.9092 0.000 1.000 0.000 0.000
#> GSM614462 2 0.0000 0.9092 0.000 1.000 0.000 0.000
#> GSM614463 2 0.0000 0.9092 0.000 1.000 0.000 0.000
#> GSM614464 2 0.0000 0.9092 0.000 1.000 0.000 0.000
#> GSM614465 2 0.0000 0.9092 0.000 1.000 0.000 0.000
#> GSM614466 2 0.0000 0.9092 0.000 1.000 0.000 0.000
#> GSM614467 2 0.2814 0.7846 0.132 0.868 0.000 0.000
#> GSM614468 2 0.2814 0.7846 0.132 0.868 0.000 0.000
#> GSM614469 2 0.0000 0.9092 0.000 1.000 0.000 0.000
#> GSM614470 2 0.0000 0.9092 0.000 1.000 0.000 0.000
#> GSM614471 2 0.0000 0.9092 0.000 1.000 0.000 0.000
#> GSM614472 2 0.0000 0.9092 0.000 1.000 0.000 0.000
#> GSM614473 2 0.0000 0.9092 0.000 1.000 0.000 0.000
#> GSM614474 2 0.0000 0.9092 0.000 1.000 0.000 0.000
#> GSM614475 2 0.0000 0.9092 0.000 1.000 0.000 0.000
#> GSM614476 2 0.6861 0.4213 0.200 0.616 0.004 0.180
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM614415 5 0.0880 0.9306 0.000 0.032 0.000 0.000 0.968
#> GSM614416 5 0.0880 0.9306 0.000 0.032 0.000 0.000 0.968
#> GSM614417 5 0.0880 0.9306 0.000 0.032 0.000 0.000 0.968
#> GSM614418 5 0.0880 0.9306 0.000 0.032 0.000 0.000 0.968
#> GSM614419 5 0.0000 0.9229 0.000 0.000 0.000 0.000 1.000
#> GSM614420 5 0.0000 0.9229 0.000 0.000 0.000 0.000 1.000
#> GSM614421 3 0.5332 0.6946 0.184 0.008 0.720 0.032 0.056
#> GSM614422 3 0.5332 0.6946 0.184 0.008 0.720 0.032 0.056
#> GSM614423 3 0.5778 0.4194 0.000 0.272 0.596 0.000 0.132
#> GSM614424 3 0.5332 0.6946 0.184 0.008 0.720 0.032 0.056
#> GSM614425 3 0.5332 0.6946 0.184 0.008 0.720 0.032 0.056
#> GSM614426 3 0.5332 0.6946 0.184 0.008 0.720 0.032 0.056
#> GSM614427 3 0.4010 0.6556 0.208 0.000 0.760 0.032 0.000
#> GSM614428 3 0.5211 0.5732 0.232 0.000 0.668 0.100 0.000
#> GSM614429 2 0.0162 0.9235 0.000 0.996 0.000 0.000 0.004
#> GSM614430 2 0.0162 0.9235 0.000 0.996 0.000 0.000 0.004
#> GSM614431 2 0.0162 0.9235 0.000 0.996 0.000 0.000 0.004
#> GSM614432 2 0.0162 0.9235 0.000 0.996 0.000 0.000 0.004
#> GSM614433 2 0.0162 0.9235 0.000 0.996 0.000 0.000 0.004
#> GSM614434 2 0.0162 0.9235 0.000 0.996 0.000 0.000 0.004
#> GSM614435 2 0.0162 0.9235 0.000 0.996 0.000 0.000 0.004
#> GSM614436 2 0.4449 0.1131 0.000 0.512 0.484 0.000 0.004
#> GSM614437 4 0.6410 0.3172 0.000 0.152 0.376 0.468 0.004
#> GSM614438 4 0.0000 0.7668 0.000 0.000 0.000 1.000 0.000
#> GSM614439 4 0.0000 0.7668 0.000 0.000 0.000 1.000 0.000
#> GSM614440 4 0.0000 0.7668 0.000 0.000 0.000 1.000 0.000
#> GSM614441 4 0.0000 0.7668 0.000 0.000 0.000 1.000 0.000
#> GSM614442 4 0.0000 0.7668 0.000 0.000 0.000 1.000 0.000
#> GSM614443 4 0.6410 0.3172 0.000 0.152 0.376 0.468 0.004
#> GSM614444 4 0.0000 0.7668 0.000 0.000 0.000 1.000 0.000
#> GSM614391 5 0.0162 0.9222 0.000 0.000 0.004 0.000 0.996
#> GSM614392 5 0.0162 0.9222 0.000 0.000 0.004 0.000 0.996
#> GSM614393 5 0.0162 0.9222 0.000 0.000 0.004 0.000 0.996
#> GSM614394 5 0.0162 0.9222 0.000 0.000 0.004 0.000 0.996
#> GSM614395 1 0.1341 0.8075 0.944 0.000 0.000 0.056 0.000
#> GSM614396 5 0.0162 0.9222 0.000 0.000 0.004 0.000 0.996
#> GSM614397 1 0.0000 0.8043 1.000 0.000 0.000 0.000 0.000
#> GSM614398 1 0.0000 0.8043 1.000 0.000 0.000 0.000 0.000
#> GSM614399 2 0.0000 0.9253 0.000 1.000 0.000 0.000 0.000
#> GSM614400 2 0.0000 0.9253 0.000 1.000 0.000 0.000 0.000
#> GSM614401 2 0.0000 0.9253 0.000 1.000 0.000 0.000 0.000
#> GSM614402 2 0.0000 0.9253 0.000 1.000 0.000 0.000 0.000
#> GSM614403 3 0.4434 -0.0169 0.000 0.460 0.536 0.000 0.004
#> GSM614404 2 0.0000 0.9253 0.000 1.000 0.000 0.000 0.000
#> GSM614405 3 0.0566 0.6190 0.000 0.012 0.984 0.000 0.004
#> GSM614406 3 0.3590 0.6178 0.080 0.000 0.828 0.092 0.000
#> GSM614407 5 0.2127 0.8900 0.000 0.108 0.000 0.000 0.892
#> GSM614408 5 0.2127 0.8900 0.000 0.108 0.000 0.000 0.892
#> GSM614409 5 0.2127 0.8900 0.000 0.108 0.000 0.000 0.892
#> GSM614410 5 0.2127 0.8900 0.000 0.108 0.000 0.000 0.892
#> GSM614411 5 0.2127 0.8900 0.000 0.108 0.000 0.000 0.892
#> GSM614412 5 0.2077 0.8995 0.000 0.084 0.008 0.000 0.908
#> GSM614413 3 0.4566 0.6132 0.268 0.000 0.700 0.016 0.016
#> GSM614414 3 0.4566 0.6132 0.268 0.000 0.700 0.016 0.016
#> GSM614445 2 0.6235 0.0758 0.000 0.500 0.344 0.000 0.156
#> GSM614446 2 0.6235 0.0758 0.000 0.500 0.344 0.000 0.156
#> GSM614447 2 0.6235 0.0758 0.000 0.500 0.344 0.000 0.156
#> GSM614448 3 0.3862 0.5815 0.000 0.104 0.808 0.000 0.088
#> GSM614449 3 0.4364 0.5467 0.000 0.148 0.764 0.000 0.088
#> GSM614450 3 0.4364 0.5467 0.000 0.148 0.764 0.000 0.088
#> GSM614451 1 0.3966 0.6751 0.664 0.000 0.000 0.336 0.000
#> GSM614452 1 0.3966 0.6751 0.664 0.000 0.000 0.336 0.000
#> GSM614453 2 0.0000 0.9253 0.000 1.000 0.000 0.000 0.000
#> GSM614454 2 0.0000 0.9253 0.000 1.000 0.000 0.000 0.000
#> GSM614455 2 0.0000 0.9253 0.000 1.000 0.000 0.000 0.000
#> GSM614456 2 0.0000 0.9253 0.000 1.000 0.000 0.000 0.000
#> GSM614457 2 0.0000 0.9253 0.000 1.000 0.000 0.000 0.000
#> GSM614458 2 0.0162 0.9235 0.000 0.996 0.000 0.000 0.004
#> GSM614459 2 0.0000 0.9253 0.000 1.000 0.000 0.000 0.000
#> GSM614460 2 0.0000 0.9253 0.000 1.000 0.000 0.000 0.000
#> GSM614461 2 0.0000 0.9253 0.000 1.000 0.000 0.000 0.000
#> GSM614462 2 0.0000 0.9253 0.000 1.000 0.000 0.000 0.000
#> GSM614463 2 0.0000 0.9253 0.000 1.000 0.000 0.000 0.000
#> GSM614464 2 0.0000 0.9253 0.000 1.000 0.000 0.000 0.000
#> GSM614465 2 0.0000 0.9253 0.000 1.000 0.000 0.000 0.000
#> GSM614466 2 0.0000 0.9253 0.000 1.000 0.000 0.000 0.000
#> GSM614467 2 0.3055 0.8032 0.000 0.864 0.064 0.000 0.072
#> GSM614468 2 0.3055 0.8032 0.000 0.864 0.064 0.000 0.072
#> GSM614469 2 0.0000 0.9253 0.000 1.000 0.000 0.000 0.000
#> GSM614470 2 0.0000 0.9253 0.000 1.000 0.000 0.000 0.000
#> GSM614471 2 0.0000 0.9253 0.000 1.000 0.000 0.000 0.000
#> GSM614472 2 0.0000 0.9253 0.000 1.000 0.000 0.000 0.000
#> GSM614473 2 0.0000 0.9253 0.000 1.000 0.000 0.000 0.000
#> GSM614474 2 0.0000 0.9253 0.000 1.000 0.000 0.000 0.000
#> GSM614475 2 0.0000 0.9253 0.000 1.000 0.000 0.000 0.000
#> GSM614476 2 0.5052 0.4152 0.000 0.612 0.340 0.000 0.048
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM614415 5 0.0790 0.8924 0.000 0.032 0.000 0.000 0.968 0.000
#> GSM614416 5 0.0790 0.8924 0.000 0.032 0.000 0.000 0.968 0.000
#> GSM614417 5 0.0790 0.8924 0.000 0.032 0.000 0.000 0.968 0.000
#> GSM614418 5 0.0790 0.8924 0.000 0.032 0.000 0.000 0.968 0.000
#> GSM614419 5 0.0000 0.8772 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM614420 5 0.0000 0.8772 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM614421 3 0.0000 0.7433 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM614422 3 0.0000 0.7433 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM614423 3 0.5563 0.0948 0.060 0.244 0.624 0.000 0.072 0.000
#> GSM614424 3 0.0000 0.7433 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM614425 3 0.0000 0.7433 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM614426 3 0.0000 0.7433 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM614427 3 0.2618 0.7141 0.116 0.000 0.860 0.000 0.000 0.024
#> GSM614428 3 0.4347 0.6578 0.120 0.000 0.768 0.064 0.000 0.048
#> GSM614429 2 0.0405 0.8609 0.008 0.988 0.000 0.000 0.004 0.000
#> GSM614430 2 0.0405 0.8609 0.008 0.988 0.000 0.000 0.004 0.000
#> GSM614431 2 0.0405 0.8609 0.008 0.988 0.000 0.000 0.004 0.000
#> GSM614432 2 0.0405 0.8609 0.008 0.988 0.000 0.000 0.004 0.000
#> GSM614433 2 0.0405 0.8609 0.008 0.988 0.000 0.000 0.004 0.000
#> GSM614434 2 0.0405 0.8609 0.008 0.988 0.000 0.000 0.004 0.000
#> GSM614435 2 0.0405 0.8609 0.008 0.988 0.000 0.000 0.004 0.000
#> GSM614436 1 0.5799 0.0000 0.448 0.368 0.184 0.000 0.000 0.000
#> GSM614437 4 0.5514 0.3761 0.424 0.008 0.100 0.468 0.000 0.000
#> GSM614438 4 0.0000 0.7941 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM614439 4 0.0000 0.7941 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM614440 4 0.0000 0.7941 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM614441 4 0.0000 0.7941 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM614442 4 0.0000 0.7941 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM614443 4 0.5514 0.3761 0.424 0.008 0.100 0.468 0.000 0.000
#> GSM614444 4 0.0000 0.7941 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM614391 5 0.2260 0.8390 0.140 0.000 0.000 0.000 0.860 0.000
#> GSM614392 5 0.2260 0.8390 0.140 0.000 0.000 0.000 0.860 0.000
#> GSM614393 5 0.2260 0.8390 0.140 0.000 0.000 0.000 0.860 0.000
#> GSM614394 5 0.2260 0.8390 0.140 0.000 0.000 0.000 0.860 0.000
#> GSM614395 6 0.1285 0.7282 0.004 0.000 0.000 0.052 0.000 0.944
#> GSM614396 5 0.2260 0.8390 0.140 0.000 0.000 0.000 0.860 0.000
#> GSM614397 6 0.0260 0.7308 0.000 0.000 0.008 0.000 0.000 0.992
#> GSM614398 6 0.0260 0.7308 0.000 0.000 0.008 0.000 0.000 0.992
#> GSM614399 2 0.0000 0.8694 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM614400 2 0.0000 0.8694 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM614401 2 0.0000 0.8694 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM614402 2 0.0000 0.8694 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM614403 2 0.5829 -0.7891 0.360 0.448 0.192 0.000 0.000 0.000
#> GSM614404 2 0.0000 0.8694 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM614405 3 0.3797 0.5608 0.420 0.000 0.580 0.000 0.000 0.000
#> GSM614406 3 0.5107 0.6233 0.296 0.000 0.620 0.060 0.000 0.024
#> GSM614407 5 0.1910 0.8615 0.000 0.108 0.000 0.000 0.892 0.000
#> GSM614408 5 0.1910 0.8615 0.000 0.108 0.000 0.000 0.892 0.000
#> GSM614409 5 0.1910 0.8615 0.000 0.108 0.000 0.000 0.892 0.000
#> GSM614410 5 0.1910 0.8615 0.000 0.108 0.000 0.000 0.892 0.000
#> GSM614411 5 0.1910 0.8615 0.000 0.108 0.000 0.000 0.892 0.000
#> GSM614412 5 0.1866 0.8706 0.000 0.084 0.008 0.000 0.908 0.000
#> GSM614413 3 0.4023 0.6810 0.124 0.000 0.780 0.000 0.016 0.080
#> GSM614414 3 0.4023 0.6810 0.124 0.000 0.780 0.000 0.016 0.080
#> GSM614445 2 0.6821 -0.4569 0.172 0.424 0.332 0.000 0.072 0.000
#> GSM614446 2 0.6821 -0.4569 0.172 0.424 0.332 0.000 0.072 0.000
#> GSM614447 2 0.6821 -0.4569 0.172 0.424 0.332 0.000 0.072 0.000
#> GSM614448 3 0.4524 0.4535 0.336 0.048 0.616 0.000 0.000 0.000
#> GSM614449 3 0.5054 0.3437 0.336 0.092 0.572 0.000 0.000 0.000
#> GSM614450 3 0.5054 0.3437 0.336 0.092 0.572 0.000 0.000 0.000
#> GSM614451 6 0.5982 0.5093 0.240 0.000 0.000 0.332 0.000 0.428
#> GSM614452 6 0.5982 0.5093 0.240 0.000 0.000 0.332 0.000 0.428
#> GSM614453 2 0.0000 0.8694 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM614454 2 0.0000 0.8694 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM614455 2 0.0000 0.8694 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM614456 2 0.0000 0.8694 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM614457 2 0.0000 0.8694 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM614458 2 0.0146 0.8662 0.000 0.996 0.000 0.000 0.004 0.000
#> GSM614459 2 0.0000 0.8694 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM614460 2 0.0000 0.8694 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM614461 2 0.0000 0.8694 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM614462 2 0.0000 0.8694 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM614463 2 0.0000 0.8694 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM614464 2 0.0000 0.8694 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM614465 2 0.0000 0.8694 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM614466 2 0.0000 0.8694 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM614467 2 0.3508 0.6019 0.064 0.832 0.032 0.000 0.072 0.000
#> GSM614468 2 0.3508 0.6019 0.064 0.832 0.032 0.000 0.072 0.000
#> GSM614469 2 0.0000 0.8694 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM614470 2 0.0000 0.8694 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM614471 2 0.0000 0.8694 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM614472 2 0.0000 0.8694 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM614473 2 0.0000 0.8694 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM614474 2 0.0000 0.8694 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM614475 2 0.0000 0.8694 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM614476 2 0.5993 -0.4171 0.236 0.580 0.136 0.000 0.048 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 individual(p) protocol(p) time(p) other(p) k
#> ATC:hclust 73 1.10e-06 0.029778 0.908 1.96e-03 2
#> ATC:hclust 55 6.76e-12 0.002925 0.959 1.53e-03 3
#> ATC:hclust 63 1.49e-20 0.000652 0.967 4.97e-04 4
#> ATC:hclust 77 1.26e-26 0.000909 0.996 2.80e-05 5
#> ATC:hclust 74 1.12e-28 0.000281 0.991 5.64e-05 6
If matrix rows can be associated to genes, consider to use functional_enrichment(res,
...) to perform function enrichment for the signature genes. See this vignette for more detailed explanations.
The object with results only for a single top-value method and a single partition method can be extracted as:
res = res_list["ATC", "kmeans"]
# you can also extract it by
# res = res_list["ATC:kmeans"]
A summary of res and all the functions that can be applied to it:
res
#> A 'ConsensusPartition' object with k = 2, 3, 4, 5, 6.
#> On a matrix with 21168 rows and 86 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 1.000 0.993 0.993 0.4596 0.540 0.540
#> 3 3 0.595 0.628 0.829 0.3084 0.810 0.663
#> 4 4 0.567 0.509 0.726 0.1519 0.856 0.666
#> 5 5 0.589 0.386 0.639 0.0903 0.869 0.655
#> 6 6 0.631 0.467 0.675 0.0574 0.801 0.436
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
#> GSM614415 2 0.1184 0.989 0.016 0.984
#> GSM614416 2 0.1184 0.989 0.016 0.984
#> GSM614417 2 0.1184 0.989 0.016 0.984
#> GSM614418 2 0.1184 0.989 0.016 0.984
#> GSM614419 1 0.0000 0.989 1.000 0.000
#> GSM614420 1 0.0000 0.989 1.000 0.000
#> GSM614421 1 0.1184 0.995 0.984 0.016
#> GSM614422 1 0.1184 0.995 0.984 0.016
#> GSM614423 2 0.0000 0.996 0.000 1.000
#> GSM614424 1 0.1184 0.995 0.984 0.016
#> GSM614425 1 0.1184 0.995 0.984 0.016
#> GSM614426 1 0.1184 0.995 0.984 0.016
#> GSM614427 1 0.1184 0.995 0.984 0.016
#> GSM614428 1 0.1184 0.995 0.984 0.016
#> GSM614429 2 0.0000 0.996 0.000 1.000
#> GSM614430 2 0.0000 0.996 0.000 1.000
#> GSM614431 2 0.0000 0.996 0.000 1.000
#> GSM614432 2 0.0000 0.996 0.000 1.000
#> GSM614433 2 0.0000 0.996 0.000 1.000
#> GSM614434 2 0.0000 0.996 0.000 1.000
#> GSM614435 2 0.0000 0.996 0.000 1.000
#> GSM614436 2 0.0000 0.996 0.000 1.000
#> GSM614437 2 0.0000 0.996 0.000 1.000
#> GSM614438 1 0.1184 0.995 0.984 0.016
#> GSM614439 1 0.1184 0.995 0.984 0.016
#> GSM614440 1 0.1184 0.995 0.984 0.016
#> GSM614441 1 0.1184 0.995 0.984 0.016
#> GSM614442 1 0.1184 0.995 0.984 0.016
#> GSM614443 1 0.1184 0.995 0.984 0.016
#> GSM614444 1 0.1184 0.995 0.984 0.016
#> GSM614391 1 0.0000 0.989 1.000 0.000
#> GSM614392 2 0.1184 0.989 0.016 0.984
#> GSM614393 2 0.1184 0.989 0.016 0.984
#> GSM614394 1 0.0000 0.989 1.000 0.000
#> GSM614395 1 0.0000 0.989 1.000 0.000
#> GSM614396 1 0.0000 0.989 1.000 0.000
#> GSM614397 1 0.0000 0.989 1.000 0.000
#> GSM614398 1 0.0000 0.989 1.000 0.000
#> GSM614399 2 0.0000 0.996 0.000 1.000
#> GSM614400 2 0.0000 0.996 0.000 1.000
#> GSM614401 2 0.0000 0.996 0.000 1.000
#> GSM614402 2 0.0000 0.996 0.000 1.000
#> GSM614403 2 0.0000 0.996 0.000 1.000
#> GSM614404 2 0.0000 0.996 0.000 1.000
#> GSM614405 1 0.1184 0.995 0.984 0.016
#> GSM614406 1 0.1184 0.995 0.984 0.016
#> GSM614407 2 0.1184 0.989 0.016 0.984
#> GSM614408 2 0.1184 0.989 0.016 0.984
#> GSM614409 2 0.1184 0.989 0.016 0.984
#> GSM614410 2 0.1184 0.989 0.016 0.984
#> GSM614411 2 0.1184 0.989 0.016 0.984
#> GSM614412 2 0.1184 0.989 0.016 0.984
#> GSM614413 1 0.0000 0.989 1.000 0.000
#> GSM614414 1 0.0000 0.989 1.000 0.000
#> GSM614445 2 0.0000 0.996 0.000 1.000
#> GSM614446 2 0.0000 0.996 0.000 1.000
#> GSM614447 2 0.0000 0.996 0.000 1.000
#> GSM614448 1 0.1184 0.995 0.984 0.016
#> GSM614449 1 0.1184 0.995 0.984 0.016
#> GSM614450 2 0.0000 0.996 0.000 1.000
#> GSM614451 1 0.1184 0.995 0.984 0.016
#> GSM614452 1 0.1184 0.995 0.984 0.016
#> GSM614453 2 0.0000 0.996 0.000 1.000
#> GSM614454 2 0.0000 0.996 0.000 1.000
#> GSM614455 2 0.0000 0.996 0.000 1.000
#> GSM614456 2 0.0000 0.996 0.000 1.000
#> GSM614457 2 0.0000 0.996 0.000 1.000
#> GSM614458 2 0.0000 0.996 0.000 1.000
#> GSM614459 2 0.0000 0.996 0.000 1.000
#> GSM614460 2 0.0000 0.996 0.000 1.000
#> GSM614461 2 0.0000 0.996 0.000 1.000
#> GSM614462 2 0.0000 0.996 0.000 1.000
#> GSM614463 2 0.0000 0.996 0.000 1.000
#> GSM614464 2 0.0000 0.996 0.000 1.000
#> GSM614465 2 0.0000 0.996 0.000 1.000
#> GSM614466 2 0.0000 0.996 0.000 1.000
#> GSM614467 2 0.0000 0.996 0.000 1.000
#> GSM614468 2 0.0000 0.996 0.000 1.000
#> GSM614469 2 0.0672 0.993 0.008 0.992
#> GSM614470 2 0.0672 0.993 0.008 0.992
#> GSM614471 2 0.0672 0.993 0.008 0.992
#> GSM614472 2 0.0672 0.993 0.008 0.992
#> GSM614473 2 0.0672 0.993 0.008 0.992
#> GSM614474 2 0.0672 0.993 0.008 0.992
#> GSM614475 2 0.0672 0.993 0.008 0.992
#> GSM614476 2 0.0376 0.994 0.004 0.996
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM614415 1 0.5926 0.3755 0.644 0.356 0.000
#> GSM614416 1 0.5988 0.3551 0.632 0.368 0.000
#> GSM614417 1 0.5988 0.3551 0.632 0.368 0.000
#> GSM614418 1 0.5988 0.3551 0.632 0.368 0.000
#> GSM614419 1 0.0237 0.5251 0.996 0.000 0.004
#> GSM614420 1 0.0237 0.5251 0.996 0.000 0.004
#> GSM614421 3 0.6307 0.5275 0.488 0.000 0.512
#> GSM614422 3 0.6305 0.5350 0.484 0.000 0.516
#> GSM614423 2 0.6291 0.1783 0.468 0.532 0.000
#> GSM614424 3 0.6305 0.5350 0.484 0.000 0.516
#> GSM614425 3 0.6305 0.5350 0.484 0.000 0.516
#> GSM614426 3 0.6305 0.5350 0.484 0.000 0.516
#> GSM614427 3 0.6305 0.5350 0.484 0.000 0.516
#> GSM614428 3 0.5058 0.6900 0.244 0.000 0.756
#> GSM614429 2 0.0000 0.8883 0.000 1.000 0.000
#> GSM614430 2 0.0000 0.8883 0.000 1.000 0.000
#> GSM614431 2 0.0000 0.8883 0.000 1.000 0.000
#> GSM614432 2 0.0000 0.8883 0.000 1.000 0.000
#> GSM614433 2 0.0000 0.8883 0.000 1.000 0.000
#> GSM614434 2 0.0000 0.8883 0.000 1.000 0.000
#> GSM614435 2 0.0000 0.8883 0.000 1.000 0.000
#> GSM614436 2 0.3573 0.7663 0.120 0.876 0.004
#> GSM614437 2 0.3816 0.7448 0.000 0.852 0.148
#> GSM614438 3 0.0000 0.6938 0.000 0.000 1.000
#> GSM614439 3 0.0000 0.6938 0.000 0.000 1.000
#> GSM614440 3 0.0000 0.6938 0.000 0.000 1.000
#> GSM614441 3 0.0000 0.6938 0.000 0.000 1.000
#> GSM614442 3 0.0000 0.6938 0.000 0.000 1.000
#> GSM614443 3 0.5237 0.6323 0.120 0.056 0.824
#> GSM614444 3 0.0000 0.6938 0.000 0.000 1.000
#> GSM614391 1 0.0000 0.5250 1.000 0.000 0.000
#> GSM614392 1 0.0237 0.5281 0.996 0.004 0.000
#> GSM614393 1 0.0237 0.5281 0.996 0.004 0.000
#> GSM614394 1 0.1964 0.4590 0.944 0.000 0.056
#> GSM614395 3 0.5431 0.6768 0.284 0.000 0.716
#> GSM614396 1 0.1964 0.4590 0.944 0.000 0.056
#> GSM614397 3 0.6252 0.5502 0.444 0.000 0.556
#> GSM614398 3 0.6260 0.5455 0.448 0.000 0.552
#> GSM614399 2 0.1411 0.8794 0.036 0.964 0.000
#> GSM614400 2 0.1411 0.8794 0.036 0.964 0.000
