Date: 2019-12-25 20:17:15 CET, cola version: 1.3.2
Document is loading...
All available functions which can be applied to this res_list object:
res_list
#> A 'ConsensusPartitionList' object with 24 methods.
#> On a matrix with 21168 rows and 100 columns.
#> Top rows are extracted by 'SD, CV, MAD, ATC' methods.
#> Subgroups are detected by 'hclust, kmeans, skmeans, pam, mclust, NMF' method.
#> Number of partitions are tried for k = 2, 3, 4, 5, 6.
#> Performed in total 30000 partitions by row resampling.
#>
#> Following methods can be applied to this 'ConsensusPartitionList' object:
#> [1] "cola_report" "collect_classes" "collect_plots" "collect_stats"
#> [5] "colnames" "functional_enrichment" "get_anno_col" "get_anno"
#> [9] "get_classes" "get_matrix" "get_membership" "get_stats"
#> [13] "is_best_k" "is_stable_k" "ncol" "nrow"
#> [17] "rownames" "show" "suggest_best_k" "test_to_known_factors"
#> [21] "top_rows_heatmap" "top_rows_overlap"
#>
#> You can get result for a single method by, e.g. object["SD", "hclust"] or object["SD:hclust"]
#> or a subset of methods by object[c("SD", "CV")], c("hclust", "kmeans")]
The call of run_all_consensus_partition_methods() was:
#> run_all_consensus_partition_methods(data = mat, mc.cores = 4, anno = anno)
Dimension of the input matrix:
mat = get_matrix(res_list)
dim(mat)
#> [1] 21168 100
The density distribution for each sample is visualized as in one column in the following heatmap. The clustering is based on the distance which is the Kolmogorov-Smirnov statistic between two distributions.
library(ComplexHeatmap)
densityHeatmap(mat, top_annotation = HeatmapAnnotation(df = get_anno(res_list),
col = get_anno_col(res_list)), ylab = "value", cluster_columns = TRUE, show_column_names = FALSE,
mc.cores = 4)
Folowing table shows the best k (number of partitions) for each combination
of top-value methods and partition methods. Clicking on the method name in
the table goes to the section for a single combination of methods.
The cola vignette explains the definition of the metrics used for determining the best number of partitions.
suggest_best_k(res_list)
| The best k | 1-PAC | Mean silhouette | Concordance | Optional k | ||
|---|---|---|---|---|---|---|
| SD:kmeans | 2 | 1.000 | 0.951 | 0.980 | ** | |
| SD:skmeans | 2 | 1.000 | 0.986 | 0.994 | ** | |
| CV:kmeans | 2 | 1.000 | 0.980 | 0.991 | ** | |
| CV:skmeans | 2 | 1.000 | 0.983 | 0.992 | ** | |
| CV:NMF | 2 | 1.000 | 0.972 | 0.988 | ** | |
| MAD:kmeans | 2 | 1.000 | 0.963 | 0.985 | ** | |
| MAD:mclust | 2 | 1.000 | 0.974 | 0.980 | ** | |
| MAD:NMF | 2 | 1.000 | 0.966 | 0.986 | ** | |
| ATC:NMF | 2 | 1.000 | 0.979 | 0.991 | ** | |
| ATC:kmeans | 3 | 0.976 | 0.966 | 0.977 | ** | 2 |
| MAD:skmeans | 3 | 0.974 | 0.953 | 0.976 | ** | 2 |
| ATC:skmeans | 6 | 0.962 | 0.925 | 0.947 | ** | 2,3 |
| ATC:mclust | 6 | 0.941 | 0.887 | 0.951 | * | 2,4 |
| SD:NMF | 3 | 0.926 | 0.915 | 0.959 | * | 2 |
| CV:mclust | 4 | 0.910 | 0.904 | 0.960 | * | |
| ATC:pam | 3 | 0.908 | 0.948 | 0.978 | * | 2 |
| MAD:pam | 2 | 0.898 | 0.946 | 0.971 | ||
| SD:pam | 2 | 0.894 | 0.935 | 0.970 | ||
| SD:mclust | 4 | 0.861 | 0.891 | 0.946 | ||
| CV:pam | 2 | 0.626 | 0.862 | 0.929 | ||
| SD:hclust | 3 | 0.580 | 0.862 | 0.908 | ||
| ATC:hclust | 3 | 0.538 | 0.735 | 0.872 | ||
| MAD:hclust | 3 | 0.488 | 0.809 | 0.885 | ||
| CV:hclust | 2 | 0.318 | 0.647 | 0.841 |
**: 1-PAC > 0.95, *: 1-PAC > 0.9
Cumulative distribution function curves of consensus matrix for all methods.
collect_plots(res_list, fun = plot_ecdf)
Consensus heatmaps for all methods. (What is a consensus heatmap?)
collect_plots(res_list, k = 2, fun = consensus_heatmap, mc.cores = 4)
collect_plots(res_list, k = 3, fun = consensus_heatmap, mc.cores = 4)
collect_plots(res_list, k = 4, fun = consensus_heatmap, mc.cores = 4)
collect_plots(res_list, k = 5, fun = consensus_heatmap, mc.cores = 4)
collect_plots(res_list, k = 6, fun = consensus_heatmap, mc.cores = 4)
Membership heatmaps for all methods. (What is a membership heatmap?)
collect_plots(res_list, k = 2, fun = membership_heatmap, mc.cores = 4)
collect_plots(res_list, k = 3, fun = membership_heatmap, mc.cores = 4)
collect_plots(res_list, k = 4, fun = membership_heatmap, mc.cores = 4)
collect_plots(res_list, k = 5, fun = membership_heatmap, mc.cores = 4)
collect_plots(res_list, k = 6, fun = membership_heatmap, mc.cores = 4)
Signature heatmaps for all methods. (What is a signature heatmap?)
Note in following heatmaps, rows are scaled.
collect_plots(res_list, k = 2, fun = get_signatures, mc.cores = 4)
collect_plots(res_list, k = 3, fun = get_signatures, mc.cores = 4)
collect_plots(res_list, k = 4, fun = get_signatures, mc.cores = 4)
collect_plots(res_list, k = 5, fun = get_signatures, mc.cores = 4)
collect_plots(res_list, k = 6, fun = get_signatures, mc.cores = 4)
The statistics used for measuring the stability of consensus partitioning. (How are they defined?)
get_stats(res_list, k = 2)
#> k 1-PAC mean_silhouette concordance area_increased Rand Jaccard
#> SD:NMF 2 0.999 0.971 0.987 0.495 0.505 0.505
#> CV:NMF 2 1.000 0.972 0.988 0.496 0.505 0.505
#> MAD:NMF 2 1.000 0.966 0.986 0.494 0.508 0.508
#> ATC:NMF 2 1.000 0.979 0.991 0.502 0.500 0.500
#> SD:skmeans 2 1.000 0.986 0.994 0.501 0.500 0.500
#> CV:skmeans 2 1.000 0.983 0.992 0.500 0.500 0.500
#> MAD:skmeans 2 1.000 0.986 0.993 0.500 0.500 0.500
#> ATC:skmeans 2 1.000 0.983 0.993 0.505 0.495 0.495
#> SD:mclust 2 0.600 0.915 0.948 0.365 0.642 0.642
#> CV:mclust 2 0.801 0.915 0.954 0.350 0.665 0.665
#> MAD:mclust 2 1.000 0.974 0.980 0.343 0.665 0.665
#> ATC:mclust 2 1.000 0.975 0.990 0.216 0.787 0.787
#> SD:kmeans 2 1.000 0.951 0.980 0.488 0.508 0.508
#> CV:kmeans 2 1.000 0.980 0.991 0.494 0.508 0.508
#> MAD:kmeans 2 1.000 0.963 0.985 0.487 0.519 0.519
#> ATC:kmeans 2 1.000 0.965 0.976 0.497 0.495 0.495
#> SD:pam 2 0.894 0.935 0.970 0.460 0.547 0.547
#> CV:pam 2 0.626 0.862 0.929 0.487 0.508 0.508
#> MAD:pam 2 0.898 0.946 0.971 0.463 0.540 0.540
#> ATC:pam 2 0.958 0.933 0.971 0.491 0.515 0.515
#> SD:hclust 2 0.466 0.884 0.925 0.387 0.653 0.653
#> CV:hclust 2 0.318 0.647 0.841 0.434 0.560 0.560
#> MAD:hclust 2 0.399 0.685 0.857 0.429 0.602 0.602
#> ATC:hclust 2 0.529 0.885 0.923 0.250 0.818 0.818
get_stats(res_list, k = 3)
#> k 1-PAC mean_silhouette concordance area_increased Rand Jaccard
#> SD:NMF 3 0.926 0.915 0.959 0.318 0.787 0.600
#> CV:NMF 3 0.529 0.701 0.840 0.325 0.807 0.634
#> MAD:NMF 3 0.889 0.900 0.952 0.322 0.791 0.609
#> ATC:NMF 3 0.846 0.862 0.933 0.284 0.699 0.478
#> SD:skmeans 3 0.837 0.942 0.966 0.307 0.788 0.600
#> CV:skmeans 3 0.671 0.414 0.731 0.322 0.754 0.543
#> MAD:skmeans 3 0.974 0.953 0.976 0.307 0.788 0.600
#> ATC:skmeans 3 0.950 0.912 0.964 0.288 0.822 0.652
#> SD:mclust 3 0.589 0.716 0.861 0.720 0.708 0.549
#> CV:mclust 3 0.469 0.437 0.711 0.726 0.792 0.692
#> MAD:mclust 3 0.862 0.879 0.946 0.811 0.659 0.507
#> ATC:mclust 3 0.587 0.791 0.903 1.614 0.600 0.498
#> SD:kmeans 3 0.630 0.793 0.885 0.332 0.751 0.551
#> CV:kmeans 3 0.487 0.385 0.682 0.305 0.772 0.606
#> MAD:kmeans 3 0.727 0.827 0.904 0.335 0.759 0.566
#> ATC:kmeans 3 0.976 0.966 0.977 0.196 0.899 0.800
#> SD:pam 3 0.435 0.435 0.738 0.365 0.872 0.775
#> CV:pam 3 0.518 0.701 0.854 0.365 0.732 0.515
#> MAD:pam 3 0.473 0.503 0.752 0.368 0.757 0.569
#> ATC:pam 3 0.908 0.948 0.978 0.194 0.899 0.804
#> SD:hclust 3 0.580 0.862 0.908 0.592 0.732 0.590
#> CV:hclust 3 0.391 0.699 0.817 0.423 0.776 0.617
#> MAD:hclust 3 0.488 0.809 0.885 0.443 0.743 0.588
#> ATC:hclust 3 0.538 0.735 0.872 1.250 0.596 0.506
get_stats(res_list, k = 4)
#> k 1-PAC mean_silhouette concordance area_increased Rand Jaccard
#> SD:NMF 4 0.614 0.570 0.755 0.137 0.805 0.505
#> CV:NMF 4 0.587 0.564 0.793 0.126 0.827 0.567
#> MAD:NMF 4 0.626 0.628 0.753 0.137 0.844 0.595
#> ATC:NMF 4 0.828 0.865 0.930 0.133 0.836 0.582
#> SD:skmeans 4 0.768 0.695 0.876 0.139 0.851 0.599
#> CV:skmeans 4 0.681 0.740 0.870 0.132 0.806 0.495
#> MAD:skmeans 4 0.761 0.672 0.848 0.140 0.844 0.585
#> ATC:skmeans 4 0.719 0.764 0.788 0.121 0.862 0.632
#> SD:mclust 4 0.861 0.891 0.946 0.155 0.889 0.705
#> CV:mclust 4 0.910 0.904 0.960 0.209 0.698 0.422
#> MAD:mclust 4 0.853 0.868 0.936 0.178 0.876 0.678
#> ATC:mclust 4 0.983 0.955 0.983 0.179 0.901 0.766
#> SD:kmeans 4 0.764 0.513 0.746 0.129 0.913 0.758
#> CV:kmeans 4 0.716 0.731 0.851 0.137 0.747 0.472
#> MAD:kmeans 4 0.801 0.744 0.864 0.139 0.827 0.549
#> ATC:kmeans 4 0.597 0.605 0.763 0.193 0.829 0.604
#> SD:pam 4 0.527 0.562 0.766 0.161 0.687 0.390
#> CV:pam 4 0.528 0.551 0.760 0.118 0.863 0.620
#> MAD:pam 4 0.580 0.654 0.809 0.155 0.812 0.529
#> ATC:pam 4 0.649 0.763 0.837 0.208 0.855 0.664
#> SD:hclust 4 0.678 0.839 0.872 0.140 0.920 0.793
#> CV:hclust 4 0.522 0.729 0.830 0.131 0.907 0.763
#> MAD:hclust 4 0.565 0.741 0.830 0.130 0.920 0.793
#> ATC:hclust 4 0.557 0.701 0.839 0.197 0.903 0.765
get_stats(res_list, k = 5)
#> k 1-PAC mean_silhouette concordance area_increased Rand Jaccard
#> SD:NMF 5 0.599 0.528 0.738 0.0717 0.842 0.474
#> CV:NMF 5 0.591 0.526 0.746 0.0726 0.817 0.439
#> MAD:NMF 5 0.600 0.549 0.737 0.0727 0.844 0.495
#> ATC:NMF 5 0.686 0.651 0.825 0.0634 0.911 0.699
#> SD:skmeans 5 0.735 0.748 0.841 0.0635 0.879 0.582
#> CV:skmeans 5 0.643 0.589 0.724 0.0591 0.904 0.647
#> MAD:skmeans 5 0.714 0.715 0.826 0.0631 0.879 0.584
#> ATC:skmeans 5 0.783 0.773 0.816 0.0712 0.937 0.769
#> SD:mclust 5 0.772 0.752 0.828 0.0727 0.937 0.785
#> CV:mclust 5 0.805 0.705 0.834 0.0585 0.956 0.846
#> MAD:mclust 5 0.791 0.860 0.877 0.0764 0.898 0.649
#> ATC:mclust 5 0.736 0.785 0.882 0.1119 0.881 0.673
#> SD:kmeans 5 0.694 0.655 0.802 0.0655 0.874 0.603
#> CV:kmeans 5 0.664 0.572 0.758 0.0676 0.865 0.584
#> MAD:kmeans 5 0.717 0.774 0.828 0.0624 0.909 0.667
#> ATC:kmeans 5 0.763 0.692 0.832 0.1012 0.805 0.429
#> SD:pam 5 0.692 0.657 0.820 0.0783 0.862 0.541
#> CV:pam 5 0.627 0.645 0.795 0.0650 0.848 0.502
#> MAD:pam 5 0.699 0.774 0.837 0.0753 0.874 0.583
#> ATC:pam 5 0.860 0.792 0.915 0.1006 0.872 0.608
#> SD:hclust 5 0.688 0.754 0.792 0.0890 0.900 0.677
#> CV:hclust 5 0.575 0.676 0.773 0.0671 1.000 1.000
#> MAD:hclust 5 0.625 0.687 0.767 0.0862 0.909 0.704
#> ATC:hclust 5 0.545 0.539 0.743 0.0892 0.939 0.808
get_stats(res_list, k = 6)
#> k 1-PAC mean_silhouette concordance area_increased Rand Jaccard
#> SD:NMF 6 0.631 0.528 0.722 0.0448 0.840 0.392
#> CV:NMF 6 0.626 0.485 0.696 0.0467 0.884 0.512
#> MAD:NMF 6 0.622 0.460 0.709 0.0441 0.852 0.417
#> ATC:NMF 6 0.652 0.538 0.752 0.0414 0.916 0.672
#> SD:skmeans 6 0.733 0.668 0.766 0.0389 0.970 0.858
#> CV:skmeans 6 0.640 0.493 0.704 0.0424 0.940 0.726
#> MAD:skmeans 6 0.705 0.548 0.720 0.0399 0.977 0.894
#> ATC:skmeans 6 0.962 0.925 0.947 0.0600 0.924 0.672
#> SD:mclust 6 0.874 0.891 0.918 0.0640 0.906 0.626
#> CV:mclust 6 0.772 0.771 0.787 0.0460 0.874 0.545
#> MAD:mclust 6 0.827 0.833 0.838 0.0429 0.948 0.756
#> ATC:mclust 6 0.941 0.887 0.951 0.1120 0.878 0.572
#> SD:kmeans 6 0.746 0.759 0.807 0.0457 0.916 0.660
#> CV:kmeans 6 0.709 0.664 0.743 0.0454 0.921 0.680
#> MAD:kmeans 6 0.747 0.731 0.787 0.0447 0.967 0.844
#> ATC:kmeans 6 0.837 0.866 0.882 0.0536 0.923 0.658
#> SD:pam 6 0.701 0.495 0.684 0.0500 0.909 0.613
#> CV:pam 6 0.627 0.495 0.686 0.0425 0.931 0.695
#> MAD:pam 6 0.723 0.598 0.750 0.0516 0.950 0.767
#> ATC:pam 6 0.854 0.754 0.887 0.0452 0.928 0.696
#> SD:hclust 6 0.742 0.806 0.868 0.0535 0.962 0.827
#> CV:hclust 6 0.607 0.579 0.741 0.0594 0.893 0.656
#> MAD:hclust 6 0.687 0.721 0.806 0.0549 0.962 0.827
#> ATC:hclust 6 0.589 0.608 0.725 0.0537 0.834 0.478
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 disease.state(p) specimen(p) cell.type(p) other(p) k
#> SD:NMF 99 1.61e-05 0.166 3.04e-14 0.13812 2
#> CV:NMF 100 2.42e-05 0.251 1.12e-13 0.18012 2
#> MAD:NMF 99 1.61e-05 0.166 3.04e-14 0.13812 2
#> ATC:NMF 99 6.29e-06 0.617 1.57e-11 0.28570 2
#> SD:skmeans 99 1.18e-04 0.225 6.66e-13 0.10663 2
#> CV:skmeans 100 9.36e-05 0.173 5.21e-13 0.09028 2
#> MAD:skmeans 100 9.36e-05 0.173 5.21e-13 0.09028 2
#> ATC:skmeans 100 1.24e-07 0.714 5.25e-14 0.10191 2
#> SD:mclust 98 2.79e-05 0.404 8.16e-12 0.01147 2
#> CV:mclust 99 6.56e-07 0.570 2.99e-10 0.03080 2
#> MAD:mclust 100 6.54e-06 0.438 1.48e-10 0.00517 2
#> ATC:mclust 99 3.22e-02 0.424 8.13e-04 0.46276 2
#> SD:kmeans 98 4.41e-06 0.222 5.40e-15 0.06905 2
#> CV:kmeans 100 1.13e-05 0.298 4.23e-14 0.12314 2
#> MAD:kmeans 99 4.79e-06 0.212 5.56e-16 0.03262 2
#> ATC:kmeans 100 1.24e-07 0.714 5.25e-14 0.10191 2
#> SD:pam 98 7.18e-09 0.530 2.47e-19 0.00887 2
#> CV:pam 95 1.87e-09 0.519 9.94e-19 0.04830 2
#> MAD:pam 100 1.39e-06 0.440 9.32e-16 0.05269 2
#> ATC:pam 94 3.02e-09 0.472 7.85e-18 0.01100 2
#> SD:hclust 99 1.20e-04 0.166 2.00e-10 0.00263 2
#> CV:hclust 79 9.28e-04 0.304 3.11e-12 0.03868 2
#> MAD:hclust 78 4.12e-04 0.199 7.19e-12 0.00845 2
#> ATC:hclust 100 5.90e-02 0.526 4.19e-04 0.45353 2
test_to_known_factors(res_list, k = 3)
#> n disease.state(p) specimen(p) cell.type(p) other(p) k
#> SD:NMF 96 3.56e-05 0.1686 1.18e-17 0.0543 3
#> CV:NMF 90 8.38e-06 0.1059 9.92e-20 0.0238 3
#> MAD:NMF 97 7.29e-05 0.2888 5.42e-18 0.0784 3
#> ATC:NMF 94 9.01e-05 0.0862 1.22e-21 0.1605 3
#> SD:skmeans 100 1.14e-04 0.2896 6.36e-17 0.0685 3
#> CV:skmeans 37 3.62e-01 1.0000 2.96e-03 0.2540 3
#> MAD:skmeans 100 1.14e-04 0.2896 6.36e-17 0.0685 3
#> ATC:skmeans 93 4.35e-06 0.5150 2.47e-18 0.3916 3
#> SD:mclust 77 8.69e-03 0.4160 3.41e-15 0.1125 3
#> CV:mclust 59 3.59e-03 0.3146 6.24e-13 0.2008 3
#> MAD:mclust 92 8.69e-05 0.4307 9.63e-18 0.0340 3
#> ATC:mclust 96 4.68e-06 0.1355 2.76e-16 0.1236 3
#> SD:kmeans 94 1.92e-04 0.4126 1.41e-18 0.0374 3
#> CV:kmeans 44 NA NA NA NA 3
#> MAD:kmeans 97 3.86e-05 0.2944 5.38e-19 0.0178 3
#> ATC:kmeans 100 1.01e-06 0.2295 3.56e-15 0.2241 3
#> SD:pam 46 3.33e-04 0.1775 1.03e-10 0.0697 3
#> CV:pam 84 3.06e-07 0.5632 3.99e-15 0.1185 3
#> MAD:pam 54 3.15e-03 0.8100 1.13e-13 0.1281 3
#> ATC:pam 100 1.61e-07 0.2208 1.13e-18 0.0321 3
#> SD:hclust 99 1.47e-03 0.1882 8.26e-14 0.0197 3
#> CV:hclust 85 6.37e-04 0.4828 2.04e-15 0.0545 3
#> MAD:hclust 95 2.70e-03 0.1882 1.66e-13 0.0301 3
#> ATC:hclust 89 6.61e-08 0.1775 4.27e-16 0.0620 3
test_to_known_factors(res_list, k = 4)
#> n disease.state(p) specimen(p) cell.type(p) other(p) k
#> SD:NMF 69 3.15e-04 0.0933 6.59e-12 0.03948 4
#> CV:NMF 70 3.14e-04 0.4693 2.72e-14 0.13855 4
#> MAD:NMF 84 1.19e-05 0.0649 5.06e-18 0.03217 4
#> ATC:NMF 96 7.69e-04 0.3426 1.63e-19 0.10578 4
#> SD:skmeans 78 2.01e-04 0.1674 3.84e-13 0.03117 4
#> CV:skmeans 87 1.23e-04 0.1849 1.40e-17 0.05833 4
#> MAD:skmeans 76 7.71e-05 0.4814 5.77e-13 0.02462 4
#> ATC:skmeans 91 6.42e-06 0.6007 1.86e-17 0.25032 4
#> SD:mclust 98 3.34e-05 0.0724 1.27e-18 0.02800 4
#> CV:mclust 96 8.49e-06 0.0410 2.31e-19 0.01710 4
#> MAD:mclust 96 1.04e-05 0.0566 5.29e-19 0.03038 4
#> ATC:mclust 98 2.37e-04 0.3213 3.62e-21 0.12438 4
#> SD:kmeans 46 4.76e-03 0.5495 1.31e-05 0.02305 4
#> CV:kmeans 85 2.83e-04 0.1406 1.07e-17 0.03455 4
#> MAD:kmeans 86 4.63e-05 0.2382 4.32e-13 0.02198 4
#> ATC:kmeans 70 6.91e-05 0.2979 8.78e-14 0.25065 4
#> SD:pam 68 1.04e-04 0.4784 5.69e-15 0.00165 4
#> CV:pam 60 7.60e-04 0.5942 2.81e-12 0.24305 4
#> MAD:pam 85 1.60e-05 0.8256 1.90e-17 0.00887 4
#> ATC:pam 95 2.77e-06 0.5445 2.06e-16 0.41837 4
#> SD:hclust 99 5.97e-04 0.0720 5.23e-16 0.02581 4
#> CV:hclust 90 3.28e-04 0.1786 6.02e-16 0.03872 4
#> MAD:hclust 95 1.09e-03 0.0924 8.74e-15 0.03654 4
#> ATC:hclust 88 6.64e-06 0.1174 1.27e-19 0.00812 4
test_to_known_factors(res_list, k = 5)
#> n disease.state(p) specimen(p) cell.type(p) other(p) k
#> SD:NMF 68 3.97e-03 0.2354 2.68e-14 0.04537 5
#> CV:NMF 72 1.45e-02 0.6117 1.16e-13 0.11178 5
#> MAD:NMF 69 2.76e-03 0.3427 4.61e-15 0.01764 5
#> ATC:NMF 78 1.26e-04 0.1133 1.89e-19 0.04313 5
#> SD:skmeans 88 6.48e-05 0.2277 3.50e-16 0.03502 5
#> CV:skmeans 74 1.74e-05 0.0624 8.83e-17 0.01379 5
#> MAD:skmeans 82 2.08e-05 0.4374 4.16e-16 0.06484 5
#> ATC:skmeans 86 3.28e-05 0.1786 7.68e-16 0.31132 5
#> SD:mclust 93 1.18e-05 0.0572 2.11e-19 0.01013 5
#> CV:mclust 93 3.60e-05 0.0441 9.45e-18 0.01237 5
#> MAD:mclust 97 1.30e-04 0.2812 4.26e-16 0.24324 5
#> ATC:mclust 83 5.25e-06 0.1483 1.34e-19 0.02197 5
#> SD:kmeans 83 5.43e-03 0.4675 2.72e-12 0.22397 5
#> CV:kmeans 79 5.99e-03 0.3902 3.70e-13 0.11470 5
#> MAD:kmeans 96 5.29e-05 0.2139 2.61e-15 0.13189 5
#> ATC:kmeans 73 5.03e-04 0.4267 5.55e-11 0.10647 5
#> SD:pam 80 1.45e-06 0.2105 6.75e-17 0.01235 5
#> CV:pam 79 3.36e-05 0.4310 7.14e-15 0.15301 5
#> MAD:pam 95 1.81e-07 0.2789 1.77e-20 0.01760 5
#> ATC:pam 88 1.57e-04 0.4898 5.32e-21 0.30149 5
#> SD:hclust 95 2.27e-04 0.1254 6.70e-15 0.08524 5
#> CV:hclust 89 1.87e-04 0.1611 2.22e-16 0.03245 5
#> MAD:hclust 90 1.91e-04 0.1509 8.12e-13 0.09718 5
#> ATC:hclust 61 4.28e-04 0.1236 1.61e-17 0.00533 5
test_to_known_factors(res_list, k = 6)
#> n disease.state(p) specimen(p) cell.type(p) other(p) k
#> SD:NMF 65 4.40e-03 0.4976 7.36e-13 0.0693 6
#> CV:NMF 50 3.71e-04 0.2514 9.45e-12 0.0525 6
#> MAD:NMF 60 1.92e-03 0.6206 9.68e-14 0.0968 6
#> ATC:NMF 61 7.20e-05 0.2446 1.92e-14 0.1045 6
#> SD:skmeans 83 4.09e-05 0.3370 3.40e-18 0.0216 6
#> CV:skmeans 60 3.04e-04 0.3256 1.60e-17 0.0982 6
#> MAD:skmeans 59 6.82e-05 0.4207 1.51e-15 0.1276 6
#> ATC:skmeans 99 1.82e-05 0.4471 5.89e-16 0.4205 6
#> SD:mclust 99 3.41e-06 0.0900 1.47e-19 0.0315 6
#> CV:mclust 94 3.88e-06 0.1120 4.63e-20 0.0189 6
#> MAD:mclust 96 8.80e-06 0.1944 1.18e-18 0.0496 6
#> ATC:mclust 94 8.30e-05 0.3993 1.53e-18 0.2559 6
#> SD:kmeans 94 1.45e-04 0.3720 2.03e-16 0.0504 6
#> CV:kmeans 82 4.78e-05 0.2056 1.42e-15 0.0045 6
#> MAD:kmeans 91 7.04e-05 0.3876 1.34e-15 0.0362 6
#> ATC:kmeans 98 8.19e-05 0.5824 5.47e-16 0.4758 6
#> SD:pam 51 4.23e-03 0.1398 2.03e-11 0.0623 6
#> CV:pam 58 3.68e-02 0.4973 2.34e-10 0.1339 6
#> MAD:pam 70 1.72e-06 0.3386 2.38e-14 0.0200 6
#> ATC:pam 84 4.04e-04 0.9368 7.19e-19 0.4925 6
#> SD:hclust 97 3.55e-04 0.1647 1.98e-17 0.0327 6
#> CV:hclust 70 4.39e-04 0.2097 3.25e-14 0.0128 6
#> MAD:hclust 95 2.67e-04 0.2186 3.97e-16 0.0232 6
#> ATC:hclust 62 5.10e-04 0.0427 9.55e-18 0.0149 6
The object with results only for a single top-value method and a single partition method can be extracted as:
res = res_list["SD", "hclust"]
# you can also extract it by
# res = res_list["SD:hclust"]
A summary of res and all the functions that can be applied to it:
res
#> A 'ConsensusPartition' object with k = 2, 3, 4, 5, 6.
#> On a matrix with 21168 rows and 100 columns.
#> Top rows (1000, 2000, 3000, 4000, 5000) are extracted by 'SD' method.
#> Subgroups are detected by 'hclust' method.
#> Performed in total 1250 partitions by row resampling.
#> Best k for subgroups seems to be 3.
#>
#> Following methods can be applied to this 'ConsensusPartition' object:
#> [1] "cola_report" "collect_classes" "collect_plots"
#> [4] "collect_stats" "colnames" "compare_signatures"
#> [7] "consensus_heatmap" "dimension_reduction" "functional_enrichment"
#> [10] "get_anno_col" "get_anno" "get_classes"
#> [13] "get_consensus" "get_matrix" "get_membership"
#> [16] "get_param" "get_signatures" "get_stats"
#> [19] "is_best_k" "is_stable_k" "membership_heatmap"
#> [22] "ncol" "nrow" "plot_ecdf"
#> [25] "rownames" "select_partition_number" "show"
#> [28] "suggest_best_k" "test_to_known_factors"
collect_plots() function collects all the plots made from res for all k (number of partitions)
into one single page to provide an easy and fast comparison between different k.
collect_plots(res)
The plots are:
k and the heatmap of
predicted classes for each k.k.k.k.All the plots in panels can be made by individual functions and they are plotted later in this section.
select_partition_number() produces several plots showing different
statistics for choosing “optimized” k. There are following statistics:
k;k, the area increased is defined as \(A_k - A_{k-1}\).The detailed explanations of these statistics can be found in the cola vignette.
Generally speaking, lower PAC score, higher mean silhouette score or higher
concordance corresponds to better partition. Rand index and Jaccard index
measure how similar the current partition is compared to partition with k-1.
If they are too similar, we won't accept k is better than k-1.
select_partition_number(res)
The numeric values for all these statistics can be obtained by get_stats().
get_stats(res)
#> k 1-PAC mean_silhouette concordance area_increased Rand Jaccard
#> 2 2 0.466 0.884 0.925 0.3875 0.653 0.653
#> 3 3 0.580 0.862 0.908 0.5923 0.732 0.590
#> 4 4 0.678 0.839 0.872 0.1405 0.920 0.793
#> 5 5 0.688 0.754 0.792 0.0890 0.900 0.677
#> 6 6 0.742 0.806 0.868 0.0535 0.962 0.827
suggest_best_k() suggests the best \(k\) based on these statistics. The rules are as follows:
suggest_best_k(res)
#> [1] 3
Following shows the table of the partitions (You need to click the show/hide
code output link to see it). The membership matrix (columns with name p*)
is inferred by
clue::cl_consensus()
function with the SE method. Basically the value in the membership matrix
represents the probability to belong to a certain group. The finall class
label for an item is determined with the group with highest probability it
belongs to.
In get_classes() function, the entropy is calculated from the membership
matrix and the silhouette score is calculated from the consensus matrix.
cbind(get_classes(res, k = 2), get_membership(res, k = 2))
#> class entropy silhouette p1 p2
#> GSM97038 2 0.6801 0.7484 0.180 0.820
#> GSM97045 2 0.0938 0.9479 0.012 0.988
#> GSM97047 1 0.7883 0.7740 0.764 0.236
#> GSM97025 2 0.0938 0.9479 0.012 0.988
#> GSM97030 1 0.6973 0.8426 0.812 0.188
#> GSM97027 2 0.0938 0.9479 0.012 0.988
#> GSM97033 2 0.0376 0.9524 0.004 0.996
#> GSM97034 1 0.6801 0.8371 0.820 0.180
#> GSM97020 2 0.0672 0.9504 0.008 0.992
#> GSM97026 1 0.6531 0.8395 0.832 0.168
#> GSM97012 2 0.0000 0.9541 0.000 1.000
#> GSM97015 1 0.6973 0.8426 0.812 0.188
#> GSM97016 2 0.0000 0.9541 0.000 1.000
#> GSM97017 1 0.5059 0.8812 0.888 0.112
#> GSM97019 2 0.0000 0.9541 0.000 1.000
#> GSM97022 2 0.0000 0.9541 0.000 1.000
#> GSM97035 2 0.0000 0.9541 0.000 1.000
#> GSM97036 1 0.2603 0.9078 0.956 0.044
#> GSM97039 2 0.0000 0.9541 0.000 1.000
#> GSM97046 2 0.0000 0.9541 0.000 1.000
#> GSM97023 1 0.0000 0.9120 1.000 0.000
#> GSM97029 1 0.6343 0.8511 0.840 0.160
#> GSM97043 1 0.8713 0.7020 0.708 0.292
#> GSM97013 1 0.0000 0.9120 1.000 0.000
#> GSM96956 1 0.8861 0.6931 0.696 0.304
#> GSM97024 2 0.2043 0.9286 0.032 0.968
#> GSM97032 1 0.6887 0.8463 0.816 0.184
#> GSM97044 1 0.6973 0.8426 0.812 0.188
#> GSM97049 2 0.0000 0.9541 0.000 1.000
#> GSM96968 1 0.6343 0.8687 0.840 0.160
#> GSM96971 1 0.4431 0.8942 0.908 0.092
#> GSM96986 1 0.5059 0.8856 0.888 0.112
#> GSM97003 1 0.0000 0.9120 1.000 0.000
#> GSM96957 1 0.3274 0.9073 0.940 0.060
#> GSM96960 1 0.0000 0.9120 1.000 0.000
#> GSM96975 1 0.0376 0.9124 0.996 0.004
#> GSM96998 1 0.0000 0.9120 1.000 0.000
#> GSM96999 1 0.3274 0.9073 0.940 0.060
#> GSM97001 1 0.3274 0.9073 0.940 0.060
#> GSM97005 1 0.1843 0.9124 0.972 0.028
#> GSM97006 1 0.0000 0.9120 1.000 0.000
#> GSM97021 1 0.5294 0.8769 0.880 0.120
#> GSM97028 1 0.5519 0.8830 0.872 0.128
#> GSM97031 1 0.2423 0.9096 0.960 0.040
#> GSM97037 1 0.8327 0.7570 0.736 0.264
#> GSM97018 1 0.6438 0.8680 0.836 0.164
#> GSM97014 1 0.7528 0.7896 0.784 0.216
#> GSM97042 2 0.0000 0.9541 0.000 1.000
#> GSM97040 1 0.5737 0.8689 0.864 0.136
#> GSM97041 1 0.5059 0.8812 0.888 0.112
#> GSM96955 2 0.5842 0.8025 0.140 0.860
#> GSM96990 1 0.6973 0.8426 0.812 0.188
#> GSM96991 2 0.0000 0.9541 0.000 1.000
#> GSM97048 2 0.0000 0.9541 0.000 1.000
#> GSM96963 2 0.0000 0.9541 0.000 1.000
#> GSM96953 2 0.0000 0.9541 0.000 1.000
#> GSM96966 1 0.0000 0.9120 1.000 0.000
#> GSM96979 1 0.5059 0.8856 0.888 0.112
#> GSM96983 1 0.5294 0.8815 0.880 0.120
#> GSM96984 1 0.5178 0.8833 0.884 0.116
#> GSM96994 1 0.5059 0.8856 0.888 0.112
#> GSM96996 1 0.0376 0.9118 0.996 0.004
#> GSM96997 1 0.5059 0.8856 0.888 0.112
#> GSM97007 1 0.5178 0.8833 0.884 0.116
#> GSM96954 1 0.4431 0.8942 0.908 0.092
#> GSM96962 1 0.5059 0.8856 0.888 0.112
#> GSM96969 1 0.0000 0.9120 1.000 0.000
#> GSM96970 1 0.0000 0.9120 1.000 0.000
#> GSM96973 1 0.0000 0.9120 1.000 0.000
#> GSM96976 1 0.4431 0.8954 0.908 0.092
#> GSM96977 1 0.1843 0.9118 0.972 0.028
#> GSM96995 1 0.6343 0.8687 0.840 0.160
#> GSM97002 1 0.0000 0.9120 1.000 0.000
#> GSM97009 1 0.7745 0.7743 0.772 0.228
#> GSM97010 1 0.3274 0.9080 0.940 0.060
#> GSM96974 1 0.2043 0.9112 0.968 0.032
#> GSM96985 1 0.5294 0.8815 0.880 0.120
#> GSM96959 2 0.9896 0.0969 0.440 0.560
#> GSM96972 1 0.0000 0.9120 1.000 0.000
#> GSM96978 1 0.5294 0.8815 0.880 0.120
#> GSM96967 1 0.0000 0.9120 1.000 0.000
#> GSM96987 1 0.0000 0.9120 1.000 0.000
#> GSM97011 1 0.7299 0.8011 0.796 0.204
#> GSM96964 1 0.1414 0.9118 0.980 0.020
#> GSM96965 1 0.4431 0.8954 0.908 0.092
#> GSM96981 1 0.0000 0.9120 1.000 0.000
#> GSM96982 1 0.0000 0.9120 1.000 0.000
#> GSM96988 1 0.5519 0.8830 0.872 0.128
#> GSM97000 1 0.5629 0.8700 0.868 0.132
#> GSM97004 1 0.0000 0.9120 1.000 0.000
#> GSM97008 1 0.4562 0.8910 0.904 0.096
#> GSM96950 1 0.1843 0.9119 0.972 0.028
#> GSM96980 1 0.0000 0.9120 1.000 0.000
#> GSM96989 1 0.0000 0.9120 1.000 0.000
#> GSM96992 1 0.0000 0.9120 1.000 0.000
#> GSM96993 1 0.2603 0.9078 0.956 0.044
#> GSM96958 1 0.1633 0.9117 0.976 0.024
#> GSM96951 1 0.0000 0.9120 1.000 0.000
#> GSM96952 1 0.0000 0.9120 1.000 0.000
#> GSM96961 1 0.0000 0.9120 1.000 0.000
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM97038 2 0.5330 0.753 0.044 0.812 0.144
#> GSM97045 2 0.0592 0.945 0.012 0.988 0.000
#> GSM97047 1 0.8521 0.629 0.608 0.228 0.164
#> GSM97025 2 0.0592 0.945 0.012 0.988 0.000
#> GSM97030 3 0.2356 0.877 0.000 0.072 0.928
#> GSM97027 2 0.0592 0.945 0.012 0.988 0.000
#> GSM97033 2 0.0237 0.950 0.004 0.996 0.000
#> GSM97034 1 0.7493 0.744 0.696 0.168 0.136
#> GSM97020 2 0.0424 0.947 0.008 0.992 0.000
#> GSM97026 1 0.6775 0.794 0.740 0.164 0.096
#> GSM97012 2 0.0000 0.951 0.000 1.000 0.000
#> GSM97015 3 0.3234 0.878 0.020 0.072 0.908
#> GSM97016 2 0.0000 0.951 0.000 1.000 0.000
#> GSM97017 1 0.4818 0.864 0.844 0.108 0.048
#> GSM97019 2 0.0000 0.951 0.000 1.000 0.000
#> GSM97022 2 0.0000 0.951 0.000 1.000 0.000
#> GSM97035 2 0.0000 0.951 0.000 1.000 0.000
#> GSM97036 1 0.3155 0.895 0.916 0.044 0.040
#> GSM97039 2 0.0000 0.951 0.000 1.000 0.000
#> GSM97046 2 0.0000 0.951 0.000 1.000 0.000
#> GSM97023 1 0.0237 0.896 0.996 0.000 0.004
#> GSM97029 1 0.6529 0.802 0.756 0.152 0.092
#> GSM97043 3 0.8388 0.582 0.140 0.248 0.612
#> GSM97013 1 0.0000 0.895 1.000 0.000 0.000
#> GSM96956 3 0.4654 0.765 0.000 0.208 0.792
#> GSM97024 2 0.1289 0.927 0.000 0.968 0.032
#> GSM97032 3 0.2680 0.880 0.008 0.068 0.924
#> GSM97044 3 0.2356 0.877 0.000 0.072 0.928
#> GSM97049 2 0.0000 0.951 0.000 1.000 0.000
#> GSM96968 3 0.6171 0.795 0.144 0.080 0.776
#> GSM96971 3 0.4291 0.764 0.180 0.000 0.820
#> GSM96986 3 0.0237 0.889 0.004 0.000 0.996
#> GSM97003 1 0.0237 0.895 0.996 0.000 0.004
#> GSM96957 1 0.3572 0.891 0.900 0.060 0.040
#> GSM96960 1 0.0237 0.895 0.996 0.000 0.004
#> GSM96975 1 0.0829 0.899 0.984 0.004 0.012
#> GSM96998 1 0.0000 0.895 1.000 0.000 0.000
#> GSM96999 1 0.3572 0.891 0.900 0.060 0.040
#> GSM97001 1 0.3572 0.891 0.900 0.060 0.040
#> GSM97005 1 0.2187 0.900 0.948 0.028 0.024
#> GSM97006 1 0.0237 0.895 0.996 0.000 0.004
#> GSM97021 1 0.5263 0.854 0.824 0.116 0.060
#> GSM97028 3 0.4137 0.853 0.096 0.032 0.872
#> GSM97031 1 0.5529 0.580 0.704 0.000 0.296
#> GSM97037 3 0.4002 0.818 0.000 0.160 0.840
#> GSM97018 3 0.6719 0.759 0.160 0.096 0.744
#> GSM97014 1 0.6977 0.758 0.712 0.212 0.076
#> GSM97042 2 0.0000 0.951 0.000 1.000 0.000
#> GSM97040 1 0.6144 0.824 0.780 0.132 0.088
#> GSM97041 1 0.4712 0.866 0.848 0.108 0.044
#> GSM96955 2 0.4565 0.812 0.064 0.860 0.076
#> GSM96990 3 0.3091 0.879 0.016 0.072 0.912
#> GSM96991 2 0.0000 0.951 0.000 1.000 0.000
#> GSM97048 2 0.0000 0.951 0.000 1.000 0.000
#> GSM96963 2 0.0000 0.951 0.000 1.000 0.000
#> GSM96953 2 0.0000 0.951 0.000 1.000 0.000
#> GSM96966 1 0.3267 0.866 0.884 0.000 0.116
#> GSM96979 3 0.0237 0.889 0.004 0.000 0.996
#> GSM96983 3 0.0237 0.888 0.000 0.004 0.996
#> GSM96984 3 0.0000 0.887 0.000 0.000 1.000
#> GSM96994 3 0.0237 0.889 0.004 0.000 0.996
#> GSM96996 1 0.1129 0.899 0.976 0.004 0.020
#> GSM96997 3 0.0237 0.889 0.004 0.000 0.996
#> GSM97007 3 0.0000 0.887 0.000 0.000 1.000
#> GSM96954 3 0.4291 0.764 0.180 0.000 0.820
#> GSM96962 3 0.0237 0.889 0.004 0.000 0.996
#> GSM96969 1 0.3038 0.872 0.896 0.000 0.104
#> GSM96970 1 0.3192 0.868 0.888 0.000 0.112
#> GSM96973 1 0.3267 0.866 0.884 0.000 0.116
#> GSM96976 1 0.6644 0.791 0.748 0.092 0.160
#> GSM96977 1 0.2318 0.898 0.944 0.028 0.028
#> GSM96995 3 0.6171 0.795 0.144 0.080 0.776
#> GSM97002 1 0.0237 0.895 0.996 0.000 0.004
#> GSM97009 1 0.7821 0.701 0.660 0.224 0.116
#> GSM97010 1 0.3683 0.890 0.896 0.060 0.044
#> GSM96974 1 0.5627 0.807 0.780 0.032 0.188
#> GSM96985 3 0.0237 0.888 0.000 0.004 0.996
#> GSM96959 2 0.8646 0.269 0.320 0.556 0.124
#> GSM96972 1 0.1411 0.891 0.964 0.000 0.036
#> GSM96978 3 0.0237 0.888 0.000 0.004 0.996
#> GSM96967 1 0.3267 0.866 0.884 0.000 0.116
#> GSM96987 1 0.0000 0.895 1.000 0.000 0.000
#> GSM97011 1 0.7569 0.731 0.684 0.200 0.116
#> GSM96964 1 0.2050 0.898 0.952 0.020 0.028
#> GSM96965 1 0.6644 0.791 0.748 0.092 0.160
#> GSM96981 1 0.0592 0.898 0.988 0.000 0.012
#> GSM96982 1 0.0592 0.898 0.988 0.000 0.012
#> GSM96988 3 0.4137 0.853 0.096 0.032 0.872
#> GSM97000 1 0.5981 0.829 0.788 0.132 0.080
#> GSM97004 1 0.0237 0.895 0.996 0.000 0.004
#> GSM97008 1 0.5253 0.858 0.828 0.096 0.076
#> GSM96950 1 0.2318 0.899 0.944 0.028 0.028
#> GSM96980 1 0.0424 0.896 0.992 0.000 0.008
#> GSM96989 1 0.0000 0.895 1.000 0.000 0.000
#> GSM96992 1 0.0237 0.895 0.996 0.000 0.004
#> GSM96993 1 0.3155 0.895 0.916 0.044 0.040
#> GSM96958 1 0.2187 0.898 0.948 0.024 0.028
#> GSM96951 1 0.0237 0.896 0.996 0.000 0.004
#> GSM96952 1 0.0237 0.895 0.996 0.000 0.004
#> GSM96961 1 0.0237 0.895 0.996 0.000 0.004
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM97038 2 0.5454 0.728 0.044 0.780 0.104 0.072
#> GSM97045 2 0.0524 0.940 0.008 0.988 0.000 0.004
#> GSM97047 1 0.7942 0.606 0.596 0.192 0.108 0.104
#> GSM97025 2 0.0524 0.940 0.008 0.988 0.000 0.004
#> GSM97030 3 0.2466 0.862 0.000 0.056 0.916 0.028
#> GSM97027 2 0.0524 0.940 0.008 0.988 0.000 0.004
#> GSM97033 2 0.0188 0.944 0.000 0.996 0.000 0.004
#> GSM97034 1 0.6895 0.710 0.688 0.132 0.076 0.104
#> GSM97020 2 0.0376 0.942 0.004 0.992 0.000 0.004
#> GSM97026 1 0.6224 0.751 0.728 0.124 0.044 0.104
#> GSM97012 2 0.0000 0.946 0.000 1.000 0.000 0.000
#> GSM97015 3 0.3161 0.860 0.020 0.056 0.896 0.028
#> GSM97016 2 0.0000 0.946 0.000 1.000 0.000 0.000
#> GSM97017 1 0.4292 0.824 0.832 0.072 0.008 0.088
#> GSM97019 2 0.0000 0.946 0.000 1.000 0.000 0.000
#> GSM97022 2 0.0000 0.946 0.000 1.000 0.000 0.000
#> GSM97035 2 0.0000 0.946 0.000 1.000 0.000 0.000
#> GSM97036 1 0.2599 0.857 0.912 0.020 0.004 0.064
#> GSM97039 2 0.0000 0.946 0.000 1.000 0.000 0.000
#> GSM97046 2 0.0000 0.946 0.000 1.000 0.000 0.000
#> GSM97023 1 0.1118 0.863 0.964 0.000 0.000 0.036
#> GSM97029 1 0.5888 0.764 0.748 0.120 0.036 0.096
#> GSM97043 3 0.7570 0.552 0.136 0.232 0.592 0.040
#> GSM97013 1 0.1022 0.863 0.968 0.000 0.000 0.032
#> GSM96956 3 0.4348 0.758 0.000 0.196 0.780 0.024
#> GSM97024 2 0.1022 0.919 0.000 0.968 0.032 0.000
#> GSM97032 3 0.2846 0.864 0.012 0.052 0.908 0.028
#> GSM97044 3 0.2466 0.862 0.000 0.056 0.916 0.028
#> GSM97049 2 0.0000 0.946 0.000 1.000 0.000 0.000
#> GSM96968 3 0.5652 0.747 0.144 0.068 0.756 0.032
#> GSM96971 3 0.3791 0.711 0.004 0.000 0.796 0.200
#> GSM96986 3 0.0376 0.871 0.004 0.000 0.992 0.004
#> GSM97003 1 0.2216 0.842 0.908 0.000 0.000 0.092
#> GSM96957 1 0.2774 0.858 0.908 0.044 0.004 0.044
#> GSM96960 1 0.2281 0.841 0.904 0.000 0.000 0.096
#> GSM96975 1 0.1902 0.860 0.932 0.004 0.000 0.064
#> GSM96998 1 0.1474 0.858 0.948 0.000 0.000 0.052
#> GSM96999 1 0.2774 0.858 0.908 0.044 0.004 0.044
#> GSM97001 1 0.2774 0.858 0.908 0.044 0.004 0.044
#> GSM97005 1 0.2441 0.868 0.920 0.020 0.004 0.056
#> GSM97006 1 0.2281 0.841 0.904 0.000 0.000 0.096
#> GSM97021 1 0.4676 0.813 0.812 0.076 0.012 0.100
#> GSM97028 3 0.3639 0.819 0.096 0.028 0.864 0.012
#> GSM97031 1 0.5815 0.507 0.652 0.000 0.288 0.060
#> GSM97037 3 0.3863 0.808 0.000 0.144 0.828 0.028
#> GSM97018 3 0.6332 0.691 0.164 0.080 0.712 0.044
#> GSM97014 1 0.6368 0.715 0.700 0.172 0.028 0.100
#> GSM97042 2 0.0000 0.946 0.000 1.000 0.000 0.000
#> GSM97040 1 0.5637 0.781 0.768 0.096 0.040 0.096
#> GSM97041 1 0.4226 0.825 0.836 0.072 0.008 0.084
#> GSM96955 2 0.4621 0.778 0.060 0.828 0.036 0.076
#> GSM96990 3 0.3146 0.861 0.016 0.056 0.896 0.032
#> GSM96991 2 0.0000 0.946 0.000 1.000 0.000 0.000
#> GSM97048 2 0.0000 0.946 0.000 1.000 0.000 0.000
#> GSM96963 2 0.0000 0.946 0.000 1.000 0.000 0.000
#> GSM96953 2 0.0000 0.946 0.000 1.000 0.000 0.000
#> GSM96966 4 0.3143 0.913 0.100 0.000 0.024 0.876
#> GSM96979 3 0.0376 0.871 0.004 0.000 0.992 0.004
#> GSM96983 3 0.0000 0.870 0.000 0.000 1.000 0.000
#> GSM96984 3 0.0188 0.869 0.000 0.000 0.996 0.004
#> GSM96994 3 0.0376 0.871 0.004 0.000 0.992 0.004
#> GSM96996 1 0.2081 0.860 0.916 0.000 0.000 0.084
#> GSM96997 3 0.0376 0.871 0.004 0.000 0.992 0.004
#> GSM97007 3 0.0188 0.869 0.000 0.000 0.996 0.004
#> GSM96954 3 0.3791 0.711 0.004 0.000 0.796 0.200
#> GSM96962 3 0.0376 0.871 0.004 0.000 0.992 0.004
#> GSM96969 4 0.2654 0.906 0.108 0.000 0.004 0.888
#> GSM96970 4 0.2266 0.917 0.084 0.000 0.004 0.912
#> GSM96973 4 0.2266 0.919 0.084 0.000 0.004 0.912
#> GSM96976 4 0.3432 0.848 0.036 0.060 0.020 0.884
#> GSM96977 1 0.1796 0.862 0.948 0.016 0.004 0.032
#> GSM96995 3 0.5652 0.747 0.144 0.068 0.756 0.032
#> GSM97002 1 0.2216 0.842 0.908 0.000 0.000 0.092
#> GSM97009 1 0.7120 0.681 0.656 0.184 0.060 0.100
#> GSM97010 1 0.3249 0.860 0.888 0.044 0.008 0.060
#> GSM96974 4 0.2908 0.851 0.040 0.000 0.064 0.896
#> GSM96985 3 0.0000 0.870 0.000 0.000 1.000 0.000
#> GSM96959 2 0.8023 0.233 0.312 0.524 0.084 0.080
#> GSM96972 4 0.3172 0.851 0.160 0.000 0.000 0.840
#> GSM96978 3 0.0000 0.870 0.000 0.000 1.000 0.000
#> GSM96967 4 0.2266 0.919 0.084 0.000 0.004 0.912
#> GSM96987 1 0.1022 0.861 0.968 0.000 0.000 0.032
#> GSM97011 1 0.6993 0.698 0.672 0.160 0.060 0.108
#> GSM96964 1 0.1543 0.861 0.956 0.008 0.004 0.032
#> GSM96965 4 0.3353 0.849 0.036 0.056 0.020 0.888
#> GSM96981 1 0.1716 0.858 0.936 0.000 0.000 0.064
#> GSM96982 1 0.1716 0.858 0.936 0.000 0.000 0.064
#> GSM96988 3 0.3639 0.819 0.096 0.028 0.864 0.012
#> GSM97000 1 0.5427 0.788 0.780 0.096 0.036 0.088
#> GSM97004 1 0.2281 0.841 0.904 0.000 0.000 0.096
#> GSM97008 1 0.4772 0.816 0.816 0.064 0.028 0.092
#> GSM96950 1 0.1771 0.863 0.948 0.012 0.004 0.036
#> GSM96980 1 0.3444 0.774 0.816 0.000 0.000 0.184
#> GSM96989 1 0.1022 0.861 0.968 0.000 0.000 0.032
#> GSM96992 1 0.2149 0.844 0.912 0.000 0.000 0.088
#> GSM96993 1 0.2521 0.858 0.916 0.020 0.004 0.060
#> GSM96958 1 0.1674 0.862 0.952 0.012 0.004 0.032
#> GSM96951 1 0.1389 0.860 0.952 0.000 0.000 0.048
#> GSM96952 1 0.2149 0.844 0.912 0.000 0.000 0.088
#> GSM96961 1 0.2149 0.844 0.912 0.000 0.000 0.088
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM97038 2 0.5343 0.7029 0.020 0.732 0.096 0.012 0.140
#> GSM97045 2 0.0880 0.9478 0.000 0.968 0.000 0.000 0.032
#> GSM97047 5 0.5505 0.5618 0.036 0.092 0.092 0.032 0.748
#> GSM97025 2 0.0880 0.9478 0.000 0.968 0.000 0.000 0.032
#> GSM97030 3 0.2659 0.7924 0.004 0.040 0.904 0.036 0.016
#> GSM97027 2 0.0880 0.9478 0.000 0.968 0.000 0.000 0.032
#> GSM97033 2 0.0771 0.9548 0.004 0.976 0.000 0.000 0.020
#> GSM97034 5 0.4979 0.6223 0.024 0.088 0.056 0.048 0.784
#> GSM97020 2 0.0794 0.9503 0.000 0.972 0.000 0.000 0.028
#> GSM97026 5 0.2902 0.6810 0.008 0.052 0.036 0.012 0.892
#> GSM97012 2 0.0000 0.9623 0.000 1.000 0.000 0.000 0.000
#> GSM97015 3 0.3102 0.7927 0.004 0.040 0.884 0.036 0.036
#> GSM97016 2 0.0290 0.9602 0.008 0.992 0.000 0.000 0.000
#> GSM97017 5 0.0833 0.7094 0.016 0.004 0.004 0.000 0.976
#> GSM97019 2 0.0000 0.9623 0.000 1.000 0.000 0.000 0.000
#> GSM97022 2 0.0000 0.9623 0.000 1.000 0.000 0.000 0.000
#> GSM97035 2 0.0000 0.9623 0.000 1.000 0.000 0.000 0.000
#> GSM97036 5 0.2770 0.6817 0.124 0.004 0.000 0.008 0.864
#> GSM97039 2 0.0162 0.9612 0.004 0.996 0.000 0.000 0.000
#> GSM97046 2 0.0162 0.9612 0.004 0.996 0.000 0.000 0.000
#> GSM97023 5 0.3480 0.5462 0.248 0.000 0.000 0.000 0.752
#> GSM97029 5 0.4361 0.6684 0.044 0.080 0.024 0.032 0.820
#> GSM97043 3 0.7013 0.5685 0.016 0.196 0.576 0.036 0.176
#> GSM97013 5 0.3336 0.5758 0.228 0.000 0.000 0.000 0.772
#> GSM96956 3 0.4266 0.7065 0.004 0.184 0.772 0.028 0.012
#> GSM97024 2 0.0880 0.9390 0.000 0.968 0.032 0.000 0.000
#> GSM97032 3 0.2942 0.7939 0.004 0.036 0.892 0.036 0.032
#> GSM97044 3 0.2659 0.7924 0.004 0.040 0.904 0.036 0.016
#> GSM97049 2 0.0162 0.9612 0.004 0.996 0.000 0.000 0.000
#> GSM96968 3 0.5110 0.7254 0.016 0.048 0.748 0.028 0.160
#> GSM96971 3 0.6272 0.5524 0.200 0.000 0.560 0.236 0.004
#> GSM96986 3 0.3897 0.7605 0.204 0.000 0.768 0.028 0.000
#> GSM97003 1 0.3661 0.8709 0.724 0.000 0.000 0.000 0.276
#> GSM96957 5 0.3488 0.6471 0.168 0.024 0.000 0.000 0.808
#> GSM96960 1 0.3741 0.8671 0.732 0.000 0.000 0.004 0.264
#> GSM96975 1 0.4047 0.8504 0.676 0.000 0.000 0.004 0.320
#> GSM96998 1 0.4403 0.6372 0.560 0.000 0.000 0.004 0.436
#> GSM96999 5 0.3488 0.6471 0.168 0.024 0.000 0.000 0.808
#> GSM97001 5 0.3488 0.6471 0.168 0.024 0.000 0.000 0.808
#> GSM97005 5 0.3003 0.6343 0.188 0.000 0.000 0.000 0.812
#> GSM97006 1 0.3741 0.8671 0.732 0.000 0.000 0.004 0.264
#> GSM97021 5 0.1016 0.7071 0.008 0.004 0.004 0.012 0.972
#> GSM97028 3 0.3260 0.7783 0.004 0.024 0.860 0.012 0.100
#> GSM97031 1 0.5941 0.1387 0.544 0.000 0.124 0.000 0.332
#> GSM97037 3 0.3958 0.7486 0.004 0.128 0.816 0.036 0.016
#> GSM97018 3 0.5669 0.6800 0.008 0.072 0.692 0.032 0.196
#> GSM97014 5 0.3439 0.6601 0.040 0.080 0.024 0.000 0.856
#> GSM97042 2 0.0000 0.9623 0.000 1.000 0.000 0.000 0.000
#> GSM97040 5 0.2637 0.6904 0.048 0.008 0.032 0.008 0.904
#> GSM97041 5 0.0932 0.7099 0.020 0.004 0.004 0.000 0.972
#> GSM96955 2 0.4933 0.7219 0.036 0.756 0.028 0.016 0.164
#> GSM96990 3 0.3145 0.7912 0.008 0.040 0.884 0.036 0.032
#> GSM96991 2 0.0000 0.9623 0.000 1.000 0.000 0.000 0.000
#> GSM97048 2 0.0162 0.9612 0.004 0.996 0.000 0.000 0.000
#> GSM96963 2 0.0000 0.9623 0.000 1.000 0.000 0.000 0.000
#> GSM96953 2 0.0000 0.9623 0.000 1.000 0.000 0.000 0.000
#> GSM96966 4 0.2865 0.9093 0.132 0.000 0.004 0.856 0.008
#> GSM96979 3 0.3897 0.7605 0.204 0.000 0.768 0.028 0.000
#> GSM96983 3 0.0794 0.7955 0.028 0.000 0.972 0.000 0.000
#> GSM96984 3 0.3897 0.7596 0.204 0.000 0.768 0.028 0.000
#> GSM96994 3 0.3897 0.7605 0.204 0.000 0.768 0.028 0.000
#> GSM96996 1 0.3999 0.8281 0.656 0.000 0.000 0.000 0.344
#> GSM96997 3 0.3993 0.7556 0.216 0.000 0.756 0.028 0.000
#> GSM97007 3 0.3897 0.7596 0.204 0.000 0.768 0.028 0.000
#> GSM96954 3 0.6272 0.5524 0.200 0.000 0.560 0.236 0.004
#> GSM96962 3 0.3897 0.7605 0.204 0.000 0.768 0.028 0.000
#> GSM96969 4 0.3224 0.8954 0.160 0.000 0.000 0.824 0.016
#> GSM96970 4 0.2648 0.9112 0.152 0.000 0.000 0.848 0.000
#> GSM96973 4 0.2561 0.9132 0.144 0.000 0.000 0.856 0.000
#> GSM96976 4 0.1809 0.8463 0.000 0.060 0.000 0.928 0.012
#> GSM96977 5 0.3210 0.6056 0.212 0.000 0.000 0.000 0.788
#> GSM96995 3 0.5110 0.7254 0.016 0.048 0.748 0.028 0.160
#> GSM97002 1 0.3661 0.8709 0.724 0.000 0.000 0.000 0.276
#> GSM97009 5 0.4592 0.6242 0.040 0.096 0.052 0.012 0.800
#> GSM97010 5 0.5143 -0.2500 0.420 0.032 0.004 0.000 0.544
#> GSM96974 4 0.1799 0.8575 0.020 0.000 0.028 0.940 0.012
#> GSM96985 3 0.0794 0.7955 0.028 0.000 0.972 0.000 0.000
#> GSM96959 5 0.6783 -0.0393 0.036 0.432 0.076 0.012 0.444
#> GSM96972 4 0.4024 0.8102 0.220 0.000 0.000 0.752 0.028
#> GSM96978 3 0.0794 0.7955 0.028 0.000 0.972 0.000 0.000
#> GSM96967 4 0.2561 0.9132 0.144 0.000 0.000 0.856 0.000
#> GSM96987 5 0.3636 0.4935 0.272 0.000 0.000 0.000 0.728
#> GSM97011 5 0.4036 0.6380 0.032 0.068 0.052 0.012 0.836
#> GSM96964 5 0.3242 0.6017 0.216 0.000 0.000 0.000 0.784
#> GSM96965 4 0.1845 0.8477 0.000 0.056 0.000 0.928 0.016
#> GSM96981 1 0.4009 0.8562 0.684 0.000 0.000 0.004 0.312
#> GSM96982 1 0.4009 0.8562 0.684 0.000 0.000 0.004 0.312
#> GSM96988 3 0.3260 0.7783 0.004 0.024 0.860 0.012 0.100
#> GSM97000 5 0.2694 0.6941 0.056 0.008 0.028 0.008 0.900
#> GSM97004 1 0.3741 0.8671 0.732 0.000 0.000 0.004 0.264
#> GSM97008 5 0.2514 0.7088 0.056 0.008 0.020 0.008 0.908
#> GSM96950 5 0.3398 0.6082 0.216 0.004 0.000 0.000 0.780
#> GSM96980 1 0.5141 0.7800 0.672 0.000 0.000 0.092 0.236
#> GSM96989 5 0.3636 0.4935 0.272 0.000 0.000 0.000 0.728
#> GSM96992 1 0.3774 0.8690 0.704 0.000 0.000 0.000 0.296
#> GSM96993 5 0.2818 0.6798 0.128 0.004 0.000 0.008 0.860
#> GSM96958 5 0.3242 0.6004 0.216 0.000 0.000 0.000 0.784
#> GSM96951 1 0.4227 0.6940 0.580 0.000 0.000 0.000 0.420
#> GSM96952 1 0.3774 0.8690 0.704 0.000 0.000 0.000 0.296
#> GSM96961 1 0.3774 0.8690 0.704 0.000 0.000 0.000 0.296
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM97038 2 0.5056 0.6576 0.000 0.708 0.136 0.012 0.124 0.020
#> GSM97045 2 0.1010 0.9448 0.000 0.960 0.004 0.000 0.036 0.000
#> GSM97047 5 0.4455 0.6671 0.000 0.064 0.148 0.008 0.756 0.024
#> GSM97025 2 0.1010 0.9448 0.000 0.960 0.004 0.000 0.036 0.000
#> GSM97030 3 0.1370 0.8470 0.000 0.012 0.948 0.000 0.004 0.036
#> GSM97027 2 0.1010 0.9448 0.000 0.960 0.004 0.000 0.036 0.000
#> GSM97033 2 0.0547 0.9518 0.000 0.980 0.000 0.000 0.020 0.000
#> GSM97034 5 0.4890 0.7174 0.032 0.060 0.120 0.020 0.756 0.012
#> GSM97020 2 0.0790 0.9475 0.000 0.968 0.000 0.000 0.032 0.000
#> GSM97026 5 0.3796 0.7696 0.028 0.044 0.064 0.020 0.836 0.008
#> GSM97012 2 0.0291 0.9581 0.000 0.992 0.004 0.000 0.004 0.000
#> GSM97015 3 0.1700 0.8523 0.000 0.012 0.936 0.000 0.024 0.028
#> GSM97016 2 0.0405 0.9540 0.000 0.988 0.004 0.000 0.000 0.008
#> GSM97017 5 0.1667 0.7888 0.032 0.000 0.012 0.008 0.940 0.008
#> GSM97019 2 0.0291 0.9581 0.000 0.992 0.004 0.000 0.004 0.000
#> GSM97022 2 0.0291 0.9581 0.000 0.992 0.004 0.000 0.004 0.000
#> GSM97035 2 0.0291 0.9581 0.000 0.992 0.004 0.000 0.004 0.000
#> GSM97036 5 0.2723 0.7808 0.128 0.000 0.000 0.016 0.852 0.004
#> GSM97039 2 0.0260 0.9546 0.000 0.992 0.000 0.000 0.000 0.008
#> GSM97046 2 0.0260 0.9546 0.000 0.992 0.000 0.000 0.000 0.008
#> GSM97023 5 0.3314 0.6959 0.256 0.000 0.000 0.004 0.740 0.000
#> GSM97029 5 0.4253 0.7570 0.040 0.056 0.068 0.020 0.808 0.008
#> GSM97043 3 0.4997 0.6369 0.000 0.160 0.660 0.000 0.176 0.004
#> GSM97013 5 0.3314 0.7224 0.224 0.000 0.000 0.012 0.764 0.000
#> GSM96956 3 0.2859 0.7604 0.000 0.156 0.828 0.000 0.000 0.016
#> GSM97024 2 0.1080 0.9396 0.000 0.960 0.032 0.000 0.004 0.004
#> GSM97032 3 0.1838 0.8512 0.000 0.012 0.928 0.000 0.020 0.040
#> GSM97044 3 0.1370 0.8470 0.000 0.012 0.948 0.000 0.004 0.036
#> GSM97049 2 0.0260 0.9546 0.000 0.992 0.000 0.000 0.000 0.008
#> GSM96968 3 0.3300 0.7929 0.000 0.016 0.812 0.000 0.156 0.016
#> GSM96971 6 0.3964 0.7389 0.000 0.000 0.044 0.232 0.000 0.724
#> GSM96986 6 0.1267 0.9299 0.000 0.000 0.060 0.000 0.000 0.940
#> GSM97003 1 0.0458 0.8231 0.984 0.000 0.000 0.000 0.016 0.000
#> GSM96957 5 0.3784 0.7582 0.180 0.020 0.012 0.004 0.780 0.004
#> GSM96960 1 0.0146 0.8146 0.996 0.000 0.000 0.000 0.000 0.004
#> GSM96975 1 0.2196 0.8286 0.884 0.000 0.004 0.004 0.108 0.000
#> GSM96998 1 0.3314 0.6851 0.764 0.000 0.000 0.012 0.224 0.000
#> GSM96999 5 0.3784 0.7582 0.180 0.020 0.012 0.004 0.780 0.004
#> GSM97001 5 0.3784 0.7582 0.180 0.020 0.012 0.004 0.780 0.004
#> GSM97005 5 0.2994 0.7474 0.208 0.000 0.000 0.004 0.788 0.000
#> GSM97006 1 0.0146 0.8146 0.996 0.000 0.000 0.000 0.000 0.004
#> GSM97021 5 0.2215 0.7881 0.032 0.000 0.024 0.020 0.916 0.008
#> GSM97028 3 0.3874 0.8130 0.000 0.012 0.804 0.008 0.096 0.080
#> GSM97031 1 0.6416 0.1555 0.404 0.000 0.016 0.000 0.304 0.276
#> GSM97037 3 0.2526 0.8125 0.000 0.096 0.876 0.000 0.004 0.024
#> GSM97018 3 0.4706 0.7352 0.000 0.044 0.728 0.012 0.184 0.032
#> GSM97014 5 0.3136 0.7484 0.000 0.068 0.032 0.012 0.864 0.024
#> GSM97042 2 0.0291 0.9581 0.000 0.992 0.004 0.000 0.004 0.000
#> GSM97040 5 0.2179 0.7751 0.012 0.000 0.040 0.008 0.916 0.024
#> GSM97041 5 0.1628 0.7899 0.036 0.000 0.012 0.004 0.940 0.008
#> GSM96955 2 0.4870 0.7059 0.000 0.736 0.064 0.020 0.148 0.032
#> GSM96990 3 0.1616 0.8516 0.000 0.012 0.940 0.000 0.020 0.028
#> GSM96991 2 0.0291 0.9581 0.000 0.992 0.004 0.000 0.004 0.000
#> GSM97048 2 0.0260 0.9546 0.000 0.992 0.000 0.000 0.000 0.008
#> GSM96963 2 0.0291 0.9581 0.000 0.992 0.004 0.000 0.004 0.000
#> GSM96953 2 0.0291 0.9581 0.000 0.992 0.004 0.000 0.004 0.000
#> GSM96966 4 0.2191 0.9033 0.120 0.000 0.000 0.876 0.000 0.004
#> GSM96979 6 0.1267 0.9299 0.000 0.000 0.060 0.000 0.000 0.940
#> GSM96983 3 0.2100 0.7995 0.000 0.000 0.884 0.004 0.000 0.112
#> GSM96984 6 0.1204 0.9252 0.000 0.000 0.056 0.000 0.000 0.944
#> GSM96994 6 0.1267 0.9299 0.000 0.000 0.060 0.000 0.000 0.940
#> GSM96996 1 0.1765 0.8226 0.904 0.000 0.000 0.000 0.096 0.000
#> GSM96997 6 0.1007 0.9257 0.000 0.000 0.044 0.000 0.000 0.956
#> GSM97007 6 0.1204 0.9252 0.000 0.000 0.056 0.000 0.000 0.944
#> GSM96954 6 0.3964 0.7389 0.000 0.000 0.044 0.232 0.000 0.724
#> GSM96962 6 0.1267 0.9299 0.000 0.000 0.060 0.000 0.000 0.940
#> GSM96969 4 0.2558 0.8957 0.156 0.000 0.000 0.840 0.000 0.004
#> GSM96970 4 0.2320 0.9069 0.132 0.000 0.000 0.864 0.000 0.004
#> GSM96973 4 0.2135 0.9083 0.128 0.000 0.000 0.872 0.000 0.000
#> GSM96976 4 0.1625 0.8311 0.000 0.060 0.012 0.928 0.000 0.000
#> GSM96977 5 0.3081 0.7303 0.220 0.000 0.000 0.004 0.776 0.000
#> GSM96995 3 0.3300 0.7929 0.000 0.016 0.812 0.000 0.156 0.016
#> GSM97002 1 0.0458 0.8231 0.984 0.000 0.000 0.000 0.016 0.000
#> GSM97009 5 0.4267 0.7168 0.008 0.076 0.076 0.016 0.800 0.024
#> GSM97010 1 0.4889 0.4548 0.620 0.028 0.012 0.008 0.328 0.004
#> GSM96974 4 0.1003 0.8384 0.004 0.000 0.028 0.964 0.000 0.004
#> GSM96985 3 0.2100 0.7995 0.000 0.000 0.884 0.004 0.000 0.112
#> GSM96959 5 0.6446 0.0362 0.000 0.404 0.116 0.016 0.432 0.032
#> GSM96972 4 0.3240 0.8046 0.244 0.000 0.000 0.752 0.000 0.004
#> GSM96978 3 0.2100 0.7995 0.000 0.000 0.884 0.004 0.000 0.112
#> GSM96967 4 0.2135 0.9083 0.128 0.000 0.000 0.872 0.000 0.000
#> GSM96987 5 0.3690 0.6501 0.288 0.000 0.000 0.012 0.700 0.000
#> GSM97011 5 0.3620 0.7288 0.000 0.052 0.072 0.016 0.836 0.024
#> GSM96964 5 0.3109 0.7297 0.224 0.000 0.000 0.004 0.772 0.000
#> GSM96965 4 0.1657 0.8326 0.000 0.056 0.016 0.928 0.000 0.000
#> GSM96981 1 0.1788 0.8328 0.916 0.000 0.004 0.004 0.076 0.000
#> GSM96982 1 0.1788 0.8328 0.916 0.000 0.004 0.004 0.076 0.000
#> GSM96988 3 0.3874 0.8130 0.000 0.012 0.804 0.008 0.096 0.080
#> GSM97000 5 0.2289 0.7781 0.020 0.000 0.036 0.008 0.912 0.024
#> GSM97004 1 0.0146 0.8146 0.996 0.000 0.000 0.000 0.000 0.004
#> GSM97008 5 0.2327 0.7909 0.044 0.000 0.028 0.008 0.908 0.012
#> GSM96950 5 0.3081 0.7323 0.220 0.000 0.000 0.004 0.776 0.000
#> GSM96980 1 0.2113 0.7425 0.896 0.000 0.000 0.092 0.008 0.004
#> GSM96989 5 0.3690 0.6501 0.288 0.000 0.000 0.012 0.700 0.000
#> GSM96992 1 0.2092 0.8091 0.876 0.000 0.000 0.000 0.124 0.000
#> GSM96993 5 0.2765 0.7797 0.132 0.000 0.000 0.016 0.848 0.004
#> GSM96958 5 0.3109 0.7266 0.224 0.000 0.000 0.004 0.772 0.000
#> GSM96951 1 0.3371 0.6138 0.708 0.000 0.000 0.000 0.292 0.000
#> GSM96952 1 0.2092 0.8091 0.876 0.000 0.000 0.000 0.124 0.000
#> GSM96961 1 0.2092 0.8091 0.876 0.000 0.000 0.000 0.124 0.000
Heatmaps for the consensus matrix. It visualizes the probability of two samples to be in a same group.
consensus_heatmap(res, k = 2)
consensus_heatmap(res, k = 3)
consensus_heatmap(res, k = 4)
consensus_heatmap(res, k = 5)
consensus_heatmap(res, k = 6)
Heatmaps for the membership of samples in all partitions to see how consistent they are:
membership_heatmap(res, k = 2)
membership_heatmap(res, k = 3)
membership_heatmap(res, k = 4)
membership_heatmap(res, k = 5)
membership_heatmap(res, k = 6)
As soon as we have had the classes for columns, we can look for signatures which are significantly different between classes which can be candidate marks for certain classes. Following are the heatmaps for signatures.
