Date: 2019-12-25 21:02:07 CET, cola version: 1.3.2
Document is loading...
All available functions which can be applied to this res_list object:
res_list
#> A 'ConsensusPartitionList' object with 24 methods.
#> On a matrix with 51941 rows and 92 columns.
#> Top rows are extracted by 'SD, CV, MAD, ATC' methods.
#> Subgroups are detected by 'hclust, kmeans, skmeans, pam, mclust, NMF' method.
#> Number of partitions are tried for k = 2, 3, 4, 5, 6.
#> Performed in total 30000 partitions by row resampling.
#>
#> Following methods can be applied to this 'ConsensusPartitionList' object:
#> [1] "cola_report" "collect_classes" "collect_plots" "collect_stats"
#> [5] "colnames" "functional_enrichment" "get_anno_col" "get_anno"
#> [9] "get_classes" "get_matrix" "get_membership" "get_stats"
#> [13] "is_best_k" "is_stable_k" "ncol" "nrow"
#> [17] "rownames" "show" "suggest_best_k" "test_to_known_factors"
#> [21] "top_rows_heatmap" "top_rows_overlap"
#>
#> You can get result for a single method by, e.g. object["SD", "hclust"] or object["SD:hclust"]
#> or a subset of methods by object[c("SD", "CV")], c("hclust", "kmeans")]
The call of run_all_consensus_partition_methods() was:
#> run_all_consensus_partition_methods(data = mat, mc.cores = 4, anno = anno)
Dimension of the input matrix:
mat = get_matrix(res_list)
dim(mat)
#> [1] 51941 92
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 | 1.000 | 1.000 | ** | |
| ATC:skmeans | 3 | 1.000 | 0.984 | 0.993 | ** | 2 |
| MAD:pam | 6 | 0.995 | 0.930 | 0.973 | ** | 3,4,5 |
| CV:mclust | 4 | 0.987 | 0.956 | 0.982 | ** | 2,3 |
| ATC:hclust | 6 | 0.983 | 0.931 | 0.960 | ** | 2,5 |
| MAD:NMF | 4 | 0.981 | 0.925 | 0.967 | ** | 2,3 |
| CV:skmeans | 5 | 0.978 | 0.950 | 0.967 | ** | 2,3,4 |
| MAD:skmeans | 5 | 0.959 | 0.924 | 0.959 | ** | 2,3,4 |
| SD:skmeans | 5 | 0.956 | 0.929 | 0.961 | ** | 2,3,4 |
| SD:pam | 6 | 0.950 | 0.916 | 0.958 | * | 4,5 |
| CV:NMF | 5 | 0.949 | 0.876 | 0.944 | * | 2,4 |
| SD:mclust | 4 | 0.947 | 0.943 | 0.976 | * | 2 |
| MAD:mclust | 2 | 0.933 | 0.925 | 0.969 | * | |
| SD:hclust | 4 | 0.932 | 0.925 | 0.962 | * | |
| ATC:pam | 6 | 0.925 | 0.910 | 0.929 | * | 2,3,4,5 |
| CV:pam | 6 | 0.921 | 0.874 | 0.943 | * | 4,5 |
| SD:NMF | 5 | 0.920 | 0.881 | 0.944 | * | 2,3,4 |
| MAD:hclust | 6 | 0.918 | 0.883 | 0.936 | * | |
| ATC:mclust | 3 | 0.914 | 0.964 | 0.976 | * | 2 |
| ATC:NMF | 5 | 0.908 | 0.891 | 0.930 | * | 2,4 |
| CV:hclust | 4 | 0.886 | 0.928 | 0.966 | ||
| CV:kmeans | 4 | 0.833 | 0.934 | 0.882 | ||
| MAD:kmeans | 2 | 0.724 | 0.896 | 0.944 | ||
| SD:kmeans | 2 | 0.688 | 0.932 | 0.952 |
**: 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 1.000 0.998 0.999 0.500 0.500 0.500
#> CV:NMF 2 1.000 0.991 0.996 0.501 0.500 0.500
#> MAD:NMF 2 1.000 0.983 0.992 0.501 0.500 0.500
#> ATC:NMF 2 1.000 0.984 0.993 0.489 0.514 0.514
#> SD:skmeans 2 1.000 0.988 0.995 0.501 0.500 0.500
#> CV:skmeans 2 1.000 0.991 0.996 0.501 0.500 0.500
#> MAD:skmeans 2 1.000 0.974 0.987 0.504 0.497 0.497
#> ATC:skmeans 2 1.000 1.000 1.000 0.487 0.514 0.514
#> SD:mclust 2 0.955 0.943 0.976 0.465 0.548 0.548
#> CV:mclust 2 0.955 0.970 0.986 0.474 0.523 0.523
#> MAD:mclust 2 0.933 0.925 0.969 0.469 0.535 0.535
#> ATC:mclust 2 1.000 1.000 1.000 0.429 0.572 0.572
#> SD:kmeans 2 0.688 0.932 0.952 0.482 0.500 0.500
#> CV:kmeans 2 0.719 0.904 0.933 0.470 0.500 0.500
#> MAD:kmeans 2 0.724 0.896 0.944 0.494 0.500 0.500
#> ATC:kmeans 2 1.000 1.000 1.000 0.477 0.523 0.523
#> SD:pam 2 0.853 0.945 0.973 0.501 0.500 0.500
#> CV:pam 2 0.749 0.950 0.973 0.499 0.500 0.500
#> MAD:pam 2 0.853 0.912 0.962 0.501 0.498 0.498
#> ATC:pam 2 1.000 0.987 0.995 0.479 0.523 0.523
#> SD:hclust 2 0.635 0.828 0.908 0.483 0.518 0.518
#> CV:hclust 2 0.653 0.840 0.923 0.482 0.514 0.514
#> MAD:hclust 2 0.669 0.903 0.941 0.484 0.518 0.518
#> ATC:hclust 2 0.912 0.947 0.965 0.466 0.541 0.541
get_stats(res_list, k = 3)
#> k 1-PAC mean_silhouette concordance area_increased Rand Jaccard
#> SD:NMF 3 0.910 0.904 0.958 0.250 0.843 0.695
#> CV:NMF 3 0.776 0.925 0.944 0.219 0.888 0.778
#> MAD:NMF 3 0.909 0.909 0.954 0.305 0.703 0.482
#> ATC:NMF 3 0.864 0.861 0.938 0.304 0.809 0.647
#> SD:skmeans 3 1.000 0.979 0.989 0.290 0.821 0.653
#> CV:skmeans 3 1.000 0.969 0.984 0.288 0.821 0.653
#> MAD:skmeans 3 1.000 0.966 0.984 0.269 0.832 0.671
#> ATC:skmeans 3 1.000 0.984 0.993 0.337 0.816 0.647
#> SD:mclust 3 0.786 0.910 0.960 0.298 0.610 0.415
#> CV:mclust 3 0.939 0.953 0.979 0.286 0.678 0.478
#> MAD:mclust 3 0.667 0.802 0.877 0.326 0.598 0.400
#> ATC:mclust 3 0.914 0.964 0.976 0.289 0.893 0.813
#> SD:kmeans 3 0.673 0.876 0.873 0.323 0.793 0.605
#> CV:kmeans 3 0.670 0.882 0.873 0.347 0.793 0.605
#> MAD:kmeans 3 0.781 0.897 0.913 0.318 0.783 0.588
#> ATC:kmeans 3 0.803 0.945 0.947 0.377 0.777 0.587
#> SD:pam 3 0.753 0.871 0.904 0.206 0.917 0.834
#> CV:pam 3 0.760 0.831 0.896 0.202 0.917 0.834
#> MAD:pam 3 0.937 0.954 0.979 0.322 0.791 0.601
#> ATC:pam 3 1.000 0.997 0.998 0.393 0.777 0.587
#> SD:hclust 3 0.676 0.868 0.901 0.305 0.832 0.676
#> CV:hclust 3 0.629 0.818 0.865 0.208 0.928 0.861
#> MAD:hclust 3 0.801 0.882 0.922 0.345 0.832 0.676
#> ATC:hclust 3 0.827 0.980 0.977 0.407 0.791 0.614
get_stats(res_list, k = 4)
#> k 1-PAC mean_silhouette concordance area_increased Rand Jaccard
#> SD:NMF 4 0.999 0.968 0.985 0.1564 0.807 0.544
#> CV:NMF 4 0.989 0.947 0.979 0.1895 0.815 0.568
#> MAD:NMF 4 0.981 0.925 0.967 0.1164 0.883 0.683
#> ATC:NMF 4 0.934 0.932 0.956 0.1451 0.841 0.604
#> SD:skmeans 4 0.982 0.949 0.975 0.1196 0.899 0.721
#> CV:skmeans 4 0.990 0.923 0.965 0.1110 0.893 0.714
#> MAD:skmeans 4 0.964 0.953 0.972 0.1338 0.890 0.704
#> ATC:skmeans 4 0.878 0.759 0.873 0.0519 0.977 0.934
#> SD:mclust 4 0.947 0.943 0.976 0.2097 0.853 0.637
#> CV:mclust 4 0.987 0.956 0.982 0.1857 0.844 0.620
#> MAD:mclust 4 0.864 0.903 0.953 0.1645 0.863 0.660
#> ATC:mclust 4 0.782 0.860 0.923 0.2452 0.842 0.659
#> SD:kmeans 4 0.854 0.926 0.885 0.1164 0.934 0.806
#> CV:kmeans 4 0.833 0.934 0.882 0.1199 0.934 0.806
#> MAD:kmeans 4 0.783 0.758 0.773 0.1030 0.922 0.770
#> ATC:kmeans 4 0.869 0.873 0.867 0.0914 0.912 0.744
#> SD:pam 4 1.000 0.949 0.982 0.2015 0.803 0.556
#> CV:pam 4 1.000 0.960 0.986 0.2139 0.805 0.558
#> MAD:pam 4 0.977 0.941 0.974 0.1015 0.899 0.716
#> ATC:pam 4 1.000 0.972 0.987 0.0914 0.937 0.810
#> SD:hclust 4 0.932 0.925 0.962 0.1436 0.928 0.796
#> CV:hclust 4 0.886 0.928 0.966 0.2426 0.817 0.593
#> MAD:hclust 4 0.831 0.859 0.890 0.0971 0.928 0.796
#> ATC:hclust 4 0.826 0.946 0.926 0.0522 0.980 0.940
get_stats(res_list, k = 5)
#> k 1-PAC mean_silhouette concordance area_increased Rand Jaccard
#> SD:NMF 5 0.920 0.881 0.944 0.0616 0.941 0.792
#> CV:NMF 5 0.949 0.876 0.944 0.0598 0.934 0.768
#> MAD:NMF 5 0.884 0.884 0.932 0.0544 0.958 0.849
#> ATC:NMF 5 0.908 0.891 0.930 0.0616 0.922 0.726
#> SD:skmeans 5 0.956 0.929 0.961 0.0358 0.963 0.868
#> CV:skmeans 5 0.978 0.950 0.967 0.0452 0.947 0.821
#> MAD:skmeans 5 0.959 0.924 0.959 0.0335 0.974 0.908
#> ATC:skmeans 5 0.807 0.801 0.853 0.0586 0.953 0.860
#> SD:mclust 5 0.887 0.796 0.887 0.0435 0.968 0.886
#> CV:mclust 5 0.898 0.833 0.908 0.0528 0.978 0.918
#> MAD:mclust 5 0.802 0.749 0.858 0.0518 0.961 0.857
#> ATC:mclust 5 0.757 0.812 0.849 0.0886 0.854 0.585
#> SD:kmeans 5 0.747 0.847 0.840 0.0682 1.000 1.000
#> CV:kmeans 5 0.759 0.858 0.837 0.0706 1.000 1.000
#> MAD:kmeans 5 0.740 0.810 0.823 0.0610 0.953 0.841
#> ATC:kmeans 5 0.761 0.823 0.812 0.0526 0.978 0.921
#> SD:pam 5 0.989 0.956 0.982 0.0463 0.948 0.817
#> CV:pam 5 0.979 0.928 0.958 0.0464 0.953 0.829
#> MAD:pam 5 0.990 0.937 0.970 0.0399 0.953 0.830
#> ATC:pam 5 1.000 0.967 0.983 0.0459 0.960 0.853
#> SD:hclust 5 0.880 0.844 0.934 0.0409 0.984 0.942
#> CV:hclust 5 0.883 0.878 0.924 0.0446 0.969 0.890
#> MAD:hclust 5 0.898 0.876 0.932 0.0629 0.969 0.890
#> ATC:hclust 5 1.000 0.982 0.990 0.0927 0.934 0.789
get_stats(res_list, k = 6)
#> k 1-PAC mean_silhouette concordance area_increased Rand Jaccard
#> SD:NMF 6 0.850 0.736 0.853 0.0432 0.957 0.819
#> CV:NMF 6 0.850 0.705 0.823 0.0369 0.952 0.799
#> MAD:NMF 6 0.811 0.699 0.838 0.0395 0.974 0.893
#> ATC:NMF 6 0.870 0.816 0.877 0.0255 0.973 0.879
#> SD:skmeans 6 0.934 0.867 0.906 0.0280 0.994 0.977
#> CV:skmeans 6 0.895 0.877 0.897 0.0298 0.989 0.957
#> MAD:skmeans 6 0.926 0.837 0.897 0.0285 0.989 0.957
#> ATC:skmeans 6 0.814 0.781 0.786 0.0452 0.920 0.737
#> SD:mclust 6 0.845 0.765 0.871 0.0440 0.951 0.811
#> CV:mclust 6 0.873 0.815 0.898 0.0552 0.954 0.820
#> MAD:mclust 6 0.799 0.733 0.855 0.0367 0.954 0.813
#> ATC:mclust 6 0.754 0.687 0.802 0.0470 0.917 0.693
#> SD:kmeans 6 0.727 0.774 0.769 0.0396 1.000 1.000
#> CV:kmeans 6 0.789 0.641 0.767 0.0436 0.984 0.942
#> MAD:kmeans 6 0.716 0.684 0.775 0.0433 0.931 0.748
#> ATC:kmeans 6 0.742 0.766 0.808 0.0421 0.955 0.831
#> SD:pam 6 0.950 0.916 0.958 0.0307 0.980 0.916
#> CV:pam 6 0.921 0.874 0.943 0.0273 0.967 0.865
#> MAD:pam 6 0.995 0.930 0.973 0.0323 0.970 0.879
#> ATC:pam 6 0.925 0.910 0.929 0.0330 0.976 0.897
#> SD:hclust 6 0.871 0.835 0.922 0.0421 0.964 0.864
#> CV:hclust 6 0.872 0.769 0.880 0.0352 0.991 0.963
#> MAD:hclust 6 0.918 0.883 0.936 0.0368 0.964 0.856
#> ATC:hclust 6 0.983 0.931 0.960 0.0203 0.987 0.948
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 specimen(p) k
#> SD:NMF 92 4.01e-14 2
#> CV:NMF 92 4.01e-14 2
#> MAD:NMF 92 4.01e-14 2
#> ATC:NMF 92 7.21e-17 2
#> SD:skmeans 92 4.01e-14 2
#> CV:skmeans 92 4.01e-14 2
#> MAD:skmeans 92 1.41e-13 2
#> ATC:skmeans 92 7.21e-17 2
#> SD:mclust 88 7.88e-17 2
#> CV:mclust 92 1.16e-17 2
#> MAD:mclust 88 7.88e-17 2
#> ATC:mclust 92 3.07e-14 2
#> SD:kmeans 92 4.01e-14 2
#> CV:kmeans 92 4.01e-14 2
#> MAD:kmeans 92 4.01e-14 2
#> ATC:kmeans 92 5.33e-17 2
#> SD:pam 92 4.01e-14 2
#> CV:pam 92 4.01e-14 2
#> MAD:pam 85 2.23e-13 2
#> ATC:pam 91 1.88e-17 2
#> SD:hclust 82 1.38e-15 2
#> CV:hclust 90 3.03e-17 2
#> MAD:hclust 92 1.16e-17 2
#> ATC:hclust 92 1.16e-17 2
test_to_known_factors(res_list, k = 3)
#> n specimen(p) k
#> SD:NMF 89 4.94e-24 3
#> CV:NMF 92 7.13e-26 3
#> MAD:NMF 89 3.87e-26 3
#> ATC:NMF 84 5.05e-28 3
#> SD:skmeans 92 3.33e-30 3
#> CV:skmeans 90 2.83e-29 3
#> MAD:skmeans 91 1.42e-27 3
#> ATC:skmeans 92 2.19e-29 3
#> SD:mclust 89 1.10e-31 3
#> CV:mclust 92 7.03e-30 3
#> MAD:mclust 91 1.66e-32 3
#> ATC:mclust 92 2.81e-22 3
#> SD:kmeans 92 7.20e-32 3
#> CV:kmeans 92 7.20e-32 3
#> MAD:kmeans 92 1.23e-30 3
#> ATC:kmeans 92 6.44e-33 3
#> SD:pam 92 1.29e-28 3
#> CV:pam 90 6.60e-29 3
#> MAD:pam 91 6.53e-28 3
#> ATC:pam 92 6.44e-33 3
#> SD:hclust 88 2.82e-31 3
#> CV:hclust 90 4.27e-32 3
#> MAD:hclust 92 4.55e-28 3
#> ATC:hclust 92 6.44e-33 3
test_to_known_factors(res_list, k = 4)
#> n specimen(p) k
#> SD:NMF 92 1.69e-42 4
#> CV:NMF 90 7.15e-46 4
#> MAD:NMF 88 1.85e-38 4
#> ATC:NMF 90 5.52e-35 4
#> SD:skmeans 90 1.98e-39 4
#> CV:skmeans 87 4.69e-45 4
#> MAD:skmeans 91 2.41e-37 4
#> ATC:skmeans 72 3.93e-14 4
#> SD:mclust 90 6.82e-47 4
#> CV:mclust 91 1.66e-47 4
#> MAD:mclust 91 1.66e-47 4
#> ATC:mclust 88 3.64e-26 4
#> SD:kmeans 92 4.48e-47 4
#> CV:kmeans 92 4.48e-47 4
#> MAD:kmeans 84 2.84e-42 4
#> ATC:kmeans 90 6.82e-47 4
#> SD:pam 89 2.80e-46 4
#> CV:pam 90 7.34e-46 4
#> MAD:pam 90 7.15e-46 4
#> ATC:pam 91 9.29e-41 4
#> SD:hclust 88 1.15e-45 4
#> CV:hclust 92 4.04e-48 4
#> MAD:hclust 85 7.83e-44 4
#> ATC:hclust 92 4.04e-48 4
test_to_known_factors(res_list, k = 5)
#> n specimen(p) k
#> SD:NMF 88 5.18e-38 5
#> CV:NMF 86 3.52e-39 5
#> MAD:NMF 91 3.78e-39 5
#> ATC:NMF 89 3.95e-34 5
#> SD:skmeans 90 1.60e-43 5
#> CV:skmeans 91 3.30e-44 5
#> MAD:skmeans 89 6.34e-43 5
#> ATC:skmeans 85 5.98e-38 5
#> SD:mclust 84 1.06e-41 5
#> CV:mclust 87 6.47e-47 5
#> MAD:mclust 82 6.99e-43 5
#> ATC:mclust 91 1.56e-24 5
#> SD:kmeans 92 4.48e-47 5
#> CV:kmeans 92 4.48e-47 5
#> MAD:kmeans 88 1.14e-44 5
#> ATC:kmeans 86 1.92e-44 5
#> SD:pam 91 2.31e-41 5
#> CV:pam 90 3.55e-41 5
#> MAD:pam 89 9.56e-41 5
#> ATC:pam 91 1.00e-38 5
#> SD:hclust 86 2.09e-58 5
#> CV:hclust 91 6.79e-45 5
#> MAD:hclust 87 3.47e-56 5
#> ATC:hclust 92 2.68e-63 5
test_to_known_factors(res_list, k = 6)
#> n specimen(p) k
#> SD:NMF 77 1.31e-34 6
#> CV:NMF 73 1.11e-31 6
#> MAD:NMF 74 2.34e-29 6
#> ATC:NMF 87 6.15e-33 6
#> SD:skmeans 88 2.73e-39 6
#> CV:skmeans 86 4.00e-41 6
#> MAD:skmeans 87 4.84e-40 6
#> ATC:skmeans 77 6.99e-48 6
#> SD:mclust 84 1.37e-53 6
#> CV:mclust 82 1.00e-45 6
#> MAD:mclust 79 1.39e-49 6
#> ATC:mclust 70 7.18e-24 6
#> SD:kmeans 92 4.48e-47 6
#> CV:kmeans 80 1.36e-37 6
#> MAD:kmeans 76 1.62e-32 6
#> ATC:kmeans 87 1.08e-41 6
#> SD:pam 91 1.54e-38 6
#> CV:pam 88 3.66e-37 6
#> MAD:pam 89 1.64e-35 6
#> ATC:pam 91 4.25e-37 6
#> SD:hclust 87 1.01e-52 6
#> CV:hclust 78 2.81e-36 6
#> MAD:hclust 84 2.67e-51 6
#> ATC:hclust 88 2.84e-59 6
The object with results only for a single top-value method and a single partition method can be extracted as:
res = res_list["SD", "hclust"]
# you can also extract it by
# res = res_list["SD:hclust"]
A summary of res and all the functions that can be applied to it:
res
#> A 'ConsensusPartition' object with k = 2, 3, 4, 5, 6.
#> On a matrix with 51941 rows and 92 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 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 0.635 0.828 0.908 0.4832 0.518 0.518
#> 3 3 0.676 0.868 0.901 0.3045 0.832 0.676
#> 4 4 0.932 0.925 0.962 0.1436 0.928 0.796
#> 5 5 0.880 0.844 0.934 0.0409 0.984 0.942
#> 6 6 0.871 0.835 0.922 0.0421 0.964 0.864
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
#> GSM587155 2 0.000 1.000 0.000 1.000
#> GSM587156 2 0.000 1.000 0.000 1.000
#> GSM587157 2 0.000 1.000 0.000 1.000
#> GSM587158 2 0.000 1.000 0.000 1.000
#> GSM587159 2 0.000 1.000 0.000 1.000
#> GSM587160 2 0.000 1.000 0.000 1.000
#> GSM587161 2 0.000 1.000 0.000 1.000
#> GSM587162 2 0.000 1.000 0.000 1.000
#> GSM587163 2 0.000 1.000 0.000 1.000
#> GSM587164 2 0.000 1.000 0.000 1.000
#> GSM587165 2 0.000 1.000 0.000 1.000
#> GSM587166 2 0.000 1.000 0.000 1.000
#> GSM587167 2 0.000 1.000 0.000 1.000
#> GSM587168 2 0.000 1.000 0.000 1.000
#> GSM587169 2 0.000 1.000 0.000 1.000
#> GSM587170 2 0.000 1.000 0.000 1.000
#> GSM587171 2 0.000 1.000 0.000 1.000
#> GSM587172 2 0.000 1.000 0.000 1.000
#> GSM587173 2 0.000 1.000 0.000 1.000
#> GSM587174 2 0.000 1.000 0.000 1.000
#> GSM587175 2 0.000 1.000 0.000 1.000
#> GSM587176 2 0.000 1.000 0.000 1.000
#> GSM587177 2 0.000 1.000 0.000 1.000
#> GSM587178 2 0.000 1.000 0.000 1.000
#> GSM587179 2 0.000 1.000 0.000 1.000
#> GSM587180 2 0.000 1.000 0.000 1.000
#> GSM587181 2 0.000 1.000 0.000 1.000
#> GSM587182 2 0.000 1.000 0.000 1.000
#> GSM587183 2 0.000 1.000 0.000 1.000
#> GSM587184 2 0.000 1.000 0.000 1.000
#> GSM587185 2 0.000 1.000 0.000 1.000
#> GSM587186 2 0.000 1.000 0.000 1.000
#> GSM587187 2 0.000 1.000 0.000 1.000
#> GSM587188 2 0.000 1.000 0.000 1.000
#> GSM587189 2 0.000 1.000 0.000 1.000
#> GSM587190 2 0.000 1.000 0.000 1.000
#> GSM587203 1 0.000 0.830 1.000 0.000
#> GSM587204 1 0.000 0.830 1.000 0.000
#> GSM587205 1 0.000 0.830 1.000 0.000
#> GSM587206 1 0.000 0.830 1.000 0.000
#> GSM587207 1 0.000 0.830 1.000 0.000
#> GSM587208 1 0.000 0.830 1.000 0.000
#> GSM587209 1 0.000 0.830 1.000 0.000
#> GSM587210 1 0.000 0.830 1.000 0.000
#> GSM587211 1 0.000 0.830 1.000 0.000
#> GSM587212 1 0.000 0.830 1.000 0.000
#> GSM587213 1 0.000 0.830 1.000 0.000
#> GSM587214 1 0.000 0.830 1.000 0.000
#> GSM587215 1 0.000 0.830 1.000 0.000
#> GSM587216 1 0.000 0.830 1.000 0.000
#> GSM587217 1 0.000 0.830 1.000 0.000
#> GSM587191 1 0.991 0.478 0.556 0.444
#> GSM587192 1 0.991 0.478 0.556 0.444
#> GSM587193 1 0.969 0.551 0.604 0.396
#> GSM587194 1 0.969 0.551 0.604 0.396
#> GSM587195 1 0.991 0.478 0.556 0.444
#> GSM587196 1 0.991 0.478 0.556 0.444
#> GSM587197 1 0.991 0.478 0.556 0.444
#> GSM587198 1 0.975 0.536 0.592 0.408
#> GSM587199 1 0.975 0.536 0.592 0.408
#> GSM587200 1 0.714 0.724 0.804 0.196
#> GSM587201 1 0.714 0.724 0.804 0.196
#> GSM587202 1 0.975 0.536 0.592 0.408
#> GSM198767 1 0.000 0.830 1.000 0.000
#> GSM198769 1 0.000 0.830 1.000 0.000
#> GSM198772 1 0.000 0.830 1.000 0.000
#> GSM198773 1 0.000 0.830 1.000 0.000
#> GSM198776 1 0.000 0.830 1.000 0.000
#> GSM198778 1 0.000 0.830 1.000 0.000
#> GSM198780 1 0.000 0.830 1.000 0.000
#> GSM198781 1 0.000 0.830 1.000 0.000
#> GSM198765 1 0.991 0.478 0.556 0.444
#> GSM198766 1 0.969 0.551 0.604 0.396
#> GSM198768 1 0.991 0.478 0.556 0.444
#> GSM198770 1 0.991 0.478 0.556 0.444
#> GSM198771 1 0.975 0.536 0.592 0.408
#> GSM198774 1 0.991 0.478 0.556 0.444
#> GSM198775 1 0.969 0.551 0.604 0.396
#> GSM198777 1 0.991 0.478 0.556 0.444
#> GSM198779 1 0.975 0.536 0.592 0.408
#> GSM587218 1 0.000 0.830 1.000 0.000
#> GSM587219 1 0.000 0.830 1.000 0.000
#> GSM587220 1 0.000 0.830 1.000 0.000
#> GSM587221 1 0.000 0.830 1.000 0.000
#> GSM587222 1 0.000 0.830 1.000 0.000
#> GSM587223 1 0.000 0.830 1.000 0.000
#> GSM587224 1 0.000 0.830 1.000 0.000
#> GSM587225 1 0.000 0.830 1.000 0.000
#> GSM587226 1 0.000 0.830 1.000 0.000
#> GSM587227 1 0.000 0.830 1.000 0.000
#> GSM587228 1 0.000 0.830 1.000 0.000
#> GSM587229 1 0.000 0.830 1.000 0.000
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM587155 2 0.0000 0.995 0.000 1.000 0.000
#> GSM587156 2 0.1411 0.961 0.000 0.964 0.036
#> GSM587157 2 0.0000 0.995 0.000 1.000 0.000
#> GSM587158 2 0.0000 0.995 0.000 1.000 0.000
#> GSM587159 2 0.0000 0.995 0.000 1.000 0.000
#> GSM587160 2 0.0000 0.995 0.000 1.000 0.000
#> GSM587161 2 0.0000 0.995 0.000 1.000 0.000
#> GSM587162 2 0.0000 0.995 0.000 1.000 0.000
#> GSM587163 2 0.0000 0.995 0.000 1.000 0.000
#> GSM587164 2 0.0000 0.995 0.000 1.000 0.000
#> GSM587165 2 0.0000 0.995 0.000 1.000 0.000
#> GSM587166 2 0.1411 0.961 0.000 0.964 0.036
#> GSM587167 2 0.1411 0.961 0.000 0.964 0.036
#> GSM587168 2 0.0000 0.995 0.000 1.000 0.000
#> GSM587169 2 0.0000 0.995 0.000 1.000 0.000
#> GSM587170 2 0.0000 0.995 0.000 1.000 0.000
#> GSM587171 2 0.0000 0.995 0.000 1.000 0.000
#> GSM587172 2 0.0000 0.995 0.000 1.000 0.000
#> GSM587173 2 0.0000 0.995 0.000 1.000 0.000
#> GSM587174 2 0.0000 0.995 0.000 1.000 0.000
#> GSM587175 2 0.0000 0.995 0.000 1.000 0.000
#> GSM587176 2 0.0000 0.995 0.000 1.000 0.000
#> GSM587177 2 0.0000 0.995 0.000 1.000 0.000
#> GSM587178 2 0.0000 0.995 0.000 1.000 0.000
#> GSM587179 2 0.0000 0.995 0.000 1.000 0.000
#> GSM587180 2 0.0000 0.995 0.000 1.000 0.000
#> GSM587181 2 0.0000 0.995 0.000 1.000 0.000
#> GSM587182 2 0.0000 0.995 0.000 1.000 0.000
#> GSM587183 2 0.0000 0.995 0.000 1.000 0.000
#> GSM587184 2 0.0000 0.995 0.000 1.000 0.000
#> GSM587185 2 0.0000 0.995 0.000 1.000 0.000
#> GSM587186 2 0.0000 0.995 0.000 1.000 0.000
#> GSM587187 2 0.0000 0.995 0.000 1.000 0.000
#> GSM587188 2 0.0000 0.995 0.000 1.000 0.000
#> GSM587189 2 0.0000 0.995 0.000 1.000 0.000
#> GSM587190 2 0.1411 0.961 0.000 0.964 0.036
#> GSM587203 1 0.0000 0.998 1.000 0.000 0.000
#> GSM587204 1 0.0000 0.998 1.000 0.000 0.000
#> GSM587205 1 0.0000 0.998 1.000 0.000 0.000
#> GSM587206 1 0.0000 0.998 1.000 0.000 0.000
#> GSM587207 1 0.0000 0.998 1.000 0.000 0.000
#> GSM587208 1 0.0000 0.998 1.000 0.000 0.000
#> GSM587209 1 0.0892 0.979 0.980 0.000 0.020
#> GSM587210 3 0.6204 0.453 0.424 0.000 0.576
#> GSM587211 1 0.0000 0.998 1.000 0.000 0.000
#> GSM587212 3 0.6215 0.449 0.428 0.000 0.572
#> GSM587213 1 0.0000 0.998 1.000 0.000 0.000
#> GSM587214 1 0.0000 0.998 1.000 0.000 0.000
#> GSM587215 1 0.0000 0.998 1.000 0.000 0.000
#> GSM587216 1 0.0000 0.998 1.000 0.000 0.000
#> GSM587217 1 0.0000 0.998 1.000 0.000 0.000
#> GSM587191 3 0.5291 0.733 0.000 0.268 0.732
#> GSM587192 3 0.5291 0.733 0.000 0.268 0.732
#> GSM587193 3 0.4796 0.760 0.000 0.220 0.780
#> GSM587194 3 0.4796 0.760 0.000 0.220 0.780
#> GSM587195 3 0.5291 0.733 0.000 0.268 0.732
#> GSM587196 3 0.5291 0.733 0.000 0.268 0.732
#> GSM587197 3 0.5291 0.733 0.000 0.268 0.732
#> GSM587198 3 0.4931 0.756 0.000 0.232 0.768
#> GSM587199 3 0.4931 0.756 0.000 0.232 0.768
#> GSM587200 3 0.5356 0.662 0.196 0.020 0.784
#> GSM587201 3 0.5356 0.662 0.196 0.020 0.784
#> GSM587202 3 0.4931 0.756 0.000 0.232 0.768
#> GSM198767 1 0.0000 0.998 1.000 0.000 0.000
#> GSM198769 1 0.0892 0.979 0.980 0.000 0.020
#> GSM198772 1 0.0000 0.998 1.000 0.000 0.000
#> GSM198773 1 0.0000 0.998 1.000 0.000 0.000
#> GSM198776 1 0.0000 0.998 1.000 0.000 0.000
#> GSM198778 3 0.6204 0.453 0.424 0.000 0.576
#> GSM198780 3 0.6215 0.449 0.428 0.000 0.572
#> GSM198781 1 0.0000 0.998 1.000 0.000 0.000
#> GSM198765 3 0.5291 0.733 0.000 0.268 0.732
#> GSM198766 3 0.4796 0.760 0.000 0.220 0.780
#> GSM198768 3 0.5291 0.733 0.000 0.268 0.732
#> GSM198770 3 0.5291 0.733 0.000 0.268 0.732
#> GSM198771 3 0.4931 0.756 0.000 0.232 0.768
#> GSM198774 3 0.5291 0.733 0.000 0.268 0.732
#> GSM198775 3 0.4796 0.760 0.000 0.220 0.780
#> GSM198777 3 0.5291 0.733 0.000 0.268 0.732
#> GSM198779 3 0.4931 0.756 0.000 0.232 0.768
#> GSM587218 3 0.4235 0.662 0.176 0.000 0.824
#> GSM587219 3 0.4235 0.662 0.176 0.000 0.824
#> GSM587220 3 0.4235 0.662 0.176 0.000 0.824
#> GSM587221 3 0.4235 0.662 0.176 0.000 0.824
#> GSM587222 3 0.4235 0.662 0.176 0.000 0.824
#> GSM587223 3 0.4235 0.662 0.176 0.000 0.824
#> GSM587224 3 0.4235 0.662 0.176 0.000 0.824
#> GSM587225 3 0.4235 0.662 0.176 0.000 0.824
#> GSM587226 3 0.4235 0.662 0.176 0.000 0.824
#> GSM587227 3 0.4235 0.662 0.176 0.000 0.824
#> GSM587228 3 0.4235 0.662 0.176 0.000 0.824
#> GSM587229 3 0.4235 0.662 0.176 0.000 0.824
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM587155 2 0.1302 0.953 0.000 0.956 0.044 0.000
#> GSM587156 2 0.2704 0.882 0.000 0.876 0.124 0.000
#> GSM587157 2 0.1302 0.953 0.000 0.956 0.044 0.000
#> GSM587158 2 0.0000 0.977 0.000 1.000 0.000 0.000
#> GSM587159 2 0.0000 0.977 0.000 1.000 0.000 0.000
#> GSM587160 2 0.0000 0.977 0.000 1.000 0.000 0.000
#> GSM587161 2 0.0000 0.977 0.000 1.000 0.000 0.000
#> GSM587162 2 0.0000 0.977 0.000 1.000 0.000 0.000
#> GSM587163 2 0.0000 0.977 0.000 1.000 0.000 0.000
#> GSM587164 2 0.1302 0.953 0.000 0.956 0.044 0.000
#> GSM587165 2 0.0000 0.977 0.000 1.000 0.000 0.000
#> GSM587166 2 0.2704 0.882 0.000 0.876 0.124 0.000
#> GSM587167 2 0.2704 0.882 0.000 0.876 0.124 0.000
#> GSM587168 2 0.0000 0.977 0.000 1.000 0.000 0.000
#> GSM587169 2 0.0000 0.977 0.000 1.000 0.000 0.000
#> GSM587170 2 0.1302 0.953 0.000 0.956 0.044 0.000
#> GSM587171 2 0.0000 0.977 0.000 1.000 0.000 0.000
#> GSM587172 2 0.0000 0.977 0.000 1.000 0.000 0.000
#> GSM587173 2 0.0000 0.977 0.000 1.000 0.000 0.000
#> GSM587174 2 0.0000 0.977 0.000 1.000 0.000 0.000
#> GSM587175 2 0.1302 0.953 0.000 0.956 0.044 0.000
#> GSM587176 2 0.0000 0.977 0.000 1.000 0.000 0.000
#> GSM587177 2 0.0000 0.977 0.000 1.000 0.000 0.000
#> GSM587178 2 0.0000 0.977 0.000 1.000 0.000 0.000
#> GSM587179 2 0.0000 0.977 0.000 1.000 0.000 0.000
#> GSM587180 2 0.0000 0.977 0.000 1.000 0.000 0.000
#> GSM587181 2 0.0000 0.977 0.000 1.000 0.000 0.000
#> GSM587182 2 0.0000 0.977 0.000 1.000 0.000 0.000
#> GSM587183 2 0.0000 0.977 0.000 1.000 0.000 0.000
#> GSM587184 2 0.0000 0.977 0.000 1.000 0.000 0.000
#> GSM587185 2 0.0000 0.977 0.000 1.000 0.000 0.000
#> GSM587186 2 0.0000 0.977 0.000 1.000 0.000 0.000
#> GSM587187 2 0.0188 0.975 0.000 0.996 0.004 0.000
#> GSM587188 2 0.0707 0.965 0.000 0.980 0.020 0.000
#> GSM587189 2 0.0817 0.964 0.000 0.976 0.024 0.000
#> GSM587190 2 0.2704 0.882 0.000 0.876 0.124 0.000
#> GSM587203 1 0.0000 0.998 1.000 0.000 0.000 0.000
#> GSM587204 1 0.0000 0.998 1.000 0.000 0.000 0.000
#> GSM587205 1 0.0000 0.998 1.000 0.000 0.000 0.000
#> GSM587206 1 0.0000 0.998 1.000 0.000 0.000 0.000
#> GSM587207 1 0.0000 0.998 1.000 0.000 0.000 0.000
#> GSM587208 1 0.0000 0.998 1.000 0.000 0.000 0.000
#> GSM587209 1 0.0779 0.982 0.980 0.000 0.016 0.004
#> GSM587210 3 0.5088 0.362 0.424 0.000 0.572 0.004
#> GSM587211 1 0.0000 0.998 1.000 0.000 0.000 0.000
#> GSM587212 3 0.4925 0.357 0.428 0.000 0.572 0.000
#> GSM587213 1 0.0000 0.998 1.000 0.000 0.000 0.000
#> GSM587214 1 0.0000 0.998 1.000 0.000 0.000 0.000
#> GSM587215 1 0.0000 0.998 1.000 0.000 0.000 0.000
#> GSM587216 1 0.0000 0.998 1.000 0.000 0.000 0.000
#> GSM587217 1 0.0000 0.998 1.000 0.000 0.000 0.000
#> GSM587191 3 0.1389 0.874 0.000 0.048 0.952 0.000
#> GSM587192 3 0.1389 0.874 0.000 0.048 0.952 0.000
#> GSM587193 3 0.0376 0.869 0.000 0.004 0.992 0.004
#> GSM587194 3 0.0376 0.869 0.000 0.004 0.992 0.004
#> GSM587195 3 0.1389 0.874 0.000 0.048 0.952 0.000
#> GSM587196 3 0.1389 0.874 0.000 0.048 0.952 0.000
#> GSM587197 3 0.1389 0.874 0.000 0.048 0.952 0.000
#> GSM587198 3 0.0469 0.874 0.000 0.012 0.988 0.000
#> GSM587199 3 0.0469 0.874 0.000 0.012 0.988 0.000
#> GSM587200 3 0.3751 0.728 0.196 0.000 0.800 0.004
#> GSM587201 3 0.3751 0.728 0.196 0.000 0.800 0.004
#> GSM587202 3 0.0469 0.874 0.000 0.012 0.988 0.000
#> GSM198767 1 0.0000 0.998 1.000 0.000 0.000 0.000
#> GSM198769 1 0.0779 0.982 0.980 0.000 0.016 0.004
#> GSM198772 1 0.0000 0.998 1.000 0.000 0.000 0.000
#> GSM198773 1 0.0000 0.998 1.000 0.000 0.000 0.000
#> GSM198776 1 0.0000 0.998 1.000 0.000 0.000 0.000
#> GSM198778 3 0.5088 0.362 0.424 0.000 0.572 0.004
#> GSM198780 3 0.4925 0.357 0.428 0.000 0.572 0.000
#> GSM198781 1 0.0000 0.998 1.000 0.000 0.000 0.000
#> GSM198765 3 0.1389 0.874 0.000 0.048 0.952 0.000
#> GSM198766 3 0.0376 0.869 0.000 0.004 0.992 0.004
#> GSM198768 3 0.1389 0.874 0.000 0.048 0.952 0.000
#> GSM198770 3 0.1389 0.874 0.000 0.048 0.952 0.000
#> GSM198771 3 0.0469 0.874 0.000 0.012 0.988 0.000
#> GSM198774 3 0.1389 0.874 0.000 0.048 0.952 0.000
#> GSM198775 3 0.0376 0.869 0.000 0.004 0.992 0.004
#> GSM198777 3 0.1389 0.874 0.000 0.048 0.952 0.000
#> GSM198779 3 0.0469 0.874 0.000 0.012 0.988 0.000
#> GSM587218 4 0.0000 1.000 0.000 0.000 0.000 1.000
#> GSM587219 4 0.0000 1.000 0.000 0.000 0.000 1.000
#> GSM587220 4 0.0000 1.000 0.000 0.000 0.000 1.000
#> GSM587221 4 0.0000 1.000 0.000 0.000 0.000 1.000
#> GSM587222 4 0.0000 1.000 0.000 0.000 0.000 1.000
#> GSM587223 4 0.0000 1.000 0.000 0.000 0.000 1.000
#> GSM587224 4 0.0000 1.000 0.000 0.000 0.000 1.000
#> GSM587225 4 0.0000 1.000 0.000 0.000 0.000 1.000
#> GSM587226 4 0.0000 1.000 0.000 0.000 0.000 1.000
#> GSM587227 4 0.0000 1.000 0.000 0.000 0.000 1.000
#> GSM587228 4 0.0000 1.000 0.000 0.000 0.000 1.000
#> GSM587229 4 0.0000 1.000 0.000 0.000 0.000 1.000
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM587155 2 0.2471 0.7852 0.000 0.864 0.000 0 0.136
#> GSM587156 2 0.4404 0.5332 0.000 0.712 0.036 0 0.252
#> GSM587157 2 0.2471 0.7852 0.000 0.864 0.000 0 0.136
#> GSM587158 2 0.0162 0.8877 0.000 0.996 0.000 0 0.004
#> GSM587159 2 0.0000 0.8884 0.000 1.000 0.000 0 0.000
#> GSM587160 2 0.0162 0.8877 0.000 0.996 0.000 0 0.004
#> GSM587161 2 0.0609 0.8823 0.000 0.980 0.000 0 0.020
#> GSM587162 2 0.0404 0.8846 0.000 0.988 0.000 0 0.012
#> GSM587163 2 0.0609 0.8823 0.000 0.980 0.000 0 0.020
#> GSM587164 2 0.2516 0.7808 0.000 0.860 0.000 0 0.140
#> GSM587165 2 0.0000 0.8884 0.000 1.000 0.000 0 0.000
#> GSM587166 2 0.4404 0.5332 0.000 0.712 0.036 0 0.252
#> GSM587167 2 0.4430 0.5244 0.000 0.708 0.036 0 0.256
#> GSM587168 2 0.0000 0.8884 0.000 1.000 0.000 0 0.000
#> GSM587169 2 0.0290 0.8868 0.000 0.992 0.000 0 0.008
#> GSM587170 2 0.2516 0.7808 0.000 0.860 0.000 0 0.140
#> GSM587171 2 0.0000 0.8884 0.000 1.000 0.000 0 0.000
#> GSM587172 2 0.0000 0.8884 0.000 1.000 0.000 0 0.000
#> GSM587173 2 0.0000 0.8884 0.000 1.000 0.000 0 0.000
#> GSM587174 2 0.0000 0.8884 0.000 1.000 0.000 0 0.000
#> GSM587175 2 0.2471 0.7852 0.000 0.864 0.000 0 0.136
#> GSM587176 2 0.1270 0.8587 0.000 0.948 0.000 0 0.052
#> GSM587177 2 0.0000 0.8884 0.000 1.000 0.000 0 0.000
#> GSM587178 2 0.0000 0.8884 0.000 1.000 0.000 0 0.000
#> GSM587179 2 0.0609 0.8823 0.000 0.980 0.000 0 0.020
#> GSM587180 2 0.0000 0.8884 0.000 1.000 0.000 0 0.000
#> GSM587181 2 0.0000 0.8884 0.000 1.000 0.000 0 0.000
#> GSM587182 2 0.0000 0.8884 0.000 1.000 0.000 0 0.000
#> GSM587183 2 0.0000 0.8884 0.000 1.000 0.000 0 0.000
#> GSM587184 2 0.0000 0.8884 0.000 1.000 0.000 0 0.000
#> GSM587185 2 0.0609 0.8823 0.000 0.980 0.000 0 0.020
#> GSM587186 2 0.0000 0.8884 0.000 1.000 0.000 0 0.000
#> GSM587187 2 0.4118 0.0523 0.000 0.660 0.004 0 0.336
#> GSM587188 5 0.4206 0.6203 0.000 0.288 0.016 0 0.696
#> GSM587189 2 0.4872 -0.4210 0.000 0.540 0.024 0 0.436
#> GSM587190 5 0.4637 0.5610 0.000 0.292 0.036 0 0.672
#> GSM587203 1 0.0000 0.9956 1.000 0.000 0.000 0 0.000
#> GSM587204 1 0.0000 0.9956 1.000 0.000 0.000 0 0.000
#> GSM587205 1 0.0000 0.9956 1.000 0.000 0.000 0 0.000
#> GSM587206 1 0.0000 0.9956 1.000 0.000 0.000 0 0.000
#> GSM587207 1 0.0000 0.9956 1.000 0.000 0.000 0 0.000
#> GSM587208 1 0.0000 0.9956 1.000 0.000 0.000 0 0.000
#> GSM587209 1 0.1041 0.9660 0.964 0.000 0.004 0 0.032
#> GSM587210 3 0.5151 0.4075 0.396 0.000 0.560 0 0.044
#> GSM587211 1 0.0000 0.9956 1.000 0.000 0.000 0 0.000
#> GSM587212 3 0.5036 0.3991 0.404 0.000 0.560 0 0.036
#> GSM587213 1 0.0000 0.9956 1.000 0.000 0.000 0 0.000
#> GSM587214 1 0.0000 0.9956 1.000 0.000 0.000 0 0.000
#> GSM587215 1 0.0000 0.9956 1.000 0.000 0.000 0 0.000
#> GSM587216 1 0.0290 0.9896 0.992 0.000 0.000 0 0.008
#> GSM587217 1 0.0000 0.9956 1.000 0.000 0.000 0 0.000
#> GSM587191 3 0.1082 0.8542 0.000 0.028 0.964 0 0.008
#> GSM587192 3 0.1082 0.8542 0.000 0.028 0.964 0 0.008
#> GSM587193 3 0.1478 0.8369 0.000 0.000 0.936 0 0.064
#> GSM587194 3 0.1478 0.8369 0.000 0.000 0.936 0 0.064
#> GSM587195 3 0.1082 0.8542 0.000 0.028 0.964 0 0.008
#> GSM587196 3 0.1082 0.8542 0.000 0.028 0.964 0 0.008
#> GSM587197 3 0.1082 0.8542 0.000 0.028 0.964 0 0.008
#> GSM587198 3 0.0290 0.8525 0.000 0.000 0.992 0 0.008
#> GSM587199 3 0.0290 0.8525 0.000 0.000 0.992 0 0.008
#> GSM587200 3 0.4021 0.7121 0.168 0.000 0.780 0 0.052
#> GSM587201 3 0.4021 0.7121 0.168 0.000 0.780 0 0.052
#> GSM587202 3 0.0290 0.8525 0.000 0.000 0.992 0 0.008
#> GSM198767 1 0.0000 0.9956 1.000 0.000 0.000 0 0.000
#> GSM198769 1 0.1041 0.9660 0.964 0.000 0.004 0 0.032
#> GSM198772 1 0.0000 0.9956 1.000 0.000 0.000 0 0.000
#> GSM198773 1 0.0000 0.9956 1.000 0.000 0.000 0 0.000
#> GSM198776 1 0.0000 0.9956 1.000 0.000 0.000 0 0.000
#> GSM198778 3 0.5151 0.4075 0.396 0.000 0.560 0 0.044
#> GSM198780 3 0.5036 0.3991 0.404 0.000 0.560 0 0.036
#> GSM198781 1 0.0000 0.9956 1.000 0.000 0.000 0 0.000
#> GSM198765 3 0.1082 0.8542 0.000 0.028 0.964 0 0.008
#> GSM198766 3 0.1478 0.8369 0.000 0.000 0.936 0 0.064
#> GSM198768 3 0.1082 0.8542 0.000 0.028 0.964 0 0.008
#> GSM198770 3 0.1082 0.8542 0.000 0.028 0.964 0 0.008
#> GSM198771 3 0.0290 0.8525 0.000 0.000 0.992 0 0.008
#> GSM198774 3 0.1082 0.8542 0.000 0.028 0.964 0 0.008
#> GSM198775 3 0.1478 0.8369 0.000 0.000 0.936 0 0.064
#> GSM198777 3 0.1082 0.8542 0.000 0.028 0.964 0 0.008
#> GSM198779 3 0.0290 0.8525 0.000 0.000 0.992 0 0.008
#> GSM587218 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000
#> GSM587219 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000
#> GSM587220 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000
#> GSM587221 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000
#> GSM587222 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000
#> GSM587223 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000
#> GSM587224 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000
#> GSM587225 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000
#> GSM587226 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000
#> GSM587227 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000
#> GSM587228 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000
#> GSM587229 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM587155 2 0.2260 0.7844 0.000 0.860 0.000 0 0.000 0.140
#> GSM587156 2 0.3713 0.5349 0.000 0.704 0.008 0 0.004 0.284
#> GSM587157 2 0.2260 0.7844 0.000 0.860 0.000 0 0.000 0.140
#> GSM587158 2 0.0146 0.8872 0.000 0.996 0.000 0 0.000 0.004
#> GSM587159 2 0.0000 0.8878 0.000 1.000 0.000 0 0.000 0.000
#> GSM587160 2 0.0146 0.8872 0.000 0.996 0.000 0 0.000 0.004
#> GSM587161 2 0.0547 0.8819 0.000 0.980 0.000 0 0.000 0.020
#> GSM587162 2 0.0363 0.8842 0.000 0.988 0.000 0 0.000 0.012
#> GSM587163 2 0.0547 0.8819 0.000 0.980 0.000 0 0.000 0.020
#> GSM587164 2 0.2300 0.7800 0.000 0.856 0.000 0 0.000 0.144
#> GSM587165 2 0.0000 0.8878 0.000 1.000 0.000 0 0.000 0.000
#> GSM587166 2 0.3713 0.5349 0.000 0.704 0.008 0 0.004 0.284
#> GSM587167 2 0.4127 0.4908 0.000 0.684 0.028 0 0.004 0.284
#> GSM587168 2 0.0000 0.8878 0.000 1.000 0.000 0 0.000 0.000
#> GSM587169 2 0.0260 0.8863 0.000 0.992 0.000 0 0.000 0.008
#> GSM587170 2 0.2300 0.7800 0.000 0.856 0.000 0 0.000 0.144
#> GSM587171 2 0.0000 0.8878 0.000 1.000 0.000 0 0.000 0.000
#> GSM587172 2 0.0000 0.8878 0.000 1.000 0.000 0 0.000 0.000
#> GSM587173 2 0.0000 0.8878 0.000 1.000 0.000 0 0.000 0.000
#> GSM587174 2 0.0000 0.8878 0.000 1.000 0.000 0 0.000 0.000
#> GSM587175 2 0.2260 0.7844 0.000 0.860 0.000 0 0.000 0.140
#> GSM587176 2 0.1141 0.8591 0.000 0.948 0.000 0 0.000 0.052
#> GSM587177 2 0.0000 0.8878 0.000 1.000 0.000 0 0.000 0.000
#> GSM587178 2 0.0000 0.8878 0.000 1.000 0.000 0 0.000 0.000
#> GSM587179 2 0.0547 0.8819 0.000 0.980 0.000 0 0.000 0.020
#> GSM587180 2 0.0000 0.8878 0.000 1.000 0.000 0 0.000 0.000
#> GSM587181 2 0.0000 0.8878 0.000 1.000 0.000 0 0.000 0.000
#> GSM587182 2 0.0000 0.8878 0.000 1.000 0.000 0 0.000 0.000
#> GSM587183 2 0.0000 0.8878 0.000 1.000 0.000 0 0.000 0.000
#> GSM587184 2 0.0000 0.8878 0.000 1.000 0.000 0 0.000 0.000
#> GSM587185 2 0.0547 0.8819 0.000 0.980 0.000 0 0.000 0.020
#> GSM587186 2 0.0000 0.8878 0.000 1.000 0.000 0 0.000 0.000
#> GSM587187 2 0.3984 0.0454 0.000 0.648 0.016 0 0.000 0.336
#> GSM587188 6 0.3534 0.5562 0.000 0.276 0.000 0 0.008 0.716
#> GSM587189 2 0.4819 -0.4612 0.000 0.512 0.044 0 0.004 0.440
#> GSM587190 6 0.4245 0.5734 0.000 0.256 0.044 0 0.004 0.696
#> GSM587203 1 0.0000 0.9677 1.000 0.000 0.000 0 0.000 0.000
#> GSM587204 1 0.0000 0.9677 1.000 0.000 0.000 0 0.000 0.000
#> GSM587205 1 0.0000 0.9677 1.000 0.000 0.000 0 0.000 0.000
#> GSM587206 1 0.0000 0.9677 1.000 0.000 0.000 0 0.000 0.000
#> GSM587207 1 0.0000 0.9677 1.000 0.000 0.000 0 0.000 0.000
#> GSM587208 1 0.0000 0.9677 1.000 0.000 0.000 0 0.000 0.000
#> GSM587209 1 0.2823 0.7560 0.796 0.000 0.000 0 0.204 0.000
#> GSM587210 5 0.3136 0.6021 0.228 0.000 0.004 0 0.768 0.000
#> GSM587211 1 0.0000 0.9677 1.000 0.000 0.000 0 0.000 0.000
#> GSM587212 5 0.3383 0.5892 0.268 0.000 0.004 0 0.728 0.000
#> GSM587213 1 0.0000 0.9677 1.000 0.000 0.000 0 0.000 0.000
#> GSM587214 1 0.0000 0.9677 1.000 0.000 0.000 0 0.000 0.000
#> GSM587215 1 0.0000 0.9677 1.000 0.000 0.000 0 0.000 0.000
#> GSM587216 1 0.2003 0.8577 0.884 0.000 0.000 0 0.116 0.000
#> GSM587217 1 0.0000 0.9677 1.000 0.000 0.000 0 0.000 0.000
#> GSM587191 3 0.0713 0.9424 0.000 0.000 0.972 0 0.028 0.000
#> GSM587192 3 0.0713 0.9424 0.000 0.000 0.972 0 0.028 0.000
#> GSM587193 5 0.4024 0.6318 0.000 0.000 0.184 0 0.744 0.072
#> GSM587194 5 0.4024 0.6318 0.000 0.000 0.184 0 0.744 0.072
#> GSM587195 3 0.0146 0.9405 0.000 0.000 0.996 0 0.000 0.004
#> GSM587196 3 0.0146 0.9405 0.000 0.000 0.996 0 0.000 0.004
#> GSM587197 3 0.0146 0.9405 0.000 0.000 0.996 0 0.000 0.004
#> GSM587198 3 0.2092 0.9017 0.000 0.000 0.876 0 0.124 0.000
#> GSM587199 3 0.2092 0.9017 0.000 0.000 0.876 0 0.124 0.000
#> GSM587200 5 0.3531 0.3500 0.000 0.000 0.328 0 0.672 0.000
#> GSM587201 5 0.3531 0.3500 0.000 0.000 0.328 0 0.672 0.000
#> GSM587202 3 0.2092 0.9017 0.000 0.000 0.876 0 0.124 0.000
#> GSM198767 1 0.0000 0.9677 1.000 0.000 0.000 0 0.000 0.000
#> GSM198769 1 0.2823 0.7560 0.796 0.000 0.000 0 0.204 0.000
#> GSM198772 1 0.0000 0.9677 1.000 0.000 0.000 0 0.000 0.000
#> GSM198773 1 0.0000 0.9677 1.000 0.000 0.000 0 0.000 0.000
#> GSM198776 1 0.0000 0.9677 1.000 0.000 0.000 0 0.000 0.000
#> GSM198778 5 0.3136 0.6021 0.228 0.000 0.004 0 0.768 0.000
#> GSM198780 5 0.3383 0.5892 0.268 0.000 0.004 0 0.728 0.000
#> GSM198781 1 0.0000 0.9677 1.000 0.000 0.000 0 0.000 0.000
#> GSM198765 3 0.0713 0.9424 0.000 0.000 0.972 0 0.028 0.000
#> GSM198766 5 0.4024 0.6318 0.000 0.000 0.184 0 0.744 0.072
#> GSM198768 3 0.0146 0.9405 0.000 0.000 0.996 0 0.000 0.004
#> GSM198770 3 0.0146 0.9405 0.000 0.000 0.996 0 0.000 0.004
#> GSM198771 3 0.2092 0.9017 0.000 0.000 0.876 0 0.124 0.000
#> GSM198774 3 0.0713 0.9424 0.000 0.000 0.972 0 0.028 0.000
#> GSM198775 5 0.4024 0.6318 0.000 0.000 0.184 0 0.744 0.072
#> GSM198777 3 0.0146 0.9405 0.000 0.000 0.996 0 0.000 0.004
#> GSM198779 3 0.2092 0.9017 0.000 0.000 0.876 0 0.124 0.000
#> GSM587218 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587219 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587220 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587221 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587222 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587223 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587224 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587225 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587226 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587227 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587228 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587229 4 0.0000 1.0000 0.000 0.000 0.000 1 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 specimen(p) k
#> SD:hclust 82 1.38e-15 2
#> SD:hclust 88 2.82e-31 3
#> SD:hclust 88 1.15e-45 4
#> SD:hclust 86 2.09e-58 5
#> SD:hclust 87 1.01e-52 6
If matrix rows can be associated to genes, consider to use functional_enrichment(res,
...) to perform function enrichment for the signature genes. See this vignette for more detailed explanations.
The object with results only for a single top-value method and a single partition method can be extracted as:
res = res_list["SD", "kmeans"]
# you can also extract it by
# res = res_list["SD:kmeans"]
A summary of res and all the functions that can be applied to it:
res
#> A 'ConsensusPartition' object with k = 2, 3, 4, 5, 6.
#> On a matrix with 51941 rows and 92 columns.
#> Top rows (1000, 2000, 3000, 4000, 5000) are extracted by 'SD' method.
#> Subgroups are detected by 'kmeans' method.
#> Performed in total 1250 partitions by row resampling.
#> Best k for subgroups seems to be 2.
#>
#> Following methods can be applied to this 'ConsensusPartition' object:
#> [1] "cola_report" "collect_classes" "collect_plots"
#> [4] "collect_stats" "colnames" "compare_signatures"
#> [7] "consensus_heatmap" "dimension_reduction" "functional_enrichment"
#> [10] "get_anno_col" "get_anno" "get_classes"
#> [13] "get_consensus" "get_matrix" "get_membership"
#> [16] "get_param" "get_signatures" "get_stats"
#> [19] "is_best_k" "is_stable_k" "membership_heatmap"
#> [22] "ncol" "nrow" "plot_ecdf"
#> [25] "rownames" "select_partition_number" "show"
#> [28] "suggest_best_k" "test_to_known_factors"
collect_plots() function collects all the plots made from res for all k (number of partitions)
into one single page to provide an easy and fast comparison between different k.
collect_plots(res)
The plots are:
k and the heatmap of
predicted classes for each k.k.k.k.All the plots in panels can be made by individual functions and they are plotted later in this section.
select_partition_number() produces several plots showing different
statistics for choosing “optimized” k. There are following statistics:
k;k, the area increased is defined as \(A_k - A_{k-1}\).The detailed explanations of these statistics can be found in the cola vignette.
Generally speaking, lower PAC score, higher mean silhouette score or higher
concordance corresponds to better partition. Rand index and Jaccard index
measure how similar the current partition is compared to partition with k-1.
If they are too similar, we won't accept k is better than k-1.
select_partition_number(res)
The numeric values for all these statistics can be obtained by get_stats().
get_stats(res)
#> k 1-PAC mean_silhouette concordance area_increased Rand Jaccard
#> 2 2 0.688 0.932 0.952 0.4823 0.500 0.500
#> 3 3 0.673 0.876 0.873 0.3235 0.793 0.605
#> 4 4 0.854 0.926 0.885 0.1164 0.934 0.806
#> 5 5 0.747 0.847 0.840 0.0682 1.000 1.000
#> 6 6 0.727 0.774 0.769 0.0396 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
#> GSM587155 2 0.000 0.940 0.000 1.000
#> GSM587156 2 0.000 0.940 0.000 1.000
#> GSM587157 2 0.000 0.940 0.000 1.000
#> GSM587158 2 0.000 0.940 0.000 1.000
#> GSM587159 2 0.000 0.940 0.000 1.000
#> GSM587160 2 0.000 0.940 0.000 1.000
#> GSM587161 2 0.000 0.940 0.000 1.000
#> GSM587162 2 0.000 0.940 0.000 1.000
#> GSM587163 2 0.000 0.940 0.000 1.000
#> GSM587164 2 0.000 0.940 0.000 1.000
#> GSM587165 2 0.000 0.940 0.000 1.000
#> GSM587166 2 0.000 0.940 0.000 1.000
#> GSM587167 2 0.000 0.940 0.000 1.000
#> GSM587168 2 0.000 0.940 0.000 1.000
#> GSM587169 2 0.000 0.940 0.000 1.000
#> GSM587170 2 0.000 0.940 0.000 1.000
#> GSM587171 2 0.000 0.940 0.000 1.000
#> GSM587172 2 0.000 0.940 0.000 1.000
#> GSM587173 2 0.000 0.940 0.000 1.000
#> GSM587174 2 0.000 0.940 0.000 1.000
#> GSM587175 2 0.000 0.940 0.000 1.000
#> GSM587176 2 0.000 0.940 0.000 1.000
#> GSM587177 2 0.000 0.940 0.000 1.000
#> GSM587178 2 0.000 0.940 0.000 1.000
#> GSM587179 2 0.000 0.940 0.000 1.000
#> GSM587180 2 0.000 0.940 0.000 1.000
#> GSM587181 2 0.000 0.940 0.000 1.000
#> GSM587182 2 0.000 0.940 0.000 1.000
#> GSM587183 2 0.000 0.940 0.000 1.000
#> GSM587184 2 0.000 0.940 0.000 1.000
#> GSM587185 2 0.000 0.940 0.000 1.000
#> GSM587186 2 0.000 0.940 0.000 1.000
#> GSM587187 2 0.000 0.940 0.000 1.000
#> GSM587188 2 0.000 0.940 0.000 1.000
#> GSM587189 2 0.000 0.940 0.000 1.000
#> GSM587190 2 0.000 0.940 0.000 1.000
#> GSM587203 1 0.295 0.975 0.948 0.052
#> GSM587204 1 0.295 0.975 0.948 0.052
#> GSM587205 1 0.295 0.975 0.948 0.052
#> GSM587206 1 0.295 0.975 0.948 0.052
#> GSM587207 1 0.295 0.975 0.948 0.052
#> GSM587208 1 0.295 0.975 0.948 0.052
#> GSM587209 1 0.295 0.975 0.948 0.052
#> GSM587210 1 0.295 0.975 0.948 0.052
#> GSM587211 1 0.295 0.975 0.948 0.052
#> GSM587212 1 0.295 0.975 0.948 0.052
#> GSM587213 1 0.295 0.975 0.948 0.052
#> GSM587214 1 0.295 0.975 0.948 0.052
#> GSM587215 1 0.295 0.975 0.948 0.052
#> GSM587216 1 0.295 0.975 0.948 0.052
#> GSM587217 1 0.295 0.975 0.948 0.052
#> GSM587191 2 0.680 0.830 0.180 0.820
#> GSM587192 1 0.662 0.831 0.828 0.172
#> GSM587193 1 0.295 0.975 0.948 0.052
#> GSM587194 2 0.680 0.830 0.180 0.820
#> GSM587195 2 0.671 0.834 0.176 0.824
#> GSM587196 2 0.680 0.830 0.180 0.820
#> GSM587197 2 0.671 0.834 0.176 0.824
#> GSM587198 2 0.680 0.830 0.180 0.820
#> GSM587199 2 0.680 0.830 0.180 0.820
#> GSM587200 1 0.295 0.975 0.948 0.052
#> GSM587201 1 0.295 0.975 0.948 0.052
#> GSM587202 2 0.680 0.830 0.180 0.820
#> GSM198767 1 0.295 0.975 0.948 0.052
#> GSM198769 1 0.295 0.975 0.948 0.052
#> GSM198772 1 0.295 0.975 0.948 0.052
#> GSM198773 1 0.295 0.975 0.948 0.052
#> GSM198776 1 0.295 0.975 0.948 0.052
#> GSM198778 1 0.295 0.975 0.948 0.052
#> GSM198780 1 0.295 0.975 0.948 0.052
#> GSM198781 1 0.295 0.975 0.948 0.052
#> GSM198765 2 0.680 0.830 0.180 0.820
#> GSM198766 1 0.295 0.975 0.948 0.052
#> GSM198768 2 0.671 0.834 0.176 0.824
#> GSM198770 2 0.671 0.834 0.176 0.824
#> GSM198771 2 0.680 0.830 0.180 0.820
#> GSM198774 1 0.662 0.831 0.828 0.172
#> GSM198775 2 0.680 0.830 0.180 0.820
#> GSM198777 2 0.680 0.830 0.180 0.820
#> GSM198779 2 0.680 0.830 0.180 0.820
#> GSM587218 1 0.000 0.951 1.000 0.000
#> GSM587219 1 0.000 0.951 1.000 0.000
#> GSM587220 1 0.000 0.951 1.000 0.000
#> GSM587221 1 0.000 0.951 1.000 0.000
#> GSM587222 1 0.000 0.951 1.000 0.000
#> GSM587223 1 0.000 0.951 1.000 0.000
#> GSM587224 1 0.000 0.951 1.000 0.000
#> GSM587225 1 0.000 0.951 1.000 0.000
#> GSM587226 1 0.000 0.951 1.000 0.000
#> GSM587227 1 0.000 0.951 1.000 0.000
#> GSM587228 1 0.000 0.951 1.000 0.000
#> GSM587229 1 0.000 0.951 1.000 0.000
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM587155 2 0.0000 0.998 0.000 1.000 0.000
#> GSM587156 2 0.0424 0.990 0.000 0.992 0.008
#> GSM587157 2 0.0000 0.998 0.000 1.000 0.000
#> GSM587158 2 0.0000 0.998 0.000 1.000 0.000
#> GSM587159 2 0.0000 0.998 0.000 1.000 0.000
#> GSM587160 2 0.0000 0.998 0.000 1.000 0.000
#> GSM587161 2 0.0000 0.998 0.000 1.000 0.000
#> GSM587162 2 0.0000 0.998 0.000 1.000 0.000
#> GSM587163 2 0.0000 0.998 0.000 1.000 0.000
#> GSM587164 2 0.0000 0.998 0.000 1.000 0.000
#> GSM587165 2 0.0237 0.996 0.000 0.996 0.004
#> GSM587166 2 0.0747 0.980 0.000 0.984 0.016
#> GSM587167 2 0.0000 0.998 0.000 1.000 0.000
#> GSM587168 2 0.0237 0.996 0.000 0.996 0.004
#> GSM587169 2 0.0000 0.998 0.000 1.000 0.000
#> GSM587170 2 0.0000 0.998 0.000 1.000 0.000
#> GSM587171 2 0.0000 0.998 0.000 1.000 0.000
#> GSM587172 2 0.0000 0.998 0.000 1.000 0.000
#> GSM587173 2 0.0237 0.996 0.000 0.996 0.004
#> GSM587174 2 0.0000 0.998 0.000 1.000 0.000
#> GSM587175 2 0.0000 0.998 0.000 1.000 0.000
#> GSM587176 2 0.0000 0.998 0.000 1.000 0.000
#> GSM587177 2 0.0237 0.996 0.000 0.996 0.004
#> GSM587178 2 0.0000 0.998 0.000 1.000 0.000
#> GSM587179 2 0.0000 0.998 0.000 1.000 0.000
#> GSM587180 2 0.0237 0.996 0.000 0.996 0.004
#> GSM587181 2 0.0000 0.998 0.000 1.000 0.000
#> GSM587182 2 0.0000 0.998 0.000 1.000 0.000
#> GSM587183 2 0.0237 0.996 0.000 0.996 0.004
#> GSM587184 2 0.0000 0.998 0.000 1.000 0.000
#> GSM587185 2 0.0000 0.998 0.000 1.000 0.000
#> GSM587186 2 0.0237 0.996 0.000 0.996 0.004
#> GSM587187 2 0.0237 0.996 0.000 0.996 0.004
#> GSM587188 2 0.0424 0.993 0.000 0.992 0.008
#> GSM587189 2 0.0424 0.993 0.000 0.992 0.008
#> GSM587190 3 0.5621 0.803 0.000 0.308 0.692
#> GSM587203 1 0.5058 0.838 0.756 0.000 0.244
#> GSM587204 1 0.5058 0.838 0.756 0.000 0.244
#> GSM587205 1 0.5058 0.838 0.756 0.000 0.244
#> GSM587206 1 0.5058 0.838 0.756 0.000 0.244
#> GSM587207 1 0.5058 0.838 0.756 0.000 0.244
#> GSM587208 1 0.5058 0.838 0.756 0.000 0.244
#> GSM587209 1 0.5058 0.838 0.756 0.000 0.244
#> GSM587210 1 0.6008 0.794 0.628 0.000 0.372
#> GSM587211 1 0.5058 0.838 0.756 0.000 0.244
#> GSM587212 1 0.5988 0.796 0.632 0.000 0.368
#> GSM587213 1 0.5058 0.838 0.756 0.000 0.244
#> GSM587214 1 0.5058 0.838 0.756 0.000 0.244
#> GSM587215 1 0.5058 0.838 0.756 0.000 0.244
#> GSM587216 1 0.5058 0.838 0.756 0.000 0.244
#> GSM587217 1 0.5058 0.838 0.756 0.000 0.244
#> GSM587191 3 0.5325 0.894 0.004 0.248 0.748
#> GSM587192 3 0.2955 0.762 0.008 0.080 0.912
#> GSM587193 3 0.0592 0.645 0.012 0.000 0.988
#> GSM587194 3 0.5406 0.886 0.012 0.224 0.764
#> GSM587195 3 0.5541 0.893 0.008 0.252 0.740
#> GSM587196 3 0.5541 0.893 0.008 0.252 0.740
#> GSM587197 3 0.5541 0.893 0.008 0.252 0.740
#> GSM587198 3 0.5502 0.895 0.008 0.248 0.744
#> GSM587199 3 0.5619 0.894 0.012 0.244 0.744
#> GSM587200 3 0.0592 0.645 0.012 0.000 0.988
#> GSM587201 3 0.0747 0.644 0.016 0.000 0.984
#> GSM587202 3 0.5502 0.895 0.008 0.248 0.744
#> GSM198767 1 0.5058 0.838 0.756 0.000 0.244
#> GSM198769 1 0.5058 0.838 0.756 0.000 0.244
#> GSM198772 1 0.5058 0.838 0.756 0.000 0.244
#> GSM198773 1 0.5058 0.838 0.756 0.000 0.244
#> GSM198776 1 0.5058 0.838 0.756 0.000 0.244
#> GSM198778 1 0.6008 0.794 0.628 0.000 0.372
#> GSM198780 1 0.5988 0.796 0.632 0.000 0.368
#> GSM198781 1 0.5058 0.838 0.756 0.000 0.244
#> GSM198765 3 0.5325 0.894 0.004 0.248 0.748
#> GSM198766 3 0.0592 0.645 0.012 0.000 0.988
#> GSM198768 3 0.5541 0.893 0.008 0.252 0.740
#> GSM198770 3 0.5541 0.893 0.008 0.252 0.740
#> GSM198771 3 0.5502 0.895 0.008 0.248 0.744
#> GSM198774 3 0.2955 0.762 0.008 0.080 0.912
#> GSM198775 3 0.5406 0.886 0.012 0.224 0.764
#> GSM198777 3 0.5541 0.893 0.008 0.252 0.740
#> GSM198779 3 0.5619 0.894 0.012 0.244 0.744
#> GSM587218 1 0.5560 0.525 0.700 0.000 0.300
#> GSM587219 1 0.4235 0.706 0.824 0.000 0.176
#> GSM587220 1 0.4178 0.709 0.828 0.000 0.172
#> GSM587221 1 0.4235 0.706 0.824 0.000 0.176
#> GSM587222 1 0.4178 0.709 0.828 0.000 0.172
#> GSM587223 1 0.4235 0.706 0.824 0.000 0.176
#> GSM587224 1 0.4235 0.706 0.824 0.000 0.176
#> GSM587225 1 0.4178 0.709 0.828 0.000 0.172
#> GSM587226 1 0.4235 0.706 0.824 0.000 0.176
#> GSM587227 1 0.4178 0.709 0.828 0.000 0.172
#> GSM587228 1 0.4178 0.709 0.828 0.000 0.172
#> GSM587229 1 0.4178 0.709 0.828 0.000 0.172
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM587155 2 0.2266 0.922 0.084 0.912 0.000 0.004
#> GSM587156 2 0.3972 0.853 0.164 0.816 0.016 0.004
#> GSM587157 2 0.2401 0.918 0.092 0.904 0.000 0.004
#> GSM587158 2 0.0188 0.949 0.004 0.996 0.000 0.000
#> GSM587159 2 0.0000 0.949 0.000 1.000 0.000 0.000
#> GSM587160 2 0.0188 0.949 0.004 0.996 0.000 0.000
#> GSM587161 2 0.1209 0.942 0.032 0.964 0.000 0.004
#> GSM587162 2 0.0779 0.947 0.016 0.980 0.000 0.004
#> GSM587163 2 0.0188 0.949 0.004 0.996 0.000 0.000
#> GSM587164 2 0.2401 0.918 0.092 0.904 0.000 0.004
#> GSM587165 2 0.2053 0.939 0.072 0.924 0.000 0.004
#> GSM587166 2 0.4559 0.827 0.164 0.792 0.040 0.004
#> GSM587167 2 0.2530 0.913 0.100 0.896 0.000 0.004
#> GSM587168 2 0.2053 0.939 0.072 0.924 0.000 0.004
#> GSM587169 2 0.0188 0.949 0.004 0.996 0.000 0.000
#> GSM587170 2 0.2401 0.918 0.092 0.904 0.000 0.004
#> GSM587171 2 0.0000 0.949 0.000 1.000 0.000 0.000
#> GSM587172 2 0.0000 0.949 0.000 1.000 0.000 0.000
#> GSM587173 2 0.2053 0.939 0.072 0.924 0.000 0.004
#> GSM587174 2 0.0817 0.949 0.024 0.976 0.000 0.000
#> GSM587175 2 0.2266 0.922 0.084 0.912 0.000 0.004
#> GSM587176 2 0.0657 0.947 0.012 0.984 0.000 0.004
#> GSM587177 2 0.2053 0.939 0.072 0.924 0.000 0.004
#> GSM587178 2 0.1978 0.940 0.068 0.928 0.000 0.004
#> GSM587179 2 0.0779 0.947 0.016 0.980 0.000 0.004
#> GSM587180 2 0.2053 0.939 0.072 0.924 0.000 0.004
#> GSM587181 2 0.0817 0.949 0.024 0.976 0.000 0.000
#> GSM587182 2 0.1978 0.940 0.068 0.928 0.000 0.004
#> GSM587183 2 0.2053 0.939 0.072 0.924 0.000 0.004
#> GSM587184 2 0.0592 0.949 0.016 0.984 0.000 0.000
#> GSM587185 2 0.0779 0.947 0.016 0.980 0.000 0.004
#> GSM587186 2 0.2053 0.939 0.072 0.924 0.000 0.004
#> GSM587187 2 0.2197 0.936 0.080 0.916 0.000 0.004
#> GSM587188 2 0.2586 0.928 0.092 0.900 0.004 0.004
#> GSM587189 2 0.2586 0.928 0.092 0.900 0.004 0.004
#> GSM587190 3 0.3521 0.922 0.084 0.052 0.864 0.000
#> GSM587203 1 0.5231 0.927 0.604 0.000 0.012 0.384
#> GSM587204 1 0.5231 0.927 0.604 0.000 0.012 0.384
#> GSM587205 1 0.5231 0.927 0.604 0.000 0.012 0.384
#> GSM587206 1 0.5231 0.927 0.604 0.000 0.012 0.384
#> GSM587207 1 0.5231 0.927 0.604 0.000 0.012 0.384
#> GSM587208 1 0.5231 0.927 0.604 0.000 0.012 0.384
#> GSM587209 1 0.5659 0.929 0.600 0.000 0.032 0.368
#> GSM587210 1 0.7504 0.668 0.464 0.000 0.192 0.344
#> GSM587211 1 0.5673 0.930 0.596 0.000 0.032 0.372
#> GSM587212 1 0.6961 0.809 0.512 0.000 0.120 0.368
#> GSM587213 1 0.5510 0.932 0.600 0.000 0.024 0.376
#> GSM587214 1 0.5510 0.932 0.600 0.000 0.024 0.376
#> GSM587215 1 0.5673 0.930 0.596 0.000 0.032 0.372
#> GSM587216 1 0.5645 0.926 0.604 0.000 0.032 0.364
#> GSM587217 1 0.5587 0.931 0.600 0.000 0.028 0.372
#> GSM587191 3 0.3160 0.937 0.060 0.040 0.892 0.008
#> GSM587192 3 0.2342 0.920 0.080 0.000 0.912 0.008
#> GSM587193 3 0.2999 0.891 0.132 0.000 0.864 0.004
#> GSM587194 3 0.3931 0.908 0.128 0.040 0.832 0.000
#> GSM587195 3 0.3385 0.932 0.072 0.040 0.880 0.008
#> GSM587196 3 0.3385 0.932 0.072 0.040 0.880 0.008
#> GSM587197 3 0.3160 0.934 0.060 0.040 0.892 0.008
#> GSM587198 3 0.2245 0.940 0.020 0.040 0.932 0.008
#> GSM587199 3 0.2245 0.940 0.020 0.040 0.932 0.008
#> GSM587200 3 0.2342 0.915 0.080 0.000 0.912 0.008
#> GSM587201 3 0.2342 0.915 0.080 0.000 0.912 0.008
#> GSM587202 3 0.2245 0.940 0.020 0.040 0.932 0.008
#> GSM198767 1 0.5231 0.927 0.604 0.000 0.012 0.384
#> GSM198769 1 0.5659 0.929 0.600 0.000 0.032 0.368
#> GSM198772 1 0.5673 0.930 0.596 0.000 0.032 0.372
#> GSM198773 1 0.5510 0.932 0.600 0.000 0.024 0.376
#> GSM198776 1 0.5231 0.927 0.604 0.000 0.012 0.384
#> GSM198778 1 0.7504 0.668 0.464 0.000 0.192 0.344
#> GSM198780 1 0.6961 0.809 0.512 0.000 0.120 0.368
#> GSM198781 1 0.5510 0.932 0.600 0.000 0.024 0.376
#> GSM198765 3 0.3160 0.937 0.060 0.040 0.892 0.008
#> GSM198766 3 0.2999 0.891 0.132 0.000 0.864 0.004
#> GSM198768 3 0.3385 0.932 0.072 0.040 0.880 0.008
#> GSM198770 3 0.3160 0.934 0.060 0.040 0.892 0.008
#> GSM198771 3 0.2245 0.940 0.020 0.040 0.932 0.008
#> GSM198774 3 0.2342 0.920 0.080 0.000 0.912 0.008
#> GSM198775 3 0.3931 0.908 0.128 0.040 0.832 0.000
#> GSM198777 3 0.3385 0.932 0.072 0.040 0.880 0.008
#> GSM198779 3 0.2245 0.940 0.020 0.040 0.932 0.008
#> GSM587218 4 0.3013 0.840 0.032 0.000 0.080 0.888
#> GSM587219 4 0.1004 0.975 0.004 0.000 0.024 0.972
#> GSM587220 4 0.1004 0.975 0.004 0.000 0.024 0.972
#> GSM587221 4 0.0707 0.975 0.000 0.000 0.020 0.980
#> GSM587222 4 0.0707 0.975 0.000 0.000 0.020 0.980
#> GSM587223 4 0.0895 0.974 0.004 0.000 0.020 0.976
#> GSM587224 4 0.0707 0.975 0.000 0.000 0.020 0.980
#> GSM587225 4 0.1398 0.970 0.004 0.000 0.040 0.956
#> GSM587226 4 0.0707 0.975 0.000 0.000 0.020 0.980
#> GSM587227 4 0.1545 0.970 0.008 0.000 0.040 0.952
#> GSM587228 4 0.1398 0.970 0.004 0.000 0.040 0.956
#> GSM587229 4 0.1545 0.970 0.008 0.000 0.040 0.952
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM587155 2 0.4218 0.736 0.000 0.660 0.000 0.008 NA
#> GSM587156 2 0.4936 0.651 0.000 0.564 0.012 0.012 NA
#> GSM587157 2 0.4182 0.723 0.000 0.644 0.000 0.004 NA
#> GSM587158 2 0.0162 0.873 0.000 0.996 0.000 0.000 NA
#> GSM587159 2 0.0404 0.873 0.000 0.988 0.000 0.000 NA
#> GSM587160 2 0.0798 0.872 0.000 0.976 0.000 0.008 NA
#> GSM587161 2 0.2929 0.831 0.000 0.840 0.000 0.008 NA
#> GSM587162 2 0.2077 0.862 0.000 0.908 0.000 0.008 NA
#> GSM587163 2 0.0912 0.872 0.000 0.972 0.000 0.012 NA
#> GSM587164 2 0.4045 0.722 0.000 0.644 0.000 0.000 NA
#> GSM587165 2 0.2735 0.856 0.000 0.880 0.000 0.036 NA
#> GSM587166 2 0.5028 0.646 0.000 0.560 0.016 0.012 NA
#> GSM587167 2 0.4074 0.715 0.000 0.636 0.000 0.000 NA
#> GSM587168 2 0.2946 0.856 0.000 0.868 0.000 0.044 NA
#> GSM587169 2 0.0912 0.872 0.000 0.972 0.000 0.012 NA
#> GSM587170 2 0.4045 0.722 0.000 0.644 0.000 0.000 NA
#> GSM587171 2 0.0404 0.873 0.000 0.988 0.000 0.000 NA
#> GSM587172 2 0.0404 0.873 0.000 0.988 0.000 0.000 NA
#> GSM587173 2 0.2983 0.855 0.000 0.864 0.000 0.040 NA
#> GSM587174 2 0.1211 0.871 0.000 0.960 0.000 0.016 NA
#> GSM587175 2 0.4101 0.737 0.000 0.664 0.000 0.004 NA
#> GSM587176 2 0.1628 0.866 0.000 0.936 0.000 0.008 NA
#> GSM587177 2 0.2735 0.856 0.000 0.880 0.000 0.036 NA
#> GSM587178 2 0.2270 0.860 0.000 0.904 0.000 0.020 NA
#> GSM587179 2 0.1597 0.868 0.000 0.940 0.000 0.012 NA
#> GSM587180 2 0.3019 0.858 0.000 0.864 0.000 0.048 NA
#> GSM587181 2 0.1211 0.871 0.000 0.960 0.000 0.016 NA
#> GSM587182 2 0.2946 0.859 0.000 0.868 0.000 0.044 NA
#> GSM587183 2 0.2735 0.856 0.000 0.880 0.000 0.036 NA
#> GSM587184 2 0.0324 0.873 0.000 0.992 0.000 0.004 NA
#> GSM587185 2 0.1597 0.868 0.000 0.940 0.000 0.012 NA
#> GSM587186 2 0.2983 0.855 0.000 0.864 0.000 0.040 NA
#> GSM587187 2 0.3323 0.848 0.000 0.844 0.000 0.056 NA
#> GSM587188 2 0.4275 0.824 0.000 0.796 0.024 0.052 NA
#> GSM587189 2 0.4364 0.826 0.000 0.788 0.024 0.052 NA
#> GSM587190 3 0.4145 0.830 0.000 0.012 0.772 0.028 NA
#> GSM587203 1 0.2179 0.840 0.896 0.000 0.000 0.004 NA
#> GSM587204 1 0.2179 0.840 0.896 0.000 0.000 0.004 NA
#> GSM587205 1 0.2179 0.840 0.896 0.000 0.000 0.004 NA
#> GSM587206 1 0.2179 0.840 0.896 0.000 0.000 0.004 NA
#> GSM587207 1 0.2179 0.840 0.896 0.000 0.000 0.004 NA
#> GSM587208 1 0.2179 0.840 0.896 0.000 0.000 0.004 NA
#> GSM587209 1 0.2052 0.852 0.912 0.000 0.004 0.004 NA
#> GSM587210 1 0.5910 0.633 0.652 0.000 0.092 0.036 NA
#> GSM587211 1 0.2445 0.841 0.884 0.000 0.004 0.004 NA
#> GSM587212 1 0.4604 0.743 0.748 0.000 0.040 0.020 NA
#> GSM587213 1 0.0290 0.859 0.992 0.000 0.000 0.008 NA
#> GSM587214 1 0.0290 0.859 0.992 0.000 0.000 0.008 NA
#> GSM587215 1 0.1928 0.852 0.920 0.000 0.004 0.004 NA
#> GSM587216 1 0.3299 0.808 0.828 0.000 0.004 0.016 NA
#> GSM587217 1 0.1768 0.853 0.924 0.000 0.000 0.004 NA
#> GSM587191 3 0.3532 0.870 0.004 0.004 0.840 0.044 NA
#> GSM587192 3 0.3881 0.864 0.008 0.000 0.812 0.052 NA
#> GSM587193 3 0.5890 0.789 0.032 0.000 0.640 0.084 NA
#> GSM587194 3 0.5395 0.811 0.004 0.004 0.664 0.084 NA
#> GSM587195 3 0.2227 0.867 0.004 0.004 0.920 0.028 NA
#> GSM587196 3 0.2227 0.867 0.004 0.004 0.920 0.028 NA
#> GSM587197 3 0.2312 0.869 0.004 0.004 0.916 0.032 NA
#> GSM587198 3 0.1471 0.878 0.000 0.004 0.952 0.024 NA
#> GSM587199 3 0.2369 0.877 0.000 0.004 0.908 0.032 NA
#> GSM587200 3 0.4848 0.832 0.012 0.000 0.736 0.076 NA
#> GSM587201 3 0.4848 0.832 0.012 0.000 0.736 0.076 NA
#> GSM587202 3 0.1560 0.878 0.000 0.004 0.948 0.028 NA
#> GSM198767 1 0.2179 0.840 0.896 0.000 0.000 0.004 NA
#> GSM198769 1 0.2052 0.852 0.912 0.000 0.004 0.004 NA
#> GSM198772 1 0.2445 0.841 0.884 0.000 0.004 0.004 NA
#> GSM198773 1 0.0290 0.859 0.992 0.000 0.000 0.008 NA
#> GSM198776 1 0.2179 0.840 0.896 0.000 0.000 0.004 NA
#> GSM198778 1 0.5910 0.633 0.652 0.000 0.092 0.036 NA
#> GSM198780 1 0.4604 0.743 0.748 0.000 0.040 0.020 NA
#> GSM198781 1 0.0290 0.859 0.992 0.000 0.000 0.008 NA
#> GSM198765 3 0.3532 0.870 0.004 0.004 0.840 0.044 NA
#> GSM198766 3 0.5890 0.789 0.032 0.000 0.640 0.084 NA
#> GSM198768 3 0.2227 0.867 0.004 0.004 0.920 0.028 NA
#> GSM198770 3 0.2312 0.869 0.004 0.004 0.916 0.032 NA
#> GSM198771 3 0.1471 0.878 0.000 0.004 0.952 0.024 NA
#> GSM198774 3 0.3881 0.864 0.008 0.000 0.812 0.052 NA
#> GSM198775 3 0.5395 0.811 0.004 0.004 0.664 0.084 NA
#> GSM198777 3 0.2227 0.867 0.004 0.004 0.920 0.028 NA
#> GSM198779 3 0.2369 0.877 0.000 0.004 0.908 0.032 NA
#> GSM587218 4 0.3327 0.888 0.144 0.000 0.028 0.828 NA
#> GSM587219 4 0.3972 0.965 0.212 0.000 0.008 0.764 NA
#> GSM587220 4 0.3883 0.964 0.216 0.000 0.004 0.764 NA
#> GSM587221 4 0.3764 0.964 0.212 0.000 0.008 0.772 NA
#> GSM587222 4 0.3675 0.963 0.216 0.000 0.004 0.772 NA
#> GSM587223 4 0.3487 0.964 0.212 0.000 0.008 0.780 NA
#> GSM587224 4 0.3764 0.964 0.212 0.000 0.008 0.772 NA
#> GSM587225 4 0.5009 0.953 0.216 0.000 0.008 0.704 NA
#> GSM587226 4 0.3764 0.964 0.212 0.000 0.008 0.772 NA
#> GSM587227 4 0.4893 0.952 0.216 0.000 0.008 0.712 NA
#> GSM587228 4 0.5009 0.953 0.216 0.000 0.008 0.704 NA
#> GSM587229 4 0.4893 0.952 0.216 0.000 0.008 0.712 NA
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM587155 2 0.3769 0.689 0.000 0.640 0.000 0.000 NA NA
#> GSM587156 2 0.5125 0.534 0.000 0.480 0.016 0.016 NA NA
#> GSM587157 2 0.3807 0.681 0.000 0.628 0.000 0.000 NA NA
#> GSM587158 2 0.1082 0.820 0.000 0.956 0.000 0.000 NA NA
#> GSM587159 2 0.0260 0.819 0.000 0.992 0.000 0.000 NA NA
#> GSM587160 2 0.0717 0.818 0.000 0.976 0.000 0.000 NA NA
#> GSM587161 2 0.2092 0.791 0.000 0.876 0.000 0.000 NA NA
#> GSM587162 2 0.1858 0.808 0.000 0.912 0.000 0.000 NA NA
#> GSM587163 2 0.0547 0.818 0.000 0.980 0.000 0.000 NA NA
#> GSM587164 2 0.3717 0.672 0.000 0.616 0.000 0.000 NA NA
#> GSM587165 2 0.3266 0.773 0.000 0.728 0.000 0.000 NA NA
#> GSM587166 2 0.5198 0.527 0.000 0.476 0.020 0.016 NA NA
#> GSM587167 2 0.4486 0.653 0.000 0.584 0.000 0.004 NA NA
#> GSM587168 2 0.3629 0.770 0.000 0.712 0.000 0.000 NA NA
#> GSM587169 2 0.0547 0.818 0.000 0.980 0.000 0.000 NA NA
#> GSM587170 2 0.3737 0.670 0.000 0.608 0.000 0.000 NA NA
#> GSM587171 2 0.0260 0.819 0.000 0.992 0.000 0.000 NA NA
#> GSM587172 2 0.0260 0.819 0.000 0.992 0.000 0.000 NA NA
#> GSM587173 2 0.3489 0.766 0.000 0.708 0.000 0.004 NA NA
#> GSM587174 2 0.1462 0.819 0.000 0.936 0.000 0.000 NA NA
#> GSM587175 2 0.3607 0.693 0.000 0.652 0.000 0.000 NA NA
#> GSM587176 2 0.1398 0.814 0.000 0.940 0.000 0.000 NA NA
#> GSM587177 2 0.3221 0.774 0.000 0.736 0.000 0.000 NA NA
#> GSM587178 2 0.2738 0.796 0.000 0.820 0.000 0.000 NA NA
#> GSM587179 2 0.1196 0.816 0.000 0.952 0.000 0.000 NA NA
#> GSM587180 2 0.3641 0.781 0.000 0.732 0.000 0.000 NA NA
#> GSM587181 2 0.1462 0.819 0.000 0.936 0.000 0.000 NA NA
#> GSM587182 2 0.3460 0.790 0.000 0.760 0.000 0.000 NA NA
#> GSM587183 2 0.3221 0.774 0.000 0.736 0.000 0.000 NA NA
#> GSM587184 2 0.0865 0.819 0.000 0.964 0.000 0.000 NA NA
#> GSM587185 2 0.1196 0.816 0.000 0.952 0.000 0.000 NA NA
#> GSM587186 2 0.3489 0.766 0.000 0.708 0.000 0.004 NA NA
#> GSM587187 2 0.3990 0.752 0.000 0.676 0.000 0.016 NA NA
#> GSM587188 2 0.5369 0.689 0.000 0.592 0.036 0.016 NA NA
#> GSM587189 2 0.5355 0.696 0.000 0.596 0.036 0.016 NA NA
#> GSM587190 3 0.5951 0.718 0.000 0.012 0.588 0.024 NA NA
#> GSM587203 1 0.3702 0.777 0.720 0.000 0.000 0.004 NA NA
#> GSM587204 1 0.3767 0.777 0.720 0.000 0.000 0.004 NA NA
#> GSM587205 1 0.3702 0.777 0.720 0.000 0.000 0.004 NA NA
#> GSM587206 1 0.3702 0.777 0.720 0.000 0.000 0.004 NA NA
#> GSM587207 1 0.3702 0.777 0.720 0.000 0.000 0.004 NA NA
#> GSM587208 1 0.3702 0.777 0.720 0.000 0.000 0.004 NA NA
#> GSM587209 1 0.0547 0.791 0.980 0.000 0.000 0.000 NA NA
#> GSM587210 1 0.6030 0.534 0.616 0.000 0.076 0.012 NA NA
#> GSM587211 1 0.1777 0.774 0.928 0.000 0.004 0.000 NA NA
#> GSM587212 1 0.5228 0.600 0.672 0.000 0.048 0.000 NA NA
#> GSM587213 1 0.2288 0.801 0.876 0.000 0.000 0.004 NA NA
#> GSM587214 1 0.2288 0.801 0.876 0.000 0.000 0.004 NA NA
#> GSM587215 1 0.0806 0.788 0.972 0.000 0.000 0.000 NA NA
#> GSM587216 1 0.2437 0.752 0.888 0.000 0.000 0.004 NA NA
#> GSM587217 1 0.0717 0.789 0.976 0.000 0.000 0.000 NA NA
#> GSM587191 3 0.5138 0.761 0.000 0.004 0.684 0.020 NA NA
#> GSM587192 3 0.5615 0.747 0.000 0.004 0.620 0.020 NA NA
#> GSM587193 3 0.7451 0.632 0.068 0.000 0.396 0.040 NA NA
#> GSM587194 3 0.6769 0.651 0.004 0.004 0.412 0.048 NA NA
#> GSM587195 3 0.2924 0.769 0.000 0.004 0.876 0.040 NA NA
#> GSM587196 3 0.2924 0.769 0.000 0.004 0.876 0.040 NA NA
#> GSM587197 3 0.3054 0.773 0.000 0.004 0.868 0.036 NA NA
#> GSM587198 3 0.1218 0.787 0.000 0.004 0.956 0.012 NA NA
#> GSM587199 3 0.3361 0.787 0.000 0.004 0.832 0.012 NA NA
#> GSM587200 3 0.6137 0.730 0.056 0.000 0.636 0.036 NA NA
#> GSM587201 3 0.6137 0.730 0.056 0.000 0.636 0.036 NA NA
#> GSM587202 3 0.1218 0.787 0.000 0.004 0.956 0.012 NA NA
#> GSM198767 1 0.3702 0.777 0.720 0.000 0.000 0.004 NA NA
#> GSM198769 1 0.0547 0.791 0.980 0.000 0.000 0.000 NA NA
#> GSM198772 1 0.1777 0.774 0.928 0.000 0.004 0.000 NA NA
#> GSM198773 1 0.2288 0.801 0.876 0.000 0.000 0.004 NA NA
#> GSM198776 1 0.3767 0.777 0.720 0.000 0.000 0.004 NA NA
#> GSM198778 1 0.6030 0.534 0.616 0.000 0.076 0.012 NA NA
#> GSM198780 1 0.5228 0.600 0.672 0.000 0.048 0.000 NA NA
#> GSM198781 1 0.2288 0.801 0.876 0.000 0.000 0.004 NA NA
#> GSM198765 3 0.5138 0.761 0.000 0.004 0.684 0.020 NA NA
#> GSM198766 3 0.7451 0.632 0.068 0.000 0.396 0.040 NA NA
#> GSM198768 3 0.2924 0.769 0.000 0.004 0.876 0.040 NA NA
#> GSM198770 3 0.3054 0.773 0.000 0.004 0.868 0.036 NA NA
#> GSM198771 3 0.1218 0.787 0.000 0.004 0.956 0.012 NA NA
#> GSM198774 3 0.5615 0.747 0.000 0.004 0.620 0.020 NA NA
#> GSM198775 3 0.6769 0.651 0.004 0.004 0.412 0.048 NA NA
#> GSM198777 3 0.2924 0.769 0.000 0.004 0.876 0.040 NA NA
#> GSM198779 3 0.3361 0.787 0.000 0.004 0.832 0.012 NA NA
#> GSM587218 4 0.2757 0.909 0.084 0.000 0.020 0.876 NA NA
#> GSM587219 4 0.3395 0.949 0.132 0.000 0.000 0.820 NA NA
#> GSM587220 4 0.3395 0.949 0.132 0.000 0.000 0.820 NA NA
#> GSM587221 4 0.2178 0.950 0.132 0.000 0.000 0.868 NA NA
#> GSM587222 4 0.2178 0.950 0.132 0.000 0.000 0.868 NA NA
#> GSM587223 4 0.2716 0.949 0.132 0.000 0.004 0.852 NA NA
#> GSM587224 4 0.2178 0.950 0.132 0.000 0.000 0.868 NA NA
#> GSM587225 4 0.4737 0.927 0.132 0.000 0.000 0.736 NA NA
#> GSM587226 4 0.2178 0.950 0.132 0.000 0.000 0.868 NA NA
#> GSM587227 4 0.4844 0.926 0.132 0.000 0.000 0.728 NA NA
#> GSM587228 4 0.4737 0.927 0.132 0.000 0.000 0.736 NA NA
#> GSM587229 4 0.4844 0.926 0.132 0.000 0.000 0.728 NA 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 specimen(p) k
#> SD:kmeans 92 4.01e-14 2
#> SD:kmeans 92 7.20e-32 3
#> SD:kmeans 92 4.48e-47 4
#> SD:kmeans 92 4.48e-47 5
#> SD:kmeans 92 4.48e-47 6
If matrix rows can be associated to genes, consider to use functional_enrichment(res,
...) to perform function enrichment for the signature genes. See this vignette for more detailed explanations.
The object with results only for a single top-value method and a single partition method can be extracted as:
res = res_list["SD", "skmeans"]
# you can also extract it by
# res = res_list["SD:skmeans"]
A summary of res and all the functions that can be applied to it:
res
#> A 'ConsensusPartition' object with k = 2, 3, 4, 5, 6.