#> GSM614401 2 0.1411 0.8794 0.036 0.964 0.000
#> GSM614402 2 0.1411 0.8794 0.036 0.964 0.000
#> GSM614403 2 0.5956 0.5003 0.324 0.672 0.004
#> GSM614404 2 0.1411 0.8794 0.036 0.964 0.000
#> GSM614405 1 0.7152 -0.4199 0.532 0.024 0.444
#> GSM614406 3 0.3267 0.7125 0.116 0.000 0.884
#> GSM614407 2 0.6309 0.0467 0.496 0.504 0.000
#> GSM614408 2 0.6309 0.0467 0.496 0.504 0.000
#> GSM614409 1 0.6045 0.3264 0.620 0.380 0.000
#> GSM614410 2 0.6309 0.0467 0.496 0.504 0.000
#> GSM614411 1 0.6280 0.0654 0.540 0.460 0.000
#> GSM614412 1 0.1289 0.5304 0.968 0.032 0.000
#> GSM614413 1 0.6274 -0.4888 0.544 0.000 0.456
#> GSM614414 1 0.6260 -0.4724 0.552 0.000 0.448
#> GSM614445 2 0.1411 0.8794 0.036 0.964 0.000
#> GSM614446 2 0.2625 0.8532 0.084 0.916 0.000
#> GSM614447 2 0.1411 0.8794 0.036 0.964 0.000
#> GSM614448 3 0.6225 0.5870 0.432 0.000 0.568
#> GSM614449 1 0.8644 -0.2987 0.496 0.104 0.400
#> GSM614450 2 0.5845 0.5329 0.308 0.688 0.004
#> GSM614451 3 0.2625 0.7117 0.084 0.000 0.916
#> GSM614452 3 0.2625 0.7117 0.084 0.000 0.916
#> GSM614453 2 0.0000 0.8883 0.000 1.000 0.000
#> GSM614454 2 0.0000 0.8883 0.000 1.000 0.000
#> GSM614455 2 0.0000 0.8883 0.000 1.000 0.000
#> GSM614456 2 0.0000 0.8883 0.000 1.000 0.000
#> GSM614457 2 0.0000 0.8883 0.000 1.000 0.000
#> GSM614458 2 0.0000 0.8883 0.000 1.000 0.000
#> GSM614459 2 0.0000 0.8883 0.000 1.000 0.000
#> GSM614460 2 0.0000 0.8883 0.000 1.000 0.000
#> GSM614461 2 0.0000 0.8883 0.000 1.000 0.000
#> GSM614462 2 0.0000 0.8883 0.000 1.000 0.000
#> GSM614463 2 0.0000 0.8883 0.000 1.000 0.000
#> GSM614464 2 0.0000 0.8883 0.000 1.000 0.000
#> GSM614465 2 0.0000 0.8883 0.000 1.000 0.000
#> GSM614466 2 0.0000 0.8883 0.000 1.000 0.000
#> GSM614467 2 0.0000 0.8883 0.000 1.000 0.000
#> GSM614468 2 0.0000 0.8883 0.000 1.000 0.000
#> GSM614469 2 0.3267 0.8312 0.116 0.884 0.000
#> GSM614470 2 0.3267 0.8312 0.116 0.884 0.000
#> GSM614471 2 0.3267 0.8312 0.116 0.884 0.000
#> GSM614472 2 0.3267 0.8312 0.116 0.884 0.000
#> GSM614473 2 0.3267 0.8312 0.116 0.884 0.000
#> GSM614474 2 0.3267 0.8312 0.116 0.884 0.000
#> GSM614475 2 0.3267 0.8312 0.116 0.884 0.000
#> GSM614476 2 0.5968 0.4760 0.364 0.636 0.000
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM614415 1 0.6387 0.5681 0.492 0.064 0.444 0.000
#> GSM614416 1 0.6387 0.5681 0.492 0.064 0.444 0.000
#> GSM614417 1 0.6387 0.5681 0.492 0.064 0.444 0.000
#> GSM614418 1 0.6387 0.5681 0.492 0.064 0.444 0.000
#> GSM614419 3 0.4040 -0.0356 0.248 0.000 0.752 0.000
#> GSM614420 3 0.4040 -0.0356 0.248 0.000 0.752 0.000
#> GSM614421 3 0.7806 0.3369 0.284 0.000 0.420 0.296
#> GSM614422 3 0.7806 0.3369 0.284 0.000 0.420 0.296
#> GSM614423 1 0.7031 0.1406 0.536 0.324 0.140 0.000
#> GSM614424 3 0.7806 0.3369 0.284 0.000 0.420 0.296
#> GSM614425 3 0.7806 0.3369 0.284 0.000 0.420 0.296
#> GSM614426 3 0.7806 0.3369 0.284 0.000 0.420 0.296
#> GSM614427 3 0.7806 0.3369 0.284 0.000 0.420 0.296
#> GSM614428 4 0.7644 0.1676 0.260 0.000 0.272 0.468
#> GSM614429 2 0.0000 0.8085 0.000 1.000 0.000 0.000
#> GSM614430 2 0.0000 0.8085 0.000 1.000 0.000 0.000
#> GSM614431 2 0.0000 0.8085 0.000 1.000 0.000 0.000
#> GSM614432 2 0.0000 0.8085 0.000 1.000 0.000 0.000
#> GSM614433 2 0.0000 0.8085 0.000 1.000 0.000 0.000
#> GSM614434 2 0.0000 0.8085 0.000 1.000 0.000 0.000
#> GSM614435 2 0.1004 0.8018 0.024 0.972 0.004 0.000
#> GSM614436 2 0.6003 0.5513 0.156 0.724 0.100 0.020
#> GSM614437 2 0.6610 0.4354 0.100 0.604 0.004 0.292
#> GSM614438 4 0.0000 0.7268 0.000 0.000 0.000 1.000
#> GSM614439 4 0.0000 0.7268 0.000 0.000 0.000 1.000
#> GSM614440 4 0.0000 0.7268 0.000 0.000 0.000 1.000
#> GSM614441 4 0.0000 0.7268 0.000 0.000 0.000 1.000
#> GSM614442 4 0.0000 0.7268 0.000 0.000 0.000 1.000
#> GSM614443 4 0.6575 0.4527 0.140 0.052 0.104 0.704
#> GSM614444 4 0.0000 0.7268 0.000 0.000 0.000 1.000
#> GSM614391 3 0.3219 0.0253 0.164 0.000 0.836 0.000
#> GSM614392 3 0.4585 -0.2986 0.332 0.000 0.668 0.000
#> GSM614393 3 0.4585 -0.2986 0.332 0.000 0.668 0.000
#> GSM614394 3 0.0469 0.2896 0.000 0.000 0.988 0.012
#> GSM614395 4 0.7475 0.1928 0.176 0.000 0.404 0.420
#> GSM614396 3 0.0592 0.2946 0.000 0.000 0.984 0.016
#> GSM614397 3 0.7103 -0.0306 0.160 0.000 0.544 0.296
#> GSM614398 3 0.7086 -0.0211 0.160 0.000 0.548 0.292
#> GSM614399 2 0.3726 0.7274 0.212 0.788 0.000 0.000
#> GSM614400 2 0.3726 0.7274 0.212 0.788 0.000 0.000
#> GSM614401 2 0.3726 0.7274 0.212 0.788 0.000 0.000
#> GSM614402 2 0.3688 0.7276 0.208 0.792 0.000 0.000
#> GSM614403 2 0.7746 0.0917 0.416 0.424 0.144 0.016
#> GSM614404 2 0.3726 0.7274 0.212 0.788 0.000 0.000
#> GSM614405 1 0.8497 -0.4703 0.424 0.036 0.328 0.212
#> GSM614406 4 0.7146 0.3284 0.212 0.000 0.228 0.560
#> GSM614407 1 0.6928 0.6132 0.556 0.136 0.308 0.000
#> GSM614408 1 0.6928 0.6132 0.556 0.136 0.308 0.000
#> GSM614409 1 0.6599 0.6134 0.564 0.096 0.340 0.000
#> GSM614410 1 0.6928 0.6132 0.556 0.136 0.308 0.000
#> GSM614411 1 0.6634 0.6152 0.564 0.100 0.336 0.000
#> GSM614412 1 0.5378 0.4316 0.540 0.012 0.448 0.000
#> GSM614413 3 0.7540 0.3601 0.304 0.000 0.480 0.216
#> GSM614414 3 0.7517 0.3622 0.304 0.000 0.484 0.212
#> GSM614445 2 0.3172 0.7542 0.160 0.840 0.000 0.000
#> GSM614446 2 0.5389 0.5723 0.308 0.660 0.032 0.000
#> GSM614447 2 0.3444 0.7479 0.184 0.816 0.000 0.000
#> GSM614448 3 0.7883 0.2918 0.316 0.000 0.384 0.300
#> GSM614449 1 0.9283 -0.4042 0.404 0.108 0.288 0.200
#> GSM614450 2 0.7889 0.1543 0.372 0.460 0.144 0.024
#> GSM614451 4 0.5637 0.5929 0.168 0.000 0.112 0.720
#> GSM614452 4 0.5637 0.5929 0.168 0.000 0.112 0.720
#> GSM614453 2 0.1743 0.7952 0.056 0.940 0.004 0.000
#> GSM614454 2 0.1743 0.7952 0.056 0.940 0.004 0.000
#> GSM614455 2 0.1743 0.7952 0.056 0.940 0.004 0.000
#> GSM614456 2 0.1743 0.7952 0.056 0.940 0.004 0.000
#> GSM614457 2 0.1743 0.7952 0.056 0.940 0.004 0.000
#> GSM614458 2 0.1743 0.7952 0.056 0.940 0.004 0.000
#> GSM614459 2 0.1743 0.7952 0.056 0.940 0.004 0.000
#> GSM614460 2 0.1743 0.7952 0.056 0.940 0.004 0.000
#> GSM614461 2 0.0707 0.8093 0.020 0.980 0.000 0.000
#> GSM614462 2 0.0707 0.8093 0.020 0.980 0.000 0.000
#> GSM614463 2 0.0707 0.8093 0.020 0.980 0.000 0.000
#> GSM614464 2 0.0707 0.8093 0.020 0.980 0.000 0.000
#> GSM614465 2 0.0707 0.8093 0.020 0.980 0.000 0.000
#> GSM614466 2 0.0707 0.8093 0.020 0.980 0.000 0.000
#> GSM614467 2 0.1302 0.8020 0.044 0.956 0.000 0.000
#> GSM614468 2 0.0707 0.8093 0.020 0.980 0.000 0.000
#> GSM614469 2 0.5730 0.5218 0.344 0.616 0.040 0.000
#> GSM614470 2 0.5730 0.5218 0.344 0.616 0.040 0.000
#> GSM614471 2 0.5730 0.5218 0.344 0.616 0.040 0.000
#> GSM614472 2 0.5730 0.5218 0.344 0.616 0.040 0.000
#> GSM614473 2 0.5730 0.5218 0.344 0.616 0.040 0.000
#> GSM614474 2 0.5713 0.5282 0.340 0.620 0.040 0.000
#> GSM614475 2 0.5713 0.5282 0.340 0.620 0.040 0.000
#> GSM614476 1 0.7384 0.0497 0.476 0.352 0.172 0.000
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM614415 1 0.0932 0.71900 0.972 0.004 0.000 0.004 0.020
#> GSM614416 1 0.1173 0.72177 0.964 0.012 0.000 0.004 0.020
#> GSM614417 1 0.1173 0.72177 0.964 0.012 0.000 0.004 0.020
#> GSM614418 1 0.1173 0.72177 0.964 0.012 0.000 0.004 0.020
#> GSM614419 1 0.4861 0.52319 0.732 0.000 0.072 0.012 0.184
#> GSM614420 1 0.4861 0.52319 0.732 0.000 0.072 0.012 0.184
#> GSM614421 3 0.4610 0.23863 0.016 0.000 0.596 0.000 0.388
#> GSM614422 3 0.4610 0.23863 0.016 0.000 0.596 0.000 0.388
#> GSM614423 3 0.7779 -0.14028 0.088 0.148 0.408 0.352 0.004
#> GSM614424 3 0.4610 0.23863 0.016 0.000 0.596 0.000 0.388
#> GSM614425 3 0.4610 0.23863 0.016 0.000 0.596 0.000 0.388
#> GSM614426 3 0.4610 0.23863 0.016 0.000 0.596 0.000 0.388
#> GSM614427 3 0.4610 0.23863 0.016 0.000 0.596 0.000 0.388
#> GSM614428 5 0.4798 0.05236 0.000 0.000 0.396 0.024 0.580
#> GSM614429 2 0.0404 0.66192 0.000 0.988 0.000 0.012 0.000
#> GSM614430 2 0.0404 0.66192 0.000 0.988 0.000 0.012 0.000
#> GSM614431 2 0.0290 0.66451 0.000 0.992 0.000 0.008 0.000
#> GSM614432 2 0.0290 0.66451 0.000 0.992 0.000 0.008 0.000
#> GSM614433 2 0.0162 0.66366 0.000 0.996 0.000 0.004 0.000
#> GSM614434 2 0.0404 0.66192 0.000 0.988 0.000 0.012 0.000
#> GSM614435 2 0.1124 0.65344 0.000 0.960 0.000 0.036 0.004
#> GSM614436 2 0.5531 0.27814 0.000 0.664 0.164 0.168 0.004
#> GSM614437 2 0.7275 -0.19159 0.000 0.400 0.192 0.372 0.036
#> GSM614438 3 0.5686 -0.01855 0.000 0.000 0.552 0.356 0.092
#> GSM614439 3 0.5686 -0.01855 0.000 0.000 0.552 0.356 0.092
#> GSM614440 3 0.5686 -0.01855 0.000 0.000 0.552 0.356 0.092
#> GSM614441 3 0.5686 -0.01855 0.000 0.000 0.552 0.356 0.092
#> GSM614442 3 0.5686 -0.01855 0.000 0.000 0.552 0.356 0.092
#> GSM614443 4 0.6319 -0.20995 0.000 0.048 0.428 0.472 0.052
#> GSM614444 3 0.5686 -0.01855 0.000 0.000 0.552 0.356 0.092
#> GSM614391 1 0.5304 0.27370 0.548 0.000 0.036 0.008 0.408
#> GSM614392 1 0.4796 0.49889 0.664 0.000 0.028 0.008 0.300
#> GSM614393 1 0.4796 0.49889 0.664 0.000 0.028 0.008 0.300
#> GSM614394 5 0.5353 0.16775 0.368 0.000 0.052 0.004 0.576
#> GSM614395 5 0.3848 0.45001 0.000 0.000 0.172 0.040 0.788
#> GSM614396 5 0.5353 0.16775 0.368 0.000 0.052 0.004 0.576
#> GSM614397 5 0.2504 0.50217 0.064 0.000 0.040 0.000 0.896
#> GSM614398 5 0.2504 0.50217 0.064 0.000 0.040 0.000 0.896
#> GSM614399 2 0.6016 0.45720 0.140 0.548 0.000 0.312 0.000
#> GSM614400 2 0.6016 0.45720 0.140 0.548 0.000 0.312 0.000
#> GSM614401 2 0.6016 0.45720 0.140 0.548 0.000 0.312 0.000
#> GSM614402 2 0.6068 0.43808 0.140 0.532 0.000 0.328 0.000
#> GSM614403 3 0.6950 -0.06598 0.040 0.128 0.452 0.380 0.000
#> GSM614404 2 0.6016 0.45720 0.140 0.548 0.000 0.312 0.000
#> GSM614405 3 0.7320 0.22217 0.020 0.040 0.504 0.308 0.128
#> GSM614406 3 0.5009 -0.08357 0.000 0.000 0.540 0.032 0.428
#> GSM614407 1 0.4775 0.64166 0.756 0.048 0.004 0.168 0.024
#> GSM614408 1 0.4775 0.64166 0.756 0.048 0.004 0.168 0.024
#> GSM614409 1 0.4483 0.65759 0.776 0.028 0.008 0.164 0.024
#> GSM614410 1 0.4775 0.64166 0.756 0.048 0.004 0.168 0.024
#> GSM614411 1 0.4775 0.64166 0.756 0.048 0.004 0.168 0.024
#> GSM614412 1 0.6127 0.58516 0.660 0.000 0.144 0.144 0.052
#> GSM614413 3 0.6068 0.10952 0.056 0.000 0.468 0.028 0.448
#> GSM614414 3 0.6068 0.10952 0.056 0.000 0.468 0.028 0.448
#> GSM614445 2 0.4756 0.53808 0.044 0.668 0.000 0.288 0.000
#> GSM614446 2 0.7515 -0.00774 0.060 0.416 0.180 0.344 0.000
#> GSM614447 2 0.5506 0.40545 0.048 0.572 0.012 0.368 0.000
#> GSM614448 3 0.6062 0.25068 0.012 0.000 0.600 0.132 0.256
#> GSM614449 3 0.7459 0.16615 0.024 0.080 0.508 0.308 0.080
#> GSM614450 3 0.7519 -0.03481 0.036 0.168 0.452 0.328 0.016
#> GSM614451 5 0.6273 0.22139 0.000 0.000 0.416 0.148 0.436
#> GSM614452 5 0.6273 0.22139 0.000 0.000 0.416 0.148 0.436
#> GSM614453 2 0.2824 0.63203 0.000 0.872 0.000 0.096 0.032
#> GSM614454 2 0.2769 0.63096 0.000 0.876 0.000 0.092 0.032
#> GSM614455 2 0.2824 0.63203 0.000 0.872 0.000 0.096 0.032
#> GSM614456 2 0.2769 0.63096 0.000 0.876 0.000 0.092 0.032
#> GSM614457 2 0.2769 0.63096 0.000 0.876 0.000 0.092 0.032
#> GSM614458 2 0.2712 0.63192 0.000 0.880 0.000 0.088 0.032
#> GSM614459 2 0.2769 0.63096 0.000 0.876 0.000 0.092 0.032
#> GSM614460 2 0.2769 0.63096 0.000 0.876 0.000 0.092 0.032
#> GSM614461 2 0.2672 0.66082 0.004 0.872 0.000 0.116 0.008
#> GSM614462 2 0.2672 0.66082 0.004 0.872 0.000 0.116 0.008
#> GSM614463 2 0.2672 0.66082 0.004 0.872 0.000 0.116 0.008
#> GSM614464 2 0.2621 0.66105 0.004 0.876 0.000 0.112 0.008
#> GSM614465 2 0.2672 0.66082 0.004 0.872 0.000 0.116 0.008
#> GSM614466 2 0.2672 0.66082 0.004 0.872 0.000 0.116 0.008
#> GSM614467 2 0.3675 0.57971 0.004 0.772 0.000 0.216 0.008
#> GSM614468 2 0.2722 0.65435 0.004 0.868 0.000 0.120 0.008
#> GSM614469 2 0.6683 0.31104 0.308 0.432 0.000 0.260 0.000
#> GSM614470 2 0.6683 0.31104 0.308 0.432 0.000 0.260 0.000
#> GSM614471 2 0.6683 0.31104 0.308 0.432 0.000 0.260 0.000
#> GSM614472 2 0.6683 0.31104 0.308 0.432 0.000 0.260 0.000
#> GSM614473 2 0.6683 0.31104 0.308 0.432 0.000 0.260 0.000
#> GSM614474 2 0.6683 0.31104 0.308 0.432 0.000 0.260 0.000
#> GSM614475 2 0.6674 0.31581 0.304 0.436 0.000 0.260 0.000
#> GSM614476 4 0.8600 -0.07479 0.224 0.212 0.216 0.344 0.004
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM614415 6 0.4197 0.3259 0.052 0.000 0.000 0.020 0.176 0.752
#> GSM614416 6 0.4197 0.3259 0.052 0.000 0.000 0.020 0.176 0.752
#> GSM614417 6 0.4197 0.3259 0.052 0.000 0.000 0.020 0.176 0.752
#> GSM614418 6 0.4197 0.3259 0.052 0.000 0.000 0.020 0.176 0.752
#> GSM614419 6 0.7011 -0.3032 0.064 0.000 0.136 0.020 0.352 0.428
#> GSM614420 6 0.7011 -0.3032 0.064 0.000 0.136 0.020 0.352 0.428
#> GSM614421 3 0.0146 0.8146 0.000 0.000 0.996 0.000 0.000 0.004
#> GSM614422 3 0.0146 0.8146 0.000 0.000 0.996 0.000 0.000 0.004
#> GSM614423 1 0.5992 0.2781 0.456 0.052 0.424 0.000 0.004 0.064
#> GSM614424 3 0.0603 0.8070 0.016 0.000 0.980 0.000 0.000 0.004
#> GSM614425 3 0.0146 0.8146 0.000 0.000 0.996 0.000 0.000 0.004
#> GSM614426 3 0.0146 0.8146 0.000 0.000 0.996 0.000 0.000 0.004
#> GSM614427 3 0.0146 0.8146 0.000 0.000 0.996 0.000 0.000 0.004
#> GSM614428 3 0.4532 0.5917 0.100 0.000 0.748 0.032 0.120 0.000
#> GSM614429 2 0.0653 0.7205 0.004 0.980 0.000 0.004 0.012 0.000
#> GSM614430 2 0.0653 0.7205 0.004 0.980 0.000 0.004 0.012 0.000
#> GSM614431 2 0.0508 0.7215 0.012 0.984 0.000 0.004 0.000 0.000
#> GSM614432 2 0.0508 0.7215 0.012 0.984 0.000 0.004 0.000 0.000
#> GSM614433 2 0.0653 0.7215 0.012 0.980 0.000 0.004 0.004 0.000
#> GSM614434 2 0.0436 0.7215 0.004 0.988 0.000 0.004 0.004 0.000
#> GSM614435 2 0.1562 0.7133 0.032 0.940 0.000 0.004 0.024 0.000
#> GSM614436 2 0.5488 0.3177 0.220 0.644 0.096 0.004 0.036 0.000
#> GSM614437 4 0.7156 0.1993 0.228 0.248 0.020 0.448 0.056 0.000
#> GSM614438 4 0.1556 0.7559 0.000 0.000 0.080 0.920 0.000 0.000
#> GSM614439 4 0.1556 0.7559 0.000 0.000 0.080 0.920 0.000 0.000
#> GSM614440 4 0.1556 0.7559 0.000 0.000 0.080 0.920 0.000 0.000
#> GSM614441 4 0.1556 0.7559 0.000 0.000 0.080 0.920 0.000 0.000
#> GSM614442 4 0.1556 0.7559 0.000 0.000 0.080 0.920 0.000 0.000
#> GSM614443 4 0.5276 0.5716 0.160 0.020 0.072 0.704 0.044 0.000
#> GSM614444 4 0.1556 0.7559 0.000 0.000 0.080 0.920 0.000 0.000
#> GSM614391 5 0.4923 0.5801 0.004 0.000 0.108 0.000 0.652 0.236
#> GSM614392 5 0.4864 0.4543 0.008 0.000 0.040 0.004 0.584 0.364
#> GSM614393 5 0.4864 0.4543 0.008 0.000 0.040 0.004 0.584 0.364
#> GSM614394 5 0.4712 0.6451 0.008 0.000 0.212 0.000 0.688 0.092
#> GSM614395 5 0.6833 0.3283 0.132 0.000 0.208 0.152 0.508 0.000
#> GSM614396 5 0.4712 0.6451 0.008 0.000 0.212 0.000 0.688 0.092
#> GSM614397 5 0.5553 0.4747 0.104 0.000 0.268 0.028 0.600 0.000
#> GSM614398 5 0.5553 0.4747 0.104 0.000 0.268 0.028 0.600 0.000
#> GSM614399 1 0.6724 0.0347 0.396 0.368 0.000 0.000 0.056 0.180
#> GSM614400 2 0.6727 -0.1330 0.380 0.384 0.000 0.000 0.056 0.180
#> GSM614401 2 0.6727 -0.1330 0.380 0.384 0.000 0.000 0.056 0.180
#> GSM614402 1 0.6676 0.1182 0.432 0.336 0.000 0.000 0.056 0.176
#> GSM614403 1 0.5389 0.2722 0.516 0.032 0.412 0.000 0.008 0.032
#> GSM614404 2 0.6727 -0.1330 0.380 0.384 0.000 0.000 0.056 0.180
#> GSM614405 1 0.5014 0.0846 0.484 0.008 0.468 0.000 0.012 0.028
#> GSM614406 3 0.4265 0.6784 0.072 0.000 0.780 0.092 0.056 0.000
#> GSM614407 6 0.1096 0.4630 0.020 0.008 0.004 0.004 0.000 0.964
#> GSM614408 6 0.1096 0.4630 0.020 0.008 0.004 0.004 0.000 0.964
#> GSM614409 6 0.1096 0.4536 0.020 0.000 0.004 0.004 0.008 0.964
#> GSM614410 6 0.1096 0.4630 0.020 0.008 0.004 0.004 0.000 0.964
#> GSM614411 6 0.0982 0.4605 0.020 0.004 0.004 0.004 0.000 0.968
#> GSM614412 6 0.4233 0.2701 0.080 0.000 0.108 0.004 0.028 0.780
#> GSM614413 3 0.4248 0.6799 0.064 0.000 0.788 0.004 0.056 0.088
#> GSM614414 3 0.4248 0.6799 0.064 0.000 0.788 0.004 0.056 0.088
#> GSM614445 2 0.5613 -0.0686 0.448 0.468 0.012 0.000 0.028 0.044
#> GSM614446 1 0.6289 0.4738 0.528 0.256 0.180 0.000 0.004 0.032
#> GSM614447 1 0.5426 0.2366 0.556 0.364 0.036 0.000 0.008 0.036
#> GSM614448 3 0.3370 0.5706 0.212 0.000 0.772 0.000 0.012 0.004
#> GSM614449 3 0.5299 -0.1922 0.432 0.032 0.504 0.000 0.012 0.020
#> GSM614450 1 0.5671 0.2486 0.476 0.060 0.432 0.000 0.012 0.020
#> GSM614451 4 0.7126 0.1682 0.128 0.000 0.344 0.384 0.144 0.000
#> GSM614452 4 0.7126 0.1682 0.128 0.000 0.344 0.384 0.144 0.000
#> GSM614453 2 0.3909 0.6780 0.116 0.796 0.000 0.028 0.060 0.000
#> GSM614454 2 0.3966 0.6775 0.116 0.792 0.000 0.028 0.064 0.000
#> GSM614455 2 0.3909 0.6780 0.116 0.796 0.000 0.028 0.060 0.000
#> GSM614456 2 0.3966 0.6775 0.116 0.792 0.000 0.028 0.064 0.000
#> GSM614457 2 0.4118 0.6727 0.120 0.780 0.000 0.028 0.072 0.000
#> GSM614458 2 0.4159 0.6713 0.124 0.776 0.000 0.028 0.072 0.000
#> GSM614459 2 0.4118 0.6727 0.120 0.780 0.000 0.028 0.072 0.000
#> GSM614460 2 0.3966 0.6775 0.116 0.792 0.000 0.028 0.064 0.000
#> GSM614461 2 0.2983 0.6742 0.136 0.832 0.000 0.000 0.032 0.000
#> GSM614462 2 0.2983 0.6742 0.136 0.832 0.000 0.000 0.032 0.000
#> GSM614463 2 0.2983 0.6742 0.136 0.832 0.000 0.000 0.032 0.000
#> GSM614464 2 0.2750 0.6797 0.136 0.844 0.000 0.000 0.020 0.000
#> GSM614465 2 0.2983 0.6742 0.136 0.832 0.000 0.000 0.032 0.000
#> GSM614466 2 0.2983 0.6742 0.136 0.832 0.000 0.000 0.032 0.000
#> GSM614467 2 0.4157 0.5171 0.276 0.688 0.000 0.004 0.032 0.000
#> GSM614468 2 0.3590 0.6363 0.188 0.776 0.000 0.004 0.032 0.000
#> GSM614469 6 0.7115 0.2697 0.184 0.316 0.000 0.008 0.076 0.416
#> GSM614470 6 0.7115 0.2697 0.184 0.316 0.000 0.008 0.076 0.416
#> GSM614471 6 0.7115 0.2697 0.184 0.316 0.000 0.008 0.076 0.416
#> GSM614472 6 0.7115 0.2697 0.184 0.316 0.000 0.008 0.076 0.416
#> GSM614473 6 0.7115 0.2697 0.184 0.316 0.000 0.008 0.076 0.416
#> GSM614474 6 0.7158 0.2668 0.188 0.308 0.000 0.008 0.080 0.416
#> GSM614475 6 0.7158 0.2668 0.188 0.308 0.000 0.008 0.080 0.416
#> GSM614476 1 0.8244 0.1677 0.372 0.156 0.160 0.008 0.044 0.260
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 individual(p) protocol(p) time(p) other(p) k
#> ATC:kmeans 86 3.56e-06 0.01542 0.845 0.0109 2
#> ATC:kmeans 69 2.03e-12 0.02503 0.959 0.0777 3
#> ATC:kmeans 56 1.54e-14 0.13126 0.981 0.1328 4
#> ATC:kmeans 38 2.38e-11 0.00294 0.971 0.0140 5
#> ATC:kmeans 44 6.73e-17 0.34496 0.997 0.0447 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 21168 rows and 86 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 0.978 0.991 0.4961 0.504 0.504
#> 3 3 0.963 0.957 0.979 0.2976 0.808 0.635
#> 4 4 0.722 0.701 0.847 0.1159 0.909 0.757
#> 5 5 0.728 0.726 0.810 0.0787 0.889 0.647
#> 6 6 0.714 0.706 0.807 0.0490 0.964 0.834
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
#> GSM614415 2 0.0000 0.993 0.000 1.000
#> GSM614416 2 0.0000 0.993 0.000 1.000
#> GSM614417 2 0.0000 0.993 0.000 1.000
#> GSM614418 2 0.0000 0.993 0.000 1.000
#> GSM614419 1 0.0000 0.987 1.000 0.000
#> GSM614420 1 0.0000 0.987 1.000 0.000
#> GSM614421 1 0.0000 0.987 1.000 0.000