Signature heatmaps where rows are scaled:
get_signatures(res, k = 2)
get_signatures(res, k = 3)
get_signatures(res, k = 4)
get_signatures(res, k = 5)
get_signatures(res, k = 6)
Signature heatmaps where rows are not scaled:
get_signatures(res, k = 2, scale_rows = FALSE)
get_signatures(res, k = 3, scale_rows = FALSE)
get_signatures(res, k = 4, scale_rows = FALSE)
get_signatures(res, k = 5, scale_rows = FALSE)
get_signatures(res, k = 6, scale_rows = FALSE)
Compare the overlap of signatures from different k:
compare_signatures(res)
get_signature() returns a data frame invisibly. TO get the list of signatures, the function
call should be assigned to a variable explicitly. In following code, if plot argument is set
to FALSE, no heatmap is plotted while only the differential analysis is performed.
# code only for demonstration
tb = get_signature(res, k = ..., plot = FALSE)
An example of the output of tb is:
#> which_row fdr mean_1 mean_2 scaled_mean_1 scaled_mean_2 km
#> 1 38 0.042760348 8.373488 9.131774 -0.5533452 0.5164555 1
#> 2 40 0.018707592 7.106213 8.469186 -0.6173731 0.5762149 1
#> 3 55 0.019134737 10.221463 11.207825 -0.6159697 0.5749050 1
#> 4 59 0.006059896 5.921854 7.869574 -0.6899429 0.6439467 1
#> 5 60 0.018055526 8.928898 10.211722 -0.6204761 0.5791110 1
#> 6 98 0.009384629 15.714769 14.887706 0.6635654 -0.6193277 2
...
The columns in tb are:
which_row: row indices corresponding to the input matrix.fdr: FDR for the differential test. mean_x: The mean value in group x.scaled_mean_x: The mean value in group x after rows are scaled.km: Row groups if k-means clustering is applied to rows.UMAP plot which shows how samples are separated.
dimension_reduction(res, k = 2, method = "UMAP")
dimension_reduction(res, k = 3, method = "UMAP")
dimension_reduction(res, k = 4, method = "UMAP")
dimension_reduction(res, k = 5, method = "UMAP")
dimension_reduction(res, k = 6, method = "UMAP")
Following heatmap shows how subgroups are split when increasing k:
collect_classes(res)
Test correlation between subgroups and known annotations. If the known annotation is numeric, one-way ANOVA test is applied, and if the known annotation is discrete, chi-squared contingency table test is applied.
test_to_known_factors(res)
#> n disease.state(p) specimen(p) cell.type(p) other(p) k
#> SD:hclust 99 0.000120 0.166 2.00e-10 0.00263 2
#> SD:hclust 99 0.001471 0.188 8.26e-14 0.01966 3
#> SD:hclust 99 0.000597 0.072 5.23e-16 0.02581 4
#> SD:hclust 95 0.000227 0.125 6.70e-15 0.08524 5
#> SD:hclust 97 0.000355 0.165 1.98e-17 0.03266 6
If matrix rows can be associated to genes, consider to use functional_enrichment(res,
...) to perform function enrichment for the signature genes. See this vignette for more detailed explanations.
The object with results only for a single top-value method and a single partition method can be extracted as:
res = res_list["SD", "kmeans"]
# you can also extract it by
# res = res_list["SD:kmeans"]
A summary of res and all the functions that can be applied to it:
res
#> A 'ConsensusPartition' object with k = 2, 3, 4, 5, 6.
#> On a matrix with 21168 rows and 100 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 1.000 0.951 0.980 0.4885 0.508 0.508
#> 3 3 0.630 0.793 0.885 0.3320 0.751 0.551
#> 4 4 0.764 0.513 0.746 0.1294 0.913 0.758
#> 5 5 0.694 0.655 0.802 0.0655 0.874 0.603
#> 6 6 0.746 0.759 0.807 0.0457 0.916 0.660
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
#> GSM97038 2 0.0376 0.9636 0.004 0.996
#> GSM97045 2 0.0376 0.9636 0.004 0.996
#> GSM97047 2 0.0376 0.9636 0.004 0.996
#> GSM97025 2 0.0376 0.9636 0.004 0.996
#> GSM97030 2 0.0000 0.9622 0.000 1.000
#> GSM97027 2 0.0376 0.9636 0.004 0.996
#> GSM97033 2 0.0376 0.9636 0.004 0.996
#> GSM97034 2 0.0000 0.9622 0.000 1.000
#> GSM97020 2 0.0376 0.9636 0.004 0.996
#> GSM97026 2 0.0376 0.9636 0.004 0.996
#> GSM97012 2 0.0376 0.9636 0.004 0.996
#> GSM97015 2 0.0000 0.9622 0.000 1.000
#> GSM97016 2 0.0376 0.9636 0.004 0.996
#> GSM97017 1 0.0000 0.9919 1.000 0.000
#> GSM97019 2 0.0376 0.9636 0.004 0.996
#> GSM97022 2 0.0376 0.9636 0.004 0.996
#> GSM97035 2 0.0376 0.9636 0.004 0.996
#> GSM97036 1 0.0000 0.9919 1.000 0.000
#> GSM97039 2 0.0376 0.9636 0.004 0.996
#> GSM97046 2 0.0376 0.9636 0.004 0.996
#> GSM97023 1 0.0000 0.9919 1.000 0.000
#> GSM97029 1 0.0000 0.9919 1.000 0.000
#> GSM97043 2 0.0376 0.9636 0.004 0.996
#> GSM97013 1 0.0000 0.9919 1.000 0.000
#> GSM96956 2 0.0000 0.9622 0.000 1.000
#> GSM97024 2 0.0000 0.9622 0.000 1.000
#> GSM97032 2 0.0000 0.9622 0.000 1.000
#> GSM97044 2 0.0000 0.9622 0.000 1.000
#> GSM97049 2 0.0376 0.9636 0.004 0.996
#> GSM96968 1 0.1633 0.9728 0.976 0.024
#> GSM96971 1 0.0376 0.9899 0.996 0.004
#> GSM96986 1 0.0376 0.9899 0.996 0.004
#> GSM97003 1 0.0376 0.9899 0.996 0.004
#> GSM96957 1 0.0000 0.9919 1.000 0.000
#> GSM96960 1 0.0376 0.9899 0.996 0.004
#> GSM96975 1 0.0000 0.9919 1.000 0.000
#> GSM96998 1 0.0000 0.9919 1.000 0.000
#> GSM96999 1 0.0000 0.9919 1.000 0.000
#> GSM97001 1 0.0000 0.9919 1.000 0.000
#> GSM97005 1 0.0000 0.9919 1.000 0.000
#> GSM97006 1 0.0000 0.9919 1.000 0.000
#> GSM97021 1 0.0000 0.9919 1.000 0.000
#> GSM97028 2 0.9998 0.0613 0.492 0.508
#> GSM97031 1 0.0376 0.9899 0.996 0.004
#> GSM97037 2 0.0000 0.9622 0.000 1.000
#> GSM97018 2 0.0000 0.9622 0.000 1.000
#> GSM97014 2 0.0376 0.9636 0.004 0.996
#> GSM97042 2 0.0376 0.9636 0.004 0.996
#> GSM97040 2 0.9248 0.5013 0.340 0.660
#> GSM97041 1 0.0000 0.9919 1.000 0.000
#> GSM96955 2 0.0376 0.9636 0.004 0.996
#> GSM96990 2 0.0000 0.9622 0.000 1.000
#> GSM96991 2 0.0376 0.9636 0.004 0.996
#> GSM97048 2 0.0376 0.9636 0.004 0.996
#> GSM96963 2 0.0376 0.9636 0.004 0.996
#> GSM96953 2 0.0376 0.9636 0.004 0.996
#> GSM96966 1 0.0000 0.9919 1.000 0.000
#> GSM96979 1 0.0376 0.9899 0.996 0.004
#> GSM96983 2 0.0000 0.9622 0.000 1.000
#> GSM96984 1 0.2603 0.9525 0.956 0.044
#> GSM96994 2 0.1843 0.9405 0.028 0.972
#> GSM96996 1 0.0000 0.9919 1.000 0.000
#> GSM96997 1 0.0376 0.9899 0.996 0.004
#> GSM97007 2 0.4022 0.8903 0.080 0.920
#> GSM96954 1 0.0376 0.9899 0.996 0.004
#> GSM96962 1 0.0376 0.9899 0.996 0.004
#> GSM96969 1 0.0000 0.9919 1.000 0.000
#> GSM96970 1 0.0000 0.9919 1.000 0.000
#> GSM96973 1 0.0000 0.9919 1.000 0.000
#> GSM96976 1 0.8499 0.6034 0.724 0.276
#> GSM96977 1 0.0000 0.9919 1.000 0.000
#> GSM96995 2 0.9993 0.0796 0.484 0.516
#> GSM97002 1 0.0000 0.9919 1.000 0.000
#> GSM97009 2 0.0376 0.9636 0.004 0.996
#> GSM97010 1 0.0000 0.9919 1.000 0.000
#> GSM96974 1 0.0376 0.9899 0.996 0.004
#> GSM96985 1 0.0376 0.9899 0.996 0.004
#> GSM96959 2 0.0000 0.9622 0.000 1.000
#> GSM96972 1 0.0000 0.9919 1.000 0.000
#> GSM96978 1 0.2948 0.9438 0.948 0.052
#> GSM96967 1 0.0000 0.9919 1.000 0.000
#> GSM96987 1 0.0000 0.9919 1.000 0.000
#> GSM97011 1 0.0000 0.9919 1.000 0.000
#> GSM96964 1 0.0000 0.9919 1.000 0.000
#> GSM96965 1 0.0000 0.9919 1.000 0.000
#> GSM96981 1 0.0000 0.9919 1.000 0.000
#> GSM96982 1 0.0000 0.9919 1.000 0.000
#> GSM96988 1 0.0376 0.9899 0.996 0.004
#> GSM97000 1 0.0376 0.9899 0.996 0.004
#> GSM97004 1 0.0000 0.9919 1.000 0.000
#> GSM97008 1 0.0000 0.9919 1.000 0.000
#> GSM96950 1 0.0000 0.9919 1.000 0.000
#> GSM96980 1 0.0000 0.9919 1.000 0.000
#> GSM96989 1 0.0000 0.9919 1.000 0.000
#> GSM96992 1 0.0000 0.9919 1.000 0.000
#> GSM96993 1 0.0000 0.9919 1.000 0.000
#> GSM96958 1 0.0000 0.9919 1.000 0.000
#> GSM96951 1 0.0000 0.9919 1.000 0.000
#> GSM96952 1 0.0000 0.9919 1.000 0.000
#> GSM96961 1 0.0000 0.9919 1.000 0.000
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM97038 2 0.0000 0.9289 0.000 1.000 0.000
#> GSM97045 2 0.0424 0.9300 0.000 0.992 0.008
#> GSM97047 2 0.6180 0.0760 0.000 0.584 0.416
#> GSM97025 2 0.0424 0.9300 0.000 0.992 0.008
#> GSM97030 3 0.5529 0.6692 0.000 0.296 0.704
#> GSM97027 2 0.0424 0.9300 0.000 0.992 0.008
#> GSM97033 2 0.0237 0.9282 0.000 0.996 0.004
#> GSM97034 3 0.5465 0.6801 0.000 0.288 0.712
#> GSM97020 2 0.0237 0.9282 0.000 0.996 0.004
#> GSM97026 2 0.5420 0.5915 0.008 0.752 0.240
#> GSM97012 2 0.0747 0.9303 0.000 0.984 0.016
#> GSM97015 3 0.5656 0.6851 0.004 0.284 0.712
#> GSM97016 2 0.0237 0.9282 0.000 0.996 0.004
#> GSM97017 1 0.1315 0.8794 0.972 0.008 0.020
#> GSM97019 2 0.0747 0.9303 0.000 0.984 0.016
#> GSM97022 2 0.0747 0.9303 0.000 0.984 0.016
#> GSM97035 2 0.0747 0.9303 0.000 0.984 0.016
#> GSM97036 1 0.0848 0.8819 0.984 0.008 0.008
#> GSM97039 2 0.0237 0.9282 0.000 0.996 0.004
#> GSM97046 2 0.0237 0.9282 0.000 0.996 0.004
#> GSM97023 1 0.0000 0.8835 1.000 0.000 0.000
#> GSM97029 1 0.1315 0.8794 0.972 0.008 0.020
#> GSM97043 2 0.0747 0.9303 0.000 0.984 0.016
#> GSM97013 1 0.0661 0.8819 0.988 0.008 0.004
#> GSM96956 2 0.3267 0.8145 0.000 0.884 0.116
#> GSM97024 2 0.0747 0.9303 0.000 0.984 0.016
#> GSM97032 3 0.5465 0.6801 0.000 0.288 0.712
#> GSM97044 3 0.5465 0.6801 0.000 0.288 0.712
#> GSM97049 2 0.0237 0.9282 0.000 0.996 0.004
#> GSM96968 3 0.4974 0.7209 0.236 0.000 0.764
#> GSM96971 3 0.1031 0.7858 0.024 0.000 0.976
#> GSM96986 3 0.2878 0.8046 0.096 0.000 0.904
#> GSM97003 1 0.3412 0.8425 0.876 0.000 0.124
#> GSM96957 1 0.1315 0.8794 0.972 0.008 0.020
#> GSM96960 1 0.3340 0.8423 0.880 0.000 0.120
#> GSM96975 1 0.0592 0.8829 0.988 0.000 0.012
#> GSM96998 1 0.0424 0.8835 0.992 0.000 0.008
#> GSM96999 1 0.0592 0.8829 0.988 0.000 0.012
#> GSM97001 1 0.1315 0.8794 0.972 0.008 0.020
#> GSM97005 1 0.0747 0.8822 0.984 0.000 0.016
#> GSM97006 1 0.3340 0.8423 0.880 0.000 0.120
#> GSM97021 1 0.2384 0.8596 0.936 0.008 0.056
#> GSM97028 3 0.4469 0.8095 0.076 0.060 0.864
#> GSM97031 1 0.3267 0.8362 0.884 0.000 0.116
#> GSM97037 3 0.6225 0.4056 0.000 0.432 0.568
#> GSM97018 3 0.5431 0.6829 0.000 0.284 0.716
#> GSM97014 2 0.3325 0.8310 0.076 0.904 0.020
#> GSM97042 2 0.0747 0.9303 0.000 0.984 0.016
#> GSM97040 1 0.6936 -0.0340 0.524 0.016 0.460
#> GSM97041 1 0.1315 0.8794 0.972 0.008 0.020
#> GSM96955 2 0.0592 0.9288 0.000 0.988 0.012
#> GSM96990 3 0.5497 0.6798 0.000 0.292 0.708
#> GSM96991 2 0.0747 0.9303 0.000 0.984 0.016
#> GSM97048 2 0.0237 0.9282 0.000 0.996 0.004
#> GSM96963 2 0.0747 0.9303 0.000 0.984 0.016
#> GSM96953 2 0.0747 0.9303 0.000 0.984 0.016
#> GSM96966 1 0.5397 0.7273 0.720 0.000 0.280
#> GSM96979 3 0.2959 0.8041 0.100 0.000 0.900
#> GSM96983 3 0.3551 0.7800 0.000 0.132 0.868
#> GSM96984 3 0.3043 0.8101 0.084 0.008 0.908
#> GSM96994 3 0.3325 0.8025 0.020 0.076 0.904
#> GSM96996 1 0.2066 0.8708 0.940 0.000 0.060
#> GSM96997 3 0.2959 0.8041 0.100 0.000 0.900
#> GSM97007 3 0.3045 0.8043 0.020 0.064 0.916
#> GSM96954 3 0.4887 0.7284 0.228 0.000 0.772
#> GSM96962 3 0.2959 0.8041 0.100 0.000 0.900
#> GSM96969 1 0.5397 0.7273 0.720 0.000 0.280
#> GSM96970 1 0.5363 0.7310 0.724 0.000 0.276
#> GSM96973 1 0.5397 0.7273 0.720 0.000 0.280
#> GSM96976 3 0.2056 0.7901 0.024 0.024 0.952
#> GSM96977 1 0.6026 0.2927 0.624 0.000 0.376
#> GSM96995 3 0.6375 0.7069 0.244 0.036 0.720
#> GSM97002 1 0.3116 0.8488 0.892 0.000 0.108
#> GSM97009 2 0.8920 0.0892 0.144 0.532 0.324
#> GSM97010 1 0.2878 0.8453 0.904 0.000 0.096
#> GSM96974 3 0.3482 0.7101 0.128 0.000 0.872
#> GSM96985 3 0.4796 0.6359 0.220 0.000 0.780
#> GSM96959 3 0.6337 0.6984 0.028 0.264 0.708
#> GSM96972 1 0.5397 0.7273 0.720 0.000 0.280
#> GSM96978 3 0.2200 0.8083 0.056 0.004 0.940
#> GSM96967 1 0.5397 0.7273 0.720 0.000 0.280
#> GSM96987 1 0.0000 0.8835 1.000 0.000 0.000
#> GSM97011 1 0.3213 0.8318 0.900 0.008 0.092
#> GSM96964 1 0.0000 0.8835 1.000 0.000 0.000
#> GSM96965 1 0.5016 0.7629 0.760 0.000 0.240
#> GSM96981 1 0.0237 0.8831 0.996 0.000 0.004
#> GSM96982 1 0.1753 0.8749 0.952 0.000 0.048
#> GSM96988 3 0.2878 0.8063 0.096 0.000 0.904
#> GSM97000 1 0.6252 0.0655 0.556 0.000 0.444
#> GSM97004 1 0.3412 0.8407 0.876 0.000 0.124
#> GSM97008 1 0.3816 0.7743 0.852 0.000 0.148
#> GSM96950 1 0.0424 0.8832 0.992 0.000 0.008
#> GSM96980 1 0.4291 0.8099 0.820 0.000 0.180
#> GSM96989 1 0.0000 0.8835 1.000 0.000 0.000
#> GSM96992 1 0.0592 0.8830 0.988 0.000 0.012
#> GSM96993 1 0.0829 0.8825 0.984 0.004 0.012
#> GSM96958 1 0.0000 0.8835 1.000 0.000 0.000
#> GSM96951 1 0.0424 0.8835 0.992 0.000 0.008
#> GSM96952 1 0.0424 0.8835 0.992 0.000 0.008
#> GSM96961 1 0.0424 0.8835 0.992 0.000 0.008
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM97038 2 0.1792 0.9371 0.000 0.932 0.000 0.068
#> GSM97045 2 0.0000 0.9487 0.000 1.000 0.000 0.000
#> GSM97047 4 0.7793 0.3328 0.036 0.356 0.112 0.496
#> GSM97025 2 0.0000 0.9487 0.000 1.000 0.000 0.000
#> GSM97030 3 0.1211 0.9036 0.000 0.040 0.960 0.000
#> GSM97027 2 0.0000 0.9487 0.000 1.000 0.000 0.000
#> GSM97033 2 0.1716 0.9378 0.000 0.936 0.000 0.064
#> GSM97034 3 0.1471 0.9066 0.004 0.024 0.960 0.012
#> GSM97020 2 0.1792 0.9371 0.000 0.932 0.000 0.068
#> GSM97026 4 0.7663 0.3058 0.052 0.396 0.072 0.480
#> GSM97012 2 0.0000 0.9487 0.000 1.000 0.000 0.000
#> GSM97015 3 0.1471 0.9066 0.004 0.024 0.960 0.012
#> GSM97016 2 0.1792 0.9371 0.000 0.932 0.000 0.068
#> GSM97017 1 0.5294 -0.1617 0.508 0.000 0.008 0.484
#> GSM97019 2 0.0000 0.9487 0.000 1.000 0.000 0.000
#> GSM97022 2 0.0000 0.9487 0.000 1.000 0.000 0.000
#> GSM97035 2 0.0000 0.9487 0.000 1.000 0.000 0.000
#> GSM97036 1 0.5288 -0.1409 0.520 0.000 0.008 0.472
#> GSM97039 2 0.1716 0.9378 0.000 0.936 0.000 0.064
#> GSM97046 2 0.1792 0.9371 0.000 0.932 0.000 0.068
#> GSM97023 1 0.3400 0.3384 0.820 0.000 0.000 0.180
#> GSM97029 1 0.5294 -0.1617 0.508 0.000 0.008 0.484
#> GSM97043 2 0.1109 0.9296 0.000 0.968 0.028 0.004
#> GSM97013 1 0.5163 -0.1426 0.516 0.000 0.004 0.480
#> GSM96956 2 0.6249 0.3867 0.000 0.580 0.352 0.068
#> GSM97024 2 0.0000 0.9487 0.000 1.000 0.000 0.000
#> GSM97032 3 0.1211 0.9036 0.000 0.040 0.960 0.000
#> GSM97044 3 0.1211 0.9036 0.000 0.040 0.960 0.000
#> GSM97049 2 0.1792 0.9371 0.000 0.932 0.000 0.068
#> GSM96968 3 0.1109 0.9060 0.004 0.000 0.968 0.028
#> GSM96971 3 0.1302 0.9024 0.000 0.000 0.956 0.044
#> GSM96986 3 0.1256 0.9082 0.008 0.000 0.964 0.028
#> GSM97003 1 0.1807 0.4482 0.940 0.000 0.008 0.052
#> GSM96957 1 0.5168 -0.1619 0.504 0.000 0.004 0.492
#> GSM96960 1 0.1635 0.4509 0.948 0.000 0.008 0.044
#> GSM96975 1 0.4999 -0.1544 0.508 0.000 0.000 0.492
#> GSM96998 1 0.0188 0.4571 0.996 0.000 0.004 0.000
#> GSM96999 1 0.4998 -0.1459 0.512 0.000 0.000 0.488
#> GSM97001 1 0.5168 -0.1619 0.504 0.000 0.004 0.492
#> GSM97005 1 0.5168 -0.1714 0.500 0.000 0.004 0.496
#> GSM97006 1 0.1452 0.4535 0.956 0.000 0.008 0.036
#> GSM97021 4 0.5781 0.1492 0.480 0.000 0.028 0.492
#> GSM97028 3 0.1059 0.9096 0.000 0.016 0.972 0.012
#> GSM97031 1 0.4831 0.2332 0.704 0.000 0.016 0.280
#> GSM97037 3 0.4499 0.7384 0.000 0.160 0.792 0.048
#> GSM97018 3 0.1471 0.9066 0.004 0.024 0.960 0.012
#> GSM97014 4 0.6513 0.2461 0.044 0.400 0.016 0.540
#> GSM97042 2 0.0000 0.9487 0.000 1.000 0.000 0.000
#> GSM97040 4 0.6758 0.2583 0.424 0.004 0.080 0.492
#> GSM97041 1 0.5294 -0.1617 0.508 0.000 0.008 0.484
#> GSM96955 2 0.3972 0.7437 0.000 0.788 0.008 0.204
#> GSM96990 3 0.1471 0.9066 0.004 0.024 0.960 0.012
#> GSM96991 2 0.0000 0.9487 0.000 1.000 0.000 0.000
#> GSM97048 2 0.1792 0.9371 0.000 0.932 0.000 0.068
#> GSM96963 2 0.0000 0.9487 0.000 1.000 0.000 0.000
#> GSM96953 2 0.0000 0.9487 0.000 1.000 0.000 0.000
#> GSM96966 1 0.5792 0.2509 0.552 0.000 0.032 0.416
#> GSM96979 3 0.1256 0.9082 0.008 0.000 0.964 0.028
#> GSM96983 3 0.0564 0.9117 0.004 0.004 0.988 0.004
#> GSM96984 3 0.1256 0.9082 0.008 0.000 0.964 0.028
#> GSM96994 3 0.1151 0.9087 0.008 0.000 0.968 0.024
#> GSM96996 1 0.0524 0.4576 0.988 0.000 0.008 0.004
#> GSM96997 3 0.1256 0.9082 0.008 0.000 0.964 0.028
#> GSM97007 3 0.1256 0.9082 0.008 0.000 0.964 0.028
#> GSM96954 3 0.1256 0.9075 0.008 0.000 0.964 0.028
#> GSM96962 3 0.1256 0.9082 0.008 0.000 0.964 0.028
#> GSM96969 1 0.5792 0.2509 0.552 0.000 0.032 0.416
#> GSM96970 1 0.5792 0.2509 0.552 0.000 0.032 0.416
#> GSM96973 1 0.5792 0.2509 0.552 0.000 0.032 0.416
#> GSM96976 3 0.6924 0.3439 0.108 0.000 0.464 0.428
#> GSM96977 4 0.6546 0.2483 0.432 0.000 0.076 0.492
#> GSM96995 3 0.1388 0.8995 0.012 0.000 0.960 0.028
#> GSM97002 1 0.1635 0.4509 0.948 0.000 0.008 0.044
#> GSM97009 4 0.8247 0.3201 0.304 0.120 0.068 0.508
#> GSM97010 1 0.5697 -0.2141 0.488 0.000 0.024 0.488
#> GSM96974 4 0.7521 -0.3152 0.184 0.000 0.396 0.420
#> GSM96985 3 0.6147 0.5903 0.200 0.000 0.672 0.128
#> GSM96959 3 0.5276 0.1918 0.004 0.004 0.560 0.432
#> GSM96972 1 0.5792 0.2509 0.552 0.000 0.032 0.416
#> GSM96978 3 0.0336 0.9107 0.008 0.000 0.992 0.000
#> GSM96967 1 0.5792 0.2509 0.552 0.000 0.032 0.416
#> GSM96987 1 0.0188 0.4555 0.996 0.000 0.000 0.004
#> GSM97011 4 0.5607 0.1366 0.484 0.000 0.020 0.496
#> GSM96964 1 0.4008 0.2762 0.756 0.000 0.000 0.244
#> GSM96965 4 0.5022 -0.0906 0.264 0.000 0.028 0.708
#> GSM96981 1 0.4790 0.0223 0.620 0.000 0.000 0.380
#> GSM96982 1 0.1356 0.4543 0.960 0.000 0.008 0.032
#> GSM96988 3 0.0336 0.9107 0.008 0.000 0.992 0.000
#> GSM97000 4 0.6491 0.2469 0.432 0.000 0.072 0.496
#> GSM97004 1 0.2048 0.4409 0.928 0.000 0.008 0.064
#> GSM97008 4 0.5607 0.1334 0.484 0.000 0.020 0.496
#> GSM96950 1 0.4994 -0.1361 0.520 0.000 0.000 0.480
#> GSM96980 1 0.5203 0.2584 0.576 0.000 0.008 0.416
#> GSM96989 1 0.0188 0.4555 0.996 0.000 0.000 0.004
#> GSM96992 1 0.0336 0.4575 0.992 0.000 0.008 0.000
#> GSM96993 1 0.5604 -0.1816 0.504 0.000 0.020 0.476
#> GSM96958 1 0.4948 -0.0642 0.560 0.000 0.000 0.440
#> GSM96951 1 0.4567 0.2416 0.716 0.000 0.008 0.276
#> GSM96952 1 0.0336 0.4575 0.992 0.000 0.008 0.000
#> GSM96961 1 0.2053 0.4224 0.924 0.000 0.004 0.072
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM97038 2 0.0932 0.8122 0.004 0.972 0.004 0.000 0.020
#> GSM97045 2 0.3430 0.8677 0.000 0.776 0.004 0.220 0.000
#> GSM97047 5 0.4451 0.6233 0.000 0.036 0.128 0.048 0.788
#> GSM97025 2 0.3461 0.8682 0.000 0.772 0.004 0.224 0.000
#> GSM97030 3 0.1597 0.8037 0.000 0.000 0.940 0.012 0.048
#> GSM97027 2 0.3430 0.8677 0.000 0.776 0.004 0.220 0.000
#> GSM97033 2 0.0324 0.8231 0.004 0.992 0.000 0.000 0.004
#> GSM97034 3 0.1701 0.8013 0.000 0.000 0.936 0.016 0.048
#> GSM97020 2 0.0324 0.8231 0.004 0.992 0.000 0.000 0.004
#> GSM97026 5 0.4894 0.6113 0.000 0.020 0.116 0.112 0.752
#> GSM97012 2 0.3612 0.8682 0.000 0.764 0.008 0.228 0.000
#> GSM97015 3 0.1740 0.8009 0.000 0.000 0.932 0.012 0.056
#> GSM97016 2 0.0324 0.8231 0.004 0.992 0.000 0.000 0.004
#> GSM97017 5 0.2104 0.8131 0.060 0.000 0.000 0.024 0.916
#> GSM97019 2 0.3612 0.8682 0.000 0.764 0.008 0.228 0.000
#> GSM97022 2 0.3612 0.8682 0.000 0.764 0.008 0.228 0.000
#> GSM97035 2 0.3612 0.8682 0.000 0.764 0.008 0.228 0.000
#> GSM97036 5 0.3914 0.7339 0.164 0.000 0.000 0.048 0.788
#> GSM97039 2 0.0324 0.8231 0.004 0.992 0.000 0.000 0.004
#> GSM97046 2 0.0324 0.8231 0.004 0.992 0.000 0.000 0.004
#> GSM97023 1 0.4602 0.4989 0.656 0.000 0.000 0.028 0.316
#> GSM97029 5 0.2754 0.8027 0.080 0.000 0.000 0.040 0.880
#> GSM97043 2 0.6112 0.7646 0.000 0.636 0.112 0.216 0.036
#> GSM97013 5 0.3736 0.7560 0.140 0.000 0.000 0.052 0.808
#> GSM96956 2 0.4655 0.4217 0.000 0.660 0.312 0.004 0.024
#> GSM97024 2 0.3612 0.8682 0.000 0.764 0.008 0.228 0.000
#> GSM97032 3 0.1845 0.7993 0.000 0.000 0.928 0.016 0.056
#> GSM97044 3 0.0912 0.8102 0.000 0.000 0.972 0.012 0.016
#> GSM97049 2 0.0324 0.8231 0.004 0.992 0.000 0.000 0.004
#> GSM96968 3 0.1704 0.7986 0.000 0.000 0.928 0.004 0.068
#> GSM96971 3 0.4270 0.6831 0.012 0.000 0.668 0.320 0.000
#> GSM96986 3 0.3989 0.7534 0.008 0.000 0.728 0.260 0.004
#> GSM97003 1 0.4686 0.6275 0.736 0.000 0.000 0.104 0.160
#> GSM96957 5 0.2236 0.8109 0.068 0.000 0.000 0.024 0.908
#> GSM96960 1 0.3278 0.6771 0.824 0.000 0.000 0.020 0.156
#> GSM96975 5 0.1830 0.8124 0.068 0.000 0.000 0.008 0.924
#> GSM96998 1 0.4190 0.6721 0.768 0.000 0.000 0.060 0.172
#> GSM96999 5 0.2813 0.7908 0.108 0.000 0.000 0.024 0.868
#> GSM97001 5 0.1697 0.8134 0.060 0.000 0.000 0.008 0.932
#> GSM97005 5 0.1557 0.8147 0.052 0.000 0.000 0.008 0.940
#> GSM97006 1 0.3278 0.6771 0.824 0.000 0.000 0.020 0.156
#> GSM97021 5 0.1106 0.8144 0.024 0.000 0.000 0.012 0.964
#> GSM97028 3 0.1211 0.8125 0.000 0.000 0.960 0.024 0.016
#> GSM97031 5 0.5114 -0.0803 0.472 0.000 0.000 0.036 0.492
#> GSM97037 3 0.4792 0.5529 0.000 0.232 0.712 0.012 0.044
#> GSM97018 3 0.1774 0.7991 0.000 0.000 0.932 0.016 0.052
#> GSM97014 5 0.3276 0.6700 0.000 0.132 0.000 0.032 0.836
#> GSM97042 2 0.3612 0.8682 0.000 0.764 0.008 0.228 0.000
#> GSM97040 5 0.1502 0.7784 0.000 0.000 0.056 0.004 0.940
#> GSM97041 5 0.2104 0.8131 0.060 0.000 0.000 0.024 0.916
#> GSM96955 2 0.6283 0.5402 0.000 0.576 0.020 0.124 0.280
#> GSM96990 3 0.1740 0.8009 0.000 0.000 0.932 0.012 0.056
#> GSM96991 2 0.3612 0.8682 0.000 0.764 0.008 0.228 0.000
#> GSM97048 2 0.0324 0.8231 0.004 0.992 0.000 0.000 0.004
#> GSM96963 2 0.3582 0.8684 0.000 0.768 0.008 0.224 0.000
#> GSM96953 2 0.3582 0.8684 0.000 0.768 0.008 0.224 0.000
#> GSM96966 1 0.4533 -0.3596 0.544 0.000 0.000 0.448 0.008
#> GSM96979 3 0.3989 0.7534 0.008 0.000 0.728 0.260 0.004
#> GSM96983 3 0.1121 0.8114 0.000 0.000 0.956 0.044 0.000
#> GSM96984 3 0.3989 0.7534 0.008 0.000 0.728 0.260 0.004
#> GSM96994 3 0.3963 0.7550 0.008 0.000 0.732 0.256 0.004
#> GSM96996 1 0.3359 0.6810 0.816 0.000 0.000 0.020 0.164
#> GSM96997 3 0.4146 0.7455 0.012 0.000 0.716 0.268 0.004
#> GSM97007 3 0.3963 0.7550 0.008 0.000 0.732 0.256 0.004
#> GSM96954 3 0.3146 0.7984 0.000 0.000 0.844 0.128 0.028
#> GSM96962 3 0.3989 0.7534 0.008 0.000 0.728 0.260 0.004
#> GSM96969 1 0.4533 -0.3596 0.544 0.000 0.000 0.448 0.008
#> GSM96970 1 0.4533 -0.3596 0.544 0.000 0.000 0.448 0.008
#> GSM96973 1 0.4533 -0.3596 0.544 0.000 0.000 0.448 0.008
#> GSM96976 4 0.6219 0.7313 0.260 0.000 0.196 0.544 0.000
#> GSM96977 5 0.1569 0.7910 0.004 0.000 0.044 0.008 0.944
#> GSM96995 3 0.2233 0.7697 0.000 0.000 0.892 0.004 0.104
#> GSM97002 1 0.3183 0.6785 0.828 0.000 0.000 0.016 0.156
#> GSM97009 5 0.2207 0.7715 0.004 0.020 0.040 0.012 0.924
#> GSM97010 5 0.2390 0.8040 0.084 0.000 0.000 0.020 0.896
#> GSM96974 4 0.6348 0.7331 0.292 0.000 0.196 0.512 0.000
#> GSM96985 3 0.5580 0.2541 0.336 0.000 0.576 0.088 0.000
#> GSM96959 5 0.4610 0.4341 0.000 0.020 0.296 0.008 0.676
#> GSM96972 1 0.4546 -0.3675 0.532 0.000 0.000 0.460 0.008
#> GSM96978 3 0.1608 0.8072 0.000 0.000 0.928 0.072 0.000
#> GSM96967 1 0.4533 -0.3596 0.544 0.000 0.000 0.448 0.008
#> GSM96987 1 0.4066 0.6669 0.768 0.000 0.000 0.044 0.188
#> GSM97011 5 0.0451 0.8090 0.008 0.000 0.000 0.004 0.988
#> GSM96964 1 0.5107 0.3882 0.596 0.000 0.000 0.048 0.356
#> GSM96965 4 0.6589 0.4591 0.312 0.000 0.000 0.456 0.232
#> GSM96981 5 0.4510 0.1876 0.432 0.000 0.000 0.008 0.560
#> GSM96982 1 0.2891 0.6818 0.824 0.000 0.000 0.000 0.176
#> GSM96988 3 0.2102 0.8092 0.004 0.000 0.916 0.068 0.012
#> GSM97000 5 0.0854 0.8046 0.008 0.000 0.012 0.004 0.976
#> GSM97004 1 0.3106 0.6677 0.840 0.000 0.000 0.020 0.140
#> GSM97008 5 0.0451 0.8090 0.008 0.000 0.000 0.004 0.988
#> GSM96950 5 0.3821 0.7482 0.148 0.000 0.000 0.052 0.800
#> GSM96980 1 0.1697 0.4303 0.932 0.000 0.000 0.060 0.008
#> GSM96989 1 0.4170 0.6630 0.760 0.000 0.000 0.048 0.192
#> GSM96992 1 0.2891 0.6818 0.824 0.000 0.000 0.000 0.176
#> GSM96993 5 0.4058 0.7560 0.144 0.000 0.008 0.052 0.796
#> GSM96958 5 0.4734 0.3543 0.372 0.000 0.000 0.024 0.604
#> GSM96951 1 0.4367 0.3975 0.620 0.000 0.000 0.008 0.372
#> GSM96952 1 0.2891 0.6818 0.824 0.000 0.000 0.000 0.176
#> GSM96961 1 0.3809 0.6165 0.736 0.000 0.000 0.008 0.256
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM97038 2 0.5154 0.777 0.008 0.680 0.216 0.048 0.048 0.000
#> GSM97045 2 0.0291 0.863 0.000 0.992 0.000 0.004 0.004 0.000
#> GSM97047 5 0.3173 0.727 0.000 0.008 0.156 0.008 0.820 0.008
#> GSM97025 2 0.0436 0.863 0.000 0.988 0.004 0.004 0.004 0.000
#> GSM97030 3 0.4245 0.771 0.000 0.004 0.604 0.000 0.016 0.376
#> GSM97027 2 0.0291 0.863 0.000 0.992 0.000 0.004 0.004 0.000
#> GSM97033 2 0.4777 0.804 0.008 0.724 0.180 0.044 0.044 0.000
#> GSM97034 3 0.4891 0.773 0.000 0.004 0.592 0.032 0.016 0.356
#> GSM97020 2 0.4777 0.804 0.008 0.724 0.180 0.044 0.044 0.000
#> GSM97026 5 0.3875 0.701 0.000 0.016 0.260 0.008 0.716 0.000
#> GSM97012 2 0.0146 0.863 0.000 0.996 0.004 0.000 0.000 0.000
#> GSM97015 3 0.4290 0.773 0.000 0.004 0.612 0.000 0.020 0.364
#> GSM97016 2 0.4745 0.803 0.008 0.724 0.184 0.040 0.044 0.000
#> GSM97017 5 0.2994 0.807 0.076 0.000 0.060 0.008 0.856 0.000
#> GSM97019 2 0.0146 0.863 0.000 0.996 0.004 0.000 0.000 0.000
#> GSM97022 2 0.0146 0.863 0.000 0.996 0.004 0.000 0.000 0.000
#> GSM97035 2 0.0146 0.863 0.000 0.996 0.004 0.000 0.000 0.000
#> GSM97036 5 0.6130 0.523 0.272 0.000 0.148 0.040 0.540 0.000
#> GSM97039 2 0.4745 0.803 0.008 0.724 0.184 0.040 0.044 0.000
#> GSM97046 2 0.4775 0.802 0.008 0.720 0.188 0.040 0.044 0.000
#> GSM97023 1 0.3151 0.791 0.848 0.000 0.076 0.012 0.064 0.000
#> GSM97029 5 0.5288 0.713 0.148 0.000 0.136 0.040 0.676 0.000
#> GSM97043 2 0.3309 0.705 0.000 0.788 0.192 0.004 0.016 0.000
#> GSM97013 5 0.6026 0.542 0.268 0.000 0.144 0.036 0.552 0.000
#> GSM96956 3 0.5907 0.272 0.004 0.212 0.640 0.028 0.036 0.080
#> GSM97024 2 0.0146 0.863 0.000 0.996 0.004 0.000 0.000 0.000
#> GSM97032 3 0.4254 0.776 0.000 0.004 0.624 0.000 0.020 0.352
#> GSM97044 3 0.3899 0.757 0.000 0.000 0.592 0.000 0.004 0.404
#> GSM97049 2 0.4745 0.803 0.008 0.724 0.184 0.040 0.044 0.000
#> GSM96968 3 0.4606 0.757 0.000 0.000 0.604 0.000 0.052 0.344
#> GSM96971 6 0.2039 0.836 0.000 0.000 0.020 0.076 0.000 0.904
#> GSM96986 6 0.0000 0.929 0.000 0.000 0.000 0.000 0.000 1.000
#> GSM97003 1 0.4013 0.689 0.768 0.000 0.016 0.052 0.000 0.164
#> GSM96957 5 0.3526 0.789 0.124 0.000 0.040 0.020 0.816 0.000
#> GSM96960 1 0.1982 0.790 0.912 0.000 0.016 0.068 0.000 0.004
#> GSM96975 5 0.3260 0.797 0.136 0.000 0.028 0.012 0.824 0.000
#> GSM96998 1 0.3300 0.790 0.840 0.000 0.096 0.036 0.028 0.000
#> GSM96999 5 0.4274 0.724 0.200 0.000 0.040 0.024 0.736 0.000
#> GSM97001 5 0.2222 0.812 0.084 0.000 0.012 0.008 0.896 0.000
#> GSM97005 5 0.1866 0.811 0.084 0.000 0.000 0.000 0.908 0.008
#> GSM97006 1 0.1888 0.791 0.916 0.000 0.012 0.068 0.000 0.004
#> GSM97021 5 0.2344 0.816 0.068 0.000 0.028 0.008 0.896 0.000
#> GSM97028 3 0.5072 0.716 0.000 0.000 0.532 0.052 0.012 0.404
#> GSM97031 1 0.6522 0.297 0.464 0.000 0.008 0.020 0.268 0.240
#> GSM97037 3 0.4274 0.614 0.000 0.040 0.736 0.000 0.024 0.200
#> GSM97018 3 0.4776 0.774 0.000 0.004 0.612 0.028 0.016 0.340
#> GSM97014 5 0.1801 0.788 0.012 0.012 0.040 0.004 0.932 0.000
#> GSM97042 2 0.0146 0.863 0.000 0.996 0.004 0.000 0.000 0.000
#> GSM97040 5 0.2231 0.801 0.028 0.000 0.068 0.000 0.900 0.004
#> GSM97041 5 0.3185 0.806 0.076 0.000 0.060 0.016 0.848 0.000
#> GSM96955 2 0.5704 0.260 0.008 0.492 0.080 0.016 0.404 0.000
#> GSM96990 3 0.4290 0.773 0.000 0.004 0.612 0.000 0.020 0.364
#> GSM96991 2 0.0405 0.862 0.000 0.988 0.008 0.000 0.004 0.000
#> GSM97048 2 0.4745 0.803 0.008 0.724 0.184 0.040 0.044 0.000
#> GSM96963 2 0.0405 0.862 0.000 0.988 0.008 0.000 0.004 0.000
#> GSM96953 2 0.0146 0.863 0.000 0.996 0.004 0.000 0.000 0.000
#> GSM96966 4 0.2378 0.898 0.152 0.000 0.000 0.848 0.000 0.000
#> GSM96979 6 0.0146 0.932 0.000 0.000 0.004 0.000 0.000 0.996
#> GSM96983 3 0.4879 0.665 0.000 0.000 0.500 0.048 0.004 0.448
#> GSM96984 6 0.0260 0.933 0.000 0.000 0.008 0.000 0.000 0.992
#> GSM96994 6 0.0260 0.933 0.000 0.000 0.008 0.000 0.000 0.992
#> GSM96996 1 0.2001 0.802 0.920 0.000 0.020 0.044 0.016 0.000
#> GSM96997 6 0.0551 0.912 0.004 0.000 0.008 0.004 0.000 0.984
#> GSM97007 6 0.0260 0.933 0.000 0.000 0.008 0.000 0.000 0.992
#> GSM96954 6 0.3296 0.546 0.000 0.000 0.188 0.008 0.012 0.792
#> GSM96962 6 0.0260 0.933 0.000 0.000 0.008 0.000 0.000 0.992
#> GSM96969 4 0.2378 0.898 0.152 0.000 0.000 0.848 0.000 0.000
#> GSM96970 4 0.2378 0.898 0.152 0.000 0.000 0.848 0.000 0.000
#> GSM96973 4 0.2378 0.898 0.152 0.000 0.000 0.848 0.000 0.000
#> GSM96976 4 0.3492 0.751 0.008 0.000 0.064 0.816 0.000 0.112
#> GSM96977 5 0.2939 0.806 0.044 0.000 0.080 0.008 0.864 0.004
#> GSM96995 3 0.4986 0.691 0.000 0.000 0.612 0.000 0.104 0.284
#> GSM97002 1 0.1982 0.793 0.912 0.000 0.016 0.068 0.004 0.000
#> GSM97009 5 0.2599 0.792 0.028 0.000 0.068 0.008 0.888 0.008
#> GSM97010 5 0.5015 0.736 0.168 0.000 0.092 0.032 0.704 0.004
#> GSM96974 4 0.3413 0.753 0.016 0.000 0.052 0.828 0.000 0.104
#> GSM96985 3 0.6968 0.382 0.156 0.000 0.424 0.072 0.008 0.340
#> GSM96959 5 0.3887 0.598 0.000 0.000 0.248 0.008 0.724 0.020
#> GSM96972 4 0.2416 0.878 0.156 0.000 0.000 0.844 0.000 0.000
#> GSM96978 3 0.4887 0.625 0.000 0.000 0.476 0.048 0.004 0.472
#> GSM96967 4 0.2378 0.898 0.152 0.000 0.000 0.848 0.000 0.000
#> GSM96987 1 0.3985 0.766 0.792 0.000 0.120 0.044 0.044 0.000
#> GSM97011 5 0.2207 0.811 0.060 0.000 0.020 0.008 0.908 0.004
#> GSM96964 1 0.4775 0.721 0.732 0.000 0.120 0.044 0.104 0.000
#> GSM96965 4 0.3433 0.762 0.040 0.000 0.012 0.816 0.132 0.000
#> GSM96981 1 0.4667 0.361 0.608 0.000 0.016 0.028 0.348 0.000
#> GSM96982 1 0.1838 0.806 0.928 0.000 0.012 0.040 0.020 0.000
#> GSM96988 3 0.4886 0.633 0.000 0.000 0.480 0.048 0.004 0.468
#> GSM97000 5 0.2172 0.810 0.044 0.000 0.024 0.000 0.912 0.020
#> GSM97004 1 0.1802 0.791 0.916 0.000 0.012 0.072 0.000 0.000
#> GSM97008 5 0.1921 0.813 0.056 0.000 0.012 0.000 0.920 0.012
#> GSM96950 5 0.6034 0.552 0.256 0.000 0.152 0.036 0.556 0.000
#> GSM96980 1 0.2912 0.679 0.816 0.000 0.012 0.172 0.000 0.000
#> GSM96989 1 0.4110 0.763 0.784 0.000 0.120 0.044 0.052 0.000
#> GSM96992 1 0.1492 0.808 0.940 0.000 0.000 0.036 0.024 0.000
#> GSM96993 5 0.6039 0.539 0.264 0.000 0.148 0.036 0.552 0.000
#> GSM96958 1 0.5229 0.406 0.596 0.000 0.052 0.032 0.320 0.000
#> GSM96951 1 0.3636 0.782 0.820 0.000 0.028 0.028 0.116 0.008
#> GSM96952 1 0.1636 0.808 0.936 0.000 0.004 0.036 0.024 0.000
#> GSM96961 1 0.2122 0.808 0.916 0.000 0.024 0.028 0.032 0.000
Heatmaps for the consensus matrix. It visualizes the probability of two samples to be in a same group.
consensus_heatmap(res, k = 2)
consensus_heatmap(res, k = 3)
consensus_heatmap(res, k = 4)
consensus_heatmap(res, k = 5)
consensus_heatmap(res, k = 6)
Heatmaps for the membership of samples in all partitions to see how consistent they are:
membership_heatmap(res, k = 2)
membership_heatmap(res, k = 3)
membership_heatmap(res, k = 4)
membership_heatmap(res, k = 5)
membership_heatmap(res, k = 6)
As soon as we have had the classes for columns, we can look for signatures which are significantly different between classes which can be candidate marks for certain classes. Following are the heatmaps for signatures.
Signature heatmaps where rows are scaled:
get_signatures(res, k = 2)
get_signatures(res, k = 3)
get_signatures(res, k = 4)
get_signatures(res, k = 5)
get_signatures(res, k = 6)
Signature heatmaps where rows are not scaled:
get_signatures(res, k = 2, scale_rows = FALSE)
get_signatures(res, k = 3, scale_rows = FALSE)
get_signatures(res, k = 4, scale_rows = FALSE)
get_signatures(res, k = 5, scale_rows = FALSE)
get_signatures(res, k = 6, scale_rows = FALSE)
Compare the overlap of signatures from different k:
compare_signatures(res)
get_signature() returns a data frame invisibly. TO get the list of signatures, the function
call should be assigned to a variable explicitly. In following code, if plot argument is set
to FALSE, no heatmap is plotted while only the differential analysis is performed.
# code only for demonstration
tb = get_signature(res, k = ..., plot = FALSE)
An example of the output of tb is:
#> which_row fdr mean_1 mean_2 scaled_mean_1 scaled_mean_2 km
#> 1 38 0.042760348 8.373488 9.131774 -0.5533452 0.5164555 1
#> 2 40 0.018707592 7.106213 8.469186 -0.6173731 0.5762149 1
#> 3 55 0.019134737 10.221463 11.207825 -0.6159697 0.5749050 1
#> 4 59 0.006059896 5.921854 7.869574 -0.6899429 0.6439467 1
#> 5 60 0.018055526 8.928898 10.211722 -0.6204761 0.5791110 1
#> 6 98 0.009384629 15.714769 14.887706 0.6635654 -0.6193277 2
...
The columns in tb are:
which_row: row indices corresponding to the input matrix.fdr: FDR for the differential test. mean_x: The mean value in group x.scaled_mean_x: The mean value in group x after rows are scaled.km: Row groups if k-means clustering is applied to rows.UMAP plot which shows how samples are separated.
dimension_reduction(res, k = 2, method = "UMAP")
dimension_reduction(res, k = 3, method = "UMAP")
dimension_reduction(res, k = 4, method = "UMAP")
dimension_reduction(res, k = 5, method = "UMAP")
dimension_reduction(res, k = 6, method = "UMAP")
Following heatmap shows how subgroups are split when increasing k:
collect_classes(res)
Test correlation between subgroups and known annotations. If the known annotation is numeric, one-way ANOVA test is applied, and if the known annotation is discrete, chi-squared contingency table test is applied.
test_to_known_factors(res)
#> n disease.state(p) specimen(p) cell.type(p) other(p) k
#> SD:kmeans 98 4.41e-06 0.222 5.40e-15 0.0691 2
#> SD:kmeans 94 1.92e-04 0.413 1.41e-18 0.0374 3
#> SD:kmeans 46 4.76e-03 0.550 1.31e-05 0.0231 4
#> SD:kmeans 83 5.43e-03 0.468 2.72e-12 0.2240 5
#> SD:kmeans 94 1.45e-04 0.372 2.03e-16 0.0504 6
If matrix rows can be associated to genes, consider to use functional_enrichment(res,
...) to perform function enrichment for the signature genes. See this vignette for more detailed explanations.
The object with results only for a single top-value method and a single partition method can be extracted as:
res = res_list["SD", "skmeans"]
# you can also extract it by
# res = res_list["SD:skmeans"]
A summary of res and all the functions that can be applied to it:
res
#> A 'ConsensusPartition' object with k = 2, 3, 4, 5, 6.
#> On a matrix with 21168 rows and 100 columns.
#> Top rows (1000, 2000, 3000, 4000, 5000) are extracted by 'SD' method.
#> Subgroups are detected by 'skmeans' method.
#> Performed in total 1250 partitions by row resampling.
#> Best k for subgroups seems to be 2.
#>
#> Following methods can be applied to this 'ConsensusPartition' object:
#> [1] "cola_report" "collect_classes" "collect_plots"
#> [4] "collect_stats" "colnames" "compare_signatures"
#> [7] "consensus_heatmap" "dimension_reduction" "functional_enrichment"
#> [10] "get_anno_col" "get_anno" "get_classes"
#> [13] "get_consensus" "get_matrix" "get_membership"
#> [16] "get_param" "get_signatures" "get_stats"
#> [19] "is_best_k" "is_stable_k" "membership_heatmap"
#> [22] "ncol" "nrow" "plot_ecdf"
#> [25] "rownames" "select_partition_number" "show"
#> [28] "suggest_best_k" "test_to_known_factors"
collect_plots() function collects all the plots made from res for all k (number of partitions)
into one single page to provide an easy and fast comparison between different k.
collect_plots(res)
The plots are:
k and the heatmap of
predicted classes for each k.k.k.k.All the plots in panels can be made by individual functions and they are plotted later in this section.
select_partition_number() produces several plots showing different
statistics for choosing “optimized” k. There are following statistics:
k;k, the area increased is defined as \(A_k - A_{k-1}\).The detailed explanations of these statistics can be found in the cola vignette.
Generally speaking, lower PAC score, higher mean silhouette score or higher
concordance corresponds to better partition. Rand index and Jaccard index
measure how similar the current partition is compared to partition with k-1.
If they are too similar, we won't accept k is better than k-1.
select_partition_number(res)
The numeric values for all these statistics can be obtained by get_stats().
get_stats(res)
#> k 1-PAC mean_silhouette concordance area_increased Rand Jaccard
#> 2 2 1.000 0.986 0.994 0.5009 0.500 0.500
#> 3 3 0.837 0.942 0.966 0.3070 0.788 0.600
#> 4 4 0.768 0.695 0.876 0.1394 0.851 0.599
#> 5 5 0.735 0.748 0.841 0.0635 0.879 0.582
#> 6 6 0.733 0.668 0.766 0.0389 0.970 0.858
suggest_best_k() suggests the best \(k\) based on these statistics. The rules are as follows:
suggest_best_k(res)
#> [1] 2
Following shows the table of the partitions (You need to click the show/hide
code output link to see it). The membership matrix (columns with name p*)
is inferred by
clue::cl_consensus()
function with the SE method. Basically the value in the membership matrix
represents the probability to belong to a certain group. The finall class
label for an item is determined with the group with highest probability it
belongs to.
In get_classes() function, the entropy is calculated from the membership
matrix and the silhouette score is calculated from the consensus matrix.