#> On a matrix with 51941 rows and 92 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 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 1.000 0.988 0.995 0.5010 0.500 0.500
#> 3 3 1.000 0.979 0.989 0.2903 0.821 0.653
#> 4 4 0.982 0.949 0.975 0.1196 0.899 0.721
#> 5 5 0.956 0.929 0.961 0.0358 0.963 0.868
#> 6 6 0.934 0.867 0.906 0.0280 0.994 0.977
suggest_best_k() suggests the best \(k\) based on these statistics. The rules are as follows:
suggest_best_k(res)
#> [1] 5
#> attr(,"optional")
#> [1] 2 3 4
There is also optional best \(k\) = 2 3 4 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
#> GSM587155 2 0.000 0.990 0.000 1.000
#> GSM587156 2 0.000 0.990 0.000 1.000
#> GSM587157 2 0.000 0.990 0.000 1.000
#> GSM587158 2 0.000 0.990 0.000 1.000
#> GSM587159 2 0.000 0.990 0.000 1.000
#> GSM587160 2 0.000 0.990 0.000 1.000
#> GSM587161 2 0.000 0.990 0.000 1.000
#> GSM587162 2 0.000 0.990 0.000 1.000
#> GSM587163 2 0.000 0.990 0.000 1.000
#> GSM587164 2 0.000 0.990 0.000 1.000
#> GSM587165 2 0.000 0.990 0.000 1.000
#> GSM587166 2 0.000 0.990 0.000 1.000
#> GSM587167 2 0.000 0.990 0.000 1.000
#> GSM587168 2 0.000 0.990 0.000 1.000
#> GSM587169 2 0.000 0.990 0.000 1.000
#> GSM587170 2 0.000 0.990 0.000 1.000
#> GSM587171 2 0.000 0.990 0.000 1.000
#> GSM587172 2 0.000 0.990 0.000 1.000
#> GSM587173 2 0.000 0.990 0.000 1.000
#> GSM587174 2 0.000 0.990 0.000 1.000
#> GSM587175 2 0.000 0.990 0.000 1.000
#> GSM587176 2 0.000 0.990 0.000 1.000
#> GSM587177 2 0.000 0.990 0.000 1.000
#> GSM587178 2 0.000 0.990 0.000 1.000
#> GSM587179 2 0.000 0.990 0.000 1.000
#> GSM587180 2 0.000 0.990 0.000 1.000
#> GSM587181 2 0.000 0.990 0.000 1.000
#> GSM587182 2 0.000 0.990 0.000 1.000
#> GSM587183 2 0.000 0.990 0.000 1.000
#> GSM587184 2 0.000 0.990 0.000 1.000
#> GSM587185 2 0.000 0.990 0.000 1.000
#> GSM587186 2 0.000 0.990 0.000 1.000
#> GSM587187 2 0.000 0.990 0.000 1.000
#> GSM587188 2 0.000 0.990 0.000 1.000
#> GSM587189 2 0.000 0.990 0.000 1.000
#> GSM587190 2 0.000 0.990 0.000 1.000
#> GSM587203 1 0.000 1.000 1.000 0.000
#> GSM587204 1 0.000 1.000 1.000 0.000
#> GSM587205 1 0.000 1.000 1.000 0.000
#> GSM587206 1 0.000 1.000 1.000 0.000
#> GSM587207 1 0.000 1.000 1.000 0.000
#> GSM587208 1 0.000 1.000 1.000 0.000
#> GSM587209 1 0.000 1.000 1.000 0.000
#> GSM587210 1 0.000 1.000 1.000 0.000
#> GSM587211 1 0.000 1.000 1.000 0.000
#> GSM587212 1 0.000 1.000 1.000 0.000
#> GSM587213 1 0.000 1.000 1.000 0.000
#> GSM587214 1 0.000 1.000 1.000 0.000
#> GSM587215 1 0.000 1.000 1.000 0.000
#> GSM587216 1 0.000 1.000 1.000 0.000
#> GSM587217 1 0.000 1.000 1.000 0.000
#> GSM587191 2 0.000 0.990 0.000 1.000
#> GSM587192 1 0.000 1.000 1.000 0.000
#> GSM587193 1 0.000 1.000 1.000 0.000
#> GSM587194 2 0.808 0.677 0.248 0.752
#> GSM587195 2 0.000 0.990 0.000 1.000
#> GSM587196 2 0.000 0.990 0.000 1.000
#> GSM587197 2 0.000 0.990 0.000 1.000
#> GSM587198 2 0.000 0.990 0.000 1.000
#> GSM587199 2 0.000 0.990 0.000 1.000
#> GSM587200 1 0.000 1.000 1.000 0.000
#> GSM587201 1 0.000 1.000 1.000 0.000
#> GSM587202 2 0.000 0.990 0.000 1.000
#> GSM198767 1 0.000 1.000 1.000 0.000
#> GSM198769 1 0.000 1.000 1.000 0.000
#> GSM198772 1 0.000 1.000 1.000 0.000
#> GSM198773 1 0.000 1.000 1.000 0.000
#> GSM198776 1 0.000 1.000 1.000 0.000
#> GSM198778 1 0.000 1.000 1.000 0.000
#> GSM198780 1 0.000 1.000 1.000 0.000
#> GSM198781 1 0.000 1.000 1.000 0.000
#> GSM198765 2 0.000 0.990 0.000 1.000
#> GSM198766 1 0.000 1.000 1.000 0.000
#> GSM198768 2 0.000 0.990 0.000 1.000
#> GSM198770 2 0.000 0.990 0.000 1.000
#> GSM198771 2 0.000 0.990 0.000 1.000
#> GSM198774 1 0.000 1.000 1.000 0.000
#> GSM198775 2 0.808 0.677 0.248 0.752
#> GSM198777 2 0.000 0.990 0.000 1.000
#> GSM198779 2 0.000 0.990 0.000 1.000
#> GSM587218 1 0.000 1.000 1.000 0.000
#> GSM587219 1 0.000 1.000 1.000 0.000
#> GSM587220 1 0.000 1.000 1.000 0.000
#> GSM587221 1 0.000 1.000 1.000 0.000
#> GSM587222 1 0.000 1.000 1.000 0.000
#> GSM587223 1 0.000 1.000 1.000 0.000
#> GSM587224 1 0.000 1.000 1.000 0.000
#> GSM587225 1 0.000 1.000 1.000 0.000
#> GSM587226 1 0.000 1.000 1.000 0.000
#> GSM587227 1 0.000 1.000 1.000 0.000
#> GSM587228 1 0.000 1.000 1.000 0.000
#> GSM587229 1 0.000 1.000 1.000 0.000
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM587155 2 0.0000 0.997 0.000 1.000 0.000
#> GSM587156 2 0.0000 0.997 0.000 1.000 0.000
#> GSM587157 2 0.0000 0.997 0.000 1.000 0.000
#> GSM587158 2 0.0000 0.997 0.000 1.000 0.000
#> GSM587159 2 0.0000 0.997 0.000 1.000 0.000
#> GSM587160 2 0.0000 0.997 0.000 1.000 0.000
#> GSM587161 2 0.0000 0.997 0.000 1.000 0.000
#> GSM587162 2 0.0000 0.997 0.000 1.000 0.000
#> GSM587163 2 0.0000 0.997 0.000 1.000 0.000
#> GSM587164 2 0.0000 0.997 0.000 1.000 0.000
#> GSM587165 2 0.0000 0.997 0.000 1.000 0.000
#> GSM587166 2 0.0000 0.997 0.000 1.000 0.000
#> GSM587167 2 0.0000 0.997 0.000 1.000 0.000
#> GSM587168 2 0.0000 0.997 0.000 1.000 0.000
#> GSM587169 2 0.0000 0.997 0.000 1.000 0.000
#> GSM587170 2 0.0000 0.997 0.000 1.000 0.000
#> GSM587171 2 0.0000 0.997 0.000 1.000 0.000
#> GSM587172 2 0.0000 0.997 0.000 1.000 0.000
#> GSM587173 2 0.0000 0.997 0.000 1.000 0.000
#> GSM587174 2 0.0000 0.997 0.000 1.000 0.000
#> GSM587175 2 0.0000 0.997 0.000 1.000 0.000
#> GSM587176 2 0.0000 0.997 0.000 1.000 0.000
#> GSM587177 2 0.0000 0.997 0.000 1.000 0.000
#> GSM587178 2 0.0000 0.997 0.000 1.000 0.000
#> GSM587179 2 0.0000 0.997 0.000 1.000 0.000
#> GSM587180 2 0.0000 0.997 0.000 1.000 0.000
#> GSM587181 2 0.0000 0.997 0.000 1.000 0.000
#> GSM587182 2 0.0000 0.997 0.000 1.000 0.000
#> GSM587183 2 0.0000 0.997 0.000 1.000 0.000
#> GSM587184 2 0.0000 0.997 0.000 1.000 0.000
#> GSM587185 2 0.0000 0.997 0.000 1.000 0.000
#> GSM587186 2 0.0000 0.997 0.000 1.000 0.000
#> GSM587187 2 0.0000 0.997 0.000 1.000 0.000
#> GSM587188 2 0.0000 0.997 0.000 1.000 0.000
#> GSM587189 2 0.0000 0.997 0.000 1.000 0.000
#> GSM587190 2 0.2711 0.899 0.000 0.912 0.088
#> GSM587203 1 0.0000 0.998 1.000 0.000 0.000
#> GSM587204 1 0.0000 0.998 1.000 0.000 0.000
#> GSM587205 1 0.0000 0.998 1.000 0.000 0.000
#> GSM587206 1 0.0000 0.998 1.000 0.000 0.000
#> GSM587207 1 0.0000 0.998 1.000 0.000 0.000
#> GSM587208 1 0.0000 0.998 1.000 0.000 0.000
#> GSM587209 1 0.0000 0.998 1.000 0.000 0.000
#> GSM587210 1 0.0000 0.998 1.000 0.000 0.000
#> GSM587211 1 0.0000 0.998 1.000 0.000 0.000
#> GSM587212 1 0.0000 0.998 1.000 0.000 0.000
#> GSM587213 1 0.0000 0.998 1.000 0.000 0.000
#> GSM587214 1 0.0000 0.998 1.000 0.000 0.000
#> GSM587215 1 0.0000 0.998 1.000 0.000 0.000
#> GSM587216 1 0.0000 0.998 1.000 0.000 0.000
#> GSM587217 1 0.0000 0.998 1.000 0.000 0.000
#> GSM587191 3 0.0000 0.956 0.000 0.000 1.000
#> GSM587192 3 0.0000 0.956 0.000 0.000 1.000
#> GSM587193 1 0.0237 0.996 0.996 0.000 0.004
#> GSM587194 3 0.6224 0.684 0.032 0.240 0.728
#> GSM587195 3 0.0000 0.956 0.000 0.000 1.000
#> GSM587196 3 0.0000 0.956 0.000 0.000 1.000
#> GSM587197 3 0.0000 0.956 0.000 0.000 1.000
#> GSM587198 3 0.0000 0.956 0.000 0.000 1.000
#> GSM587199 3 0.0000 0.956 0.000 0.000 1.000
#> GSM587200 3 0.1643 0.922 0.044 0.000 0.956
#> GSM587201 3 0.5058 0.684 0.244 0.000 0.756
#> GSM587202 3 0.0000 0.956 0.000 0.000 1.000
#> GSM198767 1 0.0000 0.998 1.000 0.000 0.000
#> GSM198769 1 0.0000 0.998 1.000 0.000 0.000
#> GSM198772 1 0.0000 0.998 1.000 0.000 0.000
#> GSM198773 1 0.0000 0.998 1.000 0.000 0.000
#> GSM198776 1 0.0000 0.998 1.000 0.000 0.000
#> GSM198778 1 0.0000 0.998 1.000 0.000 0.000
#> GSM198780 1 0.0000 0.998 1.000 0.000 0.000
#> GSM198781 1 0.0000 0.998 1.000 0.000 0.000
#> GSM198765 3 0.0000 0.956 0.000 0.000 1.000
#> GSM198766 1 0.0237 0.996 0.996 0.000 0.004
#> GSM198768 3 0.0000 0.956 0.000 0.000 1.000
#> GSM198770 3 0.0000 0.956 0.000 0.000 1.000
#> GSM198771 3 0.0000 0.956 0.000 0.000 1.000
#> GSM198774 3 0.0000 0.956 0.000 0.000 1.000
#> GSM198775 3 0.6224 0.684 0.032 0.240 0.728
#> GSM198777 3 0.0000 0.956 0.000 0.000 1.000
#> GSM198779 3 0.0000 0.956 0.000 0.000 1.000
#> GSM587218 1 0.0237 0.997 0.996 0.000 0.004
#> GSM587219 1 0.0237 0.997 0.996 0.000 0.004
#> GSM587220 1 0.0237 0.997 0.996 0.000 0.004
#> GSM587221 1 0.0237 0.997 0.996 0.000 0.004
#> GSM587222 1 0.0237 0.997 0.996 0.000 0.004
#> GSM587223 1 0.0237 0.997 0.996 0.000 0.004
#> GSM587224 1 0.0237 0.997 0.996 0.000 0.004
#> GSM587225 1 0.0237 0.997 0.996 0.000 0.004
#> GSM587226 1 0.0237 0.997 0.996 0.000 0.004
#> GSM587227 1 0.0237 0.997 0.996 0.000 0.004
#> GSM587228 1 0.0237 0.997 0.996 0.000 0.004
#> GSM587229 1 0.0237 0.997 0.996 0.000 0.004
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM587155 2 0.0000 0.997 0.000 1.000 0.000 0.000
#> GSM587156 2 0.0000 0.997 0.000 1.000 0.000 0.000
#> GSM587157 2 0.0000 0.997 0.000 1.000 0.000 0.000
#> GSM587158 2 0.0000 0.997 0.000 1.000 0.000 0.000
#> GSM587159 2 0.0000 0.997 0.000 1.000 0.000 0.000
#> GSM587160 2 0.0000 0.997 0.000 1.000 0.000 0.000
#> GSM587161 2 0.0000 0.997 0.000 1.000 0.000 0.000
#> GSM587162 2 0.0000 0.997 0.000 1.000 0.000 0.000
#> GSM587163 2 0.0000 0.997 0.000 1.000 0.000 0.000
#> GSM587164 2 0.0000 0.997 0.000 1.000 0.000 0.000
#> GSM587165 2 0.0000 0.997 0.000 1.000 0.000 0.000
#> GSM587166 2 0.0000 0.997 0.000 1.000 0.000 0.000
#> GSM587167 2 0.0000 0.997 0.000 1.000 0.000 0.000
#> GSM587168 2 0.0000 0.997 0.000 1.000 0.000 0.000
#> GSM587169 2 0.0000 0.997 0.000 1.000 0.000 0.000
#> GSM587170 2 0.0000 0.997 0.000 1.000 0.000 0.000
#> GSM587171 2 0.0000 0.997 0.000 1.000 0.000 0.000
#> GSM587172 2 0.0000 0.997 0.000 1.000 0.000 0.000
#> GSM587173 2 0.0000 0.997 0.000 1.000 0.000 0.000
#> GSM587174 2 0.0000 0.997 0.000 1.000 0.000 0.000
#> GSM587175 2 0.0000 0.997 0.000 1.000 0.000 0.000
#> GSM587176 2 0.0000 0.997 0.000 1.000 0.000 0.000
#> GSM587177 2 0.0000 0.997 0.000 1.000 0.000 0.000
#> GSM587178 2 0.0000 0.997 0.000 1.000 0.000 0.000
#> GSM587179 2 0.0000 0.997 0.000 1.000 0.000 0.000
#> GSM587180 2 0.0000 0.997 0.000 1.000 0.000 0.000
#> GSM587181 2 0.0000 0.997 0.000 1.000 0.000 0.000
#> GSM587182 2 0.0000 0.997 0.000 1.000 0.000 0.000
#> GSM587183 2 0.0000 0.997 0.000 1.000 0.000 0.000
#> GSM587184 2 0.0000 0.997 0.000 1.000 0.000 0.000
#> GSM587185 2 0.0000 0.997 0.000 1.000 0.000 0.000
#> GSM587186 2 0.0000 0.997 0.000 1.000 0.000 0.000
#> GSM587187 2 0.0000 0.997 0.000 1.000 0.000 0.000
#> GSM587188 2 0.0000 0.997 0.000 1.000 0.000 0.000
#> GSM587189 2 0.0000 0.997 0.000 1.000 0.000 0.000
#> GSM587190 2 0.2412 0.898 0.000 0.908 0.084 0.008
#> GSM587203 1 0.0000 0.979 1.000 0.000 0.000 0.000
#> GSM587204 1 0.0000 0.979 1.000 0.000 0.000 0.000
#> GSM587205 1 0.0000 0.979 1.000 0.000 0.000 0.000
#> GSM587206 1 0.0000 0.979 1.000 0.000 0.000 0.000
#> GSM587207 1 0.0000 0.979 1.000 0.000 0.000 0.000
#> GSM587208 1 0.0000 0.979 1.000 0.000 0.000 0.000
#> GSM587209 1 0.0000 0.979 1.000 0.000 0.000 0.000
#> GSM587210 1 0.0469 0.970 0.988 0.000 0.000 0.012
#> GSM587211 1 0.0000 0.979 1.000 0.000 0.000 0.000
#> GSM587212 1 0.0592 0.967 0.984 0.000 0.000 0.016
#> GSM587213 1 0.0000 0.979 1.000 0.000 0.000 0.000
#> GSM587214 1 0.0000 0.979 1.000 0.000 0.000 0.000
#> GSM587215 1 0.0000 0.979 1.000 0.000 0.000 0.000
#> GSM587216 1 0.0000 0.979 1.000 0.000 0.000 0.000
#> GSM587217 1 0.0000 0.979 1.000 0.000 0.000 0.000
#> GSM587191 3 0.0707 0.954 0.000 0.000 0.980 0.020
#> GSM587192 3 0.0707 0.954 0.000 0.000 0.980 0.020
#> GSM587193 4 0.4088 0.722 0.232 0.000 0.004 0.764
#> GSM587194 4 0.4716 0.715 0.000 0.040 0.196 0.764
#> GSM587195 3 0.0188 0.961 0.000 0.000 0.996 0.004
#> GSM587196 3 0.0188 0.961 0.000 0.000 0.996 0.004
#> GSM587197 3 0.0188 0.961 0.000 0.000 0.996 0.004
#> GSM587198 3 0.0000 0.961 0.000 0.000 1.000 0.000
#> GSM587199 3 0.0000 0.961 0.000 0.000 1.000 0.000
#> GSM587200 3 0.5193 0.267 0.412 0.000 0.580 0.008
#> GSM587201 1 0.5099 0.340 0.612 0.000 0.380 0.008
#> GSM587202 3 0.0000 0.961 0.000 0.000 1.000 0.000
#> GSM198767 1 0.0000 0.979 1.000 0.000 0.000 0.000
#> GSM198769 1 0.0000 0.979 1.000 0.000 0.000 0.000
#> GSM198772 1 0.0000 0.979 1.000 0.000 0.000 0.000
#> GSM198773 1 0.0000 0.979 1.000 0.000 0.000 0.000
#> GSM198776 1 0.0000 0.979 1.000 0.000 0.000 0.000
#> GSM198778 1 0.0469 0.970 0.988 0.000 0.000 0.012
#> GSM198780 1 0.0592 0.967 0.984 0.000 0.000 0.016
#> GSM198781 1 0.0000 0.979 1.000 0.000 0.000 0.000
#> GSM198765 3 0.0707 0.954 0.000 0.000 0.980 0.020
#> GSM198766 4 0.4088 0.722 0.232 0.000 0.004 0.764
#> GSM198768 3 0.0188 0.961 0.000 0.000 0.996 0.004
#> GSM198770 3 0.0188 0.961 0.000 0.000 0.996 0.004
#> GSM198771 3 0.0000 0.961 0.000 0.000 1.000 0.000
#> GSM198774 3 0.0707 0.954 0.000 0.000 0.980 0.020
#> GSM198775 4 0.4716 0.715 0.000 0.040 0.196 0.764
#> GSM198777 3 0.0188 0.961 0.000 0.000 0.996 0.004
#> GSM198779 3 0.0000 0.961 0.000 0.000 1.000 0.000
#> GSM587218 4 0.0817 0.933 0.024 0.000 0.000 0.976
#> GSM587219 4 0.0817 0.933 0.024 0.000 0.000 0.976
#> GSM587220 4 0.0817 0.933 0.024 0.000 0.000 0.976
#> GSM587221 4 0.0817 0.933 0.024 0.000 0.000 0.976
#> GSM587222 4 0.0817 0.933 0.024 0.000 0.000 0.976
#> GSM587223 4 0.0817 0.933 0.024 0.000 0.000 0.976
#> GSM587224 4 0.0817 0.933 0.024 0.000 0.000 0.976
#> GSM587225 4 0.0817 0.933 0.024 0.000 0.000 0.976
#> GSM587226 4 0.0817 0.933 0.024 0.000 0.000 0.976
#> GSM587227 4 0.0817 0.933 0.024 0.000 0.000 0.976
#> GSM587228 4 0.0817 0.933 0.024 0.000 0.000 0.976
#> GSM587229 4 0.0817 0.933 0.024 0.000 0.000 0.976
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM587155 2 0.0162 0.990 0.000 0.996 0.000 0.000 0.004
#> GSM587156 2 0.0609 0.978 0.000 0.980 0.000 0.000 0.020
#> GSM587157 2 0.0162 0.990 0.000 0.996 0.000 0.000 0.004
#> GSM587158 2 0.0000 0.991 0.000 1.000 0.000 0.000 0.000
#> GSM587159 2 0.0000 0.991 0.000 1.000 0.000 0.000 0.000
#> GSM587160 2 0.0000 0.991 0.000 1.000 0.000 0.000 0.000
#> GSM587161 2 0.0162 0.990 0.000 0.996 0.000 0.000 0.004
#> GSM587162 2 0.0000 0.991 0.000 1.000 0.000 0.000 0.000
#> GSM587163 2 0.0162 0.990 0.000 0.996 0.000 0.000 0.004
#> GSM587164 2 0.0162 0.990 0.000 0.996 0.000 0.000 0.004
#> GSM587165 2 0.0000 0.991 0.000 1.000 0.000 0.000 0.000
#> GSM587166 2 0.0609 0.978 0.000 0.980 0.000 0.000 0.020
#> GSM587167 2 0.0162 0.990 0.000 0.996 0.000 0.000 0.004
#> GSM587168 2 0.0000 0.991 0.000 1.000 0.000 0.000 0.000
#> GSM587169 2 0.0162 0.990 0.000 0.996 0.000 0.000 0.004
#> GSM587170 2 0.0162 0.990 0.000 0.996 0.000 0.000 0.004
#> GSM587171 2 0.0000 0.991 0.000 1.000 0.000 0.000 0.000
#> GSM587172 2 0.0000 0.991 0.000 1.000 0.000 0.000 0.000
#> GSM587173 2 0.0162 0.989 0.000 0.996 0.000 0.000 0.004
#> GSM587174 2 0.0000 0.991 0.000 1.000 0.000 0.000 0.000
#> GSM587175 2 0.0162 0.990 0.000 0.996 0.000 0.000 0.004
#> GSM587176 2 0.0162 0.990 0.000 0.996 0.000 0.000 0.004
#> GSM587177 2 0.0000 0.991 0.000 1.000 0.000 0.000 0.000
#> GSM587178 2 0.0000 0.991 0.000 1.000 0.000 0.000 0.000
#> GSM587179 2 0.0162 0.990 0.000 0.996 0.000 0.000 0.004
#> GSM587180 2 0.0000 0.991 0.000 1.000 0.000 0.000 0.000
#> GSM587181 2 0.0000 0.991 0.000 1.000 0.000 0.000 0.000
#> GSM587182 2 0.0000 0.991 0.000 1.000 0.000 0.000 0.000
#> GSM587183 2 0.0000 0.991 0.000 1.000 0.000 0.000 0.000
#> GSM587184 2 0.0000 0.991 0.000 1.000 0.000 0.000 0.000
#> GSM587185 2 0.0162 0.990 0.000 0.996 0.000 0.000 0.004
#> GSM587186 2 0.0162 0.989 0.000 0.996 0.000 0.000 0.004
#> GSM587187 2 0.0162 0.989 0.000 0.996 0.000 0.000 0.004
#> GSM587188 2 0.0162 0.989 0.000 0.996 0.000 0.000 0.004
#> GSM587189 2 0.0162 0.989 0.000 0.996 0.000 0.000 0.004
#> GSM587190 2 0.3789 0.705 0.000 0.768 0.020 0.000 0.212
#> GSM587203 1 0.0000 0.973 1.000 0.000 0.000 0.000 0.000
#> GSM587204 1 0.0000 0.973 1.000 0.000 0.000 0.000 0.000
#> GSM587205 1 0.0000 0.973 1.000 0.000 0.000 0.000 0.000
#> GSM587206 1 0.0000 0.973 1.000 0.000 0.000 0.000 0.000
#> GSM587207 1 0.0000 0.973 1.000 0.000 0.000 0.000 0.000
#> GSM587208 1 0.0000 0.973 1.000 0.000 0.000 0.000 0.000
#> GSM587209 1 0.0000 0.973 1.000 0.000 0.000 0.000 0.000
#> GSM587210 1 0.2561 0.846 0.856 0.000 0.000 0.000 0.144
#> GSM587211 1 0.0000 0.973 1.000 0.000 0.000 0.000 0.000
#> GSM587212 1 0.2377 0.864 0.872 0.000 0.000 0.000 0.128
#> GSM587213 1 0.0000 0.973 1.000 0.000 0.000 0.000 0.000
#> GSM587214 1 0.0000 0.973 1.000 0.000 0.000 0.000 0.000
#> GSM587215 1 0.0000 0.973 1.000 0.000 0.000 0.000 0.000
#> GSM587216 1 0.0000 0.973 1.000 0.000 0.000 0.000 0.000
#> GSM587217 1 0.0000 0.973 1.000 0.000 0.000 0.000 0.000
#> GSM587191 5 0.2773 0.743 0.000 0.000 0.164 0.000 0.836
#> GSM587192 5 0.2605 0.754 0.000 0.000 0.148 0.000 0.852
#> GSM587193 5 0.1628 0.759 0.008 0.000 0.000 0.056 0.936
#> GSM587194 5 0.0510 0.767 0.000 0.000 0.000 0.016 0.984
#> GSM587195 3 0.0000 0.917 0.000 0.000 1.000 0.000 0.000
#> GSM587196 3 0.0000 0.917 0.000 0.000 1.000 0.000 0.000
#> GSM587197 3 0.0162 0.916 0.000 0.000 0.996 0.000 0.004
#> GSM587198 3 0.2280 0.892 0.000 0.000 0.880 0.000 0.120
#> GSM587199 3 0.2929 0.837 0.000 0.000 0.820 0.000 0.180
#> GSM587200 5 0.6441 0.157 0.188 0.000 0.344 0.000 0.468
#> GSM587201 5 0.6766 0.144 0.300 0.000 0.300 0.000 0.400
#> GSM587202 3 0.2179 0.895 0.000 0.000 0.888 0.000 0.112
#> GSM198767 1 0.0000 0.973 1.000 0.000 0.000 0.000 0.000
#> GSM198769 1 0.0000 0.973 1.000 0.000 0.000 0.000 0.000
#> GSM198772 1 0.0000 0.973 1.000 0.000 0.000 0.000 0.000
#> GSM198773 1 0.0000 0.973 1.000 0.000 0.000 0.000 0.000
#> GSM198776 1 0.0000 0.973 1.000 0.000 0.000 0.000 0.000
#> GSM198778 1 0.2561 0.846 0.856 0.000 0.000 0.000 0.144
#> GSM198780 1 0.2377 0.864 0.872 0.000 0.000 0.000 0.128
#> GSM198781 1 0.0000 0.973 1.000 0.000 0.000 0.000 0.000
#> GSM198765 5 0.2773 0.743 0.000 0.000 0.164 0.000 0.836
#> GSM198766 5 0.1628 0.759 0.008 0.000 0.000 0.056 0.936
#> GSM198768 3 0.0000 0.917 0.000 0.000 1.000 0.000 0.000
#> GSM198770 3 0.0162 0.916 0.000 0.000 0.996 0.000 0.004
#> GSM198771 3 0.2280 0.892 0.000 0.000 0.880 0.000 0.120
#> GSM198774 5 0.2605 0.754 0.000 0.000 0.148 0.000 0.852
#> GSM198775 5 0.0510 0.767 0.000 0.000 0.000 0.016 0.984
#> GSM198777 3 0.0000 0.917 0.000 0.000 1.000 0.000 0.000
#> GSM198779 3 0.2929 0.837 0.000 0.000 0.820 0.000 0.180
#> GSM587218 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000
#> GSM587219 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000
#> GSM587220 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000
#> GSM587221 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000
#> GSM587222 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000
#> GSM587223 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000
#> GSM587224 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000
#> GSM587225 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000
#> GSM587226 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000
#> GSM587227 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000
#> GSM587228 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000
#> GSM587229 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM587155 2 0.1398 0.935 0.000 0.940 0.000 0.000 0.052 0.008
#> GSM587156 2 0.2664 0.859 0.000 0.848 0.000 0.000 0.136 0.016
#> GSM587157 2 0.1340 0.940 0.000 0.948 0.004 0.000 0.040 0.008
#> GSM587158 2 0.0146 0.959 0.000 0.996 0.000 0.000 0.004 0.000
#> GSM587159 2 0.0146 0.959 0.000 0.996 0.000 0.000 0.004 0.000
#> GSM587160 2 0.0291 0.959 0.000 0.992 0.000 0.000 0.004 0.004
#> GSM587161 2 0.0508 0.957 0.000 0.984 0.000 0.000 0.012 0.004
#> GSM587162 2 0.0146 0.959 0.000 0.996 0.000 0.000 0.000 0.004
#> GSM587163 2 0.0291 0.959 0.000 0.992 0.000 0.000 0.004 0.004
#> GSM587164 2 0.1625 0.928 0.000 0.928 0.000 0.000 0.060 0.012
#> GSM587165 2 0.0935 0.955 0.000 0.964 0.000 0.000 0.032 0.004
#> GSM587166 2 0.2704 0.855 0.000 0.844 0.000 0.000 0.140 0.016
#> GSM587167 2 0.1779 0.923 0.000 0.920 0.000 0.000 0.064 0.016
#> GSM587168 2 0.0858 0.954 0.000 0.968 0.000 0.000 0.028 0.004
#> GSM587169 2 0.0291 0.959 0.000 0.992 0.000 0.000 0.004 0.004
#> GSM587170 2 0.1779 0.922 0.000 0.920 0.000 0.000 0.064 0.016
#> GSM587171 2 0.0146 0.959 0.000 0.996 0.000 0.000 0.004 0.000
#> GSM587172 2 0.0146 0.959 0.000 0.996 0.000 0.000 0.004 0.000
#> GSM587173 2 0.0935 0.955 0.000 0.964 0.000 0.000 0.032 0.004
#> GSM587174 2 0.0458 0.958 0.000 0.984 0.000 0.000 0.016 0.000
#> GSM587175 2 0.1124 0.946 0.000 0.956 0.000 0.000 0.036 0.008
#> GSM587176 2 0.0146 0.959 0.000 0.996 0.000 0.000 0.000 0.004
#> GSM587177 2 0.0935 0.955 0.000 0.964 0.000 0.000 0.032 0.004
#> GSM587178 2 0.0458 0.958 0.000 0.984 0.000 0.000 0.016 0.000
#> GSM587179 2 0.0291 0.959 0.000 0.992 0.000 0.000 0.004 0.004
#> GSM587180 2 0.0858 0.954 0.000 0.968 0.000 0.000 0.028 0.004
#> GSM587181 2 0.0458 0.958 0.000 0.984 0.000 0.000 0.016 0.000
#> GSM587182 2 0.0458 0.958 0.000 0.984 0.000 0.000 0.016 0.000
#> GSM587183 2 0.0935 0.955 0.000 0.964 0.000 0.000 0.032 0.004
#> GSM587184 2 0.0260 0.959 0.000 0.992 0.000 0.000 0.008 0.000
#> GSM587185 2 0.0291 0.959 0.000 0.992 0.000 0.000 0.004 0.004
#> GSM587186 2 0.0935 0.955 0.000 0.964 0.000 0.000 0.032 0.004
#> GSM587187 2 0.1049 0.952 0.000 0.960 0.000 0.000 0.032 0.008
#> GSM587188 2 0.1511 0.941 0.000 0.940 0.004 0.000 0.044 0.012
#> GSM587189 2 0.1410 0.943 0.000 0.944 0.004 0.000 0.044 0.008
#> GSM587190 2 0.5727 0.504 0.000 0.604 0.028 0.000 0.200 0.168
#> GSM587203 1 0.0508 0.913 0.984 0.000 0.000 0.000 0.012 0.004
#> GSM587204 1 0.0508 0.913 0.984 0.000 0.000 0.000 0.012 0.004
#> GSM587205 1 0.0508 0.913 0.984 0.000 0.000 0.000 0.012 0.004
#> GSM587206 1 0.0508 0.913 0.984 0.000 0.000 0.000 0.012 0.004
#> GSM587207 1 0.0508 0.913 0.984 0.000 0.000 0.000 0.012 0.004
#> GSM587208 1 0.0508 0.913 0.984 0.000 0.000 0.000 0.012 0.004
#> GSM587209 1 0.0713 0.908 0.972 0.000 0.000 0.000 0.028 0.000
#> GSM587210 1 0.4210 0.578 0.636 0.000 0.000 0.000 0.336 0.028
#> GSM587211 1 0.1556 0.881 0.920 0.000 0.000 0.000 0.080 0.000
#> GSM587212 1 0.4181 0.588 0.644 0.000 0.000 0.000 0.328 0.028
#> GSM587213 1 0.0000 0.915 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM587214 1 0.0000 0.915 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM587215 1 0.0000 0.915 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM587216 1 0.1219 0.898 0.948 0.000 0.000 0.000 0.048 0.004
#> GSM587217 1 0.0458 0.911 0.984 0.000 0.000 0.000 0.016 0.000
#> GSM587191 6 0.1007 0.988 0.000 0.000 0.044 0.000 0.000 0.956
#> GSM587192 6 0.1196 0.988 0.000 0.000 0.040 0.000 0.008 0.952
#> GSM587193 5 0.4250 0.434 0.000 0.000 0.000 0.016 0.528 0.456
#> GSM587194 5 0.3684 0.526 0.000 0.000 0.000 0.000 0.628 0.372
#> GSM587195 3 0.0146 0.785 0.000 0.000 0.996 0.000 0.000 0.004
#> GSM587196 3 0.0146 0.785 0.000 0.000 0.996 0.000 0.000 0.004
#> GSM587197 3 0.0458 0.781 0.000 0.000 0.984 0.000 0.000 0.016
#> GSM587198 3 0.4977 0.707 0.000 0.000 0.648 0.000 0.188 0.164
#> GSM587199 3 0.5519 0.613 0.000 0.000 0.548 0.000 0.280 0.172
#> GSM587200 5 0.5431 0.326 0.036 0.000 0.128 0.000 0.652 0.184
#> GSM587201 5 0.5644 0.328 0.064 0.000 0.120 0.000 0.648 0.168
#> GSM587202 3 0.4977 0.707 0.000 0.000 0.648 0.000 0.188 0.164
#> GSM198767 1 0.0508 0.913 0.984 0.000 0.000 0.000 0.012 0.004
#> GSM198769 1 0.0713 0.908 0.972 0.000 0.000 0.000 0.028 0.000
#> GSM198772 1 0.1556 0.881 0.920 0.000 0.000 0.000 0.080 0.000
#> GSM198773 1 0.0000 0.915 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM198776 1 0.0508 0.913 0.984 0.000 0.000 0.000 0.012 0.004
#> GSM198778 1 0.4210 0.578 0.636 0.000 0.000 0.000 0.336 0.028
#> GSM198780 1 0.4181 0.588 0.644 0.000 0.000 0.000 0.328 0.028
#> GSM198781 1 0.0000 0.915 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM198765 6 0.1007 0.988 0.000 0.000 0.044 0.000 0.000 0.956
#> GSM198766 5 0.4250 0.434 0.000 0.000 0.000 0.016 0.528 0.456
#> GSM198768 3 0.0146 0.785 0.000 0.000 0.996 0.000 0.000 0.004
#> GSM198770 3 0.0458 0.781 0.000 0.000 0.984 0.000 0.000 0.016
#> GSM198771 3 0.4977 0.707 0.000 0.000 0.648 0.000 0.188 0.164
#> GSM198774 6 0.1196 0.988 0.000 0.000 0.040 0.000 0.008 0.952
#> GSM198775 5 0.3684 0.526 0.000 0.000 0.000 0.000 0.628 0.372
#> GSM198777 3 0.0146 0.785 0.000 0.000 0.996 0.000 0.000 0.004
#> GSM198779 3 0.5519 0.613 0.000 0.000 0.548 0.000 0.280 0.172
#> GSM587218 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM587219 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM587220 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM587221 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM587222 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM587223 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM587224 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM587225 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM587226 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM587227 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM587228 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM587229 4 0.0000 1.000 0.000 0.000 0.000 1.000 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 specimen(p) k
#> SD:skmeans 92 4.01e-14 2
#> SD:skmeans 92 3.33e-30 3
#> SD:skmeans 90 1.98e-39 4
#> SD:skmeans 90 1.60e-43 5
#> SD:skmeans 88 2.73e-39 6
If matrix rows can be associated to genes, consider to use functional_enrichment(res,
...) to perform function enrichment for the signature genes. See this vignette for more detailed explanations.
The object with results only for a single top-value method and a single partition method can be extracted as:
res = res_list["SD", "pam"]
# you can also extract it by
# res = res_list["SD:pam"]
A summary of res and all the functions that can be applied to it:
res
#> A 'ConsensusPartition' object with k = 2, 3, 4, 5, 6.
#> On a matrix with 51941 rows and 92 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 6.
#>
#> 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 0.853 0.945 0.973 0.5011 0.500 0.500
#> 3 3 0.753 0.871 0.904 0.2056 0.917 0.834
#> 4 4 1.000 0.949 0.982 0.2015 0.803 0.556
#> 5 5 0.989 0.956 0.982 0.0463 0.948 0.817
#> 6 6 0.950 0.916 0.958 0.0307 0.980 0.916
suggest_best_k() suggests the best \(k\) based on these statistics. The rules are as follows:
suggest_best_k(res)
#> [1] 6
#> attr(,"optional")
#> [1] 4 5
There is also optional best \(k\) = 4 5 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
#> GSM587155 2 0.0000 0.950 0.000 1.000
#> GSM587156 2 0.0000 0.950 0.000 1.000
#> GSM587157 2 0.0000 0.950 0.000 1.000
#> GSM587158 2 0.0000 0.950 0.000 1.000
#> GSM587159 2 0.0000 0.950 0.000 1.000
#> GSM587160 2 0.0000 0.950 0.000 1.000
#> GSM587161 2 0.0000 0.950 0.000 1.000
#> GSM587162 2 0.0000 0.950 0.000 1.000
#> GSM587163 2 0.0000 0.950 0.000 1.000
#> GSM587164 2 0.0000 0.950 0.000 1.000
#> GSM587165 2 0.0000 0.950 0.000 1.000
#> GSM587166 2 0.0000 0.950 0.000 1.000
#> GSM587167 2 0.0000 0.950 0.000 1.000
#> GSM587168 2 0.0000 0.950 0.000 1.000
#> GSM587169 2 0.0000 0.950 0.000 1.000
#> GSM587170 2 0.0000 0.950 0.000 1.000
#> GSM587171 2 0.0000 0.950 0.000 1.000
#> GSM587172 2 0.0000 0.950 0.000 1.000
#> GSM587173 2 0.0000 0.950 0.000 1.000
#> GSM587174 2 0.0000 0.950 0.000 1.000
#> GSM587175 2 0.0000 0.950 0.000 1.000
#> GSM587176 2 0.0000 0.950 0.000 1.000
#> GSM587177 2 0.0000 0.950 0.000 1.000
#> GSM587178 2 0.0000 0.950 0.000 1.000
#> GSM587179 2 0.0000 0.950 0.000 1.000
#> GSM587180 2 0.0000 0.950 0.000 1.000
#> GSM587181 2 0.0000 0.950 0.000 1.000
#> GSM587182 2 0.0000 0.950 0.000 1.000
#> GSM587183 2 0.0000 0.950 0.000 1.000
#> GSM587184 2 0.0000 0.950 0.000 1.000
#> GSM587185 2 0.0000 0.950 0.000 1.000
#> GSM587186 2 0.0000 0.950 0.000 1.000
#> GSM587187 2 0.0000 0.950 0.000 1.000
#> GSM587188 2 0.0000 0.950 0.000 1.000
#> GSM587189 2 0.0000 0.950 0.000 1.000
#> GSM587190 2 0.0000 0.950 0.000 1.000
#> GSM587203 1 0.0000 1.000 1.000 0.000
#> GSM587204 1 0.0000 1.000 1.000 0.000
#> GSM587205 1 0.0000 1.000 1.000 0.000
#> GSM587206 1 0.0000 1.000 1.000 0.000
#> GSM587207 1 0.0000 1.000 1.000 0.000
#> GSM587208 1 0.0000 1.000 1.000 0.000
#> GSM587209 1 0.0000 1.000 1.000 0.000
#> GSM587210 1 0.0000 1.000 1.000 0.000
#> GSM587211 1 0.0000 1.000 1.000 0.000
#> GSM587212 1 0.0000 1.000 1.000 0.000
#> GSM587213 1 0.0000 1.000 1.000 0.000
#> GSM587214 1 0.0000 1.000 1.000 0.000
#> GSM587215 1 0.0000 1.000 1.000 0.000
#> GSM587216 1 0.0000 1.000 1.000 0.000
#> GSM587217 1 0.0000 1.000 1.000 0.000
#> GSM587191 2 0.0000 0.950 0.000 1.000
#> GSM587192 1 0.0000 1.000 1.000 0.000
#> GSM587193 1 0.0000 1.000 1.000 0.000
#> GSM587194 2 0.5059 0.863 0.112 0.888
#> GSM587195 2 0.8443 0.680 0.272 0.728
#> GSM587196 2 0.8713 0.651 0.292 0.708
#> GSM587197 2 0.6887 0.789 0.184 0.816
#> GSM587198 2 0.8661 0.657 0.288 0.712
#> GSM587199 2 0.0000 0.950 0.000 1.000
#> GSM587200 1 0.0000 1.000 1.000 0.000
#> GSM587201 1 0.0000 1.000 1.000 0.000
#> GSM587202 2 0.8661 0.657 0.288 0.712
#> GSM198767 1 0.0000 1.000 1.000 0.000
#> GSM198769 1 0.0000 1.000 1.000 0.000
#> GSM198772 1 0.0000 1.000 1.000 0.000
#> GSM198773 1 0.0000 1.000 1.000 0.000
#> GSM198776 1 0.0000 1.000 1.000 0.000
#> GSM198778 1 0.0000 1.000 1.000 0.000
#> GSM198780 1 0.0000 1.000 1.000 0.000
#> GSM198781 1 0.0000 1.000 1.000 0.000
#> GSM198765 2 0.0000 0.950 0.000 1.000
#> GSM198766 1 0.0000 1.000 1.000 0.000
#> GSM198768 2 0.8661 0.657 0.288 0.712
#> GSM198770 2 0.0376 0.947 0.004 0.996
#> GSM198771 2 0.9427 0.521 0.360 0.640
#> GSM198774 1 0.0000 1.000 1.000 0.000
#> GSM198775 2 0.4431 0.880 0.092 0.908
#> GSM198777 2 0.8661 0.657 0.288 0.712
#> GSM198779 2 0.0000 0.950 0.000 1.000
#> GSM587218 1 0.0000 1.000 1.000 0.000
#> GSM587219 1 0.0000 1.000 1.000 0.000
#> GSM587220 1 0.0000 1.000 1.000 0.000
#> GSM587221 1 0.0000 1.000 1.000 0.000
#> GSM587222 1 0.0000 1.000 1.000 0.000
#> GSM587223 1 0.0000 1.000 1.000 0.000
#> GSM587224 1 0.0000 1.000 1.000 0.000
#> GSM587225 1 0.0000 1.000 1.000 0.000
#> GSM587226 1 0.0000 1.000 1.000 0.000
#> GSM587227 1 0.0000 1.000 1.000 0.000
#> GSM587228 1 0.0000 1.000 1.000 0.000
#> GSM587229 1 0.0000 1.000 1.000 0.000
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM587155 2 0.0000 0.918 0.000 1.000 0.000
#> GSM587156 2 0.0424 0.915 0.008 0.992 0.000
#> GSM587157 2 0.0000 0.918 0.000 1.000 0.000
#> GSM587158 2 0.0000 0.918 0.000 1.000 0.000
#> GSM587159 2 0.0000 0.918 0.000 1.000 0.000
#> GSM587160 2 0.0000 0.918 0.000 1.000 0.000
#> GSM587161 2 0.0000 0.918 0.000 1.000 0.000
#> GSM587162 2 0.0000 0.918 0.000 1.000 0.000
#> GSM587163 2 0.0000 0.918 0.000 1.000 0.000
#> GSM587164 2 0.0237 0.916 0.004 0.996 0.000
#> GSM587165 2 0.0000 0.918 0.000 1.000 0.000
#> GSM587166 2 0.3686 0.861 0.140 0.860 0.000
#> GSM587167 2 0.0000 0.918 0.000 1.000 0.000
#> GSM587168 2 0.0000 0.918 0.000 1.000 0.000
#> GSM587169 2 0.0000 0.918 0.000 1.000 0.000
#> GSM587170 2 0.0000 0.918 0.000 1.000 0.000
#> GSM587171 2 0.0000 0.918 0.000 1.000 0.000
#> GSM587172 2 0.0000 0.918 0.000 1.000 0.000
#> GSM587173 2 0.0000 0.918 0.000 1.000 0.000
#> GSM587174 2 0.0000 0.918 0.000 1.000 0.000
#> GSM587175 2 0.0000 0.918 0.000 1.000 0.000
#> GSM587176 2 0.0000 0.918 0.000 1.000 0.000
#> GSM587177 2 0.0000 0.918 0.000 1.000 0.000
#> GSM587178 2 0.0000 0.918 0.000 1.000 0.000
#> GSM587179 2 0.0000 0.918 0.000 1.000 0.000
#> GSM587180 2 0.0000 0.918 0.000 1.000 0.000
#> GSM587181 2 0.0000 0.918 0.000 1.000 0.000
#> GSM587182 2 0.0000 0.918 0.000 1.000 0.000
#> GSM587183 2 0.0000 0.918 0.000 1.000 0.000
#> GSM587184 2 0.0000 0.918 0.000 1.000 0.000
#> GSM587185 2 0.0000 0.918 0.000 1.000 0.000
#> GSM587186 2 0.0000 0.918 0.000 1.000 0.000
#> GSM587187 2 0.0000 0.918 0.000 1.000 0.000
#> GSM587188 2 0.0000 0.918 0.000 1.000 0.000
#> GSM587189 2 0.0000 0.918 0.000 1.000 0.000
#> GSM587190 2 0.4750 0.825 0.216 0.784 0.000
#> GSM587203 1 0.4796 0.899 0.780 0.000 0.220
#> GSM587204 1 0.4796 0.899 0.780 0.000 0.220
#> GSM587205 1 0.4796 0.899 0.780 0.000 0.220
#> GSM587206 1 0.4796 0.899 0.780 0.000 0.220
#> GSM587207 1 0.4796 0.899 0.780 0.000 0.220
#> GSM587208 1 0.4796 0.899 0.780 0.000 0.220
#> GSM587209 1 0.4796 0.899 0.780 0.000 0.220
#> GSM587210 1 0.5621 0.840 0.692 0.000 0.308
#> GSM587211 1 0.4796 0.899 0.780 0.000 0.220
#> GSM587212 1 0.5621 0.840 0.692 0.000 0.308
#> GSM587213 1 0.4796 0.899 0.780 0.000 0.220
#> GSM587214 1 0.4796 0.899 0.780 0.000 0.220
#> GSM587215 1 0.4796 0.899 0.780 0.000 0.220
#> GSM587216 1 0.5621 0.840 0.692 0.000 0.308
#> GSM587217 1 0.4796 0.899 0.780 0.000 0.220
#> GSM587191 2 0.4796 0.823 0.220 0.780 0.000
#> GSM587192 1 0.4316 0.600 0.868 0.044 0.088
#> GSM587193 1 0.3695 0.632 0.880 0.012 0.108
#> GSM587194 2 0.4931 0.815 0.232 0.768 0.000
#> GSM587195 2 0.5216 0.795 0.260 0.740 0.000
#> GSM587196 2 0.5254 0.791 0.264 0.736 0.000
#> GSM587197 2 0.5216 0.795 0.260 0.740 0.000
#> GSM587198 2 0.5216 0.795 0.260 0.740 0.000
#> GSM587199 2 0.4796 0.823 0.220 0.780 0.000
#> GSM587200 1 0.2711 0.656 0.912 0.000 0.088
#> GSM587201 1 0.0000 0.688 1.000 0.000 0.000
#> GSM587202 2 0.5216 0.795 0.260 0.740 0.000
#> GSM198767 1 0.4796 0.899 0.780 0.000 0.220
#> GSM198769 1 0.4796 0.899 0.780 0.000 0.220
#> GSM198772 1 0.4796 0.899 0.780 0.000 0.220
#> GSM198773 1 0.4796 0.899 0.780 0.000 0.220
#> GSM198776 1 0.4796 0.899 0.780 0.000 0.220
#> GSM198778 1 0.5621 0.840 0.692 0.000 0.308
#> GSM198780 1 0.5621 0.840 0.692 0.000 0.308
#> GSM198781 1 0.4796 0.899 0.780 0.000 0.220
#> GSM198765 2 0.4796 0.823 0.220 0.780 0.000
#> GSM198766 1 0.4346 0.735 0.816 0.000 0.184
#> GSM198768 2 0.5216 0.795 0.260 0.740 0.000
#> GSM198770 2 0.4796 0.823 0.220 0.780 0.000
#> GSM198771 2 0.6260 0.521 0.448 0.552 0.000
#> GSM198774 1 0.4316 0.600 0.868 0.044 0.088
#> GSM198775 2 0.4887 0.818 0.228 0.772 0.000
#> GSM198777 2 0.5216 0.795 0.260 0.740 0.000
#> GSM198779 2 0.4796 0.823 0.220 0.780 0.000
#> GSM587218 3 0.4796 0.681 0.220 0.000 0.780
#> GSM587219 3 0.0000 0.959 0.000 0.000 1.000
#> GSM587220 3 0.0000 0.959 0.000 0.000 1.000
#> GSM587221 3 0.0000 0.959 0.000 0.000 1.000
#> GSM587222 3 0.0000 0.959 0.000 0.000 1.000
#> GSM587223 3 0.0000 0.959 0.000 0.000 1.000
#> GSM587224 3 0.1964 0.895 0.056 0.000 0.944
#> GSM587225 3 0.0000 0.959 0.000 0.000 1.000
#> GSM587226 3 0.0000 0.959 0.000 0.000 1.000
#> GSM587227 3 0.0000 0.959 0.000 0.000 1.000
#> GSM587228 3 0.0000 0.959 0.000 0.000 1.000
#> GSM587229 3 0.0000 0.959 0.000 0.000 1.000
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM587155 2 0.0000 0.9908 0.000 1.000 0.000 0
#> GSM587156 2 0.0336 0.9831 0.000 0.992 0.008 0
#> GSM587157 2 0.0000 0.9908 0.000 1.000 0.000 0
#> GSM587158 2 0.0000 0.9908 0.000 1.000 0.000 0
#> GSM587159 2 0.0000 0.9908 0.000 1.000 0.000 0
#> GSM587160 2 0.0000 0.9908 0.000 1.000 0.000 0
#> GSM587161 2 0.0000 0.9908 0.000 1.000 0.000 0
#> GSM587162 2 0.0000 0.9908 0.000 1.000 0.000 0
#> GSM587163 2 0.0000 0.9908 0.000 1.000 0.000 0
#> GSM587164 2 0.0336 0.9831 0.000 0.992 0.008 0
#> GSM587165 2 0.0000 0.9908 0.000 1.000 0.000 0
#> GSM587166 2 0.4193 0.6236 0.000 0.732 0.268 0
#> GSM587167 2 0.0000 0.9908 0.000 1.000 0.000 0
#> GSM587168 2 0.0000 0.9908 0.000 1.000 0.000 0
#> GSM587169 2 0.0000 0.9908 0.000 1.000 0.000 0
#> GSM587170 2 0.0000 0.9908 0.000 1.000 0.000 0
#> GSM587171 2 0.0000 0.9908 0.000 1.000 0.000 0
#> GSM587172 2 0.0000 0.9908 0.000 1.000 0.000 0
#> GSM587173 2 0.0000 0.9908 0.000 1.000 0.000 0
#> GSM587174 2 0.0000 0.9908 0.000 1.000 0.000 0
#> GSM587175 2 0.0000 0.9908 0.000 1.000 0.000 0
#> GSM587176 2 0.0000 0.9908 0.000 1.000 0.000 0
#> GSM587177 2 0.0000 0.9908 0.000 1.000 0.000 0
#> GSM587178 2 0.0000 0.9908 0.000 1.000 0.000 0
#> GSM587179 2 0.0000 0.9908 0.000 1.000 0.000 0
#> GSM587180 2 0.0000 0.9908 0.000 1.000 0.000 0
#> GSM587181 2 0.0000 0.9908 0.000 1.000 0.000 0
#> GSM587182 2 0.0000 0.9908 0.000 1.000 0.000 0
#> GSM587183 2 0.0000 0.9908 0.000 1.000 0.000 0
#> GSM587184 2 0.0000 0.9908 0.000 1.000 0.000 0
#> GSM587185 2 0.0000 0.9908 0.000 1.000 0.000 0
#> GSM587186 2 0.0000 0.9908 0.000 1.000 0.000 0
#> GSM587187 2 0.0000 0.9908 0.000 1.000 0.000 0
#> GSM587188 2 0.0000 0.9908 0.000 1.000 0.000 0
#> GSM587189 2 0.0000 0.9908 0.000 1.000 0.000 0
#> GSM587190 3 0.4998 0.0111 0.000 0.488 0.512 0
#> GSM587203 1 0.0000 0.9763 1.000 0.000 0.000 0
#> GSM587204 1 0.0000 0.9763 1.000 0.000 0.000 0
#> GSM587205 1 0.0000 0.9763 1.000 0.000 0.000 0
#> GSM587206 1 0.0000 0.9763 1.000 0.000 0.000 0
#> GSM587207 1 0.0000 0.9763 1.000 0.000 0.000 0
#> GSM587208 1 0.0000 0.9763 1.000 0.000 0.000 0
#> GSM587209 1 0.0000 0.9763 1.000 0.000 0.000 0
#> GSM587210 1 0.4790 0.3768 0.620 0.000 0.380 0
#> GSM587211 1 0.1557 0.9188 0.944 0.000 0.056 0
#> GSM587212 1 0.0000 0.9763 1.000 0.000 0.000 0
#> GSM587213 1 0.0000 0.9763 1.000 0.000 0.000 0
#> GSM587214 1 0.0000 0.9763 1.000 0.000 0.000 0
#> GSM587215 1 0.0000 0.9763 1.000 0.000 0.000 0
#> GSM587216 1 0.0000 0.9763 1.000 0.000 0.000 0
#> GSM587217 1 0.0000 0.9763 1.000 0.000 0.000 0
#> GSM587191 3 0.0000 0.9488 0.000 0.000 1.000 0
#> GSM587192 3 0.0000 0.9488 0.000 0.000 1.000 0
#> GSM587193 3 0.0000 0.9488 0.000 0.000 1.000 0
#> GSM587194 3 0.0000 0.9488 0.000 0.000 1.000 0
#> GSM587195 3 0.0000 0.9488 0.000 0.000 1.000 0
#> GSM587196 3 0.0000 0.9488 0.000 0.000 1.000 0
#> GSM587197 3 0.0000 0.9488 0.000 0.000 1.000 0
#> GSM587198 3 0.0000 0.9488 0.000 0.000 1.000 0
#> GSM587199 3 0.0000 0.9488 0.000 0.000 1.000 0
#> GSM587200 3 0.0000 0.9488 0.000 0.000 1.000 0
#> GSM587201 3 0.0000 0.9488 0.000 0.000 1.000 0
#> GSM587202 3 0.0000 0.9488 0.000 0.000 1.000 0
#> GSM198767 1 0.0000 0.9763 1.000 0.000 0.000 0
#> GSM198769 1 0.0000 0.9763 1.000 0.000 0.000 0
#> GSM198772 1 0.0000 0.9763 1.000 0.000 0.000 0
#> GSM198773 1 0.0000 0.9763 1.000 0.000 0.000 0
#> GSM198776 1 0.0000 0.9763 1.000 0.000 0.000 0
#> GSM198778 3 0.4907 0.2171 0.420 0.000 0.580 0
#> GSM198780 1 0.0000 0.9763 1.000 0.000 0.000 0
#> GSM198781 1 0.0000 0.9763 1.000 0.000 0.000 0
#> GSM198765 3 0.0000 0.9488 0.000 0.000 1.000 0
#> GSM198766 3 0.0000 0.9488 0.000 0.000 1.000 0
#> GSM198768 3 0.0000 0.9488 0.000 0.000 1.000 0
#> GSM198770 3 0.0000 0.9488 0.000 0.000 1.000 0
#> GSM198771 3 0.0000 0.9488 0.000 0.000 1.000 0
#> GSM198774 3 0.0000 0.9488 0.000 0.000 1.000 0
#> GSM198775 3 0.0000 0.9488 0.000 0.000 1.000 0
#> GSM198777 3 0.0000 0.9488 0.000 0.000 1.000 0
#> GSM198779 3 0.0000 0.9488 0.000 0.000 1.000 0
#> GSM587218 4 0.0000 1.0000 0.000 0.000 0.000 1
#> GSM587219 4 0.0000 1.0000 0.000 0.000 0.000 1
#> GSM587220 4 0.0000 1.0000 0.000 0.000 0.000 1
#> GSM587221 4 0.0000 1.0000 0.000 0.000 0.000 1
#> GSM587222 4 0.0000 1.0000 0.000 0.000 0.000 1
#> GSM587223 4 0.0000 1.0000 0.000 0.000 0.000 1
#> GSM587224 4 0.0000 1.0000 0.000 0.000 0.000 1
#> GSM587225 4 0.0000 1.0000 0.000 0.000 0.000 1
#> GSM587226 4 0.0000 1.0000 0.000 0.000 0.000 1
#> GSM587227 4 0.0000 1.0000 0.000 0.000 0.000 1
#> GSM587228 4 0.0000 1.0000 0.000 0.000 0.000 1
#> GSM587229 4 0.0000 1.0000 0.000 0.000 0.000 1
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM587155 2 0.000 0.991 0.000 1.000 0.000 0 0.000
#> GSM587156 2 0.029 0.983 0.000 0.992 0.008 0 0.000
#> GSM587157 2 0.000 0.991 0.000 1.000 0.000 0 0.000
#> GSM587158 2 0.000 0.991 0.000 1.000 0.000 0 0.000
#> GSM587159 2 0.000 0.991 0.000 1.000 0.000 0 0.000
#> GSM587160 2 0.000 0.991 0.000 1.000 0.000 0 0.000
#> GSM587161 2 0.000 0.991 0.000 1.000 0.000 0 0.000
#> GSM587162 2 0.000 0.991 0.000 1.000 0.000 0 0.000
#> GSM587163 2 0.000 0.991 0.000 1.000 0.000 0 0.000
#> GSM587164 2 0.029 0.983 0.000 0.992 0.008 0 0.000
#> GSM587165 2 0.000 0.991 0.000 1.000 0.000 0 0.000
#> GSM587166 2 0.361 0.619 0.000 0.732 0.268 0 0.000
#> GSM587167 2 0.000 0.991 0.000 1.000 0.000 0 0.000
#> GSM587168 2 0.000 0.991 0.000 1.000 0.000 0 0.000
#> GSM587169 2 0.000 0.991 0.000 1.000 0.000 0 0.000
#> GSM587170 2 0.000 0.991 0.000 1.000 0.000 0 0.000
#> GSM587171 2 0.000 0.991 0.000 1.000 0.000 0 0.000
#> GSM587172 2 0.000 0.991 0.000 1.000 0.000 0 0.000
#> GSM587173 2 0.000 0.991 0.000 1.000 0.000 0 0.000
#> GSM587174 2 0.000 0.991 0.000 1.000 0.000 0 0.000
#> GSM587175 2 0.000 0.991 0.000 1.000 0.000 0 0.000
#> GSM587176 2 0.000 0.991 0.000 1.000 0.000 0 0.000
#> GSM587177 2 0.000 0.991 0.000 1.000 0.000 0 0.000
#> GSM587178 2 0.000 0.991 0.000 1.000 0.000 0 0.000
#> GSM587179 2 0.000 0.991 0.000 1.000 0.000 0 0.000
#> GSM587180 2 0.000 0.991 0.000 1.000 0.000 0 0.000
#> GSM587181 2 0.000 0.991 0.000 1.000 0.000 0 0.000
#> GSM587182 2 0.000 0.991 0.000 1.000 0.000 0 0.000
#> GSM587183 2 0.000 0.991 0.000 1.000 0.000 0 0.000
#> GSM587184 2 0.000 0.991 0.000 1.000 0.000 0 0.000
#> GSM587185 2 0.000 0.991 0.000 1.000 0.000 0 0.000
#> GSM587186 2 0.000 0.991 0.000 1.000 0.000 0 0.000
#> GSM587187 2 0.000 0.991 0.000 1.000 0.000 0 0.000
#> GSM587188 2 0.000 0.991 0.000 1.000 0.000 0 0.000
#> GSM587189 2 0.000 0.991 0.000 1.000 0.000 0 0.000
#> GSM587190 3 0.421 0.284 0.000 0.412 0.588 0 0.000
#> GSM587203 5 0.000 0.960 0.000 0.000 0.000 0 1.000
#> GSM587204 5 0.000 0.960 0.000 0.000 0.000 0 1.000
#> GSM587205 5 0.000 0.960 0.000 0.000 0.000 0 1.000
#> GSM587206 5 0.000 0.960 0.000 0.000 0.000 0 1.000
#> GSM587207 5 0.000 0.960 0.000 0.000 0.000 0 1.000
#> GSM587208 5 0.000 0.960 0.000 0.000 0.000 0 1.000
#> GSM587209 1 0.000 0.948 1.000 0.000 0.000 0 0.000
#> GSM587210 1 0.000 0.948 1.000 0.000 0.000 0 0.000
#> GSM587211 1 0.000 0.948 1.000 0.000 0.000 0 0.000
#> GSM587212 1 0.000 0.948 1.000 0.000 0.000 0 0.000
#> GSM587213 1 0.223 0.852 0.884 0.000 0.000 0 0.116
#> GSM587214 5 0.285 0.813 0.172 0.000 0.000 0 0.828
#> GSM587215 1 0.000 0.948 1.000 0.000 0.000 0 0.000
#> GSM587216 1 0.000 0.948 1.000 0.000 0.000 0 0.000
#> GSM587217 1 0.000 0.948 1.000 0.000 0.000 0 0.000
#> GSM587191 3 0.000 0.970 0.000 0.000 1.000 0 0.000
#> GSM587192 3 0.000 0.970 0.000 0.000 1.000 0 0.000
#> GSM587193 1 0.368 0.619 0.720 0.000 0.280 0 0.000
#> GSM587194 3 0.000 0.970 0.000 0.000 1.000 0 0.000
#> GSM587195 3 0.000 0.970 0.000 0.000 1.000 0 0.000
#> GSM587196 3 0.000 0.970 0.000 0.000 1.000 0 0.000
#> GSM587197 3 0.000 0.970 0.000 0.000 1.000 0 0.000
#> GSM587198 3 0.000 0.970 0.000 0.000 1.000 0 0.000
#> GSM587199 3 0.000 0.970 0.000 0.000 1.000 0 0.000
#> GSM587200 3 0.000 0.970 0.000 0.000 1.000 0 0.000
#> GSM587201 3 0.000 0.970 0.000 0.000 1.000 0 0.000
#> GSM587202 3 0.000 0.970 0.000 0.000 1.000 0 0.000
#> GSM198767 5 0.000 0.960 0.000 0.000 0.000 0 1.000
#> GSM198769 1 0.000 0.948 1.000 0.000 0.000 0 0.000
#> GSM198772 1 0.000 0.948 1.000 0.000 0.000 0 0.000
#> GSM198773 1 0.223 0.852 0.884 0.000 0.000 0 0.116
#> GSM198776 5 0.000 0.960 0.000 0.000 0.000 0 1.000
#> GSM198778 1 0.000 0.948 1.000 0.000 0.000 0 0.000
#> GSM198780 1 0.000 0.948 1.000 0.000 0.000 0 0.000
#> GSM198781 5 0.285 0.813 0.172 0.000 0.000 0 0.828
#> GSM198765 3 0.000 0.970 0.000 0.000 1.000 0 0.000
#> GSM198766 1 0.191 0.861 0.908 0.000 0.092 0 0.000
#> GSM198768 3 0.000 0.970 0.000 0.000 1.000 0 0.000
#> GSM198770 3 0.000 0.970 0.000 0.000 1.000 0 0.000
#> GSM198771 3 0.000 0.970 0.000 0.000 1.000 0 0.000
#> GSM198774 3 0.000 0.970 0.000 0.000 1.000 0 0.000
#> GSM198775 3 0.000 0.970 0.000 0.000 1.000 0 0.000
#> GSM198777 3 0.000 0.970 0.000 0.000 1.000 0 0.000
#> GSM198779 3 0.000 0.970 0.000 0.000 1.000 0 0.000
#> GSM587218 4 0.000 1.000 0.000 0.000 0.000 1 0.000
#> GSM587219 4 0.000 1.000 0.000 0.000 0.000 1 0.000
#> GSM587220 4 0.000 1.000 0.000 0.000 0.000 1 0.000
#> GSM587221 4 0.000 1.000 0.000 0.000 0.000 1 0.000
#> GSM587222 4 0.000 1.000 0.000 0.000 0.000 1 0.000
#> GSM587223 4 0.000 1.000 0.000 0.000 0.000 1 0.000
#> GSM587224 4 0.000 1.000 0.000 0.000 0.000 1 0.000
#> GSM587225 4 0.000 1.000 0.000 0.000 0.000 1 0.000
#> GSM587226 4 0.000 1.000 0.000 0.000 0.000 1 0.000
#> GSM587227 4 0.000 1.000 0.000 0.000 0.000 1 0.000
#> GSM587228 4 0.000 1.000 0.000 0.000 0.000 1 0.000
#> GSM587229 4 0.000 1.000 0.000 0.000 0.000 1 0.000
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM587155 2 0.0000 0.981 0.000 1.000 0.000 0 0.000 0.000
#> GSM587156 2 0.0146 0.978 0.000 0.996 0.004 0 0.000 0.000
#> GSM587157 2 0.3409 0.592 0.000 0.700 0.300 0 0.000 0.000
#> GSM587158 2 0.0000 0.981 0.000 1.000 0.000 0 0.000 0.000
#> GSM587159 2 0.0000 0.981 0.000 1.000 0.000 0 0.000 0.000
#> GSM587160 2 0.0000 0.981 0.000 1.000 0.000 0 0.000 0.000
#> GSM587161 2 0.0000 0.981 0.000 1.000 0.000 0 0.000 0.000
#> GSM587162 2 0.0000 0.981 0.000 1.000 0.000 0 0.000 0.000
#> GSM587163 2 0.0000 0.981 0.000 1.000 0.000 0 0.000 0.000
#> GSM587164 2 0.0363 0.972 0.000 0.988 0.012 0 0.000 0.000
#> GSM587165 2 0.0000 0.981 0.000 1.000 0.000 0 0.000 0.000
#> GSM587166 2 0.4162 0.667 0.000 0.744 0.136 0 0.120 0.000
#> GSM587167 2 0.0632 0.961 0.000 0.976 0.024 0 0.000 0.000
#> GSM587168 2 0.0000 0.981 0.000 1.000 0.000 0 0.000 0.000
#> GSM587169 2 0.0000 0.981 0.000 1.000 0.000 0 0.000 0.000
#> GSM587170 2 0.0000 0.981 0.000 1.000 0.000 0 0.000 0.000
#> GSM587171 2 0.0000 0.981 0.000 1.000 0.000 0 0.000 0.000
#> GSM587172 2 0.0000 0.981 0.000 1.000 0.000 0 0.000 0.000
#> GSM587173 2 0.0000 0.981 0.000 1.000 0.000 0 0.000 0.000
#> GSM587174 2 0.0000 0.981 0.000 1.000 0.000 0 0.000 0.000
#> GSM587175 2 0.0146 0.979 0.000 0.996 0.004 0 0.000 0.000
#> GSM587176 2 0.0000 0.981 0.000 1.000 0.000 0 0.000 0.000
#> GSM587177 2 0.0000 0.981 0.000 1.000 0.000 0 0.000 0.000
#> GSM587178 2 0.0000 0.981 0.000 1.000 0.000 0 0.000 0.000
#> GSM587179 2 0.0000 0.981 0.000 1.000 0.000 0 0.000 0.000
#> GSM587180 2 0.0000 0.981 0.000 1.000 0.000 0 0.000 0.000
#> GSM587181 2 0.0000 0.981 0.000 1.000 0.000 0 0.000 0.000
#> GSM587182 2 0.0000 0.981 0.000 1.000 0.000 0 0.000 0.000
#> GSM587183 2 0.0000 0.981 0.000 1.000 0.000 0 0.000 0.000
#> GSM587184 2 0.0000 0.981 0.000 1.000 0.000 0 0.000 0.000
#> GSM587185 2 0.0000 0.981 0.000 1.000 0.000 0 0.000 0.000
#> GSM587186 2 0.0000 0.981 0.000 1.000 0.000 0 0.000 0.000
#> GSM587187 2 0.0000 0.981 0.000 1.000 0.000 0 0.000 0.000
#> GSM587188 2 0.0000 0.981 0.000 1.000 0.000 0 0.000 0.000
#> GSM587189 2 0.0363 0.972 0.000 0.988 0.012 0 0.000 0.000
#> GSM587190 3 0.4781 0.449 0.000 0.296 0.624 0 0.080 0.000
#> GSM587203 6 0.0000 0.928 0.000 0.000 0.000 0 0.000 1.000
#> GSM587204 6 0.0000 0.928 0.000 0.000 0.000 0 0.000 1.000
#> GSM587205 6 0.0000 0.928 0.000 0.000 0.000 0 0.000 1.000
#> GSM587206 6 0.0000 0.928 0.000 0.000 0.000 0 0.000 1.000
#> GSM587207 6 0.0000 0.928 0.000 0.000 0.000 0 0.000 1.000
#> GSM587208 6 0.0000 0.928 0.000 0.000 0.000 0 0.000 1.000
#> GSM587209 1 0.0000 0.914 1.000 0.000 0.000 0 0.000 0.000
#> GSM587210 1 0.1765 0.877 0.904 0.000 0.000 0 0.096 0.000
#> GSM587211 1 0.0000 0.914 1.000 0.000 0.000 0 0.000 0.000
#> GSM587212 1 0.1444 0.891 0.928 0.000 0.000 0 0.072 0.000
#> GSM587213 1 0.2048 0.830 0.880 0.000 0.000 0 0.000 0.120
#> GSM587214 6 0.3288 0.669 0.276 0.000 0.000 0 0.000 0.724
#> GSM587215 1 0.0000 0.914 1.000 0.000 0.000 0 0.000 0.000
#> GSM587216 1 0.0000 0.914 1.000 0.000 0.000 0 0.000 0.000
#> GSM587217 1 0.0000 0.914 1.000 0.000 0.000 0 0.000 0.000
#> GSM587191 5 0.0000 0.997 0.000 0.000 0.000 0 1.000 0.000
#> GSM587192 5 0.0000 0.997 0.000 0.000 0.000 0 1.000 0.000
#> GSM587193 1 0.4967 0.528 0.640 0.000 0.132 0 0.228 0.000
#> GSM587194 5 0.0146 0.995 0.000 0.000 0.004 0 0.996 0.000
#> GSM587195 3 0.0000 0.860 0.000 0.000 1.000 0 0.000 0.000
#> GSM587196 3 0.0000 0.860 0.000 0.000 1.000 0 0.000 0.000
#> GSM587197 3 0.0000 0.860 0.000 0.000 1.000 0 0.000 0.000
#> GSM587198 3 0.1556 0.853 0.000 0.000 0.920 0 0.080 0.000
#> GSM587199 3 0.1910 0.841 0.000 0.000 0.892 0 0.108 0.000
#> GSM587200 3 0.4808 0.579 0.272 0.000 0.636 0 0.092 0.000
#> GSM587201 3 0.4729 0.571 0.284 0.000 0.636 0 0.080 0.000
#> GSM587202 3 0.1556 0.853 0.000 0.000 0.920 0 0.080 0.000
#> GSM198767 6 0.0000 0.928 0.000 0.000 0.000 0 0.000 1.000
#> GSM198769 1 0.0000 0.914 1.000 0.000 0.000 0 0.000 0.000
#> GSM198772 1 0.0000 0.914 1.000 0.000 0.000 0 0.000 0.000
#> GSM198773 1 0.2048 0.830 0.880 0.000 0.000 0 0.000 0.120
#> GSM198776 6 0.0000 0.928 0.000 0.000 0.000 0 0.000 1.000
#> GSM198778 1 0.1814 0.874 0.900 0.000 0.000 0 0.100 0.000
#> GSM198780 1 0.1444 0.891 0.928 0.000 0.000 0 0.072 0.000
#> GSM198781 6 0.3288 0.669 0.276 0.000 0.000 0 0.000 0.724
#> GSM198765 5 0.0000 0.997 0.000 0.000 0.000 0 1.000 0.000
#> GSM198766 1 0.2896 0.777 0.824 0.000 0.016 0 0.160 0.000
#> GSM198768 3 0.0000 0.860 0.000 0.000 1.000 0 0.000 0.000
#> GSM198770 3 0.0000 0.860 0.000 0.000 1.000 0 0.000 0.000
#> GSM198771 3 0.1556 0.853 0.000 0.000 0.920 0 0.080 0.000
#> GSM198774 5 0.0000 0.997 0.000 0.000 0.000 0 1.000 0.000
#> GSM198775 5 0.0260 0.991 0.000 0.000 0.008 0 0.992 0.000
#> GSM198777 3 0.0000 0.860 0.000 0.000 1.000 0 0.000 0.000
#> GSM198779 3 0.1910 0.841 0.000 0.000 0.892 0 0.108 0.000
#> GSM587218 4 0.0000 1.000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587219 4 0.0000 1.000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587220 4 0.0000 1.000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587221 4 0.0000 1.000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587222 4 0.0000 1.000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587223 4 0.0000 1.000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587224 4 0.0000 1.000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587225 4 0.0000 1.000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587226 4 0.0000 1.000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587227 4 0.0000 1.000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587228 4 0.0000 1.000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587229 4 0.0000 1.000 0.000 0.000 0.000 1 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 specimen(p) k
#> SD:pam 92 4.01e-14 2
#> SD:pam 92 1.29e-28 3
#> SD:pam 89 2.80e-46 4
#> SD:pam 91 2.31e-41 5
#> SD:pam 91 1.54e-38 6
If matrix rows can be associated to genes, consider to use functional_enrichment(res,
...) to perform function enrichment for the signature genes. See this vignette for more detailed explanations.
The object with results only for a single top-value method and a single partition method can be extracted as:
res = res_list["SD", "mclust"]
# you can also extract it by
# res = res_list["SD:mclust"]
A summary of res and all the functions that can be applied to it:
res
#> A 'ConsensusPartition' object with k = 2, 3, 4, 5, 6.
#> On a matrix with 51941 rows and 92 columns.
#> Top rows (1000, 2000, 3000, 4000, 5000) are extracted by 'SD' method.
#> Subgroups are detected by 'mclust' method.
#> Performed in total 1250 partitions by row resampling.
#> Best k for subgroups seems to be 4.
#>
#> Following methods can be applied to this 'ConsensusPartition' object:
#> [1] "cola_report" "collect_classes" "collect_plots"
#> [4] "collect_stats" "colnames" "compare_signatures"
#> [7] "consensus_heatmap" "dimension_reduction" "functional_enrichment"
#> [10] "get_anno_col" "get_anno" "get_classes"
#> [13] "get_consensus" "get_matrix" "get_membership"
#> [16] "get_param" "get_signatures" "get_stats"
#> [19] "is_best_k" "is_stable_k" "membership_heatmap"
#> [22] "ncol" "nrow" "plot_ecdf"
#> [25] "rownames" "select_partition_number" "show"
#> [28] "suggest_best_k" "test_to_known_factors"
collect_plots() function collects all the plots made from res for all k (number of partitions)
into one single page to provide an easy and fast comparison between different k.
collect_plots(res)
The plots are:
k and the heatmap of
predicted classes for each k.k.k.k.All the plots in panels can be made by individual functions and they are plotted later in this section.
select_partition_number() produces several plots showing different
statistics for choosing “optimized” k. There are following statistics:
k;k, the area increased is defined as \(A_k - A_{k-1}\).The detailed explanations of these statistics can be found in the cola vignette.
Generally speaking, lower PAC score, higher mean silhouette score or higher
concordance corresponds to better partition. Rand index and Jaccard index
measure how similar the current partition is compared to partition with k-1.
If they are too similar, we won't accept k is better than k-1.
select_partition_number(res)
The numeric values for all these statistics can be obtained by get_stats().
get_stats(res)
#> k 1-PAC mean_silhouette concordance area_increased Rand Jaccard
#> 2 2 0.955 0.943 0.976 0.4650 0.548 0.548
#> 3 3 0.786 0.910 0.960 0.2984 0.610 0.415
#> 4 4 0.947 0.943 0.976 0.2097 0.853 0.637
#> 5 5 0.887 0.796 0.887 0.0435 0.968 0.886
#> 6 6 0.845 0.765 0.871 0.0440 0.951 0.811
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
#> GSM587155 2 0.000 0.963 0.000 1.000
#> GSM587156 2 0.000 0.963 0.000 1.000
#> GSM587157 2 0.000 0.963 0.000 1.000
#> GSM587158 2 0.000 0.963 0.000 1.000
#> GSM587159 2 0.000 0.963 0.000 1.000
#> GSM587160 2 0.000 0.963 0.000 1.000
#> GSM587161 2 0.000 0.963 0.000 1.000
#> GSM587162 2 0.000 0.963 0.000 1.000
#> GSM587163 2 0.000 0.963 0.000 1.000
#> GSM587164 2 0.000 0.963 0.000 1.000
#> GSM587165 2 0.000 0.963 0.000 1.000
#> GSM587166 2 0.000 0.963 0.000 1.000
#> GSM587167 2 0.000 0.963 0.000 1.000
#> GSM587168 2 0.000 0.963 0.000 1.000
#> GSM587169 2 0.000 0.963 0.000 1.000
#> GSM587170 2 0.000 0.963 0.000 1.000
#> GSM587171 2 0.000 0.963 0.000 1.000
#> GSM587172 2 0.000 0.963 0.000 1.000
#> GSM587173 2 0.000 0.963 0.000 1.000
#> GSM587174 2 0.000 0.963 0.000 1.000
#> GSM587175 2 0.000 0.963 0.000 1.000
#> GSM587176 2 0.000 0.963 0.000 1.000
#> GSM587177 2 0.000 0.963 0.000 1.000
#> GSM587178 2 0.000 0.963 0.000 1.000
#> GSM587179 2 0.000 0.963 0.000 1.000
#> GSM587180 2 0.000 0.963 0.000 1.000
#> GSM587181 2 0.000 0.963 0.000 1.000
#> GSM587182 2 0.000 0.963 0.000 1.000
#> GSM587183 2 0.000 0.963 0.000 1.000
#> GSM587184 2 0.000 0.963 0.000 1.000
#> GSM587185 2 0.000 0.963 0.000 1.000
#> GSM587186 2 0.000 0.963 0.000 1.000
#> GSM587187 2 0.000 0.963 0.000 1.000
#> GSM587188 2 0.000 0.963 0.000 1.000
#> GSM587189 2 0.000 0.963 0.000 1.000
#> GSM587190 2 0.000 0.963 0.000 1.000
#> GSM587203 1 0.000 1.000 1.000 0.000
#> GSM587204 1 0.000 1.000 1.000 0.000
#> GSM587205 1 0.000 1.000 1.000 0.000
#> GSM587206 1 0.000 1.000 1.000 0.000
#> GSM587207 1 0.000 1.000 1.000 0.000
#> GSM587208 1 0.000 1.000 1.000 0.000
#> GSM587209 1 0.000 1.000 1.000 0.000
#> GSM587210 2 0.971 0.391 0.400 0.600
#> GSM587211 1 0.000 1.000 1.000 0.000
#> GSM587212 2 0.980 0.351 0.416 0.584
#> GSM587213 1 0.000 1.000 1.000 0.000
#> GSM587214 1 0.000 1.000 1.000 0.000
#> GSM587215 1 0.000 1.000 1.000 0.000
#> GSM587216 1 0.000 1.000 1.000 0.000
#> GSM587217 1 0.000 1.000 1.000 0.000
#> GSM587191 2 0.000 0.963 0.000 1.000
#> GSM587192 2 0.000 0.963 0.000 1.000
#> GSM587193 2 0.644 0.800 0.164 0.836
#> GSM587194 2 0.000 0.963 0.000 1.000
#> GSM587195 2 0.000 0.963 0.000 1.000
#> GSM587196 2 0.000 0.963 0.000 1.000
#> GSM587197 2 0.000 0.963 0.000 1.000
#> GSM587198 2 0.000 0.963 0.000 1.000
#> GSM587199 2 0.000 0.963 0.000 1.000
#> GSM587200 2 0.118 0.950 0.016 0.984
#> GSM587201 2 0.714 0.760 0.196 0.804
#> GSM587202 2 0.000 0.963 0.000 1.000
#> GSM198767 1 0.000 1.000 1.000 0.000
#> GSM198769 1 0.000 1.000 1.000 0.000
#> GSM198772 1 0.000 1.000 1.000 0.000
#> GSM198773 1 0.000 1.000 1.000 0.000
#> GSM198776 1 0.000 1.000 1.000 0.000
#> GSM198778 2 0.971 0.391 0.400 0.600
#> GSM198780 2 0.978 0.361 0.412 0.588
#> GSM198781 1 0.000 1.000 1.000 0.000
#> GSM198765 2 0.000 0.963 0.000 1.000
#> GSM198766 2 0.706 0.765 0.192 0.808
#> GSM198768 2 0.000 0.963 0.000 1.000
#> GSM198770 2 0.000 0.963 0.000 1.000
#> GSM198771 2 0.000 0.963 0.000 1.000
#> GSM198774 2 0.000 0.963 0.000 1.000
#> GSM198775 2 0.000 0.963 0.000 1.000
#> GSM198777 2 0.000 0.963 0.000 1.000
#> GSM198779 2 0.000 0.963 0.000 1.000
#> GSM587218 1 0.000 1.000 1.000 0.000
#> GSM587219 1 0.000 1.000 1.000 0.000
#> GSM587220 1 0.000 1.000 1.000 0.000
#> GSM587221 1 0.000 1.000 1.000 0.000
#> GSM587222 1 0.000 1.000 1.000 0.000
#> GSM587223 1 0.000 1.000 1.000 0.000
#> GSM587224 1 0.000 1.000 1.000 0.000
#> GSM587225 1 0.000 1.000 1.000 0.000
#> GSM587226 1 0.000 1.000 1.000 0.000
#> GSM587227 1 0.000 1.000 1.000 0.000
#> GSM587228 1 0.000 1.000 1.000 0.000
#> GSM587229 1 0.000 1.000 1.000 0.000
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM587155 2 0.0000 0.968 0.000 1.000 0.000
#> GSM587156 2 0.0000 0.968 0.000 1.000 0.000
#> GSM587157 2 0.0000 0.968 0.000 1.000 0.000
#> GSM587158 2 0.0000 0.968 0.000 1.000 0.000
#> GSM587159 2 0.0000 0.968 0.000 1.000 0.000
#> GSM587160 2 0.0000 0.968 0.000 1.000 0.000
#> GSM587161 2 0.0000 0.968 0.000 1.000 0.000
#> GSM587162 2 0.0000 0.968 0.000 1.000 0.000
#> GSM587163 2 0.0000 0.968 0.000 1.000 0.000
#> GSM587164 2 0.0000 0.968 0.000 1.000 0.000
#> GSM587165 2 0.0000 0.968 0.000 1.000 0.000
#> GSM587166 2 0.1163 0.936 0.000 0.972 0.028
#> GSM587167 2 0.0000 0.968 0.000 1.000 0.000
#> GSM587168 2 0.0000 0.968 0.000 1.000 0.000
#> GSM587169 2 0.0000 0.968 0.000 1.000 0.000
#> GSM587170 2 0.0000 0.968 0.000 1.000 0.000
#> GSM587171 2 0.0000 0.968 0.000 1.000 0.000
#> GSM587172 2 0.0000 0.968 0.000 1.000 0.000
#> GSM587173 2 0.0000 0.968 0.000 1.000 0.000
#> GSM587174 2 0.0000 0.968 0.000 1.000 0.000
#> GSM587175 2 0.0000 0.968 0.000 1.000 0.000
#> GSM587176 2 0.0000 0.968 0.000 1.000 0.000
#> GSM587177 2 0.0000 0.968 0.000 1.000 0.000
#> GSM587178 2 0.0000 0.968 0.000 1.000 0.000
#> GSM587179 2 0.0000 0.968 0.000 1.000 0.000
#> GSM587180 2 0.0000 0.968 0.000 1.000 0.000
#> GSM587181 2 0.0000 0.968 0.000 1.000 0.000
#> GSM587182 2 0.0000 0.968 0.000 1.000 0.000
#> GSM587183 2 0.0000 0.968 0.000 1.000 0.000
#> GSM587184 2 0.0000 0.968 0.000 1.000 0.000
#> GSM587185 2 0.0000 0.968 0.000 1.000 0.000
#> GSM587186 2 0.0000 0.968 0.000 1.000 0.000
#> GSM587187 2 0.0237 0.964 0.000 0.996 0.004
#> GSM587188 2 0.6215 0.137 0.000 0.572 0.428
#> GSM587189 2 0.6062 0.284 0.000 0.616 0.384
#> GSM587190 3 0.6008 0.498 0.000 0.372 0.628
#> GSM587203 3 0.0237 0.925 0.004 0.000 0.996
#> GSM587204 3 0.0237 0.925 0.004 0.000 0.996
#> GSM587205 3 0.0237 0.925 0.004 0.000 0.996
#> GSM587206 3 0.0237 0.925 0.004 0.000 0.996
#> GSM587207 3 0.0237 0.925 0.004 0.000 0.996
#> GSM587208 3 0.0237 0.925 0.004 0.000 0.996
#> GSM587209 3 0.0237 0.925 0.004 0.000 0.996
#> GSM587210 3 0.0000 0.924 0.000 0.000 1.000
#> GSM587211 3 0.0237 0.925 0.004 0.000 0.996
#> GSM587212 3 0.0000 0.924 0.000 0.000 1.000
#> GSM587213 3 0.0237 0.925 0.004 0.000 0.996
#> GSM587214 3 0.0237 0.925 0.004 0.000 0.996
#> GSM587215 3 0.0237 0.925 0.004 0.000 0.996
#> GSM587216 3 0.0000 0.924 0.000 0.000 1.000
#> GSM587217 3 0.0237 0.925 0.004 0.000 0.996
#> GSM587191 3 0.3116 0.871 0.000 0.108 0.892
#> GSM587192 3 0.0237 0.924 0.000 0.004 0.996
#> GSM587193 3 0.1289 0.911 0.000 0.032 0.968
#> GSM587194 3 0.5678 0.615 0.000 0.316 0.684
#> GSM587195 3 0.4452 0.792 0.000 0.192 0.808
#> GSM587196 3 0.2959 0.876 0.000 0.100 0.900
#> GSM587197 3 0.5621 0.629 0.000 0.308 0.692
#> GSM587198 3 0.2959 0.876 0.000 0.100 0.900
#> GSM587199 3 0.0237 0.924 0.000 0.004 0.996
#> GSM587200 3 0.0237 0.924 0.000 0.004 0.996
#> GSM587201 3 0.0237 0.924 0.000 0.004 0.996
#> GSM587202 3 0.2959 0.876 0.000 0.100 0.900
#> GSM198767 3 0.0237 0.925 0.004 0.000 0.996
#> GSM198769 3 0.0237 0.925 0.004 0.000 0.996
#> GSM198772 3 0.0237 0.925 0.004 0.000 0.996
#> GSM198773 3 0.0237 0.925 0.004 0.000 0.996
#> GSM198776 3 0.0237 0.925 0.004 0.000 0.996
#> GSM198778 3 0.0000 0.924 0.000 0.000 1.000
#> GSM198780 3 0.0000 0.924 0.000 0.000 1.000
#> GSM198781 3 0.0237 0.925 0.004 0.000 0.996
#> GSM198765 3 0.3038 0.874 0.000 0.104 0.896
#> GSM198766 3 0.1289 0.911 0.000 0.032 0.968
#> GSM198768 3 0.3551 0.850 0.000 0.132 0.868
#> GSM198770 3 0.5678 0.615 0.000 0.316 0.684
#> GSM198771 3 0.2959 0.876 0.000 0.100 0.900
#> GSM198774 3 0.0424 0.923 0.000 0.008 0.992
#> GSM198775 3 0.5678 0.615 0.000 0.316 0.684
#> GSM198777 3 0.2959 0.876 0.000 0.100 0.900
#> GSM198779 3 0.0237 0.924 0.000 0.004 0.996
#> GSM587218 1 0.0000 1.000 1.000 0.000 0.000
#> GSM587219 1 0.0000 1.000 1.000 0.000 0.000
#> GSM587220 1 0.0000 1.000 1.000 0.000 0.000
#> GSM587221 1 0.0000 1.000 1.000 0.000 0.000
#> GSM587222 1 0.0000 1.000 1.000 0.000 0.000
#> GSM587223 1 0.0000 1.000 1.000 0.000 0.000
#> GSM587224 1 0.0000 1.000 1.000 0.000 0.000
#> GSM587225 1 0.0000 1.000 1.000 0.000 0.000
#> GSM587226 1 0.0000 1.000 1.000 0.000 0.000
#> GSM587227 1 0.0000 1.000 1.000 0.000 0.000
#> GSM587228 1 0.0000 1.000 1.000 0.000 0.000
#> GSM587229 1 0.0000 1.000 1.000 0.000 0.000
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM587155 2 0.0000 0.981 0.000 1.000 0.000 0
#> GSM587156 2 0.0188 0.977 0.000 0.996 0.004 0
#> GSM587157 2 0.0188 0.977 0.000 0.996 0.004 0
#> GSM587158 2 0.0000 0.981 0.000 1.000 0.000 0
#> GSM587159 2 0.0000 0.981 0.000 1.000 0.000 0
#> GSM587160 2 0.0000 0.981 0.000 1.000 0.000 0
#> GSM587161 2 0.0000 0.981 0.000 1.000 0.000 0
#> GSM587162 2 0.0000 0.981 0.000 1.000 0.000 0
#> GSM587163 2 0.0000 0.981 0.000 1.000 0.000 0
#> GSM587164 2 0.0000 0.981 0.000 1.000 0.000 0
#> GSM587165 2 0.0000 0.981 0.000 1.000 0.000 0
#> GSM587166 2 0.3764 0.717 0.000 0.784 0.216 0
#> GSM587167 2 0.0188 0.977 0.000 0.996 0.004 0
#> GSM587168 2 0.0000 0.981 0.000 1.000 0.000 0
#> GSM587169 2 0.0000 0.981 0.000 1.000 0.000 0
#> GSM587170 2 0.0000 0.981 0.000 1.000 0.000 0
#> GSM587171 2 0.0000 0.981 0.000 1.000 0.000 0
#> GSM587172 2 0.0000 0.981 0.000 1.000 0.000 0
#> GSM587173 2 0.0000 0.981 0.000 1.000 0.000 0
#> GSM587174 2 0.0000 0.981 0.000 1.000 0.000 0
#> GSM587175 2 0.0188 0.977 0.000 0.996 0.004 0
#> GSM587176 2 0.0000 0.981 0.000 1.000 0.000 0
#> GSM587177 2 0.0000 0.981 0.000 1.000 0.000 0
#> GSM587178 2 0.0000 0.981 0.000 1.000 0.000 0
#> GSM587179 2 0.0000 0.981 0.000 1.000 0.000 0
#> GSM587180 2 0.0000 0.981 0.000 1.000 0.000 0
#> GSM587181 2 0.0000 0.981 0.000 1.000 0.000 0
#> GSM587182 2 0.0000 0.981 0.000 1.000 0.000 0
#> GSM587183 2 0.0000 0.981 0.000 1.000 0.000 0
#> GSM587184 2 0.0000 0.981 0.000 1.000 0.000 0
#> GSM587185 2 0.0000 0.981 0.000 1.000 0.000 0
#> GSM587186 2 0.0000 0.981 0.000 1.000 0.000 0
#> GSM587187 2 0.4543 0.488 0.000 0.676 0.324 0
#> GSM587188 3 0.4522 0.543 0.000 0.320 0.680 0
#> GSM587189 3 0.4730 0.447 0.000 0.364 0.636 0
#> GSM587190 3 0.0469 0.931 0.000 0.012 0.988 0
#> GSM587203 1 0.0000 0.970 1.000 0.000 0.000 0
#> GSM587204 1 0.0000 0.970 1.000 0.000 0.000 0
#> GSM587205 1 0.0000 0.970 1.000 0.000 0.000 0
#> GSM587206 1 0.0000 0.970 1.000 0.000 0.000 0
#> GSM587207 1 0.0000 0.970 1.000 0.000 0.000 0
#> GSM587208 1 0.0000 0.970 1.000 0.000 0.000 0
#> GSM587209 1 0.0000 0.970 1.000 0.000 0.000 0
#> GSM587210 1 0.2760 0.866 0.872 0.000 0.128 0
#> GSM587211 1 0.0000 0.970 1.000 0.000 0.000 0
#> GSM587212 1 0.2530 0.881 0.888 0.000 0.112 0
#> GSM587213 1 0.0000 0.970 1.000 0.000 0.000 0
#> GSM587214 1 0.0000 0.970 1.000 0.000 0.000 0
#> GSM587215 1 0.1716 0.919 0.936 0.000 0.064 0
#> GSM587216 1 0.0592 0.960 0.984 0.000 0.016 0
#> GSM587217 1 0.0000 0.970 1.000 0.000 0.000 0
#> GSM587191 3 0.0000 0.942 0.000 0.000 1.000 0
#> GSM587192 3 0.0000 0.942 0.000 0.000 1.000 0
#> GSM587193 3 0.0000 0.942 0.000 0.000 1.000 0
#> GSM587194 3 0.0000 0.942 0.000 0.000 1.000 0
#> GSM587195 3 0.0000 0.942 0.000 0.000 1.000 0
#> GSM587196 3 0.0000 0.942 0.000 0.000 1.000 0
#> GSM587197 3 0.0000 0.942 0.000 0.000 1.000 0
#> GSM587198 3 0.0000 0.942 0.000 0.000 1.000 0
#> GSM587199 3 0.0000 0.942 0.000 0.000 1.000 0
#> GSM587200 3 0.3610 0.735 0.200 0.000 0.800 0
#> GSM587201 3 0.3649 0.729 0.204 0.000 0.796 0
#> GSM587202 3 0.0000 0.942 0.000 0.000 1.000 0
#> GSM198767 1 0.0000 0.970 1.000 0.000 0.000 0
#> GSM198769 1 0.0000 0.970 1.000 0.000 0.000 0
#> GSM198772 1 0.0000 0.970 1.000 0.000 0.000 0
#> GSM198773 1 0.0000 0.970 1.000 0.000 0.000 0
#> GSM198776 1 0.0000 0.970 1.000 0.000 0.000 0
#> GSM198778 1 0.2760 0.866 0.872 0.000 0.128 0
#> GSM198780 1 0.2530 0.881 0.888 0.000 0.112 0
#> GSM198781 1 0.0000 0.970 1.000 0.000 0.000 0
#> GSM198765 3 0.0000 0.942 0.000 0.000 1.000 0
#> GSM198766 3 0.0000 0.942 0.000 0.000 1.000 0
#> GSM198768 3 0.0000 0.942 0.000 0.000 1.000 0
#> GSM198770 3 0.0000 0.942 0.000 0.000 1.000 0
#> GSM198771 3 0.0000 0.942 0.000 0.000 1.000 0
#> GSM198774 3 0.0000 0.942 0.000 0.000 1.000 0
#> GSM198775 3 0.0000 0.942 0.000 0.000 1.000 0
#> GSM198777 3 0.0000 0.942 0.000 0.000 1.000 0
#> GSM198779 3 0.0000 0.942 0.000 0.000 1.000 0
#> GSM587218 4 0.0000 1.000 0.000 0.000 0.000 1
#> GSM587219 4 0.0000 1.000 0.000 0.000 0.000 1
#> GSM587220 4 0.0000 1.000 0.000 0.000 0.000 1
#> GSM587221 4 0.0000 1.000 0.000 0.000 0.000 1
#> GSM587222 4 0.0000 1.000 0.000 0.000 0.000 1
#> GSM587223 4 0.0000 1.000 0.000 0.000 0.000 1
#> GSM587224 4 0.0000 1.000 0.000 0.000 0.000 1
#> GSM587225 4 0.0000 1.000 0.000 0.000 0.000 1
#> GSM587226 4 0.0000 1.000 0.000 0.000 0.000 1
#> GSM587227 4 0.0000 1.000 0.000 0.000 0.000 1
#> GSM587228 4 0.0000 1.000 0.000 0.000 0.000 1
#> GSM587229 4 0.0000 1.000 0.000 0.000 0.000 1
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM587155 2 0.0703 0.9535 0.000 0.976 0.000 0 0.024
#> GSM587156 2 0.1851 0.9113 0.000 0.912 0.000 0 0.088
#> GSM587157 2 0.1478 0.9284 0.000 0.936 0.000 0 0.064
#> GSM587158 2 0.0000 0.9595 0.000 1.000 0.000 0 0.000
#> GSM587159 2 0.0000 0.9595 0.000 1.000 0.000 0 0.000
#> GSM587160 2 0.0000 0.9595 0.000 1.000 0.000 0 0.000
#> GSM587161 2 0.0000 0.9595 0.000 1.000 0.000 0 0.000
#> GSM587162 2 0.0510 0.9566 0.000 0.984 0.000 0 0.016
#> GSM587163 2 0.0000 0.9595 0.000 1.000 0.000 0 0.000
#> GSM587164 2 0.1608 0.9228 0.000 0.928 0.000 0 0.072
#> GSM587165 2 0.0000 0.9595 0.000 1.000 0.000 0 0.000
#> GSM587166 2 0.1851 0.9113 0.000 0.912 0.000 0 0.088
#> GSM587167 2 0.1732 0.9173 0.000 0.920 0.000 0 0.080
#> GSM587168 2 0.0404 0.9577 0.000 0.988 0.000 0 0.012
#> GSM587169 2 0.0000 0.9595 0.000 1.000 0.000 0 0.000
#> GSM587170 2 0.0703 0.9535 0.000 0.976 0.000 0 0.024
#> GSM587171 2 0.0000 0.9595 0.000 1.000 0.000 0 0.000
#> GSM587172 2 0.0000 0.9595 0.000 1.000 0.000 0 0.000
#> GSM587173 2 0.0000 0.9595 0.000 1.000 0.000 0 0.000
#> GSM587174 2 0.0404 0.9577 0.000 0.988 0.000 0 0.012
#> GSM587175 2 0.0880 0.9496 0.000 0.968 0.000 0 0.032
#> GSM587176 2 0.0404 0.9577 0.000 0.988 0.000 0 0.012
#> GSM587177 2 0.0000 0.9595 0.000 1.000 0.000 0 0.000
#> GSM587178 2 0.0000 0.9595 0.000 1.000 0.000 0 0.000
#> GSM587179 2 0.0000 0.9595 0.000 1.000 0.000 0 0.000
#> GSM587180 2 0.0510 0.9566 0.000 0.984 0.000 0 0.016
#> GSM587181 2 0.0404 0.9577 0.000 0.988 0.000 0 0.012
#> GSM587182 2 0.0404 0.9577 0.000 0.988 0.000 0 0.012
#> GSM587183 2 0.0000 0.9595 0.000 1.000 0.000 0 0.000
#> GSM587184 2 0.0000 0.9595 0.000 1.000 0.000 0 0.000
#> GSM587185 2 0.0000 0.9595 0.000 1.000 0.000 0 0.000
#> GSM587186 2 0.0000 0.9595 0.000 1.000 0.000 0 0.000
#> GSM587187 2 0.3921 0.7516 0.000 0.784 0.044 0 0.172
#> GSM587188 5 0.5725 0.4505 0.000 0.204 0.172 0 0.624
#> GSM587189 2 0.6625 -0.2726 0.000 0.412 0.220 0 0.368
#> GSM587190 5 0.4114 0.6244 0.000 0.000 0.376 0 0.624
#> GSM587203 1 0.3274 0.8101 0.780 0.000 0.000 0 0.220
#> GSM587204 1 0.3210 0.8113 0.788 0.000 0.000 0 0.212
#> GSM587205 1 0.3305 0.8084 0.776 0.000 0.000 0 0.224
#> GSM587206 1 0.3305 0.8084 0.776 0.000 0.000 0 0.224
#> GSM587207 1 0.3305 0.8084 0.776 0.000 0.000 0 0.224
#> GSM587208 1 0.3305 0.8084 0.776 0.000 0.000 0 0.224
#> GSM587209 1 0.1282 0.8438 0.952 0.000 0.004 0 0.044
#> GSM587210 1 0.5025 0.6506 0.704 0.000 0.124 0 0.172
#> GSM587211 1 0.1965 0.8373 0.924 0.000 0.024 0 0.052
#> GSM587212 1 0.4926 0.6643 0.712 0.000 0.112 0 0.176
#> GSM587213 1 0.0162 0.8502 0.996 0.000 0.000 0 0.004
#> GSM587214 1 0.0794 0.8503 0.972 0.000 0.000 0 0.028
#> GSM587215 1 0.1800 0.8405 0.932 0.000 0.020 0 0.048
#> GSM587216 1 0.1981 0.8372 0.924 0.000 0.028 0 0.048
#> GSM587217 1 0.1300 0.8512 0.956 0.000 0.016 0 0.028
#> GSM587191 3 0.0703 0.7538 0.000 0.000 0.976 0 0.024
#> GSM587192 3 0.0865 0.7478 0.004 0.000 0.972 0 0.024
#> GSM587193 5 0.4367 0.6353 0.004 0.000 0.416 0 0.580
#> GSM587194 3 0.4291 -0.4172 0.000 0.000 0.536 0 0.464
#> GSM587195 3 0.1478 0.7243 0.000 0.000 0.936 0 0.064
#> GSM587196 3 0.0703 0.7538 0.000 0.000 0.976 0 0.024
#> GSM587197 3 0.4101 -0.0543 0.000 0.000 0.628 0 0.372
#> GSM587198 3 0.0510 0.7506 0.000 0.000 0.984 0 0.016
#> GSM587199 3 0.1124 0.7409 0.004 0.000 0.960 0 0.036
#> GSM587200 3 0.4250 0.3830 0.252 0.000 0.720 0 0.028
#> GSM587201 3 0.4465 0.4077 0.204 0.000 0.736 0 0.060
#> GSM587202 3 0.1043 0.7524 0.000 0.000 0.960 0 0.040
#> GSM198767 1 0.3305 0.8084 0.776 0.000 0.000 0 0.224
#> GSM198769 1 0.1701 0.8407 0.936 0.000 0.016 0 0.048
#> GSM198772 1 0.1965 0.8373 0.924 0.000 0.024 0 0.052
#> GSM198773 1 0.0404 0.8508 0.988 0.000 0.000 0 0.012
#> GSM198776 1 0.3210 0.8113 0.788 0.000 0.000 0 0.212
#> GSM198778 1 0.5025 0.6506 0.704 0.000 0.124 0 0.172
#> GSM198780 1 0.4926 0.6643 0.712 0.000 0.112 0 0.176
#> GSM198781 1 0.0794 0.8503 0.972 0.000 0.000 0 0.028
#> GSM198765 3 0.0703 0.7538 0.000 0.000 0.976 0 0.024
#> GSM198766 5 0.4367 0.6353 0.004 0.000 0.416 0 0.580
#> GSM198768 3 0.0963 0.7499 0.000 0.000 0.964 0 0.036
#> GSM198770 3 0.4101 -0.0543 0.000 0.000 0.628 0 0.372
#> GSM198771 3 0.0609 0.7488 0.000 0.000 0.980 0 0.020
#> GSM198774 3 0.0404 0.7523 0.000 0.000 0.988 0 0.012
#> GSM198775 3 0.4291 -0.4172 0.000 0.000 0.536 0 0.464
#> GSM198777 3 0.0703 0.7538 0.000 0.000 0.976 0 0.024
#> GSM198779 3 0.1124 0.7409 0.004 0.000 0.960 0 0.036
#> GSM587218 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000
#> GSM587219 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000
#> GSM587220 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000
#> GSM587221 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000
#> GSM587222 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000
#> GSM587223 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000
#> GSM587224 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000
#> GSM587225 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000
#> GSM587226 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000
#> GSM587227 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000
#> GSM587228 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000
#> GSM587229 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM587155 2 0.2969 0.78425 0.000 0.776 0.000 0 0.224 0.000
#> GSM587156 2 0.3351 0.72171 0.000 0.712 0.000 0 0.288 0.000
#> GSM587157 2 0.3101 0.76719 0.000 0.756 0.000 0 0.244 0.000
#> GSM587158 2 0.0000 0.90292 0.000 1.000 0.000 0 0.000 0.000
#> GSM587159 2 0.0000 0.90292 0.000 1.000 0.000 0 0.000 0.000
#> GSM587160 2 0.0000 0.90292 0.000 1.000 0.000 0 0.000 0.000
#> GSM587161 2 0.0146 0.90254 0.000 0.996 0.000 0 0.004 0.000
#> GSM587162 2 0.2854 0.79673 0.000 0.792 0.000 0 0.208 0.000
#> GSM587163 2 0.0000 0.90292 0.000 1.000 0.000 0 0.000 0.000
#> GSM587164 2 0.3198 0.75247 0.000 0.740 0.000 0 0.260 0.000
#> GSM587165 2 0.0000 0.90292 0.000 1.000 0.000 0 0.000 0.000
#> GSM587166 2 0.3508 0.71310 0.000 0.704 0.000 0 0.292 0.004
#> GSM587167 2 0.3351 0.72171 0.000 0.712 0.000 0 0.288 0.000
#> GSM587168 2 0.0260 0.90175 0.000 0.992 0.000 0 0.008 0.000
#> GSM587169 2 0.0260 0.90130 0.000 0.992 0.000 0 0.008 0.000
#> GSM587170 2 0.2854 0.79591 0.000 0.792 0.000 0 0.208 0.000
#> GSM587171 2 0.0000 0.90292 0.000 1.000 0.000 0 0.000 0.000
#> GSM587172 2 0.0000 0.90292 0.000 1.000 0.000 0 0.000 0.000
#> GSM587173 2 0.0000 0.90292 0.000 1.000 0.000 0 0.000 0.000
#> GSM587174 2 0.0000 0.90292 0.000 1.000 0.000 0 0.000 0.000
#> GSM587175 2 0.3023 0.77751 0.000 0.768 0.000 0 0.232 0.000
#> GSM587176 2 0.0146 0.90266 0.000 0.996 0.000 0 0.004 0.000
#> GSM587177 2 0.0000 0.90292 0.000 1.000 0.000 0 0.000 0.000
#> GSM587178 2 0.0000 0.90292 0.000 1.000 0.000 0 0.000 0.000
#> GSM587179 2 0.0000 0.90292 0.000 1.000 0.000 0 0.000 0.000
#> GSM587180 2 0.2454 0.82539 0.000 0.840 0.000 0 0.160 0.000
#> GSM587181 2 0.0146 0.90266 0.000 0.996 0.000 0 0.004 0.000
#> GSM587182 2 0.0458 0.89903 0.000 0.984 0.000 0 0.016 0.000
#> GSM587183 2 0.0000 0.90292 0.000 1.000 0.000 0 0.000 0.000
#> GSM587184 2 0.0146 0.90126 0.000 0.996 0.000 0 0.000 0.004
#> GSM587185 2 0.0291 0.90135 0.000 0.992 0.000 0 0.004 0.004
#> GSM587186 2 0.0146 0.90262 0.000 0.996 0.000 0 0.004 0.000
#> GSM587187 2 0.3833 0.44016 0.000 0.556 0.000 0 0.444 0.000
#> GSM587188 5 0.1219 0.63464 0.000 0.000 0.048 0 0.948 0.004
#> GSM587189 5 0.3516 0.56602 0.000 0.164 0.048 0 0.788 0.000
#> GSM587190 5 0.3405 0.27946 0.000 0.000 0.272 0 0.724 0.004
#> GSM587203 1 0.0363 0.80466 0.988 0.000 0.000 0 0.000 0.012
#> GSM587204 1 0.1501 0.80508 0.924 0.000 0.000 0 0.000 0.076
#> GSM587205 1 0.0260 0.80489 0.992 0.000 0.000 0 0.000 0.008
#> GSM587206 1 0.0790 0.80132 0.968 0.000 0.000 0 0.000 0.032
#> GSM587207 1 0.0260 0.80626 0.992 0.000 0.000 0 0.000 0.008
#> GSM587208 1 0.0790 0.80132 0.968 0.000 0.000 0 0.000 0.032
#> GSM587209 6 0.3351 0.70230 0.288 0.000 0.000 0 0.000 0.712
#> GSM587210 6 0.3897 0.78048 0.076 0.000 0.060 0 0.056 0.808
#> GSM587211 6 0.3078 0.79278 0.192 0.000 0.000 0 0.012 0.796
#> GSM587212 6 0.3717 0.78489 0.076 0.000 0.052 0 0.052 0.820
#> GSM587213 1 0.3464 0.59621 0.688 0.000 0.000 0 0.000 0.312
#> GSM587214 1 0.3409 0.63034 0.700 0.000 0.000 0 0.000 0.300
#> GSM587215 6 0.3410 0.72140 0.216 0.000 0.008 0 0.008 0.768
#> GSM587216 6 0.3600 0.79976 0.192 0.000 0.020 0 0.012 0.776
#> GSM587217 1 0.3547 0.60344 0.668 0.000 0.000 0 0.000 0.332
#> GSM587191 3 0.0713 0.75949 0.000 0.000 0.972 0 0.028 0.000
#> GSM587192 3 0.1780 0.74866 0.000 0.000 0.924 0 0.048 0.028
#> GSM587193 3 0.6089 0.00337 0.000 0.000 0.392 0 0.308 0.300
#> GSM587194 3 0.4338 0.16729 0.000 0.000 0.496 0 0.484 0.020
#> GSM587195 3 0.1285 0.75297 0.000 0.000 0.944 0 0.052 0.004
#> GSM587196 3 0.0508 0.75994 0.000 0.000 0.984 0 0.012 0.004
#> GSM587197 3 0.4361 0.29968 0.000 0.000 0.552 0 0.424 0.024
#> GSM587198 3 0.0260 0.75813 0.000 0.000 0.992 0 0.000 0.008
#> GSM587199 3 0.1092 0.75086 0.000 0.000 0.960 0 0.020 0.020
#> GSM587200 3 0.5198 0.52680 0.060 0.000 0.676 0 0.064 0.200
#> GSM587201 3 0.4176 0.61716 0.024 0.000 0.760 0 0.052 0.164
#> GSM587202 3 0.0405 0.75928 0.000 0.000 0.988 0 0.004 0.008
#> GSM198767 1 0.0146 0.80533 0.996 0.000 0.000 0 0.000 0.004
#> GSM198769 6 0.3499 0.65292 0.320 0.000 0.000 0 0.000 0.680
#> GSM198772 6 0.2980 0.78934 0.192 0.000 0.000 0 0.008 0.800
#> GSM198773 1 0.3499 0.60729 0.680 0.000 0.000 0 0.000 0.320
#> GSM198776 1 0.1501 0.80508 0.924 0.000 0.000 0 0.000 0.076
#> GSM198778 6 0.3897 0.78048 0.076 0.000 0.060 0 0.056 0.808
#> GSM198780 6 0.3717 0.78489 0.076 0.000 0.052 0 0.052 0.820
#> GSM198781 1 0.3351 0.63212 0.712 0.000 0.000 0 0.000 0.288
#> GSM198765 3 0.0547 0.76034 0.000 0.000 0.980 0 0.020 0.000
#> GSM198766 3 0.6089 0.00337 0.000 0.000 0.392 0 0.308 0.300
#> GSM198768 3 0.1082 0.75426 0.000 0.000 0.956 0 0.040 0.004
#> GSM198770 3 0.4372 0.28305 0.000 0.000 0.544 0 0.432 0.024
#> GSM198771 3 0.0260 0.75813 0.000 0.000 0.992 0 0.000 0.008
#> GSM198774 3 0.1908 0.74918 0.000 0.000 0.916 0 0.056 0.028
#> GSM198775 3 0.4338 0.16729 0.000 0.000 0.496 0 0.484 0.020
#> GSM198777 3 0.0458 0.76004 0.000 0.000 0.984 0 0.016 0.000
#> GSM198779 3 0.1092 0.75086 0.000 0.000 0.960 0 0.020 0.020
#> GSM587218 4 0.0000 1.00000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587219 4 0.0000 1.00000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587220 4 0.0000 1.00000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587221 4 0.0000 1.00000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587222 4 0.0000 1.00000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587223 4 0.0000 1.00000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587224 4 0.0000 1.00000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587225 4 0.0000 1.00000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587226 4 0.0000 1.00000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587227 4 0.0000 1.00000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587228 4 0.0000 1.00000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587229 4 0.0000 1.00000 0.000 0.000 0.000 1 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 specimen(p) k
#> SD:mclust 88 7.88e-17 2
#> SD:mclust 89 1.10e-31 3
#> SD:mclust 90 6.82e-47 4
#> SD:mclust 84 1.06e-41 5
#> SD:mclust 84 1.37e-53 6
If matrix rows can be associated to genes, consider to use functional_enrichment(res,
...) to perform function enrichment for the signature genes. See this vignette for more detailed explanations.
The object with results only for a single top-value method and a single partition method can be extracted as:
res = res_list["SD", "NMF"]
# you can also extract it by
# res = res_list["SD:NMF"]
A summary of res and all the functions that can be applied to it:
res
#> A 'ConsensusPartition' object with k = 2, 3, 4, 5, 6.