#> GSM614422 1 0.0000 0.987 1.000 0.000
#> GSM614423 2 0.9358 0.450 0.352 0.648
#> GSM614424 1 0.0000 0.987 1.000 0.000
#> GSM614425 1 0.0000 0.987 1.000 0.000
#> GSM614426 1 0.0000 0.987 1.000 0.000
#> GSM614427 1 0.0000 0.987 1.000 0.000
#> GSM614428 1 0.0000 0.987 1.000 0.000
#> GSM614429 2 0.0000 0.993 0.000 1.000
#> GSM614430 2 0.0000 0.993 0.000 1.000
#> GSM614431 2 0.0000 0.993 0.000 1.000
#> GSM614432 2 0.0000 0.993 0.000 1.000
#> GSM614433 2 0.0000 0.993 0.000 1.000
#> GSM614434 2 0.0000 0.993 0.000 1.000
#> GSM614435 2 0.0000 0.993 0.000 1.000
#> GSM614436 1 0.8081 0.674 0.752 0.248
#> GSM614437 2 0.0000 0.993 0.000 1.000
#> GSM614438 1 0.0000 0.987 1.000 0.000
#> GSM614439 1 0.0000 0.987 1.000 0.000
#> GSM614440 1 0.0000 0.987 1.000 0.000
#> GSM614441 1 0.0000 0.987 1.000 0.000
#> GSM614442 1 0.0000 0.987 1.000 0.000
#> GSM614443 1 0.0000 0.987 1.000 0.000
#> GSM614444 1 0.0000 0.987 1.000 0.000
#> GSM614391 1 0.0000 0.987 1.000 0.000
#> GSM614392 1 0.0376 0.984 0.996 0.004
#> GSM614393 1 0.4431 0.896 0.908 0.092
#> GSM614394 1 0.0000 0.987 1.000 0.000
#> GSM614395 1 0.0000 0.987 1.000 0.000
#> GSM614396 1 0.0000 0.987 1.000 0.000
#> GSM614397 1 0.0000 0.987 1.000 0.000
#> GSM614398 1 0.0000 0.987 1.000 0.000
#> GSM614399 2 0.0000 0.993 0.000 1.000
#> GSM614400 2 0.0000 0.993 0.000 1.000
#> GSM614401 2 0.0000 0.993 0.000 1.000
#> GSM614402 2 0.0000 0.993 0.000 1.000
#> GSM614403 1 0.0000 0.987 1.000 0.000
#> GSM614404 2 0.0000 0.993 0.000 1.000
#> GSM614405 1 0.0000 0.987 1.000 0.000
#> GSM614406 1 0.0000 0.987 1.000 0.000
#> GSM614407 2 0.0000 0.993 0.000 1.000
#> GSM614408 2 0.0000 0.993 0.000 1.000
#> GSM614409 2 0.0000 0.993 0.000 1.000
#> GSM614410 2 0.0000 0.993 0.000 1.000
#> GSM614411 2 0.0000 0.993 0.000 1.000
#> GSM614412 1 0.0000 0.987 1.000 0.000
#> GSM614413 1 0.0000 0.987 1.000 0.000
#> GSM614414 1 0.0000 0.987 1.000 0.000
#> GSM614445 2 0.0000 0.993 0.000 1.000
#> GSM614446 2 0.0000 0.993 0.000 1.000
#> GSM614447 2 0.0000 0.993 0.000 1.000
#> GSM614448 1 0.0000 0.987 1.000 0.000
#> GSM614449 1 0.0000 0.987 1.000 0.000
#> GSM614450 1 0.0000 0.987 1.000 0.000
#> GSM614451 1 0.0000 0.987 1.000 0.000
#> GSM614452 1 0.0000 0.987 1.000 0.000
#> GSM614453 2 0.0000 0.993 0.000 1.000
#> GSM614454 2 0.0000 0.993 0.000 1.000
#> GSM614455 2 0.0000 0.993 0.000 1.000
#> GSM614456 2 0.0000 0.993 0.000 1.000
#> GSM614457 2 0.0000 0.993 0.000 1.000
#> GSM614458 2 0.0000 0.993 0.000 1.000
#> GSM614459 2 0.0000 0.993 0.000 1.000
#> GSM614460 2 0.0000 0.993 0.000 1.000
#> GSM614461 2 0.0000 0.993 0.000 1.000
#> GSM614462 2 0.0000 0.993 0.000 1.000
#> GSM614463 2 0.0000 0.993 0.000 1.000
#> GSM614464 2 0.0000 0.993 0.000 1.000
#> GSM614465 2 0.0000 0.993 0.000 1.000
#> GSM614466 2 0.0000 0.993 0.000 1.000
#> GSM614467 2 0.0000 0.993 0.000 1.000
#> GSM614468 2 0.0000 0.993 0.000 1.000
#> GSM614469 2 0.0000 0.993 0.000 1.000
#> GSM614470 2 0.0000 0.993 0.000 1.000
#> GSM614471 2 0.0000 0.993 0.000 1.000
#> GSM614472 2 0.0000 0.993 0.000 1.000
#> GSM614473 2 0.0000 0.993 0.000 1.000
#> GSM614474 2 0.0000 0.993 0.000 1.000
#> GSM614475 2 0.0000 0.993 0.000 1.000
#> GSM614476 1 0.4939 0.879 0.892 0.108
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM614415 1 0.000 0.961 1.000 0.000 0.000
#> GSM614416 1 0.000 0.961 1.000 0.000 0.000
#> GSM614417 1 0.000 0.961 1.000 0.000 0.000
#> GSM614418 1 0.000 0.961 1.000 0.000 0.000
#> GSM614419 1 0.263 0.900 0.916 0.000 0.084
#> GSM614420 1 0.254 0.904 0.920 0.000 0.080
#> GSM614421 3 0.000 0.968 0.000 0.000 1.000
#> GSM614422 3 0.000 0.968 0.000 0.000 1.000
#> GSM614423 1 0.000 0.961 1.000 0.000 0.000
#> GSM614424 3 0.000 0.968 0.000 0.000 1.000
#> GSM614425 3 0.000 0.968 0.000 0.000 1.000
#> GSM614426 3 0.000 0.968 0.000 0.000 1.000
#> GSM614427 3 0.000 0.968 0.000 0.000 1.000
#> GSM614428 3 0.000 0.968 0.000 0.000 1.000
#> GSM614429 2 0.000 0.989 0.000 1.000 0.000
#> GSM614430 2 0.000 0.989 0.000 1.000 0.000
#> GSM614431 2 0.000 0.989 0.000 1.000 0.000
#> GSM614432 2 0.000 0.989 0.000 1.000 0.000
#> GSM614433 2 0.000 0.989 0.000 1.000 0.000
#> GSM614434 2 0.000 0.989 0.000 1.000 0.000
#> GSM614435 2 0.000 0.989 0.000 1.000 0.000
#> GSM614436 3 0.186 0.918 0.000 0.052 0.948
#> GSM614437 2 0.304 0.880 0.000 0.896 0.104
#> GSM614438 3 0.000 0.968 0.000 0.000 1.000
#> GSM614439 3 0.000 0.968 0.000 0.000 1.000
#> GSM614440 3 0.000 0.968 0.000 0.000 1.000
#> GSM614441 3 0.000 0.968 0.000 0.000 1.000
#> GSM614442 3 0.000 0.968 0.000 0.000 1.000
#> GSM614443 3 0.000 0.968 0.000 0.000 1.000
#> GSM614444 3 0.000 0.968 0.000 0.000 1.000
#> GSM614391 1 0.000 0.961 1.000 0.000 0.000
#> GSM614392 1 0.000 0.961 1.000 0.000 0.000
#> GSM614393 1 0.000 0.961 1.000 0.000 0.000
#> GSM614394 1 0.493 0.721 0.768 0.000 0.232
#> GSM614395 3 0.000 0.968 0.000 0.000 1.000
#> GSM614396 1 0.493 0.721 0.768 0.000 0.232
#> GSM614397 3 0.296 0.884 0.100 0.000 0.900
#> GSM614398 3 0.296 0.884 0.100 0.000 0.900
#> GSM614399 2 0.000 0.989 0.000 1.000 0.000
#> GSM614400 2 0.000 0.989 0.000 1.000 0.000
#> GSM614401 2 0.000 0.989 0.000 1.000 0.000
#> GSM614402 2 0.000 0.989 0.000 1.000 0.000
#> GSM614403 3 0.000 0.968 0.000 0.000 1.000
#> GSM614404 2 0.000 0.989 0.000 1.000 0.000
#> GSM614405 3 0.000 0.968 0.000 0.000 1.000
#> GSM614406 3 0.000 0.968 0.000 0.000 1.000
#> GSM614407 1 0.000 0.961 1.000 0.000 0.000
#> GSM614408 1 0.000 0.961 1.000 0.000 0.000
#> GSM614409 1 0.000 0.961 1.000 0.000 0.000
#> GSM614410 1 0.000 0.961 1.000 0.000 0.000
#> GSM614411 1 0.000 0.961 1.000 0.000 0.000
#> GSM614412 1 0.000 0.961 1.000 0.000 0.000
#> GSM614413 3 0.296 0.884 0.100 0.000 0.900
#> GSM614414 3 0.304 0.880 0.104 0.000 0.896
#> GSM614445 2 0.000 0.989 0.000 1.000 0.000
#> GSM614446 2 0.000 0.989 0.000 1.000 0.000
#> GSM614447 2 0.000 0.989 0.000 1.000 0.000
#> GSM614448 3 0.000 0.968 0.000 0.000 1.000
#> GSM614449 3 0.000 0.968 0.000 0.000 1.000
#> GSM614450 3 0.000 0.968 0.000 0.000 1.000
#> GSM614451 3 0.000 0.968 0.000 0.000 1.000
#> GSM614452 3 0.000 0.968 0.000 0.000 1.000
#> GSM614453 2 0.000 0.989 0.000 1.000 0.000
#> GSM614454 2 0.000 0.989 0.000 1.000 0.000
#> GSM614455 2 0.000 0.989 0.000 1.000 0.000
#> GSM614456 2 0.000 0.989 0.000 1.000 0.000
#> GSM614457 2 0.000 0.989 0.000 1.000 0.000
#> GSM614458 2 0.000 0.989 0.000 1.000 0.000
#> GSM614459 2 0.000 0.989 0.000 1.000 0.000
#> GSM614460 2 0.000 0.989 0.000 1.000 0.000
#> GSM614461 2 0.000 0.989 0.000 1.000 0.000
#> GSM614462 2 0.000 0.989 0.000 1.000 0.000
#> GSM614463 2 0.000 0.989 0.000 1.000 0.000
#> GSM614464 2 0.000 0.989 0.000 1.000 0.000
#> GSM614465 2 0.000 0.989 0.000 1.000 0.000
#> GSM614466 2 0.000 0.989 0.000 1.000 0.000
#> GSM614467 2 0.000 0.989 0.000 1.000 0.000
#> GSM614468 2 0.000 0.989 0.000 1.000 0.000
#> GSM614469 2 0.153 0.963 0.040 0.960 0.000
#> GSM614470 2 0.153 0.963 0.040 0.960 0.000
#> GSM614471 2 0.153 0.963 0.040 0.960 0.000
#> GSM614472 2 0.153 0.963 0.040 0.960 0.000
#> GSM614473 2 0.153 0.963 0.040 0.960 0.000
#> GSM614474 2 0.153 0.963 0.040 0.960 0.000
#> GSM614475 2 0.153 0.963 0.040 0.960 0.000
#> GSM614476 3 0.721 0.547 0.060 0.272 0.668
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM614415 1 0.0000 0.8376 1.000 0.000 0.000 0.000
#> GSM614416 1 0.0000 0.8376 1.000 0.000 0.000 0.000
#> GSM614417 1 0.0000 0.8376 1.000 0.000 0.000 0.000
#> GSM614418 1 0.0000 0.8376 1.000 0.000 0.000 0.000
#> GSM614419 1 0.5088 0.4831 0.572 0.000 0.424 0.004
#> GSM614420 1 0.5088 0.4831 0.572 0.000 0.424 0.004
#> GSM614421 3 0.1022 0.7024 0.000 0.000 0.968 0.032
#> GSM614422 3 0.1022 0.7024 0.000 0.000 0.968 0.032
#> GSM614423 3 0.6243 0.0661 0.392 0.000 0.548 0.060
#> GSM614424 3 0.1022 0.7024 0.000 0.000 0.968 0.032
#> GSM614425 3 0.1022 0.7024 0.000 0.000 0.968 0.032
#> GSM614426 3 0.1022 0.7024 0.000 0.000 0.968 0.032
#> GSM614427 3 0.1022 0.7024 0.000 0.000 0.968 0.032
#> GSM614428 3 0.1022 0.7024 0.000 0.000 0.968 0.032
#> GSM614429 2 0.0592 0.8812 0.000 0.984 0.000 0.016
#> GSM614430 2 0.0592 0.8812 0.000 0.984 0.000 0.016
#> GSM614431 2 0.0592 0.8812 0.000 0.984 0.000 0.016
#> GSM614432 2 0.0592 0.8812 0.000 0.984 0.000 0.016
#> GSM614433 2 0.0592 0.8812 0.000 0.984 0.000 0.016
#> GSM614434 2 0.0592 0.8812 0.000 0.984 0.000 0.016
#> GSM614435 2 0.0592 0.8812 0.000 0.984 0.000 0.016
#> GSM614436 4 0.3591 0.6639 0.000 0.168 0.008 0.824
#> GSM614437 4 0.3024 0.6838 0.000 0.148 0.000 0.852
#> GSM614438 4 0.2973 0.8251 0.000 0.000 0.144 0.856
#> GSM614439 4 0.2973 0.8251 0.000 0.000 0.144 0.856
#> GSM614440 4 0.2973 0.8251 0.000 0.000 0.144 0.856
#> GSM614441 4 0.2973 0.8251 0.000 0.000 0.144 0.856
#> GSM614442 4 0.2973 0.8251 0.000 0.000 0.144 0.856
#> GSM614443 4 0.2760 0.8163 0.000 0.000 0.128 0.872
#> GSM614444 4 0.2973 0.8251 0.000 0.000 0.144 0.856
#> GSM614391 1 0.5060 0.5035 0.584 0.000 0.412 0.004
#> GSM614392 1 0.3583 0.7590 0.816 0.000 0.180 0.004
#> GSM614393 1 0.3448 0.7670 0.828 0.000 0.168 0.004
#> GSM614394 3 0.5050 -0.1207 0.408 0.000 0.588 0.004
#> GSM614395 3 0.0592 0.6983 0.000 0.000 0.984 0.016
#> GSM614396 3 0.5028 -0.0945 0.400 0.000 0.596 0.004
#> GSM614397 3 0.0592 0.6861 0.016 0.000 0.984 0.000
#> GSM614398 3 0.0592 0.6861 0.016 0.000 0.984 0.000
#> GSM614399 2 0.5010 0.7801 0.108 0.772 0.000 0.120
#> GSM614400 2 0.5010 0.7801 0.108 0.772 0.000 0.120
#> GSM614401 2 0.5010 0.7801 0.108 0.772 0.000 0.120
#> GSM614402 2 0.5010 0.7801 0.108 0.772 0.000 0.120
#> GSM614403 3 0.4877 0.3659 0.000 0.000 0.592 0.408
#> GSM614404 2 0.5010 0.7801 0.108 0.772 0.000 0.120
#> GSM614405 3 0.4941 0.2944 0.000 0.000 0.564 0.436
#> GSM614406 3 0.4941 0.2790 0.000 0.000 0.564 0.436
#> GSM614407 1 0.0817 0.8298 0.976 0.000 0.000 0.024
#> GSM614408 1 0.0817 0.8298 0.976 0.000 0.000 0.024
#> GSM614409 1 0.0188 0.8366 0.996 0.000 0.000 0.004
#> GSM614410 1 0.0817 0.8298 0.976 0.000 0.000 0.024
#> GSM614411 1 0.0817 0.8298 0.976 0.000 0.000 0.024
#> GSM614412 1 0.3726 0.7334 0.788 0.000 0.212 0.000
#> GSM614413 3 0.0524 0.6918 0.008 0.000 0.988 0.004
#> GSM614414 3 0.0895 0.6866 0.020 0.000 0.976 0.004
#> GSM614445 2 0.1211 0.8693 0.000 0.960 0.000 0.040
#> GSM614446 2 0.2214 0.8548 0.000 0.928 0.028 0.044
#> GSM614447 2 0.1302 0.8684 0.000 0.956 0.000 0.044
#> GSM614448 3 0.4898 0.3282 0.000 0.000 0.584 0.416
#> GSM614449 3 0.4925 0.3220 0.000 0.000 0.572 0.428
#> GSM614450 3 0.4933 0.3134 0.000 0.000 0.568 0.432
#> GSM614451 3 0.4790 0.3904 0.000 0.000 0.620 0.380
#> GSM614452 3 0.4790 0.3904 0.000 0.000 0.620 0.380
#> GSM614453 2 0.0592 0.8812 0.000 0.984 0.000 0.016
#> GSM614454 2 0.0592 0.8812 0.000 0.984 0.000 0.016
#> GSM614455 2 0.0592 0.8812 0.000 0.984 0.000 0.016
#> GSM614456 2 0.0592 0.8812 0.000 0.984 0.000 0.016
#> GSM614457 2 0.0592 0.8812 0.000 0.984 0.000 0.016
#> GSM614458 2 0.0592 0.8812 0.000 0.984 0.000 0.016
#> GSM614459 2 0.0592 0.8812 0.000 0.984 0.000 0.016
#> GSM614460 2 0.0592 0.8812 0.000 0.984 0.000 0.016
#> GSM614461 2 0.0000 0.8810 0.000 1.000 0.000 0.000
#> GSM614462 2 0.0000 0.8810 0.000 1.000 0.000 0.000
#> GSM614463 2 0.0000 0.8810 0.000 1.000 0.000 0.000
#> GSM614464 2 0.0000 0.8810 0.000 1.000 0.000 0.000
#> GSM614465 2 0.0000 0.8810 0.000 1.000 0.000 0.000
#> GSM614466 2 0.0000 0.8810 0.000 1.000 0.000 0.000
#> GSM614467 2 0.0188 0.8813 0.000 0.996 0.000 0.004
#> GSM614468 2 0.0188 0.8813 0.000 0.996 0.000 0.004
#> GSM614469 2 0.6566 0.6065 0.288 0.600 0.000 0.112
#> GSM614470 2 0.6566 0.6065 0.288 0.600 0.000 0.112
#> GSM614471 2 0.6566 0.6065 0.288 0.600 0.000 0.112
#> GSM614472 2 0.6566 0.6065 0.288 0.600 0.000 0.112
#> GSM614473 2 0.6566 0.6065 0.288 0.600 0.000 0.112
#> GSM614474 2 0.6566 0.6065 0.288 0.600 0.000 0.112
#> GSM614475 2 0.6566 0.6065 0.288 0.600 0.000 0.112
#> GSM614476 4 0.8486 0.0346 0.244 0.028 0.328 0.400
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM614415 5 0.1792 0.7380 0.084 0.000 0.000 0.000 0.916
#> GSM614416 5 0.1792 0.7380 0.084 0.000 0.000 0.000 0.916
#> GSM614417 5 0.1792 0.7380 0.084 0.000 0.000 0.000 0.916
#> GSM614418 5 0.1792 0.7380 0.084 0.000 0.000 0.000 0.916
#> GSM614419 5 0.4573 0.6018 0.044 0.000 0.256 0.000 0.700
#> GSM614420 5 0.4547 0.6070 0.044 0.000 0.252 0.000 0.704
#> GSM614421 3 0.0609 0.7596 0.000 0.000 0.980 0.020 0.000
#> GSM614422 3 0.0609 0.7596 0.000 0.000 0.980 0.020 0.000
#> GSM614423 3 0.5941 0.3237 0.096 0.000 0.600 0.016 0.288
#> GSM614424 3 0.0703 0.7588 0.000 0.000 0.976 0.024 0.000
#> GSM614425 3 0.0609 0.7596 0.000 0.000 0.980 0.020 0.000
#> GSM614426 3 0.0609 0.7596 0.000 0.000 0.980 0.020 0.000
#> GSM614427 3 0.0609 0.7596 0.000 0.000 0.980 0.020 0.000
#> GSM614428 3 0.0609 0.7596 0.000 0.000 0.980 0.020 0.000
#> GSM614429 2 0.0162 0.8957 0.000 0.996 0.000 0.004 0.000
#> GSM614430 2 0.0162 0.8957 0.000 0.996 0.000 0.004 0.000
#> GSM614431 2 0.0000 0.8957 0.000 1.000 0.000 0.000 0.000
#> GSM614432 2 0.0000 0.8957 0.000 1.000 0.000 0.000 0.000
#> GSM614433 2 0.0000 0.8957 0.000 1.000 0.000 0.000 0.000
#> GSM614434 2 0.0000 0.8957 0.000 1.000 0.000 0.000 0.000
#> GSM614435 2 0.0579 0.8930 0.008 0.984 0.000 0.008 0.000
#> GSM614436 4 0.3086 0.7383 0.004 0.180 0.000 0.816 0.000
#> GSM614437 4 0.1892 0.8545 0.000 0.080 0.000 0.916 0.004
#> GSM614438 4 0.1544 0.9413 0.000 0.000 0.068 0.932 0.000
#> GSM614439 4 0.1544 0.9413 0.000 0.000 0.068 0.932 0.000
#> GSM614440 4 0.1544 0.9413 0.000 0.000 0.068 0.932 0.000
#> GSM614441 4 0.1544 0.9413 0.000 0.000 0.068 0.932 0.000
#> GSM614442 4 0.1544 0.9413 0.000 0.000 0.068 0.932 0.000
#> GSM614443 4 0.1628 0.9307 0.000 0.008 0.056 0.936 0.000
#> GSM614444 4 0.1544 0.9413 0.000 0.000 0.068 0.932 0.000
#> GSM614391 5 0.4547 0.6067 0.044 0.000 0.252 0.000 0.704
#> GSM614392 5 0.2983 0.7152 0.040 0.000 0.096 0.000 0.864
#> GSM614393 5 0.2983 0.7152 0.040 0.000 0.096 0.000 0.864
#> GSM614394 5 0.5232 0.2467 0.044 0.000 0.456 0.000 0.500
#> GSM614395 3 0.2554 0.7196 0.020 0.000 0.896 0.008 0.076
#> GSM614396 5 0.5238 0.2015 0.044 0.000 0.472 0.000 0.484
#> GSM614397 3 0.2824 0.6836 0.020 0.000 0.864 0.000 0.116
#> GSM614398 3 0.2824 0.6836 0.020 0.000 0.864 0.000 0.116
#> GSM614399 1 0.5550 0.7419 0.600 0.336 0.000 0.032 0.032
#> GSM614400 1 0.5550 0.7419 0.600 0.336 0.000 0.032 0.032
#> GSM614401 1 0.5550 0.7419 0.600 0.336 0.000 0.032 0.032
#> GSM614402 1 0.5300 0.7236 0.632 0.312 0.000 0.032 0.024
#> GSM614403 3 0.6450 0.4444 0.296 0.000 0.492 0.212 0.000
#> GSM614404 1 0.5550 0.7419 0.600 0.336 0.000 0.032 0.032
#> GSM614405 3 0.5721 0.2926 0.084 0.000 0.492 0.424 0.000
#> GSM614406 3 0.4256 0.3385 0.000 0.000 0.564 0.436 0.000
#> GSM614407 5 0.3491 0.6544 0.228 0.000 0.000 0.004 0.768
#> GSM614408 5 0.3491 0.6544 0.228 0.000 0.000 0.004 0.768
#> GSM614409 5 0.2890 0.7067 0.160 0.000 0.000 0.004 0.836
#> GSM614410 5 0.3491 0.6544 0.228 0.000 0.000 0.004 0.768
#> GSM614411 5 0.3491 0.6544 0.228 0.000 0.000 0.004 0.768
#> GSM614412 5 0.2972 0.7193 0.024 0.000 0.108 0.004 0.864
#> GSM614413 3 0.2720 0.6980 0.020 0.000 0.880 0.004 0.096
#> GSM614414 3 0.2720 0.6980 0.020 0.000 0.880 0.004 0.096
#> GSM614445 2 0.5240 0.3015 0.360 0.584 0.000 0.056 0.000
#> GSM614446 2 0.6801 0.0773 0.412 0.448 0.084 0.056 0.000
#> GSM614447 2 0.5961 0.1951 0.408 0.512 0.024 0.056 0.000
#> GSM614448 3 0.5972 0.4686 0.140 0.000 0.560 0.300 0.000
#> GSM614449 3 0.6458 0.4395 0.240 0.000 0.500 0.260 0.000
#> GSM614450 3 0.6649 0.3730 0.284 0.000 0.448 0.268 0.000
#> GSM614451 3 0.3561 0.6042 0.000 0.000 0.740 0.260 0.000
#> GSM614452 3 0.3508 0.6117 0.000 0.000 0.748 0.252 0.000
#> GSM614453 2 0.0740 0.8934 0.008 0.980 0.000 0.008 0.004
#> GSM614454 2 0.0740 0.8934 0.008 0.980 0.000 0.008 0.004
#> GSM614455 2 0.0740 0.8934 0.008 0.980 0.000 0.008 0.004
#> GSM614456 2 0.0740 0.8934 0.008 0.980 0.000 0.008 0.004
#> GSM614457 2 0.0740 0.8934 0.008 0.980 0.000 0.008 0.004
#> GSM614458 2 0.0740 0.8934 0.008 0.980 0.000 0.008 0.004
#> GSM614459 2 0.0740 0.8934 0.008 0.980 0.000 0.008 0.004
#> GSM614460 2 0.0740 0.8934 0.008 0.980 0.000 0.008 0.004
#> GSM614461 2 0.1357 0.8703 0.048 0.948 0.000 0.004 0.000
#> GSM614462 2 0.1357 0.8703 0.048 0.948 0.000 0.004 0.000
#> GSM614463 2 0.1357 0.8703 0.048 0.948 0.000 0.004 0.000
#> GSM614464 2 0.1041 0.8812 0.032 0.964 0.000 0.004 0.000
#> GSM614465 2 0.1357 0.8703 0.048 0.948 0.000 0.004 0.000
#> GSM614466 2 0.1357 0.8703 0.048 0.948 0.000 0.004 0.000
#> GSM614467 2 0.1205 0.8794 0.040 0.956 0.000 0.004 0.000
#> GSM614468 2 0.1357 0.8771 0.048 0.948 0.000 0.004 0.000
#> GSM614469 1 0.5594 0.8325 0.608 0.284 0.000 0.000 0.108
#> GSM614470 1 0.5594 0.8325 0.608 0.284 0.000 0.000 0.108
#> GSM614471 1 0.5594 0.8325 0.608 0.284 0.000 0.000 0.108
#> GSM614472 1 0.5594 0.8325 0.608 0.284 0.000 0.000 0.108
#> GSM614473 1 0.5594 0.8325 0.608 0.284 0.000 0.000 0.108
#> GSM614474 1 0.5575 0.8308 0.612 0.280 0.000 0.000 0.108
#> GSM614475 1 0.5515 0.8230 0.624 0.268 0.000 0.000 0.108
#> GSM614476 1 0.7168 0.4015 0.596 0.048 0.208 0.044 0.104
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM614415 5 0.5312 0.673 0.208 0.000 0.000 0.024 0.648 0.120
#> GSM614416 5 0.5312 0.673 0.208 0.000 0.000 0.024 0.648 0.120
#> GSM614417 5 0.5312 0.673 0.208 0.000 0.000 0.024 0.648 0.120
#> GSM614418 5 0.5312 0.673 0.208 0.000 0.000 0.024 0.648 0.120
#> GSM614419 5 0.2579 0.592 0.004 0.000 0.088 0.000 0.876 0.032
#> GSM614420 5 0.2579 0.592 0.004 0.000 0.088 0.000 0.876 0.032
#> GSM614421 3 0.0260 0.727 0.000 0.000 0.992 0.000 0.000 0.008
#> GSM614422 3 0.0260 0.727 0.000 0.000 0.992 0.000 0.000 0.008
#> GSM614423 3 0.6204 0.349 0.088 0.000 0.636 0.024 0.140 0.112
#> GSM614424 3 0.0260 0.727 0.000 0.000 0.992 0.000 0.000 0.008
#> GSM614425 3 0.0260 0.727 0.000 0.000 0.992 0.000 0.000 0.008
#> GSM614426 3 0.0260 0.727 0.000 0.000 0.992 0.000 0.000 0.008
#> GSM614427 3 0.0260 0.727 0.000 0.000 0.992 0.000 0.000 0.008
#> GSM614428 3 0.0260 0.726 0.000 0.000 0.992 0.008 0.000 0.000
#> GSM614429 2 0.0146 0.931 0.000 0.996 0.000 0.000 0.000 0.004
#> GSM614430 2 0.0000 0.931 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM614431 2 0.0260 0.931 0.000 0.992 0.000 0.000 0.000 0.008
#> GSM614432 2 0.0260 0.931 0.000 0.992 0.000 0.000 0.000 0.008
#> GSM614433 2 0.0260 0.931 0.000 0.992 0.000 0.000 0.000 0.008
#> GSM614434 2 0.0146 0.931 0.000 0.996 0.000 0.000 0.000 0.004
#> GSM614435 2 0.0146 0.931 0.000 0.996 0.000 0.000 0.000 0.004
#> GSM614436 4 0.3599 0.669 0.004 0.212 0.016 0.764 0.000 0.004
#> GSM614437 4 0.1152 0.888 0.000 0.044 0.000 0.952 0.000 0.004
#> GSM614438 4 0.1075 0.949 0.000 0.000 0.048 0.952 0.000 0.000
#> GSM614439 4 0.1075 0.949 0.000 0.000 0.048 0.952 0.000 0.000
#> GSM614440 4 0.1075 0.949 0.000 0.000 0.048 0.952 0.000 0.000
#> GSM614441 4 0.1075 0.949 0.000 0.000 0.048 0.952 0.000 0.000
#> GSM614442 4 0.1075 0.949 0.000 0.000 0.048 0.952 0.000 0.000