cbind(get_classes(res, k = 2), get_membership(res, k = 2))
#> class entropy silhouette p1 p2
#> GSM97038 2 0.000 0.997 0.000 1.000
#> GSM97045 2 0.000 0.997 0.000 1.000
#> GSM97047 2 0.000 0.997 0.000 1.000
#> GSM97025 2 0.000 0.997 0.000 1.000
#> GSM97030 2 0.000 0.997 0.000 1.000
#> GSM97027 2 0.000 0.997 0.000 1.000
#> GSM97033 2 0.000 0.997 0.000 1.000
#> GSM97034 2 0.000 0.997 0.000 1.000
#> GSM97020 2 0.000 0.997 0.000 1.000
#> GSM97026 2 0.000 0.997 0.000 1.000
#> GSM97012 2 0.000 0.997 0.000 1.000
#> GSM97015 2 0.000 0.997 0.000 1.000
#> GSM97016 2 0.000 0.997 0.000 1.000
#> GSM97017 1 0.000 0.992 1.000 0.000
#> GSM97019 2 0.000 0.997 0.000 1.000
#> GSM97022 2 0.000 0.997 0.000 1.000
#> GSM97035 2 0.000 0.997 0.000 1.000
#> GSM97036 1 0.000 0.992 1.000 0.000
#> GSM97039 2 0.000 0.997 0.000 1.000
#> GSM97046 2 0.000 0.997 0.000 1.000
#> GSM97023 1 0.000 0.992 1.000 0.000
#> GSM97029 1 0.000 0.992 1.000 0.000
#> GSM97043 2 0.000 0.997 0.000 1.000
#> GSM97013 1 0.000 0.992 1.000 0.000
#> GSM96956 2 0.000 0.997 0.000 1.000
#> GSM97024 2 0.000 0.997 0.000 1.000
#> GSM97032 2 0.000 0.997 0.000 1.000
#> GSM97044 2 0.000 0.997 0.000 1.000
#> GSM97049 2 0.000 0.997 0.000 1.000
#> GSM96968 1 0.961 0.371 0.616 0.384
#> GSM96971 1 0.000 0.992 1.000 0.000
#> GSM96986 1 0.000 0.992 1.000 0.000
#> GSM97003 1 0.000 0.992 1.000 0.000
#> GSM96957 1 0.000 0.992 1.000 0.000
#> GSM96960 1 0.000 0.992 1.000 0.000
#> GSM96975 1 0.000 0.992 1.000 0.000
#> GSM96998 1 0.000 0.992 1.000 0.000
#> GSM96999 1 0.000 0.992 1.000 0.000
#> GSM97001 1 0.000 0.992 1.000 0.000
#> GSM97005 1 0.000 0.992 1.000 0.000
#> GSM97006 1 0.000 0.992 1.000 0.000
#> GSM97021 1 0.000 0.992 1.000 0.000
#> GSM97028 2 0.000 0.997 0.000 1.000
#> GSM97031 1 0.000 0.992 1.000 0.000
#> GSM97037 2 0.000 0.997 0.000 1.000
#> GSM97018 2 0.000 0.997 0.000 1.000
#> GSM97014 2 0.000 0.997 0.000 1.000
#> GSM97042 2 0.000 0.997 0.000 1.000
#> GSM97040 2 0.000 0.997 0.000 1.000
#> GSM97041 1 0.000 0.992 1.000 0.000
#> GSM96955 2 0.000 0.997 0.000 1.000
#> GSM96990 2 0.000 0.997 0.000 1.000
#> GSM96991 2 0.000 0.997 0.000 1.000
#> GSM97048 2 0.000 0.997 0.000 1.000
#> GSM96963 2 0.000 0.997 0.000 1.000
#> GSM96953 2 0.000 0.997 0.000 1.000
#> GSM96966 1 0.000 0.992 1.000 0.000
#> GSM96979 1 0.000 0.992 1.000 0.000
#> GSM96983 2 0.000 0.997 0.000 1.000
#> GSM96984 2 0.482 0.885 0.104 0.896
#> GSM96994 2 0.000 0.997 0.000 1.000
#> GSM96996 1 0.000 0.992 1.000 0.000
#> GSM96997 1 0.000 0.992 1.000 0.000
#> GSM97007 2 0.000 0.997 0.000 1.000
#> GSM96954 1 0.000 0.992 1.000 0.000
#> GSM96962 1 0.000 0.992 1.000 0.000
#> GSM96969 1 0.000 0.992 1.000 0.000
#> GSM96970 1 0.000 0.992 1.000 0.000
#> GSM96973 1 0.000 0.992 1.000 0.000
#> GSM96976 2 0.000 0.997 0.000 1.000
#> GSM96977 1 0.000 0.992 1.000 0.000
#> GSM96995 2 0.000 0.997 0.000 1.000
#> GSM97002 1 0.000 0.992 1.000 0.000
#> GSM97009 2 0.000 0.997 0.000 1.000
#> GSM97010 1 0.000 0.992 1.000 0.000
#> GSM96974 1 0.000 0.992 1.000 0.000
#> GSM96985 1 0.000 0.992 1.000 0.000
#> GSM96959 2 0.000 0.997 0.000 1.000
#> GSM96972 1 0.000 0.992 1.000 0.000
#> GSM96978 2 0.278 0.949 0.048 0.952
#> GSM96967 1 0.000 0.992 1.000 0.000
#> GSM96987 1 0.000 0.992 1.000 0.000
#> GSM97011 1 0.141 0.973 0.980 0.020
#> GSM96964 1 0.000 0.992 1.000 0.000
#> GSM96965 1 0.000 0.992 1.000 0.000
#> GSM96981 1 0.000 0.992 1.000 0.000
#> GSM96982 1 0.000 0.992 1.000 0.000
#> GSM96988 1 0.000 0.992 1.000 0.000
#> GSM97000 1 0.000 0.992 1.000 0.000
#> GSM97004 1 0.000 0.992 1.000 0.000
#> GSM97008 1 0.000 0.992 1.000 0.000
#> GSM96950 1 0.000 0.992 1.000 0.000
#> GSM96980 1 0.000 0.992 1.000 0.000
#> GSM96989 1 0.000 0.992 1.000 0.000
#> GSM96992 1 0.000 0.992 1.000 0.000
#> GSM96993 1 0.000 0.992 1.000 0.000
#> GSM96958 1 0.000 0.992 1.000 0.000
#> GSM96951 1 0.000 0.992 1.000 0.000
#> GSM96952 1 0.000 0.992 1.000 0.000
#> GSM96961 1 0.000 0.992 1.000 0.000
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM97038 2 0.0000 0.990 0.000 1.000 0.000
#> GSM97045 2 0.0000 0.990 0.000 1.000 0.000
#> GSM97047 2 0.0000 0.990 0.000 1.000 0.000
#> GSM97025 2 0.0000 0.990 0.000 1.000 0.000
#> GSM97030 3 0.3551 0.873 0.000 0.132 0.868
#> GSM97027 2 0.0000 0.990 0.000 1.000 0.000
#> GSM97033 2 0.0000 0.990 0.000 1.000 0.000
#> GSM97034 3 0.3412 0.879 0.000 0.124 0.876
#> GSM97020 2 0.0000 0.990 0.000 1.000 0.000
#> GSM97026 2 0.0000 0.990 0.000 1.000 0.000
#> GSM97012 2 0.0000 0.990 0.000 1.000 0.000
#> GSM97015 3 0.3412 0.879 0.000 0.124 0.876
#> GSM97016 2 0.0000 0.990 0.000 1.000 0.000
#> GSM97017 1 0.0237 0.961 0.996 0.004 0.000
#> GSM97019 2 0.0000 0.990 0.000 1.000 0.000
#> GSM97022 2 0.0000 0.990 0.000 1.000 0.000
#> GSM97035 2 0.0000 0.990 0.000 1.000 0.000
#> GSM97036 1 0.0237 0.961 0.996 0.004 0.000
#> GSM97039 2 0.0000 0.990 0.000 1.000 0.000
#> GSM97046 2 0.0000 0.990 0.000 1.000 0.000
#> GSM97023 1 0.0000 0.962 1.000 0.000 0.000
#> GSM97029 1 0.0237 0.961 0.996 0.004 0.000
#> GSM97043 2 0.0000 0.990 0.000 1.000 0.000
#> GSM97013 1 0.0000 0.962 1.000 0.000 0.000
#> GSM96956 2 0.1529 0.953 0.000 0.960 0.040
#> GSM97024 2 0.0000 0.990 0.000 1.000 0.000
#> GSM97032 3 0.4750 0.781 0.000 0.216 0.784
#> GSM97044 3 0.3412 0.879 0.000 0.124 0.876
#> GSM97049 2 0.0000 0.990 0.000 1.000 0.000
#> GSM96968 3 0.0237 0.932 0.004 0.000 0.996
#> GSM96971 3 0.0000 0.933 0.000 0.000 1.000
#> GSM96986 3 0.0000 0.933 0.000 0.000 1.000
#> GSM97003 1 0.0892 0.958 0.980 0.000 0.020
#> GSM96957 1 0.0000 0.962 1.000 0.000 0.000
#> GSM96960 1 0.0892 0.958 0.980 0.000 0.020
#> GSM96975 1 0.0000 0.962 1.000 0.000 0.000
#> GSM96998 1 0.0000 0.962 1.000 0.000 0.000
#> GSM96999 1 0.0000 0.962 1.000 0.000 0.000
#> GSM97001 1 0.0000 0.962 1.000 0.000 0.000
#> GSM97005 1 0.0000 0.962 1.000 0.000 0.000
#> GSM97006 1 0.0892 0.958 0.980 0.000 0.020
#> GSM97021 1 0.0424 0.960 0.992 0.000 0.008
#> GSM97028 3 0.0592 0.931 0.000 0.012 0.988
#> GSM97031 1 0.1163 0.951 0.972 0.000 0.028
#> GSM97037 2 0.2878 0.888 0.000 0.904 0.096
#> GSM97018 3 0.4121 0.840 0.000 0.168 0.832
#> GSM97014 2 0.0000 0.990 0.000 1.000 0.000
#> GSM97042 2 0.0000 0.990 0.000 1.000 0.000
#> GSM97040 2 0.2269 0.938 0.040 0.944 0.016
#> GSM97041 1 0.0237 0.961 0.996 0.004 0.000
#> GSM96955 2 0.0000 0.990 0.000 1.000 0.000
#> GSM96990 3 0.3752 0.863 0.000 0.144 0.856
#> GSM96991 2 0.0000 0.990 0.000 1.000 0.000
#> GSM97048 2 0.0000 0.990 0.000 1.000 0.000
#> GSM96963 2 0.0000 0.990 0.000 1.000 0.000
#> GSM96953 2 0.0000 0.990 0.000 1.000 0.000
#> GSM96966 1 0.3412 0.891 0.876 0.000 0.124
#> GSM96979 3 0.0000 0.933 0.000 0.000 1.000
#> GSM96983 3 0.0237 0.933 0.000 0.004 0.996
#> GSM96984 3 0.0000 0.933 0.000 0.000 1.000
#> GSM96994 3 0.0000 0.933 0.000 0.000 1.000
#> GSM96996 1 0.0892 0.958 0.980 0.000 0.020
#> GSM96997 3 0.0000 0.933 0.000 0.000 1.000
#> GSM97007 3 0.0000 0.933 0.000 0.000 1.000
#> GSM96954 3 0.0000 0.933 0.000 0.000 1.000
#> GSM96962 3 0.0000 0.933 0.000 0.000 1.000
#> GSM96969 1 0.3412 0.891 0.876 0.000 0.124
#> GSM96970 1 0.3412 0.891 0.876 0.000 0.124
#> GSM96973 1 0.3412 0.891 0.876 0.000 0.124
#> GSM96976 3 0.5058 0.691 0.000 0.244 0.756
#> GSM96977 1 0.4555 0.752 0.800 0.000 0.200
#> GSM96995 3 0.2959 0.893 0.000 0.100 0.900
#> GSM97002 1 0.0892 0.958 0.980 0.000 0.020
#> GSM97009 2 0.0000 0.990 0.000 1.000 0.000
#> GSM97010 1 0.2959 0.909 0.900 0.000 0.100
#> GSM96974 3 0.1289 0.913 0.032 0.000 0.968
#> GSM96985 3 0.1860 0.897 0.052 0.000 0.948
#> GSM96959 2 0.2165 0.929 0.000 0.936 0.064
#> GSM96972 1 0.3412 0.891 0.876 0.000 0.124
#> GSM96978 3 0.0000 0.933 0.000 0.000 1.000
#> GSM96967 1 0.3412 0.891 0.876 0.000 0.124
#> GSM96987 1 0.0000 0.962 1.000 0.000 0.000
#> GSM97011 1 0.3816 0.824 0.852 0.148 0.000
#> GSM96964 1 0.0000 0.962 1.000 0.000 0.000
#> GSM96965 1 0.3340 0.894 0.880 0.000 0.120
#> GSM96981 1 0.0000 0.962 1.000 0.000 0.000
#> GSM96982 1 0.0592 0.960 0.988 0.000 0.012
#> GSM96988 3 0.0000 0.933 0.000 0.000 1.000
#> GSM97000 1 0.3412 0.862 0.876 0.000 0.124
#> GSM97004 1 0.0892 0.958 0.980 0.000 0.020
#> GSM97008 1 0.1031 0.951 0.976 0.000 0.024
#> GSM96950 1 0.0000 0.962 1.000 0.000 0.000
#> GSM96980 1 0.0892 0.958 0.980 0.000 0.020
#> GSM96989 1 0.0000 0.962 1.000 0.000 0.000
#> GSM96992 1 0.0000 0.962 1.000 0.000 0.000
#> GSM96993 1 0.0000 0.962 1.000 0.000 0.000
#> GSM96958 1 0.0000 0.962 1.000 0.000 0.000
#> GSM96951 1 0.0000 0.962 1.000 0.000 0.000
#> GSM96952 1 0.0000 0.962 1.000 0.000 0.000
#> GSM96961 1 0.0000 0.962 1.000 0.000 0.000
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM97038 2 0.0000 0.94081 0.000 1.000 0.000 0.000
#> GSM97045 2 0.0000 0.94081 0.000 1.000 0.000 0.000
#> GSM97047 2 0.4817 0.50390 0.388 0.612 0.000 0.000
#> GSM97025 2 0.0000 0.94081 0.000 1.000 0.000 0.000
#> GSM97030 3 0.0188 0.97176 0.004 0.000 0.996 0.000
#> GSM97027 2 0.0000 0.94081 0.000 1.000 0.000 0.000
#> GSM97033 2 0.0000 0.94081 0.000 1.000 0.000 0.000
#> GSM97034 3 0.0376 0.97033 0.004 0.004 0.992 0.000
#> GSM97020 2 0.0000 0.94081 0.000 1.000 0.000 0.000
#> GSM97026 2 0.0921 0.92254 0.028 0.972 0.000 0.000
#> GSM97012 2 0.0000 0.94081 0.000 1.000 0.000 0.000
#> GSM97015 3 0.0188 0.97176 0.004 0.000 0.996 0.000
#> GSM97016 2 0.0000 0.94081 0.000 1.000 0.000 0.000
#> GSM97017 1 0.0188 0.68936 0.996 0.000 0.000 0.004
#> GSM97019 2 0.0000 0.94081 0.000 1.000 0.000 0.000
#> GSM97022 2 0.0000 0.94081 0.000 1.000 0.000 0.000
#> GSM97035 2 0.0000 0.94081 0.000 1.000 0.000 0.000
#> GSM97036 1 0.5000 0.02944 0.504 0.000 0.000 0.496
#> GSM97039 2 0.0000 0.94081 0.000 1.000 0.000 0.000
#> GSM97046 2 0.0000 0.94081 0.000 1.000 0.000 0.000
#> GSM97023 1 0.4907 0.28883 0.580 0.000 0.000 0.420
#> GSM97029 1 0.3726 0.57928 0.788 0.000 0.000 0.212
#> GSM97043 2 0.0000 0.94081 0.000 1.000 0.000 0.000
#> GSM97013 1 0.4746 0.37781 0.632 0.000 0.000 0.368
#> GSM96956 2 0.2921 0.81276 0.000 0.860 0.140 0.000
#> GSM97024 2 0.0000 0.94081 0.000 1.000 0.000 0.000
#> GSM97032 3 0.3257 0.82510 0.004 0.152 0.844 0.000
#> GSM97044 3 0.0188 0.97176 0.004 0.000 0.996 0.000
#> GSM97049 2 0.0000 0.94081 0.000 1.000 0.000 0.000
#> GSM96968 3 0.0188 0.97176 0.004 0.000 0.996 0.000
#> GSM96971 3 0.2081 0.91142 0.000 0.000 0.916 0.084
#> GSM96986 3 0.0000 0.97226 0.000 0.000 1.000 0.000
#> GSM97003 4 0.2412 0.69319 0.084 0.000 0.008 0.908
#> GSM96957 1 0.0592 0.68959 0.984 0.000 0.000 0.016
#> GSM96960 4 0.2149 0.69218 0.088 0.000 0.000 0.912
#> GSM96975 1 0.4277 0.50530 0.720 0.000 0.000 0.280
#> GSM96998 4 0.4431 0.44127 0.304 0.000 0.000 0.696
#> GSM96999 1 0.4981 0.13744 0.536 0.000 0.000 0.464
#> GSM97001 1 0.0188 0.68936 0.996 0.000 0.000 0.004
#> GSM97005 1 0.0592 0.68956 0.984 0.000 0.000 0.016
#> GSM97006 4 0.2469 0.67914 0.108 0.000 0.000 0.892
#> GSM97021 1 0.0469 0.68989 0.988 0.000 0.000 0.012
#> GSM97028 3 0.0000 0.97226 0.000 0.000 1.000 0.000
#> GSM97031 1 0.4936 0.44050 0.652 0.000 0.008 0.340
#> GSM97037 2 0.4220 0.65561 0.004 0.748 0.248 0.000
#> GSM97018 3 0.2714 0.87437 0.004 0.112 0.884 0.000
#> GSM97014 2 0.5000 0.28772 0.496 0.504 0.000 0.000
#> GSM97042 2 0.0000 0.94081 0.000 1.000 0.000 0.000
#> GSM97040 1 0.0921 0.66997 0.972 0.028 0.000 0.000
#> GSM97041 1 0.0188 0.68936 0.996 0.000 0.000 0.004
#> GSM96955 2 0.1389 0.90752 0.048 0.952 0.000 0.000
#> GSM96990 3 0.1004 0.95641 0.004 0.024 0.972 0.000
#> GSM96991 2 0.0000 0.94081 0.000 1.000 0.000 0.000
#> GSM97048 2 0.0000 0.94081 0.000 1.000 0.000 0.000
#> GSM96963 2 0.0000 0.94081 0.000 1.000 0.000 0.000
#> GSM96953 2 0.0000 0.94081 0.000 1.000 0.000 0.000
#> GSM96966 4 0.0000 0.70249 0.000 0.000 0.000 1.000
#> GSM96979 3 0.2216 0.89869 0.000 0.000 0.908 0.092
#> GSM96983 3 0.0000 0.97226 0.000 0.000 1.000 0.000
#> GSM96984 3 0.0000 0.97226 0.000 0.000 1.000 0.000
#> GSM96994 3 0.0000 0.97226 0.000 0.000 1.000 0.000
#> GSM96996 4 0.1867 0.69881 0.072 0.000 0.000 0.928
#> GSM96997 3 0.0188 0.97065 0.000 0.000 0.996 0.004
#> GSM97007 3 0.0000 0.97226 0.000 0.000 1.000 0.000
#> GSM96954 3 0.0188 0.97176 0.004 0.000 0.996 0.000
#> GSM96962 3 0.0000 0.97226 0.000 0.000 1.000 0.000
#> GSM96969 4 0.0000 0.70249 0.000 0.000 0.000 1.000
#> GSM96970 4 0.0000 0.70249 0.000 0.000 0.000 1.000
#> GSM96973 4 0.0000 0.70249 0.000 0.000 0.000 1.000
#> GSM96976 4 0.7211 0.24593 0.000 0.248 0.204 0.548
#> GSM96977 1 0.3037 0.64411 0.880 0.000 0.020 0.100
#> GSM96995 3 0.1474 0.93743 0.052 0.000 0.948 0.000
#> GSM97002 4 0.1940 0.69753 0.076 0.000 0.000 0.924
#> GSM97009 2 0.4072 0.70086 0.252 0.748 0.000 0.000
#> GSM97010 4 0.0469 0.69912 0.012 0.000 0.000 0.988
#> GSM96974 4 0.4679 0.26588 0.000 0.000 0.352 0.648
#> GSM96985 4 0.4072 0.45574 0.000 0.000 0.252 0.748
#> GSM96959 1 0.7771 -0.04566 0.424 0.256 0.320 0.000
#> GSM96972 4 0.0000 0.70249 0.000 0.000 0.000 1.000
#> GSM96978 3 0.0000 0.97226 0.000 0.000 1.000 0.000
#> GSM96967 4 0.0000 0.70249 0.000 0.000 0.000 1.000
#> GSM96987 4 0.4941 0.14751 0.436 0.000 0.000 0.564
#> GSM97011 1 0.0657 0.68803 0.984 0.004 0.000 0.012
#> GSM96964 4 0.5000 -0.07726 0.500 0.000 0.000 0.500
#> GSM96965 4 0.1211 0.68494 0.040 0.000 0.000 0.960
#> GSM96981 4 0.4331 0.47664 0.288 0.000 0.000 0.712
#> GSM96982 4 0.1867 0.69939 0.072 0.000 0.000 0.928
#> GSM96988 3 0.0188 0.97098 0.000 0.000 0.996 0.004
#> GSM97000 1 0.1388 0.67324 0.960 0.000 0.028 0.012
#> GSM97004 4 0.1557 0.70182 0.056 0.000 0.000 0.944
#> GSM97008 1 0.0657 0.68782 0.984 0.000 0.004 0.012
#> GSM96950 1 0.4907 0.26720 0.580 0.000 0.000 0.420
#> GSM96980 4 0.0000 0.70249 0.000 0.000 0.000 1.000
#> GSM96989 4 0.4941 0.14751 0.436 0.000 0.000 0.564
#> GSM96992 4 0.4916 0.16465 0.424 0.000 0.000 0.576
#> GSM96993 1 0.4877 0.29485 0.592 0.000 0.000 0.408
#> GSM96958 4 0.4999 -0.06756 0.492 0.000 0.000 0.508
#> GSM96951 1 0.4994 0.10456 0.520 0.000 0.000 0.480
#> GSM96952 4 0.4925 0.15303 0.428 0.000 0.000 0.572
#> GSM96961 4 0.4989 0.00392 0.472 0.000 0.000 0.528
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM97038 2 0.0566 0.9477 0.000 0.984 0.000 0.012 0.004
#> GSM97045 2 0.0000 0.9507 0.000 1.000 0.000 0.000 0.000
#> GSM97047 5 0.4329 0.6743 0.000 0.224 0.028 0.008 0.740
#> GSM97025 2 0.0000 0.9507 0.000 1.000 0.000 0.000 0.000
#> GSM97030 3 0.1461 0.8625 0.000 0.028 0.952 0.004 0.016
#> GSM97027 2 0.0000 0.9507 0.000 1.000 0.000 0.000 0.000
#> GSM97033 2 0.0451 0.9494 0.000 0.988 0.000 0.008 0.004
#> GSM97034 3 0.1988 0.8529 0.000 0.048 0.928 0.008 0.016
#> GSM97020 2 0.0451 0.9494 0.000 0.988 0.000 0.008 0.004
#> GSM97026 2 0.4192 0.7986 0.040 0.828 0.052 0.012 0.068
#> GSM97012 2 0.0000 0.9507 0.000 1.000 0.000 0.000 0.000
#> GSM97015 3 0.1306 0.8648 0.000 0.016 0.960 0.008 0.016
#> GSM97016 2 0.0451 0.9494 0.000 0.988 0.000 0.008 0.004
#> GSM97017 5 0.3123 0.7530 0.184 0.000 0.000 0.004 0.812
#> GSM97019 2 0.0000 0.9507 0.000 1.000 0.000 0.000 0.000
#> GSM97022 2 0.0000 0.9507 0.000 1.000 0.000 0.000 0.000
#> GSM97035 2 0.0000 0.9507 0.000 1.000 0.000 0.000 0.000
#> GSM97036 1 0.2570 0.6984 0.888 0.000 0.000 0.028 0.084
#> GSM97039 2 0.0451 0.9494 0.000 0.988 0.000 0.008 0.004
#> GSM97046 2 0.0451 0.9494 0.000 0.988 0.000 0.008 0.004
#> GSM97023 1 0.3056 0.7258 0.864 0.000 0.000 0.068 0.068
#> GSM97029 1 0.3999 0.5665 0.740 0.000 0.000 0.020 0.240
#> GSM97043 2 0.0854 0.9360 0.000 0.976 0.012 0.004 0.008
#> GSM97013 1 0.2798 0.6659 0.852 0.000 0.000 0.008 0.140
#> GSM96956 2 0.4048 0.7079 0.000 0.764 0.208 0.012 0.016
#> GSM97024 2 0.0162 0.9492 0.000 0.996 0.000 0.000 0.004
#> GSM97032 3 0.3734 0.7198 0.000 0.184 0.792 0.008 0.016
#> GSM97044 3 0.0833 0.8695 0.000 0.004 0.976 0.004 0.016
#> GSM97049 2 0.0451 0.9494 0.000 0.988 0.000 0.008 0.004
#> GSM96968 3 0.1393 0.8664 0.008 0.000 0.956 0.012 0.024
#> GSM96971 3 0.4818 0.3849 0.000 0.000 0.520 0.460 0.020
#> GSM96986 3 0.3445 0.8530 0.000 0.000 0.824 0.140 0.036
#> GSM97003 1 0.5704 0.5039 0.592 0.000 0.016 0.328 0.064
#> GSM96957 1 0.4450 -0.0617 0.508 0.000 0.000 0.004 0.488
#> GSM96960 1 0.4594 0.5527 0.680 0.000 0.000 0.284 0.036
#> GSM96975 1 0.6500 0.2050 0.412 0.000 0.000 0.188 0.400
#> GSM96998 1 0.1410 0.7123 0.940 0.000 0.000 0.060 0.000
#> GSM96999 1 0.3495 0.6922 0.812 0.000 0.000 0.028 0.160
#> GSM97001 5 0.2732 0.7542 0.160 0.000 0.000 0.000 0.840
#> GSM97005 5 0.2144 0.7977 0.068 0.000 0.000 0.020 0.912
#> GSM97006 1 0.4644 0.5565 0.680 0.000 0.000 0.280 0.040
#> GSM97021 5 0.2361 0.7957 0.096 0.000 0.000 0.012 0.892
#> GSM97028 3 0.0798 0.8735 0.000 0.000 0.976 0.016 0.008
#> GSM97031 1 0.6947 0.4540 0.488 0.000 0.032 0.160 0.320
#> GSM97037 2 0.5061 0.3484 0.000 0.580 0.388 0.012 0.020
#> GSM97018 3 0.3731 0.7323 0.000 0.172 0.800 0.012 0.016
#> GSM97014 5 0.3421 0.7045 0.000 0.204 0.000 0.008 0.788
#> GSM97042 2 0.0000 0.9507 0.000 1.000 0.000 0.000 0.000
#> GSM97040 5 0.1605 0.8035 0.040 0.004 0.012 0.000 0.944
#> GSM97041 5 0.3333 0.7305 0.208 0.000 0.000 0.004 0.788
#> GSM96955 2 0.2771 0.8234 0.000 0.860 0.000 0.012 0.128
#> GSM96990 3 0.1949 0.8565 0.000 0.040 0.932 0.012 0.016
#> GSM96991 2 0.0000 0.9507 0.000 1.000 0.000 0.000 0.000
#> GSM97048 2 0.0451 0.9494 0.000 0.988 0.000 0.008 0.004
#> GSM96963 2 0.0162 0.9504 0.000 0.996 0.000 0.000 0.004
#> GSM96953 2 0.0162 0.9504 0.000 0.996 0.000 0.000 0.004
#> GSM96966 4 0.2648 0.8311 0.152 0.000 0.000 0.848 0.000
#> GSM96979 3 0.4350 0.7456 0.000 0.000 0.704 0.268 0.028
#> GSM96983 3 0.0955 0.8756 0.000 0.000 0.968 0.028 0.004
#> GSM96984 3 0.3099 0.8615 0.000 0.000 0.848 0.124 0.028
#> GSM96994 3 0.3051 0.8626 0.000 0.000 0.852 0.120 0.028
#> GSM96996 1 0.4152 0.5385 0.692 0.000 0.000 0.296 0.012
#> GSM96997 3 0.3799 0.8452 0.012 0.000 0.812 0.144 0.032
#> GSM97007 3 0.3051 0.8626 0.000 0.000 0.852 0.120 0.028
#> GSM96954 3 0.2209 0.8751 0.000 0.000 0.912 0.056 0.032
#> GSM96962 3 0.3146 0.8601 0.000 0.000 0.844 0.128 0.028
#> GSM96969 4 0.2690 0.8282 0.156 0.000 0.000 0.844 0.000
#> GSM96970 4 0.2648 0.8311 0.152 0.000 0.000 0.848 0.000
#> GSM96973 4 0.2648 0.8311 0.152 0.000 0.000 0.848 0.000
#> GSM96976 4 0.3315 0.6684 0.000 0.084 0.052 0.856 0.008
#> GSM96977 5 0.6569 0.4804 0.264 0.000 0.072 0.080 0.584
#> GSM96995 3 0.2522 0.8187 0.000 0.000 0.880 0.012 0.108
#> GSM97002 1 0.4135 0.4804 0.656 0.000 0.000 0.340 0.004
#> GSM97009 5 0.4528 0.2328 0.000 0.444 0.000 0.008 0.548
#> GSM97010 4 0.5052 0.5442 0.340 0.000 0.000 0.612 0.048
#> GSM96974 4 0.3161 0.6991 0.032 0.000 0.100 0.860 0.008
#> GSM96985 4 0.4471 0.7136 0.088 0.000 0.132 0.772 0.008
#> GSM96959 5 0.4403 0.6671 0.000 0.036 0.188 0.016 0.760
#> GSM96972 4 0.2648 0.8311 0.152 0.000 0.000 0.848 0.000
#> GSM96978 3 0.2818 0.8634 0.000 0.000 0.856 0.132 0.012
#> GSM96967 4 0.2648 0.8311 0.152 0.000 0.000 0.848 0.000
#> GSM96987 1 0.0771 0.7139 0.976 0.000 0.000 0.004 0.020
#> GSM97011 5 0.1408 0.8034 0.044 0.000 0.000 0.008 0.948
#> GSM96964 1 0.1043 0.7137 0.960 0.000 0.000 0.000 0.040
#> GSM96965 4 0.3760 0.7841 0.188 0.000 0.000 0.784 0.028
#> GSM96981 1 0.5552 0.4261 0.584 0.000 0.000 0.328 0.088
#> GSM96982 1 0.5019 0.2253 0.532 0.000 0.000 0.436 0.032
#> GSM96988 3 0.2193 0.8721 0.000 0.000 0.900 0.092 0.008
#> GSM97000 5 0.1701 0.7971 0.028 0.000 0.012 0.016 0.944
#> GSM97004 1 0.3999 0.4748 0.656 0.000 0.000 0.344 0.000
#> GSM97008 5 0.1740 0.8011 0.056 0.000 0.000 0.012 0.932
#> GSM96950 1 0.2358 0.6892 0.888 0.000 0.000 0.008 0.104
#> GSM96980 4 0.4074 0.4344 0.364 0.000 0.000 0.636 0.000
#> GSM96989 1 0.0771 0.7139 0.976 0.000 0.000 0.004 0.020
#> GSM96992 1 0.3875 0.6789 0.792 0.000 0.000 0.160 0.048
#> GSM96993 1 0.2645 0.6863 0.884 0.000 0.008 0.012 0.096
#> GSM96958 1 0.3192 0.7191 0.848 0.000 0.000 0.040 0.112
#> GSM96951 1 0.3806 0.7177 0.812 0.000 0.000 0.084 0.104
#> GSM96952 1 0.3365 0.7000 0.836 0.000 0.000 0.120 0.044
#> GSM96961 1 0.2782 0.7190 0.880 0.000 0.000 0.072 0.048
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM97038 2 0.2076 0.9123 0.000 0.912 0.012 0.000 0.016 0.060
#> GSM97045 2 0.0363 0.9280 0.000 0.988 0.000 0.000 0.000 0.012
#> GSM97047 5 0.3550 0.7375 0.000 0.132 0.032 0.000 0.812 0.024
#> GSM97025 2 0.0405 0.9286 0.000 0.988 0.004 0.000 0.000 0.008
#> GSM97030 3 0.1092 0.6396 0.000 0.020 0.960 0.000 0.000 0.020
#> GSM97027 2 0.0458 0.9281 0.000 0.984 0.000 0.000 0.000 0.016
#> GSM97033 2 0.1672 0.9197 0.000 0.932 0.016 0.000 0.004 0.048
#> GSM97034 3 0.2588 0.6276 0.000 0.060 0.888 0.008 0.004 0.040
#> GSM97020 2 0.1923 0.9148 0.000 0.916 0.016 0.000 0.004 0.064
#> GSM97026 2 0.6184 0.5834 0.032 0.660 0.128 0.008 0.084 0.088
#> GSM97012 2 0.0260 0.9281 0.000 0.992 0.008 0.000 0.000 0.000
#> GSM97015 3 0.0993 0.6330 0.000 0.012 0.964 0.000 0.000 0.024
#> GSM97016 2 0.1863 0.9156 0.000 0.920 0.016 0.000 0.004 0.060
#> GSM97017 5 0.4033 0.7210 0.156 0.000 0.000 0.004 0.760 0.080
#> GSM97019 2 0.0405 0.9271 0.000 0.988 0.008 0.000 0.000 0.004
#> GSM97022 2 0.0405 0.9271 0.000 0.988 0.008 0.000 0.000 0.004
#> GSM97035 2 0.0520 0.9283 0.000 0.984 0.008 0.000 0.000 0.008
#> GSM97036 1 0.4616 0.6068 0.736 0.008 0.000 0.036 0.044 0.176
#> GSM97039 2 0.1863 0.9156 0.000 0.920 0.016 0.000 0.004 0.060
#> GSM97046 2 0.1863 0.9156 0.000 0.920 0.016 0.000 0.004 0.060
#> GSM97023 1 0.3796 0.6825 0.812 0.000 0.000 0.092 0.048 0.048
#> GSM97029 1 0.5416 0.4730 0.636 0.004 0.000 0.012 0.160 0.188
#> GSM97043 2 0.1124 0.9133 0.000 0.956 0.036 0.000 0.000 0.008
#> GSM97013 1 0.4498 0.5698 0.720 0.000 0.000 0.012 0.080 0.188
#> GSM96956 2 0.4756 0.5304 0.000 0.628 0.304 0.000 0.004 0.064
#> GSM97024 2 0.0858 0.9207 0.000 0.968 0.028 0.000 0.000 0.004
#> GSM97032 3 0.3168 0.5425 0.000 0.172 0.804 0.000 0.000 0.024
#> GSM97044 3 0.1411 0.6042 0.000 0.004 0.936 0.000 0.000 0.060
#> GSM97049 2 0.1923 0.9148 0.000 0.916 0.016 0.000 0.004 0.064
#> GSM96968 3 0.3158 0.5319 0.000 0.000 0.812 0.004 0.020 0.164
#> GSM96971 6 0.5894 0.4475 0.000 0.000 0.244 0.284 0.000 0.472
#> GSM96986 6 0.4045 0.8457 0.000 0.000 0.428 0.000 0.008 0.564
#> GSM97003 1 0.6096 0.4103 0.488 0.000 0.004 0.196 0.008 0.304
#> GSM96957 1 0.5962 0.2102 0.524 0.000 0.004 0.012 0.300 0.160
#> GSM96960 1 0.4758 0.5879 0.660 0.000 0.000 0.260 0.008 0.072
#> GSM96975 1 0.7478 0.2872 0.352 0.000 0.000 0.220 0.280 0.148
#> GSM96998 1 0.3526 0.6777 0.820 0.000 0.000 0.088 0.012 0.080
#> GSM96999 1 0.4640 0.6398 0.744 0.000 0.000 0.044 0.092 0.120
#> GSM97001 5 0.4002 0.7331 0.136 0.000 0.000 0.012 0.776 0.076
#> GSM97005 5 0.1777 0.8038 0.044 0.000 0.000 0.004 0.928 0.024
#> GSM97006 1 0.4815 0.6045 0.668 0.000 0.000 0.236 0.008 0.088
#> GSM97021 5 0.2706 0.7926 0.060 0.000 0.000 0.008 0.876 0.056
#> GSM97028 3 0.2631 0.5415 0.000 0.000 0.856 0.012 0.004 0.128
#> GSM97031 1 0.7731 0.3124 0.352 0.000 0.016 0.124 0.224 0.284
#> GSM97037 3 0.4766 0.3307 0.000 0.320 0.616 0.000 0.004 0.060
#> GSM97018 3 0.3412 0.5802 0.000 0.136 0.820 0.008 0.008 0.028
#> GSM97014 5 0.2818 0.7832 0.008 0.084 0.008 0.000 0.872 0.028
#> GSM97042 2 0.0405 0.9271 0.000 0.988 0.008 0.000 0.000 0.004
#> GSM97040 5 0.1180 0.8096 0.012 0.000 0.016 0.000 0.960 0.012
#> GSM97041 5 0.4664 0.6663 0.184 0.000 0.000 0.004 0.696 0.116
#> GSM96955 2 0.4205 0.7535 0.000 0.760 0.016 0.000 0.148 0.076
#> GSM96990 3 0.1864 0.6379 0.000 0.040 0.924 0.000 0.004 0.032
#> GSM96991 2 0.0405 0.9280 0.000 0.988 0.008 0.000 0.000 0.004
#> GSM97048 2 0.1923 0.9148 0.000 0.916 0.016 0.000 0.004 0.064
#> GSM96963 2 0.0146 0.9284 0.000 0.996 0.004 0.000 0.000 0.000
#> GSM96953 2 0.0622 0.9289 0.000 0.980 0.008 0.000 0.000 0.012
#> GSM96966 4 0.0790 0.8168 0.032 0.000 0.000 0.968 0.000 0.000
#> GSM96979 6 0.4829 0.7720 0.000 0.000 0.356 0.056 0.004 0.584
#> GSM96983 3 0.3213 0.3964 0.000 0.000 0.784 0.008 0.004 0.204
#> GSM96984 6 0.3828 0.8487 0.000 0.000 0.440 0.000 0.000 0.560
#> GSM96994 6 0.3843 0.8301 0.000 0.000 0.452 0.000 0.000 0.548
#> GSM96996 1 0.5076 0.5676 0.616 0.000 0.000 0.288 0.008 0.088
#> GSM96997 6 0.3890 0.8260 0.004 0.000 0.400 0.000 0.000 0.596
#> GSM97007 6 0.3838 0.8416 0.000 0.000 0.448 0.000 0.000 0.552
#> GSM96954 3 0.4206 -0.2906 0.000 0.000 0.620 0.000 0.024 0.356
#> GSM96962 6 0.3828 0.8487 0.000 0.000 0.440 0.000 0.000 0.560
#> GSM96969 4 0.0937 0.8128 0.040 0.000 0.000 0.960 0.000 0.000
#> GSM96970 4 0.0790 0.8168 0.032 0.000 0.000 0.968 0.000 0.000
#> GSM96973 4 0.0790 0.8168 0.032 0.000 0.000 0.968 0.000 0.000
#> GSM96976 4 0.3416 0.7028 0.000 0.032 0.008 0.816 0.004 0.140
#> GSM96977 5 0.7794 0.3613 0.208 0.000 0.116 0.056 0.452 0.168
#> GSM96995 3 0.3563 0.5364 0.000 0.000 0.796 0.000 0.132 0.072
#> GSM97002 1 0.4997 0.5686 0.628 0.000 0.000 0.280 0.008 0.084
#> GSM97009 5 0.5375 0.4186 0.000 0.316 0.004 0.004 0.572 0.104
#> GSM97010 4 0.6650 0.2449 0.300 0.000 0.008 0.472 0.040 0.180
#> GSM96974 4 0.2983 0.7062 0.000 0.000 0.032 0.832 0.000 0.136
#> GSM96985 4 0.5323 0.5827 0.040 0.000 0.092 0.656 0.000 0.212
#> GSM96959 5 0.4436 0.6584 0.000 0.012 0.180 0.000 0.728 0.080
#> GSM96972 4 0.0937 0.8128 0.040 0.000 0.000 0.960 0.000 0.000
#> GSM96978 3 0.4278 -0.1777 0.000 0.000 0.632 0.032 0.000 0.336
#> GSM96967 4 0.0790 0.8168 0.032 0.000 0.000 0.968 0.000 0.000
#> GSM96987 1 0.3028 0.6646 0.848 0.000 0.000 0.040 0.008 0.104
#> GSM97011 5 0.1515 0.8071 0.020 0.000 0.000 0.008 0.944 0.028
#> GSM96964 1 0.2890 0.6610 0.856 0.000 0.000 0.020 0.016 0.108
#> GSM96965 4 0.2095 0.7698 0.076 0.000 0.000 0.904 0.016 0.004
#> GSM96981 1 0.6136 0.4731 0.524 0.000 0.000 0.316 0.056 0.104
#> GSM96982 1 0.4940 0.3963 0.532 0.000 0.000 0.400 0.000 0.068
#> GSM96988 3 0.4237 0.0364 0.000 0.000 0.660 0.028 0.004 0.308
#> GSM97000 5 0.1526 0.8053 0.008 0.000 0.004 0.008 0.944 0.036
#> GSM97004 1 0.4653 0.5680 0.644 0.000 0.000 0.292 0.004 0.060
#> GSM97008 5 0.1518 0.8071 0.024 0.000 0.000 0.008 0.944 0.024
#> GSM96950 1 0.4267 0.5900 0.740 0.000 0.000 0.016 0.056 0.188
#> GSM96980 4 0.4285 0.2470 0.320 0.000 0.000 0.644 0.000 0.036
#> GSM96989 1 0.2986 0.6623 0.852 0.000 0.000 0.032 0.012 0.104
#> GSM96992 1 0.4176 0.6413 0.740 0.000 0.000 0.200 0.016 0.044
#> GSM96993 1 0.4114 0.6029 0.756 0.000 0.008 0.012 0.036 0.188
#> GSM96958 1 0.4946 0.6634 0.724 0.000 0.000 0.120 0.076 0.080
#> GSM96951 1 0.4922 0.6682 0.724 0.000 0.000 0.128 0.080 0.068
#> GSM96952 1 0.3667 0.6610 0.788 0.000 0.000 0.164 0.012 0.036
#> GSM96961 1 0.3101 0.6746 0.832 0.000 0.000 0.136 0.012 0.020
Heatmaps for the consensus matrix. It visualizes the probability of two samples to be in a same group.
consensus_heatmap(res, k = 2)
consensus_heatmap(res, k = 3)
consensus_heatmap(res, k = 4)
consensus_heatmap(res, k = 5)
consensus_heatmap(res, k = 6)
Heatmaps for the membership of samples in all partitions to see how consistent they are:
membership_heatmap(res, k = 2)
membership_heatmap(res, k = 3)
membership_heatmap(res, k = 4)
membership_heatmap(res, k = 5)
membership_heatmap(res, k = 6)
As soon as we have had the classes for columns, we can look for signatures which are significantly different between classes which can be candidate marks for certain classes. Following are the heatmaps for signatures.
Signature heatmaps where rows are scaled:
get_signatures(res, k = 2)
get_signatures(res, k = 3)
get_signatures(res, k = 4)
get_signatures(res, k = 5)
get_signatures(res, k = 6)
Signature heatmaps where rows are not scaled:
get_signatures(res, k = 2, scale_rows = FALSE)
get_signatures(res, k = 3, scale_rows = FALSE)
get_signatures(res, k = 4, scale_rows = FALSE)
get_signatures(res, k = 5, scale_rows = FALSE)
get_signatures(res, k = 6, scale_rows = FALSE)
Compare the overlap of signatures from different k:
compare_signatures(res)
get_signature() returns a data frame invisibly. TO get the list of signatures, the function
call should be assigned to a variable explicitly. In following code, if plot argument is set
to FALSE, no heatmap is plotted while only the differential analysis is performed.
# code only for demonstration
tb = get_signature(res, k = ..., plot = FALSE)
An example of the output of tb is:
#> which_row fdr mean_1 mean_2 scaled_mean_1 scaled_mean_2 km
#> 1 38 0.042760348 8.373488 9.131774 -0.5533452 0.5164555 1
#> 2 40 0.018707592 7.106213 8.469186 -0.6173731 0.5762149 1
#> 3 55 0.019134737 10.221463 11.207825 -0.6159697 0.5749050 1
#> 4 59 0.006059896 5.921854 7.869574 -0.6899429 0.6439467 1
#> 5 60 0.018055526 8.928898 10.211722 -0.6204761 0.5791110 1
#> 6 98 0.009384629 15.714769 14.887706 0.6635654 -0.6193277 2
...
The columns in tb are:
which_row: row indices corresponding to the input matrix.fdr: FDR for the differential test. mean_x: The mean value in group x.scaled_mean_x: The mean value in group x after rows are scaled.km: Row groups if k-means clustering is applied to rows.UMAP plot which shows how samples are separated.
dimension_reduction(res, k = 2, method = "UMAP")
dimension_reduction(res, k = 3, method = "UMAP")
dimension_reduction(res, k = 4, method = "UMAP")
dimension_reduction(res, k = 5, method = "UMAP")
dimension_reduction(res, k = 6, method = "UMAP")
Following heatmap shows how subgroups are split when increasing k:
collect_classes(res)
Test correlation between subgroups and known annotations. If the known annotation is numeric, one-way ANOVA test is applied, and if the known annotation is discrete, chi-squared contingency table test is applied.
test_to_known_factors(res)
#> n disease.state(p) specimen(p) cell.type(p) other(p) k
#> SD:skmeans 99 1.18e-04 0.225 6.66e-13 0.1066 2
#> SD:skmeans 100 1.14e-04 0.290 6.36e-17 0.0685 3
#> SD:skmeans 78 2.01e-04 0.167 3.84e-13 0.0312 4
#> SD:skmeans 88 6.48e-05 0.228 3.50e-16 0.0350 5
#> SD:skmeans 83 4.09e-05 0.337 3.40e-18 0.0216 6
If matrix rows can be associated to genes, consider to use functional_enrichment(res,
...) to perform function enrichment for the signature genes. See this vignette for more detailed explanations.
The object with results only for a single top-value method and a single partition method can be extracted as:
res = res_list["SD", "pam"]
# you can also extract it by
# res = res_list["SD:pam"]
A summary of res and all the functions that can be applied to it:
res
#> A 'ConsensusPartition' object with k = 2, 3, 4, 5, 6.
#> On a matrix with 21168 rows and 100 columns.
#> Top rows (1000, 2000, 3000, 4000, 5000) are extracted by 'SD' method.
#> Subgroups are detected by 'pam' method.
#> Performed in total 1250 partitions by row resampling.
#> Best k for subgroups seems to be 2.
#>
#> Following methods can be applied to this 'ConsensusPartition' object:
#> [1] "cola_report" "collect_classes" "collect_plots"
#> [4] "collect_stats" "colnames" "compare_signatures"
#> [7] "consensus_heatmap" "dimension_reduction" "functional_enrichment"
#> [10] "get_anno_col" "get_anno" "get_classes"
#> [13] "get_consensus" "get_matrix" "get_membership"
#> [16] "get_param" "get_signatures" "get_stats"
#> [19] "is_best_k" "is_stable_k" "membership_heatmap"
#> [22] "ncol" "nrow" "plot_ecdf"
#> [25] "rownames" "select_partition_number" "show"
#> [28] "suggest_best_k" "test_to_known_factors"
collect_plots() function collects all the plots made from res for all k (number of partitions)
into one single page to provide an easy and fast comparison between different k.
collect_plots(res)
The plots are:
k and the heatmap of
predicted classes for each k.k.k.k.All the plots in panels can be made by individual functions and they are plotted later in this section.
select_partition_number() produces several plots showing different
statistics for choosing “optimized” k. There are following statistics:
k;k, the area increased is defined as \(A_k - A_{k-1}\).The detailed explanations of these statistics can be found in the cola vignette.
Generally speaking, lower PAC score, higher mean silhouette score or higher
concordance corresponds to better partition. Rand index and Jaccard index
measure how similar the current partition is compared to partition with k-1.
If they are too similar, we won't accept k is better than k-1.
select_partition_number(res)
The numeric values for all these statistics can be obtained by get_stats().
get_stats(res)
#> k 1-PAC mean_silhouette concordance area_increased Rand Jaccard
#> 2 2 0.894 0.935 0.970 0.4604 0.547 0.547
#> 3 3 0.435 0.435 0.738 0.3650 0.872 0.775
#> 4 4 0.527 0.562 0.766 0.1608 0.687 0.390
#> 5 5 0.692 0.657 0.820 0.0783 0.862 0.541
#> 6 6 0.701 0.495 0.684 0.0500 0.909 0.613
suggest_best_k() suggests the best \(k\) based on these statistics. The rules are as follows:
suggest_best_k(res)
#> [1] 2
Following shows the table of the partitions (You need to click the show/hide
code output link to see it). The membership matrix (columns with name p*)
is inferred by
clue::cl_consensus()
function with the SE method. Basically the value in the membership matrix
represents the probability to belong to a certain group. The finall class
label for an item is determined with the group with highest probability it
belongs to.
In get_classes() function, the entropy is calculated from the membership
matrix and the silhouette score is calculated from the consensus matrix.
cbind(get_classes(res, k = 2), get_membership(res, k = 2))
#> class entropy silhouette p1 p2
#> GSM97038 2 0.0000 0.973 0.000 1.000
#> GSM97045 2 0.0000 0.973 0.000 1.000
#> GSM97047 2 0.0938 0.964 0.012 0.988
#> GSM97025 2 0.0000 0.973 0.000 1.000
#> GSM97030 2 0.0000 0.973 0.000 1.000
#> GSM97027 2 0.0000 0.973 0.000 1.000
#> GSM97033 2 0.0000 0.973 0.000 1.000
#> GSM97034 2 0.3274 0.923 0.060 0.940
#> GSM97020 2 0.0000 0.973 0.000 1.000
#> GSM97026 2 0.0000 0.973 0.000 1.000
#> GSM97012 2 0.0000 0.973 0.000 1.000
#> GSM97015 2 0.6438 0.803 0.164 0.836
#> GSM97016 2 0.0000 0.973 0.000 1.000
#> GSM97017 1 0.0376 0.965 0.996 0.004
#> GSM97019 2 0.0000 0.973 0.000 1.000
#> GSM97022 2 0.0000 0.973 0.000 1.000
#> GSM97035 2 0.0000 0.973 0.000 1.000
#> GSM97036 2 0.7139 0.764 0.196 0.804
#> GSM97039 2 0.0000 0.973 0.000 1.000
#> GSM97046 2 0.0000 0.973 0.000 1.000
#> GSM97023 1 0.0000 0.967 1.000 0.000
#> GSM97029 1 0.6712 0.794 0.824 0.176
#> GSM97043 2 0.0000 0.973 0.000 1.000
#> GSM97013 1 0.7056 0.770 0.808 0.192
#> GSM96956 2 0.0000 0.973 0.000 1.000
#> GSM97024 2 0.0000 0.973 0.000 1.000
#> GSM97032 2 0.0000 0.973 0.000 1.000
#> GSM97044 2 0.1184 0.961 0.016 0.984
#> GSM97049 2 0.0000 0.973 0.000 1.000
#> GSM96968 1 0.0000 0.967 1.000 0.000
#> GSM96971 1 0.0000 0.967 1.000 0.000
#> GSM96986 1 0.0376 0.965 0.996 0.004
#> GSM97003 1 0.0000 0.967 1.000 0.000
#> GSM96957 1 0.0000 0.967 1.000 0.000
#> GSM96960 1 0.0000 0.967 1.000 0.000
#> GSM96975 1 0.0000 0.967 1.000 0.000
#> GSM96998 1 0.0000 0.967 1.000 0.000
#> GSM96999 1 0.0000 0.967 1.000 0.000
#> GSM97001 1 0.0000 0.967 1.000 0.000
#> GSM97005 1 0.0000 0.967 1.000 0.000
#> GSM97006 1 0.0000 0.967 1.000 0.000
#> GSM97021 1 0.0376 0.965 0.996 0.004
#> GSM97028 1 0.1843 0.948 0.972 0.028
#> GSM97031 1 0.0000 0.967 1.000 0.000
#> GSM97037 2 0.0000 0.973 0.000 1.000
#> GSM97018 2 0.3274 0.923 0.060 0.940
#> GSM97014 1 0.9815 0.332 0.580 0.420
#> GSM97042 2 0.0000 0.973 0.000 1.000
#> GSM97040 1 0.0938 0.960 0.988 0.012
#> GSM97041 1 0.4298 0.895 0.912 0.088
#> GSM96955 1 0.9000 0.572 0.684 0.316
#> GSM96990 2 0.0000 0.973 0.000 1.000
#> GSM96991 2 0.0000 0.973 0.000 1.000
#> GSM97048 2 0.0000 0.973 0.000 1.000
#> GSM96963 2 0.0000 0.973 0.000 1.000
#> GSM96953 2 0.0000 0.973 0.000 1.000
#> GSM96966 1 0.0000 0.967 1.000 0.000
#> GSM96979 1 0.0000 0.967 1.000 0.000
#> GSM96983 1 0.5519 0.855 0.872 0.128
#> GSM96984 1 0.1633 0.952 0.976 0.024
#> GSM96994 1 0.6712 0.796 0.824 0.176
#> GSM96996 1 0.0000 0.967 1.000 0.000
#> GSM96997 1 0.0000 0.967 1.000 0.000
#> GSM97007 2 0.9460 0.445 0.364 0.636
#> GSM96954 1 0.0000 0.967 1.000 0.000
#> GSM96962 1 0.0000 0.967 1.000 0.000
#> GSM96969 1 0.0000 0.967 1.000 0.000
#> GSM96970 1 0.0000 0.967 1.000 0.000
#> GSM96973 1 0.0000 0.967 1.000 0.000
#> GSM96976 1 0.4562 0.890 0.904 0.096
#> GSM96977 1 0.0000 0.967 1.000 0.000
#> GSM96995 1 0.3584 0.913 0.932 0.068
#> GSM97002 1 0.0000 0.967 1.000 0.000
#> GSM97009 1 0.8861 0.600 0.696 0.304
#> GSM97010 1 0.0672 0.962 0.992 0.008
#> GSM96974 1 0.0000 0.967 1.000 0.000
#> GSM96985 1 0.0000 0.967 1.000 0.000
#> GSM96959 1 0.1843 0.948 0.972 0.028
#> GSM96972 1 0.0000 0.967 1.000 0.000
#> GSM96978 1 0.0000 0.967 1.000 0.000
#> GSM96967 1 0.0000 0.967 1.000 0.000
#> GSM96987 1 0.0000 0.967 1.000 0.000
#> GSM97011 1 0.0376 0.965 0.996 0.004
#> GSM96964 1 0.0000 0.967 1.000 0.000
#> GSM96965 1 0.0938 0.960 0.988 0.012
#> GSM96981 1 0.0000 0.967 1.000 0.000
#> GSM96982 1 0.0000 0.967 1.000 0.000
#> GSM96988 1 0.0000 0.967 1.000 0.000
#> GSM97000 1 0.0000 0.967 1.000 0.000
#> GSM97004 1 0.0000 0.967 1.000 0.000
#> GSM97008 1 0.0000 0.967 1.000 0.000
#> GSM96950 1 0.0000 0.967 1.000 0.000
#> GSM96980 1 0.0000 0.967 1.000 0.000
#> GSM96989 1 0.0000 0.967 1.000 0.000
#> GSM96992 1 0.0000 0.967 1.000 0.000
#> GSM96993 1 0.0938 0.960 0.988 0.012
#> GSM96958 1 0.0000 0.967 1.000 0.000
#> GSM96951 1 0.0000 0.967 1.000 0.000
#> GSM96952 1 0.0000 0.967 1.000 0.000
#> GSM96961 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
#> GSM97038 2 0.5810 0.761 0.000 0.664 0.336
#> GSM97045 2 0.5810 0.761 0.000 0.664 0.336
#> GSM97047 2 0.1482 0.622 0.020 0.968 0.012
#> GSM97025 2 0.5810 0.761 0.000 0.664 0.336
#> GSM97030 2 0.0661 0.633 0.004 0.988 0.008
#> GSM97027 2 0.5810 0.761 0.000 0.664 0.336
#> GSM97033 2 0.5810 0.761 0.000 0.664 0.336
#> GSM97034 2 0.0661 0.633 0.004 0.988 0.008
#> GSM97020 2 0.5810 0.761 0.000 0.664 0.336
#> GSM97026 2 0.4178 0.717 0.000 0.828 0.172
#> GSM97012 2 0.5810 0.761 0.000 0.664 0.336
#> GSM97015 2 0.6548 -0.154 0.372 0.616 0.012
#> GSM97016 2 0.5810 0.761 0.000 0.664 0.336
#> GSM97017 1 0.4645 0.433 0.816 0.008 0.176
#> GSM97019 2 0.5785 0.760 0.000 0.668 0.332
#> GSM97022 2 0.5760 0.760 0.000 0.672 0.328
#> GSM97035 2 0.5810 0.761 0.000 0.664 0.336
#> GSM97036 2 0.5229 0.602 0.104 0.828 0.068
#> GSM97039 2 0.5810 0.761 0.000 0.664 0.336
#> GSM97046 2 0.5810 0.761 0.000 0.664 0.336
#> GSM97023 1 0.4887 0.459 0.772 0.000 0.228
#> GSM97029 1 0.9119 -0.121 0.484 0.368 0.148
#> GSM97043 2 0.2537 0.683 0.000 0.920 0.080
#> GSM97013 3 0.6505 -0.260 0.468 0.004 0.528
#> GSM96956 2 0.1643 0.666 0.000 0.956 0.044
#> GSM97024 2 0.2537 0.683 0.000 0.920 0.080
#> GSM97032 2 0.0000 0.641 0.000 1.000 0.000
#> GSM97044 2 0.1267 0.620 0.004 0.972 0.024
#> GSM97049 2 0.5810 0.761 0.000 0.664 0.336
#> GSM96968 1 0.5098 0.465 0.752 0.248 0.000
#> GSM96971 1 0.9559 -0.106 0.472 0.220 0.308
#> GSM96986 1 0.5486 0.487 0.780 0.196 0.024
#> GSM97003 1 0.1620 0.533 0.964 0.012 0.024
#> GSM96957 1 0.3276 0.535 0.908 0.068 0.024
#> GSM96960 1 0.4887 0.459 0.772 0.000 0.228
#> GSM96975 1 0.2261 0.537 0.932 0.068 0.000
#> GSM96998 1 0.4887 0.459 0.772 0.000 0.228
#> GSM96999 1 0.0848 0.529 0.984 0.008 0.008
#> GSM97001 1 0.4628 0.508 0.856 0.056 0.088
#> GSM97005 1 0.4808 0.490 0.804 0.008 0.188
#> GSM97006 1 0.5058 0.451 0.756 0.000 0.244
#> GSM97021 1 0.1950 0.536 0.952 0.040 0.008
#> GSM97028 1 0.6252 0.223 0.556 0.444 0.000
#> GSM97031 1 0.5098 0.454 0.752 0.000 0.248
#> GSM97037 2 0.0424 0.646 0.000 0.992 0.008
#> GSM97018 2 0.0661 0.633 0.004 0.988 0.008
#> GSM97014 2 0.9734 0.458 0.236 0.432 0.332
#> GSM97042 2 0.5810 0.761 0.000 0.664 0.336
#> GSM97040 1 0.5988 0.418 0.688 0.304 0.008
#> GSM97041 1 0.6427 0.296 0.640 0.012 0.348
#> GSM96955 2 0.9633 0.275 0.300 0.464 0.236
#> GSM96990 2 0.0983 0.627 0.004 0.980 0.016
#> GSM96991 2 0.5785 0.760 0.000 0.668 0.332
#> GSM97048 2 0.5810 0.761 0.000 0.664 0.336
#> GSM96963 2 0.5810 0.761 0.000 0.664 0.336
#> GSM96953 2 0.5810 0.761 0.000 0.664 0.336
#> GSM96966 1 0.6192 -0.268 0.580 0.000 0.420
#> GSM96979 1 0.7091 0.446 0.676 0.268 0.056
#> GSM96983 2 0.7130 -0.241 0.432 0.544 0.024
#> GSM96984 1 0.6702 0.380 0.648 0.328 0.024
#> GSM96994 1 0.6750 0.375 0.640 0.336 0.024
#> GSM96996 1 0.0237 0.526 0.996 0.000 0.004
#> GSM96997 1 0.6059 0.490 0.764 0.048 0.188
#> GSM97007 2 0.8579 -0.446 0.440 0.464 0.096
#> GSM96954 1 0.9438 0.381 0.504 0.252 0.244
#> GSM96962 1 0.9125 0.357 0.516 0.320 0.164
#> GSM96969 1 0.6235 -0.273 0.564 0.000 0.436
#> GSM96970 1 0.6168 -0.267 0.588 0.000 0.412
#> GSM96973 1 0.6192 -0.268 0.580 0.000 0.420
#> GSM96976 3 0.9806 0.308 0.276 0.292 0.432
#> GSM96977 1 0.4974 0.472 0.764 0.236 0.000
#> GSM96995 1 0.6047 0.410 0.680 0.312 0.008
#> GSM97002 1 0.2537 0.513 0.920 0.000 0.080
#> GSM97009 1 0.6726 0.198 0.644 0.024 0.332
#> GSM97010 1 0.6295 0.431 0.764 0.072 0.164
#> GSM96974 3 0.9744 0.301 0.236 0.336 0.428
#> GSM96985 1 0.9265 -0.357 0.428 0.156 0.416
#> GSM96959 1 0.5659 0.461 0.740 0.248 0.012
#> GSM96972 3 0.5948 0.287 0.360 0.000 0.640
#> GSM96978 1 0.6625 0.396 0.660 0.316 0.024
#> GSM96967 1 0.6225 -0.275 0.568 0.000 0.432
#> GSM96987 1 0.4887 0.459 0.772 0.000 0.228
#> GSM97011 1 0.5492 0.490 0.816 0.080 0.104
#> GSM96964 1 0.4887 0.459 0.772 0.000 0.228
#> GSM96965 3 0.6282 0.317 0.324 0.012 0.664
#> GSM96981 1 0.0592 0.527 0.988 0.000 0.012
#> GSM96982 1 0.2261 0.517 0.932 0.000 0.068
#> GSM96988 1 0.6726 0.384 0.644 0.332 0.024
#> GSM97000 1 0.3207 0.535 0.904 0.084 0.012
#> GSM97004 3 0.5948 0.287 0.360 0.000 0.640
#> GSM97008 1 0.2866 0.536 0.916 0.076 0.008
#> GSM96950 1 0.5816 0.502 0.788 0.056 0.156
#> GSM96980 1 0.6305 -0.310 0.516 0.000 0.484
#> GSM96989 1 0.4887 0.459 0.772 0.000 0.228
#> GSM96992 1 0.4887 0.459 0.772 0.000 0.228
#> GSM96993 1 0.8674 0.401 0.568 0.296 0.136
#> GSM96958 1 0.0424 0.526 0.992 0.000 0.008
#> GSM96951 1 0.4887 0.464 0.772 0.000 0.228
#> GSM96952 1 0.4887 0.459 0.772 0.000 0.228
#> GSM96961 1 0.4887 0.459 0.772 0.000 0.228
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM97038 2 0.0188 0.8003 0.000 0.996 0.004 0.000
#> GSM97045 2 0.0000 0.8017 0.000 1.000 0.000 0.000
#> GSM97047 2 0.7417 0.0166 0.016 0.464 0.412 0.108
#> GSM97025 2 0.0000 0.8017 0.000 1.000 0.000 0.000
#> GSM97030 3 0.6060 0.0206 0.000 0.440 0.516 0.044
#> GSM97027 2 0.0000 0.8017 0.000 1.000 0.000 0.000
#> GSM97033 2 0.0000 0.8017 0.000 1.000 0.000 0.000
#> GSM97034 2 0.6265 0.1277 0.000 0.500 0.444 0.056
#> GSM97020 2 0.0000 0.8017 0.000 1.000 0.000 0.000
#> GSM97026 2 0.3958 0.6695 0.000 0.816 0.160 0.024
#> GSM97012 2 0.0000 0.8017 0.000 1.000 0.000 0.000
#> GSM97015 3 0.7612 0.4005 0.096 0.268 0.580 0.056
#> GSM97016 2 0.0000 0.8017 0.000 1.000 0.000 0.000
#> GSM97017 4 0.7233 0.5782 0.232 0.128 0.028 0.612
#> GSM97019 2 0.0188 0.8006 0.000 0.996 0.004 0.000
#> GSM97022 2 0.0817 0.7902 0.000 0.976 0.024 0.000
#> GSM97035 2 0.0000 0.8017 0.000 1.000 0.000 0.000
#> GSM97036 2 0.7471 0.4290 0.040 0.604 0.224 0.132
#> GSM97039 2 0.0000 0.8017 0.000 1.000 0.000 0.000
#> GSM97046 2 0.0000 0.8017 0.000 1.000 0.000 0.000
#> GSM97023 1 0.0000 0.7676 1.000 0.000 0.000 0.000
#> GSM97029 4 0.7317 0.5071 0.156 0.244 0.016 0.584
#> GSM97043 2 0.4406 0.5263 0.000 0.700 0.300 0.000
#> GSM97013 1 0.4989 0.1810 0.528 0.472 0.000 0.000
#> GSM96956 2 0.4992 0.1350 0.000 0.524 0.476 0.000
#> GSM97024 2 0.5062 0.5063 0.000 0.680 0.300 0.020
#> GSM97032 2 0.6276 0.0572 0.000 0.480 0.464 0.056
#> GSM97044 3 0.3308 0.6906 0.000 0.092 0.872 0.036
#> GSM97049 2 0.0000 0.8017 0.000 1.000 0.000 0.000
#> GSM96968 4 0.6856 0.6076 0.140 0.000 0.284 0.576
#> GSM96971 3 0.4250 0.4905 0.000 0.000 0.724 0.276
#> GSM96986 3 0.4988 0.3496 0.020 0.000 0.692 0.288
#> GSM97003 4 0.7254 0.6107 0.300 0.000 0.176 0.524
#> GSM96957 4 0.6038 0.4952 0.424 0.000 0.044 0.532
#> GSM96960 1 0.0469 0.7615 0.988 0.000 0.000 0.012
#> GSM96975 4 0.6678 0.6551 0.172 0.000 0.208 0.620
#> GSM96998 1 0.0000 0.7676 1.000 0.000 0.000 0.000
#> GSM96999 4 0.5277 0.4382 0.460 0.000 0.008 0.532
#> GSM97001 4 0.6501 0.6306 0.256 0.004 0.108 0.632
#> GSM97005 1 0.5169 0.4154 0.696 0.000 0.032 0.272
#> GSM97006 1 0.0707 0.7561 0.980 0.000 0.020 0.000
#> GSM97021 4 0.6566 0.6447 0.236 0.000 0.140 0.624
#> GSM97028 3 0.5255 0.4596 0.028 0.004 0.696 0.272
#> GSM97031 1 0.2644 0.7180 0.908 0.000 0.060 0.032
#> GSM97037 2 0.4998 0.1059 0.000 0.512 0.488 0.000
#> GSM97018 3 0.6252 0.0345 0.000 0.432 0.512 0.056
#> GSM97014 2 0.6047 0.2437 0.020 0.624 0.028 0.328
#> GSM97042 2 0.0336 0.7989 0.000 0.992 0.008 0.000
#> GSM97040 4 0.6523 0.6361 0.136 0.000 0.236 0.628
#> GSM97041 1 0.7854 0.0890 0.452 0.284 0.004 0.260
#> GSM96955 4 0.7805 0.5163 0.048 0.172 0.196 0.584
#> GSM96990 3 0.3266 0.7064 0.000 0.084 0.876 0.040
#> GSM96991 2 0.0469 0.7967 0.000 0.988 0.012 0.000
#> GSM97048 2 0.0000 0.8017 0.000 1.000 0.000 0.000
#> GSM96963 2 0.0000 0.8017 0.000 1.000 0.000 0.000
#> GSM96953 2 0.0000 0.8017 0.000 1.000 0.000 0.000
#> GSM96966 4 0.2647 0.5313 0.120 0.000 0.000 0.880
#> GSM96979 3 0.2011 0.7124 0.000 0.000 0.920 0.080
#> GSM96983 3 0.1042 0.7299 0.000 0.020 0.972 0.008
#> GSM96984 3 0.0657 0.7307 0.012 0.000 0.984 0.004
#> GSM96994 3 0.0000 0.7299 0.000 0.000 1.000 0.000
#> GSM96996 4 0.5906 0.4540 0.436 0.000 0.036 0.528
#> GSM96997 3 0.5309 0.4694 0.256 0.000 0.700 0.044
#> GSM97007 3 0.0188 0.7303 0.000 0.000 0.996 0.004
#> GSM96954 3 0.3176 0.7058 0.036 0.000 0.880 0.084
#> GSM96962 3 0.0937 0.7300 0.012 0.000 0.976 0.012
#> GSM96969 4 0.3610 0.4269 0.200 0.000 0.000 0.800
#> GSM96970 4 0.2197 0.5543 0.080 0.000 0.004 0.916
#> GSM96973 4 0.2319 0.5582 0.036 0.000 0.040 0.924
#> GSM96976 4 0.1545 0.5638 0.000 0.008 0.040 0.952
#> GSM96977 4 0.6635 0.6461 0.152 0.000 0.228 0.620
#> GSM96995 4 0.6442 0.6309 0.124 0.000 0.244 0.632
#> GSM97002 1 0.4605 0.1144 0.664 0.000 0.000 0.336
#> GSM97009 2 0.8749 -0.2129 0.136 0.468 0.096 0.300
#> GSM97010 4 0.8804 0.6104 0.156 0.108 0.240 0.496
#> GSM96974 3 0.5409 0.2927 0.012 0.000 0.496 0.492
#> GSM96985 4 0.3505 0.5665 0.048 0.000 0.088 0.864
#> GSM96959 4 0.6494 0.6388 0.136 0.000 0.232 0.632
#> GSM96972 1 0.5659 0.3889 0.600 0.000 0.032 0.368
#> GSM96978 3 0.4737 0.4790 0.020 0.000 0.728 0.252
#> GSM96967 4 0.4149 0.4597 0.168 0.000 0.028 0.804
#> GSM96987 1 0.0000 0.7676 1.000 0.000 0.000 0.000
#> GSM97011 4 0.6522 0.6447 0.144 0.000 0.224 0.632
#> GSM96964 1 0.0000 0.7676 1.000 0.000 0.000 0.000
#> GSM96965 4 0.2469 0.5316 0.000 0.108 0.000 0.892
#> GSM96981 4 0.4989 0.4165 0.472 0.000 0.000 0.528
#> GSM96982 1 0.4624 0.1014 0.660 0.000 0.000 0.340
#> GSM96988 3 0.3176 0.6943 0.084 0.000 0.880 0.036
#> GSM97000 4 0.6534 0.6316 0.132 0.000 0.244 0.624
#> GSM97004 1 0.2868 0.6497 0.864 0.000 0.000 0.136
#> GSM97008 4 0.6534 0.6475 0.148 0.000 0.220 0.632
#> GSM96950 1 0.3219 0.6036 0.836 0.000 0.000 0.164
#> GSM96980 4 0.4624 0.1990 0.340 0.000 0.000 0.660
#> GSM96989 1 0.0000 0.7676 1.000 0.000 0.000 0.000
#> GSM96992 1 0.0000 0.7676 1.000 0.000 0.000 0.000
#> GSM96993 1 0.6858 0.1048 0.532 0.004 0.368 0.096
#> GSM96958 4 0.4992 0.4086 0.476 0.000 0.000 0.524
#> GSM96951 1 0.1022 0.7548 0.968 0.000 0.000 0.032
#> GSM96952 1 0.0188 0.7660 0.996 0.000 0.000 0.004
#> GSM96961 1 0.0000 0.7676 1.000 0.000 0.000 0.000
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM97038 2 0.0000 0.796 0.000 1.000 0.000 0.000 0.000
#> GSM97045 2 0.0000 0.796 0.000 1.000 0.000 0.000 0.000
#> GSM97047 5 0.4763 0.376 0.000 0.212 0.076 0.000 0.712
#> GSM97025 2 0.0000 0.796 0.000 1.000 0.000 0.000 0.000
#> GSM97030 2 0.5750 0.378 0.000 0.492 0.436 0.008 0.064
#> GSM97027 2 0.0000 0.796 0.000 1.000 0.000 0.000 0.000
#> GSM97033 2 0.0000 0.796 0.000 1.000 0.000 0.000 0.000
#> GSM97034 2 0.6221 0.397 0.000 0.492 0.400 0.016 0.092
#> GSM97020 2 0.0000 0.796 0.000 1.000 0.000 0.000 0.000
#> GSM97026 2 0.3412 0.734 0.000 0.820 0.152 0.000 0.028
#> GSM97012 2 0.2723 0.785 0.000 0.864 0.012 0.124 0.000
#> GSM97015 3 0.5959 -0.276 0.000 0.420 0.472 0.000 0.108
#> GSM97016 2 0.0000 0.796 0.000 1.000 0.000 0.000 0.000
#> GSM97017 5 0.2464 0.691 0.016 0.096 0.000 0.000 0.888
#> GSM97019 2 0.2723 0.785 0.000 0.864 0.012 0.124 0.000
#> GSM97022 2 0.3413 0.777 0.000 0.832 0.044 0.124 0.000
#> GSM97035 2 0.2723 0.785 0.000 0.864 0.012 0.124 0.000
#> GSM97036 2 0.6162 0.551 0.020 0.604 0.248 0.000 0.128
#> GSM97039 2 0.0000 0.796 0.000 1.000 0.000 0.000 0.000
#> GSM97046 2 0.0000 0.796 0.000 1.000 0.000 0.000 0.000
#> GSM97023 1 0.0000 0.827 1.000 0.000 0.000 0.000 0.000
#> GSM97029 5 0.7454 0.130 0.164 0.376 0.060 0.000 0.400
#> GSM97043 2 0.4403 0.531 0.000 0.608 0.384 0.008 0.000
#> GSM97013 1 0.4297 0.220 0.528 0.472 0.000 0.000 0.000
#> GSM96956 2 0.4482 0.524 0.000 0.612 0.376 0.000 0.012
#> GSM97024 2 0.6287 0.537 0.000 0.536 0.328 0.124 0.012
#> GSM97032 2 0.5944 0.377 0.000 0.488 0.404 0.000 0.108
#> GSM97044 3 0.1357 0.687 0.000 0.004 0.948 0.048 0.000
#> GSM97049 2 0.0000 0.796 0.000 1.000 0.000 0.000 0.000
#> GSM96968 3 0.4238 0.401 0.004 0.000 0.628 0.000 0.368
#> GSM96971 3 0.4278 0.415 0.000 0.000 0.548 0.000 0.452
#> GSM96986 3 0.4161 0.468 0.000 0.000 0.608 0.000 0.392
#> GSM97003 5 0.4197 0.532 0.244 0.000 0.028 0.000 0.728
#> GSM96957 1 0.3730 0.580 0.712 0.000 0.000 0.000 0.288
#> GSM96960 1 0.0162 0.827 0.996 0.000 0.000 0.004 0.000
#> GSM96975 5 0.0955 0.710 0.028 0.000 0.004 0.000 0.968
#> GSM96998 1 0.0000 0.827 1.000 0.000 0.000 0.000 0.000
#> GSM96999 1 0.3895 0.532 0.680 0.000 0.000 0.000 0.320
#> GSM97001 5 0.0162 0.711 0.004 0.000 0.000 0.000 0.996
#> GSM97005 5 0.3534 0.505 0.256 0.000 0.000 0.000 0.744
#> GSM97006 1 0.0162 0.827 0.996 0.000 0.000 0.004 0.000
#> GSM97021 5 0.0609 0.709 0.020 0.000 0.000 0.000 0.980
#> GSM97028 3 0.3816 0.516 0.000 0.000 0.696 0.000 0.304
#> GSM97031 1 0.3814 0.507 0.720 0.000 0.000 0.004 0.276
#> GSM97037 2 0.4547 0.491 0.000 0.588 0.400 0.000 0.012
#> GSM97018 2 0.5944 0.377 0.000 0.488 0.404 0.000 0.108
#> GSM97014 5 0.4219 0.459 0.000 0.416 0.000 0.000 0.584
#> GSM97042 2 0.2921 0.784 0.000 0.856 0.020 0.124 0.000
#> GSM97040 5 0.0000 0.710 0.000 0.000 0.000 0.000 1.000
#> GSM97041 5 0.6313 0.467 0.188 0.296 0.000 0.000 0.516
#> GSM96955 5 0.3969 0.558 0.000 0.304 0.004 0.000 0.692
#> GSM96990 3 0.3420 0.649 0.000 0.076 0.840 0.000 0.084
#> GSM96991 2 0.2921 0.784 0.000 0.856 0.020 0.124 0.000
#> GSM97048 2 0.0000 0.796 0.000 1.000 0.000 0.000 0.000
#> GSM96963 2 0.2723 0.785 0.000 0.864 0.012 0.124 0.000
#> GSM96953 2 0.2723 0.785 0.000 0.864 0.012 0.124 0.000
#> GSM96966 4 0.2859 0.909 0.068 0.000 0.000 0.876 0.056
#> GSM96979 3 0.4540 0.545 0.000 0.000 0.656 0.024 0.320
#> GSM96983 3 0.0794 0.706 0.000 0.000 0.972 0.000 0.028
#> GSM96984 3 0.1270 0.709 0.000 0.000 0.948 0.000 0.052
#> GSM96994 3 0.0404 0.701 0.000 0.000 0.988 0.000 0.012
#> GSM96996 5 0.5360 0.217 0.396 0.000 0.048 0.004 0.552
#> GSM96997 3 0.5569 0.474 0.092 0.000 0.588 0.000 0.320
#> GSM97007 3 0.0510 0.703 0.000 0.000 0.984 0.000 0.016
#> GSM96954 3 0.4789 0.494 0.024 0.000 0.584 0.000 0.392
#> GSM96962 3 0.3452 0.626 0.000 0.000 0.756 0.000 0.244
#> GSM96969 4 0.2329 0.888 0.124 0.000 0.000 0.876 0.000
#> GSM96970 4 0.2949 0.911 0.052 0.000 0.004 0.876 0.068
#> GSM96973 4 0.3009 0.902 0.008 0.000 0.052 0.876 0.064
#> GSM96976 4 0.2843 0.896 0.000 0.000 0.048 0.876 0.076
#> GSM96977 5 0.0510 0.711 0.016 0.000 0.000 0.000 0.984
#> GSM96995 5 0.1121 0.689 0.000 0.000 0.044 0.000 0.956
#> GSM97002 1 0.3452 0.636 0.756 0.000 0.000 0.000 0.244
#> GSM97009 5 0.4219 0.477 0.000 0.416 0.000 0.000 0.584
#> GSM97010 5 0.4480 0.629 0.016 0.152 0.060 0.000 0.772
#> GSM96974 4 0.2377 0.846 0.000 0.000 0.128 0.872 0.000
#> GSM96985 4 0.4356 0.830 0.016 0.000 0.060 0.784 0.140
#> GSM96959 5 0.0162 0.709 0.000 0.000 0.004 0.000 0.996
#> GSM96972 4 0.2793 0.902 0.088 0.000 0.036 0.876 0.000
#> GSM96978 3 0.2471 0.668 0.000 0.000 0.864 0.000 0.136
#> GSM96967 4 0.2990 0.910 0.080 0.000 0.032 0.876 0.012
#> GSM96987 1 0.0000 0.827 1.000 0.000 0.000 0.000 0.000
#> GSM97011 5 0.0000 0.710 0.000 0.000 0.000 0.000 1.000
#> GSM96964 1 0.0000 0.827 1.000 0.000 0.000 0.000 0.000
#> GSM96965 4 0.2989 0.878 0.000 0.072 0.000 0.868 0.060
#> GSM96981 5 0.4397 0.160 0.432 0.000 0.000 0.004 0.564
#> GSM96982 1 0.3607 0.639 0.752 0.000 0.000 0.004 0.244
#> GSM96988 3 0.1270 0.706 0.000 0.000 0.948 0.000 0.052
#> GSM97000 5 0.0162 0.708 0.000 0.000 0.004 0.000 0.996
#> GSM97004 1 0.0510 0.819 0.984 0.000 0.000 0.016 0.000
#> GSM97008 5 0.0000 0.710 0.000 0.000 0.000 0.000 1.000
#> GSM96950 1 0.1908 0.781 0.908 0.000 0.000 0.000 0.092
#> GSM96980 4 0.2921 0.889 0.124 0.000 0.000 0.856 0.020
#> GSM96989 1 0.0000 0.827 1.000 0.000 0.000 0.000 0.000
#> GSM96992 1 0.0162 0.827 0.996 0.000 0.000 0.004 0.000
#> GSM96993 1 0.6182 0.178 0.520 0.000 0.324 0.000 0.156
#> GSM96958 1 0.3730 0.583 0.712 0.000 0.000 0.000 0.288
#> GSM96951 1 0.0510 0.820 0.984 0.000 0.000 0.000 0.016
#> GSM96952 1 0.0162 0.827 0.996 0.000 0.000 0.004 0.000
#> GSM96961 1 0.0000 0.827 1.000 0.000 0.000 0.000 0.000
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM97038 2 0.4184 0.34778 0.000 0.500 0.012 0.000 0.000 0.488
#> GSM97045 6 0.5714 -0.26886 0.000 0.176 0.340 0.000 0.000 0.484
#> GSM97047 5 0.4051 0.49534 0.000 0.004 0.224 0.000 0.728 0.044
#> GSM97025 6 0.5714 -0.26886 0.000 0.176 0.340 0.000 0.000 0.484
#> GSM97030 3 0.3610 0.48128 0.000 0.200 0.768 0.000 0.028 0.004
#> GSM97027 6 0.5714 -0.26886 0.000 0.176 0.340 0.000 0.000 0.484
#> GSM97033 6 0.5691 -0.26890 0.000 0.172 0.340 0.000 0.000 0.488
#> GSM97034 3 0.2629 0.41119 0.000 0.068 0.872 0.000 0.060 0.000
#> GSM97020 6 0.5844 -0.31503 0.000 0.268 0.244 0.000 0.000 0.488
#> GSM97026 3 0.5978 -0.23922 0.000 0.152 0.532 0.000 0.024 0.292
#> GSM97012 2 0.3578 0.54843 0.000 0.660 0.340 0.000 0.000 0.000
#> GSM97015 3 0.4945 0.45647 0.000 0.344 0.584 0.000 0.068 0.004
#> GSM97016 2 0.3997 0.35429 0.000 0.508 0.004 0.000 0.000 0.488
#> GSM97017 5 0.3620 0.67219 0.036 0.000 0.056 0.000 0.824 0.084
#> GSM97019 2 0.3578 0.54843 0.000 0.660 0.340 0.000 0.000 0.000
#> GSM97022 2 0.3727 0.51045 0.000 0.612 0.388 0.000 0.000 0.000
#> GSM97035 2 0.3607 0.54628 0.000 0.652 0.348 0.000 0.000 0.000
#> GSM97036 3 0.6136 0.09076 0.040 0.088 0.656 0.000 0.100 0.116
#> GSM97039 2 0.3997 0.35429 0.000 0.508 0.004 0.000 0.000 0.488
#> GSM97046 2 0.3997 0.35429 0.000 0.508 0.004 0.000 0.000 0.488
#> GSM97023 1 0.0000 0.81994 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM97029 5 0.7860 0.12661 0.160 0.036 0.296 0.000 0.376 0.132
#> GSM97043 3 0.3916 0.30116 0.000 0.184 0.752 0.000 0.000 0.064
#> GSM97013 1 0.4513 0.19324 0.528 0.032 0.000 0.000 0.000 0.440
#> GSM96956 3 0.5384 0.35267 0.000 0.428 0.472 0.000 0.004 0.096
#> GSM97024 2 0.3737 0.50544 0.000 0.608 0.392 0.000 0.000 0.000
#> GSM97032 3 0.3752 0.48991 0.000 0.164 0.772 0.000 0.064 0.000
#> GSM97044 3 0.4246 0.46473 0.000 0.020 0.580 0.000 0.000 0.400
#> GSM97049 2 0.3997 0.35429 0.000 0.508 0.004 0.000 0.000 0.488
#> GSM96968 3 0.6511 0.37532 0.016 0.012 0.456 0.000 0.308 0.208
#> GSM96971 6 0.5147 0.19628 0.000 0.000 0.064 0.008 0.416 0.512
#> GSM96986 6 0.5361 0.21497 0.000 0.000 0.116 0.000 0.372 0.512
#> GSM97003 5 0.3732 0.53204 0.228 0.000 0.024 0.000 0.744 0.004
#> GSM96957 1 0.3221 0.62750 0.736 0.000 0.000 0.000 0.264 0.000
#> GSM96960 1 0.0363 0.81811 0.988 0.000 0.000 0.012 0.000 0.000
#> GSM96975 5 0.1141 0.68365 0.052 0.000 0.000 0.000 0.948 0.000
#> GSM96998 1 0.0000 0.81994 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM96999 1 0.3428 0.57712 0.696 0.000 0.000 0.000 0.304 0.000
#> GSM97001 5 0.0146 0.68405 0.004 0.000 0.000 0.000 0.996 0.000
#> GSM97005 5 0.4551 0.54104 0.168 0.000 0.056 0.000 0.736 0.040
#> GSM97006 1 0.0458 0.81678 0.984 0.000 0.000 0.016 0.000 0.000
#> GSM97021 5 0.2577 0.67285 0.016 0.000 0.056 0.000 0.888 0.040
#> GSM97028 3 0.5646 0.41359 0.000 0.000 0.536 0.000 0.244 0.220
#> GSM97031 1 0.4680 0.33622 0.628 0.000 0.000 0.012 0.320 0.040
#> GSM97037 3 0.5105 0.38752 0.000 0.428 0.500 0.000 0.004 0.068
#> GSM97018 3 0.2965 0.45757 0.000 0.080 0.848 0.000 0.072 0.000
#> GSM97014 5 0.5364 0.38993 0.000 0.024 0.056 0.000 0.504 0.416
#> GSM97042 2 0.3578 0.54843 0.000 0.660 0.340 0.000 0.000 0.000
#> GSM97040 5 0.2129 0.67295 0.000 0.000 0.056 0.000 0.904 0.040
#> GSM97041 5 0.6944 0.42338 0.156 0.020 0.056 0.000 0.472 0.296
#> GSM96955 5 0.4789 0.49657 0.000 0.092 0.000 0.000 0.640 0.268
#> GSM96990 3 0.4671 0.49775 0.000 0.000 0.628 0.000 0.068 0.304
#> GSM96991 2 0.3634 0.54022 0.000 0.644 0.356 0.000 0.000 0.000
#> GSM97048 2 0.3997 0.35429 0.000 0.508 0.004 0.000 0.000 0.488
#> GSM96963 2 0.3578 0.54843 0.000 0.660 0.340 0.000 0.000 0.000
#> GSM96953 2 0.0260 0.38814 0.000 0.992 0.008 0.000 0.000 0.000
#> GSM96966 4 0.0000 0.98267 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM96979 6 0.5894 0.22541 0.000 0.000 0.084 0.044 0.360 0.512
#> GSM96983 3 0.3937 0.44610 0.000 0.000 0.572 0.000 0.004 0.424
#> GSM96984 6 0.4807 -0.40948 0.000 0.000 0.464 0.000 0.052 0.484
#> GSM96994 3 0.3860 0.40679 0.000 0.000 0.528 0.000 0.000 0.472
#> GSM96996 5 0.4177 -0.00686 0.468 0.000 0.000 0.012 0.520 0.000
#> GSM96997 6 0.6009 0.22181 0.056 0.000 0.080 0.000 0.360 0.504
#> GSM97007 3 0.4260 0.39078 0.000 0.000 0.512 0.000 0.016 0.472
#> GSM96954 6 0.5368 0.16795 0.024 0.000 0.056 0.000 0.420 0.500
#> GSM96962 6 0.5787 0.00387 0.000 0.000 0.252 0.000 0.244 0.504
#> GSM96969 4 0.0000 0.98267 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM96970 4 0.0000 0.98267 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM96973 4 0.0000 0.98267 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM96976 4 0.0405 0.97589 0.000 0.008 0.000 0.988 0.004 0.000
#> GSM96977 5 0.1007 0.68392 0.044 0.000 0.000 0.000 0.956 0.000
#> GSM96995 5 0.1625 0.67432 0.000 0.000 0.060 0.000 0.928 0.012
#> GSM97002 1 0.3198 0.63209 0.740 0.000 0.000 0.000 0.260 0.000
#> GSM97009 5 0.4440 0.40789 0.000 0.008 0.016 0.000 0.556 0.420
#> GSM97010 5 0.3746 0.60048 0.048 0.000 0.000 0.000 0.760 0.192
#> GSM96974 4 0.0146 0.98037 0.000 0.000 0.004 0.996 0.000 0.000
#> GSM96985 4 0.2147 0.88602 0.000 0.000 0.020 0.896 0.084 0.000
#> GSM96959 5 0.2129 0.67295 0.000 0.000 0.056 0.000 0.904 0.040
#> GSM96972 4 0.0000 0.98267 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM96978 3 0.5486 0.40143 0.000 0.000 0.496 0.000 0.132 0.372
#> GSM96967 4 0.0000 0.98267 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM96987 1 0.0000 0.81994 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM97011 5 0.0000 0.68318 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM96964 1 0.0000 0.81994 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM96965 4 0.0520 0.97527 0.000 0.000 0.000 0.984 0.008 0.008
#> GSM96981 5 0.4169 0.02661 0.456 0.000 0.000 0.012 0.532 0.000
#> GSM96982 1 0.3608 0.62140 0.716 0.000 0.000 0.012 0.272 0.000
#> GSM96988 3 0.4400 0.46699 0.000 0.000 0.592 0.000 0.032 0.376
#> GSM97000 5 0.2129 0.67295 0.000 0.000 0.056 0.000 0.904 0.040
#> GSM97004 1 0.1075 0.79754 0.952 0.000 0.000 0.048 0.000 0.000
#> GSM97008 5 0.2066 0.67385 0.000 0.000 0.052 0.000 0.908 0.040
#> GSM96950 1 0.1765 0.77572 0.904 0.000 0.000 0.000 0.096 0.000
#> GSM96980 4 0.0692 0.96611 0.004 0.000 0.000 0.976 0.020 0.000
#> GSM96989 1 0.0000 0.81994 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM96992 1 0.0363 0.81811 0.988 0.000 0.000 0.012 0.000 0.000
#> GSM96993 1 0.5371 0.12864 0.520 0.000 0.360 0.000 0.120 0.000
#> GSM96958 1 0.3288 0.61839 0.724 0.000 0.000 0.000 0.276 0.000
#> GSM96951 1 0.0865 0.80175 0.964 0.000 0.000 0.000 0.036 0.000
#> GSM96952 1 0.0260 0.81900 0.992 0.000 0.000 0.008 0.000 0.000
#> GSM96961 1 0.0000 0.81994 1.000 0.000 0.000 0.000 0.000 0.000
Heatmaps for the consensus matrix. It visualizes the probability of two samples to be in a same group.
consensus_heatmap(res, k = 2)
consensus_heatmap(res, k = 3)
consensus_heatmap(res, k = 4)
consensus_heatmap(res, k = 5)
consensus_heatmap(res, k = 6)
Heatmaps for the membership of samples in all partitions to see how consistent they are:
membership_heatmap(res, k = 2)
membership_heatmap(res, k = 3)
membership_heatmap(res, k = 4)
membership_heatmap(res, k = 5)
membership_heatmap(res, k = 6)
As soon as we have had the classes for columns, we can look for signatures which are significantly different between classes which can be candidate marks for certain classes. Following are the heatmaps for signatures.
Signature heatmaps where rows are scaled:
get_signatures(res, k = 2)
get_signatures(res, k = 3)
get_signatures(res, k = 4)
get_signatures(res, k = 5)
get_signatures(res, k = 6)
Signature heatmaps where rows are not scaled:
get_signatures(res, k = 2, scale_rows = FALSE)
get_signatures(res, k = 3, scale_rows = FALSE)
get_signatures(res, k = 4, scale_rows = FALSE)
get_signatures(res, k = 5, scale_rows = FALSE)
get_signatures(res, k = 6, scale_rows = FALSE)
Compare the overlap of signatures from different k:
compare_signatures(res)
get_signature() returns a data frame invisibly. TO get the list of signatures, the function
call should be assigned to a variable explicitly. In following code, if plot argument is set
to FALSE, no heatmap is plotted while only the differential analysis is performed.
# code only for demonstration
tb = get_signature(res, k = ..., plot = FALSE)
An example of the output of tb is:
#> which_row fdr mean_1 mean_2 scaled_mean_1 scaled_mean_2 km
#> 1 38 0.042760348 8.373488 9.131774 -0.5533452 0.5164555 1
#> 2 40 0.018707592 7.106213 8.469186 -0.6173731 0.5762149 1
#> 3 55 0.019134737 10.221463 11.207825 -0.6159697 0.5749050 1
#> 4 59 0.006059896 5.921854 7.869574 -0.6899429 0.6439467 1
#> 5 60 0.018055526 8.928898 10.211722 -0.6204761 0.5791110 1
#> 6 98 0.009384629 15.714769 14.887706 0.6635654 -0.6193277 2
...
The columns in tb are:
which_row: row indices corresponding to the input matrix.fdr: FDR for the differential test. mean_x: The mean value in group x.scaled_mean_x: The mean value in group x after rows are scaled.km: Row groups if k-means clustering is applied to rows.UMAP plot which shows how samples are separated.
dimension_reduction(res, k = 2, method = "UMAP")
dimension_reduction(res, k = 3, method = "UMAP")
dimension_reduction(res, k = 4, method = "UMAP")
dimension_reduction(res, k = 5, method = "UMAP")
dimension_reduction(res, k = 6, method = "UMAP")
Following heatmap shows how subgroups are split when increasing k:
collect_classes(res)
Test correlation between subgroups and known annotations. If the known annotation is numeric, one-way ANOVA test is applied, and if the known annotation is discrete, chi-squared contingency table test is applied.
test_to_known_factors(res)
#> n disease.state(p) specimen(p) cell.type(p) other(p) k
#> SD:pam 98 7.18e-09 0.530 2.47e-19 0.00887 2
#> SD:pam 46 3.33e-04 0.178 1.03e-10 0.06974 3
#> SD:pam 68 1.04e-04 0.478 5.69e-15 0.00165 4
#> SD:pam 80 1.45e-06 0.211 6.75e-17 0.01235 5
#> SD:pam 51 4.23e-03 0.140 2.03e-11 0.06232 6
If matrix rows can be associated to genes, consider to use functional_enrichment(res,
...) to perform function enrichment for the signature genes. See this vignette for more detailed explanations.