#> On a matrix with 51941 rows and 92 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 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 1.000 0.998 0.999 0.5001 0.500 0.500
#> 3 3 0.910 0.904 0.958 0.2505 0.843 0.695
#> 4 4 0.999 0.968 0.985 0.1564 0.807 0.544
#> 5 5 0.920 0.881 0.944 0.0616 0.941 0.792
#> 6 6 0.850 0.736 0.853 0.0432 0.957 0.819
suggest_best_k() suggests the best \(k\) based on these statistics. The rules are as follows:
suggest_best_k(res)
#> [1] 5
#> attr(,"optional")
#> [1] 2 3 4
There is also optional best \(k\) = 2 3 4 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
#> GSM587155 2 0.0000 0.999 0.000 1.000
#> GSM587156 2 0.0000 0.999 0.000 1.000
#> GSM587157 2 0.0000 0.999 0.000 1.000
#> GSM587158 2 0.0000 0.999 0.000 1.000
#> GSM587159 2 0.0000 0.999 0.000 1.000
#> GSM587160 2 0.0000 0.999 0.000 1.000
#> GSM587161 2 0.0000 0.999 0.000 1.000
#> GSM587162 2 0.0000 0.999 0.000 1.000
#> GSM587163 2 0.0000 0.999 0.000 1.000
#> GSM587164 2 0.0000 0.999 0.000 1.000
#> GSM587165 2 0.0000 0.999 0.000 1.000
#> GSM587166 2 0.0000 0.999 0.000 1.000
#> GSM587167 2 0.0000 0.999 0.000 1.000
#> GSM587168 2 0.0000 0.999 0.000 1.000
#> GSM587169 2 0.0000 0.999 0.000 1.000
#> GSM587170 2 0.0000 0.999 0.000 1.000
#> GSM587171 2 0.0000 0.999 0.000 1.000
#> GSM587172 2 0.0000 0.999 0.000 1.000
#> GSM587173 2 0.0000 0.999 0.000 1.000
#> GSM587174 2 0.0000 0.999 0.000 1.000
#> GSM587175 2 0.0000 0.999 0.000 1.000
#> GSM587176 2 0.0000 0.999 0.000 1.000
#> GSM587177 2 0.0000 0.999 0.000 1.000
#> GSM587178 2 0.0000 0.999 0.000 1.000
#> GSM587179 2 0.0000 0.999 0.000 1.000
#> GSM587180 2 0.0000 0.999 0.000 1.000
#> GSM587181 2 0.0000 0.999 0.000 1.000
#> GSM587182 2 0.0000 0.999 0.000 1.000
#> GSM587183 2 0.0000 0.999 0.000 1.000
#> GSM587184 2 0.0000 0.999 0.000 1.000
#> GSM587185 2 0.0000 0.999 0.000 1.000
#> GSM587186 2 0.0000 0.999 0.000 1.000
#> GSM587187 2 0.0000 0.999 0.000 1.000
#> GSM587188 2 0.0000 0.999 0.000 1.000
#> GSM587189 2 0.0000 0.999 0.000 1.000
#> GSM587190 2 0.0000 0.999 0.000 1.000
#> GSM587203 1 0.0000 1.000 1.000 0.000
#> GSM587204 1 0.0000 1.000 1.000 0.000
#> GSM587205 1 0.0000 1.000 1.000 0.000
#> GSM587206 1 0.0000 1.000 1.000 0.000
#> GSM587207 1 0.0000 1.000 1.000 0.000
#> GSM587208 1 0.0000 1.000 1.000 0.000
#> GSM587209 1 0.0000 1.000 1.000 0.000
#> GSM587210 1 0.0000 1.000 1.000 0.000
#> GSM587211 1 0.0000 1.000 1.000 0.000
#> GSM587212 1 0.0000 1.000 1.000 0.000
#> GSM587213 1 0.0000 1.000 1.000 0.000
#> GSM587214 1 0.0000 1.000 1.000 0.000
#> GSM587215 1 0.0000 1.000 1.000 0.000
#> GSM587216 1 0.0000 1.000 1.000 0.000
#> GSM587217 1 0.0000 1.000 1.000 0.000
#> GSM587191 2 0.0000 0.999 0.000 1.000
#> GSM587192 1 0.0376 0.996 0.996 0.004
#> GSM587193 1 0.0000 1.000 1.000 0.000
#> GSM587194 2 0.0000 0.999 0.000 1.000
#> GSM587195 2 0.0000 0.999 0.000 1.000
#> GSM587196 2 0.1843 0.972 0.028 0.972
#> GSM587197 2 0.0000 0.999 0.000 1.000
#> GSM587198 2 0.0000 0.999 0.000 1.000
#> GSM587199 2 0.0000 0.999 0.000 1.000
#> GSM587200 1 0.0000 1.000 1.000 0.000
#> GSM587201 1 0.0000 1.000 1.000 0.000
#> GSM587202 2 0.0000 0.999 0.000 1.000
#> GSM198767 1 0.0000 1.000 1.000 0.000
#> GSM198769 1 0.0000 1.000 1.000 0.000
#> GSM198772 1 0.0000 1.000 1.000 0.000
#> GSM198773 1 0.0000 1.000 1.000 0.000
#> GSM198776 1 0.0000 1.000 1.000 0.000
#> GSM198778 1 0.0000 1.000 1.000 0.000
#> GSM198780 1 0.0000 1.000 1.000 0.000
#> GSM198781 1 0.0000 1.000 1.000 0.000
#> GSM198765 2 0.0376 0.995 0.004 0.996
#> GSM198766 1 0.0000 1.000 1.000 0.000
#> GSM198768 2 0.0000 0.999 0.000 1.000
#> GSM198770 2 0.0000 0.999 0.000 1.000
#> GSM198771 2 0.1633 0.976 0.024 0.976
#> GSM198774 1 0.1184 0.984 0.984 0.016
#> GSM198775 2 0.0000 0.999 0.000 1.000
#> GSM198777 2 0.0672 0.992 0.008 0.992
#> GSM198779 2 0.0000 0.999 0.000 1.000
#> GSM587218 1 0.0000 1.000 1.000 0.000
#> GSM587219 1 0.0000 1.000 1.000 0.000
#> GSM587220 1 0.0000 1.000 1.000 0.000
#> GSM587221 1 0.0000 1.000 1.000 0.000
#> GSM587222 1 0.0000 1.000 1.000 0.000
#> GSM587223 1 0.0000 1.000 1.000 0.000
#> GSM587224 1 0.0000 1.000 1.000 0.000
#> GSM587225 1 0.0000 1.000 1.000 0.000
#> GSM587226 1 0.0000 1.000 1.000 0.000
#> GSM587227 1 0.0000 1.000 1.000 0.000
#> GSM587228 1 0.0000 1.000 1.000 0.000
#> GSM587229 1 0.0000 1.000 1.000 0.000
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM587155 2 0.0000 0.974 0.000 1.000 0.000
#> GSM587156 2 0.0000 0.974 0.000 1.000 0.000
#> GSM587157 2 0.0000 0.974 0.000 1.000 0.000
#> GSM587158 2 0.0000 0.974 0.000 1.000 0.000
#> GSM587159 2 0.0000 0.974 0.000 1.000 0.000
#> GSM587160 2 0.0000 0.974 0.000 1.000 0.000
#> GSM587161 2 0.0000 0.974 0.000 1.000 0.000
#> GSM587162 2 0.0000 0.974 0.000 1.000 0.000
#> GSM587163 2 0.0000 0.974 0.000 1.000 0.000
#> GSM587164 2 0.0000 0.974 0.000 1.000 0.000
#> GSM587165 2 0.0000 0.974 0.000 1.000 0.000
#> GSM587166 2 0.0237 0.971 0.000 0.996 0.004
#> GSM587167 2 0.0000 0.974 0.000 1.000 0.000
#> GSM587168 2 0.0000 0.974 0.000 1.000 0.000
#> GSM587169 2 0.0000 0.974 0.000 1.000 0.000
#> GSM587170 2 0.0000 0.974 0.000 1.000 0.000
#> GSM587171 2 0.0000 0.974 0.000 1.000 0.000
#> GSM587172 2 0.0000 0.974 0.000 1.000 0.000
#> GSM587173 2 0.0000 0.974 0.000 1.000 0.000
#> GSM587174 2 0.0000 0.974 0.000 1.000 0.000
#> GSM587175 2 0.0000 0.974 0.000 1.000 0.000
#> GSM587176 2 0.0000 0.974 0.000 1.000 0.000
#> GSM587177 2 0.0000 0.974 0.000 1.000 0.000
#> GSM587178 2 0.0000 0.974 0.000 1.000 0.000
#> GSM587179 2 0.0000 0.974 0.000 1.000 0.000
#> GSM587180 2 0.0000 0.974 0.000 1.000 0.000
#> GSM587181 2 0.0000 0.974 0.000 1.000 0.000
#> GSM587182 2 0.0000 0.974 0.000 1.000 0.000
#> GSM587183 2 0.0000 0.974 0.000 1.000 0.000
#> GSM587184 2 0.0000 0.974 0.000 1.000 0.000
#> GSM587185 2 0.0000 0.974 0.000 1.000 0.000
#> GSM587186 2 0.0000 0.974 0.000 1.000 0.000
#> GSM587187 2 0.0000 0.974 0.000 1.000 0.000
#> GSM587188 2 0.0000 0.974 0.000 1.000 0.000
#> GSM587189 2 0.0000 0.974 0.000 1.000 0.000
#> GSM587190 2 0.1529 0.941 0.000 0.960 0.040
#> GSM587203 1 0.0000 0.956 1.000 0.000 0.000
#> GSM587204 1 0.0000 0.956 1.000 0.000 0.000
#> GSM587205 1 0.0000 0.956 1.000 0.000 0.000
#> GSM587206 1 0.0000 0.956 1.000 0.000 0.000
#> GSM587207 1 0.0000 0.956 1.000 0.000 0.000
#> GSM587208 1 0.0000 0.956 1.000 0.000 0.000
#> GSM587209 1 0.0000 0.956 1.000 0.000 0.000
#> GSM587210 1 0.3412 0.862 0.876 0.000 0.124
#> GSM587211 1 0.0000 0.956 1.000 0.000 0.000
#> GSM587212 1 0.3941 0.819 0.844 0.000 0.156
#> GSM587213 1 0.0000 0.956 1.000 0.000 0.000
#> GSM587214 1 0.0000 0.956 1.000 0.000 0.000
#> GSM587215 1 0.0000 0.956 1.000 0.000 0.000
#> GSM587216 1 0.0424 0.951 0.992 0.000 0.008
#> GSM587217 1 0.0000 0.956 1.000 0.000 0.000
#> GSM587191 2 0.0424 0.969 0.000 0.992 0.008
#> GSM587192 1 0.1170 0.944 0.976 0.008 0.016
#> GSM587193 1 0.5706 0.560 0.680 0.000 0.320
#> GSM587194 3 0.1411 0.868 0.000 0.036 0.964
#> GSM587195 2 0.2537 0.905 0.000 0.920 0.080
#> GSM587196 2 0.1529 0.944 0.000 0.960 0.040
#> GSM587197 3 0.6154 0.321 0.000 0.408 0.592
#> GSM587198 2 0.3267 0.861 0.000 0.884 0.116
#> GSM587199 3 0.5291 0.643 0.000 0.268 0.732
#> GSM587200 3 0.6154 0.238 0.408 0.000 0.592
#> GSM587201 1 0.0424 0.951 0.992 0.000 0.008
#> GSM587202 2 0.0592 0.966 0.000 0.988 0.012
#> GSM198767 1 0.0000 0.956 1.000 0.000 0.000
#> GSM198769 1 0.0000 0.956 1.000 0.000 0.000
#> GSM198772 1 0.0000 0.956 1.000 0.000 0.000
#> GSM198773 1 0.0000 0.956 1.000 0.000 0.000
#> GSM198776 1 0.0000 0.956 1.000 0.000 0.000
#> GSM198778 1 0.2537 0.903 0.920 0.000 0.080
#> GSM198780 1 0.3267 0.865 0.884 0.000 0.116
#> GSM198781 1 0.0000 0.956 1.000 0.000 0.000
#> GSM198765 2 0.0424 0.969 0.000 0.992 0.008
#> GSM198766 1 0.5327 0.647 0.728 0.000 0.272
#> GSM198768 2 0.2711 0.897 0.000 0.912 0.088
#> GSM198770 2 0.6244 0.173 0.000 0.560 0.440
#> GSM198771 2 0.4974 0.677 0.000 0.764 0.236
#> GSM198774 1 0.1453 0.932 0.968 0.024 0.008
#> GSM198775 3 0.2711 0.830 0.000 0.088 0.912
#> GSM198777 2 0.1411 0.948 0.000 0.964 0.036
#> GSM198779 3 0.5497 0.605 0.000 0.292 0.708
#> GSM587218 3 0.0424 0.891 0.008 0.000 0.992
#> GSM587219 3 0.0424 0.891 0.008 0.000 0.992
#> GSM587220 3 0.0424 0.891 0.008 0.000 0.992
#> GSM587221 3 0.0424 0.891 0.008 0.000 0.992
#> GSM587222 3 0.0424 0.891 0.008 0.000 0.992
#> GSM587223 3 0.0424 0.891 0.008 0.000 0.992
#> GSM587224 3 0.0424 0.891 0.008 0.000 0.992
#> GSM587225 3 0.0424 0.891 0.008 0.000 0.992
#> GSM587226 3 0.0424 0.891 0.008 0.000 0.992
#> GSM587227 3 0.0424 0.891 0.008 0.000 0.992
#> GSM587228 3 0.0424 0.891 0.008 0.000 0.992
#> GSM587229 3 0.0424 0.891 0.008 0.000 0.992
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM587155 2 0.0000 0.990 0.000 1.000 0.000 0.000
#> GSM587156 2 0.0188 0.987 0.000 0.996 0.004 0.000
#> GSM587157 2 0.1557 0.933 0.000 0.944 0.056 0.000
#> GSM587158 2 0.0000 0.990 0.000 1.000 0.000 0.000
#> GSM587159 2 0.0000 0.990 0.000 1.000 0.000 0.000
#> GSM587160 2 0.0000 0.990 0.000 1.000 0.000 0.000
#> GSM587161 2 0.0000 0.990 0.000 1.000 0.000 0.000
#> GSM587162 2 0.0000 0.990 0.000 1.000 0.000 0.000
#> GSM587163 2 0.0000 0.990 0.000 1.000 0.000 0.000
#> GSM587164 2 0.0000 0.990 0.000 1.000 0.000 0.000
#> GSM587165 2 0.0000 0.990 0.000 1.000 0.000 0.000
#> GSM587166 2 0.0592 0.976 0.000 0.984 0.016 0.000
#> GSM587167 2 0.0000 0.990 0.000 1.000 0.000 0.000
#> GSM587168 2 0.0000 0.990 0.000 1.000 0.000 0.000
#> GSM587169 2 0.0000 0.990 0.000 1.000 0.000 0.000
#> GSM587170 2 0.0000 0.990 0.000 1.000 0.000 0.000
#> GSM587171 2 0.0000 0.990 0.000 1.000 0.000 0.000
#> GSM587172 2 0.0000 0.990 0.000 1.000 0.000 0.000
#> GSM587173 2 0.0000 0.990 0.000 1.000 0.000 0.000
#> GSM587174 2 0.0000 0.990 0.000 1.000 0.000 0.000
#> GSM587175 2 0.0000 0.990 0.000 1.000 0.000 0.000
#> GSM587176 2 0.0000 0.990 0.000 1.000 0.000 0.000
#> GSM587177 2 0.0000 0.990 0.000 1.000 0.000 0.000
#> GSM587178 2 0.0000 0.990 0.000 1.000 0.000 0.000
#> GSM587179 2 0.0000 0.990 0.000 1.000 0.000 0.000
#> GSM587180 2 0.0000 0.990 0.000 1.000 0.000 0.000
#> GSM587181 2 0.0000 0.990 0.000 1.000 0.000 0.000
#> GSM587182 2 0.0000 0.990 0.000 1.000 0.000 0.000
#> GSM587183 2 0.0000 0.990 0.000 1.000 0.000 0.000
#> GSM587184 2 0.0000 0.990 0.000 1.000 0.000 0.000
#> GSM587185 2 0.0000 0.990 0.000 1.000 0.000 0.000
#> GSM587186 2 0.0000 0.990 0.000 1.000 0.000 0.000
#> GSM587187 2 0.0188 0.987 0.000 0.996 0.004 0.000
#> GSM587188 2 0.0188 0.987 0.000 0.996 0.004 0.000
#> GSM587189 2 0.0188 0.987 0.000 0.996 0.004 0.000
#> GSM587190 2 0.3837 0.713 0.000 0.776 0.224 0.000
#> GSM587203 1 0.0000 1.000 1.000 0.000 0.000 0.000
#> GSM587204 1 0.0000 1.000 1.000 0.000 0.000 0.000
#> GSM587205 1 0.0000 1.000 1.000 0.000 0.000 0.000
#> GSM587206 1 0.0000 1.000 1.000 0.000 0.000 0.000
#> GSM587207 1 0.0000 1.000 1.000 0.000 0.000 0.000
#> GSM587208 1 0.0000 1.000 1.000 0.000 0.000 0.000
#> GSM587209 1 0.0000 1.000 1.000 0.000 0.000 0.000
#> GSM587210 3 0.2408 0.876 0.104 0.000 0.896 0.000
#> GSM587211 1 0.0000 1.000 1.000 0.000 0.000 0.000
#> GSM587212 3 0.4431 0.608 0.304 0.000 0.696 0.000
#> GSM587213 1 0.0000 1.000 1.000 0.000 0.000 0.000
#> GSM587214 1 0.0000 1.000 1.000 0.000 0.000 0.000
#> GSM587215 1 0.0000 1.000 1.000 0.000 0.000 0.000
#> GSM587216 1 0.0000 1.000 1.000 0.000 0.000 0.000
#> GSM587217 1 0.0000 1.000 1.000 0.000 0.000 0.000
#> GSM587191 3 0.0000 0.951 0.000 0.000 1.000 0.000
#> GSM587192 3 0.0188 0.949 0.004 0.000 0.996 0.000
#> GSM587193 3 0.2216 0.884 0.092 0.000 0.908 0.000
#> GSM587194 3 0.0336 0.946 0.000 0.000 0.992 0.008
#> GSM587195 3 0.0000 0.951 0.000 0.000 1.000 0.000
#> GSM587196 3 0.0000 0.951 0.000 0.000 1.000 0.000
#> GSM587197 3 0.0000 0.951 0.000 0.000 1.000 0.000
#> GSM587198 3 0.0000 0.951 0.000 0.000 1.000 0.000
#> GSM587199 3 0.0000 0.951 0.000 0.000 1.000 0.000
#> GSM587200 3 0.0188 0.949 0.004 0.000 0.996 0.000
#> GSM587201 3 0.0188 0.949 0.004 0.000 0.996 0.000
#> GSM587202 3 0.0000 0.951 0.000 0.000 1.000 0.000
#> GSM198767 1 0.0000 1.000 1.000 0.000 0.000 0.000
#> GSM198769 1 0.0000 1.000 1.000 0.000 0.000 0.000
#> GSM198772 1 0.0000 1.000 1.000 0.000 0.000 0.000
#> GSM198773 1 0.0000 1.000 1.000 0.000 0.000 0.000
#> GSM198776 1 0.0000 1.000 1.000 0.000 0.000 0.000
#> GSM198778 3 0.1867 0.903 0.072 0.000 0.928 0.000
#> GSM198780 3 0.4304 0.644 0.284 0.000 0.716 0.000
#> GSM198781 1 0.0000 1.000 1.000 0.000 0.000 0.000
#> GSM198765 3 0.0000 0.951 0.000 0.000 1.000 0.000
#> GSM198766 3 0.3610 0.763 0.200 0.000 0.800 0.000
#> GSM198768 3 0.0000 0.951 0.000 0.000 1.000 0.000
#> GSM198770 3 0.0000 0.951 0.000 0.000 1.000 0.000
#> GSM198771 3 0.0000 0.951 0.000 0.000 1.000 0.000
#> GSM198774 3 0.0188 0.949 0.004 0.000 0.996 0.000
#> GSM198775 3 0.0000 0.951 0.000 0.000 1.000 0.000
#> GSM198777 3 0.0000 0.951 0.000 0.000 1.000 0.000
#> GSM198779 3 0.0000 0.951 0.000 0.000 1.000 0.000
#> GSM587218 4 0.0000 1.000 0.000 0.000 0.000 1.000
#> GSM587219 4 0.0000 1.000 0.000 0.000 0.000 1.000
#> GSM587220 4 0.0000 1.000 0.000 0.000 0.000 1.000
#> GSM587221 4 0.0000 1.000 0.000 0.000 0.000 1.000
#> GSM587222 4 0.0000 1.000 0.000 0.000 0.000 1.000
#> GSM587223 4 0.0000 1.000 0.000 0.000 0.000 1.000
#> GSM587224 4 0.0000 1.000 0.000 0.000 0.000 1.000
#> GSM587225 4 0.0000 1.000 0.000 0.000 0.000 1.000
#> GSM587226 4 0.0000 1.000 0.000 0.000 0.000 1.000
#> GSM587227 4 0.0000 1.000 0.000 0.000 0.000 1.000
#> GSM587228 4 0.0000 1.000 0.000 0.000 0.000 1.000
#> GSM587229 4 0.0000 1.000 0.000 0.000 0.000 1.000
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM587155 2 0.1124 0.9453 0.000 0.960 0.036 0 0.004
#> GSM587156 2 0.3969 0.6024 0.000 0.692 0.304 0 0.004
#> GSM587157 5 0.4323 0.3961 0.000 0.332 0.012 0 0.656
#> GSM587158 2 0.0000 0.9699 0.000 1.000 0.000 0 0.000
#> GSM587159 2 0.0000 0.9699 0.000 1.000 0.000 0 0.000
#> GSM587160 2 0.0000 0.9699 0.000 1.000 0.000 0 0.000
#> GSM587161 2 0.0290 0.9664 0.000 0.992 0.008 0 0.000
#> GSM587162 2 0.0000 0.9699 0.000 1.000 0.000 0 0.000
#> GSM587163 2 0.0000 0.9699 0.000 1.000 0.000 0 0.000
#> GSM587164 2 0.1571 0.9263 0.000 0.936 0.060 0 0.004
#> GSM587165 2 0.0000 0.9699 0.000 1.000 0.000 0 0.000
#> GSM587166 2 0.4118 0.5396 0.000 0.660 0.336 0 0.004
#> GSM587167 2 0.2124 0.8938 0.000 0.900 0.096 0 0.004
#> GSM587168 2 0.0162 0.9685 0.000 0.996 0.000 0 0.004
#> GSM587169 2 0.0000 0.9699 0.000 1.000 0.000 0 0.000
#> GSM587170 2 0.2011 0.9016 0.000 0.908 0.088 0 0.004
#> GSM587171 2 0.0000 0.9699 0.000 1.000 0.000 0 0.000
#> GSM587172 2 0.0000 0.9699 0.000 1.000 0.000 0 0.000
#> GSM587173 2 0.0162 0.9685 0.000 0.996 0.000 0 0.004
#> GSM587174 2 0.0000 0.9699 0.000 1.000 0.000 0 0.000
#> GSM587175 2 0.0579 0.9622 0.000 0.984 0.008 0 0.008
#> GSM587176 2 0.0162 0.9683 0.000 0.996 0.004 0 0.000
#> GSM587177 2 0.0162 0.9685 0.000 0.996 0.000 0 0.004
#> GSM587178 2 0.0000 0.9699 0.000 1.000 0.000 0 0.000
#> GSM587179 2 0.0000 0.9699 0.000 1.000 0.000 0 0.000
#> GSM587180 2 0.0000 0.9699 0.000 1.000 0.000 0 0.000
#> GSM587181 2 0.0000 0.9699 0.000 1.000 0.000 0 0.000
#> GSM587182 2 0.0000 0.9699 0.000 1.000 0.000 0 0.000
#> GSM587183 2 0.0000 0.9699 0.000 1.000 0.000 0 0.000
#> GSM587184 2 0.0000 0.9699 0.000 1.000 0.000 0 0.000
#> GSM587185 2 0.0000 0.9699 0.000 1.000 0.000 0 0.000
#> GSM587186 2 0.0162 0.9685 0.000 0.996 0.000 0 0.004
#> GSM587187 2 0.0162 0.9685 0.000 0.996 0.000 0 0.004
#> GSM587188 2 0.0324 0.9667 0.000 0.992 0.004 0 0.004
#> GSM587189 2 0.0290 0.9663 0.000 0.992 0.000 0 0.008
#> GSM587190 3 0.4702 0.1270 0.000 0.432 0.552 0 0.016
#> GSM587203 1 0.0000 0.9818 1.000 0.000 0.000 0 0.000
#> GSM587204 1 0.0000 0.9818 1.000 0.000 0.000 0 0.000
#> GSM587205 1 0.0000 0.9818 1.000 0.000 0.000 0 0.000
#> GSM587206 1 0.0000 0.9818 1.000 0.000 0.000 0 0.000
#> GSM587207 1 0.0000 0.9818 1.000 0.000 0.000 0 0.000
#> GSM587208 1 0.0000 0.9818 1.000 0.000 0.000 0 0.000
#> GSM587209 1 0.0162 0.9803 0.996 0.000 0.004 0 0.000
#> GSM587210 3 0.4922 0.6649 0.156 0.000 0.716 0 0.128
#> GSM587211 1 0.1648 0.9403 0.940 0.000 0.040 0 0.020
#> GSM587212 3 0.4280 0.7009 0.140 0.000 0.772 0 0.088
#> GSM587213 1 0.0000 0.9818 1.000 0.000 0.000 0 0.000
#> GSM587214 1 0.0000 0.9818 1.000 0.000 0.000 0 0.000
#> GSM587215 1 0.0324 0.9786 0.992 0.000 0.004 0 0.004
#> GSM587216 1 0.3039 0.7598 0.808 0.000 0.192 0 0.000
#> GSM587217 1 0.0324 0.9786 0.992 0.000 0.004 0 0.004
#> GSM587191 3 0.2773 0.7169 0.000 0.000 0.836 0 0.164
#> GSM587192 3 0.1197 0.7739 0.000 0.000 0.952 0 0.048
#> GSM587193 3 0.0162 0.7667 0.000 0.000 0.996 0 0.004
#> GSM587194 3 0.0404 0.7725 0.000 0.000 0.988 0 0.012
#> GSM587195 5 0.0000 0.8687 0.000 0.000 0.000 0 1.000
#> GSM587196 5 0.0162 0.8706 0.000 0.000 0.004 0 0.996
#> GSM587197 5 0.0162 0.8706 0.000 0.000 0.004 0 0.996
#> GSM587198 5 0.1544 0.8424 0.000 0.000 0.068 0 0.932
#> GSM587199 5 0.3561 0.6013 0.000 0.000 0.260 0 0.740
#> GSM587200 3 0.4182 0.3711 0.000 0.000 0.600 0 0.400
#> GSM587201 3 0.4307 0.0809 0.000 0.000 0.500 0 0.500
#> GSM587202 5 0.0703 0.8652 0.000 0.000 0.024 0 0.976
#> GSM198767 1 0.0000 0.9818 1.000 0.000 0.000 0 0.000
#> GSM198769 1 0.0162 0.9803 0.996 0.000 0.004 0 0.000
#> GSM198772 1 0.1444 0.9465 0.948 0.000 0.040 0 0.012
#> GSM198773 1 0.0000 0.9818 1.000 0.000 0.000 0 0.000
#> GSM198776 1 0.0000 0.9818 1.000 0.000 0.000 0 0.000
#> GSM198778 3 0.4805 0.6770 0.128 0.000 0.728 0 0.144
#> GSM198780 3 0.4291 0.7026 0.136 0.000 0.772 0 0.092
#> GSM198781 1 0.0000 0.9818 1.000 0.000 0.000 0 0.000
#> GSM198765 3 0.2377 0.7433 0.000 0.000 0.872 0 0.128
#> GSM198766 3 0.0162 0.7705 0.000 0.000 0.996 0 0.004
#> GSM198768 5 0.0000 0.8687 0.000 0.000 0.000 0 1.000
#> GSM198770 5 0.0162 0.8706 0.000 0.000 0.004 0 0.996
#> GSM198771 5 0.1608 0.8396 0.000 0.000 0.072 0 0.928
#> GSM198774 3 0.1197 0.7739 0.000 0.000 0.952 0 0.048
#> GSM198775 3 0.0404 0.7725 0.000 0.000 0.988 0 0.012
#> GSM198777 5 0.0162 0.8706 0.000 0.000 0.004 0 0.996
#> GSM198779 5 0.3534 0.6091 0.000 0.000 0.256 0 0.744
#> GSM587218 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000
#> GSM587219 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000
#> GSM587220 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000
#> GSM587221 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000
#> GSM587222 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000
#> GSM587223 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000
#> GSM587224 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000
#> GSM587225 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000
#> GSM587226 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000
#> GSM587227 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000
#> GSM587228 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000
#> GSM587229 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM587155 2 0.3934 0.6537 0.000 0.708 0.000 0 0.032 0.260
#> GSM587156 6 0.5823 0.0824 0.000 0.332 0.000 0 0.200 0.468
#> GSM587157 3 0.4250 0.5983 0.000 0.144 0.744 0 0.004 0.108
#> GSM587158 2 0.0000 0.9250 0.000 1.000 0.000 0 0.000 0.000
#> GSM587159 2 0.0146 0.9248 0.000 0.996 0.000 0 0.000 0.004
#> GSM587160 2 0.0632 0.9197 0.000 0.976 0.000 0 0.000 0.024
#> GSM587161 2 0.2219 0.8345 0.000 0.864 0.000 0 0.000 0.136
#> GSM587162 2 0.0632 0.9202 0.000 0.976 0.000 0 0.000 0.024
#> GSM587163 2 0.0363 0.9233 0.000 0.988 0.000 0 0.000 0.012
#> GSM587164 2 0.4350 0.5809 0.000 0.660 0.000 0 0.048 0.292
#> GSM587165 2 0.0458 0.9220 0.000 0.984 0.000 0 0.000 0.016
#> GSM587166 6 0.5813 0.0699 0.000 0.296 0.000 0 0.216 0.488
#> GSM587167 2 0.4747 0.4317 0.000 0.584 0.000 0 0.060 0.356
#> GSM587168 2 0.0458 0.9220 0.000 0.984 0.000 0 0.000 0.016
#> GSM587169 2 0.0458 0.9223 0.000 0.984 0.000 0 0.000 0.016
#> GSM587170 2 0.4436 0.5459 0.000 0.640 0.000 0 0.048 0.312
#> GSM587171 2 0.0146 0.9248 0.000 0.996 0.000 0 0.000 0.004
#> GSM587172 2 0.0146 0.9248 0.000 0.996 0.000 0 0.000 0.004
#> GSM587173 2 0.0547 0.9204 0.000 0.980 0.000 0 0.000 0.020
#> GSM587174 2 0.0000 0.9250 0.000 1.000 0.000 0 0.000 0.000
#> GSM587175 2 0.3309 0.7588 0.000 0.788 0.016 0 0.004 0.192
#> GSM587176 2 0.0865 0.9140 0.000 0.964 0.000 0 0.000 0.036
#> GSM587177 2 0.0363 0.9232 0.000 0.988 0.000 0 0.000 0.012
#> GSM587178 2 0.0260 0.9241 0.000 0.992 0.000 0 0.000 0.008
#> GSM587179 2 0.0632 0.9197 0.000 0.976 0.000 0 0.000 0.024
#> GSM587180 2 0.0260 0.9241 0.000 0.992 0.000 0 0.000 0.008
#> GSM587181 2 0.0000 0.9250 0.000 1.000 0.000 0 0.000 0.000
#> GSM587182 2 0.0260 0.9241 0.000 0.992 0.000 0 0.000 0.008
#> GSM587183 2 0.0363 0.9232 0.000 0.988 0.000 0 0.000 0.012
#> GSM587184 2 0.0260 0.9251 0.000 0.992 0.000 0 0.000 0.008
#> GSM587185 2 0.0713 0.9180 0.000 0.972 0.000 0 0.000 0.028
#> GSM587186 2 0.0547 0.9204 0.000 0.980 0.000 0 0.000 0.020
#> GSM587187 2 0.0806 0.9165 0.000 0.972 0.008 0 0.000 0.020
#> GSM587188 2 0.1434 0.8980 0.000 0.948 0.024 0 0.008 0.020
#> GSM587189 2 0.1871 0.8835 0.000 0.928 0.032 0 0.016 0.024
#> GSM587190 5 0.5929 0.3363 0.000 0.168 0.036 0 0.584 0.212
#> GSM587203 1 0.0000 0.8713 1.000 0.000 0.000 0 0.000 0.000
#> GSM587204 1 0.0000 0.8713 1.000 0.000 0.000 0 0.000 0.000
#> GSM587205 1 0.0000 0.8713 1.000 0.000 0.000 0 0.000 0.000
#> GSM587206 1 0.0000 0.8713 1.000 0.000 0.000 0 0.000 0.000
#> GSM587207 1 0.0000 0.8713 1.000 0.000 0.000 0 0.000 0.000
#> GSM587208 1 0.0000 0.8713 1.000 0.000 0.000 0 0.000 0.000
#> GSM587209 1 0.2941 0.7200 0.780 0.000 0.000 0 0.000 0.220
#> GSM587210 5 0.5128 0.2106 0.056 0.000 0.012 0 0.524 0.408
#> GSM587211 6 0.4976 -0.1733 0.428 0.000 0.032 0 0.020 0.520
#> GSM587212 6 0.5370 0.0175 0.060 0.000 0.036 0 0.312 0.592
#> GSM587213 1 0.1007 0.8614 0.956 0.000 0.000 0 0.000 0.044
#> GSM587214 1 0.0632 0.8676 0.976 0.000 0.000 0 0.000 0.024
#> GSM587215 1 0.4153 0.5279 0.636 0.000 0.024 0 0.000 0.340
#> GSM587216 1 0.5176 0.3034 0.548 0.000 0.000 0 0.100 0.352
#> GSM587217 1 0.3795 0.5108 0.632 0.000 0.004 0 0.000 0.364
#> GSM587191 5 0.1950 0.6429 0.000 0.000 0.064 0 0.912 0.024
#> GSM587192 5 0.2165 0.6251 0.000 0.000 0.008 0 0.884 0.108
#> GSM587193 5 0.3838 0.3913 0.000 0.000 0.000 0 0.552 0.448
#> GSM587194 5 0.2854 0.5802 0.000 0.000 0.000 0 0.792 0.208
#> GSM587195 3 0.0458 0.8622 0.000 0.000 0.984 0 0.000 0.016
#> GSM587196 3 0.0458 0.8621 0.000 0.000 0.984 0 0.000 0.016
#> GSM587197 3 0.0291 0.8611 0.000 0.000 0.992 0 0.004 0.004
#> GSM587198 3 0.3834 0.6649 0.000 0.000 0.732 0 0.232 0.036
#> GSM587199 5 0.4466 0.3438 0.000 0.000 0.336 0 0.620 0.044
#> GSM587200 5 0.3660 0.6000 0.000 0.000 0.160 0 0.780 0.060
#> GSM587201 5 0.4066 0.5589 0.000 0.000 0.204 0 0.732 0.064
#> GSM587202 3 0.2909 0.7736 0.000 0.000 0.836 0 0.136 0.028
#> GSM198767 1 0.0000 0.8713 1.000 0.000 0.000 0 0.000 0.000
#> GSM198769 1 0.2941 0.7200 0.780 0.000 0.000 0 0.000 0.220
#> GSM198772 6 0.4854 -0.1898 0.436 0.000 0.024 0 0.020 0.520
#> GSM198773 1 0.0937 0.8626 0.960 0.000 0.000 0 0.000 0.040
#> GSM198776 1 0.0000 0.8713 1.000 0.000 0.000 0 0.000 0.000
#> GSM198778 5 0.5147 0.2306 0.052 0.000 0.016 0 0.532 0.400
#> GSM198780 6 0.5370 0.0175 0.060 0.000 0.036 0 0.312 0.592
#> GSM198781 1 0.0547 0.8685 0.980 0.000 0.000 0 0.000 0.020
#> GSM198765 5 0.1829 0.6429 0.000 0.000 0.056 0 0.920 0.024
#> GSM198766 5 0.3810 0.4123 0.000 0.000 0.000 0 0.572 0.428
#> GSM198768 3 0.0632 0.8603 0.000 0.000 0.976 0 0.000 0.024
#> GSM198770 3 0.0405 0.8605 0.000 0.000 0.988 0 0.008 0.004
#> GSM198771 3 0.3860 0.6589 0.000 0.000 0.728 0 0.236 0.036
#> GSM198774 5 0.2118 0.6261 0.000 0.000 0.008 0 0.888 0.104
#> GSM198775 5 0.2912 0.5775 0.000 0.000 0.000 0 0.784 0.216
#> GSM198777 3 0.0458 0.8621 0.000 0.000 0.984 0 0.000 0.016
#> GSM198779 5 0.4480 0.3350 0.000 0.000 0.340 0 0.616 0.044
#> GSM587218 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587219 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587220 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587221 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587222 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587223 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587224 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587225 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587226 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587227 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587228 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587229 4 0.0000 1.0000 0.000 0.000 0.000 1 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 specimen(p) k
#> SD:NMF 92 4.01e-14 2
#> SD:NMF 89 4.94e-24 3
#> SD:NMF 92 1.69e-42 4
#> SD:NMF 88 5.18e-38 5
#> SD:NMF 77 1.31e-34 6
If matrix rows can be associated to genes, consider to use functional_enrichment(res,
...) to perform function enrichment for the signature genes. See this vignette for more detailed explanations.
The object with results only for a single top-value method and a single partition method can be extracted as:
res = res_list["CV", "hclust"]
# you can also extract it by
# res = res_list["CV:hclust"]
A summary of res and all the functions that can be applied to it:
res
#> A 'ConsensusPartition' object with k = 2, 3, 4, 5, 6.
#> On a matrix with 51941 rows and 92 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 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 0.653 0.840 0.923 0.4819 0.514 0.514
#> 3 3 0.629 0.818 0.865 0.2075 0.928 0.861
#> 4 4 0.886 0.928 0.966 0.2426 0.817 0.593
#> 5 5 0.883 0.878 0.924 0.0446 0.969 0.890
#> 6 6 0.872 0.769 0.880 0.0352 0.991 0.963
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
#> GSM587155 2 0.0000 0.871 0.000 1.000
#> GSM587156 2 0.0000 0.871 0.000 1.000
#> GSM587157 2 0.0000 0.871 0.000 1.000
#> GSM587158 2 0.0000 0.871 0.000 1.000
#> GSM587159 2 0.0000 0.871 0.000 1.000
#> GSM587160 2 0.0000 0.871 0.000 1.000
#> GSM587161 2 0.0000 0.871 0.000 1.000
#> GSM587162 2 0.0000 0.871 0.000 1.000
#> GSM587163 2 0.0000 0.871 0.000 1.000
#> GSM587164 2 0.0000 0.871 0.000 1.000
#> GSM587165 2 0.0000 0.871 0.000 1.000
#> GSM587166 2 0.0000 0.871 0.000 1.000
#> GSM587167 2 0.0000 0.871 0.000 1.000
#> GSM587168 2 0.0000 0.871 0.000 1.000
#> GSM587169 2 0.0000 0.871 0.000 1.000
#> GSM587170 2 0.0000 0.871 0.000 1.000
#> GSM587171 2 0.0000 0.871 0.000 1.000
#> GSM587172 2 0.0000 0.871 0.000 1.000
#> GSM587173 2 0.0000 0.871 0.000 1.000
#> GSM587174 2 0.0000 0.871 0.000 1.000
#> GSM587175 2 0.0000 0.871 0.000 1.000
#> GSM587176 2 0.0000 0.871 0.000 1.000
#> GSM587177 2 0.0000 0.871 0.000 1.000
#> GSM587178 2 0.0000 0.871 0.000 1.000
#> GSM587179 2 0.0000 0.871 0.000 1.000
#> GSM587180 2 0.0000 0.871 0.000 1.000
#> GSM587181 2 0.0000 0.871 0.000 1.000
#> GSM587182 2 0.0000 0.871 0.000 1.000
#> GSM587183 2 0.0000 0.871 0.000 1.000
#> GSM587184 2 0.0000 0.871 0.000 1.000
#> GSM587185 2 0.0000 0.871 0.000 1.000
#> GSM587186 2 0.0000 0.871 0.000 1.000
#> GSM587187 2 0.0000 0.871 0.000 1.000
#> GSM587188 2 0.0000 0.871 0.000 1.000
#> GSM587189 2 0.0000 0.871 0.000 1.000
#> GSM587190 2 0.0000 0.871 0.000 1.000
#> GSM587203 1 0.0000 0.967 1.000 0.000
#> GSM587204 1 0.0000 0.967 1.000 0.000
#> GSM587205 1 0.0000 0.967 1.000 0.000
#> GSM587206 1 0.0000 0.967 1.000 0.000
#> GSM587207 1 0.0000 0.967 1.000 0.000
#> GSM587208 1 0.0000 0.967 1.000 0.000
#> GSM587209 1 0.0938 0.956 0.988 0.012
#> GSM587210 1 0.1184 0.952 0.984 0.016
#> GSM587211 1 0.0000 0.967 1.000 0.000
#> GSM587212 1 0.0376 0.964 0.996 0.004
#> GSM587213 1 0.0000 0.967 1.000 0.000
#> GSM587214 1 0.0000 0.967 1.000 0.000
#> GSM587215 1 0.0000 0.967 1.000 0.000
#> GSM587216 1 0.0000 0.967 1.000 0.000
#> GSM587217 1 0.0000 0.967 1.000 0.000
#> GSM587191 2 0.9248 0.633 0.340 0.660
#> GSM587192 2 0.9248 0.633 0.340 0.660
#> GSM587193 2 0.8207 0.716 0.256 0.744
#> GSM587194 2 0.8207 0.716 0.256 0.744
#> GSM587195 2 0.9248 0.633 0.340 0.660
#> GSM587196 2 0.9248 0.633 0.340 0.660
#> GSM587197 2 0.9248 0.633 0.340 0.660
#> GSM587198 2 0.9286 0.627 0.344 0.656
#> GSM587199 2 0.9286 0.627 0.344 0.656
#> GSM587200 1 0.9944 -0.106 0.544 0.456
#> GSM587201 1 0.9944 -0.106 0.544 0.456
#> GSM587202 2 0.9286 0.627 0.344 0.656
#> GSM198767 1 0.0000 0.967 1.000 0.000
#> GSM198769 1 0.0938 0.956 0.988 0.012
#> GSM198772 1 0.0000 0.967 1.000 0.000
#> GSM198773 1 0.0000 0.967 1.000 0.000
#> GSM198776 1 0.0000 0.967 1.000 0.000
#> GSM198778 1 0.1184 0.952 0.984 0.016
#> GSM198780 1 0.0376 0.964 0.996 0.004
#> GSM198781 1 0.0000 0.967 1.000 0.000
#> GSM198765 2 0.9248 0.633 0.340 0.660
#> GSM198766 2 0.8207 0.716 0.256 0.744
#> GSM198768 2 0.9248 0.633 0.340 0.660
#> GSM198770 2 0.9248 0.633 0.340 0.660
#> GSM198771 2 0.9286 0.627 0.344 0.656
#> GSM198774 2 0.9248 0.633 0.340 0.660
#> GSM198775 2 0.8207 0.716 0.256 0.744
#> GSM198777 2 0.9248 0.633 0.340 0.660
#> GSM198779 2 0.9286 0.627 0.344 0.656
#> GSM587218 1 0.0000 0.967 1.000 0.000
#> GSM587219 1 0.0000 0.967 1.000 0.000
#> GSM587220 1 0.0000 0.967 1.000 0.000
#> GSM587221 1 0.0000 0.967 1.000 0.000
#> GSM587222 1 0.0000 0.967 1.000 0.000
#> GSM587223 1 0.0000 0.967 1.000 0.000
#> GSM587224 1 0.0000 0.967 1.000 0.000
#> GSM587225 1 0.0000 0.967 1.000 0.000
#> GSM587226 1 0.0000 0.967 1.000 0.000
#> GSM587227 1 0.0000 0.967 1.000 0.000
#> GSM587228 1 0.0000 0.967 1.000 0.000
#> GSM587229 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
#> GSM587155 2 0.0000 0.873 0.000 1.000 0.000
#> GSM587156 2 0.0237 0.872 0.004 0.996 0.000
#> GSM587157 2 0.0000 0.873 0.000 1.000 0.000
#> GSM587158 2 0.0000 0.873 0.000 1.000 0.000
#> GSM587159 2 0.0000 0.873 0.000 1.000 0.000
#> GSM587160 2 0.0000 0.873 0.000 1.000 0.000
#> GSM587161 2 0.0000 0.873 0.000 1.000 0.000
#> GSM587162 2 0.0000 0.873 0.000 1.000 0.000
#> GSM587163 2 0.0000 0.873 0.000 1.000 0.000
#> GSM587164 2 0.0000 0.873 0.000 1.000 0.000
#> GSM587165 2 0.0000 0.873 0.000 1.000 0.000
#> GSM587166 2 0.0237 0.872 0.004 0.996 0.000
#> GSM587167 2 0.0237 0.872 0.004 0.996 0.000
#> GSM587168 2 0.0000 0.873 0.000 1.000 0.000
#> GSM587169 2 0.0000 0.873 0.000 1.000 0.000
#> GSM587170 2 0.0000 0.873 0.000 1.000 0.000
#> GSM587171 2 0.0000 0.873 0.000 1.000 0.000
#> GSM587172 2 0.0000 0.873 0.000 1.000 0.000
#> GSM587173 2 0.0000 0.873 0.000 1.000 0.000
#> GSM587174 2 0.0000 0.873 0.000 1.000 0.000
#> GSM587175 2 0.0000 0.873 0.000 1.000 0.000
#> GSM587176 2 0.0000 0.873 0.000 1.000 0.000
#> GSM587177 2 0.0000 0.873 0.000 1.000 0.000
#> GSM587178 2 0.0000 0.873 0.000 1.000 0.000
#> GSM587179 2 0.0000 0.873 0.000 1.000 0.000
#> GSM587180 2 0.0000 0.873 0.000 1.000 0.000
#> GSM587181 2 0.0000 0.873 0.000 1.000 0.000
#> GSM587182 2 0.0000 0.873 0.000 1.000 0.000
#> GSM587183 2 0.0000 0.873 0.000 1.000 0.000
#> GSM587184 2 0.0000 0.873 0.000 1.000 0.000
#> GSM587185 2 0.0000 0.873 0.000 1.000 0.000
#> GSM587186 2 0.0000 0.873 0.000 1.000 0.000
#> GSM587187 2 0.0000 0.873 0.000 1.000 0.000
#> GSM587188 2 0.0000 0.873 0.000 1.000 0.000
#> GSM587189 2 0.0000 0.873 0.000 1.000 0.000
#> GSM587190 2 0.0237 0.872 0.004 0.996 0.000
#> GSM587203 1 0.4178 0.866 0.828 0.000 0.172
#> GSM587204 1 0.4178 0.866 0.828 0.000 0.172
#> GSM587205 1 0.4178 0.866 0.828 0.000 0.172
#> GSM587206 1 0.4178 0.866 0.828 0.000 0.172
#> GSM587207 1 0.4178 0.866 0.828 0.000 0.172
#> GSM587208 1 0.4178 0.866 0.828 0.000 0.172
#> GSM587209 1 0.2356 0.766 0.928 0.000 0.072
#> GSM587210 1 0.2165 0.698 0.936 0.000 0.064
#> GSM587211 1 0.4178 0.866 0.828 0.000 0.172
#> GSM587212 1 0.4062 0.813 0.836 0.000 0.164
#> GSM587213 1 0.4178 0.866 0.828 0.000 0.172
#> GSM587214 1 0.4178 0.866 0.828 0.000 0.172
#> GSM587215 1 0.4178 0.866 0.828 0.000 0.172
#> GSM587216 1 0.4178 0.866 0.828 0.000 0.172
#> GSM587217 1 0.4178 0.866 0.828 0.000 0.172
#> GSM587191 2 0.7327 0.679 0.312 0.636 0.052
#> GSM587192 2 0.7327 0.679 0.312 0.636 0.052
#> GSM587193 2 0.6586 0.743 0.216 0.728 0.056
#> GSM587194 2 0.6586 0.743 0.216 0.728 0.056
#> GSM587195 2 0.7327 0.679 0.312 0.636 0.052
#> GSM587196 2 0.7327 0.679 0.312 0.636 0.052
#> GSM587197 2 0.7327 0.679 0.312 0.636 0.052
#> GSM587198 2 0.7417 0.675 0.312 0.632 0.056
#> GSM587199 2 0.7417 0.675 0.312 0.632 0.056
#> GSM587200 1 0.7890 -0.275 0.512 0.432 0.056
#> GSM587201 1 0.7890 -0.275 0.512 0.432 0.056
#> GSM587202 2 0.7417 0.675 0.312 0.632 0.056
#> GSM198767 1 0.4178 0.866 0.828 0.000 0.172
#> GSM198769 1 0.2356 0.766 0.928 0.000 0.072
#> GSM198772 1 0.4178 0.866 0.828 0.000 0.172
#> GSM198773 1 0.4178 0.866 0.828 0.000 0.172
#> GSM198776 1 0.4178 0.866 0.828 0.000 0.172
#> GSM198778 1 0.2165 0.698 0.936 0.000 0.064
#> GSM198780 1 0.4062 0.813 0.836 0.000 0.164
#> GSM198781 1 0.4178 0.866 0.828 0.000 0.172
#> GSM198765 2 0.7327 0.679 0.312 0.636 0.052
#> GSM198766 2 0.6586 0.743 0.216 0.728 0.056
#> GSM198768 2 0.7327 0.679 0.312 0.636 0.052
#> GSM198770 2 0.7327 0.679 0.312 0.636 0.052
#> GSM198771 2 0.7417 0.675 0.312 0.632 0.056
#> GSM198774 2 0.7327 0.679 0.312 0.636 0.052
#> GSM198775 2 0.6586 0.743 0.216 0.728 0.056
#> GSM198777 2 0.7327 0.679 0.312 0.636 0.052
#> GSM198779 2 0.7417 0.675 0.312 0.632 0.056
#> GSM587218 3 0.0000 1.000 0.000 0.000 1.000
#> GSM587219 3 0.0000 1.000 0.000 0.000 1.000
#> GSM587220 3 0.0000 1.000 0.000 0.000 1.000
#> GSM587221 3 0.0000 1.000 0.000 0.000 1.000
#> GSM587222 3 0.0000 1.000 0.000 0.000 1.000
#> GSM587223 3 0.0000 1.000 0.000 0.000 1.000
#> GSM587224 3 0.0000 1.000 0.000 0.000 1.000
#> GSM587225 3 0.0000 1.000 0.000 0.000 1.000
#> GSM587226 3 0.0000 1.000 0.000 0.000 1.000
#> GSM587227 3 0.0000 1.000 0.000 0.000 1.000
#> GSM587228 3 0.0000 1.000 0.000 0.000 1.000
#> GSM587229 3 0.0000 1.000 0.000 0.000 1.000
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM587155 2 0.0188 0.968 0.000 0.996 0.004 0
#> GSM587156 2 0.3837 0.728 0.000 0.776 0.224 0
#> GSM587157 2 0.0188 0.968 0.000 0.996 0.004 0
#> GSM587158 2 0.0000 0.970 0.000 1.000 0.000 0
#> GSM587159 2 0.0000 0.970 0.000 1.000 0.000 0
#> GSM587160 2 0.0000 0.970 0.000 1.000 0.000 0
#> GSM587161 2 0.0000 0.970 0.000 1.000 0.000 0
#> GSM587162 2 0.0000 0.970 0.000 1.000 0.000 0
#> GSM587163 2 0.0000 0.970 0.000 1.000 0.000 0
#> GSM587164 2 0.0188 0.968 0.000 0.996 0.004 0
#> GSM587165 2 0.0000 0.970 0.000 1.000 0.000 0
#> GSM587166 2 0.3837 0.728 0.000 0.776 0.224 0
#> GSM587167 2 0.3873 0.722 0.000 0.772 0.228 0
#> GSM587168 2 0.0000 0.970 0.000 1.000 0.000 0
#> GSM587169 2 0.0000 0.970 0.000 1.000 0.000 0
#> GSM587170 2 0.0188 0.968 0.000 0.996 0.004 0
#> GSM587171 2 0.0000 0.970 0.000 1.000 0.000 0
#> GSM587172 2 0.0000 0.970 0.000 1.000 0.000 0
#> GSM587173 2 0.0000 0.970 0.000 1.000 0.000 0
#> GSM587174 2 0.0000 0.970 0.000 1.000 0.000 0
#> GSM587175 2 0.0188 0.968 0.000 0.996 0.004 0
#> GSM587176 2 0.0000 0.970 0.000 1.000 0.000 0
#> GSM587177 2 0.0000 0.970 0.000 1.000 0.000 0
#> GSM587178 2 0.0000 0.970 0.000 1.000 0.000 0
#> GSM587179 2 0.0000 0.970 0.000 1.000 0.000 0
#> GSM587180 2 0.0000 0.970 0.000 1.000 0.000 0
#> GSM587181 2 0.0000 0.970 0.000 1.000 0.000 0
#> GSM587182 2 0.0000 0.970 0.000 1.000 0.000 0
#> GSM587183 2 0.0000 0.970 0.000 1.000 0.000 0
#> GSM587184 2 0.0000 0.970 0.000 1.000 0.000 0
#> GSM587185 2 0.0000 0.970 0.000 1.000 0.000 0
#> GSM587186 2 0.0000 0.970 0.000 1.000 0.000 0
#> GSM587187 2 0.0336 0.966 0.000 0.992 0.008 0
#> GSM587188 2 0.0469 0.964 0.000 0.988 0.012 0
#> GSM587189 2 0.0469 0.964 0.000 0.988 0.012 0
#> GSM587190 2 0.3942 0.716 0.000 0.764 0.236 0
#> GSM587203 1 0.0000 0.937 1.000 0.000 0.000 0
#> GSM587204 1 0.0000 0.937 1.000 0.000 0.000 0
#> GSM587205 1 0.0000 0.937 1.000 0.000 0.000 0
#> GSM587206 1 0.0000 0.937 1.000 0.000 0.000 0
#> GSM587207 1 0.0000 0.937 1.000 0.000 0.000 0
#> GSM587208 1 0.0000 0.937 1.000 0.000 0.000 0
#> GSM587209 1 0.3764 0.751 0.784 0.000 0.216 0
#> GSM587210 1 0.4477 0.617 0.688 0.000 0.312 0
#> GSM587211 1 0.0000 0.937 1.000 0.000 0.000 0
#> GSM587212 1 0.2281 0.873 0.904 0.000 0.096 0
#> GSM587213 1 0.0000 0.937 1.000 0.000 0.000 0
#> GSM587214 1 0.0000 0.937 1.000 0.000 0.000 0
#> GSM587215 1 0.0000 0.937 1.000 0.000 0.000 0
#> GSM587216 1 0.0000 0.937 1.000 0.000 0.000 0
#> GSM587217 1 0.0000 0.937 1.000 0.000 0.000 0
#> GSM587191 3 0.0336 0.945 0.000 0.008 0.992 0
#> GSM587192 3 0.0336 0.945 0.000 0.008 0.992 0
#> GSM587193 3 0.2593 0.866 0.004 0.104 0.892 0
#> GSM587194 3 0.2593 0.866 0.004 0.104 0.892 0
#> GSM587195 3 0.0469 0.945 0.000 0.012 0.988 0
#> GSM587196 3 0.0469 0.945 0.000 0.012 0.988 0
#> GSM587197 3 0.0469 0.945 0.000 0.012 0.988 0
#> GSM587198 3 0.0188 0.944 0.000 0.004 0.996 0
#> GSM587199 3 0.0188 0.944 0.000 0.004 0.996 0
#> GSM587200 3 0.3610 0.716 0.200 0.000 0.800 0
#> GSM587201 3 0.3610 0.716 0.200 0.000 0.800 0
#> GSM587202 3 0.0188 0.944 0.000 0.004 0.996 0
#> GSM198767 1 0.0000 0.937 1.000 0.000 0.000 0
#> GSM198769 1 0.3764 0.751 0.784 0.000 0.216 0
#> GSM198772 1 0.0000 0.937 1.000 0.000 0.000 0
#> GSM198773 1 0.0000 0.937 1.000 0.000 0.000 0
#> GSM198776 1 0.0000 0.937 1.000 0.000 0.000 0
#> GSM198778 1 0.4477 0.617 0.688 0.000 0.312 0
#> GSM198780 1 0.2281 0.873 0.904 0.000 0.096 0
#> GSM198781 1 0.0000 0.937 1.000 0.000 0.000 0
#> GSM198765 3 0.0336 0.945 0.000 0.008 0.992 0
#> GSM198766 3 0.2593 0.866 0.004 0.104 0.892 0
#> GSM198768 3 0.0469 0.945 0.000 0.012 0.988 0
#> GSM198770 3 0.0469 0.945 0.000 0.012 0.988 0
#> GSM198771 3 0.0188 0.944 0.000 0.004 0.996 0
#> GSM198774 3 0.0336 0.945 0.000 0.008 0.992 0
#> GSM198775 3 0.2593 0.866 0.004 0.104 0.892 0
#> GSM198777 3 0.0469 0.945 0.000 0.012 0.988 0
#> GSM198779 3 0.0188 0.944 0.000 0.004 0.996 0
#> GSM587218 4 0.0000 1.000 0.000 0.000 0.000 1
#> GSM587219 4 0.0000 1.000 0.000 0.000 0.000 1
#> GSM587220 4 0.0000 1.000 0.000 0.000 0.000 1
#> GSM587221 4 0.0000 1.000 0.000 0.000 0.000 1
#> GSM587222 4 0.0000 1.000 0.000 0.000 0.000 1
#> GSM587223 4 0.0000 1.000 0.000 0.000 0.000 1
#> GSM587224 4 0.0000 1.000 0.000 0.000 0.000 1
#> GSM587225 4 0.0000 1.000 0.000 0.000 0.000 1
#> GSM587226 4 0.0000 1.000 0.000 0.000 0.000 1
#> GSM587227 4 0.0000 1.000 0.000 0.000 0.000 1
#> GSM587228 4 0.0000 1.000 0.000 0.000 0.000 1
#> GSM587229 4 0.0000 1.000 0.000 0.000 0.000 1
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM587155 2 0.1478 0.868 0.000 0.936 0.000 0 0.064
#> GSM587156 5 0.4294 0.867 0.000 0.468 0.000 0 0.532
#> GSM587157 2 0.1478 0.871 0.000 0.936 0.000 0 0.064
#> GSM587158 2 0.0162 0.941 0.000 0.996 0.000 0 0.004
#> GSM587159 2 0.0162 0.941 0.000 0.996 0.000 0 0.004
#> GSM587160 2 0.0000 0.941 0.000 1.000 0.000 0 0.000
#> GSM587161 2 0.0000 0.941 0.000 1.000 0.000 0 0.000
#> GSM587162 2 0.0000 0.941 0.000 1.000 0.000 0 0.000
#> GSM587163 2 0.0000 0.941 0.000 1.000 0.000 0 0.000
#> GSM587164 2 0.1478 0.868 0.000 0.936 0.000 0 0.064
#> GSM587165 2 0.0162 0.941 0.000 0.996 0.000 0 0.004
#> GSM587166 5 0.4294 0.867 0.000 0.468 0.000 0 0.532
#> GSM587167 5 0.4287 0.867 0.000 0.460 0.000 0 0.540
#> GSM587168 2 0.0000 0.941 0.000 1.000 0.000 0 0.000
#> GSM587169 2 0.0000 0.941 0.000 1.000 0.000 0 0.000
#> GSM587170 2 0.1478 0.868 0.000 0.936 0.000 0 0.064
#> GSM587171 2 0.0162 0.941 0.000 0.996 0.000 0 0.004
#> GSM587172 2 0.0162 0.941 0.000 0.996 0.000 0 0.004
#> GSM587173 2 0.0162 0.941 0.000 0.996 0.000 0 0.004
#> GSM587174 2 0.0000 0.941 0.000 1.000 0.000 0 0.000
#> GSM587175 2 0.1478 0.871 0.000 0.936 0.000 0 0.064
#> GSM587176 2 0.0510 0.927 0.000 0.984 0.000 0 0.016
#> GSM587177 2 0.0162 0.941 0.000 0.996 0.000 0 0.004
#> GSM587178 2 0.0162 0.941 0.000 0.996 0.000 0 0.004
#> GSM587179 2 0.0000 0.941 0.000 1.000 0.000 0 0.000
#> GSM587180 2 0.0000 0.941 0.000 1.000 0.000 0 0.000
#> GSM587181 2 0.0000 0.941 0.000 1.000 0.000 0 0.000
#> GSM587182 2 0.0000 0.941 0.000 1.000 0.000 0 0.000
#> GSM587183 2 0.0162 0.941 0.000 0.996 0.000 0 0.004
#> GSM587184 2 0.0162 0.941 0.000 0.996 0.000 0 0.004
#> GSM587185 2 0.0000 0.941 0.000 1.000 0.000 0 0.000
#> GSM587186 2 0.0162 0.941 0.000 0.996 0.000 0 0.004
#> GSM587187 2 0.2929 0.640 0.000 0.820 0.000 0 0.180
#> GSM587188 2 0.3837 0.319 0.000 0.692 0.000 0 0.308
#> GSM587189 2 0.3074 0.604 0.000 0.804 0.000 0 0.196
#> GSM587190 5 0.3636 0.704 0.000 0.272 0.000 0 0.728
#> GSM587203 1 0.0000 0.935 1.000 0.000 0.000 0 0.000
#> GSM587204 1 0.0000 0.935 1.000 0.000 0.000 0 0.000
#> GSM587205 1 0.0000 0.935 1.000 0.000 0.000 0 0.000
#> GSM587206 1 0.0000 0.935 1.000 0.000 0.000 0 0.000
#> GSM587207 1 0.0000 0.935 1.000 0.000 0.000 0 0.000
#> GSM587208 1 0.0000 0.935 1.000 0.000 0.000 0 0.000
#> GSM587209 1 0.3242 0.731 0.784 0.000 0.216 0 0.000
#> GSM587210 1 0.3857 0.583 0.688 0.000 0.312 0 0.000
#> GSM587211 1 0.0000 0.935 1.000 0.000 0.000 0 0.000
#> GSM587212 1 0.1965 0.864 0.904 0.000 0.096 0 0.000
#> GSM587213 1 0.0000 0.935 1.000 0.000 0.000 0 0.000
#> GSM587214 1 0.0000 0.935 1.000 0.000 0.000 0 0.000
#> GSM587215 1 0.0000 0.935 1.000 0.000 0.000 0 0.000
#> GSM587216 1 0.0000 0.935 1.000 0.000 0.000 0 0.000
#> GSM587217 1 0.0000 0.935 1.000 0.000 0.000 0 0.000
#> GSM587191 3 0.1043 0.862 0.000 0.000 0.960 0 0.040
#> GSM587192 3 0.1043 0.862 0.000 0.000 0.960 0 0.040
#> GSM587193 3 0.4211 0.624 0.004 0.000 0.636 0 0.360
#> GSM587194 3 0.4211 0.624 0.004 0.000 0.636 0 0.360
#> GSM587195 3 0.2377 0.840 0.000 0.000 0.872 0 0.128
#> GSM587196 3 0.2377 0.840 0.000 0.000 0.872 0 0.128
#> GSM587197 3 0.2377 0.840 0.000 0.000 0.872 0 0.128
#> GSM587198 3 0.0000 0.864 0.000 0.000 1.000 0 0.000
#> GSM587199 3 0.0000 0.864 0.000 0.000 1.000 0 0.000
#> GSM587200 3 0.3266 0.704 0.200 0.000 0.796 0 0.004
#> GSM587201 3 0.3266 0.704 0.200 0.000 0.796 0 0.004
#> GSM587202 3 0.0000 0.864 0.000 0.000 1.000 0 0.000
#> GSM198767 1 0.0000 0.935 1.000 0.000 0.000 0 0.000
#> GSM198769 1 0.3242 0.731 0.784 0.000 0.216 0 0.000
#> GSM198772 1 0.0000 0.935 1.000 0.000 0.000 0 0.000
#> GSM198773 1 0.0000 0.935 1.000 0.000 0.000 0 0.000
#> GSM198776 1 0.0000 0.935 1.000 0.000 0.000 0 0.000
#> GSM198778 1 0.3857 0.583 0.688 0.000 0.312 0 0.000
#> GSM198780 1 0.1965 0.864 0.904 0.000 0.096 0 0.000
#> GSM198781 1 0.0000 0.935 1.000 0.000 0.000 0 0.000
#> GSM198765 3 0.1043 0.862 0.000 0.000 0.960 0 0.040
#> GSM198766 3 0.4211 0.624 0.004 0.000 0.636 0 0.360
#> GSM198768 3 0.2377 0.840 0.000 0.000 0.872 0 0.128
#> GSM198770 3 0.2377 0.840 0.000 0.000 0.872 0 0.128
#> GSM198771 3 0.0000 0.864 0.000 0.000 1.000 0 0.000
#> GSM198774 3 0.1043 0.862 0.000 0.000 0.960 0 0.040
#> GSM198775 3 0.4211 0.624 0.004 0.000 0.636 0 0.360
#> GSM198777 3 0.2377 0.840 0.000 0.000 0.872 0 0.128
#> GSM198779 3 0.0000 0.864 0.000 0.000 1.000 0 0.000
#> GSM587218 4 0.0000 1.000 0.000 0.000 0.000 1 0.000
#> GSM587219 4 0.0000 1.000 0.000 0.000 0.000 1 0.000
#> GSM587220 4 0.0000 1.000 0.000 0.000 0.000 1 0.000
#> GSM587221 4 0.0000 1.000 0.000 0.000 0.000 1 0.000
#> GSM587222 4 0.0000 1.000 0.000 0.000 0.000 1 0.000
#> GSM587223 4 0.0000 1.000 0.000 0.000 0.000 1 0.000
#> GSM587224 4 0.0000 1.000 0.000 0.000 0.000 1 0.000
#> GSM587225 4 0.0000 1.000 0.000 0.000 0.000 1 0.000
#> GSM587226 4 0.0000 1.000 0.000 0.000 0.000 1 0.000
#> GSM587227 4 0.0000 1.000 0.000 0.000 0.000 1 0.000
#> GSM587228 4 0.0000 1.000 0.000 0.000 0.000 1 0.000
#> GSM587229 4 0.0000 1.000 0.000 0.000 0.000 1 0.000
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM587155 2 0.2527 0.7499 0.000 0.832 0.000 0 0.168 0.000
#> GSM587156 5 0.3330 0.8842 0.000 0.284 0.000 0 0.716 0.000
#> GSM587157 2 0.2389 0.7967 0.000 0.864 0.000 0 0.128 0.008
#> GSM587158 2 0.0146 0.9227 0.000 0.996 0.000 0 0.004 0.000
#> GSM587159 2 0.0146 0.9227 0.000 0.996 0.000 0 0.004 0.000
#> GSM587160 2 0.0146 0.9224 0.000 0.996 0.000 0 0.004 0.000
#> GSM587161 2 0.0146 0.9224 0.000 0.996 0.000 0 0.004 0.000
#> GSM587162 2 0.0363 0.9183 0.000 0.988 0.000 0 0.012 0.000
#> GSM587163 2 0.0146 0.9224 0.000 0.996 0.000 0 0.004 0.000
#> GSM587164 2 0.2527 0.7499 0.000 0.832 0.000 0 0.168 0.000
#> GSM587165 2 0.0146 0.9227 0.000 0.996 0.000 0 0.004 0.000
#> GSM587166 5 0.3330 0.8842 0.000 0.284 0.000 0 0.716 0.000
#> GSM587167 5 0.3534 0.8821 0.000 0.276 0.008 0 0.716 0.000
#> GSM587168 2 0.0146 0.9223 0.000 0.996 0.000 0 0.004 0.000
#> GSM587169 2 0.0146 0.9224 0.000 0.996 0.000 0 0.004 0.000
#> GSM587170 2 0.2527 0.7499 0.000 0.832 0.000 0 0.168 0.000
#> GSM587171 2 0.0146 0.9227 0.000 0.996 0.000 0 0.004 0.000
#> GSM587172 2 0.0146 0.9227 0.000 0.996 0.000 0 0.004 0.000
#> GSM587173 2 0.0260 0.9208 0.000 0.992 0.000 0 0.008 0.000
#> GSM587174 2 0.0146 0.9224 0.000 0.996 0.000 0 0.004 0.000
#> GSM587175 2 0.2389 0.7967 0.000 0.864 0.000 0 0.128 0.008
#> GSM587176 2 0.1501 0.8631 0.000 0.924 0.000 0 0.076 0.000
#> GSM587177 2 0.0260 0.9208 0.000 0.992 0.000 0 0.008 0.000
#> GSM587178 2 0.0146 0.9227 0.000 0.996 0.000 0 0.004 0.000
#> GSM587179 2 0.0146 0.9224 0.000 0.996 0.000 0 0.004 0.000
#> GSM587180 2 0.0000 0.9228 0.000 1.000 0.000 0 0.000 0.000
#> GSM587181 2 0.0146 0.9224 0.000 0.996 0.000 0 0.004 0.000
#> GSM587182 2 0.0000 0.9228 0.000 1.000 0.000 0 0.000 0.000
#> GSM587183 2 0.0146 0.9227 0.000 0.996 0.000 0 0.004 0.000
#> GSM587184 2 0.0146 0.9227 0.000 0.996 0.000 0 0.004 0.000
#> GSM587185 2 0.0146 0.9224 0.000 0.996 0.000 0 0.004 0.000
#> GSM587186 2 0.0260 0.9208 0.000 0.992 0.000 0 0.008 0.000
#> GSM587187 2 0.2664 0.7104 0.000 0.816 0.000 0 0.184 0.000
#> GSM587188 2 0.5640 0.0879 0.000 0.528 0.000 0 0.280 0.192
#> GSM587189 2 0.2980 0.6856 0.000 0.800 0.008 0 0.192 0.000
#> GSM587190 5 0.1918 0.6225 0.000 0.088 0.008 0 0.904 0.000
#> GSM587203 1 0.0000 0.9095 1.000 0.000 0.000 0 0.000 0.000
#> GSM587204 1 0.0000 0.9095 1.000 0.000 0.000 0 0.000 0.000
#> GSM587205 1 0.0000 0.9095 1.000 0.000 0.000 0 0.000 0.000
#> GSM587206 1 0.0000 0.9095 1.000 0.000 0.000 0 0.000 0.000
#> GSM587207 1 0.0000 0.9095 1.000 0.000 0.000 0 0.000 0.000
#> GSM587208 1 0.0000 0.9095 1.000 0.000 0.000 0 0.000 0.000
#> GSM587209 1 0.4781 0.4855 0.624 0.000 0.080 0 0.000 0.296
#> GSM587210 1 0.5059 0.3112 0.528 0.000 0.080 0 0.000 0.392
#> GSM587211 1 0.0000 0.9095 1.000 0.000 0.000 0 0.000 0.000
#> GSM587212 1 0.1765 0.8331 0.904 0.000 0.000 0 0.000 0.096
#> GSM587213 1 0.0000 0.9095 1.000 0.000 0.000 0 0.000 0.000
#> GSM587214 1 0.0000 0.9095 1.000 0.000 0.000 0 0.000 0.000
#> GSM587215 1 0.0000 0.9095 1.000 0.000 0.000 0 0.000 0.000
#> GSM587216 1 0.0260 0.9049 0.992 0.000 0.000 0 0.000 0.008
#> GSM587217 1 0.0000 0.9095 1.000 0.000 0.000 0 0.000 0.000
#> GSM587191 3 0.2362 0.5059 0.000 0.000 0.860 0 0.004 0.136
#> GSM587192 3 0.2362 0.5059 0.000 0.000 0.860 0 0.004 0.136
#> GSM587193 3 0.5962 0.0371 0.000 0.000 0.436 0 0.328 0.236
#> GSM587194 3 0.5962 0.0371 0.000 0.000 0.436 0 0.328 0.236
#> GSM587195 3 0.2340 0.5205 0.000 0.000 0.852 0 0.000 0.148
#> GSM587196 3 0.2340 0.5205 0.000 0.000 0.852 0 0.000 0.148
#> GSM587197 3 0.2340 0.5205 0.000 0.000 0.852 0 0.000 0.148
#> GSM587198 3 0.2340 0.4094 0.000 0.000 0.852 0 0.000 0.148
#> GSM587199 3 0.2340 0.4094 0.000 0.000 0.852 0 0.000 0.148
#> GSM587200 6 0.4589 1.0000 0.036 0.000 0.460 0 0.000 0.504
#> GSM587201 6 0.4589 1.0000 0.036 0.000 0.460 0 0.000 0.504
#> GSM587202 3 0.2340 0.4094 0.000 0.000 0.852 0 0.000 0.148
#> GSM198767 1 0.0000 0.9095 1.000 0.000 0.000 0 0.000 0.000
#> GSM198769 1 0.4781 0.4855 0.624 0.000 0.080 0 0.000 0.296
#> GSM198772 1 0.0000 0.9095 1.000 0.000 0.000 0 0.000 0.000
#> GSM198773 1 0.0000 0.9095 1.000 0.000 0.000 0 0.000 0.000
#> GSM198776 1 0.0000 0.9095 1.000 0.000 0.000 0 0.000 0.000
#> GSM198778 1 0.5059 0.3112 0.528 0.000 0.080 0 0.000 0.392
#> GSM198780 1 0.1765 0.8331 0.904 0.000 0.000 0 0.000 0.096
#> GSM198781 1 0.0000 0.9095 1.000 0.000 0.000 0 0.000 0.000
#> GSM198765 3 0.2362 0.5059 0.000 0.000 0.860 0 0.004 0.136
#> GSM198766 3 0.5962 0.0371 0.000 0.000 0.436 0 0.328 0.236
#> GSM198768 3 0.2340 0.5205 0.000 0.000 0.852 0 0.000 0.148
#> GSM198770 3 0.2340 0.5205 0.000 0.000 0.852 0 0.000 0.148
#> GSM198771 3 0.2340 0.4094 0.000 0.000 0.852 0 0.000 0.148
#> GSM198774 3 0.2362 0.5059 0.000 0.000 0.860 0 0.004 0.136
#> GSM198775 3 0.5962 0.0371 0.000 0.000 0.436 0 0.328 0.236
#> GSM198777 3 0.2340 0.5205 0.000 0.000 0.852 0 0.000 0.148
#> GSM198779 3 0.2340 0.4094 0.000 0.000 0.852 0 0.000 0.148
#> GSM587218 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587219 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587220 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587221 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587222 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587223 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587224 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587225 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587226 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587227 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587228 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587229 4 0.0000 1.0000 0.000 0.000 0.000 1 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 specimen(p) k
#> CV:hclust 90 3.03e-17 2
#> CV:hclust 90 4.27e-32 3
#> CV:hclust 92 4.04e-48 4
#> CV:hclust 91 6.79e-45 5
#> CV:hclust 78 2.81e-36 6
If matrix rows can be associated to genes, consider to use functional_enrichment(res,
...) to perform function enrichment for the signature genes. See this vignette for more detailed explanations.
The object with results only for a single top-value method and a single partition method can be extracted as:
res = res_list["CV", "kmeans"]
# you can also extract it by
# res = res_list["CV:kmeans"]
A summary of res and all the functions that can be applied to it:
res
#> A 'ConsensusPartition' object with k = 2, 3, 4, 5, 6.
#> On a matrix with 51941 rows and 92 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 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 0.719 0.904 0.933 0.4695 0.500 0.500
#> 3 3 0.670 0.882 0.873 0.3473 0.793 0.605
#> 4 4 0.833 0.934 0.882 0.1199 0.934 0.806
#> 5 5 0.759 0.858 0.837 0.0706 1.000 1.000
#> 6 6 0.789 0.641 0.767 0.0436 0.984 0.942
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
#> GSM587155 2 0.000 0.932 0.000 1.000
#> GSM587156 2 0.000 0.932 0.000 1.000
#> GSM587157 2 0.000 0.932 0.000 1.000
#> GSM587158 2 0.000 0.932 0.000 1.000
#> GSM587159 2 0.000 0.932 0.000 1.000
#> GSM587160 2 0.000 0.932 0.000 1.000
#> GSM587161 2 0.000 0.932 0.000 1.000
#> GSM587162 2 0.000 0.932 0.000 1.000
#> GSM587163 2 0.000 0.932 0.000 1.000
#> GSM587164 2 0.000 0.932 0.000 1.000
#> GSM587165 2 0.000 0.932 0.000 1.000
#> GSM587166 2 0.000 0.932 0.000 1.000
#> GSM587167 2 0.000 0.932 0.000 1.000
#> GSM587168 2 0.000 0.932 0.000 1.000
#> GSM587169 2 0.000 0.932 0.000 1.000
#> GSM587170 2 0.000 0.932 0.000 1.000
#> GSM587171 2 0.000 0.932 0.000 1.000
#> GSM587172 2 0.000 0.932 0.000 1.000
#> GSM587173 2 0.000 0.932 0.000 1.000
#> GSM587174 2 0.000 0.932 0.000 1.000
#> GSM587175 2 0.000 0.932 0.000 1.000
#> GSM587176 2 0.000 0.932 0.000 1.000
#> GSM587177 2 0.000 0.932 0.000 1.000
#> GSM587178 2 0.000 0.932 0.000 1.000
#> GSM587179 2 0.000 0.932 0.000 1.000
#> GSM587180 2 0.000 0.932 0.000 1.000
#> GSM587181 2 0.000 0.932 0.000 1.000
#> GSM587182 2 0.000 0.932 0.000 1.000
#> GSM587183 2 0.000 0.932 0.000 1.000
#> GSM587184 2 0.000 0.932 0.000 1.000
#> GSM587185 2 0.000 0.932 0.000 1.000
#> GSM587186 2 0.000 0.932 0.000 1.000
#> GSM587187 2 0.000 0.932 0.000 1.000
#> GSM587188 2 0.000 0.932 0.000 1.000
#> GSM587189 2 0.000 0.932 0.000 1.000
#> GSM587190 2 0.000 0.932 0.000 1.000
#> GSM587203 1 0.456 0.949 0.904 0.096
#> GSM587204 1 0.456 0.949 0.904 0.096
#> GSM587205 1 0.456 0.949 0.904 0.096
#> GSM587206 1 0.456 0.949 0.904 0.096
#> GSM587207 1 0.456 0.949 0.904 0.096
#> GSM587208 1 0.456 0.949 0.904 0.096
#> GSM587209 1 0.456 0.949 0.904 0.096
#> GSM587210 1 0.456 0.949 0.904 0.096
#> GSM587211 1 0.456 0.949 0.904 0.096
#> GSM587212 1 0.456 0.949 0.904 0.096
#> GSM587213 1 0.456 0.949 0.904 0.096
#> GSM587214 1 0.456 0.949 0.904 0.096
#> GSM587215 1 0.456 0.949 0.904 0.096
#> GSM587216 1 0.456 0.949 0.904 0.096
#> GSM587217 1 0.456 0.949 0.904 0.096
#> GSM587191 2 0.714 0.799 0.196 0.804
#> GSM587192 1 0.921 0.565 0.664 0.336
#> GSM587193 1 0.456 0.949 0.904 0.096
#> GSM587194 2 0.697 0.806 0.188 0.812
#> GSM587195 2 0.714 0.799 0.196 0.804
#> GSM587196 2 0.714 0.799 0.196 0.804
#> GSM587197 2 0.714 0.799 0.196 0.804
#> GSM587198 2 0.714 0.799 0.196 0.804
#> GSM587199 2 0.680 0.813 0.180 0.820
#> GSM587200 1 0.456 0.949 0.904 0.096
#> GSM587201 1 0.456 0.949 0.904 0.096
#> GSM587202 2 0.714 0.799 0.196 0.804
#> GSM198767 1 0.456 0.949 0.904 0.096
#> GSM198769 1 0.456 0.949 0.904 0.096
#> GSM198772 1 0.456 0.949 0.904 0.096
#> GSM198773 1 0.456 0.949 0.904 0.096
#> GSM198776 1 0.456 0.949 0.904 0.096
#> GSM198778 1 0.456 0.949 0.904 0.096
#> GSM198780 1 0.456 0.949 0.904 0.096
#> GSM198781 1 0.456 0.949 0.904 0.096
#> GSM198765 2 0.714 0.799 0.196 0.804
#> GSM198766 1 0.456 0.949 0.904 0.096
#> GSM198768 2 0.714 0.799 0.196 0.804
#> GSM198770 2 0.714 0.799 0.196 0.804
#> GSM198771 2 0.714 0.799 0.196 0.804
#> GSM198774 1 0.921 0.565 0.664 0.336
#> GSM198775 2 0.697 0.806 0.188 0.812
#> GSM198777 2 0.714 0.799 0.196 0.804
#> GSM198779 2 0.680 0.813 0.180 0.820
#> GSM587218 1 0.000 0.906 1.000 0.000
#> GSM587219 1 0.000 0.906 1.000 0.000
#> GSM587220 1 0.000 0.906 1.000 0.000
#> GSM587221 1 0.000 0.906 1.000 0.000
#> GSM587222 1 0.000 0.906 1.000 0.000
#> GSM587223 1 0.000 0.906 1.000 0.000
#> GSM587224 1 0.000 0.906 1.000 0.000
#> GSM587225 1 0.000 0.906 1.000 0.000
#> GSM587226 1 0.000 0.906 1.000 0.000
#> GSM587227 1 0.000 0.906 1.000 0.000
#> GSM587228 1 0.000 0.906 1.000 0.000
#> GSM587229 1 0.000 0.906 1.000 0.000
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM587155 2 0.0000 0.997 0.000 1.000 0.000
#> GSM587156 2 0.0000 0.997 0.000 1.000 0.000
#> GSM587157 2 0.0000 0.997 0.000 1.000 0.000
#> GSM587158 2 0.0000 0.997 0.000 1.000 0.000
#> GSM587159 2 0.0000 0.997 0.000 1.000 0.000
#> GSM587160 2 0.0000 0.997 0.000 1.000 0.000
#> GSM587161 2 0.0000 0.997 0.000 1.000 0.000
#> GSM587162 2 0.0000 0.997 0.000 1.000 0.000
#> GSM587163 2 0.0000 0.997 0.000 1.000 0.000
#> GSM587164 2 0.0000 0.997 0.000 1.000 0.000
#> GSM587165 2 0.0424 0.995 0.000 0.992 0.008
#> GSM587166 2 0.0237 0.994 0.000 0.996 0.004
#> GSM587167 2 0.0000 0.997 0.000 1.000 0.000
#> GSM587168 2 0.0424 0.995 0.000 0.992 0.008
#> GSM587169 2 0.0000 0.997 0.000 1.000 0.000
#> GSM587170 2 0.0000 0.997 0.000 1.000 0.000
#> GSM587171 2 0.0000 0.997 0.000 1.000 0.000
#> GSM587172 2 0.0000 0.997 0.000 1.000 0.000
#> GSM587173 2 0.0424 0.995 0.000 0.992 0.008
#> GSM587174 2 0.0237 0.996 0.000 0.996 0.004
#> GSM587175 2 0.0000 0.997 0.000 1.000 0.000
#> GSM587176 2 0.0000 0.997 0.000 1.000 0.000
#> GSM587177 2 0.0424 0.995 0.000 0.992 0.008
#> GSM587178 2 0.0424 0.995 0.000 0.992 0.008
#> GSM587179 2 0.0000 0.997 0.000 1.000 0.000
#> GSM587180 2 0.0424 0.995 0.000 0.992 0.008
#> GSM587181 2 0.0237 0.996 0.000 0.996 0.004
#> GSM587182 2 0.0424 0.995 0.000 0.992 0.008
#> GSM587183 2 0.0424 0.995 0.000 0.992 0.008
#> GSM587184 2 0.0000 0.997 0.000 1.000 0.000
#> GSM587185 2 0.0000 0.997 0.000 1.000 0.000
#> GSM587186 2 0.0424 0.995 0.000 0.992 0.008
#> GSM587187 2 0.0424 0.995 0.000 0.992 0.008
#> GSM587188 2 0.0424 0.995 0.000 0.992 0.008
#> GSM587189 2 0.0424 0.995 0.000 0.992 0.008
#> GSM587190 3 0.5835 0.737 0.000 0.340 0.660
#> GSM587203 1 0.0829 0.850 0.984 0.012 0.004
#> GSM587204 1 0.0829 0.850 0.984 0.012 0.004
#> GSM587205 1 0.0829 0.850 0.984 0.012 0.004
#> GSM587206 1 0.0829 0.850 0.984 0.012 0.004
#> GSM587207 1 0.0829 0.850 0.984 0.012 0.004
#> GSM587208 1 0.0829 0.850 0.984 0.012 0.004
#> GSM587209 1 0.1015 0.847 0.980 0.012 0.008
#> GSM587210 1 0.2116 0.836 0.948 0.012 0.040
#> GSM587211 1 0.1182 0.845 0.976 0.012 0.012
#> GSM587212 1 0.1999 0.838 0.952 0.012 0.036
#> GSM587213 1 0.0592 0.850 0.988 0.012 0.000
#> GSM587214 1 0.0592 0.850 0.988 0.012 0.000
#> GSM587215 1 0.1182 0.845 0.976 0.012 0.012
#> GSM587216 1 0.1182 0.845 0.976 0.012 0.012
#> GSM587217 1 0.0592 0.850 0.988 0.012 0.000
#> GSM587191 3 0.7265 0.897 0.076 0.240 0.684
#> GSM587192 3 0.7339 0.767 0.224 0.088 0.688
#> GSM587193 3 0.6143 0.657 0.304 0.012 0.684
#> GSM587194 3 0.7306 0.897 0.080 0.236 0.684
#> GSM587195 3 0.7381 0.897 0.080 0.244 0.676
#> GSM587196 3 0.7381 0.897 0.080 0.244 0.676
#> GSM587197 3 0.7381 0.897 0.080 0.244 0.676
#> GSM587198 3 0.7344 0.898 0.080 0.240 0.680
#> GSM587199 3 0.7112 0.877 0.060 0.260 0.680
#> GSM587200 3 0.6143 0.657 0.304 0.012 0.684
#> GSM587201 3 0.6143 0.657 0.304 0.012 0.684
#> GSM587202 3 0.7344 0.898 0.080 0.240 0.680
#> GSM198767 1 0.0829 0.850 0.984 0.012 0.004
#> GSM198769 1 0.1015 0.847 0.980 0.012 0.008
#> GSM198772 1 0.1182 0.845 0.976 0.012 0.012
#> GSM198773 1 0.0592 0.850 0.988 0.012 0.000
#> GSM198776 1 0.0829 0.850 0.984 0.012 0.004
#> GSM198778 1 0.2116 0.836 0.948 0.012 0.040
#> GSM198780 1 0.1999 0.838 0.952 0.012 0.036
#> GSM198781 1 0.0592 0.850 0.988 0.012 0.000
#> GSM198765 3 0.7265 0.897 0.076 0.240 0.684
#> GSM198766 3 0.6143 0.657 0.304 0.012 0.684
#> GSM198768 3 0.7381 0.897 0.080 0.244 0.676
#> GSM198770 3 0.7381 0.897 0.080 0.244 0.676
#> GSM198771 3 0.7344 0.898 0.080 0.240 0.680
#> GSM198774 3 0.7339 0.767 0.224 0.088 0.688
#> GSM198775 3 0.7306 0.897 0.080 0.236 0.684
#> GSM198777 3 0.7381 0.897 0.080 0.244 0.676
#> GSM198779 3 0.7112 0.877 0.060 0.260 0.680
#> GSM587218 1 0.6062 0.712 0.616 0.000 0.384
#> GSM587219 1 0.6079 0.712 0.612 0.000 0.388
#> GSM587220 1 0.6079 0.712 0.612 0.000 0.388
#> GSM587221 1 0.6079 0.712 0.612 0.000 0.388
#> GSM587222 1 0.6079 0.712 0.612 0.000 0.388
#> GSM587223 1 0.6062 0.712 0.616 0.000 0.384
#> GSM587224 1 0.6079 0.712 0.612 0.000 0.388
#> GSM587225 1 0.6111 0.712 0.604 0.000 0.396
#> GSM587226 1 0.6079 0.712 0.612 0.000 0.388
#> GSM587227 1 0.6095 0.712 0.608 0.000 0.392
#> GSM587228 1 0.6095 0.712 0.608 0.000 0.392
#> GSM587229 1 0.6095 0.712 0.608 0.000 0.392
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM587155 2 0.2921 0.888 0.000 0.860 0.000 0.140
#> GSM587156 2 0.3356 0.863 0.000 0.824 0.000 0.176
#> GSM587157 2 0.2921 0.888 0.000 0.860 0.000 0.140
#> GSM587158 2 0.0000 0.940 0.000 1.000 0.000 0.000
#> GSM587159 2 0.0000 0.940 0.000 1.000 0.000 0.000
#> GSM587160 2 0.0000 0.940 0.000 1.000 0.000 0.000
#> GSM587161 2 0.1389 0.929 0.000 0.952 0.000 0.048
#> GSM587162 2 0.0707 0.937 0.000 0.980 0.000 0.020
#> GSM587163 2 0.0000 0.940 0.000 1.000 0.000 0.000
#> GSM587164 2 0.2921 0.888 0.000 0.860 0.000 0.140
#> GSM587165 2 0.2197 0.928 0.000 0.916 0.004 0.080
#> GSM587166 2 0.3539 0.860 0.000 0.820 0.004 0.176
#> GSM587167 2 0.3074 0.880 0.000 0.848 0.000 0.152
#> GSM587168 2 0.2197 0.928 0.000 0.916 0.004 0.080
#> GSM587169 2 0.0000 0.940 0.000 1.000 0.000 0.000
#> GSM587170 2 0.2921 0.888 0.000 0.860 0.000 0.140
#> GSM587171 2 0.0000 0.940 0.000 1.000 0.000 0.000
#> GSM587172 2 0.0000 0.940 0.000 1.000 0.000 0.000
#> GSM587173 2 0.2197 0.928 0.000 0.916 0.004 0.080
#> GSM587174 2 0.0817 0.939 0.000 0.976 0.000 0.024
#> GSM587175 2 0.2921 0.888 0.000 0.860 0.000 0.140
#> GSM587176 2 0.0921 0.936 0.000 0.972 0.000 0.028
#> GSM587177 2 0.2197 0.928 0.000 0.916 0.004 0.080
#> GSM587178 2 0.2011 0.929 0.000 0.920 0.000 0.080
#> GSM587179 2 0.0469 0.939 0.000 0.988 0.000 0.012
#> GSM587180 2 0.2125 0.929 0.000 0.920 0.004 0.076
#> GSM587181 2 0.0817 0.939 0.000 0.976 0.000 0.024
#> GSM587182 2 0.1940 0.930 0.000 0.924 0.000 0.076
#> GSM587183 2 0.2197 0.928 0.000 0.916 0.004 0.080
#> GSM587184 2 0.0188 0.940 0.000 0.996 0.000 0.004
#> GSM587185 2 0.0469 0.939 0.000 0.988 0.000 0.012
#> GSM587186 2 0.2197 0.928 0.000 0.916 0.004 0.080
#> GSM587187 2 0.2266 0.926 0.000 0.912 0.004 0.084
#> GSM587188 2 0.2401 0.923 0.000 0.904 0.004 0.092
#> GSM587189 2 0.2654 0.924 0.000 0.888 0.004 0.108
#> GSM587190 3 0.5222 0.836 0.000 0.112 0.756 0.132
#> GSM587203 1 0.1697 0.948 0.952 0.004 0.016 0.028
#> GSM587204 1 0.1697 0.948 0.952 0.004 0.016 0.028
#> GSM587205 1 0.1697 0.948 0.952 0.004 0.016 0.028
#> GSM587206 1 0.1697 0.948 0.952 0.004 0.016 0.028
#> GSM587207 1 0.1697 0.948 0.952 0.004 0.016 0.028
#> GSM587208 1 0.1697 0.948 0.952 0.004 0.016 0.028
#> GSM587209 1 0.0524 0.956 0.988 0.004 0.008 0.000
#> GSM587210 1 0.2053 0.883 0.924 0.004 0.072 0.000
#> GSM587211 1 0.0524 0.956 0.988 0.004 0.008 0.000
#> GSM587212 1 0.1398 0.927 0.956 0.004 0.040 0.000
#> GSM587213 1 0.0844 0.956 0.980 0.004 0.004 0.012
#> GSM587214 1 0.0844 0.956 0.980 0.004 0.004 0.012
#> GSM587215 1 0.0524 0.956 0.988 0.004 0.008 0.000
#> GSM587216 1 0.0524 0.956 0.988 0.004 0.008 0.000
#> GSM587217 1 0.0524 0.956 0.988 0.004 0.008 0.000
#> GSM587191 3 0.4254 0.932 0.036 0.056 0.848 0.060
#> GSM587192 3 0.3938 0.904 0.080 0.008 0.852 0.060
#> GSM587193 3 0.4457 0.883 0.108 0.004 0.816 0.072
#> GSM587194 3 0.4918 0.922 0.040 0.064 0.812 0.084
#> GSM587195 3 0.3720 0.939 0.032 0.064 0.872 0.032
#> GSM587196 3 0.3720 0.939 0.032 0.064 0.872 0.032
#> GSM587197 3 0.3813 0.940 0.032 0.064 0.868 0.036
#> GSM587198 3 0.2722 0.941 0.032 0.064 0.904 0.000
#> GSM587199 3 0.2635 0.935 0.020 0.076 0.904 0.000
#> GSM587200 3 0.3109 0.902 0.100 0.004 0.880 0.016
#> GSM587201 3 0.3109 0.902 0.100 0.004 0.880 0.016
#> GSM587202 3 0.2722 0.941 0.032 0.064 0.904 0.000
#> GSM198767 1 0.1697 0.948 0.952 0.004 0.016 0.028
#> GSM198769 1 0.0524 0.956 0.988 0.004 0.008 0.000
#> GSM198772 1 0.0524 0.956 0.988 0.004 0.008 0.000
#> GSM198773 1 0.0844 0.956 0.980 0.004 0.004 0.012
#> GSM198776 1 0.1697 0.948 0.952 0.004 0.016 0.028
#> GSM198778 1 0.2053 0.883 0.924 0.004 0.072 0.000
#> GSM198780 1 0.1398 0.927 0.956 0.004 0.040 0.000
#> GSM198781 1 0.0844 0.956 0.980 0.004 0.004 0.012
#> GSM198765 3 0.4254 0.932 0.036 0.056 0.848 0.060
#> GSM198766 3 0.4457 0.883 0.108 0.004 0.816 0.072
#> GSM198768 3 0.3720 0.939 0.032 0.064 0.872 0.032
#> GSM198770 3 0.3813 0.940 0.032 0.064 0.868 0.036
#> GSM198771 3 0.2722 0.941 0.032 0.064 0.904 0.000
#> GSM198774 3 0.3938 0.904 0.080 0.008 0.852 0.060
#> GSM198775 3 0.4918 0.922 0.040 0.064 0.812 0.084
#> GSM198777 3 0.3720 0.939 0.032 0.064 0.872 0.032
#> GSM198779 3 0.2635 0.935 0.020 0.076 0.904 0.000
#> GSM587218 4 0.5548 0.977 0.340 0.000 0.032 0.628
#> GSM587219 4 0.5368 0.979 0.340 0.000 0.024 0.636
#> GSM587220 4 0.5368 0.979 0.340 0.000 0.024 0.636
#> GSM587221 4 0.5565 0.979 0.344 0.000 0.032 0.624
#> GSM587222 4 0.5565 0.979 0.344 0.000 0.032 0.624
#> GSM587223 4 0.5548 0.977 0.340 0.000 0.032 0.628
#> GSM587224 4 0.5730 0.977 0.344 0.000 0.040 0.616
#> GSM587225 4 0.6382 0.966 0.340 0.000 0.080 0.580
#> GSM587226 4 0.5565 0.979 0.344 0.000 0.032 0.624
#> GSM587227 4 0.6265 0.966 0.340 0.000 0.072 0.588
#> GSM587228 4 0.6382 0.966 0.340 0.000 0.080 0.580
#> GSM587229 4 0.6265 0.966 0.340 0.000 0.072 0.588
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM587155 2 0.3885 0.758 0.000 0.724 0.000 0.008 NA
#> GSM587156 2 0.4553 0.657 0.000 0.604 0.008 0.004 NA
#> GSM587157 2 0.3861 0.749 0.000 0.712 0.000 0.004 NA
#> GSM587158 2 0.0579 0.859 0.000 0.984 0.000 0.008 NA
#> GSM587159 2 0.0290 0.860 0.000 0.992 0.000 0.000 NA
#> GSM587160 2 0.0898 0.857 0.000 0.972 0.000 0.008 NA
#> GSM587161 2 0.2411 0.835 0.000 0.884 0.000 0.008 NA
#> GSM587162 2 0.1956 0.846 0.000 0.916 0.000 0.008 NA
#> GSM587163 2 0.0992 0.856 0.000 0.968 0.000 0.008 NA
#> GSM587164 2 0.3928 0.741 0.000 0.700 0.004 0.000 NA
#> GSM587165 2 0.3495 0.829 0.000 0.812 0.000 0.028 NA
#> GSM587166 2 0.4655 0.651 0.000 0.600 0.012 0.004 NA
#> GSM587167 2 0.4353 0.711 0.000 0.660 0.004 0.008 NA
#> GSM587168 2 0.3536 0.830 0.000 0.812 0.000 0.032 NA
#> GSM587169 2 0.1012 0.858 0.000 0.968 0.000 0.012 NA
#> GSM587170 2 0.3928 0.741 0.000 0.700 0.004 0.000 NA
#> GSM587171 2 0.0290 0.860 0.000 0.992 0.000 0.000 NA
#> GSM587172 2 0.0290 0.860 0.000 0.992 0.000 0.000 NA
#> GSM587173 2 0.3655 0.828 0.000 0.804 0.000 0.036 NA
#> GSM587174 2 0.1809 0.856 0.000 0.928 0.000 0.012 NA
#> GSM587175 2 0.3838 0.752 0.000 0.716 0.000 0.004 NA
#> GSM587176 2 0.1956 0.847 0.000 0.916 0.000 0.008 NA
#> GSM587177 2 0.3495 0.829 0.000 0.812 0.000 0.028 NA
#> GSM587178 2 0.3278 0.832 0.000 0.824 0.000 0.020 NA
#> GSM587179 2 0.1251 0.855 0.000 0.956 0.000 0.008 NA
#> GSM587180 2 0.3536 0.832 0.000 0.812 0.000 0.032 NA
#> GSM587181 2 0.1809 0.856 0.000 0.928 0.000 0.012 NA
#> GSM587182 2 0.3368 0.834 0.000 0.820 0.000 0.024 NA
#> GSM587183 2 0.3495 0.829 0.000 0.812 0.000 0.028 NA
#> GSM587184 2 0.0703 0.859 0.000 0.976 0.000 0.000 NA
#> GSM587185 2 0.1251 0.855 0.000 0.956 0.000 0.008 NA
#> GSM587186 2 0.3655 0.828 0.000 0.804 0.000 0.036 NA
#> GSM587187 2 0.3914 0.822 0.000 0.788 0.000 0.048 NA
#> GSM587188 2 0.4220 0.816 0.000 0.768 0.004 0.048 NA
#> GSM587189 2 0.4549 0.811 0.000 0.728 0.004 0.048 NA
#> GSM587190 3 0.6002 0.712 0.000 0.064 0.596 0.036 NA
#> GSM587203 1 0.2228 0.898 0.912 0.000 0.000 0.040 NA
#> GSM587204 1 0.2313 0.898 0.912 0.000 0.004 0.040 NA
#> GSM587205 1 0.2228 0.898 0.912 0.000 0.000 0.040 NA
#> GSM587206 1 0.2228 0.898 0.912 0.000 0.000 0.040 NA
#> GSM587207 1 0.2228 0.898 0.912 0.000 0.000 0.040 NA
#> GSM587208 1 0.2228 0.898 0.912 0.000 0.000 0.040 NA
#> GSM587209 1 0.1571 0.910 0.936 0.000 0.004 0.000 NA
#> GSM587210 1 0.3289 0.846 0.844 0.000 0.048 0.000 NA
#> GSM587211 1 0.1831 0.905 0.920 0.000 0.004 0.000 NA
#> GSM587212 1 0.2597 0.883 0.884 0.000 0.024 0.000 NA
#> GSM587213 1 0.0566 0.913 0.984 0.000 0.000 0.012 NA
#> GSM587214 1 0.0566 0.913 0.984 0.000 0.000 0.012 NA
#> GSM587215 1 0.1571 0.911 0.936 0.000 0.004 0.000 NA
#> GSM587216 1 0.2011 0.899 0.908 0.000 0.004 0.000 NA
#> GSM587217 1 0.1270 0.912 0.948 0.000 0.000 0.000 NA
#> GSM587191 3 0.5347 0.817 0.012 0.012 0.700 0.064 NA
#> GSM587192 3 0.5439 0.808 0.028 0.000 0.680 0.064 NA
#> GSM587193 3 0.5608 0.775 0.052 0.000 0.660 0.040 NA
#> GSM587194 3 0.5677 0.783 0.016 0.020 0.652 0.044 NA
#> GSM587195 3 0.3765 0.847 0.016 0.020 0.848 0.036 NA
#> GSM587196 3 0.3765 0.847 0.016 0.020 0.848 0.036 NA
#> GSM587197 3 0.4222 0.845 0.016 0.020 0.816 0.040 NA
#> GSM587198 3 0.1299 0.860 0.012 0.020 0.960 0.000 NA
#> GSM587199 3 0.1851 0.860 0.008 0.024 0.940 0.004 NA
#> GSM587200 3 0.3590 0.839 0.036 0.000 0.828 0.008 NA
#> GSM587201 3 0.3590 0.839 0.036 0.000 0.828 0.008 NA
#> GSM587202 3 0.1299 0.860 0.012 0.020 0.960 0.000 NA
#> GSM198767 1 0.2228 0.898 0.912 0.000 0.000 0.040 NA
#> GSM198769 1 0.1571 0.910 0.936 0.000 0.004 0.000 NA
#> GSM198772 1 0.1831 0.905 0.920 0.000 0.004 0.000 NA
#> GSM198773 1 0.0566 0.913 0.984 0.000 0.000 0.012 NA
#> GSM198776 1 0.2313 0.898 0.912 0.000 0.004 0.040 NA
#> GSM198778 1 0.3289 0.846 0.844 0.000 0.048 0.000 NA
#> GSM198780 1 0.2597 0.883 0.884 0.000 0.024 0.000 NA
#> GSM198781 1 0.0566 0.913 0.984 0.000 0.000 0.012 NA
#> GSM198765 3 0.5347 0.817 0.012 0.012 0.700 0.064 NA
#> GSM198766 3 0.5608 0.775 0.052 0.000 0.660 0.040 NA
#> GSM198768 3 0.3765 0.847 0.016 0.020 0.848 0.036 NA
#> GSM198770 3 0.4222 0.845 0.016 0.020 0.816 0.040 NA
#> GSM198771 3 0.1299 0.860 0.012 0.020 0.960 0.000 NA
#> GSM198774 3 0.5439 0.808 0.028 0.000 0.680 0.064 NA
#> GSM198775 3 0.5677 0.783 0.016 0.020 0.652 0.044 NA
#> GSM198777 3 0.3765 0.847 0.016 0.020 0.848 0.036 NA
#> GSM198779 3 0.1851 0.860 0.008 0.024 0.940 0.004 NA
#> GSM587218 4 0.3391 0.971 0.188 0.000 0.012 0.800 NA
#> GSM587219 4 0.3652 0.980 0.200 0.000 0.012 0.784 NA
#> GSM587220 4 0.3652 0.980 0.200 0.000 0.012 0.784 NA
#> GSM587221 4 0.3752 0.980 0.200 0.000 0.016 0.780 NA
#> GSM587222 4 0.3752 0.980 0.200 0.000 0.016 0.780 NA
#> GSM587223 4 0.3496 0.979 0.200 0.000 0.012 0.788 NA
#> GSM587224 4 0.3752 0.980 0.200 0.000 0.016 0.780 NA
#> GSM587225 4 0.4913 0.966 0.200 0.000 0.016 0.724 NA
#> GSM587226 4 0.3752 0.980 0.200 0.000 0.016 0.780 NA
#> GSM587227 4 0.4877 0.966 0.200 0.000 0.020 0.728 NA
#> GSM587228 4 0.4850 0.966 0.200 0.000 0.016 0.728 NA
#> GSM587229 4 0.4850 0.966 0.200 0.000 0.016 0.728 NA
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM587155 2 0.3944 -0.699 0.000 0.568 0.000 0.004 0.428 NA
#> GSM587156 5 0.4714 1.000 0.000 0.460 0.008 0.008 0.508 NA
#> GSM587157 2 0.4093 -0.746 0.000 0.552 0.004 0.000 0.440 NA
#> GSM587158 2 0.0767 0.598 0.000 0.976 0.000 0.004 0.012 NA
#> GSM587159 2 0.0881 0.598 0.000 0.972 0.000 0.008 0.008 NA
#> GSM587160 2 0.0779 0.591 0.000 0.976 0.000 0.008 0.008 NA
#> GSM587161 2 0.2845 0.312 0.000 0.820 0.000 0.004 0.172 NA
#> GSM587162 2 0.2320 0.424 0.000 0.864 0.000 0.000 0.132 NA
#> GSM587163 2 0.0862 0.589 0.000 0.972 0.000 0.004 0.016 NA
#> GSM587164 2 0.3851 -0.768 0.000 0.540 0.000 0.000 0.460 NA
#> GSM587165 2 0.3958 0.584 0.000 0.764 0.000 0.016 0.040 NA
#> GSM587166 5 0.4714 1.000 0.000 0.460 0.008 0.008 0.508 NA
#> GSM587167 2 0.4452 -0.859 0.000 0.508 0.004 0.008 0.472 NA
#> GSM587168 2 0.4173 0.580 0.000 0.752 0.000 0.016 0.056 NA
#> GSM587169 2 0.0653 0.593 0.000 0.980 0.000 0.004 0.012 NA
#> GSM587170 2 0.3843 -0.745 0.000 0.548 0.000 0.000 0.452 NA
#> GSM587171 2 0.0881 0.598 0.000 0.972 0.000 0.008 0.008 NA
#> GSM587172 2 0.0881 0.598 0.000 0.972 0.000 0.008 0.008 NA
#> GSM587173 2 0.4308 0.569 0.000 0.736 0.000 0.024 0.044 NA
#> GSM587174 2 0.1863 0.612 0.000 0.920 0.000 0.000 0.036 NA
#> GSM587175 2 0.4039 -0.694 0.000 0.568 0.000 0.008 0.424 NA
#> GSM587176 2 0.2737 0.348 0.000 0.832 0.000 0.004 0.160 NA
#> GSM587177 2 0.4042 0.582 0.000 0.760 0.000 0.020 0.040 NA
#> GSM587178 2 0.3309 0.600 0.000 0.824 0.000 0.004 0.056 NA
#> GSM587179 2 0.1364 0.560 0.000 0.944 0.000 0.004 0.048 NA
#> GSM587180 2 0.4197 0.584 0.000 0.752 0.000 0.016 0.060 NA
#> GSM587181 2 0.1863 0.612 0.000 0.920 0.000 0.000 0.036 NA
#> GSM587182 2 0.3603 0.601 0.000 0.808 0.000 0.012 0.056 NA
#> GSM587183 2 0.4042 0.582 0.000 0.760 0.000 0.020 0.040 NA
#> GSM587184 2 0.0748 0.604 0.000 0.976 0.000 0.004 0.004 NA
#> GSM587185 2 0.1477 0.559 0.000 0.940 0.000 0.008 0.048 NA
#> GSM587186 2 0.4308 0.569 0.000 0.736 0.000 0.024 0.044 NA
#> GSM587187 2 0.4836 0.541 0.000 0.704 0.000 0.036 0.068 NA
#> GSM587188 2 0.5379 0.492 0.000 0.664 0.004 0.044 0.084 NA
#> GSM587189 2 0.6218 0.340 0.000 0.568 0.004 0.044 0.184 NA
#> GSM587190 3 0.7122 0.421 0.000 0.072 0.440 0.028 0.336 NA
#> GSM587203 1 0.3835 0.821 0.748 0.000 0.000 0.000 0.204 NA
#> GSM587204 1 0.3867 0.821 0.748 0.000 0.000 0.000 0.200 NA
#> GSM587205 1 0.3835 0.821 0.748 0.000 0.000 0.000 0.204 NA
#> GSM587206 1 0.3835 0.821 0.748 0.000 0.000 0.000 0.204 NA
#> GSM587207 1 0.3835 0.821 0.748 0.000 0.000 0.000 0.204 NA
#> GSM587208 1 0.3835 0.821 0.748 0.000 0.000 0.000 0.204 NA
#> GSM587209 1 0.1096 0.862 0.964 0.000 0.004 0.008 0.004 NA
#> GSM587210 1 0.3523 0.784 0.816 0.000 0.048 0.008 0.004 NA
#> GSM587211 1 0.1526 0.859 0.944 0.000 0.004 0.008 0.008 NA
#> GSM587212 1 0.2480 0.838 0.884 0.000 0.008 0.008 0.008 NA
#> GSM587213 1 0.1938 0.865 0.920 0.000 0.000 0.008 0.052 NA
#> GSM587214 1 0.1938 0.865 0.920 0.000 0.000 0.008 0.052 NA
#> GSM587215 1 0.1210 0.862 0.960 0.000 0.004 0.008 0.008 NA
#> GSM587216 1 0.1988 0.845 0.912 0.000 0.004 0.008 0.004 NA
#> GSM587217 1 0.0976 0.863 0.968 0.000 0.000 0.008 0.008 NA
#> GSM587191 3 0.4015 0.728 0.004 0.000 0.596 0.000 0.004 NA
#> GSM587192 3 0.4088 0.715 0.004 0.000 0.556 0.000 0.004 NA
#> GSM587193 3 0.6485 0.665 0.084 0.000 0.484 0.004 0.088 NA
#> GSM587194 3 0.6188 0.686 0.024 0.000 0.508 0.008 0.136 NA
#> GSM587195 3 0.3355 0.757 0.008 0.000 0.840 0.016 0.036 NA
#> GSM587196 3 0.3355 0.757 0.008 0.000 0.840 0.016 0.036 NA
#> GSM587197 3 0.4077 0.753 0.008 0.000 0.772 0.012 0.048 NA
#> GSM587198 3 0.0551 0.785 0.008 0.000 0.984 0.004 0.000 NA
#> GSM587199 3 0.2026 0.786 0.008 0.000 0.916 0.004 0.012 NA
#> GSM587200 3 0.4462 0.762 0.044 0.000 0.736 0.004 0.028 NA
#> GSM587201 3 0.4462 0.762 0.044 0.000 0.736 0.004 0.028 NA
#> GSM587202 3 0.0405 0.784 0.008 0.000 0.988 0.004 0.000 NA
#> GSM198767 1 0.3835 0.821 0.748 0.000 0.000 0.000 0.204 NA
#> GSM198769 1 0.1096 0.862 0.964 0.000 0.004 0.008 0.004 NA
#> GSM198772 1 0.1526 0.859 0.944 0.000 0.004 0.008 0.008 NA
#> GSM198773 1 0.1938 0.865 0.920 0.000 0.000 0.008 0.052 NA
#> GSM198776 1 0.3867 0.821 0.748 0.000 0.000 0.000 0.200 NA
#> GSM198778 1 0.3523 0.784 0.816 0.000 0.048 0.008 0.004 NA
#> GSM198780 1 0.2480 0.838 0.884 0.000 0.008 0.008 0.008 NA
#> GSM198781 1 0.1938 0.865 0.920 0.000 0.000 0.008 0.052 NA
#> GSM198765 3 0.4015 0.728 0.004 0.000 0.596 0.000 0.004 NA
#> GSM198766 3 0.6485 0.665 0.084 0.000 0.484 0.004 0.088 NA
#> GSM198768 3 0.3355 0.757 0.008 0.000 0.840 0.016 0.036 NA
#> GSM198770 3 0.4077 0.753 0.008 0.000 0.772 0.012 0.048 NA
#> GSM198771 3 0.0551 0.785 0.008 0.000 0.984 0.004 0.000 NA
#> GSM198774 3 0.4088 0.715 0.004 0.000 0.556 0.000 0.004 NA
#> GSM198775 3 0.6188 0.686 0.024 0.000 0.508 0.008 0.136 NA
#> GSM198777 3 0.3355 0.757 0.008 0.000 0.840 0.016 0.036 NA
#> GSM198779 3 0.2026 0.786 0.008 0.000 0.916 0.004 0.012 NA
#> GSM587218 4 0.2290 0.952 0.084 0.000 0.000 0.892 0.004 NA
#> GSM587219 4 0.1806 0.955 0.088 0.000 0.000 0.908 0.000 NA
#> GSM587220 4 0.1806 0.955 0.088 0.000 0.000 0.908 0.000 NA
#> GSM587221 4 0.2816 0.954 0.088 0.000 0.000 0.868 0.020 NA
#> GSM587222 4 0.2816 0.954 0.088 0.000 0.000 0.868 0.020 NA
#> GSM587223 4 0.2255 0.953 0.088 0.000 0.000 0.892 0.004 NA
#> GSM587224 4 0.3051 0.953 0.088 0.000 0.000 0.856 0.024 NA
#> GSM587225 4 0.4517 0.930 0.088 0.000 0.004 0.764 0.100 NA
#> GSM587226 4 0.2816 0.954 0.088 0.000 0.000 0.868 0.020 NA
#> GSM587227 4 0.4124 0.931 0.088 0.000 0.004 0.792 0.084 NA
#> GSM587228 4 0.4339 0.931 0.088 0.000 0.004 0.776 0.096 NA
#> GSM587229 4 0.4124 0.931 0.088 0.000 0.004 0.792 0.084 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 specimen(p) k
#> CV:kmeans 92 4.01e-14 2
#> CV:kmeans 92 7.20e-32 3
#> CV:kmeans 92 4.48e-47 4
#> CV:kmeans 92 4.48e-47 5
#> CV:kmeans 80 1.36e-37 6
If matrix rows can be associated to genes, consider to use functional_enrichment(res,
...) to perform function enrichment for the signature genes. See this vignette for more detailed explanations.
The object with results only for a single top-value method and a single partition method can be extracted as:
res = res_list["CV", "skmeans"]
# you can also extract it by
# res = res_list["CV:skmeans"]
A summary of res and all the functions that can be applied to it:
res
#> A 'ConsensusPartition' object with k = 2, 3, 4, 5, 6.
#> On a matrix with 51941 rows and 92 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 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 1.000 0.991 0.996 0.5007 0.500 0.500
#> 3 3 1.000 0.969 0.984 0.2879 0.821 0.653
#> 4 4 0.990 0.923 0.965 0.1110 0.893 0.714
#> 5 5 0.978 0.950 0.967 0.0452 0.947 0.821
#> 6 6 0.895 0.877 0.897 0.0298 0.989 0.957
suggest_best_k() suggests the best \(k\) based on these statistics. The rules are as follows:
suggest_best_k(res)
#> [1] 5
#> attr(,"optional")
#> [1] 2 3 4
There is also optional best \(k\) = 2 3 4 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
#> GSM587155 2 0.000 0.993 0.000 1.000
#> GSM587156 2 0.000 0.993 0.000 1.000
#> GSM587157 2 0.000 0.993 0.000 1.000
#> GSM587158 2 0.000 0.993 0.000 1.000
#> GSM587159 2 0.000 0.993 0.000 1.000
#> GSM587160 2 0.000 0.993 0.000 1.000
#> GSM587161 2 0.000 0.993 0.000 1.000
#> GSM587162 2 0.000 0.993 0.000 1.000
#> GSM587163 2 0.000 0.993 0.000 1.000
#> GSM587164 2 0.000 0.993 0.000 1.000
#> GSM587165 2 0.000 0.993 0.000 1.000
#> GSM587166 2 0.000 0.993 0.000 1.000
#> GSM587167 2 0.000 0.993 0.000 1.000
#> GSM587168 2 0.000 0.993 0.000 1.000
#> GSM587169 2 0.000 0.993 0.000 1.000
#> GSM587170 2 0.000 0.993 0.000 1.000
#> GSM587171 2 0.000 0.993 0.000 1.000
#> GSM587172 2 0.000 0.993 0.000 1.000
#> GSM587173 2 0.000 0.993 0.000 1.000
#> GSM587174 2 0.000 0.993 0.000 1.000
#> GSM587175 2 0.000 0.993 0.000 1.000
#> GSM587176 2 0.000 0.993 0.000 1.000
#> GSM587177 2 0.000 0.993 0.000 1.000
#> GSM587178 2 0.000 0.993 0.000 1.000
#> GSM587179 2 0.000 0.993 0.000 1.000
#> GSM587180 2 0.000 0.993 0.000 1.000
#> GSM587181 2 0.000 0.993 0.000 1.000
#> GSM587182 2 0.000 0.993 0.000 1.000
#> GSM587183 2 0.000 0.993 0.000 1.000
#> GSM587184 2 0.000 0.993 0.000 1.000
#> GSM587185 2 0.000 0.993 0.000 1.000
#> GSM587186 2 0.000 0.993 0.000 1.000
#> GSM587187 2 0.000 0.993 0.000 1.000
#> GSM587188 2 0.000 0.993 0.000 1.000
#> GSM587189 2 0.000 0.993 0.000 1.000
#> GSM587190 2 0.000 0.993 0.000 1.000
#> GSM587203 1 0.000 1.000 1.000 0.000
#> GSM587204 1 0.000 1.000 1.000 0.000
#> GSM587205 1 0.000 1.000 1.000 0.000
#> GSM587206 1 0.000 1.000 1.000 0.000
#> GSM587207 1 0.000 1.000 1.000 0.000
#> GSM587208 1 0.000 1.000 1.000 0.000
#> GSM587209 1 0.000 1.000 1.000 0.000
#> GSM587210 1 0.000 1.000 1.000 0.000
#> GSM587211 1 0.000 1.000 1.000 0.000
#> GSM587212 1 0.000 1.000 1.000 0.000
#> GSM587213 1 0.000 1.000 1.000 0.000
#> GSM587214 1 0.000 1.000 1.000 0.000
#> GSM587215 1 0.000 1.000 1.000 0.000
#> GSM587216 1 0.000 1.000 1.000 0.000
#> GSM587217 1 0.000 1.000 1.000 0.000
#> GSM587191 2 0.000 0.993 0.000 1.000
#> GSM587192 1 0.000 1.000 1.000 0.000
#> GSM587193 1 0.000 1.000 1.000 0.000
#> GSM587194 2 0.689 0.779 0.184 0.816
#> GSM587195 2 0.000 0.993 0.000 1.000
#> GSM587196 2 0.000 0.993 0.000 1.000
#> GSM587197 2 0.000 0.993 0.000 1.000
#> GSM587198 2 0.000 0.993 0.000 1.000
#> GSM587199 2 0.000 0.993 0.000 1.000
#> GSM587200 1 0.000 1.000 1.000 0.000
#> GSM587201 1 0.000 1.000 1.000 0.000
#> GSM587202 2 0.000 0.993 0.000 1.000
#> GSM198767 1 0.000 1.000 1.000 0.000
#> GSM198769 1 0.000 1.000 1.000 0.000
#> GSM198772 1 0.000 1.000 1.000 0.000
#> GSM198773 1 0.000 1.000 1.000 0.000
#> GSM198776 1 0.000 1.000 1.000 0.000
#> GSM198778 1 0.000 1.000 1.000 0.000
#> GSM198780 1 0.000 1.000 1.000 0.000
#> GSM198781 1 0.000 1.000 1.000 0.000
#> GSM198765 2 0.000 0.993 0.000 1.000
#> GSM198766 1 0.000 1.000 1.000 0.000
#> GSM198768 2 0.000 0.993 0.000 1.000
#> GSM198770 2 0.000 0.993 0.000 1.000
#> GSM198771 2 0.000 0.993 0.000 1.000
#> GSM198774 1 0.000 1.000 1.000 0.000
#> GSM198775 2 0.689 0.779 0.184 0.816
#> GSM198777 2 0.000 0.993 0.000 1.000
#> GSM198779 2 0.000 0.993 0.000 1.000
#> GSM587218 1 0.000 1.000 1.000 0.000
#> GSM587219 1 0.000 1.000 1.000 0.000
#> GSM587220 1 0.000 1.000 1.000 0.000
#> GSM587221 1 0.000 1.000 1.000 0.000
#> GSM587222 1 0.000 1.000 1.000 0.000
#> GSM587223 1 0.000 1.000 1.000 0.000
#> GSM587224 1 0.000 1.000 1.000 0.000
#> GSM587225 1 0.000 1.000 1.000 0.000
#> GSM587226 1 0.000 1.000 1.000 0.000
#> GSM587227 1 0.000 1.000 1.000 0.000
#> GSM587228 1 0.000 1.000 1.000 0.000
#> GSM587229 1 0.000 1.000 1.000 0.000
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM587155 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587156 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587157 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587158 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587159 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587160 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587161 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587162 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587163 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587164 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587165 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587166 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587167 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587168 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587169 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587170 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587171 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587172 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587173 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587174 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587175 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587176 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587177 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587178 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587179 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587180 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587181 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587182 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587183 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587184 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587185 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587186 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587187 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587188 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587189 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587190 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587203 1 0.0000 0.994 1.000 0.000 0.000
#> GSM587204 1 0.0000 0.994 1.000 0.000 0.000
#> GSM587205 1 0.0000 0.994 1.000 0.000 0.000
#> GSM587206 1 0.0000 0.994 1.000 0.000 0.000
#> GSM587207 1 0.0000 0.994 1.000 0.000 0.000
#> GSM587208 1 0.0000 0.994 1.000 0.000 0.000
#> GSM587209 1 0.0000 0.994 1.000 0.000 0.000
#> GSM587210 1 0.0000 0.994 1.000 0.000 0.000
#> GSM587211 1 0.0000 0.994 1.000 0.000 0.000
#> GSM587212 1 0.0000 0.994 1.000 0.000 0.000
#> GSM587213 1 0.0000 0.994 1.000 0.000 0.000
#> GSM587214 1 0.0000 0.994 1.000 0.000 0.000
#> GSM587215 1 0.0000 0.994 1.000 0.000 0.000
#> GSM587216 1 0.0000 0.994 1.000 0.000 0.000
#> GSM587217 1 0.0000 0.994 1.000 0.000 0.000
#> GSM587191 3 0.0747 0.942 0.000 0.016 0.984
#> GSM587192 3 0.0747 0.932 0.016 0.000 0.984
#> GSM587193 1 0.0237 0.993 0.996 0.000 0.004
#> GSM587194 3 0.6274 0.232 0.000 0.456 0.544
#> GSM587195 3 0.0892 0.943 0.000 0.020 0.980
#> GSM587196 3 0.0892 0.943 0.000 0.020 0.980
#> GSM587197 3 0.0747 0.942 0.000 0.016 0.984
#> GSM587198 3 0.0892 0.943 0.000 0.020 0.980
#> GSM587199 3 0.0892 0.943 0.000 0.020 0.980
#> GSM587200 3 0.0892 0.931 0.020 0.000 0.980
#> GSM587201 3 0.0892 0.931 0.020 0.000 0.980
#> GSM587202 3 0.0892 0.943 0.000 0.020 0.980
#> GSM198767 1 0.0000 0.994 1.000 0.000 0.000
#> GSM198769 1 0.0000 0.994 1.000 0.000 0.000
#> GSM198772 1 0.0000 0.994 1.000 0.000 0.000
#> GSM198773 1 0.0000 0.994 1.000 0.000 0.000
#> GSM198776 1 0.0000 0.994 1.000 0.000 0.000
#> GSM198778 1 0.0000 0.994 1.000 0.000 0.000
#> GSM198780 1 0.0000 0.994 1.000 0.000 0.000
#> GSM198781 1 0.0000 0.994 1.000 0.000 0.000
#> GSM198765 3 0.0747 0.942 0.000 0.016 0.984
#> GSM198766 1 0.0237 0.993 0.996 0.000 0.004
#> GSM198768 3 0.0892 0.943 0.000 0.020 0.980
#> GSM198770 3 0.0747 0.942 0.000 0.016 0.984
#> GSM198771 3 0.0892 0.943 0.000 0.020 0.980
#> GSM198774 3 0.0747 0.932 0.016 0.000 0.984
#> GSM198775 3 0.6274 0.232 0.000 0.456 0.544
#> GSM198777 3 0.0892 0.943 0.000 0.020 0.980
#> GSM198779 3 0.0892 0.943 0.000 0.020 0.980
#> GSM587218 1 0.0892 0.988 0.980 0.000 0.020
#> GSM587219 1 0.0892 0.988 0.980 0.000 0.020
#> GSM587220 1 0.0892 0.988 0.980 0.000 0.020
#> GSM587221 1 0.0892 0.988 0.980 0.000 0.020
#> GSM587222 1 0.0892 0.988 0.980 0.000 0.020
#> GSM587223 1 0.0892 0.988 0.980 0.000 0.020
#> GSM587224 1 0.0892 0.988 0.980 0.000 0.020
#> GSM587225 1 0.0892 0.988 0.980 0.000 0.020
#> GSM587226 1 0.0892 0.988 0.980 0.000 0.020
#> GSM587227 1 0.0892 0.988 0.980 0.000 0.020
#> GSM587228 1 0.0892 0.988 0.980 0.000 0.020
#> GSM587229 1 0.0892 0.988 0.980 0.000 0.020
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM587155 2 0.0000 0.9703 0.000 1.000 0.000 0.000
#> GSM587156 2 0.0000 0.9703 0.000 1.000 0.000 0.000
#> GSM587157 2 0.0000 0.9703 0.000 1.000 0.000 0.000
#> GSM587158 2 0.0000 0.9703 0.000 1.000 0.000 0.000
#> GSM587159 2 0.0000 0.9703 0.000 1.000 0.000 0.000
#> GSM587160 2 0.0000 0.9703 0.000 1.000 0.000 0.000
#> GSM587161 2 0.0000 0.9703 0.000 1.000 0.000 0.000
#> GSM587162 2 0.0000 0.9703 0.000 1.000 0.000 0.000
#> GSM587163 2 0.0000 0.9703 0.000 1.000 0.000 0.000
#> GSM587164 2 0.0000 0.9703 0.000 1.000 0.000 0.000
#> GSM587165 2 0.0000 0.9703 0.000 1.000 0.000 0.000
#> GSM587166 2 0.0000 0.9703 0.000 1.000 0.000 0.000
#> GSM587167 2 0.0000 0.9703 0.000 1.000 0.000 0.000
#> GSM587168 2 0.0000 0.9703 0.000 1.000 0.000 0.000
#> GSM587169 2 0.0000 0.9703 0.000 1.000 0.000 0.000
#> GSM587170 2 0.0000 0.9703 0.000 1.000 0.000 0.000
#> GSM587171 2 0.0000 0.9703 0.000 1.000 0.000 0.000
#> GSM587172 2 0.0000 0.9703 0.000 1.000 0.000 0.000
#> GSM587173 2 0.0000 0.9703 0.000 1.000 0.000 0.000
#> GSM587174 2 0.0000 0.9703 0.000 1.000 0.000 0.000
#> GSM587175 2 0.0000 0.9703 0.000 1.000 0.000 0.000
#> GSM587176 2 0.0000 0.9703 0.000 1.000 0.000 0.000
#> GSM587177 2 0.0000 0.9703 0.000 1.000 0.000 0.000
#> GSM587178 2 0.0000 0.9703 0.000 1.000 0.000 0.000
#> GSM587179 2 0.0000 0.9703 0.000 1.000 0.000 0.000
#> GSM587180 2 0.0000 0.9703 0.000 1.000 0.000 0.000
#> GSM587181 2 0.0000 0.9703 0.000 1.000 0.000 0.000
#> GSM587182 2 0.0000 0.9703 0.000 1.000 0.000 0.000
#> GSM587183 2 0.0000 0.9703 0.000 1.000 0.000 0.000
#> GSM587184 2 0.0000 0.9703 0.000 1.000 0.000 0.000
#> GSM587185 2 0.0000 0.9703 0.000 1.000 0.000 0.000
#> GSM587186 2 0.0000 0.9703 0.000 1.000 0.000 0.000
#> GSM587187 2 0.0000 0.9703 0.000 1.000 0.000 0.000
#> GSM587188 2 0.0000 0.9703 0.000 1.000 0.000 0.000
#> GSM587189 2 0.0000 0.9703 0.000 1.000 0.000 0.000
#> GSM587190 2 0.0524 0.9599 0.000 0.988 0.004 0.008
#> GSM587203 1 0.0000 0.9501 1.000 0.000 0.000 0.000
#> GSM587204 1 0.0000 0.9501 1.000 0.000 0.000 0.000
#> GSM587205 1 0.0000 0.9501 1.000 0.000 0.000 0.000
#> GSM587206 1 0.0000 0.9501 1.000 0.000 0.000 0.000
#> GSM587207 1 0.0000 0.9501 1.000 0.000 0.000 0.000
#> GSM587208 1 0.0000 0.9501 1.000 0.000 0.000 0.000
#> GSM587209 1 0.0000 0.9501 1.000 0.000 0.000 0.000
#> GSM587210 1 0.0000 0.9501 1.000 0.000 0.000 0.000
#> GSM587211 1 0.0000 0.9501 1.000 0.000 0.000 0.000
#> GSM587212 1 0.0000 0.9501 1.000 0.000 0.000 0.000
#> GSM587213 1 0.0000 0.9501 1.000 0.000 0.000 0.000
#> GSM587214 1 0.0000 0.9501 1.000 0.000 0.000 0.000
#> GSM587215 1 0.0000 0.9501 1.000 0.000 0.000 0.000
#> GSM587216 1 0.0000 0.9501 1.000 0.000 0.000 0.000
#> GSM587217 1 0.0000 0.9501 1.000 0.000 0.000 0.000
#> GSM587191 3 0.1118 0.9534 0.000 0.000 0.964 0.036
#> GSM587192 3 0.1118 0.9534 0.000 0.000 0.964 0.036
#> GSM587193 1 0.4790 0.4344 0.620 0.000 0.000 0.380
#> GSM587194 2 0.7823 0.0367 0.004 0.440 0.332 0.224
#> GSM587195 3 0.0000 0.9665 0.000 0.000 1.000 0.000
#> GSM587196 3 0.0000 0.9665 0.000 0.000 1.000 0.000
#> GSM587197 3 0.0000 0.9665 0.000 0.000 1.000 0.000
#> GSM587198 3 0.0000 0.9665 0.000 0.000 1.000 0.000
#> GSM587199 3 0.0188 0.9655 0.000 0.000 0.996 0.004
#> GSM587200 3 0.4769 0.5286 0.308 0.000 0.684 0.008
#> GSM587201 1 0.5229 0.2098 0.564 0.000 0.428 0.008
#> GSM587202 3 0.0000 0.9665 0.000 0.000 1.000 0.000
#> GSM198767 1 0.0000 0.9501 1.000 0.000 0.000 0.000
#> GSM198769 1 0.0000 0.9501 1.000 0.000 0.000 0.000
#> GSM198772 1 0.0000 0.9501 1.000 0.000 0.000 0.000
#> GSM198773 1 0.0000 0.9501 1.000 0.000 0.000 0.000
#> GSM198776 1 0.0000 0.9501 1.000 0.000 0.000 0.000
#> GSM198778 1 0.0000 0.9501 1.000 0.000 0.000 0.000
#> GSM198780 1 0.0000 0.9501 1.000 0.000 0.000 0.000
#> GSM198781 1 0.0000 0.9501 1.000 0.000 0.000 0.000
#> GSM198765 3 0.1118 0.9534 0.000 0.000 0.964 0.036
#> GSM198766 1 0.4790 0.4344 0.620 0.000 0.000 0.380
#> GSM198768 3 0.0000 0.9665 0.000 0.000 1.000 0.000
#> GSM198770 3 0.0000 0.9665 0.000 0.000 1.000 0.000
#> GSM198771 3 0.0000 0.9665 0.000 0.000 1.000 0.000
#> GSM198774 3 0.1118 0.9534 0.000 0.000 0.964 0.036
#> GSM198775 2 0.7823 0.0367 0.004 0.440 0.332 0.224
#> GSM198777 3 0.0000 0.9665 0.000 0.000 1.000 0.000
#> GSM198779 3 0.0188 0.9655 0.000 0.000 0.996 0.004
#> GSM587218 4 0.1118 1.0000 0.036 0.000 0.000 0.964
#> GSM587219 4 0.1118 1.0000 0.036 0.000 0.000 0.964
#> GSM587220 4 0.1118 1.0000 0.036 0.000 0.000 0.964
#> GSM587221 4 0.1118 1.0000 0.036 0.000 0.000 0.964
#> GSM587222 4 0.1118 1.0000 0.036 0.000 0.000 0.964
#> GSM587223 4 0.1118 1.0000 0.036 0.000 0.000 0.964
#> GSM587224 4 0.1118 1.0000 0.036 0.000 0.000 0.964
#> GSM587225 4 0.1118 1.0000 0.036 0.000 0.000 0.964
#> GSM587226 4 0.1118 1.0000 0.036 0.000 0.000 0.964
#> GSM587227 4 0.1118 1.0000 0.036 0.000 0.000 0.964
#> GSM587228 4 0.1118 1.0000 0.036 0.000 0.000 0.964
#> GSM587229 4 0.1118 1.0000 0.036 0.000 0.000 0.964
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM587155 2 0.0000 0.993 0.000 1.000 0.000 0.000 0.000
#> GSM587156 2 0.0703 0.973 0.000 0.976 0.000 0.000 0.024
#> GSM587157 2 0.0162 0.990 0.000 0.996 0.004 0.000 0.000
#> GSM587158 2 0.0000 0.993 0.000 1.000 0.000 0.000 0.000
#> GSM587159 2 0.0000 0.993 0.000 1.000 0.000 0.000 0.000
#> GSM587160 2 0.0000 0.993 0.000 1.000 0.000 0.000 0.000
#> GSM587161 2 0.0000 0.993 0.000 1.000 0.000 0.000 0.000
#> GSM587162 2 0.0000 0.993 0.000 1.000 0.000 0.000 0.000
#> GSM587163 2 0.0000 0.993 0.000 1.000 0.000 0.000 0.000
#> GSM587164 2 0.0000 0.993 0.000 1.000 0.000 0.000 0.000
#> GSM587165 2 0.0000 0.993 0.000 1.000 0.000 0.000 0.000
#> GSM587166 2 0.0703 0.973 0.000 0.976 0.000 0.000 0.024
#> GSM587167 2 0.0000 0.993 0.000 1.000 0.000 0.000 0.000
#> GSM587168 2 0.0000 0.993 0.000 1.000 0.000 0.000 0.000
#> GSM587169 2 0.0000 0.993 0.000 1.000 0.000 0.000 0.000
#> GSM587170 2 0.0000 0.993 0.000 1.000 0.000 0.000 0.000
#> GSM587171 2 0.0000 0.993 0.000 1.000 0.000 0.000 0.000
#> GSM587172 2 0.0000 0.993 0.000 1.000 0.000 0.000 0.000
#> GSM587173 2 0.0000 0.993 0.000 1.000 0.000 0.000 0.000
#> GSM587174 2 0.0000 0.993 0.000 1.000 0.000 0.000 0.000
#> GSM587175 2 0.0000 0.993 0.000 1.000 0.000 0.000 0.000
#> GSM587176 2 0.0000 0.993 0.000 1.000 0.000 0.000 0.000
#> GSM587177 2 0.0000 0.993 0.000 1.000 0.000 0.000 0.000
#> GSM587178 2 0.0000 0.993 0.000 1.000 0.000 0.000 0.000
#> GSM587179 2 0.0000 0.993 0.000 1.000 0.000 0.000 0.000
#> GSM587180 2 0.0000 0.993 0.000 1.000 0.000 0.000 0.000
#> GSM587181 2 0.0000 0.993 0.000 1.000 0.000 0.000 0.000
#> GSM587182 2 0.0000 0.993 0.000 1.000 0.000 0.000 0.000
#> GSM587183 2 0.0000 0.993 0.000 1.000 0.000 0.000 0.000
#> GSM587184 2 0.0000 0.993 0.000 1.000 0.000 0.000 0.000
#> GSM587185 2 0.0000 0.993 0.000 1.000 0.000 0.000 0.000
#> GSM587186 2 0.0000 0.993 0.000 1.000 0.000 0.000 0.000
#> GSM587187 2 0.0162 0.990 0.000 0.996 0.000 0.000 0.004
#> GSM587188 2 0.0162 0.990 0.000 0.996 0.000 0.000 0.004
#> GSM587189 2 0.0162 0.990 0.000 0.996 0.000 0.000 0.004
#> GSM587190 2 0.3246 0.767 0.000 0.808 0.008 0.000 0.184
#> GSM587203 1 0.0404 0.989 0.988 0.000 0.000 0.000 0.012
#> GSM587204 1 0.0404 0.989 0.988 0.000 0.000 0.000 0.012
#> GSM587205 1 0.0404 0.989 0.988 0.000 0.000 0.000 0.012
#> GSM587206 1 0.0404 0.989 0.988 0.000 0.000 0.000 0.012
#> GSM587207 1 0.0404 0.989 0.988 0.000 0.000 0.000 0.012
#> GSM587208 1 0.0404 0.989 0.988 0.000 0.000 0.000 0.012
#> GSM587209 1 0.0000 0.991 1.000 0.000 0.000 0.000 0.000
#> GSM587210 1 0.0963 0.966 0.964 0.000 0.000 0.000 0.036
#> GSM587211 1 0.0000 0.991 1.000 0.000 0.000 0.000 0.000
#> GSM587212 1 0.0609 0.979 0.980 0.000 0.000 0.000 0.020
#> GSM587213 1 0.0000 0.991 1.000 0.000 0.000 0.000 0.000
#> GSM587214 1 0.0000 0.991 1.000 0.000 0.000 0.000 0.000
#> GSM587215 1 0.0000 0.991 1.000 0.000 0.000 0.000 0.000
#> GSM587216 1 0.0000 0.991 1.000 0.000 0.000 0.000 0.000
#> GSM587217 1 0.0000 0.991 1.000 0.000 0.000 0.000 0.000
#> GSM587191 5 0.2852 0.882 0.000 0.000 0.172 0.000 0.828
#> GSM587192 5 0.2852 0.882 0.000 0.000 0.172 0.000 0.828
#> GSM587193 5 0.0865 0.876 0.024 0.000 0.000 0.004 0.972
#> GSM587194 5 0.0451 0.881 0.000 0.008 0.004 0.000 0.988
#> GSM587195 3 0.0000 0.844 0.000 0.000 1.000 0.000 0.000
#> GSM587196 3 0.0000 0.844 0.000 0.000 1.000 0.000 0.000
#> GSM587197 3 0.0290 0.843 0.000 0.000 0.992 0.000 0.008
#> GSM587198 3 0.2930 0.829 0.000 0.000 0.832 0.004 0.164
#> GSM587199 3 0.3048 0.823 0.000 0.000 0.820 0.004 0.176
#> GSM587200 3 0.5761 0.550 0.092 0.000 0.572 0.004 0.332
#> GSM587201 3 0.6527 0.402 0.184 0.000 0.484 0.004 0.328
#> GSM587202 3 0.2583 0.837 0.000 0.000 0.864 0.004 0.132
#> GSM198767 1 0.0404 0.989 0.988 0.000 0.000 0.000 0.012
#> GSM198769 1 0.0000 0.991 1.000 0.000 0.000 0.000 0.000
#> GSM198772 1 0.0000 0.991 1.000 0.000 0.000 0.000 0.000
#> GSM198773 1 0.0000 0.991 1.000 0.000 0.000 0.000 0.000
#> GSM198776 1 0.0404 0.989 0.988 0.000 0.000 0.000 0.012
#> GSM198778 1 0.0963 0.966 0.964 0.000 0.000 0.000 0.036
#> GSM198780 1 0.0609 0.979 0.980 0.000 0.000 0.000 0.020
#> GSM198781 1 0.0000 0.991 1.000 0.000 0.000 0.000 0.000
#> GSM198765 5 0.2852 0.882 0.000 0.000 0.172 0.000 0.828
#> GSM198766 5 0.0865 0.876 0.024 0.000 0.000 0.004 0.972
#> GSM198768 3 0.0000 0.844 0.000 0.000 1.000 0.000 0.000
#> GSM198770 3 0.0290 0.843 0.000 0.000 0.992 0.000 0.008
#> GSM198771 3 0.2930 0.829 0.000 0.000 0.832 0.004 0.164
#> GSM198774 5 0.2852 0.882 0.000 0.000 0.172 0.000 0.828
#> GSM198775 5 0.0451 0.881 0.000 0.008 0.004 0.000 0.988
#> GSM198777 3 0.0000 0.844 0.000 0.000 1.000 0.000 0.000
#> GSM198779 3 0.3048 0.823 0.000 0.000 0.820 0.004 0.176
#> GSM587218 4 0.0162 1.000 0.004 0.000 0.000 0.996 0.000
#> GSM587219 4 0.0162 1.000 0.004 0.000 0.000 0.996 0.000
#> GSM587220 4 0.0162 1.000 0.004 0.000 0.000 0.996 0.000
#> GSM587221 4 0.0162 1.000 0.004 0.000 0.000 0.996 0.000
#> GSM587222 4 0.0162 1.000 0.004 0.000 0.000 0.996 0.000
#> GSM587223 4 0.0162 1.000 0.004 0.000 0.000 0.996 0.000
#> GSM587224 4 0.0162 1.000 0.004 0.000 0.000 0.996 0.000
#> GSM587225 4 0.0162 1.000 0.004 0.000 0.000 0.996 0.000
#> GSM587226 4 0.0162 1.000 0.004 0.000 0.000 0.996 0.000
#> GSM587227 4 0.0162 1.000 0.004 0.000 0.000 0.996 0.000
#> GSM587228 4 0.0162 1.000 0.004 0.000 0.000 0.996 0.000
#> GSM587229 4 0.0162 1.000 0.004 0.000 0.000 0.996 0.000
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM587155 2 0.2020 0.899 0.000 0.896 0.000 0 0.096 0.008
#> GSM587156 2 0.3629 0.705 0.000 0.712 0.000 0 0.276 0.012
#> GSM587157 2 0.1866 0.906 0.000 0.908 0.000 0 0.084 0.008
#> GSM587158 2 0.0000 0.947 0.000 1.000 0.000 0 0.000 0.000
#> GSM587159 2 0.0000 0.947 0.000 1.000 0.000 0 0.000 0.000
#> GSM587160 2 0.0000 0.947 0.000 1.000 0.000 0 0.000 0.000
#> GSM587161 2 0.1010 0.934 0.000 0.960 0.000 0 0.036 0.004
#> GSM587162 2 0.0458 0.944 0.000 0.984 0.000 0 0.016 0.000
#> GSM587163 2 0.0000 0.947 0.000 1.000 0.000 0 0.000 0.000
#> GSM587164 2 0.2165 0.891 0.000 0.884 0.000 0 0.108 0.008
#> GSM587165 2 0.0858 0.941 0.000 0.968 0.000 0 0.028 0.004
#> GSM587166 2 0.3629 0.705 0.000 0.712 0.000 0 0.276 0.012
#> GSM587167 2 0.2257 0.885 0.000 0.876 0.000 0 0.116 0.008
#> GSM587168 2 0.0858 0.941 0.000 0.968 0.000 0 0.028 0.004
#> GSM587169 2 0.0000 0.947 0.000 1.000 0.000 0 0.000 0.000
#> GSM587170 2 0.2165 0.891 0.000 0.884 0.000 0 0.108 0.008
#> GSM587171 2 0.0000 0.947 0.000 1.000 0.000 0 0.000 0.000
#> GSM587172 2 0.0000 0.947 0.000 1.000 0.000 0 0.000 0.000
#> GSM587173 2 0.0858 0.941 0.000 0.968 0.000 0 0.028 0.004
#> GSM587174 2 0.0000 0.947 0.000 1.000 0.000 0 0.000 0.000
#> GSM587175 2 0.1812 0.909 0.000 0.912 0.000 0 0.080 0.008
#> GSM587176 2 0.0363 0.945 0.000 0.988 0.000 0 0.012 0.000
#> GSM587177 2 0.0858 0.941 0.000 0.968 0.000 0 0.028 0.004
#> GSM587178 2 0.0260 0.946 0.000 0.992 0.000 0 0.008 0.000
#> GSM587179 2 0.0000 0.947 0.000 1.000 0.000 0 0.000 0.000
#> GSM587180 2 0.0777 0.942 0.000 0.972 0.000 0 0.024 0.004
#> GSM587181 2 0.0000 0.947 0.000 1.000 0.000 0 0.000 0.000
#> GSM587182 2 0.0405 0.946 0.000 0.988 0.000 0 0.008 0.004
#> GSM587183 2 0.0858 0.941 0.000 0.968 0.000 0 0.028 0.004
#> GSM587184 2 0.0000 0.947 0.000 1.000 0.000 0 0.000 0.000
#> GSM587185 2 0.0000 0.947 0.000 1.000 0.000 0 0.000 0.000
#> GSM587186 2 0.0858 0.941 0.000 0.968 0.000 0 0.028 0.004
#> GSM587187 2 0.0935 0.940 0.000 0.964 0.000 0 0.032 0.004
#> GSM587188 2 0.1010 0.938 0.000 0.960 0.000 0 0.036 0.004
#> GSM587189 2 0.1082 0.939 0.000 0.956 0.000 0 0.040 0.004
#> GSM587190 2 0.5331 0.578 0.000 0.616 0.016 0 0.260 0.108
#> GSM587203 1 0.1007 0.929 0.956 0.000 0.000 0 0.044 0.000
#> GSM587204 1 0.1007 0.929 0.956 0.000 0.000 0 0.044 0.000
#> GSM587205 1 0.1007 0.929 0.956 0.000 0.000 0 0.044 0.000
#> GSM587206 1 0.1007 0.929 0.956 0.000 0.000 0 0.044 0.000
#> GSM587207 1 0.1007 0.929 0.956 0.000 0.000 0 0.044 0.000
#> GSM587208 1 0.1007 0.929 0.956 0.000 0.000 0 0.044 0.000
#> GSM587209 1 0.1700 0.921 0.916 0.000 0.000 0 0.080 0.004
#> GSM587210 1 0.2968 0.851 0.816 0.000 0.000 0 0.168 0.016
#> GSM587211 1 0.1644 0.922 0.920 0.000 0.000 0 0.076 0.004
#> GSM587212 1 0.2653 0.873 0.844 0.000 0.000 0 0.144 0.012
#> GSM587213 1 0.0000 0.936 1.000 0.000 0.000 0 0.000 0.000
#> GSM587214 1 0.0000 0.936 1.000 0.000 0.000 0 0.000 0.000
#> GSM587215 1 0.0632 0.934 0.976 0.000 0.000 0 0.024 0.000
#> GSM587216 1 0.1588 0.924 0.924 0.000 0.000 0 0.072 0.004
#> GSM587217 1 0.1219 0.930 0.948 0.000 0.000 0 0.048 0.004
#> GSM587191 6 0.0790 1.000 0.000 0.000 0.032 0 0.000 0.968
#> GSM587192 6 0.0790 1.000 0.000 0.000 0.032 0 0.000 0.968
#> GSM587193 5 0.3996 0.410 0.004 0.000 0.000 0 0.512 0.484
#> GSM587194 5 0.3828 0.452 0.000 0.000 0.000 0 0.560 0.440
#> GSM587195 3 0.0547 0.812 0.000 0.000 0.980 0 0.000 0.020
#> GSM587196 3 0.0547 0.812 0.000 0.000 0.980 0 0.000 0.020
#> GSM587197 3 0.1524 0.790 0.000 0.000 0.932 0 0.008 0.060
#> GSM587198 3 0.4358 0.762 0.000 0.000 0.712 0 0.196 0.092
#> GSM587199 3 0.4441 0.753 0.000 0.000 0.700 0 0.208 0.092
#> GSM587200 5 0.6269 0.176 0.060 0.000 0.248 0 0.548 0.144
#> GSM587201 5 0.6479 0.227 0.096 0.000 0.216 0 0.548 0.140
#> GSM587202 3 0.4311 0.764 0.000 0.000 0.716 0 0.196 0.088
#> GSM198767 1 0.1007 0.929 0.956 0.000 0.000 0 0.044 0.000
#> GSM198769 1 0.1700 0.921 0.916 0.000 0.000 0 0.080 0.004
#> GSM198772 1 0.1644 0.922 0.920 0.000 0.000 0 0.076 0.004
#> GSM198773 1 0.0000 0.936 1.000 0.000 0.000 0 0.000 0.000
#> GSM198776 1 0.1007 0.929 0.956 0.000 0.000 0 0.044 0.000
#> GSM198778 1 0.2968 0.851 0.816 0.000 0.000 0 0.168 0.016
#> GSM198780 1 0.2653 0.873 0.844 0.000 0.000 0 0.144 0.012
#> GSM198781 1 0.0000 0.936 1.000 0.000 0.000 0 0.000 0.000
#> GSM198765 6 0.0790 1.000 0.000 0.000 0.032 0 0.000 0.968
#> GSM198766 5 0.3996 0.410 0.004 0.000 0.000 0 0.512 0.484
#> GSM198768 3 0.0547 0.812 0.000 0.000 0.980 0 0.000 0.020
#> GSM198770 3 0.1524 0.790 0.000 0.000 0.932 0 0.008 0.060
#> GSM198771 3 0.4358 0.762 0.000 0.000 0.712 0 0.196 0.092
#> GSM198774 6 0.0790 1.000 0.000 0.000 0.032 0 0.000 0.968
#> GSM198775 5 0.3828 0.452 0.000 0.000 0.000 0 0.560 0.440
#> GSM198777 3 0.0547 0.812 0.000 0.000 0.980 0 0.000 0.020
#> GSM198779 3 0.4441 0.753 0.000 0.000 0.700 0 0.208 0.092
#> GSM587218 4 0.0000 1.000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587219 4 0.0000 1.000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587220 4 0.0000 1.000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587221 4 0.0000 1.000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587222 4 0.0000 1.000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587223 4 0.0000 1.000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587224 4 0.0000 1.000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587225 4 0.0000 1.000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587226 4 0.0000 1.000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587227 4 0.0000 1.000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587228 4 0.0000 1.000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587229 4 0.0000 1.000 0.000 0.000 0.000 1 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 specimen(p) k
#> CV:skmeans 92 4.01e-14 2
#> CV:skmeans 90 2.83e-29 3
#> CV:skmeans 87 4.69e-45 4
#> CV:skmeans 91 3.30e-44 5
#> CV:skmeans 86 4.00e-41 6
If matrix rows can be associated to genes, consider to use functional_enrichment(res,
...) to perform function enrichment for the signature genes. See this vignette for more detailed explanations.