#> GSM614443 4 0.0937 0.943 0.000 0.000 0.040 0.960 0.000 0.000
#> GSM614444 4 0.1075 0.949 0.000 0.000 0.048 0.952 0.000 0.000
#> GSM614391 5 0.1556 0.578 0.000 0.000 0.080 0.000 0.920 0.000
#> GSM614392 5 0.0551 0.626 0.004 0.000 0.008 0.000 0.984 0.004
#> GSM614393 5 0.0551 0.626 0.004 0.000 0.008 0.000 0.984 0.004
#> GSM614394 5 0.3288 0.264 0.000 0.000 0.276 0.000 0.724 0.000
#> GSM614395 3 0.3349 0.651 0.000 0.000 0.748 0.008 0.244 0.000
#> GSM614396 5 0.3371 0.225 0.000 0.000 0.292 0.000 0.708 0.000
#> GSM614397 3 0.3758 0.597 0.000 0.000 0.668 0.008 0.324 0.000
#> GSM614398 3 0.3668 0.594 0.000 0.000 0.668 0.004 0.328 0.000
#> GSM614399 1 0.5778 0.555 0.516 0.184 0.000 0.000 0.004 0.296
#> GSM614400 1 0.5778 0.555 0.516 0.184 0.000 0.000 0.004 0.296
#> GSM614401 1 0.5778 0.555 0.516 0.184 0.000 0.000 0.004 0.296
#> GSM614402 1 0.5729 0.544 0.516 0.168 0.000 0.000 0.004 0.312
#> GSM614403 6 0.4670 0.558 0.012 0.000 0.264 0.040 0.008 0.676
#> GSM614404 1 0.5778 0.555 0.516 0.184 0.000 0.000 0.004 0.296
#> GSM614405 3 0.6192 0.208 0.012 0.000 0.484 0.300 0.004 0.200
#> GSM614406 3 0.3727 0.389 0.000 0.000 0.612 0.388 0.000 0.000
#> GSM614407 5 0.6633 0.545 0.372 0.000 0.000 0.040 0.384 0.204
#> GSM614408 5 0.6633 0.545 0.372 0.000 0.000 0.040 0.384 0.204
#> GSM614409 5 0.6590 0.585 0.324 0.000 0.000 0.040 0.432 0.204
#> GSM614410 5 0.6633 0.545 0.372 0.000 0.000 0.040 0.384 0.204
#> GSM614411 5 0.6633 0.545 0.372 0.000 0.000 0.040 0.384 0.204
#> GSM614412 5 0.5680 0.643 0.104 0.000 0.012 0.040 0.648 0.196
#> GSM614413 3 0.4425 0.634 0.000 0.000 0.704 0.012 0.232 0.052
#> GSM614414 3 0.4450 0.632 0.000 0.000 0.700 0.012 0.236 0.052
#> GSM614445 6 0.4757 0.393 0.084 0.280 0.000 0.000 0.000 0.636
#> GSM614446 6 0.4440 0.603 0.028 0.132 0.088 0.000 0.000 0.752
#> GSM614447 6 0.4077 0.530 0.044 0.212 0.008 0.000 0.000 0.736
#> GSM614448 3 0.5292 -0.107 0.000 0.000 0.520 0.108 0.000 0.372
#> GSM614449 6 0.5110 0.235 0.000 0.000 0.440 0.080 0.000 0.480
#> GSM614450 6 0.4652 0.514 0.000 0.000 0.312 0.064 0.000 0.624
#> GSM614451 3 0.2260 0.668 0.000 0.000 0.860 0.140 0.000 0.000
#> GSM614452 3 0.2219 0.669 0.000 0.000 0.864 0.136 0.000 0.000
#> GSM614453 2 0.1223 0.925 0.012 0.960 0.000 0.008 0.004 0.016
#> GSM614454 2 0.1223 0.925 0.012 0.960 0.000 0.008 0.004 0.016
#> GSM614455 2 0.1223 0.925 0.012 0.960 0.000 0.008 0.004 0.016
#> GSM614456 2 0.1223 0.925 0.012 0.960 0.000 0.008 0.004 0.016
#> GSM614457 2 0.1223 0.925 0.012 0.960 0.000 0.008 0.004 0.016
#> GSM614458 2 0.1223 0.925 0.012 0.960 0.000 0.008 0.004 0.016
#> GSM614459 2 0.1223 0.925 0.012 0.960 0.000 0.008 0.004 0.016
#> GSM614460 2 0.1223 0.925 0.012 0.960 0.000 0.008 0.004 0.016
#> GSM614461 2 0.2542 0.885 0.044 0.876 0.000 0.000 0.000 0.080
#> GSM614462 2 0.2542 0.885 0.044 0.876 0.000 0.000 0.000 0.080
#> GSM614463 2 0.2542 0.885 0.044 0.876 0.000 0.000 0.000 0.080
#> GSM614464 2 0.2331 0.892 0.032 0.888 0.000 0.000 0.000 0.080
#> GSM614465 2 0.2542 0.885 0.044 0.876 0.000 0.000 0.000 0.080
#> GSM614466 2 0.2542 0.885 0.044 0.876 0.000 0.000 0.000 0.080
#> GSM614467 2 0.2457 0.890 0.036 0.880 0.000 0.000 0.000 0.084
#> GSM614468 2 0.2660 0.883 0.048 0.868 0.000 0.000 0.000 0.084
#> GSM614469 1 0.1958 0.768 0.896 0.100 0.000 0.000 0.004 0.000
#> GSM614470 1 0.1958 0.768 0.896 0.100 0.000 0.000 0.004 0.000
#> GSM614471 1 0.1958 0.768 0.896 0.100 0.000 0.000 0.004 0.000
#> GSM614472 1 0.1958 0.768 0.896 0.100 0.000 0.000 0.004 0.000
#> GSM614473 1 0.1958 0.768 0.896 0.100 0.000 0.000 0.004 0.000
#> GSM614474 1 0.1958 0.768 0.896 0.100 0.000 0.000 0.004 0.000
#> GSM614475 1 0.1858 0.763 0.904 0.092 0.000 0.000 0.004 0.000
#> GSM614476 1 0.3797 0.598 0.820 0.016 0.112 0.028 0.012 0.012
Heatmaps for the consensus matrix. It visualizes the probability of two samples to be in a same group.
consensus_heatmap(res, k = 2)
consensus_heatmap(res, k = 3)
consensus_heatmap(res, k = 4)
consensus_heatmap(res, k = 5)
consensus_heatmap(res, k = 6)
Heatmaps for the membership of samples in all partitions to see how consistent they are:
membership_heatmap(res, k = 2)
membership_heatmap(res, k = 3)
membership_heatmap(res, k = 4)
membership_heatmap(res, k = 5)
membership_heatmap(res, k = 6)
As soon as we have had the classes for columns, we can look for signatures which are significantly different between classes which can be candidate marks for certain classes. Following are the heatmaps for signatures.
Signature heatmaps where rows are scaled:
get_signatures(res, k = 2)
get_signatures(res, k = 3)
get_signatures(res, k = 4)
get_signatures(res, k = 5)
get_signatures(res, k = 6)
Signature heatmaps where rows are not scaled:
get_signatures(res, k = 2, scale_rows = FALSE)
get_signatures(res, k = 3, scale_rows = FALSE)
get_signatures(res, k = 4, scale_rows = FALSE)
get_signatures(res, k = 5, scale_rows = FALSE)
get_signatures(res, k = 6, scale_rows = FALSE)
Compare the overlap of signatures from different k:
compare_signatures(res)
get_signature() returns a data frame invisibly. TO get the list of signatures, the function
call should be assigned to a variable explicitly. In following code, if plot argument is set
to FALSE, no heatmap is plotted while only the differential analysis is performed.
# code only for demonstration
tb = get_signature(res, k = ..., plot = FALSE)
An example of the output of tb is:
#> which_row fdr mean_1 mean_2 scaled_mean_1 scaled_mean_2 km
#> 1 38 0.042760348 8.373488 9.131774 -0.5533452 0.5164555 1
#> 2 40 0.018707592 7.106213 8.469186 -0.6173731 0.5762149 1
#> 3 55 0.019134737 10.221463 11.207825 -0.6159697 0.5749050 1
#> 4 59 0.006059896 5.921854 7.869574 -0.6899429 0.6439467 1
#> 5 60 0.018055526 8.928898 10.211722 -0.6204761 0.5791110 1
#> 6 98 0.009384629 15.714769 14.887706 0.6635654 -0.6193277 2
...
The columns in tb are:
which_row: row indices corresponding to the input matrix.fdr: FDR for the differential test. mean_x: The mean value in group x.scaled_mean_x: The mean value in group x after rows are scaled.km: Row groups if k-means clustering is applied to rows.UMAP plot which shows how samples are separated.
dimension_reduction(res, k = 2, method = "UMAP")
dimension_reduction(res, k = 3, method = "UMAP")
dimension_reduction(res, k = 4, method = "UMAP")
dimension_reduction(res, k = 5, method = "UMAP")
dimension_reduction(res, k = 6, method = "UMAP")
Following heatmap shows how subgroups are split when increasing k:
collect_classes(res)
Test correlation between subgroups and known annotations. If the known annotation is numeric, one-way ANOVA test is applied, and if the known annotation is discrete, chi-squared contingency table test is applied.
test_to_known_factors(res)
#> n individual(p) protocol(p) time(p) other(p) k
#> ATC:skmeans 85 1.43e-06 0.013416 0.686 0.00765 2
#> ATC:skmeans 86 9.93e-14 0.000205 0.945 0.26856 3
#> ATC:skmeans 72 5.45e-22 0.011462 1.000 0.04118 4
#> ATC:skmeans 73 1.56e-32 0.003642 1.000 0.01719 5
#> ATC:skmeans 78 1.35e-36 0.007309 1.000 0.02286 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 21168 rows and 86 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 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 0.927 0.936 0.973 0.4751 0.521 0.521
#> 3 3 0.702 0.790 0.886 0.1950 0.904 0.821
#> 4 4 0.916 0.871 0.952 0.1853 0.830 0.647
#> 5 5 0.831 0.793 0.906 0.1292 0.841 0.567
#> 6 6 0.871 0.844 0.934 0.0276 0.960 0.838
suggest_best_k() suggests the best \(k\) based on these statistics. The rules are as follows:
suggest_best_k(res)
#> [1] 4
#> 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
#> GSM614415 2 0.0000 0.982 0.000 1.000
#> GSM614416 2 0.0000 0.982 0.000 1.000
#> GSM614417 2 0.0000 0.982 0.000 1.000
#> GSM614418 2 0.0000 0.982 0.000 1.000
#> GSM614419 1 0.0000 0.954 1.000 0.000
#> GSM614420 1 0.0000 0.954 1.000 0.000
#> GSM614421 1 0.0000 0.954 1.000 0.000
#> GSM614422 1 0.0000 0.954 1.000 0.000
#> GSM614423 1 0.9393 0.475 0.644 0.356
#> GSM614424 1 0.0000 0.954 1.000 0.000
#> GSM614425 1 0.0000 0.954 1.000 0.000
#> GSM614426 1 0.0000 0.954 1.000 0.000
#> GSM614427 1 0.0000 0.954 1.000 0.000
#> GSM614428 1 0.0000 0.954 1.000 0.000
#> GSM614429 2 0.0000 0.982 0.000 1.000
#> GSM614430 2 0.0000 0.982 0.000 1.000
#> GSM614431 2 0.0000 0.982 0.000 1.000
#> GSM614432 2 0.0000 0.982 0.000 1.000
#> GSM614433 2 0.0000 0.982 0.000 1.000
#> GSM614434 2 0.0000 0.982 0.000 1.000
#> GSM614435 2 0.0000 0.982 0.000 1.000
#> GSM614436 2 0.0938 0.971 0.012 0.988
#> GSM614437 2 0.0000 0.982 0.000 1.000
#> GSM614438 1 0.0672 0.948 0.992 0.008
#> GSM614439 1 0.0000 0.954 1.000 0.000
#> GSM614440 1 0.0000 0.954 1.000 0.000
#> GSM614441 1 0.0000 0.954 1.000 0.000
#> GSM614442 1 0.6438 0.786 0.836 0.164
#> GSM614443 2 0.7139 0.751 0.196 0.804
#> GSM614444 1 0.0000 0.954 1.000 0.000
#> GSM614391 1 0.0000 0.954 1.000 0.000
#> GSM614392 1 0.3879 0.889 0.924 0.076
#> GSM614393 1 0.9323 0.494 0.652 0.348
#> GSM614394 1 0.0000 0.954 1.000 0.000
#> GSM614395 1 0.0000 0.954 1.000 0.000
#> GSM614396 1 0.0000 0.954 1.000 0.000
#> GSM614397 1 0.0000 0.954 1.000 0.000
#> GSM614398 1 0.0000 0.954 1.000 0.000
#> GSM614399 2 0.0000 0.982 0.000 1.000
#> GSM614400 2 0.0000 0.982 0.000 1.000
#> GSM614401 2 0.0000 0.982 0.000 1.000
#> GSM614402 2 0.0000 0.982 0.000 1.000
#> GSM614403 2 0.5737 0.836 0.136 0.864
#> GSM614404 2 0.0000 0.982 0.000 1.000
#> GSM614405 1 0.0000 0.954 1.000 0.000
#> GSM614406 1 0.0000 0.954 1.000 0.000
#> GSM614407 2 0.0000 0.982 0.000 1.000
#> GSM614408 2 0.0000 0.982 0.000 1.000
#> GSM614409 2 0.0000 0.982 0.000 1.000
#> GSM614410 2 0.0000 0.982 0.000 1.000
#> GSM614411 2 0.0000 0.982 0.000 1.000
#> GSM614412 1 0.0376 0.951 0.996 0.004
#> GSM614413 1 0.0000 0.954 1.000 0.000
#> GSM614414 1 0.0000 0.954 1.000 0.000
#> GSM614445 2 0.0000 0.982 0.000 1.000
#> GSM614446 2 0.0000 0.982 0.000 1.000
#> GSM614447 2 0.0000 0.982 0.000 1.000
#> GSM614448 1 0.0000 0.954 1.000 0.000
#> GSM614449 1 0.9977 0.104 0.528 0.472
#> GSM614450 2 0.8327 0.638 0.264 0.736
#> GSM614451 1 0.0000 0.954 1.000 0.000
#> GSM614452 1 0.0000 0.954 1.000 0.000
#> GSM614453 2 0.0000 0.982 0.000 1.000
#> GSM614454 2 0.0000 0.982 0.000 1.000
#> GSM614455 2 0.0000 0.982 0.000 1.000
#> GSM614456 2 0.0000 0.982 0.000 1.000
#> GSM614457 2 0.0000 0.982 0.000 1.000
#> GSM614458 2 0.0000 0.982 0.000 1.000
#> GSM614459 2 0.0000 0.982 0.000 1.000
#> GSM614460 2 0.0000 0.982 0.000 1.000
#> GSM614461 2 0.0000 0.982 0.000 1.000
#> GSM614462 2 0.0000 0.982 0.000 1.000
#> GSM614463 2 0.0000 0.982 0.000 1.000
#> GSM614464 2 0.0000 0.982 0.000 1.000
#> GSM614465 2 0.0000 0.982 0.000 1.000
#> GSM614466 2 0.0000 0.982 0.000 1.000
#> GSM614467 2 0.0000 0.982 0.000 1.000
#> GSM614468 2 0.0000 0.982 0.000 1.000
#> GSM614469 2 0.0000 0.982 0.000 1.000
#> GSM614470 2 0.0000 0.982 0.000 1.000
#> GSM614471 2 0.0000 0.982 0.000 1.000
#> GSM614472 2 0.0000 0.982 0.000 1.000
#> GSM614473 2 0.0000 0.982 0.000 1.000
#> GSM614474 2 0.0000 0.982 0.000 1.000
#> GSM614475 2 0.0000 0.982 0.000 1.000
#> GSM614476 2 0.8144 0.661 0.252 0.748
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM614415 2 0.5098 0.7379 0.248 0.752 0.000
#> GSM614416 2 0.5098 0.7379 0.248 0.752 0.000
#> GSM614417 2 0.5098 0.7379 0.248 0.752 0.000
#> GSM614418 2 0.5098 0.7379 0.248 0.752 0.000
#> GSM614419 1 0.0000 0.7190 1.000 0.000 0.000
#> GSM614420 1 0.0000 0.7190 1.000 0.000 0.000
#> GSM614421 1 0.5285 0.8202 0.752 0.004 0.244
#> GSM614422 1 0.5285 0.8202 0.752 0.004 0.244
#> GSM614423 1 0.6518 0.6049 0.752 0.168 0.080
#> GSM614424 1 0.5285 0.8202 0.752 0.004 0.244
#> GSM614425 1 0.5285 0.8202 0.752 0.004 0.244
#> GSM614426 1 0.5285 0.8202 0.752 0.004 0.244
#> GSM614427 1 0.5285 0.8202 0.752 0.004 0.244
#> GSM614428 1 0.5098 0.8177 0.752 0.000 0.248
#> GSM614429 2 0.0000 0.9187 0.000 1.000 0.000
#> GSM614430 2 0.0000 0.9187 0.000 1.000 0.000
#> GSM614431 2 0.0000 0.9187 0.000 1.000 0.000
#> GSM614432 2 0.0000 0.9187 0.000 1.000 0.000
#> GSM614433 2 0.0000 0.9187 0.000 1.000 0.000
#> GSM614434 2 0.0000 0.9187 0.000 1.000 0.000
#> GSM614435 2 0.0000 0.9187 0.000 1.000 0.000
#> GSM614436 2 0.1411 0.8918 0.036 0.964 0.000
#> GSM614437 2 0.5859 0.4742 0.000 0.656 0.344
#> GSM614438 3 0.0000 0.7694 0.000 0.000 1.000
#> GSM614439 3 0.0000 0.7694 0.000 0.000 1.000
#> GSM614440 3 0.0000 0.7694 0.000 0.000 1.000
#> GSM614441 3 0.0000 0.7694 0.000 0.000 1.000
#> GSM614442 3 0.0000 0.7694 0.000 0.000 1.000
#> GSM614443 3 0.5058 0.4963 0.000 0.244 0.756
#> GSM614444 3 0.0000 0.7694 0.000 0.000 1.000
#> GSM614391 1 0.0000 0.7190 1.000 0.000 0.000
#> GSM614392 1 0.0000 0.7190 1.000 0.000 0.000
#> GSM614393 1 0.0000 0.7190 1.000 0.000 0.000
#> GSM614394 1 0.0237 0.7212 0.996 0.000 0.004
#> GSM614395 3 0.6307 -0.2749 0.488 0.000 0.512
#> GSM614396 1 0.2711 0.7607 0.912 0.000 0.088
#> GSM614397 1 0.5058 0.8202 0.756 0.000 0.244
#> GSM614398 1 0.5058 0.8202 0.756 0.000 0.244
#> GSM614399 2 0.0000 0.9187 0.000 1.000 0.000
#> GSM614400 2 0.0000 0.9187 0.000 1.000 0.000
#> GSM614401 2 0.0000 0.9187 0.000 1.000 0.000
#> GSM614402 2 0.0000 0.9187 0.000 1.000 0.000
#> GSM614403 2 0.6187 0.6120 0.248 0.724 0.028
#> GSM614404 2 0.0000 0.9187 0.000 1.000 0.000
#> GSM614405 1 0.5285 0.8202 0.752 0.004 0.244
#> GSM614406 1 0.6045 0.5915 0.620 0.000 0.380
#> GSM614407 2 0.4974 0.7485 0.236 0.764 0.000
#> GSM614408 2 0.5098 0.7379 0.248 0.752 0.000
#> GSM614409 2 0.5098 0.7379 0.248 0.752 0.000
#> GSM614410 2 0.5098 0.7379 0.248 0.752 0.000
#> GSM614411 2 0.5058 0.7417 0.244 0.756 0.000
#> GSM614412 1 0.0000 0.7190 1.000 0.000 0.000
#> GSM614413 1 0.5058 0.8202 0.756 0.000 0.244
#> GSM614414 1 0.5058 0.8202 0.756 0.000 0.244
#> GSM614445 2 0.0000 0.9187 0.000 1.000 0.000
#> GSM614446 2 0.0000 0.9187 0.000 1.000 0.000
#> GSM614447 2 0.0000 0.9187 0.000 1.000 0.000
#> GSM614448 1 0.5365 0.8128 0.744 0.004 0.252
#> GSM614449 2 0.9441 0.0578 0.316 0.484 0.200
#> GSM614450 2 0.5551 0.6792 0.212 0.768 0.020
#> GSM614451 3 0.5785 0.3110 0.332 0.000 0.668
#> GSM614452 3 0.5859 0.2778 0.344 0.000 0.656
#> GSM614453 2 0.0000 0.9187 0.000 1.000 0.000
#> GSM614454 2 0.0000 0.9187 0.000 1.000 0.000
#> GSM614455 2 0.0000 0.9187 0.000 1.000 0.000
#> GSM614456 2 0.0000 0.9187 0.000 1.000 0.000
#> GSM614457 2 0.0000 0.9187 0.000 1.000 0.000
#> GSM614458 2 0.0000 0.9187 0.000 1.000 0.000
#> GSM614459 2 0.0000 0.9187 0.000 1.000 0.000
#> GSM614460 2 0.0000 0.9187 0.000 1.000 0.000
#> GSM614461 2 0.0000 0.9187 0.000 1.000 0.000
#> GSM614462 2 0.0000 0.9187 0.000 1.000 0.000
#> GSM614463 2 0.0000 0.9187 0.000 1.000 0.000
#> GSM614464 2 0.0000 0.9187 0.000 1.000 0.000
#> GSM614465 2 0.0000 0.9187 0.000 1.000 0.000
#> GSM614466 2 0.0000 0.9187 0.000 1.000 0.000
#> GSM614467 2 0.0000 0.9187 0.000 1.000 0.000
#> GSM614468 2 0.0000 0.9187 0.000 1.000 0.000
#> GSM614469 2 0.0000 0.9187 0.000 1.000 0.000
#> GSM614470 2 0.0000 0.9187 0.000 1.000 0.000
#> GSM614471 2 0.0000 0.9187 0.000 1.000 0.000
#> GSM614472 2 0.0000 0.9187 0.000 1.000 0.000
#> GSM614473 2 0.0000 0.9187 0.000 1.000 0.000
#> GSM614474 2 0.0000 0.9187 0.000 1.000 0.000
#> GSM614475 2 0.0000 0.9187 0.000 1.000 0.000
#> GSM614476 2 0.7232 0.6104 0.172 0.712 0.116
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM614415 1 0.0000 0.9731 1.000 0.000 0.000 0.000
#> GSM614416 1 0.0000 0.9731 1.000 0.000 0.000 0.000
#> GSM614417 1 0.0000 0.9731 1.000 0.000 0.000 0.000
#> GSM614418 1 0.0000 0.9731 1.000 0.000 0.000 0.000
#> GSM614419 1 0.0469 0.9636 0.988 0.000 0.012 0.000
#> GSM614420 1 0.0469 0.9636 0.988 0.000 0.012 0.000
#> GSM614421 3 0.0000 0.8996 0.000 0.000 1.000 0.000
#> GSM614422 3 0.0000 0.8996 0.000 0.000 1.000 0.000
#> GSM614423 3 0.0336 0.8921 0.000 0.008 0.992 0.000
#> GSM614424 3 0.0000 0.8996 0.000 0.000 1.000 0.000
#> GSM614425 3 0.0000 0.8996 0.000 0.000 1.000 0.000
#> GSM614426 3 0.0000 0.8996 0.000 0.000 1.000 0.000
#> GSM614427 3 0.0000 0.8996 0.000 0.000 1.000 0.000
#> GSM614428 3 0.0000 0.8996 0.000 0.000 1.000 0.000
#> GSM614429 2 0.0000 0.9573 0.000 1.000 0.000 0.000
#> GSM614430 2 0.0000 0.9573 0.000 1.000 0.000 0.000
#> GSM614431 2 0.0000 0.9573 0.000 1.000 0.000 0.000
#> GSM614432 2 0.0000 0.9573 0.000 1.000 0.000 0.000
#> GSM614433 2 0.0000 0.9573 0.000 1.000 0.000 0.000
#> GSM614434 2 0.0000 0.9573 0.000 1.000 0.000 0.000
#> GSM614435 2 0.0000 0.9573 0.000 1.000 0.000 0.000
#> GSM614436 2 0.0469 0.9484 0.000 0.988 0.012 0.000
#> GSM614437 2 0.4933 0.2651 0.000 0.568 0.000 0.432
#> GSM614438 4 0.0000 0.8783 0.000 0.000 0.000 1.000
#> GSM614439 4 0.0000 0.8783 0.000 0.000 0.000 1.000
#> GSM614440 4 0.0000 0.8783 0.000 0.000 0.000 1.000
#> GSM614441 4 0.0000 0.8783 0.000 0.000 0.000 1.000
#> GSM614442 4 0.0000 0.8783 0.000 0.000 0.000 1.000
#> GSM614443 4 0.0000 0.8783 0.000 0.000 0.000 1.000
#> GSM614444 4 0.0000 0.8783 0.000 0.000 0.000 1.000
#> GSM614391 1 0.0469 0.9636 0.988 0.000 0.012 0.000
#> GSM614392 1 0.0000 0.9731 1.000 0.000 0.000 0.000
#> GSM614393 1 0.0000 0.9731 1.000 0.000 0.000 0.000
#> GSM614394 3 0.2589 0.7915 0.116 0.000 0.884 0.000
#> GSM614395 3 0.4406 0.4690 0.000 0.000 0.700 0.300
#> GSM614396 3 0.1022 0.8750 0.032 0.000 0.968 0.000
#> GSM614397 3 0.0000 0.8996 0.000 0.000 1.000 0.000
#> GSM614398 3 0.0000 0.8996 0.000 0.000 1.000 0.000
#> GSM614399 2 0.0000 0.9573 0.000 1.000 0.000 0.000
#> GSM614400 2 0.0469 0.9519 0.012 0.988 0.000 0.000
#> GSM614401 2 0.0469 0.9519 0.012 0.988 0.000 0.000
#> GSM614402 2 0.0000 0.9573 0.000 1.000 0.000 0.000
#> GSM614403 2 0.4999 0.0466 0.000 0.508 0.492 0.000
#> GSM614404 2 0.0469 0.9519 0.012 0.988 0.000 0.000
#> GSM614405 3 0.0000 0.8996 0.000 0.000 1.000 0.000
#> GSM614406 3 0.2921 0.7523 0.000 0.000 0.860 0.140
#> GSM614407 1 0.3486 0.6792 0.812 0.188 0.000 0.000
#> GSM614408 1 0.0000 0.9731 1.000 0.000 0.000 0.000
#> GSM614409 1 0.0000 0.9731 1.000 0.000 0.000 0.000
#> GSM614410 1 0.0000 0.9731 1.000 0.000 0.000 0.000
#> GSM614411 1 0.0000 0.9731 1.000 0.000 0.000 0.000
#> GSM614412 3 0.4790 0.4000 0.380 0.000 0.620 0.000
#> GSM614413 3 0.0000 0.8996 0.000 0.000 1.000 0.000
#> GSM614414 3 0.0000 0.8996 0.000 0.000 1.000 0.000
#> GSM614445 2 0.0000 0.9573 0.000 1.000 0.000 0.000
#> GSM614446 2 0.0000 0.9573 0.000 1.000 0.000 0.000
#> GSM614447 2 0.0000 0.9573 0.000 1.000 0.000 0.000
#> GSM614448 3 0.0000 0.8996 0.000 0.000 1.000 0.000
#> GSM614449 3 0.4855 0.2651 0.000 0.400 0.600 0.000
#> GSM614450 2 0.4643 0.4800 0.000 0.656 0.344 0.000
#> GSM614451 4 0.4888 0.3327 0.000 0.000 0.412 0.588
#> GSM614452 4 0.4933 0.2798 0.000 0.000 0.432 0.568
#> GSM614453 2 0.0000 0.9573 0.000 1.000 0.000 0.000
#> GSM614454 2 0.0000 0.9573 0.000 1.000 0.000 0.000
#> GSM614455 2 0.0000 0.9573 0.000 1.000 0.000 0.000
#> GSM614456 2 0.0000 0.9573 0.000 1.000 0.000 0.000