The object with results only for a single top-value method and a single partition method can be extracted as:
res = res_list["SD", "mclust"]
# you can also extract it by
# res = res_list["SD:mclust"]
A summary of res and all the functions that can be applied to it:
res
#> A 'ConsensusPartition' object with k = 2, 3, 4, 5, 6.
#> On a matrix with 21168 rows and 100 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 0.600 0.915 0.948 0.3655 0.642 0.642
#> 3 3 0.589 0.716 0.861 0.7197 0.708 0.549
#> 4 4 0.861 0.891 0.946 0.1553 0.889 0.705
#> 5 5 0.772 0.752 0.828 0.0727 0.937 0.785
#> 6 6 0.874 0.891 0.918 0.0640 0.906 0.626
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
#> GSM97038 2 0.6531 0.802 0.168 0.832
#> GSM97045 2 0.0376 0.945 0.004 0.996
#> GSM97047 1 0.5059 0.915 0.888 0.112
#> GSM97025 2 0.0376 0.945 0.004 0.996
#> GSM97030 1 0.5059 0.915 0.888 0.112
#> GSM97027 2 0.0376 0.945 0.004 0.996
#> GSM97033 2 0.0376 0.945 0.004 0.996
#> GSM97034 1 0.5059 0.915 0.888 0.112
#> GSM97020 2 0.0376 0.945 0.004 0.996
#> GSM97026 1 0.5059 0.915 0.888 0.112
#> GSM97012 2 0.0376 0.945 0.004 0.996
#> GSM97015 1 0.5059 0.915 0.888 0.112
#> GSM97016 2 0.0376 0.945 0.004 0.996
#> GSM97017 1 0.0000 0.942 1.000 0.000
#> GSM97019 2 0.0376 0.945 0.004 0.996
#> GSM97022 2 0.0376 0.945 0.004 0.996
#> GSM97035 2 0.0376 0.945 0.004 0.996
#> GSM97036 1 0.0000 0.942 1.000 0.000
#> GSM97039 2 0.0376 0.945 0.004 0.996
#> GSM97046 2 0.0376 0.945 0.004 0.996
#> GSM97023 1 0.0000 0.942 1.000 0.000
#> GSM97029 1 0.0000 0.942 1.000 0.000
#> GSM97043 2 0.6801 0.788 0.180 0.820
#> GSM97013 1 0.0000 0.942 1.000 0.000
#> GSM96956 2 0.9393 0.435 0.356 0.644
#> GSM97024 2 0.7056 0.773 0.192 0.808
#> GSM97032 1 0.5059 0.915 0.888 0.112
#> GSM97044 1 0.5178 0.915 0.884 0.116
#> GSM97049 2 0.0376 0.945 0.004 0.996
#> GSM96968 1 0.5059 0.915 0.888 0.112
#> GSM96971 1 0.5178 0.915 0.884 0.116
#> GSM96986 1 0.5178 0.915 0.884 0.116
#> GSM97003 1 0.0938 0.941 0.988 0.012
#> GSM96957 1 0.0000 0.942 1.000 0.000
#> GSM96960 1 0.0376 0.941 0.996 0.004
#> GSM96975 1 0.0000 0.942 1.000 0.000
#> GSM96998 1 0.0000 0.942 1.000 0.000
#> GSM96999 1 0.0000 0.942 1.000 0.000
#> GSM97001 1 0.0000 0.942 1.000 0.000
#> GSM97005 1 0.0000 0.942 1.000 0.000
#> GSM97006 1 0.0000 0.942 1.000 0.000
#> GSM97021 1 0.0000 0.942 1.000 0.000
#> GSM97028 1 0.5059 0.915 0.888 0.112
#> GSM97031 1 0.4298 0.923 0.912 0.088
#> GSM97037 1 0.9970 0.177 0.532 0.468
#> GSM97018 1 0.5059 0.915 0.888 0.112
#> GSM97014 1 0.4298 0.923 0.912 0.088
#> GSM97042 2 0.0376 0.945 0.004 0.996
#> GSM97040 1 0.4690 0.919 0.900 0.100
#> GSM97041 1 0.0000 0.942 1.000 0.000
#> GSM96955 2 0.7139 0.767 0.196 0.804
#> GSM96990 1 0.5059 0.915 0.888 0.112
#> GSM96991 2 0.0672 0.943 0.008 0.992
#> GSM97048 2 0.0376 0.945 0.004 0.996
#> GSM96963 2 0.0376 0.945 0.004 0.996
#> GSM96953 2 0.0376 0.945 0.004 0.996
#> GSM96966 1 0.0376 0.941 0.996 0.004
#> GSM96979 1 0.5178 0.915 0.884 0.116
#> GSM96983 1 0.5178 0.915 0.884 0.116
#> GSM96984 1 0.5178 0.915 0.884 0.116
#> GSM96994 1 0.5178 0.915 0.884 0.116
#> GSM96996 1 0.0000 0.942 1.000 0.000
#> GSM96997 1 0.5178 0.915 0.884 0.116
#> GSM97007 1 0.5178 0.915 0.884 0.116
#> GSM96954 1 0.4815 0.918 0.896 0.104
#> GSM96962 1 0.5178 0.915 0.884 0.116
#> GSM96969 1 0.0376 0.941 0.996 0.004
#> GSM96970 1 0.0376 0.941 0.996 0.004
#> GSM96973 1 0.0376 0.941 0.996 0.004
#> GSM96976 1 0.1843 0.938 0.972 0.028
#> GSM96977 1 0.0938 0.941 0.988 0.012
#> GSM96995 1 0.5059 0.915 0.888 0.112
#> GSM97002 1 0.0000 0.942 1.000 0.000
#> GSM97009 1 0.5059 0.915 0.888 0.112
#> GSM97010 1 0.0000 0.942 1.000 0.000
#> GSM96974 1 0.1843 0.938 0.972 0.028
#> GSM96985 1 0.1633 0.939 0.976 0.024
#> GSM96959 1 0.4815 0.918 0.896 0.104
#> GSM96972 1 0.0376 0.941 0.996 0.004
#> GSM96978 1 0.5178 0.915 0.884 0.116
#> GSM96967 1 0.0376 0.941 0.996 0.004
#> GSM96987 1 0.0000 0.942 1.000 0.000
#> GSM97011 1 0.0000 0.942 1.000 0.000
#> GSM96964 1 0.0000 0.942 1.000 0.000
#> GSM96965 1 0.0376 0.941 0.996 0.004
#> GSM96981 1 0.0000 0.942 1.000 0.000
#> GSM96982 1 0.0376 0.941 0.996 0.004
#> GSM96988 1 0.5178 0.915 0.884 0.116
#> GSM97000 1 0.4690 0.919 0.900 0.100
#> GSM97004 1 0.0376 0.941 0.996 0.004
#> GSM97008 1 0.4298 0.923 0.912 0.088
#> GSM96950 1 0.0000 0.942 1.000 0.000
#> GSM96980 1 0.0376 0.941 0.996 0.004
#> GSM96989 1 0.0000 0.942 1.000 0.000
#> GSM96992 1 0.0000 0.942 1.000 0.000
#> GSM96993 1 0.0000 0.942 1.000 0.000
#> GSM96958 1 0.0000 0.942 1.000 0.000
#> GSM96951 1 0.0000 0.942 1.000 0.000
#> GSM96952 1 0.0000 0.942 1.000 0.000
#> GSM96961 1 0.0000 0.942 1.000 0.000
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM97038 2 0.0747 0.9255 0.000 0.984 0.016
#> GSM97045 2 0.0000 0.9374 0.000 1.000 0.000
#> GSM97047 3 0.7458 0.6114 0.088 0.236 0.676
#> GSM97025 2 0.0000 0.9374 0.000 1.000 0.000
#> GSM97030 3 0.1711 0.8681 0.008 0.032 0.960
#> GSM97027 2 0.0000 0.9374 0.000 1.000 0.000
#> GSM97033 2 0.0000 0.9374 0.000 1.000 0.000
#> GSM97034 3 0.1585 0.8689 0.008 0.028 0.964
#> GSM97020 2 0.0000 0.9374 0.000 1.000 0.000
#> GSM97026 3 0.9547 0.2819 0.228 0.292 0.480
#> GSM97012 2 0.0000 0.9374 0.000 1.000 0.000
#> GSM97015 3 0.1711 0.8681 0.008 0.032 0.960
#> GSM97016 2 0.0000 0.9374 0.000 1.000 0.000
#> GSM97017 1 0.2796 0.7659 0.908 0.000 0.092
#> GSM97019 2 0.0000 0.9374 0.000 1.000 0.000
#> GSM97022 2 0.0000 0.9374 0.000 1.000 0.000
#> GSM97035 2 0.0000 0.9374 0.000 1.000 0.000
#> GSM97036 1 0.2448 0.7677 0.924 0.000 0.076
#> GSM97039 2 0.0000 0.9374 0.000 1.000 0.000
#> GSM97046 2 0.0000 0.9374 0.000 1.000 0.000
#> GSM97023 1 0.0747 0.7551 0.984 0.000 0.016
#> GSM97029 1 0.2796 0.7659 0.908 0.000 0.092
#> GSM97043 2 0.3192 0.8412 0.000 0.888 0.112
#> GSM97013 1 0.2711 0.7668 0.912 0.000 0.088
#> GSM96956 2 0.6398 0.2400 0.004 0.580 0.416
#> GSM97024 2 0.4110 0.8044 0.004 0.844 0.152
#> GSM97032 3 0.2866 0.8414 0.008 0.076 0.916
#> GSM97044 3 0.1585 0.8689 0.008 0.028 0.964
#> GSM97049 2 0.0000 0.9374 0.000 1.000 0.000
#> GSM96968 3 0.2187 0.8644 0.024 0.028 0.948
#> GSM96971 3 0.0424 0.8667 0.008 0.000 0.992
#> GSM96986 3 0.0424 0.8667 0.008 0.000 0.992
#> GSM97003 1 0.6267 0.3926 0.548 0.000 0.452
#> GSM96957 1 0.2796 0.7659 0.908 0.000 0.092
#> GSM96960 1 0.6062 0.4678 0.616 0.000 0.384
#> GSM96975 1 0.2711 0.7669 0.912 0.000 0.088
#> GSM96998 1 0.0000 0.7482 1.000 0.000 0.000
#> GSM96999 1 0.2796 0.7659 0.908 0.000 0.092
#> GSM97001 1 0.2796 0.7659 0.908 0.000 0.092
#> GSM97005 1 0.2625 0.7674 0.916 0.000 0.084
#> GSM97006 1 0.5988 0.4722 0.632 0.000 0.368
#> GSM97021 1 0.3610 0.7575 0.888 0.016 0.096
#> GSM97028 3 0.1453 0.8692 0.008 0.024 0.968
#> GSM97031 1 0.6925 0.3767 0.532 0.016 0.452
#> GSM97037 2 0.6633 0.2356 0.008 0.548 0.444
#> GSM97018 3 0.2845 0.8470 0.012 0.068 0.920
#> GSM97014 1 0.9555 0.3079 0.480 0.232 0.288
#> GSM97042 2 0.0000 0.9374 0.000 1.000 0.000
#> GSM97040 1 0.4995 0.7060 0.824 0.032 0.144
#> GSM97041 1 0.2796 0.7659 0.908 0.000 0.092
#> GSM96955 2 0.3941 0.7938 0.000 0.844 0.156
#> GSM96990 3 0.1711 0.8681 0.008 0.032 0.960
#> GSM96991 2 0.0424 0.9317 0.000 0.992 0.008
#> GSM97048 2 0.0000 0.9374 0.000 1.000 0.000
#> GSM96963 2 0.0000 0.9374 0.000 1.000 0.000
#> GSM96953 2 0.0000 0.9374 0.000 1.000 0.000
#> GSM96966 1 0.6026 0.4710 0.624 0.000 0.376
#> GSM96979 3 0.4399 0.6606 0.188 0.000 0.812
#> GSM96983 3 0.0424 0.8667 0.008 0.000 0.992
#> GSM96984 3 0.0424 0.8667 0.008 0.000 0.992
#> GSM96994 3 0.0424 0.8667 0.008 0.000 0.992
#> GSM96996 1 0.0000 0.7482 1.000 0.000 0.000
#> GSM96997 3 0.0424 0.8667 0.008 0.000 0.992
#> GSM97007 3 0.0424 0.8667 0.008 0.000 0.992
#> GSM96954 3 0.2031 0.8679 0.016 0.032 0.952
#> GSM96962 3 0.0424 0.8667 0.008 0.000 0.992
#> GSM96969 1 0.6026 0.4710 0.624 0.000 0.376
#> GSM96970 1 0.6026 0.4710 0.624 0.000 0.376
#> GSM96973 1 0.6026 0.4710 0.624 0.000 0.376
#> GSM96976 3 0.6852 0.3807 0.300 0.036 0.664
#> GSM96977 1 0.4999 0.7037 0.820 0.028 0.152
#> GSM96995 3 0.2176 0.8661 0.020 0.032 0.948
#> GSM97002 1 0.3752 0.7020 0.856 0.000 0.144
#> GSM97009 1 0.9027 0.1561 0.440 0.132 0.428
#> GSM97010 1 0.5591 0.6186 0.696 0.000 0.304
#> GSM96974 3 0.6235 -0.0878 0.436 0.000 0.564
#> GSM96985 3 0.6309 -0.3083 0.496 0.000 0.504
#> GSM96959 3 0.1832 0.8667 0.008 0.036 0.956
#> GSM96972 1 0.6026 0.4710 0.624 0.000 0.376
#> GSM96978 3 0.0424 0.8667 0.008 0.000 0.992
#> GSM96967 1 0.6026 0.4710 0.624 0.000 0.376
#> GSM96987 1 0.0592 0.7544 0.988 0.000 0.012
#> GSM97011 1 0.3670 0.7560 0.888 0.020 0.092
#> GSM96964 1 0.1411 0.7616 0.964 0.000 0.036
#> GSM96965 1 0.6244 0.4139 0.560 0.000 0.440
#> GSM96981 1 0.0892 0.7579 0.980 0.000 0.020
#> GSM96982 1 0.6026 0.4778 0.624 0.000 0.376
#> GSM96988 3 0.3340 0.7615 0.120 0.000 0.880
#> GSM97000 1 0.7286 0.3301 0.508 0.028 0.464
#> GSM97004 1 0.6008 0.4715 0.628 0.000 0.372
#> GSM97008 1 0.3966 0.7499 0.876 0.024 0.100
#> GSM96950 1 0.2625 0.7673 0.916 0.000 0.084
#> GSM96980 1 0.6026 0.4710 0.624 0.000 0.376
#> GSM96989 1 0.0747 0.7564 0.984 0.000 0.016
#> GSM96992 1 0.0747 0.7551 0.984 0.000 0.016
#> GSM96993 1 0.2796 0.7659 0.908 0.000 0.092
#> GSM96958 1 0.2165 0.7670 0.936 0.000 0.064
#> GSM96951 1 0.2066 0.7665 0.940 0.000 0.060
#> GSM96952 1 0.0747 0.7551 0.984 0.000 0.016
#> GSM96961 1 0.0747 0.7551 0.984 0.000 0.016
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM97038 2 0.0000 0.9542 0.000 1.000 0.000 0.000
#> GSM97045 2 0.0000 0.9542 0.000 1.000 0.000 0.000
#> GSM97047 2 0.5204 0.4346 0.376 0.612 0.012 0.000
#> GSM97025 2 0.0000 0.9542 0.000 1.000 0.000 0.000
#> GSM97030 3 0.0000 0.9550 0.000 0.000 1.000 0.000
#> GSM97027 2 0.0000 0.9542 0.000 1.000 0.000 0.000
#> GSM97033 2 0.0000 0.9542 0.000 1.000 0.000 0.000
#> GSM97034 3 0.0000 0.9550 0.000 0.000 1.000 0.000
#> GSM97020 2 0.0000 0.9542 0.000 1.000 0.000 0.000
#> GSM97026 2 0.4250 0.6261 0.276 0.724 0.000 0.000
#> GSM97012 2 0.0000 0.9542 0.000 1.000 0.000 0.000
#> GSM97015 3 0.0000 0.9550 0.000 0.000 1.000 0.000
#> GSM97016 2 0.0000 0.9542 0.000 1.000 0.000 0.000
#> GSM97017 1 0.0000 0.9162 1.000 0.000 0.000 0.000
#> GSM97019 2 0.0000 0.9542 0.000 1.000 0.000 0.000
#> GSM97022 2 0.0000 0.9542 0.000 1.000 0.000 0.000
#> GSM97035 2 0.0000 0.9542 0.000 1.000 0.000 0.000
#> GSM97036 1 0.0000 0.9162 1.000 0.000 0.000 0.000
#> GSM97039 2 0.0000 0.9542 0.000 1.000 0.000 0.000
#> GSM97046 2 0.0000 0.9542 0.000 1.000 0.000 0.000
#> GSM97023 1 0.2345 0.8862 0.900 0.000 0.000 0.100
#> GSM97029 1 0.0000 0.9162 1.000 0.000 0.000 0.000
#> GSM97043 2 0.0000 0.9542 0.000 1.000 0.000 0.000
#> GSM97013 1 0.0000 0.9162 1.000 0.000 0.000 0.000
#> GSM96956 2 0.0921 0.9308 0.000 0.972 0.028 0.000
#> GSM97024 2 0.0592 0.9423 0.000 0.984 0.016 0.000
#> GSM97032 3 0.3726 0.7087 0.000 0.212 0.788 0.000
#> GSM97044 3 0.0000 0.9550 0.000 0.000 1.000 0.000
#> GSM97049 2 0.0000 0.9542 0.000 1.000 0.000 0.000
#> GSM96968 3 0.1474 0.9134 0.052 0.000 0.948 0.000
#> GSM96971 3 0.0000 0.9550 0.000 0.000 1.000 0.000
#> GSM96986 3 0.0000 0.9550 0.000 0.000 1.000 0.000
#> GSM97003 1 0.3356 0.8449 0.824 0.000 0.000 0.176
#> GSM96957 1 0.0000 0.9162 1.000 0.000 0.000 0.000
#> GSM96960 1 0.3610 0.8270 0.800 0.000 0.000 0.200
#> GSM96975 1 0.0000 0.9162 1.000 0.000 0.000 0.000
#> GSM96998 1 0.3610 0.8270 0.800 0.000 0.000 0.200
#> GSM96999 1 0.0000 0.9162 1.000 0.000 0.000 0.000
#> GSM97001 1 0.0000 0.9162 1.000 0.000 0.000 0.000
#> GSM97005 1 0.0000 0.9162 1.000 0.000 0.000 0.000
#> GSM97006 1 0.3649 0.8233 0.796 0.000 0.000 0.204
#> GSM97021 1 0.0000 0.9162 1.000 0.000 0.000 0.000
#> GSM97028 3 0.0000 0.9550 0.000 0.000 1.000 0.000
#> GSM97031 1 0.1557 0.9043 0.944 0.000 0.000 0.056
#> GSM97037 2 0.3649 0.7280 0.000 0.796 0.204 0.000
#> GSM97018 3 0.3610 0.7257 0.000 0.200 0.800 0.000
#> GSM97014 1 0.4967 0.0513 0.548 0.452 0.000 0.000
#> GSM97042 2 0.0000 0.9542 0.000 1.000 0.000 0.000
#> GSM97040 1 0.0000 0.9162 1.000 0.000 0.000 0.000
#> GSM97041 1 0.0000 0.9162 1.000 0.000 0.000 0.000
#> GSM96955 2 0.1302 0.9150 0.044 0.956 0.000 0.000
#> GSM96990 3 0.0000 0.9550 0.000 0.000 1.000 0.000
#> GSM96991 2 0.0000 0.9542 0.000 1.000 0.000 0.000
#> GSM97048 2 0.0000 0.9542 0.000 1.000 0.000 0.000
#> GSM96963 2 0.0000 0.9542 0.000 1.000 0.000 0.000
#> GSM96953 2 0.0000 0.9542 0.000 1.000 0.000 0.000
#> GSM96966 4 0.0000 0.9213 0.000 0.000 0.000 1.000
#> GSM96979 3 0.0469 0.9460 0.000 0.000 0.988 0.012
#> GSM96983 3 0.0000 0.9550 0.000 0.000 1.000 0.000
#> GSM96984 3 0.0000 0.9550 0.000 0.000 1.000 0.000
#> GSM96994 3 0.0000 0.9550 0.000 0.000 1.000 0.000
#> GSM96996 1 0.3610 0.8270 0.800 0.000 0.000 0.200
#> GSM96997 3 0.0000 0.9550 0.000 0.000 1.000 0.000
#> GSM97007 3 0.0000 0.9550 0.000 0.000 1.000 0.000
#> GSM96954 3 0.1637 0.9053 0.060 0.000 0.940 0.000
#> GSM96962 3 0.0000 0.9550 0.000 0.000 1.000 0.000
#> GSM96969 4 0.0000 0.9213 0.000 0.000 0.000 1.000
#> GSM96970 4 0.0000 0.9213 0.000 0.000 0.000 1.000
#> GSM96973 4 0.0000 0.9213 0.000 0.000 0.000 1.000
#> GSM96976 4 0.3668 0.7759 0.000 0.004 0.188 0.808
#> GSM96977 1 0.0000 0.9162 1.000 0.000 0.000 0.000
#> GSM96995 3 0.1302 0.9212 0.044 0.000 0.956 0.000
#> GSM97002 1 0.3764 0.8114 0.784 0.000 0.000 0.216
#> GSM97009 1 0.0921 0.9037 0.972 0.028 0.000 0.000
#> GSM97010 1 0.0469 0.9123 0.988 0.000 0.000 0.012
#> GSM96974 4 0.3486 0.7779 0.000 0.000 0.188 0.812
#> GSM96985 4 0.3219 0.8046 0.000 0.000 0.164 0.836
#> GSM96959 3 0.3444 0.7421 0.184 0.000 0.816 0.000
#> GSM96972 4 0.0000 0.9213 0.000 0.000 0.000 1.000
#> GSM96978 3 0.0000 0.9550 0.000 0.000 1.000 0.000
#> GSM96967 4 0.0000 0.9213 0.000 0.000 0.000 1.000
#> GSM96987 1 0.3219 0.8521 0.836 0.000 0.000 0.164
#> GSM97011 1 0.0000 0.9162 1.000 0.000 0.000 0.000
#> GSM96964 1 0.1118 0.9098 0.964 0.000 0.000 0.036
#> GSM96965 4 0.3528 0.7590 0.192 0.000 0.000 0.808
#> GSM96981 1 0.0707 0.9135 0.980 0.000 0.000 0.020
#> GSM96982 1 0.3356 0.8448 0.824 0.000 0.000 0.176
#> GSM96988 3 0.0000 0.9550 0.000 0.000 1.000 0.000
#> GSM97000 1 0.0000 0.9162 1.000 0.000 0.000 0.000
#> GSM97004 4 0.0000 0.9213 0.000 0.000 0.000 1.000
#> GSM97008 1 0.0000 0.9162 1.000 0.000 0.000 0.000
#> GSM96950 1 0.0000 0.9162 1.000 0.000 0.000 0.000
#> GSM96980 4 0.0000 0.9213 0.000 0.000 0.000 1.000
#> GSM96989 1 0.1867 0.8989 0.928 0.000 0.000 0.072
#> GSM96992 1 0.3486 0.8363 0.812 0.000 0.000 0.188
#> GSM96993 1 0.0000 0.9162 1.000 0.000 0.000 0.000
#> GSM96958 1 0.0188 0.9159 0.996 0.000 0.000 0.004
#> GSM96951 1 0.1940 0.8969 0.924 0.000 0.000 0.076
#> GSM96952 1 0.3486 0.8363 0.812 0.000 0.000 0.188
#> GSM96961 1 0.3444 0.8391 0.816 0.000 0.000 0.184
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM97038 2 0.0162 0.9468 0.000 0.996 0.000 0.000 0.004
#> GSM97045 2 0.0000 0.9491 0.000 1.000 0.000 0.000 0.000
#> GSM97047 5 0.4118 0.5128 0.336 0.000 0.004 0.000 0.660
#> GSM97025 2 0.0000 0.9491 0.000 1.000 0.000 0.000 0.000
#> GSM97030 5 0.4015 0.6468 0.000 0.000 0.348 0.000 0.652
#> GSM97027 2 0.0000 0.9491 0.000 1.000 0.000 0.000 0.000
#> GSM97033 2 0.0000 0.9491 0.000 1.000 0.000 0.000 0.000
#> GSM97034 5 0.4045 0.6386 0.000 0.000 0.356 0.000 0.644
#> GSM97020 2 0.0000 0.9491 0.000 1.000 0.000 0.000 0.000
#> GSM97026 2 0.6711 0.0785 0.336 0.444 0.004 0.000 0.216
#> GSM97012 2 0.0290 0.9479 0.000 0.992 0.000 0.000 0.008
#> GSM97015 5 0.3983 0.6514 0.000 0.000 0.340 0.000 0.660
#> GSM97016 2 0.0000 0.9491 0.000 1.000 0.000 0.000 0.000
#> GSM97017 1 0.0000 0.7587 1.000 0.000 0.000 0.000 0.000
#> GSM97019 2 0.0290 0.9479 0.000 0.992 0.000 0.000 0.008
#> GSM97022 2 0.0290 0.9479 0.000 0.992 0.000 0.000 0.008
#> GSM97035 2 0.0290 0.9479 0.000 0.992 0.000 0.000 0.008
#> GSM97036 1 0.0609 0.7594 0.980 0.000 0.000 0.000 0.020
#> GSM97039 2 0.0000 0.9491 0.000 1.000 0.000 0.000 0.000
#> GSM97046 2 0.0000 0.9491 0.000 1.000 0.000 0.000 0.000
#> GSM97023 1 0.4925 0.6987 0.632 0.000 0.000 0.044 0.324
#> GSM97029 1 0.0162 0.7593 0.996 0.000 0.000 0.000 0.004
#> GSM97043 2 0.2690 0.7746 0.000 0.844 0.000 0.000 0.156
#> GSM97013 1 0.0162 0.7593 0.996 0.000 0.000 0.000 0.004
#> GSM96956 5 0.4403 0.3855 0.000 0.384 0.008 0.000 0.608
#> GSM97024 5 0.4151 0.4610 0.000 0.344 0.004 0.000 0.652
#> GSM97032 5 0.4836 0.6630 0.000 0.044 0.304 0.000 0.652
#> GSM97044 5 0.4287 0.4640 0.000 0.000 0.460 0.000 0.540
#> GSM97049 2 0.0000 0.9491 0.000 1.000 0.000 0.000 0.000
#> GSM96968 5 0.5708 0.5657 0.096 0.000 0.348 0.000 0.556
#> GSM96971 3 0.0000 0.9374 0.000 0.000 1.000 0.000 0.000
#> GSM96986 3 0.0000 0.9374 0.000 0.000 1.000 0.000 0.000
#> GSM97003 1 0.6461 0.6105 0.492 0.000 0.004 0.172 0.332
#> GSM96957 1 0.0162 0.7577 0.996 0.000 0.000 0.000 0.004
#> GSM96960 1 0.6482 0.5861 0.468 0.000 0.000 0.200 0.332
#> GSM96975 1 0.1197 0.7582 0.952 0.000 0.000 0.000 0.048
#> GSM96998 1 0.6461 0.5911 0.472 0.000 0.000 0.196 0.332
#> GSM96999 1 0.0000 0.7587 1.000 0.000 0.000 0.000 0.000
#> GSM97001 1 0.0162 0.7577 0.996 0.000 0.000 0.000 0.004
#> GSM97005 1 0.0162 0.7577 0.996 0.000 0.000 0.000 0.004
#> GSM97006 1 0.6461 0.5911 0.472 0.000 0.000 0.196 0.332
#> GSM97021 1 0.0162 0.7577 0.996 0.000 0.000 0.000 0.004
#> GSM97028 3 0.3932 0.2593 0.000 0.000 0.672 0.000 0.328
#> GSM97031 1 0.4602 0.7086 0.656 0.000 0.000 0.028 0.316
#> GSM97037 5 0.4339 0.4745 0.000 0.336 0.012 0.000 0.652
#> GSM97018 5 0.4716 0.6641 0.000 0.036 0.308 0.000 0.656
#> GSM97014 1 0.3530 0.5117 0.784 0.204 0.000 0.000 0.012
#> GSM97042 2 0.0290 0.9479 0.000 0.992 0.000 0.000 0.008
#> GSM97040 1 0.0609 0.7493 0.980 0.000 0.000 0.000 0.020
#> GSM97041 1 0.0000 0.7587 1.000 0.000 0.000 0.000 0.000
#> GSM96955 2 0.3317 0.7186 0.004 0.804 0.004 0.000 0.188
#> GSM96990 5 0.4015 0.6468 0.000 0.000 0.348 0.000 0.652
#> GSM96991 2 0.0290 0.9479 0.000 0.992 0.000 0.000 0.008
#> GSM97048 2 0.0000 0.9491 0.000 1.000 0.000 0.000 0.000
#> GSM96963 2 0.0162 0.9485 0.000 0.996 0.000 0.000 0.004
#> GSM96953 2 0.0290 0.9479 0.000 0.992 0.000 0.000 0.008
#> GSM96966 4 0.0162 0.8608 0.000 0.000 0.000 0.996 0.004
#> GSM96979 3 0.0000 0.9374 0.000 0.000 1.000 0.000 0.000
#> GSM96983 3 0.0000 0.9374 0.000 0.000 1.000 0.000 0.000
#> GSM96984 3 0.0000 0.9374 0.000 0.000 1.000 0.000 0.000
#> GSM96994 3 0.0000 0.9374 0.000 0.000 1.000 0.000 0.000
#> GSM96996 1 0.6461 0.5911 0.472 0.000 0.000 0.196 0.332
#> GSM96997 3 0.0000 0.9374 0.000 0.000 1.000 0.000 0.000
#> GSM97007 3 0.0000 0.9374 0.000 0.000 1.000 0.000 0.000
#> GSM96954 3 0.4057 0.6387 0.120 0.000 0.792 0.000 0.088
#> GSM96962 3 0.0000 0.9374 0.000 0.000 1.000 0.000 0.000
#> GSM96969 4 0.0000 0.8614 0.000 0.000 0.000 1.000 0.000
#> GSM96970 4 0.0162 0.8608 0.000 0.000 0.000 0.996 0.004
#> GSM96973 4 0.0000 0.8614 0.000 0.000 0.000 1.000 0.000
#> GSM96976 4 0.3391 0.7323 0.000 0.000 0.188 0.800 0.012
#> GSM96977 1 0.0162 0.7577 0.996 0.000 0.000 0.000 0.004
#> GSM96995 5 0.4642 0.6612 0.032 0.000 0.308 0.000 0.660
#> GSM97002 1 0.6482 0.5861 0.468 0.000 0.000 0.200 0.332
#> GSM97009 5 0.4545 0.4011 0.432 0.004 0.004 0.000 0.560
#> GSM97010 1 0.0609 0.7594 0.980 0.000 0.000 0.000 0.020
#> GSM96974 4 0.3109 0.7281 0.000 0.000 0.200 0.800 0.000
#> GSM96985 4 0.3109 0.7281 0.000 0.000 0.200 0.800 0.000
#> GSM96959 5 0.5414 0.5993 0.200 0.000 0.140 0.000 0.660
#> GSM96972 4 0.0000 0.8614 0.000 0.000 0.000 1.000 0.000
#> GSM96978 3 0.0000 0.9374 0.000 0.000 1.000 0.000 0.000
#> GSM96967 4 0.0000 0.8614 0.000 0.000 0.000 1.000 0.000
#> GSM96987 1 0.5701 0.6658 0.568 0.000 0.000 0.100 0.332
#> GSM97011 1 0.0162 0.7577 0.996 0.000 0.000 0.000 0.004
#> GSM96964 1 0.4201 0.7102 0.664 0.000 0.000 0.008 0.328
#> GSM96965 4 0.3109 0.6871 0.200 0.000 0.000 0.800 0.000
#> GSM96981 1 0.3949 0.7113 0.668 0.000 0.000 0.000 0.332
#> GSM96982 1 0.5865 0.6557 0.552 0.000 0.000 0.116 0.332
#> GSM96988 3 0.0000 0.9374 0.000 0.000 1.000 0.000 0.000
#> GSM97000 1 0.0162 0.7577 0.996 0.000 0.000 0.000 0.004
#> GSM97004 4 0.3816 0.6045 0.000 0.000 0.000 0.696 0.304
#> GSM97008 1 0.0162 0.7577 0.996 0.000 0.000 0.000 0.004
#> GSM96950 1 0.0609 0.7602 0.980 0.000 0.000 0.000 0.020
#> GSM96980 4 0.2605 0.7640 0.000 0.000 0.000 0.852 0.148
#> GSM96989 1 0.4747 0.7011 0.636 0.000 0.000 0.032 0.332
#> GSM96992 1 0.6461 0.5911 0.472 0.000 0.000 0.196 0.332
#> GSM96993 1 0.0000 0.7587 1.000 0.000 0.000 0.000 0.000
#> GSM96958 1 0.3932 0.7121 0.672 0.000 0.000 0.000 0.328
#> GSM96951 1 0.4522 0.7088 0.660 0.000 0.000 0.024 0.316
#> GSM96952 1 0.6461 0.5911 0.472 0.000 0.000 0.196 0.332
#> GSM96961 1 0.6205 0.6262 0.512 0.000 0.000 0.156 0.332
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM97038 2 0.1332 0.947 0.012 0.952 0.028 0.008 0.000 0.000
#> GSM97045 2 0.0363 0.960 0.012 0.988 0.000 0.000 0.000 0.000
#> GSM97047 3 0.2070 0.859 0.000 0.000 0.892 0.008 0.100 0.000
#> GSM97025 2 0.0363 0.960 0.012 0.988 0.000 0.000 0.000 0.000
#> GSM97030 3 0.0937 0.938 0.000 0.000 0.960 0.000 0.000 0.040
#> GSM97027 2 0.0508 0.960 0.012 0.984 0.000 0.004 0.000 0.000
#> GSM97033 2 0.0622 0.958 0.012 0.980 0.000 0.008 0.000 0.000
#> GSM97034 3 0.1007 0.937 0.000 0.000 0.956 0.000 0.000 0.044
#> GSM97020 2 0.0622 0.958 0.012 0.980 0.000 0.008 0.000 0.000
#> GSM97026 5 0.3679 0.781 0.012 0.012 0.160 0.020 0.796 0.000
#> GSM97012 2 0.1138 0.957 0.004 0.960 0.012 0.024 0.000 0.000
#> GSM97015 3 0.0937 0.938 0.000 0.000 0.960 0.000 0.000 0.040
#> GSM97016 2 0.0260 0.960 0.000 0.992 0.000 0.008 0.000 0.000
#> GSM97017 5 0.0000 0.941 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM97019 2 0.1036 0.958 0.004 0.964 0.008 0.024 0.000 0.000
#> GSM97022 2 0.1036 0.958 0.004 0.964 0.008 0.024 0.000 0.000
#> GSM97035 2 0.1036 0.958 0.004 0.964 0.008 0.024 0.000 0.000
#> GSM97036 5 0.0551 0.939 0.004 0.000 0.004 0.008 0.984 0.000
#> GSM97039 2 0.0260 0.960 0.000 0.992 0.000 0.008 0.000 0.000
#> GSM97046 2 0.0665 0.960 0.004 0.980 0.008 0.008 0.000 0.000
#> GSM97023 1 0.2178 0.882 0.868 0.000 0.000 0.000 0.132 0.000
#> GSM97029 5 0.0551 0.939 0.004 0.000 0.004 0.008 0.984 0.000
#> GSM97043 2 0.3046 0.785 0.012 0.800 0.188 0.000 0.000 0.000
#> GSM97013 5 0.0405 0.940 0.000 0.000 0.004 0.008 0.988 0.000
#> GSM96956 3 0.1007 0.913 0.000 0.044 0.956 0.000 0.000 0.000
#> GSM97024 3 0.1010 0.917 0.000 0.036 0.960 0.004 0.000 0.000
#> GSM97032 3 0.0937 0.938 0.000 0.000 0.960 0.000 0.000 0.040
#> GSM97044 3 0.1007 0.937 0.000 0.000 0.956 0.000 0.000 0.044
#> GSM97049 2 0.0622 0.958 0.012 0.980 0.000 0.008 0.000 0.000
#> GSM96968 3 0.5239 0.616 0.000 0.000 0.640 0.116 0.016 0.228
#> GSM96971 6 0.0000 0.944 0.000 0.000 0.000 0.000 0.000 1.000
#> GSM96986 6 0.0000 0.944 0.000 0.000 0.000 0.000 0.000 1.000
#> GSM97003 1 0.1036 0.894 0.964 0.000 0.000 0.004 0.024 0.008
#> GSM96957 5 0.0000 0.941 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM96960 1 0.0717 0.889 0.976 0.000 0.000 0.008 0.016 0.000
#> GSM96975 5 0.1895 0.875 0.072 0.000 0.016 0.000 0.912 0.000
#> GSM96998 1 0.1321 0.892 0.952 0.000 0.024 0.004 0.020 0.000
#> GSM96999 5 0.0146 0.941 0.004 0.000 0.000 0.000 0.996 0.000
#> GSM97001 5 0.0000 0.941 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM97005 5 0.0972 0.929 0.028 0.000 0.000 0.008 0.964 0.000
#> GSM97006 1 0.1245 0.878 0.952 0.000 0.000 0.032 0.016 0.000
#> GSM97021 5 0.1349 0.922 0.004 0.000 0.000 0.056 0.940 0.000
#> GSM97028 6 0.4025 0.211 0.000 0.000 0.416 0.008 0.000 0.576
#> GSM97031 1 0.2821 0.879 0.860 0.000 0.004 0.040 0.096 0.000
#> GSM97037 3 0.0891 0.926 0.000 0.024 0.968 0.000 0.000 0.008
#> GSM97018 3 0.1049 0.937 0.000 0.008 0.960 0.000 0.000 0.032
#> GSM97014 5 0.0622 0.933 0.012 0.008 0.000 0.000 0.980 0.000
#> GSM97042 2 0.1138 0.957 0.004 0.960 0.012 0.024 0.000 0.000
#> GSM97040 5 0.2146 0.887 0.000 0.000 0.004 0.116 0.880 0.000
#> GSM97041 5 0.0000 0.941 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM96955 2 0.3292 0.770 0.008 0.784 0.200 0.008 0.000 0.000
#> GSM96990 3 0.0937 0.938 0.000 0.000 0.960 0.000 0.000 0.040
#> GSM96991 2 0.1138 0.957 0.004 0.960 0.012 0.024 0.000 0.000
#> GSM97048 2 0.0622 0.958 0.012 0.980 0.000 0.008 0.000 0.000
#> GSM96963 2 0.1138 0.957 0.004 0.960 0.012 0.024 0.000 0.000
#> GSM96953 2 0.1036 0.958 0.004 0.964 0.008 0.024 0.000 0.000
#> GSM96966 4 0.2491 0.873 0.164 0.000 0.000 0.836 0.000 0.000
#> GSM96979 6 0.0000 0.944 0.000 0.000 0.000 0.000 0.000 1.000
#> GSM96983 6 0.0000 0.944 0.000 0.000 0.000 0.000 0.000 1.000
#> GSM96984 6 0.0000 0.944 0.000 0.000 0.000 0.000 0.000 1.000
#> GSM96994 6 0.0000 0.944 0.000 0.000 0.000 0.000 0.000 1.000
#> GSM96996 1 0.1630 0.886 0.940 0.000 0.024 0.016 0.020 0.000
#> GSM96997 6 0.0000 0.944 0.000 0.000 0.000 0.000 0.000 1.000
#> GSM97007 6 0.0000 0.944 0.000 0.000 0.000 0.000 0.000 1.000
#> GSM96954 6 0.3441 0.785 0.004 0.000 0.020 0.116 0.032 0.828
#> GSM96962 6 0.0000 0.944 0.000 0.000 0.000 0.000 0.000 1.000
#> GSM96969 4 0.2416 0.876 0.156 0.000 0.000 0.844 0.000 0.000
#> GSM96970 4 0.2454 0.875 0.160 0.000 0.000 0.840 0.000 0.000
#> GSM96973 4 0.2416 0.876 0.156 0.000 0.000 0.844 0.000 0.000
#> GSM96976 4 0.3037 0.762 0.000 0.000 0.016 0.808 0.000 0.176
#> GSM96977 5 0.2146 0.888 0.004 0.000 0.000 0.116 0.880 0.000
#> GSM96995 3 0.2579 0.894 0.000 0.000 0.872 0.088 0.000 0.040
#> GSM97002 1 0.1542 0.885 0.944 0.000 0.024 0.016 0.016 0.000
#> GSM97009 5 0.3445 0.688 0.000 0.000 0.244 0.012 0.744 0.000
#> GSM97010 5 0.0436 0.940 0.004 0.000 0.004 0.004 0.988 0.000
#> GSM96974 4 0.2793 0.751 0.000 0.000 0.000 0.800 0.000 0.200
#> GSM96985 4 0.2793 0.751 0.000 0.000 0.000 0.800 0.000 0.200
#> GSM96959 3 0.2624 0.894 0.000 0.000 0.880 0.080 0.024 0.016
#> GSM96972 4 0.2416 0.876 0.156 0.000 0.000 0.844 0.000 0.000
#> GSM96978 6 0.0000 0.944 0.000 0.000 0.000 0.000 0.000 1.000
#> GSM96967 4 0.2416 0.876 0.156 0.000 0.000 0.844 0.000 0.000
#> GSM96987 1 0.2804 0.885 0.852 0.000 0.024 0.004 0.120 0.000
#> GSM97011 5 0.0000 0.941 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM96964 1 0.2877 0.856 0.820 0.000 0.012 0.000 0.168 0.000
#> GSM96965 4 0.2730 0.727 0.000 0.000 0.000 0.808 0.192 0.000
#> GSM96981 1 0.3394 0.823 0.776 0.000 0.024 0.000 0.200 0.000
#> GSM96982 1 0.2206 0.899 0.904 0.000 0.024 0.008 0.064 0.000
#> GSM96988 6 0.0000 0.944 0.000 0.000 0.000 0.000 0.000 1.000
#> GSM97000 5 0.2243 0.889 0.004 0.000 0.004 0.112 0.880 0.000
#> GSM97004 1 0.3221 0.678 0.792 0.000 0.020 0.188 0.000 0.000
#> GSM97008 5 0.2006 0.896 0.004 0.000 0.000 0.104 0.892 0.000
#> GSM96950 5 0.0291 0.940 0.004 0.000 0.004 0.000 0.992 0.000
#> GSM96980 4 0.3634 0.612 0.356 0.000 0.000 0.644 0.000 0.000
#> GSM96989 1 0.2988 0.866 0.824 0.000 0.024 0.000 0.152 0.000
#> GSM96992 1 0.0603 0.891 0.980 0.000 0.000 0.004 0.016 0.000
#> GSM96993 5 0.0291 0.940 0.004 0.000 0.004 0.000 0.992 0.000
#> GSM96958 1 0.2703 0.849 0.824 0.000 0.000 0.004 0.172 0.000
#> GSM96951 1 0.2346 0.883 0.868 0.000 0.000 0.008 0.124 0.000
#> GSM96952 1 0.0692 0.893 0.976 0.000 0.000 0.004 0.020 0.000
#> GSM96961 1 0.1141 0.899 0.948 0.000 0.000 0.000 0.052 0.000
Heatmaps for the consensus matrix. It visualizes the probability of two samples to be in a same group.
consensus_heatmap(res, k = 2)
consensus_heatmap(res, k = 3)
consensus_heatmap(res, k = 4)
consensus_heatmap(res, k = 5)
consensus_heatmap(res, k = 6)
Heatmaps for the membership of samples in all partitions to see how consistent they are:
membership_heatmap(res, k = 2)
membership_heatmap(res, k = 3)
membership_heatmap(res, k = 4)
membership_heatmap(res, k = 5)
membership_heatmap(res, k = 6)
As soon as we have had the classes for columns, we can look for signatures which are significantly different between classes which can be candidate marks for certain classes. Following are the heatmaps for signatures.
Signature heatmaps where rows are scaled:
get_signatures(res, k = 2)
get_signatures(res, k = 3)
get_signatures(res, k = 4)
get_signatures(res, k = 5)
get_signatures(res, k = 6)
Signature heatmaps where rows are not scaled:
get_signatures(res, k = 2, scale_rows = FALSE)
get_signatures(res, k = 3, scale_rows = FALSE)
get_signatures(res, k = 4, scale_rows = FALSE)
get_signatures(res, k = 5, scale_rows = FALSE)
get_signatures(res, k = 6, scale_rows = FALSE)
Compare the overlap of signatures from different k:
compare_signatures(res)
get_signature() returns a data frame invisibly. TO get the list of signatures, the function
call should be assigned to a variable explicitly. In following code, if plot argument is set
to FALSE, no heatmap is plotted while only the differential analysis is performed.
# code only for demonstration
tb = get_signature(res, k = ..., plot = FALSE)
An example of the output of tb is:
#> which_row fdr mean_1 mean_2 scaled_mean_1 scaled_mean_2 km
#> 1 38 0.042760348 8.373488 9.131774 -0.5533452 0.5164555 1
#> 2 40 0.018707592 7.106213 8.469186 -0.6173731 0.5762149 1
#> 3 55 0.019134737 10.221463 11.207825 -0.6159697 0.5749050 1
#> 4 59 0.006059896 5.921854 7.869574 -0.6899429 0.6439467 1
#> 5 60 0.018055526 8.928898 10.211722 -0.6204761 0.5791110 1
#> 6 98 0.009384629 15.714769 14.887706 0.6635654 -0.6193277 2
...
The columns in tb are:
which_row: row indices corresponding to the input matrix.fdr: FDR for the differential test. mean_x: The mean value in group x.scaled_mean_x: The mean value in group x after rows are scaled.km: Row groups if k-means clustering is applied to rows.UMAP plot which shows how samples are separated.
dimension_reduction(res, k = 2, method = "UMAP")
dimension_reduction(res, k = 3, method = "UMAP")
dimension_reduction(res, k = 4, method = "UMAP")
dimension_reduction(res, k = 5, method = "UMAP")
dimension_reduction(res, k = 6, method = "UMAP")
Following heatmap shows how subgroups are split when increasing k:
collect_classes(res)
Test correlation between subgroups and known annotations. If the known annotation is numeric, one-way ANOVA test is applied, and if the known annotation is discrete, chi-squared contingency table test is applied.
test_to_known_factors(res)
#> n disease.state(p) specimen(p) cell.type(p) other(p) k
#> SD:mclust 98 2.79e-05 0.4038 8.16e-12 0.0115 2
#> SD:mclust 77 8.69e-03 0.4160 3.41e-15 0.1125 3
#> SD:mclust 98 3.34e-05 0.0724 1.27e-18 0.0280 4
#> SD:mclust 93 1.18e-05 0.0572 2.11e-19 0.0101 5
#> SD:mclust 99 3.41e-06 0.0900 1.47e-19 0.0315 6
If matrix rows can be associated to genes, consider to use functional_enrichment(res,
...) to perform function enrichment for the signature genes. See this vignette for more detailed explanations.
The object with results only for a single top-value method and a single partition method can be extracted as:
res = res_list["SD", "NMF"]
# you can also extract it by
# res = res_list["SD:NMF"]
A summary of res and all the functions that can be applied to it:
res
#> A 'ConsensusPartition' object with k = 2, 3, 4, 5, 6.
#> On a matrix with 21168 rows and 100 columns.
#> Top rows (1000, 2000, 3000, 4000, 5000) are extracted by 'SD' method.
#> Subgroups are detected by 'NMF' method.
#> Performed in total 1250 partitions by row resampling.
#> Best k for subgroups seems to be 3.
#>
#> Following methods can be applied to this 'ConsensusPartition' object:
#> [1] "cola_report" "collect_classes" "collect_plots"
#> [4] "collect_stats" "colnames" "compare_signatures"
#> [7] "consensus_heatmap" "dimension_reduction" "functional_enrichment"
#> [10] "get_anno_col" "get_anno" "get_classes"
#> [13] "get_consensus" "get_matrix" "get_membership"
#> [16] "get_param" "get_signatures" "get_stats"
#> [19] "is_best_k" "is_stable_k" "membership_heatmap"
#> [22] "ncol" "nrow" "plot_ecdf"
#> [25] "rownames" "select_partition_number" "show"
#> [28] "suggest_best_k" "test_to_known_factors"
collect_plots() function collects all the plots made from res for all k (number of partitions)
into one single page to provide an easy and fast comparison between different k.
collect_plots(res)
The plots are:
k and the heatmap of
predicted classes for each k.k.k.k.All the plots in panels can be made by individual functions and they are plotted later in this section.
select_partition_number() produces several plots showing different
statistics for choosing “optimized” k. There are following statistics:
k;k, the area increased is defined as \(A_k - A_{k-1}\).The detailed explanations of these statistics can be found in the cola vignette.
Generally speaking, lower PAC score, higher mean silhouette score or higher
concordance corresponds to better partition. Rand index and Jaccard index
measure how similar the current partition is compared to partition with k-1.
If they are too similar, we won't accept k is better than k-1.
select_partition_number(res)
The numeric values for all these statistics can be obtained by get_stats().
get_stats(res)
#> k 1-PAC mean_silhouette concordance area_increased Rand Jaccard
#> 2 2 0.999 0.971 0.987 0.4949 0.505 0.505
#> 3 3 0.926 0.915 0.959 0.3180 0.787 0.600
#> 4 4 0.614 0.570 0.755 0.1373 0.805 0.505
#> 5 5 0.599 0.528 0.738 0.0717 0.842 0.474
#> 6 6 0.631 0.528 0.722 0.0448 0.840 0.392
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
#> GSM97038 2 0.0000 0.983 0.000 1.000
#> GSM97045 2 0.0000 0.983 0.000 1.000
#> GSM97047 2 0.0000 0.983 0.000 1.000
#> GSM97025 2 0.0000 0.983 0.000 1.000
#> GSM97030 2 0.0000 0.983 0.000 1.000
#> GSM97027 2 0.0000 0.983 0.000 1.000
#> GSM97033 2 0.0000 0.983 0.000 1.000
#> GSM97034 2 0.0000 0.983 0.000 1.000
#> GSM97020 2 0.0000 0.983 0.000 1.000
#> GSM97026 2 0.0000 0.983 0.000 1.000
#> GSM97012 2 0.0000 0.983 0.000 1.000
#> GSM97015 2 0.0000 0.983 0.000 1.000
#> GSM97016 2 0.0000 0.983 0.000 1.000
#> GSM97017 1 0.0376 0.986 0.996 0.004
#> GSM97019 2 0.0000 0.983 0.000 1.000
#> GSM97022 2 0.0000 0.983 0.000 1.000
#> GSM97035 2 0.0000 0.983 0.000 1.000
#> GSM97036 1 0.1414 0.974 0.980 0.020
#> GSM97039 2 0.0000 0.983 0.000 1.000
#> GSM97046 2 0.0000 0.983 0.000 1.000
#> GSM97023 1 0.0000 0.989 1.000 0.000
#> GSM97029 1 0.2043 0.963 0.968 0.032
#> GSM97043 2 0.0000 0.983 0.000 1.000
#> GSM97013 1 0.0000 0.989 1.000 0.000
#> GSM96956 2 0.0000 0.983 0.000 1.000
#> GSM97024 2 0.0000 0.983 0.000 1.000
#> GSM97032 2 0.0000 0.983 0.000 1.000
#> GSM97044 2 0.0000 0.983 0.000 1.000
#> GSM97049 2 0.0000 0.983 0.000 1.000
#> GSM96968 1 0.7883 0.695 0.764 0.236
#> GSM96971 1 0.0000 0.989 1.000 0.000
#> GSM96986 1 0.0000 0.989 1.000 0.000
#> GSM97003 1 0.0000 0.989 1.000 0.000
#> GSM96957 1 0.0938 0.981 0.988 0.012
#> GSM96960 1 0.0000 0.989 1.000 0.000
#> GSM96975 1 0.0000 0.989 1.000 0.000
#> GSM96998 1 0.0000 0.989 1.000 0.000
#> GSM96999 1 0.0000 0.989 1.000 0.000
#> GSM97001 1 0.0000 0.989 1.000 0.000
#> GSM97005 1 0.0000 0.989 1.000 0.000
#> GSM97006 1 0.0000 0.989 1.000 0.000
#> GSM97021 1 0.0376 0.986 0.996 0.004
#> GSM97028 2 0.9686 0.346 0.396 0.604
#> GSM97031 1 0.0000 0.989 1.000 0.000
#> GSM97037 2 0.0000 0.983 0.000 1.000
#> GSM97018 2 0.0000 0.983 0.000 1.000
#> GSM97014 2 0.0000 0.983 0.000 1.000
#> GSM97042 2 0.0000 0.983 0.000 1.000
#> GSM97040 2 0.1843 0.960 0.028 0.972
#> GSM97041 1 0.2236 0.960 0.964 0.036
#> GSM96955 2 0.0000 0.983 0.000 1.000
#> GSM96990 2 0.0000 0.983 0.000 1.000
#> GSM96991 2 0.0000 0.983 0.000 1.000
#> GSM97048 2 0.0000 0.983 0.000 1.000
#> GSM96963 2 0.0000 0.983 0.000 1.000
#> GSM96953 2 0.0000 0.983 0.000 1.000
#> GSM96966 1 0.0000 0.989 1.000 0.000
#> GSM96979 1 0.0000 0.989 1.000 0.000
#> GSM96983 2 0.0000 0.983 0.000 1.000
#> GSM96984 1 0.1414 0.974 0.980 0.020
#> GSM96994 2 0.0000 0.983 0.000 1.000
#> GSM96996 1 0.0000 0.989 1.000 0.000
#> GSM96997 1 0.0000 0.989 1.000 0.000
#> GSM97007 2 0.4815 0.881 0.104 0.896
#> GSM96954 1 0.0000 0.989 1.000 0.000
#> GSM96962 1 0.0000 0.989 1.000 0.000
#> GSM96969 1 0.0000 0.989 1.000 0.000
#> GSM96970 1 0.0000 0.989 1.000 0.000
#> GSM96973 1 0.0000 0.989 1.000 0.000
#> GSM96976 2 0.4022 0.909 0.080 0.920
#> GSM96977 1 0.0000 0.989 1.000 0.000
#> GSM96995 2 0.3879 0.914 0.076 0.924
#> GSM97002 1 0.0000 0.989 1.000 0.000
#> GSM97009 2 0.0000 0.983 0.000 1.000
#> GSM97010 1 0.0376 0.986 0.996 0.004
#> GSM96974 1 0.0000 0.989 1.000 0.000
#> GSM96985 1 0.0000 0.989 1.000 0.000
#> GSM96959 2 0.0000 0.983 0.000 1.000
#> GSM96972 1 0.0000 0.989 1.000 0.000
#> GSM96978 1 0.6712 0.790 0.824 0.176
#> GSM96967 1 0.0000 0.989 1.000 0.000
#> GSM96987 1 0.0000 0.989 1.000 0.000
#> GSM97011 1 0.3274 0.935 0.940 0.060
#> GSM96964 1 0.0000 0.989 1.000 0.000
#> GSM96965 1 0.0376 0.986 0.996 0.004
#> GSM96981 1 0.0000 0.989 1.000 0.000
#> GSM96982 1 0.0000 0.989 1.000 0.000
#> GSM96988 1 0.0000 0.989 1.000 0.000
#> GSM97000 1 0.0000 0.989 1.000 0.000
#> GSM97004 1 0.0000 0.989 1.000 0.000
#> GSM97008 1 0.0000 0.989 1.000 0.000
#> GSM96950 1 0.0000 0.989 1.000 0.000
#> GSM96980 1 0.0000 0.989 1.000 0.000
#> GSM96989 1 0.0000 0.989 1.000 0.000
#> GSM96992 1 0.0000 0.989 1.000 0.000
#> GSM96993 1 0.0672 0.984 0.992 0.008
#> GSM96958 1 0.0000 0.989 1.000 0.000
#> GSM96951 1 0.0000 0.989 1.000 0.000
#> GSM96952 1 0.0000 0.989 1.000 0.000
#> GSM96961 1 0.0000 0.989 1.000 0.000
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM97038 2 0.0424 0.9497 0.000 0.992 0.008
#> GSM97045 2 0.0237 0.9479 0.004 0.996 0.000
#> GSM97047 2 0.0237 0.9479 0.004 0.996 0.000
#> GSM97025 2 0.0237 0.9479 0.004 0.996 0.000
#> GSM97030 3 0.3340 0.8564 0.000 0.120 0.880
#> GSM97027 2 0.0237 0.9479 0.004 0.996 0.000
#> GSM97033 2 0.0237 0.9479 0.004 0.996 0.000
#> GSM97034 3 0.3412 0.8518 0.000 0.124 0.876
#> GSM97020 2 0.0237 0.9479 0.004 0.996 0.000
#> GSM97026 2 0.0424 0.9452 0.008 0.992 0.000
#> GSM97012 2 0.0424 0.9497 0.000 0.992 0.008
#> GSM97015 3 0.3941 0.8168 0.000 0.156 0.844
#> GSM97016 2 0.0424 0.9497 0.000 0.992 0.008
#> GSM97017 1 0.2356 0.9338 0.928 0.072 0.000
#> GSM97019 2 0.0424 0.9497 0.000 0.992 0.008
#> GSM97022 2 0.0592 0.9476 0.000 0.988 0.012
#> GSM97035 2 0.0592 0.9476 0.000 0.988 0.012
#> GSM97036 1 0.2796 0.9154 0.908 0.092 0.000
#> GSM97039 2 0.0424 0.9497 0.000 0.992 0.008
#> GSM97046 2 0.0424 0.9497 0.000 0.992 0.008
#> GSM97023 1 0.0000 0.9721 1.000 0.000 0.000
#> GSM97029 1 0.2878 0.9116 0.904 0.096 0.000
#> GSM97043 2 0.0424 0.9497 0.000 0.992 0.008
#> GSM97013 1 0.1529 0.9568 0.960 0.040 0.000
#> GSM96956 2 0.6280 0.0848 0.000 0.540 0.460
#> GSM97024 2 0.1031 0.9388 0.000 0.976 0.024
#> GSM97032 2 0.6309 -0.0635 0.000 0.504 0.496
#> GSM97044 3 0.1411 0.9212 0.000 0.036 0.964
#> GSM97049 2 0.0237 0.9479 0.004 0.996 0.000
#> GSM96968 3 0.0592 0.9322 0.012 0.000 0.988
#> GSM96971 3 0.0237 0.9345 0.004 0.000 0.996
#> GSM96986 3 0.0424 0.9335 0.008 0.000 0.992
#> GSM97003 1 0.3941 0.8401 0.844 0.000 0.156
#> GSM96957 1 0.1860 0.9498 0.948 0.052 0.000
#> GSM96960 1 0.1753 0.9525 0.952 0.000 0.048
#> GSM96975 1 0.0237 0.9716 0.996 0.004 0.000
#> GSM96998 1 0.0000 0.9721 1.000 0.000 0.000
#> GSM96999 1 0.0424 0.9708 0.992 0.008 0.000
#> GSM97001 1 0.1753 0.9519 0.952 0.048 0.000
#> GSM97005 1 0.0424 0.9708 0.992 0.008 0.000
#> GSM97006 1 0.0747 0.9697 0.984 0.000 0.016
#> GSM97021 1 0.0747 0.9684 0.984 0.016 0.000
#> GSM97028 3 0.0237 0.9354 0.000 0.004 0.996
#> GSM97031 1 0.2261 0.9362 0.932 0.000 0.068
#> GSM97037 2 0.3879 0.7940 0.000 0.848 0.152
#> GSM97018 3 0.5948 0.4685 0.000 0.360 0.640
#> GSM97014 2 0.1031 0.9312 0.024 0.976 0.000
#> GSM97042 2 0.0424 0.9497 0.000 0.992 0.008
#> GSM97040 2 0.1753 0.9055 0.048 0.952 0.000
#> GSM97041 1 0.3340 0.8866 0.880 0.120 0.000
#> GSM96955 2 0.0237 0.9494 0.000 0.996 0.004
#> GSM96990 3 0.5968 0.4572 0.000 0.364 0.636
#> GSM96991 2 0.0424 0.9497 0.000 0.992 0.008
#> GSM97048 2 0.0237 0.9479 0.004 0.996 0.000
#> GSM96963 2 0.0424 0.9497 0.000 0.992 0.008
#> GSM96953 2 0.0424 0.9497 0.000 0.992 0.008
#> GSM96966 1 0.0747 0.9697 0.984 0.000 0.016
#> GSM96979 3 0.0592 0.9321 0.012 0.000 0.988
#> GSM96983 3 0.0237 0.9354 0.000 0.004 0.996
#> GSM96984 3 0.0237 0.9354 0.000 0.004 0.996
#> GSM96994 3 0.0237 0.9354 0.000 0.004 0.996
#> GSM96996 1 0.0237 0.9721 0.996 0.000 0.004
#> GSM96997 3 0.0424 0.9335 0.008 0.000 0.992
#> GSM97007 3 0.0237 0.9354 0.000 0.004 0.996
#> GSM96954 3 0.1289 0.9177 0.032 0.000 0.968
#> GSM96962 3 0.0237 0.9354 0.000 0.004 0.996
#> GSM96969 1 0.1529 0.9575 0.960 0.000 0.040
#> GSM96970 1 0.0747 0.9697 0.984 0.000 0.016
#> GSM96973 1 0.1753 0.9525 0.952 0.000 0.048
#> GSM96976 3 0.1163 0.9272 0.000 0.028 0.972
#> GSM96977 1 0.0237 0.9724 0.996 0.000 0.004
#> GSM96995 3 0.2066 0.9064 0.000 0.060 0.940
#> GSM97002 1 0.0424 0.9717 0.992 0.000 0.008
#> GSM97009 2 0.0237 0.9479 0.004 0.996 0.000
#> GSM97010 1 0.0848 0.9710 0.984 0.008 0.008
#> GSM96974 3 0.0592 0.9323 0.012 0.000 0.988
#> GSM96985 3 0.0892 0.9272 0.020 0.000 0.980
#> GSM96959 2 0.1411 0.9293 0.000 0.964 0.036
#> GSM96972 1 0.1031 0.9663 0.976 0.000 0.024
#> GSM96978 3 0.0237 0.9354 0.000 0.004 0.996
#> GSM96967 1 0.1411 0.9599 0.964 0.000 0.036
#> GSM96987 1 0.0424 0.9708 0.992 0.008 0.000
#> GSM97011 1 0.2066 0.9437 0.940 0.060 0.000
#> GSM96964 1 0.0000 0.9721 1.000 0.000 0.000
#> GSM96965 1 0.1989 0.9523 0.948 0.048 0.004
#> GSM96981 1 0.0000 0.9721 1.000 0.000 0.000
#> GSM96982 1 0.0592 0.9709 0.988 0.000 0.012
#> GSM96988 3 0.0424 0.9335 0.008 0.000 0.992
#> GSM97000 1 0.0592 0.9712 0.988 0.000 0.012
#> GSM97004 1 0.0592 0.9709 0.988 0.000 0.012
#> GSM97008 1 0.0424 0.9708 0.992 0.008 0.000
#> GSM96950 1 0.0424 0.9708 0.992 0.008 0.000
#> GSM96980 1 0.0424 0.9717 0.992 0.000 0.008
#> GSM96989 1 0.0000 0.9721 1.000 0.000 0.000
#> GSM96992 1 0.0237 0.9721 0.996 0.000 0.004
#> GSM96993 1 0.1411 0.9598 0.964 0.036 0.000
#> GSM96958 1 0.0237 0.9721 0.996 0.000 0.004
#> GSM96951 1 0.0237 0.9721 0.996 0.000 0.004
#> GSM96952 1 0.0237 0.9721 0.996 0.000 0.004
#> GSM96961 1 0.0237 0.9721 0.996 0.000 0.004
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM97038 2 0.4535 0.8197 0.292 0.704 0.004 0.000
#> GSM97045 2 0.4643 0.8022 0.344 0.656 0.000 0.000
#> GSM97047 1 0.5088 -0.5031 0.572 0.424 0.004 0.000
#> GSM97025 2 0.4454 0.8154 0.308 0.692 0.000 0.000
#> GSM97030 3 0.1118 0.8736 0.000 0.036 0.964 0.000
#> GSM97027 2 0.4800 0.8032 0.340 0.656 0.004 0.000
#> GSM97033 2 0.4761 0.8074 0.332 0.664 0.004 0.000
#> GSM97034 3 0.2867 0.8419 0.012 0.104 0.884 0.000
#> GSM97020 2 0.4643 0.8022 0.344 0.656 0.000 0.000
#> GSM97026 2 0.4804 0.7551 0.384 0.616 0.000 0.000
#> GSM97012 2 0.1520 0.7475 0.020 0.956 0.000 0.024
#> GSM97015 3 0.2227 0.8692 0.036 0.036 0.928 0.000
#> GSM97016 2 0.4594 0.8209 0.280 0.712 0.008 0.000
#> GSM97017 1 0.3647 0.5424 0.832 0.016 0.000 0.152
#> GSM97019 2 0.3208 0.8075 0.148 0.848 0.004 0.000
#> GSM97022 2 0.2124 0.7852 0.068 0.924 0.008 0.000
#> GSM97035 2 0.1824 0.7830 0.060 0.936 0.004 0.000
#> GSM97036 4 0.6665 0.1908 0.360 0.096 0.000 0.544
#> GSM97039 2 0.4655 0.8144 0.312 0.684 0.004 0.000
#> GSM97046 2 0.4008 0.8210 0.244 0.756 0.000 0.000
#> GSM97023 1 0.4866 0.3213 0.596 0.000 0.000 0.404
#> GSM97029 1 0.4610 0.5062 0.744 0.020 0.000 0.236
#> GSM97043 2 0.4382 0.8194 0.296 0.704 0.000 0.000
#> GSM97013 1 0.4516 0.4983 0.736 0.012 0.000 0.252
#> GSM96956 2 0.5413 0.5107 0.048 0.712 0.236 0.004
#> GSM97024 2 0.4501 0.8181 0.212 0.764 0.024 0.000
#> GSM97032 3 0.6421 0.2875 0.076 0.368 0.556 0.000
#> GSM97044 3 0.0927 0.8779 0.008 0.016 0.976 0.000
#> GSM97049 2 0.4624 0.8043 0.340 0.660 0.000 0.000
#> GSM96968 3 0.1109 0.8739 0.028 0.000 0.968 0.004
#> GSM96971 3 0.4341 0.7994 0.024 0.020 0.820 0.136
#> GSM96986 3 0.0895 0.8752 0.020 0.000 0.976 0.004
#> GSM97003 3 0.7544 0.0189 0.200 0.000 0.460 0.340
#> GSM96957 1 0.1767 0.5296 0.944 0.044 0.000 0.012
#> GSM96960 4 0.4054 0.5709 0.188 0.000 0.016 0.796
#> GSM96975 4 0.4998 -0.0549 0.488 0.000 0.000 0.512
#> GSM96998 4 0.4761 0.3088 0.372 0.000 0.000 0.628
#> GSM96999 1 0.4888 0.3082 0.588 0.000 0.000 0.412
#> GSM97001 1 0.2300 0.5346 0.924 0.048 0.000 0.028
#> GSM97005 1 0.5300 0.4377 0.664 0.000 0.028 0.308
#> GSM97006 4 0.5040 0.3209 0.364 0.000 0.008 0.628
#> GSM97021 1 0.2814 0.5499 0.908 0.008 0.032 0.052
#> GSM97028 3 0.1593 0.8783 0.016 0.024 0.956 0.004
#> GSM97031 1 0.6557 0.1308 0.476 0.000 0.448 0.076
#> GSM97037 2 0.6400 0.7062 0.168 0.652 0.180 0.000
#> GSM97018 2 0.5626 0.1861 0.024 0.612 0.360 0.004
#> GSM97014 1 0.4907 -0.4816 0.580 0.420 0.000 0.000
#> GSM97042 2 0.1042 0.7429 0.008 0.972 0.000 0.020
#> GSM97040 1 0.2334 0.4674 0.908 0.088 0.000 0.004
#> GSM97041 1 0.2363 0.5296 0.920 0.056 0.000 0.024
#> GSM96955 2 0.4008 0.7700 0.148 0.820 0.000 0.032
#> GSM96990 3 0.3803 0.7972 0.032 0.132 0.836 0.000
#> GSM96991 2 0.2660 0.6970 0.024 0.916 0.012 0.048
#> GSM97048 2 0.4781 0.8060 0.336 0.660 0.004 0.000
#> GSM96963 2 0.1545 0.7247 0.008 0.952 0.000 0.040
#> GSM96953 2 0.1398 0.7729 0.040 0.956 0.004 0.000
#> GSM96966 4 0.0927 0.6392 0.016 0.008 0.000 0.976
#> GSM96979 3 0.3606 0.8159 0.008 0.020 0.856 0.116
#> GSM96983 3 0.2099 0.8661 0.012 0.044 0.936 0.008
#> GSM96984 3 0.0188 0.8771 0.000 0.000 0.996 0.004
#> GSM96994 3 0.0524 0.8769 0.008 0.000 0.988 0.004
#> GSM96996 4 0.3444 0.5768 0.184 0.000 0.000 0.816
#> GSM96997 3 0.0779 0.8758 0.016 0.000 0.980 0.004
#> GSM97007 3 0.0188 0.8771 0.000 0.000 0.996 0.004
#> GSM96954 3 0.2799 0.8209 0.108 0.000 0.884 0.008
#> GSM96962 3 0.0657 0.8765 0.012 0.000 0.984 0.004
#> GSM96969 4 0.1305 0.6460 0.036 0.000 0.004 0.960
#> GSM96970 4 0.0469 0.6315 0.000 0.012 0.000 0.988
#> GSM96973 4 0.1824 0.6004 0.004 0.060 0.000 0.936
#> GSM96976 4 0.6473 0.0561 0.024 0.472 0.028 0.476
#> GSM96977 1 0.5953 0.4404 0.656 0.000 0.076 0.268
#> GSM96995 3 0.1635 0.8689 0.044 0.008 0.948 0.000
#> GSM97002 4 0.2149 0.6401 0.088 0.000 0.000 0.912
#> GSM97009 2 0.5097 0.7196 0.428 0.568 0.004 0.000
#> GSM97010 4 0.2412 0.6435 0.084 0.008 0.000 0.908
#> GSM96974 4 0.6510 0.3461 0.024 0.320 0.048 0.608
#> GSM96985 4 0.6358 0.3551 0.024 0.320 0.040 0.616
#> GSM96959 3 0.6005 0.5333 0.324 0.060 0.616 0.000
#> GSM96972 4 0.1452 0.6458 0.036 0.000 0.008 0.956
#> GSM96978 3 0.5881 0.7068 0.020 0.200 0.716 0.064
#> GSM96967 4 0.0592 0.6293 0.000 0.016 0.000 0.984
#> GSM96987 4 0.4948 0.1258 0.440 0.000 0.000 0.560
#> GSM97011 1 0.2908 0.5466 0.896 0.040 0.000 0.064
#> GSM96964 1 0.4933 0.2642 0.568 0.000 0.000 0.432
#> GSM96965 4 0.3157 0.5404 0.004 0.144 0.000 0.852
#> GSM96981 4 0.4477 0.4177 0.312 0.000 0.000 0.688
#> GSM96982 4 0.2011 0.6443 0.080 0.000 0.000 0.920
#> GSM96988 3 0.3529 0.8467 0.012 0.068 0.876 0.044
#> GSM97000 1 0.5404 0.2780 0.600 0.004 0.384 0.012
#> GSM97004 4 0.1716 0.6467 0.064 0.000 0.000 0.936
#> GSM97008 1 0.5011 0.5048 0.764 0.000 0.160 0.076
#> GSM96950 1 0.4967 0.2024 0.548 0.000 0.000 0.452
#> GSM96980 4 0.1474 0.6474 0.052 0.000 0.000 0.948
#> GSM96989 4 0.4925 0.1742 0.428 0.000 0.000 0.572
#> GSM96992 4 0.4961 0.1068 0.448 0.000 0.000 0.552
#> GSM96993 1 0.5252 0.4129 0.644 0.020 0.000 0.336
#> GSM96958 1 0.4925 0.2722 0.572 0.000 0.000 0.428
#> GSM96951 1 0.5728 0.3530 0.600 0.000 0.036 0.364
#> GSM96952 4 0.4977 0.0603 0.460 0.000 0.000 0.540
#> GSM96961 1 0.4972 0.1953 0.544 0.000 0.000 0.456
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM97038 2 0.0566 0.6871 0.012 0.984 0.000 0.000 0.004
#> GSM97045 2 0.4150 0.5054 0.036 0.748 0.000 0.000 0.216
#> GSM97047 2 0.5514 0.4481 0.292 0.628 0.012 0.000 0.068
#> GSM97025 2 0.4812 0.1233 0.028 0.600 0.000 0.000 0.372
#> GSM97030 3 0.2295 0.7558 0.000 0.008 0.900 0.004 0.088
#> GSM97027 2 0.4240 0.4910 0.036 0.736 0.000 0.000 0.228
#> GSM97033 2 0.0671 0.6896 0.004 0.980 0.000 0.000 0.016
#> GSM97034 5 0.5470 0.4194 0.040 0.020 0.340 0.000 0.600
#> GSM97020 2 0.0703 0.6878 0.000 0.976 0.000 0.000 0.024
#> GSM97026 5 0.6066 0.4637 0.188 0.240 0.000 0.000 0.572
#> GSM97012 5 0.4444 0.4947 0.000 0.364 0.000 0.012 0.624
#> GSM97015 3 0.3996 0.6965 0.044 0.016 0.808 0.000 0.132
#> GSM97016 2 0.1043 0.6811 0.000 0.960 0.000 0.000 0.040
#> GSM97017 1 0.3379 0.6823 0.860 0.040 0.000 0.024 0.076
#> GSM97019 5 0.4309 0.5927 0.016 0.308 0.000 0.000 0.676
#> GSM97022 5 0.4402 0.5193 0.004 0.372 0.000 0.004 0.620
#> GSM97035 2 0.4448 -0.2012 0.000 0.516 0.000 0.004 0.480
#> GSM97036 1 0.5877 0.4792 0.596 0.008 0.000 0.108 0.288
#> GSM97039 2 0.0703 0.6873 0.000 0.976 0.000 0.000 0.024
#> GSM97046 2 0.1121 0.6804 0.000 0.956 0.000 0.000 0.044
#> GSM97023 1 0.3112 0.6794 0.856 0.000 0.000 0.100 0.044
#> GSM97029 1 0.3964 0.6389 0.796 0.032 0.000 0.012 0.160
#> GSM97043 5 0.5006 0.5962 0.048 0.272 0.008 0.000 0.672
#> GSM97013 1 0.6147 0.6313 0.668 0.132 0.000 0.128 0.072
#> GSM96956 2 0.4796 0.4772 0.000 0.740 0.164 0.008 0.088
#> GSM97024 5 0.5220 0.5029 0.020 0.380 0.020 0.000 0.580
#> GSM97032 5 0.6052 0.5808 0.028 0.100 0.252 0.000 0.620
#> GSM97044 3 0.2389 0.7428 0.004 0.000 0.880 0.000 0.116
#> GSM97049 2 0.0000 0.6889 0.000 1.000 0.000 0.000 0.000
#> GSM96968 3 0.0932 0.7805 0.000 0.004 0.972 0.004 0.020
#> GSM96971 4 0.4449 -0.1104 0.004 0.000 0.484 0.512 0.000
#> GSM96986 3 0.2170 0.7578 0.036 0.000 0.924 0.020 0.020
#> GSM97003 3 0.6717 0.1930 0.136 0.000 0.508 0.328 0.028
#> GSM96957 1 0.5619 0.4393 0.592 0.336 0.016 0.000 0.056
#> GSM96960 4 0.5686 0.4589 0.284 0.000 0.024 0.628 0.064
#> GSM96975 1 0.6089 0.1778 0.500 0.012 0.000 0.400 0.088
#> GSM96998 1 0.5474 0.3730 0.576 0.000 0.000 0.348 0.076
#> GSM96999 1 0.4739 0.5748 0.704 0.008 0.008 0.256 0.024
#> GSM97001 1 0.6355 0.3256 0.536 0.360 0.012 0.020 0.072
#> GSM97005 1 0.5377 0.6397 0.760 0.032 0.048 0.096 0.064
#> GSM97006 4 0.5732 -0.0601 0.464 0.000 0.016 0.472 0.048
#> GSM97021 1 0.3359 0.6663 0.868 0.028 0.036 0.004 0.064
#> GSM97028 5 0.4953 0.1049 0.028 0.000 0.440 0.000 0.532
#> GSM97031 3 0.6179 0.3795 0.312 0.000 0.580 0.052 0.056
#> GSM97037 2 0.6977 -0.2747 0.012 0.420 0.232 0.000 0.336
#> GSM97018 5 0.5213 0.6036 0.020 0.072 0.204 0.000 0.704
#> GSM97014 2 0.4068 0.5811 0.144 0.792 0.000 0.004 0.060
#> GSM97042 5 0.4232 0.5759 0.000 0.312 0.000 0.012 0.676
#> GSM97040 1 0.5065 0.5840 0.732 0.172 0.016 0.004 0.076
#> GSM97041 1 0.3559 0.6674 0.836 0.096 0.000 0.004 0.064
#> GSM96955 2 0.5591 0.4969 0.040 0.656 0.000 0.048 0.256
#> GSM96990 3 0.5733 0.2620 0.016 0.064 0.584 0.000 0.336
#> GSM96991 5 0.3690 0.6157 0.000 0.200 0.000 0.020 0.780
#> GSM97048 2 0.0000 0.6889 0.000 1.000 0.000 0.000 0.000
#> GSM96963 5 0.4506 0.5463 0.000 0.296 0.000 0.028 0.676
#> GSM96953 2 0.4118 0.3159 0.000 0.660 0.000 0.004 0.336
#> GSM96966 4 0.0671 0.7050 0.016 0.000 0.000 0.980 0.004
#> GSM96979 3 0.4326 0.5931 0.000 0.000 0.708 0.264 0.028
#> GSM96983 3 0.2929 0.7198 0.000 0.000 0.840 0.008 0.152
#> GSM96984 3 0.0798 0.7798 0.000 0.000 0.976 0.008 0.016
#> GSM96994 3 0.1168 0.7772 0.000 0.000 0.960 0.008 0.032
#> GSM96996 4 0.5524 0.1342 0.416 0.000 0.000 0.516 0.068
#> GSM96997 3 0.0968 0.7752 0.012 0.000 0.972 0.004 0.012
#> GSM97007 3 0.0955 0.7781 0.000 0.000 0.968 0.004 0.028
#> GSM96954 3 0.1956 0.7622 0.076 0.000 0.916 0.000 0.008
#> GSM96962 3 0.0566 0.7793 0.000 0.000 0.984 0.004 0.012
#> GSM96969 4 0.0771 0.7047 0.020 0.000 0.000 0.976 0.004
#> GSM96970 4 0.0798 0.7044 0.016 0.000 0.000 0.976 0.008
#> GSM96973 4 0.0566 0.7000 0.000 0.000 0.004 0.984 0.012
#> GSM96976 4 0.4640 0.5701 0.000 0.076 0.016 0.764 0.144
#> GSM96977 1 0.3700 0.6777 0.840 0.000 0.020 0.060 0.080
#> GSM96995 3 0.2209 0.7665 0.056 0.000 0.912 0.000 0.032
#> GSM97002 4 0.5192 0.4644 0.280 0.000 0.000 0.644 0.076
#> GSM97009 2 0.3682 0.6146 0.096 0.832 0.000 0.008 0.064
#> GSM97010 4 0.5177 0.6125 0.044 0.172 0.008 0.736 0.040
#> GSM96974 4 0.3720 0.5665 0.000 0.000 0.012 0.760 0.228
#> GSM96985 4 0.4954 0.2933 0.004 0.000 0.020 0.528 0.448
#> GSM96959 2 0.7537 0.2652 0.172 0.484 0.272 0.004 0.068
#> GSM96972 4 0.1728 0.6990 0.036 0.000 0.004 0.940 0.020
#> GSM96978 3 0.5606 0.3571 0.000 0.000 0.568 0.088 0.344
#> GSM96967 4 0.0807 0.7052 0.012 0.000 0.000 0.976 0.012
#> GSM96987 1 0.4934 0.5946 0.708 0.000 0.000 0.188 0.104
#> GSM97011 1 0.6601 0.0248 0.448 0.436 0.008 0.028 0.080
#> GSM96964 1 0.4016 0.6568 0.796 0.000 0.000 0.112 0.092
#> GSM96965 4 0.1990 0.6902 0.004 0.040 0.000 0.928 0.028
#> GSM96981 4 0.5744 0.2085 0.380 0.000 0.000 0.528 0.092
#> GSM96982 4 0.5112 0.5102 0.256 0.000 0.000 0.664 0.080
#> GSM96988 5 0.4920 0.1784 0.012 0.000 0.404 0.012 0.572
#> GSM97000 3 0.6783 0.2846 0.344 0.036 0.528 0.020 0.072
#> GSM97004 4 0.4865 0.5147 0.252 0.000 0.000 0.684 0.064
#> GSM97008 1 0.6411 0.5530 0.672 0.104 0.132 0.016 0.076
#> GSM96950 1 0.4307 0.6467 0.772 0.000 0.000 0.128 0.100
#> GSM96980 4 0.2580 0.6861 0.064 0.000 0.000 0.892 0.044
#> GSM96989 1 0.4914 0.5926 0.712 0.000 0.000 0.180 0.108
#> GSM96992 1 0.4960 0.5251 0.680 0.000 0.008 0.264 0.048
#> GSM96993 1 0.4087 0.6403 0.784 0.008 0.000 0.040 0.168
#> GSM96958 1 0.4681 0.6354 0.748 0.000 0.016 0.180 0.056
#> GSM96951 1 0.3872 0.6720 0.836 0.000 0.064 0.060 0.040
#> GSM96952 1 0.4602 0.5609 0.708 0.000 0.000 0.240 0.052
#> GSM96961 1 0.3975 0.6534 0.792 0.000 0.000 0.144 0.064
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM97038 3 0.1760 0.7756 0.000 0.020 0.928 0.004 0.048 0.000
#> GSM97045 2 0.6946 0.3382 0.052 0.384 0.364 0.008 0.192 0.000
#> GSM97047 5 0.4683 0.5145 0.036 0.032 0.212 0.000 0.712 0.008
#> GSM97025 2 0.6191 0.5116 0.052 0.524 0.320 0.004 0.100 0.000
#> GSM97030 6 0.2101 0.7589 0.008 0.072 0.000 0.008 0.004 0.908
#> GSM97027 2 0.6736 0.3554 0.056 0.408 0.376 0.004 0.156 0.000
#> GSM97033 3 0.2680 0.7076 0.000 0.076 0.868 0.000 0.056 0.000
#> GSM97034 2 0.5902 0.3531 0.072 0.576 0.004 0.004 0.048 0.296
#> GSM97020 3 0.1750 0.7727 0.012 0.040 0.932 0.000 0.016 0.000
#> GSM97026 2 0.6280 0.4516 0.340 0.520 0.072 0.004 0.052 0.012
#> GSM97012 2 0.3736 0.6434 0.000 0.788 0.160 0.020 0.032 0.000
#> GSM97015 6 0.3672 0.7197 0.028 0.140 0.004 0.004 0.016 0.808
#> GSM97016 3 0.0713 0.7945 0.000 0.028 0.972 0.000 0.000 0.000
#> GSM97017 5 0.4609 0.4909 0.352 0.016 0.024 0.000 0.608 0.000
#> GSM97019 2 0.4374 0.6604 0.024 0.768 0.136 0.004 0.064 0.004
#> GSM97022 2 0.4887 0.6485 0.020 0.716 0.188 0.012 0.060 0.004
#> GSM97035 2 0.4508 0.5845 0.000 0.668 0.280 0.012 0.040 0.000
#> GSM97036 1 0.3100 0.5720 0.836 0.128 0.000 0.024 0.012 0.000
#> GSM97039 3 0.0777 0.7951 0.000 0.024 0.972 0.000 0.004 0.000
#> GSM97046 3 0.0508 0.7974 0.000 0.012 0.984 0.004 0.000 0.000
#> GSM97023 1 0.3275 0.5983 0.828 0.008 0.000 0.044 0.120 0.000
#> GSM97029 1 0.5141 0.3362 0.660 0.200 0.016 0.000 0.124 0.000
#> GSM97043 2 0.5374 0.6337 0.104 0.712 0.120 0.004 0.036 0.024
#> GSM97013 1 0.4526 0.5567 0.744 0.004 0.156 0.072 0.024 0.000
#> GSM96956 3 0.2201 0.7422 0.000 0.028 0.896 0.000 0.000 0.076
#> GSM97024 2 0.5874 0.6355 0.032 0.648 0.176 0.000 0.112 0.032
#> GSM97032 2 0.5587 0.3259 0.056 0.584 0.028 0.004 0.008 0.320
#> GSM97044 6 0.2642 0.7425 0.012 0.116 0.000 0.004 0.004 0.864
#> GSM97049 3 0.0000 0.7977 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM96968 6 0.1843 0.7668 0.008 0.040 0.004 0.004 0.012 0.932
#> GSM96971 4 0.4823 0.2780 0.000 0.012 0.000 0.600 0.044 0.344
#> GSM96986 6 0.3291 0.7431 0.008 0.012 0.000 0.060 0.072 0.848
#> GSM97003 6 0.7082 0.1821 0.044 0.028 0.000 0.280 0.192 0.456
#> GSM96957 3 0.5118 0.3242 0.148 0.004 0.640 0.000 0.208 0.000
#> GSM96960 1 0.7072 0.1789 0.388 0.028 0.000 0.384 0.152 0.048
#> GSM96975 5 0.5032 0.5011 0.120 0.008 0.004 0.196 0.672 0.000
#> GSM96998 1 0.3507 0.5901 0.764 0.012 0.000 0.216 0.008 0.000
#> GSM96999 1 0.6792 0.3150 0.484 0.016 0.020 0.144 0.316 0.020
#> GSM97001 5 0.4621 0.6164 0.128 0.000 0.140 0.012 0.720 0.000
#> GSM97005 5 0.4353 0.6054 0.164 0.004 0.000 0.060 0.752 0.020
#> GSM97006 1 0.6329 0.1826 0.452 0.024 0.000 0.412 0.068 0.044
#> GSM97021 5 0.4450 0.5508 0.264 0.028 0.016 0.000 0.688 0.004
#> GSM97028 2 0.5069 -0.2194 0.020 0.476 0.000 0.004 0.028 0.472
#> GSM97031 5 0.5554 0.1846 0.032 0.008 0.000 0.044 0.512 0.404
#> GSM97037 3 0.5216 0.2758 0.004 0.080 0.568 0.004 0.000 0.344
#> GSM97018 2 0.4502 0.4713 0.032 0.720 0.016 0.004 0.008 0.220
#> GSM97014 5 0.4578 0.2473 0.004 0.008 0.424 0.016 0.548 0.000
#> GSM97042 2 0.3390 0.6505 0.000 0.816 0.140 0.016 0.028 0.000
#> GSM97040 5 0.4495 0.6070 0.180 0.024 0.048 0.000 0.740 0.008
#> GSM97041 5 0.5452 0.4078 0.380 0.036 0.052 0.000 0.532 0.000
#> GSM96955 5 0.6970 0.2425 0.004 0.240 0.172 0.100 0.484 0.000
#> GSM96990 6 0.4452 0.6571 0.020 0.176 0.060 0.004 0.000 0.740
#> GSM96991 2 0.2867 0.6298 0.004 0.872 0.076 0.032 0.016 0.000
#> GSM97048 3 0.0146 0.7973 0.000 0.000 0.996 0.004 0.000 0.000
#> GSM96963 2 0.4384 0.6143 0.004 0.764 0.140 0.040 0.052 0.000
#> GSM96953 2 0.5212 0.4776 0.000 0.572 0.348 0.020 0.060 0.000
#> GSM96966 4 0.1644 0.8042 0.052 0.004 0.000 0.932 0.012 0.000
#> GSM96979 6 0.4582 0.5010 0.012 0.020 0.000 0.280 0.016 0.672
#> GSM96983 6 0.4339 0.6374 0.004 0.216 0.000 0.012 0.044 0.724
#> GSM96984 6 0.1874 0.7695 0.008 0.012 0.000 0.020 0.028 0.932
#> GSM96994 6 0.2007 0.7736 0.008 0.016 0.000 0.012 0.040 0.924
#> GSM96996 1 0.6122 0.3007 0.492 0.028 0.000 0.376 0.088 0.016
#> GSM96997 6 0.3136 0.7449 0.008 0.024 0.000 0.052 0.052 0.864
#> GSM97007 6 0.1332 0.7738 0.008 0.000 0.000 0.012 0.028 0.952
#> GSM96954 6 0.2807 0.7581 0.016 0.028 0.000 0.000 0.088 0.868
#> GSM96962 6 0.1490 0.7730 0.008 0.004 0.000 0.016 0.024 0.948
#> GSM96969 4 0.1666 0.8059 0.036 0.008 0.000 0.936 0.020 0.000
#> GSM96970 4 0.1334 0.8052 0.032 0.000 0.000 0.948 0.020 0.000
#> GSM96973 4 0.0993 0.8058 0.024 0.000 0.000 0.964 0.012 0.000
#> GSM96976 4 0.3064 0.7305 0.004 0.092 0.004 0.860 0.024 0.016
#> GSM96977 5 0.5389 0.5236 0.268 0.016 0.004 0.040 0.640 0.032
#> GSM96995 6 0.3571 0.6426 0.000 0.020 0.004 0.000 0.216 0.760
#> GSM97002 1 0.6011 0.1091 0.452 0.036 0.000 0.432 0.068 0.012
#> GSM97009 5 0.5340 0.0405 0.012 0.020 0.448 0.020 0.492 0.008
#> GSM97010 3 0.5952 0.1989 0.028 0.024 0.540 0.360 0.016 0.032
#> GSM96974 4 0.3013 0.7377 0.008 0.116 0.000 0.848 0.004 0.024
#> GSM96985 2 0.6522 -0.1339 0.044 0.456 0.000 0.392 0.076 0.032
#> GSM96959 5 0.5182 0.5492 0.000 0.000 0.220 0.012 0.644 0.124
#> GSM96972 4 0.2587 0.7459 0.120 0.004 0.000 0.864 0.004 0.008
#> GSM96978 6 0.6585 0.3314 0.012 0.348 0.000 0.100 0.064 0.476
#> GSM96967 4 0.1668 0.7990 0.060 0.008 0.000 0.928 0.004 0.000
#> GSM96987 1 0.2002 0.6539 0.908 0.004 0.000 0.076 0.012 0.000
#> GSM97011 5 0.3230 0.6276 0.032 0.004 0.064 0.044 0.856 0.000
#> GSM96964 1 0.2152 0.6399 0.912 0.012 0.000 0.040 0.036 0.000
#> GSM96965 4 0.1823 0.7932 0.016 0.004 0.012 0.932 0.036 0.000
#> GSM96981 5 0.4988 0.4792 0.096 0.016 0.000 0.220 0.668 0.000
#> GSM96982 5 0.6339 0.0959 0.152 0.036 0.000 0.376 0.436 0.000
#> GSM96988 6 0.5679 0.1880 0.016 0.452 0.000 0.024 0.048 0.460
#> GSM97000 5 0.4409 0.5742 0.032 0.004 0.000 0.028 0.728 0.208
#> GSM97004 4 0.5072 -0.1738 0.472 0.024 0.000 0.476 0.024 0.004
#> GSM97008 5 0.2955 0.6229 0.084 0.000 0.016 0.016 0.868 0.016
#> GSM96950 1 0.2419 0.6398 0.896 0.016 0.000 0.060 0.028 0.000
#> GSM96980 4 0.4274 0.6126 0.200 0.024 0.000 0.736 0.040 0.000
#> GSM96989 1 0.1946 0.6534 0.912 0.004 0.000 0.072 0.012 0.000
#> GSM96992 1 0.6141 0.2224 0.444 0.016 0.000 0.176 0.364 0.000
#> GSM96993 1 0.2152 0.5755 0.904 0.068 0.000 0.004 0.024 0.000
#> GSM96958 5 0.5011 0.2440 0.392 0.000 0.000 0.064 0.540 0.004
#> GSM96951 5 0.4710 0.3790 0.360 0.000 0.000 0.024 0.596 0.020
#> GSM96952 1 0.5724 0.3969 0.560 0.020 0.000 0.128 0.292 0.000
#> GSM96961 1 0.3931 0.5595 0.756 0.008 0.000 0.044 0.192 0.000
Heatmaps for the consensus matrix. It visualizes the probability of two samples to be in a same group.