The object with results only for a single top-value method and a single partition method can be extracted as:
res = res_list["CV", "pam"]
# you can also extract it by
# res = res_list["CV:pam"]
A summary of res and all the functions that can be applied to it:
res
#> A 'ConsensusPartition' object with k = 2, 3, 4, 5, 6.
#> On a matrix with 51941 rows and 92 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 6.
#>
#> 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 0.749 0.950 0.973 0.4989 0.500 0.500
#> 3 3 0.760 0.831 0.896 0.2020 0.917 0.834
#> 4 4 1.000 0.960 0.986 0.2139 0.805 0.558
#> 5 5 0.979 0.928 0.958 0.0464 0.953 0.829
#> 6 6 0.921 0.874 0.943 0.0273 0.967 0.865
suggest_best_k() suggests the best \(k\) based on these statistics. The rules are as follows:
suggest_best_k(res)
#> [1] 6
#> attr(,"optional")
#> [1] 4 5
There is also optional best \(k\) = 4 5 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
#> GSM587155 2 0.000 0.953 0.000 1.000
#> GSM587156 2 0.000 0.953 0.000 1.000
#> GSM587157 2 0.000 0.953 0.000 1.000
#> GSM587158 2 0.000 0.953 0.000 1.000
#> GSM587159 2 0.000 0.953 0.000 1.000
#> GSM587160 2 0.000 0.953 0.000 1.000
#> GSM587161 2 0.000 0.953 0.000 1.000
#> GSM587162 2 0.000 0.953 0.000 1.000
#> GSM587163 2 0.000 0.953 0.000 1.000
#> GSM587164 2 0.000 0.953 0.000 1.000
#> GSM587165 2 0.000 0.953 0.000 1.000
#> GSM587166 2 0.000 0.953 0.000 1.000
#> GSM587167 2 0.000 0.953 0.000 1.000
#> GSM587168 2 0.000 0.953 0.000 1.000
#> GSM587169 2 0.000 0.953 0.000 1.000
#> GSM587170 2 0.000 0.953 0.000 1.000
#> GSM587171 2 0.000 0.953 0.000 1.000
#> GSM587172 2 0.000 0.953 0.000 1.000
#> GSM587173 2 0.000 0.953 0.000 1.000
#> GSM587174 2 0.000 0.953 0.000 1.000
#> GSM587175 2 0.000 0.953 0.000 1.000
#> GSM587176 2 0.000 0.953 0.000 1.000
#> GSM587177 2 0.000 0.953 0.000 1.000
#> GSM587178 2 0.000 0.953 0.000 1.000
#> GSM587179 2 0.000 0.953 0.000 1.000
#> GSM587180 2 0.000 0.953 0.000 1.000
#> GSM587181 2 0.000 0.953 0.000 1.000
#> GSM587182 2 0.000 0.953 0.000 1.000
#> GSM587183 2 0.000 0.953 0.000 1.000
#> GSM587184 2 0.000 0.953 0.000 1.000
#> GSM587185 2 0.000 0.953 0.000 1.000
#> GSM587186 2 0.000 0.953 0.000 1.000
#> GSM587187 2 0.000 0.953 0.000 1.000
#> GSM587188 2 0.000 0.953 0.000 1.000
#> GSM587189 2 0.000 0.953 0.000 1.000
#> GSM587190 2 0.000 0.953 0.000 1.000
#> GSM587203 1 0.000 0.995 1.000 0.000
#> GSM587204 1 0.000 0.995 1.000 0.000
#> GSM587205 1 0.000 0.995 1.000 0.000
#> GSM587206 1 0.000 0.995 1.000 0.000
#> GSM587207 1 0.000 0.995 1.000 0.000
#> GSM587208 1 0.000 0.995 1.000 0.000
#> GSM587209 1 0.000 0.995 1.000 0.000
#> GSM587210 1 0.000 0.995 1.000 0.000
#> GSM587211 1 0.000 0.995 1.000 0.000
#> GSM587212 1 0.000 0.995 1.000 0.000
#> GSM587213 1 0.000 0.995 1.000 0.000
#> GSM587214 1 0.000 0.995 1.000 0.000
#> GSM587215 1 0.000 0.995 1.000 0.000
#> GSM587216 1 0.000 0.995 1.000 0.000
#> GSM587217 1 0.000 0.995 1.000 0.000
#> GSM587191 2 0.000 0.953 0.000 1.000
#> GSM587192 1 0.469 0.882 0.900 0.100
#> GSM587193 1 0.000 0.995 1.000 0.000
#> GSM587194 2 0.738 0.789 0.208 0.792
#> GSM587195 2 0.722 0.798 0.200 0.800
#> GSM587196 2 0.738 0.789 0.208 0.792
#> GSM587197 2 0.722 0.798 0.200 0.800
#> GSM587198 2 0.738 0.789 0.208 0.792
#> GSM587199 2 0.000 0.953 0.000 1.000
#> GSM587200 1 0.000 0.995 1.000 0.000
#> GSM587201 1 0.000 0.995 1.000 0.000
#> GSM587202 2 0.722 0.798 0.200 0.800
#> GSM198767 1 0.000 0.995 1.000 0.000
#> GSM198769 1 0.000 0.995 1.000 0.000
#> GSM198772 1 0.000 0.995 1.000 0.000
#> GSM198773 1 0.000 0.995 1.000 0.000
#> GSM198776 1 0.000 0.995 1.000 0.000
#> GSM198778 1 0.000 0.995 1.000 0.000
#> GSM198780 1 0.000 0.995 1.000 0.000
#> GSM198781 1 0.000 0.995 1.000 0.000
#> GSM198765 2 0.278 0.922 0.048 0.952
#> GSM198766 1 0.000 0.995 1.000 0.000
#> GSM198768 2 0.738 0.789 0.208 0.792
#> GSM198770 2 0.644 0.832 0.164 0.836
#> GSM198771 2 0.738 0.789 0.208 0.792
#> GSM198774 1 0.469 0.882 0.900 0.100
#> GSM198775 2 0.738 0.789 0.208 0.792
#> GSM198777 2 0.738 0.789 0.208 0.792
#> GSM198779 2 0.000 0.953 0.000 1.000
#> GSM587218 1 0.000 0.995 1.000 0.000
#> GSM587219 1 0.000 0.995 1.000 0.000
#> GSM587220 1 0.000 0.995 1.000 0.000
#> GSM587221 1 0.000 0.995 1.000 0.000
#> GSM587222 1 0.000 0.995 1.000 0.000
#> GSM587223 1 0.000 0.995 1.000 0.000
#> GSM587224 1 0.000 0.995 1.000 0.000
#> GSM587225 1 0.000 0.995 1.000 0.000
#> GSM587226 1 0.000 0.995 1.000 0.000
#> GSM587227 1 0.000 0.995 1.000 0.000
#> GSM587228 1 0.000 0.995 1.000 0.000
#> GSM587229 1 0.000 0.995 1.000 0.000
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM587155 2 0.0000 0.903 0.000 1.000 0.000
#> GSM587156 2 0.0000 0.903 0.000 1.000 0.000
#> GSM587157 2 0.0000 0.903 0.000 1.000 0.000
#> GSM587158 2 0.0000 0.903 0.000 1.000 0.000
#> GSM587159 2 0.0000 0.903 0.000 1.000 0.000
#> GSM587160 2 0.0000 0.903 0.000 1.000 0.000
#> GSM587161 2 0.0000 0.903 0.000 1.000 0.000
#> GSM587162 2 0.0000 0.903 0.000 1.000 0.000
#> GSM587163 2 0.0000 0.903 0.000 1.000 0.000
#> GSM587164 2 0.0000 0.903 0.000 1.000 0.000
#> GSM587165 2 0.0000 0.903 0.000 1.000 0.000
#> GSM587166 2 0.3192 0.858 0.112 0.888 0.000
#> GSM587167 2 0.0000 0.903 0.000 1.000 0.000
#> GSM587168 2 0.0000 0.903 0.000 1.000 0.000
#> GSM587169 2 0.0000 0.903 0.000 1.000 0.000
#> GSM587170 2 0.0000 0.903 0.000 1.000 0.000
#> GSM587171 2 0.0000 0.903 0.000 1.000 0.000
#> GSM587172 2 0.0000 0.903 0.000 1.000 0.000
#> GSM587173 2 0.0000 0.903 0.000 1.000 0.000
#> GSM587174 2 0.0000 0.903 0.000 1.000 0.000
#> GSM587175 2 0.0000 0.903 0.000 1.000 0.000
#> GSM587176 2 0.0000 0.903 0.000 1.000 0.000
#> GSM587177 2 0.0000 0.903 0.000 1.000 0.000
#> GSM587178 2 0.0000 0.903 0.000 1.000 0.000
#> GSM587179 2 0.0000 0.903 0.000 1.000 0.000
#> GSM587180 2 0.0000 0.903 0.000 1.000 0.000
#> GSM587181 2 0.0000 0.903 0.000 1.000 0.000
#> GSM587182 2 0.0000 0.903 0.000 1.000 0.000
#> GSM587183 2 0.0000 0.903 0.000 1.000 0.000
#> GSM587184 2 0.0000 0.903 0.000 1.000 0.000
#> GSM587185 2 0.0000 0.903 0.000 1.000 0.000
#> GSM587186 2 0.0000 0.903 0.000 1.000 0.000
#> GSM587187 2 0.0000 0.903 0.000 1.000 0.000
#> GSM587188 2 0.0000 0.903 0.000 1.000 0.000
#> GSM587189 2 0.0892 0.896 0.020 0.980 0.000
#> GSM587190 2 0.5016 0.795 0.240 0.760 0.000
#> GSM587203 1 0.5016 0.874 0.760 0.000 0.240
#> GSM587204 1 0.5016 0.874 0.760 0.000 0.240
#> GSM587205 1 0.5016 0.874 0.760 0.000 0.240
#> GSM587206 1 0.5016 0.874 0.760 0.000 0.240
#> GSM587207 1 0.5016 0.874 0.760 0.000 0.240
#> GSM587208 1 0.5016 0.874 0.760 0.000 0.240
#> GSM587209 1 0.5016 0.874 0.760 0.000 0.240
#> GSM587210 1 0.5058 0.871 0.756 0.000 0.244
#> GSM587211 1 0.5016 0.874 0.760 0.000 0.240
#> GSM587212 1 0.5058 0.871 0.756 0.000 0.244
#> GSM587213 1 0.5016 0.874 0.760 0.000 0.240
#> GSM587214 1 0.5016 0.874 0.760 0.000 0.240
#> GSM587215 1 0.5016 0.874 0.760 0.000 0.240
#> GSM587216 1 0.5058 0.871 0.756 0.000 0.244
#> GSM587217 1 0.5016 0.874 0.760 0.000 0.240
#> GSM587191 2 0.5016 0.795 0.240 0.760 0.000
#> GSM587192 1 0.6509 -0.393 0.524 0.472 0.004
#> GSM587193 1 0.0747 0.620 0.984 0.000 0.016
#> GSM587194 2 0.5678 0.745 0.316 0.684 0.000
#> GSM587195 2 0.5678 0.745 0.316 0.684 0.000
#> GSM587196 2 0.5706 0.741 0.320 0.680 0.000
#> GSM587197 2 0.5678 0.745 0.316 0.684 0.000
#> GSM587198 2 0.5678 0.745 0.316 0.684 0.000
#> GSM587199 2 0.5016 0.795 0.240 0.760 0.000
#> GSM587200 1 0.0237 0.614 0.996 0.000 0.004
#> GSM587201 1 0.0000 0.615 1.000 0.000 0.000
#> GSM587202 2 0.5678 0.745 0.316 0.684 0.000
#> GSM198767 1 0.5016 0.874 0.760 0.000 0.240
#> GSM198769 1 0.5016 0.874 0.760 0.000 0.240
#> GSM198772 1 0.5016 0.874 0.760 0.000 0.240
#> GSM198773 1 0.5016 0.874 0.760 0.000 0.240
#> GSM198776 1 0.5016 0.874 0.760 0.000 0.240
#> GSM198778 1 0.5058 0.871 0.756 0.000 0.244
#> GSM198780 1 0.5058 0.871 0.756 0.000 0.244
#> GSM198781 1 0.5016 0.874 0.760 0.000 0.240
#> GSM198765 2 0.5216 0.783 0.260 0.740 0.000
#> GSM198766 1 0.3879 0.778 0.848 0.000 0.152
#> GSM198768 2 0.5706 0.741 0.320 0.680 0.000
#> GSM198770 2 0.5621 0.751 0.308 0.692 0.000
#> GSM198771 2 0.5706 0.741 0.320 0.680 0.000
#> GSM198774 1 0.6516 -0.412 0.516 0.480 0.004
#> GSM198775 2 0.5678 0.745 0.316 0.684 0.000
#> GSM198777 2 0.5706 0.741 0.320 0.680 0.000
#> GSM198779 2 0.5016 0.795 0.240 0.760 0.000
#> GSM587218 3 0.5016 0.685 0.240 0.000 0.760
#> GSM587219 3 0.0000 0.941 0.000 0.000 1.000
#> GSM587220 3 0.0000 0.941 0.000 0.000 1.000
#> GSM587221 3 0.0000 0.941 0.000 0.000 1.000
#> GSM587222 3 0.0000 0.941 0.000 0.000 1.000
#> GSM587223 3 0.0000 0.941 0.000 0.000 1.000
#> GSM587224 3 0.4235 0.758 0.176 0.000 0.824
#> GSM587225 3 0.0000 0.941 0.000 0.000 1.000
#> GSM587226 3 0.0000 0.941 0.000 0.000 1.000
#> GSM587227 3 0.0000 0.941 0.000 0.000 1.000
#> GSM587228 3 0.0000 0.941 0.000 0.000 1.000
#> GSM587229 3 0.0000 0.941 0.000 0.000 1.000
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM587155 2 0.0000 1.000 0.000 1.000 0.000 0
#> GSM587156 2 0.0000 1.000 0.000 1.000 0.000 0
#> GSM587157 2 0.0000 1.000 0.000 1.000 0.000 0
#> GSM587158 2 0.0000 1.000 0.000 1.000 0.000 0
#> GSM587159 2 0.0000 1.000 0.000 1.000 0.000 0
#> GSM587160 2 0.0000 1.000 0.000 1.000 0.000 0
#> GSM587161 2 0.0000 1.000 0.000 1.000 0.000 0
#> GSM587162 2 0.0000 1.000 0.000 1.000 0.000 0
#> GSM587163 2 0.0000 1.000 0.000 1.000 0.000 0
#> GSM587164 2 0.0000 1.000 0.000 1.000 0.000 0
#> GSM587165 2 0.0000 1.000 0.000 1.000 0.000 0
#> GSM587166 3 0.4998 0.039 0.000 0.488 0.512 0
#> GSM587167 2 0.0000 1.000 0.000 1.000 0.000 0
#> GSM587168 2 0.0000 1.000 0.000 1.000 0.000 0
#> GSM587169 2 0.0000 1.000 0.000 1.000 0.000 0
#> GSM587170 2 0.0000 1.000 0.000 1.000 0.000 0
#> GSM587171 2 0.0000 1.000 0.000 1.000 0.000 0
#> GSM587172 2 0.0000 1.000 0.000 1.000 0.000 0
#> GSM587173 2 0.0000 1.000 0.000 1.000 0.000 0
#> GSM587174 2 0.0000 1.000 0.000 1.000 0.000 0
#> GSM587175 2 0.0000 1.000 0.000 1.000 0.000 0
#> GSM587176 2 0.0000 1.000 0.000 1.000 0.000 0
#> GSM587177 2 0.0000 1.000 0.000 1.000 0.000 0
#> GSM587178 2 0.0000 1.000 0.000 1.000 0.000 0
#> GSM587179 2 0.0000 1.000 0.000 1.000 0.000 0
#> GSM587180 2 0.0000 1.000 0.000 1.000 0.000 0
#> GSM587181 2 0.0000 1.000 0.000 1.000 0.000 0
#> GSM587182 2 0.0000 1.000 0.000 1.000 0.000 0
#> GSM587183 2 0.0000 1.000 0.000 1.000 0.000 0
#> GSM587184 2 0.0000 1.000 0.000 1.000 0.000 0
#> GSM587185 2 0.0000 1.000 0.000 1.000 0.000 0
#> GSM587186 2 0.0000 1.000 0.000 1.000 0.000 0
#> GSM587187 2 0.0000 1.000 0.000 1.000 0.000 0
#> GSM587188 2 0.0000 1.000 0.000 1.000 0.000 0
#> GSM587189 2 0.0188 0.996 0.000 0.996 0.004 0
#> GSM587190 3 0.0921 0.938 0.000 0.028 0.972 0
#> GSM587203 1 0.0000 0.961 1.000 0.000 0.000 0
#> GSM587204 1 0.0000 0.961 1.000 0.000 0.000 0
#> GSM587205 1 0.0000 0.961 1.000 0.000 0.000 0
#> GSM587206 1 0.0000 0.961 1.000 0.000 0.000 0
#> GSM587207 1 0.0000 0.961 1.000 0.000 0.000 0
#> GSM587208 1 0.0000 0.961 1.000 0.000 0.000 0
#> GSM587209 1 0.0000 0.961 1.000 0.000 0.000 0
#> GSM587210 1 0.4564 0.526 0.672 0.000 0.328 0
#> GSM587211 1 0.0000 0.961 1.000 0.000 0.000 0
#> GSM587212 1 0.0000 0.961 1.000 0.000 0.000 0
#> GSM587213 1 0.0000 0.961 1.000 0.000 0.000 0
#> GSM587214 1 0.0000 0.961 1.000 0.000 0.000 0
#> GSM587215 1 0.0000 0.961 1.000 0.000 0.000 0
#> GSM587216 1 0.0000 0.961 1.000 0.000 0.000 0
#> GSM587217 1 0.0000 0.961 1.000 0.000 0.000 0
#> GSM587191 3 0.0000 0.968 0.000 0.000 1.000 0
#> GSM587192 3 0.0000 0.968 0.000 0.000 1.000 0
#> GSM587193 3 0.0000 0.968 0.000 0.000 1.000 0
#> GSM587194 3 0.0000 0.968 0.000 0.000 1.000 0
#> GSM587195 3 0.0000 0.968 0.000 0.000 1.000 0
#> GSM587196 3 0.0000 0.968 0.000 0.000 1.000 0
#> GSM587197 3 0.0000 0.968 0.000 0.000 1.000 0
#> GSM587198 3 0.0000 0.968 0.000 0.000 1.000 0
#> GSM587199 3 0.0000 0.968 0.000 0.000 1.000 0
#> GSM587200 3 0.0000 0.968 0.000 0.000 1.000 0
#> GSM587201 3 0.0000 0.968 0.000 0.000 1.000 0
#> GSM587202 3 0.0000 0.968 0.000 0.000 1.000 0
#> GSM198767 1 0.0000 0.961 1.000 0.000 0.000 0
#> GSM198769 1 0.0000 0.961 1.000 0.000 0.000 0
#> GSM198772 1 0.0000 0.961 1.000 0.000 0.000 0
#> GSM198773 1 0.0000 0.961 1.000 0.000 0.000 0
#> GSM198776 1 0.0000 0.961 1.000 0.000 0.000 0
#> GSM198778 1 0.4916 0.299 0.576 0.000 0.424 0
#> GSM198780 1 0.0000 0.961 1.000 0.000 0.000 0
#> GSM198781 1 0.0000 0.961 1.000 0.000 0.000 0
#> GSM198765 3 0.0000 0.968 0.000 0.000 1.000 0
#> GSM198766 3 0.0817 0.944 0.024 0.000 0.976 0
#> GSM198768 3 0.0000 0.968 0.000 0.000 1.000 0
#> GSM198770 3 0.0000 0.968 0.000 0.000 1.000 0
#> GSM198771 3 0.0000 0.968 0.000 0.000 1.000 0
#> GSM198774 3 0.0000 0.968 0.000 0.000 1.000 0
#> GSM198775 3 0.0000 0.968 0.000 0.000 1.000 0
#> GSM198777 3 0.0000 0.968 0.000 0.000 1.000 0
#> GSM198779 3 0.0000 0.968 0.000 0.000 1.000 0
#> GSM587218 4 0.0000 1.000 0.000 0.000 0.000 1
#> GSM587219 4 0.0000 1.000 0.000 0.000 0.000 1
#> GSM587220 4 0.0000 1.000 0.000 0.000 0.000 1
#> GSM587221 4 0.0000 1.000 0.000 0.000 0.000 1
#> GSM587222 4 0.0000 1.000 0.000 0.000 0.000 1
#> GSM587223 4 0.0000 1.000 0.000 0.000 0.000 1
#> GSM587224 4 0.0000 1.000 0.000 0.000 0.000 1
#> GSM587225 4 0.0000 1.000 0.000 0.000 0.000 1
#> GSM587226 4 0.0000 1.000 0.000 0.000 0.000 1
#> GSM587227 4 0.0000 1.000 0.000 0.000 0.000 1
#> GSM587228 4 0.0000 1.000 0.000 0.000 0.000 1
#> GSM587229 4 0.0000 1.000 0.000 0.000 0.000 1
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM587155 2 0.0000 0.995 0.000 1.000 0.000 0 0.000
#> GSM587156 2 0.0000 0.995 0.000 1.000 0.000 0 0.000
#> GSM587157 2 0.1764 0.930 0.000 0.928 0.008 0 0.064
#> GSM587158 2 0.0000 0.995 0.000 1.000 0.000 0 0.000
#> GSM587159 2 0.0000 0.995 0.000 1.000 0.000 0 0.000
#> GSM587160 2 0.0000 0.995 0.000 1.000 0.000 0 0.000
#> GSM587161 2 0.0000 0.995 0.000 1.000 0.000 0 0.000
#> GSM587162 2 0.0000 0.995 0.000 1.000 0.000 0 0.000
#> GSM587163 2 0.0000 0.995 0.000 1.000 0.000 0 0.000
#> GSM587164 2 0.0000 0.995 0.000 1.000 0.000 0 0.000
#> GSM587165 2 0.0000 0.995 0.000 1.000 0.000 0 0.000
#> GSM587166 3 0.4294 0.111 0.000 0.468 0.532 0 0.000
#> GSM587167 2 0.0000 0.995 0.000 1.000 0.000 0 0.000
#> GSM587168 2 0.0000 0.995 0.000 1.000 0.000 0 0.000
#> GSM587169 2 0.0000 0.995 0.000 1.000 0.000 0 0.000
#> GSM587170 2 0.0000 0.995 0.000 1.000 0.000 0 0.000
#> GSM587171 2 0.0000 0.995 0.000 1.000 0.000 0 0.000
#> GSM587172 2 0.0000 0.995 0.000 1.000 0.000 0 0.000
#> GSM587173 2 0.0000 0.995 0.000 1.000 0.000 0 0.000
#> GSM587174 2 0.0000 0.995 0.000 1.000 0.000 0 0.000
#> GSM587175 2 0.1430 0.946 0.000 0.944 0.004 0 0.052
#> GSM587176 2 0.0000 0.995 0.000 1.000 0.000 0 0.000
#> GSM587177 2 0.0000 0.995 0.000 1.000 0.000 0 0.000
#> GSM587178 2 0.0000 0.995 0.000 1.000 0.000 0 0.000
#> GSM587179 2 0.0000 0.995 0.000 1.000 0.000 0 0.000
#> GSM587180 2 0.0000 0.995 0.000 1.000 0.000 0 0.000
#> GSM587181 2 0.0000 0.995 0.000 1.000 0.000 0 0.000
#> GSM587182 2 0.0000 0.995 0.000 1.000 0.000 0 0.000
#> GSM587183 2 0.0000 0.995 0.000 1.000 0.000 0 0.000
#> GSM587184 2 0.0000 0.995 0.000 1.000 0.000 0 0.000
#> GSM587185 2 0.0000 0.995 0.000 1.000 0.000 0 0.000
#> GSM587186 2 0.0000 0.995 0.000 1.000 0.000 0 0.000
#> GSM587187 2 0.0000 0.995 0.000 1.000 0.000 0 0.000
#> GSM587188 2 0.0000 0.995 0.000 1.000 0.000 0 0.000
#> GSM587189 2 0.0963 0.958 0.000 0.964 0.036 0 0.000
#> GSM587190 3 0.0963 0.918 0.000 0.036 0.964 0 0.000
#> GSM587203 5 0.1732 0.930 0.080 0.000 0.000 0 0.920
#> GSM587204 5 0.1732 0.930 0.080 0.000 0.000 0 0.920
#> GSM587205 5 0.1732 0.930 0.080 0.000 0.000 0 0.920
#> GSM587206 5 0.1732 0.930 0.080 0.000 0.000 0 0.920
#> GSM587207 5 0.1732 0.930 0.080 0.000 0.000 0 0.920
#> GSM587208 5 0.1732 0.930 0.080 0.000 0.000 0 0.920
#> GSM587209 1 0.1270 0.865 0.948 0.000 0.000 0 0.052
#> GSM587210 1 0.0000 0.898 1.000 0.000 0.000 0 0.000
#> GSM587211 1 0.0000 0.898 1.000 0.000 0.000 0 0.000
#> GSM587212 1 0.0000 0.898 1.000 0.000 0.000 0 0.000
#> GSM587213 1 0.2732 0.750 0.840 0.000 0.000 0 0.160
#> GSM587214 5 0.3999 0.622 0.344 0.000 0.000 0 0.656
#> GSM587215 1 0.0000 0.898 1.000 0.000 0.000 0 0.000
#> GSM587216 1 0.0000 0.898 1.000 0.000 0.000 0 0.000
#> GSM587217 1 0.0000 0.898 1.000 0.000 0.000 0 0.000
#> GSM587191 3 0.0510 0.939 0.000 0.000 0.984 0 0.016
#> GSM587192 3 0.0510 0.939 0.000 0.000 0.984 0 0.016
#> GSM587193 1 0.4457 0.403 0.620 0.000 0.368 0 0.012
#> GSM587194 3 0.1251 0.921 0.036 0.000 0.956 0 0.008
#> GSM587195 3 0.1478 0.923 0.000 0.000 0.936 0 0.064
#> GSM587196 3 0.1478 0.923 0.000 0.000 0.936 0 0.064
#> GSM587197 3 0.1478 0.923 0.000 0.000 0.936 0 0.064
#> GSM587198 3 0.0000 0.940 0.000 0.000 1.000 0 0.000
#> GSM587199 3 0.0000 0.940 0.000 0.000 1.000 0 0.000
#> GSM587200 3 0.0290 0.939 0.000 0.000 0.992 0 0.008
#> GSM587201 3 0.0290 0.938 0.008 0.000 0.992 0 0.000
#> GSM587202 3 0.0000 0.940 0.000 0.000 1.000 0 0.000
#> GSM198767 5 0.1732 0.930 0.080 0.000 0.000 0 0.920
#> GSM198769 1 0.1732 0.842 0.920 0.000 0.000 0 0.080
#> GSM198772 1 0.0000 0.898 1.000 0.000 0.000 0 0.000
#> GSM198773 1 0.3143 0.681 0.796 0.000 0.000 0 0.204
#> GSM198776 5 0.1732 0.930 0.080 0.000 0.000 0 0.920
#> GSM198778 1 0.0000 0.898 1.000 0.000 0.000 0 0.000
#> GSM198780 1 0.0000 0.898 1.000 0.000 0.000 0 0.000
#> GSM198781 5 0.3999 0.622 0.344 0.000 0.000 0 0.656
#> GSM198765 3 0.0510 0.939 0.000 0.000 0.984 0 0.016
#> GSM198766 1 0.3318 0.699 0.808 0.000 0.180 0 0.012
#> GSM198768 3 0.1478 0.923 0.000 0.000 0.936 0 0.064
#> GSM198770 3 0.1478 0.923 0.000 0.000 0.936 0 0.064
#> GSM198771 3 0.0000 0.940 0.000 0.000 1.000 0 0.000
#> GSM198774 3 0.0510 0.939 0.000 0.000 0.984 0 0.016
#> GSM198775 3 0.1251 0.921 0.036 0.000 0.956 0 0.008
#> GSM198777 3 0.1478 0.923 0.000 0.000 0.936 0 0.064
#> GSM198779 3 0.0000 0.940 0.000 0.000 1.000 0 0.000
#> GSM587218 4 0.0000 1.000 0.000 0.000 0.000 1 0.000
#> GSM587219 4 0.0000 1.000 0.000 0.000 0.000 1 0.000
#> GSM587220 4 0.0000 1.000 0.000 0.000 0.000 1 0.000
#> GSM587221 4 0.0000 1.000 0.000 0.000 0.000 1 0.000
#> GSM587222 4 0.0000 1.000 0.000 0.000 0.000 1 0.000
#> GSM587223 4 0.0000 1.000 0.000 0.000 0.000 1 0.000
#> GSM587224 4 0.0000 1.000 0.000 0.000 0.000 1 0.000
#> GSM587225 4 0.0000 1.000 0.000 0.000 0.000 1 0.000
#> GSM587226 4 0.0000 1.000 0.000 0.000 0.000 1 0.000
#> GSM587227 4 0.0000 1.000 0.000 0.000 0.000 1 0.000
#> GSM587228 4 0.0000 1.000 0.000 0.000 0.000 1 0.000
#> GSM587229 4 0.0000 1.000 0.000 0.000 0.000 1 0.000
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM587155 2 0.0000 0.969 0.000 1.000 0.000 0 0.000 0.000
#> GSM587156 2 0.0000 0.969 0.000 1.000 0.000 0 0.000 0.000
#> GSM587157 2 0.3629 0.625 0.000 0.712 0.276 0 0.000 0.012
#> GSM587158 2 0.0000 0.969 0.000 1.000 0.000 0 0.000 0.000
#> GSM587159 2 0.0000 0.969 0.000 1.000 0.000 0 0.000 0.000
#> GSM587160 2 0.0000 0.969 0.000 1.000 0.000 0 0.000 0.000
#> GSM587161 2 0.0000 0.969 0.000 1.000 0.000 0 0.000 0.000
#> GSM587162 2 0.0000 0.969 0.000 1.000 0.000 0 0.000 0.000
#> GSM587163 2 0.0000 0.969 0.000 1.000 0.000 0 0.000 0.000
#> GSM587164 2 0.0000 0.969 0.000 1.000 0.000 0 0.000 0.000
#> GSM587165 2 0.0000 0.969 0.000 1.000 0.000 0 0.000 0.000
#> GSM587166 2 0.4926 0.286 0.000 0.584 0.336 0 0.080 0.000
#> GSM587167 2 0.0000 0.969 0.000 1.000 0.000 0 0.000 0.000
#> GSM587168 2 0.0000 0.969 0.000 1.000 0.000 0 0.000 0.000
#> GSM587169 2 0.0000 0.969 0.000 1.000 0.000 0 0.000 0.000
#> GSM587170 2 0.0000 0.969 0.000 1.000 0.000 0 0.000 0.000
#> GSM587171 2 0.0000 0.969 0.000 1.000 0.000 0 0.000 0.000
#> GSM587172 2 0.0000 0.969 0.000 1.000 0.000 0 0.000 0.000
#> GSM587173 2 0.0000 0.969 0.000 1.000 0.000 0 0.000 0.000
#> GSM587174 2 0.0000 0.969 0.000 1.000 0.000 0 0.000 0.000
#> GSM587175 2 0.2738 0.779 0.000 0.820 0.176 0 0.000 0.004
#> GSM587176 2 0.0000 0.969 0.000 1.000 0.000 0 0.000 0.000
#> GSM587177 2 0.0000 0.969 0.000 1.000 0.000 0 0.000 0.000
#> GSM587178 2 0.0000 0.969 0.000 1.000 0.000 0 0.000 0.000
#> GSM587179 2 0.0000 0.969 0.000 1.000 0.000 0 0.000 0.000
#> GSM587180 2 0.0000 0.969 0.000 1.000 0.000 0 0.000 0.000
#> GSM587181 2 0.0000 0.969 0.000 1.000 0.000 0 0.000 0.000
#> GSM587182 2 0.0000 0.969 0.000 1.000 0.000 0 0.000 0.000
#> GSM587183 2 0.0000 0.969 0.000 1.000 0.000 0 0.000 0.000
#> GSM587184 2 0.0000 0.969 0.000 1.000 0.000 0 0.000 0.000
#> GSM587185 2 0.0000 0.969 0.000 1.000 0.000 0 0.000 0.000
#> GSM587186 2 0.0000 0.969 0.000 1.000 0.000 0 0.000 0.000
#> GSM587187 2 0.0000 0.969 0.000 1.000 0.000 0 0.000 0.000
#> GSM587188 2 0.0000 0.969 0.000 1.000 0.000 0 0.000 0.000
#> GSM587189 2 0.1765 0.874 0.000 0.904 0.096 0 0.000 0.000
#> GSM587190 3 0.3680 0.660 0.000 0.144 0.784 0 0.072 0.000
#> GSM587203 6 0.0363 0.919 0.012 0.000 0.000 0 0.000 0.988
#> GSM587204 6 0.0363 0.919 0.012 0.000 0.000 0 0.000 0.988
#> GSM587205 6 0.0363 0.919 0.012 0.000 0.000 0 0.000 0.988
#> GSM587206 6 0.0363 0.919 0.012 0.000 0.000 0 0.000 0.988
#> GSM587207 6 0.0363 0.919 0.012 0.000 0.000 0 0.000 0.988
#> GSM587208 6 0.0363 0.919 0.012 0.000 0.000 0 0.000 0.988
#> GSM587209 1 0.1556 0.848 0.920 0.000 0.000 0 0.000 0.080
#> GSM587210 1 0.1471 0.856 0.932 0.000 0.000 0 0.064 0.004
#> GSM587211 1 0.0000 0.888 1.000 0.000 0.000 0 0.000 0.000
#> GSM587212 1 0.0000 0.888 1.000 0.000 0.000 0 0.000 0.000
#> GSM587213 1 0.2941 0.695 0.780 0.000 0.000 0 0.000 0.220
#> GSM587214 6 0.3428 0.598 0.304 0.000 0.000 0 0.000 0.696
#> GSM587215 1 0.0000 0.888 1.000 0.000 0.000 0 0.000 0.000
#> GSM587216 1 0.0000 0.888 1.000 0.000 0.000 0 0.000 0.000
#> GSM587217 1 0.0000 0.888 1.000 0.000 0.000 0 0.000 0.000
#> GSM587191 5 0.0000 0.783 0.000 0.000 0.000 0 1.000 0.000
#> GSM587192 5 0.0000 0.783 0.000 0.000 0.000 0 1.000 0.000
#> GSM587193 1 0.3905 0.453 0.668 0.000 0.316 0 0.016 0.000
#> GSM587194 5 0.4372 0.237 0.024 0.000 0.432 0 0.544 0.000
#> GSM587195 3 0.0363 0.873 0.000 0.000 0.988 0 0.000 0.012
#> GSM587196 3 0.0363 0.873 0.000 0.000 0.988 0 0.000 0.012
#> GSM587197 3 0.0363 0.873 0.000 0.000 0.988 0 0.000 0.012
#> GSM587198 3 0.1556 0.869 0.000 0.000 0.920 0 0.080 0.000
#> GSM587199 3 0.1556 0.869 0.000 0.000 0.920 0 0.080 0.000
#> GSM587200 3 0.4215 0.650 0.196 0.000 0.724 0 0.080 0.000
#> GSM587201 3 0.4215 0.650 0.196 0.000 0.724 0 0.080 0.000
#> GSM587202 3 0.1556 0.869 0.000 0.000 0.920 0 0.080 0.000
#> GSM198767 6 0.0363 0.919 0.012 0.000 0.000 0 0.000 0.988
#> GSM198769 1 0.2135 0.808 0.872 0.000 0.000 0 0.000 0.128
#> GSM198772 1 0.0000 0.888 1.000 0.000 0.000 0 0.000 0.000
#> GSM198773 1 0.3221 0.622 0.736 0.000 0.000 0 0.000 0.264
#> GSM198776 6 0.0363 0.919 0.012 0.000 0.000 0 0.000 0.988
#> GSM198778 1 0.1806 0.837 0.908 0.000 0.000 0 0.088 0.004
#> GSM198780 1 0.0000 0.888 1.000 0.000 0.000 0 0.000 0.000
#> GSM198781 6 0.3428 0.598 0.304 0.000 0.000 0 0.000 0.696
#> GSM198765 5 0.0000 0.783 0.000 0.000 0.000 0 1.000 0.000
#> GSM198766 1 0.2212 0.785 0.880 0.000 0.112 0 0.008 0.000
#> GSM198768 3 0.0363 0.873 0.000 0.000 0.988 0 0.000 0.012
#> GSM198770 3 0.0363 0.873 0.000 0.000 0.988 0 0.000 0.012
#> GSM198771 3 0.1556 0.869 0.000 0.000 0.920 0 0.080 0.000
#> GSM198774 5 0.0000 0.783 0.000 0.000 0.000 0 1.000 0.000
#> GSM198775 5 0.4305 0.224 0.020 0.000 0.436 0 0.544 0.000
#> GSM198777 3 0.0363 0.873 0.000 0.000 0.988 0 0.000 0.012
#> GSM198779 3 0.1556 0.869 0.000 0.000 0.920 0 0.080 0.000
#> GSM587218 4 0.0000 1.000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587219 4 0.0000 1.000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587220 4 0.0000 1.000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587221 4 0.0000 1.000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587222 4 0.0000 1.000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587223 4 0.0000 1.000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587224 4 0.0000 1.000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587225 4 0.0000 1.000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587226 4 0.0000 1.000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587227 4 0.0000 1.000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587228 4 0.0000 1.000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587229 4 0.0000 1.000 0.000 0.000 0.000 1 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 specimen(p) k
#> CV:pam 92 4.01e-14 2
#> CV:pam 90 6.60e-29 3
#> CV:pam 90 7.34e-46 4
#> CV:pam 90 3.55e-41 5
#> CV:pam 88 3.66e-37 6
If matrix rows can be associated to genes, consider to use functional_enrichment(res,
...) to perform function enrichment for the signature genes. See this vignette for more detailed explanations.
The object with results only for a single top-value method and a single partition method can be extracted as:
res = res_list["CV", "mclust"]
# you can also extract it by
# res = res_list["CV:mclust"]
A summary of res and all the functions that can be applied to it:
res
#> A 'ConsensusPartition' object with k = 2, 3, 4, 5, 6.
#> On a matrix with 51941 rows and 92 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 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 0.955 0.970 0.986 0.4745 0.523 0.523
#> 3 3 0.939 0.953 0.979 0.2856 0.678 0.478
#> 4 4 0.987 0.956 0.982 0.1857 0.844 0.620
#> 5 5 0.898 0.833 0.908 0.0528 0.978 0.918
#> 6 6 0.873 0.815 0.898 0.0552 0.954 0.820
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 3
There is also optional best \(k\) = 2 3 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
#> GSM587155 2 0.000 0.990 0.000 1.000
#> GSM587156 2 0.000 0.990 0.000 1.000
#> GSM587157 2 0.000 0.990 0.000 1.000
#> GSM587158 2 0.000 0.990 0.000 1.000
#> GSM587159 2 0.000 0.990 0.000 1.000
#> GSM587160 2 0.000 0.990 0.000 1.000
#> GSM587161 2 0.000 0.990 0.000 1.000
#> GSM587162 2 0.000 0.990 0.000 1.000
#> GSM587163 2 0.000 0.990 0.000 1.000
#> GSM587164 2 0.000 0.990 0.000 1.000
#> GSM587165 2 0.000 0.990 0.000 1.000
#> GSM587166 2 0.000 0.990 0.000 1.000
#> GSM587167 2 0.000 0.990 0.000 1.000
#> GSM587168 2 0.000 0.990 0.000 1.000
#> GSM587169 2 0.000 0.990 0.000 1.000
#> GSM587170 2 0.000 0.990 0.000 1.000
#> GSM587171 2 0.000 0.990 0.000 1.000
#> GSM587172 2 0.000 0.990 0.000 1.000
#> GSM587173 2 0.000 0.990 0.000 1.000
#> GSM587174 2 0.000 0.990 0.000 1.000
#> GSM587175 2 0.000 0.990 0.000 1.000
#> GSM587176 2 0.000 0.990 0.000 1.000
#> GSM587177 2 0.000 0.990 0.000 1.000
#> GSM587178 2 0.000 0.990 0.000 1.000
#> GSM587179 2 0.000 0.990 0.000 1.000
#> GSM587180 2 0.000 0.990 0.000 1.000
#> GSM587181 2 0.000 0.990 0.000 1.000
#> GSM587182 2 0.000 0.990 0.000 1.000
#> GSM587183 2 0.000 0.990 0.000 1.000
#> GSM587184 2 0.000 0.990 0.000 1.000
#> GSM587185 2 0.000 0.990 0.000 1.000
#> GSM587186 2 0.000 0.990 0.000 1.000
#> GSM587187 2 0.000 0.990 0.000 1.000
#> GSM587188 2 0.000 0.990 0.000 1.000
#> GSM587189 2 0.000 0.990 0.000 1.000
#> GSM587190 2 0.000 0.990 0.000 1.000
#> GSM587203 1 0.000 0.976 1.000 0.000
#> GSM587204 1 0.000 0.976 1.000 0.000
#> GSM587205 1 0.000 0.976 1.000 0.000
#> GSM587206 1 0.000 0.976 1.000 0.000
#> GSM587207 1 0.000 0.976 1.000 0.000
#> GSM587208 1 0.000 0.976 1.000 0.000
#> GSM587209 1 0.000 0.976 1.000 0.000
#> GSM587210 1 0.745 0.748 0.788 0.212
#> GSM587211 1 0.000 0.976 1.000 0.000
#> GSM587212 1 0.697 0.782 0.812 0.188
#> GSM587213 1 0.000 0.976 1.000 0.000
#> GSM587214 1 0.000 0.976 1.000 0.000
#> GSM587215 1 0.000 0.976 1.000 0.000
#> GSM587216 1 0.000 0.976 1.000 0.000
#> GSM587217 1 0.000 0.976 1.000 0.000
#> GSM587191 2 0.000 0.990 0.000 1.000
#> GSM587192 2 0.000 0.990 0.000 1.000
#> GSM587193 2 0.584 0.839 0.140 0.860
#> GSM587194 2 0.000 0.990 0.000 1.000
#> GSM587195 2 0.000 0.990 0.000 1.000
#> GSM587196 2 0.000 0.990 0.000 1.000
#> GSM587197 2 0.000 0.990 0.000 1.000
#> GSM587198 2 0.000 0.990 0.000 1.000
#> GSM587199 2 0.000 0.990 0.000 1.000
#> GSM587200 2 0.430 0.901 0.088 0.912
#> GSM587201 2 0.615 0.823 0.152 0.848
#> GSM587202 2 0.000 0.990 0.000 1.000
#> GSM198767 1 0.000 0.976 1.000 0.000
#> GSM198769 1 0.000 0.976 1.000 0.000
#> GSM198772 1 0.000 0.976 1.000 0.000
#> GSM198773 1 0.000 0.976 1.000 0.000
#> GSM198776 1 0.000 0.976 1.000 0.000
#> GSM198778 1 0.738 0.754 0.792 0.208
#> GSM198780 1 0.697 0.782 0.812 0.188
#> GSM198781 1 0.000 0.976 1.000 0.000
#> GSM198765 2 0.000 0.990 0.000 1.000
#> GSM198766 2 0.584 0.839 0.140 0.860
#> GSM198768 2 0.000 0.990 0.000 1.000
#> GSM198770 2 0.000 0.990 0.000 1.000
#> GSM198771 2 0.000 0.990 0.000 1.000
#> GSM198774 2 0.000 0.990 0.000 1.000
#> GSM198775 2 0.000 0.990 0.000 1.000
#> GSM198777 2 0.000 0.990 0.000 1.000
#> GSM198779 2 0.000 0.990 0.000 1.000
#> GSM587218 1 0.000 0.976 1.000 0.000
#> GSM587219 1 0.000 0.976 1.000 0.000
#> GSM587220 1 0.000 0.976 1.000 0.000
#> GSM587221 1 0.000 0.976 1.000 0.000
#> GSM587222 1 0.000 0.976 1.000 0.000
#> GSM587223 1 0.000 0.976 1.000 0.000
#> GSM587224 1 0.000 0.976 1.000 0.000
#> GSM587225 1 0.000 0.976 1.000 0.000
#> GSM587226 1 0.000 0.976 1.000 0.000
#> GSM587227 1 0.000 0.976 1.000 0.000
#> GSM587228 1 0.000 0.976 1.000 0.000
#> GSM587229 1 0.000 0.976 1.000 0.000
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM587155 2 0.0000 0.951 0.000 1.000 0.000
#> GSM587156 2 0.0000 0.951 0.000 1.000 0.000
#> GSM587157 2 0.0000 0.951 0.000 1.000 0.000
#> GSM587158 2 0.0000 0.951 0.000 1.000 0.000
#> GSM587159 2 0.0000 0.951 0.000 1.000 0.000
#> GSM587160 2 0.0000 0.951 0.000 1.000 0.000
#> GSM587161 2 0.0000 0.951 0.000 1.000 0.000
#> GSM587162 2 0.0000 0.951 0.000 1.000 0.000
#> GSM587163 2 0.0000 0.951 0.000 1.000 0.000
#> GSM587164 2 0.0000 0.951 0.000 1.000 0.000
#> GSM587165 2 0.0000 0.951 0.000 1.000 0.000
#> GSM587166 2 0.0000 0.951 0.000 1.000 0.000
#> GSM587167 2 0.0000 0.951 0.000 1.000 0.000
#> GSM587168 2 0.0000 0.951 0.000 1.000 0.000
#> GSM587169 2 0.0000 0.951 0.000 1.000 0.000
#> GSM587170 2 0.0000 0.951 0.000 1.000 0.000
#> GSM587171 2 0.0000 0.951 0.000 1.000 0.000
#> GSM587172 2 0.0000 0.951 0.000 1.000 0.000
#> GSM587173 2 0.0000 0.951 0.000 1.000 0.000
#> GSM587174 2 0.0000 0.951 0.000 1.000 0.000
#> GSM587175 2 0.0000 0.951 0.000 1.000 0.000
#> GSM587176 2 0.0000 0.951 0.000 1.000 0.000
#> GSM587177 2 0.0000 0.951 0.000 1.000 0.000
#> GSM587178 2 0.0000 0.951 0.000 1.000 0.000
#> GSM587179 2 0.0000 0.951 0.000 1.000 0.000
#> GSM587180 2 0.0000 0.951 0.000 1.000 0.000
#> GSM587181 2 0.0000 0.951 0.000 1.000 0.000
#> GSM587182 2 0.0000 0.951 0.000 1.000 0.000
#> GSM587183 2 0.0000 0.951 0.000 1.000 0.000
#> GSM587184 2 0.0000 0.951 0.000 1.000 0.000
#> GSM587185 2 0.0000 0.951 0.000 1.000 0.000
#> GSM587186 2 0.0000 0.951 0.000 1.000 0.000
#> GSM587187 2 0.0000 0.951 0.000 1.000 0.000
#> GSM587188 2 0.4121 0.793 0.000 0.832 0.168
#> GSM587189 2 0.4121 0.793 0.000 0.832 0.168
#> GSM587190 2 0.4235 0.784 0.000 0.824 0.176
#> GSM587203 3 0.0592 0.982 0.012 0.000 0.988
#> GSM587204 3 0.0592 0.982 0.012 0.000 0.988
#> GSM587205 3 0.0592 0.982 0.012 0.000 0.988
#> GSM587206 3 0.0592 0.982 0.012 0.000 0.988
#> GSM587207 3 0.0592 0.982 0.012 0.000 0.988
#> GSM587208 3 0.0592 0.982 0.012 0.000 0.988
#> GSM587209 3 0.0000 0.987 0.000 0.000 1.000
#> GSM587210 3 0.0000 0.987 0.000 0.000 1.000
#> GSM587211 3 0.0000 0.987 0.000 0.000 1.000
#> GSM587212 3 0.0000 0.987 0.000 0.000 1.000
#> GSM587213 3 0.0000 0.987 0.000 0.000 1.000
#> GSM587214 3 0.0000 0.987 0.000 0.000 1.000
#> GSM587215 3 0.0000 0.987 0.000 0.000 1.000
#> GSM587216 3 0.0000 0.987 0.000 0.000 1.000
#> GSM587217 3 0.0000 0.987 0.000 0.000 1.000
#> GSM587191 3 0.3816 0.807 0.000 0.148 0.852
#> GSM587192 3 0.0000 0.987 0.000 0.000 1.000
#> GSM587193 3 0.1031 0.967 0.000 0.024 0.976
#> GSM587194 2 0.4974 0.713 0.000 0.764 0.236
#> GSM587195 3 0.0592 0.979 0.000 0.012 0.988
#> GSM587196 3 0.0000 0.987 0.000 0.000 1.000
#> GSM587197 2 0.5254 0.677 0.000 0.736 0.264
#> GSM587198 3 0.0000 0.987 0.000 0.000 1.000
#> GSM587199 3 0.0424 0.983 0.000 0.008 0.992
#> GSM587200 3 0.0000 0.987 0.000 0.000 1.000
#> GSM587201 3 0.0000 0.987 0.000 0.000 1.000
#> GSM587202 3 0.0237 0.985 0.000 0.004 0.996
#> GSM198767 3 0.0592 0.982 0.012 0.000 0.988
#> GSM198769 3 0.0000 0.987 0.000 0.000 1.000
#> GSM198772 3 0.0000 0.987 0.000 0.000 1.000
#> GSM198773 3 0.0000 0.987 0.000 0.000 1.000
#> GSM198776 3 0.0592 0.982 0.012 0.000 0.988
#> GSM198778 3 0.0000 0.987 0.000 0.000 1.000
#> GSM198780 3 0.0000 0.987 0.000 0.000 1.000
#> GSM198781 3 0.0000 0.987 0.000 0.000 1.000
#> GSM198765 3 0.2711 0.889 0.000 0.088 0.912
#> GSM198766 3 0.1031 0.967 0.000 0.024 0.976
#> GSM198768 3 0.0237 0.985 0.000 0.004 0.996
#> GSM198770 2 0.5254 0.677 0.000 0.736 0.264
#> GSM198771 3 0.0000 0.987 0.000 0.000 1.000
#> GSM198774 3 0.0000 0.987 0.000 0.000 1.000
#> GSM198775 2 0.4974 0.713 0.000 0.764 0.236
#> GSM198777 3 0.0000 0.987 0.000 0.000 1.000
#> GSM198779 3 0.0424 0.983 0.000 0.008 0.992
#> GSM587218 1 0.0000 1.000 1.000 0.000 0.000
#> GSM587219 1 0.0000 1.000 1.000 0.000 0.000
#> GSM587220 1 0.0000 1.000 1.000 0.000 0.000
#> GSM587221 1 0.0000 1.000 1.000 0.000 0.000
#> GSM587222 1 0.0000 1.000 1.000 0.000 0.000
#> GSM587223 1 0.0000 1.000 1.000 0.000 0.000
#> GSM587224 1 0.0000 1.000 1.000 0.000 0.000
#> GSM587225 1 0.0000 1.000 1.000 0.000 0.000
#> GSM587226 1 0.0000 1.000 1.000 0.000 0.000
#> GSM587227 1 0.0000 1.000 1.000 0.000 0.000
#> GSM587228 1 0.0000 1.000 1.000 0.000 0.000
#> GSM587229 1 0.0000 1.000 1.000 0.000 0.000
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM587155 2 0.0000 0.991 0.000 1.000 0.000 0
#> GSM587156 2 0.0000 0.991 0.000 1.000 0.000 0
#> GSM587157 2 0.0000 0.991 0.000 1.000 0.000 0
#> GSM587158 2 0.0000 0.991 0.000 1.000 0.000 0
#> GSM587159 2 0.0000 0.991 0.000 1.000 0.000 0
#> GSM587160 2 0.0000 0.991 0.000 1.000 0.000 0
#> GSM587161 2 0.0000 0.991 0.000 1.000 0.000 0
#> GSM587162 2 0.0000 0.991 0.000 1.000 0.000 0
#> GSM587163 2 0.0000 0.991 0.000 1.000 0.000 0
#> GSM587164 2 0.0000 0.991 0.000 1.000 0.000 0
#> GSM587165 2 0.0000 0.991 0.000 1.000 0.000 0
#> GSM587166 2 0.0592 0.977 0.000 0.984 0.016 0
#> GSM587167 2 0.0336 0.984 0.000 0.992 0.008 0
#> GSM587168 2 0.0000 0.991 0.000 1.000 0.000 0
#> GSM587169 2 0.0000 0.991 0.000 1.000 0.000 0
#> GSM587170 2 0.0000 0.991 0.000 1.000 0.000 0
#> GSM587171 2 0.0000 0.991 0.000 1.000 0.000 0
#> GSM587172 2 0.0000 0.991 0.000 1.000 0.000 0
#> GSM587173 2 0.0000 0.991 0.000 1.000 0.000 0
#> GSM587174 2 0.0000 0.991 0.000 1.000 0.000 0
#> GSM587175 2 0.0000 0.991 0.000 1.000 0.000 0
#> GSM587176 2 0.0000 0.991 0.000 1.000 0.000 0
#> GSM587177 2 0.0000 0.991 0.000 1.000 0.000 0
#> GSM587178 2 0.0000 0.991 0.000 1.000 0.000 0
#> GSM587179 2 0.0000 0.991 0.000 1.000 0.000 0
#> GSM587180 2 0.0000 0.991 0.000 1.000 0.000 0
#> GSM587181 2 0.0000 0.991 0.000 1.000 0.000 0
#> GSM587182 2 0.0000 0.991 0.000 1.000 0.000 0
#> GSM587183 2 0.0000 0.991 0.000 1.000 0.000 0
#> GSM587184 2 0.0000 0.991 0.000 1.000 0.000 0
#> GSM587185 2 0.0000 0.991 0.000 1.000 0.000 0
#> GSM587186 2 0.0000 0.991 0.000 1.000 0.000 0
#> GSM587187 2 0.1557 0.937 0.000 0.944 0.056 0
#> GSM587188 2 0.3024 0.819 0.000 0.852 0.148 0
#> GSM587189 2 0.1792 0.924 0.000 0.932 0.068 0
#> GSM587190 3 0.4925 0.300 0.000 0.428 0.572 0
#> GSM587203 1 0.0000 1.000 1.000 0.000 0.000 0
#> GSM587204 1 0.0000 1.000 1.000 0.000 0.000 0
#> GSM587205 1 0.0000 1.000 1.000 0.000 0.000 0
#> GSM587206 1 0.0000 1.000 1.000 0.000 0.000 0
#> GSM587207 1 0.0000 1.000 1.000 0.000 0.000 0
#> GSM587208 1 0.0000 1.000 1.000 0.000 0.000 0
#> GSM587209 1 0.0000 1.000 1.000 0.000 0.000 0
#> GSM587210 1 0.0000 1.000 1.000 0.000 0.000 0
#> GSM587211 1 0.0000 1.000 1.000 0.000 0.000 0
#> GSM587212 1 0.0000 1.000 1.000 0.000 0.000 0
#> GSM587213 1 0.0000 1.000 1.000 0.000 0.000 0
#> GSM587214 1 0.0000 1.000 1.000 0.000 0.000 0
#> GSM587215 1 0.0000 1.000 1.000 0.000 0.000 0
#> GSM587216 1 0.0000 1.000 1.000 0.000 0.000 0
#> GSM587217 1 0.0000 1.000 1.000 0.000 0.000 0
#> GSM587191 3 0.0000 0.917 0.000 0.000 1.000 0
#> GSM587192 3 0.0469 0.909 0.012 0.000 0.988 0
#> GSM587193 3 0.0000 0.917 0.000 0.000 1.000 0
#> GSM587194 3 0.4454 0.576 0.000 0.308 0.692 0
#> GSM587195 3 0.0000 0.917 0.000 0.000 1.000 0
#> GSM587196 3 0.0000 0.917 0.000 0.000 1.000 0
#> GSM587197 3 0.0000 0.917 0.000 0.000 1.000 0
#> GSM587198 3 0.0000 0.917 0.000 0.000 1.000 0
#> GSM587199 3 0.0000 0.917 0.000 0.000 1.000 0
#> GSM587200 3 0.3123 0.774 0.156 0.000 0.844 0
#> GSM587201 3 0.2921 0.792 0.140 0.000 0.860 0
#> GSM587202 3 0.0000 0.917 0.000 0.000 1.000 0
#> GSM198767 1 0.0000 1.000 1.000 0.000 0.000 0
#> GSM198769 1 0.0000 1.000 1.000 0.000 0.000 0
#> GSM198772 1 0.0000 1.000 1.000 0.000 0.000 0
#> GSM198773 1 0.0000 1.000 1.000 0.000 0.000 0
#> GSM198776 1 0.0000 1.000 1.000 0.000 0.000 0
#> GSM198778 1 0.0000 1.000 1.000 0.000 0.000 0
#> GSM198780 1 0.0000 1.000 1.000 0.000 0.000 0
#> GSM198781 1 0.0000 1.000 1.000 0.000 0.000 0
#> GSM198765 3 0.0000 0.917 0.000 0.000 1.000 0
#> GSM198766 3 0.0000 0.917 0.000 0.000 1.000 0
#> GSM198768 3 0.0000 0.917 0.000 0.000 1.000 0
#> GSM198770 3 0.0000 0.917 0.000 0.000 1.000 0
#> GSM198771 3 0.0000 0.917 0.000 0.000 1.000 0
#> GSM198774 3 0.0188 0.915 0.004 0.000 0.996 0
#> GSM198775 3 0.4454 0.576 0.000 0.308 0.692 0
#> GSM198777 3 0.0000 0.917 0.000 0.000 1.000 0
#> GSM198779 3 0.0000 0.917 0.000 0.000 1.000 0
#> GSM587218 4 0.0000 1.000 0.000 0.000 0.000 1
#> GSM587219 4 0.0000 1.000 0.000 0.000 0.000 1
#> GSM587220 4 0.0000 1.000 0.000 0.000 0.000 1
#> GSM587221 4 0.0000 1.000 0.000 0.000 0.000 1
#> GSM587222 4 0.0000 1.000 0.000 0.000 0.000 1
#> GSM587223 4 0.0000 1.000 0.000 0.000 0.000 1
#> GSM587224 4 0.0000 1.000 0.000 0.000 0.000 1
#> GSM587225 4 0.0000 1.000 0.000 0.000 0.000 1
#> GSM587226 4 0.0000 1.000 0.000 0.000 0.000 1
#> GSM587227 4 0.0000 1.000 0.000 0.000 0.000 1
#> GSM587228 4 0.0000 1.000 0.000 0.000 0.000 1
#> GSM587229 4 0.0000 1.000 0.000 0.000 0.000 1
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM587155 2 0.0000 0.943 0.000 1.000 0.000 0 0.000
#> GSM587156 2 0.2648 0.798 0.000 0.848 0.000 0 0.152
#> GSM587157 2 0.1121 0.912 0.000 0.956 0.000 0 0.044
#> GSM587158 2 0.0000 0.943 0.000 1.000 0.000 0 0.000
#> GSM587159 2 0.0000 0.943 0.000 1.000 0.000 0 0.000
#> GSM587160 2 0.0000 0.943 0.000 1.000 0.000 0 0.000
#> GSM587161 2 0.0000 0.943 0.000 1.000 0.000 0 0.000
#> GSM587162 2 0.0404 0.935 0.000 0.988 0.000 0 0.012
#> GSM587163 2 0.0000 0.943 0.000 1.000 0.000 0 0.000
#> GSM587164 2 0.0794 0.924 0.000 0.972 0.000 0 0.028
#> GSM587165 2 0.0000 0.943 0.000 1.000 0.000 0 0.000
#> GSM587166 2 0.2930 0.777 0.000 0.832 0.004 0 0.164
#> GSM587167 2 0.2561 0.810 0.000 0.856 0.000 0 0.144
#> GSM587168 2 0.0000 0.943 0.000 1.000 0.000 0 0.000
#> GSM587169 2 0.0000 0.943 0.000 1.000 0.000 0 0.000
#> GSM587170 2 0.0404 0.935 0.000 0.988 0.000 0 0.012
#> GSM587171 2 0.0000 0.943 0.000 1.000 0.000 0 0.000
#> GSM587172 2 0.0000 0.943 0.000 1.000 0.000 0 0.000
#> GSM587173 2 0.0000 0.943 0.000 1.000 0.000 0 0.000
#> GSM587174 2 0.0000 0.943 0.000 1.000 0.000 0 0.000
#> GSM587175 2 0.1732 0.874 0.000 0.920 0.000 0 0.080
#> GSM587176 2 0.0000 0.943 0.000 1.000 0.000 0 0.000
#> GSM587177 2 0.0000 0.943 0.000 1.000 0.000 0 0.000
#> GSM587178 2 0.0000 0.943 0.000 1.000 0.000 0 0.000
#> GSM587179 2 0.0000 0.943 0.000 1.000 0.000 0 0.000
#> GSM587180 2 0.0000 0.943 0.000 1.000 0.000 0 0.000
#> GSM587181 2 0.0000 0.943 0.000 1.000 0.000 0 0.000
#> GSM587182 2 0.0000 0.943 0.000 1.000 0.000 0 0.000
#> GSM587183 2 0.0000 0.943 0.000 1.000 0.000 0 0.000
#> GSM587184 2 0.0000 0.943 0.000 1.000 0.000 0 0.000
#> GSM587185 2 0.0000 0.943 0.000 1.000 0.000 0 0.000
#> GSM587186 2 0.0000 0.943 0.000 1.000 0.000 0 0.000
#> GSM587187 2 0.5646 -0.118 0.000 0.520 0.080 0 0.400
#> GSM587188 5 0.5845 0.453 0.000 0.352 0.108 0 0.540
#> GSM587189 2 0.5775 -0.277 0.000 0.472 0.088 0 0.440
#> GSM587190 5 0.5373 0.676 0.000 0.092 0.276 0 0.632
#> GSM587203 1 0.3452 0.804 0.756 0.000 0.000 0 0.244
#> GSM587204 1 0.3424 0.803 0.760 0.000 0.000 0 0.240
#> GSM587205 1 0.3534 0.798 0.744 0.000 0.000 0 0.256
#> GSM587206 1 0.3534 0.798 0.744 0.000 0.000 0 0.256
#> GSM587207 1 0.3534 0.798 0.744 0.000 0.000 0 0.256
#> GSM587208 1 0.3534 0.798 0.744 0.000 0.000 0 0.256
#> GSM587209 1 0.1608 0.865 0.928 0.000 0.000 0 0.072
#> GSM587210 1 0.2964 0.837 0.856 0.000 0.024 0 0.120
#> GSM587211 1 0.1608 0.865 0.928 0.000 0.000 0 0.072
#> GSM587212 1 0.2915 0.839 0.860 0.000 0.024 0 0.116
#> GSM587213 1 0.0000 0.871 1.000 0.000 0.000 0 0.000
#> GSM587214 1 0.0000 0.871 1.000 0.000 0.000 0 0.000
#> GSM587215 1 0.2304 0.854 0.892 0.000 0.008 0 0.100
#> GSM587216 1 0.2017 0.860 0.912 0.000 0.008 0 0.080
#> GSM587217 1 0.0000 0.871 1.000 0.000 0.000 0 0.000
#> GSM587191 3 0.0609 0.857 0.000 0.000 0.980 0 0.020
#> GSM587192 3 0.1197 0.848 0.000 0.000 0.952 0 0.048
#> GSM587193 3 0.4397 -0.058 0.004 0.000 0.564 0 0.432
#> GSM587194 5 0.5423 0.641 0.000 0.064 0.388 0 0.548
#> GSM587195 3 0.0404 0.859 0.000 0.000 0.988 0 0.012
#> GSM587196 3 0.0162 0.859 0.000 0.000 0.996 0 0.004
#> GSM587197 3 0.2020 0.803 0.000 0.000 0.900 0 0.100
#> GSM587198 3 0.0609 0.854 0.000 0.000 0.980 0 0.020
#> GSM587199 3 0.0609 0.854 0.000 0.000 0.980 0 0.020
#> GSM587200 3 0.4333 0.583 0.188 0.000 0.752 0 0.060
#> GSM587201 3 0.4367 0.575 0.192 0.000 0.748 0 0.060
#> GSM587202 3 0.0290 0.858 0.000 0.000 0.992 0 0.008
#> GSM198767 1 0.3534 0.798 0.744 0.000 0.000 0 0.256
#> GSM198769 1 0.1608 0.865 0.928 0.000 0.000 0 0.072
#> GSM198772 1 0.1608 0.865 0.928 0.000 0.000 0 0.072
#> GSM198773 1 0.0000 0.871 1.000 0.000 0.000 0 0.000
#> GSM198776 1 0.3424 0.803 0.760 0.000 0.000 0 0.240
#> GSM198778 1 0.3002 0.836 0.856 0.000 0.028 0 0.116
#> GSM198780 1 0.2964 0.837 0.856 0.000 0.024 0 0.120
#> GSM198781 1 0.0000 0.871 1.000 0.000 0.000 0 0.000
#> GSM198765 3 0.0609 0.857 0.000 0.000 0.980 0 0.020
#> GSM198766 3 0.4397 -0.058 0.004 0.000 0.564 0 0.432
#> GSM198768 3 0.0290 0.858 0.000 0.000 0.992 0 0.008
#> GSM198770 3 0.2127 0.795 0.000 0.000 0.892 0 0.108
#> GSM198771 3 0.0510 0.855 0.000 0.000 0.984 0 0.016
#> GSM198774 3 0.1197 0.848 0.000 0.000 0.952 0 0.048
#> GSM198775 5 0.5423 0.641 0.000 0.064 0.388 0 0.548
#> GSM198777 3 0.0000 0.859 0.000 0.000 1.000 0 0.000
#> GSM198779 3 0.0609 0.854 0.000 0.000 0.980 0 0.020
#> GSM587218 4 0.0000 1.000 0.000 0.000 0.000 1 0.000
#> GSM587219 4 0.0000 1.000 0.000 0.000 0.000 1 0.000
#> GSM587220 4 0.0000 1.000 0.000 0.000 0.000 1 0.000
#> GSM587221 4 0.0000 1.000 0.000 0.000 0.000 1 0.000
#> GSM587222 4 0.0000 1.000 0.000 0.000 0.000 1 0.000
#> GSM587223 4 0.0000 1.000 0.000 0.000 0.000 1 0.000
#> GSM587224 4 0.0000 1.000 0.000 0.000 0.000 1 0.000
#> GSM587225 4 0.0000 1.000 0.000 0.000 0.000 1 0.000
#> GSM587226 4 0.0000 1.000 0.000 0.000 0.000 1 0.000
#> GSM587227 4 0.0000 1.000 0.000 0.000 0.000 1 0.000
#> GSM587228 4 0.0000 1.000 0.000 0.000 0.000 1 0.000
#> GSM587229 4 0.0000 1.000 0.000 0.000 0.000 1 0.000
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM587155 2 0.1765 0.8588 0.000 0.904 0.000 0 0.096 0.000
#> GSM587156 2 0.3817 0.4355 0.000 0.568 0.000 0 0.432 0.000
#> GSM587157 2 0.3819 0.5956 0.000 0.652 0.000 0 0.340 0.008
#> GSM587158 2 0.0000 0.9059 0.000 1.000 0.000 0 0.000 0.000
#> GSM587159 2 0.0000 0.9059 0.000 1.000 0.000 0 0.000 0.000
#> GSM587160 2 0.0000 0.9059 0.000 1.000 0.000 0 0.000 0.000
#> GSM587161 2 0.0000 0.9059 0.000 1.000 0.000 0 0.000 0.000
#> GSM587162 2 0.2491 0.8025 0.000 0.836 0.000 0 0.164 0.000
#> GSM587163 2 0.0000 0.9059 0.000 1.000 0.000 0 0.000 0.000
#> GSM587164 2 0.3330 0.6812 0.000 0.716 0.000 0 0.284 0.000
#> GSM587165 2 0.0000 0.9059 0.000 1.000 0.000 0 0.000 0.000
#> GSM587166 2 0.3828 0.4181 0.000 0.560 0.000 0 0.440 0.000
#> GSM587167 2 0.3823 0.4281 0.000 0.564 0.000 0 0.436 0.000
#> GSM587168 2 0.0000 0.9059 0.000 1.000 0.000 0 0.000 0.000
#> GSM587169 2 0.0000 0.9059 0.000 1.000 0.000 0 0.000 0.000
#> GSM587170 2 0.1444 0.8740 0.000 0.928 0.000 0 0.072 0.000
#> GSM587171 2 0.0000 0.9059 0.000 1.000 0.000 0 0.000 0.000
#> GSM587172 2 0.0000 0.9059 0.000 1.000 0.000 0 0.000 0.000
#> GSM587173 2 0.0000 0.9059 0.000 1.000 0.000 0 0.000 0.000
#> GSM587174 2 0.0000 0.9059 0.000 1.000 0.000 0 0.000 0.000
#> GSM587175 2 0.3684 0.5398 0.000 0.628 0.000 0 0.372 0.000
#> GSM587176 2 0.0000 0.9059 0.000 1.000 0.000 0 0.000 0.000
#> GSM587177 2 0.0000 0.9059 0.000 1.000 0.000 0 0.000 0.000
#> GSM587178 2 0.0000 0.9059 0.000 1.000 0.000 0 0.000 0.000
#> GSM587179 2 0.0000 0.9059 0.000 1.000 0.000 0 0.000 0.000
#> GSM587180 2 0.1714 0.8618 0.000 0.908 0.000 0 0.092 0.000
#> GSM587181 2 0.0260 0.9037 0.000 0.992 0.000 0 0.008 0.000
#> GSM587182 2 0.0000 0.9059 0.000 1.000 0.000 0 0.000 0.000
#> GSM587183 2 0.1387 0.8771 0.000 0.932 0.000 0 0.068 0.000
#> GSM587184 2 0.0146 0.9045 0.000 0.996 0.000 0 0.004 0.000
#> GSM587185 2 0.0632 0.8974 0.000 0.976 0.000 0 0.024 0.000
#> GSM587186 2 0.0363 0.9025 0.000 0.988 0.000 0 0.012 0.000
#> GSM587187 5 0.1204 0.8230 0.000 0.056 0.000 0 0.944 0.000
#> GSM587188 5 0.1152 0.8309 0.000 0.044 0.000 0 0.952 0.004
#> GSM587189 5 0.0865 0.8301 0.000 0.036 0.000 0 0.964 0.000
#> GSM587190 5 0.2213 0.8052 0.000 0.004 0.100 0 0.888 0.008
#> GSM587203 6 0.1556 0.9880 0.080 0.000 0.000 0 0.000 0.920
#> GSM587204 6 0.1753 0.9817 0.084 0.000 0.000 0 0.004 0.912
#> GSM587205 6 0.1444 0.9940 0.072 0.000 0.000 0 0.000 0.928
#> GSM587206 6 0.1444 0.9940 0.072 0.000 0.000 0 0.000 0.928
#> GSM587207 6 0.1444 0.9940 0.072 0.000 0.000 0 0.000 0.928
#> GSM587208 6 0.1444 0.9940 0.072 0.000 0.000 0 0.000 0.928
#> GSM587209 1 0.0858 0.8125 0.968 0.000 0.000 0 0.004 0.028
#> GSM587210 1 0.0000 0.8090 1.000 0.000 0.000 0 0.000 0.000
#> GSM587211 1 0.0632 0.8141 0.976 0.000 0.000 0 0.000 0.024
#> GSM587212 1 0.0000 0.8090 1.000 0.000 0.000 0 0.000 0.000
#> GSM587213 1 0.3830 0.4894 0.620 0.000 0.000 0 0.004 0.376
#> GSM587214 1 0.3955 0.3823 0.560 0.000 0.000 0 0.004 0.436
#> GSM587215 1 0.1082 0.8106 0.956 0.000 0.000 0 0.004 0.040
#> GSM587216 1 0.0363 0.8126 0.988 0.000 0.000 0 0.000 0.012
#> GSM587217 1 0.3747 0.4564 0.604 0.000 0.000 0 0.000 0.396
#> GSM587191 3 0.0777 0.8440 0.004 0.000 0.972 0 0.024 0.000
#> GSM587192 3 0.3537 0.7873 0.040 0.000 0.824 0 0.104 0.032
#> GSM587193 3 0.5613 0.0683 0.036 0.000 0.468 0 0.436 0.060
#> GSM587194 5 0.4120 0.6968 0.008 0.000 0.228 0 0.724 0.040
#> GSM587195 3 0.0458 0.8458 0.000 0.000 0.984 0 0.016 0.000
#> GSM587196 3 0.0260 0.8451 0.008 0.000 0.992 0 0.000 0.000
#> GSM587197 3 0.3788 0.6686 0.008 0.000 0.740 0 0.232 0.020
#> GSM587198 3 0.0260 0.8451 0.008 0.000 0.992 0 0.000 0.000
#> GSM587199 3 0.0547 0.8433 0.020 0.000 0.980 0 0.000 0.000
#> GSM587200 3 0.4661 0.7122 0.136 0.000 0.732 0 0.104 0.028
#> GSM587201 3 0.4842 0.6949 0.148 0.000 0.716 0 0.104 0.032
#> GSM587202 3 0.0000 0.8446 0.000 0.000 1.000 0 0.000 0.000
#> GSM198767 6 0.1444 0.9940 0.072 0.000 0.000 0 0.000 0.928
#> GSM198769 1 0.0777 0.8136 0.972 0.000 0.000 0 0.004 0.024
#> GSM198772 1 0.0632 0.8141 0.976 0.000 0.000 0 0.000 0.024
#> GSM198773 1 0.3955 0.3823 0.560 0.000 0.000 0 0.004 0.436
#> GSM198776 6 0.1644 0.9890 0.076 0.000 0.000 0 0.004 0.920
#> GSM198778 1 0.0000 0.8090 1.000 0.000 0.000 0 0.000 0.000
#> GSM198780 1 0.0000 0.8090 1.000 0.000 0.000 0 0.000 0.000
#> GSM198781 1 0.3955 0.3823 0.560 0.000 0.000 0 0.004 0.436
#> GSM198765 3 0.0692 0.8447 0.004 0.000 0.976 0 0.020 0.000
#> GSM198766 3 0.5613 0.0683 0.036 0.000 0.468 0 0.436 0.060
#> GSM198768 3 0.0146 0.8454 0.000 0.000 0.996 0 0.004 0.000
#> GSM198770 3 0.3788 0.6686 0.008 0.000 0.740 0 0.232 0.020
#> GSM198771 3 0.0260 0.8451 0.008 0.000 0.992 0 0.000 0.000
#> GSM198774 3 0.3463 0.7886 0.040 0.000 0.828 0 0.104 0.028
#> GSM198775 5 0.4120 0.6968 0.008 0.000 0.228 0 0.724 0.040
#> GSM198777 3 0.0000 0.8446 0.000 0.000 1.000 0 0.000 0.000
#> GSM198779 3 0.0547 0.8433 0.020 0.000 0.980 0 0.000 0.000
#> GSM587218 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587219 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587220 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587221 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587222 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587223 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587224 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587225 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587226 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587227 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587228 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587229 4 0.0000 1.0000 0.000 0.000 0.000 1 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 specimen(p) k
#> CV:mclust 92 1.16e-17 2
#> CV:mclust 92 7.03e-30 3
#> CV:mclust 91 1.66e-47 4
#> CV:mclust 87 6.47e-47 5
#> CV:mclust 82 1.00e-45 6
If matrix rows can be associated to genes, consider to use functional_enrichment(res,
...) to perform function enrichment for the signature genes. See this vignette for more detailed explanations.
The object with results only for a single top-value method and a single partition method can be extracted as:
res = res_list["CV", "NMF"]
# you can also extract it by
# res = res_list["CV:NMF"]
A summary of res and all the functions that can be applied to it:
res
#> A 'ConsensusPartition' object with k = 2, 3, 4, 5, 6.
#> On a matrix with 51941 rows and 92 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 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 1.000 0.991 0.996 0.5007 0.500 0.500
#> 3 3 0.776 0.925 0.944 0.2185 0.888 0.778
#> 4 4 0.989 0.947 0.979 0.1895 0.815 0.568
#> 5 5 0.949 0.876 0.944 0.0598 0.934 0.768
#> 6 6 0.850 0.705 0.823 0.0369 0.952 0.799
suggest_best_k() suggests the best \(k\) based on these statistics. The rules are as follows:
suggest_best_k(res)
#> [1] 5
#> attr(,"optional")
#> [1] 2 4
There is also optional best \(k\) = 2 4 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
#> GSM587155 2 0.0000 0.993 0.000 1.000
#> GSM587156 2 0.0000 0.993 0.000 1.000
#> GSM587157 2 0.0000 0.993 0.000 1.000
#> GSM587158 2 0.0000 0.993 0.000 1.000
#> GSM587159 2 0.0000 0.993 0.000 1.000
#> GSM587160 2 0.0000 0.993 0.000 1.000
#> GSM587161 2 0.0000 0.993 0.000 1.000
#> GSM587162 2 0.0000 0.993 0.000 1.000
#> GSM587163 2 0.0000 0.993 0.000 1.000
#> GSM587164 2 0.0000 0.993 0.000 1.000
#> GSM587165 2 0.0000 0.993 0.000 1.000
#> GSM587166 2 0.0000 0.993 0.000 1.000
#> GSM587167 2 0.0000 0.993 0.000 1.000
#> GSM587168 2 0.0000 0.993 0.000 1.000
#> GSM587169 2 0.0000 0.993 0.000 1.000
#> GSM587170 2 0.0000 0.993 0.000 1.000
#> GSM587171 2 0.0000 0.993 0.000 1.000
#> GSM587172 2 0.0000 0.993 0.000 1.000
#> GSM587173 2 0.0000 0.993 0.000 1.000
#> GSM587174 2 0.0000 0.993 0.000 1.000
#> GSM587175 2 0.0000 0.993 0.000 1.000
#> GSM587176 2 0.0000 0.993 0.000 1.000
#> GSM587177 2 0.0000 0.993 0.000 1.000
#> GSM587178 2 0.0000 0.993 0.000 1.000
#> GSM587179 2 0.0000 0.993 0.000 1.000
#> GSM587180 2 0.0000 0.993 0.000 1.000
#> GSM587181 2 0.0000 0.993 0.000 1.000
#> GSM587182 2 0.0000 0.993 0.000 1.000
#> GSM587183 2 0.0000 0.993 0.000 1.000
#> GSM587184 2 0.0000 0.993 0.000 1.000
#> GSM587185 2 0.0000 0.993 0.000 1.000
#> GSM587186 2 0.0000 0.993 0.000 1.000
#> GSM587187 2 0.0000 0.993 0.000 1.000
#> GSM587188 2 0.0000 0.993 0.000 1.000
#> GSM587189 2 0.0000 0.993 0.000 1.000
#> GSM587190 2 0.0000 0.993 0.000 1.000
#> GSM587203 1 0.0000 1.000 1.000 0.000
#> GSM587204 1 0.0000 1.000 1.000 0.000
#> GSM587205 1 0.0000 1.000 1.000 0.000
#> GSM587206 1 0.0000 1.000 1.000 0.000
#> GSM587207 1 0.0000 1.000 1.000 0.000
#> GSM587208 1 0.0000 1.000 1.000 0.000
#> GSM587209 1 0.0000 1.000 1.000 0.000
#> GSM587210 1 0.0000 1.000 1.000 0.000
#> GSM587211 1 0.0000 1.000 1.000 0.000
#> GSM587212 1 0.0000 1.000 1.000 0.000
#> GSM587213 1 0.0000 1.000 1.000 0.000
#> GSM587214 1 0.0000 1.000 1.000 0.000
#> GSM587215 1 0.0000 1.000 1.000 0.000
#> GSM587216 1 0.0000 1.000 1.000 0.000
#> GSM587217 1 0.0000 1.000 1.000 0.000
#> GSM587191 2 0.0000 0.993 0.000 1.000
#> GSM587192 1 0.0000 1.000 1.000 0.000
#> GSM587193 1 0.0000 1.000 1.000 0.000
#> GSM587194 2 0.0672 0.986 0.008 0.992
#> GSM587195 2 0.0000 0.993 0.000 1.000
#> GSM587196 2 0.8443 0.631 0.272 0.728
#> GSM587197 2 0.0000 0.993 0.000 1.000
#> GSM587198 2 0.0000 0.993 0.000 1.000
#> GSM587199 2 0.0000 0.993 0.000 1.000
#> GSM587200 1 0.0000 1.000 1.000 0.000
#> GSM587201 1 0.0000 1.000 1.000 0.000
#> GSM587202 2 0.0000 0.993 0.000 1.000
#> GSM198767 1 0.0000 1.000 1.000 0.000
#> GSM198769 1 0.0000 1.000 1.000 0.000
#> GSM198772 1 0.0000 1.000 1.000 0.000
#> GSM198773 1 0.0000 1.000 1.000 0.000
#> GSM198776 1 0.0000 1.000 1.000 0.000
#> GSM198778 1 0.0000 1.000 1.000 0.000
#> GSM198780 1 0.0000 1.000 1.000 0.000
#> GSM198781 1 0.0000 1.000 1.000 0.000
#> GSM198765 2 0.0000 0.993 0.000 1.000
#> GSM198766 1 0.0000 1.000 1.000 0.000
#> GSM198768 2 0.0000 0.993 0.000 1.000
#> GSM198770 2 0.0000 0.993 0.000 1.000
#> GSM198771 2 0.0000 0.993 0.000 1.000
#> GSM198774 1 0.0000 1.000 1.000 0.000
#> GSM198775 2 0.0376 0.989 0.004 0.996
#> GSM198777 2 0.3733 0.920 0.072 0.928
#> GSM198779 2 0.0000 0.993 0.000 1.000
#> GSM587218 1 0.0000 1.000 1.000 0.000
#> GSM587219 1 0.0000 1.000 1.000 0.000
#> GSM587220 1 0.0000 1.000 1.000 0.000
#> GSM587221 1 0.0000 1.000 1.000 0.000
#> GSM587222 1 0.0000 1.000 1.000 0.000
#> GSM587223 1 0.0000 1.000 1.000 0.000
#> GSM587224 1 0.0000 1.000 1.000 0.000
#> GSM587225 1 0.0000 1.000 1.000 0.000
#> GSM587226 1 0.0000 1.000 1.000 0.000
#> GSM587227 1 0.0000 1.000 1.000 0.000
#> GSM587228 1 0.0000 1.000 1.000 0.000
#> GSM587229 1 0.0000 1.000 1.000 0.000
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM587155 2 0.0000 0.954 0.000 1.000 0.000
#> GSM587156 2 0.0000 0.954 0.000 1.000 0.000
#> GSM587157 2 0.0237 0.952 0.000 0.996 0.004
#> GSM587158 2 0.0000 0.954 0.000 1.000 0.000
#> GSM587159 2 0.0000 0.954 0.000 1.000 0.000
#> GSM587160 2 0.0000 0.954 0.000 1.000 0.000
#> GSM587161 2 0.0000 0.954 0.000 1.000 0.000
#> GSM587162 2 0.0000 0.954 0.000 1.000 0.000
#> GSM587163 2 0.0000 0.954 0.000 1.000 0.000
#> GSM587164 2 0.0000 0.954 0.000 1.000 0.000
#> GSM587165 2 0.0000 0.954 0.000 1.000 0.000
#> GSM587166 2 0.0000 0.954 0.000 1.000 0.000
#> GSM587167 2 0.0000 0.954 0.000 1.000 0.000
#> GSM587168 2 0.0000 0.954 0.000 1.000 0.000
#> GSM587169 2 0.0000 0.954 0.000 1.000 0.000
#> GSM587170 2 0.0000 0.954 0.000 1.000 0.000
#> GSM587171 2 0.0000 0.954 0.000 1.000 0.000
#> GSM587172 2 0.0000 0.954 0.000 1.000 0.000
#> GSM587173 2 0.0000 0.954 0.000 1.000 0.000
#> GSM587174 2 0.0000 0.954 0.000 1.000 0.000
#> GSM587175 2 0.0000 0.954 0.000 1.000 0.000
#> GSM587176 2 0.0000 0.954 0.000 1.000 0.000
#> GSM587177 2 0.0000 0.954 0.000 1.000 0.000
#> GSM587178 2 0.0000 0.954 0.000 1.000 0.000
#> GSM587179 2 0.0000 0.954 0.000 1.000 0.000
#> GSM587180 2 0.0000 0.954 0.000 1.000 0.000
#> GSM587181 2 0.0000 0.954 0.000 1.000 0.000
#> GSM587182 2 0.0000 0.954 0.000 1.000 0.000
#> GSM587183 2 0.0000 0.954 0.000 1.000 0.000
#> GSM587184 2 0.0000 0.954 0.000 1.000 0.000
#> GSM587185 2 0.0000 0.954 0.000 1.000 0.000
#> GSM587186 2 0.0000 0.954 0.000 1.000 0.000
#> GSM587187 2 0.0000 0.954 0.000 1.000 0.000
#> GSM587188 2 0.0000 0.954 0.000 1.000 0.000
#> GSM587189 2 0.0000 0.954 0.000 1.000 0.000
#> GSM587190 2 0.1031 0.942 0.000 0.976 0.024
#> GSM587203 1 0.0000 0.946 1.000 0.000 0.000
#> GSM587204 1 0.0000 0.946 1.000 0.000 0.000
#> GSM587205 1 0.0000 0.946 1.000 0.000 0.000
#> GSM587206 1 0.0000 0.946 1.000 0.000 0.000
#> GSM587207 1 0.0000 0.946 1.000 0.000 0.000
#> GSM587208 1 0.0000 0.946 1.000 0.000 0.000
#> GSM587209 1 0.0000 0.946 1.000 0.000 0.000
#> GSM587210 1 0.3192 0.889 0.888 0.000 0.112
#> GSM587211 1 0.0000 0.946 1.000 0.000 0.000
#> GSM587212 1 0.0424 0.944 0.992 0.000 0.008
#> GSM587213 1 0.0000 0.946 1.000 0.000 0.000
#> GSM587214 1 0.0000 0.946 1.000 0.000 0.000
#> GSM587215 1 0.0000 0.946 1.000 0.000 0.000
#> GSM587216 1 0.2537 0.908 0.920 0.000 0.080
#> GSM587217 1 0.0000 0.946 1.000 0.000 0.000
#> GSM587191 2 0.3619 0.879 0.000 0.864 0.136
#> GSM587192 1 0.3619 0.872 0.864 0.000 0.136
#> GSM587193 1 0.3619 0.872 0.864 0.000 0.136
#> GSM587194 3 0.1643 0.800 0.000 0.044 0.956
#> GSM587195 2 0.3619 0.879 0.000 0.864 0.136
#> GSM587196 2 0.8399 0.524 0.256 0.608 0.136
#> GSM587197 2 0.3619 0.879 0.000 0.864 0.136
#> GSM587198 2 0.3619 0.879 0.000 0.864 0.136
#> GSM587199 2 0.4002 0.862 0.000 0.840 0.160
#> GSM587200 1 0.3619 0.872 0.864 0.000 0.136
#> GSM587201 1 0.3619 0.872 0.864 0.000 0.136
#> GSM587202 2 0.3619 0.879 0.000 0.864 0.136
#> GSM198767 1 0.0000 0.946 1.000 0.000 0.000
#> GSM198769 1 0.0000 0.946 1.000 0.000 0.000
#> GSM198772 1 0.0000 0.946 1.000 0.000 0.000
#> GSM198773 1 0.0000 0.946 1.000 0.000 0.000
#> GSM198776 1 0.0000 0.946 1.000 0.000 0.000
#> GSM198778 1 0.3412 0.881 0.876 0.000 0.124
#> GSM198780 1 0.0892 0.938 0.980 0.000 0.020
#> GSM198781 1 0.0000 0.946 1.000 0.000 0.000
#> GSM198765 2 0.3851 0.877 0.004 0.860 0.136
#> GSM198766 1 0.3619 0.872 0.864 0.000 0.136
#> GSM198768 2 0.3619 0.879 0.000 0.864 0.136
#> GSM198770 2 0.3619 0.879 0.000 0.864 0.136
#> GSM198771 2 0.4345 0.867 0.016 0.848 0.136
#> GSM198774 1 0.3619 0.872 0.864 0.000 0.136
#> GSM198775 3 0.3038 0.742 0.000 0.104 0.896
#> GSM198777 2 0.7104 0.719 0.140 0.724 0.136
#> GSM198779 2 0.3686 0.877 0.000 0.860 0.140
#> GSM587218 3 0.3619 0.963 0.136 0.000 0.864
#> GSM587219 3 0.3619 0.963 0.136 0.000 0.864
#> GSM587220 3 0.3619 0.963 0.136 0.000 0.864
#> GSM587221 3 0.3619 0.963 0.136 0.000 0.864
#> GSM587222 3 0.3619 0.963 0.136 0.000 0.864
#> GSM587223 3 0.3619 0.963 0.136 0.000 0.864
#> GSM587224 3 0.3619 0.963 0.136 0.000 0.864
#> GSM587225 3 0.3619 0.963 0.136 0.000 0.864
#> GSM587226 3 0.3619 0.963 0.136 0.000 0.864
#> GSM587227 3 0.3619 0.963 0.136 0.000 0.864
#> GSM587228 3 0.3619 0.963 0.136 0.000 0.864
#> GSM587229 3 0.3619 0.963 0.136 0.000 0.864
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM587155 2 0.0000 0.993 0.000 1.000 0.000 0.000
#> GSM587156 2 0.0000 0.993 0.000 1.000 0.000 0.000
#> GSM587157 2 0.3610 0.748 0.000 0.800 0.200 0.000
#> GSM587158 2 0.0000 0.993 0.000 1.000 0.000 0.000
#> GSM587159 2 0.0000 0.993 0.000 1.000 0.000 0.000
#> GSM587160 2 0.0000 0.993 0.000 1.000 0.000 0.000
#> GSM587161 2 0.0000 0.993 0.000 1.000 0.000 0.000
#> GSM587162 2 0.0000 0.993 0.000 1.000 0.000 0.000
#> GSM587163 2 0.0000 0.993 0.000 1.000 0.000 0.000
#> GSM587164 2 0.0469 0.981 0.000 0.988 0.012 0.000
#> GSM587165 2 0.0000 0.993 0.000 1.000 0.000 0.000
#> GSM587166 2 0.0000 0.993 0.000 1.000 0.000 0.000
#> GSM587167 2 0.0000 0.993 0.000 1.000 0.000 0.000
#> GSM587168 2 0.0000 0.993 0.000 1.000 0.000 0.000
#> GSM587169 2 0.0000 0.993 0.000 1.000 0.000 0.000
#> GSM587170 2 0.0000 0.993 0.000 1.000 0.000 0.000
#> GSM587171 2 0.0000 0.993 0.000 1.000 0.000 0.000
#> GSM587172 2 0.0000 0.993 0.000 1.000 0.000 0.000
#> GSM587173 2 0.0000 0.993 0.000 1.000 0.000 0.000
#> GSM587174 2 0.0000 0.993 0.000 1.000 0.000 0.000
#> GSM587175 2 0.0000 0.993 0.000 1.000 0.000 0.000
#> GSM587176 2 0.0000 0.993 0.000 1.000 0.000 0.000
#> GSM587177 2 0.0000 0.993 0.000 1.000 0.000 0.000
#> GSM587178 2 0.0000 0.993 0.000 1.000 0.000 0.000
#> GSM587179 2 0.0000 0.993 0.000 1.000 0.000 0.000
#> GSM587180 2 0.0000 0.993 0.000 1.000 0.000 0.000
#> GSM587181 2 0.0000 0.993 0.000 1.000 0.000 0.000
#> GSM587182 2 0.0000 0.993 0.000 1.000 0.000 0.000
#> GSM587183 2 0.0000 0.993 0.000 1.000 0.000 0.000
#> GSM587184 2 0.0000 0.993 0.000 1.000 0.000 0.000
#> GSM587185 2 0.0000 0.993 0.000 1.000 0.000 0.000
#> GSM587186 2 0.0000 0.993 0.000 1.000 0.000 0.000
#> GSM587187 2 0.0000 0.993 0.000 1.000 0.000 0.000
#> GSM587188 2 0.0000 0.993 0.000 1.000 0.000 0.000
#> GSM587189 2 0.0000 0.993 0.000 1.000 0.000 0.000
#> GSM587190 3 0.4431 0.557 0.000 0.304 0.696 0.000
#> GSM587203 1 0.0000 0.930 1.000 0.000 0.000 0.000
#> GSM587204 1 0.0000 0.930 1.000 0.000 0.000 0.000
#> GSM587205 1 0.0000 0.930 1.000 0.000 0.000 0.000
#> GSM587206 1 0.0000 0.930 1.000 0.000 0.000 0.000
#> GSM587207 1 0.0000 0.930 1.000 0.000 0.000 0.000
#> GSM587208 1 0.0000 0.930 1.000 0.000 0.000 0.000
#> GSM587209 1 0.0000 0.930 1.000 0.000 0.000 0.000
#> GSM587210 1 0.4955 0.276 0.556 0.000 0.444 0.000
#> GSM587211 1 0.0000 0.930 1.000 0.000 0.000 0.000
#> GSM587212 1 0.3444 0.765 0.816 0.000 0.184 0.000
#> GSM587213 1 0.0000 0.930 1.000 0.000 0.000 0.000
#> GSM587214 1 0.0000 0.930 1.000 0.000 0.000 0.000
#> GSM587215 1 0.0000 0.930 1.000 0.000 0.000 0.000
#> GSM587216 1 0.0000 0.930 1.000 0.000 0.000 0.000
#> GSM587217 1 0.0000 0.930 1.000 0.000 0.000 0.000
#> GSM587191 3 0.0000 0.978 0.000 0.000 1.000 0.000
#> GSM587192 3 0.0000 0.978 0.000 0.000 1.000 0.000
#> GSM587193 3 0.0000 0.978 0.000 0.000 1.000 0.000
#> GSM587194 3 0.0592 0.965 0.000 0.000 0.984 0.016
#> GSM587195 3 0.0000 0.978 0.000 0.000 1.000 0.000
#> GSM587196 3 0.0000 0.978 0.000 0.000 1.000 0.000
#> GSM587197 3 0.0000 0.978 0.000 0.000 1.000 0.000
#> GSM587198 3 0.0000 0.978 0.000 0.000 1.000 0.000
#> GSM587199 3 0.0000 0.978 0.000 0.000 1.000 0.000
#> GSM587200 3 0.0000 0.978 0.000 0.000 1.000 0.000
#> GSM587201 3 0.0000 0.978 0.000 0.000 1.000 0.000
#> GSM587202 3 0.0000 0.978 0.000 0.000 1.000 0.000
#> GSM198767 1 0.0000 0.930 1.000 0.000 0.000 0.000
#> GSM198769 1 0.0000 0.930 1.000 0.000 0.000 0.000
#> GSM198772 1 0.0000 0.930 1.000 0.000 0.000 0.000
#> GSM198773 1 0.0000 0.930 1.000 0.000 0.000 0.000
#> GSM198776 1 0.0000 0.930 1.000 0.000 0.000 0.000
#> GSM198778 1 0.4996 0.153 0.516 0.000 0.484 0.000
#> GSM198780 1 0.3873 0.708 0.772 0.000 0.228 0.000
#> GSM198781 1 0.0000 0.930 1.000 0.000 0.000 0.000
#> GSM198765 3 0.0000 0.978 0.000 0.000 1.000 0.000
#> GSM198766 3 0.1118 0.943 0.036 0.000 0.964 0.000
#> GSM198768 3 0.0000 0.978 0.000 0.000 1.000 0.000
#> GSM198770 3 0.0000 0.978 0.000 0.000 1.000 0.000
#> GSM198771 3 0.0000 0.978 0.000 0.000 1.000 0.000
#> GSM198774 3 0.0000 0.978 0.000 0.000 1.000 0.000
#> GSM198775 3 0.0336 0.972 0.000 0.000 0.992 0.008
#> GSM198777 3 0.0000 0.978 0.000 0.000 1.000 0.000
#> GSM198779 3 0.0000 0.978 0.000 0.000 1.000 0.000
#> GSM587218 4 0.0000 1.000 0.000 0.000 0.000 1.000
#> GSM587219 4 0.0000 1.000 0.000 0.000 0.000 1.000
#> GSM587220 4 0.0000 1.000 0.000 0.000 0.000 1.000
#> GSM587221 4 0.0000 1.000 0.000 0.000 0.000 1.000
#> GSM587222 4 0.0000 1.000 0.000 0.000 0.000 1.000
#> GSM587223 4 0.0000 1.000 0.000 0.000 0.000 1.000
#> GSM587224 4 0.0000 1.000 0.000 0.000 0.000 1.000
#> GSM587225 4 0.0000 1.000 0.000 0.000 0.000 1.000
#> GSM587226 4 0.0000 1.000 0.000 0.000 0.000 1.000
#> GSM587227 4 0.0000 1.000 0.000 0.000 0.000 1.000
#> GSM587228 4 0.0000 1.000 0.000 0.000 0.000 1.000
#> GSM587229 4 0.0000 1.000 0.000 0.000 0.000 1.000
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM587155 2 0.1082 0.9537 0.000 0.964 0.028 0 0.008
#> GSM587156 2 0.4632 0.1767 0.000 0.540 0.448 0 0.012
#> GSM587157 5 0.2971 0.7157 0.000 0.156 0.008 0 0.836
#> GSM587158 2 0.0000 0.9782 0.000 1.000 0.000 0 0.000
#> GSM587159 2 0.0000 0.9782 0.000 1.000 0.000 0 0.000
#> GSM587160 2 0.0000 0.9782 0.000 1.000 0.000 0 0.000
#> GSM587161 2 0.0290 0.9731 0.000 0.992 0.008 0 0.000
#> GSM587162 2 0.0000 0.9782 0.000 1.000 0.000 0 0.000
#> GSM587163 2 0.0000 0.9782 0.000 1.000 0.000 0 0.000
#> GSM587164 2 0.1364 0.9447 0.000 0.952 0.036 0 0.012
#> GSM587165 2 0.0000 0.9782 0.000 1.000 0.000 0 0.000
#> GSM587166 3 0.4632 0.0601 0.000 0.448 0.540 0 0.012
#> GSM587167 2 0.1740 0.9256 0.000 0.932 0.056 0 0.012
#> GSM587168 2 0.0000 0.9782 0.000 1.000 0.000 0 0.000
#> GSM587169 2 0.0000 0.9782 0.000 1.000 0.000 0 0.000
#> GSM587170 2 0.1331 0.9442 0.000 0.952 0.040 0 0.008
#> GSM587171 2 0.0000 0.9782 0.000 1.000 0.000 0 0.000
#> GSM587172 2 0.0000 0.9782 0.000 1.000 0.000 0 0.000
#> GSM587173 2 0.0000 0.9782 0.000 1.000 0.000 0 0.000
#> GSM587174 2 0.0000 0.9782 0.000 1.000 0.000 0 0.000
#> GSM587175 2 0.0865 0.9597 0.000 0.972 0.004 0 0.024
#> GSM587176 2 0.0000 0.9782 0.000 1.000 0.000 0 0.000
#> GSM587177 2 0.0000 0.9782 0.000 1.000 0.000 0 0.000
#> GSM587178 2 0.0000 0.9782 0.000 1.000 0.000 0 0.000
#> GSM587179 2 0.0000 0.9782 0.000 1.000 0.000 0 0.000
#> GSM587180 2 0.0000 0.9782 0.000 1.000 0.000 0 0.000
#> GSM587181 2 0.0000 0.9782 0.000 1.000 0.000 0 0.000
#> GSM587182 2 0.0000 0.9782 0.000 1.000 0.000 0 0.000
#> GSM587183 2 0.0000 0.9782 0.000 1.000 0.000 0 0.000
#> GSM587184 2 0.0000 0.9782 0.000 1.000 0.000 0 0.000
#> GSM587185 2 0.0000 0.9782 0.000 1.000 0.000 0 0.000
#> GSM587186 2 0.0000 0.9782 0.000 1.000 0.000 0 0.000
#> GSM587187 2 0.0000 0.9782 0.000 1.000 0.000 0 0.000
#> GSM587188 2 0.0000 0.9782 0.000 1.000 0.000 0 0.000
#> GSM587189 2 0.0000 0.9782 0.000 1.000 0.000 0 0.000
#> GSM587190 3 0.2300 0.7532 0.000 0.052 0.908 0 0.040
#> GSM587203 1 0.0794 0.9294 0.972 0.000 0.028 0 0.000
#> GSM587204 1 0.0794 0.9294 0.972 0.000 0.028 0 0.000
#> GSM587205 1 0.0794 0.9294 0.972 0.000 0.028 0 0.000
#> GSM587206 1 0.0794 0.9294 0.972 0.000 0.028 0 0.000
#> GSM587207 1 0.0794 0.9294 0.972 0.000 0.028 0 0.000
#> GSM587208 1 0.0794 0.9294 0.972 0.000 0.028 0 0.000
#> GSM587209 1 0.0162 0.9321 0.996 0.000 0.004 0 0.000
#> GSM587210 3 0.5109 0.0287 0.460 0.000 0.504 0 0.036
#> GSM587211 1 0.0404 0.9292 0.988 0.000 0.012 0 0.000
#> GSM587212 1 0.4182 0.4468 0.644 0.000 0.352 0 0.004
#> GSM587213 1 0.0162 0.9321 0.996 0.000 0.004 0 0.000
#> GSM587214 1 0.0000 0.9323 1.000 0.000 0.000 0 0.000
#> GSM587215 1 0.0162 0.9321 0.996 0.000 0.004 0 0.000
#> GSM587216 1 0.3336 0.6863 0.772 0.000 0.228 0 0.000
#> GSM587217 1 0.0290 0.9309 0.992 0.000 0.008 0 0.000
#> GSM587191 3 0.1608 0.7848 0.000 0.000 0.928 0 0.072
#> GSM587192 3 0.1544 0.7859 0.000 0.000 0.932 0 0.068
#> GSM587193 3 0.1195 0.7813 0.012 0.000 0.960 0 0.028
#> GSM587194 3 0.0963 0.7874 0.000 0.000 0.964 0 0.036
#> GSM587195 5 0.0000 0.9202 0.000 0.000 0.000 0 1.000
#> GSM587196 5 0.0162 0.9220 0.000 0.000 0.004 0 0.996
#> GSM587197 5 0.0290 0.9220 0.000 0.000 0.008 0 0.992
#> GSM587198 5 0.1732 0.8932 0.000 0.000 0.080 0 0.920
#> GSM587199 5 0.2929 0.7950 0.000 0.000 0.180 0 0.820
#> GSM587200 3 0.2929 0.6906 0.000 0.000 0.820 0 0.180
#> GSM587201 3 0.3661 0.5553 0.000 0.000 0.724 0 0.276
#> GSM587202 5 0.0510 0.9207 0.000 0.000 0.016 0 0.984
#> GSM198767 1 0.0794 0.9294 0.972 0.000 0.028 0 0.000
#> GSM198769 1 0.0162 0.9321 0.996 0.000 0.004 0 0.000
#> GSM198772 1 0.0404 0.9292 0.988 0.000 0.012 0 0.000
#> GSM198773 1 0.0162 0.9321 0.996 0.000 0.004 0 0.000
#> GSM198776 1 0.0794 0.9294 0.972 0.000 0.028 0 0.000
#> GSM198778 3 0.5161 0.0828 0.444 0.000 0.516 0 0.040
#> GSM198780 1 0.4392 0.3697 0.612 0.000 0.380 0 0.008
#> GSM198781 1 0.0000 0.9323 1.000 0.000 0.000 0 0.000
#> GSM198765 3 0.1608 0.7848 0.000 0.000 0.928 0 0.072
#> GSM198766 3 0.1106 0.7828 0.012 0.000 0.964 0 0.024
#> GSM198768 5 0.0000 0.9202 0.000 0.000 0.000 0 1.000
#> GSM198770 5 0.0290 0.9220 0.000 0.000 0.008 0 0.992
#> GSM198771 5 0.1792 0.8906 0.000 0.000 0.084 0 0.916
#> GSM198774 3 0.1544 0.7859 0.000 0.000 0.932 0 0.068
#> GSM198775 3 0.0963 0.7874 0.000 0.000 0.964 0 0.036
#> GSM198777 5 0.0162 0.9220 0.000 0.000 0.004 0 0.996
#> GSM198779 5 0.2891 0.8004 0.000 0.000 0.176 0 0.824
#> GSM587218 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000
#> GSM587219 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000
#> GSM587220 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000
#> GSM587221 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000
#> GSM587222 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000
#> GSM587223 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000
#> GSM587224 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000
#> GSM587225 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000
#> GSM587226 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000
#> GSM587227 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000
#> GSM587228 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000
#> GSM587229 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM587155 2 0.3046 0.7972 0.000 0.800 0.000 0 0.188 0.012
#> GSM587156 5 0.3871 0.3534 0.000 0.308 0.000 0 0.676 0.016
#> GSM587157 3 0.2882 0.7280 0.000 0.076 0.860 0 0.060 0.004
#> GSM587158 2 0.0000 0.9543 0.000 1.000 0.000 0 0.000 0.000
#> GSM587159 2 0.0000 0.9543 0.000 1.000 0.000 0 0.000 0.000
#> GSM587160 2 0.0000 0.9543 0.000 1.000 0.000 0 0.000 0.000
#> GSM587161 2 0.2489 0.8516 0.000 0.860 0.000 0 0.128 0.012
#> GSM587162 2 0.0260 0.9499 0.000 0.992 0.000 0 0.000 0.008
#> GSM587163 2 0.0000 0.9543 0.000 1.000 0.000 0 0.000 0.000
#> GSM587164 2 0.3509 0.7341 0.000 0.744 0.000 0 0.240 0.016
#> GSM587165 2 0.0000 0.9543 0.000 1.000 0.000 0 0.000 0.000
#> GSM587166 5 0.3766 0.4191 0.000 0.256 0.000 0 0.720 0.024
#> GSM587167 2 0.3927 0.5724 0.000 0.644 0.000 0 0.344 0.012
#> GSM587168 2 0.0000 0.9543 0.000 1.000 0.000 0 0.000 0.000
#> GSM587169 2 0.0000 0.9543 0.000 1.000 0.000 0 0.000 0.000
#> GSM587170 2 0.3230 0.7720 0.000 0.776 0.000 0 0.212 0.012
#> GSM587171 2 0.0000 0.9543 0.000 1.000 0.000 0 0.000 0.000
#> GSM587172 2 0.0000 0.9543 0.000 1.000 0.000 0 0.000 0.000
#> GSM587173 2 0.0000 0.9543 0.000 1.000 0.000 0 0.000 0.000
#> GSM587174 2 0.0000 0.9543 0.000 1.000 0.000 0 0.000 0.000
#> GSM587175 2 0.4403 0.7365 0.000 0.740 0.100 0 0.148 0.012
#> GSM587176 2 0.0508 0.9451 0.000 0.984 0.000 0 0.004 0.012
#> GSM587177 2 0.0000 0.9543 0.000 1.000 0.000 0 0.000 0.000
#> GSM587178 2 0.0000 0.9543 0.000 1.000 0.000 0 0.000 0.000
#> GSM587179 2 0.0000 0.9543 0.000 1.000 0.000 0 0.000 0.000
#> GSM587180 2 0.0000 0.9543 0.000 1.000 0.000 0 0.000 0.000
#> GSM587181 2 0.0000 0.9543 0.000 1.000 0.000 0 0.000 0.000
#> GSM587182 2 0.0000 0.9543 0.000 1.000 0.000 0 0.000 0.000
#> GSM587183 2 0.0000 0.9543 0.000 1.000 0.000 0 0.000 0.000
#> GSM587184 2 0.0000 0.9543 0.000 1.000 0.000 0 0.000 0.000
#> GSM587185 2 0.0000 0.9543 0.000 1.000 0.000 0 0.000 0.000
#> GSM587186 2 0.0000 0.9543 0.000 1.000 0.000 0 0.000 0.000
#> GSM587187 2 0.0000 0.9543 0.000 1.000 0.000 0 0.000 0.000
#> GSM587188 2 0.0000 0.9543 0.000 1.000 0.000 0 0.000 0.000
#> GSM587189 2 0.0146 0.9520 0.000 0.996 0.000 0 0.004 0.000
#> GSM587190 5 0.2604 0.7535 0.000 0.028 0.004 0 0.872 0.096
#> GSM587203 1 0.0000 0.6367 1.000 0.000 0.000 0 0.000 0.000
#> GSM587204 1 0.0000 0.6367 1.000 0.000 0.000 0 0.000 0.000
#> GSM587205 1 0.0000 0.6367 1.000 0.000 0.000 0 0.000 0.000
#> GSM587206 1 0.0000 0.6367 1.000 0.000 0.000 0 0.000 0.000
#> GSM587207 1 0.0000 0.6367 1.000 0.000 0.000 0 0.000 0.000
#> GSM587208 1 0.0000 0.6367 1.000 0.000 0.000 0 0.000 0.000
#> GSM587209 1 0.3868 -0.2100 0.504 0.000 0.000 0 0.000 0.496
#> GSM587210 6 0.2908 0.3827 0.104 0.000 0.000 0 0.048 0.848
#> GSM587211 6 0.4086 0.2090 0.464 0.000 0.000 0 0.008 0.528
#> GSM587212 6 0.4269 0.4170 0.316 0.000 0.000 0 0.036 0.648
#> GSM587213 1 0.3847 -0.0613 0.544 0.000 0.000 0 0.000 0.456
#> GSM587214 1 0.3833 -0.0230 0.556 0.000 0.000 0 0.000 0.444
#> GSM587215 6 0.3862 0.1730 0.476 0.000 0.000 0 0.000 0.524
#> GSM587216 6 0.4004 0.3751 0.368 0.000 0.000 0 0.012 0.620
#> GSM587217 6 0.3860 0.1857 0.472 0.000 0.000 0 0.000 0.528
#> GSM587191 5 0.3672 0.7672 0.000 0.000 0.008 0 0.688 0.304
#> GSM587192 5 0.3565 0.7687 0.000 0.000 0.004 0 0.692 0.304
#> GSM587193 5 0.1863 0.7527 0.000 0.000 0.000 0 0.896 0.104
#> GSM587194 5 0.2597 0.7783 0.000 0.000 0.000 0 0.824 0.176
#> GSM587195 3 0.0000 0.8431 0.000 0.000 1.000 0 0.000 0.000
#> GSM587196 3 0.0000 0.8431 0.000 0.000 1.000 0 0.000 0.000
#> GSM587197 3 0.0000 0.8431 0.000 0.000 1.000 0 0.000 0.000
#> GSM587198 3 0.4149 0.7129 0.000 0.000 0.720 0 0.064 0.216
#> GSM587199 6 0.6117 -0.4471 0.000 0.000 0.344 0 0.300 0.356
#> GSM587200 5 0.4514 0.6372 0.000 0.000 0.040 0 0.588 0.372
#> GSM587201 5 0.4868 0.5713 0.000 0.000 0.060 0 0.524 0.416
#> GSM587202 3 0.2838 0.7708 0.000 0.000 0.808 0 0.004 0.188
#> GSM198767 1 0.0000 0.6367 1.000 0.000 0.000 0 0.000 0.000
#> GSM198769 1 0.3868 -0.2100 0.504 0.000 0.000 0 0.000 0.496
#> GSM198772 6 0.4086 0.2090 0.464 0.000 0.000 0 0.008 0.528
#> GSM198773 1 0.3843 -0.0483 0.548 0.000 0.000 0 0.000 0.452
#> GSM198776 1 0.0000 0.6367 1.000 0.000 0.000 0 0.000 0.000
#> GSM198778 6 0.2762 0.3772 0.092 0.000 0.000 0 0.048 0.860
#> GSM198780 6 0.4316 0.4183 0.312 0.000 0.000 0 0.040 0.648
#> GSM198781 1 0.3828 -0.0120 0.560 0.000 0.000 0 0.000 0.440
#> GSM198765 5 0.3672 0.7672 0.000 0.000 0.008 0 0.688 0.304
#> GSM198766 5 0.2300 0.7693 0.000 0.000 0.000 0 0.856 0.144
#> GSM198768 3 0.0000 0.8431 0.000 0.000 1.000 0 0.000 0.000
#> GSM198770 3 0.0000 0.8431 0.000 0.000 1.000 0 0.000 0.000
#> GSM198771 3 0.4500 0.6814 0.000 0.000 0.688 0 0.088 0.224
#> GSM198774 5 0.3565 0.7687 0.000 0.000 0.004 0 0.692 0.304
#> GSM198775 5 0.2597 0.7783 0.000 0.000 0.000 0 0.824 0.176
#> GSM198777 3 0.0000 0.8431 0.000 0.000 1.000 0 0.000 0.000
#> GSM198779 3 0.6113 -0.0386 0.000 0.000 0.356 0 0.296 0.348
#> GSM587218 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587219 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587220 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587221 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587222 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587223 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587224 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587225 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587226 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587227 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587228 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587229 4 0.0000 1.0000 0.000 0.000 0.000 1 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 specimen(p) k
#> CV:NMF 92 4.01e-14 2
#> CV:NMF 92 7.13e-26 3
#> CV:NMF 90 7.15e-46 4
#> CV:NMF 86 3.52e-39 5
#> CV:NMF 73 1.11e-31 6
If matrix rows can be associated to genes, consider to use functional_enrichment(res,
...) to perform function enrichment for the signature genes. See this vignette for more detailed explanations.