#> GSM614457 2 0.0000 0.9573 0.000 1.000 0.000 0.000
#> GSM614458 2 0.0000 0.9573 0.000 1.000 0.000 0.000
#> GSM614459 2 0.0000 0.9573 0.000 1.000 0.000 0.000
#> GSM614460 2 0.0000 0.9573 0.000 1.000 0.000 0.000
#> GSM614461 2 0.0000 0.9573 0.000 1.000 0.000 0.000
#> GSM614462 2 0.0000 0.9573 0.000 1.000 0.000 0.000
#> GSM614463 2 0.0000 0.9573 0.000 1.000 0.000 0.000
#> GSM614464 2 0.0000 0.9573 0.000 1.000 0.000 0.000
#> GSM614465 2 0.0000 0.9573 0.000 1.000 0.000 0.000
#> GSM614466 2 0.0000 0.9573 0.000 1.000 0.000 0.000
#> GSM614467 2 0.0000 0.9573 0.000 1.000 0.000 0.000
#> GSM614468 2 0.0000 0.9573 0.000 1.000 0.000 0.000
#> GSM614469 2 0.0469 0.9519 0.012 0.988 0.000 0.000
#> GSM614470 2 0.0469 0.9519 0.012 0.988 0.000 0.000
#> GSM614471 2 0.0469 0.9519 0.012 0.988 0.000 0.000
#> GSM614472 2 0.0469 0.9519 0.012 0.988 0.000 0.000
#> GSM614473 2 0.0469 0.9519 0.012 0.988 0.000 0.000
#> GSM614474 2 0.0469 0.9519 0.012 0.988 0.000 0.000
#> GSM614475 2 0.0469 0.9519 0.012 0.988 0.000 0.000
#> GSM614476 2 0.4546 0.6427 0.012 0.732 0.256 0.000
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM614415 1 0.3816 0.6949 0.696 0.000 0.000 0.000 0.304
#> GSM614416 1 0.3816 0.6949 0.696 0.000 0.000 0.000 0.304
#> GSM614417 1 0.3816 0.6949 0.696 0.000 0.000 0.000 0.304
#> GSM614418 1 0.3816 0.6949 0.696 0.000 0.000 0.000 0.304
#> GSM614419 5 0.0912 0.7670 0.012 0.000 0.016 0.000 0.972
#> GSM614420 5 0.0404 0.7702 0.012 0.000 0.000 0.000 0.988
#> GSM614421 3 0.0000 0.8673 0.000 0.000 1.000 0.000 0.000
#> GSM614422 3 0.0000 0.8673 0.000 0.000 1.000 0.000 0.000
#> GSM614423 3 0.1908 0.7710 0.000 0.092 0.908 0.000 0.000
#> GSM614424 3 0.0000 0.8673 0.000 0.000 1.000 0.000 0.000
#> GSM614425 3 0.0000 0.8673 0.000 0.000 1.000 0.000 0.000
#> GSM614426 3 0.0000 0.8673 0.000 0.000 1.000 0.000 0.000
#> GSM614427 3 0.0000 0.8673 0.000 0.000 1.000 0.000 0.000
#> GSM614428 3 0.0162 0.8646 0.000 0.000 0.996 0.000 0.004
#> GSM614429 2 0.0000 0.9364 0.000 1.000 0.000 0.000 0.000
#> GSM614430 2 0.0000 0.9364 0.000 1.000 0.000 0.000 0.000
#> GSM614431 2 0.0000 0.9364 0.000 1.000 0.000 0.000 0.000
#> GSM614432 2 0.0000 0.9364 0.000 1.000 0.000 0.000 0.000
#> GSM614433 2 0.0000 0.9364 0.000 1.000 0.000 0.000 0.000
#> GSM614434 2 0.0000 0.9364 0.000 1.000 0.000 0.000 0.000
#> GSM614435 2 0.0000 0.9364 0.000 1.000 0.000 0.000 0.000
#> GSM614436 2 0.0000 0.9364 0.000 1.000 0.000 0.000 0.000
#> GSM614437 2 0.4201 0.3169 0.000 0.592 0.000 0.408 0.000
#> GSM614438 4 0.0000 0.8495 0.000 0.000 0.000 1.000 0.000
#> GSM614439 4 0.0000 0.8495 0.000 0.000 0.000 1.000 0.000
#> GSM614440 4 0.0000 0.8495 0.000 0.000 0.000 1.000 0.000
#> GSM614441 4 0.0000 0.8495 0.000 0.000 0.000 1.000 0.000
#> GSM614442 4 0.0000 0.8495 0.000 0.000 0.000 1.000 0.000
#> GSM614443 4 0.2074 0.7333 0.000 0.104 0.000 0.896 0.000
#> GSM614444 4 0.0000 0.8495 0.000 0.000 0.000 1.000 0.000
#> GSM614391 5 0.0566 0.7714 0.012 0.000 0.004 0.000 0.984
#> GSM614392 5 0.0404 0.7702 0.012 0.000 0.000 0.000 0.988
#> GSM614393 5 0.0404 0.7702 0.012 0.000 0.000 0.000 0.988
#> GSM614394 5 0.3816 0.7050 0.000 0.000 0.304 0.000 0.696
#> GSM614395 5 0.5240 0.6189 0.000 0.000 0.120 0.204 0.676
#> GSM614396 5 0.3876 0.6934 0.000 0.000 0.316 0.000 0.684
#> GSM614397 5 0.3816 0.7029 0.000 0.000 0.304 0.000 0.696
#> GSM614398 5 0.3816 0.7029 0.000 0.000 0.304 0.000 0.696
#> GSM614399 2 0.0000 0.9364 0.000 1.000 0.000 0.000 0.000
#> GSM614400 1 0.3274 0.6196 0.780 0.220 0.000 0.000 0.000
#> GSM614401 2 0.4074 0.4705 0.364 0.636 0.000 0.000 0.000
#> GSM614402 2 0.0000 0.9364 0.000 1.000 0.000 0.000 0.000
#> GSM614403 2 0.4302 -0.0261 0.000 0.520 0.480 0.000 0.000
#> GSM614404 2 0.3508 0.6737 0.252 0.748 0.000 0.000 0.000
#> GSM614405 3 0.0000 0.8673 0.000 0.000 1.000 0.000 0.000
#> GSM614406 3 0.1732 0.7990 0.000 0.000 0.920 0.080 0.000
#> GSM614407 1 0.0404 0.8450 0.988 0.012 0.000 0.000 0.000
#> GSM614408 1 0.0000 0.8406 1.000 0.000 0.000 0.000 0.000
#> GSM614409 1 0.4067 0.6928 0.692 0.008 0.000 0.000 0.300
#> GSM614410 1 0.0000 0.8406 1.000 0.000 0.000 0.000 0.000
#> GSM614411 1 0.2813 0.7764 0.832 0.000 0.000 0.000 0.168
#> GSM614412 3 0.6494 0.1416 0.256 0.000 0.492 0.000 0.252
#> GSM614413 3 0.0000 0.8673 0.000 0.000 1.000 0.000 0.000
#> GSM614414 3 0.0000 0.8673 0.000 0.000 1.000 0.000 0.000
#> GSM614445 2 0.0000 0.9364 0.000 1.000 0.000 0.000 0.000
#> GSM614446 2 0.0000 0.9364 0.000 1.000 0.000 0.000 0.000
#> GSM614447 2 0.0000 0.9364 0.000 1.000 0.000 0.000 0.000
#> GSM614448 3 0.0000 0.8673 0.000 0.000 1.000 0.000 0.000
#> GSM614449 3 0.2852 0.6525 0.000 0.172 0.828 0.000 0.000
#> GSM614450 3 0.4300 0.0956 0.000 0.476 0.524 0.000 0.000
#> GSM614451 4 0.4505 0.4130 0.000 0.000 0.384 0.604 0.012
#> GSM614452 4 0.4565 0.3582 0.000 0.000 0.408 0.580 0.012
#> GSM614453 2 0.2852 0.7770 0.172 0.828 0.000 0.000 0.000
#> GSM614454 2 0.0000 0.9364 0.000 1.000 0.000 0.000 0.000
#> GSM614455 2 0.0000 0.9364 0.000 1.000 0.000 0.000 0.000
#> GSM614456 2 0.0000 0.9364 0.000 1.000 0.000 0.000 0.000
#> GSM614457 2 0.0000 0.9364 0.000 1.000 0.000 0.000 0.000
#> GSM614458 2 0.0000 0.9364 0.000 1.000 0.000 0.000 0.000
#> GSM614459 2 0.0000 0.9364 0.000 1.000 0.000 0.000 0.000
#> GSM614460 2 0.0000 0.9364 0.000 1.000 0.000 0.000 0.000
#> GSM614461 2 0.0000 0.9364 0.000 1.000 0.000 0.000 0.000
#> GSM614462 2 0.0000 0.9364 0.000 1.000 0.000 0.000 0.000
#> GSM614463 2 0.2891 0.7723 0.176 0.824 0.000 0.000 0.000
#> GSM614464 2 0.0000 0.9364 0.000 1.000 0.000 0.000 0.000
#> GSM614465 2 0.0000 0.9364 0.000 1.000 0.000 0.000 0.000
#> GSM614466 2 0.0000 0.9364 0.000 1.000 0.000 0.000 0.000
#> GSM614467 2 0.0000 0.9364 0.000 1.000 0.000 0.000 0.000
#> GSM614468 2 0.0000 0.9364 0.000 1.000 0.000 0.000 0.000
#> GSM614469 1 0.0703 0.8478 0.976 0.024 0.000 0.000 0.000
#> GSM614470 1 0.0703 0.8478 0.976 0.024 0.000 0.000 0.000
#> GSM614471 1 0.0703 0.8478 0.976 0.024 0.000 0.000 0.000
#> GSM614472 1 0.0703 0.8478 0.976 0.024 0.000 0.000 0.000
#> GSM614473 1 0.0703 0.8478 0.976 0.024 0.000 0.000 0.000
#> GSM614474 1 0.0703 0.8478 0.976 0.024 0.000 0.000 0.000
#> GSM614475 1 0.0703 0.8478 0.976 0.024 0.000 0.000 0.000
#> GSM614476 1 0.3565 0.6974 0.800 0.176 0.024 0.000 0.000
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM614415 6 0.0000 0.93239 0.000 0.000 0.000 0.000 0.000 1.000
#> GSM614416 6 0.0000 0.93239 0.000 0.000 0.000 0.000 0.000 1.000
#> GSM614417 6 0.0000 0.93239 0.000 0.000 0.000 0.000 0.000 1.000
#> GSM614418 6 0.0000 0.93239 0.000 0.000 0.000 0.000 0.000 1.000
#> GSM614419 6 0.0146 0.93050 0.000 0.000 0.000 0.000 0.004 0.996
#> GSM614420 6 0.0146 0.93050 0.000 0.000 0.000 0.000 0.004 0.996
#> GSM614421 3 0.0000 0.90839 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM614422 3 0.0000 0.90839 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM614423 3 0.0000 0.90839 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM614424 3 0.0000 0.90839 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM614425 3 0.0000 0.90839 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM614426 3 0.0000 0.90839 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM614427 3 0.0000 0.90839 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM614428 3 0.0790 0.88589 0.000 0.000 0.968 0.000 0.032 0.000
#> GSM614429 2 0.0000 0.95649 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM614430 2 0.0000 0.95649 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM614431 2 0.0000 0.95649 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM614432 2 0.0000 0.95649 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM614433 2 0.0000 0.95649 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM614434 2 0.0000 0.95649 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM614435 2 0.0000 0.95649 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM614436 2 0.0000 0.95649 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM614437 2 0.3774 0.32050 0.000 0.592 0.000 0.408 0.000 0.000
#> GSM614438 4 0.0000 0.84205 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM614439 4 0.0000 0.84205 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM614440 4 0.0000 0.84205 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM614441 4 0.0000 0.84205 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM614442 4 0.0000 0.84205 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM614443 4 0.1327 0.77029 0.000 0.064 0.000 0.936 0.000 0.000
#> GSM614444 4 0.0000 0.84205 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM614391 5 0.1765 0.89617 0.000 0.000 0.000 0.000 0.904 0.096
#> GSM614392 5 0.1863 0.89148 0.000 0.000 0.000 0.000 0.896 0.104
#> GSM614393 5 0.2006 0.89740 0.016 0.000 0.000 0.000 0.904 0.080
#> GSM614394 5 0.1765 0.88021 0.000 0.000 0.096 0.000 0.904 0.000
#> GSM614395 5 0.1863 0.82735 0.000 0.000 0.000 0.104 0.896 0.000
#> GSM614396 5 0.1765 0.88021 0.000 0.000 0.096 0.000 0.904 0.000
#> GSM614397 5 0.0000 0.90290 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM614398 5 0.0000 0.90290 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM614399 2 0.0000 0.95649 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM614400 1 0.3151 0.60028 0.748 0.252 0.000 0.000 0.000 0.000
#> GSM614401 2 0.3446 0.56026 0.308 0.692 0.000 0.000 0.000 0.000
#> GSM614402 2 0.0000 0.95649 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM614403 3 0.3737 0.39196 0.000 0.392 0.608 0.000 0.000 0.000
#> GSM614404 2 0.2883 0.72813 0.212 0.788 0.000 0.000 0.000 0.000
#> GSM614405 3 0.0000 0.90839 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM614406 3 0.2135 0.78009 0.000 0.000 0.872 0.128 0.000 0.000
#> GSM614407 1 0.0146 0.84607 0.996 0.000 0.000 0.000 0.000 0.004
#> GSM614408 1 0.2631 0.71237 0.820 0.000 0.000 0.000 0.000 0.180
#> GSM614409 1 0.3866 0.00335 0.516 0.000 0.000 0.000 0.000 0.484
#> GSM614410 1 0.0146 0.84607 0.996 0.000 0.000 0.000 0.000 0.004
#> GSM614411 1 0.2527 0.70390 0.832 0.000 0.000 0.000 0.000 0.168
#> GSM614412 6 0.5020 0.56983 0.128 0.000 0.244 0.000 0.000 0.628
#> GSM614413 3 0.0000 0.90839 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM614414 3 0.0000 0.90839 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM614445 2 0.0000 0.95649 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM614446 2 0.0000 0.95649 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM614447 2 0.0000 0.95649 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM614448 3 0.0000 0.90839 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM614449 3 0.2219 0.75616 0.000 0.136 0.864 0.000 0.000 0.000
#> GSM614450 3 0.2969 0.63863 0.000 0.224 0.776 0.000 0.000 0.000
#> GSM614451 4 0.5205 0.33356 0.000 0.000 0.384 0.520 0.096 0.000
#> GSM614452 4 0.5238 0.27267 0.000 0.000 0.408 0.496 0.096 0.000
#> GSM614453 2 0.2454 0.79613 0.160 0.840 0.000 0.000 0.000 0.000
#> GSM614454 2 0.0000 0.95649 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM614455 2 0.0000 0.95649 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM614456 2 0.0000 0.95649 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM614457 2 0.0000 0.95649 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM614458 2 0.0000 0.95649 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM614459 2 0.0000 0.95649 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM614460 2 0.0000 0.95649 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM614461 2 0.0000 0.95649 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM614462 2 0.0000 0.95649 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM614463 2 0.2491 0.79127 0.164 0.836 0.000 0.000 0.000 0.000
#> GSM614464 2 0.0000 0.95649 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM614465 2 0.0000 0.95649 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM614466 2 0.0000 0.95649 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM614467 2 0.0000 0.95649 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM614468 2 0.0000 0.95649 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM614469 1 0.0790 0.86888 0.968 0.032 0.000 0.000 0.000 0.000
#> GSM614470 1 0.0790 0.86888 0.968 0.032 0.000 0.000 0.000 0.000
#> GSM614471 1 0.0790 0.86888 0.968 0.032 0.000 0.000 0.000 0.000
#> GSM614472 1 0.0790 0.86888 0.968 0.032 0.000 0.000 0.000 0.000
#> GSM614473 1 0.0790 0.86888 0.968 0.032 0.000 0.000 0.000 0.000
#> GSM614474 1 0.0790 0.86888 0.968 0.032 0.000 0.000 0.000 0.000
#> GSM614475 1 0.0790 0.86888 0.968 0.032 0.000 0.000 0.000 0.000
#> GSM614476 1 0.3319 0.68877 0.800 0.164 0.036 0.000 0.000 0.000
Heatmaps for the consensus matrix. It visualizes the probability of two samples to be in a same group.
consensus_heatmap(res, k = 2)
consensus_heatmap(res, k = 3)
consensus_heatmap(res, k = 4)
consensus_heatmap(res, k = 5)
consensus_heatmap(res, k = 6)
Heatmaps for the membership of samples in all partitions to see how consistent they are:
membership_heatmap(res, k = 2)
membership_heatmap(res, k = 3)
membership_heatmap(res, k = 4)
membership_heatmap(res, k = 5)
membership_heatmap(res, k = 6)
As soon as we have had the classes for columns, we can look for signatures which are significantly different between classes which can be candidate marks for certain classes. Following are the heatmaps for signatures.
Signature heatmaps where rows are scaled:
get_signatures(res, k = 2)
get_signatures(res, k = 3)
get_signatures(res, k = 4)
get_signatures(res, k = 5)
get_signatures(res, k = 6)
Signature heatmaps where rows are not scaled:
get_signatures(res, k = 2, scale_rows = FALSE)
get_signatures(res, k = 3, scale_rows = FALSE)
get_signatures(res, k = 4, scale_rows = FALSE)
get_signatures(res, k = 5, scale_rows = FALSE)
get_signatures(res, k = 6, scale_rows = FALSE)
Compare the overlap of signatures from different k:
compare_signatures(res)
get_signature() returns a data frame invisibly. TO get the list of signatures, the function
call should be assigned to a variable explicitly. In following code, if plot argument is set
to FALSE, no heatmap is plotted while only the differential analysis is performed.
# code only for demonstration
tb = get_signature(res, k = ..., plot = FALSE)
An example of the output of tb is:
#> which_row fdr mean_1 mean_2 scaled_mean_1 scaled_mean_2 km
#> 1 38 0.042760348 8.373488 9.131774 -0.5533452 0.5164555 1
#> 2 40 0.018707592 7.106213 8.469186 -0.6173731 0.5762149 1
#> 3 55 0.019134737 10.221463 11.207825 -0.6159697 0.5749050 1
#> 4 59 0.006059896 5.921854 7.869574 -0.6899429 0.6439467 1
#> 5 60 0.018055526 8.928898 10.211722 -0.6204761 0.5791110 1
#> 6 98 0.009384629 15.714769 14.887706 0.6635654 -0.6193277 2
...
The columns in tb are:
which_row: row indices corresponding to the input matrix.fdr: FDR for the differential test. mean_x: The mean value in group x.scaled_mean_x: The mean value in group x after rows are scaled.km: Row groups if k-means clustering is applied to rows.UMAP plot which shows how samples are separated.
dimension_reduction(res, k = 2, method = "UMAP")
dimension_reduction(res, k = 3, method = "UMAP")
dimension_reduction(res, k = 4, method = "UMAP")
dimension_reduction(res, k = 5, method = "UMAP")
dimension_reduction(res, k = 6, method = "UMAP")
Following heatmap shows how subgroups are split when increasing k:
collect_classes(res)
Test correlation between subgroups and known annotations. If the known annotation is numeric, one-way ANOVA test is applied, and if the known annotation is discrete, chi-squared contingency table test is applied.
test_to_known_factors(res)
#> n individual(p) protocol(p) time(p) other(p) k
#> ATC:pam 83 4.67e-07 0.0805 0.968 6.07e-03 2
#> ATC:pam 80 1.51e-18 0.1707 0.979 4.03e-05 3
#> ATC:pam 78 9.14e-25 0.0182 0.999 5.53e-03 4
#> ATC:pam 79 4.45e-34 0.2967 1.000 6.15e-03 5
#> ATC:pam 81 4.36e-46 0.3645 1.000 2.09e-02 6
If matrix rows can be associated to genes, consider to use functional_enrichment(res,
...) to perform function enrichment for the signature genes. See this vignette for more detailed explanations.
The object with results only for a single top-value method and a single partition method can be extracted as:
res = res_list["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 21168 rows and 86 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 5.
#>
#> Following methods can be applied to this 'ConsensusPartition' object:
#> [1] "cola_report" "collect_classes" "collect_plots"
#> [4] "collect_stats" "colnames" "compare_signatures"
#> [7] "consensus_heatmap" "dimension_reduction" "functional_enrichment"
#> [10] "get_anno_col" "get_anno" "get_classes"
#> [13] "get_consensus" "get_matrix" "get_membership"
#> [16] "get_param" "get_signatures" "get_stats"
#> [19] "is_best_k" "is_stable_k" "membership_heatmap"
#> [22] "ncol" "nrow" "plot_ecdf"
#> [25] "rownames" "select_partition_number" "show"
#> [28] "suggest_best_k" "test_to_known_factors"
collect_plots() function collects all the plots made from res for all k (number of partitions)
into one single page to provide an easy and fast comparison between different k.
collect_plots(res)
The plots are:
k and the heatmap of
predicted classes for each k.k.k.k.All the plots in panels can be made by individual functions and they are plotted later in this section.
select_partition_number() produces several plots showing different
statistics for choosing “optimized” k. There are following statistics:
k;k, the area increased is defined as \(A_k - A_{k-1}\).The detailed explanations of these statistics can be found in the cola vignette.
Generally speaking, lower PAC score, higher mean silhouette score or higher
concordance corresponds to better partition. Rand index and Jaccard index
measure how similar the current partition is compared to partition with k-1.
If they are too similar, we won't accept k is better than k-1.
select_partition_number(res)
The numeric values for all these statistics can be obtained by get_stats().
get_stats(res)
#> k 1-PAC mean_silhouette concordance area_increased Rand Jaccard
#> 2 2 0.276 0.752 0.812 0.446 0.497 0.497
#> 3 3 0.421 0.697 0.824 0.282 0.854 0.726
#> 4 4 0.516 0.746 0.756 0.222 0.819 0.595
#> 5 5 0.750 0.769 0.872 0.114 0.923 0.735
#> 6 6 0.752 0.690 0.804 0.039 0.966 0.853
suggest_best_k() suggests the best \(k\) based on these statistics. The rules are as follows:
suggest_best_k(res)
#> [1] 5
Following shows the table of the partitions (You need to click the show/hide
code output link to see it). The membership matrix (columns with name p*)
is inferred by
clue::cl_consensus()
function with the SE method. Basically the value in the membership matrix
represents the probability to belong to a certain group. The finall class
label for an item is determined with the group with highest probability it
belongs to.