consensus_heatmap(res, k = 2)
consensus_heatmap(res, k = 3)
consensus_heatmap(res, k = 4)
consensus_heatmap(res, k = 5)
consensus_heatmap(res, k = 6)
Heatmaps for the membership of samples in all partitions to see how consistent they are:
membership_heatmap(res, k = 2)
membership_heatmap(res, k = 3)
membership_heatmap(res, k = 4)
membership_heatmap(res, k = 5)
membership_heatmap(res, k = 6)
As soon as we have had the classes for columns, we can look for signatures which are significantly different between classes which can be candidate marks for certain classes. Following are the heatmaps for signatures.
Signature heatmaps where rows are scaled:
get_signatures(res, k = 2)
get_signatures(res, k = 3)
get_signatures(res, k = 4)
get_signatures(res, k = 5)
get_signatures(res, k = 6)
Signature heatmaps where rows are not scaled:
get_signatures(res, k = 2, scale_rows = FALSE)
get_signatures(res, k = 3, scale_rows = FALSE)
get_signatures(res, k = 4, scale_rows = FALSE)
get_signatures(res, k = 5, scale_rows = FALSE)
get_signatures(res, k = 6, scale_rows = FALSE)
Compare the overlap of signatures from different k:
compare_signatures(res)
get_signature() returns a data frame invisibly. TO get the list of signatures, the function
call should be assigned to a variable explicitly. In following code, if plot argument is set
to FALSE, no heatmap is plotted while only the differential analysis is performed.
# code only for demonstration
tb = get_signature(res, k = ..., plot = FALSE)
An example of the output of tb is:
#> which_row fdr mean_1 mean_2 scaled_mean_1 scaled_mean_2 km
#> 1 38 0.042760348 8.373488 9.131774 -0.5533452 0.5164555 1
#> 2 40 0.018707592 7.106213 8.469186 -0.6173731 0.5762149 1
#> 3 55 0.019134737 10.221463 11.207825 -0.6159697 0.5749050 1
#> 4 59 0.006059896 5.921854 7.869574 -0.6899429 0.6439467 1
#> 5 60 0.018055526 8.928898 10.211722 -0.6204761 0.5791110 1
#> 6 98 0.009384629 15.714769 14.887706 0.6635654 -0.6193277 2
...
The columns in tb are:
which_row: row indices corresponding to the input matrix.fdr: FDR for the differential test. mean_x: The mean value in group x.scaled_mean_x: The mean value in group x after rows are scaled.km: Row groups if k-means clustering is applied to rows.UMAP plot which shows how samples are separated.
dimension_reduction(res, k = 2, method = "UMAP")
dimension_reduction(res, k = 3, method = "UMAP")
dimension_reduction(res, k = 4, method = "UMAP")
dimension_reduction(res, k = 5, method = "UMAP")
dimension_reduction(res, k = 6, method = "UMAP")
Following heatmap shows how subgroups are split when increasing k:
collect_classes(res)
Test correlation between subgroups and known annotations. If the known annotation is numeric, one-way ANOVA test is applied, and if the known annotation is discrete, chi-squared contingency table test is applied.
test_to_known_factors(res)
#> n disease.state(p) specimen(p) cell.type(p) other(p) k
#> SD:NMF 99 1.61e-05 0.1658 3.04e-14 0.1381 2
#> SD:NMF 96 3.56e-05 0.1686 1.18e-17 0.0543 3
#> SD:NMF 69 3.15e-04 0.0933 6.59e-12 0.0395 4
#> SD:NMF 68 3.97e-03 0.2354 2.68e-14 0.0454 5
#> SD:NMF 65 4.40e-03 0.4976 7.36e-13 0.0693 6
If matrix rows can be associated to genes, consider to use functional_enrichment(res,
...) to perform function enrichment for the signature genes. See this vignette for more detailed explanations.
The object with results only for a single top-value method and a single partition method can be extracted as:
res = res_list["CV", "hclust"]
# you can also extract it by
# res = res_list["CV:hclust"]
A summary of res and all the functions that can be applied to it:
res
#> A 'ConsensusPartition' object with k = 2, 3, 4, 5, 6.
#> On a matrix with 21168 rows and 100 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 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 0.318 0.647 0.841 0.4340 0.560 0.560
#> 3 3 0.391 0.699 0.817 0.4231 0.776 0.617
#> 4 4 0.522 0.729 0.830 0.1307 0.907 0.763
#> 5 5 0.575 0.676 0.773 0.0671 1.000 1.000
#> 6 6 0.607 0.579 0.741 0.0594 0.893 0.656
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
#> GSM97038 2 0.2236 0.7873 0.036 0.964
#> GSM97045 2 0.2603 0.7843 0.044 0.956
#> GSM97047 1 0.9209 0.5626 0.664 0.336
#> GSM97025 2 0.2603 0.7843 0.044 0.956
#> GSM97030 2 0.9833 0.1949 0.424 0.576
#> GSM97027 2 0.2423 0.7851 0.040 0.960
#> GSM97033 2 0.1184 0.7916 0.016 0.984
#> GSM97034 1 0.8499 0.6521 0.724 0.276
#> GSM97020 2 0.2236 0.7871 0.036 0.964
#> GSM97026 1 0.9286 0.5502 0.656 0.344
#> GSM97012 2 0.0376 0.7928 0.004 0.996
#> GSM97015 2 0.9866 0.1670 0.432 0.568
#> GSM97016 2 0.0376 0.7928 0.004 0.996
#> GSM97017 1 0.3733 0.8025 0.928 0.072
#> GSM97019 2 0.0672 0.7933 0.008 0.992
#> GSM97022 2 0.0376 0.7928 0.004 0.996
#> GSM97035 2 0.0376 0.7928 0.004 0.996
#> GSM97036 1 0.3274 0.8012 0.940 0.060
#> GSM97039 2 0.0376 0.7928 0.004 0.996
#> GSM97046 2 0.0376 0.7928 0.004 0.996
#> GSM97023 1 0.1414 0.8012 0.980 0.020
#> GSM97029 1 0.7139 0.7403 0.804 0.196
#> GSM97043 1 0.9922 0.2896 0.552 0.448
#> GSM97013 1 0.1843 0.8015 0.972 0.028
#> GSM96956 2 0.4939 0.7391 0.108 0.892
#> GSM97024 2 0.3733 0.7664 0.072 0.928
#> GSM97032 2 0.9850 0.1811 0.428 0.572
#> GSM97044 2 0.9850 0.1811 0.428 0.572
#> GSM97049 2 0.0376 0.7928 0.004 0.996
#> GSM96968 1 0.9248 0.5424 0.660 0.340
#> GSM96971 1 0.8327 0.6328 0.736 0.264
#> GSM96986 1 0.9833 0.3221 0.576 0.424
#> GSM97003 1 0.0376 0.7957 0.996 0.004
#> GSM96957 1 0.5842 0.7778 0.860 0.140
#> GSM96960 1 0.0376 0.7957 0.996 0.004
#> GSM96975 1 0.3733 0.8037 0.928 0.072
#> GSM96998 1 0.0938 0.7969 0.988 0.012
#> GSM96999 1 0.5842 0.7778 0.860 0.140
#> GSM97001 1 0.5842 0.7778 0.860 0.140
#> GSM97005 1 0.4690 0.7954 0.900 0.100
#> GSM97006 1 0.0376 0.7957 0.996 0.004
#> GSM97021 1 0.5178 0.7887 0.884 0.116
#> GSM97028 1 0.9833 0.3340 0.576 0.424
#> GSM97031 1 0.7745 0.6794 0.772 0.228
#> GSM97037 2 0.6712 0.6723 0.176 0.824
#> GSM97018 1 0.9944 0.2400 0.544 0.456
#> GSM97014 1 0.9248 0.5506 0.660 0.340
#> GSM97042 2 0.0376 0.7928 0.004 0.996
#> GSM97040 1 0.7950 0.6987 0.760 0.240
#> GSM97041 1 0.3733 0.8025 0.928 0.072
#> GSM96955 2 0.7602 0.6247 0.220 0.780
#> GSM96990 2 0.9833 0.1947 0.424 0.576
#> GSM96991 2 0.0938 0.7928 0.012 0.988
#> GSM97048 2 0.0376 0.7928 0.004 0.996
#> GSM96963 2 0.0938 0.7928 0.012 0.988
#> GSM96953 2 0.0376 0.7928 0.004 0.996
#> GSM96966 1 0.1633 0.8030 0.976 0.024
#> GSM96979 1 0.9815 0.3337 0.580 0.420
#> GSM96983 2 1.0000 -0.0668 0.496 0.504
#> GSM96984 1 0.9933 0.2319 0.548 0.452
#> GSM96994 1 0.9896 0.2736 0.560 0.440
#> GSM96996 1 0.0938 0.7969 0.988 0.012
#> GSM96997 1 0.9922 0.2461 0.552 0.448
#> GSM97007 1 0.9933 0.2319 0.548 0.452
#> GSM96954 1 0.9286 0.5084 0.656 0.344
#> GSM96962 1 0.9815 0.3337 0.580 0.420
#> GSM96969 1 0.1633 0.8026 0.976 0.024
#> GSM96970 1 0.1843 0.8028 0.972 0.028
#> GSM96973 1 0.2603 0.8020 0.956 0.044
#> GSM96976 1 0.3733 0.7997 0.928 0.072
#> GSM96977 1 0.6623 0.7609 0.828 0.172
#> GSM96995 2 0.9815 0.2078 0.420 0.580
#> GSM97002 1 0.0376 0.7957 0.996 0.004
#> GSM97009 1 0.8499 0.6543 0.724 0.276
#> GSM97010 1 0.2603 0.8038 0.956 0.044
#> GSM96974 1 0.3584 0.8002 0.932 0.068
#> GSM96985 1 1.0000 0.0542 0.504 0.496
#> GSM96959 2 0.9754 0.2167 0.408 0.592
#> GSM96972 1 0.0376 0.7957 0.996 0.004
#> GSM96978 2 1.0000 -0.0668 0.496 0.504
#> GSM96967 1 0.1414 0.8025 0.980 0.020
#> GSM96987 1 0.1633 0.8011 0.976 0.024
#> GSM97011 1 0.8499 0.6543 0.724 0.276
#> GSM96964 1 0.1843 0.8015 0.972 0.028
#> GSM96965 1 0.3733 0.8016 0.928 0.072
#> GSM96981 1 0.1414 0.8035 0.980 0.020
#> GSM96982 1 0.0938 0.8016 0.988 0.012
#> GSM96988 1 0.9881 0.2883 0.564 0.436
#> GSM97000 1 0.6438 0.7645 0.836 0.164
#> GSM97004 1 0.0376 0.7957 0.996 0.004
#> GSM97008 1 0.4690 0.7954 0.900 0.100
#> GSM96950 1 0.5294 0.7897 0.880 0.120
#> GSM96980 1 0.0376 0.7957 0.996 0.004
#> GSM96989 1 0.1633 0.8011 0.976 0.024
#> GSM96992 1 0.0000 0.7972 1.000 0.000
#> GSM96993 1 0.3274 0.8012 0.940 0.060
#> GSM96958 1 0.5178 0.7901 0.884 0.116
#> GSM96951 1 0.1843 0.8039 0.972 0.028
#> GSM96952 1 0.0000 0.7972 1.000 0.000
#> GSM96961 1 0.0000 0.7972 1.000 0.000
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM97038 2 0.2982 0.844 0.024 0.920 0.056
#> GSM97045 2 0.1765 0.873 0.040 0.956 0.004
#> GSM97047 1 0.9203 0.434 0.536 0.248 0.216
#> GSM97025 2 0.1765 0.873 0.040 0.956 0.004
#> GSM97030 3 0.6059 0.748 0.048 0.188 0.764
#> GSM97027 2 0.1647 0.875 0.036 0.960 0.004
#> GSM97033 2 0.0829 0.886 0.012 0.984 0.004
#> GSM97034 1 0.8961 0.299 0.504 0.136 0.360
#> GSM97020 2 0.1525 0.878 0.032 0.964 0.004
#> GSM97026 1 0.9424 0.234 0.472 0.188 0.340
#> GSM97012 2 0.0237 0.889 0.000 0.996 0.004
#> GSM97015 3 0.6098 0.760 0.056 0.176 0.768
#> GSM97016 2 0.0237 0.889 0.000 0.996 0.004
#> GSM97017 1 0.4563 0.774 0.852 0.036 0.112
#> GSM97019 2 0.0475 0.889 0.004 0.992 0.004
#> GSM97022 2 0.0237 0.889 0.000 0.996 0.004
#> GSM97035 2 0.0237 0.889 0.000 0.996 0.004
#> GSM97036 1 0.4058 0.777 0.880 0.044 0.076
#> GSM97039 2 0.0000 0.889 0.000 1.000 0.000
#> GSM97046 2 0.0000 0.889 0.000 1.000 0.000
#> GSM97023 1 0.2860 0.778 0.912 0.004 0.084
#> GSM97029 1 0.7596 0.650 0.672 0.100 0.228
#> GSM97043 3 0.9962 0.141 0.344 0.292 0.364
#> GSM97013 1 0.3207 0.778 0.904 0.012 0.084
#> GSM96956 2 0.5591 0.527 0.000 0.696 0.304
#> GSM97024 2 0.4413 0.778 0.024 0.852 0.124
#> GSM97032 3 0.6054 0.756 0.052 0.180 0.768
#> GSM97044 3 0.6001 0.760 0.052 0.176 0.772
#> GSM97049 2 0.0000 0.889 0.000 1.000 0.000
#> GSM96968 3 0.8058 0.257 0.376 0.072 0.552
#> GSM96971 3 0.5285 0.585 0.244 0.004 0.752
#> GSM96986 3 0.3500 0.790 0.116 0.004 0.880
#> GSM97003 1 0.1163 0.759 0.972 0.000 0.028
#> GSM96957 1 0.6354 0.721 0.744 0.052 0.204
#> GSM96960 1 0.1411 0.757 0.964 0.000 0.036
#> GSM96975 1 0.4921 0.770 0.816 0.020 0.164
#> GSM96998 1 0.1267 0.759 0.972 0.004 0.024
#> GSM96999 1 0.6354 0.721 0.744 0.052 0.204
#> GSM97001 1 0.6354 0.721 0.744 0.052 0.204
#> GSM97005 1 0.5058 0.763 0.820 0.032 0.148
#> GSM97006 1 0.1289 0.757 0.968 0.000 0.032
#> GSM97021 1 0.5719 0.753 0.792 0.052 0.156
#> GSM97028 3 0.5618 0.760 0.156 0.048 0.796
#> GSM97031 1 0.6260 0.271 0.552 0.000 0.448
#> GSM97037 2 0.6483 0.114 0.004 0.544 0.452
#> GSM97018 3 0.6902 0.747 0.148 0.116 0.736
#> GSM97014 1 0.8600 0.508 0.580 0.284 0.136
#> GSM97042 2 0.0237 0.889 0.000 0.996 0.004
#> GSM97040 1 0.8171 0.612 0.644 0.172 0.184
#> GSM97041 1 0.4563 0.774 0.852 0.036 0.112
#> GSM96955 2 0.8016 0.445 0.108 0.632 0.260
#> GSM96990 3 0.6447 0.741 0.060 0.196 0.744
#> GSM96991 2 0.1529 0.875 0.000 0.960 0.040
#> GSM97048 2 0.0000 0.889 0.000 1.000 0.000
#> GSM96963 2 0.1529 0.875 0.000 0.960 0.040
#> GSM96953 2 0.0237 0.889 0.000 0.996 0.004
#> GSM96966 1 0.5859 0.544 0.656 0.000 0.344
#> GSM96979 3 0.3425 0.788 0.112 0.004 0.884
#> GSM96983 3 0.2031 0.793 0.016 0.032 0.952
#> GSM96984 3 0.1878 0.800 0.044 0.004 0.952
#> GSM96994 3 0.2845 0.804 0.068 0.012 0.920
#> GSM96996 1 0.1399 0.761 0.968 0.004 0.028
#> GSM96997 3 0.2096 0.801 0.052 0.004 0.944
#> GSM97007 3 0.1878 0.800 0.044 0.004 0.952
#> GSM96954 3 0.5202 0.686 0.220 0.008 0.772
#> GSM96962 3 0.3425 0.788 0.112 0.004 0.884
#> GSM96969 1 0.6008 0.498 0.628 0.000 0.372
#> GSM96970 1 0.6008 0.492 0.628 0.000 0.372
#> GSM96973 1 0.6140 0.453 0.596 0.000 0.404
#> GSM96976 1 0.6659 0.363 0.532 0.008 0.460
#> GSM96977 1 0.7157 0.641 0.668 0.056 0.276
#> GSM96995 3 0.6922 0.729 0.080 0.200 0.720
#> GSM97002 1 0.1163 0.759 0.972 0.000 0.028
#> GSM97009 1 0.8525 0.571 0.612 0.200 0.188
#> GSM97010 1 0.4677 0.779 0.840 0.028 0.132
#> GSM96974 1 0.6291 0.357 0.532 0.000 0.468
#> GSM96985 3 0.3237 0.801 0.056 0.032 0.912
#> GSM96959 2 0.9820 -0.237 0.244 0.396 0.360
#> GSM96972 1 0.4399 0.670 0.812 0.000 0.188
#> GSM96978 3 0.2031 0.793 0.016 0.032 0.952
#> GSM96967 1 0.5905 0.535 0.648 0.000 0.352
#> GSM96987 1 0.2955 0.779 0.912 0.008 0.080
#> GSM97011 1 0.8525 0.571 0.612 0.200 0.188
#> GSM96964 1 0.3120 0.781 0.908 0.012 0.080
#> GSM96965 1 0.6735 0.450 0.564 0.012 0.424
#> GSM96981 1 0.3272 0.779 0.892 0.004 0.104
#> GSM96982 1 0.2590 0.775 0.924 0.004 0.072
#> GSM96988 3 0.4489 0.803 0.108 0.036 0.856
#> GSM97000 1 0.6746 0.712 0.732 0.076 0.192
#> GSM97004 1 0.1289 0.757 0.968 0.000 0.032
#> GSM97008 1 0.5058 0.763 0.820 0.032 0.148
#> GSM96950 1 0.5643 0.732 0.760 0.020 0.220
#> GSM96980 1 0.2356 0.751 0.928 0.000 0.072
#> GSM96989 1 0.2955 0.779 0.912 0.008 0.080
#> GSM96992 1 0.1964 0.772 0.944 0.000 0.056
#> GSM96993 1 0.4232 0.778 0.872 0.044 0.084
#> GSM96958 1 0.5521 0.748 0.788 0.032 0.180
#> GSM96951 1 0.3192 0.773 0.888 0.000 0.112
#> GSM96952 1 0.1964 0.772 0.944 0.000 0.056
#> GSM96961 1 0.1964 0.772 0.944 0.000 0.056
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM97038 2 0.2652 0.8448 0.028 0.912 0.056 0.004
#> GSM97045 2 0.1398 0.8722 0.040 0.956 0.004 0.000
#> GSM97047 1 0.7091 0.5007 0.592 0.244 0.156 0.008
#> GSM97025 2 0.1398 0.8722 0.040 0.956 0.004 0.000
#> GSM97030 3 0.5200 0.7454 0.052 0.188 0.752 0.008
#> GSM97027 2 0.1305 0.8744 0.036 0.960 0.004 0.000
#> GSM97033 2 0.0657 0.8837 0.012 0.984 0.004 0.000
#> GSM97034 1 0.7362 0.3156 0.540 0.136 0.312 0.012
#> GSM97020 2 0.1209 0.8767 0.032 0.964 0.004 0.000
#> GSM97026 1 0.7430 0.3530 0.548 0.184 0.260 0.008
#> GSM97012 2 0.0188 0.8858 0.000 0.996 0.004 0.000
#> GSM97015 3 0.5159 0.7527 0.064 0.176 0.756 0.004
#> GSM97016 2 0.0376 0.8854 0.000 0.992 0.004 0.004
#> GSM97017 1 0.2297 0.7900 0.932 0.032 0.024 0.012
#> GSM97019 2 0.0376 0.8859 0.004 0.992 0.004 0.000
#> GSM97022 2 0.0188 0.8858 0.000 0.996 0.004 0.000
#> GSM97035 2 0.0188 0.8858 0.000 0.996 0.004 0.000
#> GSM97036 1 0.1913 0.7893 0.940 0.040 0.000 0.020
#> GSM97039 2 0.0188 0.8855 0.000 0.996 0.000 0.004
#> GSM97046 2 0.0188 0.8855 0.000 0.996 0.000 0.004
#> GSM97023 1 0.0592 0.7874 0.984 0.000 0.000 0.016
#> GSM97029 1 0.5585 0.6816 0.748 0.096 0.144 0.012
#> GSM97043 1 0.8170 -0.1074 0.360 0.292 0.340 0.008
#> GSM97013 1 0.0927 0.7884 0.976 0.008 0.000 0.016
#> GSM96956 2 0.4608 0.5046 0.000 0.692 0.304 0.004
#> GSM97024 2 0.3447 0.7760 0.020 0.852 0.128 0.000
#> GSM97032 3 0.5055 0.7512 0.056 0.180 0.760 0.004
#> GSM97044 3 0.5013 0.7539 0.056 0.176 0.764 0.004
#> GSM97049 2 0.0188 0.8855 0.000 0.996 0.000 0.004
#> GSM96968 3 0.6863 0.2206 0.404 0.072 0.512 0.012
#> GSM96971 3 0.5130 0.3245 0.016 0.000 0.652 0.332
#> GSM96986 3 0.3354 0.7844 0.084 0.000 0.872 0.044
#> GSM97003 1 0.4245 0.7309 0.784 0.000 0.020 0.196
#> GSM96957 1 0.4316 0.7556 0.824 0.048 0.120 0.008
#> GSM96960 1 0.4387 0.7266 0.776 0.000 0.024 0.200
#> GSM96975 1 0.4107 0.7924 0.848 0.020 0.088 0.044
#> GSM96998 1 0.3946 0.7496 0.812 0.004 0.012 0.172
#> GSM96999 1 0.4316 0.7556 0.824 0.048 0.120 0.008
#> GSM97001 1 0.4316 0.7556 0.824 0.048 0.120 0.008
#> GSM97005 1 0.2884 0.7855 0.900 0.028 0.068 0.004
#> GSM97006 1 0.4323 0.7251 0.776 0.000 0.020 0.204
#> GSM97021 1 0.3629 0.7743 0.868 0.048 0.076 0.008
#> GSM97028 3 0.4466 0.7498 0.156 0.040 0.800 0.004
#> GSM97031 1 0.6360 0.2689 0.516 0.000 0.420 0.064
#> GSM97037 2 0.5263 0.0630 0.008 0.544 0.448 0.000
#> GSM97018 3 0.5680 0.7376 0.148 0.108 0.736 0.008
#> GSM97014 1 0.5862 0.5849 0.664 0.280 0.048 0.008
#> GSM97042 2 0.0188 0.8858 0.000 0.996 0.004 0.000
#> GSM97040 1 0.5968 0.6593 0.716 0.168 0.104 0.012
#> GSM97041 1 0.2297 0.7900 0.932 0.032 0.024 0.012
#> GSM96955 2 0.6919 0.4873 0.176 0.608 0.212 0.004
#> GSM96990 3 0.5828 0.7284 0.084 0.196 0.712 0.008
#> GSM96991 2 0.1637 0.8578 0.000 0.940 0.060 0.000
#> GSM97048 2 0.0188 0.8855 0.000 0.996 0.000 0.004
#> GSM96963 2 0.1637 0.8578 0.000 0.940 0.060 0.000
#> GSM96953 2 0.0188 0.8858 0.000 0.996 0.004 0.000
#> GSM96966 4 0.3617 0.8870 0.076 0.000 0.064 0.860
#> GSM96979 3 0.3176 0.7831 0.084 0.000 0.880 0.036
#> GSM96983 3 0.0804 0.7689 0.000 0.008 0.980 0.012
#> GSM96984 3 0.1722 0.7722 0.008 0.000 0.944 0.048
#> GSM96994 3 0.2831 0.7885 0.044 0.008 0.908 0.040
#> GSM96996 1 0.4063 0.7511 0.808 0.004 0.016 0.172
#> GSM96997 3 0.1975 0.7751 0.016 0.000 0.936 0.048
#> GSM97007 3 0.1722 0.7722 0.008 0.000 0.944 0.048
#> GSM96954 3 0.5302 0.7015 0.164 0.004 0.752 0.080
#> GSM96962 3 0.3176 0.7831 0.084 0.000 0.880 0.036
#> GSM96969 4 0.3900 0.9034 0.072 0.000 0.084 0.844
#> GSM96970 4 0.3900 0.9035 0.072 0.000 0.084 0.844
#> GSM96973 4 0.3919 0.8947 0.056 0.000 0.104 0.840
#> GSM96976 4 0.4381 0.8425 0.032 0.008 0.152 0.808
#> GSM96977 1 0.5441 0.6779 0.736 0.052 0.200 0.012
#> GSM96995 3 0.6154 0.7115 0.104 0.200 0.688 0.008
#> GSM97002 1 0.4245 0.7309 0.784 0.000 0.020 0.196
#> GSM97009 1 0.6118 0.6305 0.692 0.196 0.104 0.008
#> GSM97010 1 0.3725 0.7952 0.872 0.024 0.052 0.052
#> GSM96974 4 0.4152 0.8397 0.032 0.000 0.160 0.808
#> GSM96985 3 0.2010 0.7789 0.040 0.008 0.940 0.012
#> GSM96959 2 0.8023 -0.0963 0.296 0.392 0.308 0.004
#> GSM96972 4 0.2973 0.7443 0.144 0.000 0.000 0.856
#> GSM96978 3 0.0804 0.7689 0.000 0.008 0.980 0.012
#> GSM96967 4 0.3691 0.8950 0.076 0.000 0.068 0.856
#> GSM96987 1 0.1909 0.7888 0.940 0.004 0.008 0.048
#> GSM97011 1 0.6118 0.6305 0.692 0.196 0.104 0.008
#> GSM96964 1 0.2186 0.7922 0.932 0.008 0.012 0.048
#> GSM96965 4 0.5102 0.8491 0.116 0.008 0.096 0.780
#> GSM96981 1 0.3128 0.7846 0.888 0.004 0.032 0.076
#> GSM96982 1 0.3932 0.7689 0.836 0.004 0.032 0.128
#> GSM96988 3 0.3806 0.7949 0.092 0.028 0.860 0.020
#> GSM97000 1 0.4768 0.7372 0.800 0.072 0.120 0.008
#> GSM97004 1 0.4323 0.7251 0.776 0.000 0.020 0.204
#> GSM97008 1 0.2884 0.7855 0.900 0.028 0.068 0.004
#> GSM96950 1 0.3898 0.7631 0.836 0.016 0.136 0.012
#> GSM96980 1 0.5105 0.4049 0.564 0.000 0.004 0.432
#> GSM96989 1 0.1909 0.7888 0.940 0.004 0.008 0.048
#> GSM96992 1 0.3278 0.7702 0.864 0.000 0.020 0.116
#> GSM96993 1 0.1863 0.7913 0.944 0.040 0.004 0.012
#> GSM96958 1 0.4413 0.7755 0.828 0.028 0.112 0.032
#> GSM96951 1 0.3903 0.7792 0.844 0.000 0.076 0.080
#> GSM96952 1 0.3278 0.7702 0.864 0.000 0.020 0.116
#> GSM96961 1 0.3278 0.7702 0.864 0.000 0.020 0.116
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM97038 2 0.2534 0.8413 0.020 0.908 0.052 0.004 NA
#> GSM97045 2 0.1569 0.8645 0.044 0.944 0.004 0.000 NA
#> GSM97047 1 0.6659 0.4568 0.588 0.216 0.148 0.000 NA
#> GSM97025 2 0.1569 0.8645 0.044 0.944 0.004 0.000 NA
#> GSM97030 3 0.5045 0.6733 0.052 0.164 0.744 0.004 NA
#> GSM97027 2 0.1365 0.8688 0.040 0.952 0.004 0.000 NA
#> GSM97033 2 0.0566 0.8812 0.012 0.984 0.004 0.000 NA
#> GSM97034 1 0.6900 0.2853 0.524 0.108 0.308 0.000 NA
#> GSM97020 2 0.1124 0.8724 0.036 0.960 0.004 0.000 NA
#> GSM97026 1 0.6972 0.3343 0.536 0.160 0.256 0.000 NA
#> GSM97012 2 0.0324 0.8830 0.000 0.992 0.000 0.004 NA
#> GSM97015 3 0.5043 0.6820 0.064 0.152 0.748 0.004 NA
#> GSM97016 2 0.0324 0.8825 0.000 0.992 0.004 0.000 NA
#> GSM97017 1 0.2234 0.7204 0.920 0.012 0.032 0.000 NA
#> GSM97019 2 0.0727 0.8825 0.004 0.980 0.000 0.004 NA
#> GSM97022 2 0.0324 0.8830 0.000 0.992 0.000 0.004 NA
#> GSM97035 2 0.0324 0.8830 0.000 0.992 0.000 0.004 NA
#> GSM97036 1 0.2264 0.7293 0.912 0.024 0.004 0.000 NA
#> GSM97039 2 0.0162 0.8828 0.000 0.996 0.000 0.000 NA
#> GSM97046 2 0.0162 0.8828 0.000 0.996 0.000 0.000 NA
#> GSM97023 1 0.1357 0.7329 0.948 0.000 0.004 0.000 NA
#> GSM97029 1 0.5258 0.6183 0.732 0.072 0.148 0.000 NA
#> GSM97043 1 0.7735 -0.0724 0.356 0.264 0.332 0.004 NA
#> GSM97013 1 0.1357 0.7320 0.948 0.000 0.004 0.000 NA
#> GSM96956 2 0.4067 0.5251 0.000 0.692 0.300 0.000 NA
#> GSM97024 2 0.3142 0.7756 0.016 0.852 0.124 0.004 NA
#> GSM97032 3 0.4957 0.6799 0.056 0.156 0.752 0.004 NA
#> GSM97044 3 0.4995 0.6867 0.056 0.152 0.752 0.004 NA
#> GSM97049 2 0.0162 0.8828 0.000 0.996 0.000 0.000 NA
#> GSM96968 3 0.7060 0.1797 0.368 0.048 0.468 0.004 NA
#> GSM96971 3 0.6301 0.2944 0.004 0.000 0.512 0.336 NA
#> GSM96986 3 0.4633 0.6741 0.036 0.000 0.696 0.004 NA
#> GSM97003 1 0.4917 0.6014 0.588 0.000 0.004 0.024 NA
#> GSM96957 1 0.4855 0.6778 0.764 0.032 0.100 0.000 NA
#> GSM96960 1 0.5146 0.5829 0.564 0.000 0.008 0.028 NA
#> GSM96975 1 0.4546 0.7231 0.756 0.012 0.056 0.000 NA
#> GSM96998 1 0.4526 0.6599 0.672 0.000 0.000 0.028 NA
#> GSM96999 1 0.4855 0.6778 0.764 0.032 0.100 0.000 NA
#> GSM97001 1 0.4855 0.6778 0.764 0.032 0.100 0.000 NA
#> GSM97005 1 0.2679 0.7147 0.892 0.004 0.048 0.000 NA
#> GSM97006 1 0.5041 0.5820 0.564 0.000 0.004 0.028 NA
#> GSM97021 1 0.3575 0.7021 0.848 0.020 0.056 0.000 NA
#> GSM97028 3 0.5220 0.6821 0.144 0.036 0.744 0.008 NA
#> GSM97031 1 0.6952 0.1754 0.412 0.000 0.320 0.008 NA
#> GSM97037 2 0.4882 0.1260 0.008 0.540 0.440 0.000 NA
#> GSM97018 3 0.5444 0.6716 0.136 0.096 0.728 0.008 NA
#> GSM97014 1 0.5570 0.5384 0.660 0.252 0.044 0.000 NA
#> GSM97042 2 0.0324 0.8830 0.000 0.992 0.000 0.004 NA
#> GSM97040 1 0.5630 0.5956 0.708 0.140 0.096 0.000 NA
#> GSM97041 1 0.2234 0.7204 0.920 0.012 0.032 0.000 NA
#> GSM96955 2 0.7079 0.4839 0.156 0.588 0.076 0.008 NA
#> GSM96990 3 0.5515 0.6581 0.084 0.172 0.708 0.004 NA
#> GSM96991 2 0.1571 0.8547 0.000 0.936 0.000 0.004 NA
#> GSM97048 2 0.0162 0.8828 0.000 0.996 0.000 0.000 NA
#> GSM96963 2 0.1571 0.8547 0.000 0.936 0.000 0.004 NA
#> GSM96953 2 0.0324 0.8830 0.000 0.992 0.000 0.004 NA
#> GSM96966 4 0.3117 0.8915 0.036 0.000 0.004 0.860 NA
#> GSM96979 3 0.4552 0.6790 0.040 0.000 0.716 0.004 NA
#> GSM96983 3 0.4492 0.6034 0.000 0.004 0.680 0.020 NA
#> GSM96984 3 0.3550 0.6677 0.000 0.000 0.760 0.004 NA
#> GSM96994 3 0.4063 0.6963 0.016 0.008 0.768 0.004 NA
#> GSM96996 1 0.4678 0.6614 0.668 0.000 0.004 0.028 NA
#> GSM96997 3 0.3607 0.6654 0.000 0.000 0.752 0.004 NA
#> GSM97007 3 0.3491 0.6705 0.000 0.000 0.768 0.004 NA
#> GSM96954 3 0.6087 0.6461 0.140 0.000 0.672 0.068 NA
#> GSM96962 3 0.4552 0.6790 0.040 0.000 0.716 0.004 NA
#> GSM96969 4 0.2236 0.9065 0.024 0.000 0.000 0.908 NA
#> GSM96970 4 0.2236 0.9066 0.024 0.000 0.000 0.908 NA
#> GSM96973 4 0.1386 0.8996 0.016 0.000 0.000 0.952 NA
#> GSM96976 4 0.1095 0.8635 0.000 0.008 0.012 0.968 NA
#> GSM96977 1 0.5662 0.6081 0.692 0.036 0.164 0.000 NA
#> GSM96995 3 0.5798 0.6406 0.104 0.176 0.684 0.004 NA
#> GSM97002 1 0.4917 0.6014 0.588 0.000 0.004 0.024 NA
#> GSM97009 1 0.5764 0.5737 0.688 0.168 0.096 0.000 NA
#> GSM97010 1 0.4522 0.7276 0.768 0.012 0.028 0.016 NA
#> GSM96974 4 0.0912 0.8639 0.000 0.000 0.016 0.972 NA
#> GSM96985 3 0.5221 0.6005 0.024 0.004 0.652 0.024 NA
#> GSM96959 2 0.7589 -0.1257 0.288 0.368 0.308 0.004 NA
#> GSM96972 4 0.4906 0.7511 0.076 0.000 0.000 0.692 NA
#> GSM96978 3 0.4405 0.6138 0.000 0.004 0.696 0.020 NA
#> GSM96967 4 0.2769 0.8971 0.032 0.000 0.000 0.876 NA
#> GSM96987 1 0.2338 0.7323 0.884 0.000 0.000 0.004 NA
#> GSM97011 1 0.5764 0.5737 0.688 0.168 0.096 0.000 NA
#> GSM96964 1 0.2536 0.7353 0.868 0.000 0.000 0.004 NA
#> GSM96965 4 0.2408 0.8292 0.096 0.008 0.000 0.892 NA
#> GSM96981 1 0.4146 0.6962 0.716 0.000 0.004 0.012 NA
#> GSM96982 1 0.4607 0.6629 0.656 0.000 0.004 0.020 NA
#> GSM96988 3 0.4417 0.7214 0.076 0.024 0.800 0.004 NA
#> GSM97000 1 0.4581 0.6750 0.788 0.040 0.096 0.000 NA
#> GSM97004 1 0.5041 0.5820 0.564 0.000 0.004 0.028 NA
#> GSM97008 1 0.2679 0.7147 0.892 0.004 0.048 0.000 NA
#> GSM96950 1 0.4298 0.6911 0.788 0.008 0.096 0.000 NA
#> GSM96980 1 0.6728 0.2588 0.404 0.000 0.000 0.260 NA
#> GSM96989 1 0.2338 0.7323 0.884 0.000 0.000 0.004 NA
#> GSM96992 1 0.3934 0.6998 0.748 0.000 0.004 0.012 NA
#> GSM96993 1 0.2104 0.7335 0.916 0.024 0.000 0.000 NA
#> GSM96958 1 0.4766 0.6998 0.760 0.012 0.084 0.004 NA
#> GSM96951 1 0.4579 0.7096 0.744 0.000 0.056 0.008 NA
#> GSM96952 1 0.3934 0.6998 0.748 0.000 0.004 0.012 NA
#> GSM96961 1 0.3934 0.6998 0.748 0.000 0.004 0.012 NA
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM97038 2 0.2774 0.8108 0.012 0.872 0.076 0.000 0.040 0.000
#> GSM97045 2 0.1398 0.8513 0.000 0.940 0.008 0.000 0.052 0.000
#> GSM97047 5 0.5298 0.4431 0.004 0.164 0.180 0.000 0.644 0.008
#> GSM97025 2 0.1398 0.8513 0.000 0.940 0.008 0.000 0.052 0.000
#> GSM97030 3 0.3449 0.6152 0.000 0.116 0.808 0.000 0.076 0.000
#> GSM97027 2 0.1265 0.8556 0.000 0.948 0.008 0.000 0.044 0.000
#> GSM97033 2 0.0508 0.8684 0.000 0.984 0.004 0.000 0.012 0.000
#> GSM97034 5 0.6030 0.2567 0.048 0.068 0.332 0.000 0.540 0.012
#> GSM97020 2 0.1124 0.8594 0.000 0.956 0.008 0.000 0.036 0.000
#> GSM97026 5 0.5937 0.3233 0.024 0.128 0.268 0.000 0.572 0.008
#> GSM97012 2 0.0260 0.8700 0.000 0.992 0.000 0.000 0.000 0.008
#> GSM97015 3 0.3810 0.6206 0.004 0.104 0.800 0.000 0.084 0.008
#> GSM97016 2 0.0922 0.8661 0.024 0.968 0.004 0.000 0.000 0.004
#> GSM97017 5 0.1606 0.5854 0.056 0.008 0.004 0.000 0.932 0.000
#> GSM97019 2 0.0622 0.8698 0.000 0.980 0.000 0.000 0.012 0.008
#> GSM97022 2 0.0260 0.8700 0.000 0.992 0.000 0.000 0.000 0.008
#> GSM97035 2 0.0260 0.8700 0.000 0.992 0.000 0.000 0.000 0.008
#> GSM97036 5 0.2551 0.5517 0.108 0.012 0.004 0.000 0.872 0.004
#> GSM97039 2 0.0777 0.8660 0.024 0.972 0.000 0.000 0.000 0.004
#> GSM97046 2 0.0858 0.8652 0.028 0.968 0.000 0.000 0.000 0.004
#> GSM97023 5 0.2260 0.5303 0.140 0.000 0.000 0.000 0.860 0.000
#> GSM97029 5 0.4808 0.5684 0.060 0.048 0.148 0.000 0.736 0.008
#> GSM97043 5 0.6599 -0.1379 0.024 0.224 0.372 0.000 0.376 0.004
#> GSM97013 5 0.2278 0.5393 0.128 0.000 0.004 0.000 0.868 0.000
#> GSM96956 2 0.3986 0.4932 0.020 0.664 0.316 0.000 0.000 0.000
#> GSM97024 2 0.2868 0.7627 0.000 0.852 0.112 0.000 0.032 0.004
#> GSM97032 3 0.3664 0.6174 0.000 0.108 0.804 0.000 0.080 0.008
#> GSM97044 3 0.3972 0.6110 0.000 0.104 0.792 0.000 0.080 0.024
#> GSM97049 2 0.0858 0.8652 0.028 0.968 0.000 0.000 0.000 0.004
#> GSM96968 3 0.6533 0.1663 0.092 0.008 0.488 0.000 0.336 0.076
#> GSM96971 6 0.6586 0.2838 0.024 0.000 0.224 0.344 0.004 0.404
#> GSM96986 6 0.4982 0.7403 0.024 0.000 0.292 0.000 0.052 0.632
#> GSM97003 1 0.2562 0.7481 0.828 0.000 0.000 0.000 0.172 0.000
#> GSM96957 5 0.5431 0.4888 0.184 0.000 0.128 0.000 0.652 0.036
#> GSM96960 1 0.3013 0.7387 0.828 0.000 0.004 0.004 0.152 0.012
#> GSM96975 5 0.5667 0.0131 0.412 0.000 0.060 0.004 0.492 0.032
#> GSM96998 1 0.3930 0.6271 0.628 0.000 0.004 0.004 0.364 0.000
#> GSM96999 5 0.5431 0.4888 0.184 0.000 0.128 0.000 0.652 0.036
#> GSM97001 5 0.5431 0.4888 0.184 0.000 0.128 0.000 0.652 0.036
#> GSM97005 5 0.2386 0.5863 0.064 0.000 0.028 0.000 0.896 0.012
#> GSM97006 1 0.2914 0.7398 0.832 0.000 0.004 0.004 0.152 0.008
#> GSM97021 5 0.2420 0.5964 0.028 0.008 0.044 0.000 0.904 0.016
#> GSM97028 3 0.6035 0.5388 0.040 0.028 0.656 0.008 0.148 0.120
#> GSM97031 5 0.7539 0.1027 0.204 0.000 0.180 0.000 0.356 0.260
#> GSM97037 2 0.4486 0.0429 0.008 0.512 0.464 0.000 0.016 0.000
#> GSM97018 3 0.5506 0.5863 0.020 0.068 0.696 0.004 0.156 0.056
#> GSM97014 5 0.4259 0.4948 0.004 0.228 0.040 0.000 0.720 0.008
#> GSM97042 2 0.0260 0.8700 0.000 0.992 0.000 0.000 0.000 0.008
#> GSM97040 5 0.4201 0.5642 0.008 0.108 0.104 0.000 0.772 0.008
#> GSM97041 5 0.1606 0.5854 0.056 0.008 0.004 0.000 0.932 0.000
#> GSM96955 2 0.7097 0.3987 0.048 0.552 0.184 0.008 0.152 0.056
#> GSM96990 3 0.4166 0.6085 0.000 0.124 0.760 0.000 0.108 0.008
#> GSM96991 2 0.1462 0.8473 0.000 0.936 0.056 0.000 0.000 0.008
#> GSM97048 2 0.0858 0.8652 0.028 0.968 0.000 0.000 0.000 0.004
#> GSM96963 2 0.1462 0.8473 0.000 0.936 0.056 0.000 0.000 0.008
#> GSM96953 2 0.0260 0.8700 0.000 0.992 0.000 0.000 0.000 0.008
#> GSM96966 4 0.2704 0.8806 0.140 0.000 0.000 0.844 0.016 0.000
#> GSM96979 6 0.4872 0.7571 0.020 0.000 0.284 0.000 0.052 0.644
#> GSM96983 3 0.4962 0.3246 0.064 0.004 0.628 0.008 0.000 0.296
#> GSM96984 6 0.3101 0.7645 0.000 0.000 0.244 0.000 0.000 0.756
#> GSM96994 6 0.5079 0.7152 0.012 0.008 0.332 0.000 0.048 0.600
#> GSM96996 1 0.4145 0.6294 0.628 0.000 0.008 0.004 0.356 0.004
#> GSM96997 6 0.3215 0.7646 0.004 0.000 0.240 0.000 0.000 0.756
#> GSM97007 6 0.3151 0.7623 0.000 0.000 0.252 0.000 0.000 0.748
#> GSM96954 3 0.7648 -0.1686 0.056 0.000 0.372 0.080 0.136 0.356
#> GSM96962 6 0.4872 0.7571 0.020 0.000 0.284 0.000 0.052 0.644
#> GSM96969 4 0.2263 0.8964 0.100 0.000 0.000 0.884 0.016 0.000
#> GSM96970 4 0.2263 0.8967 0.100 0.000 0.000 0.884 0.016 0.000
#> GSM96973 4 0.1584 0.8891 0.064 0.000 0.000 0.928 0.008 0.000
#> GSM96976 4 0.0665 0.8488 0.000 0.008 0.008 0.980 0.000 0.004
#> GSM96977 5 0.6272 0.4641 0.188 0.000 0.200 0.000 0.556 0.056
#> GSM96995 3 0.4450 0.5933 0.000 0.132 0.732 0.000 0.128 0.008
#> GSM97002 1 0.2562 0.7481 0.828 0.000 0.000 0.000 0.172 0.000
#> GSM97009 5 0.4684 0.5509 0.008 0.132 0.104 0.004 0.740 0.012
#> GSM97010 5 0.5748 -0.2863 0.448 0.008 0.036 0.020 0.468 0.020
#> GSM96974 4 0.0405 0.8492 0.000 0.000 0.008 0.988 0.000 0.004
#> GSM96985 3 0.5480 0.3225 0.084 0.004 0.608 0.012 0.008 0.284
#> GSM96959 2 0.6731 -0.2637 0.004 0.332 0.316 0.004 0.328 0.016
#> GSM96972 4 0.4459 0.7071 0.288 0.000 0.004 0.668 0.032 0.008
#> GSM96978 3 0.4822 0.3316 0.056 0.004 0.644 0.008 0.000 0.288
#> GSM96967 4 0.2623 0.8857 0.132 0.000 0.000 0.852 0.016 0.000
#> GSM96987 5 0.3330 0.3046 0.284 0.000 0.000 0.000 0.716 0.000
#> GSM97011 5 0.4684 0.5509 0.008 0.132 0.104 0.004 0.740 0.012
#> GSM96964 5 0.3808 0.3102 0.284 0.000 0.012 0.000 0.700 0.004
#> GSM96965 4 0.2256 0.8086 0.000 0.008 0.004 0.892 0.092 0.004
#> GSM96981 1 0.4590 0.6203 0.632 0.000 0.008 0.008 0.328 0.024
#> GSM96982 1 0.4138 0.6967 0.700 0.000 0.008 0.004 0.268 0.020
#> GSM96988 3 0.5809 0.3796 0.024 0.016 0.632 0.008 0.092 0.228
#> GSM97000 5 0.3911 0.5921 0.048 0.024 0.100 0.000 0.812 0.016
#> GSM97004 1 0.2914 0.7398 0.832 0.000 0.004 0.004 0.152 0.008
#> GSM97008 5 0.2386 0.5863 0.064 0.000 0.028 0.000 0.896 0.012
#> GSM96950 5 0.5465 0.4733 0.208 0.000 0.108 0.000 0.644 0.040
#> GSM96980 1 0.5082 0.4371 0.652 0.000 0.004 0.236 0.100 0.008
#> GSM96989 5 0.3330 0.3046 0.284 0.000 0.000 0.000 0.716 0.000
#> GSM96992 1 0.4242 0.5231 0.572 0.000 0.004 0.000 0.412 0.012
#> GSM96993 5 0.3018 0.5022 0.168 0.012 0.004 0.000 0.816 0.000
#> GSM96958 5 0.5695 0.3511 0.280 0.000 0.100 0.000 0.584 0.036
#> GSM96951 5 0.5247 -0.3127 0.460 0.000 0.016 0.000 0.468 0.056
#> GSM96952 1 0.4242 0.5231 0.572 0.000 0.004 0.000 0.412 0.012
#> GSM96961 1 0.4242 0.5231 0.572 0.000 0.004 0.000 0.412 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 disease.state(p) specimen(p) cell.type(p) other(p) k
#> CV:hclust 79 0.000928 0.304 3.11e-12 0.0387 2
#> CV:hclust 85 0.000637 0.483 2.04e-15 0.0545 3
#> CV:hclust 90 0.000328 0.179 6.02e-16 0.0387 4
#> CV:hclust 89 0.000187 0.161 2.22e-16 0.0325 5
#> CV:hclust 70 0.000439 0.210 3.25e-14 0.0128 6
If matrix rows can be associated to genes, consider to use functional_enrichment(res,
...) to perform function enrichment for the signature genes. See this vignette for more detailed explanations.
The object with results only for a single top-value method and a single partition method can be extracted as:
res = res_list["CV", "kmeans"]
# you can also extract it by
# res = res_list["CV:kmeans"]
A summary of res and all the functions that can be applied to it:
res
#> A 'ConsensusPartition' object with k = 2, 3, 4, 5, 6.
#> On a matrix with 21168 rows and 100 columns.
#> Top rows (1000, 2000, 3000, 4000, 5000) are extracted by 'CV' method.
#> Subgroups are detected by 'kmeans' method.
#> Performed in total 1250 partitions by row resampling.
#> Best k for subgroups seems to be 2.
#>
#> Following methods can be applied to this 'ConsensusPartition' object:
#> [1] "cola_report" "collect_classes" "collect_plots"
#> [4] "collect_stats" "colnames" "compare_signatures"
#> [7] "consensus_heatmap" "dimension_reduction" "functional_enrichment"
#> [10] "get_anno_col" "get_anno" "get_classes"
#> [13] "get_consensus" "get_matrix" "get_membership"
#> [16] "get_param" "get_signatures" "get_stats"
#> [19] "is_best_k" "is_stable_k" "membership_heatmap"
#> [22] "ncol" "nrow" "plot_ecdf"
#> [25] "rownames" "select_partition_number" "show"
#> [28] "suggest_best_k" "test_to_known_factors"
collect_plots() function collects all the plots made from res for all k (number of partitions)
into one single page to provide an easy and fast comparison between different k.
collect_plots(res)
The plots are:
k and the heatmap of
predicted classes for each k.k.k.k.All the plots in panels can be made by individual functions and they are plotted later in this section.
select_partition_number() produces several plots showing different
statistics for choosing “optimized” k. There are following statistics:
k;k, the area increased is defined as \(A_k - A_{k-1}\).The detailed explanations of these statistics can be found in the cola vignette.
Generally speaking, lower PAC score, higher mean silhouette score or higher
concordance corresponds to better partition. Rand index and Jaccard index
measure how similar the current partition is compared to partition with k-1.