The object with results only for a single top-value method and a single partition method can be extracted as:
res = res_list["MAD", "hclust"]
# you can also extract it by
# res = res_list["MAD:hclust"]
A summary of res and all the functions that can be applied to it:
res
#> A 'ConsensusPartition' object with k = 2, 3, 4, 5, 6.
#> On a matrix with 51941 rows and 92 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 6.
#>
#> 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 0.669 0.903 0.941 0.4836 0.518 0.518
#> 3 3 0.801 0.882 0.922 0.3454 0.832 0.676
#> 4 4 0.831 0.859 0.890 0.0971 0.928 0.796
#> 5 5 0.898 0.876 0.932 0.0629 0.969 0.890
#> 6 6 0.918 0.883 0.936 0.0368 0.964 0.856
suggest_best_k() suggests the best \(k\) based on these statistics. The rules are as follows:
suggest_best_k(res)
#> [1] 6
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
#> GSM587155 2 0.000 1.000 0.000 1.000
#> GSM587156 2 0.000 1.000 0.000 1.000
#> GSM587157 2 0.000 1.000 0.000 1.000
#> GSM587158 2 0.000 1.000 0.000 1.000
#> GSM587159 2 0.000 1.000 0.000 1.000
#> GSM587160 2 0.000 1.000 0.000 1.000
#> GSM587161 2 0.000 1.000 0.000 1.000
#> GSM587162 2 0.000 1.000 0.000 1.000
#> GSM587163 2 0.000 1.000 0.000 1.000
#> GSM587164 2 0.000 1.000 0.000 1.000
#> GSM587165 2 0.000 1.000 0.000 1.000
#> GSM587166 2 0.000 1.000 0.000 1.000
#> GSM587167 2 0.000 1.000 0.000 1.000
#> GSM587168 2 0.000 1.000 0.000 1.000
#> GSM587169 2 0.000 1.000 0.000 1.000
#> GSM587170 2 0.000 1.000 0.000 1.000
#> GSM587171 2 0.000 1.000 0.000 1.000
#> GSM587172 2 0.000 1.000 0.000 1.000
#> GSM587173 2 0.000 1.000 0.000 1.000
#> GSM587174 2 0.000 1.000 0.000 1.000
#> GSM587175 2 0.000 1.000 0.000 1.000
#> GSM587176 2 0.000 1.000 0.000 1.000
#> GSM587177 2 0.000 1.000 0.000 1.000
#> GSM587178 2 0.000 1.000 0.000 1.000
#> GSM587179 2 0.000 1.000 0.000 1.000
#> GSM587180 2 0.000 1.000 0.000 1.000
#> GSM587181 2 0.000 1.000 0.000 1.000
#> GSM587182 2 0.000 1.000 0.000 1.000
#> GSM587183 2 0.000 1.000 0.000 1.000
#> GSM587184 2 0.000 1.000 0.000 1.000
#> GSM587185 2 0.000 1.000 0.000 1.000
#> GSM587186 2 0.000 1.000 0.000 1.000
#> GSM587187 2 0.000 1.000 0.000 1.000
#> GSM587188 2 0.000 1.000 0.000 1.000
#> GSM587189 2 0.000 1.000 0.000 1.000
#> GSM587190 2 0.000 1.000 0.000 1.000
#> GSM587203 1 0.000 0.894 1.000 0.000
#> GSM587204 1 0.000 0.894 1.000 0.000
#> GSM587205 1 0.000 0.894 1.000 0.000
#> GSM587206 1 0.000 0.894 1.000 0.000
#> GSM587207 1 0.000 0.894 1.000 0.000
#> GSM587208 1 0.000 0.894 1.000 0.000
#> GSM587209 1 0.000 0.894 1.000 0.000
#> GSM587210 1 0.000 0.894 1.000 0.000
#> GSM587211 1 0.000 0.894 1.000 0.000
#> GSM587212 1 0.000 0.894 1.000 0.000
#> GSM587213 1 0.000 0.894 1.000 0.000
#> GSM587214 1 0.000 0.894 1.000 0.000
#> GSM587215 1 0.000 0.894 1.000 0.000
#> GSM587216 1 0.000 0.894 1.000 0.000
#> GSM587217 1 0.000 0.894 1.000 0.000
#> GSM587191 1 0.866 0.731 0.712 0.288
#> GSM587192 1 0.866 0.731 0.712 0.288
#> GSM587193 1 0.795 0.770 0.760 0.240
#> GSM587194 1 0.795 0.770 0.760 0.240
#> GSM587195 1 0.866 0.731 0.712 0.288
#> GSM587196 1 0.866 0.731 0.712 0.288
#> GSM587197 1 0.866 0.731 0.712 0.288
#> GSM587198 1 0.866 0.731 0.712 0.288
#> GSM587199 1 0.866 0.731 0.712 0.288
#> GSM587200 1 0.295 0.874 0.948 0.052
#> GSM587201 1 0.295 0.874 0.948 0.052
#> GSM587202 1 0.866 0.731 0.712 0.288
#> GSM198767 1 0.000 0.894 1.000 0.000
#> GSM198769 1 0.000 0.894 1.000 0.000
#> GSM198772 1 0.000 0.894 1.000 0.000
#> GSM198773 1 0.000 0.894 1.000 0.000
#> GSM198776 1 0.000 0.894 1.000 0.000
#> GSM198778 1 0.000 0.894 1.000 0.000
#> GSM198780 1 0.000 0.894 1.000 0.000
#> GSM198781 1 0.000 0.894 1.000 0.000
#> GSM198765 1 0.866 0.731 0.712 0.288
#> GSM198766 1 0.795 0.770 0.760 0.240
#> GSM198768 1 0.866 0.731 0.712 0.288
#> GSM198770 1 0.866 0.731 0.712 0.288
#> GSM198771 1 0.866 0.731 0.712 0.288
#> GSM198774 1 0.866 0.731 0.712 0.288
#> GSM198775 1 0.795 0.770 0.760 0.240
#> GSM198777 1 0.866 0.731 0.712 0.288
#> GSM198779 1 0.866 0.731 0.712 0.288
#> GSM587218 1 0.000 0.894 1.000 0.000
#> GSM587219 1 0.000 0.894 1.000 0.000
#> GSM587220 1 0.000 0.894 1.000 0.000
#> GSM587221 1 0.000 0.894 1.000 0.000
#> GSM587222 1 0.000 0.894 1.000 0.000
#> GSM587223 1 0.000 0.894 1.000 0.000
#> GSM587224 1 0.000 0.894 1.000 0.000
#> GSM587225 1 0.000 0.894 1.000 0.000
#> GSM587226 1 0.000 0.894 1.000 0.000
#> GSM587227 1 0.000 0.894 1.000 0.000
#> GSM587228 1 0.000 0.894 1.000 0.000
#> GSM587229 1 0.000 0.894 1.000 0.000
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM587155 2 0.0000 0.983 0.000 1.000 0.000
#> GSM587156 2 0.3941 0.843 0.000 0.844 0.156
#> GSM587157 2 0.0000 0.983 0.000 1.000 0.000
#> GSM587158 2 0.0000 0.983 0.000 1.000 0.000
#> GSM587159 2 0.0000 0.983 0.000 1.000 0.000
#> GSM587160 2 0.0000 0.983 0.000 1.000 0.000
#> GSM587161 2 0.0000 0.983 0.000 1.000 0.000
#> GSM587162 2 0.0000 0.983 0.000 1.000 0.000
#> GSM587163 2 0.0000 0.983 0.000 1.000 0.000
#> GSM587164 2 0.0000 0.983 0.000 1.000 0.000
#> GSM587165 2 0.0000 0.983 0.000 1.000 0.000
#> GSM587166 2 0.3941 0.843 0.000 0.844 0.156
#> GSM587167 2 0.3941 0.843 0.000 0.844 0.156
#> GSM587168 2 0.0000 0.983 0.000 1.000 0.000
#> GSM587169 2 0.0000 0.983 0.000 1.000 0.000
#> GSM587170 2 0.0000 0.983 0.000 1.000 0.000
#> GSM587171 2 0.0000 0.983 0.000 1.000 0.000
#> GSM587172 2 0.0000 0.983 0.000 1.000 0.000
#> GSM587173 2 0.0000 0.983 0.000 1.000 0.000
#> GSM587174 2 0.0000 0.983 0.000 1.000 0.000
#> GSM587175 2 0.0000 0.983 0.000 1.000 0.000
#> GSM587176 2 0.0000 0.983 0.000 1.000 0.000
#> GSM587177 2 0.0000 0.983 0.000 1.000 0.000
#> GSM587178 2 0.0000 0.983 0.000 1.000 0.000
#> GSM587179 2 0.0000 0.983 0.000 1.000 0.000
#> GSM587180 2 0.0000 0.983 0.000 1.000 0.000
#> GSM587181 2 0.0000 0.983 0.000 1.000 0.000
#> GSM587182 2 0.0000 0.983 0.000 1.000 0.000
#> GSM587183 2 0.0000 0.983 0.000 1.000 0.000
#> GSM587184 2 0.0000 0.983 0.000 1.000 0.000
#> GSM587185 2 0.0000 0.983 0.000 1.000 0.000
#> GSM587186 2 0.0000 0.983 0.000 1.000 0.000
#> GSM587187 2 0.0000 0.983 0.000 1.000 0.000
#> GSM587188 2 0.0000 0.983 0.000 1.000 0.000
#> GSM587189 2 0.0000 0.983 0.000 1.000 0.000
#> GSM587190 2 0.3412 0.876 0.000 0.876 0.124
#> GSM587203 1 0.0000 0.994 1.000 0.000 0.000
#> GSM587204 1 0.0000 0.994 1.000 0.000 0.000
#> GSM587205 1 0.0000 0.994 1.000 0.000 0.000
#> GSM587206 1 0.0000 0.994 1.000 0.000 0.000
#> GSM587207 1 0.0000 0.994 1.000 0.000 0.000
#> GSM587208 1 0.0000 0.994 1.000 0.000 0.000
#> GSM587209 1 0.1643 0.944 0.956 0.000 0.044
#> GSM587210 3 0.5968 0.640 0.364 0.000 0.636
#> GSM587211 1 0.0000 0.994 1.000 0.000 0.000
#> GSM587212 3 0.6126 0.584 0.400 0.000 0.600
#> GSM587213 1 0.0000 0.994 1.000 0.000 0.000
#> GSM587214 1 0.0000 0.994 1.000 0.000 0.000
#> GSM587215 1 0.0000 0.994 1.000 0.000 0.000
#> GSM587216 1 0.0000 0.994 1.000 0.000 0.000
#> GSM587217 1 0.0000 0.994 1.000 0.000 0.000
#> GSM587191 3 0.1860 0.792 0.000 0.052 0.948
#> GSM587192 3 0.1860 0.792 0.000 0.052 0.948
#> GSM587193 3 0.0661 0.789 0.004 0.008 0.988
#> GSM587194 3 0.0661 0.789 0.004 0.008 0.988
#> GSM587195 3 0.1860 0.792 0.000 0.052 0.948
#> GSM587196 3 0.1860 0.792 0.000 0.052 0.948
#> GSM587197 3 0.1860 0.792 0.000 0.052 0.948
#> GSM587198 3 0.1860 0.792 0.000 0.052 0.948
#> GSM587199 3 0.1860 0.792 0.000 0.052 0.948
#> GSM587200 3 0.4346 0.756 0.184 0.000 0.816
#> GSM587201 3 0.4346 0.756 0.184 0.000 0.816
#> GSM587202 3 0.1860 0.792 0.000 0.052 0.948
#> GSM198767 1 0.0000 0.994 1.000 0.000 0.000
#> GSM198769 1 0.1643 0.944 0.956 0.000 0.044
#> GSM198772 1 0.0000 0.994 1.000 0.000 0.000
#> GSM198773 1 0.0000 0.994 1.000 0.000 0.000
#> GSM198776 1 0.0000 0.994 1.000 0.000 0.000
#> GSM198778 3 0.5968 0.640 0.364 0.000 0.636
#> GSM198780 3 0.6126 0.584 0.400 0.000 0.600
#> GSM198781 1 0.0000 0.994 1.000 0.000 0.000
#> GSM198765 3 0.1860 0.792 0.000 0.052 0.948
#> GSM198766 3 0.0661 0.789 0.004 0.008 0.988
#> GSM198768 3 0.1860 0.792 0.000 0.052 0.948
#> GSM198770 3 0.1860 0.792 0.000 0.052 0.948
#> GSM198771 3 0.1860 0.792 0.000 0.052 0.948
#> GSM198774 3 0.1860 0.792 0.000 0.052 0.948
#> GSM198775 3 0.0661 0.789 0.004 0.008 0.988
#> GSM198777 3 0.1860 0.792 0.000 0.052 0.948
#> GSM198779 3 0.1860 0.792 0.000 0.052 0.948
#> GSM587218 3 0.5650 0.709 0.312 0.000 0.688
#> GSM587219 3 0.5650 0.709 0.312 0.000 0.688
#> GSM587220 3 0.5650 0.709 0.312 0.000 0.688
#> GSM587221 3 0.5650 0.709 0.312 0.000 0.688
#> GSM587222 3 0.5650 0.709 0.312 0.000 0.688
#> GSM587223 3 0.5650 0.709 0.312 0.000 0.688
#> GSM587224 3 0.5650 0.709 0.312 0.000 0.688
#> GSM587225 3 0.5650 0.709 0.312 0.000 0.688
#> GSM587226 3 0.5650 0.709 0.312 0.000 0.688
#> GSM587227 3 0.5650 0.709 0.312 0.000 0.688
#> GSM587228 3 0.5650 0.709 0.312 0.000 0.688
#> GSM587229 3 0.5650 0.709 0.312 0.000 0.688
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM587155 2 0.000 0.9456 0.000 1.000 0.000 0.000
#> GSM587156 2 0.370 0.8014 0.000 0.828 0.156 0.016
#> GSM587157 2 0.000 0.9456 0.000 1.000 0.000 0.000
#> GSM587158 2 0.000 0.9456 0.000 1.000 0.000 0.000
#> GSM587159 2 0.000 0.9456 0.000 1.000 0.000 0.000
#> GSM587160 2 0.000 0.9456 0.000 1.000 0.000 0.000
#> GSM587161 2 0.000 0.9456 0.000 1.000 0.000 0.000
#> GSM587162 2 0.000 0.9456 0.000 1.000 0.000 0.000
#> GSM587163 2 0.000 0.9456 0.000 1.000 0.000 0.000
#> GSM587164 2 0.000 0.9456 0.000 1.000 0.000 0.000
#> GSM587165 2 0.000 0.9456 0.000 1.000 0.000 0.000
#> GSM587166 2 0.370 0.8014 0.000 0.828 0.156 0.016
#> GSM587167 2 0.370 0.8014 0.000 0.828 0.156 0.016
#> GSM587168 2 0.000 0.9456 0.000 1.000 0.000 0.000
#> GSM587169 2 0.000 0.9456 0.000 1.000 0.000 0.000
#> GSM587170 2 0.000 0.9456 0.000 1.000 0.000 0.000
#> GSM587171 2 0.000 0.9456 0.000 1.000 0.000 0.000
#> GSM587172 2 0.000 0.9456 0.000 1.000 0.000 0.000
#> GSM587173 2 0.000 0.9456 0.000 1.000 0.000 0.000
#> GSM587174 2 0.000 0.9456 0.000 1.000 0.000 0.000
#> GSM587175 2 0.000 0.9456 0.000 1.000 0.000 0.000
#> GSM587176 2 0.000 0.9456 0.000 1.000 0.000 0.000
#> GSM587177 2 0.000 0.9456 0.000 1.000 0.000 0.000
#> GSM587178 2 0.000 0.9456 0.000 1.000 0.000 0.000
#> GSM587179 2 0.000 0.9456 0.000 1.000 0.000 0.000
#> GSM587180 2 0.000 0.9456 0.000 1.000 0.000 0.000
#> GSM587181 2 0.000 0.9456 0.000 1.000 0.000 0.000
#> GSM587182 2 0.000 0.9456 0.000 1.000 0.000 0.000
#> GSM587183 2 0.000 0.9456 0.000 1.000 0.000 0.000
#> GSM587184 2 0.000 0.9456 0.000 1.000 0.000 0.000
#> GSM587185 2 0.000 0.9456 0.000 1.000 0.000 0.000
#> GSM587186 2 0.000 0.9456 0.000 1.000 0.000 0.000
#> GSM587187 2 0.583 0.6401 0.000 0.632 0.052 0.316
#> GSM587188 2 0.583 0.6401 0.000 0.632 0.052 0.316
#> GSM587189 2 0.583 0.6401 0.000 0.632 0.052 0.316
#> GSM587190 2 0.736 0.4688 0.000 0.492 0.176 0.332
#> GSM587203 1 0.000 0.9709 1.000 0.000 0.000 0.000
#> GSM587204 1 0.000 0.9709 1.000 0.000 0.000 0.000
#> GSM587205 1 0.000 0.9709 1.000 0.000 0.000 0.000
#> GSM587206 1 0.000 0.9709 1.000 0.000 0.000 0.000
#> GSM587207 1 0.000 0.9709 1.000 0.000 0.000 0.000
#> GSM587208 1 0.000 0.9709 1.000 0.000 0.000 0.000
#> GSM587209 1 0.409 0.7288 0.776 0.000 0.008 0.216
#> GSM587210 3 0.717 0.0118 0.184 0.000 0.548 0.268
#> GSM587211 1 0.000 0.9709 1.000 0.000 0.000 0.000
#> GSM587212 3 0.631 0.2182 0.392 0.000 0.544 0.064
#> GSM587213 1 0.000 0.9709 1.000 0.000 0.000 0.000
#> GSM587214 1 0.000 0.9709 1.000 0.000 0.000 0.000
#> GSM587215 1 0.000 0.9709 1.000 0.000 0.000 0.000
#> GSM587216 1 0.147 0.9275 0.948 0.000 0.000 0.052
#> GSM587217 1 0.000 0.9709 1.000 0.000 0.000 0.000
#> GSM587191 3 0.000 0.8360 0.000 0.000 1.000 0.000
#> GSM587192 3 0.000 0.8360 0.000 0.000 1.000 0.000
#> GSM587193 3 0.139 0.7999 0.000 0.000 0.952 0.048
#> GSM587194 3 0.139 0.7999 0.000 0.000 0.952 0.048
#> GSM587195 3 0.000 0.8360 0.000 0.000 1.000 0.000
#> GSM587196 3 0.000 0.8360 0.000 0.000 1.000 0.000
#> GSM587197 3 0.000 0.8360 0.000 0.000 1.000 0.000
#> GSM587198 3 0.000 0.8360 0.000 0.000 1.000 0.000
#> GSM587199 3 0.000 0.8360 0.000 0.000 1.000 0.000
#> GSM587200 3 0.419 0.3577 0.000 0.000 0.732 0.268
#> GSM587201 3 0.419 0.3577 0.000 0.000 0.732 0.268
#> GSM587202 3 0.000 0.8360 0.000 0.000 1.000 0.000
#> GSM198767 1 0.000 0.9709 1.000 0.000 0.000 0.000
#> GSM198769 1 0.409 0.7288 0.776 0.000 0.008 0.216
#> GSM198772 1 0.000 0.9709 1.000 0.000 0.000 0.000
#> GSM198773 1 0.000 0.9709 1.000 0.000 0.000 0.000
#> GSM198776 1 0.000 0.9709 1.000 0.000 0.000 0.000
#> GSM198778 3 0.717 0.0118 0.184 0.000 0.548 0.268
#> GSM198780 3 0.631 0.2182 0.392 0.000 0.544 0.064
#> GSM198781 1 0.000 0.9709 1.000 0.000 0.000 0.000
#> GSM198765 3 0.000 0.8360 0.000 0.000 1.000 0.000
#> GSM198766 3 0.139 0.7999 0.000 0.000 0.952 0.048
#> GSM198768 3 0.000 0.8360 0.000 0.000 1.000 0.000
#> GSM198770 3 0.000 0.8360 0.000 0.000 1.000 0.000
#> GSM198771 3 0.000 0.8360 0.000 0.000 1.000 0.000
#> GSM198774 3 0.000 0.8360 0.000 0.000 1.000 0.000
#> GSM198775 3 0.139 0.7999 0.000 0.000 0.952 0.048
#> GSM198777 3 0.000 0.8360 0.000 0.000 1.000 0.000
#> GSM198779 3 0.000 0.8360 0.000 0.000 1.000 0.000
#> GSM587218 4 0.458 1.0000 0.000 0.000 0.332 0.668
#> GSM587219 4 0.458 1.0000 0.000 0.000 0.332 0.668
#> GSM587220 4 0.458 1.0000 0.000 0.000 0.332 0.668
#> GSM587221 4 0.458 1.0000 0.000 0.000 0.332 0.668
#> GSM587222 4 0.458 1.0000 0.000 0.000 0.332 0.668
#> GSM587223 4 0.458 1.0000 0.000 0.000 0.332 0.668
#> GSM587224 4 0.458 1.0000 0.000 0.000 0.332 0.668
#> GSM587225 4 0.458 1.0000 0.000 0.000 0.332 0.668
#> GSM587226 4 0.458 1.0000 0.000 0.000 0.332 0.668
#> GSM587227 4 0.458 1.0000 0.000 0.000 0.332 0.668
#> GSM587228 4 0.458 1.0000 0.000 0.000 0.332 0.668
#> GSM587229 4 0.458 1.0000 0.000 0.000 0.332 0.668
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM587155 2 0.1270 0.909 0.000 0.948 0.000 0.000 0.052
#> GSM587156 2 0.4066 0.475 0.000 0.672 0.004 0.000 0.324
#> GSM587157 2 0.1197 0.911 0.000 0.952 0.000 0.000 0.048
#> GSM587158 2 0.0000 0.945 0.000 1.000 0.000 0.000 0.000
#> GSM587159 2 0.0000 0.945 0.000 1.000 0.000 0.000 0.000
#> GSM587160 2 0.0290 0.941 0.000 0.992 0.000 0.000 0.008
#> GSM587161 2 0.0162 0.943 0.000 0.996 0.000 0.000 0.004
#> GSM587162 2 0.0290 0.941 0.000 0.992 0.000 0.000 0.008
#> GSM587163 2 0.0000 0.945 0.000 1.000 0.000 0.000 0.000
#> GSM587164 2 0.1197 0.911 0.000 0.952 0.000 0.000 0.048
#> GSM587165 2 0.0000 0.945 0.000 1.000 0.000 0.000 0.000
#> GSM587166 2 0.4066 0.475 0.000 0.672 0.004 0.000 0.324
#> GSM587167 2 0.4084 0.470 0.000 0.668 0.004 0.000 0.328
#> GSM587168 2 0.0000 0.945 0.000 1.000 0.000 0.000 0.000
#> GSM587169 2 0.0000 0.945 0.000 1.000 0.000 0.000 0.000
#> GSM587170 2 0.1197 0.911 0.000 0.952 0.000 0.000 0.048
#> GSM587171 2 0.0000 0.945 0.000 1.000 0.000 0.000 0.000
#> GSM587172 2 0.0000 0.945 0.000 1.000 0.000 0.000 0.000
#> GSM587173 2 0.0000 0.945 0.000 1.000 0.000 0.000 0.000
#> GSM587174 2 0.0290 0.941 0.000 0.992 0.000 0.000 0.008
#> GSM587175 2 0.1197 0.911 0.000 0.952 0.000 0.000 0.048
#> GSM587176 2 0.0290 0.941 0.000 0.992 0.000 0.000 0.008
#> GSM587177 2 0.0000 0.945 0.000 1.000 0.000 0.000 0.000
#> GSM587178 2 0.0000 0.945 0.000 1.000 0.000 0.000 0.000
#> GSM587179 2 0.0162 0.943 0.000 0.996 0.000 0.000 0.004
#> GSM587180 2 0.0000 0.945 0.000 1.000 0.000 0.000 0.000
#> GSM587181 2 0.0290 0.941 0.000 0.992 0.000 0.000 0.008
#> GSM587182 2 0.0000 0.945 0.000 1.000 0.000 0.000 0.000
#> GSM587183 2 0.0000 0.945 0.000 1.000 0.000 0.000 0.000
#> GSM587184 2 0.0000 0.945 0.000 1.000 0.000 0.000 0.000
#> GSM587185 2 0.0162 0.943 0.000 0.996 0.000 0.000 0.004
#> GSM587186 2 0.0000 0.945 0.000 1.000 0.000 0.000 0.000
#> GSM587187 5 0.4957 0.852 0.000 0.332 0.044 0.000 0.624
#> GSM587188 5 0.4972 0.849 0.000 0.336 0.044 0.000 0.620
#> GSM587189 5 0.4957 0.852 0.000 0.332 0.044 0.000 0.624
#> GSM587190 5 0.2228 0.564 0.000 0.040 0.048 0.000 0.912
#> GSM587203 1 0.0000 0.966 1.000 0.000 0.000 0.000 0.000
#> GSM587204 1 0.0000 0.966 1.000 0.000 0.000 0.000 0.000
#> GSM587205 1 0.0000 0.966 1.000 0.000 0.000 0.000 0.000
#> GSM587206 1 0.0000 0.966 1.000 0.000 0.000 0.000 0.000
#> GSM587207 1 0.0000 0.966 1.000 0.000 0.000 0.000 0.000
#> GSM587208 1 0.0000 0.966 1.000 0.000 0.000 0.000 0.000
#> GSM587209 1 0.3305 0.712 0.776 0.000 0.000 0.224 0.000
#> GSM587210 3 0.6983 0.502 0.184 0.000 0.536 0.236 0.044
#> GSM587211 1 0.0000 0.966 1.000 0.000 0.000 0.000 0.000
#> GSM587212 3 0.5834 0.405 0.392 0.000 0.536 0.028 0.044
#> GSM587213 1 0.0000 0.966 1.000 0.000 0.000 0.000 0.000
#> GSM587214 1 0.0000 0.966 1.000 0.000 0.000 0.000 0.000
#> GSM587215 1 0.0000 0.966 1.000 0.000 0.000 0.000 0.000
#> GSM587216 1 0.1270 0.917 0.948 0.000 0.000 0.052 0.000
#> GSM587217 1 0.0000 0.966 1.000 0.000 0.000 0.000 0.000
#> GSM587191 3 0.0000 0.856 0.000 0.000 1.000 0.000 0.000
#> GSM587192 3 0.0000 0.856 0.000 0.000 1.000 0.000 0.000
#> GSM587193 3 0.3177 0.768 0.000 0.000 0.792 0.000 0.208
#> GSM587194 3 0.3177 0.768 0.000 0.000 0.792 0.000 0.208
#> GSM587195 3 0.0000 0.856 0.000 0.000 1.000 0.000 0.000
#> GSM587196 3 0.0000 0.856 0.000 0.000 1.000 0.000 0.000
#> GSM587197 3 0.0000 0.856 0.000 0.000 1.000 0.000 0.000
#> GSM587198 3 0.0000 0.856 0.000 0.000 1.000 0.000 0.000
#> GSM587199 3 0.0000 0.856 0.000 0.000 1.000 0.000 0.000
#> GSM587200 3 0.4584 0.680 0.000 0.000 0.716 0.228 0.056
#> GSM587201 3 0.4584 0.680 0.000 0.000 0.716 0.228 0.056
#> GSM587202 3 0.0000 0.856 0.000 0.000 1.000 0.000 0.000
#> GSM198767 1 0.0000 0.966 1.000 0.000 0.000 0.000 0.000
#> GSM198769 1 0.3305 0.712 0.776 0.000 0.000 0.224 0.000
#> GSM198772 1 0.0000 0.966 1.000 0.000 0.000 0.000 0.000
#> GSM198773 1 0.0000 0.966 1.000 0.000 0.000 0.000 0.000
#> GSM198776 1 0.0000 0.966 1.000 0.000 0.000 0.000 0.000
#> GSM198778 3 0.6983 0.502 0.184 0.000 0.536 0.236 0.044
#> GSM198780 3 0.5834 0.405 0.392 0.000 0.536 0.028 0.044
#> GSM198781 1 0.0000 0.966 1.000 0.000 0.000 0.000 0.000
#> GSM198765 3 0.0000 0.856 0.000 0.000 1.000 0.000 0.000
#> GSM198766 3 0.3177 0.768 0.000 0.000 0.792 0.000 0.208
#> GSM198768 3 0.0000 0.856 0.000 0.000 1.000 0.000 0.000
#> GSM198770 3 0.0000 0.856 0.000 0.000 1.000 0.000 0.000
#> GSM198771 3 0.0000 0.856 0.000 0.000 1.000 0.000 0.000
#> GSM198774 3 0.0000 0.856 0.000 0.000 1.000 0.000 0.000
#> GSM198775 3 0.3177 0.768 0.000 0.000 0.792 0.000 0.208
#> GSM198777 3 0.0000 0.856 0.000 0.000 1.000 0.000 0.000
#> GSM198779 3 0.0000 0.856 0.000 0.000 1.000 0.000 0.000
#> GSM587218 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000
#> GSM587219 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000
#> GSM587220 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000
#> GSM587221 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000
#> GSM587222 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000
#> GSM587223 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000
#> GSM587224 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000
#> GSM587225 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000
#> GSM587226 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000
#> GSM587227 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000
#> GSM587228 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000
#> GSM587229 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM587155 2 0.1267 0.904 0.000 0.940 0.000 0.000 0.000 0.060
#> GSM587156 2 0.3905 0.482 0.000 0.668 0.000 0.000 0.016 0.316
#> GSM587157 2 0.1141 0.908 0.000 0.948 0.000 0.000 0.000 0.052
#> GSM587158 2 0.0000 0.944 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587159 2 0.0000 0.944 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587160 2 0.0363 0.938 0.000 0.988 0.000 0.000 0.000 0.012
#> GSM587161 2 0.0146 0.942 0.000 0.996 0.000 0.000 0.000 0.004
#> GSM587162 2 0.0363 0.938 0.000 0.988 0.000 0.000 0.000 0.012
#> GSM587163 2 0.0000 0.944 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587164 2 0.1141 0.908 0.000 0.948 0.000 0.000 0.000 0.052
#> GSM587165 2 0.0000 0.944 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587166 2 0.3905 0.482 0.000 0.668 0.000 0.000 0.016 0.316
#> GSM587167 2 0.3938 0.472 0.000 0.660 0.000 0.000 0.016 0.324
#> GSM587168 2 0.0000 0.944 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587169 2 0.0000 0.944 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587170 2 0.1141 0.908 0.000 0.948 0.000 0.000 0.000 0.052
#> GSM587171 2 0.0000 0.944 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587172 2 0.0000 0.944 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587173 2 0.0000 0.944 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587174 2 0.0363 0.938 0.000 0.988 0.000 0.000 0.000 0.012
#> GSM587175 2 0.1141 0.908 0.000 0.948 0.000 0.000 0.000 0.052
#> GSM587176 2 0.0363 0.938 0.000 0.988 0.000 0.000 0.000 0.012
#> GSM587177 2 0.0000 0.944 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587178 2 0.0000 0.944 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587179 2 0.0146 0.942 0.000 0.996 0.000 0.000 0.000 0.004
#> GSM587180 2 0.0000 0.944 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587181 2 0.0363 0.938 0.000 0.988 0.000 0.000 0.000 0.012
#> GSM587182 2 0.0000 0.944 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587183 2 0.0000 0.944 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587184 2 0.0000 0.944 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587185 2 0.0146 0.942 0.000 0.996 0.000 0.000 0.000 0.004
#> GSM587186 2 0.0000 0.944 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587187 6 0.3464 0.838 0.000 0.312 0.000 0.000 0.000 0.688
#> GSM587188 6 0.3482 0.835 0.000 0.316 0.000 0.000 0.000 0.684
#> GSM587189 6 0.3464 0.838 0.000 0.312 0.000 0.000 0.000 0.688
#> GSM587190 6 0.0891 0.408 0.000 0.024 0.000 0.000 0.008 0.968
#> GSM587203 1 0.0000 0.967 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM587204 1 0.0000 0.967 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM587205 1 0.0000 0.967 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM587206 1 0.0000 0.967 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM587207 1 0.0000 0.967 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM587208 1 0.0000 0.967 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM587209 1 0.3796 0.705 0.776 0.000 0.000 0.084 0.140 0.000
#> GSM587210 5 0.4314 0.624 0.184 0.000 0.000 0.096 0.720 0.000
#> GSM587211 1 0.0000 0.967 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM587212 5 0.4228 0.475 0.392 0.000 0.000 0.020 0.588 0.000
#> GSM587213 1 0.0000 0.967 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM587214 1 0.0000 0.967 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM587215 1 0.0000 0.967 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM587216 1 0.1334 0.918 0.948 0.000 0.000 0.020 0.032 0.000
#> GSM587217 1 0.0000 0.967 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM587191 3 0.0000 0.969 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM587192 3 0.0000 0.969 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM587193 5 0.2362 0.653 0.000 0.000 0.004 0.000 0.860 0.136
#> GSM587194 5 0.2362 0.653 0.000 0.000 0.004 0.000 0.860 0.136
#> GSM587195 3 0.0000 0.969 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM587196 3 0.0000 0.969 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM587197 3 0.0000 0.969 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM587198 3 0.0000 0.969 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM587199 3 0.2664 0.762 0.000 0.000 0.816 0.000 0.184 0.000
#> GSM587200 5 0.4705 0.494 0.000 0.000 0.260 0.088 0.652 0.000
#> GSM587201 5 0.4705 0.494 0.000 0.000 0.260 0.088 0.652 0.000
#> GSM587202 3 0.0000 0.969 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM198767 1 0.0000 0.967 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM198769 1 0.3796 0.705 0.776 0.000 0.000 0.084 0.140 0.000
#> GSM198772 1 0.0000 0.967 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM198773 1 0.0000 0.967 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM198776 1 0.0000 0.967 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM198778 5 0.4314 0.624 0.184 0.000 0.000 0.096 0.720 0.000
#> GSM198780 5 0.4228 0.475 0.392 0.000 0.000 0.020 0.588 0.000
#> GSM198781 1 0.0000 0.967 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM198765 3 0.0000 0.969 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM198766 5 0.2362 0.653 0.000 0.000 0.004 0.000 0.860 0.136
#> GSM198768 3 0.0000 0.969 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM198770 3 0.0000 0.969 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM198771 3 0.0000 0.969 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM198774 3 0.0000 0.969 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM198775 5 0.2362 0.653 0.000 0.000 0.004 0.000 0.860 0.136
#> GSM198777 3 0.0000 0.969 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM198779 3 0.2664 0.762 0.000 0.000 0.816 0.000 0.184 0.000
#> GSM587218 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM587219 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM587220 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM587221 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM587222 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM587223 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM587224 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM587225 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM587226 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM587227 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM587228 4 0.0000 1.000 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM587229 4 0.0000 1.000 0.000 0.000 0.000 1.000 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 specimen(p) k
#> MAD:hclust 92 1.16e-17 2
#> MAD:hclust 92 4.55e-28 3
#> MAD:hclust 85 7.83e-44 4
#> MAD:hclust 87 3.47e-56 5
#> MAD:hclust 84 2.67e-51 6
If matrix rows can be associated to genes, consider to use functional_enrichment(res,
...) to perform function enrichment for the signature genes. See this vignette for more detailed explanations.
The object with results only for a single top-value method and a single partition method can be extracted as:
res = res_list["MAD", "kmeans"]
# you can also extract it by
# res = res_list["MAD:kmeans"]
A summary of res and all the functions that can be applied to it:
res
#> A 'ConsensusPartition' object with k = 2, 3, 4, 5, 6.
#> On a matrix with 51941 rows and 92 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 0.724 0.896 0.944 0.4940 0.500 0.500
#> 3 3 0.781 0.897 0.913 0.3181 0.783 0.588
#> 4 4 0.783 0.758 0.773 0.1030 0.922 0.770
#> 5 5 0.740 0.810 0.823 0.0610 0.953 0.841
#> 6 6 0.716 0.684 0.775 0.0433 0.931 0.748
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
#> GSM587155 2 0.0000 0.889 0.000 1.000
#> GSM587156 2 0.0000 0.889 0.000 1.000
#> GSM587157 2 0.0000 0.889 0.000 1.000
#> GSM587158 2 0.0000 0.889 0.000 1.000
#> GSM587159 2 0.0000 0.889 0.000 1.000
#> GSM587160 2 0.0000 0.889 0.000 1.000
#> GSM587161 2 0.0000 0.889 0.000 1.000
#> GSM587162 2 0.0000 0.889 0.000 1.000
#> GSM587163 2 0.0000 0.889 0.000 1.000
#> GSM587164 2 0.0000 0.889 0.000 1.000
#> GSM587165 2 0.0000 0.889 0.000 1.000
#> GSM587166 2 0.0000 0.889 0.000 1.000
#> GSM587167 2 0.0000 0.889 0.000 1.000
#> GSM587168 2 0.0000 0.889 0.000 1.000
#> GSM587169 2 0.0000 0.889 0.000 1.000
#> GSM587170 2 0.0000 0.889 0.000 1.000
#> GSM587171 2 0.0000 0.889 0.000 1.000
#> GSM587172 2 0.0000 0.889 0.000 1.000
#> GSM587173 2 0.0000 0.889 0.000 1.000
#> GSM587174 2 0.0000 0.889 0.000 1.000
#> GSM587175 2 0.0000 0.889 0.000 1.000
#> GSM587176 2 0.0000 0.889 0.000 1.000
#> GSM587177 2 0.0000 0.889 0.000 1.000
#> GSM587178 2 0.0000 0.889 0.000 1.000
#> GSM587179 2 0.0000 0.889 0.000 1.000
#> GSM587180 2 0.0000 0.889 0.000 1.000
#> GSM587181 2 0.0000 0.889 0.000 1.000
#> GSM587182 2 0.0000 0.889 0.000 1.000
#> GSM587183 2 0.0000 0.889 0.000 1.000
#> GSM587184 2 0.0000 0.889 0.000 1.000
#> GSM587185 2 0.0000 0.889 0.000 1.000
#> GSM587186 2 0.0000 0.889 0.000 1.000
#> GSM587187 2 0.0000 0.889 0.000 1.000
#> GSM587188 2 0.0000 0.889 0.000 1.000
#> GSM587189 2 0.0000 0.889 0.000 1.000
#> GSM587190 2 0.0672 0.885 0.008 0.992
#> GSM587203 1 0.0000 1.000 1.000 0.000
#> GSM587204 1 0.0000 1.000 1.000 0.000
#> GSM587205 1 0.0000 1.000 1.000 0.000
#> GSM587206 1 0.0000 1.000 1.000 0.000
#> GSM587207 1 0.0000 1.000 1.000 0.000
#> GSM587208 1 0.0000 1.000 1.000 0.000
#> GSM587209 1 0.0000 1.000 1.000 0.000
#> GSM587210 1 0.0000 1.000 1.000 0.000
#> GSM587211 1 0.0000 1.000 1.000 0.000
#> GSM587212 1 0.0000 1.000 1.000 0.000
#> GSM587213 1 0.0000 1.000 1.000 0.000
#> GSM587214 1 0.0000 1.000 1.000 0.000
#> GSM587215 1 0.0000 1.000 1.000 0.000
#> GSM587216 1 0.0000 1.000 1.000 0.000
#> GSM587217 1 0.0000 1.000 1.000 0.000
#> GSM587191 2 0.9286 0.623 0.344 0.656
#> GSM587192 1 0.0000 1.000 1.000 0.000
#> GSM587193 1 0.0000 1.000 1.000 0.000
#> GSM587194 2 0.9286 0.623 0.344 0.656
#> GSM587195 2 0.9170 0.637 0.332 0.668
#> GSM587196 2 0.9286 0.623 0.344 0.656
#> GSM587197 2 0.9286 0.623 0.344 0.656
#> GSM587198 2 0.9286 0.623 0.344 0.656
#> GSM587199 2 0.9129 0.641 0.328 0.672
#> GSM587200 1 0.0000 1.000 1.000 0.000
#> GSM587201 1 0.0000 1.000 1.000 0.000
#> GSM587202 2 0.9286 0.623 0.344 0.656
#> GSM198767 1 0.0000 1.000 1.000 0.000
#> GSM198769 1 0.0000 1.000 1.000 0.000
#> GSM198772 1 0.0000 1.000 1.000 0.000
#> GSM198773 1 0.0000 1.000 1.000 0.000
#> GSM198776 1 0.0000 1.000 1.000 0.000
#> GSM198778 1 0.0000 1.000 1.000 0.000
#> GSM198780 1 0.0000 1.000 1.000 0.000
#> GSM198781 1 0.0000 1.000 1.000 0.000
#> GSM198765 2 0.9286 0.623 0.344 0.656
#> GSM198766 1 0.0000 1.000 1.000 0.000
#> GSM198768 2 0.9286 0.623 0.344 0.656
#> GSM198770 2 0.9286 0.623 0.344 0.656
#> GSM198771 2 0.9286 0.623 0.344 0.656
#> GSM198774 1 0.0000 1.000 1.000 0.000
#> GSM198775 2 0.9286 0.623 0.344 0.656
#> GSM198777 2 0.9286 0.623 0.344 0.656
#> GSM198779 2 0.9129 0.641 0.328 0.672
#> GSM587218 1 0.0000 1.000 1.000 0.000
#> GSM587219 1 0.0000 1.000 1.000 0.000
#> GSM587220 1 0.0000 1.000 1.000 0.000
#> GSM587221 1 0.0000 1.000 1.000 0.000
#> GSM587222 1 0.0000 1.000 1.000 0.000
#> GSM587223 1 0.0000 1.000 1.000 0.000
#> GSM587224 1 0.0000 1.000 1.000 0.000
#> GSM587225 1 0.0000 1.000 1.000 0.000
#> GSM587226 1 0.0000 1.000 1.000 0.000
#> GSM587227 1 0.0000 1.000 1.000 0.000
#> GSM587228 1 0.0000 1.000 1.000 0.000
#> GSM587229 1 0.0000 1.000 1.000 0.000
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM587155 2 0.0592 0.985 0.000 0.988 0.012
#> GSM587156 2 0.0592 0.985 0.000 0.988 0.012
#> GSM587157 2 0.0592 0.985 0.000 0.988 0.012
#> GSM587158 2 0.0000 0.988 0.000 1.000 0.000
#> GSM587159 2 0.0000 0.988 0.000 1.000 0.000
#> GSM587160 2 0.0000 0.988 0.000 1.000 0.000
#> GSM587161 2 0.0592 0.985 0.000 0.988 0.012
#> GSM587162 2 0.0000 0.988 0.000 1.000 0.000
#> GSM587163 2 0.0000 0.988 0.000 1.000 0.000
#> GSM587164 2 0.0592 0.985 0.000 0.988 0.012
#> GSM587165 2 0.0592 0.986 0.000 0.988 0.012
#> GSM587166 2 0.0592 0.985 0.000 0.988 0.012
#> GSM587167 2 0.0592 0.985 0.000 0.988 0.012
#> GSM587168 2 0.0592 0.986 0.000 0.988 0.012
#> GSM587169 2 0.0000 0.988 0.000 1.000 0.000
#> GSM587170 2 0.0592 0.985 0.000 0.988 0.012
#> GSM587171 2 0.0000 0.988 0.000 1.000 0.000
#> GSM587172 2 0.0000 0.988 0.000 1.000 0.000
#> GSM587173 2 0.0592 0.986 0.000 0.988 0.012
#> GSM587174 2 0.0424 0.987 0.000 0.992 0.008
#> GSM587175 2 0.0592 0.985 0.000 0.988 0.012
#> GSM587176 2 0.0000 0.988 0.000 1.000 0.000
#> GSM587177 2 0.0592 0.986 0.000 0.988 0.012
#> GSM587178 2 0.0424 0.987 0.000 0.992 0.008
#> GSM587179 2 0.0000 0.988 0.000 1.000 0.000
#> GSM587180 2 0.0592 0.986 0.000 0.988 0.012
#> GSM587181 2 0.0424 0.987 0.000 0.992 0.008
#> GSM587182 2 0.0424 0.987 0.000 0.992 0.008
#> GSM587183 2 0.0592 0.986 0.000 0.988 0.012
#> GSM587184 2 0.0000 0.988 0.000 1.000 0.000
#> GSM587185 2 0.0000 0.988 0.000 1.000 0.000
#> GSM587186 2 0.0592 0.986 0.000 0.988 0.012
#> GSM587187 2 0.0592 0.986 0.000 0.988 0.012
#> GSM587188 2 0.2959 0.896 0.000 0.900 0.100
#> GSM587189 2 0.2959 0.896 0.000 0.900 0.100
#> GSM587190 3 0.3482 0.879 0.000 0.128 0.872
#> GSM587203 1 0.1289 0.855 0.968 0.000 0.032
#> GSM587204 1 0.1411 0.854 0.964 0.000 0.036
#> GSM587205 1 0.1289 0.855 0.968 0.000 0.032
#> GSM587206 1 0.1289 0.855 0.968 0.000 0.032
#> GSM587207 1 0.1289 0.855 0.968 0.000 0.032
#> GSM587208 1 0.1289 0.855 0.968 0.000 0.032
#> GSM587209 1 0.1031 0.857 0.976 0.000 0.024
#> GSM587210 1 0.4399 0.802 0.812 0.000 0.188
#> GSM587211 1 0.1031 0.857 0.976 0.000 0.024
#> GSM587212 1 0.4062 0.815 0.836 0.000 0.164
#> GSM587213 1 0.1031 0.857 0.976 0.000 0.024
#> GSM587214 1 0.1031 0.857 0.976 0.000 0.024
#> GSM587215 1 0.1031 0.857 0.976 0.000 0.024
#> GSM587216 1 0.1031 0.857 0.976 0.000 0.024
#> GSM587217 1 0.1031 0.857 0.976 0.000 0.024
#> GSM587191 3 0.4007 0.943 0.036 0.084 0.880
#> GSM587192 3 0.2878 0.886 0.096 0.000 0.904
#> GSM587193 3 0.2878 0.886 0.096 0.000 0.904
#> GSM587194 3 0.3550 0.939 0.024 0.080 0.896
#> GSM587195 3 0.4174 0.942 0.036 0.092 0.872
#> GSM587196 3 0.4174 0.942 0.036 0.092 0.872
#> GSM587197 3 0.4174 0.942 0.036 0.092 0.872
#> GSM587198 3 0.4007 0.943 0.036 0.084 0.880
#> GSM587199 3 0.3805 0.939 0.024 0.092 0.884
#> GSM587200 3 0.3267 0.879 0.116 0.000 0.884
#> GSM587201 3 0.3267 0.879 0.116 0.000 0.884
#> GSM587202 3 0.4174 0.942 0.036 0.092 0.872
#> GSM198767 1 0.1289 0.855 0.968 0.000 0.032
#> GSM198769 1 0.1031 0.857 0.976 0.000 0.024
#> GSM198772 1 0.1031 0.857 0.976 0.000 0.024
#> GSM198773 1 0.1031 0.857 0.976 0.000 0.024
#> GSM198776 1 0.1411 0.854 0.964 0.000 0.036
#> GSM198778 1 0.4399 0.802 0.812 0.000 0.188
#> GSM198780 1 0.4062 0.815 0.836 0.000 0.164
#> GSM198781 1 0.1031 0.857 0.976 0.000 0.024
#> GSM198765 3 0.4007 0.943 0.036 0.084 0.880
#> GSM198766 3 0.2878 0.886 0.096 0.000 0.904
#> GSM198768 3 0.4174 0.942 0.036 0.092 0.872
#> GSM198770 3 0.4174 0.942 0.036 0.092 0.872
#> GSM198771 3 0.4007 0.943 0.036 0.084 0.880
#> GSM198774 3 0.2878 0.886 0.096 0.000 0.904
#> GSM198775 3 0.3550 0.939 0.024 0.080 0.896
#> GSM198777 3 0.4174 0.942 0.036 0.092 0.872
#> GSM198779 3 0.3805 0.939 0.024 0.092 0.884
#> GSM587218 3 0.4452 0.678 0.192 0.000 0.808
#> GSM587219 1 0.5810 0.695 0.664 0.000 0.336
#> GSM587220 1 0.5785 0.700 0.668 0.000 0.332
#> GSM587221 1 0.5810 0.695 0.664 0.000 0.336
#> GSM587222 1 0.5785 0.700 0.668 0.000 0.332
#> GSM587223 1 0.5810 0.695 0.664 0.000 0.336
#> GSM587224 1 0.5810 0.695 0.664 0.000 0.336
#> GSM587225 1 0.5810 0.699 0.664 0.000 0.336
#> GSM587226 1 0.5810 0.695 0.664 0.000 0.336
#> GSM587227 1 0.5810 0.699 0.664 0.000 0.336
#> GSM587228 1 0.5810 0.699 0.664 0.000 0.336
#> GSM587229 1 0.5327 0.750 0.728 0.000 0.272
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM587155 2 0.2760 0.894 0.128 0.872 0.000 0.000
#> GSM587156 2 0.3649 0.847 0.204 0.796 0.000 0.000
#> GSM587157 2 0.2868 0.890 0.136 0.864 0.000 0.000
#> GSM587158 2 0.0000 0.936 0.000 1.000 0.000 0.000
#> GSM587159 2 0.0000 0.936 0.000 1.000 0.000 0.000
#> GSM587160 2 0.0336 0.936 0.008 0.992 0.000 0.000
#> GSM587161 2 0.2408 0.905 0.104 0.896 0.000 0.000
#> GSM587162 2 0.0336 0.936 0.008 0.992 0.000 0.000
#> GSM587163 2 0.0336 0.936 0.008 0.992 0.000 0.000
#> GSM587164 2 0.3074 0.882 0.152 0.848 0.000 0.000
#> GSM587165 2 0.1940 0.922 0.076 0.924 0.000 0.000
#> GSM587166 2 0.3649 0.847 0.204 0.796 0.000 0.000
#> GSM587167 2 0.3123 0.879 0.156 0.844 0.000 0.000
#> GSM587168 2 0.1940 0.922 0.076 0.924 0.000 0.000
#> GSM587169 2 0.0336 0.936 0.008 0.992 0.000 0.000
#> GSM587170 2 0.3074 0.882 0.152 0.848 0.000 0.000
#> GSM587171 2 0.0000 0.936 0.000 1.000 0.000 0.000
#> GSM587172 2 0.0000 0.936 0.000 1.000 0.000 0.000
#> GSM587173 2 0.1940 0.922 0.076 0.924 0.000 0.000
#> GSM587174 2 0.0336 0.936 0.008 0.992 0.000 0.000
#> GSM587175 2 0.2868 0.890 0.136 0.864 0.000 0.000
#> GSM587176 2 0.0336 0.936 0.008 0.992 0.000 0.000
#> GSM587177 2 0.1940 0.922 0.076 0.924 0.000 0.000
#> GSM587178 2 0.1716 0.926 0.064 0.936 0.000 0.000
#> GSM587179 2 0.0336 0.936 0.008 0.992 0.000 0.000
#> GSM587180 2 0.1867 0.924 0.072 0.928 0.000 0.000
#> GSM587181 2 0.0336 0.936 0.008 0.992 0.000 0.000
#> GSM587182 2 0.1792 0.925 0.068 0.932 0.000 0.000
#> GSM587183 2 0.1940 0.922 0.076 0.924 0.000 0.000
#> GSM587184 2 0.0000 0.936 0.000 1.000 0.000 0.000
#> GSM587185 2 0.0336 0.936 0.008 0.992 0.000 0.000
#> GSM587186 2 0.1940 0.922 0.076 0.924 0.000 0.000
#> GSM587187 2 0.2345 0.911 0.100 0.900 0.000 0.000
#> GSM587188 2 0.4780 0.815 0.116 0.788 0.096 0.000
#> GSM587189 2 0.4780 0.815 0.116 0.788 0.096 0.000
#> GSM587190 3 0.3328 0.912 0.100 0.024 0.872 0.004
#> GSM587203 1 0.5774 0.932 0.508 0.000 0.028 0.464
#> GSM587204 1 0.5678 0.926 0.524 0.000 0.024 0.452
#> GSM587205 1 0.5774 0.932 0.508 0.000 0.028 0.464
#> GSM587206 1 0.5774 0.932 0.508 0.000 0.028 0.464
#> GSM587207 1 0.5774 0.932 0.508 0.000 0.028 0.464
#> GSM587208 1 0.5774 0.932 0.508 0.000 0.028 0.464
#> GSM587209 1 0.5861 0.920 0.492 0.000 0.032 0.476
#> GSM587210 4 0.6972 -0.207 0.356 0.000 0.124 0.520
#> GSM587211 4 0.5862 -0.932 0.484 0.000 0.032 0.484
#> GSM587212 4 0.6677 -0.341 0.364 0.000 0.096 0.540
#> GSM587213 4 0.5861 -0.930 0.476 0.000 0.032 0.492
#> GSM587214 1 0.5861 0.924 0.488 0.000 0.032 0.480
#> GSM587215 1 0.5859 0.922 0.496 0.000 0.032 0.472
#> GSM587216 1 0.5858 0.920 0.500 0.000 0.032 0.468
#> GSM587217 1 0.5861 0.922 0.492 0.000 0.032 0.476
#> GSM587191 3 0.1722 0.928 0.048 0.008 0.944 0.000
#> GSM587192 3 0.2179 0.924 0.064 0.000 0.924 0.012
#> GSM587193 3 0.5280 0.808 0.124 0.000 0.752 0.124
#> GSM587194 3 0.4406 0.880 0.184 0.004 0.788 0.024
#> GSM587195 3 0.2048 0.920 0.064 0.008 0.928 0.000
#> GSM587196 3 0.2048 0.920 0.064 0.008 0.928 0.000
#> GSM587197 3 0.1994 0.924 0.052 0.008 0.936 0.004
#> GSM587198 3 0.1339 0.929 0.024 0.008 0.964 0.004
#> GSM587199 3 0.2530 0.924 0.072 0.008 0.912 0.008
#> GSM587200 3 0.2741 0.917 0.096 0.000 0.892 0.012
#> GSM587201 3 0.2610 0.917 0.088 0.000 0.900 0.012
#> GSM587202 3 0.1339 0.929 0.024 0.008 0.964 0.004
#> GSM198767 1 0.5774 0.932 0.508 0.000 0.028 0.464
#> GSM198769 1 0.5861 0.920 0.492 0.000 0.032 0.476
#> GSM198772 1 0.5862 0.924 0.484 0.000 0.032 0.484
#> GSM198773 4 0.5861 -0.930 0.476 0.000 0.032 0.492
#> GSM198776 1 0.5678 0.926 0.524 0.000 0.024 0.452
#> GSM198778 4 0.6972 -0.207 0.356 0.000 0.124 0.520
#> GSM198780 4 0.6677 -0.341 0.364 0.000 0.096 0.540
#> GSM198781 1 0.5861 0.924 0.488 0.000 0.032 0.480
#> GSM198765 3 0.1722 0.928 0.048 0.008 0.944 0.000
#> GSM198766 3 0.5280 0.808 0.124 0.000 0.752 0.124
#> GSM198768 3 0.2048 0.920 0.064 0.008 0.928 0.000
#> GSM198770 3 0.1994 0.924 0.052 0.008 0.936 0.004
#> GSM198771 3 0.1339 0.929 0.024 0.008 0.964 0.004
#> GSM198774 3 0.2179 0.924 0.064 0.000 0.924 0.012
#> GSM198775 3 0.4406 0.880 0.184 0.004 0.788 0.024
#> GSM198777 3 0.2048 0.920 0.064 0.008 0.928 0.000
#> GSM198779 3 0.2530 0.924 0.072 0.008 0.912 0.008
#> GSM587218 4 0.6100 0.223 0.084 0.000 0.272 0.644
#> GSM587219 4 0.1022 0.617 0.000 0.000 0.032 0.968
#> GSM587220 4 0.1022 0.617 0.000 0.000 0.032 0.968
#> GSM587221 4 0.1022 0.617 0.000 0.000 0.032 0.968
#> GSM587222 4 0.1022 0.617 0.000 0.000 0.032 0.968
#> GSM587223 4 0.1022 0.617 0.000 0.000 0.032 0.968
#> GSM587224 4 0.1022 0.617 0.000 0.000 0.032 0.968
#> GSM587225 4 0.1284 0.612 0.012 0.000 0.024 0.964
#> GSM587226 4 0.1022 0.617 0.000 0.000 0.032 0.968
#> GSM587227 4 0.1284 0.612 0.012 0.000 0.024 0.964
#> GSM587228 4 0.1284 0.612 0.012 0.000 0.024 0.964
#> GSM587229 4 0.1059 0.603 0.012 0.000 0.016 0.972
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM587155 2 0.4074 0.711 0.000 0.636 0.000 0.000 NA
#> GSM587156 2 0.4510 0.653 0.008 0.560 0.000 0.000 NA
#> GSM587157 2 0.4088 0.708 0.000 0.632 0.000 0.000 NA
#> GSM587158 2 0.0566 0.862 0.004 0.984 0.000 0.000 NA
#> GSM587159 2 0.0693 0.862 0.012 0.980 0.000 0.000 NA
#> GSM587160 2 0.0703 0.860 0.000 0.976 0.000 0.000 NA
#> GSM587161 2 0.3612 0.762 0.000 0.732 0.000 0.000 NA
#> GSM587162 2 0.1121 0.858 0.000 0.956 0.000 0.000 NA
#> GSM587163 2 0.0992 0.860 0.008 0.968 0.000 0.000 NA
#> GSM587164 2 0.4150 0.696 0.000 0.612 0.000 0.000 NA
#> GSM587165 2 0.2914 0.841 0.052 0.872 0.000 0.000 NA
#> GSM587166 2 0.4658 0.648 0.008 0.556 0.004 0.000 NA
#> GSM587167 2 0.4192 0.684 0.000 0.596 0.000 0.000 NA
#> GSM587168 2 0.3085 0.841 0.060 0.868 0.000 0.004 NA
#> GSM587169 2 0.0992 0.860 0.008 0.968 0.000 0.000 NA
#> GSM587170 2 0.4150 0.697 0.000 0.612 0.000 0.000 NA
#> GSM587171 2 0.0693 0.862 0.012 0.980 0.000 0.000 NA
#> GSM587172 2 0.0693 0.862 0.012 0.980 0.000 0.000 NA
#> GSM587173 2 0.3461 0.837 0.068 0.848 0.000 0.008 NA
#> GSM587174 2 0.0671 0.862 0.004 0.980 0.000 0.000 NA
#> GSM587175 2 0.4060 0.713 0.000 0.640 0.000 0.000 NA
#> GSM587176 2 0.0794 0.860 0.000 0.972 0.000 0.000 NA
#> GSM587177 2 0.2914 0.841 0.052 0.872 0.000 0.000 NA
#> GSM587178 2 0.2632 0.844 0.040 0.888 0.000 0.000 NA
#> GSM587179 2 0.0771 0.861 0.004 0.976 0.000 0.000 NA
#> GSM587180 2 0.3018 0.842 0.056 0.872 0.000 0.004 NA
#> GSM587181 2 0.0671 0.862 0.004 0.980 0.000 0.000 NA
#> GSM587182 2 0.2949 0.843 0.052 0.876 0.000 0.004 NA
#> GSM587183 2 0.2914 0.841 0.052 0.872 0.000 0.000 NA
#> GSM587184 2 0.0671 0.862 0.016 0.980 0.000 0.000 NA
#> GSM587185 2 0.0771 0.861 0.004 0.976 0.000 0.000 NA
#> GSM587186 2 0.3461 0.837 0.068 0.848 0.000 0.008 NA
#> GSM587187 2 0.3436 0.835 0.056 0.852 0.000 0.012 NA
#> GSM587188 2 0.6182 0.702 0.072 0.688 0.116 0.012 NA
#> GSM587189 2 0.6125 0.704 0.068 0.692 0.116 0.012 NA
#> GSM587190 3 0.4261 0.837 0.048 0.000 0.780 0.012 NA
#> GSM587203 1 0.5358 0.811 0.648 0.000 0.000 0.248 NA
#> GSM587204 1 0.5423 0.809 0.644 0.000 0.000 0.244 NA
#> GSM587205 1 0.5358 0.811 0.648 0.000 0.000 0.248 NA
#> GSM587206 1 0.5358 0.811 0.648 0.000 0.000 0.248 NA
#> GSM587207 1 0.5358 0.811 0.648 0.000 0.000 0.248 NA
#> GSM587208 1 0.5358 0.811 0.648 0.000 0.000 0.248 NA
#> GSM587209 1 0.4668 0.831 0.684 0.000 0.000 0.272 NA
#> GSM587210 1 0.7263 0.391 0.400 0.000 0.048 0.396 NA
#> GSM587211 1 0.4800 0.827 0.676 0.000 0.000 0.272 NA
#> GSM587212 1 0.6781 0.465 0.424 0.000 0.028 0.420 NA
#> GSM587213 1 0.3636 0.837 0.728 0.000 0.000 0.272 NA
#> GSM587214 1 0.3612 0.837 0.732 0.000 0.000 0.268 NA
#> GSM587215 1 0.4268 0.835 0.708 0.000 0.000 0.268 NA
#> GSM587216 1 0.5470 0.788 0.628 0.000 0.000 0.268 NA
#> GSM587217 1 0.4268 0.835 0.708 0.000 0.000 0.268 NA
#> GSM587191 3 0.4014 0.842 0.060 0.000 0.804 0.008 NA
#> GSM587192 3 0.4314 0.838 0.068 0.000 0.780 0.008 NA
#> GSM587193 3 0.7312 0.653 0.088 0.000 0.512 0.132 NA
#> GSM587194 3 0.6167 0.754 0.068 0.000 0.600 0.048 NA
#> GSM587195 3 0.2313 0.837 0.044 0.000 0.912 0.004 NA
#> GSM587196 3 0.2313 0.837 0.044 0.000 0.912 0.004 NA
#> GSM587197 3 0.2751 0.839 0.056 0.000 0.888 0.004 NA
#> GSM587198 3 0.1331 0.852 0.008 0.000 0.952 0.000 NA
#> GSM587199 3 0.2561 0.852 0.020 0.000 0.884 0.000 NA
#> GSM587200 3 0.5047 0.800 0.056 0.000 0.724 0.028 NA
#> GSM587201 3 0.5229 0.791 0.068 0.000 0.712 0.028 NA
#> GSM587202 3 0.1408 0.851 0.008 0.000 0.948 0.000 NA
#> GSM198767 1 0.5358 0.811 0.648 0.000 0.000 0.248 NA
#> GSM198769 1 0.4668 0.831 0.684 0.000 0.000 0.272 NA
#> GSM198772 1 0.4800 0.827 0.676 0.000 0.000 0.272 NA
#> GSM198773 1 0.3636 0.837 0.728 0.000 0.000 0.272 NA
#> GSM198776 1 0.5423 0.809 0.644 0.000 0.000 0.244 NA
#> GSM198778 1 0.7263 0.391 0.400 0.000 0.048 0.396 NA
#> GSM198780 1 0.6781 0.465 0.424 0.000 0.028 0.420 NA
#> GSM198781 1 0.3612 0.837 0.732 0.000 0.000 0.268 NA
#> GSM198765 3 0.4014 0.842 0.060 0.000 0.804 0.008 NA
#> GSM198766 3 0.7312 0.653 0.088 0.000 0.512 0.132 NA
#> GSM198768 3 0.2313 0.837 0.044 0.000 0.912 0.004 NA
#> GSM198770 3 0.2751 0.839 0.056 0.000 0.888 0.004 NA
#> GSM198771 3 0.1331 0.852 0.008 0.000 0.952 0.000 NA
#> GSM198774 3 0.4314 0.838 0.068 0.000 0.780 0.008 NA
#> GSM198775 3 0.6167 0.754 0.068 0.000 0.600 0.048 NA
#> GSM198777 3 0.2313 0.837 0.044 0.000 0.912 0.004 NA
#> GSM198779 3 0.2561 0.852 0.020 0.000 0.884 0.000 NA
#> GSM587218 4 0.4282 0.703 0.064 0.000 0.112 0.800 NA
#> GSM587219 4 0.0609 0.955 0.000 0.000 0.020 0.980 NA
#> GSM587220 4 0.0609 0.955 0.000 0.000 0.020 0.980 NA
#> GSM587221 4 0.0609 0.955 0.000 0.000 0.020 0.980 NA
#> GSM587222 4 0.0609 0.955 0.000 0.000 0.020 0.980 NA
#> GSM587223 4 0.0609 0.955 0.000 0.000 0.020 0.980 NA
#> GSM587224 4 0.0771 0.952 0.004 0.000 0.020 0.976 NA
#> GSM587225 4 0.1461 0.943 0.004 0.000 0.016 0.952 NA
#> GSM587226 4 0.0609 0.955 0.000 0.000 0.020 0.980 NA
#> GSM587227 4 0.1461 0.943 0.004 0.000 0.016 0.952 NA
#> GSM587228 4 0.1461 0.943 0.004 0.000 0.016 0.952 NA
#> GSM587229 4 0.1483 0.939 0.008 0.000 0.012 0.952 NA
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM587155 6 0.3989 0.91900 0.000 0.468 0.000 0.004 0.000 0.528
#> GSM587156 6 0.5233 0.86719 0.000 0.384 0.000 0.012 0.068 0.536
#> GSM587157 6 0.3851 0.93042 0.000 0.460 0.000 0.000 0.000 0.540
#> GSM587158 2 0.0551 0.74351 0.000 0.984 0.000 0.004 0.004 0.008
#> GSM587159 2 0.0551 0.74351 0.000 0.984 0.000 0.004 0.004 0.008
#> GSM587160 2 0.1074 0.72474 0.000 0.960 0.000 0.012 0.000 0.028
#> GSM587161 2 0.3714 -0.42252 0.000 0.656 0.000 0.004 0.000 0.340
#> GSM587162 2 0.2006 0.66514 0.000 0.904 0.000 0.016 0.000 0.080
#> GSM587163 2 0.0632 0.73003 0.000 0.976 0.000 0.000 0.000 0.024
#> GSM587164 6 0.3833 0.93680 0.000 0.444 0.000 0.000 0.000 0.556
#> GSM587165 2 0.4055 0.70416 0.000 0.792 0.000 0.068 0.040 0.100
#> GSM587166 6 0.5233 0.86719 0.000 0.384 0.000 0.012 0.068 0.536
#> GSM587167 6 0.4199 0.93539 0.000 0.444 0.000 0.004 0.008 0.544
#> GSM587168 2 0.4637 0.68215 0.000 0.752 0.000 0.080 0.072 0.096
#> GSM587169 2 0.0713 0.73187 0.000 0.972 0.000 0.000 0.000 0.028
#> GSM587170 6 0.4103 0.93706 0.000 0.448 0.000 0.004 0.004 0.544
#> GSM587171 2 0.0551 0.74351 0.000 0.984 0.000 0.004 0.004 0.008
#> GSM587172 2 0.0551 0.74351 0.000 0.984 0.000 0.004 0.004 0.008
#> GSM587173 2 0.5020 0.65040 0.000 0.720 0.000 0.084 0.092 0.104
#> GSM587174 2 0.0914 0.74833 0.000 0.968 0.000 0.016 0.000 0.016
#> GSM587175 6 0.3854 0.92712 0.000 0.464 0.000 0.000 0.000 0.536
#> GSM587176 2 0.1367 0.71094 0.000 0.944 0.000 0.012 0.000 0.044
#> GSM587177 2 0.3861 0.71035 0.000 0.804 0.000 0.064 0.032 0.100
#> GSM587178 2 0.2402 0.73896 0.000 0.888 0.000 0.020 0.008 0.084
#> GSM587179 2 0.1605 0.72375 0.000 0.940 0.000 0.016 0.012 0.032
#> GSM587180 2 0.4435 0.69328 0.000 0.768 0.000 0.080 0.064 0.088
#> GSM587181 2 0.0914 0.74833 0.000 0.968 0.000 0.016 0.000 0.016
#> GSM587182 2 0.4284 0.70003 0.000 0.780 0.000 0.080 0.064 0.076
#> GSM587183 2 0.3861 0.71035 0.000 0.804 0.000 0.064 0.032 0.100
#> GSM587184 2 0.0767 0.74816 0.000 0.976 0.000 0.012 0.004 0.008
#> GSM587185 2 0.1605 0.72375 0.000 0.940 0.000 0.016 0.012 0.032
#> GSM587186 2 0.5020 0.65040 0.000 0.720 0.000 0.084 0.092 0.104
#> GSM587187 2 0.5998 0.55177 0.000 0.640 0.016 0.068 0.108 0.168
#> GSM587188 2 0.7664 0.31057 0.000 0.488 0.108 0.072 0.144 0.188
#> GSM587189 2 0.7614 0.31586 0.000 0.496 0.108 0.072 0.140 0.184
#> GSM587190 3 0.6133 0.12518 0.000 0.004 0.540 0.024 0.256 0.176
#> GSM587203 1 0.2712 0.80328 0.864 0.000 0.000 0.016 0.012 0.108
#> GSM587204 1 0.2880 0.80207 0.856 0.000 0.000 0.012 0.024 0.108
#> GSM587205 1 0.2712 0.80328 0.864 0.000 0.000 0.016 0.012 0.108
#> GSM587206 1 0.2712 0.80328 0.864 0.000 0.000 0.016 0.012 0.108
#> GSM587207 1 0.2712 0.80328 0.864 0.000 0.000 0.016 0.012 0.108
#> GSM587208 1 0.2712 0.80328 0.864 0.000 0.000 0.016 0.012 0.108
#> GSM587209 1 0.2351 0.79901 0.900 0.000 0.000 0.012 0.052 0.036
#> GSM587210 1 0.6918 0.32705 0.452 0.000 0.008 0.164 0.308 0.068
#> GSM587211 1 0.2384 0.80019 0.896 0.000 0.000 0.008 0.056 0.040
#> GSM587212 1 0.6500 0.43068 0.536 0.000 0.004 0.156 0.240 0.064
#> GSM587213 1 0.0458 0.81806 0.984 0.000 0.000 0.016 0.000 0.000
#> GSM587214 1 0.0458 0.81806 0.984 0.000 0.000 0.016 0.000 0.000
#> GSM587215 1 0.1562 0.81060 0.940 0.000 0.000 0.004 0.024 0.032
#> GSM587216 1 0.3822 0.73786 0.800 0.000 0.000 0.020 0.112 0.068
#> GSM587217 1 0.1675 0.81122 0.936 0.000 0.000 0.008 0.024 0.032
#> GSM587191 3 0.4603 0.34482 0.008 0.000 0.664 0.016 0.288 0.024
#> GSM587192 3 0.4965 0.00625 0.008 0.000 0.552 0.020 0.400 0.020
#> GSM587193 5 0.6080 0.82226 0.032 0.000 0.280 0.104 0.568 0.016
#> GSM587194 5 0.4928 0.81265 0.000 0.000 0.288 0.040 0.640 0.032
#> GSM587195 3 0.1109 0.61845 0.004 0.000 0.964 0.012 0.016 0.004
#> GSM587196 3 0.1109 0.61845 0.004 0.000 0.964 0.012 0.016 0.004
#> GSM587197 3 0.1419 0.61508 0.004 0.000 0.952 0.016 0.012 0.016
#> GSM587198 3 0.3655 0.58813 0.004 0.000 0.804 0.020 0.144 0.028
#> GSM587199 3 0.4545 0.45838 0.000 0.000 0.668 0.020 0.280 0.032
#> GSM587200 3 0.6252 0.02119 0.016 0.000 0.476 0.044 0.388 0.076
#> GSM587201 3 0.6325 0.00790 0.020 0.000 0.472 0.044 0.388 0.076
#> GSM587202 3 0.3577 0.59018 0.004 0.000 0.812 0.020 0.136 0.028
#> GSM198767 1 0.2712 0.80328 0.864 0.000 0.000 0.016 0.012 0.108
#> GSM198769 1 0.2351 0.79901 0.900 0.000 0.000 0.012 0.052 0.036
#> GSM198772 1 0.2384 0.80019 0.896 0.000 0.000 0.008 0.056 0.040
#> GSM198773 1 0.0458 0.81806 0.984 0.000 0.000 0.016 0.000 0.000
#> GSM198776 1 0.2880 0.80207 0.856 0.000 0.000 0.012 0.024 0.108
#> GSM198778 1 0.6918 0.32705 0.452 0.000 0.008 0.164 0.308 0.068
#> GSM198780 1 0.6500 0.43068 0.536 0.000 0.004 0.156 0.240 0.064
#> GSM198781 1 0.0458 0.81806 0.984 0.000 0.000 0.016 0.000 0.000
#> GSM198765 3 0.4603 0.34482 0.008 0.000 0.664 0.016 0.288 0.024
#> GSM198766 5 0.6080 0.82226 0.032 0.000 0.280 0.104 0.568 0.016
#> GSM198768 3 0.1109 0.61845 0.004 0.000 0.964 0.012 0.016 0.004
#> GSM198770 3 0.1419 0.61508 0.004 0.000 0.952 0.016 0.012 0.016
#> GSM198771 3 0.3655 0.58813 0.004 0.000 0.804 0.020 0.144 0.028
#> GSM198774 3 0.4965 0.00625 0.008 0.000 0.552 0.020 0.400 0.020
#> GSM198775 5 0.4928 0.81265 0.000 0.000 0.288 0.040 0.640 0.032
#> GSM198777 3 0.1109 0.61845 0.004 0.000 0.964 0.012 0.016 0.004
#> GSM198779 3 0.4545 0.45838 0.000 0.000 0.668 0.020 0.280 0.032
#> GSM587218 4 0.3755 0.71676 0.044 0.000 0.056 0.816 0.084 0.000
#> GSM587219 4 0.2915 0.95268 0.184 0.000 0.008 0.808 0.000 0.000
#> GSM587220 4 0.2915 0.95268 0.184 0.000 0.008 0.808 0.000 0.000
#> GSM587221 4 0.2915 0.95268 0.184 0.000 0.008 0.808 0.000 0.000
#> GSM587222 4 0.2915 0.95268 0.184 0.000 0.008 0.808 0.000 0.000
#> GSM587223 4 0.2915 0.95268 0.184 0.000 0.008 0.808 0.000 0.000
#> GSM587224 4 0.2882 0.95020 0.180 0.000 0.008 0.812 0.000 0.000
#> GSM587225 4 0.4138 0.93246 0.184 0.000 0.000 0.752 0.020 0.044
#> GSM587226 4 0.2915 0.95268 0.184 0.000 0.008 0.808 0.000 0.000
#> GSM587227 4 0.4138 0.93246 0.184 0.000 0.000 0.752 0.020 0.044
#> GSM587228 4 0.4138 0.93246 0.184 0.000 0.000 0.752 0.020 0.044
#> GSM587229 4 0.4138 0.93246 0.184 0.000 0.000 0.752 0.020 0.044
Heatmaps for the consensus matrix. It visualizes the probability of two samples to be in a same group.
consensus_heatmap(res, k = 2)
consensus_heatmap(res, k = 3)
consensus_heatmap(res, k = 4)
consensus_heatmap(res, k = 5)
consensus_heatmap(res, k = 6)
Heatmaps for the membership of samples in all partitions to see how consistent they are:
membership_heatmap(res, k = 2)
membership_heatmap(res, k = 3)
membership_heatmap(res, k = 4)
membership_heatmap(res, k = 5)
membership_heatmap(res, k = 6)
As soon as we have had the classes for columns, we can look for signatures which are significantly different between classes which can be candidate marks for certain classes. Following are the heatmaps for signatures.
Signature heatmaps where rows are scaled:
get_signatures(res, k = 2)
get_signatures(res, k = 3)
get_signatures(res, k = 4)
get_signatures(res, k = 5)
get_signatures(res, k = 6)
Signature heatmaps where rows are not scaled:
get_signatures(res, k = 2, scale_rows = FALSE)
get_signatures(res, k = 3, scale_rows = FALSE)
get_signatures(res, k = 4, scale_rows = FALSE)
get_signatures(res, k = 5, scale_rows = FALSE)
get_signatures(res, k = 6, scale_rows = FALSE)
Compare the overlap of signatures from different k:
compare_signatures(res)
get_signature() returns a data frame invisibly. TO get the list of signatures, the function
call should be assigned to a variable explicitly. In following code, if plot argument is set
to FALSE, no heatmap is plotted while only the differential analysis is performed.
# code only for demonstration
tb = get_signature(res, k = ..., plot = FALSE)
An example of the output of tb is:
#> which_row fdr mean_1 mean_2 scaled_mean_1 scaled_mean_2 km
#> 1 38 0.042760348 8.373488 9.131774 -0.5533452 0.5164555 1
#> 2 40 0.018707592 7.106213 8.469186 -0.6173731 0.5762149 1
#> 3 55 0.019134737 10.221463 11.207825 -0.6159697 0.5749050 1
#> 4 59 0.006059896 5.921854 7.869574 -0.6899429 0.6439467 1
#> 5 60 0.018055526 8.928898 10.211722 -0.6204761 0.5791110 1
#> 6 98 0.009384629 15.714769 14.887706 0.6635654 -0.6193277 2
...
The columns in tb are:
which_row: row indices corresponding to the input matrix.fdr: FDR for the differential test. mean_x: The mean value in group x.scaled_mean_x: The mean value in group x after rows are scaled.km: Row groups if k-means clustering is applied to rows.UMAP plot which shows how samples are separated.
dimension_reduction(res, k = 2, method = "UMAP")
dimension_reduction(res, k = 3, method = "UMAP")
dimension_reduction(res, k = 4, method = "UMAP")
dimension_reduction(res, k = 5, method = "UMAP")
dimension_reduction(res, k = 6, method = "UMAP")
Following heatmap shows how subgroups are split when increasing k:
collect_classes(res)
Test correlation between subgroups and known annotations. If the known annotation is numeric, one-way ANOVA test is applied, and if the known annotation is discrete, chi-squared contingency table test is applied.
test_to_known_factors(res)
#> n specimen(p) k
#> MAD:kmeans 92 4.01e-14 2
#> MAD:kmeans 92 1.23e-30 3
#> MAD:kmeans 84 2.84e-42 4
#> MAD:kmeans 88 1.14e-44 5
#> MAD:kmeans 76 1.62e-32 6
If matrix rows can be associated to genes, consider to use functional_enrichment(res,
...) to perform function enrichment for the signature genes. See this vignette for more detailed explanations.
The object with results only for a single top-value method and a single partition method can be extracted as:
res = res_list["MAD", "skmeans"]
# you can also extract it by
# res = res_list["MAD:skmeans"]
A summary of res and all the functions that can be applied to it:
res
#> A 'ConsensusPartition' object with k = 2, 3, 4, 5, 6.
#> On a matrix with 51941 rows and 92 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 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 1.000 0.974 0.987 0.5038 0.497 0.497
#> 3 3 1.000 0.966 0.984 0.2693 0.832 0.671
#> 4 4 0.964 0.953 0.972 0.1338 0.890 0.704
#> 5 5 0.959 0.924 0.959 0.0335 0.974 0.908
#> 6 6 0.926 0.837 0.897 0.0285 0.989 0.957
suggest_best_k() suggests the best \(k\) based on these statistics. The rules are as follows:
suggest_best_k(res)
#> [1] 5
#> attr(,"optional")
#> [1] 2 3 4
There is also optional best \(k\) = 2 3 4 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
#> GSM587155 2 0.000 0.977 0.000 1.000
#> GSM587156 2 0.000 0.977 0.000 1.000
#> GSM587157 2 0.000 0.977 0.000 1.000
#> GSM587158 2 0.000 0.977 0.000 1.000
#> GSM587159 2 0.000 0.977 0.000 1.000
#> GSM587160 2 0.000 0.977 0.000 1.000
#> GSM587161 2 0.000 0.977 0.000 1.000
#> GSM587162 2 0.000 0.977 0.000 1.000
#> GSM587163 2 0.000 0.977 0.000 1.000
#> GSM587164 2 0.000 0.977 0.000 1.000
#> GSM587165 2 0.000 0.977 0.000 1.000
#> GSM587166 2 0.000 0.977 0.000 1.000
#> GSM587167 2 0.000 0.977 0.000 1.000
#> GSM587168 2 0.000 0.977 0.000 1.000
#> GSM587169 2 0.000 0.977 0.000 1.000
#> GSM587170 2 0.000 0.977 0.000 1.000
#> GSM587171 2 0.000 0.977 0.000 1.000
#> GSM587172 2 0.000 0.977 0.000 1.000
#> GSM587173 2 0.000 0.977 0.000 1.000
#> GSM587174 2 0.000 0.977 0.000 1.000
#> GSM587175 2 0.000 0.977 0.000 1.000
#> GSM587176 2 0.000 0.977 0.000 1.000
#> GSM587177 2 0.000 0.977 0.000 1.000
#> GSM587178 2 0.000 0.977 0.000 1.000
#> GSM587179 2 0.000 0.977 0.000 1.000
#> GSM587180 2 0.000 0.977 0.000 1.000
#> GSM587181 2 0.000 0.977 0.000 1.000
#> GSM587182 2 0.000 0.977 0.000 1.000
#> GSM587183 2 0.000 0.977 0.000 1.000
#> GSM587184 2 0.000 0.977 0.000 1.000
#> GSM587185 2 0.000 0.977 0.000 1.000
#> GSM587186 2 0.000 0.977 0.000 1.000
#> GSM587187 2 0.000 0.977 0.000 1.000
#> GSM587188 2 0.000 0.977 0.000 1.000
#> GSM587189 2 0.000 0.977 0.000 1.000
#> GSM587190 2 0.000 0.977 0.000 1.000
#> GSM587203 1 0.000 0.998 1.000 0.000
#> GSM587204 1 0.000 0.998 1.000 0.000
#> GSM587205 1 0.000 0.998 1.000 0.000
#> GSM587206 1 0.000 0.998 1.000 0.000
#> GSM587207 1 0.000 0.998 1.000 0.000
#> GSM587208 1 0.000 0.998 1.000 0.000
#> GSM587209 1 0.000 0.998 1.000 0.000
#> GSM587210 1 0.000 0.998 1.000 0.000
#> GSM587211 1 0.000 0.998 1.000 0.000
#> GSM587212 1 0.000 0.998 1.000 0.000
#> GSM587213 1 0.000 0.998 1.000 0.000
#> GSM587214 1 0.000 0.998 1.000 0.000
#> GSM587215 1 0.000 0.998 1.000 0.000
#> GSM587216 1 0.000 0.998 1.000 0.000
#> GSM587217 1 0.000 0.998 1.000 0.000
#> GSM587191 2 0.000 0.977 0.000 1.000
#> GSM587192 1 0.000 0.998 1.000 0.000
#> GSM587193 1 0.000 0.998 1.000 0.000
#> GSM587194 1 0.278 0.950 0.952 0.048
#> GSM587195 2 0.000 0.977 0.000 1.000
#> GSM587196 2 0.260 0.943 0.044 0.956
#> GSM587197 2 0.000 0.977 0.000 1.000
#> GSM587198 2 0.706 0.786 0.192 0.808
#> GSM587199 2 0.722 0.776 0.200 0.800
#> GSM587200 1 0.000 0.998 1.000 0.000
#> GSM587201 1 0.000 0.998 1.000 0.000
#> GSM587202 2 0.722 0.776 0.200 0.800
#> GSM198767 1 0.000 0.998 1.000 0.000
#> GSM198769 1 0.000 0.998 1.000 0.000
#> GSM198772 1 0.000 0.998 1.000 0.000
#> GSM198773 1 0.000 0.998 1.000 0.000
#> GSM198776 1 0.000 0.998 1.000 0.000
#> GSM198778 1 0.000 0.998 1.000 0.000
#> GSM198780 1 0.000 0.998 1.000 0.000
#> GSM198781 1 0.000 0.998 1.000 0.000
#> GSM198765 2 0.224 0.950 0.036 0.964
#> GSM198766 1 0.000 0.998 1.000 0.000
#> GSM198768 2 0.000 0.977 0.000 1.000
#> GSM198770 2 0.000 0.977 0.000 1.000
#> GSM198771 2 0.722 0.776 0.200 0.800
#> GSM198774 1 0.000 0.998 1.000 0.000
#> GSM198775 1 0.278 0.950 0.952 0.048
#> GSM198777 2 0.260 0.943 0.044 0.956
#> GSM198779 2 0.722 0.776 0.200 0.800
#> GSM587218 1 0.000 0.998 1.000 0.000
#> GSM587219 1 0.000 0.998 1.000 0.000
#> GSM587220 1 0.000 0.998 1.000 0.000
#> GSM587221 1 0.000 0.998 1.000 0.000
#> GSM587222 1 0.000 0.998 1.000 0.000
#> GSM587223 1 0.000 0.998 1.000 0.000
#> GSM587224 1 0.000 0.998 1.000 0.000
#> GSM587225 1 0.000 0.998 1.000 0.000
#> GSM587226 1 0.000 0.998 1.000 0.000
#> GSM587227 1 0.000 0.998 1.000 0.000
#> GSM587228 1 0.000 0.998 1.000 0.000
#> GSM587229 1 0.000 0.998 1.000 0.000
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM587155 2 0.000 1.000 0.000 1.00 0.000
#> GSM587156 2 0.000 1.000 0.000 1.00 0.000
#> GSM587157 2 0.000 1.000 0.000 1.00 0.000
#> GSM587158 2 0.000 1.000 0.000 1.00 0.000
#> GSM587159 2 0.000 1.000 0.000 1.00 0.000
#> GSM587160 2 0.000 1.000 0.000 1.00 0.000
#> GSM587161 2 0.000 1.000 0.000 1.00 0.000
#> GSM587162 2 0.000 1.000 0.000 1.00 0.000
#> GSM587163 2 0.000 1.000 0.000 1.00 0.000
#> GSM587164 2 0.000 1.000 0.000 1.00 0.000
#> GSM587165 2 0.000 1.000 0.000 1.00 0.000
#> GSM587166 2 0.000 1.000 0.000 1.00 0.000
#> GSM587167 2 0.000 1.000 0.000 1.00 0.000
#> GSM587168 2 0.000 1.000 0.000 1.00 0.000
#> GSM587169 2 0.000 1.000 0.000 1.00 0.000
#> GSM587170 2 0.000 1.000 0.000 1.00 0.000
#> GSM587171 2 0.000 1.000 0.000 1.00 0.000
#> GSM587172 2 0.000 1.000 0.000 1.00 0.000
#> GSM587173 2 0.000 1.000 0.000 1.00 0.000
#> GSM587174 2 0.000 1.000 0.000 1.00 0.000
#> GSM587175 2 0.000 1.000 0.000 1.00 0.000
#> GSM587176 2 0.000 1.000 0.000 1.00 0.000
#> GSM587177 2 0.000 1.000 0.000 1.00 0.000
#> GSM587178 2 0.000 1.000 0.000 1.00 0.000
#> GSM587179 2 0.000 1.000 0.000 1.00 0.000
#> GSM587180 2 0.000 1.000 0.000 1.00 0.000
#> GSM587181 2 0.000 1.000 0.000 1.00 0.000
#> GSM587182 2 0.000 1.000 0.000 1.00 0.000
#> GSM587183 2 0.000 1.000 0.000 1.00 0.000
#> GSM587184 2 0.000 1.000 0.000 1.00 0.000
#> GSM587185 2 0.000 1.000 0.000 1.00 0.000
#> GSM587186 2 0.000 1.000 0.000 1.00 0.000
#> GSM587187 2 0.000 1.000 0.000 1.00 0.000
#> GSM587188 2 0.000 1.000 0.000 1.00 0.000
#> GSM587189 2 0.000 1.000 0.000 1.00 0.000
#> GSM587190 3 0.628 0.224 0.000 0.46 0.540
#> GSM587203 1 0.000 0.995 1.000 0.00 0.000
#> GSM587204 1 0.000 0.995 1.000 0.00 0.000
#> GSM587205 1 0.000 0.995 1.000 0.00 0.000
#> GSM587206 1 0.000 0.995 1.000 0.00 0.000
#> GSM587207 1 0.000 0.995 1.000 0.00 0.000
#> GSM587208 1 0.000 0.995 1.000 0.00 0.000
#> GSM587209 1 0.000 0.995 1.000 0.00 0.000
#> GSM587210 1 0.000 0.995 1.000 0.00 0.000
#> GSM587211 1 0.000 0.995 1.000 0.00 0.000
#> GSM587212 1 0.000 0.995 1.000 0.00 0.000
#> GSM587213 1 0.000 0.995 1.000 0.00 0.000
#> GSM587214 1 0.000 0.995 1.000 0.00 0.000
#> GSM587215 1 0.000 0.995 1.000 0.00 0.000
#> GSM587216 1 0.000 0.995 1.000 0.00 0.000
#> GSM587217 1 0.000 0.995 1.000 0.00 0.000
#> GSM587191 3 0.000 0.931 0.000 0.00 1.000
#> GSM587192 3 0.000 0.931 0.000 0.00 1.000
#> GSM587193 1 0.000 0.995 1.000 0.00 0.000
#> GSM587194 3 0.811 0.537 0.096 0.30 0.604
#> GSM587195 3 0.000 0.931 0.000 0.00 1.000
#> GSM587196 3 0.000 0.931 0.000 0.00 1.000
#> GSM587197 3 0.000 0.931 0.000 0.00 1.000
#> GSM587198 3 0.000 0.931 0.000 0.00 1.000
#> GSM587199 3 0.000 0.931 0.000 0.00 1.000
#> GSM587200 1 0.388 0.817 0.848 0.00 0.152
#> GSM587201 1 0.141 0.959 0.964 0.00 0.036
#> GSM587202 3 0.000 0.931 0.000 0.00 1.000
#> GSM198767 1 0.000 0.995 1.000 0.00 0.000
#> GSM198769 1 0.000 0.995 1.000 0.00 0.000
#> GSM198772 1 0.000 0.995 1.000 0.00 0.000
#> GSM198773 1 0.000 0.995 1.000 0.00 0.000
#> GSM198776 1 0.000 0.995 1.000 0.00 0.000
#> GSM198778 1 0.000 0.995 1.000 0.00 0.000
#> GSM198780 1 0.000 0.995 1.000 0.00 0.000
#> GSM198781 1 0.000 0.995 1.000 0.00 0.000
#> GSM198765 3 0.000 0.931 0.000 0.00 1.000
#> GSM198766 1 0.000 0.995 1.000 0.00 0.000
#> GSM198768 3 0.000 0.931 0.000 0.00 1.000
#> GSM198770 3 0.000 0.931 0.000 0.00 1.000
#> GSM198771 3 0.000 0.931 0.000 0.00 1.000
#> GSM198774 3 0.000 0.931 0.000 0.00 1.000
#> GSM198775 3 0.811 0.537 0.096 0.30 0.604
#> GSM198777 3 0.000 0.931 0.000 0.00 1.000
#> GSM198779 3 0.000 0.931 0.000 0.00 1.000
#> GSM587218 1 0.000 0.995 1.000 0.00 0.000
#> GSM587219 1 0.000 0.995 1.000 0.00 0.000
#> GSM587220 1 0.000 0.995 1.000 0.00 0.000
#> GSM587221 1 0.000 0.995 1.000 0.00 0.000
#> GSM587222 1 0.000 0.995 1.000 0.00 0.000
#> GSM587223 1 0.000 0.995 1.000 0.00 0.000
#> GSM587224 1 0.000 0.995 1.000 0.00 0.000
#> GSM587225 1 0.000 0.995 1.000 0.00 0.000
#> GSM587226 1 0.000 0.995 1.000 0.00 0.000
#> GSM587227 1 0.000 0.995 1.000 0.00 0.000
#> GSM587228 1 0.000 0.995 1.000 0.00 0.000
#> GSM587229 1 0.000 0.995 1.000 0.00 0.000
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM587155 2 0.0000 0.990 0.000 1.000 0.000 0.000
#> GSM587156 2 0.0000 0.990 0.000 1.000 0.000 0.000
#> GSM587157 2 0.0000 0.990 0.000 1.000 0.000 0.000
#> GSM587158 2 0.0000 0.990 0.000 1.000 0.000 0.000
#> GSM587159 2 0.0000 0.990 0.000 1.000 0.000 0.000
#> GSM587160 2 0.0000 0.990 0.000 1.000 0.000 0.000
#> GSM587161 2 0.0000 0.990 0.000 1.000 0.000 0.000
#> GSM587162 2 0.0000 0.990 0.000 1.000 0.000 0.000
#> GSM587163 2 0.0000 0.990 0.000 1.000 0.000 0.000
#> GSM587164 2 0.0000 0.990 0.000 1.000 0.000 0.000
#> GSM587165 2 0.0000 0.990 0.000 1.000 0.000 0.000
#> GSM587166 2 0.0000 0.990 0.000 1.000 0.000 0.000
#> GSM587167 2 0.0000 0.990 0.000 1.000 0.000 0.000
#> GSM587168 2 0.0000 0.990 0.000 1.000 0.000 0.000
#> GSM587169 2 0.0000 0.990 0.000 1.000 0.000 0.000
#> GSM587170 2 0.0000 0.990 0.000 1.000 0.000 0.000
#> GSM587171 2 0.0000 0.990 0.000 1.000 0.000 0.000
#> GSM587172 2 0.0000 0.990 0.000 1.000 0.000 0.000
#> GSM587173 2 0.0000 0.990 0.000 1.000 0.000 0.000
#> GSM587174 2 0.0000 0.990 0.000 1.000 0.000 0.000
#> GSM587175 2 0.0000 0.990 0.000 1.000 0.000 0.000
#> GSM587176 2 0.0000 0.990 0.000 1.000 0.000 0.000
#> GSM587177 2 0.0000 0.990 0.000 1.000 0.000 0.000
#> GSM587178 2 0.0000 0.990 0.000 1.000 0.000 0.000
#> GSM587179 2 0.0000 0.990 0.000 1.000 0.000 0.000
#> GSM587180 2 0.0000 0.990 0.000 1.000 0.000 0.000
#> GSM587181 2 0.0000 0.990 0.000 1.000 0.000 0.000
#> GSM587182 2 0.0000 0.990 0.000 1.000 0.000 0.000
#> GSM587183 2 0.0000 0.990 0.000 1.000 0.000 0.000
#> GSM587184 2 0.0000 0.990 0.000 1.000 0.000 0.000
#> GSM587185 2 0.0000 0.990 0.000 1.000 0.000 0.000
#> GSM587186 2 0.0000 0.990 0.000 1.000 0.000 0.000
#> GSM587187 2 0.0000 0.990 0.000 1.000 0.000 0.000
#> GSM587188 2 0.0000 0.990 0.000 1.000 0.000 0.000
#> GSM587189 2 0.0000 0.990 0.000 1.000 0.000 0.000
#> GSM587190 2 0.5271 0.483 0.000 0.656 0.320 0.024
#> GSM587203 1 0.0000 0.942 1.000 0.000 0.000 0.000
#> GSM587204 1 0.0000 0.942 1.000 0.000 0.000 0.000
#> GSM587205 1 0.0000 0.942 1.000 0.000 0.000 0.000
#> GSM587206 1 0.0000 0.942 1.000 0.000 0.000 0.000
#> GSM587207 1 0.0000 0.942 1.000 0.000 0.000 0.000
#> GSM587208 1 0.0000 0.942 1.000 0.000 0.000 0.000
#> GSM587209 1 0.0000 0.942 1.000 0.000 0.000 0.000
#> GSM587210 1 0.3873 0.732 0.772 0.000 0.000 0.228
#> GSM587211 1 0.0000 0.942 1.000 0.000 0.000 0.000
#> GSM587212 1 0.3873 0.732 0.772 0.000 0.000 0.228
#> GSM587213 1 0.0000 0.942 1.000 0.000 0.000 0.000
#> GSM587214 1 0.0000 0.942 1.000 0.000 0.000 0.000
#> GSM587215 1 0.0000 0.942 1.000 0.000 0.000 0.000
#> GSM587216 1 0.0000 0.942 1.000 0.000 0.000 0.000
#> GSM587217 1 0.0000 0.942 1.000 0.000 0.000 0.000
#> GSM587191 3 0.1474 0.964 0.000 0.000 0.948 0.052
#> GSM587192 3 0.1867 0.951 0.000 0.000 0.928 0.072
#> GSM587193 4 0.0592 0.946 0.016 0.000 0.000 0.984
#> GSM587194 4 0.0524 0.943 0.000 0.004 0.008 0.988
#> GSM587195 3 0.0000 0.986 0.000 0.000 1.000 0.000
#> GSM587196 3 0.0000 0.986 0.000 0.000 1.000 0.000
#> GSM587197 3 0.0000 0.986 0.000 0.000 1.000 0.000
#> GSM587198 3 0.0000 0.986 0.000 0.000 1.000 0.000
#> GSM587199 3 0.0000 0.986 0.000 0.000 1.000 0.000
#> GSM587200 1 0.6429 0.611 0.644 0.000 0.144 0.212
#> GSM587201 1 0.2048 0.889 0.928 0.000 0.064 0.008
#> GSM587202 3 0.0000 0.986 0.000 0.000 1.000 0.000
#> GSM198767 1 0.0000 0.942 1.000 0.000 0.000 0.000
#> GSM198769 1 0.0000 0.942 1.000 0.000 0.000 0.000
#> GSM198772 1 0.0000 0.942 1.000 0.000 0.000 0.000
#> GSM198773 1 0.0000 0.942 1.000 0.000 0.000 0.000
#> GSM198776 1 0.0000 0.942 1.000 0.000 0.000 0.000
#> GSM198778 1 0.3873 0.732 0.772 0.000 0.000 0.228
#> GSM198780 1 0.3873 0.732 0.772 0.000 0.000 0.228
#> GSM198781 1 0.0000 0.942 1.000 0.000 0.000 0.000
#> GSM198765 3 0.1474 0.964 0.000 0.000 0.948 0.052
#> GSM198766 4 0.0592 0.946 0.016 0.000 0.000 0.984
#> GSM198768 3 0.0000 0.986 0.000 0.000 1.000 0.000
#> GSM198770 3 0.0000 0.986 0.000 0.000 1.000 0.000
#> GSM198771 3 0.0000 0.986 0.000 0.000 1.000 0.000
#> GSM198774 3 0.1867 0.951 0.000 0.000 0.928 0.072
#> GSM198775 4 0.0524 0.943 0.000 0.004 0.008 0.988
#> GSM198777 3 0.0000 0.986 0.000 0.000 1.000 0.000
#> GSM198779 3 0.0000 0.986 0.000 0.000 1.000 0.000
#> GSM587218 4 0.1118 0.973 0.036 0.000 0.000 0.964
#> GSM587219 4 0.1474 0.981 0.052 0.000 0.000 0.948
#> GSM587220 4 0.1474 0.981 0.052 0.000 0.000 0.948
#> GSM587221 4 0.1474 0.981 0.052 0.000 0.000 0.948
#> GSM587222 4 0.1474 0.981 0.052 0.000 0.000 0.948
#> GSM587223 4 0.1474 0.981 0.052 0.000 0.000 0.948
#> GSM587224 4 0.1474 0.981 0.052 0.000 0.000 0.948
#> GSM587225 4 0.1474 0.981 0.052 0.000 0.000 0.948
#> GSM587226 4 0.1474 0.981 0.052 0.000 0.000 0.948
#> GSM587227 4 0.1474 0.981 0.052 0.000 0.000 0.948
#> GSM587228 4 0.1474 0.981 0.052 0.000 0.000 0.948
#> GSM587229 4 0.1474 0.981 0.052 0.000 0.000 0.948
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM587155 2 0.0000 0.9890 0.000 1.000 0.000 0.000 0.000
#> GSM587156 2 0.0000 0.9890 0.000 1.000 0.000 0.000 0.000
#> GSM587157 2 0.0000 0.9890 0.000 1.000 0.000 0.000 0.000
#> GSM587158 2 0.0000 0.9890 0.000 1.000 0.000 0.000 0.000
#> GSM587159 2 0.0000 0.9890 0.000 1.000 0.000 0.000 0.000
#> GSM587160 2 0.0000 0.9890 0.000 1.000 0.000 0.000 0.000
#> GSM587161 2 0.0000 0.9890 0.000 1.000 0.000 0.000 0.000
#> GSM587162 2 0.0000 0.9890 0.000 1.000 0.000 0.000 0.000
#> GSM587163 2 0.0000 0.9890 0.000 1.000 0.000 0.000 0.000
#> GSM587164 2 0.0000 0.9890 0.000 1.000 0.000 0.000 0.000
#> GSM587165 2 0.0000 0.9890 0.000 1.000 0.000 0.000 0.000
#> GSM587166 2 0.0000 0.9890 0.000 1.000 0.000 0.000 0.000
#> GSM587167 2 0.0000 0.9890 0.000 1.000 0.000 0.000 0.000
#> GSM587168 2 0.0000 0.9890 0.000 1.000 0.000 0.000 0.000
#> GSM587169 2 0.0000 0.9890 0.000 1.000 0.000 0.000 0.000
#> GSM587170 2 0.0000 0.9890 0.000 1.000 0.000 0.000 0.000
#> GSM587171 2 0.0000 0.9890 0.000 1.000 0.000 0.000 0.000
#> GSM587172 2 0.0000 0.9890 0.000 1.000 0.000 0.000 0.000
#> GSM587173 2 0.0000 0.9890 0.000 1.000 0.000 0.000 0.000
#> GSM587174 2 0.0000 0.9890 0.000 1.000 0.000 0.000 0.000
#> GSM587175 2 0.0000 0.9890 0.000 1.000 0.000 0.000 0.000
#> GSM587176 2 0.0000 0.9890 0.000 1.000 0.000 0.000 0.000
#> GSM587177 2 0.0000 0.9890 0.000 1.000 0.000 0.000 0.000
#> GSM587178 2 0.0000 0.9890 0.000 1.000 0.000 0.000 0.000
#> GSM587179 2 0.0000 0.9890 0.000 1.000 0.000 0.000 0.000
#> GSM587180 2 0.0000 0.9890 0.000 1.000 0.000 0.000 0.000
#> GSM587181 2 0.0000 0.9890 0.000 1.000 0.000 0.000 0.000
#> GSM587182 2 0.0000 0.9890 0.000 1.000 0.000 0.000 0.000
#> GSM587183 2 0.0000 0.9890 0.000 1.000 0.000 0.000 0.000
#> GSM587184 2 0.0000 0.9890 0.000 1.000 0.000 0.000 0.000
#> GSM587185 2 0.0000 0.9890 0.000 1.000 0.000 0.000 0.000
#> GSM587186 2 0.0000 0.9890 0.000 1.000 0.000 0.000 0.000
#> GSM587187 2 0.0000 0.9890 0.000 1.000 0.000 0.000 0.000
#> GSM587188 2 0.0162 0.9855 0.000 0.996 0.004 0.000 0.000
#> GSM587189 2 0.0162 0.9855 0.000 0.996 0.004 0.000 0.000
#> GSM587190 2 0.5708 0.4120 0.000 0.616 0.112 0.004 0.268
#> GSM587203 1 0.0000 0.9172 1.000 0.000 0.000 0.000 0.000
#> GSM587204 1 0.0162 0.9151 0.996 0.000 0.000 0.000 0.004
#> GSM587205 1 0.0000 0.9172 1.000 0.000 0.000 0.000 0.000
#> GSM587206 1 0.0000 0.9172 1.000 0.000 0.000 0.000 0.000
#> GSM587207 1 0.0000 0.9172 1.000 0.000 0.000 0.000 0.000
#> GSM587208 1 0.0000 0.9172 1.000 0.000 0.000 0.000 0.000
#> GSM587209 1 0.0000 0.9172 1.000 0.000 0.000 0.000 0.000
#> GSM587210 1 0.4617 0.7194 0.744 0.000 0.000 0.108 0.148
#> GSM587211 1 0.0000 0.9172 1.000 0.000 0.000 0.000 0.000
#> GSM587212 1 0.4361 0.7437 0.768 0.000 0.000 0.108 0.124
#> GSM587213 1 0.0000 0.9172 1.000 0.000 0.000 0.000 0.000
#> GSM587214 1 0.0000 0.9172 1.000 0.000 0.000 0.000 0.000
#> GSM587215 1 0.0000 0.9172 1.000 0.000 0.000 0.000 0.000
#> GSM587216 1 0.0000 0.9172 1.000 0.000 0.000 0.000 0.000
#> GSM587217 1 0.0000 0.9172 1.000 0.000 0.000 0.000 0.000
#> GSM587191 5 0.1608 0.9001 0.000 0.000 0.072 0.000 0.928
#> GSM587192 5 0.0963 0.9132 0.000 0.000 0.036 0.000 0.964
#> GSM587193 5 0.3011 0.8616 0.016 0.000 0.000 0.140 0.844
#> GSM587194 5 0.1270 0.9119 0.000 0.000 0.000 0.052 0.948
#> GSM587195 3 0.0162 0.9355 0.000 0.000 0.996 0.000 0.004
#> GSM587196 3 0.0162 0.9355 0.000 0.000 0.996 0.000 0.004
#> GSM587197 3 0.0162 0.9355 0.000 0.000 0.996 0.000 0.004
#> GSM587198 3 0.1892 0.9193 0.000 0.000 0.916 0.004 0.080
#> GSM587199 3 0.3231 0.8262 0.000 0.000 0.800 0.004 0.196
#> GSM587200 1 0.7197 0.0885 0.412 0.000 0.292 0.020 0.276
#> GSM587201 1 0.6202 0.3997 0.564 0.000 0.260 0.004 0.172
#> GSM587202 3 0.1892 0.9193 0.000 0.000 0.916 0.004 0.080
#> GSM198767 1 0.0000 0.9172 1.000 0.000 0.000 0.000 0.000
#> GSM198769 1 0.0000 0.9172 1.000 0.000 0.000 0.000 0.000
#> GSM198772 1 0.0000 0.9172 1.000 0.000 0.000 0.000 0.000
#> GSM198773 1 0.0000 0.9172 1.000 0.000 0.000 0.000 0.000
#> GSM198776 1 0.0162 0.9151 0.996 0.000 0.000 0.000 0.004
#> GSM198778 1 0.4617 0.7194 0.744 0.000 0.000 0.108 0.148
#> GSM198780 1 0.4361 0.7437 0.768 0.000 0.000 0.108 0.124
#> GSM198781 1 0.0000 0.9172 1.000 0.000 0.000 0.000 0.000
#> GSM198765 5 0.1608 0.9001 0.000 0.000 0.072 0.000 0.928
#> GSM198766 5 0.3011 0.8616 0.016 0.000 0.000 0.140 0.844
#> GSM198768 3 0.0162 0.9355 0.000 0.000 0.996 0.000 0.004
#> GSM198770 3 0.0162 0.9355 0.000 0.000 0.996 0.000 0.004
#> GSM198771 3 0.1892 0.9193 0.000 0.000 0.916 0.004 0.080
#> GSM198774 5 0.0963 0.9132 0.000 0.000 0.036 0.000 0.964
#> GSM198775 5 0.1270 0.9119 0.000 0.000 0.000 0.052 0.948
#> GSM198777 3 0.0162 0.9355 0.000 0.000 0.996 0.000 0.004
#> GSM198779 3 0.3231 0.8262 0.000 0.000 0.800 0.004 0.196
#> GSM587218 4 0.0162 1.0000 0.004 0.000 0.000 0.996 0.000
#> GSM587219 4 0.0162 1.0000 0.004 0.000 0.000 0.996 0.000
#> GSM587220 4 0.0162 1.0000 0.004 0.000 0.000 0.996 0.000
#> GSM587221 4 0.0162 1.0000 0.004 0.000 0.000 0.996 0.000
#> GSM587222 4 0.0162 1.0000 0.004 0.000 0.000 0.996 0.000
#> GSM587223 4 0.0162 1.0000 0.004 0.000 0.000 0.996 0.000
#> GSM587224 4 0.0162 1.0000 0.004 0.000 0.000 0.996 0.000
#> GSM587225 4 0.0162 1.0000 0.004 0.000 0.000 0.996 0.000
#> GSM587226 4 0.0162 1.0000 0.004 0.000 0.000 0.996 0.000
#> GSM587227 4 0.0162 1.0000 0.004 0.000 0.000 0.996 0.000
#> GSM587228 4 0.0162 1.0000 0.004 0.000 0.000 0.996 0.000
#> GSM587229 4 0.0162 1.0000 0.004 0.000 0.000 0.996 0.000
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM587155 2 0.1349 0.9293 0.000 0.940 0.000 0.000 0.004 0.056
#> GSM587156 2 0.3252 0.8283 0.000 0.824 0.000 0.000 0.108 0.068
#> GSM587157 2 0.1285 0.9315 0.000 0.944 0.000 0.000 0.004 0.052
#> GSM587158 2 0.0000 0.9523 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587159 2 0.0000 0.9523 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587160 2 0.0146 0.9521 0.000 0.996 0.000 0.000 0.000 0.004
#> GSM587161 2 0.0363 0.9503 0.000 0.988 0.000 0.000 0.000 0.012
#> GSM587162 2 0.0146 0.9521 0.000 0.996 0.000 0.000 0.000 0.004
#> GSM587163 2 0.0146 0.9521 0.000 0.996 0.000 0.000 0.000 0.004
#> GSM587164 2 0.1471 0.9248 0.000 0.932 0.000 0.000 0.004 0.064
#> GSM587165 2 0.0713 0.9470 0.000 0.972 0.000 0.000 0.000 0.028
#> GSM587166 2 0.3252 0.8283 0.000 0.824 0.000 0.000 0.108 0.068
#> GSM587167 2 0.1531 0.9223 0.000 0.928 0.000 0.000 0.004 0.068
#> GSM587168 2 0.0547 0.9495 0.000 0.980 0.000 0.000 0.000 0.020
#> GSM587169 2 0.0146 0.9521 0.000 0.996 0.000 0.000 0.000 0.004
#> GSM587170 2 0.1471 0.9248 0.000 0.932 0.000 0.000 0.004 0.064
#> GSM587171 2 0.0000 0.9523 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587172 2 0.0000 0.9523 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587173 2 0.0713 0.9470 0.000 0.972 0.000 0.000 0.000 0.028
#> GSM587174 2 0.0260 0.9518 0.000 0.992 0.000 0.000 0.000 0.008
#> GSM587175 2 0.1082 0.9375 0.000 0.956 0.000 0.000 0.004 0.040
#> GSM587176 2 0.0146 0.9521 0.000 0.996 0.000 0.000 0.000 0.004
#> GSM587177 2 0.0713 0.9470 0.000 0.972 0.000 0.000 0.000 0.028
#> GSM587178 2 0.0363 0.9512 0.000 0.988 0.000 0.000 0.000 0.012
#> GSM587179 2 0.0146 0.9521 0.000 0.996 0.000 0.000 0.000 0.004
#> GSM587180 2 0.0547 0.9495 0.000 0.980 0.000 0.000 0.000 0.020
#> GSM587181 2 0.0260 0.9518 0.000 0.992 0.000 0.000 0.000 0.008
#> GSM587182 2 0.0458 0.9504 0.000 0.984 0.000 0.000 0.000 0.016
#> GSM587183 2 0.0713 0.9470 0.000 0.972 0.000 0.000 0.000 0.028
#> GSM587184 2 0.0000 0.9523 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587185 2 0.0146 0.9521 0.000 0.996 0.000 0.000 0.000 0.004
#> GSM587186 2 0.0713 0.9470 0.000 0.972 0.000 0.000 0.000 0.028
#> GSM587187 2 0.1204 0.9325 0.000 0.944 0.000 0.000 0.000 0.056
#> GSM587188 2 0.2361 0.8853 0.000 0.884 0.028 0.000 0.000 0.088
#> GSM587189 2 0.2361 0.8853 0.000 0.884 0.028 0.000 0.000 0.088
#> GSM587190 2 0.7097 0.0888 0.000 0.440 0.108 0.000 0.204 0.248
#> GSM587203 1 0.0363 0.8821 0.988 0.000 0.000 0.000 0.000 0.012
#> GSM587204 1 0.0458 0.8817 0.984 0.000 0.000 0.000 0.000 0.016
#> GSM587205 1 0.0363 0.8821 0.988 0.000 0.000 0.000 0.000 0.012
#> GSM587206 1 0.0363 0.8821 0.988 0.000 0.000 0.000 0.000 0.012
#> GSM587207 1 0.0363 0.8821 0.988 0.000 0.000 0.000 0.000 0.012
#> GSM587208 1 0.0363 0.8821 0.988 0.000 0.000 0.000 0.000 0.012
#> GSM587209 1 0.0713 0.8752 0.972 0.000 0.000 0.000 0.000 0.028
#> GSM587210 1 0.6134 0.2562 0.588 0.000 0.000 0.068 0.156 0.188
#> GSM587211 1 0.1584 0.8470 0.928 0.000 0.000 0.000 0.008 0.064
#> GSM587212 1 0.5839 0.3649 0.624 0.000 0.000 0.060 0.156 0.160
#> GSM587213 1 0.0291 0.8827 0.992 0.000 0.000 0.000 0.004 0.004
#> GSM587214 1 0.0291 0.8827 0.992 0.000 0.000 0.000 0.004 0.004
#> GSM587215 1 0.0405 0.8822 0.988 0.000 0.000 0.000 0.004 0.008
#> GSM587216 1 0.1462 0.8569 0.936 0.000 0.000 0.000 0.008 0.056
#> GSM587217 1 0.0508 0.8812 0.984 0.000 0.000 0.000 0.004 0.012
#> GSM587191 5 0.4636 0.6573 0.000 0.000 0.040 0.000 0.516 0.444
#> GSM587192 5 0.4377 0.6672 0.000 0.000 0.024 0.000 0.540 0.436
#> GSM587193 5 0.2288 0.6172 0.016 0.000 0.000 0.068 0.900 0.016
#> GSM587194 5 0.2301 0.5780 0.000 0.000 0.000 0.020 0.884 0.096
#> GSM587195 3 0.0000 0.7619 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM587196 3 0.0000 0.7619 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM587197 3 0.1219 0.7421 0.000 0.000 0.948 0.000 0.004 0.048
#> GSM587198 3 0.3592 0.6498 0.000 0.000 0.656 0.000 0.000 0.344
#> GSM587199 3 0.4802 0.5215 0.000 0.000 0.540 0.000 0.056 0.404
#> GSM587200 6 0.7127 0.8640 0.196 0.000 0.112 0.012 0.184 0.496
#> GSM587201 6 0.6961 0.8707 0.264 0.000 0.092 0.004 0.168 0.472
#> GSM587202 3 0.3592 0.6498 0.000 0.000 0.656 0.000 0.000 0.344
#> GSM198767 1 0.0363 0.8821 0.988 0.000 0.000 0.000 0.000 0.012
#> GSM198769 1 0.0713 0.8752 0.972 0.000 0.000 0.000 0.000 0.028
#> GSM198772 1 0.1584 0.8470 0.928 0.000 0.000 0.000 0.008 0.064
#> GSM198773 1 0.0291 0.8827 0.992 0.000 0.000 0.000 0.004 0.004
#> GSM198776 1 0.0458 0.8817 0.984 0.000 0.000 0.000 0.000 0.016
#> GSM198778 1 0.6134 0.2562 0.588 0.000 0.000 0.068 0.156 0.188
#> GSM198780 1 0.5839 0.3649 0.624 0.000 0.000 0.060 0.156 0.160
#> GSM198781 1 0.0291 0.8827 0.992 0.000 0.000 0.000 0.004 0.004
#> GSM198765 5 0.4636 0.6573 0.000 0.000 0.040 0.000 0.516 0.444
#> GSM198766 5 0.2288 0.6172 0.016 0.000 0.000 0.068 0.900 0.016
#> GSM198768 3 0.0000 0.7619 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM198770 3 0.1219 0.7421 0.000 0.000 0.948 0.000 0.004 0.048
#> GSM198771 3 0.3592 0.6498 0.000 0.000 0.656 0.000 0.000 0.344
#> GSM198774 5 0.4377 0.6672 0.000 0.000 0.024 0.000 0.540 0.436
#> GSM198775 5 0.2301 0.5780 0.000 0.000 0.000 0.020 0.884 0.096
#> GSM198777 3 0.0000 0.7619 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM198779 3 0.4802 0.5215 0.000 0.000 0.540 0.000 0.056 0.404
#> GSM587218 4 0.0000 1.0000 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM587219 4 0.0000 1.0000 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM587220 4 0.0000 1.0000 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM587221 4 0.0000 1.0000 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM587222 4 0.0000 1.0000 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM587223 4 0.0000 1.0000 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM587224 4 0.0000 1.0000 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM587225 4 0.0000 1.0000 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM587226 4 0.0000 1.0000 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM587227 4 0.0000 1.0000 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM587228 4 0.0000 1.0000 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM587229 4 0.0000 1.0000 0.000 0.000 0.000 1.000 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 specimen(p) k
#> MAD:skmeans 92 1.41e-13 2
#> MAD:skmeans 91 1.42e-27 3
#> MAD:skmeans 91 2.41e-37 4
#> MAD:skmeans 89 6.34e-43 5
#> MAD:skmeans 87 4.84e-40 6
If matrix rows can be associated to genes, consider to use functional_enrichment(res,
...) to perform function enrichment for the signature genes. See this vignette for more detailed explanations.