In get_classes() function, the entropy is calculated from the membership
matrix and the silhouette score is calculated from the consensus matrix.
cbind(get_classes(res, k = 2), get_membership(res, k = 2))
#> class entropy silhouette p1 p2
#> GSM614415 1 0.4022 0.818 0.920 0.080
#> GSM614416 1 0.4022 0.818 0.920 0.080
#> GSM614417 1 0.4022 0.818 0.920 0.080
#> GSM614418 1 0.4022 0.818 0.920 0.080
#> GSM614419 1 0.4022 0.818 0.920 0.080
#> GSM614420 1 0.4022 0.818 0.920 0.080
#> GSM614421 1 0.9522 0.350 0.628 0.372
#> GSM614422 1 0.9522 0.350 0.628 0.372
#> GSM614423 1 0.9427 0.366 0.640 0.360
#> GSM614424 1 0.9552 0.338 0.624 0.376
#> GSM614425 1 0.9491 0.354 0.632 0.368
#> GSM614426 1 0.9686 0.271 0.604 0.396
#> GSM614427 1 0.9580 0.332 0.620 0.380
#> GSM614428 1 0.9661 0.300 0.608 0.392
#> GSM614429 2 0.8386 0.850 0.268 0.732
#> GSM614430 2 0.8386 0.850 0.268 0.732
#> GSM614431 2 0.9087 0.827 0.324 0.676
#> GSM614432 2 0.9129 0.823 0.328 0.672
#> GSM614433 2 0.8909 0.838 0.308 0.692
#> GSM614434 2 0.8955 0.835 0.312 0.688
#> GSM614435 2 0.8955 0.836 0.312 0.688
#> GSM614436 2 0.9460 0.806 0.364 0.636
#> GSM614437 2 0.1633 0.695 0.024 0.976
#> GSM614438 2 0.1633 0.695 0.024 0.976
#> GSM614439 2 0.1633 0.695 0.024 0.976
#> GSM614440 2 0.1633 0.695 0.024 0.976
#> GSM614441 2 0.1633 0.695 0.024 0.976
#> GSM614442 2 0.1633 0.695 0.024 0.976
#> GSM614443 2 0.1633 0.695 0.024 0.976
#> GSM614444 2 0.1633 0.695 0.024 0.976
#> GSM614391 1 0.4298 0.814 0.912 0.088
#> GSM614392 1 0.4298 0.814 0.912 0.088
#> GSM614393 1 0.4298 0.814 0.912 0.088
#> GSM614394 1 0.4298 0.814 0.912 0.088
#> GSM614395 1 0.4298 0.814 0.912 0.088
#> GSM614396 1 0.4298 0.814 0.912 0.088
#> GSM614397 1 0.4298 0.814 0.912 0.088
#> GSM614398 1 0.4298 0.814 0.912 0.088
#> GSM614399 1 0.1633 0.832 0.976 0.024
#> GSM614400 1 0.1633 0.832 0.976 0.024
#> GSM614401 1 0.1633 0.832 0.976 0.024
#> GSM614402 1 0.2603 0.821 0.956 0.044
#> GSM614403 1 0.5629 0.704 0.868 0.132
#> GSM614404 1 0.1633 0.832 0.976 0.024
#> GSM614405 1 0.6531 0.696 0.832 0.168
#> GSM614406 2 0.7453 0.801 0.212 0.788
#> GSM614407 1 0.0672 0.836 0.992 0.008
#> GSM614408 1 0.0376 0.836 0.996 0.004
#> GSM614409 1 0.0000 0.836 1.000 0.000
#> GSM614410 1 0.0376 0.836 0.996 0.004
#> GSM614411 1 0.0000 0.836 1.000 0.000
#> GSM614412 1 0.0000 0.836 1.000 0.000
#> GSM614413 1 0.4161 0.794 0.916 0.084
#> GSM614414 1 0.0938 0.834 0.988 0.012
#> GSM614445 2 0.9954 0.530 0.460 0.540
#> GSM614446 2 0.9460 0.779 0.364 0.636
#> GSM614447 2 0.9286 0.803 0.344 0.656
#> GSM614448 2 0.8016 0.797 0.244 0.756
#> GSM614449 2 0.8813 0.838 0.300 0.700
#> GSM614450 1 0.9323 0.230 0.652 0.348
#> GSM614451 2 0.6712 0.789 0.176 0.824
#> GSM614452 2 0.6712 0.789 0.176 0.824
#> GSM614453 2 0.8443 0.850 0.272 0.728
#> GSM614454 2 0.8386 0.850 0.268 0.732
#> GSM614455 2 0.8386 0.850 0.268 0.732
#> GSM614456 2 0.8386 0.850 0.268 0.732
#> GSM614457 2 0.8386 0.850 0.268 0.732
#> GSM614458 2 0.8386 0.850 0.268 0.732
#> GSM614459 2 0.8386 0.850 0.268 0.732
#> GSM614460 2 0.8386 0.850 0.268 0.732
#> GSM614461 2 0.9209 0.815 0.336 0.664
#> GSM614462 2 0.9248 0.811 0.340 0.660
#> GSM614463 2 0.9427 0.786 0.360 0.640
#> GSM614464 2 0.9087 0.819 0.324 0.676
#> GSM614465 2 0.8661 0.845 0.288 0.712
#> GSM614466 2 0.9087 0.827 0.324 0.676
#> GSM614467 2 0.9983 0.588 0.476 0.524
#> GSM614468 2 0.8763 0.843 0.296 0.704
#> GSM614469 1 0.1633 0.832 0.976 0.024
#> GSM614470 1 0.1414 0.833 0.980 0.020
#> GSM614471 1 0.1633 0.832 0.976 0.024
#> GSM614472 1 0.1633 0.832 0.976 0.024
#> GSM614473 1 0.1633 0.832 0.976 0.024
#> GSM614474 1 0.1633 0.832 0.976 0.024
#> GSM614475 1 0.1633 0.832 0.976 0.024
#> GSM614476 1 0.0000 0.836 1.000 0.000
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM614415 1 0.0592 0.7340 0.988 0.000 0.012
#> GSM614416 1 0.0592 0.7340 0.988 0.000 0.012
#> GSM614417 1 0.0592 0.7340 0.988 0.000 0.012
#> GSM614418 1 0.0592 0.7340 0.988 0.000 0.012
#> GSM614419 1 0.0592 0.7340 0.988 0.000 0.012
#> GSM614420 1 0.0592 0.7340 0.988 0.000 0.012
#> GSM614421 1 0.8937 0.5819 0.540 0.308 0.152
#> GSM614422 1 0.8937 0.5819 0.540 0.308 0.152
#> GSM614423 1 0.8790 0.5769 0.540 0.328 0.132
#> GSM614424 1 0.8957 0.5774 0.536 0.312 0.152
#> GSM614425 1 0.8937 0.5819 0.540 0.308 0.152
#> GSM614426 1 0.9002 0.5747 0.532 0.312 0.156
#> GSM614427 1 0.9046 0.5728 0.528 0.312 0.160
#> GSM614428 1 0.9092 0.5727 0.532 0.296 0.172
#> GSM614429 2 0.0000 0.8856 0.000 1.000 0.000
#> GSM614430 2 0.0000 0.8856 0.000 1.000 0.000
#> GSM614431 2 0.0000 0.8856 0.000 1.000 0.000
#> GSM614432 2 0.0000 0.8856 0.000 1.000 0.000
#> GSM614433 2 0.0424 0.8821 0.000 0.992 0.008
#> GSM614434 2 0.0000 0.8856 0.000 1.000 0.000
#> GSM614435 2 0.0000 0.8856 0.000 1.000 0.000
#> GSM614436 2 0.6295 0.6078 0.164 0.764 0.072
#> GSM614437 3 0.3918 0.7440 0.004 0.140 0.856
#> GSM614438 3 0.3267 0.7656 0.000 0.116 0.884
#> GSM614439 3 0.3267 0.7656 0.000 0.116 0.884
#> GSM614440 3 0.3267 0.7656 0.000 0.116 0.884
#> GSM614441 3 0.3267 0.7656 0.000 0.116 0.884
#> GSM614442 3 0.3267 0.7656 0.000 0.116 0.884
#> GSM614443 3 0.3500 0.7640 0.004 0.116 0.880
#> GSM614444 3 0.3267 0.7656 0.000 0.116 0.884
#> GSM614391 1 0.0592 0.7340 0.988 0.000 0.012
#> GSM614392 1 0.0592 0.7340 0.988 0.000 0.012
#> GSM614393 1 0.0592 0.7340 0.988 0.000 0.012
#> GSM614394 1 0.0237 0.7337 0.996 0.000 0.004
#> GSM614395 1 0.4002 0.6323 0.840 0.000 0.160
#> GSM614396 1 0.0424 0.7326 0.992 0.000 0.008
#> GSM614397 1 0.4002 0.6323 0.840 0.000 0.160
#> GSM614398 1 0.3686 0.6511 0.860 0.000 0.140
#> GSM614399 1 0.7298 0.7228 0.700 0.200 0.100
#> GSM614400 1 0.6424 0.7414 0.752 0.180 0.068
#> GSM614401 1 0.7323 0.7233 0.700 0.196 0.104
#> GSM614402 1 0.7923 0.6991 0.652 0.228 0.120
#> GSM614403 1 0.8784 0.5935 0.548 0.316 0.136
#> GSM614404 1 0.7368 0.7211 0.696 0.200 0.104
#> GSM614405 1 0.8872 0.6011 0.552 0.296 0.152
#> GSM614406 3 0.9857 -0.0560 0.308 0.276 0.416
#> GSM614407 1 0.2356 0.7653 0.928 0.072 0.000
#> GSM614408 1 0.2356 0.7653 0.928 0.072 0.000
#> GSM614409 1 0.2356 0.7653 0.928 0.072 0.000
#> GSM614410 1 0.2356 0.7653 0.928 0.072 0.000
#> GSM614411 1 0.2356 0.7653 0.928 0.072 0.000
#> GSM614412 1 0.2711 0.7677 0.912 0.088 0.000
#> GSM614413 1 0.3445 0.7690 0.896 0.088 0.016
#> GSM614414 1 0.3112 0.7692 0.900 0.096 0.004
#> GSM614445 2 0.8690 -0.3963 0.440 0.456 0.104
#> GSM614446 1 0.8586 0.5205 0.520 0.376 0.104
#> GSM614447 2 0.8425 -0.0805 0.348 0.552 0.100
#> GSM614448 1 0.9111 0.5562 0.532 0.292 0.176
#> GSM614449 1 0.8890 0.5603 0.532 0.328 0.140
#> GSM614450 1 0.8841 0.5541 0.528 0.340 0.132
#> GSM614451 3 0.9531 0.0472 0.344 0.200 0.456
#> GSM614452 3 0.9541 0.0333 0.348 0.200 0.452
#> GSM614453 2 0.1170 0.8776 0.008 0.976 0.016
#> GSM614454 2 0.1170 0.8776 0.008 0.976 0.016
#> GSM614455 2 0.1170 0.8776 0.008 0.976 0.016
#> GSM614456 2 0.1170 0.8776 0.008 0.976 0.016
#> GSM614457 2 0.1832 0.8554 0.008 0.956 0.036
#> GSM614458 2 0.0424 0.8850 0.008 0.992 0.000
#> GSM614459 2 0.1832 0.8554 0.008 0.956 0.036
#> GSM614460 2 0.1170 0.8776 0.008 0.976 0.016
#> GSM614461 2 0.0475 0.8853 0.004 0.992 0.004
#> GSM614462 2 0.0661 0.8838 0.008 0.988 0.004
#> GSM614463 2 0.0661 0.8838 0.008 0.988 0.004
#> GSM614464 2 0.0661 0.8839 0.008 0.988 0.004
#> GSM614465 2 0.0661 0.8833 0.008 0.988 0.004
#> GSM614466 2 0.0592 0.8822 0.000 0.988 0.012
#> GSM614467 2 0.3921 0.7472 0.112 0.872 0.016
#> GSM614468 2 0.5538 0.6767 0.116 0.812 0.072
#> GSM614469 1 0.5514 0.7515 0.800 0.156 0.044
#> GSM614470 1 0.5454 0.7534 0.804 0.152 0.044
#> GSM614471 1 0.5514 0.7515 0.800 0.156 0.044
#> GSM614472 1 0.5454 0.7534 0.804 0.152 0.044
#> GSM614473 1 0.5514 0.7515 0.800 0.156 0.044
#> GSM614474 1 0.5514 0.7515 0.800 0.156 0.044
#> GSM614475 1 0.5514 0.7515 0.800 0.156 0.044
#> GSM614476 1 0.4558 0.7667 0.856 0.100 0.044
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM614415 1 0.0817 0.7398 0.976 0.000 0.024 0.000
#> GSM614416 1 0.0707 0.7404 0.980 0.000 0.020 0.000
#> GSM614417 1 0.0707 0.7404 0.980 0.000 0.020 0.000
#> GSM614418 1 0.0707 0.7404 0.980 0.000 0.020 0.000
#> GSM614419 1 0.1022 0.7376 0.968 0.000 0.032 0.000
#> GSM614420 1 0.1022 0.7376 0.968 0.000 0.032 0.000
#> GSM614421 3 0.1716 0.8230 0.000 0.064 0.936 0.000
#> GSM614422 3 0.1716 0.8234 0.000 0.064 0.936 0.000
#> GSM614423 3 0.1867 0.8223 0.000 0.072 0.928 0.000
#> GSM614424 3 0.1792 0.8232 0.000 0.068 0.932 0.000
#> GSM614425 3 0.1557 0.8207 0.000 0.056 0.944 0.000
#> GSM614426 3 0.1637 0.8221 0.000 0.060 0.940 0.000
#> GSM614427 3 0.1716 0.8235 0.000 0.064 0.936 0.000
#> GSM614428 3 0.1637 0.8227 0.000 0.060 0.940 0.000
#> GSM614429 2 0.4037 0.8411 0.000 0.832 0.112 0.056
#> GSM614430 2 0.3384 0.8509 0.000 0.860 0.116 0.024
#> GSM614431 2 0.2593 0.8482 0.000 0.892 0.104 0.004
#> GSM614432 2 0.2589 0.8519 0.000 0.884 0.116 0.000
#> GSM614433 2 0.2589 0.8519 0.000 0.884 0.116 0.000
#> GSM614434 2 0.2773 0.8525 0.000 0.880 0.116 0.004
#> GSM614435 2 0.4401 0.8322 0.000 0.812 0.112 0.076
#> GSM614436 2 0.8898 0.1374 0.152 0.408 0.352 0.088
#> GSM614437 4 0.1557 1.0000 0.000 0.056 0.000 0.944
#> GSM614438 4 0.1557 1.0000 0.000 0.056 0.000 0.944
#> GSM614439 4 0.1557 1.0000 0.000 0.056 0.000 0.944
#> GSM614440 4 0.1557 1.0000 0.000 0.056 0.000 0.944
#> GSM614441 4 0.1557 1.0000 0.000 0.056 0.000 0.944
#> GSM614442 4 0.1557 1.0000 0.000 0.056 0.000 0.944
#> GSM614443 4 0.1557 1.0000 0.000 0.056 0.000 0.944
#> GSM614444 4 0.1557 1.0000 0.000 0.056 0.000 0.944
#> GSM614391 1 0.1022 0.7376 0.968 0.000 0.032 0.000
#> GSM614392 1 0.0817 0.7398 0.976 0.000 0.024 0.000
#> GSM614393 1 0.0921 0.7388 0.972 0.000 0.028 0.000
#> GSM614394 1 0.1118 0.7358 0.964 0.000 0.036 0.000
#> GSM614395 1 0.2345 0.7105 0.900 0.000 0.100 0.000
#> GSM614396 1 0.1474 0.7319 0.948 0.000 0.052 0.000
#> GSM614397 1 0.2345 0.7105 0.900 0.000 0.100 0.000
#> GSM614398 1 0.2345 0.7105 0.900 0.000 0.100 0.000
#> GSM614399 1 0.8322 0.5941 0.492 0.300 0.156 0.052
#> GSM614400 1 0.8226 0.6206 0.512 0.284 0.152 0.052
#> GSM614401 1 0.8878 0.4295 0.408 0.244 0.292 0.056
#> GSM614402 3 0.8600 -0.0346 0.312 0.200 0.440 0.048
#> GSM614403 3 0.5979 0.6668 0.156 0.064 0.736 0.044
#> GSM614404 1 0.8483 0.5637 0.488 0.260 0.200 0.052
#> GSM614405 3 0.8126 0.3191 0.336 0.136 0.484 0.044
#> GSM614406 3 0.6822 0.6128 0.248 0.140 0.608 0.004
#> GSM614407 1 0.7100 0.7245 0.620 0.240 0.112 0.028
#> GSM614408 1 0.6396 0.7402 0.696 0.168 0.112 0.024
#> GSM614409 1 0.5653 0.7459 0.756 0.092 0.128 0.024
#> GSM614410 1 0.7100 0.7245 0.620 0.240 0.112 0.028
#> GSM614411 1 0.6373 0.7417 0.700 0.156 0.120 0.024
#> GSM614412 1 0.5685 0.6870 0.712 0.036 0.228 0.024
#> GSM614413 1 0.5870 0.6628 0.688 0.036 0.252 0.024
#> GSM614414 1 0.5870 0.6628 0.688 0.036 0.252 0.024
#> GSM614445 3 0.3903 0.7710 0.008 0.156 0.824 0.012
#> GSM614446 3 0.3508 0.7893 0.004 0.136 0.848 0.012
#> GSM614447 3 0.3672 0.7900 0.012 0.128 0.848 0.012
#> GSM614448 3 0.3547 0.7775 0.000 0.144 0.840 0.016
#> GSM614449 3 0.3662 0.7759 0.004 0.148 0.836 0.012
#> GSM614450 3 0.3805 0.7765 0.008 0.148 0.832 0.012
#> GSM614451 3 0.6070 0.6788 0.188 0.076 0.712 0.024
#> GSM614452 3 0.6031 0.6842 0.184 0.076 0.716 0.024
#> GSM614453 2 0.2402 0.8067 0.000 0.912 0.012 0.076
#> GSM614454 2 0.2522 0.8107 0.000 0.908 0.016 0.076
#> GSM614455 2 0.2522 0.8107 0.000 0.908 0.016 0.076
#> GSM614456 2 0.2775 0.8114 0.000 0.896 0.020 0.084
#> GSM614457 2 0.2882 0.8142 0.000 0.892 0.024 0.084
#> GSM614458 2 0.3037 0.8290 0.000 0.888 0.036 0.076
#> GSM614459 2 0.2882 0.8142 0.000 0.892 0.024 0.084
#> GSM614460 2 0.2775 0.8114 0.000 0.896 0.020 0.084
#> GSM614461 2 0.2589 0.8519 0.000 0.884 0.116 0.000
#> GSM614462 2 0.2589 0.8519 0.000 0.884 0.116 0.000
#> GSM614463 2 0.2888 0.8460 0.004 0.872 0.124 0.000
#> GSM614464 2 0.2773 0.8512 0.004 0.880 0.116 0.000
#> GSM614465 2 0.3224 0.8519 0.000 0.864 0.120 0.016
#> GSM614466 2 0.2773 0.8502 0.000 0.880 0.116 0.004
#> GSM614467 2 0.8152 0.3275 0.096 0.492 0.340 0.072
#> GSM614468 2 0.6827 0.2775 0.068 0.548 0.368 0.016
#> GSM614469 1 0.7603 0.6959 0.576 0.276 0.092 0.056
#> GSM614470 1 0.7603 0.6959 0.576 0.276 0.092 0.056
#> GSM614471 1 0.7603 0.6959 0.576 0.276 0.092 0.056
#> GSM614472 1 0.7603 0.6959 0.576 0.276 0.092 0.056
#> GSM614473 1 0.7603 0.6959 0.576 0.276 0.092 0.056
#> GSM614474 1 0.7559 0.6984 0.584 0.268 0.092 0.056
#> GSM614475 1 0.7603 0.6940 0.576 0.276 0.092 0.056
#> GSM614476 1 0.8133 0.6349 0.552 0.184 0.208 0.056
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM614415 5 0.1662 0.814 0.056 0.004 0.004 0.000 0.936
#> GSM614416 5 0.1731 0.813 0.060 0.004 0.004 0.000 0.932
#> GSM614417 5 0.1731 0.813 0.060 0.004 0.004 0.000 0.932
#> GSM614418 5 0.1731 0.813 0.060 0.004 0.004 0.000 0.932
#> GSM614419 5 0.1568 0.824 0.036 0.000 0.020 0.000 0.944
#> GSM614420 5 0.1485 0.823 0.032 0.000 0.020 0.000 0.948
#> GSM614421 3 0.1893 0.830 0.048 0.024 0.928 0.000 0.000
#> GSM614422 3 0.1893 0.830 0.048 0.024 0.928 0.000 0.000
#> GSM614423 3 0.1471 0.834 0.020 0.024 0.952 0.000 0.004
#> GSM614424 3 0.1661 0.833 0.036 0.024 0.940 0.000 0.000
#> GSM614425 3 0.1893 0.830 0.048 0.024 0.928 0.000 0.000
#> GSM614426 3 0.1893 0.830 0.048 0.024 0.928 0.000 0.000
#> GSM614427 3 0.1243 0.833 0.008 0.028 0.960 0.000 0.004
#> GSM614428 3 0.1026 0.832 0.004 0.024 0.968 0.000 0.004
#> GSM614429 2 0.0000 0.924 0.000 1.000 0.000 0.000 0.000
#> GSM614430 2 0.0000 0.924 0.000 1.000 0.000 0.000 0.000
#> GSM614431 2 0.0703 0.918 0.024 0.976 0.000 0.000 0.000
#> GSM614432 2 0.0162 0.924 0.004 0.996 0.000 0.000 0.000
#> GSM614433 2 0.0794 0.919 0.000 0.972 0.028 0.000 0.000
#> GSM614434 2 0.0000 0.924 0.000 1.000 0.000 0.000 0.000
#> GSM614435 2 0.1282 0.912 0.000 0.952 0.044 0.004 0.000
#> GSM614436 2 0.4100 0.707 0.004 0.760 0.212 0.004 0.020
#> GSM614437 4 0.0162 0.995 0.000 0.004 0.000 0.996 0.000
#> GSM614438 4 0.0000 0.999 0.000 0.000 0.000 1.000 0.000
#> GSM614439 4 0.0000 0.999 0.000 0.000 0.000 1.000 0.000
#> GSM614440 4 0.0000 0.999 0.000 0.000 0.000 1.000 0.000
#> GSM614441 4 0.0000 0.999 0.000 0.000 0.000 1.000 0.000
#> GSM614442 4 0.0000 0.999 0.000 0.000 0.000 1.000 0.000
#> GSM614443 4 0.0000 0.999 0.000 0.000 0.000 1.000 0.000
#> GSM614444 4 0.0000 0.999 0.000 0.000 0.000 1.000 0.000
#> GSM614391 5 0.1012 0.813 0.012 0.000 0.020 0.000 0.968
#> GSM614392 5 0.1728 0.823 0.036 0.004 0.020 0.000 0.940
#> GSM614393 5 0.1646 0.824 0.032 0.004 0.020 0.000 0.944
#> GSM614394 5 0.1399 0.823 0.028 0.000 0.020 0.000 0.952
#> GSM614395 5 0.1484 0.801 0.008 0.000 0.048 0.000 0.944
#> GSM614396 5 0.1725 0.823 0.044 0.000 0.020 0.000 0.936
#> GSM614397 5 0.1251 0.806 0.008 0.000 0.036 0.000 0.956
#> GSM614398 5 0.1251 0.806 0.008 0.000 0.036 0.000 0.956
#> GSM614399 1 0.5466 0.558 0.628 0.072 0.292 0.000 0.008
#> GSM614400 1 0.5424 0.593 0.652 0.072 0.264 0.000 0.012
#> GSM614401 1 0.6037 0.205 0.472 0.040 0.448 0.000 0.040
#> GSM614402 3 0.6028 -0.139 0.432 0.040 0.488 0.000 0.040
#> GSM614403 3 0.2857 0.802 0.064 0.028 0.888 0.000 0.020
#> GSM614404 1 0.5732 0.561 0.620 0.072 0.288 0.000 0.020
#> GSM614405 3 0.3348 0.787 0.068 0.036 0.864 0.000 0.032
#> GSM614406 3 0.2644 0.818 0.008 0.036 0.896 0.000 0.060
#> GSM614407 1 0.4347 0.227 0.636 0.004 0.004 0.000 0.356
#> GSM614408 5 0.4594 0.237 0.484 0.004 0.004 0.000 0.508
#> GSM614409 5 0.4670 0.348 0.440 0.004 0.008 0.000 0.548
#> GSM614410 1 0.4478 0.212 0.628 0.008 0.004 0.000 0.360
#> GSM614411 5 0.4583 0.291 0.464 0.004 0.004 0.000 0.528
#> GSM614412 5 0.5691 0.503 0.296 0.000 0.112 0.000 0.592
#> GSM614413 5 0.6142 0.477 0.184 0.004 0.232 0.000 0.580
#> GSM614414 5 0.6123 0.506 0.224 0.004 0.188 0.000 0.584
#> GSM614445 3 0.4235 0.562 0.000 0.336 0.656 0.000 0.008
#> GSM614446 3 0.4397 0.665 0.024 0.264 0.708 0.000 0.004
#> GSM614447 3 0.5449 0.543 0.068 0.328 0.600 0.000 0.004
#> GSM614448 3 0.2646 0.802 0.004 0.124 0.868 0.000 0.004
#> GSM614449 3 0.2536 0.800 0.000 0.128 0.868 0.000 0.004
#> GSM614450 3 0.2646 0.802 0.004 0.124 0.868 0.000 0.004
#> GSM614451 3 0.2948 0.783 0.004 0.008 0.884 0.040 0.064
#> GSM614452 3 0.2948 0.783 0.004 0.008 0.884 0.040 0.064
#> GSM614453 2 0.2604 0.897 0.108 0.880 0.004 0.004 0.004
#> GSM614454 2 0.2445 0.897 0.108 0.884 0.000 0.004 0.004
#> GSM614455 2 0.2445 0.897 0.108 0.884 0.000 0.004 0.004
#> GSM614456 2 0.2445 0.897 0.108 0.884 0.000 0.004 0.004
#> GSM614457 2 0.2445 0.897 0.108 0.884 0.000 0.004 0.004
#> GSM614458 2 0.2747 0.908 0.060 0.888 0.048 0.004 0.000
#> GSM614459 2 0.2445 0.897 0.108 0.884 0.000 0.004 0.004
#> GSM614460 2 0.2445 0.897 0.108 0.884 0.000 0.004 0.004
#> GSM614461 2 0.0162 0.924 0.004 0.996 0.000 0.000 0.000
#> GSM614462 2 0.0162 0.924 0.004 0.996 0.000 0.000 0.000
#> GSM614463 2 0.0798 0.923 0.008 0.976 0.016 0.000 0.000
#> GSM614464 2 0.0404 0.924 0.012 0.988 0.000 0.000 0.000
#> GSM614465 2 0.1357 0.912 0.004 0.948 0.048 0.000 0.000
#> GSM614466 2 0.0566 0.924 0.004 0.984 0.012 0.000 0.000
#> GSM614467 2 0.3409 0.783 0.024 0.816 0.160 0.000 0.000
#> GSM614468 2 0.3651 0.783 0.032 0.812 0.152 0.000 0.004
#> GSM614469 1 0.0324 0.757 0.992 0.004 0.004 0.000 0.000
#> GSM614470 1 0.0324 0.757 0.992 0.004 0.004 0.000 0.000
#> GSM614471 1 0.0671 0.759 0.980 0.004 0.016 0.000 0.000
#> GSM614472 1 0.0324 0.757 0.992 0.004 0.004 0.000 0.000
#> GSM614473 1 0.0324 0.757 0.992 0.004 0.004 0.000 0.000
#> GSM614474 1 0.1059 0.758 0.968 0.008 0.020 0.000 0.004
#> GSM614475 1 0.1569 0.756 0.944 0.008 0.044 0.000 0.004
#> GSM614476 1 0.5162 0.577 0.692 0.048 0.236 0.000 0.024
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM614415 5 0.1334 0.8015 0.020 0.000 0.000 0.000 0.948 NA