If they are too similar, we won't accept k is better than k-1.
select_partition_number(res)
The numeric values for all these statistics can be obtained by get_stats().
get_stats(res)
#> k 1-PAC mean_silhouette concordance area_increased Rand Jaccard
#> 2 2 1.000 0.980 0.991 0.4938 0.508 0.508
#> 3 3 0.487 0.385 0.682 0.3047 0.772 0.606
#> 4 4 0.716 0.731 0.851 0.1370 0.747 0.472
#> 5 5 0.664 0.572 0.758 0.0676 0.865 0.584
#> 6 6 0.709 0.664 0.743 0.0454 0.921 0.680
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
#> GSM97038 2 0.000 0.995 0.000 1.000
#> GSM97045 2 0.000 0.995 0.000 1.000
#> GSM97047 2 0.000 0.995 0.000 1.000
#> GSM97025 2 0.000 0.995 0.000 1.000
#> GSM97030 2 0.000 0.995 0.000 1.000
#> GSM97027 2 0.000 0.995 0.000 1.000
#> GSM97033 2 0.000 0.995 0.000 1.000
#> GSM97034 2 0.000 0.995 0.000 1.000
#> GSM97020 2 0.000 0.995 0.000 1.000
#> GSM97026 2 0.000 0.995 0.000 1.000
#> GSM97012 2 0.000 0.995 0.000 1.000
#> GSM97015 2 0.000 0.995 0.000 1.000
#> GSM97016 2 0.000 0.995 0.000 1.000
#> GSM97017 1 0.000 0.988 1.000 0.000
#> GSM97019 2 0.000 0.995 0.000 1.000
#> GSM97022 2 0.000 0.995 0.000 1.000
#> GSM97035 2 0.000 0.995 0.000 1.000
#> GSM97036 1 0.000 0.988 1.000 0.000
#> GSM97039 2 0.000 0.995 0.000 1.000
#> GSM97046 2 0.000 0.995 0.000 1.000
#> GSM97023 1 0.000 0.988 1.000 0.000
#> GSM97029 1 0.000 0.988 1.000 0.000
#> GSM97043 2 0.000 0.995 0.000 1.000
#> GSM97013 1 0.000 0.988 1.000 0.000
#> GSM96956 2 0.000 0.995 0.000 1.000
#> GSM97024 2 0.000 0.995 0.000 1.000
#> GSM97032 2 0.000 0.995 0.000 1.000
#> GSM97044 2 0.000 0.995 0.000 1.000
#> GSM97049 2 0.000 0.995 0.000 1.000
#> GSM96968 1 0.584 0.837 0.860 0.140
#> GSM96971 1 0.000 0.988 1.000 0.000
#> GSM96986 1 0.000 0.988 1.000 0.000
#> GSM97003 1 0.000 0.988 1.000 0.000
#> GSM96957 1 0.000 0.988 1.000 0.000
#> GSM96960 1 0.000 0.988 1.000 0.000
#> GSM96975 1 0.000 0.988 1.000 0.000
#> GSM96998 1 0.000 0.988 1.000 0.000
#> GSM96999 1 0.000 0.988 1.000 0.000
#> GSM97001 1 0.000 0.988 1.000 0.000
#> GSM97005 1 0.000 0.988 1.000 0.000
#> GSM97006 1 0.000 0.988 1.000 0.000
#> GSM97021 1 0.000 0.988 1.000 0.000
#> GSM97028 2 0.141 0.978 0.020 0.980
#> GSM97031 1 0.000 0.988 1.000 0.000
#> GSM97037 2 0.000 0.995 0.000 1.000
#> GSM97018 2 0.000 0.995 0.000 1.000
#> GSM97014 2 0.000 0.995 0.000 1.000
#> GSM97042 2 0.000 0.995 0.000 1.000
#> GSM97040 2 0.000 0.995 0.000 1.000
#> GSM97041 1 0.000 0.988 1.000 0.000
#> GSM96955 2 0.000 0.995 0.000 1.000
#> GSM96990 2 0.000 0.995 0.000 1.000
#> GSM96991 2 0.000 0.995 0.000 1.000
#> GSM97048 2 0.000 0.995 0.000 1.000
#> GSM96963 2 0.000 0.995 0.000 1.000
#> GSM96953 2 0.000 0.995 0.000 1.000
#> GSM96966 1 0.000 0.988 1.000 0.000
#> GSM96979 1 0.000 0.988 1.000 0.000
#> GSM96983 2 0.000 0.995 0.000 1.000
#> GSM96984 1 0.184 0.961 0.972 0.028
#> GSM96994 2 0.224 0.963 0.036 0.964
#> GSM96996 1 0.000 0.988 1.000 0.000
#> GSM96997 1 0.000 0.988 1.000 0.000
#> GSM97007 2 0.295 0.947 0.052 0.948
#> GSM96954 1 0.000 0.988 1.000 0.000
#> GSM96962 1 0.000 0.988 1.000 0.000
#> GSM96969 1 0.000 0.988 1.000 0.000
#> GSM96970 1 0.000 0.988 1.000 0.000
#> GSM96973 1 0.000 0.988 1.000 0.000
#> GSM96976 1 0.871 0.599 0.708 0.292
#> GSM96977 1 0.000 0.988 1.000 0.000
#> GSM96995 2 0.443 0.900 0.092 0.908
#> GSM97002 1 0.000 0.988 1.000 0.000
#> GSM97009 2 0.000 0.995 0.000 1.000
#> GSM97010 1 0.000 0.988 1.000 0.000
#> GSM96974 1 0.000 0.988 1.000 0.000
#> GSM96985 1 0.000 0.988 1.000 0.000
#> GSM96959 2 0.000 0.995 0.000 1.000
#> GSM96972 1 0.000 0.988 1.000 0.000
#> GSM96978 1 0.802 0.685 0.756 0.244
#> GSM96967 1 0.000 0.988 1.000 0.000
#> GSM96987 1 0.000 0.988 1.000 0.000
#> GSM97011 1 0.000 0.988 1.000 0.000
#> GSM96964 1 0.000 0.988 1.000 0.000
#> GSM96965 1 0.000 0.988 1.000 0.000
#> GSM96981 1 0.000 0.988 1.000 0.000
#> GSM96982 1 0.000 0.988 1.000 0.000
#> GSM96988 1 0.000 0.988 1.000 0.000
#> GSM97000 1 0.000 0.988 1.000 0.000
#> GSM97004 1 0.000 0.988 1.000 0.000
#> GSM97008 1 0.000 0.988 1.000 0.000
#> GSM96950 1 0.000 0.988 1.000 0.000
#> GSM96980 1 0.000 0.988 1.000 0.000
#> GSM96989 1 0.000 0.988 1.000 0.000
#> GSM96992 1 0.000 0.988 1.000 0.000
#> GSM96993 1 0.000 0.988 1.000 0.000
#> GSM96958 1 0.000 0.988 1.000 0.000
#> GSM96951 1 0.000 0.988 1.000 0.000
#> GSM96952 1 0.000 0.988 1.000 0.000
#> GSM96961 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
#> GSM97038 3 0.6309 -0.2149 0.000 0.500 0.500
#> GSM97045 2 0.6302 0.1149 0.000 0.520 0.480
#> GSM97047 2 0.5860 0.2246 0.024 0.748 0.228
#> GSM97025 2 0.6302 0.1149 0.000 0.520 0.480
#> GSM97030 2 0.1031 0.2466 0.000 0.976 0.024
#> GSM97027 2 0.6302 0.1149 0.000 0.520 0.480
#> GSM97033 3 0.6309 -0.2149 0.000 0.500 0.500
#> GSM97034 2 0.0475 0.2492 0.004 0.992 0.004
#> GSM97020 3 0.6309 -0.2149 0.000 0.500 0.500
#> GSM97026 2 0.6075 0.1906 0.008 0.676 0.316
#> GSM97012 2 0.6299 0.1180 0.000 0.524 0.476
#> GSM97015 2 0.0475 0.2503 0.004 0.992 0.004
#> GSM97016 2 0.6309 0.0832 0.000 0.500 0.500
#> GSM97017 1 0.2066 0.8196 0.940 0.060 0.000
#> GSM97019 2 0.6299 0.1180 0.000 0.524 0.476
#> GSM97022 2 0.6299 0.1180 0.000 0.524 0.476
#> GSM97035 2 0.6299 0.1180 0.000 0.524 0.476
#> GSM97036 1 0.1337 0.8307 0.972 0.016 0.012
#> GSM97039 2 0.6309 0.0832 0.000 0.500 0.500
#> GSM97046 3 0.6309 -0.2149 0.000 0.500 0.500
#> GSM97023 1 0.0848 0.8292 0.984 0.008 0.008
#> GSM97029 1 0.2878 0.8059 0.904 0.096 0.000
#> GSM97043 2 0.6299 0.1180 0.000 0.524 0.476
#> GSM97013 1 0.0892 0.8290 0.980 0.020 0.000
#> GSM96956 2 0.6308 0.0939 0.000 0.508 0.492
#> GSM97024 2 0.6299 0.1180 0.000 0.524 0.476
#> GSM97032 2 0.0237 0.2490 0.000 0.996 0.004
#> GSM97044 2 0.1031 0.2414 0.000 0.976 0.024
#> GSM97049 2 0.6309 0.0832 0.000 0.500 0.500
#> GSM96968 2 0.9277 0.0452 0.328 0.496 0.176
#> GSM96971 3 0.7841 0.0637 0.052 0.468 0.480
#> GSM96986 2 0.9581 0.0369 0.288 0.476 0.236
#> GSM97003 1 0.5737 0.7938 0.804 0.092 0.104
#> GSM96957 1 0.3272 0.8001 0.892 0.104 0.004
#> GSM96960 1 0.3482 0.7992 0.872 0.000 0.128
#> GSM96975 1 0.1129 0.8299 0.976 0.020 0.004
#> GSM96998 1 0.1643 0.8248 0.956 0.000 0.044
#> GSM96999 1 0.3038 0.8010 0.896 0.104 0.000
#> GSM97001 1 0.3038 0.8010 0.896 0.104 0.000
#> GSM97005 1 0.3038 0.8010 0.896 0.104 0.000
#> GSM97006 1 0.2878 0.8121 0.904 0.000 0.096
#> GSM97021 1 0.4291 0.7376 0.820 0.180 0.000
#> GSM97028 2 0.5730 0.1538 0.060 0.796 0.144
#> GSM97031 1 0.4293 0.7540 0.832 0.164 0.004
#> GSM97037 2 0.6260 0.1289 0.000 0.552 0.448
#> GSM97018 2 0.0237 0.2501 0.004 0.996 0.000
#> GSM97014 2 0.8720 0.1579 0.108 0.480 0.412
#> GSM97042 2 0.6299 0.1180 0.000 0.524 0.476
#> GSM97040 2 0.6793 -0.0810 0.452 0.536 0.012
#> GSM97041 1 0.2796 0.8072 0.908 0.092 0.000
#> GSM96955 2 0.6267 0.1291 0.000 0.548 0.452
#> GSM96990 2 0.0983 0.2495 0.004 0.980 0.016
#> GSM96991 2 0.6299 0.1180 0.000 0.524 0.476
#> GSM97048 2 0.6309 0.0832 0.000 0.500 0.500
#> GSM96963 2 0.6299 0.1180 0.000 0.524 0.476
#> GSM96953 2 0.6299 0.1180 0.000 0.524 0.476
#> GSM96966 1 0.6260 0.5308 0.552 0.000 0.448
#> GSM96979 2 0.9598 0.0304 0.276 0.476 0.248
#> GSM96983 2 0.4834 0.1187 0.004 0.792 0.204
#> GSM96984 2 0.9557 0.0326 0.268 0.484 0.248
#> GSM96994 2 0.8175 0.0615 0.132 0.632 0.236
#> GSM96996 1 0.2796 0.8135 0.908 0.000 0.092
#> GSM96997 2 0.9598 0.0304 0.276 0.476 0.248
#> GSM97007 2 0.8058 0.0631 0.124 0.640 0.236
#> GSM96954 2 0.8865 -0.0804 0.404 0.476 0.120
#> GSM96962 2 0.9563 0.0386 0.284 0.480 0.236
#> GSM96969 1 0.6267 0.5267 0.548 0.000 0.452
#> GSM96970 1 0.6267 0.5267 0.548 0.000 0.452
#> GSM96973 1 0.6267 0.5267 0.548 0.000 0.452
#> GSM96976 3 0.7913 0.0978 0.056 0.452 0.492
#> GSM96977 1 0.5958 0.5758 0.692 0.300 0.008
#> GSM96995 2 0.7633 0.1323 0.264 0.652 0.084
#> GSM97002 1 0.3482 0.7992 0.872 0.000 0.128
#> GSM97009 2 0.9148 0.1937 0.236 0.544 0.220
#> GSM97010 1 0.5500 0.7848 0.816 0.084 0.100
#> GSM96974 3 0.9303 0.1350 0.184 0.316 0.500
#> GSM96985 3 0.9357 0.1134 0.196 0.304 0.500
#> GSM96959 2 0.4802 0.2105 0.156 0.824 0.020
#> GSM96972 1 0.6260 0.5308 0.552 0.000 0.448
#> GSM96978 2 0.8018 -0.1178 0.064 0.520 0.416
#> GSM96967 1 0.6267 0.5267 0.548 0.000 0.452
#> GSM96987 1 0.1411 0.8253 0.964 0.000 0.036
#> GSM97011 1 0.3619 0.7783 0.864 0.136 0.000
#> GSM96964 1 0.0848 0.8292 0.984 0.008 0.008
#> GSM96965 1 0.6754 0.5387 0.556 0.012 0.432
#> GSM96981 1 0.2066 0.8220 0.940 0.000 0.060
#> GSM96982 1 0.4346 0.7682 0.816 0.000 0.184
#> GSM96988 2 0.9464 0.0332 0.248 0.500 0.252
#> GSM97000 1 0.6026 0.4507 0.624 0.376 0.000
#> GSM97004 1 0.4504 0.7570 0.804 0.000 0.196
#> GSM97008 1 0.4235 0.7416 0.824 0.176 0.000
#> GSM96950 1 0.0892 0.8290 0.980 0.020 0.000
#> GSM96980 1 0.6111 0.5807 0.604 0.000 0.396
#> GSM96989 1 0.1411 0.8253 0.964 0.000 0.036
#> GSM96992 1 0.1643 0.8248 0.956 0.000 0.044
#> GSM96993 1 0.1860 0.8238 0.948 0.052 0.000
#> GSM96958 1 0.0424 0.8294 0.992 0.008 0.000
#> GSM96951 1 0.0592 0.8297 0.988 0.012 0.000
#> GSM96952 1 0.1529 0.8250 0.960 0.000 0.040
#> GSM96961 1 0.0661 0.8289 0.988 0.004 0.008
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM97038 2 0.1302 0.9104 0.000 0.956 0.000 0.044
#> GSM97045 2 0.1305 0.9240 0.000 0.960 0.036 0.004
#> GSM97047 1 0.7987 -0.1569 0.412 0.392 0.180 0.016
#> GSM97025 2 0.1305 0.9240 0.000 0.960 0.036 0.004
#> GSM97030 3 0.2803 0.8696 0.008 0.080 0.900 0.012
#> GSM97027 2 0.1118 0.9239 0.000 0.964 0.036 0.000
#> GSM97033 2 0.1302 0.9104 0.000 0.956 0.000 0.044
#> GSM97034 3 0.2706 0.8801 0.024 0.064 0.908 0.004
#> GSM97020 2 0.1302 0.9104 0.000 0.956 0.000 0.044
#> GSM97026 2 0.6549 0.5081 0.308 0.604 0.080 0.008
#> GSM97012 2 0.1398 0.9236 0.000 0.956 0.040 0.004
#> GSM97015 3 0.3026 0.8784 0.032 0.056 0.900 0.012
#> GSM97016 2 0.1302 0.9104 0.000 0.956 0.000 0.044
#> GSM97017 1 0.0188 0.7529 0.996 0.000 0.004 0.000
#> GSM97019 2 0.1398 0.9236 0.000 0.956 0.040 0.004
#> GSM97022 2 0.1398 0.9236 0.000 0.956 0.040 0.004
#> GSM97035 2 0.1398 0.9236 0.000 0.956 0.040 0.004
#> GSM97036 1 0.1296 0.7530 0.964 0.004 0.004 0.028
#> GSM97039 2 0.1302 0.9104 0.000 0.956 0.000 0.044
#> GSM97046 2 0.1302 0.9104 0.000 0.956 0.000 0.044
#> GSM97023 1 0.3306 0.6940 0.840 0.000 0.004 0.156
#> GSM97029 1 0.0779 0.7496 0.980 0.004 0.016 0.000
#> GSM97043 2 0.1994 0.9122 0.008 0.936 0.052 0.004
#> GSM97013 1 0.0469 0.7536 0.988 0.000 0.000 0.012
#> GSM96956 2 0.4638 0.7395 0.000 0.776 0.180 0.044
#> GSM97024 2 0.1545 0.9225 0.000 0.952 0.040 0.008
#> GSM97032 3 0.2803 0.8714 0.012 0.080 0.900 0.008
#> GSM97044 3 0.2207 0.8859 0.012 0.056 0.928 0.004
#> GSM97049 2 0.1302 0.9104 0.000 0.956 0.000 0.044
#> GSM96968 3 0.2125 0.8792 0.076 0.004 0.920 0.000
#> GSM96971 3 0.4220 0.6736 0.004 0.000 0.748 0.248
#> GSM96986 3 0.1888 0.8902 0.016 0.000 0.940 0.044
#> GSM97003 1 0.5837 0.4104 0.564 0.000 0.036 0.400
#> GSM96957 1 0.0937 0.7523 0.976 0.000 0.012 0.012
#> GSM96960 1 0.5594 0.2689 0.520 0.000 0.020 0.460
#> GSM96975 1 0.0804 0.7540 0.980 0.000 0.012 0.008
#> GSM96998 1 0.5269 0.4812 0.620 0.000 0.016 0.364
#> GSM96999 1 0.0779 0.7537 0.980 0.000 0.016 0.004
#> GSM97001 1 0.0804 0.7528 0.980 0.000 0.012 0.008
#> GSM97005 1 0.0927 0.7530 0.976 0.000 0.016 0.008
#> GSM97006 1 0.5517 0.3865 0.568 0.000 0.020 0.412
#> GSM97021 1 0.1256 0.7425 0.964 0.000 0.028 0.008
#> GSM97028 3 0.1958 0.8972 0.028 0.008 0.944 0.020
#> GSM97031 1 0.2214 0.7490 0.928 0.000 0.028 0.044
#> GSM97037 2 0.6087 0.4335 0.004 0.596 0.352 0.048
#> GSM97018 3 0.3144 0.8755 0.020 0.072 0.892 0.016
#> GSM97014 1 0.6466 -0.0945 0.496 0.452 0.024 0.028
#> GSM97042 2 0.1398 0.9236 0.000 0.956 0.040 0.004
#> GSM97040 1 0.3882 0.6502 0.848 0.028 0.112 0.012
#> GSM97041 1 0.0336 0.7524 0.992 0.000 0.008 0.000
#> GSM96955 2 0.4534 0.8403 0.072 0.832 0.064 0.032
#> GSM96990 3 0.2718 0.8827 0.020 0.056 0.912 0.012
#> GSM96991 2 0.1545 0.9227 0.000 0.952 0.040 0.008
#> GSM97048 2 0.1302 0.9104 0.000 0.956 0.000 0.044
#> GSM96963 2 0.1356 0.9233 0.000 0.960 0.032 0.008
#> GSM96953 2 0.1209 0.9237 0.000 0.964 0.032 0.004
#> GSM96966 4 0.2813 0.8193 0.080 0.000 0.024 0.896
#> GSM96979 3 0.1888 0.8902 0.016 0.000 0.940 0.044
#> GSM96983 3 0.1396 0.8969 0.004 0.004 0.960 0.032
#> GSM96984 3 0.1888 0.8902 0.016 0.000 0.940 0.044
#> GSM96994 3 0.1109 0.8976 0.004 0.000 0.968 0.028
#> GSM96996 1 0.5414 0.4574 0.604 0.000 0.020 0.376
#> GSM96997 3 0.1888 0.8902 0.016 0.000 0.940 0.044
#> GSM97007 3 0.1109 0.8976 0.004 0.000 0.968 0.028
#> GSM96954 3 0.2197 0.8751 0.080 0.000 0.916 0.004
#> GSM96962 3 0.1888 0.8902 0.016 0.000 0.940 0.044
#> GSM96969 4 0.2813 0.8193 0.080 0.000 0.024 0.896
#> GSM96970 4 0.2813 0.8193 0.080 0.000 0.024 0.896
#> GSM96973 4 0.2670 0.8171 0.072 0.000 0.024 0.904
#> GSM96976 4 0.4134 0.5915 0.000 0.000 0.260 0.740
#> GSM96977 1 0.2944 0.6794 0.868 0.000 0.128 0.004
#> GSM96995 3 0.2660 0.8799 0.072 0.008 0.908 0.012
#> GSM97002 1 0.5590 0.2797 0.524 0.000 0.020 0.456
#> GSM97009 1 0.6926 0.4133 0.636 0.216 0.128 0.020
#> GSM97010 1 0.2926 0.7347 0.896 0.000 0.048 0.056
#> GSM96974 4 0.4008 0.6160 0.000 0.000 0.244 0.756
#> GSM96985 4 0.4741 0.5093 0.004 0.000 0.328 0.668
#> GSM96959 3 0.6042 0.4931 0.348 0.024 0.608 0.020
#> GSM96972 4 0.2593 0.8107 0.080 0.000 0.016 0.904
#> GSM96978 3 0.2156 0.8899 0.008 0.004 0.928 0.060
#> GSM96967 4 0.2813 0.8193 0.080 0.000 0.024 0.896
#> GSM96987 1 0.4855 0.5080 0.644 0.000 0.004 0.352
#> GSM97011 1 0.1284 0.7447 0.964 0.000 0.024 0.012
#> GSM96964 1 0.2654 0.7241 0.888 0.000 0.004 0.108
#> GSM96965 4 0.4426 0.7013 0.204 0.000 0.024 0.772
#> GSM96981 1 0.4621 0.5610 0.708 0.000 0.008 0.284
#> GSM96982 1 0.5277 0.2997 0.532 0.000 0.008 0.460
#> GSM96988 3 0.1545 0.8961 0.008 0.000 0.952 0.040
#> GSM97000 1 0.3161 0.6669 0.864 0.000 0.124 0.012
#> GSM97004 4 0.5511 -0.2391 0.484 0.000 0.016 0.500
#> GSM97008 1 0.1488 0.7405 0.956 0.000 0.032 0.012
#> GSM96950 1 0.0592 0.7533 0.984 0.000 0.000 0.016
#> GSM96980 4 0.2149 0.7949 0.088 0.000 0.000 0.912
#> GSM96989 1 0.4837 0.5128 0.648 0.000 0.004 0.348
#> GSM96992 1 0.5143 0.4904 0.628 0.000 0.012 0.360
#> GSM96993 1 0.1059 0.7535 0.972 0.000 0.012 0.016
#> GSM96958 1 0.0927 0.7538 0.976 0.000 0.008 0.016
#> GSM96951 1 0.2142 0.7434 0.928 0.000 0.016 0.056
#> GSM96952 1 0.5040 0.4883 0.628 0.000 0.008 0.364
#> GSM96961 1 0.3852 0.6752 0.808 0.000 0.012 0.180
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM97038 2 0.3305 0.8313 0.000 0.776 0.000 0.224 0.000
#> GSM97045 2 0.0579 0.8708 0.000 0.984 0.008 0.008 0.000
#> GSM97047 5 0.5824 0.4855 0.000 0.104 0.248 0.016 0.632
#> GSM97025 2 0.0579 0.8708 0.000 0.984 0.008 0.008 0.000
#> GSM97030 3 0.2127 0.6989 0.000 0.108 0.892 0.000 0.000
#> GSM97027 2 0.0579 0.8708 0.000 0.984 0.008 0.008 0.000
#> GSM97033 2 0.3305 0.8313 0.000 0.776 0.000 0.224 0.000
#> GSM97034 3 0.2568 0.7018 0.000 0.092 0.888 0.004 0.016
#> GSM97020 2 0.3305 0.8313 0.000 0.776 0.000 0.224 0.000
#> GSM97026 5 0.7079 0.1530 0.000 0.316 0.276 0.012 0.396
#> GSM97012 2 0.0579 0.8695 0.000 0.984 0.008 0.008 0.000
#> GSM97015 3 0.2570 0.7046 0.000 0.084 0.888 0.000 0.028
#> GSM97016 2 0.3305 0.8313 0.000 0.776 0.000 0.224 0.000
#> GSM97017 5 0.0932 0.7560 0.020 0.000 0.004 0.004 0.972
#> GSM97019 2 0.0579 0.8695 0.000 0.984 0.008 0.008 0.000
#> GSM97022 2 0.0579 0.8695 0.000 0.984 0.008 0.008 0.000
#> GSM97035 2 0.0579 0.8695 0.000 0.984 0.008 0.008 0.000
#> GSM97036 5 0.4108 0.6597 0.188 0.004 0.024 0.008 0.776
#> GSM97039 2 0.3305 0.8313 0.000 0.776 0.000 0.224 0.000
#> GSM97046 2 0.3305 0.8313 0.000 0.776 0.000 0.224 0.000
#> GSM97023 5 0.4151 0.3726 0.344 0.000 0.000 0.004 0.652
#> GSM97029 5 0.1518 0.7521 0.048 0.000 0.004 0.004 0.944
#> GSM97043 2 0.3422 0.7030 0.000 0.792 0.200 0.004 0.004
#> GSM97013 5 0.2228 0.7334 0.092 0.000 0.004 0.004 0.900
#> GSM96956 2 0.6285 0.5637 0.000 0.536 0.244 0.220 0.000
#> GSM97024 2 0.1341 0.8442 0.000 0.944 0.056 0.000 0.000
#> GSM97032 3 0.2304 0.7009 0.000 0.100 0.892 0.000 0.008
#> GSM97044 3 0.2069 0.7159 0.000 0.076 0.912 0.012 0.000
#> GSM97049 2 0.3305 0.8313 0.000 0.776 0.000 0.224 0.000
#> GSM96968 3 0.2989 0.7146 0.000 0.000 0.868 0.072 0.060
#> GSM96971 4 0.5651 -0.2006 0.044 0.000 0.428 0.512 0.016
#> GSM96986 3 0.5033 0.6247 0.016 0.000 0.660 0.292 0.032
#> GSM97003 1 0.6228 0.4622 0.592 0.000 0.012 0.176 0.220
#> GSM96957 5 0.1430 0.7534 0.052 0.000 0.004 0.000 0.944
#> GSM96960 1 0.4438 0.5725 0.732 0.000 0.004 0.040 0.224
#> GSM96975 5 0.3805 0.6456 0.192 0.000 0.016 0.008 0.784
#> GSM96998 1 0.4130 0.5375 0.696 0.000 0.000 0.012 0.292
#> GSM96999 5 0.1430 0.7534 0.052 0.000 0.004 0.000 0.944
#> GSM97001 5 0.0703 0.7561 0.024 0.000 0.000 0.000 0.976
#> GSM97005 5 0.0693 0.7557 0.012 0.000 0.000 0.008 0.980
#> GSM97006 1 0.4552 0.5664 0.716 0.000 0.004 0.040 0.240
#> GSM97021 5 0.1557 0.7469 0.000 0.000 0.052 0.008 0.940
#> GSM97028 3 0.1960 0.7006 0.000 0.020 0.928 0.048 0.004
#> GSM97031 5 0.3021 0.7199 0.064 0.000 0.004 0.060 0.872
#> GSM97037 3 0.6477 -0.0911 0.000 0.352 0.456 0.192 0.000
#> GSM97018 3 0.2734 0.6986 0.000 0.076 0.888 0.028 0.008
#> GSM97014 5 0.4759 0.6434 0.000 0.088 0.072 0.060 0.780
#> GSM97042 2 0.0579 0.8695 0.000 0.984 0.008 0.008 0.000
#> GSM97040 5 0.3622 0.6642 0.000 0.016 0.172 0.008 0.804
#> GSM97041 5 0.0932 0.7560 0.020 0.000 0.004 0.004 0.972
#> GSM96955 2 0.5964 0.6492 0.004 0.692 0.144 0.076 0.084
#> GSM96990 3 0.2450 0.7086 0.000 0.076 0.896 0.000 0.028
#> GSM96991 2 0.1012 0.8646 0.000 0.968 0.020 0.012 0.000
#> GSM97048 2 0.3305 0.8313 0.000 0.776 0.000 0.224 0.000
#> GSM96963 2 0.1216 0.8655 0.000 0.960 0.020 0.020 0.000
#> GSM96953 2 0.0992 0.8694 0.000 0.968 0.008 0.024 0.000
#> GSM96966 1 0.4464 -0.2072 0.584 0.000 0.000 0.408 0.008
#> GSM96979 3 0.5102 0.6230 0.020 0.000 0.660 0.288 0.032
#> GSM96983 3 0.2497 0.7017 0.000 0.004 0.880 0.112 0.004
#> GSM96984 3 0.5012 0.6263 0.016 0.000 0.664 0.288 0.032
#> GSM96994 3 0.4673 0.6251 0.012 0.000 0.680 0.288 0.020
#> GSM96996 1 0.4016 0.5503 0.716 0.000 0.000 0.012 0.272
#> GSM96997 3 0.5226 0.6116 0.024 0.000 0.648 0.296 0.032
#> GSM97007 3 0.4935 0.6277 0.016 0.000 0.668 0.288 0.028
#> GSM96954 3 0.5139 0.6364 0.004 0.000 0.680 0.236 0.080
#> GSM96962 3 0.5012 0.6263 0.016 0.000 0.664 0.288 0.032
#> GSM96969 1 0.4350 -0.2156 0.588 0.000 0.000 0.408 0.004
#> GSM96970 1 0.4350 -0.2156 0.588 0.000 0.000 0.408 0.004
#> GSM96973 1 0.4499 -0.2235 0.584 0.000 0.004 0.408 0.004
#> GSM96976 4 0.6195 0.5502 0.360 0.000 0.128 0.508 0.004
#> GSM96977 5 0.4107 0.7201 0.072 0.000 0.120 0.008 0.800
#> GSM96995 3 0.1704 0.7042 0.000 0.000 0.928 0.004 0.068
#> GSM97002 1 0.3934 0.5708 0.740 0.000 0.000 0.016 0.244
#> GSM97009 5 0.4458 0.6669 0.008 0.044 0.132 0.024 0.792
#> GSM97010 5 0.4037 0.6921 0.176 0.000 0.028 0.012 0.784
#> GSM96974 4 0.6281 0.5245 0.388 0.000 0.152 0.460 0.000
#> GSM96985 1 0.6477 -0.3816 0.464 0.000 0.340 0.196 0.000
#> GSM96959 5 0.5198 0.3499 0.000 0.020 0.372 0.020 0.588
#> GSM96972 1 0.4359 -0.2183 0.584 0.000 0.000 0.412 0.004
#> GSM96978 3 0.2911 0.6935 0.004 0.000 0.852 0.136 0.008
#> GSM96967 1 0.4350 -0.2156 0.588 0.000 0.000 0.408 0.004
#> GSM96987 1 0.4009 0.5226 0.684 0.000 0.000 0.004 0.312
#> GSM97011 5 0.1949 0.7490 0.012 0.000 0.040 0.016 0.932
#> GSM96964 5 0.4430 0.0800 0.456 0.000 0.000 0.004 0.540
#> GSM96965 1 0.6100 -0.3098 0.472 0.000 0.004 0.416 0.108
#> GSM96981 1 0.5020 0.2969 0.564 0.000 0.012 0.016 0.408
#> GSM96982 1 0.4684 0.5686 0.720 0.000 0.028 0.020 0.232
#> GSM96988 3 0.2228 0.7029 0.004 0.000 0.900 0.092 0.004
#> GSM97000 5 0.2642 0.7264 0.004 0.000 0.084 0.024 0.888
#> GSM97004 1 0.3663 0.5800 0.776 0.000 0.000 0.016 0.208
#> GSM97008 5 0.1596 0.7533 0.012 0.000 0.028 0.012 0.948
#> GSM96950 5 0.2964 0.6929 0.152 0.000 0.004 0.004 0.840
#> GSM96980 1 0.2462 0.2622 0.880 0.000 0.000 0.112 0.008
#> GSM96989 1 0.4029 0.5178 0.680 0.000 0.000 0.004 0.316
#> GSM96992 1 0.4059 0.5354 0.700 0.000 0.004 0.004 0.292
#> GSM96993 5 0.3691 0.6856 0.164 0.000 0.028 0.004 0.804
#> GSM96958 5 0.3969 0.4891 0.304 0.000 0.004 0.000 0.692
#> GSM96951 5 0.4426 0.3199 0.380 0.000 0.004 0.004 0.612
#> GSM96952 1 0.3906 0.5327 0.704 0.000 0.004 0.000 0.292
#> GSM96961 1 0.4545 0.2076 0.560 0.000 0.004 0.004 0.432
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM97038 2 0.5438 0.7301 0.008 0.636 0.004 0.088 0.016 0.248
#> GSM97045 2 0.0363 0.7987 0.000 0.988 0.012 0.000 0.000 0.000
#> GSM97047 5 0.3505 0.6496 0.000 0.048 0.136 0.008 0.808 0.000
#> GSM97025 2 0.0363 0.7987 0.000 0.988 0.012 0.000 0.000 0.000
#> GSM97030 3 0.3521 0.6021 0.000 0.120 0.812 0.000 0.060 0.008
#> GSM97027 2 0.0363 0.7987 0.000 0.988 0.012 0.000 0.000 0.000
#> GSM97033 2 0.5281 0.7332 0.008 0.644 0.000 0.088 0.016 0.244
#> GSM97034 3 0.3618 0.6150 0.000 0.104 0.808 0.008 0.080 0.000
#> GSM97020 2 0.5281 0.7332 0.008 0.644 0.000 0.088 0.016 0.244
#> GSM97026 5 0.7378 0.0255 0.008 0.272 0.272 0.008 0.380 0.060
#> GSM97012 2 0.0146 0.8004 0.000 0.996 0.000 0.004 0.000 0.000
#> GSM97015 3 0.3325 0.6111 0.000 0.084 0.820 0.000 0.096 0.000
#> GSM97016 2 0.5303 0.7322 0.008 0.640 0.000 0.088 0.016 0.248
#> GSM97017 5 0.3053 0.7487 0.100 0.000 0.008 0.004 0.852 0.036
#> GSM97019 2 0.0363 0.7987 0.000 0.988 0.012 0.000 0.000 0.000
#> GSM97022 2 0.0363 0.7987 0.000 0.988 0.012 0.000 0.000 0.000
#> GSM97035 2 0.0000 0.8005 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97036 5 0.5988 0.3581 0.368 0.000 0.024 0.008 0.500 0.100
#> GSM97039 2 0.5303 0.7322 0.008 0.640 0.000 0.088 0.016 0.248
#> GSM97046 2 0.5323 0.7307 0.008 0.636 0.000 0.088 0.016 0.252
#> GSM97023 1 0.4865 0.4565 0.632 0.000 0.000 0.004 0.284 0.080
#> GSM97029 5 0.3892 0.7363 0.116 0.000 0.008 0.004 0.792 0.080
#> GSM97043 2 0.3422 0.5980 0.000 0.788 0.176 0.000 0.036 0.000
#> GSM97013 5 0.5099 0.6324 0.232 0.000 0.000 0.012 0.648 0.108
#> GSM96956 2 0.7814 0.3399 0.008 0.360 0.272 0.088 0.020 0.252
#> GSM97024 2 0.1152 0.7767 0.000 0.952 0.044 0.000 0.004 0.000
#> GSM97032 3 0.3285 0.6084 0.000 0.116 0.820 0.000 0.064 0.000
#> GSM97044 3 0.3819 0.5796 0.000 0.084 0.812 0.000 0.048 0.056
#> GSM97049 2 0.5303 0.7322 0.008 0.640 0.000 0.088 0.016 0.248
#> GSM96968 3 0.3992 0.4883 0.008 0.000 0.788 0.008 0.120 0.076
#> GSM96971 6 0.6252 0.3247 0.004 0.000 0.236 0.368 0.004 0.388
#> GSM96986 6 0.4644 0.8505 0.012 0.000 0.456 0.000 0.020 0.512
#> GSM97003 1 0.4666 0.6006 0.688 0.000 0.012 0.020 0.028 0.252
#> GSM96957 5 0.3821 0.7229 0.156 0.000 0.000 0.004 0.776 0.064
#> GSM96960 1 0.2112 0.7806 0.916 0.000 0.000 0.036 0.020 0.028
#> GSM96975 5 0.5230 0.4959 0.312 0.000 0.020 0.004 0.604 0.060
#> GSM96998 1 0.2803 0.7970 0.872 0.000 0.000 0.012 0.052 0.064
#> GSM96999 5 0.3959 0.7126 0.172 0.000 0.000 0.004 0.760 0.064
#> GSM97001 5 0.2333 0.7454 0.120 0.000 0.000 0.004 0.872 0.004
#> GSM97005 5 0.2275 0.7528 0.096 0.000 0.000 0.008 0.888 0.008
#> GSM97006 1 0.1851 0.7815 0.928 0.000 0.000 0.036 0.012 0.024
#> GSM97021 5 0.2170 0.7574 0.044 0.000 0.016 0.008 0.916 0.016
#> GSM97028 3 0.3540 0.5481 0.004 0.020 0.848 0.040 0.024 0.064
#> GSM97031 5 0.4394 0.6762 0.148 0.000 0.000 0.008 0.736 0.108
#> GSM97037 3 0.7637 0.0402 0.004 0.244 0.420 0.056 0.044 0.232
#> GSM97018 3 0.4266 0.6162 0.000 0.080 0.796 0.020 0.068 0.036
#> GSM97014 5 0.1743 0.7452 0.000 0.004 0.028 0.008 0.936 0.024
#> GSM97042 2 0.0000 0.8005 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97040 5 0.2312 0.7336 0.012 0.000 0.080 0.008 0.896 0.004
#> GSM97041 5 0.3170 0.7482 0.104 0.000 0.008 0.004 0.844 0.040
#> GSM96955 2 0.7340 0.3041 0.004 0.504 0.232 0.036 0.124 0.100
#> GSM96990 3 0.3112 0.6128 0.000 0.068 0.836 0.000 0.096 0.000
#> GSM96991 2 0.1053 0.7908 0.000 0.964 0.020 0.004 0.000 0.012
#> GSM97048 2 0.5303 0.7322 0.008 0.640 0.000 0.088 0.016 0.248
#> GSM96963 2 0.1237 0.7915 0.000 0.956 0.020 0.004 0.000 0.020
#> GSM96953 2 0.0405 0.8002 0.000 0.988 0.000 0.004 0.000 0.008
#> GSM96966 4 0.2378 0.9252 0.152 0.000 0.000 0.848 0.000 0.000
#> GSM96979 6 0.4540 0.8717 0.012 0.000 0.472 0.004 0.008 0.504
#> GSM96983 3 0.3685 0.4461 0.004 0.008 0.812 0.040 0.008 0.128
#> GSM96984 6 0.4541 0.8703 0.012 0.000 0.476 0.004 0.008 0.500
#> GSM96994 6 0.4639 0.8694 0.012 0.000 0.472 0.008 0.008 0.500
#> GSM96996 1 0.2434 0.7992 0.896 0.000 0.000 0.016 0.056 0.032
#> GSM96997 6 0.4624 0.8491 0.024 0.000 0.452 0.000 0.008 0.516
#> GSM97007 6 0.4541 0.8703 0.012 0.000 0.476 0.004 0.008 0.500
#> GSM96954 3 0.5748 -0.4116 0.004 0.000 0.512 0.008 0.124 0.352
#> GSM96962 6 0.4540 0.8717 0.012 0.000 0.472 0.004 0.008 0.504
#> GSM96969 4 0.2378 0.9252 0.152 0.000 0.000 0.848 0.000 0.000
#> GSM96970 4 0.2378 0.9252 0.152 0.000 0.000 0.848 0.000 0.000
#> GSM96973 4 0.2378 0.9252 0.152 0.000 0.000 0.848 0.000 0.000
#> GSM96976 4 0.3038 0.8074 0.012 0.000 0.060 0.856 0.000 0.072
#> GSM96977 5 0.5459 0.6819 0.076 0.000 0.128 0.012 0.696 0.088
#> GSM96995 3 0.2982 0.5438 0.000 0.000 0.820 0.004 0.164 0.012
#> GSM97002 1 0.2024 0.7902 0.920 0.000 0.000 0.036 0.028 0.016
#> GSM97009 5 0.2423 0.7346 0.004 0.016 0.064 0.004 0.900 0.012
#> GSM97010 5 0.5455 0.6398 0.216 0.000 0.024 0.012 0.652 0.096
#> GSM96974 4 0.2858 0.8122 0.016 0.000 0.092 0.864 0.000 0.028
#> GSM96985 3 0.7536 0.0926 0.204 0.004 0.440 0.188 0.008 0.156
#> GSM96959 5 0.3627 0.5988 0.004 0.000 0.200 0.012 0.772 0.012
#> GSM96972 4 0.2814 0.9041 0.172 0.000 0.000 0.820 0.000 0.008
#> GSM96978 3 0.4060 0.4118 0.004 0.008 0.772 0.040 0.008 0.168
#> GSM96967 4 0.2378 0.9252 0.152 0.000 0.000 0.848 0.000 0.000
#> GSM96987 1 0.3204 0.7684 0.836 0.000 0.000 0.004 0.068 0.092
#> GSM97011 5 0.1484 0.7587 0.040 0.000 0.008 0.004 0.944 0.004
#> GSM96964 1 0.4633 0.6163 0.704 0.000 0.000 0.008 0.188 0.100
#> GSM96965 4 0.3516 0.8411 0.072 0.000 0.008 0.832 0.076 0.012
#> GSM96981 1 0.5007 0.5461 0.660 0.000 0.024 0.012 0.264 0.040
#> GSM96982 1 0.4273 0.7156 0.800 0.000 0.076 0.028 0.052 0.044
#> GSM96988 3 0.3833 0.4640 0.004 0.008 0.808 0.040 0.016 0.124
#> GSM97000 5 0.1819 0.7483 0.024 0.000 0.032 0.004 0.932 0.008
#> GSM97004 1 0.1850 0.7779 0.924 0.000 0.000 0.052 0.008 0.016
#> GSM97008 5 0.1872 0.7549 0.064 0.000 0.004 0.004 0.920 0.008
#> GSM96950 5 0.5378 0.5254 0.304 0.000 0.000 0.008 0.576 0.112
#> GSM96980 1 0.4083 0.3490 0.668 0.000 0.000 0.304 0.000 0.028
#> GSM96989 1 0.3260 0.7666 0.832 0.000 0.000 0.004 0.072 0.092
#> GSM96992 1 0.1942 0.8023 0.916 0.000 0.000 0.012 0.064 0.008
#> GSM96993 5 0.6054 0.3941 0.348 0.000 0.024 0.008 0.508 0.112
#> GSM96958 5 0.5302 0.1667 0.448 0.000 0.000 0.012 0.472 0.068
#> GSM96951 1 0.4728 0.3724 0.616 0.000 0.000 0.004 0.324 0.056
#> GSM96952 1 0.1728 0.8014 0.924 0.000 0.000 0.004 0.064 0.008
#> GSM96961 1 0.2752 0.7697 0.856 0.000 0.000 0.000 0.108 0.036
Heatmaps for the consensus matrix. It visualizes the probability of two samples to be in a same group.
consensus_heatmap(res, k = 2)
consensus_heatmap(res, k = 3)
consensus_heatmap(res, k = 4)
consensus_heatmap(res, k = 5)
consensus_heatmap(res, k = 6)
Heatmaps for the membership of samples in all partitions to see how consistent they are:
membership_heatmap(res, k = 2)
membership_heatmap(res, k = 3)
membership_heatmap(res, k = 4)
membership_heatmap(res, k = 5)
membership_heatmap(res, k = 6)
As soon as we have had the classes for columns, we can look for signatures which are significantly different between classes which can be candidate marks for certain classes. Following are the heatmaps for signatures.
Signature heatmaps where rows are scaled:
get_signatures(res, k = 2)
get_signatures(res, k = 3)
get_signatures(res, k = 4)
get_signatures(res, k = 5)
get_signatures(res, k = 6)
Signature heatmaps where rows are not scaled:
get_signatures(res, k = 2, scale_rows = FALSE)
get_signatures(res, k = 3, scale_rows = FALSE)
get_signatures(res, k = 4, scale_rows = FALSE)
get_signatures(res, k = 5, scale_rows = FALSE)
get_signatures(res, k = 6, scale_rows = FALSE)
Compare the overlap of signatures from different k:
compare_signatures(res)
get_signature() returns a data frame invisibly. TO get the list of signatures, the function
call should be assigned to a variable explicitly. In following code, if plot argument is set
to FALSE, no heatmap is plotted while only the differential analysis is performed.
# code only for demonstration
tb = get_signature(res, k = ..., plot = FALSE)
An example of the output of tb is:
#> which_row fdr mean_1 mean_2 scaled_mean_1 scaled_mean_2 km
#> 1 38 0.042760348 8.373488 9.131774 -0.5533452 0.5164555 1
#> 2 40 0.018707592 7.106213 8.469186 -0.6173731 0.5762149 1
#> 3 55 0.019134737 10.221463 11.207825 -0.6159697 0.5749050 1
#> 4 59 0.006059896 5.921854 7.869574 -0.6899429 0.6439467 1
#> 5 60 0.018055526 8.928898 10.211722 -0.6204761 0.5791110 1
#> 6 98 0.009384629 15.714769 14.887706 0.6635654 -0.6193277 2
...
The columns in tb are:
which_row: row indices corresponding to the input matrix.fdr: FDR for the differential test. mean_x: The mean value in group x.scaled_mean_x: The mean value in group x after rows are scaled.km: Row groups if k-means clustering is applied to rows.UMAP plot which shows how samples are separated.
dimension_reduction(res, k = 2, method = "UMAP")
dimension_reduction(res, k = 3, method = "UMAP")
dimension_reduction(res, k = 4, method = "UMAP")
dimension_reduction(res, k = 5, method = "UMAP")
dimension_reduction(res, k = 6, method = "UMAP")
Following heatmap shows how subgroups are split when increasing k:
collect_classes(res)
Test correlation between subgroups and known annotations. If the known annotation is numeric, one-way ANOVA test is applied, and if the known annotation is discrete, chi-squared contingency table test is applied.
test_to_known_factors(res)
#> n disease.state(p) specimen(p) cell.type(p) other(p) k
#> CV:kmeans 100 1.13e-05 0.298 4.23e-14 0.1231 2
#> CV:kmeans 44 NA NA NA NA 3
#> CV:kmeans 85 2.83e-04 0.141 1.07e-17 0.0345 4
#> CV:kmeans 79 5.99e-03 0.390 3.70e-13 0.1147 5
#> CV:kmeans 82 4.78e-05 0.206 1.42e-15 0.0045 6
If matrix rows can be associated to genes, consider to use functional_enrichment(res,
...) to perform function enrichment for the signature genes. See this vignette for more detailed explanations.
The object with results only for a single top-value method and a single partition method can be extracted as:
res = res_list["CV", "skmeans"]
# you can also extract it by
# res = res_list["CV:skmeans"]
A summary of res and all the functions that can be applied to it:
res
#> A 'ConsensusPartition' object with k = 2, 3, 4, 5, 6.
#> On a matrix with 21168 rows and 100 columns.
#> Top rows (1000, 2000, 3000, 4000, 5000) are extracted by 'CV' method.
#> Subgroups are detected by 'skmeans' method.
#> Performed in total 1250 partitions by row resampling.
#> Best k for subgroups seems to be 2.
#>
#> Following methods can be applied to this 'ConsensusPartition' object:
#> [1] "cola_report" "collect_classes" "collect_plots"
#> [4] "collect_stats" "colnames" "compare_signatures"
#> [7] "consensus_heatmap" "dimension_reduction" "functional_enrichment"
#> [10] "get_anno_col" "get_anno" "get_classes"
#> [13] "get_consensus" "get_matrix" "get_membership"
#> [16] "get_param" "get_signatures" "get_stats"
#> [19] "is_best_k" "is_stable_k" "membership_heatmap"
#> [22] "ncol" "nrow" "plot_ecdf"
#> [25] "rownames" "select_partition_number" "show"
#> [28] "suggest_best_k" "test_to_known_factors"
collect_plots() function collects all the plots made from res for all k (number of partitions)
into one single page to provide an easy and fast comparison between different k.
collect_plots(res)
The plots are:
k and the heatmap of
predicted classes for each k.k.k.k.All the plots in panels can be made by individual functions and they are plotted later in this section.
select_partition_number() produces several plots showing different
statistics for choosing “optimized” k. There are following statistics:
k;k, the area increased is defined as \(A_k - A_{k-1}\).The detailed explanations of these statistics can be found in the cola vignette.
Generally speaking, lower PAC score, higher mean silhouette score or higher
concordance corresponds to better partition. Rand index and Jaccard index
measure how similar the current partition is compared to partition with k-1.
If they are too similar, we won't accept k is better than k-1.
select_partition_number(res)
The numeric values for all these statistics can be obtained by get_stats().
get_stats(res)
#> k 1-PAC mean_silhouette concordance area_increased Rand Jaccard
#> 2 2 1.000 0.983 0.992 0.5002 0.500 0.500
#> 3 3 0.671 0.414 0.731 0.3219 0.754 0.543
#> 4 4 0.681 0.740 0.870 0.1324 0.806 0.495
#> 5 5 0.643 0.589 0.724 0.0591 0.904 0.647
#> 6 6 0.640 0.493 0.704 0.0424 0.940 0.726
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
#> GSM97038 2 0.0000 0.991 0.000 1.000
#> GSM97045 2 0.0000 0.991 0.000 1.000
#> GSM97047 2 0.0000 0.991 0.000 1.000
#> GSM97025 2 0.0000 0.991 0.000 1.000
#> GSM97030 2 0.0000 0.991 0.000 1.000
#> GSM97027 2 0.0000 0.991 0.000 1.000
#> GSM97033 2 0.0000 0.991 0.000 1.000
#> GSM97034 2 0.0000 0.991 0.000 1.000
#> GSM97020 2 0.0000 0.991 0.000 1.000
#> GSM97026 2 0.0000 0.991 0.000 1.000
#> GSM97012 2 0.0000 0.991 0.000 1.000
#> GSM97015 2 0.0000 0.991 0.000 1.000
#> GSM97016 2 0.0000 0.991 0.000 1.000
#> GSM97017 1 0.0672 0.988 0.992 0.008
#> GSM97019 2 0.0000 0.991 0.000 1.000
#> GSM97022 2 0.0000 0.991 0.000 1.000
#> GSM97035 2 0.0000 0.991 0.000 1.000
#> GSM97036 1 0.1633 0.974 0.976 0.024
#> GSM97039 2 0.0000 0.991 0.000 1.000
#> GSM97046 2 0.0000 0.991 0.000 1.000
#> GSM97023 1 0.0000 0.993 1.000 0.000
#> GSM97029 1 0.0938 0.984 0.988 0.012
#> GSM97043 2 0.0000 0.991 0.000 1.000
#> GSM97013 1 0.0000 0.993 1.000 0.000
#> GSM96956 2 0.0000 0.991 0.000 1.000
#> GSM97024 2 0.0000 0.991 0.000 1.000
#> GSM97032 2 0.0000 0.991 0.000 1.000
#> GSM97044 2 0.0000 0.991 0.000 1.000
#> GSM97049 2 0.0000 0.991 0.000 1.000
#> GSM96968 1 0.7299 0.744 0.796 0.204
#> GSM96971 1 0.0000 0.993 1.000 0.000
#> GSM96986 1 0.0000 0.993 1.000 0.000
#> GSM97003 1 0.0000 0.993 1.000 0.000
#> GSM96957 1 0.0000 0.993 1.000 0.000
#> GSM96960 1 0.0000 0.993 1.000 0.000
#> GSM96975 1 0.0000 0.993 1.000 0.000
#> GSM96998 1 0.0000 0.993 1.000 0.000
#> GSM96999 1 0.0000 0.993 1.000 0.000
#> GSM97001 1 0.0000 0.993 1.000 0.000
#> GSM97005 1 0.0000 0.993 1.000 0.000
#> GSM97006 1 0.0000 0.993 1.000 0.000
#> GSM97021 1 0.1184 0.981 0.984 0.016
#> GSM97028 2 0.0000 0.991 0.000 1.000
#> GSM97031 1 0.0000 0.993 1.000 0.000
#> GSM97037 2 0.0000 0.991 0.000 1.000
#> GSM97018 2 0.0000 0.991 0.000 1.000
#> GSM97014 2 0.0000 0.991 0.000 1.000
#> GSM97042 2 0.0000 0.991 0.000 1.000
#> GSM97040 2 0.0000 0.991 0.000 1.000
#> GSM97041 1 0.0672 0.988 0.992 0.008
#> GSM96955 2 0.0000 0.991 0.000 1.000
#> GSM96990 2 0.0000 0.991 0.000 1.000
#> GSM96991 2 0.0000 0.991 0.000 1.000
#> GSM97048 2 0.0000 0.991 0.000 1.000
#> GSM96963 2 0.0000 0.991 0.000 1.000
#> GSM96953 2 0.0000 0.991 0.000 1.000
#> GSM96966 1 0.0000 0.993 1.000 0.000
#> GSM96979 1 0.0000 0.993 1.000 0.000
#> GSM96983 2 0.0000 0.991 0.000 1.000
#> GSM96984 2 0.7674 0.718 0.224 0.776
#> GSM96994 2 0.0000 0.991 0.000 1.000
#> GSM96996 1 0.0000 0.993 1.000 0.000
#> GSM96997 1 0.0000 0.993 1.000 0.000
#> GSM97007 2 0.0000 0.991 0.000 1.000
#> GSM96954 1 0.0000 0.993 1.000 0.000
#> GSM96962 1 0.0000 0.993 1.000 0.000
#> GSM96969 1 0.0000 0.993 1.000 0.000
#> GSM96970 1 0.0000 0.993 1.000 0.000
#> GSM96973 1 0.0000 0.993 1.000 0.000
#> GSM96976 2 0.1843 0.965 0.028 0.972
#> GSM96977 1 0.0000 0.993 1.000 0.000
#> GSM96995 2 0.0000 0.991 0.000 1.000
#> GSM97002 1 0.0000 0.993 1.000 0.000
#> GSM97009 2 0.0000 0.991 0.000 1.000
#> GSM97010 1 0.0000 0.993 1.000 0.000
#> GSM96974 1 0.1633 0.973 0.976 0.024
#> GSM96985 1 0.0000 0.993 1.000 0.000
#> GSM96959 2 0.0000 0.991 0.000 1.000
#> GSM96972 1 0.0000 0.993 1.000 0.000
#> GSM96978 2 0.6438 0.807 0.164 0.836
#> GSM96967 1 0.0000 0.993 1.000 0.000
#> GSM96987 1 0.0000 0.993 1.000 0.000
#> GSM97011 1 0.0672 0.988 0.992 0.008
#> GSM96964 1 0.0000 0.993 1.000 0.000
#> GSM96965 1 0.0000 0.993 1.000 0.000
#> GSM96981 1 0.0000 0.993 1.000 0.000
#> GSM96982 1 0.0000 0.993 1.000 0.000
#> GSM96988 1 0.2948 0.944 0.948 0.052
#> GSM97000 1 0.0000 0.993 1.000 0.000
#> GSM97004 1 0.0000 0.993 1.000 0.000
#> GSM97008 1 0.0000 0.993 1.000 0.000
#> GSM96950 1 0.0000 0.993 1.000 0.000
#> GSM96980 1 0.0000 0.993 1.000 0.000
#> GSM96989 1 0.0000 0.993 1.000 0.000
#> GSM96992 1 0.0000 0.993 1.000 0.000
#> GSM96993 1 0.0000 0.993 1.000 0.000
#> GSM96958 1 0.0000 0.993 1.000 0.000
#> GSM96951 1 0.0000 0.993 1.000 0.000
#> GSM96952 1 0.0000 0.993 1.000 0.000
#> GSM96961 1 0.0000 0.993 1.000 0.000
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM97038 2 0.0000 0.9099 0.000 1.000 0.000
#> GSM97045 2 0.0000 0.9099 0.000 1.000 0.000
#> GSM97047 2 0.0424 0.9066 0.000 0.992 0.008
#> GSM97025 2 0.0000 0.9099 0.000 1.000 0.000
#> GSM97030 2 0.5733 0.6573 0.000 0.676 0.324
#> GSM97027 2 0.0000 0.9099 0.000 1.000 0.000
#> GSM97033 2 0.0000 0.9099 0.000 1.000 0.000
#> GSM97034 2 0.6168 0.5472 0.000 0.588 0.412
#> GSM97020 2 0.0000 0.9099 0.000 1.000 0.000
#> GSM97026 2 0.0000 0.9099 0.000 1.000 0.000
#> GSM97012 2 0.0000 0.9099 0.000 1.000 0.000
#> GSM97015 2 0.6062 0.5862 0.000 0.616 0.384
#> GSM97016 2 0.0000 0.9099 0.000 1.000 0.000
#> GSM97017 3 0.6683 -0.4860 0.492 0.008 0.500
#> GSM97019 2 0.0000 0.9099 0.000 1.000 0.000
#> GSM97022 2 0.0000 0.9099 0.000 1.000 0.000
#> GSM97035 2 0.0000 0.9099 0.000 1.000 0.000
#> GSM97036 3 0.7493 -0.4751 0.480 0.036 0.484
#> GSM97039 2 0.0000 0.9099 0.000 1.000 0.000
#> GSM97046 2 0.0000 0.9099 0.000 1.000 0.000
#> GSM97023 3 0.6309 -0.4931 0.500 0.000 0.500
#> GSM97029 3 0.6954 -0.4807 0.484 0.016 0.500
#> GSM97043 2 0.0000 0.9099 0.000 1.000 0.000
#> GSM97013 3 0.6309 -0.4931 0.500 0.000 0.500
#> GSM96956 2 0.0592 0.9045 0.000 0.988 0.012
#> GSM97024 2 0.0000 0.9099 0.000 1.000 0.000
#> GSM97032 2 0.5859 0.6348 0.000 0.656 0.344
#> GSM97044 2 0.6305 0.4335 0.000 0.516 0.484
#> GSM97049 2 0.0000 0.9099 0.000 1.000 0.000
#> GSM96968 3 0.6587 0.3144 0.424 0.008 0.568
#> GSM96971 3 0.6309 0.3298 0.500 0.000 0.500
#> GSM96986 3 0.6307 0.3316 0.488 0.000 0.512
#> GSM97003 1 0.4842 0.4624 0.776 0.000 0.224
#> GSM96957 1 0.6309 0.4591 0.500 0.000 0.500
#> GSM96960 1 0.4796 0.4869 0.780 0.000 0.220
#> GSM96975 1 0.6180 0.4891 0.584 0.000 0.416
#> GSM96998 1 0.6299 0.4717 0.524 0.000 0.476
#> GSM96999 1 0.6309 0.4591 0.500 0.000 0.500
#> GSM97001 3 0.6309 -0.4931 0.500 0.000 0.500
#> GSM97005 1 0.6309 0.4591 0.500 0.000 0.500
#> GSM97006 1 0.6204 0.4884 0.576 0.000 0.424
#> GSM97021 3 0.6669 -0.4705 0.468 0.008 0.524
#> GSM97028 3 0.9303 -0.0435 0.184 0.316 0.500
#> GSM97031 3 0.6252 -0.4559 0.444 0.000 0.556
#> GSM97037 2 0.0892 0.9002 0.000 0.980 0.020
#> GSM97018 2 0.5785 0.6459 0.000 0.668 0.332
#> GSM97014 2 0.1163 0.8900 0.000 0.972 0.028
#> GSM97042 2 0.0000 0.9099 0.000 1.000 0.000
#> GSM97040 2 0.5650 0.5742 0.000 0.688 0.312
#> GSM97041 3 0.6683 -0.4860 0.492 0.008 0.500
#> GSM96955 2 0.0000 0.9099 0.000 1.000 0.000
#> GSM96990 2 0.6008 0.6022 0.000 0.628 0.372
#> GSM96991 2 0.0000 0.9099 0.000 1.000 0.000
#> GSM97048 2 0.0000 0.9099 0.000 1.000 0.000
#> GSM96963 2 0.0000 0.9099 0.000 1.000 0.000
#> GSM96953 2 0.0000 0.9099 0.000 1.000 0.000
#> GSM96966 1 0.0000 0.3754 1.000 0.000 0.000
#> GSM96979 3 0.6309 0.3298 0.500 0.000 0.500
#> GSM96983 3 0.7578 0.3273 0.460 0.040 0.500
#> GSM96984 3 0.6521 0.3322 0.492 0.004 0.504
#> GSM96994 3 0.6825 0.3324 0.488 0.012 0.500
#> GSM96996 1 0.5926 0.5003 0.644 0.000 0.356
#> GSM96997 3 0.6307 0.3316 0.488 0.000 0.512
#> GSM97007 3 0.6954 0.3322 0.484 0.016 0.500
#> GSM96954 3 0.5621 0.2468 0.308 0.000 0.692
#> GSM96962 3 0.6307 0.3316 0.488 0.000 0.512
#> GSM96969 1 0.0000 0.3754 1.000 0.000 0.000
#> GSM96970 1 0.0000 0.3754 1.000 0.000 0.000
#> GSM96973 1 0.0000 0.3754 1.000 0.000 0.000
#> GSM96976 1 0.8721 -0.3143 0.504 0.112 0.384
#> GSM96977 3 0.5760 -0.1383 0.328 0.000 0.672
#> GSM96995 3 0.8807 -0.1853 0.120 0.376 0.504
#> GSM97002 1 0.5058 0.4937 0.756 0.000 0.244
#> GSM97009 2 0.0592 0.9027 0.000 0.988 0.012
#> GSM97010 1 0.2599 0.3252 0.932 0.016 0.052
#> GSM96974 1 0.6267 -0.3310 0.548 0.000 0.452
#> GSM96985 1 0.6260 -0.3275 0.552 0.000 0.448
#> GSM96959 2 0.4605 0.7769 0.000 0.796 0.204
#> GSM96972 1 0.0000 0.3754 1.000 0.000 0.000
#> GSM96978 3 0.6521 0.3311 0.496 0.004 0.500
#> GSM96967 1 0.0000 0.3754 1.000 0.000 0.000
#> GSM96987 1 0.6307 0.4666 0.512 0.000 0.488
#> GSM97011 3 0.7392 -0.4683 0.468 0.032 0.500
#> GSM96964 1 0.6309 0.4629 0.504 0.000 0.496
#> GSM96965 1 0.0747 0.3583 0.984 0.016 0.000
#> GSM96981 1 0.6008 0.4989 0.628 0.000 0.372
#> GSM96982 1 0.3482 0.4532 0.872 0.000 0.128
#> GSM96988 3 0.6309 0.3298 0.500 0.000 0.500
#> GSM97000 3 0.5560 -0.3387 0.300 0.000 0.700
#> GSM97004 1 0.5497 0.4999 0.708 0.000 0.292
#> GSM97008 3 0.6260 -0.4572 0.448 0.000 0.552
#> GSM96950 1 0.6309 0.4629 0.504 0.000 0.496
#> GSM96980 1 0.2625 0.4294 0.916 0.000 0.084
#> GSM96989 1 0.6305 0.4686 0.516 0.000 0.484
#> GSM96992 1 0.6308 0.4655 0.508 0.000 0.492
#> GSM96993 1 0.6307 0.4666 0.512 0.000 0.488
#> GSM96958 1 0.6309 0.4629 0.504 0.000 0.496
#> GSM96951 3 0.6309 -0.4931 0.500 0.000 0.500
#> GSM96952 1 0.6309 0.4629 0.504 0.000 0.496
#> GSM96961 1 0.6309 0.4591 0.500 0.000 0.500
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM97038 2 0.0000 0.9486 0.000 1.000 0.000 0.000
#> GSM97045 2 0.0188 0.9481 0.000 0.996 0.004 0.000
#> GSM97047 2 0.5337 0.6441 0.260 0.696 0.044 0.000
#> GSM97025 2 0.0188 0.9481 0.000 0.996 0.004 0.000
#> GSM97030 3 0.4134 0.7034 0.000 0.260 0.740 0.000
#> GSM97027 2 0.0188 0.9481 0.000 0.996 0.004 0.000
#> GSM97033 2 0.0000 0.9486 0.000 1.000 0.000 0.000
#> GSM97034 3 0.3764 0.7552 0.000 0.216 0.784 0.000
#> GSM97020 2 0.0000 0.9486 0.000 1.000 0.000 0.000
#> GSM97026 2 0.0937 0.9371 0.012 0.976 0.012 0.000
#> GSM97012 2 0.0000 0.9486 0.000 1.000 0.000 0.000
#> GSM97015 3 0.3569 0.7711 0.000 0.196 0.804 0.000
#> GSM97016 2 0.0000 0.9486 0.000 1.000 0.000 0.000
#> GSM97017 1 0.0000 0.7920 1.000 0.000 0.000 0.000
#> GSM97019 2 0.0188 0.9481 0.000 0.996 0.004 0.000
#> GSM97022 2 0.0188 0.9481 0.000 0.996 0.004 0.000
#> GSM97035 2 0.0188 0.9481 0.000 0.996 0.004 0.000
#> GSM97036 4 0.6499 0.2500 0.400 0.076 0.000 0.524
#> GSM97039 2 0.0000 0.9486 0.000 1.000 0.000 0.000
#> GSM97046 2 0.0000 0.9486 0.000 1.000 0.000 0.000
#> GSM97023 1 0.3074 0.7442 0.848 0.000 0.000 0.152
#> GSM97029 1 0.1706 0.7894 0.948 0.016 0.000 0.036
#> GSM97043 2 0.0188 0.9481 0.000 0.996 0.004 0.000
#> GSM97013 1 0.2760 0.7578 0.872 0.000 0.000 0.128
#> GSM96956 2 0.2704 0.8234 0.000 0.876 0.124 0.000
#> GSM97024 2 0.0188 0.9481 0.000 0.996 0.004 0.000
#> GSM97032 3 0.4643 0.5725 0.000 0.344 0.656 0.000
#> GSM97044 3 0.2081 0.8387 0.000 0.084 0.916 0.000
#> GSM97049 2 0.0000 0.9486 0.000 1.000 0.000 0.000
#> GSM96968 3 0.0524 0.8537 0.004 0.000 0.988 0.008
#> GSM96971 3 0.3610 0.7242 0.000 0.000 0.800 0.200
#> GSM96986 3 0.1059 0.8521 0.012 0.000 0.972 0.016
#> GSM97003 4 0.5993 0.6258 0.148 0.000 0.160 0.692
#> GSM96957 1 0.0657 0.7935 0.984 0.000 0.004 0.012
#> GSM96960 4 0.3479 0.7161 0.148 0.000 0.012 0.840
#> GSM96975 4 0.4888 0.2650 0.412 0.000 0.000 0.588
#> GSM96998 4 0.4356 0.5679 0.292 0.000 0.000 0.708
#> GSM96999 1 0.3810 0.7072 0.804 0.000 0.008 0.188
#> GSM97001 1 0.0336 0.7922 0.992 0.000 0.008 0.000
#> GSM97005 1 0.0469 0.7914 0.988 0.000 0.012 0.000
#> GSM97006 4 0.4699 0.5259 0.320 0.000 0.004 0.676
#> GSM97021 1 0.0336 0.7912 0.992 0.000 0.008 0.000
#> GSM97028 3 0.1256 0.8544 0.000 0.028 0.964 0.008
#> GSM97031 1 0.3687 0.7697 0.856 0.000 0.064 0.080
#> GSM97037 2 0.3123 0.7776 0.000 0.844 0.156 0.000
#> GSM97018 3 0.4790 0.4945 0.000 0.380 0.620 0.000
#> GSM97014 2 0.4877 0.4138 0.408 0.592 0.000 0.000
#> GSM97042 2 0.0188 0.9481 0.000 0.996 0.004 0.000
#> GSM97040 1 0.3959 0.6722 0.840 0.092 0.068 0.000
#> GSM97041 1 0.0000 0.7920 1.000 0.000 0.000 0.000
#> GSM96955 2 0.0804 0.9361 0.012 0.980 0.000 0.008
#> GSM96990 3 0.3975 0.7291 0.000 0.240 0.760 0.000
#> GSM96991 2 0.0000 0.9486 0.000 1.000 0.000 0.000
#> GSM97048 2 0.0000 0.9486 0.000 1.000 0.000 0.000
#> GSM96963 2 0.0000 0.9486 0.000 1.000 0.000 0.000
#> GSM96953 2 0.0000 0.9486 0.000 1.000 0.000 0.000
#> GSM96966 4 0.0336 0.7592 0.000 0.000 0.008 0.992
#> GSM96979 3 0.3024 0.7683 0.000 0.000 0.852 0.148
#> GSM96983 3 0.1042 0.8540 0.000 0.008 0.972 0.020
#> GSM96984 3 0.0592 0.8527 0.000 0.000 0.984 0.016
#> GSM96994 3 0.0188 0.8537 0.000 0.000 0.996 0.004
#> GSM96996 4 0.3402 0.7055 0.164 0.000 0.004 0.832
#> GSM96997 3 0.0817 0.8513 0.000 0.000 0.976 0.024
#> GSM97007 3 0.0336 0.8531 0.000 0.000 0.992 0.008
#> GSM96954 3 0.2401 0.8066 0.092 0.000 0.904 0.004
#> GSM96962 3 0.0336 0.8531 0.000 0.000 0.992 0.008
#> GSM96969 4 0.0336 0.7592 0.000 0.000 0.008 0.992
#> GSM96970 4 0.0336 0.7592 0.000 0.000 0.008 0.992
#> GSM96973 4 0.0336 0.7592 0.000 0.000 0.008 0.992
#> GSM96976 4 0.6192 0.4125 0.000 0.104 0.244 0.652
#> GSM96977 1 0.7369 0.3665 0.524 0.000 0.228 0.248
#> GSM96995 3 0.1575 0.8501 0.028 0.012 0.956 0.004
#> GSM97002 4 0.2149 0.7477 0.088 0.000 0.000 0.912
#> GSM97009 2 0.4508 0.7509 0.184 0.780 0.036 0.000
#> GSM97010 4 0.2500 0.7481 0.040 0.000 0.044 0.916
#> GSM96974 4 0.3837 0.5772 0.000 0.000 0.224 0.776
#> GSM96985 4 0.2868 0.6761 0.000 0.000 0.136 0.864
#> GSM96959 3 0.7741 0.3396 0.296 0.264 0.440 0.000
#> GSM96972 4 0.0524 0.7598 0.004 0.000 0.008 0.988
#> GSM96978 3 0.2530 0.8091 0.000 0.000 0.888 0.112
#> GSM96967 4 0.0336 0.7592 0.000 0.000 0.008 0.992
#> GSM96987 4 0.4804 0.3964 0.384 0.000 0.000 0.616
#> GSM97011 1 0.1816 0.7776 0.948 0.004 0.024 0.024
#> GSM96964 1 0.4817 0.3648 0.612 0.000 0.000 0.388
#> GSM96965 4 0.1339 0.7530 0.024 0.004 0.008 0.964
#> GSM96981 4 0.2647 0.7277 0.120 0.000 0.000 0.880
#> GSM96982 4 0.0707 0.7602 0.020 0.000 0.000 0.980
#> GSM96988 3 0.1637 0.8416 0.000 0.000 0.940 0.060
#> GSM97000 1 0.2760 0.7119 0.872 0.000 0.128 0.000
#> GSM97004 4 0.2345 0.7436 0.100 0.000 0.000 0.900
#> GSM97008 1 0.0817 0.7879 0.976 0.000 0.024 0.000
#> GSM96950 1 0.4564 0.5215 0.672 0.000 0.000 0.328
#> GSM96980 4 0.0592 0.7597 0.016 0.000 0.000 0.984
#> GSM96989 4 0.4730 0.4419 0.364 0.000 0.000 0.636
#> GSM96992 4 0.5147 0.1727 0.460 0.000 0.004 0.536
#> GSM96993 1 0.4428 0.6074 0.720 0.000 0.004 0.276
#> GSM96958 1 0.4699 0.5315 0.676 0.000 0.004 0.320
#> GSM96951 1 0.4391 0.6380 0.740 0.000 0.008 0.252
#> GSM96952 4 0.5167 0.0552 0.488 0.000 0.004 0.508
#> GSM96961 1 0.4837 0.4668 0.648 0.000 0.004 0.348
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM97038 2 0.1965 0.90261 0.024 0.924 0.000 0.000 0.052
#> GSM97045 2 0.0693 0.91115 0.008 0.980 0.000 0.000 0.012
#> GSM97047 5 0.5988 0.12639 0.044 0.400 0.036 0.000 0.520
#> GSM97025 2 0.0451 0.91200 0.004 0.988 0.000 0.000 0.008
#> GSM97030 3 0.6138 0.57741 0.044 0.280 0.604 0.000 0.072
#> GSM97027 2 0.0693 0.91173 0.008 0.980 0.000 0.000 0.012
#> GSM97033 2 0.1741 0.90526 0.024 0.936 0.000 0.000 0.040
#> GSM97034 3 0.6530 0.59160 0.084 0.268 0.584 0.000 0.064
#> GSM97020 2 0.2067 0.90240 0.032 0.920 0.000 0.000 0.048
#> GSM97026 2 0.5132 0.68090 0.152 0.728 0.020 0.000 0.100
#> GSM97012 2 0.0451 0.91136 0.004 0.988 0.000 0.000 0.008
#> GSM97015 3 0.6439 0.63812 0.076 0.172 0.636 0.000 0.116
#> GSM97016 2 0.1981 0.90266 0.028 0.924 0.000 0.000 0.048
#> GSM97017 5 0.4273 0.37946 0.448 0.000 0.000 0.000 0.552
#> GSM97019 2 0.0671 0.90936 0.004 0.980 0.000 0.000 0.016
#> GSM97022 2 0.0451 0.91027 0.004 0.988 0.000 0.000 0.008
#> GSM97035 2 0.0162 0.91191 0.000 0.996 0.000 0.000 0.004
#> GSM97036 1 0.6033 0.53360 0.668 0.056 0.000 0.168 0.108
#> GSM97039 2 0.2054 0.90124 0.028 0.920 0.000 0.000 0.052
#> GSM97046 2 0.2054 0.90124 0.028 0.920 0.000 0.000 0.052
#> GSM97023 1 0.5224 0.44635 0.644 0.000 0.000 0.080 0.276
#> GSM97029 1 0.5605 -0.21541 0.488 0.036 0.004 0.012 0.460
#> GSM97043 2 0.1498 0.89865 0.016 0.952 0.008 0.000 0.024
#> GSM97013 1 0.4221 0.39080 0.732 0.000 0.000 0.032 0.236
#> GSM96956 2 0.4683 0.77701 0.036 0.776 0.120 0.000 0.068
#> GSM97024 2 0.1095 0.90447 0.012 0.968 0.012 0.000 0.008
#> GSM97032 3 0.6586 0.45077 0.072 0.356 0.516 0.000 0.056
#> GSM97044 3 0.4380 0.71259 0.052 0.116 0.796 0.000 0.036
#> GSM97049 2 0.2139 0.90089 0.032 0.916 0.000 0.000 0.052
#> GSM96968 3 0.3802 0.71699 0.096 0.000 0.820 0.004 0.080
#> GSM96971 3 0.5459 0.15746 0.012 0.000 0.496 0.456 0.036
#> GSM96986 3 0.4075 0.67918 0.024 0.000 0.804 0.036 0.136
#> GSM97003 4 0.8113 0.00600 0.308 0.000 0.164 0.384 0.144
#> GSM96957 1 0.4306 -0.18346 0.508 0.000 0.000 0.000 0.492
#> GSM96960 4 0.6054 -0.03420 0.408 0.000 0.016 0.500 0.076
#> GSM96975 4 0.6674 -0.09533 0.324 0.000 0.000 0.428 0.248
#> GSM96998 1 0.4482 0.42233 0.636 0.000 0.000 0.348 0.016
#> GSM96999 1 0.6328 0.40008 0.540 0.000 0.016 0.120 0.324
#> GSM97001 5 0.3932 0.50963 0.328 0.000 0.000 0.000 0.672
#> GSM97005 5 0.3662 0.57678 0.252 0.000 0.004 0.000 0.744
#> GSM97006 1 0.6282 0.25931 0.492 0.000 0.028 0.404 0.076
#> GSM97021 5 0.3684 0.57345 0.280 0.000 0.000 0.000 0.720
#> GSM97028 3 0.4862 0.71801 0.076 0.056 0.784 0.008 0.076
#> GSM97031 5 0.6582 0.25921 0.292 0.000 0.084 0.060 0.564
#> GSM97037 2 0.5801 0.61416 0.052 0.676 0.196 0.000 0.076
#> GSM97018 3 0.7167 0.38364 0.084 0.368 0.456 0.000 0.092
#> GSM97014 5 0.5357 0.32561 0.068 0.344 0.000 0.000 0.588
#> GSM97042 2 0.0693 0.91005 0.008 0.980 0.000 0.000 0.012
#> GSM97040 5 0.4607 0.59410 0.136 0.036 0.052 0.000 0.776
#> GSM97041 5 0.4219 0.45084 0.416 0.000 0.000 0.000 0.584
#> GSM96955 2 0.4126 0.81900 0.036 0.808 0.008 0.016 0.132
#> GSM96990 3 0.6015 0.65937 0.084 0.160 0.676 0.000 0.080
#> GSM96991 2 0.0693 0.90965 0.008 0.980 0.000 0.000 0.012
#> GSM97048 2 0.2139 0.90089 0.032 0.916 0.000 0.000 0.052
#> GSM96963 2 0.1211 0.91045 0.016 0.960 0.000 0.000 0.024
#> GSM96953 2 0.0807 0.91180 0.012 0.976 0.000 0.000 0.012
#> GSM96966 4 0.0609 0.70472 0.020 0.000 0.000 0.980 0.000
#> GSM96979 3 0.4889 0.62148 0.024 0.000 0.740 0.176 0.060
#> GSM96983 3 0.2893 0.73090 0.052 0.008 0.888 0.004 0.048
#> GSM96984 3 0.2291 0.72371 0.012 0.000 0.916 0.024 0.048
#> GSM96994 3 0.1967 0.72806 0.012 0.000 0.932 0.020 0.036
#> GSM96996 1 0.5496 0.08574 0.492 0.000 0.020 0.460 0.028
#> GSM96997 3 0.3748 0.69588 0.020 0.000 0.836 0.056 0.088
#> GSM97007 3 0.1921 0.72649 0.012 0.000 0.932 0.012 0.044
#> GSM96954 3 0.5311 0.57920 0.096 0.000 0.692 0.012 0.200
#> GSM96962 3 0.2302 0.72504 0.016 0.000 0.916 0.020 0.048
#> GSM96969 4 0.0510 0.70468 0.016 0.000 0.000 0.984 0.000
#> GSM96970 4 0.0404 0.70492 0.012 0.000 0.000 0.988 0.000
#> GSM96973 4 0.0290 0.70416 0.008 0.000 0.000 0.992 0.000
#> GSM96976 4 0.4940 0.54147 0.032 0.040 0.140 0.768 0.020
#> GSM96977 1 0.8274 0.00916 0.376 0.000 0.172 0.176 0.276
#> GSM96995 3 0.4938 0.65368 0.064 0.012 0.716 0.000 0.208
#> GSM97002 4 0.4697 0.32600 0.360 0.000 0.008 0.620 0.012
#> GSM97009 2 0.6491 0.26663 0.052 0.524 0.048 0.008 0.368
#> GSM97010 4 0.6882 0.37208 0.280 0.000 0.092 0.548 0.080
#> GSM96974 4 0.2669 0.63110 0.020 0.000 0.104 0.876 0.000
#> GSM96985 4 0.3844 0.63218 0.044 0.000 0.104 0.828 0.024
#> GSM96959 5 0.6896 0.18060 0.060 0.132 0.252 0.000 0.556
#> GSM96972 4 0.0880 0.70235 0.032 0.000 0.000 0.968 0.000
#> GSM96978 3 0.4986 0.63225 0.036 0.000 0.720 0.208 0.036
#> GSM96967 4 0.0404 0.70492 0.012 0.000 0.000 0.988 0.000
#> GSM96987 1 0.3849 0.58242 0.752 0.000 0.000 0.232 0.016
#> GSM97011 5 0.3770 0.61332 0.160 0.012 0.004 0.016 0.808
#> GSM96964 1 0.3573 0.63237 0.812 0.000 0.000 0.152 0.036
#> GSM96965 4 0.1243 0.68930 0.028 0.004 0.000 0.960 0.008
#> GSM96981 4 0.5215 0.45345 0.240 0.000 0.000 0.664 0.096
#> GSM96982 4 0.3910 0.58585 0.196 0.000 0.000 0.772 0.032
#> GSM96988 3 0.4106 0.72177 0.048 0.004 0.824 0.088 0.036
#> GSM97000 5 0.3575 0.59781 0.120 0.000 0.056 0.000 0.824
#> GSM97004 4 0.4564 0.25140 0.388 0.000 0.004 0.600 0.008
#> GSM97008 5 0.3266 0.60471 0.200 0.000 0.004 0.000 0.796
#> GSM96950 1 0.4444 0.60288 0.760 0.000 0.000 0.136 0.104
#> GSM96980 4 0.2674 0.64515 0.140 0.000 0.000 0.856 0.004
#> GSM96989 1 0.3807 0.57526 0.748 0.000 0.000 0.240 0.012
#> GSM96992 1 0.5781 0.52921 0.596 0.000 0.004 0.292 0.108
#> GSM96993 1 0.3982 0.52244 0.812 0.000 0.012 0.060 0.116
#> GSM96958 1 0.5831 0.56633 0.608 0.000 0.000 0.172 0.220
#> GSM96951 1 0.6208 0.55008 0.592 0.000 0.016 0.140 0.252
#> GSM96952 1 0.5575 0.54736 0.612 0.000 0.000 0.280 0.108
#> GSM96961 1 0.5072 0.62315 0.696 0.000 0.000 0.188 0.116
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM97038 2 0.3824 0.76607 0.016 0.792 0.156 0.016 0.020 0.000
#> GSM97045 2 0.1265 0.79966 0.008 0.948 0.044 0.000 0.000 0.000
#> GSM97047 5 0.6703 0.21870 0.020 0.320 0.196 0.000 0.444 0.020
#> GSM97025 2 0.1155 0.80007 0.004 0.956 0.036 0.004 0.000 0.000
#> GSM97030 3 0.5950 0.65692 0.000 0.172 0.480 0.004 0.004 0.340
#> GSM97027 2 0.1606 0.80094 0.008 0.932 0.056 0.004 0.000 0.000
#> GSM97033 2 0.3155 0.78459 0.012 0.844 0.116 0.016 0.012 0.000
#> GSM97034 3 0.6313 0.68062 0.008 0.220 0.456 0.000 0.008 0.308
#> GSM97020 2 0.3425 0.78024 0.016 0.828 0.124 0.016 0.016 0.000
#> GSM97026 2 0.6330 0.43153 0.100 0.588 0.224 0.004 0.076 0.008
#> GSM97012 2 0.1349 0.79269 0.004 0.940 0.056 0.000 0.000 0.000
#> GSM97015 3 0.5587 0.62533 0.000 0.084 0.532 0.000 0.024 0.360
#> GSM97016 2 0.3673 0.77238 0.016 0.808 0.140 0.016 0.020 0.000
#> GSM97017 5 0.4506 0.41131 0.348 0.000 0.044 0.000 0.608 0.000
#> GSM97019 2 0.1588 0.78729 0.004 0.924 0.072 0.000 0.000 0.000
#> GSM97022 2 0.1285 0.79168 0.004 0.944 0.052 0.000 0.000 0.000
#> GSM97035 2 0.0935 0.80026 0.004 0.964 0.032 0.000 0.000 0.000
#> GSM97036 1 0.6131 0.42269 0.668 0.072 0.116 0.088 0.052 0.004
#> GSM97039 2 0.3508 0.77606 0.016 0.820 0.132 0.016 0.016 0.000
#> GSM97046 2 0.3824 0.76344 0.016 0.792 0.156 0.016 0.020 0.000
#> GSM97023 1 0.5082 0.38602 0.624 0.000 0.048 0.032 0.296 0.000
#> GSM97029 1 0.6509 -0.20010 0.424 0.036 0.128 0.012 0.400 0.000
#> GSM97043 2 0.2553 0.73253 0.008 0.848 0.144 0.000 0.000 0.000
#> GSM97013 1 0.4774 0.36469 0.700 0.000 0.084 0.020 0.196 0.000
#> GSM96956 2 0.6001 0.55602 0.016 0.608 0.256 0.016 0.024 0.080
#> GSM97024 2 0.2809 0.72653 0.004 0.848 0.128 0.000 0.000 0.020
#> GSM97032 3 0.6161 0.65451 0.012 0.264 0.476 0.000 0.000 0.248
#> GSM97044 6 0.5297 -0.55985 0.004 0.088 0.412 0.000 0.000 0.496
#> GSM97049 2 0.4024 0.75968 0.020 0.780 0.160 0.016 0.024 0.000
#> GSM96968 6 0.5654 0.21878 0.044 0.000 0.264 0.016 0.056 0.620
#> GSM96971 6 0.4884 0.08964 0.000 0.000 0.048 0.460 0.004 0.488
#> GSM96986 6 0.2683 0.51298 0.020 0.000 0.024 0.008 0.060 0.888
#> GSM97003 6 0.8570 -0.28636 0.260 0.000 0.092 0.216 0.140 0.292
#> GSM96957 5 0.5935 0.14047 0.352 0.000 0.124 0.012 0.504 0.008
#> GSM96960 1 0.7383 0.35278 0.448 0.000 0.100 0.312 0.084 0.056
#> GSM96975 4 0.7535 -0.14460 0.280 0.000 0.104 0.340 0.268 0.008
#> GSM96998 1 0.4161 0.57857 0.752 0.000 0.040 0.188 0.016 0.004
#> GSM96999 1 0.6610 0.26032 0.444 0.000 0.072 0.088 0.384 0.012
#> GSM97001 5 0.3472 0.55609 0.136 0.000 0.044 0.004 0.812 0.004
#> GSM97005 5 0.2829 0.59014 0.096 0.000 0.024 0.000 0.864 0.016
#> GSM97006 1 0.6958 0.47652 0.524 0.000 0.076 0.252 0.112 0.036
#> GSM97021 5 0.3817 0.57674 0.152 0.000 0.052 0.000 0.784 0.012
#> GSM97028 3 0.5406 0.41546 0.004 0.048 0.504 0.008 0.012 0.424
#> GSM97031 5 0.7039 0.20357 0.220 0.000 0.048 0.048 0.520 0.164
#> GSM97037 2 0.6281 0.34781 0.016 0.508 0.352 0.008 0.024 0.092
#> GSM97018 3 0.6194 0.65786 0.004 0.208 0.524 0.000 0.020 0.244
#> GSM97014 5 0.6143 0.34378 0.040 0.272 0.120 0.008 0.560 0.000
#> GSM97042 2 0.1471 0.78907 0.004 0.932 0.064 0.000 0.000 0.000
#> GSM97040 5 0.4302 0.59615 0.060 0.036 0.116 0.000 0.780 0.008
#> GSM97041 5 0.4332 0.44754 0.316 0.000 0.040 0.000 0.644 0.000
#> GSM96955 2 0.6049 0.57269 0.024 0.592 0.256 0.032 0.096 0.000
#> GSM96990 3 0.5879 0.57270 0.004 0.104 0.488 0.000 0.020 0.384
#> GSM96991 2 0.2062 0.78390 0.008 0.900 0.088 0.000 0.004 0.000
#> GSM97048 2 0.3908 0.76270 0.020 0.788 0.156 0.016 0.020 0.000
#> GSM96963 2 0.2001 0.79245 0.008 0.912 0.068 0.000 0.012 0.000
#> GSM96953 2 0.1219 0.80104 0.004 0.948 0.048 0.000 0.000 0.000
#> GSM96966 4 0.1141 0.74789 0.052 0.000 0.000 0.948 0.000 0.000
#> GSM96979 6 0.4456 0.46916 0.044 0.000 0.052 0.120 0.012 0.772
#> GSM96983 6 0.4420 -0.00186 0.000 0.012 0.348 0.008 0.008 0.624
#> GSM96984 6 0.0405 0.51320 0.004 0.000 0.000 0.008 0.000 0.988
#> GSM96994 6 0.1493 0.48879 0.000 0.000 0.056 0.004 0.004 0.936
#> GSM96996 1 0.5926 0.41509 0.576 0.000 0.064 0.300 0.040 0.020
#> GSM96997 6 0.2666 0.51372 0.024 0.000 0.044 0.008 0.032 0.892
#> GSM97007 6 0.0777 0.50298 0.000 0.000 0.024 0.000 0.004 0.972
#> GSM96954 6 0.5787 0.33376 0.056 0.000 0.136 0.004 0.164 0.640
#> GSM96962 6 0.1036 0.50874 0.004 0.000 0.024 0.000 0.008 0.964
#> GSM96969 4 0.1152 0.75039 0.044 0.000 0.004 0.952 0.000 0.000
#> GSM96970 4 0.0713 0.75266 0.028 0.000 0.000 0.972 0.000 0.000
#> GSM96973 4 0.0858 0.75240 0.028 0.000 0.000 0.968 0.000 0.004
#> GSM96976 4 0.3306 0.67301 0.012 0.008 0.028 0.844 0.004 0.104
#> GSM96977 5 0.8725 0.06496 0.268 0.000 0.180 0.132 0.276 0.144
#> GSM96995 6 0.5806 -0.23627 0.008 0.004 0.428 0.004 0.104 0.452
#> GSM97002 1 0.5949 0.19187 0.472 0.000 0.056 0.420 0.036 0.016
#> GSM97009 2 0.7286 -0.07324 0.016 0.400 0.116 0.012 0.376 0.080
#> GSM97010 4 0.7688 0.17563 0.288 0.000 0.124 0.432 0.072 0.084
#> GSM96974 4 0.2434 0.71089 0.008 0.000 0.036 0.892 0.000 0.064
#> GSM96985 4 0.5133 0.61821 0.028 0.000 0.128 0.708 0.012 0.124
#> GSM96959 5 0.7395 0.16279 0.016 0.108 0.312 0.008 0.436 0.120
#> GSM96972 4 0.1866 0.72867 0.084 0.000 0.008 0.908 0.000 0.000
#> GSM96978 6 0.5356 0.32373 0.000 0.000 0.196 0.152 0.016 0.636
#> GSM96967 4 0.0713 0.75266 0.028 0.000 0.000 0.972 0.000 0.000
#> GSM96987 1 0.3042 0.61315 0.836 0.000 0.032 0.128 0.004 0.000
#> GSM97011 5 0.3403 0.60764 0.048 0.008 0.068 0.012 0.852 0.012
#> GSM96964 1 0.2945 0.58852 0.868 0.000 0.040 0.064 0.028 0.000
#> GSM96965 4 0.1649 0.72635 0.040 0.000 0.016 0.936 0.008 0.000
#> GSM96981 4 0.6582 0.26548 0.268 0.000 0.100 0.524 0.104 0.004
#> GSM96982 4 0.6128 0.32659 0.260 0.000 0.100 0.572 0.064 0.004
#> GSM96988 6 0.5573 -0.00132 0.004 0.008 0.336 0.084 0.008 0.560
#> GSM97000 5 0.2976 0.60159 0.020 0.000 0.020 0.000 0.852 0.108
#> GSM97004 1 0.5498 0.14414 0.460 0.000 0.060 0.452 0.028 0.000
#> GSM97008 5 0.2318 0.60134 0.048 0.000 0.020 0.000 0.904 0.028
#> GSM96950 1 0.5000 0.52200 0.728 0.000 0.068 0.092 0.108 0.004
#> GSM96980 4 0.3242 0.65221 0.148 0.000 0.032 0.816 0.004 0.000
#> GSM96989 1 0.3072 0.61628 0.836 0.000 0.036 0.124 0.004 0.000
#> GSM96992 1 0.6310 0.57591 0.584 0.000 0.076 0.156 0.180 0.004
#> GSM96993 1 0.4396 0.46879 0.780 0.004 0.084 0.028 0.096 0.008
#> GSM96958 1 0.6352 0.48403 0.556 0.000 0.080 0.092 0.264 0.008
#> GSM96951 1 0.6450 0.43082 0.512 0.000 0.056 0.076 0.332 0.024
#> GSM96952 1 0.5771 0.59233 0.640 0.000 0.052 0.148 0.156 0.004
#> GSM96961 1 0.5171 0.59327 0.696 0.000 0.040 0.100 0.160 0.004
Heatmaps for the consensus matrix. It visualizes the probability of two samples to be in a same group.