The object with results only for a single top-value method and a single partition method can be extracted as:
res = res_list["MAD", "pam"]
# you can also extract it by
# res = res_list["MAD:pam"]
A summary of res and all the functions that can be applied to it:
res
#> A 'ConsensusPartition' object with k = 2, 3, 4, 5, 6.
#> On a matrix with 51941 rows and 92 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 6.
#>
#> 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 0.853 0.912 0.962 0.5015 0.498 0.498
#> 3 3 0.937 0.954 0.979 0.3219 0.791 0.601
#> 4 4 0.977 0.941 0.974 0.1015 0.899 0.716
#> 5 5 0.990 0.937 0.970 0.0399 0.953 0.830
#> 6 6 0.995 0.930 0.973 0.0323 0.970 0.879
suggest_best_k() suggests the best \(k\) based on these statistics. The rules are as follows:
suggest_best_k(res)
#> [1] 6
#> attr(,"optional")
#> [1] 3 4 5
There is also optional best \(k\) = 3 4 5 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
#> GSM587155 2 0.000 0.930 0.000 1.000
#> GSM587156 2 0.000 0.930 0.000 1.000
#> GSM587157 2 0.000 0.930 0.000 1.000
#> GSM587158 2 0.000 0.930 0.000 1.000
#> GSM587159 2 0.000 0.930 0.000 1.000
#> GSM587160 2 0.000 0.930 0.000 1.000
#> GSM587161 2 0.000 0.930 0.000 1.000
#> GSM587162 2 0.000 0.930 0.000 1.000
#> GSM587163 2 0.000 0.930 0.000 1.000
#> GSM587164 2 0.000 0.930 0.000 1.000
#> GSM587165 2 0.000 0.930 0.000 1.000
#> GSM587166 2 0.000 0.930 0.000 1.000
#> GSM587167 2 0.000 0.930 0.000 1.000
#> GSM587168 2 0.000 0.930 0.000 1.000
#> GSM587169 2 0.000 0.930 0.000 1.000
#> GSM587170 2 0.000 0.930 0.000 1.000
#> GSM587171 2 0.000 0.930 0.000 1.000
#> GSM587172 2 0.000 0.930 0.000 1.000
#> GSM587173 2 0.000 0.930 0.000 1.000
#> GSM587174 2 0.000 0.930 0.000 1.000
#> GSM587175 2 0.000 0.930 0.000 1.000
#> GSM587176 2 0.000 0.930 0.000 1.000
#> GSM587177 2 0.000 0.930 0.000 1.000
#> GSM587178 2 0.000 0.930 0.000 1.000
#> GSM587179 2 0.000 0.930 0.000 1.000
#> GSM587180 2 0.000 0.930 0.000 1.000
#> GSM587181 2 0.000 0.930 0.000 1.000
#> GSM587182 2 0.000 0.930 0.000 1.000
#> GSM587183 2 0.000 0.930 0.000 1.000
#> GSM587184 2 0.000 0.930 0.000 1.000
#> GSM587185 2 0.000 0.930 0.000 1.000
#> GSM587186 2 0.000 0.930 0.000 1.000
#> GSM587187 2 0.000 0.930 0.000 1.000
#> GSM587188 2 0.000 0.930 0.000 1.000
#> GSM587189 2 0.000 0.930 0.000 1.000
#> GSM587190 2 0.000 0.930 0.000 1.000
#> GSM587203 1 0.000 0.993 1.000 0.000
#> GSM587204 1 0.000 0.993 1.000 0.000
#> GSM587205 1 0.000 0.993 1.000 0.000
#> GSM587206 1 0.000 0.993 1.000 0.000
#> GSM587207 1 0.000 0.993 1.000 0.000
#> GSM587208 1 0.000 0.993 1.000 0.000
#> GSM587209 1 0.000 0.993 1.000 0.000
#> GSM587210 1 0.000 0.993 1.000 0.000
#> GSM587211 1 0.000 0.993 1.000 0.000
#> GSM587212 1 0.000 0.993 1.000 0.000
#> GSM587213 1 0.000 0.993 1.000 0.000
#> GSM587214 1 0.000 0.993 1.000 0.000
#> GSM587215 1 0.000 0.993 1.000 0.000
#> GSM587216 1 0.000 0.993 1.000 0.000
#> GSM587217 1 0.000 0.993 1.000 0.000
#> GSM587191 2 0.000 0.930 0.000 1.000
#> GSM587192 1 0.000 0.993 1.000 0.000
#> GSM587193 1 0.000 0.993 1.000 0.000
#> GSM587194 2 0.430 0.869 0.088 0.912
#> GSM587195 2 0.978 0.399 0.412 0.588
#> GSM587196 2 0.978 0.399 0.412 0.588
#> GSM587197 2 0.969 0.433 0.396 0.604
#> GSM587198 2 0.980 0.389 0.416 0.584
#> GSM587199 2 0.184 0.913 0.028 0.972
#> GSM587200 1 0.000 0.993 1.000 0.000
#> GSM587201 1 0.000 0.993 1.000 0.000
#> GSM587202 2 0.978 0.399 0.412 0.588
#> GSM198767 1 0.000 0.993 1.000 0.000
#> GSM198769 1 0.000 0.993 1.000 0.000
#> GSM198772 1 0.000 0.993 1.000 0.000
#> GSM198773 1 0.000 0.993 1.000 0.000
#> GSM198776 1 0.000 0.993 1.000 0.000
#> GSM198778 1 0.000 0.993 1.000 0.000
#> GSM198780 1 0.000 0.993 1.000 0.000
#> GSM198781 1 0.000 0.993 1.000 0.000
#> GSM198765 2 0.388 0.878 0.076 0.924
#> GSM198766 1 0.000 0.993 1.000 0.000
#> GSM198768 2 0.980 0.389 0.416 0.584
#> GSM198770 2 0.506 0.846 0.112 0.888
#> GSM198771 1 0.795 0.638 0.760 0.240
#> GSM198774 1 0.000 0.993 1.000 0.000
#> GSM198775 2 0.443 0.866 0.092 0.908
#> GSM198777 2 0.978 0.399 0.412 0.588
#> GSM198779 2 0.184 0.913 0.028 0.972
#> GSM587218 1 0.000 0.993 1.000 0.000
#> GSM587219 1 0.000 0.993 1.000 0.000
#> GSM587220 1 0.000 0.993 1.000 0.000
#> GSM587221 1 0.000 0.993 1.000 0.000
#> GSM587222 1 0.000 0.993 1.000 0.000
#> GSM587223 1 0.000 0.993 1.000 0.000
#> GSM587224 1 0.000 0.993 1.000 0.000
#> GSM587225 1 0.000 0.993 1.000 0.000
#> GSM587226 1 0.000 0.993 1.000 0.000
#> GSM587227 1 0.000 0.993 1.000 0.000
#> GSM587228 1 0.000 0.993 1.000 0.000
#> GSM587229 1 0.000 0.993 1.000 0.000
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM587155 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587156 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587157 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587158 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587159 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587160 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587161 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587162 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587163 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587164 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587165 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587166 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587167 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587168 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587169 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587170 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587171 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587172 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587173 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587174 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587175 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587176 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587177 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587178 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587179 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587180 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587181 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587182 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587183 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587184 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587185 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587186 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587187 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587188 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587189 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587190 3 0.0237 0.981 0.000 0.004 0.996
#> GSM587203 1 0.0000 0.951 1.000 0.000 0.000
#> GSM587204 1 0.0000 0.951 1.000 0.000 0.000
#> GSM587205 1 0.0000 0.951 1.000 0.000 0.000
#> GSM587206 1 0.0000 0.951 1.000 0.000 0.000
#> GSM587207 1 0.0000 0.951 1.000 0.000 0.000
#> GSM587208 1 0.0000 0.951 1.000 0.000 0.000
#> GSM587209 1 0.0000 0.951 1.000 0.000 0.000
#> GSM587210 1 0.1289 0.929 0.968 0.000 0.032
#> GSM587211 1 0.4178 0.815 0.828 0.000 0.172
#> GSM587212 1 0.0000 0.951 1.000 0.000 0.000
#> GSM587213 1 0.0000 0.951 1.000 0.000 0.000
#> GSM587214 1 0.0000 0.951 1.000 0.000 0.000
#> GSM587215 1 0.0000 0.951 1.000 0.000 0.000
#> GSM587216 1 0.0000 0.951 1.000 0.000 0.000
#> GSM587217 1 0.0000 0.951 1.000 0.000 0.000
#> GSM587191 3 0.0000 0.985 0.000 0.000 1.000
#> GSM587192 3 0.0000 0.985 0.000 0.000 1.000
#> GSM587193 1 0.6305 0.140 0.516 0.000 0.484
#> GSM587194 3 0.0000 0.985 0.000 0.000 1.000
#> GSM587195 3 0.0000 0.985 0.000 0.000 1.000
#> GSM587196 3 0.0000 0.985 0.000 0.000 1.000
#> GSM587197 3 0.0000 0.985 0.000 0.000 1.000
#> GSM587198 3 0.0000 0.985 0.000 0.000 1.000
#> GSM587199 3 0.0000 0.985 0.000 0.000 1.000
#> GSM587200 3 0.0000 0.985 0.000 0.000 1.000
#> GSM587201 3 0.0000 0.985 0.000 0.000 1.000
#> GSM587202 3 0.0000 0.985 0.000 0.000 1.000
#> GSM198767 1 0.0000 0.951 1.000 0.000 0.000
#> GSM198769 1 0.0000 0.951 1.000 0.000 0.000
#> GSM198772 1 0.2165 0.909 0.936 0.000 0.064
#> GSM198773 1 0.0000 0.951 1.000 0.000 0.000
#> GSM198776 1 0.0000 0.951 1.000 0.000 0.000
#> GSM198778 1 0.4235 0.808 0.824 0.000 0.176
#> GSM198780 1 0.0000 0.951 1.000 0.000 0.000
#> GSM198781 1 0.0000 0.951 1.000 0.000 0.000
#> GSM198765 3 0.0000 0.985 0.000 0.000 1.000
#> GSM198766 1 0.4235 0.810 0.824 0.000 0.176
#> GSM198768 3 0.0000 0.985 0.000 0.000 1.000
#> GSM198770 3 0.0000 0.985 0.000 0.000 1.000
#> GSM198771 3 0.0000 0.985 0.000 0.000 1.000
#> GSM198774 3 0.0000 0.985 0.000 0.000 1.000
#> GSM198775 3 0.0000 0.985 0.000 0.000 1.000
#> GSM198777 3 0.0000 0.985 0.000 0.000 1.000
#> GSM198779 3 0.0000 0.985 0.000 0.000 1.000
#> GSM587218 3 0.0000 0.985 0.000 0.000 1.000
#> GSM587219 1 0.4121 0.819 0.832 0.000 0.168
#> GSM587220 1 0.0000 0.951 1.000 0.000 0.000
#> GSM587221 1 0.4235 0.810 0.824 0.000 0.176
#> GSM587222 1 0.0237 0.948 0.996 0.000 0.004
#> GSM587223 1 0.0000 0.951 1.000 0.000 0.000
#> GSM587224 3 0.5497 0.541 0.292 0.000 0.708
#> GSM587225 1 0.0000 0.951 1.000 0.000 0.000
#> GSM587226 1 0.4235 0.810 0.824 0.000 0.176
#> GSM587227 1 0.0000 0.951 1.000 0.000 0.000
#> GSM587228 1 0.0000 0.951 1.000 0.000 0.000
#> GSM587229 1 0.0000 0.951 1.000 0.000 0.000
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM587155 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587156 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587157 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587158 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587159 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587160 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587161 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587162 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587163 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587164 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587165 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587166 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587167 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587168 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587169 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587170 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587171 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587172 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587173 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587174 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587175 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587176 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587177 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587178 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587179 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587180 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587181 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587182 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587183 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587184 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587185 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587186 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587187 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587188 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587189 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587190 3 0.0188 0.970 0.000 0.004 0.996 0.000
#> GSM587203 1 0.0000 0.948 1.000 0.000 0.000 0.000
#> GSM587204 1 0.0000 0.948 1.000 0.000 0.000 0.000
#> GSM587205 1 0.0000 0.948 1.000 0.000 0.000 0.000
#> GSM587206 1 0.0000 0.948 1.000 0.000 0.000 0.000
#> GSM587207 1 0.0000 0.948 1.000 0.000 0.000 0.000
#> GSM587208 1 0.0000 0.948 1.000 0.000 0.000 0.000
#> GSM587209 1 0.0000 0.948 1.000 0.000 0.000 0.000
#> GSM587210 4 0.6206 0.232 0.404 0.000 0.056 0.540
#> GSM587211 1 0.4163 0.736 0.792 0.000 0.188 0.020
#> GSM587212 1 0.4103 0.663 0.744 0.000 0.000 0.256
#> GSM587213 1 0.0000 0.948 1.000 0.000 0.000 0.000
#> GSM587214 1 0.0000 0.948 1.000 0.000 0.000 0.000
#> GSM587215 1 0.0000 0.948 1.000 0.000 0.000 0.000
#> GSM587216 1 0.2011 0.889 0.920 0.000 0.000 0.080
#> GSM587217 1 0.1389 0.916 0.952 0.000 0.000 0.048
#> GSM587191 3 0.0000 0.974 0.000 0.000 1.000 0.000
#> GSM587192 3 0.1557 0.924 0.000 0.000 0.944 0.056
#> GSM587193 3 0.3978 0.758 0.012 0.000 0.796 0.192
#> GSM587194 3 0.0000 0.974 0.000 0.000 1.000 0.000
#> GSM587195 3 0.0000 0.974 0.000 0.000 1.000 0.000
#> GSM587196 3 0.0000 0.974 0.000 0.000 1.000 0.000
#> GSM587197 3 0.0000 0.974 0.000 0.000 1.000 0.000
#> GSM587198 3 0.0000 0.974 0.000 0.000 1.000 0.000
#> GSM587199 3 0.0000 0.974 0.000 0.000 1.000 0.000
#> GSM587200 3 0.0000 0.974 0.000 0.000 1.000 0.000
#> GSM587201 3 0.0000 0.974 0.000 0.000 1.000 0.000
#> GSM587202 3 0.0000 0.974 0.000 0.000 1.000 0.000
#> GSM198767 1 0.0000 0.948 1.000 0.000 0.000 0.000
#> GSM198769 1 0.0000 0.948 1.000 0.000 0.000 0.000
#> GSM198772 1 0.2742 0.872 0.900 0.000 0.076 0.024
#> GSM198773 1 0.0000 0.948 1.000 0.000 0.000 0.000
#> GSM198776 1 0.0000 0.948 1.000 0.000 0.000 0.000
#> GSM198778 4 0.7020 0.328 0.332 0.000 0.136 0.532
#> GSM198780 1 0.4103 0.663 0.744 0.000 0.000 0.256
#> GSM198781 1 0.0000 0.948 1.000 0.000 0.000 0.000
#> GSM198765 3 0.0000 0.974 0.000 0.000 1.000 0.000
#> GSM198766 3 0.5109 0.698 0.060 0.000 0.744 0.196
#> GSM198768 3 0.0000 0.974 0.000 0.000 1.000 0.000
#> GSM198770 3 0.0000 0.974 0.000 0.000 1.000 0.000
#> GSM198771 3 0.0000 0.974 0.000 0.000 1.000 0.000
#> GSM198774 3 0.0707 0.958 0.000 0.000 0.980 0.020
#> GSM198775 3 0.0000 0.974 0.000 0.000 1.000 0.000
#> GSM198777 3 0.0000 0.974 0.000 0.000 1.000 0.000
#> GSM198779 3 0.0000 0.974 0.000 0.000 1.000 0.000
#> GSM587218 4 0.0000 0.930 0.000 0.000 0.000 1.000
#> GSM587219 4 0.0000 0.930 0.000 0.000 0.000 1.000
#> GSM587220 4 0.0000 0.930 0.000 0.000 0.000 1.000
#> GSM587221 4 0.0000 0.930 0.000 0.000 0.000 1.000
#> GSM587222 4 0.0000 0.930 0.000 0.000 0.000 1.000
#> GSM587223 4 0.0000 0.930 0.000 0.000 0.000 1.000
#> GSM587224 4 0.0000 0.930 0.000 0.000 0.000 1.000
#> GSM587225 4 0.0000 0.930 0.000 0.000 0.000 1.000
#> GSM587226 4 0.0000 0.930 0.000 0.000 0.000 1.000
#> GSM587227 4 0.0000 0.930 0.000 0.000 0.000 1.000
#> GSM587228 4 0.0000 0.930 0.000 0.000 0.000 1.000
#> GSM587229 4 0.0000 0.930 0.000 0.000 0.000 1.000
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM587155 2 0.0000 0.9999 0.000 1.000 0.000 0.000 0.000
#> GSM587156 2 0.0000 0.9999 0.000 1.000 0.000 0.000 0.000
#> GSM587157 2 0.0000 0.9999 0.000 1.000 0.000 0.000 0.000
#> GSM587158 2 0.0000 0.9999 0.000 1.000 0.000 0.000 0.000
#> GSM587159 2 0.0000 0.9999 0.000 1.000 0.000 0.000 0.000
#> GSM587160 2 0.0000 0.9999 0.000 1.000 0.000 0.000 0.000
#> GSM587161 2 0.0000 0.9999 0.000 1.000 0.000 0.000 0.000
#> GSM587162 2 0.0000 0.9999 0.000 1.000 0.000 0.000 0.000
#> GSM587163 2 0.0000 0.9999 0.000 1.000 0.000 0.000 0.000
#> GSM587164 2 0.0000 0.9999 0.000 1.000 0.000 0.000 0.000
#> GSM587165 2 0.0000 0.9999 0.000 1.000 0.000 0.000 0.000
#> GSM587166 2 0.0000 0.9999 0.000 1.000 0.000 0.000 0.000
#> GSM587167 2 0.0000 0.9999 0.000 1.000 0.000 0.000 0.000
#> GSM587168 2 0.0000 0.9999 0.000 1.000 0.000 0.000 0.000
#> GSM587169 2 0.0000 0.9999 0.000 1.000 0.000 0.000 0.000
#> GSM587170 2 0.0000 0.9999 0.000 1.000 0.000 0.000 0.000
#> GSM587171 2 0.0000 0.9999 0.000 1.000 0.000 0.000 0.000
#> GSM587172 2 0.0000 0.9999 0.000 1.000 0.000 0.000 0.000
#> GSM587173 2 0.0000 0.9999 0.000 1.000 0.000 0.000 0.000
#> GSM587174 2 0.0000 0.9999 0.000 1.000 0.000 0.000 0.000
#> GSM587175 2 0.0000 0.9999 0.000 1.000 0.000 0.000 0.000
#> GSM587176 2 0.0000 0.9999 0.000 1.000 0.000 0.000 0.000
#> GSM587177 2 0.0000 0.9999 0.000 1.000 0.000 0.000 0.000
#> GSM587178 2 0.0000 0.9999 0.000 1.000 0.000 0.000 0.000
#> GSM587179 2 0.0000 0.9999 0.000 1.000 0.000 0.000 0.000
#> GSM587180 2 0.0000 0.9999 0.000 1.000 0.000 0.000 0.000
#> GSM587181 2 0.0000 0.9999 0.000 1.000 0.000 0.000 0.000
#> GSM587182 2 0.0000 0.9999 0.000 1.000 0.000 0.000 0.000
#> GSM587183 2 0.0000 0.9999 0.000 1.000 0.000 0.000 0.000
#> GSM587184 2 0.0000 0.9999 0.000 1.000 0.000 0.000 0.000
#> GSM587185 2 0.0000 0.9999 0.000 1.000 0.000 0.000 0.000
#> GSM587186 2 0.0000 0.9999 0.000 1.000 0.000 0.000 0.000
#> GSM587187 2 0.0000 0.9999 0.000 1.000 0.000 0.000 0.000
#> GSM587188 2 0.0000 0.9999 0.000 1.000 0.000 0.000 0.000
#> GSM587189 2 0.0162 0.9957 0.000 0.996 0.004 0.000 0.000
#> GSM587190 3 0.0162 0.9888 0.000 0.004 0.996 0.000 0.000
#> GSM587203 5 0.0000 1.0000 0.000 0.000 0.000 0.000 1.000
#> GSM587204 5 0.0000 1.0000 0.000 0.000 0.000 0.000 1.000
#> GSM587205 5 0.0000 1.0000 0.000 0.000 0.000 0.000 1.000
#> GSM587206 5 0.0000 1.0000 0.000 0.000 0.000 0.000 1.000
#> GSM587207 5 0.0000 1.0000 0.000 0.000 0.000 0.000 1.000
#> GSM587208 5 0.0000 1.0000 0.000 0.000 0.000 0.000 1.000
#> GSM587209 1 0.0703 0.8703 0.976 0.000 0.000 0.000 0.024
#> GSM587210 4 0.6372 -0.0640 0.408 0.000 0.000 0.428 0.164
#> GSM587211 1 0.0798 0.8682 0.976 0.000 0.008 0.000 0.016
#> GSM587212 1 0.3039 0.7122 0.808 0.000 0.000 0.000 0.192
#> GSM587213 1 0.0703 0.8703 0.976 0.000 0.000 0.000 0.024
#> GSM587214 1 0.1544 0.8485 0.932 0.000 0.000 0.000 0.068
#> GSM587215 1 0.0703 0.8703 0.976 0.000 0.000 0.000 0.024
#> GSM587216 1 0.0000 0.8593 1.000 0.000 0.000 0.000 0.000
#> GSM587217 1 0.0703 0.8703 0.976 0.000 0.000 0.000 0.024
#> GSM587191 3 0.0703 0.9827 0.024 0.000 0.976 0.000 0.000
#> GSM587192 3 0.0703 0.9827 0.024 0.000 0.976 0.000 0.000
#> GSM587193 1 0.6357 0.3609 0.512 0.000 0.288 0.200 0.000
#> GSM587194 3 0.0703 0.9827 0.024 0.000 0.976 0.000 0.000
#> GSM587195 3 0.0000 0.9923 0.000 0.000 1.000 0.000 0.000
#> GSM587196 3 0.0000 0.9923 0.000 0.000 1.000 0.000 0.000
#> GSM587197 3 0.0000 0.9923 0.000 0.000 1.000 0.000 0.000
#> GSM587198 3 0.0000 0.9923 0.000 0.000 1.000 0.000 0.000
#> GSM587199 3 0.0000 0.9923 0.000 0.000 1.000 0.000 0.000
#> GSM587200 3 0.0000 0.9923 0.000 0.000 1.000 0.000 0.000
#> GSM587201 3 0.0000 0.9923 0.000 0.000 1.000 0.000 0.000
#> GSM587202 3 0.0000 0.9923 0.000 0.000 1.000 0.000 0.000
#> GSM198767 5 0.0000 1.0000 0.000 0.000 0.000 0.000 1.000
#> GSM198769 1 0.0703 0.8703 0.976 0.000 0.000 0.000 0.024
#> GSM198772 1 0.0798 0.8682 0.976 0.000 0.008 0.000 0.016
#> GSM198773 1 0.0703 0.8703 0.976 0.000 0.000 0.000 0.024
#> GSM198776 5 0.0000 1.0000 0.000 0.000 0.000 0.000 1.000
#> GSM198778 1 0.7149 0.0574 0.440 0.000 0.084 0.388 0.088
#> GSM198780 1 0.3039 0.7122 0.808 0.000 0.000 0.000 0.192
#> GSM198781 1 0.1544 0.8485 0.932 0.000 0.000 0.000 0.068
#> GSM198765 3 0.0703 0.9827 0.024 0.000 0.976 0.000 0.000
#> GSM198766 1 0.2966 0.7062 0.816 0.000 0.000 0.184 0.000
#> GSM198768 3 0.0000 0.9923 0.000 0.000 1.000 0.000 0.000
#> GSM198770 3 0.0000 0.9923 0.000 0.000 1.000 0.000 0.000
#> GSM198771 3 0.0000 0.9923 0.000 0.000 1.000 0.000 0.000
#> GSM198774 3 0.0703 0.9827 0.024 0.000 0.976 0.000 0.000
#> GSM198775 3 0.0703 0.9827 0.024 0.000 0.976 0.000 0.000
#> GSM198777 3 0.0000 0.9923 0.000 0.000 1.000 0.000 0.000
#> GSM198779 3 0.0000 0.9923 0.000 0.000 1.000 0.000 0.000
#> GSM587218 4 0.0000 0.9512 0.000 0.000 0.000 1.000 0.000
#> GSM587219 4 0.0000 0.9512 0.000 0.000 0.000 1.000 0.000
#> GSM587220 4 0.0000 0.9512 0.000 0.000 0.000 1.000 0.000
#> GSM587221 4 0.0000 0.9512 0.000 0.000 0.000 1.000 0.000
#> GSM587222 4 0.0000 0.9512 0.000 0.000 0.000 1.000 0.000
#> GSM587223 4 0.0000 0.9512 0.000 0.000 0.000 1.000 0.000
#> GSM587224 4 0.0000 0.9512 0.000 0.000 0.000 1.000 0.000
#> GSM587225 4 0.0000 0.9512 0.000 0.000 0.000 1.000 0.000
#> GSM587226 4 0.0000 0.9512 0.000 0.000 0.000 1.000 0.000
#> GSM587227 4 0.0000 0.9512 0.000 0.000 0.000 1.000 0.000
#> GSM587228 4 0.0000 0.9512 0.000 0.000 0.000 1.000 0.000
#> GSM587229 4 0.0000 0.9512 0.000 0.000 0.000 1.000 0.000
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM587155 2 0.0000 0.99627 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587156 2 0.0000 0.99627 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587157 2 0.0458 0.98142 0.000 0.984 0.016 0.000 0.000 0.000
#> GSM587158 2 0.0000 0.99627 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587159 2 0.0000 0.99627 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587160 2 0.0000 0.99627 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587161 2 0.0000 0.99627 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587162 2 0.0000 0.99627 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587163 2 0.0000 0.99627 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587164 2 0.0000 0.99627 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587165 2 0.0000 0.99627 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587166 2 0.1444 0.92459 0.000 0.928 0.000 0.000 0.072 0.000
#> GSM587167 2 0.0000 0.99627 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587168 2 0.0000 0.99627 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587169 2 0.0000 0.99627 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587170 2 0.0000 0.99627 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587171 2 0.0000 0.99627 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587172 2 0.0000 0.99627 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587173 2 0.0000 0.99627 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587174 2 0.0000 0.99627 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587175 2 0.0000 0.99627 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587176 2 0.0000 0.99627 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587177 2 0.0000 0.99627 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587178 2 0.0000 0.99627 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587179 2 0.0000 0.99627 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587180 2 0.0000 0.99627 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587181 2 0.0000 0.99627 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587182 2 0.0000 0.99627 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587183 2 0.0000 0.99627 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587184 2 0.0000 0.99627 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587185 2 0.0000 0.99627 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587186 2 0.0000 0.99627 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587187 2 0.0000 0.99627 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587188 2 0.0000 0.99627 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587189 2 0.0865 0.96004 0.000 0.964 0.036 0.000 0.000 0.000
#> GSM587190 3 0.0000 0.99740 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM587203 6 0.0000 1.00000 0.000 0.000 0.000 0.000 0.000 1.000
#> GSM587204 6 0.0000 1.00000 0.000 0.000 0.000 0.000 0.000 1.000
#> GSM587205 6 0.0000 1.00000 0.000 0.000 0.000 0.000 0.000 1.000
#> GSM587206 6 0.0000 1.00000 0.000 0.000 0.000 0.000 0.000 1.000
#> GSM587207 6 0.0000 1.00000 0.000 0.000 0.000 0.000 0.000 1.000
#> GSM587208 6 0.0000 1.00000 0.000 0.000 0.000 0.000 0.000 1.000
#> GSM587209 1 0.0000 0.85584 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM587210 5 0.4117 0.65626 0.228 0.000 0.000 0.056 0.716 0.000
#> GSM587211 1 0.0000 0.85584 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM587212 1 0.3860 0.00741 0.528 0.000 0.000 0.000 0.472 0.000
#> GSM587213 1 0.0000 0.85584 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM587214 1 0.0458 0.84666 0.984 0.000 0.000 0.000 0.000 0.016
#> GSM587215 1 0.0000 0.85584 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM587216 1 0.0000 0.85584 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM587217 1 0.0000 0.85584 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM587191 5 0.0000 0.90966 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM587192 5 0.0000 0.90966 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM587193 1 0.6653 0.06971 0.416 0.000 0.176 0.052 0.356 0.000
#> GSM587194 5 0.0000 0.90966 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM587195 3 0.0000 0.99740 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM587196 3 0.0000 0.99740 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM587197 3 0.0000 0.99740 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM587198 3 0.0000 0.99740 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM587199 3 0.0000 0.99740 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM587200 3 0.0508 0.98497 0.004 0.000 0.984 0.000 0.012 0.000
#> GSM587201 3 0.0458 0.98155 0.016 0.000 0.984 0.000 0.000 0.000
#> GSM587202 3 0.0000 0.99740 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM198767 6 0.0000 1.00000 0.000 0.000 0.000 0.000 0.000 1.000
#> GSM198769 1 0.0000 0.85584 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM198772 1 0.0000 0.85584 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM198773 1 0.0000 0.85584 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM198776 6 0.0000 1.00000 0.000 0.000 0.000 0.000 0.000 1.000
#> GSM198778 5 0.3841 0.62957 0.256 0.000 0.000 0.028 0.716 0.000
#> GSM198780 1 0.3864 -0.02214 0.520 0.000 0.000 0.000 0.480 0.000
#> GSM198781 1 0.0458 0.84666 0.984 0.000 0.000 0.000 0.000 0.016
#> GSM198765 5 0.0000 0.90966 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM198766 1 0.3821 0.64725 0.772 0.000 0.000 0.080 0.148 0.000
#> GSM198768 3 0.0000 0.99740 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM198770 3 0.0000 0.99740 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM198771 3 0.0000 0.99740 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM198774 5 0.0000 0.90966 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM198775 5 0.0000 0.90966 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM198777 3 0.0000 0.99740 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM198779 3 0.0000 0.99740 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM587218 4 0.0000 1.00000 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM587219 4 0.0000 1.00000 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM587220 4 0.0000 1.00000 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM587221 4 0.0000 1.00000 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM587222 4 0.0000 1.00000 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM587223 4 0.0000 1.00000 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM587224 4 0.0000 1.00000 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM587225 4 0.0000 1.00000 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM587226 4 0.0000 1.00000 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM587227 4 0.0000 1.00000 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM587228 4 0.0000 1.00000 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM587229 4 0.0000 1.00000 0.000 0.000 0.000 1.000 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 specimen(p) k
#> MAD:pam 85 2.23e-13 2
#> MAD:pam 91 6.53e-28 3
#> MAD:pam 90 7.15e-46 4
#> MAD:pam 89 9.56e-41 5
#> MAD:pam 89 1.64e-35 6
If matrix rows can be associated to genes, consider to use functional_enrichment(res,
...) to perform function enrichment for the signature genes. See this vignette for more detailed explanations.
The object with results only for a single top-value method and a single partition method can be extracted as:
res = res_list["MAD", "mclust"]
# you can also extract it by
# res = res_list["MAD:mclust"]
A summary of res and all the functions that can be applied to it:
res
#> A 'ConsensusPartition' object with k = 2, 3, 4, 5, 6.
#> On a matrix with 51941 rows and 92 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 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 0.933 0.925 0.969 0.4686 0.535 0.535
#> 3 3 0.667 0.802 0.877 0.3263 0.598 0.400
#> 4 4 0.864 0.903 0.953 0.1645 0.863 0.660
#> 5 5 0.802 0.749 0.858 0.0518 0.961 0.857
#> 6 6 0.799 0.733 0.855 0.0367 0.954 0.813
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
#> GSM587155 2 0.0000 0.964 0.000 1.000
#> GSM587156 2 0.0000 0.964 0.000 1.000
#> GSM587157 2 0.0000 0.964 0.000 1.000
#> GSM587158 2 0.0000 0.964 0.000 1.000
#> GSM587159 2 0.0000 0.964 0.000 1.000
#> GSM587160 2 0.0000 0.964 0.000 1.000
#> GSM587161 2 0.0000 0.964 0.000 1.000
#> GSM587162 2 0.0000 0.964 0.000 1.000
#> GSM587163 2 0.0000 0.964 0.000 1.000
#> GSM587164 2 0.0000 0.964 0.000 1.000
#> GSM587165 2 0.0000 0.964 0.000 1.000
#> GSM587166 2 0.0000 0.964 0.000 1.000
#> GSM587167 2 0.0000 0.964 0.000 1.000
#> GSM587168 2 0.0000 0.964 0.000 1.000
#> GSM587169 2 0.0000 0.964 0.000 1.000
#> GSM587170 2 0.0000 0.964 0.000 1.000
#> GSM587171 2 0.0000 0.964 0.000 1.000
#> GSM587172 2 0.0000 0.964 0.000 1.000
#> GSM587173 2 0.0000 0.964 0.000 1.000
#> GSM587174 2 0.0000 0.964 0.000 1.000
#> GSM587175 2 0.0000 0.964 0.000 1.000
#> GSM587176 2 0.0000 0.964 0.000 1.000
#> GSM587177 2 0.0000 0.964 0.000 1.000
#> GSM587178 2 0.0000 0.964 0.000 1.000
#> GSM587179 2 0.0000 0.964 0.000 1.000
#> GSM587180 2 0.0000 0.964 0.000 1.000
#> GSM587181 2 0.0000 0.964 0.000 1.000
#> GSM587182 2 0.0000 0.964 0.000 1.000
#> GSM587183 2 0.0000 0.964 0.000 1.000
#> GSM587184 2 0.0000 0.964 0.000 1.000
#> GSM587185 2 0.0000 0.964 0.000 1.000
#> GSM587186 2 0.0000 0.964 0.000 1.000
#> GSM587187 2 0.0000 0.964 0.000 1.000
#> GSM587188 2 0.0672 0.963 0.008 0.992
#> GSM587189 2 0.0672 0.963 0.008 0.992
#> GSM587190 2 0.0938 0.962 0.012 0.988
#> GSM587203 1 0.0000 0.972 1.000 0.000
#> GSM587204 1 0.0000 0.972 1.000 0.000
#> GSM587205 1 0.0000 0.972 1.000 0.000
#> GSM587206 1 0.0000 0.972 1.000 0.000
#> GSM587207 1 0.0000 0.972 1.000 0.000
#> GSM587208 1 0.0000 0.972 1.000 0.000
#> GSM587209 1 0.0000 0.972 1.000 0.000
#> GSM587210 2 0.9815 0.305 0.420 0.580
#> GSM587211 1 0.0000 0.972 1.000 0.000
#> GSM587212 1 0.9795 0.242 0.584 0.416
#> GSM587213 1 0.0000 0.972 1.000 0.000
#> GSM587214 1 0.0000 0.972 1.000 0.000
#> GSM587215 1 0.0000 0.972 1.000 0.000
#> GSM587216 1 0.0000 0.972 1.000 0.000
#> GSM587217 1 0.0000 0.972 1.000 0.000
#> GSM587191 2 0.0938 0.962 0.012 0.988
#> GSM587192 2 0.0938 0.962 0.012 0.988
#> GSM587193 2 0.7815 0.708 0.232 0.768
#> GSM587194 2 0.0938 0.962 0.012 0.988
#> GSM587195 2 0.0938 0.962 0.012 0.988
#> GSM587196 2 0.0938 0.962 0.012 0.988
#> GSM587197 2 0.0938 0.962 0.012 0.988
#> GSM587198 2 0.0938 0.962 0.012 0.988
#> GSM587199 2 0.0938 0.962 0.012 0.988
#> GSM587200 2 0.6438 0.802 0.164 0.836
#> GSM587201 2 0.8608 0.620 0.284 0.716
#> GSM587202 2 0.0938 0.962 0.012 0.988
#> GSM198767 1 0.0000 0.972 1.000 0.000
#> GSM198769 1 0.0000 0.972 1.000 0.000
#> GSM198772 1 0.0000 0.972 1.000 0.000
#> GSM198773 1 0.0000 0.972 1.000 0.000
#> GSM198776 1 0.0000 0.972 1.000 0.000
#> GSM198778 2 0.9815 0.305 0.420 0.580
#> GSM198780 1 0.9815 0.229 0.580 0.420
#> GSM198781 1 0.0000 0.972 1.000 0.000
#> GSM198765 2 0.0938 0.962 0.012 0.988
#> GSM198766 2 0.7815 0.708 0.232 0.768
#> GSM198768 2 0.0938 0.962 0.012 0.988
#> GSM198770 2 0.0938 0.962 0.012 0.988
#> GSM198771 2 0.0938 0.962 0.012 0.988
#> GSM198774 2 0.0938 0.962 0.012 0.988
#> GSM198775 2 0.0938 0.962 0.012 0.988
#> GSM198777 2 0.0938 0.962 0.012 0.988
#> GSM198779 2 0.0938 0.962 0.012 0.988
#> GSM587218 1 0.0000 0.972 1.000 0.000
#> GSM587219 1 0.0000 0.972 1.000 0.000
#> GSM587220 1 0.0000 0.972 1.000 0.000
#> GSM587221 1 0.0000 0.972 1.000 0.000
#> GSM587222 1 0.0000 0.972 1.000 0.000
#> GSM587223 1 0.0000 0.972 1.000 0.000
#> GSM587224 1 0.0000 0.972 1.000 0.000
#> GSM587225 1 0.0000 0.972 1.000 0.000
#> GSM587226 1 0.0000 0.972 1.000 0.000
#> GSM587227 1 0.0000 0.972 1.000 0.000
#> GSM587228 1 0.0000 0.972 1.000 0.000
#> GSM587229 1 0.0000 0.972 1.000 0.000
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM587155 2 0.0000 0.9938 0.000 1.000 0.000
#> GSM587156 2 0.0592 0.9822 0.000 0.988 0.012
#> GSM587157 2 0.2056 0.9509 0.024 0.952 0.024
#> GSM587158 2 0.0000 0.9938 0.000 1.000 0.000
#> GSM587159 2 0.0000 0.9938 0.000 1.000 0.000
#> GSM587160 2 0.0000 0.9938 0.000 1.000 0.000
#> GSM587161 2 0.0000 0.9938 0.000 1.000 0.000
#> GSM587162 2 0.0000 0.9938 0.000 1.000 0.000
#> GSM587163 2 0.0000 0.9938 0.000 1.000 0.000
#> GSM587164 2 0.1031 0.9685 0.000 0.976 0.024
#> GSM587165 2 0.0000 0.9938 0.000 1.000 0.000
#> GSM587166 2 0.2165 0.9249 0.000 0.936 0.064
#> GSM587167 2 0.0747 0.9778 0.000 0.984 0.016
#> GSM587168 2 0.0000 0.9938 0.000 1.000 0.000
#> GSM587169 2 0.0000 0.9938 0.000 1.000 0.000
#> GSM587170 2 0.0000 0.9938 0.000 1.000 0.000
#> GSM587171 2 0.0000 0.9938 0.000 1.000 0.000
#> GSM587172 2 0.0000 0.9938 0.000 1.000 0.000
#> GSM587173 2 0.0000 0.9938 0.000 1.000 0.000
#> GSM587174 2 0.0000 0.9938 0.000 1.000 0.000
#> GSM587175 2 0.0000 0.9938 0.000 1.000 0.000
#> GSM587176 2 0.0000 0.9938 0.000 1.000 0.000
#> GSM587177 2 0.0000 0.9938 0.000 1.000 0.000
#> GSM587178 2 0.0000 0.9938 0.000 1.000 0.000
#> GSM587179 2 0.0000 0.9938 0.000 1.000 0.000
#> GSM587180 2 0.0000 0.9938 0.000 1.000 0.000
#> GSM587181 2 0.0000 0.9938 0.000 1.000 0.000
#> GSM587182 2 0.0000 0.9938 0.000 1.000 0.000
#> GSM587183 2 0.0000 0.9938 0.000 1.000 0.000
#> GSM587184 2 0.0000 0.9938 0.000 1.000 0.000
#> GSM587185 2 0.0000 0.9938 0.000 1.000 0.000
#> GSM587186 2 0.0000 0.9938 0.000 1.000 0.000
#> GSM587187 3 0.7841 0.0872 0.052 0.468 0.480
#> GSM587188 3 0.6850 0.5342 0.072 0.208 0.720
#> GSM587189 3 0.6673 0.5299 0.056 0.224 0.720
#> GSM587190 3 0.6168 0.6276 0.096 0.124 0.780
#> GSM587203 3 0.5497 0.6412 0.292 0.000 0.708
#> GSM587204 3 0.5497 0.6412 0.292 0.000 0.708
#> GSM587205 3 0.5497 0.6412 0.292 0.000 0.708
#> GSM587206 3 0.5497 0.6412 0.292 0.000 0.708
#> GSM587207 3 0.5497 0.6412 0.292 0.000 0.708
#> GSM587208 3 0.5497 0.6412 0.292 0.000 0.708
#> GSM587209 3 0.5465 0.6423 0.288 0.000 0.712
#> GSM587210 3 0.0592 0.7202 0.012 0.000 0.988
#> GSM587211 3 0.5465 0.6423 0.288 0.000 0.712
#> GSM587212 3 0.0592 0.7202 0.012 0.000 0.988
#> GSM587213 3 0.5497 0.6412 0.292 0.000 0.708
#> GSM587214 3 0.5497 0.6412 0.292 0.000 0.708
#> GSM587215 3 0.5465 0.6423 0.288 0.000 0.712
#> GSM587216 3 0.4062 0.6831 0.164 0.000 0.836
#> GSM587217 3 0.5497 0.6412 0.292 0.000 0.708
#> GSM587191 3 0.5260 0.6793 0.092 0.080 0.828
#> GSM587192 3 0.4174 0.7017 0.092 0.036 0.872
#> GSM587193 3 0.4291 0.6370 0.180 0.000 0.820
#> GSM587194 3 0.6176 0.6284 0.100 0.120 0.780
#> GSM587195 3 0.4443 0.7021 0.084 0.052 0.864
#> GSM587196 3 0.4179 0.7080 0.072 0.052 0.876
#> GSM587197 3 0.5416 0.6730 0.100 0.080 0.820
#> GSM587198 3 0.3896 0.7123 0.060 0.052 0.888
#> GSM587199 3 0.2066 0.7166 0.060 0.000 0.940
#> GSM587200 3 0.1643 0.7197 0.044 0.000 0.956
#> GSM587201 3 0.1753 0.7192 0.048 0.000 0.952
#> GSM587202 3 0.3896 0.7123 0.060 0.052 0.888
#> GSM198767 3 0.5497 0.6412 0.292 0.000 0.708
#> GSM198769 3 0.5465 0.6423 0.288 0.000 0.712
#> GSM198772 3 0.5465 0.6423 0.288 0.000 0.712
#> GSM198773 3 0.5497 0.6412 0.292 0.000 0.708
#> GSM198776 3 0.5497 0.6412 0.292 0.000 0.708
#> GSM198778 3 0.0592 0.7202 0.012 0.000 0.988
#> GSM198780 3 0.0592 0.7202 0.012 0.000 0.988
#> GSM198781 3 0.5497 0.6412 0.292 0.000 0.708
#> GSM198765 3 0.5260 0.6793 0.092 0.080 0.828
#> GSM198766 3 0.4291 0.6370 0.180 0.000 0.820
#> GSM198768 3 0.4269 0.7062 0.076 0.052 0.872
#> GSM198770 3 0.5416 0.6730 0.100 0.080 0.820
#> GSM198771 3 0.3896 0.7123 0.060 0.052 0.888
#> GSM198774 3 0.4505 0.6980 0.092 0.048 0.860
#> GSM198775 3 0.6176 0.6284 0.100 0.120 0.780
#> GSM198777 3 0.4179 0.7080 0.072 0.052 0.876
#> GSM198779 3 0.2066 0.7166 0.060 0.000 0.940
#> GSM587218 1 0.2959 0.9317 0.900 0.000 0.100
#> GSM587219 1 0.2959 0.9317 0.900 0.000 0.100
#> GSM587220 1 0.2959 0.9317 0.900 0.000 0.100
#> GSM587221 1 0.2959 0.9317 0.900 0.000 0.100
#> GSM587222 1 0.2959 0.9317 0.900 0.000 0.100
#> GSM587223 1 0.2959 0.9317 0.900 0.000 0.100
#> GSM587224 1 0.2959 0.9317 0.900 0.000 0.100
#> GSM587225 1 0.4796 0.8397 0.780 0.000 0.220
#> GSM587226 1 0.2959 0.9317 0.900 0.000 0.100
#> GSM587227 1 0.4796 0.8397 0.780 0.000 0.220
#> GSM587228 1 0.4796 0.8397 0.780 0.000 0.220
#> GSM587229 1 0.4796 0.8397 0.780 0.000 0.220
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM587155 2 0.0000 0.995 0.000 1.000 0.000 0.000
#> GSM587156 2 0.0592 0.982 0.000 0.984 0.016 0.000
#> GSM587157 2 0.0707 0.979 0.000 0.980 0.020 0.000
#> GSM587158 2 0.0000 0.995 0.000 1.000 0.000 0.000
#> GSM587159 2 0.0000 0.995 0.000 1.000 0.000 0.000
#> GSM587160 2 0.0000 0.995 0.000 1.000 0.000 0.000
#> GSM587161 2 0.0000 0.995 0.000 1.000 0.000 0.000
#> GSM587162 2 0.0000 0.995 0.000 1.000 0.000 0.000
#> GSM587163 2 0.0000 0.995 0.000 1.000 0.000 0.000
#> GSM587164 2 0.0592 0.982 0.000 0.984 0.016 0.000
#> GSM587165 2 0.0000 0.995 0.000 1.000 0.000 0.000
#> GSM587166 2 0.1389 0.944 0.000 0.952 0.048 0.000
#> GSM587167 2 0.0817 0.974 0.000 0.976 0.024 0.000
#> GSM587168 2 0.0000 0.995 0.000 1.000 0.000 0.000
#> GSM587169 2 0.0000 0.995 0.000 1.000 0.000 0.000
#> GSM587170 2 0.0000 0.995 0.000 1.000 0.000 0.000
#> GSM587171 2 0.0000 0.995 0.000 1.000 0.000 0.000
#> GSM587172 2 0.0000 0.995 0.000 1.000 0.000 0.000
#> GSM587173 2 0.0000 0.995 0.000 1.000 0.000 0.000
#> GSM587174 2 0.0000 0.995 0.000 1.000 0.000 0.000
#> GSM587175 2 0.0000 0.995 0.000 1.000 0.000 0.000
#> GSM587176 2 0.0000 0.995 0.000 1.000 0.000 0.000
#> GSM587177 2 0.0000 0.995 0.000 1.000 0.000 0.000
#> GSM587178 2 0.0000 0.995 0.000 1.000 0.000 0.000
#> GSM587179 2 0.0000 0.995 0.000 1.000 0.000 0.000
#> GSM587180 2 0.0000 0.995 0.000 1.000 0.000 0.000
#> GSM587181 2 0.0000 0.995 0.000 1.000 0.000 0.000
#> GSM587182 2 0.0000 0.995 0.000 1.000 0.000 0.000
#> GSM587183 2 0.0000 0.995 0.000 1.000 0.000 0.000
#> GSM587184 2 0.0000 0.995 0.000 1.000 0.000 0.000
#> GSM587185 2 0.0000 0.995 0.000 1.000 0.000 0.000
#> GSM587186 2 0.0000 0.995 0.000 1.000 0.000 0.000
#> GSM587187 3 0.4877 0.383 0.000 0.408 0.592 0.000
#> GSM587188 3 0.3486 0.749 0.000 0.188 0.812 0.000
#> GSM587189 3 0.3486 0.749 0.000 0.188 0.812 0.000
#> GSM587190 3 0.1022 0.905 0.000 0.032 0.968 0.000
#> GSM587203 1 0.0000 0.879 1.000 0.000 0.000 0.000
#> GSM587204 1 0.0000 0.879 1.000 0.000 0.000 0.000
#> GSM587205 1 0.0000 0.879 1.000 0.000 0.000 0.000
#> GSM587206 1 0.0000 0.879 1.000 0.000 0.000 0.000
#> GSM587207 1 0.0000 0.879 1.000 0.000 0.000 0.000
#> GSM587208 1 0.0000 0.879 1.000 0.000 0.000 0.000
#> GSM587209 1 0.1940 0.863 0.924 0.000 0.076 0.000
#> GSM587210 1 0.4564 0.620 0.672 0.000 0.328 0.000
#> GSM587211 1 0.3688 0.775 0.792 0.000 0.208 0.000
#> GSM587212 1 0.4564 0.620 0.672 0.000 0.328 0.000
#> GSM587213 1 0.0188 0.880 0.996 0.000 0.004 0.000
#> GSM587214 1 0.0188 0.880 0.996 0.000 0.004 0.000
#> GSM587215 1 0.1940 0.863 0.924 0.000 0.076 0.000
#> GSM587216 1 0.3688 0.775 0.792 0.000 0.208 0.000
#> GSM587217 1 0.0188 0.880 0.996 0.000 0.004 0.000
#> GSM587191 3 0.0000 0.923 0.000 0.000 1.000 0.000
#> GSM587192 3 0.0000 0.923 0.000 0.000 1.000 0.000
#> GSM587193 3 0.1807 0.890 0.000 0.008 0.940 0.052
#> GSM587194 3 0.1302 0.897 0.000 0.044 0.956 0.000
#> GSM587195 3 0.0000 0.923 0.000 0.000 1.000 0.000
#> GSM587196 3 0.0000 0.923 0.000 0.000 1.000 0.000
#> GSM587197 3 0.0000 0.923 0.000 0.000 1.000 0.000
#> GSM587198 3 0.0000 0.923 0.000 0.000 1.000 0.000
#> GSM587199 3 0.0000 0.923 0.000 0.000 1.000 0.000
#> GSM587200 3 0.4103 0.597 0.256 0.000 0.744 0.000
#> GSM587201 3 0.3649 0.693 0.204 0.000 0.796 0.000
#> GSM587202 3 0.0000 0.923 0.000 0.000 1.000 0.000
#> GSM198767 1 0.0000 0.879 1.000 0.000 0.000 0.000
#> GSM198769 1 0.1940 0.863 0.924 0.000 0.076 0.000
#> GSM198772 1 0.3528 0.790 0.808 0.000 0.192 0.000
#> GSM198773 1 0.0188 0.880 0.996 0.000 0.004 0.000
#> GSM198776 1 0.0000 0.879 1.000 0.000 0.000 0.000
#> GSM198778 1 0.4585 0.613 0.668 0.000 0.332 0.000
#> GSM198780 1 0.4564 0.620 0.672 0.000 0.328 0.000
#> GSM198781 1 0.0188 0.880 0.996 0.000 0.004 0.000
#> GSM198765 3 0.0000 0.923 0.000 0.000 1.000 0.000
#> GSM198766 3 0.1807 0.890 0.000 0.008 0.940 0.052
#> GSM198768 3 0.0000 0.923 0.000 0.000 1.000 0.000
#> GSM198770 3 0.0000 0.923 0.000 0.000 1.000 0.000
#> GSM198771 3 0.0000 0.923 0.000 0.000 1.000 0.000
#> GSM198774 3 0.0000 0.923 0.000 0.000 1.000 0.000
#> GSM198775 3 0.1302 0.897 0.000 0.044 0.956 0.000
#> GSM198777 3 0.0000 0.923 0.000 0.000 1.000 0.000
#> GSM198779 3 0.0000 0.923 0.000 0.000 1.000 0.000
#> GSM587218 4 0.0000 0.941 0.000 0.000 0.000 1.000
#> GSM587219 4 0.0000 0.941 0.000 0.000 0.000 1.000
#> GSM587220 4 0.0000 0.941 0.000 0.000 0.000 1.000
#> GSM587221 4 0.0000 0.941 0.000 0.000 0.000 1.000
#> GSM587222 4 0.0000 0.941 0.000 0.000 0.000 1.000
#> GSM587223 4 0.0000 0.941 0.000 0.000 0.000 1.000
#> GSM587224 4 0.0000 0.941 0.000 0.000 0.000 1.000
#> GSM587225 4 0.2814 0.872 0.000 0.000 0.132 0.868
#> GSM587226 4 0.0000 0.941 0.000 0.000 0.000 1.000
#> GSM587227 4 0.2814 0.872 0.000 0.000 0.132 0.868
#> GSM587228 4 0.2814 0.872 0.000 0.000 0.132 0.868
#> GSM587229 4 0.2814 0.872 0.000 0.000 0.132 0.868
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM587155 2 0.0404 0.95351 0.000 0.988 0.000 0.000 0.012
#> GSM587156 2 0.3409 0.79799 0.000 0.836 0.112 0.000 0.052
#> GSM587157 2 0.2889 0.84405 0.000 0.872 0.084 0.000 0.044
#> GSM587158 2 0.0000 0.95900 0.000 1.000 0.000 0.000 0.000
#> GSM587159 2 0.0000 0.95900 0.000 1.000 0.000 0.000 0.000
#> GSM587160 2 0.0000 0.95900 0.000 1.000 0.000 0.000 0.000
#> GSM587161 2 0.0000 0.95900 0.000 1.000 0.000 0.000 0.000
#> GSM587162 2 0.0162 0.95744 0.000 0.996 0.000 0.000 0.004
#> GSM587163 2 0.0000 0.95900 0.000 1.000 0.000 0.000 0.000
#> GSM587164 2 0.2067 0.89552 0.000 0.920 0.048 0.000 0.032
#> GSM587165 2 0.0162 0.95744 0.000 0.996 0.000 0.000 0.004
#> GSM587166 2 0.4298 0.67446 0.000 0.756 0.184 0.000 0.060
#> GSM587167 2 0.1992 0.89963 0.000 0.924 0.044 0.000 0.032
#> GSM587168 2 0.0162 0.95744 0.000 0.996 0.000 0.000 0.004
#> GSM587169 2 0.0000 0.95900 0.000 1.000 0.000 0.000 0.000
#> GSM587170 2 0.0290 0.95527 0.000 0.992 0.000 0.000 0.008
#> GSM587171 2 0.0000 0.95900 0.000 1.000 0.000 0.000 0.000
#> GSM587172 2 0.0000 0.95900 0.000 1.000 0.000 0.000 0.000
#> GSM587173 2 0.0000 0.95900 0.000 1.000 0.000 0.000 0.000
#> GSM587174 2 0.0000 0.95900 0.000 1.000 0.000 0.000 0.000
#> GSM587175 2 0.0290 0.95527 0.000 0.992 0.000 0.000 0.008
#> GSM587176 2 0.0000 0.95900 0.000 1.000 0.000 0.000 0.000
#> GSM587177 2 0.0000 0.95900 0.000 1.000 0.000 0.000 0.000
#> GSM587178 2 0.0000 0.95900 0.000 1.000 0.000 0.000 0.000
#> GSM587179 2 0.0000 0.95900 0.000 1.000 0.000 0.000 0.000
#> GSM587180 2 0.0000 0.95900 0.000 1.000 0.000 0.000 0.000
#> GSM587181 2 0.0000 0.95900 0.000 1.000 0.000 0.000 0.000
#> GSM587182 2 0.0162 0.95744 0.000 0.996 0.000 0.000 0.004
#> GSM587183 2 0.0000 0.95900 0.000 1.000 0.000 0.000 0.000
#> GSM587184 2 0.0000 0.95900 0.000 1.000 0.000 0.000 0.000
#> GSM587185 2 0.0000 0.95900 0.000 1.000 0.000 0.000 0.000
#> GSM587186 2 0.0000 0.95900 0.000 1.000 0.000 0.000 0.000
#> GSM587187 2 0.6458 0.00563 0.000 0.500 0.260 0.000 0.240
#> GSM587188 5 0.5819 0.63144 0.004 0.088 0.368 0.000 0.540
#> GSM587189 3 0.6181 -0.52186 0.004 0.120 0.484 0.000 0.392
#> GSM587190 5 0.4450 0.71204 0.004 0.000 0.488 0.000 0.508
#> GSM587203 1 0.3039 0.77507 0.808 0.000 0.000 0.000 0.192
#> GSM587204 1 0.2471 0.78398 0.864 0.000 0.000 0.000 0.136
#> GSM587205 1 0.3039 0.77507 0.808 0.000 0.000 0.000 0.192
#> GSM587206 1 0.3039 0.77507 0.808 0.000 0.000 0.000 0.192
#> GSM587207 1 0.3039 0.77507 0.808 0.000 0.000 0.000 0.192
#> GSM587208 1 0.3039 0.77507 0.808 0.000 0.000 0.000 0.192
#> GSM587209 1 0.0290 0.79897 0.992 0.000 0.000 0.000 0.008
#> GSM587210 1 0.6695 0.24142 0.432 0.000 0.284 0.000 0.284
#> GSM587211 1 0.4696 0.64622 0.736 0.000 0.108 0.000 0.156
#> GSM587212 1 0.6417 0.37785 0.504 0.000 0.216 0.000 0.280
#> GSM587213 1 0.0290 0.79926 0.992 0.000 0.000 0.000 0.008
#> GSM587214 1 0.0000 0.79998 1.000 0.000 0.000 0.000 0.000
#> GSM587215 1 0.0703 0.79424 0.976 0.000 0.000 0.000 0.024
#> GSM587216 1 0.5481 0.57095 0.656 0.000 0.172 0.000 0.172
#> GSM587217 1 0.0000 0.79998 1.000 0.000 0.000 0.000 0.000
#> GSM587191 3 0.2583 0.59460 0.004 0.000 0.864 0.000 0.132
#> GSM587192 3 0.2763 0.55844 0.004 0.000 0.848 0.000 0.148
#> GSM587193 5 0.4473 0.72269 0.008 0.000 0.412 0.000 0.580
#> GSM587194 5 0.4425 0.66409 0.004 0.000 0.452 0.000 0.544
#> GSM587195 3 0.1478 0.61180 0.000 0.000 0.936 0.000 0.064
#> GSM587196 3 0.1608 0.61820 0.000 0.000 0.928 0.000 0.072
#> GSM587197 3 0.3160 0.40784 0.004 0.000 0.808 0.000 0.188
#> GSM587198 3 0.0703 0.65094 0.000 0.000 0.976 0.000 0.024
#> GSM587199 3 0.2732 0.54292 0.000 0.000 0.840 0.000 0.160
#> GSM587200 3 0.6193 0.11495 0.192 0.000 0.548 0.000 0.260
#> GSM587201 3 0.5702 0.16954 0.192 0.000 0.628 0.000 0.180
#> GSM587202 3 0.1908 0.61415 0.000 0.000 0.908 0.000 0.092
#> GSM198767 1 0.3039 0.77507 0.808 0.000 0.000 0.000 0.192
#> GSM198769 1 0.0290 0.79897 0.992 0.000 0.000 0.000 0.008
#> GSM198772 1 0.4117 0.69066 0.788 0.000 0.096 0.000 0.116
#> GSM198773 1 0.0162 0.79955 0.996 0.000 0.000 0.000 0.004
#> GSM198776 1 0.2471 0.78398 0.864 0.000 0.000 0.000 0.136
#> GSM198778 1 0.6695 0.24142 0.432 0.000 0.284 0.000 0.284
#> GSM198780 1 0.6417 0.37785 0.504 0.000 0.216 0.000 0.280
#> GSM198781 1 0.0000 0.79998 1.000 0.000 0.000 0.000 0.000
#> GSM198765 3 0.2536 0.59851 0.004 0.000 0.868 0.000 0.128
#> GSM198766 5 0.4464 0.71782 0.008 0.000 0.408 0.000 0.584
#> GSM198768 3 0.0880 0.63566 0.000 0.000 0.968 0.000 0.032
#> GSM198770 3 0.3160 0.40784 0.004 0.000 0.808 0.000 0.188
#> GSM198771 3 0.0880 0.65112 0.000 0.000 0.968 0.000 0.032
#> GSM198774 3 0.2719 0.55962 0.004 0.000 0.852 0.000 0.144
#> GSM198775 5 0.4420 0.66609 0.004 0.000 0.448 0.000 0.548
#> GSM198777 3 0.1671 0.61665 0.000 0.000 0.924 0.000 0.076
#> GSM198779 3 0.2732 0.54292 0.000 0.000 0.840 0.000 0.160
#> GSM587218 4 0.3507 0.79151 0.000 0.000 0.120 0.828 0.052
#> GSM587219 4 0.0000 0.91924 0.000 0.000 0.000 1.000 0.000
#> GSM587220 4 0.0000 0.91924 0.000 0.000 0.000 1.000 0.000
#> GSM587221 4 0.0000 0.91924 0.000 0.000 0.000 1.000 0.000
#> GSM587222 4 0.0000 0.91924 0.000 0.000 0.000 1.000 0.000
#> GSM587223 4 0.0000 0.91924 0.000 0.000 0.000 1.000 0.000
#> GSM587224 4 0.0000 0.91924 0.000 0.000 0.000 1.000 0.000
#> GSM587225 4 0.3362 0.85603 0.000 0.000 0.076 0.844 0.080
#> GSM587226 4 0.0000 0.91924 0.000 0.000 0.000 1.000 0.000
#> GSM587227 4 0.3119 0.86764 0.000 0.000 0.072 0.860 0.068
#> GSM587228 4 0.3119 0.86764 0.000 0.000 0.072 0.860 0.068
#> GSM587229 4 0.3119 0.86764 0.000 0.000 0.072 0.860 0.068
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM587155 2 0.0777 0.947 0.000 0.972 0.000 0.000 0.024 0.004
#> GSM587156 2 0.2871 0.765 0.000 0.804 0.000 0.000 0.192 0.004
#> GSM587157 2 0.2442 0.827 0.000 0.852 0.000 0.000 0.144 0.004
#> GSM587158 2 0.0000 0.965 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587159 2 0.0000 0.965 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587160 2 0.0000 0.965 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587161 2 0.0000 0.965 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587162 2 0.0000 0.965 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587163 2 0.0000 0.965 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587164 2 0.2362 0.838 0.000 0.860 0.000 0.000 0.136 0.004
#> GSM587165 2 0.0000 0.965 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587166 2 0.3915 0.587 0.000 0.704 0.020 0.000 0.272 0.004
#> GSM587167 2 0.2362 0.839 0.000 0.860 0.000 0.000 0.136 0.004
#> GSM587168 2 0.0000 0.965 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587169 2 0.0260 0.960 0.000 0.992 0.000 0.000 0.008 0.000
#> GSM587170 2 0.0603 0.953 0.000 0.980 0.000 0.000 0.016 0.004
#> GSM587171 2 0.0000 0.965 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587172 2 0.0000 0.965 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587173 2 0.0000 0.965 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587174 2 0.0000 0.965 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587175 2 0.0260 0.961 0.000 0.992 0.000 0.000 0.008 0.000
#> GSM587176 2 0.0000 0.965 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587177 2 0.0000 0.965 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587178 2 0.0000 0.965 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587179 2 0.0000 0.965 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587180 2 0.0000 0.965 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587181 2 0.0000 0.965 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587182 2 0.0146 0.963 0.000 0.996 0.000 0.000 0.004 0.000
#> GSM587183 2 0.0000 0.965 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587184 2 0.0000 0.965 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587185 2 0.0000 0.965 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587186 2 0.0000 0.965 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587187 5 0.5022 0.107 0.000 0.432 0.072 0.000 0.496 0.000
#> GSM587188 5 0.3516 0.524 0.000 0.088 0.096 0.000 0.812 0.004
#> GSM587189 5 0.3611 0.520 0.000 0.108 0.096 0.000 0.796 0.000
#> GSM587190 5 0.3076 0.479 0.000 0.000 0.240 0.000 0.760 0.000
#> GSM587203 1 0.0000 0.700 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM587204 1 0.3288 0.676 0.724 0.000 0.000 0.000 0.000 0.276
#> GSM587205 1 0.0363 0.708 0.988 0.000 0.000 0.000 0.000 0.012
#> GSM587206 1 0.0363 0.708 0.988 0.000 0.000 0.000 0.000 0.012
#> GSM587207 1 0.0363 0.708 0.988 0.000 0.000 0.000 0.000 0.012
#> GSM587208 1 0.0363 0.708 0.988 0.000 0.000 0.000 0.000 0.012
#> GSM587209 6 0.2883 0.352 0.212 0.000 0.000 0.000 0.000 0.788
#> GSM587210 6 0.5328 0.384 0.000 0.000 0.200 0.000 0.204 0.596
#> GSM587211 6 0.1225 0.586 0.036 0.000 0.012 0.000 0.000 0.952
#> GSM587212 6 0.4316 0.547 0.000 0.000 0.128 0.000 0.144 0.728
#> GSM587213 1 0.3854 0.526 0.536 0.000 0.000 0.000 0.000 0.464
#> GSM587214 1 0.3857 0.542 0.532 0.000 0.000 0.000 0.000 0.468
#> GSM587215 6 0.3175 0.247 0.256 0.000 0.000 0.000 0.000 0.744
#> GSM587216 6 0.3628 0.626 0.036 0.000 0.044 0.000 0.100 0.820
#> GSM587217 1 0.3860 0.536 0.528 0.000 0.000 0.000 0.000 0.472
#> GSM587191 3 0.2491 0.708 0.000 0.000 0.836 0.000 0.164 0.000
#> GSM587192 3 0.4161 0.604 0.036 0.000 0.696 0.000 0.264 0.004
#> GSM587193 5 0.5817 0.314 0.000 0.000 0.312 0.000 0.480 0.208
#> GSM587194 5 0.3930 0.264 0.000 0.000 0.420 0.000 0.576 0.004
#> GSM587195 3 0.0935 0.756 0.000 0.000 0.964 0.000 0.032 0.004
#> GSM587196 3 0.0865 0.739 0.000 0.000 0.964 0.000 0.036 0.000
#> GSM587197 3 0.3023 0.664 0.000 0.000 0.784 0.000 0.212 0.004
#> GSM587198 3 0.0508 0.755 0.000 0.000 0.984 0.000 0.012 0.004
#> GSM587199 3 0.2831 0.690 0.024 0.000 0.840 0.000 0.136 0.000
#> GSM587200 3 0.5817 0.385 0.016 0.000 0.544 0.000 0.288 0.152
#> GSM587201 3 0.6256 0.302 0.032 0.000 0.508 0.000 0.276 0.184
#> GSM587202 3 0.0865 0.739 0.000 0.000 0.964 0.000 0.036 0.000
#> GSM198767 1 0.0363 0.708 0.988 0.000 0.000 0.000 0.000 0.012
#> GSM198769 6 0.3076 0.290 0.240 0.000 0.000 0.000 0.000 0.760
#> GSM198772 6 0.1082 0.576 0.040 0.000 0.004 0.000 0.000 0.956
#> GSM198773 1 0.3843 0.545 0.548 0.000 0.000 0.000 0.000 0.452
#> GSM198776 1 0.3288 0.676 0.724 0.000 0.000 0.000 0.000 0.276
#> GSM198778 6 0.5328 0.384 0.000 0.000 0.200 0.000 0.204 0.596
#> GSM198780 6 0.4316 0.547 0.000 0.000 0.128 0.000 0.144 0.728
#> GSM198781 1 0.3857 0.542 0.532 0.000 0.000 0.000 0.000 0.468
#> GSM198765 3 0.2491 0.708 0.000 0.000 0.836 0.000 0.164 0.000
#> GSM198766 5 0.5784 0.328 0.000 0.000 0.260 0.000 0.504 0.236
#> GSM198768 3 0.0858 0.756 0.000 0.000 0.968 0.000 0.028 0.004
#> GSM198770 3 0.3109 0.649 0.000 0.000 0.772 0.000 0.224 0.004
#> GSM198771 3 0.0146 0.758 0.000 0.000 0.996 0.000 0.000 0.004
#> GSM198774 3 0.4161 0.604 0.036 0.000 0.696 0.000 0.264 0.004
#> GSM198775 5 0.3923 0.269 0.000 0.000 0.416 0.000 0.580 0.004
#> GSM198777 3 0.0865 0.739 0.000 0.000 0.964 0.000 0.036 0.000
#> GSM198779 3 0.2831 0.690 0.024 0.000 0.840 0.000 0.136 0.000
#> GSM587218 4 0.2697 0.808 0.000 0.000 0.000 0.812 0.188 0.000
#> GSM587219 4 0.0000 0.919 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM587220 4 0.0000 0.919 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM587221 4 0.0000 0.919 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM587222 4 0.0000 0.919 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM587223 4 0.0000 0.919 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM587224 4 0.0000 0.919 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM587225 4 0.2658 0.872 0.000 0.000 0.036 0.864 0.100 0.000
#> GSM587226 4 0.0000 0.919 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM587227 4 0.2658 0.872 0.000 0.000 0.036 0.864 0.100 0.000
#> GSM587228 4 0.2658 0.872 0.000 0.000 0.036 0.864 0.100 0.000
#> GSM587229 4 0.2658 0.872 0.000 0.000 0.036 0.864 0.100 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 specimen(p) k
#> MAD:mclust 88 7.88e-17 2
#> MAD:mclust 91 1.66e-32 3
#> MAD:mclust 91 1.66e-47 4
#> MAD:mclust 82 6.99e-43 5
#> MAD:mclust 79 1.39e-49 6
If matrix rows can be associated to genes, consider to use functional_enrichment(res,
...) to perform function enrichment for the signature genes. See this vignette for more detailed explanations.
The object with results only for a single top-value method and a single partition method can be extracted as:
res = res_list["MAD", "NMF"]
# you can also extract it by
# res = res_list["MAD:NMF"]
A summary of res and all the functions that can be applied to it:
res
#> A 'ConsensusPartition' object with k = 2, 3, 4, 5, 6.
#> On a matrix with 51941 rows and 92 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 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 1.000 0.983 0.992 0.5012 0.500 0.500
#> 3 3 0.909 0.909 0.954 0.3050 0.703 0.482
#> 4 4 0.981 0.925 0.967 0.1164 0.883 0.683
#> 5 5 0.884 0.884 0.932 0.0544 0.958 0.849
#> 6 6 0.811 0.699 0.838 0.0395 0.974 0.893
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 3
There is also optional best \(k\) = 2 3 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
#> GSM587155 2 0.0000 0.985 0.000 1.000
#> GSM587156 2 0.0000 0.985 0.000 1.000
#> GSM587157 2 0.0000 0.985 0.000 1.000
#> GSM587158 2 0.0000 0.985 0.000 1.000
#> GSM587159 2 0.0000 0.985 0.000 1.000
#> GSM587160 2 0.0000 0.985 0.000 1.000
#> GSM587161 2 0.0000 0.985 0.000 1.000
#> GSM587162 2 0.0000 0.985 0.000 1.000
#> GSM587163 2 0.0000 0.985 0.000 1.000
#> GSM587164 2 0.0000 0.985 0.000 1.000
#> GSM587165 2 0.0000 0.985 0.000 1.000
#> GSM587166 2 0.0000 0.985 0.000 1.000
#> GSM587167 2 0.0000 0.985 0.000 1.000
#> GSM587168 2 0.0000 0.985 0.000 1.000
#> GSM587169 2 0.0000 0.985 0.000 1.000
#> GSM587170 2 0.0000 0.985 0.000 1.000
#> GSM587171 2 0.0000 0.985 0.000 1.000
#> GSM587172 2 0.0000 0.985 0.000 1.000
#> GSM587173 2 0.0000 0.985 0.000 1.000
#> GSM587174 2 0.0000 0.985 0.000 1.000
#> GSM587175 2 0.0000 0.985 0.000 1.000
#> GSM587176 2 0.0000 0.985 0.000 1.000
#> GSM587177 2 0.0000 0.985 0.000 1.000
#> GSM587178 2 0.0000 0.985 0.000 1.000
#> GSM587179 2 0.0000 0.985 0.000 1.000
#> GSM587180 2 0.0000 0.985 0.000 1.000
#> GSM587181 2 0.0000 0.985 0.000 1.000
#> GSM587182 2 0.0000 0.985 0.000 1.000
#> GSM587183 2 0.0000 0.985 0.000 1.000
#> GSM587184 2 0.0000 0.985 0.000 1.000
#> GSM587185 2 0.0000 0.985 0.000 1.000
#> GSM587186 2 0.0000 0.985 0.000 1.000
#> GSM587187 2 0.0000 0.985 0.000 1.000
#> GSM587188 2 0.0000 0.985 0.000 1.000
#> GSM587189 2 0.0000 0.985 0.000 1.000
#> GSM587190 2 0.0000 0.985 0.000 1.000
#> GSM587203 1 0.0000 1.000 1.000 0.000
#> GSM587204 1 0.0000 1.000 1.000 0.000
#> GSM587205 1 0.0000 1.000 1.000 0.000
#> GSM587206 1 0.0000 1.000 1.000 0.000
#> GSM587207 1 0.0000 1.000 1.000 0.000
#> GSM587208 1 0.0000 1.000 1.000 0.000
#> GSM587209 1 0.0000 1.000 1.000 0.000
#> GSM587210 1 0.0000 1.000 1.000 0.000
#> GSM587211 1 0.0000 1.000 1.000 0.000
#> GSM587212 1 0.0000 1.000 1.000 0.000
#> GSM587213 1 0.0000 1.000 1.000 0.000
#> GSM587214 1 0.0000 1.000 1.000 0.000
#> GSM587215 1 0.0000 1.000 1.000 0.000
#> GSM587216 1 0.0000 1.000 1.000 0.000
#> GSM587217 1 0.0000 1.000 1.000 0.000
#> GSM587191 2 0.0000 0.985 0.000 1.000
#> GSM587192 1 0.0000 1.000 1.000 0.000
#> GSM587193 1 0.0000 1.000 1.000 0.000
#> GSM587194 2 0.0672 0.979 0.008 0.992
#> GSM587195 2 0.0000 0.985 0.000 1.000
#> GSM587196 2 0.7139 0.766 0.196 0.804
#> GSM587197 2 0.0000 0.985 0.000 1.000
#> GSM587198 2 0.0376 0.982 0.004 0.996
#> GSM587199 2 0.0000 0.985 0.000 1.000
#> GSM587200 1 0.0000 1.000 1.000 0.000
#> GSM587201 1 0.0000 1.000 1.000 0.000
#> GSM587202 2 0.4690 0.889 0.100 0.900
#> GSM198767 1 0.0000 1.000 1.000 0.000
#> GSM198769 1 0.0000 1.000 1.000 0.000
#> GSM198772 1 0.0000 1.000 1.000 0.000
#> GSM198773 1 0.0000 1.000 1.000 0.000
#> GSM198776 1 0.0000 1.000 1.000 0.000
#> GSM198778 1 0.0000 1.000 1.000 0.000
#> GSM198780 1 0.0000 1.000 1.000 0.000
#> GSM198781 1 0.0000 1.000 1.000 0.000
#> GSM198765 2 0.1184 0.972 0.016 0.984
#> GSM198766 1 0.0000 1.000 1.000 0.000
#> GSM198768 2 0.0000 0.985 0.000 1.000
#> GSM198770 2 0.0000 0.985 0.000 1.000
#> GSM198771 2 0.8955 0.566 0.312 0.688
#> GSM198774 1 0.0000 1.000 1.000 0.000
#> GSM198775 2 0.0000 0.985 0.000 1.000
#> GSM198777 2 0.4562 0.894 0.096 0.904
#> GSM198779 2 0.0000 0.985 0.000 1.000
#> GSM587218 1 0.0000 1.000 1.000 0.000
#> GSM587219 1 0.0000 1.000 1.000 0.000
#> GSM587220 1 0.0000 1.000 1.000 0.000
#> GSM587221 1 0.0000 1.000 1.000 0.000
#> GSM587222 1 0.0000 1.000 1.000 0.000
#> GSM587223 1 0.0000 1.000 1.000 0.000
#> GSM587224 1 0.0000 1.000 1.000 0.000
#> GSM587225 1 0.0000 1.000 1.000 0.000
#> GSM587226 1 0.0000 1.000 1.000 0.000
#> GSM587227 1 0.0000 1.000 1.000 0.000
#> GSM587228 1 0.0000 1.000 1.000 0.000
#> GSM587229 1 0.0000 1.000 1.000 0.000
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM587155 2 0.0000 0.973 0.000 1.000 0.000
#> GSM587156 2 0.0000 0.973 0.000 1.000 0.000
#> GSM587157 2 0.0237 0.970 0.000 0.996 0.004
#> GSM587158 2 0.0000 0.973 0.000 1.000 0.000
#> GSM587159 2 0.0000 0.973 0.000 1.000 0.000
#> GSM587160 2 0.0000 0.973 0.000 1.000 0.000
#> GSM587161 2 0.0000 0.973 0.000 1.000 0.000
#> GSM587162 2 0.0000 0.973 0.000 1.000 0.000
#> GSM587163 2 0.0000 0.973 0.000 1.000 0.000
#> GSM587164 2 0.0000 0.973 0.000 1.000 0.000
#> GSM587165 2 0.0000 0.973 0.000 1.000 0.000
#> GSM587166 2 0.0000 0.973 0.000 1.000 0.000
#> GSM587167 2 0.0000 0.973 0.000 1.000 0.000
#> GSM587168 2 0.0000 0.973 0.000 1.000 0.000
#> GSM587169 2 0.0000 0.973 0.000 1.000 0.000
#> GSM587170 2 0.0000 0.973 0.000 1.000 0.000
#> GSM587171 2 0.0000 0.973 0.000 1.000 0.000
#> GSM587172 2 0.0000 0.973 0.000 1.000 0.000
#> GSM587173 2 0.0000 0.973 0.000 1.000 0.000
#> GSM587174 2 0.0000 0.973 0.000 1.000 0.000
#> GSM587175 2 0.0000 0.973 0.000 1.000 0.000
#> GSM587176 2 0.0000 0.973 0.000 1.000 0.000
#> GSM587177 2 0.0000 0.973 0.000 1.000 0.000
#> GSM587178 2 0.0000 0.973 0.000 1.000 0.000
#> GSM587179 2 0.0000 0.973 0.000 1.000 0.000
#> GSM587180 2 0.0000 0.973 0.000 1.000 0.000
#> GSM587181 2 0.0000 0.973 0.000 1.000 0.000
#> GSM587182 2 0.0000 0.973 0.000 1.000 0.000
#> GSM587183 2 0.0000 0.973 0.000 1.000 0.000
#> GSM587184 2 0.0000 0.973 0.000 1.000 0.000
#> GSM587185 2 0.0000 0.973 0.000 1.000 0.000
#> GSM587186 2 0.0000 0.973 0.000 1.000 0.000
#> GSM587187 2 0.0424 0.967 0.000 0.992 0.008
#> GSM587188 2 0.0892 0.957 0.000 0.980 0.020
#> GSM587189 2 0.0747 0.961 0.000 0.984 0.016
#> GSM587190 3 0.1289 0.898 0.000 0.032 0.968
#> GSM587203 1 0.0000 1.000 1.000 0.000 0.000
#> GSM587204 1 0.0000 1.000 1.000 0.000 0.000
#> GSM587205 1 0.0000 1.000 1.000 0.000 0.000
#> GSM587206 1 0.0000 1.000 1.000 0.000 0.000
#> GSM587207 1 0.0000 1.000 1.000 0.000 0.000
#> GSM587208 1 0.0000 1.000 1.000 0.000 0.000
#> GSM587209 1 0.0000 1.000 1.000 0.000 0.000
#> GSM587210 3 0.2448 0.887 0.076 0.000 0.924
#> GSM587211 1 0.0000 1.000 1.000 0.000 0.000
#> GSM587212 3 0.5497 0.673 0.292 0.000 0.708
#> GSM587213 1 0.0000 1.000 1.000 0.000 0.000
#> GSM587214 1 0.0000 1.000 1.000 0.000 0.000
#> GSM587215 1 0.0000 1.000 1.000 0.000 0.000
#> GSM587216 1 0.0000 1.000 1.000 0.000 0.000
#> GSM587217 1 0.0000 1.000 1.000 0.000 0.000
#> GSM587191 2 0.6008 0.368 0.000 0.628 0.372
#> GSM587192 3 0.4654 0.779 0.208 0.000 0.792
#> GSM587193 3 0.4974 0.742 0.236 0.000 0.764
#> GSM587194 3 0.0000 0.903 0.000 0.000 1.000
#> GSM587195 2 0.6280 0.106 0.000 0.540 0.460
#> GSM587196 3 0.2537 0.867 0.000 0.080 0.920
#> GSM587197 3 0.0892 0.902 0.000 0.020 0.980
#> GSM587198 3 0.0592 0.903 0.000 0.012 0.988
#> GSM587199 3 0.0592 0.903 0.000 0.012 0.988
#> GSM587200 3 0.0892 0.905 0.020 0.000 0.980
#> GSM587201 3 0.5905 0.562 0.352 0.000 0.648
#> GSM587202 3 0.1411 0.895 0.000 0.036 0.964
#> GSM198767 1 0.0000 1.000 1.000 0.000 0.000
#> GSM198769 1 0.0000 1.000 1.000 0.000 0.000
#> GSM198772 1 0.0000 1.000 1.000 0.000 0.000
#> GSM198773 1 0.0000 1.000 1.000 0.000 0.000
#> GSM198776 1 0.0000 1.000 1.000 0.000 0.000
#> GSM198778 3 0.2448 0.887 0.076 0.000 0.924
#> GSM198780 3 0.5591 0.656 0.304 0.000 0.696
#> GSM198781 1 0.0000 1.000 1.000 0.000 0.000
#> GSM198765 3 0.6126 0.363 0.000 0.400 0.600
#> GSM198766 3 0.5835 0.579 0.340 0.000 0.660
#> GSM198768 3 0.4178 0.778 0.000 0.172 0.828
#> GSM198770 3 0.1163 0.899 0.000 0.028 0.972
#> GSM198771 3 0.0592 0.903 0.000 0.012 0.988
#> GSM198774 3 0.4702 0.774 0.212 0.000 0.788
#> GSM198775 3 0.0237 0.904 0.000 0.004 0.996
#> GSM198777 3 0.2878 0.854 0.000 0.096 0.904
#> GSM198779 3 0.0592 0.903 0.000 0.012 0.988
#> GSM587218 3 0.0237 0.904 0.004 0.000 0.996
#> GSM587219 3 0.0747 0.905 0.016 0.000 0.984
#> GSM587220 3 0.1411 0.901 0.036 0.000 0.964
#> GSM587221 3 0.0747 0.905 0.016 0.000 0.984
#> GSM587222 3 0.1031 0.904 0.024 0.000 0.976
#> GSM587223 3 0.0592 0.905 0.012 0.000 0.988
#> GSM587224 3 0.0592 0.905 0.012 0.000 0.988
#> GSM587225 3 0.1289 0.902 0.032 0.000 0.968
#> GSM587226 3 0.0747 0.905 0.016 0.000 0.984
#> GSM587227 3 0.1031 0.904 0.024 0.000 0.976
#> GSM587228 3 0.1289 0.902 0.032 0.000 0.968
#> GSM587229 3 0.2625 0.877 0.084 0.000 0.916
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM587155 2 0.0000 0.999 0.000 1.000 0.000 0.000
#> GSM587156 2 0.0188 0.995 0.000 0.996 0.004 0.000
#> GSM587157 2 0.0188 0.995 0.000 0.996 0.004 0.000
#> GSM587158 2 0.0000 0.999 0.000 1.000 0.000 0.000
#> GSM587159 2 0.0000 0.999 0.000 1.000 0.000 0.000
#> GSM587160 2 0.0000 0.999 0.000 1.000 0.000 0.000
#> GSM587161 2 0.0000 0.999 0.000 1.000 0.000 0.000
#> GSM587162 2 0.0000 0.999 0.000 1.000 0.000 0.000
#> GSM587163 2 0.0000 0.999 0.000 1.000 0.000 0.000
#> GSM587164 2 0.0000 0.999 0.000 1.000 0.000 0.000
#> GSM587165 2 0.0000 0.999 0.000 1.000 0.000 0.000
#> GSM587166 2 0.0188 0.995 0.000 0.996 0.004 0.000
#> GSM587167 2 0.0000 0.999 0.000 1.000 0.000 0.000
#> GSM587168 2 0.0000 0.999 0.000 1.000 0.000 0.000
#> GSM587169 2 0.0000 0.999 0.000 1.000 0.000 0.000
#> GSM587170 2 0.0000 0.999 0.000 1.000 0.000 0.000
#> GSM587171 2 0.0000 0.999 0.000 1.000 0.000 0.000
#> GSM587172 2 0.0000 0.999 0.000 1.000 0.000 0.000
#> GSM587173 2 0.0000 0.999 0.000 1.000 0.000 0.000
#> GSM587174 2 0.0000 0.999 0.000 1.000 0.000 0.000
#> GSM587175 2 0.0000 0.999 0.000 1.000 0.000 0.000
#> GSM587176 2 0.0000 0.999 0.000 1.000 0.000 0.000
#> GSM587177 2 0.0000 0.999 0.000 1.000 0.000 0.000
#> GSM587178 2 0.0000 0.999 0.000 1.000 0.000 0.000
#> GSM587179 2 0.0000 0.999 0.000 1.000 0.000 0.000
#> GSM587180 2 0.0000 0.999 0.000 1.000 0.000 0.000
#> GSM587181 2 0.0000 0.999 0.000 1.000 0.000 0.000
#> GSM587182 2 0.0000 0.999 0.000 1.000 0.000 0.000
#> GSM587183 2 0.0000 0.999 0.000 1.000 0.000 0.000
#> GSM587184 2 0.0000 0.999 0.000 1.000 0.000 0.000
#> GSM587185 2 0.0000 0.999 0.000 1.000 0.000 0.000
#> GSM587186 2 0.0000 0.999 0.000 1.000 0.000 0.000
#> GSM587187 2 0.0336 0.992 0.000 0.992 0.008 0.000
#> GSM587188 2 0.0657 0.986 0.000 0.984 0.012 0.004
#> GSM587189 2 0.0657 0.986 0.000 0.984 0.012 0.004
#> GSM587190 4 0.2983 0.880 0.000 0.068 0.040 0.892
#> GSM587203 1 0.0000 0.953 1.000 0.000 0.000 0.000
#> GSM587204 1 0.0000 0.953 1.000 0.000 0.000 0.000
#> GSM587205 1 0.0000 0.953 1.000 0.000 0.000 0.000
#> GSM587206 1 0.0000 0.953 1.000 0.000 0.000 0.000
#> GSM587207 1 0.0000 0.953 1.000 0.000 0.000 0.000
#> GSM587208 1 0.0000 0.953 1.000 0.000 0.000 0.000
#> GSM587209 1 0.0000 0.953 1.000 0.000 0.000 0.000
#> GSM587210 3 0.0895 0.903 0.020 0.000 0.976 0.004
#> GSM587211 1 0.0188 0.950 0.996 0.000 0.004 0.000
#> GSM587212 3 0.4454 0.589 0.308 0.000 0.692 0.000
#> GSM587213 1 0.0000 0.953 1.000 0.000 0.000 0.000
#> GSM587214 1 0.0000 0.953 1.000 0.000 0.000 0.000
#> GSM587215 1 0.0188 0.950 0.996 0.000 0.004 0.000
#> GSM587216 1 0.0188 0.950 0.996 0.000 0.004 0.000
#> GSM587217 1 0.0000 0.953 1.000 0.000 0.000 0.000
#> GSM587191 3 0.1867 0.856 0.000 0.072 0.928 0.000
#> GSM587192 3 0.0707 0.903 0.020 0.000 0.980 0.000
#> GSM587193 1 0.6392 0.198 0.528 0.000 0.068 0.404
#> GSM587194 3 0.4877 0.344 0.000 0.000 0.592 0.408
#> GSM587195 3 0.1902 0.862 0.000 0.064 0.932 0.004
#> GSM587196 3 0.0188 0.906 0.000 0.000 0.996 0.004
#> GSM587197 3 0.2334 0.857 0.000 0.004 0.908 0.088
#> GSM587198 3 0.0336 0.906 0.000 0.000 0.992 0.008
#> GSM587199 3 0.0336 0.906 0.000 0.000 0.992 0.008
#> GSM587200 3 0.0524 0.906 0.008 0.000 0.988 0.004
#> GSM587201 3 0.0707 0.903 0.020 0.000 0.980 0.000
#> GSM587202 3 0.0336 0.906 0.000 0.000 0.992 0.008
#> GSM198767 1 0.0000 0.953 1.000 0.000 0.000 0.000
#> GSM198769 1 0.0000 0.953 1.000 0.000 0.000 0.000
#> GSM198772 1 0.0000 0.953 1.000 0.000 0.000 0.000
#> GSM198773 1 0.0000 0.953 1.000 0.000 0.000 0.000
#> GSM198776 1 0.0000 0.953 1.000 0.000 0.000 0.000
#> GSM198778 3 0.0779 0.905 0.016 0.000 0.980 0.004
#> GSM198780 3 0.4356 0.616 0.292 0.000 0.708 0.000
#> GSM198781 1 0.0000 0.953 1.000 0.000 0.000 0.000
#> GSM198765 3 0.0817 0.896 0.000 0.024 0.976 0.000
#> GSM198766 1 0.6031 0.280 0.564 0.000 0.048 0.388
#> GSM198768 3 0.0376 0.905 0.000 0.004 0.992 0.004
#> GSM198770 3 0.2845 0.853 0.000 0.028 0.896 0.076
#> GSM198771 3 0.0336 0.906 0.000 0.000 0.992 0.008
#> GSM198774 3 0.0707 0.903 0.020 0.000 0.980 0.000
#> GSM198775 3 0.4877 0.344 0.000 0.000 0.592 0.408
#> GSM198777 3 0.0188 0.906 0.000 0.000 0.996 0.004
#> GSM198779 3 0.0336 0.906 0.000 0.000 0.992 0.008
#> GSM587218 4 0.0000 0.987 0.000 0.000 0.000 1.000
#> GSM587219 4 0.0000 0.987 0.000 0.000 0.000 1.000
#> GSM587220 4 0.0336 0.985 0.008 0.000 0.000 0.992
#> GSM587221 4 0.0000 0.987 0.000 0.000 0.000 1.000
#> GSM587222 4 0.0188 0.986 0.004 0.000 0.000 0.996
#> GSM587223 4 0.0000 0.987 0.000 0.000 0.000 1.000
#> GSM587224 4 0.0000 0.987 0.000 0.000 0.000 1.000
#> GSM587225 4 0.0336 0.985 0.008 0.000 0.000 0.992
#> GSM587226 4 0.0000 0.987 0.000 0.000 0.000 1.000
#> GSM587227 4 0.0188 0.986 0.004 0.000 0.000 0.996
#> GSM587228 4 0.0336 0.985 0.008 0.000 0.000 0.992
#> GSM587229 4 0.0469 0.982 0.012 0.000 0.000 0.988
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM587155 2 0.1732 0.905 0.000 0.920 0.000 0.000 0.080
#> GSM587156 2 0.3838 0.683 0.000 0.716 0.000 0.004 0.280
#> GSM587157 2 0.2677 0.863 0.000 0.872 0.112 0.000 0.016
#> GSM587158 2 0.0000 0.943 0.000 1.000 0.000 0.000 0.000
#> GSM587159 2 0.0000 0.943 0.000 1.000 0.000 0.000 0.000
#> GSM587160 2 0.0290 0.942 0.000 0.992 0.000 0.000 0.008
#> GSM587161 2 0.1197 0.925 0.000 0.952 0.000 0.000 0.048
#> GSM587162 2 0.0290 0.942 0.000 0.992 0.000 0.000 0.008
#> GSM587163 2 0.0290 0.942 0.000 0.992 0.000 0.000 0.008
#> GSM587164 2 0.1908 0.897 0.000 0.908 0.000 0.000 0.092
#> GSM587165 2 0.0404 0.940 0.000 0.988 0.012 0.000 0.000
#> GSM587166 2 0.3990 0.638 0.000 0.688 0.000 0.004 0.308
#> GSM587167 2 0.2127 0.886 0.000 0.892 0.000 0.000 0.108
#> GSM587168 2 0.0404 0.940 0.000 0.988 0.012 0.000 0.000
#> GSM587169 2 0.0290 0.942 0.000 0.992 0.000 0.000 0.008
#> GSM587170 2 0.2605 0.850 0.000 0.852 0.000 0.000 0.148
#> GSM587171 2 0.0000 0.943 0.000 1.000 0.000 0.000 0.000
#> GSM587172 2 0.0162 0.943 0.000 0.996 0.000 0.000 0.004
#> GSM587173 2 0.0404 0.940 0.000 0.988 0.012 0.000 0.000
#> GSM587174 2 0.0290 0.942 0.000 0.992 0.000 0.000 0.008
#> GSM587175 2 0.0912 0.938 0.000 0.972 0.016 0.000 0.012
#> GSM587176 2 0.0290 0.942 0.000 0.992 0.000 0.000 0.008
#> GSM587177 2 0.0290 0.942 0.000 0.992 0.008 0.000 0.000
#> GSM587178 2 0.0162 0.942 0.000 0.996 0.004 0.000 0.000
#> GSM587179 2 0.0162 0.942 0.000 0.996 0.004 0.000 0.000
#> GSM587180 2 0.0290 0.942 0.000 0.992 0.000 0.000 0.008
#> GSM587181 2 0.0162 0.942 0.000 0.996 0.004 0.000 0.000
#> GSM587182 2 0.0324 0.943 0.000 0.992 0.004 0.000 0.004
#> GSM587183 2 0.0162 0.942 0.000 0.996 0.004 0.000 0.000
#> GSM587184 2 0.0162 0.942 0.000 0.996 0.004 0.000 0.000
#> GSM587185 2 0.0290 0.942 0.000 0.992 0.000 0.000 0.008
#> GSM587186 2 0.0671 0.938 0.000 0.980 0.016 0.000 0.004
#> GSM587187 2 0.1965 0.880 0.000 0.904 0.096 0.000 0.000
#> GSM587188 2 0.2964 0.839 0.000 0.856 0.120 0.000 0.024
#> GSM587189 2 0.4126 0.403 0.000 0.620 0.380 0.000 0.000
#> GSM587190 4 0.3509 0.755 0.000 0.132 0.020 0.832 0.016
#> GSM587203 1 0.0000 0.966 1.000 0.000 0.000 0.000 0.000
#> GSM587204 1 0.0609 0.962 0.980 0.000 0.000 0.000 0.020
#> GSM587205 1 0.0000 0.966 1.000 0.000 0.000 0.000 0.000
#> GSM587206 1 0.0290 0.966 0.992 0.000 0.000 0.000 0.008
#> GSM587207 1 0.0000 0.966 1.000 0.000 0.000 0.000 0.000
#> GSM587208 1 0.0290 0.966 0.992 0.000 0.000 0.000 0.008
#> GSM587209 1 0.0324 0.965 0.992 0.000 0.004 0.000 0.004
#> GSM587210 5 0.4535 0.606 0.024 0.000 0.288 0.004 0.684
#> GSM587211 1 0.2228 0.908 0.912 0.000 0.048 0.000 0.040
#> GSM587212 5 0.4904 0.722 0.176 0.000 0.080 0.012 0.732
#> GSM587213 1 0.0162 0.966 0.996 0.000 0.000 0.000 0.004
#> GSM587214 1 0.0404 0.965 0.988 0.000 0.000 0.000 0.012
#> GSM587215 1 0.1106 0.956 0.964 0.000 0.012 0.000 0.024
#> GSM587216 1 0.3730 0.573 0.712 0.000 0.000 0.000 0.288
#> GSM587217 1 0.0880 0.958 0.968 0.000 0.000 0.000 0.032
#> GSM587191 3 0.3934 0.732 0.000 0.016 0.740 0.000 0.244
#> GSM587192 5 0.2624 0.791 0.000 0.000 0.116 0.012 0.872
#> GSM587193 5 0.2260 0.793 0.028 0.000 0.000 0.064 0.908
#> GSM587194 5 0.1710 0.802 0.000 0.004 0.016 0.040 0.940
#> GSM587195 3 0.0693 0.864 0.000 0.008 0.980 0.000 0.012
#> GSM587196 3 0.0510 0.866 0.000 0.000 0.984 0.000 0.016
#> GSM587197 3 0.0609 0.860 0.000 0.020 0.980 0.000 0.000
#> GSM587198 3 0.0880 0.869 0.000 0.000 0.968 0.000 0.032
#> GSM587199 3 0.3210 0.770 0.000 0.000 0.788 0.000 0.212
#> GSM587200 3 0.3586 0.711 0.000 0.000 0.736 0.000 0.264
#> GSM587201 3 0.4199 0.759 0.068 0.000 0.772 0.000 0.160
#> GSM587202 3 0.0880 0.869 0.000 0.000 0.968 0.000 0.032
#> GSM198767 1 0.0000 0.966 1.000 0.000 0.000 0.000 0.000
#> GSM198769 1 0.0162 0.966 0.996 0.000 0.000 0.000 0.004
#> GSM198772 1 0.1753 0.929 0.936 0.000 0.032 0.000 0.032
#> GSM198773 1 0.0162 0.966 0.996 0.000 0.000 0.000 0.004
#> GSM198776 1 0.0609 0.962 0.980 0.000 0.000 0.000 0.020
#> GSM198778 5 0.4445 0.583 0.024 0.000 0.300 0.000 0.676
#> GSM198780 5 0.4903 0.731 0.164 0.000 0.088 0.012 0.736
#> GSM198781 1 0.0290 0.966 0.992 0.000 0.000 0.000 0.008
#> GSM198765 3 0.3949 0.666 0.000 0.004 0.696 0.000 0.300
#> GSM198766 5 0.2278 0.794 0.032 0.000 0.000 0.060 0.908
#> GSM198768 3 0.0404 0.867 0.000 0.000 0.988 0.000 0.012
#> GSM198770 3 0.0794 0.852 0.000 0.028 0.972 0.000 0.000
#> GSM198771 3 0.0880 0.869 0.000 0.000 0.968 0.000 0.032
#> GSM198774 5 0.2624 0.791 0.000 0.000 0.116 0.012 0.872
#> GSM198775 5 0.1710 0.802 0.000 0.004 0.016 0.040 0.940
#> GSM198777 3 0.0510 0.866 0.000 0.000 0.984 0.000 0.016
#> GSM198779 3 0.3074 0.786 0.000 0.000 0.804 0.000 0.196
#> GSM587218 4 0.0000 0.981 0.000 0.000 0.000 1.000 0.000
#> GSM587219 4 0.0000 0.981 0.000 0.000 0.000 1.000 0.000
#> GSM587220 4 0.0000 0.981 0.000 0.000 0.000 1.000 0.000
#> GSM587221 4 0.0000 0.981 0.000 0.000 0.000 1.000 0.000
#> GSM587222 4 0.0000 0.981 0.000 0.000 0.000 1.000 0.000
#> GSM587223 4 0.0000 0.981 0.000 0.000 0.000 1.000 0.000
#> GSM587224 4 0.0000 0.981 0.000 0.000 0.000 1.000 0.000
#> GSM587225 4 0.0000 0.981 0.000 0.000 0.000 1.000 0.000
#> GSM587226 4 0.0000 0.981 0.000 0.000 0.000 1.000 0.000
#> GSM587227 4 0.0000 0.981 0.000 0.000 0.000 1.000 0.000
#> GSM587228 4 0.0000 0.981 0.000 0.000 0.000 1.000 0.000
#> GSM587229 4 0.0000 0.981 0.000 0.000 0.000 1.000 0.000
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM587155 2 0.3428 0.469 0.000 0.696 0.000 0.000 0.000 0.304
#> GSM587156 6 0.4067 0.226 0.000 0.444 0.000 0.000 0.008 0.548
#> GSM587157 2 0.5410 0.229 0.000 0.576 0.248 0.000 0.000 0.176
#> GSM587158 2 0.0000 0.869 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587159 2 0.0000 0.869 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587160 2 0.0146 0.868 0.000 0.996 0.000 0.000 0.000 0.004
#> GSM587161 2 0.2416 0.727 0.000 0.844 0.000 0.000 0.000 0.156
#> GSM587162 2 0.0260 0.868 0.000 0.992 0.000 0.000 0.000 0.008
#> GSM587163 2 0.0260 0.868 0.000 0.992 0.000 0.000 0.000 0.008
#> GSM587164 2 0.3578 0.378 0.000 0.660 0.000 0.000 0.000 0.340
#> GSM587165 2 0.0632 0.862 0.000 0.976 0.000 0.000 0.000 0.024
#> GSM587166 6 0.4118 0.344 0.000 0.396 0.000 0.004 0.008 0.592
#> GSM587167 2 0.3737 0.215 0.000 0.608 0.000 0.000 0.000 0.392
#> GSM587168 2 0.0713 0.860 0.000 0.972 0.000 0.000 0.000 0.028
#> GSM587169 2 0.0260 0.868 0.000 0.992 0.000 0.000 0.000 0.008
#> GSM587170 2 0.3782 0.136 0.000 0.588 0.000 0.000 0.000 0.412
#> GSM587171 2 0.0000 0.869 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587172 2 0.0000 0.869 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587173 2 0.1082 0.850 0.000 0.956 0.004 0.000 0.000 0.040
#> GSM587174 2 0.0000 0.869 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587175 2 0.2826 0.736 0.000 0.844 0.028 0.000 0.000 0.128
#> GSM587176 2 0.0260 0.868 0.000 0.992 0.000 0.000 0.000 0.008
#> GSM587177 2 0.0547 0.864 0.000 0.980 0.000 0.000 0.000 0.020
#> GSM587178 2 0.0363 0.867 0.000 0.988 0.000 0.000 0.000 0.012
#> GSM587179 2 0.0260 0.868 0.000 0.992 0.000 0.000 0.000 0.008
#> GSM587180 2 0.0405 0.868 0.000 0.988 0.000 0.000 0.004 0.008
#> GSM587181 2 0.0000 0.869 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587182 2 0.0146 0.869 0.000 0.996 0.000 0.000 0.000 0.004
#> GSM587183 2 0.0632 0.862 0.000 0.976 0.000 0.000 0.000 0.024
#> GSM587184 2 0.0000 0.869 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587185 2 0.0363 0.866 0.000 0.988 0.000 0.000 0.000 0.012
#> GSM587186 2 0.1152 0.847 0.000 0.952 0.004 0.000 0.000 0.044
#> GSM587187 2 0.2905 0.743 0.000 0.852 0.084 0.000 0.000 0.064
#> GSM587188 2 0.4798 0.533 0.000 0.708 0.168 0.012 0.004 0.108
#> GSM587189 2 0.4330 0.503 0.000 0.696 0.236 0.000 0.000 0.068
#> GSM587190 4 0.4424 0.684 0.000 0.124 0.076 0.764 0.004 0.032
#> GSM587203 1 0.0000 0.842 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM587204 1 0.0458 0.838 0.984 0.000 0.000 0.000 0.016 0.000
#> GSM587205 1 0.0000 0.842 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM587206 1 0.0000 0.842 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM587207 1 0.0000 0.842 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM587208 1 0.0000 0.842 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM587209 1 0.3535 0.822 0.832 0.000 0.040 0.000 0.060 0.068
#> GSM587210 5 0.2971 0.663 0.028 0.000 0.076 0.000 0.864 0.032
#> GSM587211 1 0.7649 0.339 0.360 0.000 0.144 0.012 0.180 0.304
#> GSM587212 5 0.4813 0.554 0.052 0.000 0.052 0.012 0.740 0.144
#> GSM587213 1 0.4324 0.782 0.748 0.000 0.020 0.000 0.068 0.164
#> GSM587214 1 0.2231 0.837 0.900 0.000 0.004 0.000 0.068 0.028
#> GSM587215 1 0.5197 0.741 0.696 0.000 0.056 0.000 0.120 0.128
#> GSM587216 1 0.4884 0.271 0.488 0.000 0.004 0.000 0.460 0.048
#> GSM587217 1 0.3782 0.788 0.788 0.000 0.008 0.000 0.140 0.064
#> GSM587191 3 0.5439 0.325 0.000 0.000 0.472 0.000 0.408 0.120
#> GSM587192 5 0.2669 0.665 0.000 0.000 0.008 0.000 0.836 0.156
#> GSM587193 6 0.4625 -0.126 0.000 0.004 0.000 0.032 0.424 0.540
#> GSM587194 5 0.3240 0.580 0.000 0.000 0.000 0.004 0.752 0.244
#> GSM587195 3 0.2163 0.680 0.000 0.000 0.892 0.000 0.016 0.092
#> GSM587196 3 0.1498 0.695 0.000 0.000 0.940 0.000 0.028 0.032
#> GSM587197 3 0.1409 0.707 0.000 0.008 0.948 0.000 0.012 0.032
#> GSM587198 3 0.3325 0.711 0.000 0.000 0.820 0.000 0.096 0.084
#> GSM587199 3 0.5368 0.434 0.000 0.000 0.488 0.000 0.400 0.112
#> GSM587200 3 0.5585 0.367 0.000 0.000 0.444 0.000 0.416 0.140
#> GSM587201 3 0.6320 0.470 0.032 0.000 0.472 0.000 0.324 0.172
#> GSM587202 3 0.3274 0.711 0.000 0.000 0.824 0.000 0.080 0.096
#> GSM198767 1 0.0000 0.842 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM198769 1 0.2941 0.833 0.868 0.000 0.024 0.000 0.060 0.048
#> GSM198772 1 0.7389 0.401 0.400 0.000 0.112 0.008 0.188 0.292
#> GSM198773 1 0.3930 0.802 0.784 0.000 0.016 0.000 0.064 0.136
#> GSM198776 1 0.0260 0.841 0.992 0.000 0.000 0.000 0.008 0.000
#> GSM198778 5 0.2818 0.662 0.028 0.000 0.076 0.000 0.872 0.024
#> GSM198780 5 0.4812 0.555 0.048 0.000 0.056 0.012 0.740 0.144
#> GSM198781 1 0.2307 0.837 0.896 0.000 0.004 0.000 0.068 0.032
#> GSM198765 5 0.5387 -0.351 0.000 0.000 0.424 0.000 0.464 0.112
#> GSM198766 6 0.4520 -0.178 0.000 0.000 0.000 0.032 0.448 0.520
#> GSM198768 3 0.2301 0.675 0.000 0.000 0.884 0.000 0.020 0.096
#> GSM198770 3 0.1478 0.702 0.000 0.020 0.944 0.000 0.004 0.032
#> GSM198771 3 0.3563 0.706 0.000 0.000 0.800 0.000 0.108 0.092
#> GSM198774 5 0.2706 0.664 0.000 0.000 0.008 0.000 0.832 0.160
#> GSM198775 5 0.3265 0.576 0.000 0.000 0.000 0.004 0.748 0.248
#> GSM198777 3 0.1498 0.695 0.000 0.000 0.940 0.000 0.028 0.032
#> GSM198779 3 0.5374 0.462 0.000 0.000 0.504 0.000 0.380 0.116
#> GSM587218 4 0.0260 0.969 0.000 0.000 0.000 0.992 0.000 0.008
#> GSM587219 4 0.0146 0.971 0.000 0.000 0.000 0.996 0.000 0.004
#> GSM587220 4 0.0000 0.973 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM587221 4 0.0000 0.973 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM587222 4 0.0146 0.973 0.000 0.000 0.000 0.996 0.000 0.004
#> GSM587223 4 0.0000 0.973 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM587224 4 0.0000 0.973 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM587225 4 0.0363 0.970 0.000 0.000 0.000 0.988 0.000 0.012
#> GSM587226 4 0.0146 0.973 0.000 0.000 0.000 0.996 0.000 0.004
#> GSM587227 4 0.0363 0.970 0.000 0.000 0.000 0.988 0.000 0.012
#> GSM587228 4 0.0260 0.972 0.000 0.000 0.000 0.992 0.000 0.008
#> GSM587229 4 0.0363 0.970 0.000 0.000 0.000 0.988 0.000 0.012
Heatmaps for the consensus matrix. It visualizes the probability of two samples to be in a same group.
consensus_heatmap(res, k = 2)
consensus_heatmap(res, k = 3)
consensus_heatmap(res, k = 4)
consensus_heatmap(res, k = 5)
consensus_heatmap(res, k = 6)
Heatmaps for the membership of samples in all partitions to see how consistent they are:
membership_heatmap(res, k = 2)
membership_heatmap(res, k = 3)
membership_heatmap(res, k = 4)
membership_heatmap(res, k = 5)
membership_heatmap(res, k = 6)
As soon as we have had the classes for columns, we can look for signatures which are significantly different between classes which can be candidate marks for certain classes. Following are the heatmaps for signatures.