#> GSM614416 5 0.1408 0.8014 0.020 0.000 0.000 0.000 0.944 NA
#> GSM614417 5 0.1408 0.7997 0.020 0.000 0.000 0.000 0.944 NA
#> GSM614418 5 0.1408 0.7997 0.020 0.000 0.000 0.000 0.944 NA
#> GSM614419 5 0.0748 0.8135 0.004 0.000 0.004 0.000 0.976 NA
#> GSM614420 5 0.0551 0.8127 0.004 0.000 0.004 0.000 0.984 NA
#> GSM614421 3 0.1053 0.7891 0.000 0.012 0.964 0.000 0.004 NA
#> GSM614422 3 0.1053 0.7891 0.000 0.012 0.964 0.000 0.004 NA
#> GSM614423 3 0.1837 0.7909 0.004 0.012 0.932 0.000 0.020 NA
#> GSM614424 3 0.0767 0.7915 0.000 0.012 0.976 0.000 0.004 NA
#> GSM614425 3 0.1053 0.7891 0.000 0.012 0.964 0.000 0.004 NA
#> GSM614426 3 0.1218 0.7901 0.000 0.012 0.956 0.000 0.004 NA
#> GSM614427 3 0.1180 0.7951 0.000 0.016 0.960 0.000 0.012 NA
#> GSM614428 3 0.1180 0.7933 0.000 0.016 0.960 0.000 0.012 NA
#> GSM614429 2 0.0291 0.8331 0.000 0.992 0.004 0.000 0.000 NA
#> GSM614430 2 0.0291 0.8331 0.000 0.992 0.004 0.000 0.000 NA
#> GSM614431 2 0.1616 0.8320 0.012 0.940 0.020 0.000 0.000 NA
#> GSM614432 2 0.1088 0.8299 0.000 0.960 0.016 0.000 0.000 NA
#> GSM614433 2 0.0603 0.8315 0.000 0.980 0.016 0.000 0.000 NA
#> GSM614434 2 0.0508 0.8333 0.000 0.984 0.012 0.000 0.000 NA
#> GSM614435 2 0.2471 0.8171 0.004 0.888 0.056 0.000 0.000 NA
#> GSM614436 2 0.4715 0.6210 0.012 0.692 0.212 0.000 0.000 NA
#> GSM614437 4 0.1267 0.9559 0.000 0.000 0.000 0.940 0.000 NA
#> GSM614438 4 0.0146 0.9838 0.000 0.000 0.000 0.996 0.000 NA
#> GSM614439 4 0.0000 0.9850 0.000 0.000 0.000 1.000 0.000 NA
#> GSM614440 4 0.0000 0.9850 0.000 0.000 0.000 1.000 0.000 NA
#> GSM614441 4 0.0000 0.9850 0.000 0.000 0.000 1.000 0.000 NA
#> GSM614442 4 0.0000 0.9850 0.000 0.000 0.000 1.000 0.000 NA
#> GSM614443 4 0.1267 0.9559 0.000 0.000 0.000 0.940 0.000 NA
#> GSM614444 4 0.0000 0.9850 0.000 0.000 0.000 1.000 0.000 NA
#> GSM614391 5 0.1152 0.8121 0.000 0.000 0.004 0.000 0.952 NA
#> GSM614392 5 0.1152 0.8121 0.000 0.000 0.004 0.000 0.952 NA
#> GSM614393 5 0.0748 0.8152 0.004 0.000 0.004 0.000 0.976 NA
#> GSM614394 5 0.1265 0.8117 0.000 0.000 0.008 0.000 0.948 NA
#> GSM614395 5 0.3543 0.7264 0.000 0.000 0.032 0.000 0.768 NA
#> GSM614396 5 0.1633 0.8099 0.000 0.000 0.024 0.000 0.932 NA
#> GSM614397 5 0.3418 0.7354 0.000 0.000 0.032 0.000 0.784 NA
#> GSM614398 5 0.3319 0.7516 0.000 0.000 0.036 0.000 0.800 NA
#> GSM614399 1 0.7139 0.1519 0.452 0.100 0.232 0.000 0.004 NA
#> GSM614400 1 0.7055 0.1872 0.472 0.100 0.220 0.000 0.004 NA
#> GSM614401 3 0.7234 0.1523 0.340 0.060 0.376 0.000 0.016 NA
#> GSM614402 3 0.6798 0.4100 0.280 0.048 0.496 0.000 0.020 NA
#> GSM614403 3 0.4595 0.7326 0.124 0.012 0.740 0.000 0.008 NA
#> GSM614404 1 0.7088 0.1699 0.464 0.100 0.228 0.000 0.004 NA
#> GSM614405 3 0.4701 0.7401 0.108 0.020 0.748 0.000 0.016 NA
#> GSM614406 3 0.4048 0.7649 0.072 0.016 0.800 0.000 0.016 NA
#> GSM614407 1 0.5817 0.2742 0.500 0.000 0.012 0.000 0.348 NA
#> GSM614408 1 0.5902 0.1934 0.456 0.000 0.012 0.000 0.388 NA
#> GSM614409 1 0.5876 0.1999 0.460 0.000 0.012 0.000 0.388 NA
#> GSM614410 1 0.5809 0.2808 0.504 0.000 0.012 0.000 0.344 NA
#> GSM614411 1 0.5902 0.1934 0.456 0.000 0.012 0.000 0.388 NA
#> GSM614412 5 0.7091 0.0366 0.296 0.000 0.164 0.000 0.424 NA
#> GSM614413 5 0.7132 0.1250 0.236 0.000 0.252 0.000 0.416 NA
#> GSM614414 5 0.7222 0.1209 0.244 0.000 0.224 0.000 0.416 NA
#> GSM614445 3 0.5385 0.5289 0.016 0.296 0.608 0.000 0.012 NA
#> GSM614446 3 0.5434 0.6197 0.032 0.244 0.644 0.000 0.012 NA
#> GSM614447 3 0.6156 0.5486 0.072 0.280 0.568 0.000 0.012 NA
#> GSM614448 3 0.4184 0.7752 0.072 0.048 0.800 0.000 0.012 NA
#> GSM614449 3 0.4305 0.7732 0.072 0.056 0.792 0.000 0.012 NA
#> GSM614450 3 0.4361 0.7730 0.076 0.060 0.788 0.000 0.012 NA
#> GSM614451 3 0.4743 0.7084 0.060 0.004 0.700 0.000 0.020 NA
#> GSM614452 3 0.4738 0.7097 0.060 0.004 0.692 0.000 0.016 NA
#> GSM614453 2 0.3922 0.7446 0.016 0.664 0.000 0.000 0.000 NA
#> GSM614454 2 0.3953 0.7408 0.016 0.656 0.000 0.000 0.000 NA
#> GSM614455 2 0.3953 0.7408 0.016 0.656 0.000 0.000 0.000 NA
#> GSM614456 2 0.4176 0.6964 0.016 0.580 0.000 0.000 0.000 NA
#> GSM614457 2 0.4176 0.6964 0.016 0.580 0.000 0.000 0.000 NA
#> GSM614458 2 0.3837 0.7913 0.008 0.768 0.044 0.000 0.000 NA
#> GSM614459 2 0.4176 0.6964 0.016 0.580 0.000 0.000 0.000 NA
#> GSM614460 2 0.4176 0.6964 0.016 0.580 0.000 0.000 0.000 NA
#> GSM614461 2 0.0914 0.8308 0.000 0.968 0.016 0.000 0.000 NA
#> GSM614462 2 0.1528 0.8263 0.012 0.944 0.016 0.000 0.000 NA
#> GSM614463 2 0.1630 0.8250 0.020 0.940 0.016 0.000 0.000 NA
#> GSM614464 2 0.2432 0.8035 0.020 0.892 0.016 0.000 0.000 NA
#> GSM614465 2 0.3343 0.7561 0.004 0.812 0.144 0.000 0.000 NA
#> GSM614466 2 0.1616 0.8247 0.000 0.932 0.020 0.000 0.000 NA
#> GSM614467 2 0.2437 0.8007 0.004 0.888 0.072 0.000 0.000 NA
#> GSM614468 2 0.3730 0.7508 0.052 0.812 0.104 0.000 0.000 NA
#> GSM614469 1 0.0146 0.6758 0.996 0.000 0.000 0.000 0.000 NA
#> GSM614470 1 0.0146 0.6758 0.996 0.000 0.000 0.000 0.000 NA
#> GSM614471 1 0.0146 0.6758 0.996 0.000 0.000 0.000 0.000 NA
#> GSM614472 1 0.0146 0.6758 0.996 0.000 0.000 0.000 0.000 NA
#> GSM614473 1 0.0146 0.6758 0.996 0.000 0.000 0.000 0.000 NA
#> GSM614474 1 0.0291 0.6744 0.992 0.004 0.004 0.000 0.000 NA
#> GSM614475 1 0.1116 0.6649 0.960 0.028 0.008 0.000 0.000 NA
#> GSM614476 1 0.3505 0.5568 0.808 0.008 0.136 0.000 0.000 NA
Heatmaps for the consensus matrix. It visualizes the probability of two samples to be in a same group.
consensus_heatmap(res, k = 2)
consensus_heatmap(res, k = 3)
consensus_heatmap(res, k = 4)
consensus_heatmap(res, k = 5)
consensus_heatmap(res, k = 6)
Heatmaps for the membership of samples in all partitions to see how consistent they are:
membership_heatmap(res, k = 2)
membership_heatmap(res, k = 3)
membership_heatmap(res, k = 4)
membership_heatmap(res, k = 5)
membership_heatmap(res, k = 6)
As soon as we have had the classes for columns, we can look for signatures which are significantly different between classes which can be candidate marks for certain classes. Following are the heatmaps for signatures.
Signature heatmaps where rows are scaled:
get_signatures(res, k = 2)
get_signatures(res, k = 3)
get_signatures(res, k = 4)
get_signatures(res, k = 5)
get_signatures(res, k = 6)
Signature heatmaps where rows are not scaled:
get_signatures(res, k = 2, scale_rows = FALSE)
get_signatures(res, k = 3, scale_rows = FALSE)
get_signatures(res, k = 4, scale_rows = FALSE)
get_signatures(res, k = 5, scale_rows = FALSE)
get_signatures(res, k = 6, scale_rows = FALSE)
Compare the overlap of signatures from different k:
compare_signatures(res)
get_signature() returns a data frame invisibly. TO get the list of signatures, the function
call should be assigned to a variable explicitly. In following code, if plot argument is set
to FALSE, no heatmap is plotted while only the differential analysis is performed.
# code only for demonstration
tb = get_signature(res, k = ..., plot = FALSE)
An example of the output of tb is:
#> which_row fdr mean_1 mean_2 scaled_mean_1 scaled_mean_2 km
#> 1 38 0.042760348 8.373488 9.131774 -0.5533452 0.5164555 1
#> 2 40 0.018707592 7.106213 8.469186 -0.6173731 0.5762149 1
#> 3 55 0.019134737 10.221463 11.207825 -0.6159697 0.5749050 1
#> 4 59 0.006059896 5.921854 7.869574 -0.6899429 0.6439467 1
#> 5 60 0.018055526 8.928898 10.211722 -0.6204761 0.5791110 1
#> 6 98 0.009384629 15.714769 14.887706 0.6635654 -0.6193277 2
...
The columns in tb are:
which_row: row indices corresponding to the input matrix.fdr: FDR for the differential test. mean_x: The mean value in group x.scaled_mean_x: The mean value in group x after rows are scaled.km: Row groups if k-means clustering is applied to rows.UMAP plot which shows how samples are separated.
dimension_reduction(res, k = 2, method = "UMAP")
dimension_reduction(res, k = 3, method = "UMAP")
dimension_reduction(res, k = 4, method = "UMAP")
dimension_reduction(res, k = 5, method = "UMAP")
dimension_reduction(res, k = 6, method = "UMAP")
Following heatmap shows how subgroups are split when increasing k:
collect_classes(res)
Test correlation between subgroups and known annotations. If the known annotation is numeric, one-way ANOVA test is applied, and if the known annotation is discrete, chi-squared contingency table test is applied.
test_to_known_factors(res)
#> n individual(p) protocol(p) time(p) other(p) k
#> ATC:mclust 77 3.13e-12 0.536 0.991 0.46354 2
#> ATC:mclust 81 3.08e-24 0.774 1.000 0.00234 3
#> ATC:mclust 80 4.85e-33 0.729 1.000 0.03001 4
#> ATC:mclust 78 1.21e-40 0.903 1.000 0.04818 5
#> ATC:mclust 73 7.75e-42 0.861 1.000 0.05138 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 21168 rows and 86 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 0.999 0.962 0.984 0.4681 0.534 0.534
#> 3 3 0.723 0.814 0.915 0.3622 0.735 0.541
#> 4 4 0.766 0.804 0.912 0.1146 0.873 0.673
#> 5 5 0.610 0.498 0.713 0.0676 0.880 0.650
#> 6 6 0.704 0.650 0.771 0.0367 0.907 0.703
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
#> GSM614415 2 0.0000 0.984 0.000 1.000
#> GSM614416 2 0.0000 0.984 0.000 1.000
#> GSM614417 2 0.0000 0.984 0.000 1.000
#> GSM614418 2 0.0000 0.984 0.000 1.000
#> GSM614419 1 0.0000 0.981 1.000 0.000
#> GSM614420 1 0.0000 0.981 1.000 0.000
#> GSM614421 1 0.0000 0.981 1.000 0.000
#> GSM614422 1 0.0000 0.981 1.000 0.000
#> GSM614423 2 0.0000 0.984 0.000 1.000
#> GSM614424 1 0.0000 0.981 1.000 0.000
#> GSM614425 1 0.0000 0.981 1.000 0.000
#> GSM614426 1 0.0000 0.981 1.000 0.000
#> GSM614427 1 0.0000 0.981 1.000 0.000
#> GSM614428 1 0.0000 0.981 1.000 0.000
#> GSM614429 2 0.0000 0.984 0.000 1.000
#> GSM614430 2 0.0000 0.984 0.000 1.000
#> GSM614431 2 0.0000 0.984 0.000 1.000
#> GSM614432 2 0.0000 0.984 0.000 1.000
#> GSM614433 2 0.0000 0.984 0.000 1.000
#> GSM614434 2 0.0000 0.984 0.000 1.000
#> GSM614435 2 0.0000 0.984 0.000 1.000
#> GSM614436 2 0.6048 0.828 0.148 0.852
#> GSM614437 2 0.0376 0.981 0.004 0.996
#> GSM614438 1 0.0000 0.981 1.000 0.000
#> GSM614439 1 0.0000 0.981 1.000 0.000
#> GSM614440 1 0.0000 0.981 1.000 0.000
#> GSM614441 1 0.0000 0.981 1.000 0.000
#> GSM614442 1 0.0000 0.981 1.000 0.000
#> GSM614443 1 0.3274 0.926 0.940 0.060
#> GSM614444 1 0.0000 0.981 1.000 0.000
#> GSM614391 1 0.0000 0.981 1.000 0.000
#> GSM614392 1 0.9732 0.309 0.596 0.404
#> GSM614393 2 0.3431 0.925 0.064 0.936
#> GSM614394 1 0.0000 0.981 1.000 0.000
#> GSM614395 1 0.0000 0.981 1.000 0.000
#> GSM614396 1 0.0000 0.981 1.000 0.000
#> GSM614397 1 0.0000 0.981 1.000 0.000
#> GSM614398 1 0.0000 0.981 1.000 0.000
#> GSM614399 2 0.0000 0.984 0.000 1.000
#> GSM614400 2 0.0000 0.984 0.000 1.000
#> GSM614401 2 0.0000 0.984 0.000 1.000
#> GSM614402 2 0.0000 0.984 0.000 1.000
#> GSM614403 2 0.4562 0.891 0.096 0.904
#> GSM614404 2 0.0000 0.984 0.000 1.000
#> GSM614405 1 0.0672 0.975 0.992 0.008
#> GSM614406 1 0.0000 0.981 1.000 0.000
#> GSM614407 2 0.0000 0.984 0.000 1.000
#> GSM614408 2 0.0000 0.984 0.000 1.000
#> GSM614409 2 0.0000 0.984 0.000 1.000
#> GSM614410 2 0.0000 0.984 0.000 1.000
#> GSM614411 2 0.0000 0.984 0.000 1.000
#> GSM614412 2 0.8813 0.582 0.300 0.700
#> GSM614413 1 0.0000 0.981 1.000 0.000
#> GSM614414 1 0.0000 0.981 1.000 0.000
#> GSM614445 2 0.0000 0.984 0.000 1.000
#> GSM614446 2 0.0000 0.984 0.000 1.000
#> GSM614447 2 0.0000 0.984 0.000 1.000
#> GSM614448 1 0.0000 0.981 1.000 0.000
#> GSM614449 1 0.3733 0.913 0.928 0.072
#> GSM614450 2 0.7453 0.738 0.212 0.788
#> GSM614451 1 0.0000 0.981 1.000 0.000
#> GSM614452 1 0.0000 0.981 1.000 0.000
#> GSM614453 2 0.0000 0.984 0.000 1.000
#> GSM614454 2 0.0000 0.984 0.000 1.000
#> GSM614455 2 0.0000 0.984 0.000 1.000
#> GSM614456 2 0.0000 0.984 0.000 1.000
#> GSM614457 2 0.0000 0.984 0.000 1.000
#> GSM614458 2 0.0000 0.984 0.000 1.000
#> GSM614459 2 0.0000 0.984 0.000 1.000
#> GSM614460 2 0.0000 0.984 0.000 1.000
#> GSM614461 2 0.0000 0.984 0.000 1.000
#> GSM614462 2 0.0000 0.984 0.000 1.000
#> GSM614463 2 0.0000 0.984 0.000 1.000
#> GSM614464 2 0.0000 0.984 0.000 1.000
#> GSM614465 2 0.0000 0.984 0.000 1.000
#> GSM614466 2 0.0000 0.984 0.000 1.000
#> GSM614467 2 0.0000 0.984 0.000 1.000
#> GSM614468 2 0.0000 0.984 0.000 1.000
#> GSM614469 2 0.0000 0.984 0.000 1.000
#> GSM614470 2 0.0000 0.984 0.000 1.000
#> GSM614471 2 0.0000 0.984 0.000 1.000
#> GSM614472 2 0.0000 0.984 0.000 1.000
#> GSM614473 2 0.0000 0.984 0.000 1.000
#> GSM614474 2 0.0000 0.984 0.000 1.000
#> GSM614475 2 0.0000 0.984 0.000 1.000
#> GSM614476 2 0.1633 0.963 0.024 0.976
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM614415 1 0.0237 0.868 0.996 0.004 0.000
#> GSM614416 1 0.1031 0.865 0.976 0.024 0.000
#> GSM614417 1 0.1163 0.864 0.972 0.028 0.000
#> GSM614418 1 0.1163 0.864 0.972 0.028 0.000
#> GSM614419 1 0.0747 0.863 0.984 0.000 0.016
#> GSM614420 1 0.0592 0.864 0.988 0.000 0.012
#> GSM614421 3 0.6302 0.112 0.480 0.000 0.520
#> GSM614422 3 0.6302 0.115 0.480 0.000 0.520
#> GSM614423 1 0.5968 0.387 0.636 0.364 0.000
#> GSM614424 3 0.4654 0.666 0.208 0.000 0.792
#> GSM614425 3 0.6204 0.281 0.424 0.000 0.576
#> GSM614426 3 0.3879 0.724 0.152 0.000 0.848
#> GSM614427 3 0.1289 0.806 0.032 0.000 0.968
#> GSM614428 3 0.1411 0.804 0.036 0.000 0.964
#> GSM614429 2 0.0237 0.953 0.000 0.996 0.004
#> GSM614430 2 0.0237 0.953 0.000 0.996 0.004
#> GSM614431 2 0.0237 0.955 0.004 0.996 0.000
#> GSM614432 2 0.0237 0.955 0.004 0.996 0.000
#> GSM614433 2 0.0000 0.954 0.000 1.000 0.000
#> GSM614434 2 0.0000 0.954 0.000 1.000 0.000
#> GSM614435 2 0.0747 0.947 0.000 0.984 0.016
#> GSM614436 3 0.6062 0.379 0.000 0.384 0.616
#> GSM614437 3 0.6280 0.160 0.000 0.460 0.540
#> GSM614438 3 0.1031 0.810 0.000 0.024 0.976
#> GSM614439 3 0.0592 0.814 0.000 0.012 0.988
#> GSM614440 3 0.0424 0.814 0.000 0.008 0.992
#> GSM614441 3 0.0592 0.814 0.000 0.012 0.988
#> GSM614442 3 0.1411 0.804 0.000 0.036 0.964
#> GSM614443 3 0.4346 0.679 0.000 0.184 0.816
#> GSM614444 3 0.0592 0.814 0.000 0.012 0.988
#> GSM614391 1 0.0592 0.864 0.988 0.000 0.012
#> GSM614392 1 0.0237 0.868 0.996 0.004 0.000
#> GSM614393 1 0.0237 0.868 0.996 0.004 0.000
#> GSM614394 1 0.1529 0.850 0.960 0.000 0.040
#> GSM614395 3 0.5835 0.469 0.340 0.000 0.660
#> GSM614396 1 0.1753 0.844 0.952 0.000 0.048
#> GSM614397 1 0.5905 0.395 0.648 0.000 0.352
#> GSM614398 1 0.4974 0.637 0.764 0.000 0.236
#> GSM614399 2 0.0747 0.953 0.016 0.984 0.000
#> GSM614400 2 0.1163 0.949 0.028 0.972 0.000
#> GSM614401 2 0.1860 0.937 0.052 0.948 0.000
#> GSM614402 2 0.1289 0.948 0.032 0.968 0.000
#> GSM614403 2 0.4277 0.834 0.016 0.852 0.132
#> GSM614404 2 0.1031 0.951 0.024 0.976 0.000
#> GSM614405 3 0.0661 0.815 0.004 0.008 0.988
#> GSM614406 3 0.0237 0.814 0.004 0.000 0.996
#> GSM614407 1 0.4121 0.731 0.832 0.168 0.000
#> GSM614408 1 0.3038 0.804 0.896 0.104 0.000
#> GSM614409 1 0.0592 0.868 0.988 0.012 0.000
#> GSM614410 1 0.3551 0.774 0.868 0.132 0.000
#> GSM614411 1 0.1964 0.845 0.944 0.056 0.000
#> GSM614412 1 0.0475 0.868 0.992 0.004 0.004
#> GSM614413 1 0.5178 0.604 0.744 0.000 0.256
#> GSM614414 1 0.4235 0.720 0.824 0.000 0.176
#> GSM614445 2 0.0892 0.952 0.020 0.980 0.000
#> GSM614446 2 0.0747 0.953 0.016 0.984 0.000
#> GSM614447 2 0.0892 0.952 0.020 0.980 0.000
#> GSM614448 3 0.0237 0.814 0.004 0.000 0.996
#> GSM614449 3 0.2682 0.779 0.004 0.076 0.920
#> GSM614450 2 0.5465 0.563 0.000 0.712 0.288
#> GSM614451 3 0.0424 0.813 0.008 0.000 0.992
#> GSM614452 3 0.0424 0.813 0.008 0.000 0.992
#> GSM614453 2 0.0000 0.954 0.000 1.000 0.000
#> GSM614454 2 0.0237 0.953 0.000 0.996 0.004
#> GSM614455 2 0.0237 0.953 0.000 0.996 0.004
#> GSM614456 2 0.0237 0.953 0.000 0.996 0.004
#> GSM614457 2 0.0592 0.950 0.000 0.988 0.012
#> GSM614458 2 0.0747 0.948 0.000 0.984 0.016
#> GSM614459 2 0.0747 0.947 0.000 0.984 0.016
#> GSM614460 2 0.0237 0.953 0.000 0.996 0.004
#> GSM614461 2 0.0424 0.954 0.008 0.992 0.000
#> GSM614462 2 0.0424 0.954 0.008 0.992 0.000
#> GSM614463 2 0.0747 0.953 0.016 0.984 0.000
#> GSM614464 2 0.0000 0.954 0.000 1.000 0.000
#> GSM614465 2 0.0424 0.954 0.008 0.992 0.000
#> GSM614466 2 0.0424 0.954 0.008 0.992 0.000
#> GSM614467 2 0.0747 0.948 0.000 0.984 0.016
#> GSM614468 2 0.0000 0.954 0.000 1.000 0.000
#> GSM614469 2 0.3192 0.892 0.112 0.888 0.000
#> GSM614470 2 0.3879 0.851 0.152 0.848 0.000
#> GSM614471 2 0.3038 0.899 0.104 0.896 0.000
#> GSM614472 2 0.3267 0.889 0.116 0.884 0.000
#> GSM614473 2 0.3686 0.864 0.140 0.860 0.000
#> GSM614474 2 0.3816 0.855 0.148 0.852 0.000
#> GSM614475 2 0.3192 0.893 0.112 0.888 0.000
#> GSM614476 2 0.3572 0.911 0.060 0.900 0.040
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM614415 1 0.0000 0.9488 1.000 0.000 0.000 0.000
#> GSM614416 1 0.0000 0.9488 1.000 0.000 0.000 0.000
#> GSM614417 1 0.0000 0.9488 1.000 0.000 0.000 0.000
#> GSM614418 1 0.0000 0.9488 1.000 0.000 0.000 0.000
#> GSM614419 1 0.0000 0.9488 1.000 0.000 0.000 0.000
#> GSM614420 1 0.0000 0.9488 1.000 0.000 0.000 0.000
#> GSM614421 3 0.0707 0.9134 0.000 0.020 0.980 0.000
#> GSM614422 3 0.0592 0.9146 0.000 0.016 0.984 0.000
#> GSM614423 2 0.3123 0.7577 0.000 0.844 0.156 0.000
#> GSM614424 3 0.1389 0.8944 0.000 0.048 0.952 0.000
#> GSM614425 3 0.0707 0.9134 0.000 0.020 0.980 0.000
#> GSM614426 3 0.0469 0.9154 0.000 0.012 0.988 0.000
#> GSM614427 3 0.0188 0.9151 0.000 0.004 0.996 0.000
#> GSM614428 3 0.0336 0.9143 0.000 0.000 0.992 0.008
#> GSM614429 2 0.0336 0.8849 0.000 0.992 0.000 0.008
#> GSM614430 2 0.0000 0.8877 0.000 1.000 0.000 0.000
#> GSM614431 2 0.0000 0.8877 0.000 1.000 0.000 0.000
#> GSM614432 2 0.0000 0.8877 0.000 1.000 0.000 0.000
#> GSM614433 2 0.0000 0.8877 0.000 1.000 0.000 0.000
#> GSM614434 2 0.0000 0.8877 0.000 1.000 0.000 0.000
#> GSM614435 2 0.2345 0.8237 0.000 0.900 0.000 0.100