consensus_heatmap(res, k = 2)
consensus_heatmap(res, k = 3)
consensus_heatmap(res, k = 4)
consensus_heatmap(res, k = 5)
consensus_heatmap(res, k = 6)
Heatmaps for the membership of samples in all partitions to see how consistent they are:
membership_heatmap(res, k = 2)
membership_heatmap(res, k = 3)
membership_heatmap(res, k = 4)
membership_heatmap(res, k = 5)
membership_heatmap(res, k = 6)
As soon as we have had the classes for columns, we can look for signatures which are significantly different between classes which can be candidate marks for certain classes. Following are the heatmaps for signatures.
Signature heatmaps where rows are scaled:
get_signatures(res, k = 2)
get_signatures(res, k = 3)
get_signatures(res, k = 4)
get_signatures(res, k = 5)
get_signatures(res, k = 6)
Signature heatmaps where rows are not scaled:
get_signatures(res, k = 2, scale_rows = FALSE)
get_signatures(res, k = 3, scale_rows = FALSE)
get_signatures(res, k = 4, scale_rows = FALSE)
get_signatures(res, k = 5, scale_rows = FALSE)
get_signatures(res, k = 6, scale_rows = FALSE)
Compare the overlap of signatures from different k:
compare_signatures(res)
get_signature() returns a data frame invisibly. TO get the list of signatures, the function
call should be assigned to a variable explicitly. In following code, if plot argument is set
to FALSE, no heatmap is plotted while only the differential analysis is performed.
# code only for demonstration
tb = get_signature(res, k = ..., plot = FALSE)
An example of the output of tb is:
#> which_row fdr mean_1 mean_2 scaled_mean_1 scaled_mean_2 km
#> 1 38 0.042760348 8.373488 9.131774 -0.5533452 0.5164555 1
#> 2 40 0.018707592 7.106213 8.469186 -0.6173731 0.5762149 1
#> 3 55 0.019134737 10.221463 11.207825 -0.6159697 0.5749050 1
#> 4 59 0.006059896 5.921854 7.869574 -0.6899429 0.6439467 1
#> 5 60 0.018055526 8.928898 10.211722 -0.6204761 0.5791110 1
#> 6 98 0.009384629 15.714769 14.887706 0.6635654 -0.6193277 2
...
The columns in tb are:
which_row: row indices corresponding to the input matrix.fdr: FDR for the differential test. mean_x: The mean value in group x.scaled_mean_x: The mean value in group x after rows are scaled.km: Row groups if k-means clustering is applied to rows.UMAP plot which shows how samples are separated.
dimension_reduction(res, k = 2, method = "UMAP")
dimension_reduction(res, k = 3, method = "UMAP")
dimension_reduction(res, k = 4, method = "UMAP")
dimension_reduction(res, k = 5, method = "UMAP")
dimension_reduction(res, k = 6, method = "UMAP")
Following heatmap shows how subgroups are split when increasing k:
collect_classes(res)
Test correlation between subgroups and known annotations. If the known annotation is numeric, one-way ANOVA test is applied, and if the known annotation is discrete, chi-squared contingency table test is applied.
test_to_known_factors(res)
#> n disease.state(p) specimen(p) cell.type(p) other(p) k
#> CV:skmeans 100 9.36e-05 0.1734 5.21e-13 0.0903 2
#> CV:skmeans 37 3.62e-01 1.0000 2.96e-03 0.2540 3
#> CV:skmeans 87 1.23e-04 0.1849 1.40e-17 0.0583 4
#> CV:skmeans 74 1.74e-05 0.0624 8.83e-17 0.0138 5
#> CV:skmeans 60 3.04e-04 0.3256 1.60e-17 0.0982 6
If matrix rows can be associated to genes, consider to use functional_enrichment(res,
...) to perform function enrichment for the signature genes. See this vignette for more detailed explanations.
The object with results only for a single top-value method and a single partition method can be extracted as:
res = res_list["CV", "pam"]
# you can also extract it by
# res = res_list["CV:pam"]
A summary of res and all the functions that can be applied to it:
res
#> A 'ConsensusPartition' object with k = 2, 3, 4, 5, 6.
#> On a matrix with 21168 rows and 100 columns.
#> Top rows (1000, 2000, 3000, 4000, 5000) are extracted by 'CV' method.
#> Subgroups are detected by 'pam' method.
#> Performed in total 1250 partitions by row resampling.
#> Best k for subgroups seems to be 2.
#>
#> Following methods can be applied to this 'ConsensusPartition' object:
#> [1] "cola_report" "collect_classes" "collect_plots"
#> [4] "collect_stats" "colnames" "compare_signatures"
#> [7] "consensus_heatmap" "dimension_reduction" "functional_enrichment"
#> [10] "get_anno_col" "get_anno" "get_classes"
#> [13] "get_consensus" "get_matrix" "get_membership"
#> [16] "get_param" "get_signatures" "get_stats"
#> [19] "is_best_k" "is_stable_k" "membership_heatmap"
#> [22] "ncol" "nrow" "plot_ecdf"
#> [25] "rownames" "select_partition_number" "show"
#> [28] "suggest_best_k" "test_to_known_factors"
collect_plots() function collects all the plots made from res for all k (number of partitions)
into one single page to provide an easy and fast comparison between different k.
collect_plots(res)
The plots are:
k and the heatmap of
predicted classes for each k.k.k.k.All the plots in panels can be made by individual functions and they are plotted later in this section.
select_partition_number() produces several plots showing different
statistics for choosing “optimized” k. There are following statistics:
k;k, the area increased is defined as \(A_k - A_{k-1}\).The detailed explanations of these statistics can be found in the cola vignette.
Generally speaking, lower PAC score, higher mean silhouette score or higher
concordance corresponds to better partition. Rand index and Jaccard index
measure how similar the current partition is compared to partition with k-1.
If they are too similar, we won't accept k is better than k-1.
select_partition_number(res)
The numeric values for all these statistics can be obtained by get_stats().
get_stats(res)
#> k 1-PAC mean_silhouette concordance area_increased Rand Jaccard
#> 2 2 0.626 0.862 0.929 0.4867 0.508 0.508
#> 3 3 0.518 0.701 0.854 0.3648 0.732 0.515
#> 4 4 0.528 0.551 0.760 0.1183 0.863 0.620
#> 5 5 0.627 0.645 0.795 0.0650 0.848 0.502
#> 6 6 0.627 0.495 0.686 0.0425 0.931 0.695
suggest_best_k() suggests the best \(k\) based on these statistics. The rules are as follows:
suggest_best_k(res)
#> [1] 2
Following shows the table of the partitions (You need to click the show/hide
code output link to see it). The membership matrix (columns with name p*)
is inferred by
clue::cl_consensus()
function with the SE method. Basically the value in the membership matrix
represents the probability to belong to a certain group. The finall class
label for an item is determined with the group with highest probability it
belongs to.
In get_classes() function, the entropy is calculated from the membership
matrix and the silhouette score is calculated from the consensus matrix.
cbind(get_classes(res, k = 2), get_membership(res, k = 2))
#> class entropy silhouette p1 p2
#> GSM97038 2 0.0000 0.931 0.000 1.000
#> GSM97045 2 0.0376 0.929 0.004 0.996
#> GSM97047 2 0.0000 0.931 0.000 1.000
#> GSM97025 2 0.0000 0.931 0.000 1.000
#> GSM97030 2 0.0000 0.931 0.000 1.000
#> GSM97027 2 0.0000 0.931 0.000 1.000
#> GSM97033 2 0.0000 0.931 0.000 1.000
#> GSM97034 2 0.0376 0.929 0.004 0.996
#> GSM97020 2 0.0000 0.931 0.000 1.000
#> GSM97026 2 0.3879 0.876 0.076 0.924
#> GSM97012 2 0.0000 0.931 0.000 1.000
#> GSM97015 2 0.4562 0.869 0.096 0.904
#> GSM97016 2 0.0000 0.931 0.000 1.000
#> GSM97017 1 0.0672 0.916 0.992 0.008
#> GSM97019 2 0.0000 0.931 0.000 1.000
#> GSM97022 2 0.0000 0.931 0.000 1.000
#> GSM97035 2 0.0000 0.931 0.000 1.000
#> GSM97036 2 0.4022 0.874 0.080 0.920
#> GSM97039 2 0.0000 0.931 0.000 1.000
#> GSM97046 2 0.0000 0.931 0.000 1.000
#> GSM97023 1 0.0000 0.916 1.000 0.000
#> GSM97029 2 0.7139 0.773 0.196 0.804
#> GSM97043 2 0.2043 0.912 0.032 0.968
#> GSM97013 2 0.9944 0.244 0.456 0.544
#> GSM96956 2 0.0000 0.931 0.000 1.000
#> GSM97024 2 0.0000 0.931 0.000 1.000
#> GSM97032 2 0.0000 0.931 0.000 1.000
#> GSM97044 2 0.0000 0.931 0.000 1.000
#> GSM97049 2 0.0000 0.931 0.000 1.000
#> GSM96968 1 0.3431 0.903 0.936 0.064
#> GSM96971 1 0.2948 0.909 0.948 0.052
#> GSM96986 1 0.5519 0.877 0.872 0.128
#> GSM97003 1 0.3879 0.901 0.924 0.076
#> GSM96957 1 0.0000 0.916 1.000 0.000
#> GSM96960 1 0.0000 0.916 1.000 0.000
#> GSM96975 1 0.2778 0.910 0.952 0.048
#> GSM96998 1 0.0000 0.916 1.000 0.000
#> GSM96999 1 0.0000 0.916 1.000 0.000
#> GSM97001 1 0.5059 0.886 0.888 0.112
#> GSM97005 1 0.3584 0.903 0.932 0.068
#> GSM97006 1 0.0000 0.916 1.000 0.000
#> GSM97021 1 0.3274 0.907 0.940 0.060
#> GSM97028 1 0.4298 0.881 0.912 0.088
#> GSM97031 1 0.3733 0.902 0.928 0.072
#> GSM97037 2 0.0000 0.931 0.000 1.000
#> GSM97018 2 0.0000 0.931 0.000 1.000
#> GSM97014 2 0.3431 0.885 0.064 0.936
#> GSM97042 2 0.0000 0.931 0.000 1.000
#> GSM97040 2 0.4815 0.847 0.104 0.896
#> GSM97041 1 0.6623 0.838 0.828 0.172
#> GSM96955 2 0.8267 0.628 0.260 0.740
#> GSM96990 2 0.0000 0.931 0.000 1.000
#> GSM96991 2 0.0000 0.931 0.000 1.000
#> GSM97048 2 0.0000 0.931 0.000 1.000
#> GSM96963 2 0.0000 0.931 0.000 1.000
#> GSM96953 2 0.0000 0.931 0.000 1.000
#> GSM96966 1 0.0000 0.916 1.000 0.000
#> GSM96979 1 0.5519 0.877 0.872 0.128
#> GSM96983 1 0.9580 0.503 0.620 0.380
#> GSM96984 1 0.6148 0.859 0.848 0.152
#> GSM96994 1 0.9954 0.260 0.540 0.460
#> GSM96996 1 0.5408 0.879 0.876 0.124
#> GSM96997 1 0.5519 0.877 0.872 0.128
#> GSM97007 1 0.7139 0.821 0.804 0.196
#> GSM96954 1 0.0376 0.916 0.996 0.004
#> GSM96962 1 0.5519 0.877 0.872 0.128
#> GSM96969 1 0.0000 0.916 1.000 0.000
#> GSM96970 1 0.0000 0.916 1.000 0.000
#> GSM96973 1 0.0000 0.916 1.000 0.000
#> GSM96976 2 0.9323 0.424 0.348 0.652
#> GSM96977 1 0.4562 0.885 0.904 0.096
#> GSM96995 1 0.8608 0.690 0.716 0.284
#> GSM97002 1 0.0000 0.916 1.000 0.000
#> GSM97009 2 0.3274 0.889 0.060 0.940
#> GSM97010 1 0.7950 0.760 0.760 0.240
#> GSM96974 1 0.7056 0.757 0.808 0.192
#> GSM96985 1 0.0000 0.916 1.000 0.000
#> GSM96959 2 0.9754 0.234 0.408 0.592
#> GSM96972 1 0.0000 0.916 1.000 0.000
#> GSM96978 1 0.8016 0.753 0.756 0.244
#> GSM96967 1 0.0000 0.916 1.000 0.000
#> GSM96987 1 0.0000 0.916 1.000 0.000
#> GSM97011 1 0.5519 0.877 0.872 0.128
#> GSM96964 1 0.0000 0.916 1.000 0.000
#> GSM96965 2 0.9552 0.351 0.376 0.624
#> GSM96981 1 0.4161 0.899 0.916 0.084
#> GSM96982 1 0.0000 0.916 1.000 0.000
#> GSM96988 1 0.0000 0.916 1.000 0.000
#> GSM97000 1 0.5519 0.877 0.872 0.128
#> GSM97004 1 0.0000 0.916 1.000 0.000
#> GSM97008 1 0.4022 0.900 0.920 0.080
#> GSM96950 1 0.4690 0.882 0.900 0.100
#> GSM96980 1 0.0000 0.916 1.000 0.000
#> GSM96989 1 0.0000 0.916 1.000 0.000
#> GSM96992 1 0.0000 0.916 1.000 0.000
#> GSM96993 1 0.8386 0.667 0.732 0.268
#> GSM96958 1 0.0000 0.916 1.000 0.000
#> GSM96951 1 0.0376 0.916 0.996 0.004
#> GSM96952 1 0.0000 0.916 1.000 0.000
#> GSM96961 1 0.0000 0.916 1.000 0.000
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM97038 2 0.4002 0.7394 0.000 0.840 0.160
#> GSM97045 2 0.0000 0.8499 0.000 1.000 0.000
#> GSM97047 3 0.6307 -0.0606 0.000 0.488 0.512
#> GSM97025 2 0.0000 0.8499 0.000 1.000 0.000
#> GSM97030 3 0.6079 0.3630 0.000 0.388 0.612
#> GSM97027 2 0.0000 0.8499 0.000 1.000 0.000
#> GSM97033 2 0.0000 0.8499 0.000 1.000 0.000
#> GSM97034 3 0.6192 0.2785 0.000 0.420 0.580
#> GSM97020 2 0.0000 0.8499 0.000 1.000 0.000
#> GSM97026 2 0.6979 0.6403 0.140 0.732 0.128
#> GSM97012 2 0.0000 0.8499 0.000 1.000 0.000
#> GSM97015 3 0.6000 0.6538 0.040 0.200 0.760
#> GSM97016 2 0.0237 0.8495 0.000 0.996 0.004
#> GSM97017 1 0.2689 0.8567 0.932 0.036 0.032
#> GSM97019 2 0.0747 0.8472 0.000 0.984 0.016
#> GSM97022 2 0.1031 0.8437 0.000 0.976 0.024
#> GSM97035 2 0.0000 0.8499 0.000 1.000 0.000
#> GSM97036 2 0.9334 -0.0968 0.164 0.428 0.408
#> GSM97039 2 0.0000 0.8499 0.000 1.000 0.000
#> GSM97046 2 0.1411 0.8343 0.000 0.964 0.036
#> GSM97023 1 0.0000 0.8760 1.000 0.000 0.000
#> GSM97029 2 0.7493 0.5946 0.136 0.696 0.168
#> GSM97043 2 0.1129 0.8426 0.020 0.976 0.004
#> GSM97013 2 0.5896 0.5434 0.292 0.700 0.008
#> GSM96956 2 0.5948 0.3575 0.000 0.640 0.360
#> GSM97024 2 0.3816 0.7391 0.000 0.852 0.148
#> GSM97032 3 0.4654 0.6417 0.000 0.208 0.792
#> GSM97044 3 0.4399 0.6615 0.000 0.188 0.812
#> GSM97049 2 0.0424 0.8483 0.000 0.992 0.008
#> GSM96968 3 0.5216 0.6510 0.260 0.000 0.740
#> GSM96971 3 0.2261 0.7453 0.068 0.000 0.932
#> GSM96986 3 0.1163 0.7454 0.028 0.000 0.972
#> GSM97003 1 0.4504 0.7669 0.804 0.000 0.196
#> GSM96957 1 0.2711 0.8360 0.912 0.000 0.088
#> GSM96960 1 0.0424 0.8755 0.992 0.000 0.008
#> GSM96975 1 0.3340 0.8295 0.880 0.000 0.120
#> GSM96998 1 0.0000 0.8760 1.000 0.000 0.000
#> GSM96999 1 0.0000 0.8760 1.000 0.000 0.000
#> GSM97001 1 0.5178 0.7101 0.744 0.000 0.256
#> GSM97005 1 0.3551 0.8132 0.868 0.000 0.132
#> GSM97006 1 0.0237 0.8764 0.996 0.000 0.004
#> GSM97021 1 0.4235 0.7825 0.824 0.000 0.176
#> GSM97028 3 0.4399 0.7145 0.188 0.000 0.812
#> GSM97031 1 0.4121 0.7862 0.832 0.000 0.168
#> GSM97037 2 0.5859 0.3911 0.000 0.656 0.344
#> GSM97018 3 0.4555 0.6474 0.000 0.200 0.800
#> GSM97014 2 0.5363 0.5999 0.000 0.724 0.276
#> GSM97042 2 0.1031 0.8437 0.000 0.976 0.024
#> GSM97040 3 0.6301 0.5232 0.028 0.260 0.712
#> GSM97041 1 0.7673 0.4898 0.652 0.260 0.088
#> GSM96955 3 0.6543 0.6732 0.076 0.176 0.748
#> GSM96990 3 0.3038 0.7202 0.000 0.104 0.896
#> GSM96991 2 0.1411 0.8372 0.000 0.964 0.036
#> GSM97048 2 0.0000 0.8499 0.000 1.000 0.000
#> GSM96963 2 0.0000 0.8499 0.000 1.000 0.000
#> GSM96953 2 0.0747 0.8472 0.000 0.984 0.016
#> GSM96966 1 0.1643 0.8664 0.956 0.000 0.044
#> GSM96979 3 0.1643 0.7468 0.044 0.000 0.956
#> GSM96983 3 0.3607 0.7401 0.112 0.008 0.880
#> GSM96984 3 0.0892 0.7438 0.020 0.000 0.980
#> GSM96994 3 0.0892 0.7438 0.020 0.000 0.980
#> GSM96996 1 0.6111 0.2871 0.604 0.000 0.396
#> GSM96997 1 0.4931 0.7319 0.768 0.000 0.232
#> GSM97007 3 0.1585 0.7443 0.028 0.008 0.964
#> GSM96954 1 0.6008 0.4327 0.628 0.000 0.372
#> GSM96962 3 0.4654 0.6385 0.208 0.000 0.792
#> GSM96969 1 0.1163 0.8719 0.972 0.000 0.028
#> GSM96970 1 0.1753 0.8711 0.952 0.000 0.048
#> GSM96973 1 0.1964 0.8656 0.944 0.000 0.056
#> GSM96976 3 0.3377 0.7240 0.012 0.092 0.896
#> GSM96977 3 0.6267 0.2647 0.452 0.000 0.548
#> GSM96995 3 0.0892 0.7438 0.020 0.000 0.980
#> GSM97002 1 0.0747 0.8745 0.984 0.000 0.016
#> GSM97009 2 0.6195 0.5703 0.020 0.704 0.276
#> GSM97010 3 0.9512 0.4233 0.248 0.260 0.492
#> GSM96974 3 0.5366 0.6840 0.208 0.016 0.776
#> GSM96985 1 0.6008 0.3313 0.628 0.000 0.372
#> GSM96959 3 0.5581 0.6573 0.036 0.176 0.788
#> GSM96972 1 0.0892 0.8701 0.980 0.000 0.020
#> GSM96978 3 0.2448 0.7488 0.076 0.000 0.924
#> GSM96967 1 0.0892 0.8701 0.980 0.000 0.020
#> GSM96987 1 0.0000 0.8760 1.000 0.000 0.000
#> GSM97011 3 0.6641 0.0530 0.448 0.008 0.544
#> GSM96964 1 0.0000 0.8760 1.000 0.000 0.000
#> GSM96965 2 0.8914 0.3434 0.164 0.556 0.280
#> GSM96981 1 0.2448 0.8570 0.924 0.000 0.076
#> GSM96982 1 0.0592 0.8753 0.988 0.000 0.012
#> GSM96988 3 0.6215 0.3608 0.428 0.000 0.572
#> GSM97000 3 0.6215 0.1410 0.428 0.000 0.572
#> GSM97004 1 0.0892 0.8701 0.980 0.000 0.020
#> GSM97008 1 0.4974 0.7252 0.764 0.000 0.236
#> GSM96950 1 0.6027 0.5506 0.712 0.016 0.272
#> GSM96980 1 0.1163 0.8719 0.972 0.000 0.028
#> GSM96989 1 0.0592 0.8746 0.988 0.000 0.012
#> GSM96992 1 0.0000 0.8760 1.000 0.000 0.000
#> GSM96993 3 0.5062 0.7097 0.184 0.016 0.800
#> GSM96958 1 0.2356 0.8571 0.928 0.000 0.072
#> GSM96951 1 0.3941 0.7964 0.844 0.000 0.156
#> GSM96952 1 0.0000 0.8760 1.000 0.000 0.000
#> GSM96961 1 0.0000 0.8760 1.000 0.000 0.000
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM97038 2 0.1938 0.8462 0.000 0.936 0.052 0.012
#> GSM97045 2 0.0336 0.8835 0.000 0.992 0.000 0.008
#> GSM97047 4 0.7472 0.3360 0.000 0.232 0.264 0.504
#> GSM97025 2 0.0188 0.8835 0.000 0.996 0.000 0.004
#> GSM97030 3 0.5360 0.0864 0.000 0.436 0.552 0.012
#> GSM97027 2 0.0188 0.8831 0.000 0.996 0.000 0.004
#> GSM97033 2 0.0188 0.8831 0.000 0.996 0.000 0.004
#> GSM97034 3 0.3810 0.5700 0.000 0.188 0.804 0.008
#> GSM97020 2 0.0188 0.8831 0.000 0.996 0.000 0.004
#> GSM97026 2 0.6736 0.4160 0.092 0.588 0.312 0.008
#> GSM97012 2 0.0707 0.8817 0.000 0.980 0.000 0.020
#> GSM97015 3 0.5166 0.4855 0.004 0.216 0.736 0.044
#> GSM97016 2 0.0188 0.8831 0.000 0.996 0.000 0.004
#> GSM97017 1 0.5488 0.1651 0.576 0.008 0.008 0.408
#> GSM97019 2 0.3335 0.8036 0.000 0.860 0.120 0.020
#> GSM97022 2 0.3708 0.7757 0.000 0.832 0.148 0.020
#> GSM97035 2 0.0707 0.8817 0.000 0.980 0.000 0.020
#> GSM97036 3 0.7279 0.3862 0.128 0.288 0.568 0.016
#> GSM97039 2 0.0188 0.8831 0.000 0.996 0.000 0.004
#> GSM97046 2 0.0524 0.8811 0.000 0.988 0.008 0.004
#> GSM97023 1 0.0000 0.7319 1.000 0.000 0.000 0.000
#> GSM97029 2 0.8725 0.1222 0.160 0.440 0.328 0.072
#> GSM97043 2 0.1229 0.8802 0.008 0.968 0.004 0.020
#> GSM97013 2 0.6421 0.2242 0.368 0.556 0.000 0.076
#> GSM96956 2 0.4756 0.6791 0.000 0.784 0.144 0.072
#> GSM97024 2 0.4610 0.6669 0.000 0.744 0.236 0.020
#> GSM97032 3 0.2401 0.6336 0.000 0.092 0.904 0.004
#> GSM97044 3 0.2654 0.6241 0.000 0.108 0.888 0.004
#> GSM97049 2 0.0469 0.8808 0.000 0.988 0.000 0.012
#> GSM96968 3 0.7577 0.1324 0.216 0.000 0.468 0.316
#> GSM96971 3 0.4988 0.3267 0.020 0.000 0.692 0.288
#> GSM96986 3 0.5085 0.3102 0.008 0.000 0.616 0.376
#> GSM97003 1 0.6224 0.4871 0.668 0.000 0.144 0.188
#> GSM96957 1 0.3529 0.6826 0.836 0.000 0.012 0.152
#> GSM96960 1 0.0927 0.7310 0.976 0.000 0.008 0.016
#> GSM96975 1 0.6483 0.1464 0.532 0.000 0.076 0.392
#> GSM96998 1 0.0469 0.7301 0.988 0.000 0.000 0.012
#> GSM96999 1 0.2814 0.6913 0.868 0.000 0.000 0.132
#> GSM97001 4 0.6324 0.3461 0.340 0.000 0.076 0.584
#> GSM97005 1 0.5972 0.4057 0.640 0.000 0.068 0.292
#> GSM97006 1 0.0188 0.7321 0.996 0.000 0.004 0.000
#> GSM97021 1 0.6326 0.4269 0.636 0.000 0.108 0.256
#> GSM97028 3 0.2174 0.6370 0.052 0.000 0.928 0.020
#> GSM97031 1 0.4231 0.6462 0.824 0.000 0.080 0.096
#> GSM97037 2 0.3351 0.7471 0.000 0.844 0.148 0.008
#> GSM97018 3 0.2053 0.6382 0.000 0.072 0.924 0.004
#> GSM97014 4 0.6918 0.2928 0.000 0.420 0.108 0.472
#> GSM97042 2 0.1042 0.8811 0.000 0.972 0.008 0.020
#> GSM97040 4 0.6929 0.2641 0.008 0.084 0.416 0.492
#> GSM97041 4 0.6785 0.4165 0.360 0.092 0.004 0.544
#> GSM96955 4 0.6766 0.3667 0.036 0.052 0.304 0.608
#> GSM96990 3 0.4982 0.5717 0.000 0.092 0.772 0.136
#> GSM96991 2 0.1042 0.8818 0.000 0.972 0.008 0.020
#> GSM97048 2 0.0188 0.8831 0.000 0.996 0.000 0.004
#> GSM96963 2 0.0707 0.8817 0.000 0.980 0.000 0.020
#> GSM96953 2 0.1022 0.8795 0.000 0.968 0.000 0.032
#> GSM96966 1 0.6575 0.4532 0.560 0.000 0.092 0.348
#> GSM96979 3 0.4328 0.5247 0.008 0.000 0.748 0.244
#> GSM96983 3 0.1975 0.6390 0.048 0.000 0.936 0.016
#> GSM96984 3 0.3982 0.5382 0.004 0.000 0.776 0.220
#> GSM96994 3 0.3528 0.5677 0.000 0.000 0.808 0.192
#> GSM96996 4 0.6895 0.4746 0.276 0.000 0.148 0.576
#> GSM96997 1 0.7357 0.1402 0.524 0.000 0.216 0.260
#> GSM97007 3 0.1302 0.6295 0.000 0.000 0.956 0.044
#> GSM96954 1 0.6682 0.2979 0.576 0.000 0.312 0.112
#> GSM96962 3 0.5292 0.5252 0.064 0.000 0.728 0.208
#> GSM96969 1 0.6746 0.4586 0.580 0.000 0.124 0.296
#> GSM96970 4 0.4989 -0.3621 0.472 0.000 0.000 0.528
#> GSM96973 1 0.6549 0.4519 0.556 0.000 0.088 0.356
#> GSM96976 4 0.3743 0.3157 0.000 0.016 0.160 0.824
#> GSM96977 4 0.7031 0.4591 0.200 0.000 0.224 0.576
#> GSM96995 3 0.4948 0.1400 0.000 0.000 0.560 0.440
#> GSM97002 1 0.2999 0.6927 0.864 0.000 0.004 0.132
#> GSM97009 4 0.6670 0.4269 0.004 0.304 0.100 0.592
#> GSM97010 4 0.7919 0.5121 0.148 0.112 0.132 0.608
#> GSM96974 3 0.4883 0.4704 0.016 0.000 0.696 0.288
#> GSM96985 3 0.7660 0.1346 0.324 0.000 0.448 0.228
#> GSM96959 4 0.5961 0.4042 0.004 0.052 0.308 0.636
#> GSM96972 1 0.4277 0.5623 0.720 0.000 0.000 0.280
#> GSM96978 3 0.1584 0.6409 0.012 0.000 0.952 0.036
#> GSM96967 1 0.6229 0.4942 0.628 0.000 0.088 0.284
#> GSM96987 1 0.0469 0.7301 0.988 0.000 0.000 0.012
#> GSM97011 4 0.5982 0.5239 0.112 0.000 0.204 0.684
#> GSM96964 1 0.1211 0.7265 0.960 0.000 0.000 0.040
#> GSM96965 4 0.3427 0.4406 0.028 0.112 0.000 0.860
#> GSM96981 4 0.5512 0.0192 0.488 0.000 0.016 0.496
#> GSM96982 1 0.1824 0.7237 0.936 0.000 0.004 0.060
#> GSM96988 3 0.3401 0.5860 0.152 0.000 0.840 0.008
#> GSM97000 4 0.6216 0.5186 0.120 0.000 0.220 0.660
#> GSM97004 1 0.0188 0.7317 0.996 0.000 0.000 0.004
#> GSM97008 4 0.7210 0.2920 0.360 0.000 0.148 0.492
#> GSM96950 1 0.7583 -0.2367 0.432 0.004 0.168 0.396
#> GSM96980 1 0.3074 0.6799 0.848 0.000 0.000 0.152
#> GSM96989 1 0.1174 0.7278 0.968 0.000 0.020 0.012
#> GSM96992 1 0.0000 0.7319 1.000 0.000 0.000 0.000
#> GSM96993 3 0.7179 0.3182 0.180 0.000 0.544 0.276
#> GSM96958 1 0.3404 0.7026 0.864 0.000 0.032 0.104
#> GSM96951 1 0.5226 0.5960 0.744 0.000 0.076 0.180
#> GSM96952 1 0.0000 0.7319 1.000 0.000 0.000 0.000
#> GSM96961 1 0.0000 0.7319 1.000 0.000 0.000 0.000
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM97038 2 0.0693 0.8779 0.000 0.980 0.008 0.000 0.012
#> GSM97045 2 0.1121 0.8850 0.000 0.956 0.000 0.044 0.000
#> GSM97047 5 0.4091 0.6182 0.000 0.076 0.124 0.004 0.796
#> GSM97025 2 0.1410 0.8842 0.000 0.940 0.000 0.060 0.000
#> GSM97030 2 0.5355 0.3246 0.000 0.536 0.420 0.012 0.032
#> GSM97027 2 0.0000 0.8829 0.000 1.000 0.000 0.000 0.000
#> GSM97033 2 0.0000 0.8829 0.000 1.000 0.000 0.000 0.000
#> GSM97034 3 0.2102 0.7102 0.000 0.068 0.916 0.004 0.012
#> GSM97020 2 0.0000 0.8829 0.000 1.000 0.000 0.000 0.000
#> GSM97026 2 0.6067 0.4115 0.096 0.560 0.332 0.004 0.008
#> GSM97012 2 0.2674 0.8728 0.000 0.856 0.000 0.140 0.004
#> GSM97015 3 0.5877 0.4745 0.008 0.236 0.632 0.004 0.120
#> GSM97016 2 0.0162 0.8827 0.000 0.996 0.000 0.000 0.004
#> GSM97017 5 0.4858 0.3200 0.424 0.008 0.000 0.012 0.556
#> GSM97019 2 0.4382 0.8255 0.000 0.772 0.084 0.140 0.004
#> GSM97022 2 0.4489 0.8189 0.000 0.764 0.092 0.140 0.004
#> GSM97035 2 0.2798 0.8732 0.000 0.852 0.000 0.140 0.008
#> GSM97036 3 0.6289 0.4973 0.164 0.216 0.600 0.000 0.020
#> GSM97039 2 0.0162 0.8827 0.000 0.996 0.000 0.000 0.004
#> GSM97046 2 0.0162 0.8827 0.000 0.996 0.000 0.000 0.004
#> GSM97023 1 0.0963 0.7789 0.964 0.000 0.000 0.036 0.000
#> GSM97029 3 0.7623 -0.0329 0.240 0.360 0.360 0.008 0.032
#> GSM97043 2 0.3078 0.8713 0.016 0.848 0.000 0.132 0.004
#> GSM97013 1 0.5162 0.3988 0.600 0.360 0.000 0.020 0.020
#> GSM96956 2 0.3202 0.7999 0.000 0.860 0.056 0.004 0.080
#> GSM97024 2 0.4641 0.7663 0.000 0.744 0.172 0.080 0.004
#> GSM97032 3 0.1329 0.7215 0.000 0.032 0.956 0.008 0.004
#> GSM97044 3 0.1830 0.7176 0.000 0.012 0.932 0.004 0.052
#> GSM97049 2 0.0162 0.8827 0.000 0.996 0.000 0.000 0.004
#> GSM96968 3 0.7438 0.2195 0.264 0.000 0.452 0.048 0.236
#> GSM96971 5 0.4891 0.0989 0.012 0.000 0.448 0.008 0.532
#> GSM96986 5 0.4808 -0.0241 0.000 0.000 0.400 0.024 0.576
#> GSM97003 5 0.6432 0.5247 0.256 0.000 0.120 0.036 0.588
#> GSM96957 1 0.3781 0.7169 0.828 0.000 0.016 0.048 0.108
#> GSM96960 1 0.2291 0.7579 0.908 0.000 0.056 0.036 0.000
#> GSM96975 5 0.6827 0.3887 0.316 0.000 0.084 0.072 0.528
#> GSM96998 1 0.0798 0.7801 0.976 0.000 0.000 0.016 0.008
#> GSM96999 1 0.5112 0.4554 0.664 0.000 0.004 0.064 0.268
#> GSM97001 5 0.2253 0.6818 0.036 0.000 0.016 0.028 0.920
#> GSM97005 5 0.3779 0.5822 0.236 0.000 0.000 0.012 0.752
#> GSM97006 1 0.1124 0.7784 0.960 0.000 0.004 0.036 0.000
#> GSM97021 5 0.4575 0.5931 0.212 0.000 0.040 0.012 0.736
#> GSM97028 3 0.0727 0.7215 0.012 0.000 0.980 0.004 0.004
#> GSM97031 1 0.4354 0.5470 0.712 0.000 0.000 0.032 0.256
#> GSM97037 2 0.2369 0.8325 0.000 0.908 0.056 0.004 0.032
#> GSM97018 3 0.1116 0.7223 0.000 0.028 0.964 0.004 0.004
#> GSM97014 5 0.3878 0.5802 0.000 0.236 0.000 0.016 0.748
#> GSM97042 2 0.2674 0.8728 0.000 0.856 0.000 0.140 0.004
#> GSM97040 5 0.2516 0.6243 0.000 0.000 0.140 0.000 0.860
#> GSM97041 5 0.4591 0.5015 0.332 0.012 0.000 0.008 0.648
#> GSM96955 5 0.6153 0.5085 0.000 0.072 0.228 0.064 0.636
#> GSM96990 3 0.4703 0.6382 0.000 0.096 0.744 0.004 0.156
#> GSM96991 2 0.3080 0.8712 0.000 0.844 0.008 0.140 0.008
#> GSM97048 2 0.0162 0.8827 0.000 0.996 0.000 0.000 0.004
#> GSM96963 2 0.2833 0.8723 0.000 0.852 0.004 0.140 0.004
#> GSM96953 2 0.2909 0.8727 0.000 0.848 0.000 0.140 0.012
#> GSM96966 4 0.3495 0.8436 0.160 0.000 0.000 0.812 0.028
#> GSM96979 3 0.4872 0.3159 0.000 0.000 0.540 0.024 0.436
#> GSM96983 3 0.1018 0.7217 0.000 0.000 0.968 0.016 0.016
#> GSM96984 3 0.4540 0.5150 0.000 0.000 0.656 0.024 0.320
#> GSM96994 3 0.3495 0.6433 0.000 0.000 0.812 0.028 0.160
#> GSM96996 5 0.6499 0.5177 0.248 0.000 0.100 0.056 0.596
#> GSM96997 5 0.7130 0.3159 0.156 0.000 0.260 0.060 0.524
#> GSM97007 3 0.2813 0.6956 0.000 0.000 0.868 0.024 0.108
#> GSM96954 1 0.6268 0.1448 0.484 0.000 0.156 0.000 0.360
#> GSM96962 3 0.4809 0.4964 0.008 0.000 0.648 0.024 0.320
#> GSM96969 4 0.3242 0.8426 0.216 0.000 0.000 0.784 0.000
#> GSM96970 4 0.3622 0.8354 0.136 0.000 0.000 0.816 0.048
#> GSM96973 4 0.3438 0.8502 0.172 0.000 0.000 0.808 0.020
#> GSM96976 4 0.3681 0.7093 0.000 0.008 0.036 0.820 0.136
#> GSM96977 5 0.4724 0.6463 0.152 0.000 0.064 0.024 0.760
#> GSM96995 5 0.4410 0.0442 0.000 0.000 0.440 0.004 0.556
#> GSM97002 1 0.5696 0.2337 0.604 0.000 0.044 0.032 0.320
#> GSM97009 5 0.3155 0.6511 0.000 0.128 0.016 0.008 0.848
#> GSM97010 5 0.5230 0.6401 0.164 0.056 0.008 0.036 0.736
#> GSM96974 4 0.3508 0.5948 0.000 0.000 0.252 0.748 0.000
#> GSM96985 3 0.6631 0.2996 0.160 0.000 0.568 0.240 0.032
#> GSM96959 5 0.1591 0.6617 0.000 0.004 0.052 0.004 0.940
#> GSM96972 4 0.3242 0.8426 0.216 0.000 0.000 0.784 0.000
#> GSM96978 3 0.0162 0.7222 0.000 0.000 0.996 0.000 0.004
#> GSM96967 4 0.3242 0.8426 0.216 0.000 0.000 0.784 0.000
#> GSM96987 1 0.0162 0.7777 0.996 0.000 0.000 0.000 0.004
#> GSM97011 5 0.1195 0.6689 0.000 0.000 0.028 0.012 0.960
#> GSM96964 1 0.1106 0.7719 0.964 0.000 0.000 0.012 0.024
#> GSM96965 4 0.4149 0.7063 0.000 0.080 0.004 0.792 0.124
#> GSM96981 5 0.5816 0.4162 0.320 0.000 0.020 0.068 0.592
#> GSM96982 1 0.2703 0.7703 0.896 0.000 0.024 0.060 0.020
#> GSM96988 3 0.1605 0.7188 0.040 0.000 0.944 0.012 0.004
#> GSM97000 5 0.0404 0.6639 0.000 0.000 0.012 0.000 0.988
#> GSM97004 1 0.1043 0.7777 0.960 0.000 0.000 0.040 0.000
#> GSM97008 5 0.2392 0.6730 0.104 0.000 0.004 0.004 0.888
#> GSM96950 1 0.5158 0.6109 0.748 0.004 0.108 0.032 0.108
#> GSM96980 4 0.3612 0.7858 0.268 0.000 0.000 0.732 0.000
#> GSM96989 1 0.0486 0.7788 0.988 0.000 0.004 0.004 0.004
#> GSM96992 1 0.0963 0.7789 0.964 0.000 0.000 0.036 0.000
#> GSM96993 1 0.7430 -0.1105 0.400 0.000 0.352 0.044 0.204
#> GSM96958 1 0.3130 0.7473 0.872 0.000 0.016 0.040 0.072
#> GSM96951 1 0.3967 0.5862 0.724 0.000 0.000 0.012 0.264
#> GSM96952 1 0.0963 0.7789 0.964 0.000 0.000 0.036 0.000
#> GSM96961 1 0.0963 0.7789 0.964 0.000 0.000 0.036 0.000
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM97038 6 0.4088 0.4249 0.000 0.436 0.004 0.000 0.004 0.556
#> GSM97045 2 0.3464 0.2949 0.000 0.688 0.000 0.000 0.000 0.312
#> GSM97047 5 0.3294 0.5567 0.000 0.032 0.128 0.004 0.828 0.008
#> GSM97025 2 0.3428 0.3143 0.000 0.696 0.000 0.000 0.000 0.304
#> GSM97030 3 0.5976 0.0907 0.000 0.364 0.508 0.004 0.040 0.084
#> GSM97027 2 0.3797 -0.0520 0.000 0.580 0.000 0.000 0.000 0.420
#> GSM97033 2 0.3804 -0.0668 0.000 0.576 0.000 0.000 0.000 0.424
#> GSM97034 3 0.1615 0.6643 0.000 0.064 0.928 0.000 0.004 0.004
#> GSM97020 2 0.3804 -0.0634 0.000 0.576 0.000 0.000 0.000 0.424
#> GSM97026 3 0.7353 -0.0121 0.140 0.336 0.344 0.000 0.000 0.180
#> GSM97012 2 0.0146 0.6983 0.000 0.996 0.000 0.000 0.000 0.004
#> GSM97015 3 0.6414 0.3963 0.004 0.096 0.568 0.000 0.216 0.116
#> GSM97016 6 0.3838 0.4284 0.000 0.448 0.000 0.000 0.000 0.552
#> GSM97017 5 0.6033 0.0970 0.436 0.020 0.000 0.020 0.444 0.080
#> GSM97019 2 0.0146 0.6985 0.000 0.996 0.004 0.000 0.000 0.000
#> GSM97022 2 0.0146 0.6985 0.000 0.996 0.004 0.000 0.000 0.000
#> GSM97035 2 0.2003 0.5912 0.000 0.884 0.000 0.000 0.000 0.116
#> GSM97036 3 0.8152 0.3152 0.196 0.156 0.436 0.032 0.024 0.156
#> GSM97039 6 0.3838 0.4284 0.000 0.448 0.000 0.000 0.000 0.552
#> GSM97046 6 0.3838 0.4284 0.000 0.448 0.000 0.000 0.000 0.552
#> GSM97023 1 0.0632 0.7559 0.976 0.000 0.000 0.024 0.000 0.000
#> GSM97029 3 0.7885 0.2030 0.128 0.240 0.348 0.004 0.016 0.264
#> GSM97043 2 0.1074 0.6855 0.012 0.960 0.000 0.000 0.000 0.028
#> GSM97013 6 0.4172 -0.2165 0.460 0.012 0.000 0.000 0.000 0.528
#> GSM96956 6 0.6074 0.3082 0.000 0.404 0.064 0.004 0.060 0.468
#> GSM97024 2 0.2416 0.5506 0.000 0.844 0.156 0.000 0.000 0.000
#> GSM97032 3 0.0837 0.6700 0.000 0.020 0.972 0.000 0.004 0.004
#> GSM97044 3 0.2568 0.6565 0.000 0.016 0.888 0.000 0.036 0.060
#> GSM97049 6 0.3838 0.4284 0.000 0.448 0.000 0.000 0.000 0.552
#> GSM96968 6 0.8222 -0.4214 0.136 0.000 0.264 0.068 0.168 0.364
#> GSM96971 5 0.4688 0.2892 0.012 0.000 0.352 0.020 0.608 0.008
#> GSM96986 5 0.6357 0.2149 0.004 0.000 0.216 0.072 0.568 0.140
#> GSM97003 5 0.7292 0.3906 0.284 0.000 0.084 0.028 0.456 0.148
#> GSM96957 1 0.4983 0.6389 0.676 0.000 0.000 0.068 0.032 0.224
#> GSM96960 1 0.3129 0.6983 0.820 0.000 0.004 0.024 0.000 0.152
#> GSM96975 5 0.7640 0.1239 0.332 0.000 0.040 0.072 0.364 0.192
#> GSM96998 1 0.2346 0.7316 0.868 0.000 0.000 0.008 0.000 0.124
#> GSM96999 1 0.5268 0.4428 0.656 0.000 0.000 0.060 0.228 0.056
#> GSM97001 5 0.5057 0.5673 0.088 0.000 0.000 0.040 0.692 0.180
#> GSM97005 5 0.3298 0.5245 0.236 0.000 0.000 0.008 0.756 0.000
#> GSM97006 1 0.1294 0.7541 0.956 0.000 0.004 0.024 0.008 0.008
#> GSM97021 5 0.4043 0.5336 0.212 0.000 0.036 0.012 0.740 0.000
#> GSM97028 3 0.0603 0.6676 0.004 0.000 0.980 0.000 0.000 0.016
#> GSM97031 1 0.4547 0.3312 0.628 0.000 0.000 0.020 0.332 0.020
#> GSM97037 6 0.5667 0.3401 0.000 0.412 0.060 0.004 0.032 0.492
#> GSM97018 3 0.0603 0.6697 0.000 0.016 0.980 0.000 0.000 0.004
#> GSM97014 5 0.4293 0.4884 0.000 0.032 0.000 0.016 0.704 0.248
#> GSM97042 2 0.0000 0.6985 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97040 5 0.1663 0.5746 0.000 0.000 0.088 0.000 0.912 0.000
#> GSM97041 5 0.5511 0.3149 0.352 0.016 0.000 0.004 0.548 0.080
#> GSM96955 5 0.7287 0.4390 0.000 0.100 0.116 0.060 0.528 0.196
#> GSM96990 3 0.4356 0.5847 0.000 0.024 0.764 0.004 0.132 0.076
#> GSM96991 2 0.0260 0.6962 0.000 0.992 0.008 0.000 0.000 0.000
#> GSM97048 6 0.3838 0.4284 0.000 0.448 0.000 0.000 0.000 0.552
#> GSM96963 2 0.0146 0.6975 0.000 0.996 0.000 0.000 0.000 0.004
#> GSM96953 2 0.2135 0.5761 0.000 0.872 0.000 0.000 0.000 0.128
#> GSM96966 4 0.2006 0.8745 0.080 0.000 0.000 0.904 0.016 0.000
#> GSM96979 5 0.6270 0.1256 0.000 0.000 0.280 0.068 0.536 0.116
#> GSM96983 3 0.1991 0.6632 0.000 0.000 0.920 0.044 0.012 0.024
#> GSM96984 3 0.6653 0.1781 0.000 0.000 0.432 0.068 0.356 0.144
#> GSM96994 3 0.3843 0.6048 0.000 0.000 0.804 0.068 0.100 0.028
#> GSM96996 5 0.7105 0.3138 0.292 0.000 0.100 0.076 0.488 0.044
#> GSM96997 5 0.7504 0.3851 0.168 0.000 0.108 0.076 0.516 0.132
#> GSM97007 3 0.4683 0.5941 0.000 0.000 0.744 0.060 0.076 0.120
#> GSM96954 5 0.6860 0.1929 0.348 0.000 0.132 0.000 0.420 0.100
#> GSM96962 3 0.6370 0.2641 0.000 0.000 0.504 0.068 0.312 0.116
#> GSM96969 4 0.2219 0.8733 0.136 0.000 0.000 0.864 0.000 0.000
#> GSM96970 4 0.2066 0.8583 0.040 0.000 0.000 0.908 0.052 0.000
#> GSM96973 4 0.1753 0.8802 0.084 0.000 0.000 0.912 0.004 0.000
#> GSM96976 4 0.1667 0.8250 0.000 0.008 0.008 0.940 0.032 0.012
#> GSM96977 5 0.6310 0.5236 0.064 0.000 0.056 0.036 0.572 0.272
#> GSM96995 5 0.5287 -0.0233 0.000 0.000 0.448 0.024 0.480 0.048
#> GSM97002 1 0.6660 0.3479 0.532 0.000 0.016 0.048 0.192 0.212
#> GSM97009 5 0.3494 0.5928 0.000 0.036 0.000 0.004 0.792 0.168
#> GSM97010 5 0.5935 0.4658 0.196 0.000 0.000 0.028 0.572 0.204
#> GSM96974 4 0.2595 0.7500 0.000 0.000 0.160 0.836 0.000 0.004
#> GSM96985 3 0.5409 0.2790 0.068 0.000 0.572 0.336 0.020 0.004
#> GSM96959 5 0.2034 0.5817 0.000 0.000 0.060 0.004 0.912 0.024
#> GSM96972 4 0.2219 0.8733 0.136 0.000 0.000 0.864 0.000 0.000
#> GSM96978 3 0.2056 0.6526 0.000 0.000 0.904 0.012 0.004 0.080
#> GSM96967 4 0.2219 0.8733 0.136 0.000 0.000 0.864 0.000 0.000
#> GSM96987 1 0.2048 0.7283 0.880 0.000 0.000 0.000 0.000 0.120
#> GSM97011 5 0.1866 0.6046 0.000 0.000 0.000 0.008 0.908 0.084
#> GSM96964 1 0.2976 0.7195 0.844 0.000 0.000 0.020 0.012 0.124
#> GSM96965 4 0.3447 0.7717 0.000 0.008 0.000 0.820 0.064 0.108
#> GSM96981 5 0.6901 0.2320 0.288 0.000 0.004 0.076 0.460 0.172
#> GSM96982 1 0.4434 0.6755 0.748 0.000 0.008 0.064 0.016 0.164
#> GSM96988 3 0.2332 0.6653 0.020 0.000 0.904 0.040 0.000 0.036
#> GSM97000 5 0.0000 0.5920 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM97004 1 0.1257 0.7555 0.952 0.000 0.000 0.028 0.000 0.020
#> GSM97008 5 0.1531 0.6085 0.068 0.000 0.000 0.004 0.928 0.000
#> GSM96950 1 0.6005 0.6162 0.636 0.004 0.040 0.044 0.060 0.216
#> GSM96980 4 0.3428 0.6704 0.304 0.000 0.000 0.696 0.000 0.000
#> GSM96989 1 0.2333 0.7292 0.872 0.000 0.004 0.004 0.000 0.120
#> GSM96992 1 0.0632 0.7559 0.976 0.000 0.000 0.024 0.000 0.000
#> GSM96993 1 0.7905 0.1669 0.416 0.000 0.244 0.048 0.140 0.152
#> GSM96958 1 0.4434 0.6692 0.740 0.000 0.000 0.060 0.028 0.172
#> GSM96951 1 0.3668 0.5540 0.728 0.000 0.000 0.008 0.256 0.008
#> GSM96952 1 0.0777 0.7562 0.972 0.000 0.000 0.024 0.000 0.004
#> GSM96961 1 0.0632 0.7559 0.976 0.000 0.000 0.024 0.000 0.000
Heatmaps for the consensus matrix. It visualizes the probability of two samples to be in a same group.
consensus_heatmap(res, k = 2)
consensus_heatmap(res, k = 3)
consensus_heatmap(res, k = 4)
consensus_heatmap(res, k = 5)
consensus_heatmap(res, k = 6)
Heatmaps for the membership of samples in all partitions to see how consistent they are:
membership_heatmap(res, k = 2)
membership_heatmap(res, k = 3)
membership_heatmap(res, k = 4)
membership_heatmap(res, k = 5)
membership_heatmap(res, k = 6)
As soon as we have had the classes for columns, we can look for signatures which are significantly different between classes which can be candidate marks for certain classes. Following are the heatmaps for signatures.
Signature heatmaps where rows are scaled:
get_signatures(res, k = 2)
get_signatures(res, k = 3)
get_signatures(res, k = 4)
get_signatures(res, k = 5)
get_signatures(res, k = 6)
Signature heatmaps where rows are not scaled:
get_signatures(res, k = 2, scale_rows = FALSE)
get_signatures(res, k = 3, scale_rows = FALSE)
get_signatures(res, k = 4, scale_rows = FALSE)
get_signatures(res, k = 5, scale_rows = FALSE)
get_signatures(res, k = 6, scale_rows = FALSE)
Compare the overlap of signatures from different k:
compare_signatures(res)
get_signature() returns a data frame invisibly. TO get the list of signatures, the function
call should be assigned to a variable explicitly. In following code, if plot argument is set
to FALSE, no heatmap is plotted while only the differential analysis is performed.
# code only for demonstration
tb = get_signature(res, k = ..., plot = FALSE)
An example of the output of tb is:
#> which_row fdr mean_1 mean_2 scaled_mean_1 scaled_mean_2 km
#> 1 38 0.042760348 8.373488 9.131774 -0.5533452 0.5164555 1
#> 2 40 0.018707592 7.106213 8.469186 -0.6173731 0.5762149 1
#> 3 55 0.019134737 10.221463 11.207825 -0.6159697 0.5749050 1
#> 4 59 0.006059896 5.921854 7.869574 -0.6899429 0.6439467 1
#> 5 60 0.018055526 8.928898 10.211722 -0.6204761 0.5791110 1
#> 6 98 0.009384629 15.714769 14.887706 0.6635654 -0.6193277 2
...
The columns in tb are:
which_row: row indices corresponding to the input matrix.fdr: FDR for the differential test. mean_x: The mean value in group x.scaled_mean_x: The mean value in group x after rows are scaled.km: Row groups if k-means clustering is applied to rows.UMAP plot which shows how samples are separated.
dimension_reduction(res, k = 2, method = "UMAP")
dimension_reduction(res, k = 3, method = "UMAP")
dimension_reduction(res, k = 4, method = "UMAP")
dimension_reduction(res, k = 5, method = "UMAP")
dimension_reduction(res, k = 6, method = "UMAP")
Following heatmap shows how subgroups are split when increasing k:
collect_classes(res)
Test correlation between subgroups and known annotations. If the known annotation is numeric, one-way ANOVA test is applied, and if the known annotation is discrete, chi-squared contingency table test is applied.
test_to_known_factors(res)
#> n disease.state(p) specimen(p) cell.type(p) other(p) k
#> CV:pam 95 1.87e-09 0.519 9.94e-19 0.0483 2
#> CV:pam 84 3.06e-07 0.563 3.99e-15 0.1185 3
#> CV:pam 60 7.60e-04 0.594 2.81e-12 0.2431 4
#> CV:pam 79 3.36e-05 0.431 7.14e-15 0.1530 5
#> CV:pam 58 3.68e-02 0.497 2.34e-10 0.1339 6
If matrix rows can be associated to genes, consider to use functional_enrichment(res,
...) to perform function enrichment for the signature genes. See this vignette for more detailed explanations.
The object with results only for a single top-value method and a single partition method can be extracted as:
res = res_list["CV", "mclust"]
# you can also extract it by
# res = res_list["CV:mclust"]
A summary of res and all the functions that can be applied to it:
res
#> A 'ConsensusPartition' object with k = 2, 3, 4, 5, 6.