Signature heatmaps where rows are scaled:
get_signatures(res, k = 2)
get_signatures(res, k = 3)
get_signatures(res, k = 4)
get_signatures(res, k = 5)
get_signatures(res, k = 6)
Signature heatmaps where rows are not scaled:
get_signatures(res, k = 2, scale_rows = FALSE)
get_signatures(res, k = 3, scale_rows = FALSE)
get_signatures(res, k = 4, scale_rows = FALSE)
get_signatures(res, k = 5, scale_rows = FALSE)
get_signatures(res, k = 6, scale_rows = FALSE)
Compare the overlap of signatures from different k:
compare_signatures(res)
get_signature() returns a data frame invisibly. TO get the list of signatures, the function
call should be assigned to a variable explicitly. In following code, if plot argument is set
to FALSE, no heatmap is plotted while only the differential analysis is performed.
# code only for demonstration
tb = get_signature(res, k = ..., plot = FALSE)
An example of the output of tb is:
#> which_row fdr mean_1 mean_2 scaled_mean_1 scaled_mean_2 km
#> 1 38 0.042760348 8.373488 9.131774 -0.5533452 0.5164555 1
#> 2 40 0.018707592 7.106213 8.469186 -0.6173731 0.5762149 1
#> 3 55 0.019134737 10.221463 11.207825 -0.6159697 0.5749050 1
#> 4 59 0.006059896 5.921854 7.869574 -0.6899429 0.6439467 1
#> 5 60 0.018055526 8.928898 10.211722 -0.6204761 0.5791110 1
#> 6 98 0.009384629 15.714769 14.887706 0.6635654 -0.6193277 2
...
The columns in tb are:
which_row: row indices corresponding to the input matrix.fdr: FDR for the differential test. mean_x: The mean value in group x.scaled_mean_x: The mean value in group x after rows are scaled.km: Row groups if k-means clustering is applied to rows.UMAP plot which shows how samples are separated.
dimension_reduction(res, k = 2, method = "UMAP")
dimension_reduction(res, k = 3, method = "UMAP")
dimension_reduction(res, k = 4, method = "UMAP")
dimension_reduction(res, k = 5, method = "UMAP")
dimension_reduction(res, k = 6, method = "UMAP")
Following heatmap shows how subgroups are split when increasing k:
collect_classes(res)
Test correlation between subgroups and known annotations. If the known annotation is numeric, one-way ANOVA test is applied, and if the known annotation is discrete, chi-squared contingency table test is applied.
test_to_known_factors(res)
#> n specimen(p) k
#> MAD:NMF 92 4.01e-14 2
#> MAD:NMF 89 3.87e-26 3
#> MAD:NMF 88 1.85e-38 4
#> MAD:NMF 91 3.78e-39 5
#> MAD:NMF 74 2.34e-29 6
If matrix rows can be associated to genes, consider to use functional_enrichment(res,
...) to perform function enrichment for the signature genes. See this vignette for more detailed explanations.
The object with results only for a single top-value method and a single partition method can be extracted as:
res = res_list["ATC", "hclust"]
# you can also extract it by
# res = res_list["ATC:hclust"]
A summary of res and all the functions that can be applied to it:
res
#> A 'ConsensusPartition' object with k = 2, 3, 4, 5, 6.
#> On a matrix with 51941 rows and 92 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 6.
#>
#> 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 0.912 0.947 0.965 0.4663 0.541 0.541
#> 3 3 0.827 0.980 0.977 0.4072 0.791 0.614
#> 4 4 0.826 0.946 0.926 0.0522 0.980 0.940
#> 5 5 1.000 0.982 0.990 0.0927 0.934 0.789
#> 6 6 0.983 0.931 0.960 0.0203 0.987 0.948
suggest_best_k() suggests the best \(k\) based on these statistics. The rules are as follows:
suggest_best_k(res)
#> [1] 6
#> attr(,"optional")
#> [1] 2 5
There is also optional best \(k\) = 2 5 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
#> GSM587155 2 0.000 1.000 0.000 1.000
#> GSM587156 2 0.000 1.000 0.000 1.000
#> GSM587157 2 0.000 1.000 0.000 1.000
#> GSM587158 2 0.000 1.000 0.000 1.000
#> GSM587159 2 0.000 1.000 0.000 1.000
#> GSM587160 2 0.000 1.000 0.000 1.000
#> GSM587161 2 0.000 1.000 0.000 1.000
#> GSM587162 2 0.000 1.000 0.000 1.000
#> GSM587163 2 0.000 1.000 0.000 1.000
#> GSM587164 2 0.000 1.000 0.000 1.000
#> GSM587165 2 0.000 1.000 0.000 1.000
#> GSM587166 2 0.000 1.000 0.000 1.000
#> GSM587167 2 0.000 1.000 0.000 1.000
#> GSM587168 2 0.000 1.000 0.000 1.000
#> GSM587169 2 0.000 1.000 0.000 1.000
#> GSM587170 2 0.000 1.000 0.000 1.000
#> GSM587171 2 0.000 1.000 0.000 1.000
#> GSM587172 2 0.000 1.000 0.000 1.000
#> GSM587173 2 0.000 1.000 0.000 1.000
#> GSM587174 2 0.000 1.000 0.000 1.000
#> GSM587175 2 0.000 1.000 0.000 1.000
#> GSM587176 2 0.000 1.000 0.000 1.000
#> GSM587177 2 0.000 1.000 0.000 1.000
#> GSM587178 2 0.000 1.000 0.000 1.000
#> GSM587179 2 0.000 1.000 0.000 1.000
#> GSM587180 2 0.000 1.000 0.000 1.000
#> GSM587181 2 0.000 1.000 0.000 1.000
#> GSM587182 2 0.000 1.000 0.000 1.000
#> GSM587183 2 0.000 1.000 0.000 1.000
#> GSM587184 2 0.000 1.000 0.000 1.000
#> GSM587185 2 0.000 1.000 0.000 1.000
#> GSM587186 2 0.000 1.000 0.000 1.000
#> GSM587187 1 0.814 0.753 0.748 0.252
#> GSM587188 1 0.814 0.753 0.748 0.252
#> GSM587189 1 0.814 0.753 0.748 0.252
#> GSM587190 1 0.788 0.775 0.764 0.236
#> GSM587203 1 0.000 0.946 1.000 0.000
#> GSM587204 1 0.000 0.946 1.000 0.000
#> GSM587205 1 0.000 0.946 1.000 0.000
#> GSM587206 1 0.000 0.946 1.000 0.000
#> GSM587207 1 0.000 0.946 1.000 0.000
#> GSM587208 1 0.000 0.946 1.000 0.000
#> GSM587209 1 0.000 0.946 1.000 0.000
#> GSM587210 1 0.000 0.946 1.000 0.000
#> GSM587211 1 0.000 0.946 1.000 0.000
#> GSM587212 1 0.000 0.946 1.000 0.000
#> GSM587213 1 0.000 0.946 1.000 0.000
#> GSM587214 1 0.000 0.946 1.000 0.000
#> GSM587215 1 0.000 0.946 1.000 0.000
#> GSM587216 1 0.000 0.946 1.000 0.000
#> GSM587217 1 0.000 0.946 1.000 0.000
#> GSM587191 1 0.430 0.921 0.912 0.088
#> GSM587192 1 0.430 0.921 0.912 0.088
#> GSM587193 1 0.430 0.921 0.912 0.088
#> GSM587194 1 0.706 0.829 0.808 0.192
#> GSM587195 1 0.430 0.921 0.912 0.088
#> GSM587196 1 0.430 0.921 0.912 0.088
#> GSM587197 1 0.430 0.921 0.912 0.088
#> GSM587198 1 0.430 0.921 0.912 0.088
#> GSM587199 1 0.706 0.829 0.808 0.192
#> GSM587200 1 0.358 0.928 0.932 0.068
#> GSM587201 1 0.358 0.928 0.932 0.068
#> GSM587202 1 0.358 0.928 0.932 0.068
#> GSM198767 1 0.000 0.946 1.000 0.000
#> GSM198769 1 0.000 0.946 1.000 0.000
#> GSM198772 1 0.000 0.946 1.000 0.000
#> GSM198773 1 0.000 0.946 1.000 0.000
#> GSM198776 1 0.000 0.946 1.000 0.000
#> GSM198778 1 0.000 0.946 1.000 0.000
#> GSM198780 1 0.000 0.946 1.000 0.000
#> GSM198781 1 0.000 0.946 1.000 0.000
#> GSM198765 1 0.430 0.921 0.912 0.088
#> GSM198766 1 0.430 0.921 0.912 0.088
#> GSM198768 1 0.430 0.921 0.912 0.088
#> GSM198770 1 0.430 0.921 0.912 0.088
#> GSM198771 1 0.430 0.921 0.912 0.088
#> GSM198774 1 0.430 0.921 0.912 0.088
#> GSM198775 1 0.706 0.829 0.808 0.192
#> GSM198777 1 0.430 0.921 0.912 0.088
#> GSM198779 1 0.706 0.829 0.808 0.192
#> GSM587218 1 0.000 0.946 1.000 0.000
#> GSM587219 1 0.000 0.946 1.000 0.000
#> GSM587220 1 0.000 0.946 1.000 0.000
#> GSM587221 1 0.000 0.946 1.000 0.000
#> GSM587222 1 0.000 0.946 1.000 0.000
#> GSM587223 1 0.000 0.946 1.000 0.000
#> GSM587224 1 0.000 0.946 1.000 0.000
#> GSM587225 1 0.000 0.946 1.000 0.000
#> GSM587226 1 0.000 0.946 1.000 0.000
#> GSM587227 1 0.000 0.946 1.000 0.000
#> GSM587228 1 0.000 0.946 1.000 0.000
#> GSM587229 1 0.000 0.946 1.000 0.000
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM587155 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587156 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587157 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587158 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587159 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587160 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587161 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587162 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587163 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587164 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587165 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587166 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587167 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587168 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587169 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587170 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587171 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587172 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587173 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587174 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587175 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587176 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587177 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587178 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587179 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587180 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587181 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587182 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587183 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587184 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587185 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587186 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587187 3 0.2066 0.880 0.000 0.060 0.940
#> GSM587188 3 0.2066 0.880 0.000 0.060 0.940
#> GSM587189 3 0.2066 0.880 0.000 0.060 0.940
#> GSM587190 3 0.1643 0.890 0.000 0.044 0.956
#> GSM587203 1 0.0000 0.997 1.000 0.000 0.000
#> GSM587204 1 0.0747 0.985 0.984 0.000 0.016
#> GSM587205 1 0.0000 0.997 1.000 0.000 0.000
#> GSM587206 1 0.0000 0.997 1.000 0.000 0.000
#> GSM587207 1 0.0000 0.997 1.000 0.000 0.000
#> GSM587208 1 0.0000 0.997 1.000 0.000 0.000
#> GSM587209 1 0.0000 0.997 1.000 0.000 0.000
#> GSM587210 1 0.0747 0.985 0.984 0.000 0.016
#> GSM587211 1 0.0000 0.997 1.000 0.000 0.000
#> GSM587212 1 0.0747 0.985 0.984 0.000 0.016
#> GSM587213 1 0.0000 0.997 1.000 0.000 0.000
#> GSM587214 1 0.0000 0.997 1.000 0.000 0.000
#> GSM587215 1 0.0000 0.997 1.000 0.000 0.000
#> GSM587216 1 0.0000 0.997 1.000 0.000 0.000
#> GSM587217 1 0.0000 0.997 1.000 0.000 0.000
#> GSM587191 3 0.3038 0.953 0.104 0.000 0.896
#> GSM587192 3 0.3038 0.953 0.104 0.000 0.896
#> GSM587193 3 0.3038 0.953 0.104 0.000 0.896
#> GSM587194 3 0.0000 0.910 0.000 0.000 1.000
#> GSM587195 3 0.3038 0.953 0.104 0.000 0.896
#> GSM587196 3 0.3038 0.953 0.104 0.000 0.896
#> GSM587197 3 0.3038 0.953 0.104 0.000 0.896
#> GSM587198 3 0.3038 0.953 0.104 0.000 0.896
#> GSM587199 3 0.0000 0.910 0.000 0.000 1.000
#> GSM587200 3 0.3412 0.937 0.124 0.000 0.876
#> GSM587201 3 0.3412 0.937 0.124 0.000 0.876
#> GSM587202 3 0.3412 0.937 0.124 0.000 0.876
#> GSM198767 1 0.0000 0.997 1.000 0.000 0.000
#> GSM198769 1 0.0000 0.997 1.000 0.000 0.000
#> GSM198772 1 0.0000 0.997 1.000 0.000 0.000
#> GSM198773 1 0.0000 0.997 1.000 0.000 0.000
#> GSM198776 1 0.0747 0.985 0.984 0.000 0.016
#> GSM198778 1 0.0747 0.985 0.984 0.000 0.016
#> GSM198780 1 0.0747 0.985 0.984 0.000 0.016
#> GSM198781 1 0.0000 0.997 1.000 0.000 0.000
#> GSM198765 3 0.3038 0.953 0.104 0.000 0.896
#> GSM198766 3 0.3038 0.953 0.104 0.000 0.896
#> GSM198768 3 0.3038 0.953 0.104 0.000 0.896
#> GSM198770 3 0.3038 0.953 0.104 0.000 0.896
#> GSM198771 3 0.3038 0.953 0.104 0.000 0.896
#> GSM198774 3 0.3038 0.953 0.104 0.000 0.896
#> GSM198775 3 0.0000 0.910 0.000 0.000 1.000
#> GSM198777 3 0.3038 0.953 0.104 0.000 0.896
#> GSM198779 3 0.0000 0.910 0.000 0.000 1.000
#> GSM587218 1 0.0000 0.997 1.000 0.000 0.000
#> GSM587219 1 0.0000 0.997 1.000 0.000 0.000
#> GSM587220 1 0.0000 0.997 1.000 0.000 0.000
#> GSM587221 1 0.0000 0.997 1.000 0.000 0.000
#> GSM587222 1 0.0000 0.997 1.000 0.000 0.000
#> GSM587223 1 0.0000 0.997 1.000 0.000 0.000
#> GSM587224 1 0.0000 0.997 1.000 0.000 0.000
#> GSM587225 1 0.0000 0.997 1.000 0.000 0.000
#> GSM587226 1 0.0000 0.997 1.000 0.000 0.000
#> GSM587227 1 0.0000 0.997 1.000 0.000 0.000
#> GSM587228 1 0.0000 0.997 1.000 0.000 0.000
#> GSM587229 1 0.0000 0.997 1.000 0.000 0.000
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM587155 2 0.0000 1.000 0.000 1.00 0.000 0.000
#> GSM587156 2 0.0000 1.000 0.000 1.00 0.000 0.000
#> GSM587157 2 0.0000 1.000 0.000 1.00 0.000 0.000
#> GSM587158 2 0.0000 1.000 0.000 1.00 0.000 0.000
#> GSM587159 2 0.0000 1.000 0.000 1.00 0.000 0.000
#> GSM587160 2 0.0000 1.000 0.000 1.00 0.000 0.000
#> GSM587161 2 0.0000 1.000 0.000 1.00 0.000 0.000
#> GSM587162 2 0.0000 1.000 0.000 1.00 0.000 0.000
#> GSM587163 2 0.0000 1.000 0.000 1.00 0.000 0.000
#> GSM587164 2 0.0000 1.000 0.000 1.00 0.000 0.000
#> GSM587165 2 0.0000 1.000 0.000 1.00 0.000 0.000
#> GSM587166 2 0.0000 1.000 0.000 1.00 0.000 0.000
#> GSM587167 2 0.0000 1.000 0.000 1.00 0.000 0.000
#> GSM587168 2 0.0000 1.000 0.000 1.00 0.000 0.000
#> GSM587169 2 0.0000 1.000 0.000 1.00 0.000 0.000
#> GSM587170 2 0.0000 1.000 0.000 1.00 0.000 0.000
#> GSM587171 2 0.0000 1.000 0.000 1.00 0.000 0.000
#> GSM587172 2 0.0000 1.000 0.000 1.00 0.000 0.000
#> GSM587173 2 0.0000 1.000 0.000 1.00 0.000 0.000
#> GSM587174 2 0.0000 1.000 0.000 1.00 0.000 0.000
#> GSM587175 2 0.0000 1.000 0.000 1.00 0.000 0.000
#> GSM587176 2 0.0000 1.000 0.000 1.00 0.000 0.000
#> GSM587177 2 0.0000 1.000 0.000 1.00 0.000 0.000
#> GSM587178 2 0.0000 1.000 0.000 1.00 0.000 0.000
#> GSM587179 2 0.0000 1.000 0.000 1.00 0.000 0.000
#> GSM587180 2 0.0000 1.000 0.000 1.00 0.000 0.000
#> GSM587181 2 0.0000 1.000 0.000 1.00 0.000 0.000
#> GSM587182 2 0.0000 1.000 0.000 1.00 0.000 0.000
#> GSM587183 2 0.0000 1.000 0.000 1.00 0.000 0.000
#> GSM587184 2 0.0000 1.000 0.000 1.00 0.000 0.000
#> GSM587185 2 0.0000 1.000 0.000 1.00 0.000 0.000
#> GSM587186 2 0.0000 1.000 0.000 1.00 0.000 0.000
#> GSM587187 4 0.2751 0.994 0.000 0.04 0.056 0.904
#> GSM587188 4 0.2751 0.994 0.000 0.04 0.056 0.904
#> GSM587189 4 0.2751 0.994 0.000 0.04 0.056 0.904
#> GSM587190 4 0.3056 0.982 0.000 0.04 0.072 0.888
#> GSM587203 1 0.2408 0.922 0.896 0.00 0.104 0.000
#> GSM587204 1 0.2647 0.916 0.880 0.00 0.120 0.000
#> GSM587205 1 0.2408 0.922 0.896 0.00 0.104 0.000
#> GSM587206 1 0.2408 0.922 0.896 0.00 0.104 0.000
#> GSM587207 1 0.2408 0.922 0.896 0.00 0.104 0.000
#> GSM587208 1 0.2408 0.922 0.896 0.00 0.104 0.000
#> GSM587209 1 0.2408 0.922 0.896 0.00 0.104 0.000
#> GSM587210 1 0.2647 0.916 0.880 0.00 0.120 0.000
#> GSM587211 1 0.2408 0.922 0.896 0.00 0.104 0.000
#> GSM587212 1 0.2647 0.916 0.880 0.00 0.120 0.000
#> GSM587213 1 0.2408 0.922 0.896 0.00 0.104 0.000
#> GSM587214 1 0.2408 0.922 0.896 0.00 0.104 0.000
#> GSM587215 1 0.2408 0.922 0.896 0.00 0.104 0.000
#> GSM587216 1 0.2408 0.922 0.896 0.00 0.104 0.000
#> GSM587217 1 0.2408 0.922 0.896 0.00 0.104 0.000
#> GSM587191 3 0.0000 0.963 0.000 0.00 1.000 0.000
#> GSM587192 3 0.0000 0.963 0.000 0.00 1.000 0.000
#> GSM587193 3 0.0000 0.963 0.000 0.00 1.000 0.000
#> GSM587194 3 0.2469 0.847 0.000 0.00 0.892 0.108
#> GSM587195 3 0.0000 0.963 0.000 0.00 1.000 0.000
#> GSM587196 3 0.0000 0.963 0.000 0.00 1.000 0.000
#> GSM587197 3 0.0000 0.963 0.000 0.00 1.000 0.000
#> GSM587198 3 0.0000 0.963 0.000 0.00 1.000 0.000
#> GSM587199 3 0.2469 0.847 0.000 0.00 0.892 0.108
#> GSM587200 3 0.0707 0.943 0.020 0.00 0.980 0.000
#> GSM587201 3 0.0707 0.943 0.020 0.00 0.980 0.000
#> GSM587202 3 0.0707 0.943 0.020 0.00 0.980 0.000
#> GSM198767 1 0.2408 0.922 0.896 0.00 0.104 0.000
#> GSM198769 1 0.2408 0.922 0.896 0.00 0.104 0.000
#> GSM198772 1 0.2408 0.922 0.896 0.00 0.104 0.000
#> GSM198773 1 0.2408 0.922 0.896 0.00 0.104 0.000
#> GSM198776 1 0.2647 0.916 0.880 0.00 0.120 0.000
#> GSM198778 1 0.2647 0.916 0.880 0.00 0.120 0.000
#> GSM198780 1 0.2647 0.916 0.880 0.00 0.120 0.000
#> GSM198781 1 0.2408 0.922 0.896 0.00 0.104 0.000
#> GSM198765 3 0.0000 0.963 0.000 0.00 1.000 0.000
#> GSM198766 3 0.0000 0.963 0.000 0.00 1.000 0.000
#> GSM198768 3 0.0000 0.963 0.000 0.00 1.000 0.000
#> GSM198770 3 0.0000 0.963 0.000 0.00 1.000 0.000
#> GSM198771 3 0.0000 0.963 0.000 0.00 1.000 0.000
#> GSM198774 3 0.0000 0.963 0.000 0.00 1.000 0.000
#> GSM198775 3 0.2469 0.847 0.000 0.00 0.892 0.108
#> GSM198777 3 0.0000 0.963 0.000 0.00 1.000 0.000
#> GSM198779 3 0.2469 0.847 0.000 0.00 0.892 0.108
#> GSM587218 1 0.2281 0.848 0.904 0.00 0.000 0.096
#> GSM587219 1 0.2281 0.848 0.904 0.00 0.000 0.096
#> GSM587220 1 0.2281 0.848 0.904 0.00 0.000 0.096
#> GSM587221 1 0.2281 0.848 0.904 0.00 0.000 0.096
#> GSM587222 1 0.2281 0.848 0.904 0.00 0.000 0.096
#> GSM587223 1 0.2281 0.848 0.904 0.00 0.000 0.096
#> GSM587224 1 0.2281 0.848 0.904 0.00 0.000 0.096
#> GSM587225 1 0.2281 0.848 0.904 0.00 0.000 0.096
#> GSM587226 1 0.2281 0.848 0.904 0.00 0.000 0.096
#> GSM587227 1 0.2281 0.848 0.904 0.00 0.000 0.096
#> GSM587228 1 0.2281 0.848 0.904 0.00 0.000 0.096
#> GSM587229 1 0.2281 0.848 0.904 0.00 0.000 0.096
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM587155 2 0.0000 1.000 0.000 1 0.000 0.000 0.000
#> GSM587156 2 0.0000 1.000 0.000 1 0.000 0.000 0.000
#> GSM587157 2 0.0000 1.000 0.000 1 0.000 0.000 0.000
#> GSM587158 2 0.0000 1.000 0.000 1 0.000 0.000 0.000
#> GSM587159 2 0.0000 1.000 0.000 1 0.000 0.000 0.000
#> GSM587160 2 0.0000 1.000 0.000 1 0.000 0.000 0.000
#> GSM587161 2 0.0000 1.000 0.000 1 0.000 0.000 0.000
#> GSM587162 2 0.0000 1.000 0.000 1 0.000 0.000 0.000
#> GSM587163 2 0.0000 1.000 0.000 1 0.000 0.000 0.000
#> GSM587164 2 0.0000 1.000 0.000 1 0.000 0.000 0.000
#> GSM587165 2 0.0000 1.000 0.000 1 0.000 0.000 0.000
#> GSM587166 2 0.0000 1.000 0.000 1 0.000 0.000 0.000
#> GSM587167 2 0.0000 1.000 0.000 1 0.000 0.000 0.000
#> GSM587168 2 0.0000 1.000 0.000 1 0.000 0.000 0.000
#> GSM587169 2 0.0000 1.000 0.000 1 0.000 0.000 0.000
#> GSM587170 2 0.0000 1.000 0.000 1 0.000 0.000 0.000
#> GSM587171 2 0.0000 1.000 0.000 1 0.000 0.000 0.000
#> GSM587172 2 0.0000 1.000 0.000 1 0.000 0.000 0.000
#> GSM587173 2 0.0000 1.000 0.000 1 0.000 0.000 0.000
#> GSM587174 2 0.0000 1.000 0.000 1 0.000 0.000 0.000
#> GSM587175 2 0.0000 1.000 0.000 1 0.000 0.000 0.000
#> GSM587176 2 0.0000 1.000 0.000 1 0.000 0.000 0.000
#> GSM587177 2 0.0000 1.000 0.000 1 0.000 0.000 0.000
#> GSM587178 2 0.0000 1.000 0.000 1 0.000 0.000 0.000
#> GSM587179 2 0.0000 1.000 0.000 1 0.000 0.000 0.000
#> GSM587180 2 0.0000 1.000 0.000 1 0.000 0.000 0.000
#> GSM587181 2 0.0000 1.000 0.000 1 0.000 0.000 0.000
#> GSM587182 2 0.0000 1.000 0.000 1 0.000 0.000 0.000
#> GSM587183 2 0.0000 1.000 0.000 1 0.000 0.000 0.000
#> GSM587184 2 0.0000 1.000 0.000 1 0.000 0.000 0.000
#> GSM587185 2 0.0000 1.000 0.000 1 0.000 0.000 0.000
#> GSM587186 2 0.0000 1.000 0.000 1 0.000 0.000 0.000
#> GSM587187 5 0.0162 0.993 0.000 0 0.004 0.000 0.996
#> GSM587188 5 0.0162 0.993 0.000 0 0.004 0.000 0.996
#> GSM587189 5 0.0162 0.993 0.000 0 0.004 0.000 0.996
#> GSM587190 5 0.0609 0.980 0.000 0 0.020 0.000 0.980
#> GSM587203 1 0.0000 0.993 1.000 0 0.000 0.000 0.000
#> GSM587204 1 0.0671 0.982 0.980 0 0.016 0.004 0.000
#> GSM587205 1 0.0000 0.993 1.000 0 0.000 0.000 0.000
#> GSM587206 1 0.0000 0.993 1.000 0 0.000 0.000 0.000
#> GSM587207 1 0.0000 0.993 1.000 0 0.000 0.000 0.000
#> GSM587208 1 0.0000 0.993 1.000 0 0.000 0.000 0.000
#> GSM587209 1 0.0000 0.993 1.000 0 0.000 0.000 0.000
#> GSM587210 1 0.0671 0.982 0.980 0 0.016 0.004 0.000
#> GSM587211 1 0.0000 0.993 1.000 0 0.000 0.000 0.000
#> GSM587212 1 0.0671 0.982 0.980 0 0.016 0.004 0.000
#> GSM587213 1 0.0000 0.993 1.000 0 0.000 0.000 0.000
#> GSM587214 1 0.0000 0.993 1.000 0 0.000 0.000 0.000
#> GSM587215 1 0.0000 0.993 1.000 0 0.000 0.000 0.000
#> GSM587216 1 0.0162 0.991 0.996 0 0.000 0.004 0.000
#> GSM587217 1 0.0162 0.991 0.996 0 0.000 0.004 0.000
#> GSM587191 3 0.0000 0.973 0.000 0 1.000 0.000 0.000
#> GSM587192 3 0.0000 0.973 0.000 0 1.000 0.000 0.000
#> GSM587193 3 0.0000 0.973 0.000 0 1.000 0.000 0.000
#> GSM587194 3 0.2179 0.892 0.000 0 0.888 0.000 0.112
#> GSM587195 3 0.0000 0.973 0.000 0 1.000 0.000 0.000
#> GSM587196 3 0.0000 0.973 0.000 0 1.000 0.000 0.000
#> GSM587197 3 0.0000 0.973 0.000 0 1.000 0.000 0.000
#> GSM587198 3 0.0000 0.973 0.000 0 1.000 0.000 0.000
#> GSM587199 3 0.2179 0.892 0.000 0 0.888 0.000 0.112
#> GSM587200 3 0.0671 0.959 0.016 0 0.980 0.000 0.004
#> GSM587201 3 0.0671 0.959 0.016 0 0.980 0.000 0.004
#> GSM587202 3 0.0671 0.959 0.016 0 0.980 0.000 0.004
#> GSM198767 1 0.0000 0.993 1.000 0 0.000 0.000 0.000
#> GSM198769 1 0.0000 0.993 1.000 0 0.000 0.000 0.000
#> GSM198772 1 0.0000 0.993 1.000 0 0.000 0.000 0.000
#> GSM198773 1 0.0000 0.993 1.000 0 0.000 0.000 0.000
#> GSM198776 1 0.0671 0.982 0.980 0 0.016 0.004 0.000
#> GSM198778 1 0.0671 0.982 0.980 0 0.016 0.004 0.000
#> GSM198780 1 0.0671 0.982 0.980 0 0.016 0.004 0.000
#> GSM198781 1 0.0000 0.993 1.000 0 0.000 0.000 0.000
#> GSM198765 3 0.0000 0.973 0.000 0 1.000 0.000 0.000
#> GSM198766 3 0.0000 0.973 0.000 0 1.000 0.000 0.000
#> GSM198768 3 0.0000 0.973 0.000 0 1.000 0.000 0.000
#> GSM198770 3 0.0000 0.973 0.000 0 1.000 0.000 0.000
#> GSM198771 3 0.0000 0.973 0.000 0 1.000 0.000 0.000
#> GSM198774 3 0.0000 0.973 0.000 0 1.000 0.000 0.000
#> GSM198775 3 0.2179 0.892 0.000 0 0.888 0.000 0.112
#> GSM198777 3 0.0000 0.973 0.000 0 1.000 0.000 0.000
#> GSM198779 3 0.2179 0.892 0.000 0 0.888 0.000 0.112
#> GSM587218 4 0.0000 0.970 0.000 0 0.000 1.000 0.000
#> GSM587219 4 0.0000 0.970 0.000 0 0.000 1.000 0.000
#> GSM587220 4 0.0000 0.970 0.000 0 0.000 1.000 0.000
#> GSM587221 4 0.0000 0.970 0.000 0 0.000 1.000 0.000
#> GSM587222 4 0.0000 0.970 0.000 0 0.000 1.000 0.000
#> GSM587223 4 0.0000 0.970 0.000 0 0.000 1.000 0.000
#> GSM587224 4 0.0000 0.970 0.000 0 0.000 1.000 0.000
#> GSM587225 4 0.1671 0.907 0.076 0 0.000 0.924 0.000
#> GSM587226 4 0.0000 0.970 0.000 0 0.000 1.000 0.000
#> GSM587227 4 0.1197 0.943 0.048 0 0.000 0.952 0.000
#> GSM587228 4 0.1197 0.943 0.048 0 0.000 0.952 0.000
#> GSM587229 4 0.1197 0.943 0.048 0 0.000 0.952 0.000
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM587155 2 0.0000 0.995 0.000 1.00 0.000 0.000 0.000 0.000
#> GSM587156 2 0.0000 0.995 0.000 1.00 0.000 0.000 0.000 0.000
#> GSM587157 2 0.0000 0.995 0.000 1.00 0.000 0.000 0.000 0.000
#> GSM587158 2 0.0000 0.995 0.000 1.00 0.000 0.000 0.000 0.000
#> GSM587159 2 0.0000 0.995 0.000 1.00 0.000 0.000 0.000 0.000
#> GSM587160 2 0.0000 0.995 0.000 1.00 0.000 0.000 0.000 0.000
#> GSM587161 2 0.0000 0.995 0.000 1.00 0.000 0.000 0.000 0.000
#> GSM587162 2 0.0000 0.995 0.000 1.00 0.000 0.000 0.000 0.000
#> GSM587163 2 0.0000 0.995 0.000 1.00 0.000 0.000 0.000 0.000
#> GSM587164 2 0.0000 0.995 0.000 1.00 0.000 0.000 0.000 0.000
#> GSM587165 2 0.0000 0.995 0.000 1.00 0.000 0.000 0.000 0.000
#> GSM587166 2 0.0000 0.995 0.000 1.00 0.000 0.000 0.000 0.000
#> GSM587167 2 0.0000 0.995 0.000 1.00 0.000 0.000 0.000 0.000
#> GSM587168 2 0.0000 0.995 0.000 1.00 0.000 0.000 0.000 0.000
#> GSM587169 2 0.0000 0.995 0.000 1.00 0.000 0.000 0.000 0.000
#> GSM587170 2 0.0000 0.995 0.000 1.00 0.000 0.000 0.000 0.000
#> GSM587171 2 0.0000 0.995 0.000 1.00 0.000 0.000 0.000 0.000
#> GSM587172 2 0.0000 0.995 0.000 1.00 0.000 0.000 0.000 0.000
#> GSM587173 2 0.0000 0.995 0.000 1.00 0.000 0.000 0.000 0.000
#> GSM587174 2 0.0000 0.995 0.000 1.00 0.000 0.000 0.000 0.000
#> GSM587175 2 0.0000 0.995 0.000 1.00 0.000 0.000 0.000 0.000
#> GSM587176 2 0.0000 0.995 0.000 1.00 0.000 0.000 0.000 0.000
#> GSM587177 2 0.0000 0.995 0.000 1.00 0.000 0.000 0.000 0.000
#> GSM587178 2 0.0000 0.995 0.000 1.00 0.000 0.000 0.000 0.000
#> GSM587179 2 0.0000 0.995 0.000 1.00 0.000 0.000 0.000 0.000
#> GSM587180 2 0.0000 0.995 0.000 1.00 0.000 0.000 0.000 0.000
#> GSM587181 2 0.0000 0.995 0.000 1.00 0.000 0.000 0.000 0.000
#> GSM587182 2 0.0000 0.995 0.000 1.00 0.000 0.000 0.000 0.000
#> GSM587183 2 0.0937 0.965 0.000 0.96 0.000 0.000 0.000 0.040
#> GSM587184 2 0.0937 0.965 0.000 0.96 0.000 0.000 0.000 0.040
#> GSM587185 2 0.0937 0.965 0.000 0.96 0.000 0.000 0.000 0.040
#> GSM587186 2 0.0937 0.965 0.000 0.96 0.000 0.000 0.000 0.040
#> GSM587187 5 0.0000 0.992 0.000 0.00 0.000 0.000 1.000 0.000
#> GSM587188 5 0.0000 0.992 0.000 0.00 0.000 0.000 1.000 0.000
#> GSM587189 5 0.0000 0.992 0.000 0.00 0.000 0.000 1.000 0.000
#> GSM587190 5 0.0508 0.975 0.000 0.00 0.012 0.000 0.984 0.004
#> GSM587203 1 0.0000 0.994 1.000 0.00 0.000 0.000 0.000 0.000
#> GSM587204 1 0.0603 0.985 0.980 0.00 0.016 0.004 0.000 0.000
#> GSM587205 1 0.0000 0.994 1.000 0.00 0.000 0.000 0.000 0.000
#> GSM587206 1 0.0000 0.994 1.000 0.00 0.000 0.000 0.000 0.000
#> GSM587207 1 0.0000 0.994 1.000 0.00 0.000 0.000 0.000 0.000
#> GSM587208 1 0.0000 0.994 1.000 0.00 0.000 0.000 0.000 0.000
#> GSM587209 1 0.0000 0.994 1.000 0.00 0.000 0.000 0.000 0.000
#> GSM587210 1 0.0603 0.985 0.980 0.00 0.016 0.004 0.000 0.000
#> GSM587211 1 0.0000 0.994 1.000 0.00 0.000 0.000 0.000 0.000
#> GSM587212 1 0.0603 0.985 0.980 0.00 0.016 0.004 0.000 0.000
#> GSM587213 1 0.0000 0.994 1.000 0.00 0.000 0.000 0.000 0.000
#> GSM587214 1 0.0000 0.994 1.000 0.00 0.000 0.000 0.000 0.000
#> GSM587215 1 0.0000 0.994 1.000 0.00 0.000 0.000 0.000 0.000
#> GSM587216 1 0.0146 0.992 0.996 0.00 0.000 0.004 0.000 0.000
#> GSM587217 1 0.0146 0.992 0.996 0.00 0.000 0.004 0.000 0.000
#> GSM587191 3 0.3515 0.812 0.000 0.00 0.676 0.000 0.000 0.324
#> GSM587192 3 0.3515 0.812 0.000 0.00 0.676 0.000 0.000 0.324
#> GSM587193 3 0.3515 0.812 0.000 0.00 0.676 0.000 0.000 0.324
#> GSM587194 3 0.1910 0.401 0.000 0.00 0.892 0.000 0.108 0.000
#> GSM587195 3 0.3515 0.812 0.000 0.00 0.676 0.000 0.000 0.324
#> GSM587196 3 0.3515 0.812 0.000 0.00 0.676 0.000 0.000 0.324
#> GSM587197 3 0.3515 0.812 0.000 0.00 0.676 0.000 0.000 0.324
#> GSM587198 3 0.3828 0.641 0.000 0.00 0.560 0.000 0.000 0.440
#> GSM587199 3 0.1910 0.401 0.000 0.00 0.892 0.000 0.108 0.000
#> GSM587200 6 0.0937 1.000 0.000 0.00 0.040 0.000 0.000 0.960
#> GSM587201 6 0.0937 1.000 0.000 0.00 0.040 0.000 0.000 0.960
#> GSM587202 6 0.0937 1.000 0.000 0.00 0.040 0.000 0.000 0.960
#> GSM198767 1 0.0000 0.994 1.000 0.00 0.000 0.000 0.000 0.000
#> GSM198769 1 0.0000 0.994 1.000 0.00 0.000 0.000 0.000 0.000
#> GSM198772 1 0.0000 0.994 1.000 0.00 0.000 0.000 0.000 0.000
#> GSM198773 1 0.0000 0.994 1.000 0.00 0.000 0.000 0.000 0.000
#> GSM198776 1 0.0603 0.985 0.980 0.00 0.016 0.004 0.000 0.000
#> GSM198778 1 0.0603 0.985 0.980 0.00 0.016 0.004 0.000 0.000
#> GSM198780 1 0.0603 0.985 0.980 0.00 0.016 0.004 0.000 0.000
#> GSM198781 1 0.0000 0.994 1.000 0.00 0.000 0.000 0.000 0.000
#> GSM198765 3 0.3515 0.812 0.000 0.00 0.676 0.000 0.000 0.324
#> GSM198766 3 0.3515 0.812 0.000 0.00 0.676 0.000 0.000 0.324
#> GSM198768 3 0.3515 0.812 0.000 0.00 0.676 0.000 0.000 0.324
#> GSM198770 3 0.3515 0.812 0.000 0.00 0.676 0.000 0.000 0.324
#> GSM198771 3 0.3828 0.641 0.000 0.00 0.560 0.000 0.000 0.440
#> GSM198774 3 0.3515 0.812 0.000 0.00 0.676 0.000 0.000 0.324
#> GSM198775 3 0.1910 0.401 0.000 0.00 0.892 0.000 0.108 0.000
#> GSM198777 3 0.3515 0.812 0.000 0.00 0.676 0.000 0.000 0.324
#> GSM198779 3 0.1910 0.401 0.000 0.00 0.892 0.000 0.108 0.000
#> GSM587218 4 0.0000 0.970 0.000 0.00 0.000 1.000 0.000 0.000
#> GSM587219 4 0.0000 0.970 0.000 0.00 0.000 1.000 0.000 0.000
#> GSM587220 4 0.0000 0.970 0.000 0.00 0.000 1.000 0.000 0.000
#> GSM587221 4 0.0000 0.970 0.000 0.00 0.000 1.000 0.000 0.000
#> GSM587222 4 0.0000 0.970 0.000 0.00 0.000 1.000 0.000 0.000
#> GSM587223 4 0.0000 0.970 0.000 0.00 0.000 1.000 0.000 0.000
#> GSM587224 4 0.0000 0.970 0.000 0.00 0.000 1.000 0.000 0.000
#> GSM587225 4 0.1501 0.907 0.076 0.00 0.000 0.924 0.000 0.000
#> GSM587226 4 0.0000 0.970 0.000 0.00 0.000 1.000 0.000 0.000
#> GSM587227 4 0.1075 0.943 0.048 0.00 0.000 0.952 0.000 0.000
#> GSM587228 4 0.1075 0.943 0.048 0.00 0.000 0.952 0.000 0.000
#> GSM587229 4 0.1075 0.943 0.048 0.00 0.000 0.952 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 specimen(p) k
#> ATC:hclust 92 1.16e-17 2
#> ATC:hclust 92 6.44e-33 3
#> ATC:hclust 92 4.04e-48 4
#> ATC:hclust 92 2.68e-63 5
#> ATC:hclust 88 2.84e-59 6
If matrix rows can be associated to genes, consider to use functional_enrichment(res,
...) to perform function enrichment for the signature genes. See this vignette for more detailed explanations.
The object with results only for a single top-value method and a single partition method can be extracted as:
res = res_list["ATC", "kmeans"]
# you can also extract it by
# res = res_list["ATC:kmeans"]
A summary of res and all the functions that can be applied to it:
res
#> A 'ConsensusPartition' object with k = 2, 3, 4, 5, 6.
#> On a matrix with 51941 rows and 92 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 1.000 1.000 0.4771 0.523 0.523
#> 3 3 0.803 0.945 0.947 0.3769 0.777 0.587
#> 4 4 0.869 0.873 0.867 0.0914 0.912 0.744
#> 5 5 0.761 0.823 0.812 0.0526 0.978 0.921
#> 6 6 0.742 0.766 0.808 0.0421 0.955 0.831
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
#> GSM587155 2 0 1 0 1
#> GSM587156 2 0 1 0 1
#> GSM587157 2 0 1 0 1
#> GSM587158 2 0 1 0 1
#> GSM587159 2 0 1 0 1
#> GSM587160 2 0 1 0 1
#> GSM587161 2 0 1 0 1
#> GSM587162 2 0 1 0 1
#> GSM587163 2 0 1 0 1
#> GSM587164 2 0 1 0 1
#> GSM587165 2 0 1 0 1
#> GSM587166 2 0 1 0 1
#> GSM587167 2 0 1 0 1
#> GSM587168 2 0 1 0 1
#> GSM587169 2 0 1 0 1
#> GSM587170 2 0 1 0 1
#> GSM587171 2 0 1 0 1
#> GSM587172 2 0 1 0 1
#> GSM587173 2 0 1 0 1
#> GSM587174 2 0 1 0 1
#> GSM587175 2 0 1 0 1
#> GSM587176 2 0 1 0 1
#> GSM587177 2 0 1 0 1
#> GSM587178 2 0 1 0 1
#> GSM587179 2 0 1 0 1
#> GSM587180 2 0 1 0 1
#> GSM587181 2 0 1 0 1
#> GSM587182 2 0 1 0 1
#> GSM587183 2 0 1 0 1
#> GSM587184 2 0 1 0 1
#> GSM587185 2 0 1 0 1
#> GSM587186 2 0 1 0 1
#> GSM587187 2 0 1 0 1
#> GSM587188 2 0 1 0 1
#> GSM587189 2 0 1 0 1
#> GSM587190 1 0 1 1 0
#> GSM587203 1 0 1 1 0
#> GSM587204 1 0 1 1 0
#> GSM587205 1 0 1 1 0
#> GSM587206 1 0 1 1 0
#> GSM587207 1 0 1 1 0
#> GSM587208 1 0 1 1 0
#> GSM587209 1 0 1 1 0
#> GSM587210 1 0 1 1 0
#> GSM587211 1 0 1 1 0
#> GSM587212 1 0 1 1 0
#> GSM587213 1 0 1 1 0
#> GSM587214 1 0 1 1 0
#> GSM587215 1 0 1 1 0
#> GSM587216 1 0 1 1 0
#> GSM587217 1 0 1 1 0
#> GSM587191 1 0 1 1 0
#> GSM587192 1 0 1 1 0
#> GSM587193 1 0 1 1 0
#> GSM587194 1 0 1 1 0
#> GSM587195 1 0 1 1 0
#> GSM587196 1 0 1 1 0
#> GSM587197 1 0 1 1 0
#> GSM587198 1 0 1 1 0
#> GSM587199 1 0 1 1 0
#> GSM587200 1 0 1 1 0
#> GSM587201 1 0 1 1 0
#> GSM587202 1 0 1 1 0
#> GSM198767 1 0 1 1 0
#> GSM198769 1 0 1 1 0
#> GSM198772 1 0 1 1 0
#> GSM198773 1 0 1 1 0
#> GSM198776 1 0 1 1 0
#> GSM198778 1 0 1 1 0
#> GSM198780 1 0 1 1 0
#> GSM198781 1 0 1 1 0
#> GSM198765 1 0 1 1 0
#> GSM198766 1 0 1 1 0
#> GSM198768 1 0 1 1 0
#> GSM198770 1 0 1 1 0
#> GSM198771 1 0 1 1 0
#> GSM198774 1 0 1 1 0
#> GSM198775 1 0 1 1 0
#> GSM198777 1 0 1 1 0
#> GSM198779 1 0 1 1 0
#> GSM587218 1 0 1 1 0
#> GSM587219 1 0 1 1 0
#> GSM587220 1 0 1 1 0
#> GSM587221 1 0 1 1 0
#> GSM587222 1 0 1 1 0
#> GSM587223 1 0 1 1 0
#> GSM587224 1 0 1 1 0
#> GSM587225 1 0 1 1 0
#> GSM587226 1 0 1 1 0
#> GSM587227 1 0 1 1 0
#> GSM587228 1 0 1 1 0
#> GSM587229 1 0 1 1 0
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM587155 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587156 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587157 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587158 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587159 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587160 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587161 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587162 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587163 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587164 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587165 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587166 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587167 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587168 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587169 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587170 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587171 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587172 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587173 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587174 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587175 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587176 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587177 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587178 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587179 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587180 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587181 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587182 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587183 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587184 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587185 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587186 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587187 3 0.4504 0.740 0.000 0.196 0.804
#> GSM587188 3 0.4452 0.746 0.000 0.192 0.808
#> GSM587189 3 0.4178 0.776 0.000 0.172 0.828
#> GSM587190 3 0.0747 0.966 0.016 0.000 0.984
#> GSM587203 1 0.2066 0.881 0.940 0.000 0.060
#> GSM587204 1 0.4291 0.915 0.820 0.000 0.180
#> GSM587205 1 0.4346 0.912 0.816 0.000 0.184
#> GSM587206 1 0.4346 0.912 0.816 0.000 0.184
#> GSM587207 1 0.4346 0.912 0.816 0.000 0.184
#> GSM587208 1 0.4346 0.912 0.816 0.000 0.184
#> GSM587209 1 0.4291 0.915 0.820 0.000 0.180
#> GSM587210 1 0.4235 0.916 0.824 0.000 0.176
#> GSM587211 1 0.4291 0.915 0.820 0.000 0.180
#> GSM587212 1 0.4291 0.915 0.820 0.000 0.180
#> GSM587213 1 0.4291 0.915 0.820 0.000 0.180
#> GSM587214 1 0.4291 0.915 0.820 0.000 0.180
#> GSM587215 1 0.4291 0.915 0.820 0.000 0.180
#> GSM587216 1 0.4291 0.915 0.820 0.000 0.180
#> GSM587217 1 0.4291 0.915 0.820 0.000 0.180
#> GSM587191 3 0.0747 0.966 0.016 0.000 0.984
#> GSM587192 3 0.0747 0.966 0.016 0.000 0.984
#> GSM587193 3 0.0747 0.966 0.016 0.000 0.984
#> GSM587194 3 0.0747 0.966 0.016 0.000 0.984
#> GSM587195 3 0.0747 0.966 0.016 0.000 0.984
#> GSM587196 3 0.0747 0.966 0.016 0.000 0.984
#> GSM587197 3 0.0747 0.966 0.016 0.000 0.984
#> GSM587198 3 0.0747 0.966 0.016 0.000 0.984
#> GSM587199 3 0.0747 0.966 0.016 0.000 0.984
#> GSM587200 3 0.0237 0.953 0.004 0.000 0.996
#> GSM587201 3 0.0237 0.953 0.004 0.000 0.996
#> GSM587202 3 0.0000 0.954 0.000 0.000 1.000
#> GSM198767 1 0.4346 0.912 0.816 0.000 0.184
#> GSM198769 1 0.4291 0.915 0.820 0.000 0.180
#> GSM198772 1 0.4291 0.915 0.820 0.000 0.180
#> GSM198773 1 0.4291 0.915 0.820 0.000 0.180
#> GSM198776 1 0.4291 0.915 0.820 0.000 0.180
#> GSM198778 1 0.4235 0.916 0.824 0.000 0.176
#> GSM198780 1 0.4291 0.915 0.820 0.000 0.180
#> GSM198781 1 0.4291 0.915 0.820 0.000 0.180
#> GSM198765 3 0.0747 0.966 0.016 0.000 0.984
#> GSM198766 3 0.0747 0.966 0.016 0.000 0.984
#> GSM198768 3 0.0747 0.966 0.016 0.000 0.984
#> GSM198770 3 0.0747 0.966 0.016 0.000 0.984
#> GSM198771 3 0.0747 0.966 0.016 0.000 0.984
#> GSM198774 3 0.0747 0.966 0.016 0.000 0.984
#> GSM198775 3 0.0747 0.966 0.016 0.000 0.984
#> GSM198777 3 0.0747 0.966 0.016 0.000 0.984
#> GSM198779 3 0.0747 0.966 0.016 0.000 0.984
#> GSM587218 1 0.0000 0.872 1.000 0.000 0.000
#> GSM587219 1 0.0000 0.872 1.000 0.000 0.000
#> GSM587220 1 0.0000 0.872 1.000 0.000 0.000
#> GSM587221 1 0.0000 0.872 1.000 0.000 0.000
#> GSM587222 1 0.0000 0.872 1.000 0.000 0.000
#> GSM587223 1 0.0000 0.872 1.000 0.000 0.000
#> GSM587224 1 0.0000 0.872 1.000 0.000 0.000
#> GSM587225 1 0.0000 0.872 1.000 0.000 0.000
#> GSM587226 1 0.0000 0.872 1.000 0.000 0.000
#> GSM587227 1 0.0000 0.872 1.000 0.000 0.000
#> GSM587228 1 0.0000 0.872 1.000 0.000 0.000
#> GSM587229 1 0.0000 0.872 1.000 0.000 0.000
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM587155 2 0.1302 0.939 0.044 0.956 0.000 0.000
#> GSM587156 2 0.1302 0.939 0.044 0.956 0.000 0.000
#> GSM587157 2 0.1302 0.939 0.044 0.956 0.000 0.000
#> GSM587158 2 0.0336 0.949 0.008 0.992 0.000 0.000
#> GSM587159 2 0.0336 0.949 0.008 0.992 0.000 0.000
#> GSM587160 2 0.0336 0.949 0.008 0.992 0.000 0.000
#> GSM587161 2 0.1302 0.939 0.044 0.956 0.000 0.000
#> GSM587162 2 0.1302 0.939 0.044 0.956 0.000 0.000
#> GSM587163 2 0.0336 0.949 0.008 0.992 0.000 0.000
#> GSM587164 2 0.1302 0.939 0.044 0.956 0.000 0.000
#> GSM587165 2 0.0336 0.949 0.008 0.992 0.000 0.000
#> GSM587166 2 0.1302 0.939 0.044 0.956 0.000 0.000
#> GSM587167 2 0.1302 0.939 0.044 0.956 0.000 0.000
#> GSM587168 2 0.0336 0.949 0.008 0.992 0.000 0.000
#> GSM587169 2 0.0000 0.950 0.000 1.000 0.000 0.000
#> GSM587170 2 0.0000 0.950 0.000 1.000 0.000 0.000
#> GSM587171 2 0.0000 0.950 0.000 1.000 0.000 0.000
#> GSM587172 2 0.0000 0.950 0.000 1.000 0.000 0.000
#> GSM587173 2 0.0000 0.950 0.000 1.000 0.000 0.000
#> GSM587174 2 0.0000 0.950 0.000 1.000 0.000 0.000
#> GSM587175 2 0.0000 0.950 0.000 1.000 0.000 0.000
#> GSM587176 2 0.0000 0.950 0.000 1.000 0.000 0.000
#> GSM587177 2 0.0000 0.950 0.000 1.000 0.000 0.000
#> GSM587178 2 0.0000 0.950 0.000 1.000 0.000 0.000
#> GSM587179 2 0.0000 0.950 0.000 1.000 0.000 0.000
#> GSM587180 2 0.0000 0.950 0.000 1.000 0.000 0.000
#> GSM587181 2 0.0000 0.950 0.000 1.000 0.000 0.000
#> GSM587182 2 0.0000 0.950 0.000 1.000 0.000 0.000
#> GSM587183 2 0.5384 0.671 0.028 0.648 0.324 0.000
#> GSM587184 2 0.5384 0.671 0.028 0.648 0.324 0.000
#> GSM587185 2 0.5384 0.671 0.028 0.648 0.324 0.000
#> GSM587186 2 0.5384 0.671 0.028 0.648 0.324 0.000
#> GSM587187 3 0.0592 0.675 0.000 0.016 0.984 0.000
#> GSM587188 3 0.0592 0.675 0.000 0.016 0.984 0.000
#> GSM587189 3 0.0592 0.675 0.000 0.016 0.984 0.000
#> GSM587190 3 0.4250 0.889 0.276 0.000 0.724 0.000
#> GSM587203 1 0.5000 0.761 0.504 0.000 0.000 0.496
#> GSM587204 1 0.4967 0.859 0.548 0.000 0.000 0.452
#> GSM587205 1 0.5126 0.847 0.552 0.000 0.004 0.444
#> GSM587206 1 0.5126 0.847 0.552 0.000 0.004 0.444
#> GSM587207 1 0.5126 0.847 0.552 0.000 0.004 0.444
#> GSM587208 1 0.5126 0.847 0.552 0.000 0.004 0.444
#> GSM587209 1 0.5155 0.857 0.528 0.000 0.004 0.468
#> GSM587210 1 0.4961 0.858 0.552 0.000 0.000 0.448
#> GSM587211 1 0.5155 0.857 0.528 0.000 0.004 0.468
#> GSM587212 1 0.4961 0.858 0.552 0.000 0.000 0.448
#> GSM587213 1 0.5155 0.857 0.528 0.000 0.004 0.468
#> GSM587214 1 0.4967 0.859 0.548 0.000 0.000 0.452
#> GSM587215 1 0.4605 0.676 0.664 0.000 0.000 0.336
#> GSM587216 1 0.4967 0.859 0.548 0.000 0.000 0.452
#> GSM587217 1 0.4967 0.859 0.548 0.000 0.000 0.452
#> GSM587191 3 0.4382 0.893 0.296 0.000 0.704 0.000
#> GSM587192 3 0.4877 0.930 0.408 0.000 0.592 0.000
#> GSM587193 3 0.4817 0.931 0.388 0.000 0.612 0.000
#> GSM587194 3 0.4804 0.926 0.384 0.000 0.616 0.000
#> GSM587195 3 0.4746 0.930 0.368 0.000 0.632 0.000
#> GSM587196 3 0.4776 0.932 0.376 0.000 0.624 0.000
#> GSM587197 3 0.4730 0.929 0.364 0.000 0.636 0.000
#> GSM587198 3 0.4888 0.928 0.412 0.000 0.588 0.000
#> GSM587199 3 0.4877 0.926 0.408 0.000 0.592 0.000
#> GSM587200 1 0.3870 -0.282 0.788 0.000 0.208 0.004
#> GSM587201 1 0.3870 -0.282 0.788 0.000 0.208 0.004
#> GSM587202 3 0.4898 0.926 0.416 0.000 0.584 0.000
#> GSM198767 1 0.5132 0.849 0.548 0.000 0.004 0.448
#> GSM198769 1 0.5155 0.857 0.528 0.000 0.004 0.468
#> GSM198772 1 0.5155 0.857 0.528 0.000 0.004 0.468
#> GSM198773 1 0.5155 0.857 0.528 0.000 0.004 0.468
#> GSM198776 1 0.4967 0.859 0.548 0.000 0.000 0.452
#> GSM198778 1 0.4961 0.858 0.552 0.000 0.000 0.448
#> GSM198780 1 0.4961 0.858 0.552 0.000 0.000 0.448
#> GSM198781 1 0.4967 0.859 0.548 0.000 0.000 0.452
#> GSM198765 3 0.4730 0.930 0.364 0.000 0.636 0.000
#> GSM198766 3 0.4817 0.931 0.388 0.000 0.612 0.000
#> GSM198768 3 0.4776 0.931 0.376 0.000 0.624 0.000
#> GSM198770 3 0.4730 0.929 0.364 0.000 0.636 0.000
#> GSM198771 3 0.4888 0.928 0.412 0.000 0.588 0.000
#> GSM198774 3 0.4877 0.930 0.408 0.000 0.592 0.000
#> GSM198775 3 0.4804 0.926 0.384 0.000 0.616 0.000
#> GSM198777 3 0.4776 0.932 0.376 0.000 0.624 0.000
#> GSM198779 3 0.4877 0.926 0.408 0.000 0.592 0.000
#> GSM587218 4 0.0000 0.985 0.000 0.000 0.000 1.000
#> GSM587219 4 0.0000 0.985 0.000 0.000 0.000 1.000
#> GSM587220 4 0.0000 0.985 0.000 0.000 0.000 1.000
#> GSM587221 4 0.0000 0.985 0.000 0.000 0.000 1.000
#> GSM587222 4 0.0000 0.985 0.000 0.000 0.000 1.000
#> GSM587223 4 0.0000 0.985 0.000 0.000 0.000 1.000
#> GSM587224 4 0.0000 0.985 0.000 0.000 0.000 1.000
#> GSM587225 4 0.0937 0.973 0.012 0.000 0.012 0.976
#> GSM587226 4 0.0000 0.985 0.000 0.000 0.000 1.000
#> GSM587227 4 0.1284 0.965 0.024 0.000 0.012 0.964
#> GSM587228 4 0.1284 0.965 0.024 0.000 0.012 0.964
#> GSM587229 4 0.1284 0.965 0.024 0.000 0.012 0.964
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM587155 2 0.3274 0.812 0.000 0.780 0.000 0.000 NA
#> GSM587156 2 0.3430 0.813 0.004 0.776 0.000 0.000 NA
#> GSM587157 2 0.3274 0.812 0.000 0.780 0.000 0.000 NA
#> GSM587158 2 0.0162 0.884 0.004 0.996 0.000 0.000 NA
#> GSM587159 2 0.0162 0.884 0.004 0.996 0.000 0.000 NA
#> GSM587160 2 0.0162 0.884 0.004 0.996 0.000 0.000 NA
#> GSM587161 2 0.3274 0.812 0.000 0.780 0.000 0.000 NA
#> GSM587162 2 0.3039 0.824 0.000 0.808 0.000 0.000 NA
#> GSM587163 2 0.0162 0.884 0.004 0.996 0.000 0.000 NA
#> GSM587164 2 0.3274 0.812 0.000 0.780 0.000 0.000 NA
#> GSM587165 2 0.0000 0.884 0.000 1.000 0.000 0.000 NA
#> GSM587166 2 0.3430 0.813 0.004 0.776 0.000 0.000 NA
#> GSM587167 2 0.3274 0.812 0.000 0.780 0.000 0.000 NA
#> GSM587168 2 0.0693 0.884 0.008 0.980 0.000 0.000 NA
#> GSM587169 2 0.0290 0.884 0.008 0.992 0.000 0.000 NA
#> GSM587170 2 0.0324 0.884 0.004 0.992 0.000 0.000 NA
#> GSM587171 2 0.0162 0.884 0.004 0.996 0.000 0.000 NA
#> GSM587172 2 0.0162 0.884 0.004 0.996 0.000 0.000 NA
#> GSM587173 2 0.0290 0.883 0.008 0.992 0.000 0.000 NA
#> GSM587174 2 0.0693 0.884 0.008 0.980 0.000 0.000 NA
#> GSM587175 2 0.0324 0.884 0.004 0.992 0.000 0.000 NA
#> GSM587176 2 0.0404 0.883 0.012 0.988 0.000 0.000 NA
#> GSM587177 2 0.0290 0.883 0.008 0.992 0.000 0.000 NA
#> GSM587178 2 0.0404 0.883 0.012 0.988 0.000 0.000 NA
#> GSM587179 2 0.0693 0.884 0.008 0.980 0.000 0.000 NA
#> GSM587180 2 0.0693 0.884 0.008 0.980 0.000 0.000 NA
#> GSM587181 2 0.0693 0.884 0.008 0.980 0.000 0.000 NA
#> GSM587182 2 0.0693 0.884 0.008 0.980 0.000 0.000 NA
#> GSM587183 2 0.6635 0.412 0.224 0.416 0.000 0.000 NA
#> GSM587184 2 0.6635 0.412 0.224 0.416 0.000 0.000 NA
#> GSM587185 2 0.6635 0.412 0.224 0.416 0.000 0.000 NA
#> GSM587186 2 0.6650 0.408 0.228 0.412 0.000 0.000 NA
#> GSM587187 3 0.6587 0.547 0.168 0.008 0.472 0.000 NA
#> GSM587188 3 0.6478 0.551 0.168 0.004 0.476 0.000 NA
#> GSM587189 3 0.6478 0.551 0.168 0.004 0.476 0.000 NA
#> GSM587190 3 0.3601 0.815 0.052 0.000 0.820 0.000 NA
#> GSM587203 1 0.5711 0.832 0.572 0.000 0.020 0.356 NA
#> GSM587204 1 0.6881 0.847 0.500 0.000 0.052 0.340 NA
#> GSM587205 1 0.6128 0.831 0.580 0.000 0.044 0.316 NA
#> GSM587206 1 0.6128 0.831 0.580 0.000 0.044 0.316 NA
#> GSM587207 1 0.6128 0.831 0.580 0.000 0.044 0.316 NA
#> GSM587208 1 0.6128 0.831 0.580 0.000 0.044 0.316 NA
#> GSM587209 1 0.5230 0.879 0.600 0.000 0.048 0.348 NA
#> GSM587210 1 0.7095 0.824 0.484 0.000 0.052 0.328 NA
#> GSM587211 1 0.5230 0.879 0.600 0.000 0.048 0.348 NA
#> GSM587212 1 0.7075 0.819 0.492 0.000 0.052 0.320 NA
#> GSM587213 1 0.5230 0.879 0.600 0.000 0.048 0.348 NA
#> GSM587214 1 0.6017 0.876 0.572 0.000 0.052 0.336 NA
#> GSM587215 1 0.6479 0.804 0.588 0.000 0.120 0.252 NA
#> GSM587216 1 0.6811 0.848 0.516 0.000 0.052 0.328 NA
#> GSM587217 1 0.6437 0.868 0.540 0.000 0.052 0.340 NA
#> GSM587191 3 0.2983 0.833 0.056 0.000 0.868 0.000 NA
#> GSM587192 3 0.2729 0.841 0.056 0.000 0.884 0.000 NA
#> GSM587193 3 0.1750 0.851 0.036 0.000 0.936 0.000 NA
#> GSM587194 3 0.3301 0.833 0.080 0.000 0.848 0.000 NA
#> GSM587195 3 0.2448 0.843 0.020 0.000 0.892 0.000 NA
#> GSM587196 3 0.1195 0.853 0.012 0.000 0.960 0.000 NA
#> GSM587197 3 0.2505 0.843 0.020 0.000 0.888 0.000 NA
#> GSM587198 3 0.1965 0.845 0.024 0.000 0.924 0.000 NA
#> GSM587199 3 0.2928 0.835 0.064 0.000 0.872 0.000 NA
#> GSM587200 3 0.6113 0.320 0.372 0.000 0.508 0.004 NA
#> GSM587201 3 0.6113 0.320 0.372 0.000 0.508 0.004 NA
#> GSM587202 3 0.2719 0.831 0.048 0.000 0.884 0.000 NA
#> GSM198767 1 0.6017 0.847 0.572 0.000 0.040 0.336 NA
#> GSM198769 1 0.5230 0.879 0.600 0.000 0.048 0.348 NA
#> GSM198772 1 0.5230 0.879 0.600 0.000 0.048 0.348 NA
#> GSM198773 1 0.5230 0.879 0.600 0.000 0.048 0.348 NA
#> GSM198776 1 0.6881 0.847 0.500 0.000 0.052 0.340 NA
#> GSM198778 1 0.7095 0.824 0.484 0.000 0.052 0.328 NA
#> GSM198780 1 0.7075 0.819 0.492 0.000 0.052 0.320 NA
#> GSM198781 1 0.6017 0.876 0.572 0.000 0.052 0.336 NA
#> GSM198765 3 0.1741 0.850 0.024 0.000 0.936 0.000 NA
#> GSM198766 3 0.1750 0.851 0.036 0.000 0.936 0.000 NA
#> GSM198768 3 0.2208 0.845 0.020 0.000 0.908 0.000 NA
#> GSM198770 3 0.2505 0.843 0.020 0.000 0.888 0.000 NA
#> GSM198771 3 0.1893 0.844 0.024 0.000 0.928 0.000 NA
#> GSM198774 3 0.2729 0.841 0.056 0.000 0.884 0.000 NA
#> GSM198775 3 0.3301 0.833 0.080 0.000 0.848 0.000 NA
#> GSM198777 3 0.1195 0.853 0.012 0.000 0.960 0.000 NA
#> GSM198779 3 0.2928 0.835 0.064 0.000 0.872 0.000 NA
#> GSM587218 4 0.0000 0.953 0.000 0.000 0.000 1.000 NA
#> GSM587219 4 0.0000 0.953 0.000 0.000 0.000 1.000 NA
#> GSM587220 4 0.0000 0.953 0.000 0.000 0.000 1.000 NA
#> GSM587221 4 0.0000 0.953 0.000 0.000 0.000 1.000 NA
#> GSM587222 4 0.0000 0.953 0.000 0.000 0.000 1.000 NA
#> GSM587223 4 0.0000 0.953 0.000 0.000 0.000 1.000 NA
#> GSM587224 4 0.0000 0.953 0.000 0.000 0.000 1.000 NA
#> GSM587225 4 0.2249 0.905 0.008 0.000 0.000 0.896 NA
#> GSM587226 4 0.0000 0.953 0.000 0.000 0.000 1.000 NA
#> GSM587227 4 0.2573 0.895 0.016 0.000 0.000 0.880 NA
#> GSM587228 4 0.2573 0.895 0.016 0.000 0.000 0.880 NA
#> GSM587229 4 0.2573 0.895 0.016 0.000 0.000 0.880 NA
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM587155 2 0.3706 0.639 0.000 0.620 0.000 0.000 NA 0.000
#> GSM587156 2 0.4193 0.635 0.024 0.624 0.000 0.000 NA 0.000
#> GSM587157 2 0.3706 0.639 0.000 0.620 0.000 0.000 NA 0.000
#> GSM587158 2 0.0622 0.851 0.008 0.980 0.000 0.000 NA 0.000
#> GSM587159 2 0.0520 0.851 0.008 0.984 0.000 0.000 NA 0.000
#> GSM587160 2 0.0622 0.851 0.008 0.980 0.000 0.000 NA 0.000
#> GSM587161 2 0.3706 0.639 0.000 0.620 0.000 0.000 NA 0.000
#> GSM587162 2 0.3578 0.668 0.000 0.660 0.000 0.000 NA 0.000
#> GSM587163 2 0.0622 0.851 0.008 0.980 0.000 0.000 NA 0.000
#> GSM587164 2 0.3695 0.640 0.000 0.624 0.000 0.000 NA 0.000
#> GSM587165 2 0.0520 0.851 0.008 0.984 0.000 0.000 NA 0.000
#> GSM587166 2 0.4193 0.635 0.024 0.624 0.000 0.000 NA 0.000
#> GSM587167 2 0.3819 0.639 0.004 0.624 0.000 0.000 NA 0.000
#> GSM587168 2 0.2003 0.843 0.044 0.912 0.000 0.000 NA 0.000
#> GSM587169 2 0.0260 0.851 0.008 0.992 0.000 0.000 NA 0.000
#> GSM587170 2 0.0260 0.851 0.008 0.992 0.000 0.000 NA 0.000
#> GSM587171 2 0.0363 0.851 0.012 0.988 0.000 0.000 NA 0.000
#> GSM587172 2 0.0363 0.851 0.012 0.988 0.000 0.000 NA 0.000
#> GSM587173 2 0.0937 0.847 0.040 0.960 0.000 0.000 NA 0.000
#> GSM587174 2 0.1649 0.844 0.036 0.932 0.000 0.000 NA 0.000
#> GSM587175 2 0.0260 0.851 0.008 0.992 0.000 0.000 NA 0.000
#> GSM587176 2 0.0000 0.851 0.000 1.000 0.000 0.000 NA 0.000
#> GSM587177 2 0.1010 0.846 0.036 0.960 0.000 0.000 NA 0.000
#> GSM587178 2 0.0146 0.851 0.004 0.996 0.000 0.000 NA 0.000
#> GSM587179 2 0.1649 0.844 0.036 0.932 0.000 0.000 NA 0.000
#> GSM587180 2 0.1720 0.844 0.040 0.928 0.000 0.000 NA 0.000
#> GSM587181 2 0.1649 0.844 0.036 0.932 0.000 0.000 NA 0.000
#> GSM587182 2 0.1720 0.844 0.040 0.928 0.000 0.000 NA 0.000
#> GSM587183 6 0.6328 0.569 0.088 0.312 0.000 0.000 NA 0.512
#> GSM587184 6 0.6328 0.569 0.088 0.312 0.000 0.000 NA 0.512
#> GSM587185 6 0.6327 0.568 0.092 0.312 0.000 0.000 NA 0.512
#> GSM587186 6 0.6328 0.569 0.088 0.312 0.000 0.000 NA 0.512
#> GSM587187 6 0.3756 0.254 0.000 0.004 0.352 0.000 NA 0.644
#> GSM587188 6 0.3620 0.248 0.000 0.000 0.352 0.000 NA 0.648
#> GSM587189 6 0.3620 0.248 0.000 0.000 0.352 0.000 NA 0.648
#> GSM587190 3 0.3690 0.624 0.000 0.000 0.700 0.000 NA 0.288
#> GSM587203 1 0.5293 0.781 0.644 0.000 0.004 0.216 NA 0.012
#> GSM587204 1 0.6092 0.807 0.612 0.000 0.016 0.216 NA 0.096
#> GSM587205 1 0.5426 0.771 0.632 0.000 0.012 0.204 NA 0.004
#> GSM587206 1 0.5426 0.771 0.632 0.000 0.012 0.204 NA 0.004
#> GSM587207 1 0.5426 0.771 0.632 0.000 0.012 0.204 NA 0.004
#> GSM587208 1 0.5426 0.771 0.632 0.000 0.012 0.204 NA 0.004
#> GSM587209 1 0.4891 0.835 0.692 0.000 0.024 0.228 NA 0.016
#> GSM587210 1 0.6591 0.777 0.568 0.000 0.016 0.212 NA 0.112
#> GSM587211 1 0.4600 0.836 0.708 0.000 0.024 0.228 NA 0.012
#> GSM587212 1 0.6798 0.758 0.552 0.000 0.020 0.208 NA 0.116
#> GSM587213 1 0.4891 0.834 0.692 0.000 0.024 0.228 NA 0.016
#> GSM587214 1 0.5229 0.833 0.680 0.000 0.020 0.216 NA 0.048
#> GSM587215 1 0.5597 0.752 0.696 0.000 0.116 0.104 NA 0.048
#> GSM587216 1 0.5930 0.813 0.628 0.000 0.016 0.212 NA 0.088
#> GSM587217 1 0.4958 0.834 0.696 0.000 0.020 0.216 NA 0.044
#> GSM587191 3 0.2793 0.779 0.004 0.000 0.856 0.000 NA 0.112
#> GSM587192 3 0.3571 0.778 0.024 0.000 0.824 0.000 NA 0.068
#> GSM587193 3 0.2051 0.800 0.008 0.000 0.916 0.000 NA 0.040
#> GSM587194 3 0.4591 0.738 0.028 0.000 0.740 0.000 NA 0.112
#> GSM587195 3 0.2834 0.763 0.008 0.000 0.852 0.000 NA 0.120
#> GSM587196 3 0.1719 0.800 0.008 0.000 0.928 0.000 NA 0.056
#> GSM587197 3 0.2834 0.762 0.008 0.000 0.848 0.000 NA 0.128
#> GSM587198 3 0.2362 0.785 0.016 0.000 0.892 0.000 NA 0.012
#> GSM587199 3 0.4239 0.749 0.028 0.000 0.768 0.000 NA 0.072
#> GSM587200 3 0.6003 0.359 0.272 0.000 0.496 0.000 NA 0.008
#> GSM587201 3 0.6003 0.359 0.272 0.000 0.496 0.000 NA 0.008
#> GSM587202 3 0.3043 0.757 0.024 0.000 0.836 0.000 NA 0.008
#> GSM198767 1 0.5442 0.787 0.644 0.000 0.012 0.208 NA 0.012
#> GSM198769 1 0.4891 0.835 0.692 0.000 0.024 0.228 NA 0.016
#> GSM198772 1 0.4600 0.836 0.708 0.000 0.024 0.228 NA 0.012
#> GSM198773 1 0.4891 0.834 0.692 0.000 0.024 0.228 NA 0.016
#> GSM198776 1 0.6092 0.807 0.612 0.000 0.016 0.216 NA 0.096
#> GSM198778 1 0.6591 0.777 0.568 0.000 0.016 0.212 NA 0.112
#> GSM198780 1 0.6798 0.758 0.552 0.000 0.020 0.208 NA 0.116
#> GSM198781 1 0.5229 0.833 0.680 0.000 0.020 0.216 NA 0.048
#> GSM198765 3 0.2052 0.798 0.004 0.000 0.912 0.000 NA 0.056
#> GSM198766 3 0.2051 0.800 0.008 0.000 0.916 0.000 NA 0.040
#> GSM198768 3 0.2454 0.781 0.008 0.000 0.884 0.000 NA 0.088
#> GSM198770 3 0.2834 0.762 0.008 0.000 0.848 0.000 NA 0.128
#> GSM198771 3 0.2149 0.785 0.016 0.000 0.900 0.000 NA 0.004
#> GSM198774 3 0.3571 0.778 0.024 0.000 0.824 0.000 NA 0.068
#> GSM198775 3 0.4591 0.738 0.028 0.000 0.740 0.000 NA 0.112
#> GSM198777 3 0.1719 0.800 0.008 0.000 0.928 0.000 NA 0.056
#> GSM198779 3 0.4239 0.749 0.028 0.000 0.768 0.000 NA 0.072
#> GSM587218 4 0.0000 0.935 0.000 0.000 0.000 1.000 NA 0.000
#> GSM587219 4 0.0000 0.935 0.000 0.000 0.000 1.000 NA 0.000
#> GSM587220 4 0.0000 0.935 0.000 0.000 0.000 1.000 NA 0.000
#> GSM587221 4 0.0000 0.935 0.000 0.000 0.000 1.000 NA 0.000
#> GSM587222 4 0.0000 0.935 0.000 0.000 0.000 1.000 NA 0.000
#> GSM587223 4 0.0000 0.935 0.000 0.000 0.000 1.000 NA 0.000
#> GSM587224 4 0.0000 0.935 0.000 0.000 0.000 1.000 NA 0.000
#> GSM587225 4 0.3821 0.855 0.028 0.000 0.000 0.804 NA 0.060
#> GSM587226 4 0.0000 0.935 0.000 0.000 0.000 1.000 NA 0.000
#> GSM587227 4 0.3776 0.859 0.028 0.000 0.000 0.808 NA 0.060
#> GSM587228 4 0.3776 0.859 0.028 0.000 0.000 0.808 NA 0.060
#> GSM587229 4 0.3776 0.859 0.028 0.000 0.000 0.808 NA 0.060
Heatmaps for the consensus matrix. It visualizes the probability of two samples to be in a same group.
consensus_heatmap(res, k = 2)
consensus_heatmap(res, k = 3)
consensus_heatmap(res, k = 4)
consensus_heatmap(res, k = 5)
consensus_heatmap(res, k = 6)
Heatmaps for the membership of samples in all partitions to see how consistent they are:
membership_heatmap(res, k = 2)
membership_heatmap(res, k = 3)
membership_heatmap(res, k = 4)
membership_heatmap(res, k = 5)
membership_heatmap(res, k = 6)
As soon as we have had the classes for columns, we can look for signatures which are significantly different between classes which can be candidate marks for certain classes. Following are the heatmaps for signatures.
Signature heatmaps where rows are scaled:
get_signatures(res, k = 2)
get_signatures(res, k = 3)
get_signatures(res, k = 4)
get_signatures(res, k = 5)
get_signatures(res, k = 6)
Signature heatmaps where rows are not scaled:
get_signatures(res, k = 2, scale_rows = FALSE)
get_signatures(res, k = 3, scale_rows = FALSE)
get_signatures(res, k = 4, scale_rows = FALSE)
get_signatures(res, k = 5, scale_rows = FALSE)
get_signatures(res, k = 6, scale_rows = FALSE)
Compare the overlap of signatures from different k:
compare_signatures(res)
get_signature() returns a data frame invisibly. TO get the list of signatures, the function
call should be assigned to a variable explicitly. In following code, if plot argument is set
to FALSE, no heatmap is plotted while only the differential analysis is performed.
# code only for demonstration
tb = get_signature(res, k = ..., plot = FALSE)
An example of the output of tb is:
#> which_row fdr mean_1 mean_2 scaled_mean_1 scaled_mean_2 km
#> 1 38 0.042760348 8.373488 9.131774 -0.5533452 0.5164555 1
#> 2 40 0.018707592 7.106213 8.469186 -0.6173731 0.5762149 1
#> 3 55 0.019134737 10.221463 11.207825 -0.6159697 0.5749050 1
#> 4 59 0.006059896 5.921854 7.869574 -0.6899429 0.6439467 1
#> 5 60 0.018055526 8.928898 10.211722 -0.6204761 0.5791110 1
#> 6 98 0.009384629 15.714769 14.887706 0.6635654 -0.6193277 2
...
The columns in tb are:
which_row: row indices corresponding to the input matrix.fdr: FDR for the differential test. mean_x: The mean value in group x.scaled_mean_x: The mean value in group x after rows are scaled.km: Row groups if k-means clustering is applied to rows.UMAP plot which shows how samples are separated.
dimension_reduction(res, k = 2, method = "UMAP")
dimension_reduction(res, k = 3, method = "UMAP")
dimension_reduction(res, k = 4, method = "UMAP")
dimension_reduction(res, k = 5, method = "UMAP")
dimension_reduction(res, k = 6, method = "UMAP")
Following heatmap shows how subgroups are split when increasing k:
collect_classes(res)
Test correlation between subgroups and known annotations. If the known annotation is numeric, one-way ANOVA test is applied, and if the known annotation is discrete, chi-squared contingency table test is applied.
test_to_known_factors(res)
#> n specimen(p) k
#> ATC:kmeans 92 5.33e-17 2
#> ATC:kmeans 92 6.44e-33 3
#> ATC:kmeans 90 6.82e-47 4
#> ATC:kmeans 86 1.92e-44 5
#> ATC:kmeans 87 1.08e-41 6
If matrix rows can be associated to genes, consider to use functional_enrichment(res,
...) to perform function enrichment for the signature genes. See this vignette for more detailed explanations.
The object with results only for a single top-value method and a single partition method can be extracted as:
res = res_list["ATC", "skmeans"]
# you can also extract it by
# res = res_list["ATC:skmeans"]
A summary of res and all the functions that can be applied to it:
res
#> A 'ConsensusPartition' object with k = 2, 3, 4, 5, 6.
#> On a matrix with 51941 rows and 92 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 1.000 1.000 0.4867 0.514 0.514
#> 3 3 1.000 0.984 0.993 0.3373 0.816 0.647
#> 4 4 0.878 0.759 0.873 0.0519 0.977 0.934
#> 5 5 0.807 0.801 0.853 0.0586 0.953 0.860
#> 6 6 0.814 0.781 0.786 0.0452 0.920 0.737
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
#> GSM587155 2 0 1 0 1
#> GSM587156 2 0 1 0 1
#> GSM587157 2 0 1 0 1
#> GSM587158 2 0 1 0 1
#> GSM587159 2 0 1 0 1
#> GSM587160 2 0 1 0 1
#> GSM587161 2 0 1 0 1
#> GSM587162 2 0 1 0 1
#> GSM587163 2 0 1 0 1
#> GSM587164 2 0 1 0 1
#> GSM587165 2 0 1 0 1
#> GSM587166 2 0 1 0 1
#> GSM587167 2 0 1 0 1
#> GSM587168 2 0 1 0 1
#> GSM587169 2 0 1 0 1
#> GSM587170 2 0 1 0 1
#> GSM587171 2 0 1 0 1
#> GSM587172 2 0 1 0 1
#> GSM587173 2 0 1 0 1
#> GSM587174 2 0 1 0 1
#> GSM587175 2 0 1 0 1
#> GSM587176 2 0 1 0 1
#> GSM587177 2 0 1 0 1
#> GSM587178 2 0 1 0 1
#> GSM587179 2 0 1 0 1
#> GSM587180 2 0 1 0 1
#> GSM587181 2 0 1 0 1
#> GSM587182 2 0 1 0 1
#> GSM587183 2 0 1 0 1
#> GSM587184 2 0 1 0 1
#> GSM587185 2 0 1 0 1
#> GSM587186 2 0 1 0 1
#> GSM587187 2 0 1 0 1
#> GSM587188 2 0 1 0 1
#> GSM587189 2 0 1 0 1
#> GSM587190 2 0 1 0 1
#> GSM587203 1 0 1 1 0
#> GSM587204 1 0 1 1 0
#> GSM587205 1 0 1 1 0
#> GSM587206 1 0 1 1 0
#> GSM587207 1 0 1 1 0
#> GSM587208 1 0 1 1 0
#> GSM587209 1 0 1 1 0
#> GSM587210 1 0 1 1 0
#> GSM587211 1 0 1 1 0
#> GSM587212 1 0 1 1 0
#> GSM587213 1 0 1 1 0
#> GSM587214 1 0 1 1 0
#> GSM587215 1 0 1 1 0
#> GSM587216 1 0 1 1 0
#> GSM587217 1 0 1 1 0
#> GSM587191 2 0 1 0 1
#> GSM587192 1 0 1 1 0
#> GSM587193 1 0 1 1 0
#> GSM587194 1 0 1 1 0
#> GSM587195 1 0 1 1 0
#> GSM587196 1 0 1 1 0
#> GSM587197 1 0 1 1 0
#> GSM587198 1 0 1 1 0
#> GSM587199 1 0 1 1 0
#> GSM587200 1 0 1 1 0
#> GSM587201 1 0 1 1 0
#> GSM587202 1 0 1 1 0
#> GSM198767 1 0 1 1 0
#> GSM198769 1 0 1 1 0
#> GSM198772 1 0 1 1 0
#> GSM198773 1 0 1 1 0
#> GSM198776 1 0 1 1 0
#> GSM198778 1 0 1 1 0
#> GSM198780 1 0 1 1 0
#> GSM198781 1 0 1 1 0
#> GSM198765 1 0 1 1 0
#> GSM198766 1 0 1 1 0
#> GSM198768 1 0 1 1 0
#> GSM198770 1 0 1 1 0
#> GSM198771 1 0 1 1 0
#> GSM198774 1 0 1 1 0
#> GSM198775 1 0 1 1 0
#> GSM198777 1 0 1 1 0
#> GSM198779 1 0 1 1 0
#> GSM587218 1 0 1 1 0
#> GSM587219 1 0 1 1 0
#> GSM587220 1 0 1 1 0
#> GSM587221 1 0 1 1 0
#> GSM587222 1 0 1 1 0
#> GSM587223 1 0 1 1 0
#> GSM587224 1 0 1 1 0
#> GSM587225 1 0 1 1 0
#> GSM587226 1 0 1 1 0
#> GSM587227 1 0 1 1 0
#> GSM587228 1 0 1 1 0
#> GSM587229 1 0 1 1 0
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM587155 2 0.000 1.000 0.000 1 0.000
#> GSM587156 2 0.000 1.000 0.000 1 0.000
#> GSM587157 2 0.000 1.000 0.000 1 0.000
#> GSM587158 2 0.000 1.000 0.000 1 0.000
#> GSM587159 2 0.000 1.000 0.000 1 0.000
#> GSM587160 2 0.000 1.000 0.000 1 0.000
#> GSM587161 2 0.000 1.000 0.000 1 0.000
#> GSM587162 2 0.000 1.000 0.000 1 0.000
#> GSM587163 2 0.000 1.000 0.000 1 0.000
#> GSM587164 2 0.000 1.000 0.000 1 0.000
#> GSM587165 2 0.000 1.000 0.000 1 0.000
#> GSM587166 2 0.000 1.000 0.000 1 0.000
#> GSM587167 2 0.000 1.000 0.000 1 0.000
#> GSM587168 2 0.000 1.000 0.000 1 0.000
#> GSM587169 2 0.000 1.000 0.000 1 0.000
#> GSM587170 2 0.000 1.000 0.000 1 0.000
#> GSM587171 2 0.000 1.000 0.000 1 0.000
#> GSM587172 2 0.000 1.000 0.000 1 0.000
#> GSM587173 2 0.000 1.000 0.000 1 0.000
#> GSM587174 2 0.000 1.000 0.000 1 0.000
#> GSM587175 2 0.000 1.000 0.000 1 0.000
#> GSM587176 2 0.000 1.000 0.000 1 0.000
#> GSM587177 2 0.000 1.000 0.000 1 0.000
#> GSM587178 2 0.000 1.000 0.000 1 0.000
#> GSM587179 2 0.000 1.000 0.000 1 0.000
#> GSM587180 2 0.000 1.000 0.000 1 0.000
#> GSM587181 2 0.000 1.000 0.000 1 0.000
#> GSM587182 2 0.000 1.000 0.000 1 0.000
#> GSM587183 2 0.000 1.000 0.000 1 0.000
#> GSM587184 2 0.000 1.000 0.000 1 0.000
#> GSM587185 2 0.000 1.000 0.000 1 0.000
#> GSM587186 2 0.000 1.000 0.000 1 0.000
#> GSM587187 2 0.000 1.000 0.000 1 0.000
#> GSM587188 2 0.000 1.000 0.000 1 0.000
#> GSM587189 2 0.000 1.000 0.000 1 0.000
#> GSM587190 3 0.000 0.965 0.000 0 1.000
#> GSM587203 1 0.000 1.000 1.000 0 0.000
#> GSM587204 1 0.000 1.000 1.000 0 0.000
#> GSM587205 1 0.000 1.000 1.000 0 0.000
#> GSM587206 1 0.000 1.000 1.000 0 0.000
#> GSM587207 1 0.000 1.000 1.000 0 0.000
#> GSM587208 1 0.000 1.000 1.000 0 0.000
#> GSM587209 1 0.000 1.000 1.000 0 0.000
#> GSM587210 1 0.000 1.000 1.000 0 0.000
#> GSM587211 1 0.000 1.000 1.000 0 0.000
#> GSM587212 1 0.000 1.000 1.000 0 0.000
#> GSM587213 1 0.000 1.000 1.000 0 0.000
#> GSM587214 1 0.000 1.000 1.000 0 0.000
#> GSM587215 1 0.000 1.000 1.000 0 0.000
#> GSM587216 1 0.000 1.000 1.000 0 0.000
#> GSM587217 1 0.000 1.000 1.000 0 0.000
#> GSM587191 3 0.000 0.965 0.000 0 1.000
#> GSM587192 3 0.000 0.965 0.000 0 1.000
#> GSM587193 3 0.559 0.588 0.304 0 0.696
#> GSM587194 3 0.000 0.965 0.000 0 1.000
#> GSM587195 3 0.000 0.965 0.000 0 1.000
#> GSM587196 3 0.000 0.965 0.000 0 1.000
#> GSM587197 3 0.000 0.965 0.000 0 1.000
#> GSM587198 3 0.000 0.965 0.000 0 1.000
#> GSM587199 3 0.000 0.965 0.000 0 1.000
#> GSM587200 1 0.000 1.000 1.000 0 0.000
#> GSM587201 1 0.000 1.000 1.000 0 0.000
#> GSM587202 3 0.129 0.940 0.032 0 0.968
#> GSM198767 1 0.000 1.000 1.000 0 0.000
#> GSM198769 1 0.000 1.000 1.000 0 0.000
#> GSM198772 1 0.000 1.000 1.000 0 0.000
#> GSM198773 1 0.000 1.000 1.000 0 0.000
#> GSM198776 1 0.000 1.000 1.000 0 0.000
#> GSM198778 1 0.000 1.000 1.000 0 0.000
#> GSM198780 1 0.000 1.000 1.000 0 0.000
#> GSM198781 1 0.000 1.000 1.000 0 0.000
#> GSM198765 3 0.000 0.965 0.000 0 1.000
#> GSM198766 3 0.559 0.588 0.304 0 0.696
#> GSM198768 3 0.000 0.965 0.000 0 1.000
#> GSM198770 3 0.000 0.965 0.000 0 1.000
#> GSM198771 3 0.000 0.965 0.000 0 1.000
#> GSM198774 3 0.000 0.965 0.000 0 1.000
#> GSM198775 3 0.000 0.965 0.000 0 1.000
#> GSM198777 3 0.000 0.965 0.000 0 1.000
#> GSM198779 3 0.000 0.965 0.000 0 1.000
#> GSM587218 1 0.000 1.000 1.000 0 0.000
#> GSM587219 1 0.000 1.000 1.000 0 0.000
#> GSM587220 1 0.000 1.000 1.000 0 0.000
#> GSM587221 1 0.000 1.000 1.000 0 0.000
#> GSM587222 1 0.000 1.000 1.000 0 0.000
#> GSM587223 1 0.000 1.000 1.000 0 0.000
#> GSM587224 1 0.000 1.000 1.000 0 0.000
#> GSM587225 1 0.000 1.000 1.000 0 0.000
#> GSM587226 1 0.000 1.000 1.000 0 0.000
#> GSM587227 1 0.000 1.000 1.000 0 0.000
#> GSM587228 1 0.000 1.000 1.000 0 0.000
#> GSM587229 1 0.000 1.000 1.000 0 0.000
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM587155 2 0.0000 0.98052 0.000 1.000 0.000 0.000
#> GSM587156 2 0.0000 0.98052 0.000 1.000 0.000 0.000
#> GSM587157 2 0.0000 0.98052 0.000 1.000 0.000 0.000
#> GSM587158 2 0.0000 0.98052 0.000 1.000 0.000 0.000
#> GSM587159 2 0.0000 0.98052 0.000 1.000 0.000 0.000
#> GSM587160 2 0.0000 0.98052 0.000 1.000 0.000 0.000
#> GSM587161 2 0.0000 0.98052 0.000 1.000 0.000 0.000
#> GSM587162 2 0.0000 0.98052 0.000 1.000 0.000 0.000
#> GSM587163 2 0.0000 0.98052 0.000 1.000 0.000 0.000
#> GSM587164 2 0.0000 0.98052 0.000 1.000 0.000 0.000
#> GSM587165 2 0.0000 0.98052 0.000 1.000 0.000 0.000
#> GSM587166 2 0.0000 0.98052 0.000 1.000 0.000 0.000
#> GSM587167 2 0.0000 0.98052 0.000 1.000 0.000 0.000
#> GSM587168 2 0.0000 0.98052 0.000 1.000 0.000 0.000
#> GSM587169 2 0.0000 0.98052 0.000 1.000 0.000 0.000
#> GSM587170 2 0.0000 0.98052 0.000 1.000 0.000 0.000
#> GSM587171 2 0.0000 0.98052 0.000 1.000 0.000 0.000
#> GSM587172 2 0.0000 0.98052 0.000 1.000 0.000 0.000
#> GSM587173 2 0.0000 0.98052 0.000 1.000 0.000 0.000
#> GSM587174 2 0.0000 0.98052 0.000 1.000 0.000 0.000
#> GSM587175 2 0.0000 0.98052 0.000 1.000 0.000 0.000
#> GSM587176 2 0.0000 0.98052 0.000 1.000 0.000 0.000
#> GSM587177 2 0.0000 0.98052 0.000 1.000 0.000 0.000
#> GSM587178 2 0.0000 0.98052 0.000 1.000 0.000 0.000
#> GSM587179 2 0.0000 0.98052 0.000 1.000 0.000 0.000
#> GSM587180 2 0.0000 0.98052 0.000 1.000 0.000 0.000
#> GSM587181 2 0.0000 0.98052 0.000 1.000 0.000 0.000
#> GSM587182 2 0.0000 0.98052 0.000 1.000 0.000 0.000
#> GSM587183 2 0.0000 0.98052 0.000 1.000 0.000 0.000
#> GSM587184 2 0.0000 0.98052 0.000 1.000 0.000 0.000
#> GSM587185 2 0.0000 0.98052 0.000 1.000 0.000 0.000
#> GSM587186 2 0.0000 0.98052 0.000 1.000 0.000 0.000
#> GSM587187 2 0.3610 0.77294 0.000 0.800 0.000 0.200
#> GSM587188 2 0.3649 0.76863 0.000 0.796 0.000 0.204
#> GSM587189 2 0.4867 0.68057 0.000 0.736 0.032 0.232
#> GSM587190 4 0.4998 0.03195 0.000 0.000 0.488 0.512
#> GSM587203 1 0.0000 0.95775 1.000 0.000 0.000 0.000
#> GSM587204 1 0.1824 0.93797 0.936 0.000 0.060 0.004
#> GSM587205 1 0.0188 0.95766 0.996 0.000 0.000 0.004
#> GSM587206 1 0.0188 0.95766 0.996 0.000 0.000 0.004
#> GSM587207 1 0.0188 0.95766 0.996 0.000 0.000 0.004
#> GSM587208 1 0.0188 0.95766 0.996 0.000 0.000 0.004
#> GSM587209 1 0.0188 0.95801 0.996 0.000 0.000 0.004
#> GSM587210 1 0.2255 0.93349 0.920 0.000 0.068 0.012
#> GSM587211 1 0.0188 0.95801 0.996 0.000 0.000 0.004
#> GSM587212 1 0.2741 0.91117 0.892 0.000 0.096 0.012
#> GSM587213 1 0.0188 0.95801 0.996 0.000 0.000 0.004
#> GSM587214 1 0.1743 0.94005 0.940 0.000 0.056 0.004
#> GSM587215 1 0.1824 0.93797 0.936 0.000 0.060 0.004
#> GSM587216 1 0.1824 0.93797 0.936 0.000 0.060 0.004
#> GSM587217 1 0.1743 0.94005 0.940 0.000 0.056 0.004
#> GSM587191 4 0.4661 0.13046 0.000 0.000 0.348 0.652
#> GSM587192 3 0.5016 0.06384 0.004 0.000 0.600 0.396
#> GSM587193 4 0.6603 0.11163 0.100 0.000 0.328 0.572
#> GSM587194 3 0.4855 0.05803 0.000 0.000 0.600 0.400
#> GSM587195 4 0.5000 -0.03098 0.000 0.000 0.496 0.504
#> GSM587196 3 0.3311 0.21921 0.000 0.000 0.828 0.172
#> GSM587197 4 0.4989 0.00663 0.000 0.000 0.472 0.528
#> GSM587198 3 0.4994 -0.14173 0.000 0.000 0.520 0.480
#> GSM587199 3 0.0000 0.23968 0.000 0.000 1.000 0.000
#> GSM587200 1 0.0657 0.95489 0.984 0.000 0.004 0.012
#> GSM587201 1 0.0657 0.95489 0.984 0.000 0.004 0.012
#> GSM587202 3 0.7315 0.03219 0.216 0.000 0.532 0.252
#> GSM198767 1 0.0188 0.95801 0.996 0.000 0.000 0.004
#> GSM198769 1 0.0188 0.95801 0.996 0.000 0.000 0.004
#> GSM198772 1 0.0188 0.95801 0.996 0.000 0.000 0.004
#> GSM198773 1 0.0188 0.95801 0.996 0.000 0.000 0.004
#> GSM198776 1 0.1824 0.93797 0.936 0.000 0.060 0.004
#> GSM198778 1 0.2255 0.93349 0.920 0.000 0.068 0.012
#> GSM198780 1 0.2741 0.91117 0.892 0.000 0.096 0.012
#> GSM198781 1 0.1743 0.94005 0.940 0.000 0.056 0.004
#> GSM198765 4 0.4776 0.11266 0.000 0.000 0.376 0.624
#> GSM198766 4 0.6898 0.05883 0.116 0.000 0.360 0.524
#> GSM198768 3 0.5000 -0.15731 0.000 0.000 0.504 0.496
#> GSM198770 4 0.4989 0.00663 0.000 0.000 0.472 0.528
#> GSM198771 3 0.4989 -0.13211 0.000 0.000 0.528 0.472
#> GSM198774 3 0.5016 0.06384 0.004 0.000 0.600 0.396
#> GSM198775 3 0.4855 0.05803 0.000 0.000 0.600 0.400
#> GSM198777 3 0.3311 0.21921 0.000 0.000 0.828 0.172
#> GSM198779 3 0.0000 0.23968 0.000 0.000 1.000 0.000
#> GSM587218 1 0.1557 0.94847 0.944 0.000 0.000 0.056
#> GSM587219 1 0.1557 0.94847 0.944 0.000 0.000 0.056
#> GSM587220 1 0.1557 0.94847 0.944 0.000 0.000 0.056
#> GSM587221 1 0.1557 0.94847 0.944 0.000 0.000 0.056
#> GSM587222 1 0.1557 0.94847 0.944 0.000 0.000 0.056
#> GSM587223 1 0.1557 0.94847 0.944 0.000 0.000 0.056
#> GSM587224 1 0.1557 0.94847 0.944 0.000 0.000 0.056
#> GSM587225 1 0.1557 0.94847 0.944 0.000 0.000 0.056
#> GSM587226 1 0.1557 0.94847 0.944 0.000 0.000 0.056
#> GSM587227 1 0.1557 0.94847 0.944 0.000 0.000 0.056
#> GSM587228 1 0.1557 0.94847 0.944 0.000 0.000 0.056
#> GSM587229 1 0.1557 0.94847 0.944 0.000 0.000 0.056
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM587155 2 0.0000 1.0000 0.000 1.000 0.000 0.000 0.000
#> GSM587156 2 0.0000 1.0000 0.000 1.000 0.000 0.000 0.000
#> GSM587157 2 0.0000 1.0000 0.000 1.000 0.000 0.000 0.000
#> GSM587158 2 0.0000 1.0000 0.000 1.000 0.000 0.000 0.000
#> GSM587159 2 0.0000 1.0000 0.000 1.000 0.000 0.000 0.000
#> GSM587160 2 0.0000 1.0000 0.000 1.000 0.000 0.000 0.000
#> GSM587161 2 0.0000 1.0000 0.000 1.000 0.000 0.000 0.000
#> GSM587162 2 0.0000 1.0000 0.000 1.000 0.000 0.000 0.000
#> GSM587163 2 0.0000 1.0000 0.000 1.000 0.000 0.000 0.000
#> GSM587164 2 0.0000 1.0000 0.000 1.000 0.000 0.000 0.000
#> GSM587165 2 0.0000 1.0000 0.000 1.000 0.000 0.000 0.000
#> GSM587166 2 0.0000 1.0000 0.000 1.000 0.000 0.000 0.000
#> GSM587167 2 0.0000 1.0000 0.000 1.000 0.000 0.000 0.000
#> GSM587168 2 0.0000 1.0000 0.000 1.000 0.000 0.000 0.000
#> GSM587169 2 0.0000 1.0000 0.000 1.000 0.000 0.000 0.000
#> GSM587170 2 0.0000 1.0000 0.000 1.000 0.000 0.000 0.000
#> GSM587171 2 0.0000 1.0000 0.000 1.000 0.000 0.000 0.000
#> GSM587172 2 0.0000 1.0000 0.000 1.000 0.000 0.000 0.000
#> GSM587173 2 0.0000 1.0000 0.000 1.000 0.000 0.000 0.000
#> GSM587174 2 0.0000 1.0000 0.000 1.000 0.000 0.000 0.000
#> GSM587175 2 0.0000 1.0000 0.000 1.000 0.000 0.000 0.000
#> GSM587176 2 0.0000 1.0000 0.000 1.000 0.000 0.000 0.000
#> GSM587177 2 0.0000 1.0000 0.000 1.000 0.000 0.000 0.000
#> GSM587178 2 0.0000 1.0000 0.000 1.000 0.000 0.000 0.000
#> GSM587179 2 0.0000 1.0000 0.000 1.000 0.000 0.000 0.000
#> GSM587180 2 0.0000 1.0000 0.000 1.000 0.000 0.000 0.000
#> GSM587181 2 0.0000 1.0000 0.000 1.000 0.000 0.000 0.000
#> GSM587182 2 0.0000 1.0000 0.000 1.000 0.000 0.000 0.000
#> GSM587183 2 0.0000 1.0000 0.000 1.000 0.000 0.000 0.000
#> GSM587184 2 0.0000 1.0000 0.000 1.000 0.000 0.000 0.000
#> GSM587185 2 0.0000 1.0000 0.000 1.000 0.000 0.000 0.000
#> GSM587186 2 0.0000 1.0000 0.000 1.000 0.000 0.000 0.000
#> GSM587187 5 0.7235 0.7297 0.000 0.352 0.032 0.200 0.416
#> GSM587188 5 0.7235 0.7297 0.000 0.352 0.032 0.200 0.416
#> GSM587189 5 0.7363 0.6991 0.000 0.288 0.032 0.264 0.416
#> GSM587190 5 0.5816 -0.0724 0.000 0.000 0.092 0.440 0.468
#> GSM587203 1 0.0703 0.8236 0.976 0.000 0.000 0.000 0.024
#> GSM587204 1 0.1364 0.8157 0.952 0.000 0.036 0.000 0.012
#> GSM587205 1 0.1732 0.8065 0.920 0.000 0.000 0.000 0.080
#> GSM587206 1 0.1732 0.8065 0.920 0.000 0.000 0.000 0.080
#> GSM587207 1 0.1732 0.8065 0.920 0.000 0.000 0.000 0.080
#> GSM587208 1 0.1732 0.8065 0.920 0.000 0.000 0.000 0.080
#> GSM587209 1 0.0162 0.8242 0.996 0.000 0.000 0.000 0.004
#> GSM587210 1 0.4394 0.7634 0.764 0.000 0.136 0.000 0.100
#> GSM587211 1 0.0162 0.8242 0.996 0.000 0.000 0.000 0.004
#> GSM587212 1 0.4428 0.7587 0.760 0.000 0.144 0.000 0.096
#> GSM587213 1 0.0162 0.8242 0.996 0.000 0.000 0.000 0.004
#> GSM587214 1 0.1300 0.8153 0.956 0.000 0.028 0.000 0.016
#> GSM587215 1 0.1386 0.8149 0.952 0.000 0.032 0.000 0.016
#> GSM587216 1 0.1469 0.8143 0.948 0.000 0.036 0.000 0.016
#> GSM587217 1 0.1300 0.8153 0.956 0.000 0.028 0.000 0.016
#> GSM587191 3 0.6092 0.4406 0.000 0.000 0.564 0.256 0.180
#> GSM587192 3 0.1617 0.6637 0.012 0.000 0.948 0.020 0.020
#> GSM587193 3 0.6551 0.5357 0.136 0.000 0.632 0.144 0.088
#> GSM587194 3 0.1809 0.6544 0.000 0.000 0.928 0.012 0.060
#> GSM587195 4 0.0324 0.7551 0.000 0.000 0.004 0.992 0.004
#> GSM587196 4 0.4675 0.3561 0.000 0.000 0.380 0.600 0.020
#> GSM587197 4 0.0671 0.7493 0.000 0.000 0.004 0.980 0.016
#> GSM587198 4 0.2124 0.7387 0.000 0.000 0.004 0.900 0.096
#> GSM587199 3 0.5488 0.2439 0.000 0.000 0.608 0.300 0.092
#> GSM587200 1 0.3562 0.7220 0.788 0.000 0.000 0.016 0.196
#> GSM587201 1 0.3381 0.7197 0.808 0.000 0.000 0.016 0.176
#> GSM587202 4 0.6311 0.4294 0.188 0.000 0.020 0.600 0.192
#> GSM198767 1 0.0000 0.8245 1.000 0.000 0.000 0.000 0.000
#> GSM198769 1 0.0162 0.8242 0.996 0.000 0.000 0.000 0.004
#> GSM198772 1 0.0162 0.8242 0.996 0.000 0.000 0.000 0.004
#> GSM198773 1 0.0162 0.8242 0.996 0.000 0.000 0.000 0.004
#> GSM198776 1 0.1364 0.8157 0.952 0.000 0.036 0.000 0.012
#> GSM198778 1 0.4394 0.7634 0.764 0.000 0.136 0.000 0.100
#> GSM198780 1 0.4428 0.7587 0.760 0.000 0.144 0.000 0.096
#> GSM198781 1 0.1300 0.8153 0.956 0.000 0.028 0.000 0.016
#> GSM198765 3 0.5490 0.5101 0.000 0.000 0.644 0.228 0.128
#> GSM198766 3 0.6411 0.5277 0.172 0.000 0.640 0.104 0.084
#> GSM198768 4 0.0290 0.7574 0.000 0.000 0.008 0.992 0.000
#> GSM198770 4 0.0671 0.7493 0.000 0.000 0.004 0.980 0.016
#> GSM198771 4 0.2573 0.7313 0.000 0.000 0.016 0.880 0.104
#> GSM198774 3 0.1617 0.6637 0.012 0.000 0.948 0.020 0.020
#> GSM198775 3 0.1809 0.6544 0.000 0.000 0.928 0.012 0.060
#> GSM198777 4 0.4663 0.3640 0.000 0.000 0.376 0.604 0.020
#> GSM198779 3 0.5488 0.2439 0.000 0.000 0.608 0.300 0.092
#> GSM587218 1 0.4030 0.7371 0.648 0.000 0.000 0.000 0.352
#> GSM587219 1 0.4030 0.7371 0.648 0.000 0.000 0.000 0.352
#> GSM587220 1 0.4030 0.7371 0.648 0.000 0.000 0.000 0.352
#> GSM587221 1 0.4030 0.7371 0.648 0.000 0.000 0.000 0.352
#> GSM587222 1 0.4030 0.7371 0.648 0.000 0.000 0.000 0.352
#> GSM587223 1 0.4030 0.7371 0.648 0.000 0.000 0.000 0.352
#> GSM587224 1 0.4030 0.7371 0.648 0.000 0.000 0.000 0.352
#> GSM587225 1 0.4045 0.7363 0.644 0.000 0.000 0.000 0.356
#> GSM587226 1 0.4030 0.7371 0.648 0.000 0.000 0.000 0.352
#> GSM587227 1 0.4045 0.7363 0.644 0.000 0.000 0.000 0.356
#> GSM587228 1 0.4045 0.7363 0.644 0.000 0.000 0.000 0.356
#> GSM587229 1 0.4045 0.7363 0.644 0.000 0.000 0.000 0.356
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM587155 2 0.0000 0.9994 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587156 2 0.0000 0.9994 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587157 2 0.0000 0.9994 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587158 2 0.0000 0.9994 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587159 2 0.0000 0.9994 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587160 2 0.0000 0.9994 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587161 2 0.0000 0.9994 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587162 2 0.0000 0.9994 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587163 2 0.0000 0.9994 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587164 2 0.0000 0.9994 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587165 2 0.0000 0.9994 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587166 2 0.0000 0.9994 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587167 2 0.0000 0.9994 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587168 2 0.0000 0.9994 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587169 2 0.0000 0.9994 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587170 2 0.0000 0.9994 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587171 2 0.0000 0.9994 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587172 2 0.0000 0.9994 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587173 2 0.0000 0.9994 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587174 2 0.0000 0.9994 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587175 2 0.0000 0.9994 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587176 2 0.0000 0.9994 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587177 2 0.0000 0.9994 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587178 2 0.0000 0.9994 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587179 2 0.0000 0.9994 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587180 2 0.0000 0.9994 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587181 2 0.0000 0.9994 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587182 2 0.0000 0.9994 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587183 2 0.0146 0.9960 0.000 0.996 0.000 0.000 0.000 0.004
#> GSM587184 2 0.0146 0.9960 0.000 0.996 0.000 0.000 0.000 0.004
#> GSM587185 2 0.0146 0.9960 0.000 0.996 0.000 0.000 0.000 0.004
#> GSM587186 2 0.0146 0.9960 0.000 0.996 0.000 0.000 0.000 0.004
#> GSM587187 6 0.3010 0.8661 0.000 0.148 0.020 0.000 0.004 0.828
#> GSM587188 6 0.2971 0.8714 0.000 0.144 0.020 0.000 0.004 0.832
#> GSM587189 6 0.3072 0.8678 0.000 0.124 0.036 0.000 0.004 0.836
#> GSM587190 6 0.1950 0.6445 0.000 0.000 0.064 0.000 0.024 0.912
#> GSM587203 1 0.1141 0.7403 0.948 0.000 0.000 0.052 0.000 0.000
#> GSM587204 1 0.2009 0.7535 0.908 0.000 0.000 0.068 0.024 0.000
#> GSM587205 1 0.2964 0.7136 0.868 0.000 0.036 0.060 0.000 0.036
#> GSM587206 1 0.2964 0.7136 0.868 0.000 0.036 0.060 0.000 0.036
#> GSM587207 1 0.2964 0.7136 0.868 0.000 0.036 0.060 0.000 0.036
#> GSM587208 1 0.2964 0.7136 0.868 0.000 0.036 0.060 0.000 0.036
#> GSM587209 1 0.0363 0.7674 0.988 0.000 0.000 0.012 0.000 0.000
#> GSM587210 1 0.5411 0.1320 0.572 0.000 0.000 0.260 0.168 0.000
#> GSM587211 1 0.0363 0.7674 0.988 0.000 0.000 0.012 0.000 0.000
#> GSM587212 1 0.5579 0.1287 0.548 0.000 0.000 0.248 0.204 0.000
#> GSM587213 1 0.0458 0.7674 0.984 0.000 0.000 0.016 0.000 0.000
#> GSM587214 1 0.1594 0.7614 0.932 0.000 0.000 0.052 0.016 0.000
#> GSM587215 1 0.1765 0.7576 0.924 0.000 0.000 0.052 0.024 0.000
#> GSM587216 1 0.1829 0.7581 0.920 0.000 0.000 0.056 0.024 0.000
#> GSM587217 1 0.1594 0.7623 0.932 0.000 0.000 0.052 0.016 0.000
#> GSM587191 5 0.7142 0.2045 0.000 0.000 0.100 0.188 0.364 0.348
#> GSM587192 5 0.3110 0.5579 0.020 0.000 0.000 0.128 0.836 0.016
#> GSM587193 5 0.8252 0.3259 0.136 0.000 0.092 0.200 0.404 0.168
#> GSM587194 5 0.2216 0.5567 0.000 0.000 0.024 0.016 0.908 0.052
#> GSM587195 3 0.2706 0.7804 0.000 0.000 0.832 0.000 0.008 0.160
#> GSM587196 5 0.4962 0.0840 0.004 0.000 0.464 0.004 0.484 0.044
#> GSM587197 3 0.3198 0.7664 0.000 0.000 0.796 0.008 0.008 0.188
#> GSM587198 3 0.3330 0.7203 0.000 0.000 0.828 0.108 0.008 0.056
#> GSM587199 5 0.5031 0.4642 0.004 0.000 0.160 0.056 0.712 0.068
#> GSM587200 1 0.6200 0.3852 0.576 0.000 0.152 0.216 0.004 0.052
#> GSM587201 1 0.5979 0.4335 0.612 0.000 0.152 0.180 0.004 0.052
#> GSM587202 3 0.5779 0.4653 0.108 0.000 0.660 0.160 0.016 0.056
#> GSM198767 1 0.0547 0.7641 0.980 0.000 0.000 0.020 0.000 0.000
#> GSM198769 1 0.0363 0.7674 0.988 0.000 0.000 0.012 0.000 0.000
#> GSM198772 1 0.0363 0.7674 0.988 0.000 0.000 0.012 0.000 0.000
#> GSM198773 1 0.0458 0.7674 0.984 0.000 0.000 0.016 0.000 0.000
#> GSM198776 1 0.2009 0.7535 0.908 0.000 0.000 0.068 0.024 0.000
#> GSM198778 1 0.5411 0.1320 0.572 0.000 0.000 0.260 0.168 0.000
#> GSM198780 1 0.5579 0.1287 0.548 0.000 0.000 0.248 0.204 0.000
#> GSM198781 1 0.1594 0.7614 0.932 0.000 0.000 0.052 0.016 0.000
#> GSM198765 5 0.7086 0.2773 0.000 0.000 0.100 0.188 0.416 0.296
#> GSM198766 5 0.8207 0.3333 0.164 0.000 0.092 0.200 0.412 0.132
#> GSM198768 3 0.2730 0.7803 0.000 0.000 0.836 0.000 0.012 0.152
#> GSM198770 3 0.3198 0.7664 0.000 0.000 0.796 0.008 0.008 0.188
#> GSM198771 3 0.3269 0.7178 0.000 0.000 0.832 0.108 0.008 0.052
#> GSM198774 5 0.3110 0.5579 0.020 0.000 0.000 0.128 0.836 0.016
#> GSM198775 5 0.2216 0.5567 0.000 0.000 0.024 0.016 0.908 0.052
#> GSM198777 5 0.5015 0.0667 0.004 0.000 0.468 0.004 0.476 0.048
#> GSM198779 5 0.4997 0.4680 0.004 0.000 0.156 0.056 0.716 0.068
#> GSM587218 4 0.3765 0.9749 0.404 0.000 0.000 0.596 0.000 0.000
#> GSM587219 4 0.3765 0.9749 0.404 0.000 0.000 0.596 0.000 0.000
#> GSM587220 4 0.3765 0.9749 0.404 0.000 0.000 0.596 0.000 0.000
#> GSM587221 4 0.3765 0.9749 0.404 0.000 0.000 0.596 0.000 0.000
#> GSM587222 4 0.3765 0.9749 0.404 0.000 0.000 0.596 0.000 0.000
#> GSM587223 4 0.3765 0.9749 0.404 0.000 0.000 0.596 0.000 0.000
#> GSM587224 4 0.3765 0.9749 0.404 0.000 0.000 0.596 0.000 0.000
#> GSM587225 4 0.3756 0.9717 0.400 0.000 0.000 0.600 0.000 0.000
#> GSM587226 4 0.3765 0.9749 0.404 0.000 0.000 0.596 0.000 0.000
#> GSM587227 4 0.3672 0.9270 0.368 0.000 0.000 0.632 0.000 0.000
#> GSM587228 4 0.3672 0.9270 0.368 0.000 0.000 0.632 0.000 0.000
#> GSM587229 4 0.3672 0.9270 0.368 0.000 0.000 0.632 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 specimen(p) k
#> ATC:skmeans 92 7.21e-17 2
#> ATC:skmeans 92 2.19e-29 3
#> ATC:skmeans 72 3.93e-14 4
#> ATC:skmeans 85 5.98e-38 5
#> ATC:skmeans 77 6.99e-48 6
If matrix rows can be associated to genes, consider to use functional_enrichment(res,
...) to perform function enrichment for the signature genes. See this vignette for more detailed explanations.