#> GSM614436 4 0.5750 0.2849 0.000 0.440 0.028 0.532
#> GSM614437 4 0.0336 0.7738 0.000 0.008 0.000 0.992
#> GSM614438 4 0.1118 0.7725 0.000 0.000 0.036 0.964
#> GSM614439 4 0.1637 0.7642 0.000 0.000 0.060 0.940
#> GSM614440 4 0.2216 0.7345 0.000 0.000 0.092 0.908
#> GSM614441 4 0.1557 0.7663 0.000 0.000 0.056 0.944
#> GSM614442 4 0.0817 0.7739 0.000 0.000 0.024 0.976
#> GSM614443 4 0.0000 0.7732 0.000 0.000 0.000 1.000
#> GSM614444 4 0.1637 0.7642 0.000 0.000 0.060 0.940
#> GSM614391 1 0.0000 0.9488 1.000 0.000 0.000 0.000
#> GSM614392 1 0.0000 0.9488 1.000 0.000 0.000 0.000
#> GSM614393 1 0.0000 0.9488 1.000 0.000 0.000 0.000
#> GSM614394 1 0.2011 0.8906 0.920 0.000 0.080 0.000
#> GSM614395 3 0.0707 0.9112 0.000 0.000 0.980 0.020
#> GSM614396 1 0.2704 0.8480 0.876 0.000 0.124 0.000
#> GSM614397 3 0.5488 0.0848 0.452 0.000 0.532 0.016
#> GSM614398 1 0.5329 0.2437 0.568 0.000 0.420 0.012
#> GSM614399 2 0.0188 0.8865 0.000 0.996 0.000 0.004
#> GSM614400 2 0.0188 0.8865 0.000 0.996 0.000 0.004
#> GSM614401 2 0.0000 0.8877 0.000 1.000 0.000 0.000
#> GSM614402 2 0.0188 0.8868 0.000 0.996 0.004 0.000
#> GSM614403 2 0.3257 0.7581 0.000 0.844 0.152 0.004
#> GSM614404 2 0.0000 0.8877 0.000 1.000 0.000 0.000
#> GSM614405 3 0.1182 0.9146 0.000 0.016 0.968 0.016
#> GSM614406 3 0.1302 0.9042 0.000 0.000 0.956 0.044
#> GSM614407 1 0.0000 0.9488 1.000 0.000 0.000 0.000
#> GSM614408 1 0.0000 0.9488 1.000 0.000 0.000 0.000
#> GSM614409 1 0.0188 0.9467 0.996 0.000 0.004 0.000
#> GSM614410 1 0.0000 0.9488 1.000 0.000 0.000 0.000
#> GSM614411 1 0.0376 0.9436 0.992 0.004 0.004 0.000
#> GSM614412 1 0.3123 0.8147 0.844 0.000 0.156 0.000
#> GSM614413 3 0.0336 0.9130 0.008 0.000 0.992 0.000
#> GSM614414 3 0.1792 0.8743 0.068 0.000 0.932 0.000
#> GSM614445 2 0.0592 0.8832 0.000 0.984 0.016 0.000
#> GSM614446 2 0.0707 0.8813 0.000 0.980 0.020 0.000
#> GSM614447 2 0.0707 0.8813 0.000 0.980 0.020 0.000
#> GSM614448 3 0.2843 0.8743 0.000 0.020 0.892 0.088
#> GSM614449 3 0.5160 0.6977 0.000 0.180 0.748 0.072
#> GSM614450 2 0.6592 0.3656 0.000 0.600 0.284 0.116
#> GSM614451 3 0.1867 0.8881 0.000 0.000 0.928 0.072
#> GSM614452 3 0.1792 0.8906 0.000 0.000 0.932 0.068
#> GSM614453 2 0.3486 0.7218 0.000 0.812 0.000 0.188
#> GSM614454 2 0.4866 0.2275 0.000 0.596 0.000 0.404
#> GSM614455 2 0.4406 0.5189 0.000 0.700 0.000 0.300
#> GSM614456 4 0.4898 0.3736 0.000 0.416 0.000 0.584
#> GSM614457 4 0.3907 0.6837 0.000 0.232 0.000 0.768
#> GSM614458 2 0.4040 0.6258 0.000 0.752 0.000 0.248
#> GSM614459 4 0.3649 0.7104 0.000 0.204 0.000 0.796
#> GSM614460 4 0.4907 0.3633 0.000 0.420 0.000 0.580
#> GSM614461 2 0.0000 0.8877 0.000 1.000 0.000 0.000
#> GSM614462 2 0.0000 0.8877 0.000 1.000 0.000 0.000
#> GSM614463 2 0.0000 0.8877 0.000 1.000 0.000 0.000
#> GSM614464 2 0.0000 0.8877 0.000 1.000 0.000 0.000
#> GSM614465 2 0.0188 0.8868 0.000 0.996 0.004 0.000
#> GSM614466 2 0.0000 0.8877 0.000 1.000 0.000 0.000
#> GSM614467 2 0.0469 0.8846 0.000 0.988 0.012 0.000
#> GSM614468 2 0.0707 0.8813 0.000 0.980 0.020 0.000
#> GSM614469 2 0.3428 0.7821 0.144 0.844 0.000 0.012
#> GSM614470 2 0.4456 0.6136 0.280 0.716 0.000 0.004
#> GSM614471 2 0.2101 0.8568 0.060 0.928 0.000 0.012
#> GSM614472 2 0.2805 0.8252 0.100 0.888 0.000 0.012
#> GSM614473 2 0.4933 0.3385 0.432 0.568 0.000 0.000
#> GSM614474 2 0.2216 0.8393 0.092 0.908 0.000 0.000
#> GSM614475 2 0.2053 0.8513 0.072 0.924 0.000 0.004
#> GSM614476 2 0.3674 0.8220 0.084 0.868 0.028 0.020
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM614415 5 0.0798 0.6533 0.016 0.000 0.008 0.000 0.976
#> GSM614416 5 0.0404 0.6557 0.012 0.000 0.000 0.000 0.988
#> GSM614417 5 0.1121 0.6562 0.044 0.000 0.000 0.000 0.956
#> GSM614418 5 0.1043 0.6567 0.040 0.000 0.000 0.000 0.960
#> GSM614419 5 0.2149 0.6377 0.048 0.000 0.036 0.000 0.916
#> GSM614420 5 0.1568 0.6485 0.036 0.000 0.020 0.000 0.944
#> GSM614421 3 0.4193 0.2570 0.304 0.012 0.684 0.000 0.000
#> GSM614422 3 0.4444 0.2315 0.364 0.012 0.624 0.000 0.000
#> GSM614423 2 0.6009 0.3105 0.240 0.580 0.180 0.000 0.000
#> GSM614424 3 0.5261 0.0762 0.424 0.048 0.528 0.000 0.000
#> GSM614425 3 0.4940 0.1691 0.392 0.032 0.576 0.000 0.000
#> GSM614426 3 0.4900 0.0817 0.464 0.024 0.512 0.000 0.000
#> GSM614427 3 0.4825 0.1878 0.408 0.024 0.568 0.000 0.000
#> GSM614428 3 0.4645 0.1905 0.424 0.004 0.564 0.008 0.000
#> GSM614429 2 0.0404 0.7912 0.012 0.988 0.000 0.000 0.000
#> GSM614430 2 0.0290 0.7912 0.008 0.992 0.000 0.000 0.000
#> GSM614431 2 0.0162 0.7912 0.004 0.996 0.000 0.000 0.000
#> GSM614432 2 0.0290 0.7913 0.008 0.992 0.000 0.000 0.000
#> GSM614433 2 0.0794 0.7886 0.028 0.972 0.000 0.000 0.000
#> GSM614434 2 0.0404 0.7912 0.012 0.988 0.000 0.000 0.000
#> GSM614435 2 0.1836 0.7823 0.032 0.932 0.000 0.036 0.000
#> GSM614436 2 0.3814 0.7178 0.064 0.816 0.004 0.116 0.000
#> GSM614437 4 0.0609 0.7620 0.000 0.020 0.000 0.980 0.000
#> GSM614438 4 0.1740 0.7854 0.056 0.000 0.012 0.932 0.000
#> GSM614439 4 0.2331 0.7780 0.080 0.000 0.020 0.900 0.000
#> GSM614440 4 0.2482 0.7706 0.084 0.000 0.024 0.892 0.000
#> GSM614441 4 0.2270 0.7796 0.076 0.000 0.020 0.904 0.000
#> GSM614442 4 0.1251 0.7835 0.036 0.000 0.008 0.956 0.000
#> GSM614443 4 0.0290 0.7680 0.000 0.008 0.000 0.992 0.000
#> GSM614444 4 0.2331 0.7784 0.080 0.000 0.020 0.900 0.000
#> GSM614391 5 0.4304 0.2229 0.000 0.000 0.484 0.000 0.516
#> GSM614392 5 0.4235 0.3197 0.000 0.000 0.424 0.000 0.576
#> GSM614393 5 0.4150 0.3664 0.000 0.000 0.388 0.000 0.612
#> GSM614394 3 0.4306 -0.2705 0.000 0.000 0.508 0.000 0.492
#> GSM614395 3 0.1443 0.2869 0.044 0.000 0.948 0.004 0.004
#> GSM614396 3 0.4300 -0.2437 0.000 0.000 0.524 0.000 0.476
#> GSM614397 3 0.2929 0.2737 0.000 0.000 0.820 0.000 0.180
#> GSM614398 3 0.3452 0.2065 0.000 0.000 0.756 0.000 0.244
#> GSM614399 2 0.4995 0.5702 0.264 0.668 0.000 0.000 0.068
#> GSM614400 2 0.5726 0.5120 0.248 0.612 0.000 0.000 0.140
#> GSM614401 2 0.6191 0.3943 0.292 0.536 0.000 0.000 0.172
#> GSM614402 2 0.5507 0.4645 0.316 0.596 0.000 0.000 0.088
#> GSM614403 1 0.5311 0.5414 0.700 0.212 0.064 0.004 0.020
#> GSM614404 2 0.4509 0.6154 0.236 0.716 0.000 0.000 0.048
#> GSM614405 1 0.5261 0.1820 0.572 0.044 0.380 0.004 0.000
#> GSM614406 3 0.5985 0.0917 0.408 0.000 0.480 0.112 0.000
#> GSM614407 5 0.4088 0.5987 0.304 0.000 0.008 0.000 0.688
#> GSM614408 5 0.4003 0.6061 0.288 0.000 0.008 0.000 0.704
#> GSM614409 5 0.5554 0.5693 0.328 0.004 0.076 0.000 0.592
#> GSM614410 5 0.4346 0.5987 0.304 0.004 0.012 0.000 0.680
#> GSM614411 5 0.5619 0.5653 0.332 0.004 0.080 0.000 0.584
#> GSM614412 5 0.7033 0.3701 0.352 0.012 0.244 0.000 0.392
#> GSM614413 3 0.5314 0.1579 0.420 0.000 0.528 0.000 0.052
#> GSM614414 3 0.6199 0.0898 0.392 0.000 0.468 0.000 0.140
#> GSM614445 2 0.4270 0.5315 0.320 0.668 0.000 0.000 0.012
#> GSM614446 2 0.4909 0.1302 0.472 0.508 0.008 0.000 0.012
#> GSM614447 2 0.4663 0.4244 0.376 0.604 0.000 0.000 0.020
#> GSM614448 1 0.5200 0.2997 0.628 0.000 0.304 0.068 0.000
#> GSM614449 1 0.5319 0.5484 0.704 0.072 0.196 0.028 0.000
#> GSM614450 1 0.5820 0.5787 0.684 0.188 0.088 0.032 0.008
#> GSM614451 3 0.5861 0.1117 0.400 0.000 0.500 0.100 0.000
#> GSM614452 3 0.5785 0.1181 0.404 0.000 0.504 0.092 0.000
#> GSM614453 2 0.2074 0.7621 0.000 0.896 0.000 0.104 0.000
#> GSM614454 2 0.3274 0.6685 0.000 0.780 0.000 0.220 0.000
#> GSM614455 2 0.2929 0.7086 0.000 0.820 0.000 0.180 0.000
#> GSM614456 2 0.3796 0.5545 0.000 0.700 0.000 0.300 0.000
#> GSM614457 4 0.4304 -0.0662 0.000 0.484 0.000 0.516 0.000
#> GSM614458 2 0.2377 0.7521 0.000 0.872 0.000 0.128 0.000
#> GSM614459 4 0.4192 0.2130 0.000 0.404 0.000 0.596 0.000
#> GSM614460 2 0.3876 0.5274 0.000 0.684 0.000 0.316 0.000
#> GSM614461 2 0.0000 0.7910 0.000 1.000 0.000 0.000 0.000
#> GSM614462 2 0.0609 0.7894 0.020 0.980 0.000 0.000 0.000
#> GSM614463 2 0.0510 0.7902 0.016 0.984 0.000 0.000 0.000
#> GSM614464 2 0.0794 0.7886 0.028 0.972 0.000 0.000 0.000
#> GSM614465 2 0.1341 0.7821 0.056 0.944 0.000 0.000 0.000
#> GSM614466 2 0.0963 0.7871 0.036 0.964 0.000 0.000 0.000
#> GSM614467 2 0.0880 0.7905 0.032 0.968 0.000 0.000 0.000
#> GSM614468 2 0.0963 0.7902 0.036 0.964 0.000 0.000 0.000
#> GSM614469 2 0.4640 0.3608 0.016 0.584 0.000 0.000 0.400
#> GSM614470 5 0.4738 -0.0738 0.016 0.464 0.000 0.000 0.520
#> GSM614471 2 0.3727 0.6757 0.016 0.768 0.000 0.000 0.216
#> GSM614472 2 0.4697 0.3807 0.020 0.592 0.000 0.000 0.388
#> GSM614473 5 0.4436 0.1509 0.008 0.396 0.000 0.000 0.596
#> GSM614474 2 0.3154 0.7426 0.012 0.836 0.004 0.000 0.148
#> GSM614475 2 0.2409 0.7775 0.000 0.912 0.016 0.028 0.044
#> GSM614476 2 0.4894 0.7075 0.020 0.784 0.104 0.044 0.048
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM614415 1 0.0858 0.7420 0.968 0.004 0.000 0.000 0.000 0.028
#> GSM614416 1 0.1010 0.7438 0.960 0.004 0.000 0.000 0.000 0.036
#> GSM614417 1 0.1793 0.7433 0.928 0.004 0.000 0.000 0.032 0.036
#> GSM614418 1 0.1370 0.7460 0.948 0.004 0.000 0.000 0.012 0.036
#> GSM614419 1 0.1429 0.7179 0.940 0.000 0.000 0.004 0.052 0.004
#> GSM614420 1 0.1225 0.7282 0.952 0.000 0.000 0.000 0.036 0.012
#> GSM614421 3 0.3663 0.5677 0.000 0.004 0.776 0.000 0.180 0.040
#> GSM614422 3 0.2560 0.6739 0.000 0.000 0.872 0.000 0.092 0.036
#> GSM614423 2 0.7017 0.0999 0.008 0.448 0.296 0.000 0.172 0.076
#> GSM614424 3 0.2685 0.7030 0.000 0.052 0.884 0.000 0.040 0.024
#> GSM614425 3 0.2593 0.6956 0.000 0.012 0.884 0.000 0.068 0.036
#> GSM614426 3 0.1887 0.7190 0.000 0.012 0.924 0.000 0.048 0.016
#> GSM614427 3 0.1768 0.7066 0.000 0.004 0.932 0.004 0.040 0.020
#> GSM614428 3 0.1275 0.7079 0.000 0.000 0.956 0.016 0.016 0.012
#> GSM614429 2 0.0551 0.7550 0.000 0.984 0.008 0.000 0.004 0.004
#> GSM614430 2 0.0405 0.7552 0.000 0.988 0.008 0.000 0.000 0.004
#> GSM614431 2 0.0363 0.7552 0.000 0.988 0.012 0.000 0.000 0.000
#> GSM614432 2 0.0458 0.7548 0.000 0.984 0.016 0.000 0.000 0.000
#> GSM614433 2 0.0891 0.7545 0.000 0.968 0.024 0.000 0.008 0.000
#> GSM614434 2 0.0653 0.7555 0.000 0.980 0.012 0.000 0.004 0.004
#> GSM614435 2 0.2619 0.7418 0.004 0.896 0.008 0.044 0.012 0.036
#> GSM614436 2 0.3445 0.7252 0.000 0.840 0.040 0.088 0.008 0.024
#> GSM614437 4 0.1406 0.8967 0.004 0.020 0.000 0.952 0.016 0.008
#> GSM614438 4 0.1219 0.9472 0.000 0.000 0.048 0.948 0.004 0.000
#> GSM614439 4 0.1738 0.9441 0.000 0.000 0.052 0.928 0.016 0.004
#> GSM614440 4 0.1946 0.9294 0.000 0.000 0.072 0.912 0.012 0.004
#> GSM614441 4 0.1398 0.9474 0.000 0.000 0.052 0.940 0.008 0.000
#> GSM614442 4 0.0922 0.9370 0.000 0.004 0.024 0.968 0.004 0.000
#> GSM614443 4 0.1406 0.8967 0.004 0.020 0.000 0.952 0.016 0.008
#> GSM614444 4 0.1707 0.9441 0.000 0.000 0.056 0.928 0.012 0.004
#> GSM614391 5 0.4684 0.7489 0.380 0.000 0.016 0.000 0.580 0.024
#> GSM614392 5 0.4554 0.7285 0.400 0.000 0.008 0.000 0.568 0.024
#> GSM614393 5 0.4377 0.6773 0.436 0.000 0.000 0.000 0.540 0.024
#> GSM614394 5 0.4886 0.7643 0.348 0.000 0.032 0.000 0.596 0.024
#> GSM614395 5 0.5351 0.3044 0.044 0.000 0.428 0.004 0.500 0.024
#> GSM614396 5 0.5160 0.7676 0.324 0.000 0.056 0.000 0.596 0.024
#> GSM614397 5 0.5649 0.6434 0.120 0.000 0.272 0.000 0.584 0.024
#> GSM614398 5 0.5762 0.6820 0.152 0.000 0.240 0.000 0.584 0.024
#> GSM614399 2 0.6038 0.4009 0.008 0.532 0.004 0.008 0.152 0.296
#> GSM614400 2 0.6559 0.4679 0.092 0.552 0.004 0.000 0.208 0.144
#> GSM614401 2 0.6666 0.4191 0.084 0.524 0.008 0.000 0.260 0.124
#> GSM614402 2 0.6471 0.4293 0.052 0.540 0.024 0.000 0.288 0.096
#> GSM614403 3 0.7316 0.4452 0.028 0.096 0.440 0.008 0.336 0.092
#> GSM614404 2 0.5435 0.5411 0.020 0.644 0.008 0.000 0.220 0.108
#> GSM614405 6 0.6574 -0.1767 0.000 0.032 0.380 0.004 0.184 0.400
#> GSM614406 3 0.4641 0.6310 0.000 0.000 0.740 0.144 0.052 0.064
#> GSM614407 6 0.2473 0.8333 0.136 0.000 0.000 0.000 0.008 0.856
#> GSM614408 6 0.2841 0.8083 0.164 0.000 0.000 0.000 0.012 0.824
#> GSM614409 6 0.1970 0.8505 0.092 0.000 0.000 0.000 0.008 0.900
#> GSM614410 6 0.2346 0.8398 0.124 0.000 0.000 0.000 0.008 0.868
#> GSM614411 6 0.1858 0.8500 0.092 0.004 0.000 0.000 0.000 0.904
#> GSM614412 6 0.1845 0.8482 0.072 0.000 0.004 0.000 0.008 0.916
#> GSM614413 6 0.2507 0.7825 0.004 0.000 0.072 0.000 0.040 0.884
#> GSM614414 6 0.2147 0.8072 0.012 0.000 0.044 0.000 0.032 0.912
#> GSM614445 2 0.5912 0.4248 0.008 0.556 0.064 0.000 0.320 0.052
#> GSM614446 2 0.7468 -0.0170 0.036 0.364 0.188 0.000 0.352 0.060
#> GSM614447 2 0.7139 0.2341 0.056 0.444 0.088 0.000 0.348 0.064
#> GSM614448 3 0.5665 0.5756 0.016 0.000 0.580 0.060 0.316 0.028
#> GSM614449 3 0.6128 0.5394 0.024 0.040 0.532 0.016 0.356 0.032
#> GSM614450 3 0.6863 0.4946 0.036 0.080 0.472 0.016 0.360 0.036
#> GSM614451 3 0.3173 0.6774 0.000 0.000 0.848 0.092 0.036 0.024
#> GSM614452 3 0.2999 0.6824 0.000 0.000 0.860 0.084 0.032 0.024
#> GSM614453 2 0.2220 0.7393 0.004 0.908 0.000 0.060 0.012 0.016
#> GSM614454 2 0.3283 0.7078 0.004 0.824 0.000 0.140 0.012 0.020
#> GSM614455 2 0.2900 0.7225 0.004 0.856 0.000 0.112 0.012 0.016
#> GSM614456 2 0.3422 0.6928 0.004 0.804 0.000 0.164 0.012 0.016
#> GSM614457 2 0.4434 0.4752 0.004 0.616 0.000 0.356 0.012 0.012
#> GSM614458 2 0.2518 0.7350 0.004 0.880 0.000 0.096 0.012 0.008
#> GSM614459 2 0.4691 0.2803 0.004 0.524 0.000 0.444 0.012 0.016
#> GSM614460 2 0.3592 0.6787 0.004 0.784 0.000 0.184 0.012 0.016
#> GSM614461 2 0.0405 0.7554 0.000 0.988 0.008 0.000 0.000 0.004
#> GSM614462 2 0.0603 0.7542 0.000 0.980 0.016 0.000 0.000 0.004
#> GSM614463 2 0.0603 0.7556 0.000 0.980 0.004 0.000 0.000 0.016
#> GSM614464 2 0.0862 0.7545 0.000 0.972 0.016 0.000 0.004 0.008
#> GSM614465 2 0.2172 0.7362 0.000 0.912 0.020 0.000 0.044 0.024
#> GSM614466 2 0.0748 0.7544 0.000 0.976 0.016 0.000 0.004 0.004
#> GSM614467 2 0.0909 0.7558 0.000 0.968 0.020 0.000 0.000 0.012
#> GSM614468 2 0.1003 0.7562 0.000 0.964 0.020 0.000 0.000 0.016
#> GSM614469 2 0.3989 0.0462 0.468 0.528 0.000 0.000 0.000 0.004
#> GSM614470 1 0.4275 0.3037 0.592 0.388 0.000 0.000 0.004 0.016
#> GSM614471 2 0.3559 0.5789 0.240 0.744 0.000 0.000 0.004 0.012
#> GSM614472 2 0.4128 -0.0533 0.492 0.500 0.000 0.000 0.004 0.004
#> GSM614473 1 0.3684 0.4510 0.664 0.332 0.000 0.000 0.000 0.004
#> GSM614474 2 0.3230 0.6219 0.212 0.776 0.000 0.000 0.000 0.012
#> GSM614475 2 0.2613 0.7370 0.008 0.892 0.000 0.028 0.056 0.016
#> GSM614476 2 0.4532 0.6953 0.020 0.796 0.052 0.044 0.068 0.020
Heatmaps for the consensus matrix. It visualizes the probability of two samples to be in a same group.
consensus_heatmap(res, k = 2)
consensus_heatmap(res, k = 3)
consensus_heatmap(res, k = 4)
consensus_heatmap(res, k = 5)
consensus_heatmap(res, k = 6)
Heatmaps for the membership of samples in all partitions to see how consistent they are:
membership_heatmap(res, k = 2)
membership_heatmap(res, k = 3)
membership_heatmap(res, k = 4)
membership_heatmap(res, k = 5)
membership_heatmap(res, k = 6)
As soon as we have had the classes for columns, we can look for signatures which are significantly different between classes which can be candidate marks for certain classes. Following are the heatmaps for signatures.
Signature heatmaps where rows are scaled:
get_signatures(res, k = 2)
get_signatures(res, k = 3)
get_signatures(res, k = 4)
get_signatures(res, k = 5)
get_signatures(res, k = 6)
Signature heatmaps where rows are not scaled:
get_signatures(res, k = 2, scale_rows = FALSE)
get_signatures(res, k = 3, scale_rows = FALSE)
get_signatures(res, k = 4, scale_rows = FALSE)
get_signatures(res, k = 5, scale_rows = FALSE)
get_signatures(res, k = 6, scale_rows = FALSE)
Compare the overlap of signatures from different k:
compare_signatures(res)
get_signature() returns a data frame invisibly. TO get the list of signatures, the function
call should be assigned to a variable explicitly. In following code, if plot argument is set
to FALSE, no heatmap is plotted while only the differential analysis is performed.
# code only for demonstration
tb = get_signature(res, k = ..., plot = FALSE)
An example of the output of tb is:
#> which_row fdr mean_1 mean_2 scaled_mean_1 scaled_mean_2 km
#> 1 38 0.042760348 8.373488 9.131774 -0.5533452 0.5164555 1
#> 2 40 0.018707592 7.106213 8.469186 -0.6173731 0.5762149 1
#> 3 55 0.019134737 10.221463 11.207825 -0.6159697 0.5749050 1
#> 4 59 0.006059896 5.921854 7.869574 -0.6899429 0.6439467 1
#> 5 60 0.018055526 8.928898 10.211722 -0.6204761 0.5791110 1
#> 6 98 0.009384629 15.714769 14.887706 0.6635654 -0.6193277 2
...
The columns in tb are:
which_row: row indices corresponding to the input matrix.fdr: FDR for the differential test. mean_x: The mean value in group x.scaled_mean_x: The mean value in group x after rows are scaled.km: Row groups if k-means clustering is applied to rows.UMAP plot which shows how samples are separated.
dimension_reduction(res, k = 2, method = "UMAP")
dimension_reduction(res, k = 3, method = "UMAP")
dimension_reduction(res, k = 4, method = "UMAP")
dimension_reduction(res, k = 5, method = "UMAP")
dimension_reduction(res, k = 6, method = "UMAP")
Following heatmap shows how subgroups are split when increasing k:
collect_classes(res)
Test correlation between subgroups and known annotations. If the known annotation is numeric, one-way ANOVA test is applied, and if the known annotation is discrete, chi-squared contingency table test is applied.
test_to_known_factors(res)
#> n individual(p) protocol(p) time(p) other(p) k
#> ATC:NMF 85 1.53e-06 0.01748 0.896 0.0121 2
#> ATC:NMF 78 9.42e-19 0.02908 0.999 0.2106 3
#> ATC:NMF 78 1.21e-21 0.00252 0.994 0.1799 4
#> ATC:NMF 52 2.36e-16 0.25909 0.686 0.0334 5
#> ATC:NMF 68 5.46e-44 0.60509 1.000 0.0963 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