The object with results only for a single top-value method and a single partition method can be extracted as:
res = res_list["ATC", "pam"]
# you can also extract it by
# res = res_list["ATC:pam"]
A summary of res and all the functions that can be applied to it:
res
#> A 'ConsensusPartition' object with k = 2, 3, 4, 5, 6.
#> On a matrix with 51941 rows and 92 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 6.
#>
#> 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 0.987 0.995 0.4793 0.523 0.523
#> 3 3 1.000 0.997 0.998 0.3933 0.777 0.587
#> 4 4 1.000 0.972 0.987 0.0914 0.937 0.810
#> 5 5 1.000 0.967 0.983 0.0459 0.960 0.853
#> 6 6 0.925 0.910 0.929 0.0330 0.976 0.897
suggest_best_k() suggests the best \(k\) based on these statistics. The rules are as follows:
suggest_best_k(res)
#> [1] 6
#> attr(,"optional")
#> [1] 2 3 4 5
There is also optional best \(k\) = 2 3 4 5 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
#> GSM587155 2 0.00 1.000 0.00 1.00
#> GSM587156 2 0.00 1.000 0.00 1.00
#> GSM587157 2 0.00 1.000 0.00 1.00
#> GSM587158 2 0.00 1.000 0.00 1.00
#> GSM587159 2 0.00 1.000 0.00 1.00
#> GSM587160 2 0.00 1.000 0.00 1.00
#> GSM587161 2 0.00 1.000 0.00 1.00
#> GSM587162 2 0.00 1.000 0.00 1.00
#> GSM587163 2 0.00 1.000 0.00 1.00
#> GSM587164 2 0.00 1.000 0.00 1.00
#> GSM587165 2 0.00 1.000 0.00 1.00
#> GSM587166 2 0.00 1.000 0.00 1.00
#> GSM587167 2 0.00 1.000 0.00 1.00
#> GSM587168 2 0.00 1.000 0.00 1.00
#> GSM587169 2 0.00 1.000 0.00 1.00
#> GSM587170 2 0.00 1.000 0.00 1.00
#> GSM587171 2 0.00 1.000 0.00 1.00
#> GSM587172 2 0.00 1.000 0.00 1.00
#> GSM587173 2 0.00 1.000 0.00 1.00
#> GSM587174 2 0.00 1.000 0.00 1.00
#> GSM587175 2 0.00 1.000 0.00 1.00
#> GSM587176 2 0.00 1.000 0.00 1.00
#> GSM587177 2 0.00 1.000 0.00 1.00
#> GSM587178 2 0.00 1.000 0.00 1.00
#> GSM587179 2 0.00 1.000 0.00 1.00
#> GSM587180 2 0.00 1.000 0.00 1.00
#> GSM587181 2 0.00 1.000 0.00 1.00
#> GSM587182 2 0.00 1.000 0.00 1.00
#> GSM587183 2 0.00 1.000 0.00 1.00
#> GSM587184 2 0.00 1.000 0.00 1.00
#> GSM587185 2 0.00 1.000 0.00 1.00
#> GSM587186 2 0.00 1.000 0.00 1.00
#> GSM587187 2 0.00 1.000 0.00 1.00
#> GSM587188 2 0.00 1.000 0.00 1.00
#> GSM587189 2 0.00 1.000 0.00 1.00
#> GSM587190 1 0.99 0.214 0.56 0.44
#> GSM587203 1 0.00 0.992 1.00 0.00
#> GSM587204 1 0.00 0.992 1.00 0.00
#> GSM587205 1 0.00 0.992 1.00 0.00
#> GSM587206 1 0.00 0.992 1.00 0.00
#> GSM587207 1 0.00 0.992 1.00 0.00
#> GSM587208 1 0.00 0.992 1.00 0.00
#> GSM587209 1 0.00 0.992 1.00 0.00
#> GSM587210 1 0.00 0.992 1.00 0.00
#> GSM587211 1 0.00 0.992 1.00 0.00
#> GSM587212 1 0.00 0.992 1.00 0.00
#> GSM587213 1 0.00 0.992 1.00 0.00
#> GSM587214 1 0.00 0.992 1.00 0.00
#> GSM587215 1 0.00 0.992 1.00 0.00
#> GSM587216 1 0.00 0.992 1.00 0.00
#> GSM587217 1 0.00 0.992 1.00 0.00
#> GSM587191 1 0.00 0.992 1.00 0.00
#> GSM587192 1 0.00 0.992 1.00 0.00
#> GSM587193 1 0.00 0.992 1.00 0.00
#> GSM587194 1 0.00 0.992 1.00 0.00
#> GSM587195 1 0.00 0.992 1.00 0.00
#> GSM587196 1 0.00 0.992 1.00 0.00
#> GSM587197 1 0.00 0.992 1.00 0.00
#> GSM587198 1 0.00 0.992 1.00 0.00
#> GSM587199 1 0.00 0.992 1.00 0.00
#> GSM587200 1 0.00 0.992 1.00 0.00
#> GSM587201 1 0.00 0.992 1.00 0.00
#> GSM587202 1 0.00 0.992 1.00 0.00
#> GSM198767 1 0.00 0.992 1.00 0.00
#> GSM198769 1 0.00 0.992 1.00 0.00
#> GSM198772 1 0.00 0.992 1.00 0.00
#> GSM198773 1 0.00 0.992 1.00 0.00
#> GSM198776 1 0.00 0.992 1.00 0.00
#> GSM198778 1 0.00 0.992 1.00 0.00
#> GSM198780 1 0.00 0.992 1.00 0.00
#> GSM198781 1 0.00 0.992 1.00 0.00
#> GSM198765 1 0.00 0.992 1.00 0.00
#> GSM198766 1 0.00 0.992 1.00 0.00
#> GSM198768 1 0.00 0.992 1.00 0.00
#> GSM198770 1 0.00 0.992 1.00 0.00
#> GSM198771 1 0.00 0.992 1.00 0.00
#> GSM198774 1 0.00 0.992 1.00 0.00
#> GSM198775 1 0.00 0.992 1.00 0.00
#> GSM198777 1 0.00 0.992 1.00 0.00
#> GSM198779 1 0.00 0.992 1.00 0.00
#> GSM587218 1 0.00 0.992 1.00 0.00
#> GSM587219 1 0.00 0.992 1.00 0.00
#> GSM587220 1 0.00 0.992 1.00 0.00
#> GSM587221 1 0.00 0.992 1.00 0.00
#> GSM587222 1 0.00 0.992 1.00 0.00
#> GSM587223 1 0.00 0.992 1.00 0.00
#> GSM587224 1 0.00 0.992 1.00 0.00
#> GSM587225 1 0.00 0.992 1.00 0.00
#> GSM587226 1 0.00 0.992 1.00 0.00
#> GSM587227 1 0.00 0.992 1.00 0.00
#> GSM587228 1 0.00 0.992 1.00 0.00
#> GSM587229 1 0.00 0.992 1.00 0.00
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM587155 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587156 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587157 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587158 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587159 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587160 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587161 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587162 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587163 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587164 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587165 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587166 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587167 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587168 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587169 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587170 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587171 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587172 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587173 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587174 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587175 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587176 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587177 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587178 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587179 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587180 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587181 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587182 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587183 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587184 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587185 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587186 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587187 3 0.0424 0.989 0.000 0.008 0.992
#> GSM587188 3 0.0424 0.989 0.000 0.008 0.992
#> GSM587189 3 0.0000 0.995 0.000 0.000 1.000
#> GSM587190 3 0.0000 0.995 0.000 0.000 1.000
#> GSM587203 1 0.0000 0.999 1.000 0.000 0.000
#> GSM587204 1 0.0000 0.999 1.000 0.000 0.000
#> GSM587205 1 0.0000 0.999 1.000 0.000 0.000
#> GSM587206 1 0.0000 0.999 1.000 0.000 0.000
#> GSM587207 1 0.0000 0.999 1.000 0.000 0.000
#> GSM587208 1 0.0000 0.999 1.000 0.000 0.000
#> GSM587209 1 0.0000 0.999 1.000 0.000 0.000
#> GSM587210 1 0.0000 0.999 1.000 0.000 0.000
#> GSM587211 1 0.1163 0.972 0.972 0.000 0.028
#> GSM587212 1 0.0000 0.999 1.000 0.000 0.000
#> GSM587213 1 0.0000 0.999 1.000 0.000 0.000
#> GSM587214 1 0.0000 0.999 1.000 0.000 0.000
#> GSM587215 1 0.0747 0.984 0.984 0.000 0.016
#> GSM587216 1 0.0000 0.999 1.000 0.000 0.000
#> GSM587217 1 0.0000 0.999 1.000 0.000 0.000
#> GSM587191 3 0.0000 0.995 0.000 0.000 1.000
#> GSM587192 3 0.1529 0.962 0.040 0.000 0.960
#> GSM587193 3 0.0000 0.995 0.000 0.000 1.000
#> GSM587194 3 0.0000 0.995 0.000 0.000 1.000
#> GSM587195 3 0.0000 0.995 0.000 0.000 1.000
#> GSM587196 3 0.0000 0.995 0.000 0.000 1.000
#> GSM587197 3 0.0000 0.995 0.000 0.000 1.000
#> GSM587198 3 0.0000 0.995 0.000 0.000 1.000
#> GSM587199 3 0.0000 0.995 0.000 0.000 1.000
#> GSM587200 3 0.1031 0.977 0.024 0.000 0.976
#> GSM587201 3 0.1031 0.977 0.024 0.000 0.976
#> GSM587202 3 0.0000 0.995 0.000 0.000 1.000
#> GSM198767 1 0.0000 0.999 1.000 0.000 0.000
#> GSM198769 1 0.0000 0.999 1.000 0.000 0.000
#> GSM198772 1 0.0000 0.999 1.000 0.000 0.000
#> GSM198773 1 0.0000 0.999 1.000 0.000 0.000
#> GSM198776 1 0.0000 0.999 1.000 0.000 0.000
#> GSM198778 1 0.0000 0.999 1.000 0.000 0.000
#> GSM198780 1 0.0000 0.999 1.000 0.000 0.000
#> GSM198781 1 0.0000 0.999 1.000 0.000 0.000
#> GSM198765 3 0.0000 0.995 0.000 0.000 1.000
#> GSM198766 3 0.0000 0.995 0.000 0.000 1.000
#> GSM198768 3 0.0000 0.995 0.000 0.000 1.000
#> GSM198770 3 0.0000 0.995 0.000 0.000 1.000
#> GSM198771 3 0.0000 0.995 0.000 0.000 1.000
#> GSM198774 3 0.0892 0.980 0.020 0.000 0.980
#> GSM198775 3 0.0000 0.995 0.000 0.000 1.000
#> GSM198777 3 0.0000 0.995 0.000 0.000 1.000
#> GSM198779 3 0.0000 0.995 0.000 0.000 1.000
#> GSM587218 1 0.0000 0.999 1.000 0.000 0.000
#> GSM587219 1 0.0000 0.999 1.000 0.000 0.000
#> GSM587220 1 0.0000 0.999 1.000 0.000 0.000
#> GSM587221 1 0.0000 0.999 1.000 0.000 0.000
#> GSM587222 1 0.0000 0.999 1.000 0.000 0.000
#> GSM587223 1 0.0000 0.999 1.000 0.000 0.000
#> GSM587224 1 0.0000 0.999 1.000 0.000 0.000
#> GSM587225 1 0.0000 0.999 1.000 0.000 0.000
#> GSM587226 1 0.0000 0.999 1.000 0.000 0.000
#> GSM587227 1 0.0000 0.999 1.000 0.000 0.000
#> GSM587228 1 0.0000 0.999 1.000 0.000 0.000
#> GSM587229 1 0.0000 0.999 1.000 0.000 0.000
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM587155 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587156 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587157 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587158 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587159 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587160 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587161 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587162 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587163 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587164 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587165 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587166 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587167 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587168 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587169 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587170 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587171 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587172 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587173 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587174 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587175 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587176 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587177 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587178 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587179 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587180 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587181 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587182 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587183 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587184 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587185 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587186 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587187 3 0.0336 0.986 0.000 0.008 0.992 0.000
#> GSM587188 3 0.0336 0.986 0.000 0.008 0.992 0.000
#> GSM587189 3 0.0000 0.993 0.000 0.000 1.000 0.000
#> GSM587190 3 0.0000 0.993 0.000 0.000 1.000 0.000
#> GSM587203 4 0.3400 0.778 0.180 0.000 0.000 0.820
#> GSM587204 1 0.0000 0.990 1.000 0.000 0.000 0.000
#> GSM587205 1 0.0000 0.990 1.000 0.000 0.000 0.000
#> GSM587206 1 0.0000 0.990 1.000 0.000 0.000 0.000
#> GSM587207 1 0.0000 0.990 1.000 0.000 0.000 0.000
#> GSM587208 1 0.0000 0.990 1.000 0.000 0.000 0.000
#> GSM587209 1 0.0000 0.990 1.000 0.000 0.000 0.000
#> GSM587210 1 0.0000 0.990 1.000 0.000 0.000 0.000
#> GSM587211 1 0.0000 0.990 1.000 0.000 0.000 0.000
#> GSM587212 1 0.0000 0.990 1.000 0.000 0.000 0.000
#> GSM587213 1 0.0000 0.990 1.000 0.000 0.000 0.000
#> GSM587214 1 0.0000 0.990 1.000 0.000 0.000 0.000
#> GSM587215 1 0.0000 0.990 1.000 0.000 0.000 0.000
#> GSM587216 1 0.0000 0.990 1.000 0.000 0.000 0.000
#> GSM587217 1 0.0000 0.990 1.000 0.000 0.000 0.000
#> GSM587191 3 0.0000 0.993 0.000 0.000 1.000 0.000
#> GSM587192 3 0.1118 0.960 0.036 0.000 0.964 0.000
#> GSM587193 3 0.0000 0.993 0.000 0.000 1.000 0.000
#> GSM587194 3 0.0000 0.993 0.000 0.000 1.000 0.000
#> GSM587195 3 0.0000 0.993 0.000 0.000 1.000 0.000
#> GSM587196 3 0.0000 0.993 0.000 0.000 1.000 0.000
#> GSM587197 3 0.0000 0.993 0.000 0.000 1.000 0.000
#> GSM587198 3 0.0000 0.993 0.000 0.000 1.000 0.000
#> GSM587199 3 0.0000 0.993 0.000 0.000 1.000 0.000
#> GSM587200 3 0.0921 0.968 0.028 0.000 0.972 0.000
#> GSM587201 3 0.0817 0.972 0.024 0.000 0.976 0.000
#> GSM587202 3 0.0000 0.993 0.000 0.000 1.000 0.000
#> GSM198767 1 0.0000 0.990 1.000 0.000 0.000 0.000
#> GSM198769 1 0.0000 0.990 1.000 0.000 0.000 0.000
#> GSM198772 1 0.0000 0.990 1.000 0.000 0.000 0.000
#> GSM198773 1 0.0000 0.990 1.000 0.000 0.000 0.000
#> GSM198776 1 0.0000 0.990 1.000 0.000 0.000 0.000
#> GSM198778 1 0.0000 0.990 1.000 0.000 0.000 0.000
#> GSM198780 1 0.0000 0.990 1.000 0.000 0.000 0.000
#> GSM198781 1 0.0000 0.990 1.000 0.000 0.000 0.000
#> GSM198765 3 0.0000 0.993 0.000 0.000 1.000 0.000
#> GSM198766 3 0.0000 0.993 0.000 0.000 1.000 0.000
#> GSM198768 3 0.0000 0.993 0.000 0.000 1.000 0.000
#> GSM198770 3 0.0000 0.993 0.000 0.000 1.000 0.000
#> GSM198771 3 0.0000 0.993 0.000 0.000 1.000 0.000
#> GSM198774 3 0.1022 0.964 0.032 0.000 0.968 0.000
#> GSM198775 3 0.0000 0.993 0.000 0.000 1.000 0.000
#> GSM198777 3 0.0000 0.993 0.000 0.000 1.000 0.000
#> GSM198779 3 0.0000 0.993 0.000 0.000 1.000 0.000
#> GSM587218 4 0.0000 0.916 0.000 0.000 0.000 1.000
#> GSM587219 4 0.0000 0.916 0.000 0.000 0.000 1.000
#> GSM587220 4 0.0000 0.916 0.000 0.000 0.000 1.000
#> GSM587221 4 0.0000 0.916 0.000 0.000 0.000 1.000
#> GSM587222 4 0.0000 0.916 0.000 0.000 0.000 1.000
#> GSM587223 4 0.0000 0.916 0.000 0.000 0.000 1.000
#> GSM587224 4 0.0000 0.916 0.000 0.000 0.000 1.000
#> GSM587225 4 0.4866 0.395 0.404 0.000 0.000 0.596
#> GSM587226 4 0.0000 0.916 0.000 0.000 0.000 1.000
#> GSM587227 1 0.2530 0.866 0.888 0.000 0.000 0.112
#> GSM587228 4 0.3975 0.697 0.240 0.000 0.000 0.760
#> GSM587229 1 0.2345 0.882 0.900 0.000 0.000 0.100
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM587155 2 0.0000 1.000 0.000 1.000 0.000 0.000 0.000
#> GSM587156 2 0.0000 1.000 0.000 1.000 0.000 0.000 0.000
#> GSM587157 2 0.0000 1.000 0.000 1.000 0.000 0.000 0.000
#> GSM587158 2 0.0000 1.000 0.000 1.000 0.000 0.000 0.000
#> GSM587159 2 0.0000 1.000 0.000 1.000 0.000 0.000 0.000
#> GSM587160 2 0.0000 1.000 0.000 1.000 0.000 0.000 0.000
#> GSM587161 2 0.0000 1.000 0.000 1.000 0.000 0.000 0.000
#> GSM587162 2 0.0000 1.000 0.000 1.000 0.000 0.000 0.000
#> GSM587163 2 0.0000 1.000 0.000 1.000 0.000 0.000 0.000
#> GSM587164 2 0.0000 1.000 0.000 1.000 0.000 0.000 0.000
#> GSM587165 2 0.0000 1.000 0.000 1.000 0.000 0.000 0.000
#> GSM587166 2 0.0000 1.000 0.000 1.000 0.000 0.000 0.000
#> GSM587167 2 0.0000 1.000 0.000 1.000 0.000 0.000 0.000
#> GSM587168 2 0.0000 1.000 0.000 1.000 0.000 0.000 0.000
#> GSM587169 2 0.0000 1.000 0.000 1.000 0.000 0.000 0.000
#> GSM587170 2 0.0000 1.000 0.000 1.000 0.000 0.000 0.000
#> GSM587171 2 0.0000 1.000 0.000 1.000 0.000 0.000 0.000
#> GSM587172 2 0.0000 1.000 0.000 1.000 0.000 0.000 0.000
#> GSM587173 2 0.0000 1.000 0.000 1.000 0.000 0.000 0.000
#> GSM587174 2 0.0000 1.000 0.000 1.000 0.000 0.000 0.000
#> GSM587175 2 0.0000 1.000 0.000 1.000 0.000 0.000 0.000
#> GSM587176 2 0.0000 1.000 0.000 1.000 0.000 0.000 0.000
#> GSM587177 2 0.0000 1.000 0.000 1.000 0.000 0.000 0.000
#> GSM587178 2 0.0000 1.000 0.000 1.000 0.000 0.000 0.000
#> GSM587179 2 0.0000 1.000 0.000 1.000 0.000 0.000 0.000
#> GSM587180 2 0.0000 1.000 0.000 1.000 0.000 0.000 0.000
#> GSM587181 2 0.0000 1.000 0.000 1.000 0.000 0.000 0.000
#> GSM587182 2 0.0000 1.000 0.000 1.000 0.000 0.000 0.000
#> GSM587183 2 0.0000 1.000 0.000 1.000 0.000 0.000 0.000
#> GSM587184 2 0.0000 1.000 0.000 1.000 0.000 0.000 0.000
#> GSM587185 2 0.0000 1.000 0.000 1.000 0.000 0.000 0.000
#> GSM587186 2 0.0404 0.988 0.000 0.988 0.000 0.000 0.012
#> GSM587187 3 0.0703 0.984 0.000 0.000 0.976 0.000 0.024
#> GSM587188 3 0.0703 0.984 0.000 0.000 0.976 0.000 0.024
#> GSM587189 3 0.0703 0.984 0.000 0.000 0.976 0.000 0.024
#> GSM587190 3 0.0703 0.984 0.000 0.000 0.976 0.000 0.024
#> GSM587203 5 0.0865 0.965 0.004 0.000 0.000 0.024 0.972
#> GSM587204 1 0.0162 0.971 0.996 0.000 0.000 0.000 0.004
#> GSM587205 5 0.0794 0.980 0.028 0.000 0.000 0.000 0.972
#> GSM587206 5 0.0794 0.980 0.028 0.000 0.000 0.000 0.972
#> GSM587207 5 0.0794 0.980 0.028 0.000 0.000 0.000 0.972
#> GSM587208 5 0.0794 0.980 0.028 0.000 0.000 0.000 0.972
#> GSM587209 1 0.1478 0.933 0.936 0.000 0.000 0.000 0.064
#> GSM587210 1 0.0162 0.971 0.996 0.000 0.000 0.000 0.004
#> GSM587211 1 0.0865 0.961 0.972 0.000 0.004 0.000 0.024
#> GSM587212 1 0.0162 0.971 0.996 0.000 0.000 0.000 0.004
#> GSM587213 1 0.1121 0.949 0.956 0.000 0.000 0.000 0.044
#> GSM587214 1 0.0162 0.970 0.996 0.000 0.000 0.000 0.004
#> GSM587215 1 0.0000 0.971 1.000 0.000 0.000 0.000 0.000
#> GSM587216 1 0.0162 0.971 0.996 0.000 0.000 0.000 0.004
#> GSM587217 1 0.0162 0.971 0.996 0.000 0.000 0.000 0.004
#> GSM587191 3 0.0609 0.984 0.000 0.000 0.980 0.000 0.020
#> GSM587192 3 0.0324 0.988 0.004 0.000 0.992 0.000 0.004
#> GSM587193 3 0.0162 0.994 0.000 0.000 0.996 0.000 0.004
#> GSM587194 3 0.0000 0.994 0.000 0.000 1.000 0.000 0.000
#> GSM587195 3 0.0162 0.994 0.000 0.000 0.996 0.000 0.004
#> GSM587196 3 0.0000 0.994 0.000 0.000 1.000 0.000 0.000
#> GSM587197 3 0.0162 0.994 0.000 0.000 0.996 0.000 0.004
#> GSM587198 3 0.0162 0.994 0.000 0.000 0.996 0.000 0.004
#> GSM587199 3 0.0000 0.994 0.000 0.000 1.000 0.000 0.000
#> GSM587200 5 0.0794 0.960 0.000 0.000 0.028 0.000 0.972
#> GSM587201 5 0.0794 0.960 0.000 0.000 0.028 0.000 0.972
#> GSM587202 3 0.0162 0.994 0.000 0.000 0.996 0.000 0.004
#> GSM198767 5 0.0794 0.976 0.028 0.000 0.000 0.000 0.972
#> GSM198769 1 0.0510 0.966 0.984 0.000 0.000 0.000 0.016
#> GSM198772 1 0.0162 0.970 0.996 0.000 0.000 0.000 0.004
#> GSM198773 1 0.0510 0.966 0.984 0.000 0.000 0.000 0.016
#> GSM198776 1 0.0162 0.971 0.996 0.000 0.000 0.000 0.004
#> GSM198778 1 0.0162 0.971 0.996 0.000 0.000 0.000 0.004
#> GSM198780 1 0.0162 0.971 0.996 0.000 0.000 0.000 0.004
#> GSM198781 1 0.0162 0.970 0.996 0.000 0.000 0.000 0.004
#> GSM198765 3 0.0000 0.994 0.000 0.000 1.000 0.000 0.000
#> GSM198766 3 0.0000 0.994 0.000 0.000 1.000 0.000 0.000
#> GSM198768 3 0.0162 0.994 0.000 0.000 0.996 0.000 0.004
#> GSM198770 3 0.0162 0.994 0.000 0.000 0.996 0.000 0.004
#> GSM198771 3 0.0162 0.994 0.000 0.000 0.996 0.000 0.004
#> GSM198774 3 0.0000 0.994 0.000 0.000 1.000 0.000 0.000
#> GSM198775 3 0.0000 0.994 0.000 0.000 1.000 0.000 0.000
#> GSM198777 3 0.0000 0.994 0.000 0.000 1.000 0.000 0.000
#> GSM198779 3 0.0000 0.994 0.000 0.000 1.000 0.000 0.000
#> GSM587218 4 0.0000 0.913 0.000 0.000 0.000 1.000 0.000
#> GSM587219 4 0.0000 0.913 0.000 0.000 0.000 1.000 0.000
#> GSM587220 4 0.0000 0.913 0.000 0.000 0.000 1.000 0.000
#> GSM587221 4 0.0000 0.913 0.000 0.000 0.000 1.000 0.000
#> GSM587222 4 0.0000 0.913 0.000 0.000 0.000 1.000 0.000
#> GSM587223 4 0.0000 0.913 0.000 0.000 0.000 1.000 0.000
#> GSM587224 4 0.0000 0.913 0.000 0.000 0.000 1.000 0.000
#> GSM587225 4 0.4321 0.386 0.396 0.000 0.000 0.600 0.004
#> GSM587226 4 0.0000 0.913 0.000 0.000 0.000 1.000 0.000
#> GSM587227 1 0.3109 0.746 0.800 0.000 0.000 0.200 0.000
#> GSM587228 4 0.3305 0.698 0.224 0.000 0.000 0.776 0.000
#> GSM587229 1 0.2179 0.881 0.896 0.000 0.000 0.100 0.004
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM587155 2 0.0000 0.994 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587156 2 0.0000 0.994 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587157 2 0.0000 0.994 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587158 2 0.0000 0.994 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587159 2 0.0000 0.994 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587160 2 0.0000 0.994 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587161 2 0.0000 0.994 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587162 2 0.0000 0.994 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587163 2 0.0000 0.994 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587164 2 0.0000 0.994 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587165 2 0.0000 0.994 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587166 2 0.0000 0.994 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587167 2 0.0000 0.994 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587168 2 0.0000 0.994 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587169 2 0.0000 0.994 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587170 2 0.0000 0.994 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587171 2 0.0000 0.994 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587172 2 0.0000 0.994 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587173 2 0.0000 0.994 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587174 2 0.0000 0.994 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587175 2 0.0000 0.994 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587176 2 0.0000 0.994 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587177 2 0.0000 0.994 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587178 2 0.0000 0.994 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587179 2 0.0000 0.994 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587180 2 0.0000 0.994 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587181 2 0.0000 0.994 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587182 2 0.0000 0.994 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587183 2 0.0000 0.994 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587184 2 0.0000 0.994 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587185 2 0.0000 0.994 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587186 2 0.2871 0.772 0.000 0.804 0.000 0.000 0.192 0.004
#> GSM587187 3 0.3136 0.678 0.000 0.000 0.768 0.000 0.228 0.004
#> GSM587188 3 0.3360 0.662 0.000 0.000 0.732 0.000 0.264 0.004
#> GSM587189 3 0.3314 0.668 0.000 0.000 0.740 0.000 0.256 0.004
#> GSM587190 3 0.3468 0.656 0.000 0.000 0.712 0.000 0.284 0.004
#> GSM587203 6 0.0146 0.958 0.000 0.000 0.000 0.004 0.000 0.996
#> GSM587204 1 0.0000 0.930 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM587205 6 0.0146 0.961 0.004 0.000 0.000 0.000 0.000 0.996
#> GSM587206 6 0.0146 0.961 0.004 0.000 0.000 0.000 0.000 0.996
#> GSM587207 6 0.0146 0.961 0.004 0.000 0.000 0.000 0.000 0.996
#> GSM587208 6 0.0146 0.961 0.004 0.000 0.000 0.000 0.000 0.996
#> GSM587209 1 0.3078 0.899 0.836 0.000 0.000 0.000 0.108 0.056
#> GSM587210 1 0.0000 0.930 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM587211 1 0.3075 0.904 0.844 0.000 0.008 0.000 0.108 0.040
#> GSM587212 1 0.0000 0.930 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM587213 1 0.2822 0.907 0.852 0.000 0.000 0.000 0.108 0.040
#> GSM587214 1 0.1910 0.917 0.892 0.000 0.000 0.000 0.108 0.000
#> GSM587215 1 0.0260 0.930 0.992 0.000 0.000 0.000 0.008 0.000
#> GSM587216 1 0.0000 0.930 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM587217 1 0.0000 0.930 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM587191 5 0.2092 0.674 0.000 0.000 0.124 0.000 0.876 0.000
#> GSM587192 5 0.3782 0.889 0.004 0.000 0.360 0.000 0.636 0.000
#> GSM587193 5 0.3592 0.875 0.000 0.000 0.344 0.000 0.656 0.000
#> GSM587194 3 0.0260 0.872 0.000 0.000 0.992 0.000 0.008 0.000
#> GSM587195 3 0.0146 0.871 0.000 0.000 0.996 0.000 0.004 0.000
#> GSM587196 3 0.0260 0.872 0.000 0.000 0.992 0.000 0.008 0.000
#> GSM587197 3 0.1267 0.846 0.000 0.000 0.940 0.000 0.060 0.000
#> GSM587198 3 0.1267 0.845 0.000 0.000 0.940 0.000 0.060 0.000
#> GSM587199 3 0.0260 0.872 0.000 0.000 0.992 0.000 0.008 0.000
#> GSM587200 6 0.2176 0.879 0.000 0.000 0.080 0.000 0.024 0.896
#> GSM587201 6 0.2255 0.875 0.000 0.000 0.080 0.000 0.028 0.892
#> GSM587202 3 0.0713 0.861 0.000 0.000 0.972 0.000 0.028 0.000
#> GSM198767 6 0.0291 0.959 0.004 0.000 0.000 0.000 0.004 0.992
#> GSM198769 1 0.2822 0.907 0.852 0.000 0.000 0.000 0.108 0.040
#> GSM198772 1 0.2822 0.907 0.852 0.000 0.000 0.000 0.108 0.040
#> GSM198773 1 0.2822 0.907 0.852 0.000 0.000 0.000 0.108 0.040
#> GSM198776 1 0.0000 0.930 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM198778 1 0.0000 0.930 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM198780 1 0.0000 0.930 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM198781 1 0.1910 0.917 0.892 0.000 0.000 0.000 0.108 0.000
#> GSM198765 5 0.3659 0.888 0.000 0.000 0.364 0.000 0.636 0.000
#> GSM198766 5 0.3351 0.852 0.000 0.000 0.288 0.000 0.712 0.000
#> GSM198768 3 0.0000 0.871 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM198770 3 0.2378 0.715 0.000 0.000 0.848 0.000 0.152 0.000
#> GSM198771 3 0.0790 0.859 0.000 0.000 0.968 0.000 0.032 0.000
#> GSM198774 5 0.3672 0.886 0.000 0.000 0.368 0.000 0.632 0.000
#> GSM198775 3 0.0260 0.872 0.000 0.000 0.992 0.000 0.008 0.000
#> GSM198777 3 0.0260 0.872 0.000 0.000 0.992 0.000 0.008 0.000
#> GSM198779 3 0.0260 0.872 0.000 0.000 0.992 0.000 0.008 0.000
#> GSM587218 4 0.0000 0.918 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM587219 4 0.0000 0.918 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM587220 4 0.0000 0.918 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM587221 4 0.0000 0.918 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM587222 4 0.0000 0.918 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM587223 4 0.0000 0.918 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM587224 4 0.0000 0.918 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM587225 4 0.3672 0.486 0.368 0.000 0.000 0.632 0.000 0.000
#> GSM587226 4 0.0000 0.918 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM587227 1 0.2631 0.777 0.820 0.000 0.000 0.180 0.000 0.000
#> GSM587228 4 0.2996 0.697 0.228 0.000 0.000 0.772 0.000 0.000
#> GSM587229 1 0.1141 0.898 0.948 0.000 0.000 0.052 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 specimen(p) k
#> ATC:pam 91 1.88e-17 2
#> ATC:pam 92 6.44e-33 3
#> ATC:pam 91 9.29e-41 4
#> ATC:pam 91 1.00e-38 5
#> ATC:pam 91 4.25e-37 6
If matrix rows can be associated to genes, consider to use functional_enrichment(res,
...) to perform function enrichment for the signature genes. See this vignette for more detailed explanations.
The object with results only for a single top-value method and a single partition method can be extracted as:
res = res_list["ATC", "mclust"]
# you can also extract it by
# res = res_list["ATC:mclust"]
A summary of res and all the functions that can be applied to it:
res
#> A 'ConsensusPartition' object with k = 2, 3, 4, 5, 6.
#> On a matrix with 51941 rows and 92 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 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 1.000 1.000 0.4288 0.572 0.572
#> 3 3 0.914 0.964 0.976 0.2890 0.893 0.813
#> 4 4 0.782 0.860 0.923 0.2452 0.842 0.659
#> 5 5 0.757 0.812 0.849 0.0886 0.854 0.585
#> 6 6 0.754 0.687 0.802 0.0470 0.917 0.693
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
#> GSM587155 2 0.0000 1.000 0.000 1.000
#> GSM587156 2 0.0000 1.000 0.000 1.000
#> GSM587157 2 0.0000 1.000 0.000 1.000
#> GSM587158 2 0.0000 1.000 0.000 1.000
#> GSM587159 2 0.0000 1.000 0.000 1.000
#> GSM587160 2 0.0000 1.000 0.000 1.000
#> GSM587161 2 0.0000 1.000 0.000 1.000
#> GSM587162 2 0.0000 1.000 0.000 1.000
#> GSM587163 2 0.0000 1.000 0.000 1.000
#> GSM587164 2 0.0000 1.000 0.000 1.000
#> GSM587165 2 0.0000 1.000 0.000 1.000
#> GSM587166 2 0.0000 1.000 0.000 1.000
#> GSM587167 2 0.0000 1.000 0.000 1.000
#> GSM587168 2 0.0000 1.000 0.000 1.000
#> GSM587169 2 0.0000 1.000 0.000 1.000
#> GSM587170 2 0.0000 1.000 0.000 1.000
#> GSM587171 2 0.0000 1.000 0.000 1.000
#> GSM587172 2 0.0000 1.000 0.000 1.000
#> GSM587173 2 0.0000 1.000 0.000 1.000
#> GSM587174 2 0.0000 1.000 0.000 1.000
#> GSM587175 2 0.0000 1.000 0.000 1.000
#> GSM587176 2 0.0000 1.000 0.000 1.000
#> GSM587177 2 0.0000 1.000 0.000 1.000
#> GSM587178 2 0.0000 1.000 0.000 1.000
#> GSM587179 2 0.0000 1.000 0.000 1.000
#> GSM587180 2 0.0000 1.000 0.000 1.000
#> GSM587181 2 0.0000 1.000 0.000 1.000
#> GSM587182 2 0.0000 1.000 0.000 1.000
#> GSM587183 1 0.0376 0.996 0.996 0.004
#> GSM587184 1 0.0376 0.996 0.996 0.004
#> GSM587185 1 0.0376 0.996 0.996 0.004
#> GSM587186 1 0.0376 0.996 0.996 0.004
#> GSM587187 1 0.0000 1.000 1.000 0.000
#> GSM587188 1 0.0000 1.000 1.000 0.000
#> GSM587189 1 0.0000 1.000 1.000 0.000
#> GSM587190 1 0.0000 1.000 1.000 0.000
#> GSM587203 1 0.0000 1.000 1.000 0.000
#> GSM587204 1 0.0000 1.000 1.000 0.000
#> GSM587205 1 0.0000 1.000 1.000 0.000
#> GSM587206 1 0.0000 1.000 1.000 0.000
#> GSM587207 1 0.0000 1.000 1.000 0.000
#> GSM587208 1 0.0000 1.000 1.000 0.000
#> GSM587209 1 0.0000 1.000 1.000 0.000
#> GSM587210 1 0.0000 1.000 1.000 0.000
#> GSM587211 1 0.0000 1.000 1.000 0.000
#> GSM587212 1 0.0000 1.000 1.000 0.000
#> GSM587213 1 0.0000 1.000 1.000 0.000
#> GSM587214 1 0.0000 1.000 1.000 0.000
#> GSM587215 1 0.0000 1.000 1.000 0.000
#> GSM587216 1 0.0000 1.000 1.000 0.000
#> GSM587217 1 0.0000 1.000 1.000 0.000
#> GSM587191 1 0.0000 1.000 1.000 0.000
#> GSM587192 1 0.0000 1.000 1.000 0.000
#> GSM587193 1 0.0000 1.000 1.000 0.000
#> GSM587194 1 0.0000 1.000 1.000 0.000
#> GSM587195 1 0.0000 1.000 1.000 0.000
#> GSM587196 1 0.0000 1.000 1.000 0.000
#> GSM587197 1 0.0000 1.000 1.000 0.000
#> GSM587198 1 0.0000 1.000 1.000 0.000
#> GSM587199 1 0.0000 1.000 1.000 0.000
#> GSM587200 1 0.0000 1.000 1.000 0.000
#> GSM587201 1 0.0000 1.000 1.000 0.000
#> GSM587202 1 0.0000 1.000 1.000 0.000
#> GSM198767 1 0.0000 1.000 1.000 0.000
#> GSM198769 1 0.0000 1.000 1.000 0.000
#> GSM198772 1 0.0000 1.000 1.000 0.000
#> GSM198773 1 0.0000 1.000 1.000 0.000
#> GSM198776 1 0.0000 1.000 1.000 0.000
#> GSM198778 1 0.0000 1.000 1.000 0.000
#> GSM198780 1 0.0000 1.000 1.000 0.000
#> GSM198781 1 0.0000 1.000 1.000 0.000
#> GSM198765 1 0.0000 1.000 1.000 0.000
#> GSM198766 1 0.0000 1.000 1.000 0.000
#> GSM198768 1 0.0000 1.000 1.000 0.000
#> GSM198770 1 0.0000 1.000 1.000 0.000
#> GSM198771 1 0.0000 1.000 1.000 0.000
#> GSM198774 1 0.0000 1.000 1.000 0.000
#> GSM198775 1 0.0000 1.000 1.000 0.000
#> GSM198777 1 0.0000 1.000 1.000 0.000
#> GSM198779 1 0.0000 1.000 1.000 0.000
#> GSM587218 1 0.0000 1.000 1.000 0.000
#> GSM587219 1 0.0000 1.000 1.000 0.000
#> GSM587220 1 0.0000 1.000 1.000 0.000
#> GSM587221 1 0.0000 1.000 1.000 0.000
#> GSM587222 1 0.0000 1.000 1.000 0.000
#> GSM587223 1 0.0000 1.000 1.000 0.000
#> GSM587224 1 0.0000 1.000 1.000 0.000
#> GSM587225 1 0.0000 1.000 1.000 0.000
#> GSM587226 1 0.0000 1.000 1.000 0.000
#> GSM587227 1 0.0000 1.000 1.000 0.000
#> GSM587228 1 0.0000 1.000 1.000 0.000
#> GSM587229 1 0.0000 1.000 1.000 0.000
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM587155 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587156 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587157 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587158 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587159 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587160 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587161 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587162 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587163 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587164 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587165 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587166 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587167 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587168 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587169 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587170 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587171 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587172 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587173 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587174 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587175 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587176 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587177 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587178 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587179 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587180 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587181 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587182 2 0.0000 1.000 0.000 1.000 0.000
#> GSM587183 3 0.4047 0.871 0.148 0.004 0.848
#> GSM587184 3 0.4047 0.871 0.148 0.004 0.848
#> GSM587185 3 0.4047 0.871 0.148 0.004 0.848
#> GSM587186 3 0.4047 0.871 0.148 0.004 0.848
#> GSM587187 3 0.2537 0.924 0.080 0.000 0.920
#> GSM587188 3 0.2537 0.924 0.080 0.000 0.920
#> GSM587189 3 0.2537 0.924 0.080 0.000 0.920
#> GSM587190 3 0.2165 0.933 0.064 0.000 0.936
#> GSM587203 3 0.3116 0.910 0.108 0.000 0.892
#> GSM587204 3 0.0237 0.960 0.004 0.000 0.996
#> GSM587205 3 0.0424 0.960 0.008 0.000 0.992
#> GSM587206 3 0.0424 0.960 0.008 0.000 0.992
#> GSM587207 3 0.0424 0.960 0.008 0.000 0.992
#> GSM587208 3 0.0424 0.960 0.008 0.000 0.992
#> GSM587209 3 0.0424 0.960 0.008 0.000 0.992
#> GSM587210 3 0.3619 0.887 0.136 0.000 0.864
#> GSM587211 3 0.0237 0.960 0.004 0.000 0.996
#> GSM587212 3 0.0237 0.960 0.004 0.000 0.996
#> GSM587213 3 0.0237 0.960 0.004 0.000 0.996
#> GSM587214 3 0.0424 0.960 0.008 0.000 0.992
#> GSM587215 3 0.0000 0.960 0.000 0.000 1.000
#> GSM587216 3 0.0237 0.960 0.004 0.000 0.996
#> GSM587217 3 0.0424 0.960 0.008 0.000 0.992
#> GSM587191 3 0.0000 0.960 0.000 0.000 1.000
#> GSM587192 3 0.0000 0.960 0.000 0.000 1.000
#> GSM587193 3 0.0000 0.960 0.000 0.000 1.000
#> GSM587194 3 0.0000 0.960 0.000 0.000 1.000
#> GSM587195 3 0.1163 0.950 0.028 0.000 0.972
#> GSM587196 3 0.0000 0.960 0.000 0.000 1.000
#> GSM587197 3 0.2448 0.926 0.076 0.000 0.924
#> GSM587198 3 0.0000 0.960 0.000 0.000 1.000
#> GSM587199 3 0.0000 0.960 0.000 0.000 1.000
#> GSM587200 3 0.0000 0.960 0.000 0.000 1.000
#> GSM587201 3 0.0000 0.960 0.000 0.000 1.000
#> GSM587202 3 0.0000 0.960 0.000 0.000 1.000
#> GSM198767 3 0.0424 0.960 0.008 0.000 0.992
#> GSM198769 3 0.0424 0.960 0.008 0.000 0.992
#> GSM198772 3 0.0424 0.960 0.008 0.000 0.992
#> GSM198773 3 0.0424 0.960 0.008 0.000 0.992
#> GSM198776 3 0.0237 0.960 0.004 0.000 0.996
#> GSM198778 3 0.3551 0.890 0.132 0.000 0.868
#> GSM198780 3 0.0237 0.960 0.004 0.000 0.996
#> GSM198781 3 0.0424 0.960 0.008 0.000 0.992
#> GSM198765 3 0.0000 0.960 0.000 0.000 1.000
#> GSM198766 3 0.0000 0.960 0.000 0.000 1.000
#> GSM198768 3 0.0000 0.960 0.000 0.000 1.000
#> GSM198770 3 0.2356 0.929 0.072 0.000 0.928
#> GSM198771 3 0.0000 0.960 0.000 0.000 1.000
#> GSM198774 3 0.0000 0.960 0.000 0.000 1.000
#> GSM198775 3 0.0000 0.960 0.000 0.000 1.000
#> GSM198777 3 0.0000 0.960 0.000 0.000 1.000
#> GSM198779 3 0.0000 0.960 0.000 0.000 1.000
#> GSM587218 1 0.0000 1.000 1.000 0.000 0.000
#> GSM587219 1 0.0000 1.000 1.000 0.000 0.000
#> GSM587220 1 0.0000 1.000 1.000 0.000 0.000
#> GSM587221 1 0.0000 1.000 1.000 0.000 0.000
#> GSM587222 1 0.0000 1.000 1.000 0.000 0.000
#> GSM587223 1 0.0000 1.000 1.000 0.000 0.000
#> GSM587224 1 0.0000 1.000 1.000 0.000 0.000
#> GSM587225 3 0.3941 0.871 0.156 0.000 0.844
#> GSM587226 1 0.0000 1.000 1.000 0.000 0.000
#> GSM587227 3 0.3941 0.871 0.156 0.000 0.844
#> GSM587228 3 0.3941 0.871 0.156 0.000 0.844
#> GSM587229 3 0.3941 0.871 0.156 0.000 0.844
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM587155 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587156 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587157 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587158 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587159 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587160 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587161 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587162 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587163 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587164 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587165 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587166 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587167 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587168 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587169 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587170 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587171 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587172 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587173 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587174 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587175 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587176 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587177 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587178 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587179 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587180 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587181 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587182 2 0.0000 1.000 0.000 1.000 0.000 0.000
#> GSM587183 3 0.0657 0.715 0.000 0.012 0.984 0.004
#> GSM587184 3 0.0657 0.715 0.000 0.012 0.984 0.004
#> GSM587185 3 0.0657 0.715 0.000 0.012 0.984 0.004
#> GSM587186 3 0.0657 0.715 0.000 0.012 0.984 0.004
#> GSM587187 3 0.3942 0.703 0.236 0.000 0.764 0.000
#> GSM587188 3 0.3942 0.703 0.236 0.000 0.764 0.000
#> GSM587189 3 0.3942 0.703 0.236 0.000 0.764 0.000
#> GSM587190 3 0.4008 0.694 0.244 0.000 0.756 0.000
#> GSM587203 1 0.4283 0.495 0.740 0.000 0.256 0.004
#> GSM587204 1 0.1022 0.873 0.968 0.000 0.032 0.000
#> GSM587205 1 0.0188 0.868 0.996 0.000 0.004 0.000
#> GSM587206 1 0.0188 0.868 0.996 0.000 0.004 0.000
#> GSM587207 1 0.0188 0.868 0.996 0.000 0.004 0.000
#> GSM587208 1 0.0188 0.868 0.996 0.000 0.004 0.000
#> GSM587209 1 0.0188 0.868 0.996 0.000 0.004 0.000
#> GSM587210 1 0.4319 0.571 0.760 0.000 0.228 0.012
#> GSM587211 1 0.1302 0.873 0.956 0.000 0.044 0.000
#> GSM587212 1 0.0927 0.867 0.976 0.000 0.016 0.008
#> GSM587213 1 0.0921 0.872 0.972 0.000 0.028 0.000
#> GSM587214 1 0.0188 0.869 0.996 0.000 0.004 0.000
#> GSM587215 1 0.1211 0.873 0.960 0.000 0.040 0.000
#> GSM587216 1 0.0188 0.869 0.996 0.000 0.004 0.000
#> GSM587217 1 0.0188 0.869 0.996 0.000 0.004 0.000
#> GSM587191 1 0.3172 0.843 0.840 0.000 0.160 0.000
#> GSM587192 1 0.3123 0.844 0.844 0.000 0.156 0.000
#> GSM587193 1 0.3172 0.843 0.840 0.000 0.160 0.000
#> GSM587194 1 0.3172 0.843 0.840 0.000 0.160 0.000
#> GSM587195 3 0.4382 0.617 0.296 0.000 0.704 0.000
#> GSM587196 1 0.3266 0.838 0.832 0.000 0.168 0.000
#> GSM587197 3 0.2281 0.753 0.096 0.000 0.904 0.000
#> GSM587198 3 0.4730 0.462 0.364 0.000 0.636 0.000
#> GSM587199 1 0.3266 0.838 0.832 0.000 0.168 0.000
#> GSM587200 1 0.3123 0.846 0.844 0.000 0.156 0.000
#> GSM587201 1 0.3123 0.846 0.844 0.000 0.156 0.000
#> GSM587202 1 0.3219 0.843 0.836 0.000 0.164 0.000
#> GSM198767 1 0.0188 0.868 0.996 0.000 0.004 0.000
#> GSM198769 1 0.0188 0.868 0.996 0.000 0.004 0.000
#> GSM198772 1 0.0188 0.868 0.996 0.000 0.004 0.000
#> GSM198773 1 0.0188 0.868 0.996 0.000 0.004 0.000
#> GSM198776 1 0.1022 0.873 0.968 0.000 0.032 0.000
#> GSM198778 1 0.4387 0.556 0.752 0.000 0.236 0.012
#> GSM198780 1 0.1151 0.866 0.968 0.000 0.024 0.008
#> GSM198781 1 0.0188 0.869 0.996 0.000 0.004 0.000
#> GSM198765 1 0.3172 0.843 0.840 0.000 0.160 0.000
#> GSM198766 1 0.3172 0.843 0.840 0.000 0.160 0.000
#> GSM198768 3 0.4967 0.196 0.452 0.000 0.548 0.000
#> GSM198770 3 0.2281 0.753 0.096 0.000 0.904 0.000
#> GSM198771 1 0.4776 0.448 0.624 0.000 0.376 0.000
#> GSM198774 1 0.3123 0.844 0.844 0.000 0.156 0.000
#> GSM198775 1 0.3172 0.843 0.840 0.000 0.160 0.000
#> GSM198777 1 0.3266 0.838 0.832 0.000 0.168 0.000
#> GSM198779 1 0.3266 0.838 0.832 0.000 0.168 0.000
#> GSM587218 4 0.0000 1.000 0.000 0.000 0.000 1.000
#> GSM587219 4 0.0000 1.000 0.000 0.000 0.000 1.000
#> GSM587220 4 0.0000 1.000 0.000 0.000 0.000 1.000
#> GSM587221 4 0.0000 1.000 0.000 0.000 0.000 1.000
#> GSM587222 4 0.0000 1.000 0.000 0.000 0.000 1.000
#> GSM587223 4 0.0000 1.000 0.000 0.000 0.000 1.000
#> GSM587224 4 0.0000 1.000 0.000 0.000 0.000 1.000
#> GSM587225 3 0.3428 0.657 0.144 0.000 0.844 0.012
#> GSM587226 4 0.0000 1.000 0.000 0.000 0.000 1.000
#> GSM587227 3 0.3428 0.657 0.144 0.000 0.844 0.012
#> GSM587228 3 0.3428 0.657 0.144 0.000 0.844 0.012
#> GSM587229 3 0.3428 0.657 0.144 0.000 0.844 0.012
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM587155 2 0.0000 0.999 0.000 1.000 0.000 0 0.000
#> GSM587156 2 0.0404 0.984 0.000 0.988 0.012 0 0.000
#> GSM587157 2 0.0000 0.999 0.000 1.000 0.000 0 0.000
#> GSM587158 2 0.0000 0.999 0.000 1.000 0.000 0 0.000
#> GSM587159 2 0.0000 0.999 0.000 1.000 0.000 0 0.000
#> GSM587160 2 0.0000 0.999 0.000 1.000 0.000 0 0.000
#> GSM587161 2 0.0000 0.999 0.000 1.000 0.000 0 0.000
#> GSM587162 2 0.0000 0.999 0.000 1.000 0.000 0 0.000
#> GSM587163 2 0.0000 0.999 0.000 1.000 0.000 0 0.000
#> GSM587164 2 0.0000 0.999 0.000 1.000 0.000 0 0.000
#> GSM587165 2 0.0000 0.999 0.000 1.000 0.000 0 0.000
#> GSM587166 2 0.0404 0.984 0.000 0.988 0.012 0 0.000
#> GSM587167 2 0.0000 0.999 0.000 1.000 0.000 0 0.000
#> GSM587168 2 0.0000 0.999 0.000 1.000 0.000 0 0.000
#> GSM587169 2 0.0000 0.999 0.000 1.000 0.000 0 0.000
#> GSM587170 2 0.0000 0.999 0.000 1.000 0.000 0 0.000
#> GSM587171 2 0.0000 0.999 0.000 1.000 0.000 0 0.000
#> GSM587172 2 0.0000 0.999 0.000 1.000 0.000 0 0.000
#> GSM587173 2 0.0000 0.999 0.000 1.000 0.000 0 0.000
#> GSM587174 2 0.0000 0.999 0.000 1.000 0.000 0 0.000
#> GSM587175 2 0.0000 0.999 0.000 1.000 0.000 0 0.000
#> GSM587176 2 0.0000 0.999 0.000 1.000 0.000 0 0.000
#> GSM587177 2 0.0000 0.999 0.000 1.000 0.000 0 0.000
#> GSM587178 2 0.0000 0.999 0.000 1.000 0.000 0 0.000
#> GSM587179 2 0.0000 0.999 0.000 1.000 0.000 0 0.000
#> GSM587180 2 0.0000 0.999 0.000 1.000 0.000 0 0.000
#> GSM587181 2 0.0000 0.999 0.000 1.000 0.000 0 0.000
#> GSM587182 2 0.0000 0.999 0.000 1.000 0.000 0 0.000
#> GSM587183 3 0.3333 0.697 0.000 0.004 0.788 0 0.208
#> GSM587184 3 0.3333 0.697 0.000 0.004 0.788 0 0.208
#> GSM587185 3 0.3333 0.697 0.000 0.004 0.788 0 0.208
#> GSM587186 3 0.3333 0.697 0.000 0.004 0.788 0 0.208
#> GSM587187 3 0.1281 0.789 0.012 0.000 0.956 0 0.032
#> GSM587188 3 0.1106 0.789 0.012 0.000 0.964 0 0.024
#> GSM587189 3 0.1106 0.789 0.012 0.000 0.964 0 0.024
#> GSM587190 3 0.0963 0.794 0.036 0.000 0.964 0 0.000
#> GSM587203 5 0.5336 0.762 0.252 0.000 0.100 0 0.648
#> GSM587204 1 0.3862 0.694 0.808 0.000 0.088 0 0.104
#> GSM587205 5 0.4150 0.935 0.388 0.000 0.000 0 0.612
#> GSM587206 5 0.4150 0.935 0.388 0.000 0.000 0 0.612
#> GSM587207 5 0.4150 0.935 0.388 0.000 0.000 0 0.612
#> GSM587208 5 0.4150 0.935 0.388 0.000 0.000 0 0.612
#> GSM587209 1 0.1579 0.629 0.944 0.000 0.024 0 0.032
#> GSM587210 3 0.4612 0.752 0.084 0.000 0.736 0 0.180
#> GSM587211 1 0.1981 0.676 0.920 0.000 0.064 0 0.016
#> GSM587212 3 0.5379 0.716 0.164 0.000 0.668 0 0.168
#> GSM587213 1 0.1484 0.664 0.944 0.000 0.048 0 0.008
#> GSM587214 1 0.3868 0.662 0.800 0.000 0.060 0 0.140
#> GSM587215 1 0.4680 0.685 0.740 0.000 0.132 0 0.128
#> GSM587216 1 0.3857 0.692 0.808 0.000 0.084 0 0.108
#> GSM587217 1 0.3759 0.663 0.808 0.000 0.056 0 0.136
#> GSM587191 1 0.4925 0.571 0.632 0.000 0.324 0 0.044
#> GSM587192 1 0.6101 0.545 0.528 0.000 0.328 0 0.144
#> GSM587193 1 0.4046 0.621 0.696 0.000 0.296 0 0.008
#> GSM587194 3 0.4855 0.676 0.168 0.000 0.720 0 0.112
#> GSM587195 3 0.1830 0.793 0.028 0.000 0.932 0 0.040
#> GSM587196 3 0.4748 0.683 0.172 0.000 0.728 0 0.100
#> GSM587197 3 0.1800 0.793 0.020 0.000 0.932 0 0.048
#> GSM587198 3 0.3477 0.767 0.136 0.000 0.824 0 0.040
#> GSM587199 3 0.4457 0.728 0.116 0.000 0.760 0 0.124
#> GSM587200 1 0.2771 0.692 0.860 0.000 0.128 0 0.012
#> GSM587201 1 0.2674 0.693 0.868 0.000 0.120 0 0.012
#> GSM587202 1 0.3861 0.649 0.728 0.000 0.264 0 0.008
#> GSM198767 1 0.4313 -0.307 0.636 0.000 0.008 0 0.356
#> GSM198769 1 0.1386 0.622 0.952 0.000 0.016 0 0.032
#> GSM198772 1 0.1211 0.636 0.960 0.000 0.024 0 0.016
#> GSM198773 1 0.1082 0.611 0.964 0.000 0.008 0 0.028
#> GSM198776 1 0.3862 0.694 0.808 0.000 0.088 0 0.104
#> GSM198778 3 0.4612 0.752 0.084 0.000 0.736 0 0.180
#> GSM198780 3 0.5379 0.716 0.164 0.000 0.668 0 0.168
#> GSM198781 1 0.4054 0.671 0.788 0.000 0.072 0 0.140
#> GSM198765 1 0.4957 0.575 0.624 0.000 0.332 0 0.044
#> GSM198766 1 0.3934 0.639 0.716 0.000 0.276 0 0.008
#> GSM198768 3 0.3810 0.744 0.176 0.000 0.788 0 0.036
#> GSM198770 3 0.1549 0.791 0.016 0.000 0.944 0 0.040
#> GSM198771 3 0.3942 0.697 0.232 0.000 0.748 0 0.020
#> GSM198774 1 0.6068 0.549 0.532 0.000 0.328 0 0.140
#> GSM198775 3 0.4743 0.692 0.156 0.000 0.732 0 0.112
#> GSM198777 3 0.4599 0.704 0.156 0.000 0.744 0 0.100
#> GSM198779 3 0.4457 0.729 0.116 0.000 0.760 0 0.124
#> GSM587218 4 0.0000 1.000 0.000 0.000 0.000 1 0.000
#> GSM587219 4 0.0000 1.000 0.000 0.000 0.000 1 0.000
#> GSM587220 4 0.0000 1.000 0.000 0.000 0.000 1 0.000
#> GSM587221 4 0.0000 1.000 0.000 0.000 0.000 1 0.000
#> GSM587222 4 0.0000 1.000 0.000 0.000 0.000 1 0.000
#> GSM587223 4 0.0000 1.000 0.000 0.000 0.000 1 0.000
#> GSM587224 4 0.0000 1.000 0.000 0.000 0.000 1 0.000
#> GSM587225 3 0.4786 0.696 0.188 0.000 0.720 0 0.092
#> GSM587226 4 0.0000 1.000 0.000 0.000 0.000 1 0.000
#> GSM587227 3 0.4803 0.695 0.184 0.000 0.720 0 0.096
#> GSM587228 3 0.4836 0.695 0.188 0.000 0.716 0 0.096
#> GSM587229 3 0.4836 0.695 0.188 0.000 0.716 0 0.096
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM587155 2 0.0000 0.9877 0.000 1.000 0.000 0 0.000 0.000
#> GSM587156 2 0.1714 0.8810 0.000 0.908 0.000 0 0.000 0.092
#> GSM587157 2 0.0146 0.9848 0.000 0.996 0.000 0 0.000 0.004
#> GSM587158 2 0.0000 0.9877 0.000 1.000 0.000 0 0.000 0.000
#> GSM587159 2 0.0000 0.9877 0.000 1.000 0.000 0 0.000 0.000
#> GSM587160 2 0.0000 0.9877 0.000 1.000 0.000 0 0.000 0.000
#> GSM587161 2 0.0146 0.9848 0.000 0.996 0.000 0 0.000 0.004
#> GSM587162 2 0.0000 0.9877 0.000 1.000 0.000 0 0.000 0.000
#> GSM587163 2 0.0000 0.9877 0.000 1.000 0.000 0 0.000 0.000
#> GSM587164 2 0.0000 0.9877 0.000 1.000 0.000 0 0.000 0.000
#> GSM587165 2 0.0000 0.9877 0.000 1.000 0.000 0 0.000 0.000
#> GSM587166 2 0.1714 0.8810 0.000 0.908 0.000 0 0.000 0.092
#> GSM587167 2 0.1387 0.9107 0.000 0.932 0.000 0 0.000 0.068
#> GSM587168 2 0.0000 0.9877 0.000 1.000 0.000 0 0.000 0.000
#> GSM587169 2 0.0000 0.9877 0.000 1.000 0.000 0 0.000 0.000
#> GSM587170 2 0.0000 0.9877 0.000 1.000 0.000 0 0.000 0.000
#> GSM587171 2 0.0000 0.9877 0.000 1.000 0.000 0 0.000 0.000
#> GSM587172 2 0.0000 0.9877 0.000 1.000 0.000 0 0.000 0.000
#> GSM587173 2 0.0000 0.9877 0.000 1.000 0.000 0 0.000 0.000
#> GSM587174 2 0.0000 0.9877 0.000 1.000 0.000 0 0.000 0.000
#> GSM587175 2 0.0000 0.9877 0.000 1.000 0.000 0 0.000 0.000
#> GSM587176 2 0.0000 0.9877 0.000 1.000 0.000 0 0.000 0.000
#> GSM587177 2 0.0000 0.9877 0.000 1.000 0.000 0 0.000 0.000
#> GSM587178 2 0.0000 0.9877 0.000 1.000 0.000 0 0.000 0.000
#> GSM587179 2 0.0000 0.9877 0.000 1.000 0.000 0 0.000 0.000
#> GSM587180 2 0.0000 0.9877 0.000 1.000 0.000 0 0.000 0.000
#> GSM587181 2 0.0000 0.9877 0.000 1.000 0.000 0 0.000 0.000
#> GSM587182 2 0.0000 0.9877 0.000 1.000 0.000 0 0.000 0.000
#> GSM587183 6 0.0363 0.6191 0.000 0.000 0.012 0 0.000 0.988
#> GSM587184 6 0.0363 0.6191 0.000 0.000 0.012 0 0.000 0.988
#> GSM587185 6 0.0363 0.6191 0.000 0.000 0.012 0 0.000 0.988
#> GSM587186 6 0.0363 0.6191 0.000 0.000 0.012 0 0.000 0.988
#> GSM587187 3 0.4595 0.5534 0.020 0.000 0.676 0 0.040 0.264
#> GSM587188 3 0.4146 0.5698 0.008 0.000 0.720 0 0.040 0.232
#> GSM587189 3 0.4146 0.5698 0.008 0.000 0.720 0 0.040 0.232
#> GSM587190 3 0.3804 0.5820 0.012 0.000 0.748 0 0.020 0.220
#> GSM587203 5 0.4923 0.2911 0.072 0.000 0.000 0 0.560 0.368
#> GSM587204 1 0.1958 0.5604 0.896 0.000 0.100 0 0.004 0.000
#> GSM587205 5 0.2009 0.7770 0.084 0.000 0.004 0 0.904 0.008
#> GSM587206 5 0.1753 0.7811 0.084 0.000 0.004 0 0.912 0.000
#> GSM587207 5 0.1753 0.7789 0.084 0.000 0.004 0 0.912 0.000
#> GSM587208 5 0.1753 0.7811 0.084 0.000 0.004 0 0.912 0.000
#> GSM587209 1 0.6071 0.4000 0.520 0.000 0.152 0 0.028 0.300
#> GSM587210 3 0.6034 0.0655 0.392 0.000 0.472 0 0.048 0.088
#> GSM587211 1 0.5988 0.3764 0.536 0.000 0.132 0 0.032 0.300
#> GSM587212 3 0.3983 0.5405 0.120 0.000 0.792 0 0.048 0.040
#> GSM587213 1 0.6405 0.4470 0.508 0.000 0.140 0 0.060 0.292
#> GSM587214 1 0.2791 0.5508 0.864 0.000 0.096 0 0.032 0.008
#> GSM587215 1 0.2588 0.5472 0.860 0.000 0.124 0 0.004 0.012
#> GSM587216 1 0.3500 0.5215 0.768 0.000 0.204 0 0.028 0.000
#> GSM587217 1 0.2201 0.5561 0.896 0.000 0.076 0 0.028 0.000
#> GSM587191 3 0.5789 0.3477 0.348 0.000 0.520 0 0.024 0.108
#> GSM587192 3 0.3010 0.5883 0.148 0.000 0.828 0 0.004 0.020
#> GSM587193 1 0.5501 0.3307 0.580 0.000 0.300 0 0.020 0.100
#> GSM587194 3 0.2680 0.6440 0.056 0.000 0.868 0 0.000 0.076
#> GSM587195 3 0.6185 0.4365 0.220 0.000 0.500 0 0.020 0.260
#> GSM587196 3 0.2066 0.6425 0.072 0.000 0.904 0 0.000 0.024
#> GSM587197 6 0.6434 -0.3694 0.232 0.000 0.372 0 0.020 0.376
#> GSM587198 3 0.6305 0.3835 0.236 0.000 0.468 0 0.020 0.276
#> GSM587199 3 0.1003 0.6487 0.004 0.000 0.964 0 0.004 0.028
#> GSM587200 1 0.6496 0.4021 0.436 0.000 0.236 0 0.028 0.300
#> GSM587201 1 0.6109 0.4649 0.524 0.000 0.152 0 0.032 0.292
#> GSM587202 1 0.6361 0.4000 0.452 0.000 0.252 0 0.020 0.276
#> GSM198767 5 0.6530 0.2448 0.232 0.000 0.036 0 0.460 0.272
#> GSM198769 1 0.6467 0.4246 0.492 0.000 0.164 0 0.052 0.292
#> GSM198772 1 0.5111 0.4756 0.672 0.000 0.124 0 0.020 0.184
#> GSM198773 1 0.6399 0.4511 0.516 0.000 0.116 0 0.076 0.292
#> GSM198776 1 0.2053 0.5605 0.888 0.000 0.108 0 0.004 0.000
#> GSM198778 3 0.5865 0.2205 0.360 0.000 0.516 0 0.048 0.076
#> GSM198780 3 0.4026 0.5501 0.112 0.000 0.792 0 0.048 0.048
#> GSM198781 1 0.2604 0.5523 0.872 0.000 0.100 0 0.020 0.008
#> GSM198765 3 0.3883 0.5723 0.196 0.000 0.760 0 0.024 0.020
#> GSM198766 1 0.5347 0.3644 0.600 0.000 0.292 0 0.020 0.088
#> GSM198768 3 0.6141 0.4474 0.224 0.000 0.512 0 0.020 0.244
#> GSM198770 3 0.6434 0.2607 0.232 0.000 0.380 0 0.020 0.368
#> GSM198771 3 0.6107 0.4447 0.260 0.000 0.516 0 0.020 0.204
#> GSM198774 3 0.3053 0.5917 0.144 0.000 0.828 0 0.004 0.024
#> GSM198775 3 0.2876 0.6432 0.056 0.000 0.860 0 0.004 0.080
#> GSM198777 3 0.2189 0.6465 0.060 0.000 0.904 0 0.004 0.032
#> GSM198779 3 0.1003 0.6487 0.004 0.000 0.964 0 0.004 0.028
#> GSM587218 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587219 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587220 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587221 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587222 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587223 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587224 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587225 6 0.5429 0.5296 0.272 0.000 0.044 0 0.068 0.616
#> GSM587226 4 0.0000 1.0000 0.000 0.000 0.000 1 0.000 0.000
#> GSM587227 6 0.5447 0.5227 0.276 0.000 0.044 0 0.068 0.612
#> GSM587228 6 0.5415 0.5244 0.280 0.000 0.044 0 0.064 0.612
#> GSM587229 6 0.5415 0.5244 0.280 0.000 0.044 0 0.064 0.612
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 specimen(p) k
#> ATC:mclust 92 3.07e-14 2
#> ATC:mclust 92 2.81e-22 3
#> ATC:mclust 88 3.64e-26 4
#> ATC:mclust 91 1.56e-24 5
#> ATC:mclust 70 7.18e-24 6
If matrix rows can be associated to genes, consider to use functional_enrichment(res,
...) to perform function enrichment for the signature genes. See this vignette for more detailed explanations.
The object with results only for a single top-value method and a single partition method can be extracted as:
res = res_list["ATC", "NMF"]
# you can also extract it by
# res = res_list["ATC:NMF"]
A summary of res and all the functions that can be applied to it:
res
#> A 'ConsensusPartition' object with k = 2, 3, 4, 5, 6.
#> On a matrix with 51941 rows and 92 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 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 1.000 0.984 0.993 0.4890 0.514 0.514
#> 3 3 0.864 0.861 0.938 0.3039 0.809 0.647
#> 4 4 0.934 0.932 0.956 0.1451 0.841 0.604
#> 5 5 0.908 0.891 0.930 0.0616 0.922 0.726
#> 6 6 0.870 0.816 0.877 0.0255 0.973 0.879
suggest_best_k() suggests the best \(k\) based on these statistics. The rules are as follows:
suggest_best_k(res)
#> [1] 5
#> attr(,"optional")
#> [1] 2 4
There is also optional best \(k\) = 2 4 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
#> GSM587155 2 0.000 1.000 0.000 1.000
#> GSM587156 2 0.000 1.000 0.000 1.000
#> GSM587157 2 0.000 1.000 0.000 1.000
#> GSM587158 2 0.000 1.000 0.000 1.000
#> GSM587159 2 0.000 1.000 0.000 1.000
#> GSM587160 2 0.000 1.000 0.000 1.000
#> GSM587161 2 0.000 1.000 0.000 1.000
#> GSM587162 2 0.000 1.000 0.000 1.000
#> GSM587163 2 0.000 1.000 0.000 1.000
#> GSM587164 2 0.000 1.000 0.000 1.000
#> GSM587165 2 0.000 1.000 0.000 1.000
#> GSM587166 2 0.000 1.000 0.000 1.000
#> GSM587167 2 0.000 1.000 0.000 1.000
#> GSM587168 2 0.000 1.000 0.000 1.000
#> GSM587169 2 0.000 1.000 0.000 1.000
#> GSM587170 2 0.000 1.000 0.000 1.000
#> GSM587171 2 0.000 1.000 0.000 1.000
#> GSM587172 2 0.000 1.000 0.000 1.000
#> GSM587173 2 0.000 1.000 0.000 1.000
#> GSM587174 2 0.000 1.000 0.000 1.000
#> GSM587175 2 0.000 1.000 0.000 1.000
#> GSM587176 2 0.000 1.000 0.000 1.000
#> GSM587177 2 0.000 1.000 0.000 1.000
#> GSM587178 2 0.000 1.000 0.000 1.000
#> GSM587179 2 0.000 1.000 0.000 1.000
#> GSM587180 2 0.000 1.000 0.000 1.000
#> GSM587181 2 0.000 1.000 0.000 1.000
#> GSM587182 2 0.000 1.000 0.000 1.000
#> GSM587183 2 0.000 1.000 0.000 1.000
#> GSM587184 2 0.000 1.000 0.000 1.000
#> GSM587185 2 0.000 1.000 0.000 1.000
#> GSM587186 2 0.000 1.000 0.000 1.000
#> GSM587187 2 0.000 1.000 0.000 1.000
#> GSM587188 2 0.000 1.000 0.000 1.000
#> GSM587189 2 0.000 1.000 0.000 1.000
#> GSM587190 2 0.000 1.000 0.000 1.000
#> GSM587203 1 0.000 0.988 1.000 0.000
#> GSM587204 1 0.000 0.988 1.000 0.000
#> GSM587205 1 0.000 0.988 1.000 0.000
#> GSM587206 1 0.000 0.988 1.000 0.000
#> GSM587207 1 0.000 0.988 1.000 0.000
#> GSM587208 1 0.000 0.988 1.000 0.000
#> GSM587209 1 0.000 0.988 1.000 0.000
#> GSM587210 1 0.000 0.988 1.000 0.000
#> GSM587211 1 0.000 0.988 1.000 0.000
#> GSM587212 1 0.000 0.988 1.000 0.000
#> GSM587213 1 0.000 0.988 1.000 0.000
#> GSM587214 1 0.000 0.988 1.000 0.000
#> GSM587215 1 0.000 0.988 1.000 0.000
#> GSM587216 1 0.000 0.988 1.000 0.000
#> GSM587217 1 0.000 0.988 1.000 0.000
#> GSM587191 2 0.000 1.000 0.000 1.000
#> GSM587192 1 0.000 0.988 1.000 0.000
#> GSM587193 1 0.118 0.974 0.984 0.016
#> GSM587194 1 0.118 0.974 0.984 0.016
#> GSM587195 1 0.278 0.944 0.952 0.048
#> GSM587196 1 0.000 0.988 1.000 0.000
#> GSM587197 1 0.000 0.988 1.000 0.000
#> GSM587198 1 0.000 0.988 1.000 0.000
#> GSM587199 1 0.000 0.988 1.000 0.000
#> GSM587200 1 0.000 0.988 1.000 0.000
#> GSM587201 1 0.000 0.988 1.000 0.000
#> GSM587202 1 0.000 0.988 1.000 0.000
#> GSM198767 1 0.000 0.988 1.000 0.000
#> GSM198769 1 0.000 0.988 1.000 0.000
#> GSM198772 1 0.000 0.988 1.000 0.000
#> GSM198773 1 0.000 0.988 1.000 0.000
#> GSM198776 1 0.000 0.988 1.000 0.000
#> GSM198778 1 0.000 0.988 1.000 0.000
#> GSM198780 1 0.000 0.988 1.000 0.000
#> GSM198781 1 0.000 0.988 1.000 0.000
#> GSM198765 1 0.802 0.690 0.756 0.244
#> GSM198766 1 0.000 0.988 1.000 0.000
#> GSM198768 1 0.000 0.988 1.000 0.000
#> GSM198770 1 0.833 0.656 0.736 0.264
#> GSM198771 1 0.000 0.988 1.000 0.000
#> GSM198774 1 0.000 0.988 1.000 0.000
#> GSM198775 1 0.402 0.911 0.920 0.080
#> GSM198777 1 0.000 0.988 1.000 0.000
#> GSM198779 1 0.000 0.988 1.000 0.000
#> GSM587218 1 0.000 0.988 1.000 0.000
#> GSM587219 1 0.000 0.988 1.000 0.000
#> GSM587220 1 0.000 0.988 1.000 0.000
#> GSM587221 1 0.000 0.988 1.000 0.000
#> GSM587222 1 0.000 0.988 1.000 0.000
#> GSM587223 1 0.000 0.988 1.000 0.000
#> GSM587224 1 0.000 0.988 1.000 0.000
#> GSM587225 1 0.000 0.988 1.000 0.000
#> GSM587226 1 0.000 0.988 1.000 0.000
#> GSM587227 1 0.000 0.988 1.000 0.000
#> GSM587228 1 0.000 0.988 1.000 0.000
#> GSM587229 1 0.000 0.988 1.000 0.000
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM587155 2 0.0000 0.988 0.000 1.000 0.000
#> GSM587156 2 0.0000 0.988 0.000 1.000 0.000
#> GSM587157 2 0.0000 0.988 0.000 1.000 0.000
#> GSM587158 2 0.0000 0.988 0.000 1.000 0.000
#> GSM587159 2 0.0000 0.988 0.000 1.000 0.000
#> GSM587160 2 0.0000 0.988 0.000 1.000 0.000
#> GSM587161 2 0.0000 0.988 0.000 1.000 0.000
#> GSM587162 2 0.0000 0.988 0.000 1.000 0.000
#> GSM587163 2 0.0000 0.988 0.000 1.000 0.000
#> GSM587164 2 0.0000 0.988 0.000 1.000 0.000
#> GSM587165 2 0.0000 0.988 0.000 1.000 0.000
#> GSM587166 2 0.0000 0.988 0.000 1.000 0.000
#> GSM587167 2 0.0000 0.988 0.000 1.000 0.000
#> GSM587168 2 0.0000 0.988 0.000 1.000 0.000
#> GSM587169 2 0.0000 0.988 0.000 1.000 0.000
#> GSM587170 2 0.0000 0.988 0.000 1.000 0.000
#> GSM587171 2 0.0000 0.988 0.000 1.000 0.000
#> GSM587172 2 0.0000 0.988 0.000 1.000 0.000
#> GSM587173 2 0.0000 0.988 0.000 1.000 0.000
#> GSM587174 2 0.0000 0.988 0.000 1.000 0.000
#> GSM587175 2 0.0000 0.988 0.000 1.000 0.000
#> GSM587176 2 0.0000 0.988 0.000 1.000 0.000
#> GSM587177 2 0.0000 0.988 0.000 1.000 0.000
#> GSM587178 2 0.0000 0.988 0.000 1.000 0.000
#> GSM587179 2 0.0000 0.988 0.000 1.000 0.000
#> GSM587180 2 0.0000 0.988 0.000 1.000 0.000
#> GSM587181 2 0.0000 0.988 0.000 1.000 0.000
#> GSM587182 2 0.0000 0.988 0.000 1.000 0.000
#> GSM587183 2 0.0000 0.988 0.000 1.000 0.000
#> GSM587184 2 0.0000 0.988 0.000 1.000 0.000
#> GSM587185 2 0.0000 0.988 0.000 1.000 0.000
#> GSM587186 2 0.0000 0.988 0.000 1.000 0.000
#> GSM587187 2 0.0237 0.984 0.000 0.996 0.004
#> GSM587188 2 0.0424 0.980 0.000 0.992 0.008
#> GSM587189 2 0.0592 0.976 0.000 0.988 0.012
#> GSM587190 3 0.6045 0.352 0.000 0.380 0.620
#> GSM587203 1 0.2066 0.892 0.940 0.000 0.060
#> GSM587204 3 0.1289 0.870 0.032 0.000 0.968
#> GSM587205 3 0.5363 0.651 0.276 0.000 0.724
#> GSM587206 1 0.6307 -0.187 0.512 0.000 0.488
#> GSM587207 3 0.6274 0.271 0.456 0.000 0.544
#> GSM587208 3 0.6309 0.162 0.496 0.000 0.504
#> GSM587209 3 0.6154 0.428 0.408 0.000 0.592
#> GSM587210 3 0.4654 0.736 0.208 0.000 0.792
#> GSM587211 3 0.5560 0.627 0.300 0.000 0.700
#> GSM587212 3 0.0892 0.873 0.020 0.000 0.980
#> GSM587213 3 0.6154 0.401 0.408 0.000 0.592
#> GSM587214 3 0.0237 0.874 0.004 0.000 0.996
#> GSM587215 3 0.0424 0.873 0.008 0.000 0.992
#> GSM587216 3 0.0424 0.874 0.008 0.000 0.992
#> GSM587217 3 0.1031 0.873 0.024 0.000 0.976
#> GSM587191 3 0.1031 0.864 0.000 0.024 0.976
#> GSM587192 3 0.0237 0.873 0.004 0.000 0.996
#> GSM587193 3 0.0237 0.874 0.000 0.004 0.996
#> GSM587194 3 0.0592 0.874 0.012 0.000 0.988
#> GSM587195 3 0.2681 0.843 0.028 0.040 0.932
#> GSM587196 3 0.0237 0.874 0.004 0.000 0.996
#> GSM587197 3 0.4370 0.810 0.076 0.056 0.868
#> GSM587198 3 0.1781 0.864 0.020 0.020 0.960
#> GSM587199 3 0.0424 0.874 0.008 0.000 0.992
#> GSM587200 3 0.1289 0.869 0.032 0.000 0.968
#> GSM587201 3 0.1289 0.870 0.032 0.000 0.968
#> GSM587202 3 0.0592 0.874 0.012 0.000 0.988
#> GSM198767 3 0.6244 0.337 0.440 0.000 0.560
#> GSM198769 3 0.5058 0.703 0.244 0.000 0.756
#> GSM198772 3 0.4121 0.778 0.168 0.000 0.832
#> GSM198773 3 0.4974 0.699 0.236 0.000 0.764
#> GSM198776 3 0.0592 0.874 0.012 0.000 0.988
#> GSM198778 3 0.3941 0.790 0.156 0.000 0.844
#> GSM198780 3 0.0747 0.874 0.016 0.000 0.984
#> GSM198781 3 0.0424 0.873 0.008 0.000 0.992
#> GSM198765 3 0.0237 0.873 0.004 0.000 0.996
#> GSM198766 3 0.0237 0.873 0.004 0.000 0.996
#> GSM198768 3 0.1031 0.872 0.024 0.000 0.976
#> GSM198770 2 0.6497 0.450 0.016 0.648 0.336
#> GSM198771 3 0.0747 0.874 0.016 0.000 0.984
#> GSM198774 3 0.0237 0.873 0.004 0.000 0.996
#> GSM198775 3 0.0475 0.872 0.004 0.004 0.992
#> GSM198777 3 0.0237 0.874 0.004 0.000 0.996
#> GSM198779 3 0.0424 0.874 0.008 0.000 0.992
#> GSM587218 1 0.0237 0.949 0.996 0.000 0.004
#> GSM587219 1 0.0237 0.949 0.996 0.000 0.004
#> GSM587220 1 0.0237 0.949 0.996 0.000 0.004
#> GSM587221 1 0.0237 0.949 0.996 0.000 0.004
#> GSM587222 1 0.0237 0.949 0.996 0.000 0.004
#> GSM587223 1 0.0237 0.949 0.996 0.000 0.004
#> GSM587224 1 0.0237 0.949 0.996 0.000 0.004
#> GSM587225 1 0.0237 0.949 0.996 0.000 0.004
#> GSM587226 1 0.0237 0.949 0.996 0.000 0.004
#> GSM587227 1 0.0237 0.949 0.996 0.000 0.004
#> GSM587228 1 0.0237 0.949 0.996 0.000 0.004
#> GSM587229 1 0.0237 0.949 0.996 0.000 0.004
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM587155 2 0.0000 0.993 0.000 1.000 0.000 0.000
#> GSM587156 2 0.0707 0.979 0.000 0.980 0.000 0.020
#> GSM587157 2 0.0188 0.990 0.000 0.996 0.000 0.004
#> GSM587158 2 0.0000 0.993 0.000 1.000 0.000 0.000
#> GSM587159 2 0.0000 0.993 0.000 1.000 0.000 0.000
#> GSM587160 2 0.0000 0.993 0.000 1.000 0.000 0.000
#> GSM587161 2 0.0000 0.993 0.000 1.000 0.000 0.000
#> GSM587162 2 0.0000 0.993 0.000 1.000 0.000 0.000
#> GSM587163 2 0.0000 0.993 0.000 1.000 0.000 0.000
#> GSM587164 2 0.0000 0.993 0.000 1.000 0.000 0.000
#> GSM587165 2 0.0000 0.993 0.000 1.000 0.000 0.000
#> GSM587166 2 0.0817 0.976 0.000 0.976 0.000 0.024
#> GSM587167 2 0.0336 0.988 0.000 0.992 0.000 0.008
#> GSM587168 2 0.0000 0.993 0.000 1.000 0.000 0.000
#> GSM587169 2 0.0000 0.993 0.000 1.000 0.000 0.000
#> GSM587170 2 0.0188 0.990 0.000 0.996 0.000 0.004
#> GSM587171 2 0.0000 0.993 0.000 1.000 0.000 0.000
#> GSM587172 2 0.0000 0.993 0.000 1.000 0.000 0.000
#> GSM587173 2 0.0000 0.993 0.000 1.000 0.000 0.000
#> GSM587174 2 0.0000 0.993 0.000 1.000 0.000 0.000
#> GSM587175 2 0.0000 0.993 0.000 1.000 0.000 0.000
#> GSM587176 2 0.0000 0.993 0.000 1.000 0.000 0.000
#> GSM587177 2 0.0000 0.993 0.000 1.000 0.000 0.000
#> GSM587178 2 0.0000 0.993 0.000 1.000 0.000 0.000
#> GSM587179 2 0.0000 0.993 0.000 1.000 0.000 0.000
#> GSM587180 2 0.0000 0.993 0.000 1.000 0.000 0.000
#> GSM587181 2 0.0000 0.993 0.000 1.000 0.000 0.000
#> GSM587182 2 0.0000 0.993 0.000 1.000 0.000 0.000
#> GSM587183 2 0.0000 0.993 0.000 1.000 0.000 0.000
#> GSM587184 2 0.0000 0.993 0.000 1.000 0.000 0.000
#> GSM587185 2 0.0000 0.993 0.000 1.000 0.000 0.000
#> GSM587186 2 0.0000 0.993 0.000 1.000 0.000 0.000
#> GSM587187 2 0.3311 0.794 0.000 0.828 0.172 0.000
#> GSM587188 3 0.4981 0.132 0.000 0.464 0.536 0.000
#> GSM587189 3 0.1716 0.878 0.000 0.064 0.936 0.000
#> GSM587190 3 0.0000 0.920 0.000 0.000 1.000 0.000
#> GSM587203 1 0.4804 0.476 0.616 0.000 0.000 0.384
#> GSM587204 1 0.4595 0.805 0.776 0.000 0.040 0.184
#> GSM587205 1 0.0336 0.925 0.992 0.000 0.000 0.008
#> GSM587206 1 0.1637 0.922 0.940 0.000 0.000 0.060
#> GSM587207 1 0.0817 0.927 0.976 0.000 0.000 0.024
#> GSM587208 1 0.1118 0.927 0.964 0.000 0.000 0.036
#> GSM587209 1 0.2814 0.879 0.868 0.000 0.000 0.132
#> GSM587210 3 0.3047 0.851 0.012 0.000 0.872 0.116
#> GSM587211 1 0.1940 0.917 0.924 0.000 0.000 0.076
#> GSM587212 3 0.1833 0.915 0.032 0.000 0.944 0.024
#> GSM587213 1 0.1022 0.928 0.968 0.000 0.000 0.032
#> GSM587214 1 0.0524 0.924 0.988 0.000 0.008 0.004
#> GSM587215 1 0.1109 0.911 0.968 0.000 0.028 0.004
#> GSM587216 1 0.3047 0.853 0.872 0.000 0.116 0.012
#> GSM587217 1 0.1824 0.923 0.936 0.000 0.004 0.060
#> GSM587191 3 0.2334 0.895 0.088 0.004 0.908 0.000
#> GSM587192 3 0.1557 0.914 0.056 0.000 0.944 0.000
#> GSM587193 1 0.0921 0.914 0.972 0.000 0.028 0.000
#> GSM587194 3 0.0000 0.920 0.000 0.000 1.000 0.000
#> GSM587195 3 0.0469 0.922 0.012 0.000 0.988 0.000
#> GSM587196 3 0.0707 0.922 0.020 0.000 0.980 0.000
#> GSM587197 3 0.0469 0.918 0.000 0.000 0.988 0.012
#> GSM587198 3 0.2345 0.878 0.100 0.000 0.900 0.000
#> GSM587199 3 0.0000 0.920 0.000 0.000 1.000 0.000
#> GSM587200 1 0.0336 0.926 0.992 0.000 0.000 0.008
#> GSM587201 1 0.0188 0.922 0.996 0.000 0.000 0.004
#> GSM587202 1 0.1109 0.915 0.968 0.000 0.028 0.004
#> GSM198767 1 0.2281 0.905 0.904 0.000 0.000 0.096
#> GSM198769 1 0.2589 0.892 0.884 0.000 0.000 0.116
#> GSM198772 1 0.1743 0.924 0.940 0.000 0.004 0.056
#> GSM198773 1 0.1022 0.928 0.968 0.000 0.000 0.032
#> GSM198776 1 0.4590 0.824 0.792 0.000 0.060 0.148
#> GSM198778 3 0.2376 0.893 0.016 0.000 0.916 0.068
#> GSM198780 3 0.1677 0.917 0.040 0.000 0.948 0.012
#> GSM198781 1 0.0524 0.924 0.988 0.000 0.008 0.004
#> GSM198765 3 0.1557 0.914 0.056 0.000 0.944 0.000
#> GSM198766 1 0.1716 0.895 0.936 0.000 0.064 0.000
#> GSM198768 3 0.0817 0.922 0.024 0.000 0.976 0.000
#> GSM198770 3 0.3895 0.733 0.000 0.184 0.804 0.012
#> GSM198771 3 0.2216 0.887 0.092 0.000 0.908 0.000
#> GSM198774 3 0.1389 0.917 0.048 0.000 0.952 0.000
#> GSM198775 3 0.0000 0.920 0.000 0.000 1.000 0.000
#> GSM198777 3 0.0469 0.922 0.012 0.000 0.988 0.000
#> GSM198779 3 0.0000 0.920 0.000 0.000 1.000 0.000
#> GSM587218 4 0.0921 1.000 0.028 0.000 0.000 0.972
#> GSM587219 4 0.0921 1.000 0.028 0.000 0.000 0.972
#> GSM587220 4 0.0921 1.000 0.028 0.000 0.000 0.972
#> GSM587221 4 0.0921 1.000 0.028 0.000 0.000 0.972
#> GSM587222 4 0.0921 1.000 0.028 0.000 0.000 0.972
#> GSM587223 4 0.0921 1.000 0.028 0.000 0.000 0.972
#> GSM587224 4 0.0921 1.000 0.028 0.000 0.000 0.972
#> GSM587225 4 0.0921 1.000 0.028 0.000 0.000 0.972
#> GSM587226 4 0.0921 1.000 0.028 0.000 0.000 0.972
#> GSM587227 4 0.0921 1.000 0.028 0.000 0.000 0.972
#> GSM587228 4 0.0921 1.000 0.028 0.000 0.000 0.972
#> GSM587229 4 0.0921 1.000 0.028 0.000 0.000 0.972
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM587155 2 0.0000 0.993 0.000 1.000 0.000 0.000 0.000
#> GSM587156 2 0.0671 0.985 0.016 0.980 0.004 0.000 0.000
#> GSM587157 2 0.0451 0.990 0.008 0.988 0.004 0.000 0.000
#> GSM587158 2 0.0162 0.993 0.000 0.996 0.004 0.000 0.000
#> GSM587159 2 0.0162 0.993 0.000 0.996 0.004 0.000 0.000
#> GSM587160 2 0.0162 0.993 0.000 0.996 0.004 0.000 0.000
#> GSM587161 2 0.0162 0.992 0.004 0.996 0.000 0.000 0.000
#> GSM587162 2 0.0162 0.993 0.000 0.996 0.004 0.000 0.000
#> GSM587163 2 0.0000 0.993 0.000 1.000 0.000 0.000 0.000
#> GSM587164 2 0.0162 0.992 0.004 0.996 0.000 0.000 0.000
#> GSM587165 2 0.0162 0.993 0.000 0.996 0.004 0.000 0.000
#> GSM587166 2 0.1026 0.977 0.024 0.968 0.004 0.000 0.004
#> GSM587167 2 0.0451 0.990 0.008 0.988 0.004 0.000 0.000
#> GSM587168 2 0.0162 0.993 0.000 0.996 0.004 0.000 0.000
#> GSM587169 2 0.0324 0.991 0.004 0.992 0.004 0.000 0.000
#> GSM587170 2 0.0451 0.990 0.008 0.988 0.004 0.000 0.000
#> GSM587171 2 0.0162 0.992 0.004 0.996 0.000 0.000 0.000
#> GSM587172 2 0.0324 0.991 0.004 0.992 0.004 0.000 0.000
#> GSM587173 2 0.0162 0.993 0.000 0.996 0.004 0.000 0.000
#> GSM587174 2 0.0324 0.992 0.000 0.992 0.004 0.000 0.004
#> GSM587175 2 0.0324 0.991 0.004 0.992 0.004 0.000 0.000
#> GSM587176 2 0.0162 0.993 0.000 0.996 0.004 0.000 0.000
#> GSM587177 2 0.0162 0.993 0.000 0.996 0.004 0.000 0.000
#> GSM587178 2 0.0000 0.993 0.000 1.000 0.000 0.000 0.000
#> GSM587179 2 0.0162 0.992 0.004 0.996 0.000 0.000 0.000
#> GSM587180 2 0.0162 0.992 0.000 0.996 0.000 0.000 0.004
#> GSM587181 2 0.0162 0.993 0.000 0.996 0.004 0.000 0.000
#> GSM587182 2 0.0486 0.990 0.004 0.988 0.004 0.000 0.004
#> GSM587183 2 0.0609 0.985 0.000 0.980 0.020 0.000 0.000
#> GSM587184 2 0.0609 0.985 0.000 0.980 0.020 0.000 0.000
#> GSM587185 2 0.0609 0.985 0.000 0.980 0.020 0.000 0.000
#> GSM587186 2 0.0609 0.985 0.000 0.980 0.020 0.000 0.000
#> GSM587187 3 0.1872 0.885 0.000 0.052 0.928 0.000 0.020
#> GSM587188 3 0.2928 0.854 0.000 0.064 0.872 0.000 0.064
#> GSM587189 3 0.1741 0.903 0.000 0.024 0.936 0.000 0.040
#> GSM587190 3 0.1121 0.910 0.000 0.000 0.956 0.000 0.044
#> GSM587203 1 0.4542 0.297 0.536 0.000 0.000 0.456 0.008
#> GSM587204 5 0.3112 0.800 0.100 0.000 0.000 0.044 0.856
#> GSM587205 1 0.1547 0.865 0.948 0.000 0.004 0.032 0.016
#> GSM587206 1 0.2608 0.852 0.888 0.000 0.004 0.088 0.020
#> GSM587207 1 0.1630 0.865 0.944 0.000 0.004 0.036 0.016
#> GSM587208 1 0.2364 0.861 0.908 0.000 0.008 0.064 0.020
#> GSM587209 1 0.2976 0.853 0.880 0.000 0.064 0.044 0.012
#> GSM587210 5 0.1547 0.856 0.004 0.000 0.016 0.032 0.948
#> GSM587211 1 0.2802 0.827 0.876 0.000 0.100 0.016 0.008
#> GSM587212 5 0.0671 0.860 0.004 0.000 0.016 0.000 0.980
#> GSM587213 1 0.1787 0.865 0.940 0.000 0.032 0.016 0.012
#> GSM587214 1 0.4846 0.366 0.588 0.000 0.000 0.028 0.384
#> GSM587215 5 0.4173 0.533 0.300 0.000 0.000 0.012 0.688
#> GSM587216 5 0.2464 0.817 0.096 0.000 0.000 0.016 0.888
#> GSM587217 5 0.6525 0.172 0.308 0.000 0.000 0.220 0.472
#> GSM587191 3 0.2677 0.880 0.112 0.000 0.872 0.000 0.016
#> GSM587192 5 0.0798 0.860 0.008 0.000 0.016 0.000 0.976
#> GSM587193 1 0.2351 0.839 0.896 0.000 0.088 0.000 0.016
#> GSM587194 5 0.1544 0.843 0.000 0.000 0.068 0.000 0.932
#> GSM587195 3 0.0693 0.918 0.012 0.000 0.980 0.000 0.008
#> GSM587196 5 0.3395 0.684 0.000 0.000 0.236 0.000 0.764
#> GSM587197 3 0.0693 0.918 0.012 0.000 0.980 0.000 0.008
#> GSM587198 3 0.1965 0.888 0.096 0.000 0.904 0.000 0.000
#> GSM587199 5 0.1965 0.827 0.000 0.000 0.096 0.000 0.904
#> GSM587200 1 0.1455 0.863 0.952 0.000 0.032 0.008 0.008
#> GSM587201 1 0.1186 0.863 0.964 0.000 0.020 0.008 0.008
#> GSM587202 1 0.1671 0.844 0.924 0.000 0.076 0.000 0.000
#> GSM198767 1 0.3495 0.809 0.816 0.000 0.000 0.152 0.032
#> GSM198769 1 0.5086 0.632 0.660 0.000 0.016 0.288 0.036
#> GSM198772 1 0.2910 0.863 0.888 0.000 0.052 0.036 0.024
#> GSM198773 1 0.2072 0.867 0.928 0.000 0.016 0.036 0.020
#> GSM198776 5 0.3141 0.796 0.108 0.000 0.000 0.040 0.852
#> GSM198778 5 0.1278 0.859 0.004 0.000 0.016 0.020 0.960
#> GSM198780 5 0.0671 0.860 0.004 0.000 0.016 0.000 0.980
#> GSM198781 1 0.4268 0.625 0.708 0.000 0.000 0.024 0.268
#> GSM198765 3 0.3841 0.793 0.188 0.000 0.780 0.000 0.032
#> GSM198766 1 0.2712 0.839 0.880 0.000 0.032 0.000 0.088
#> GSM198768 3 0.1018 0.918 0.016 0.000 0.968 0.000 0.016
#> GSM198770 3 0.0693 0.918 0.012 0.000 0.980 0.000 0.008
#> GSM198771 3 0.2806 0.842 0.152 0.000 0.844 0.000 0.004
#> GSM198774 5 0.0955 0.859 0.004 0.000 0.028 0.000 0.968
#> GSM198775 5 0.1908 0.830 0.000 0.000 0.092 0.000 0.908
#> GSM198777 3 0.2020 0.879 0.000 0.000 0.900 0.000 0.100
#> GSM198779 5 0.2852 0.752 0.000 0.000 0.172 0.000 0.828
#> GSM587218 4 0.0162 0.994 0.004 0.000 0.000 0.996 0.000
#> GSM587219 4 0.0162 0.994 0.004 0.000 0.000 0.996 0.000
#> GSM587220 4 0.0162 0.994 0.004 0.000 0.000 0.996 0.000
#> GSM587221 4 0.0162 0.994 0.004 0.000 0.000 0.996 0.000
#> GSM587222 4 0.0162 0.994 0.004 0.000 0.000 0.996 0.000
#> GSM587223 4 0.0162 0.994 0.004 0.000 0.000 0.996 0.000
#> GSM587224 4 0.0162 0.994 0.004 0.000 0.000 0.996 0.000
#> GSM587225 4 0.0162 0.994 0.004 0.000 0.000 0.996 0.000
#> GSM587226 4 0.0162 0.994 0.004 0.000 0.000 0.996 0.000
#> GSM587227 4 0.0579 0.986 0.008 0.000 0.000 0.984 0.008
#> GSM587228 4 0.0290 0.991 0.008 0.000 0.000 0.992 0.000
#> GSM587229 4 0.1251 0.958 0.008 0.000 0.000 0.956 0.036
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM587155 2 0.0000 0.985 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587156 2 0.0405 0.980 0.000 0.988 0.000 0.000 0.004 0.008
#> GSM587157 2 0.0000 0.985 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587158 2 0.0000 0.985 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587159 2 0.0000 0.985 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587160 2 0.0000 0.985 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587161 2 0.0146 0.984 0.000 0.996 0.000 0.000 0.000 0.004
#> GSM587162 2 0.0000 0.985 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587163 2 0.0000 0.985 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587164 2 0.0000 0.985 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587165 2 0.0000 0.985 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587166 2 0.0622 0.975 0.000 0.980 0.000 0.000 0.008 0.012
#> GSM587167 2 0.0000 0.985 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587168 2 0.0000 0.985 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587169 2 0.0146 0.984 0.000 0.996 0.000 0.000 0.000 0.004
#> GSM587170 2 0.0146 0.984 0.000 0.996 0.000 0.000 0.000 0.004
#> GSM587171 2 0.0000 0.985 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587172 2 0.0000 0.985 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587173 2 0.0405 0.979 0.000 0.988 0.008 0.000 0.000 0.004
#> GSM587174 2 0.0000 0.985 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587175 2 0.0000 0.985 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587176 2 0.0000 0.985 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587177 2 0.0000 0.985 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587178 2 0.0000 0.985 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587179 2 0.0146 0.984 0.000 0.996 0.000 0.000 0.000 0.004
#> GSM587180 2 0.0000 0.985 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587181 2 0.0000 0.985 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM587182 2 0.0146 0.984 0.000 0.996 0.000 0.000 0.000 0.004
#> GSM587183 2 0.2173 0.914 0.000 0.904 0.028 0.000 0.064 0.004
#> GSM587184 2 0.2034 0.921 0.000 0.912 0.024 0.000 0.060 0.004
#> GSM587185 2 0.2492 0.901 0.000 0.888 0.036 0.000 0.068 0.008
#> GSM587186 2 0.2744 0.871 0.000 0.864 0.072 0.000 0.064 0.000
#> GSM587187 3 0.1736 0.718 0.008 0.020 0.936 0.000 0.032 0.004
#> GSM587188 3 0.3075 0.679 0.028 0.036 0.864 0.000 0.068 0.004
#> GSM587189 3 0.1129 0.723 0.008 0.012 0.964 0.000 0.012 0.004
#> GSM587190 3 0.1630 0.727 0.020 0.000 0.940 0.000 0.024 0.016
#> GSM587203 6 0.6000 0.282 0.004 0.000 0.000 0.368 0.200 0.428
#> GSM587204 1 0.3919 0.739 0.796 0.000 0.000 0.044 0.120 0.040
#> GSM587205 6 0.2432 0.792 0.000 0.000 0.000 0.024 0.100 0.876
#> GSM587206 6 0.2579 0.795 0.000 0.000 0.000 0.040 0.088 0.872
#> GSM587207 6 0.2487 0.797 0.000 0.000 0.000 0.032 0.092 0.876
#> GSM587208 6 0.2457 0.799 0.000 0.000 0.000 0.036 0.084 0.880
#> GSM587209 5 0.4493 0.796 0.008 0.000 0.040 0.060 0.764 0.128
#> GSM587210 1 0.1708 0.799 0.932 0.000 0.000 0.040 0.024 0.004
#> GSM587211 5 0.4430 0.688 0.004 0.000 0.116 0.008 0.744 0.128
#> GSM587212 1 0.0891 0.803 0.968 0.000 0.000 0.008 0.024 0.000
#> GSM587213 5 0.3879 0.774 0.008 0.000 0.020 0.012 0.764 0.196
#> GSM587214 1 0.6351 -0.123 0.412 0.000 0.000 0.036 0.400 0.152
#> GSM587215 1 0.4948 0.521 0.648 0.000 0.000 0.012 0.260 0.080
#> GSM587216 1 0.3092 0.756 0.840 0.000 0.000 0.028 0.120 0.012
#> GSM587217 1 0.6795 0.323 0.500 0.000 0.000 0.164 0.232 0.104
#> GSM587191 3 0.4101 0.554 0.000 0.000 0.580 0.000 0.408 0.012
#> GSM587192 1 0.0858 0.803 0.968 0.000 0.000 0.000 0.028 0.004
#> GSM587193 5 0.3972 0.752 0.016 0.000 0.076 0.000 0.784 0.124
#> GSM587194 1 0.2220 0.778 0.908 0.000 0.044 0.000 0.036 0.012
#> GSM587195 3 0.3607 0.656 0.000 0.000 0.652 0.000 0.348 0.000
#> GSM587196 1 0.2912 0.716 0.816 0.000 0.172 0.000 0.012 0.000
#> GSM587197 3 0.2854 0.736 0.000 0.000 0.792 0.000 0.208 0.000
#> GSM587198 3 0.4232 0.701 0.000 0.000 0.732 0.000 0.168 0.100
#> GSM587199 1 0.3416 0.747 0.836 0.000 0.088 0.000 0.040 0.036
#> GSM587200 6 0.1604 0.756 0.008 0.000 0.024 0.008 0.016 0.944
#> GSM587201 6 0.1829 0.775 0.004 0.000 0.012 0.000 0.064 0.920
#> GSM587202 6 0.2146 0.725 0.008 0.000 0.060 0.000 0.024 0.908
#> GSM198767 6 0.5860 0.272 0.016 0.000 0.000 0.144 0.320 0.520
#> GSM198769 5 0.5641 0.610 0.052 0.000 0.000 0.208 0.632 0.108
#> GSM198772 5 0.4400 0.799 0.028 0.000 0.020 0.064 0.780 0.108
#> GSM198773 5 0.4104 0.777 0.020 0.000 0.000 0.048 0.760 0.172
#> GSM198776 1 0.4016 0.726 0.784 0.000 0.000 0.040 0.136 0.040
#> GSM198778 1 0.1296 0.802 0.952 0.000 0.004 0.032 0.012 0.000
#> GSM198780 1 0.0748 0.802 0.976 0.000 0.004 0.004 0.016 0.000
#> GSM198781 5 0.6028 0.505 0.248 0.000 0.000 0.036 0.560 0.156
#> GSM198765 3 0.4928 0.366 0.032 0.000 0.480 0.000 0.472 0.016
#> GSM198766 5 0.4104 0.760 0.092 0.000 0.028 0.000 0.784 0.096
#> GSM198768 3 0.3966 0.529 0.000 0.000 0.552 0.000 0.444 0.004
#> GSM198770 3 0.3126 0.723 0.000 0.000 0.752 0.000 0.248 0.000
#> GSM198771 3 0.4958 0.659 0.008 0.000 0.660 0.000 0.224 0.108
#> GSM198774 1 0.0508 0.801 0.984 0.000 0.004 0.000 0.012 0.000
#> GSM198775 1 0.2787 0.763 0.872 0.000 0.072 0.000 0.044 0.012
#> GSM198777 3 0.3291 0.706 0.104 0.000 0.828 0.000 0.064 0.004
#> GSM198779 1 0.4823 0.604 0.700 0.000 0.204 0.000 0.056 0.040
#> GSM587218 4 0.0363 0.974 0.000 0.000 0.000 0.988 0.012 0.000
#> GSM587219 4 0.0000 0.982 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM587220 4 0.0000 0.982 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM587221 4 0.0000 0.982 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM587222 4 0.0000 0.982 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM587223 4 0.0000 0.982 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM587224 4 0.0000 0.982 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM587225 4 0.0146 0.980 0.000 0.000 0.000 0.996 0.004 0.000
#> GSM587226 4 0.0000 0.982 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM587227 4 0.0717 0.968 0.016 0.000 0.000 0.976 0.008 0.000
#> GSM587228 4 0.0881 0.965 0.008 0.000 0.000 0.972 0.012 0.008
#> GSM587229 4 0.2321 0.879 0.052 0.000 0.000 0.900 0.040 0.008
Heatmaps for the consensus matrix. It visualizes the probability of two samples to be in a same group.
consensus_heatmap(res, k = 2)
consensus_heatmap(res, k = 3)
consensus_heatmap(res, k = 4)
consensus_heatmap(res, k = 5)
consensus_heatmap(res, k = 6)
Heatmaps for the membership of samples in all partitions to see how consistent they are:
membership_heatmap(res, k = 2)
membership_heatmap(res, k = 3)
membership_heatmap(res, k = 4)
membership_heatmap(res, k = 5)
membership_heatmap(res, k = 6)
As soon as we have had the classes for columns, we can look for signatures which are significantly different between classes which can be candidate marks for certain classes. Following are the heatmaps for signatures.
Signature heatmaps where rows are scaled:
get_signatures(res, k = 2)
get_signatures(res, k = 3)
get_signatures(res, k = 4)
get_signatures(res, k = 5)
get_signatures(res, k = 6)
Signature heatmaps where rows are not scaled:
get_signatures(res, k = 2, scale_rows = FALSE)
get_signatures(res, k = 3, scale_rows = FALSE)
get_signatures(res, k = 4, scale_rows = FALSE)
get_signatures(res, k = 5, scale_rows = FALSE)
get_signatures(res, k = 6, scale_rows = FALSE)
Compare the overlap of signatures from different k:
compare_signatures(res)
get_signature() returns a data frame invisibly. TO get the list of signatures, the function
call should be assigned to a variable explicitly. In following code, if plot argument is set
to FALSE, no heatmap is plotted while only the differential analysis is performed.
# code only for demonstration
tb = get_signature(res, k = ..., plot = FALSE)
An example of the output of tb is:
#> which_row fdr mean_1 mean_2 scaled_mean_1 scaled_mean_2 km
#> 1 38 0.042760348 8.373488 9.131774 -0.5533452 0.5164555 1
#> 2 40 0.018707592 7.106213 8.469186 -0.6173731 0.5762149 1
#> 3 55 0.019134737 10.221463 11.207825 -0.6159697 0.5749050 1
#> 4 59 0.006059896 5.921854 7.869574 -0.6899429 0.6439467 1
#> 5 60 0.018055526 8.928898 10.211722 -0.6204761 0.5791110 1
#> 6 98 0.009384629 15.714769 14.887706 0.6635654 -0.6193277 2
...
The columns in tb are:
which_row: row indices corresponding to the input matrix.fdr: FDR for the differential test. mean_x: The mean value in group x.scaled_mean_x: The mean value in group x after rows are scaled.km: Row groups if k-means clustering is applied to rows.UMAP plot which shows how samples are separated.
dimension_reduction(res, k = 2, method = "UMAP")
dimension_reduction(res, k = 3, method = "UMAP")
dimension_reduction(res, k = 4, method = "UMAP")
dimension_reduction(res, k = 5, method = "UMAP")
dimension_reduction(res, k = 6, method = "UMAP")
Following heatmap shows how subgroups are split when increasing k:
collect_classes(res)
Test correlation between subgroups and known annotations. If the known annotation is numeric, one-way ANOVA test is applied, and if the known annotation is discrete, chi-squared contingency table test is applied.
test_to_known_factors(res)
#> n specimen(p) k
#> ATC:NMF 92 7.21e-17 2
#> ATC:NMF 84 5.05e-28 3
#> ATC:NMF 90 5.52e-35 4
#> ATC:NMF 89 3.95e-34 5
#> ATC:NMF 87 6.15e-33 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