#> On a matrix with 21168 rows and 100 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 0.801 0.915 0.954 0.3505 0.665 0.665
#> 3 3 0.469 0.437 0.711 0.7263 0.792 0.692
#> 4 4 0.910 0.904 0.960 0.2091 0.698 0.422
#> 5 5 0.805 0.705 0.834 0.0585 0.956 0.846
#> 6 6 0.772 0.771 0.787 0.0460 0.874 0.545
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
#> GSM97038 2 0.0000 0.9497 0.000 1.000
#> GSM97045 2 0.0376 0.9466 0.004 0.996
#> GSM97047 1 0.6438 0.8536 0.836 0.164
#> GSM97025 2 0.0000 0.9497 0.000 1.000
#> GSM97030 1 0.6247 0.8620 0.844 0.156
#> GSM97027 2 0.0000 0.9497 0.000 1.000
#> GSM97033 2 0.0000 0.9497 0.000 1.000
#> GSM97034 1 0.5629 0.8855 0.868 0.132
#> GSM97020 2 0.0000 0.9497 0.000 1.000
#> GSM97026 1 0.5629 0.8855 0.868 0.132
#> GSM97012 2 0.0000 0.9497 0.000 1.000
#> GSM97015 1 0.5519 0.8890 0.872 0.128
#> GSM97016 2 0.0000 0.9497 0.000 1.000
#> GSM97017 1 0.0000 0.9504 1.000 0.000
#> GSM97019 2 0.0000 0.9497 0.000 1.000
#> GSM97022 2 0.0000 0.9497 0.000 1.000
#> GSM97035 2 0.0000 0.9497 0.000 1.000
#> GSM97036 1 0.0000 0.9504 1.000 0.000
#> GSM97039 2 0.0000 0.9497 0.000 1.000
#> GSM97046 2 0.0000 0.9497 0.000 1.000
#> GSM97023 1 0.0000 0.9504 1.000 0.000
#> GSM97029 1 0.0000 0.9504 1.000 0.000
#> GSM97043 2 0.0000 0.9497 0.000 1.000
#> GSM97013 1 0.0000 0.9504 1.000 0.000
#> GSM96956 2 0.9963 0.0253 0.464 0.536
#> GSM97024 2 0.7056 0.7353 0.192 0.808
#> GSM97032 1 0.6531 0.8488 0.832 0.168
#> GSM97044 1 0.5629 0.8855 0.868 0.132
#> GSM97049 2 0.0000 0.9497 0.000 1.000
#> GSM96968 1 0.4022 0.9233 0.920 0.080
#> GSM96971 1 0.4022 0.9233 0.920 0.080
#> GSM96986 1 0.4022 0.9233 0.920 0.080
#> GSM97003 1 0.0000 0.9504 1.000 0.000
#> GSM96957 1 0.0000 0.9504 1.000 0.000
#> GSM96960 1 0.0000 0.9504 1.000 0.000
#> GSM96975 1 0.0000 0.9504 1.000 0.000
#> GSM96998 1 0.0000 0.9504 1.000 0.000
#> GSM96999 1 0.0000 0.9504 1.000 0.000
#> GSM97001 1 0.0000 0.9504 1.000 0.000
#> GSM97005 1 0.0000 0.9504 1.000 0.000
#> GSM97006 1 0.0000 0.9504 1.000 0.000
#> GSM97021 1 0.0000 0.9504 1.000 0.000
#> GSM97028 1 0.4022 0.9233 0.920 0.080
#> GSM97031 1 0.0000 0.9504 1.000 0.000
#> GSM97037 2 0.8081 0.6419 0.248 0.752
#> GSM97018 1 0.5737 0.8818 0.864 0.136
#> GSM97014 1 0.4815 0.9078 0.896 0.104
#> GSM97042 2 0.0000 0.9497 0.000 1.000
#> GSM97040 1 0.0672 0.9486 0.992 0.008
#> GSM97041 1 0.0000 0.9504 1.000 0.000
#> GSM96955 1 0.7815 0.7634 0.768 0.232
#> GSM96990 1 0.5946 0.8745 0.856 0.144
#> GSM96991 1 0.8327 0.7120 0.736 0.264
#> GSM97048 2 0.0000 0.9497 0.000 1.000
#> GSM96963 1 0.9129 0.5899 0.672 0.328
#> GSM96953 2 0.0000 0.9497 0.000 1.000
#> GSM96966 1 0.0000 0.9504 1.000 0.000
#> GSM96979 1 0.4022 0.9233 0.920 0.080
#> GSM96983 1 0.4690 0.9108 0.900 0.100
#> GSM96984 1 0.4022 0.9233 0.920 0.080
#> GSM96994 1 0.4022 0.9233 0.920 0.080
#> GSM96996 1 0.0000 0.9504 1.000 0.000
#> GSM96997 1 0.4022 0.9233 0.920 0.080
#> GSM97007 1 0.4022 0.9233 0.920 0.080
#> GSM96954 1 0.4022 0.9233 0.920 0.080
#> GSM96962 1 0.4022 0.9233 0.920 0.080
#> GSM96969 1 0.0000 0.9504 1.000 0.000
#> GSM96970 1 0.0000 0.9504 1.000 0.000
#> GSM96973 1 0.0000 0.9504 1.000 0.000
#> GSM96976 1 0.0938 0.9476 0.988 0.012
#> GSM96977 1 0.0376 0.9495 0.996 0.004
#> GSM96995 1 0.4161 0.9211 0.916 0.084
#> GSM97002 1 0.0000 0.9504 1.000 0.000
#> GSM97009 1 0.4562 0.9135 0.904 0.096
#> GSM97010 1 0.0000 0.9504 1.000 0.000
#> GSM96974 1 0.0672 0.9485 0.992 0.008
#> GSM96985 1 0.0672 0.9485 0.992 0.008
#> GSM96959 1 0.5178 0.8989 0.884 0.116
#> GSM96972 1 0.0000 0.9504 1.000 0.000
#> GSM96978 1 0.4022 0.9233 0.920 0.080
#> GSM96967 1 0.0000 0.9504 1.000 0.000
#> GSM96987 1 0.0000 0.9504 1.000 0.000
#> GSM97011 1 0.0000 0.9504 1.000 0.000
#> GSM96964 1 0.0000 0.9504 1.000 0.000
#> GSM96965 1 0.0000 0.9504 1.000 0.000
#> GSM96981 1 0.0000 0.9504 1.000 0.000
#> GSM96982 1 0.0000 0.9504 1.000 0.000
#> GSM96988 1 0.4022 0.9233 0.920 0.080
#> GSM97000 1 0.0000 0.9504 1.000 0.000
#> GSM97004 1 0.0000 0.9504 1.000 0.000
#> GSM97008 1 0.0000 0.9504 1.000 0.000
#> GSM96950 1 0.0000 0.9504 1.000 0.000
#> GSM96980 1 0.0000 0.9504 1.000 0.000
#> GSM96989 1 0.0000 0.9504 1.000 0.000
#> GSM96992 1 0.0000 0.9504 1.000 0.000
#> GSM96993 1 0.0000 0.9504 1.000 0.000
#> GSM96958 1 0.0000 0.9504 1.000 0.000
#> GSM96951 1 0.0000 0.9504 1.000 0.000
#> GSM96952 1 0.0000 0.9504 1.000 0.000
#> GSM96961 1 0.0000 0.9504 1.000 0.000
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM97038 2 0.6079 0.9095 0.000 0.612 0.388
#> GSM97045 2 0.6079 0.9095 0.000 0.612 0.388
#> GSM97047 1 0.8637 -0.3868 0.588 0.260 0.152
#> GSM97025 2 0.6079 0.9095 0.000 0.612 0.388
#> GSM97030 3 0.8069 0.9165 0.460 0.064 0.476
#> GSM97027 2 0.6079 0.9095 0.000 0.612 0.388
#> GSM97033 2 0.6079 0.9095 0.000 0.612 0.388
#> GSM97034 3 0.6683 0.9448 0.492 0.008 0.500
#> GSM97020 2 0.6079 0.9095 0.000 0.612 0.388
#> GSM97026 1 0.7208 -0.1066 0.644 0.308 0.048
#> GSM97012 2 0.6079 0.9095 0.000 0.612 0.388
#> GSM97015 1 0.6309 -0.9451 0.500 0.000 0.500
#> GSM97016 2 0.6079 0.9095 0.000 0.612 0.388
#> GSM97017 1 0.6282 0.5669 0.612 0.384 0.004
#> GSM97019 2 0.6079 0.9095 0.000 0.612 0.388
#> GSM97022 2 0.6079 0.9095 0.000 0.612 0.388
#> GSM97035 2 0.6079 0.9095 0.000 0.612 0.388
#> GSM97036 1 0.2584 0.4629 0.928 0.064 0.008
#> GSM97039 2 0.6079 0.9095 0.000 0.612 0.388
#> GSM97046 2 0.6079 0.9095 0.000 0.612 0.388
#> GSM97023 1 0.6282 0.5663 0.612 0.384 0.004
#> GSM97029 1 0.6451 0.5668 0.608 0.384 0.008
#> GSM97043 2 0.6079 0.9095 0.000 0.612 0.388
#> GSM97013 1 0.6282 0.5669 0.612 0.384 0.004
#> GSM96956 2 0.9234 0.1101 0.364 0.476 0.160
#> GSM97024 2 0.8130 0.8165 0.072 0.528 0.400
#> GSM97032 3 0.8138 0.9033 0.452 0.068 0.480
#> GSM97044 3 0.7487 0.9374 0.464 0.036 0.500
#> GSM97049 2 0.6079 0.9095 0.000 0.612 0.388
#> GSM96968 1 0.6302 -0.9128 0.520 0.000 0.480
#> GSM96971 1 0.6280 -0.8740 0.540 0.000 0.460
#> GSM96986 3 0.6309 0.9440 0.496 0.000 0.504
#> GSM97003 1 0.1860 0.3479 0.948 0.000 0.052
#> GSM96957 1 0.7505 0.5560 0.572 0.384 0.044
#> GSM96960 1 0.0747 0.4139 0.984 0.000 0.016
#> GSM96975 1 0.5291 0.5518 0.732 0.268 0.000
#> GSM96998 1 0.6865 0.5619 0.596 0.384 0.020
#> GSM96999 1 0.6600 0.5658 0.604 0.384 0.012
#> GSM97001 1 0.7311 0.5594 0.580 0.384 0.036
#> GSM97005 1 0.7311 0.5589 0.580 0.384 0.036
#> GSM97006 1 0.3091 0.4735 0.912 0.072 0.016
#> GSM97021 1 0.7932 0.5469 0.552 0.384 0.064
#> GSM97028 3 0.6309 0.9396 0.500 0.000 0.500
#> GSM97031 1 0.2982 0.3756 0.920 0.024 0.056
#> GSM97037 2 0.8708 0.7198 0.108 0.488 0.404
#> GSM97018 3 0.7759 0.9281 0.476 0.048 0.476
#> GSM97014 1 0.5986 0.2469 0.736 0.240 0.024
#> GSM97042 2 0.6079 0.9095 0.000 0.612 0.388
#> GSM97040 1 0.8297 0.5139 0.560 0.348 0.092
#> GSM97041 1 0.6451 0.5665 0.608 0.384 0.008
#> GSM96955 1 0.7184 -0.2627 0.504 0.472 0.024
#> GSM96990 3 0.8277 0.9031 0.460 0.076 0.464
#> GSM96991 2 0.7029 0.0191 0.440 0.540 0.020
#> GSM97048 2 0.6079 0.9095 0.000 0.612 0.388
#> GSM96963 2 0.6786 0.0047 0.448 0.540 0.012
#> GSM96953 2 0.6079 0.9095 0.000 0.612 0.388
#> GSM96966 1 0.3112 0.4011 0.900 0.004 0.096
#> GSM96979 1 0.5529 -0.4157 0.704 0.000 0.296
#> GSM96983 3 0.7487 0.9374 0.464 0.036 0.500
#> GSM96984 3 0.6309 0.9440 0.496 0.000 0.504
#> GSM96994 3 0.6309 0.9440 0.496 0.000 0.504
#> GSM96996 1 0.6814 0.5655 0.608 0.372 0.020
#> GSM96997 1 0.6260 -0.8442 0.552 0.000 0.448
#> GSM97007 3 0.6309 0.9440 0.496 0.000 0.504
#> GSM96954 1 0.6215 -0.8039 0.572 0.000 0.428
#> GSM96962 3 0.6309 0.9440 0.496 0.000 0.504
#> GSM96969 1 0.3425 0.3959 0.884 0.004 0.112
#> GSM96970 1 0.3425 0.3959 0.884 0.004 0.112
#> GSM96973 1 0.3425 0.3959 0.884 0.004 0.112
#> GSM96976 1 0.6276 -0.1404 0.736 0.040 0.224
#> GSM96977 1 0.3910 0.2756 0.876 0.020 0.104
#> GSM96995 1 0.6307 -0.9266 0.512 0.000 0.488
#> GSM97002 1 0.1411 0.4142 0.964 0.000 0.036
#> GSM97009 1 0.6981 -0.1583 0.732 0.132 0.136
#> GSM97010 1 0.0747 0.3896 0.984 0.000 0.016
#> GSM96974 1 0.5138 -0.1566 0.748 0.000 0.252
#> GSM96985 1 0.4178 0.1414 0.828 0.000 0.172
#> GSM96959 1 0.7054 -0.9031 0.524 0.020 0.456
#> GSM96972 1 0.3425 0.3959 0.884 0.004 0.112
#> GSM96978 1 0.6302 -0.9125 0.520 0.000 0.480
#> GSM96967 1 0.3425 0.3959 0.884 0.004 0.112
#> GSM96987 1 0.6773 0.5647 0.636 0.340 0.024
#> GSM97011 1 0.7311 0.5598 0.580 0.384 0.036
#> GSM96964 1 0.6451 0.5654 0.608 0.384 0.008
#> GSM96965 1 0.1753 0.3992 0.952 0.000 0.048
#> GSM96981 1 0.1170 0.4218 0.976 0.008 0.016
#> GSM96982 1 0.2066 0.4086 0.940 0.000 0.060
#> GSM96988 1 0.5882 -0.5823 0.652 0.000 0.348
#> GSM97000 1 0.6325 0.3799 0.772 0.112 0.116
#> GSM97004 1 0.2356 0.4076 0.928 0.000 0.072
#> GSM97008 1 0.7932 0.5469 0.552 0.384 0.064
#> GSM96950 1 0.6062 0.5667 0.616 0.384 0.000
#> GSM96980 1 0.3425 0.3959 0.884 0.004 0.112
#> GSM96989 1 0.6600 0.5646 0.604 0.384 0.012
#> GSM96992 1 0.6721 0.5645 0.604 0.380 0.016
#> GSM96993 1 0.6282 0.5669 0.612 0.384 0.004
#> GSM96958 1 0.6062 0.5667 0.616 0.384 0.000
#> GSM96951 1 0.6737 0.5649 0.600 0.384 0.016
#> GSM96952 1 0.6721 0.5645 0.604 0.380 0.016
#> GSM96961 1 0.6282 0.5663 0.612 0.384 0.004
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM97038 2 0.0000 0.9557 0.000 1.000 0.000 0.000
#> GSM97045 2 0.0000 0.9557 0.000 1.000 0.000 0.000
#> GSM97047 2 0.0895 0.9330 0.020 0.976 0.004 0.000
#> GSM97025 2 0.0000 0.9557 0.000 1.000 0.000 0.000
#> GSM97030 3 0.0707 0.9406 0.000 0.020 0.980 0.000
#> GSM97027 2 0.0000 0.9557 0.000 1.000 0.000 0.000
#> GSM97033 2 0.0000 0.9557 0.000 1.000 0.000 0.000
#> GSM97034 3 0.0000 0.9561 0.000 0.000 1.000 0.000
#> GSM97020 2 0.0000 0.9557 0.000 1.000 0.000 0.000
#> GSM97026 2 0.0000 0.9557 0.000 1.000 0.000 0.000
#> GSM97012 2 0.0000 0.9557 0.000 1.000 0.000 0.000
#> GSM97015 3 0.0336 0.9508 0.000 0.008 0.992 0.000
#> GSM97016 2 0.0000 0.9557 0.000 1.000 0.000 0.000
#> GSM97017 1 0.0000 0.9594 1.000 0.000 0.000 0.000
#> GSM97019 2 0.0000 0.9557 0.000 1.000 0.000 0.000
#> GSM97022 2 0.0000 0.9557 0.000 1.000 0.000 0.000
#> GSM97035 2 0.0000 0.9557 0.000 1.000 0.000 0.000
#> GSM97036 1 0.0469 0.9544 0.988 0.000 0.000 0.012
#> GSM97039 2 0.0000 0.9557 0.000 1.000 0.000 0.000
#> GSM97046 2 0.0000 0.9557 0.000 1.000 0.000 0.000
#> GSM97023 1 0.0000 0.9594 1.000 0.000 0.000 0.000
#> GSM97029 1 0.0000 0.9594 1.000 0.000 0.000 0.000
#> GSM97043 2 0.0000 0.9557 0.000 1.000 0.000 0.000
#> GSM97013 1 0.0000 0.9594 1.000 0.000 0.000 0.000
#> GSM96956 2 0.0000 0.9557 0.000 1.000 0.000 0.000
#> GSM97024 2 0.0188 0.9525 0.000 0.996 0.004 0.000
#> GSM97032 3 0.4877 0.3111 0.000 0.408 0.592 0.000
#> GSM97044 3 0.0000 0.9561 0.000 0.000 1.000 0.000
#> GSM97049 2 0.0000 0.9557 0.000 1.000 0.000 0.000
#> GSM96968 3 0.0188 0.9534 0.004 0.000 0.996 0.000
#> GSM96971 3 0.0000 0.9561 0.000 0.000 1.000 0.000
#> GSM96986 3 0.0000 0.9561 0.000 0.000 1.000 0.000
#> GSM97003 1 0.0804 0.9514 0.980 0.000 0.012 0.008
#> GSM96957 1 0.0000 0.9594 1.000 0.000 0.000 0.000
#> GSM96960 1 0.3311 0.8284 0.828 0.000 0.000 0.172
#> GSM96975 1 0.0000 0.9594 1.000 0.000 0.000 0.000
#> GSM96998 1 0.2647 0.8828 0.880 0.000 0.000 0.120
#> GSM96999 1 0.0000 0.9594 1.000 0.000 0.000 0.000
#> GSM97001 1 0.0000 0.9594 1.000 0.000 0.000 0.000
#> GSM97005 1 0.0000 0.9594 1.000 0.000 0.000 0.000
#> GSM97006 1 0.3074 0.8509 0.848 0.000 0.000 0.152
#> GSM97021 1 0.0000 0.9594 1.000 0.000 0.000 0.000
#> GSM97028 3 0.0000 0.9561 0.000 0.000 1.000 0.000
#> GSM97031 1 0.0188 0.9574 0.996 0.000 0.004 0.000
#> GSM97037 2 0.0188 0.9525 0.000 0.996 0.004 0.000
#> GSM97018 2 0.5000 -0.0453 0.000 0.504 0.496 0.000
#> GSM97014 2 0.5000 0.0336 0.496 0.504 0.000 0.000
#> GSM97042 2 0.0000 0.9557 0.000 1.000 0.000 0.000
#> GSM97040 1 0.0000 0.9594 1.000 0.000 0.000 0.000
#> GSM97041 1 0.0000 0.9594 1.000 0.000 0.000 0.000
#> GSM96955 2 0.0000 0.9557 0.000 1.000 0.000 0.000
#> GSM96990 3 0.3688 0.7289 0.000 0.208 0.792 0.000
#> GSM96991 2 0.0000 0.9557 0.000 1.000 0.000 0.000
#> GSM97048 2 0.0000 0.9557 0.000 1.000 0.000 0.000
#> GSM96963 2 0.0000 0.9557 0.000 1.000 0.000 0.000
#> GSM96953 2 0.0000 0.9557 0.000 1.000 0.000 0.000
#> GSM96966 4 0.0000 0.9276 0.000 0.000 0.000 1.000
#> GSM96979 3 0.0000 0.9561 0.000 0.000 1.000 0.000
#> GSM96983 3 0.0000 0.9561 0.000 0.000 1.000 0.000
#> GSM96984 3 0.0000 0.9561 0.000 0.000 1.000 0.000
#> GSM96994 3 0.0000 0.9561 0.000 0.000 1.000 0.000
#> GSM96996 1 0.3024 0.8561 0.852 0.000 0.000 0.148
#> GSM96997 3 0.0000 0.9561 0.000 0.000 1.000 0.000
#> GSM97007 3 0.0000 0.9561 0.000 0.000 1.000 0.000
#> GSM96954 3 0.0817 0.9352 0.024 0.000 0.976 0.000
#> GSM96962 3 0.0000 0.9561 0.000 0.000 1.000 0.000
#> GSM96969 4 0.0000 0.9276 0.000 0.000 0.000 1.000
#> GSM96970 4 0.0000 0.9276 0.000 0.000 0.000 1.000
#> GSM96973 4 0.0000 0.9276 0.000 0.000 0.000 1.000
#> GSM96976 4 0.3528 0.7347 0.000 0.000 0.192 0.808
#> GSM96977 1 0.0817 0.9444 0.976 0.000 0.024 0.000
#> GSM96995 3 0.0000 0.9561 0.000 0.000 1.000 0.000
#> GSM97002 4 0.4746 0.3553 0.368 0.000 0.000 0.632
#> GSM97009 1 0.4244 0.7516 0.800 0.168 0.032 0.000
#> GSM97010 1 0.0000 0.9594 1.000 0.000 0.000 0.000
#> GSM96974 4 0.1867 0.8739 0.000 0.000 0.072 0.928
#> GSM96985 4 0.0336 0.9235 0.000 0.000 0.008 0.992
#> GSM96959 3 0.2563 0.8723 0.072 0.020 0.908 0.000
#> GSM96972 4 0.0000 0.9276 0.000 0.000 0.000 1.000
#> GSM96978 3 0.0000 0.9561 0.000 0.000 1.000 0.000
#> GSM96967 4 0.0000 0.9276 0.000 0.000 0.000 1.000
#> GSM96987 1 0.3172 0.8433 0.840 0.000 0.000 0.160
#> GSM97011 1 0.0000 0.9594 1.000 0.000 0.000 0.000
#> GSM96964 1 0.0000 0.9594 1.000 0.000 0.000 0.000
#> GSM96965 4 0.2704 0.8377 0.124 0.000 0.000 0.876
#> GSM96981 1 0.2589 0.8776 0.884 0.000 0.000 0.116
#> GSM96982 4 0.1389 0.8977 0.048 0.000 0.000 0.952
#> GSM96988 3 0.0000 0.9561 0.000 0.000 1.000 0.000
#> GSM97000 1 0.0188 0.9574 0.996 0.000 0.004 0.000
#> GSM97004 4 0.0000 0.9276 0.000 0.000 0.000 1.000
#> GSM97008 1 0.0000 0.9594 1.000 0.000 0.000 0.000
#> GSM96950 1 0.0000 0.9594 1.000 0.000 0.000 0.000
#> GSM96980 4 0.0000 0.9276 0.000 0.000 0.000 1.000
#> GSM96989 1 0.1302 0.9367 0.956 0.000 0.000 0.044
#> GSM96992 1 0.2589 0.8849 0.884 0.000 0.000 0.116
#> GSM96993 1 0.0000 0.9594 1.000 0.000 0.000 0.000
#> GSM96958 1 0.0000 0.9594 1.000 0.000 0.000 0.000
#> GSM96951 1 0.0000 0.9594 1.000 0.000 0.000 0.000
#> GSM96952 1 0.1940 0.9155 0.924 0.000 0.000 0.076
#> GSM96961 1 0.0000 0.9594 1.000 0.000 0.000 0.000
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM97038 2 0.0000 0.931 0.000 1.000 0.000 0.000 0.000
#> GSM97045 2 0.0290 0.931 0.000 0.992 0.000 0.000 0.008
#> GSM97047 5 0.4859 0.506 0.020 0.288 0.020 0.000 0.672
#> GSM97025 2 0.0162 0.931 0.000 0.996 0.000 0.000 0.004
#> GSM97030 3 0.4497 0.643 0.000 0.016 0.632 0.000 0.352
#> GSM97027 2 0.0162 0.931 0.000 0.996 0.000 0.000 0.004
#> GSM97033 2 0.0290 0.931 0.000 0.992 0.000 0.000 0.008
#> GSM97034 3 0.4298 0.650 0.000 0.008 0.640 0.000 0.352
#> GSM97020 2 0.0290 0.931 0.000 0.992 0.000 0.000 0.008
#> GSM97026 2 0.1901 0.886 0.004 0.928 0.012 0.000 0.056
#> GSM97012 2 0.1121 0.917 0.000 0.956 0.000 0.000 0.044
#> GSM97015 3 0.4402 0.647 0.000 0.012 0.636 0.000 0.352
#> GSM97016 2 0.0162 0.931 0.000 0.996 0.000 0.000 0.004
#> GSM97017 1 0.3999 0.632 0.656 0.000 0.000 0.000 0.344
#> GSM97019 2 0.0510 0.929 0.000 0.984 0.000 0.000 0.016
#> GSM97022 2 0.0703 0.927 0.000 0.976 0.000 0.000 0.024
#> GSM97035 2 0.0404 0.930 0.000 0.988 0.000 0.000 0.012
#> GSM97036 1 0.4624 0.620 0.636 0.000 0.000 0.024 0.340
#> GSM97039 2 0.0162 0.931 0.000 0.996 0.000 0.000 0.004
#> GSM97046 2 0.0162 0.931 0.000 0.996 0.000 0.000 0.004
#> GSM97023 1 0.0798 0.634 0.976 0.000 0.000 0.008 0.016
#> GSM97029 1 0.3999 0.632 0.656 0.000 0.000 0.000 0.344
#> GSM97043 2 0.0579 0.928 0.000 0.984 0.008 0.000 0.008
#> GSM97013 1 0.3999 0.632 0.656 0.000 0.000 0.000 0.344
#> GSM96956 2 0.3890 0.599 0.000 0.736 0.012 0.000 0.252
#> GSM97024 2 0.4283 0.413 0.000 0.644 0.008 0.000 0.348
#> GSM97032 3 0.6455 0.385 0.000 0.188 0.460 0.000 0.352
#> GSM97044 3 0.4030 0.655 0.000 0.000 0.648 0.000 0.352
#> GSM97049 2 0.0290 0.931 0.000 0.992 0.000 0.000 0.008
#> GSM96968 3 0.1124 0.787 0.004 0.000 0.960 0.000 0.036
#> GSM96971 3 0.0000 0.794 0.000 0.000 1.000 0.000 0.000
#> GSM96986 3 0.0162 0.793 0.000 0.000 0.996 0.000 0.004
#> GSM97003 1 0.1661 0.603 0.940 0.000 0.036 0.024 0.000
#> GSM96957 1 0.3999 0.632 0.656 0.000 0.000 0.000 0.344
#> GSM96960 1 0.2813 0.538 0.832 0.000 0.000 0.168 0.000
#> GSM96975 1 0.1608 0.641 0.928 0.000 0.000 0.000 0.072
#> GSM96998 1 0.2605 0.558 0.852 0.000 0.000 0.148 0.000
#> GSM96999 1 0.3999 0.632 0.656 0.000 0.000 0.000 0.344
#> GSM97001 1 0.3999 0.632 0.656 0.000 0.000 0.000 0.344
#> GSM97005 1 0.3999 0.632 0.656 0.000 0.000 0.000 0.344
#> GSM97006 1 0.2605 0.556 0.852 0.000 0.000 0.148 0.000
#> GSM97021 1 0.3999 0.632 0.656 0.000 0.000 0.000 0.344
#> GSM97028 3 0.3177 0.728 0.000 0.000 0.792 0.000 0.208
#> GSM97031 1 0.3366 0.643 0.784 0.000 0.000 0.004 0.212
#> GSM97037 2 0.4269 0.503 0.000 0.684 0.016 0.000 0.300
#> GSM97018 3 0.6742 0.223 0.000 0.260 0.388 0.000 0.352
#> GSM97014 1 0.6606 0.167 0.460 0.192 0.004 0.000 0.344
#> GSM97042 2 0.0963 0.921 0.000 0.964 0.000 0.000 0.036
#> GSM97040 1 0.4464 0.509 0.584 0.000 0.008 0.000 0.408
#> GSM97041 1 0.3999 0.632 0.656 0.000 0.000 0.000 0.344
#> GSM96955 2 0.1195 0.917 0.000 0.960 0.012 0.000 0.028
#> GSM96990 3 0.5804 0.537 0.000 0.104 0.544 0.000 0.352
#> GSM96991 2 0.2017 0.889 0.000 0.912 0.008 0.000 0.080
#> GSM97048 2 0.0290 0.931 0.000 0.992 0.000 0.000 0.008
#> GSM96963 2 0.2017 0.889 0.000 0.912 0.008 0.000 0.080
#> GSM96953 2 0.0609 0.928 0.000 0.980 0.000 0.000 0.020
#> GSM96966 4 0.0162 0.890 0.004 0.000 0.000 0.996 0.000
#> GSM96979 3 0.0000 0.794 0.000 0.000 1.000 0.000 0.000
#> GSM96983 3 0.0794 0.782 0.000 0.000 0.972 0.000 0.028
#> GSM96984 3 0.0000 0.794 0.000 0.000 1.000 0.000 0.000
#> GSM96994 3 0.0000 0.794 0.000 0.000 1.000 0.000 0.000
#> GSM96996 1 0.2813 0.543 0.832 0.000 0.000 0.168 0.000
#> GSM96997 3 0.0162 0.793 0.000 0.000 0.996 0.000 0.004
#> GSM97007 3 0.0000 0.794 0.000 0.000 1.000 0.000 0.000
#> GSM96954 3 0.0703 0.781 0.024 0.000 0.976 0.000 0.000
#> GSM96962 3 0.0162 0.793 0.000 0.000 0.996 0.000 0.004
#> GSM96969 4 0.0000 0.891 0.000 0.000 0.000 1.000 0.000
#> GSM96970 4 0.0000 0.891 0.000 0.000 0.000 1.000 0.000
#> GSM96973 4 0.0290 0.890 0.000 0.000 0.000 0.992 0.008
#> GSM96976 4 0.5167 0.711 0.000 0.000 0.088 0.664 0.248
#> GSM96977 1 0.4419 0.623 0.668 0.000 0.020 0.000 0.312
#> GSM96995 3 0.4030 0.655 0.000 0.000 0.648 0.000 0.352
#> GSM97002 1 0.4302 -0.161 0.520 0.000 0.000 0.480 0.000
#> GSM97009 5 0.4997 0.393 0.248 0.044 0.016 0.000 0.692
#> GSM97010 1 0.4135 0.629 0.656 0.000 0.004 0.000 0.340
#> GSM96974 4 0.4555 0.760 0.000 0.000 0.056 0.720 0.224
#> GSM96985 4 0.3882 0.789 0.000 0.000 0.020 0.756 0.224
#> GSM96959 3 0.5719 0.417 0.048 0.016 0.500 0.000 0.436
#> GSM96972 4 0.0000 0.891 0.000 0.000 0.000 1.000 0.000
#> GSM96978 3 0.0963 0.778 0.000 0.000 0.964 0.000 0.036
#> GSM96967 4 0.0290 0.890 0.000 0.000 0.000 0.992 0.008
#> GSM96987 1 0.2966 0.523 0.816 0.000 0.000 0.184 0.000
#> GSM97011 1 0.3999 0.632 0.656 0.000 0.000 0.000 0.344
#> GSM96964 1 0.0451 0.632 0.988 0.000 0.000 0.004 0.008
#> GSM96965 4 0.3124 0.748 0.144 0.000 0.008 0.840 0.008
#> GSM96981 1 0.3318 0.512 0.800 0.000 0.000 0.192 0.008
#> GSM96982 4 0.1478 0.853 0.064 0.000 0.000 0.936 0.000
#> GSM96988 3 0.0162 0.793 0.000 0.000 0.996 0.004 0.000
#> GSM97000 1 0.4371 0.619 0.644 0.000 0.012 0.000 0.344
#> GSM97004 4 0.3395 0.674 0.236 0.000 0.000 0.764 0.000
#> GSM97008 1 0.3999 0.632 0.656 0.000 0.000 0.000 0.344
#> GSM96950 1 0.3452 0.644 0.756 0.000 0.000 0.000 0.244
#> GSM96980 4 0.0000 0.891 0.000 0.000 0.000 1.000 0.000
#> GSM96989 1 0.2068 0.596 0.904 0.000 0.000 0.092 0.004
#> GSM96992 1 0.2516 0.562 0.860 0.000 0.000 0.140 0.000
#> GSM96993 1 0.3999 0.632 0.656 0.000 0.000 0.000 0.344
#> GSM96958 1 0.0771 0.635 0.976 0.000 0.000 0.004 0.020
#> GSM96951 1 0.0324 0.631 0.992 0.000 0.000 0.004 0.004
#> GSM96952 1 0.2329 0.573 0.876 0.000 0.000 0.124 0.000
#> GSM96961 1 0.0609 0.625 0.980 0.000 0.000 0.020 0.000
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM97038 2 0.0260 0.9332 0.000 0.992 0.008 0.000 0.000 0.000
#> GSM97045 2 0.1049 0.9316 0.000 0.960 0.008 0.000 0.000 0.032
#> GSM97047 3 0.4518 0.5981 0.000 0.072 0.688 0.000 0.236 0.004
#> GSM97025 2 0.0790 0.9329 0.000 0.968 0.000 0.000 0.000 0.032
#> GSM97030 3 0.1367 0.7695 0.000 0.012 0.944 0.000 0.000 0.044
#> GSM97027 2 0.0790 0.9329 0.000 0.968 0.000 0.000 0.000 0.032
#> GSM97033 2 0.1320 0.9285 0.000 0.948 0.016 0.000 0.000 0.036
#> GSM97034 3 0.1152 0.7643 0.000 0.004 0.952 0.000 0.000 0.044
#> GSM97020 2 0.1320 0.9285 0.000 0.948 0.016 0.000 0.000 0.036
#> GSM97026 2 0.3698 0.7664 0.000 0.788 0.116 0.000 0.096 0.000
#> GSM97012 2 0.2287 0.9179 0.048 0.904 0.012 0.000 0.000 0.036
#> GSM97015 3 0.1219 0.7655 0.000 0.004 0.948 0.000 0.000 0.048
#> GSM97016 2 0.0260 0.9341 0.000 0.992 0.008 0.000 0.000 0.000
#> GSM97017 5 0.0000 0.8456 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM97019 2 0.1483 0.9290 0.036 0.944 0.008 0.000 0.000 0.012
#> GSM97022 2 0.2046 0.9224 0.044 0.916 0.008 0.000 0.000 0.032
#> GSM97035 2 0.1382 0.9300 0.036 0.948 0.008 0.000 0.000 0.008
#> GSM97036 5 0.1007 0.8278 0.004 0.000 0.004 0.016 0.968 0.008
#> GSM97039 2 0.1245 0.9297 0.000 0.952 0.016 0.000 0.000 0.032
#> GSM97046 2 0.1168 0.9310 0.000 0.956 0.016 0.000 0.000 0.028
#> GSM97023 1 0.3986 0.7146 0.532 0.000 0.000 0.004 0.464 0.000
#> GSM97029 5 0.0291 0.8432 0.004 0.000 0.004 0.000 0.992 0.000
#> GSM97043 2 0.1448 0.9278 0.016 0.948 0.024 0.000 0.000 0.012
#> GSM97013 5 0.0000 0.8456 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM96956 3 0.3983 0.5191 0.004 0.348 0.640 0.000 0.000 0.008
#> GSM97024 3 0.3398 0.6704 0.000 0.252 0.740 0.000 0.000 0.008
#> GSM97032 3 0.1531 0.7822 0.000 0.068 0.928 0.000 0.000 0.004
#> GSM97044 3 0.1327 0.7562 0.000 0.000 0.936 0.000 0.000 0.064
#> GSM97049 2 0.1320 0.9285 0.000 0.948 0.016 0.000 0.000 0.036
#> GSM96968 6 0.3819 0.7902 0.000 0.000 0.372 0.000 0.004 0.624
#> GSM96971 6 0.4451 0.9147 0.072 0.000 0.248 0.000 0.000 0.680
#> GSM96986 6 0.2996 0.9401 0.000 0.000 0.228 0.000 0.000 0.772
#> GSM97003 1 0.5423 0.7830 0.572 0.000 0.020 0.012 0.344 0.052
#> GSM96957 5 0.0291 0.8455 0.004 0.000 0.004 0.000 0.992 0.000
#> GSM96960 1 0.5343 0.8187 0.580 0.000 0.000 0.156 0.264 0.000
#> GSM96975 5 0.4129 -0.5095 0.424 0.000 0.000 0.000 0.564 0.012
#> GSM96998 1 0.5720 0.8294 0.548 0.000 0.000 0.148 0.292 0.012
#> GSM96999 5 0.0146 0.8452 0.004 0.000 0.000 0.000 0.996 0.000
#> GSM97001 5 0.0291 0.8455 0.004 0.000 0.004 0.000 0.992 0.000
#> GSM97005 5 0.0520 0.8435 0.008 0.000 0.008 0.000 0.984 0.000
#> GSM97006 1 0.5289 0.8337 0.576 0.000 0.000 0.136 0.288 0.000
#> GSM97021 5 0.0405 0.8447 0.004 0.000 0.008 0.000 0.988 0.000
#> GSM97028 3 0.3489 0.2539 0.000 0.004 0.708 0.000 0.000 0.288
#> GSM97031 5 0.4250 -0.1847 0.360 0.000 0.012 0.004 0.620 0.004
#> GSM97037 3 0.3555 0.6380 0.000 0.280 0.712 0.000 0.000 0.008
#> GSM97018 3 0.1918 0.7754 0.000 0.088 0.904 0.000 0.000 0.008
#> GSM97014 5 0.2170 0.7050 0.000 0.100 0.012 0.000 0.888 0.000
#> GSM97042 2 0.2113 0.9209 0.048 0.912 0.008 0.000 0.000 0.032
#> GSM97040 5 0.1501 0.7768 0.000 0.000 0.076 0.000 0.924 0.000
#> GSM97041 5 0.0000 0.8456 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM96955 2 0.3288 0.8767 0.052 0.848 0.064 0.000 0.000 0.036
#> GSM96990 3 0.1333 0.7836 0.000 0.048 0.944 0.000 0.000 0.008
#> GSM96991 2 0.3643 0.8684 0.088 0.820 0.028 0.000 0.000 0.064
#> GSM97048 2 0.1320 0.9285 0.000 0.948 0.016 0.000 0.000 0.036
#> GSM96963 2 0.3643 0.8684 0.088 0.820 0.028 0.000 0.000 0.064
#> GSM96953 2 0.1649 0.9272 0.040 0.936 0.008 0.000 0.000 0.016
#> GSM96966 4 0.0363 0.8116 0.012 0.000 0.000 0.988 0.000 0.000
#> GSM96979 6 0.3265 0.9358 0.000 0.004 0.248 0.000 0.000 0.748
#> GSM96983 6 0.4843 0.8767 0.116 0.000 0.232 0.000 0.000 0.652
#> GSM96984 6 0.3023 0.9410 0.000 0.000 0.232 0.000 0.000 0.768
#> GSM96994 6 0.3050 0.9408 0.000 0.000 0.236 0.000 0.000 0.764
#> GSM96996 1 0.5804 0.8238 0.532 0.000 0.000 0.156 0.300 0.012
#> GSM96997 6 0.2996 0.9401 0.000 0.000 0.228 0.000 0.000 0.772
#> GSM97007 6 0.3023 0.9410 0.000 0.000 0.232 0.000 0.000 0.768
#> GSM96954 6 0.3420 0.9318 0.000 0.000 0.240 0.000 0.012 0.748
#> GSM96962 6 0.2996 0.9401 0.000 0.000 0.228 0.000 0.000 0.772
#> GSM96969 4 0.0260 0.8116 0.008 0.000 0.000 0.992 0.000 0.000
#> GSM96970 4 0.0363 0.8116 0.012 0.000 0.000 0.988 0.000 0.000
#> GSM96973 4 0.0000 0.8109 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM96976 4 0.5789 0.5590 0.364 0.000 0.012 0.492 0.000 0.132
#> GSM96977 5 0.2747 0.7016 0.108 0.000 0.028 0.000 0.860 0.004
#> GSM96995 3 0.1082 0.7640 0.000 0.000 0.956 0.000 0.004 0.040
#> GSM97002 4 0.4783 -0.0736 0.460 0.000 0.000 0.500 0.028 0.012
#> GSM97009 5 0.4534 -0.1169 0.000 0.032 0.472 0.000 0.496 0.000
#> GSM97010 5 0.0777 0.8292 0.024 0.000 0.004 0.000 0.972 0.000
#> GSM96974 4 0.5557 0.5897 0.340 0.000 0.008 0.532 0.000 0.120
#> GSM96985 4 0.5502 0.5985 0.332 0.000 0.008 0.544 0.000 0.116
#> GSM96959 3 0.2095 0.7488 0.000 0.016 0.904 0.000 0.076 0.004
#> GSM96972 4 0.0458 0.8106 0.016 0.000 0.000 0.984 0.000 0.000
#> GSM96978 6 0.5054 0.8653 0.124 0.004 0.232 0.000 0.000 0.640
#> GSM96967 4 0.0000 0.8109 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM96987 1 0.5945 0.7874 0.524 0.000 0.000 0.200 0.264 0.012
#> GSM97011 5 0.0000 0.8456 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM96964 1 0.4303 0.7330 0.524 0.000 0.000 0.004 0.460 0.012
#> GSM96965 4 0.3319 0.7135 0.020 0.000 0.008 0.836 0.116 0.020
#> GSM96981 1 0.6316 0.6150 0.448 0.000 0.000 0.268 0.268 0.016
#> GSM96982 4 0.1867 0.7713 0.064 0.000 0.000 0.916 0.020 0.000
#> GSM96988 6 0.4454 0.9144 0.060 0.004 0.252 0.000 0.000 0.684
#> GSM97000 5 0.0622 0.8418 0.008 0.000 0.012 0.000 0.980 0.000
#> GSM97004 4 0.3729 0.4599 0.296 0.000 0.000 0.692 0.000 0.012
#> GSM97008 5 0.0622 0.8418 0.008 0.000 0.012 0.000 0.980 0.000
#> GSM96950 5 0.2631 0.5369 0.180 0.000 0.000 0.000 0.820 0.000
#> GSM96980 4 0.0458 0.8106 0.016 0.000 0.000 0.984 0.000 0.000
#> GSM96989 1 0.5406 0.8063 0.520 0.000 0.000 0.084 0.384 0.012
#> GSM96992 1 0.5271 0.8350 0.576 0.000 0.000 0.132 0.292 0.000
#> GSM96993 5 0.0146 0.8448 0.000 0.000 0.004 0.000 0.996 0.000
#> GSM96958 1 0.3851 0.7183 0.540 0.000 0.000 0.000 0.460 0.000
#> GSM96951 1 0.4172 0.7555 0.564 0.000 0.008 0.004 0.424 0.000
#> GSM96952 1 0.5405 0.8355 0.572 0.000 0.000 0.132 0.292 0.004
#> GSM96961 1 0.4184 0.7848 0.576 0.000 0.000 0.016 0.408 0.000
Heatmaps for the consensus matrix. It visualizes the probability of two samples to be in a same group.
consensus_heatmap(res, k = 2)
consensus_heatmap(res, k = 3)
consensus_heatmap(res, k = 4)
consensus_heatmap(res, k = 5)
consensus_heatmap(res, k = 6)
Heatmaps for the membership of samples in all partitions to see how consistent they are:
membership_heatmap(res, k = 2)
membership_heatmap(res, k = 3)
membership_heatmap(res, k = 4)
membership_heatmap(res, k = 5)
membership_heatmap(res, k = 6)
As soon as we have had the classes for columns, we can look for signatures which are significantly different between classes which can be candidate marks for certain classes. Following are the heatmaps for signatures.
Signature heatmaps where rows are scaled:
get_signatures(res, k = 2)
get_signatures(res, k = 3)
get_signatures(res, k = 4)
get_signatures(res, k = 5)
get_signatures(res, k = 6)
Signature heatmaps where rows are not scaled:
get_signatures(res, k = 2, scale_rows = FALSE)
get_signatures(res, k = 3, scale_rows = FALSE)
get_signatures(res, k = 4, scale_rows = FALSE)
get_signatures(res, k = 5, scale_rows = FALSE)
get_signatures(res, k = 6, scale_rows = FALSE)
Compare the overlap of signatures from different k:
compare_signatures(res)
get_signature() returns a data frame invisibly. TO get the list of signatures, the function
call should be assigned to a variable explicitly. In following code, if plot argument is set
to FALSE, no heatmap is plotted while only the differential analysis is performed.
# code only for demonstration
tb = get_signature(res, k = ..., plot = FALSE)
An example of the output of tb is:
#> which_row fdr mean_1 mean_2 scaled_mean_1 scaled_mean_2 km
#> 1 38 0.042760348 8.373488 9.131774 -0.5533452 0.5164555 1
#> 2 40 0.018707592 7.106213 8.469186 -0.6173731 0.5762149 1
#> 3 55 0.019134737 10.221463 11.207825 -0.6159697 0.5749050 1
#> 4 59 0.006059896 5.921854 7.869574 -0.6899429 0.6439467 1
#> 5 60 0.018055526 8.928898 10.211722 -0.6204761 0.5791110 1
#> 6 98 0.009384629 15.714769 14.887706 0.6635654 -0.6193277 2
...
The columns in tb are:
which_row: row indices corresponding to the input matrix.fdr: FDR for the differential test. mean_x: The mean value in group x.scaled_mean_x: The mean value in group x after rows are scaled.km: Row groups if k-means clustering is applied to rows.UMAP plot which shows how samples are separated.
dimension_reduction(res, k = 2, method = "UMAP")
dimension_reduction(res, k = 3, method = "UMAP")
dimension_reduction(res, k = 4, method = "UMAP")
dimension_reduction(res, k = 5, method = "UMAP")
dimension_reduction(res, k = 6, method = "UMAP")
Following heatmap shows how subgroups are split when increasing k:
collect_classes(res)
Test correlation between subgroups and known annotations. If the known annotation is numeric, one-way ANOVA test is applied, and if the known annotation is discrete, chi-squared contingency table test is applied.
test_to_known_factors(res)
#> n disease.state(p) specimen(p) cell.type(p) other(p) k
#> CV:mclust 99 6.56e-07 0.5696 2.99e-10 0.0308 2
#> CV:mclust 59 3.59e-03 0.3146 6.24e-13 0.2008 3
#> CV:mclust 96 8.49e-06 0.0410 2.31e-19 0.0171 4
#> CV:mclust 93 3.60e-05 0.0441 9.45e-18 0.0124 5
#> CV:mclust 94 3.88e-06 0.1120 4.63e-20 0.0189 6
If matrix rows can be associated to genes, consider to use functional_enrichment(res,
...) to perform function enrichment for the signature genes. See this vignette for more detailed explanations.
The object with results only for a single top-value method and a single partition method can be extracted as:
res = res_list["CV", "NMF"]
# you can also extract it by
# res = res_list["CV:NMF"]
A summary of res and all the functions that can be applied to it:
res
#> A 'ConsensusPartition' object with k = 2, 3, 4, 5, 6.
#> On a matrix with 21168 rows and 100 columns.
#> Top rows (1000, 2000, 3000, 4000, 5000) are extracted by 'CV' method.
#> Subgroups are detected by 'NMF' method.
#> Performed in total 1250 partitions by row resampling.
#> Best k for subgroups seems to be 2.
#>
#> Following methods can be applied to this 'ConsensusPartition' object:
#> [1] "cola_report" "collect_classes" "collect_plots"
#> [4] "collect_stats" "colnames" "compare_signatures"
#> [7] "consensus_heatmap" "dimension_reduction" "functional_enrichment"
#> [10] "get_anno_col" "get_anno" "get_classes"
#> [13] "get_consensus" "get_matrix" "get_membership"
#> [16] "get_param" "get_signatures" "get_stats"
#> [19] "is_best_k" "is_stable_k" "membership_heatmap"
#> [22] "ncol" "nrow" "plot_ecdf"
#> [25] "rownames" "select_partition_number" "show"
#> [28] "suggest_best_k" "test_to_known_factors"
collect_plots() function collects all the plots made from res for all k (number of partitions)
into one single page to provide an easy and fast comparison between different k.
collect_plots(res)
The plots are:
k and the heatmap of
predicted classes for each k.k.k.k.All the plots in panels can be made by individual functions and they are plotted later in this section.
select_partition_number() produces several plots showing different
statistics for choosing “optimized” k. There are following statistics:
k;k, the area increased is defined as \(A_k - A_{k-1}\).The detailed explanations of these statistics can be found in the cola vignette.
Generally speaking, lower PAC score, higher mean silhouette score or higher
concordance corresponds to better partition. Rand index and Jaccard index
measure how similar the current partition is compared to partition with k-1.
If they are too similar, we won't accept k is better than k-1.
select_partition_number(res)
The numeric values for all these statistics can be obtained by get_stats().
get_stats(res)
#> k 1-PAC mean_silhouette concordance area_increased Rand Jaccard
#> 2 2 1.000 0.972 0.988 0.4964 0.505 0.505
#> 3 3 0.529 0.701 0.840 0.3248 0.807 0.634
#> 4 4 0.587 0.564 0.793 0.1264 0.827 0.567
#> 5 5 0.591 0.526 0.746 0.0726 0.817 0.439
#> 6 6 0.626 0.485 0.696 0.0467 0.884 0.512
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
#> GSM97038 2 0.0000 0.990 0.000 1.000
#> GSM97045 2 0.0000 0.990 0.000 1.000
#> GSM97047 2 0.0000 0.990 0.000 1.000
#> GSM97025 2 0.0000 0.990 0.000 1.000
#> GSM97030 2 0.0000 0.990 0.000 1.000
#> GSM97027 2 0.0000 0.990 0.000 1.000
#> GSM97033 2 0.0000 0.990 0.000 1.000
#> GSM97034 2 0.0000 0.990 0.000 1.000
#> GSM97020 2 0.0000 0.990 0.000 1.000
#> GSM97026 2 0.0000 0.990 0.000 1.000
#> GSM97012 2 0.0000 0.990 0.000 1.000
#> GSM97015 2 0.0000 0.990 0.000 1.000
#> GSM97016 2 0.0000 0.990 0.000 1.000
#> GSM97017 1 0.0376 0.983 0.996 0.004
#> GSM97019 2 0.0000 0.990 0.000 1.000
#> GSM97022 2 0.0000 0.990 0.000 1.000
#> GSM97035 2 0.0000 0.990 0.000 1.000
#> GSM97036 1 0.1843 0.963 0.972 0.028
#> GSM97039 2 0.0000 0.990 0.000 1.000
#> GSM97046 2 0.0000 0.990 0.000 1.000
#> GSM97023 1 0.0000 0.985 1.000 0.000
#> GSM97029 1 0.2603 0.948 0.956 0.044
#> GSM97043 2 0.0000 0.990 0.000 1.000
#> GSM97013 1 0.0000 0.985 1.000 0.000
#> GSM96956 2 0.0000 0.990 0.000 1.000
#> GSM97024 2 0.0000 0.990 0.000 1.000
#> GSM97032 2 0.0000 0.990 0.000 1.000
#> GSM97044 2 0.0000 0.990 0.000 1.000
#> GSM97049 2 0.0000 0.990 0.000 1.000
#> GSM96968 1 0.7376 0.741 0.792 0.208
#> GSM96971 1 0.0000 0.985 1.000 0.000
#> GSM96986 1 0.0000 0.985 1.000 0.000
#> GSM97003 1 0.0000 0.985 1.000 0.000
#> GSM96957 1 0.0376 0.983 0.996 0.004
#> GSM96960 1 0.0000 0.985 1.000 0.000
#> GSM96975 1 0.0000 0.985 1.000 0.000
#> GSM96998 1 0.0000 0.985 1.000 0.000
#> GSM96999 1 0.0000 0.985 1.000 0.000
#> GSM97001 1 0.0000 0.985 1.000 0.000
#> GSM97005 1 0.0000 0.985 1.000 0.000
#> GSM97006 1 0.0000 0.985 1.000 0.000
#> GSM97021 1 0.0938 0.977 0.988 0.012
#> GSM97028 2 0.1843 0.966 0.028 0.972
#> GSM97031 1 0.0000 0.985 1.000 0.000
#> GSM97037 2 0.0000 0.990 0.000 1.000
#> GSM97018 2 0.0000 0.990 0.000 1.000
#> GSM97014 2 0.0000 0.990 0.000 1.000
#> GSM97042 2 0.0000 0.990 0.000 1.000
#> GSM97040 2 0.0672 0.984 0.008 0.992
#> GSM97041 1 0.5842 0.838 0.860 0.140
#> GSM96955 2 0.0000 0.990 0.000 1.000
#> GSM96990 2 0.0000 0.990 0.000 1.000
#> GSM96991 2 0.0000 0.990 0.000 1.000
#> GSM97048 2 0.0000 0.990 0.000 1.000
#> GSM96963 2 0.0000 0.990 0.000 1.000
#> GSM96953 2 0.0000 0.990 0.000 1.000
#> GSM96966 1 0.0000 0.985 1.000 0.000
#> GSM96979 1 0.0000 0.985 1.000 0.000
#> GSM96983 2 0.0000 0.990 0.000 1.000
#> GSM96984 1 0.1843 0.963 0.972 0.028
#> GSM96994 2 0.0938 0.980 0.012 0.988
#> GSM96996 1 0.0000 0.985 1.000 0.000
#> GSM96997 1 0.0000 0.985 1.000 0.000
#> GSM97007 2 0.2423 0.954 0.040 0.960
#> GSM96954 1 0.0000 0.985 1.000 0.000
#> GSM96962 1 0.0000 0.985 1.000 0.000
#> GSM96969 1 0.0000 0.985 1.000 0.000
#> GSM96970 1 0.0000 0.985 1.000 0.000
#> GSM96973 1 0.0000 0.985 1.000 0.000
#> GSM96976 2 0.8763 0.569 0.296 0.704
#> GSM96977 1 0.0000 0.985 1.000 0.000
#> GSM96995 2 0.1843 0.966 0.028 0.972
#> GSM97002 1 0.0000 0.985 1.000 0.000
#> GSM97009 2 0.0000 0.990 0.000 1.000
#> GSM97010 1 0.0672 0.980 0.992 0.008
#> GSM96974 1 0.0000 0.985 1.000 0.000
#> GSM96985 1 0.0000 0.985 1.000 0.000
#> GSM96959 2 0.0000 0.990 0.000 1.000
#> GSM96972 1 0.0000 0.985 1.000 0.000
#> GSM96978 1 0.9044 0.536 0.680 0.320
#> GSM96967 1 0.0000 0.985 1.000 0.000
#> GSM96987 1 0.0000 0.985 1.000 0.000
#> GSM97011 1 0.0938 0.977 0.988 0.012
#> GSM96964 1 0.0000 0.985 1.000 0.000
#> GSM96965 1 0.0938 0.977 0.988 0.012
#> GSM96981 1 0.0000 0.985 1.000 0.000
#> GSM96982 1 0.0000 0.985 1.000 0.000
#> GSM96988 1 0.0000 0.985 1.000 0.000
#> GSM97000 1 0.0000 0.985 1.000 0.000
#> GSM97004 1 0.0000 0.985 1.000 0.000
#> GSM97008 1 0.0000 0.985 1.000 0.000
#> GSM96950 1 0.0000 0.985 1.000 0.000
#> GSM96980 1 0.0000 0.985 1.000 0.000
#> GSM96989 1 0.0000 0.985 1.000 0.000
#> GSM96992 1 0.0000 0.985 1.000 0.000
#> GSM96993 1 0.0000 0.985 1.000 0.000
#> GSM96958 1 0.0000 0.985 1.000 0.000
#> GSM96951 1 0.0000 0.985 1.000 0.000
#> GSM96952 1 0.0000 0.985 1.000 0.000
#> GSM96961 1 0.0000 0.985 1.000 0.000
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM97038 2 0.1753 0.8202 0.000 0.952 0.048
#> GSM97045 2 0.0237 0.8110 0.004 0.996 0.000
#> GSM97047 2 0.1999 0.7856 0.036 0.952 0.012
#> GSM97025 2 0.0892 0.8175 0.000 0.980 0.020
#> GSM97030 2 0.6309 0.2294 0.000 0.504 0.496
#> GSM97027 2 0.0475 0.8102 0.004 0.992 0.004
#> GSM97033 2 0.0237 0.8120 0.000 0.996 0.004
#> GSM97034 3 0.6308 -0.2311 0.000 0.492 0.508
#> GSM97020 2 0.0237 0.8110 0.004 0.996 0.000
#> GSM97026 2 0.0424 0.8151 0.000 0.992 0.008
#> GSM97012 2 0.4399 0.7779 0.000 0.812 0.188
#> GSM97015 2 0.5020 0.7561 0.012 0.796 0.192
#> GSM97016 2 0.2066 0.8204 0.000 0.940 0.060
#> GSM97017 1 0.4861 0.7222 0.800 0.192 0.008
#> GSM97019 2 0.3192 0.8105 0.000 0.888 0.112
#> GSM97022 2 0.4121 0.7897 0.000 0.832 0.168
#> GSM97035 2 0.4062 0.7909 0.000 0.836 0.164
#> GSM97036 1 0.4514 0.7633 0.832 0.156 0.012
#> GSM97039 2 0.1411 0.8193 0.000 0.964 0.036
#> GSM97046 2 0.1964 0.8207 0.000 0.944 0.056
#> GSM97023 1 0.1453 0.8235 0.968 0.024 0.008
#> GSM97029 1 0.4784 0.7165 0.796 0.200 0.004
#> GSM97043 2 0.2448 0.8186 0.000 0.924 0.076
#> GSM97013 1 0.5461 0.6686 0.748 0.244 0.008
#> GSM96956 2 0.6140 0.4971 0.000 0.596 0.404
#> GSM97024 2 0.4178 0.7907 0.000 0.828 0.172
#> GSM97032 2 0.6244 0.4044 0.000 0.560 0.440
#> GSM97044 3 0.6111 0.0984 0.000 0.396 0.604
#> GSM97049 2 0.0424 0.8092 0.008 0.992 0.000
#> GSM96968 3 0.4539 0.7058 0.148 0.016 0.836
#> GSM96971 3 0.1860 0.7556 0.052 0.000 0.948
#> GSM96986 3 0.4842 0.5902 0.224 0.000 0.776
#> GSM97003 1 0.4974 0.7510 0.764 0.000 0.236
#> GSM96957 1 0.6217 0.6359 0.712 0.264 0.024
#> GSM96960 1 0.4291 0.7849 0.820 0.000 0.180
#> GSM96975 1 0.2796 0.8232 0.908 0.000 0.092
#> GSM96998 1 0.1411 0.8330 0.964 0.000 0.036
#> GSM96999 1 0.1832 0.8216 0.956 0.036 0.008
#> GSM97001 1 0.5680 0.6950 0.764 0.212 0.024
#> GSM97005 1 0.2313 0.8179 0.944 0.032 0.024
#> GSM97006 1 0.1964 0.8325 0.944 0.000 0.056
#> GSM97021 1 0.5402 0.7235 0.792 0.180 0.028
#> GSM97028 3 0.3941 0.6328 0.000 0.156 0.844
#> GSM97031 1 0.1964 0.8270 0.944 0.000 0.056
#> GSM97037 2 0.5363 0.6973 0.000 0.724 0.276
#> GSM97018 3 0.6267 -0.1189 0.000 0.452 0.548
#> GSM97014 2 0.4099 0.6736 0.140 0.852 0.008
#> GSM97042 2 0.4399 0.7779 0.000 0.812 0.188
#> GSM97040 2 0.5331 0.5999 0.184 0.792 0.024
#> GSM97041 1 0.6387 0.5887 0.680 0.300 0.020
#> GSM96955 2 0.5591 0.6651 0.000 0.696 0.304
#> GSM96990 2 0.6111 0.5014 0.000 0.604 0.396
#> GSM96991 2 0.6045 0.5460 0.000 0.620 0.380
#> GSM97048 2 0.0237 0.8140 0.000 0.996 0.004
#> GSM96963 2 0.5138 0.7284 0.000 0.748 0.252
#> GSM96953 2 0.4974 0.7398 0.000 0.764 0.236
#> GSM96966 1 0.5363 0.6948 0.724 0.000 0.276
#> GSM96979 3 0.4235 0.6479 0.176 0.000 0.824
#> GSM96983 3 0.2261 0.7242 0.000 0.068 0.932
#> GSM96984 3 0.0829 0.7496 0.004 0.012 0.984
#> GSM96994 3 0.2711 0.7058 0.000 0.088 0.912
#> GSM96996 1 0.2625 0.8262 0.916 0.000 0.084
#> GSM96997 3 0.4605 0.6107 0.204 0.000 0.796
#> GSM97007 3 0.3192 0.6800 0.000 0.112 0.888
#> GSM96954 3 0.6302 -0.1059 0.480 0.000 0.520
#> GSM96962 3 0.2796 0.7466 0.092 0.000 0.908
#> GSM96969 1 0.5948 0.5679 0.640 0.000 0.360
#> GSM96970 1 0.5327 0.6996 0.728 0.000 0.272
#> GSM96973 1 0.6280 0.3458 0.540 0.000 0.460
#> GSM96976 3 0.2297 0.7453 0.020 0.036 0.944
#> GSM96977 1 0.3941 0.8045 0.844 0.000 0.156
#> GSM96995 3 0.5958 0.3759 0.008 0.300 0.692
#> GSM97002 1 0.3686 0.8053 0.860 0.000 0.140
#> GSM97009 2 0.1170 0.8019 0.016 0.976 0.008
#> GSM97010 1 0.4293 0.7919 0.832 0.004 0.164
#> GSM96974 3 0.2400 0.7539 0.064 0.004 0.932
#> GSM96985 3 0.2356 0.7520 0.072 0.000 0.928
#> GSM96959 2 0.2681 0.7881 0.028 0.932 0.040
#> GSM96972 1 0.5098 0.7264 0.752 0.000 0.248
#> GSM96978 3 0.1453 0.7489 0.008 0.024 0.968
#> GSM96967 1 0.6154 0.4722 0.592 0.000 0.408
#> GSM96987 1 0.1411 0.8330 0.964 0.000 0.036
#> GSM97011 1 0.3550 0.7977 0.896 0.080 0.024
#> GSM96964 1 0.0661 0.8281 0.988 0.008 0.004
#> GSM96965 1 0.6526 0.6981 0.704 0.036 0.260
#> GSM96981 1 0.3192 0.8170 0.888 0.000 0.112
#> GSM96982 1 0.5327 0.7000 0.728 0.000 0.272
#> GSM96988 3 0.1753 0.7563 0.048 0.000 0.952
#> GSM97000 1 0.3337 0.8107 0.908 0.032 0.060
#> GSM97004 1 0.3816 0.8010 0.852 0.000 0.148
#> GSM97008 1 0.3832 0.7970 0.888 0.076 0.036
#> GSM96950 1 0.1170 0.8316 0.976 0.008 0.016
#> GSM96980 1 0.4702 0.7580 0.788 0.000 0.212
#> GSM96989 1 0.1411 0.8330 0.964 0.000 0.036
#> GSM96992 1 0.1753 0.8331 0.952 0.000 0.048
#> GSM96993 1 0.2878 0.7994 0.904 0.096 0.000
#> GSM96958 1 0.1289 0.8330 0.968 0.000 0.032
#> GSM96951 1 0.1643 0.8311 0.956 0.000 0.044
#> GSM96952 1 0.1529 0.8328 0.960 0.000 0.040
#> GSM96961 1 0.0892 0.8310 0.980 0.000 0.020
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM97038 2 0.0707 0.8122 0.000 0.980 0.000 0.020
#> GSM97045 2 0.0524 0.8095 0.008 0.988 0.000 0.004
#> GSM97047 2 0.2489 0.7556 0.068 0.912 0.020 0.000
#> GSM97025 2 0.0707 0.8125 0.000 0.980 0.000 0.020
#> GSM97030 3 0.2450 0.7944 0.000 0.072 0.912 0.016
#> GSM97027 2 0.0188 0.8092 0.004 0.996 0.000 0.000
#> GSM97033 2 0.0188 0.8092 0.004 0.996 0.000 0.000
#> GSM97034 3 0.3732 0.7694 0.000 0.092 0.852 0.056
#> GSM97020 2 0.0188 0.8092 0.004 0.996 0.000 0.000
#> GSM97026 2 0.2973 0.7825 0.000 0.856 0.000 0.144
#> GSM97012 2 0.4898 0.5309 0.000 0.584 0.000 0.416
#> GSM97015 3 0.2412 0.7924 0.008 0.084 0.908 0.000
#> GSM97016 2 0.0921 0.8119 0.000 0.972 0.000 0.028
#> GSM97017 1 0.3355 0.6200 0.836 0.160 0.000 0.004
#> GSM97019 2 0.3764 0.7421 0.000 0.784 0.000 0.216
#> GSM97022 2 0.4304 0.6882 0.000 0.716 0.000 0.284
#> GSM97035 2 0.4277 0.6891 0.000 0.720 0.000 0.280
#> GSM97036 1 0.5766 0.5004 0.704 0.104 0.000 0.192
#> GSM97039 2 0.0188 0.8107 0.000 0.996 0.000 0.004
#> GSM97046 2 0.1302 0.8103 0.000 0.956 0.000 0.044
#> GSM97023 1 0.0188 0.6904 0.996 0.000 0.004 0.000
#> GSM97029 1 0.3486 0.6002 0.812 0.188 0.000 0.000
#> GSM97043 2 0.2921 0.7838 0.000 0.860 0.000 0.140
#> GSM97013 1 0.4313 0.5365 0.736 0.260 0.000 0.004
#> GSM96956 2 0.6386 0.5962 0.000 0.640 0.124 0.236
#> GSM97024 2 0.3895 0.7777 0.000 0.832 0.036 0.132
#> GSM97032 3 0.7228 0.2318 0.000 0.340 0.504 0.156
#> GSM97044 3 0.0524 0.8118 0.000 0.008 0.988 0.004
#> GSM97049 2 0.0188 0.8092 0.004 0.996 0.000 0.000
#> GSM96968 3 0.0376 0.8119 0.004 0.004 0.992 0.000
#> GSM96971 3 0.4040 0.6801 0.000 0.000 0.752 0.248
#> GSM96986 3 0.0469 0.8099 0.012 0.000 0.988 0.000
#> GSM97003 3 0.5881 0.5206 0.240 0.000 0.676 0.084
#> GSM96957 1 0.5290 0.3133 0.584 0.404 0.012 0.000
#> GSM96960 1 0.5231 0.4667 0.676 0.000 0.028 0.296
#> GSM96975 1 0.3791 0.5979 0.796 0.000 0.004 0.200
#> GSM96998 1 0.2704 0.6606 0.876 0.000 0.000 0.124
#> GSM96999 1 0.0927 0.6878 0.976 0.016 0.008 0.000
#> GSM97001 1 0.4769 0.4784 0.684 0.308 0.008 0.000
#> GSM97005 1 0.1975 0.6762 0.936 0.016 0.048 0.000
#> GSM97006 1 0.3037 0.6810 0.888 0.000 0.036 0.076
#> GSM97021 1 0.6124 0.4495 0.640 0.276 0.084 0.000
#> GSM97028 3 0.4158 0.7204 0.000 0.008 0.768 0.224
#> GSM97031 3 0.4898 0.3016 0.416 0.000 0.584 0.000
#> GSM97037 2 0.4100 0.7659 0.000 0.824 0.048 0.128
#> GSM97018 2 0.7412 0.3600 0.000 0.444 0.168 0.388
#> GSM97014 2 0.2973 0.6831 0.144 0.856 0.000 0.000
#> GSM97042 2 0.4790 0.5806 0.000 0.620 0.000 0.380
#> GSM97040 2 0.5498 0.1794 0.404 0.576 0.020 0.000
#> GSM97041 1 0.4624 0.4465 0.660 0.340 0.000 0.000
#> GSM96955 4 0.4999 -0.4283 0.000 0.492 0.000 0.508
#> GSM96990 3 0.6148 0.5016 0.000 0.280 0.636 0.084
#> GSM96991 4 0.4961 -0.3492 0.000 0.448 0.000 0.552
#> GSM97048 2 0.0000 0.8100 0.000 1.000 0.000 0.000
#> GSM96963 4 0.4977 -0.3681 0.000 0.460 0.000 0.540
#> GSM96953 2 0.4776 0.5872 0.000 0.624 0.000 0.376
#> GSM96966 4 0.4955 0.0883 0.444 0.000 0.000 0.556
#> GSM96979 3 0.1637 0.8013 0.000 0.000 0.940 0.060
#> GSM96983 3 0.4456 0.6750 0.000 0.004 0.716 0.280
#> GSM96984 3 0.0707 0.8101 0.000 0.000 0.980 0.020
#> GSM96994 3 0.1109 0.8094 0.000 0.004 0.968 0.028
#> GSM96996 1 0.4401 0.5186 0.724 0.000 0.004 0.272
#> GSM96997 3 0.0657 0.8101 0.012 0.000 0.984 0.004
#> GSM97007 3 0.0657 0.8111 0.000 0.004 0.984 0.012
#> GSM96954 3 0.0592 0.8091 0.016 0.000 0.984 0.000
#> GSM96962 3 0.0000 0.8110 0.000 0.000 1.000 0.000
#> GSM96969 4 0.5263 0.0623 0.448 0.000 0.008 0.544
#> GSM96970 4 0.4941 0.1108 0.436 0.000 0.000 0.564
#> GSM96973 4 0.4697 0.2555 0.356 0.000 0.000 0.644
#> GSM96976 4 0.0895 0.4680 0.000 0.020 0.004 0.976
#> GSM96977 1 0.4950 0.5604 0.760 0.008 0.196 0.036
#> GSM96995 3 0.0657 0.8117 0.004 0.012 0.984 0.000
#> GSM97002 1 0.5016 0.3069 0.600 0.000 0.004 0.396
#> GSM97009 2 0.1305 0.7890 0.036 0.960 0.004 0.000
#> GSM97010 1 0.5364 0.2911 0.592 0.016 0.000 0.392
#> GSM96974 4 0.0188 0.4793 0.000 0.000 0.004 0.996
#> GSM96985 4 0.0188 0.4793 0.000 0.000 0.004 0.996
#> GSM96959 3 0.4728 0.6796 0.032 0.216 0.752 0.000
#> GSM96972 1 0.5290 0.0785 0.516 0.000 0.008 0.476
#> GSM96978 3 0.4994 0.4218 0.000 0.000 0.520 0.480
#> GSM96967 4 0.4843 0.1944 0.396 0.000 0.000 0.604
#> GSM96987 1 0.2868 0.6534 0.864 0.000 0.000 0.136
#> GSM97011 1 0.2714 0.6481 0.884 0.112 0.004 0.000
#> GSM96964 1 0.0817 0.6913 0.976 0.000 0.000 0.024
#> GSM96965 4 0.4511 0.3496 0.268 0.008 0.000 0.724
#> GSM96981 1 0.4830 0.3186 0.608 0.000 0.000 0.392
#> GSM96982 1 0.5167 0.0472 0.508 0.000 0.004 0.488
#> GSM96988 3 0.4866 0.5457 0.000 0.000 0.596 0.404
#> GSM97000 3 0.5291 0.5020 0.324 0.024 0.652 0.000
#> GSM97004 1 0.5050 0.2787 0.588 0.000 0.004 0.408
#> GSM97008 1 0.6123 0.4797 0.676 0.132 0.192 0.000
#> GSM96950 1 0.0921 0.6910 0.972 0.000 0.000 0.028
#> GSM96980 1 0.4989 0.1064 0.528 0.000 0.000 0.472
#> GSM96989 1 0.2814 0.6559 0.868 0.000 0.000 0.132
#> GSM96992 1 0.2124 0.6846 0.924 0.000 0.008 0.068
#> GSM96993 1 0.1151 0.6902 0.968 0.024 0.000 0.008
#> GSM96958 1 0.1174 0.6917 0.968 0.000 0.012 0.020
#> GSM96951 1 0.2342 0.6703 0.912 0.000 0.080 0.008
#> GSM96952 1 0.1824 0.6859 0.936 0.000 0.004 0.060
#> GSM96961 1 0.0804 0.6913 0.980 0.000 0.008 0.012
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM97038 2 0.1124 0.73604 0.004 0.960 0.000 0.000 0.036
#> GSM97045 2 0.6301 0.12828 0.180 0.512 0.000 0.000 0.308
#> GSM97047 1 0.7068 -0.30896 0.400 0.388 0.024 0.000 0.188
#> GSM97025 2 0.6188 -0.13706 0.136 0.448 0.000 0.000 0.416
#> GSM97030 3 0.3511 0.71463 0.020 0.020 0.836 0.000 0.124
#> GSM97027 2 0.5948 0.27409 0.156 0.580 0.000 0.000 0.264
#> GSM97033 2 0.0693 0.74116 0.008 0.980 0.000 0.000 0.012
#> GSM97034 5 0.6537 0.42703 0.156 0.016 0.296 0.000 0.532
#> GSM97020 2 0.0898 0.74034 0.008 0.972 0.000 0.000 0.020
#> GSM97026 5 0.5624 0.57013 0.208 0.140 0.000 0.004 0.648
#> GSM97012 5 0.3340 0.65234 0.004 0.156 0.000 0.016 0.824
#> GSM97015 3 0.5890 0.53911 0.120 0.032 0.664 0.000 0.184
#> GSM97016 2 0.0609 0.73991 0.000 0.980 0.000 0.000 0.020
#> GSM97017 1 0.3033 0.64392 0.880 0.032 0.000 0.064 0.024
#> GSM97019 5 0.4814 0.63358 0.080 0.192 0.000 0.004 0.724
#> GSM97022 5 0.4595 0.61163 0.044 0.236 0.000 0.004 0.716
#> GSM97035 5 0.4235 0.49497 0.008 0.336 0.000 0.000 0.656
#> GSM97036 1 0.6763 0.16399 0.432 0.008 0.000 0.196 0.364
#> GSM97039 2 0.0609 0.74006 0.000 0.980 0.000 0.000 0.020
#> GSM97046 2 0.0609 0.73934 0.000 0.980 0.000 0.000 0.020
#> GSM97023 1 0.3496 0.63529 0.788 0.000 0.000 0.200 0.012
#> GSM97029 1 0.3086 0.60691 0.876 0.036 0.000 0.020 0.068
#> GSM97043 5 0.5270 0.62640 0.112 0.172 0.012 0.000 0.704
#> GSM97013 1 0.6449 0.54449 0.580 0.196 0.000 0.204 0.020
#> GSM96956 2 0.2067 0.70789 0.000 0.920 0.048 0.000 0.032
#> GSM97024 5 0.6717 0.56754 0.108 0.232 0.072 0.000 0.588
#> GSM97032 5 0.6072 0.49158 0.080 0.032 0.288 0.000 0.600
#> GSM97044 3 0.3283 0.71003 0.028 0.000 0.832 0.000 0.140
#> GSM97049 2 0.0000 0.73936 0.000 1.000 0.000 0.000 0.000
#> GSM96968 3 0.1805 0.77955 0.020 0.008 0.944 0.016 0.012
#> GSM96971 3 0.4991 0.53034 0.004 0.000 0.636 0.320 0.040
#> GSM96986 3 0.2362 0.76206 0.040 0.000 0.912 0.040 0.008
#> GSM97003 3 0.5352 0.58245 0.096 0.000 0.676 0.220 0.008
#> GSM96957 2 0.5419 0.00266 0.432 0.528 0.012 0.012 0.016
#> GSM96960 4 0.4935 0.55904 0.188 0.000 0.068 0.728 0.016
#> GSM96975 4 0.5532 0.19656 0.392 0.008 0.008 0.556 0.036
#> GSM96998 4 0.4449 -0.17600 0.484 0.000 0.000 0.512 0.004
#> GSM96999 1 0.4975 0.55438 0.648 0.008 0.020 0.316 0.008
#> GSM97001 1 0.6194 0.49065 0.604 0.288 0.020 0.072 0.016
#> GSM97005 1 0.4661 0.63295 0.776 0.012 0.064 0.136 0.012
#> GSM97006 4 0.5178 -0.22771 0.476 0.000 0.040 0.484 0.000
#> GSM97021 1 0.1964 0.62422 0.936 0.012 0.024 0.004 0.024
#> GSM97028 5 0.3690 0.56616 0.012 0.000 0.224 0.000 0.764
#> GSM97031 3 0.5294 0.40153 0.352 0.000 0.596 0.044 0.008
#> GSM97037 2 0.4159 0.58143 0.000 0.776 0.068 0.000 0.156
#> GSM97018 5 0.3817 0.66668 0.020 0.032 0.128 0.000 0.820
#> GSM97014 2 0.1732 0.70939 0.080 0.920 0.000 0.000 0.000
#> GSM97042 5 0.3365 0.64960 0.004 0.180 0.000 0.008 0.808
#> GSM97040 1 0.4286 0.54659 0.800 0.096 0.020 0.000 0.084
#> GSM97041 1 0.2521 0.61924 0.900 0.068 0.000 0.008 0.024
#> GSM96955 5 0.5932 0.16901 0.016 0.368 0.000 0.072 0.544
#> GSM96990 3 0.5725 0.28316 0.016 0.048 0.572 0.004 0.360
#> GSM96991 5 0.2954 0.64057 0.000 0.056 0.004 0.064 0.876
#> GSM97048 2 0.0000 0.73936 0.000 1.000 0.000 0.000 0.000
#> GSM96963 5 0.4075 0.59959 0.000 0.100 0.004 0.096 0.800
#> GSM96953 2 0.4684 0.04814 0.004 0.536 0.000 0.008 0.452
#> GSM96966 4 0.1300 0.69441 0.016 0.000 0.000 0.956 0.028
#> GSM96979 3 0.2997 0.72523 0.000 0.000 0.840 0.148 0.012
#> GSM96983 3 0.4677 0.52172 0.000 0.000 0.664 0.036 0.300
#> GSM96984 3 0.0955 0.77688 0.000 0.000 0.968 0.028 0.004
#> GSM96994 3 0.1124 0.77090 0.000 0.000 0.960 0.004 0.036
#> GSM96996 4 0.4425 0.19057 0.392 0.000 0.000 0.600 0.008
#> GSM96997 3 0.1483 0.77480 0.012 0.000 0.952 0.028 0.008
#> GSM97007 3 0.0771 0.77441 0.004 0.000 0.976 0.000 0.020
#> GSM96954 3 0.0703 0.77834 0.024 0.000 0.976 0.000 0.000
#> GSM96962 3 0.0727 0.77613 0.004 0.000 0.980 0.004 0.012
#> GSM96969 4 0.1399 0.69389 0.020 0.000 0.000 0.952 0.028
#> GSM96970 4 0.1082 0.69260 0.008 0.000 0.000 0.964 0.028
#> GSM96973 4 0.0955 0.68758 0.000 0.000 0.004 0.968 0.028
#> GSM96976 4 0.5181 0.38239 0.000 0.032 0.028 0.668 0.272
#> GSM96977 1 0.6229 0.52456 0.612 0.004 0.092 0.260 0.032
#> GSM96995 3 0.2078 0.77501 0.036 0.004 0.924 0.000 0.036
#> GSM97002 4 0.3675 0.56635 0.216 0.000 0.004 0.772 0.008
#> GSM97009 2 0.2304 0.70232 0.100 0.892 0.000 0.000 0.008
#> GSM97010 4 0.5918 0.29075 0.044 0.376 0.016 0.552 0.012
#> GSM96974 4 0.4473 0.34958 0.000 0.000 0.020 0.656 0.324
#> GSM96985 5 0.4702 -0.11028 0.004 0.000 0.008 0.476 0.512
#> GSM96959 2 0.6182 0.03008 0.096 0.484 0.408 0.000 0.012
#> GSM96972 4 0.1356 0.68969 0.028 0.000 0.012 0.956 0.004
#> GSM96978 3 0.5612 0.51122 0.000 0.000 0.624 0.128 0.248
#> GSM96967 4 0.1012 0.69405 0.012 0.000 0.000 0.968 0.020
#> GSM96987 1 0.4833 0.38072 0.564 0.000 0.000 0.412 0.024
#> GSM97011 1 0.5309 0.60231 0.736 0.132 0.012 0.100 0.020
#> GSM96964 1 0.4181 0.59779 0.712 0.000 0.000 0.268 0.020
#> GSM96965 4 0.2770 0.65118 0.000 0.044 0.000 0.880 0.076
#> GSM96981 4 0.4820 0.55847 0.232 0.008 0.000 0.708 0.052
#> GSM96982 4 0.4149 0.64347 0.128 0.000 0.000 0.784 0.088
#> GSM96988 5 0.4106 0.50463 0.000 0.000 0.256 0.020 0.724
#> GSM97000 3 0.5557 0.31419 0.408 0.020 0.544 0.016 0.012
#> GSM97004 4 0.3388 0.59209 0.200 0.000 0.000 0.792 0.008
#> GSM97008 1 0.4893 0.54529 0.756 0.032 0.168 0.028 0.016
#> GSM96950 1 0.4491 0.54737 0.652 0.000 0.000 0.328 0.020
#> GSM96980 4 0.2046 0.68014 0.068 0.000 0.000 0.916 0.016
#> GSM96989 1 0.4746 0.45855 0.600 0.000 0.000 0.376 0.024
#> GSM96992 1 0.4517 0.43125 0.616 0.000 0.004 0.372 0.008
#> GSM96993 1 0.3948 0.63658 0.808 0.008 0.000 0.128 0.056
#> GSM96958 1 0.5308 0.52647 0.624 0.004 0.036 0.324 0.012
#> GSM96951 1 0.5196 0.60462 0.712 0.000 0.092 0.180 0.016
#> GSM96952 1 0.4464 0.46951 0.632 0.000 0.004 0.356 0.008
#> GSM96961 1 0.3883 0.60579 0.744 0.000 0.004 0.244 0.008
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM97038 3 0.1408 0.7840 0.000 0.020 0.944 0.000 0.036 0.000
#> GSM97045 5 0.6085 -0.2418 0.004 0.388 0.168 0.008 0.432 0.000
#> GSM97047 5 0.5296 0.3195 0.024 0.180 0.104 0.000 0.680 0.012
#> GSM97025 2 0.5645 0.3836 0.000 0.508 0.172 0.000 0.320 0.000
#> GSM97030 6 0.2463 0.7464 0.004 0.080 0.004 0.000 0.024 0.888
#> GSM97027 5 0.6132 -0.2204 0.004 0.356 0.240 0.000 0.400 0.000
#> GSM97033 3 0.2412 0.7473 0.000 0.028 0.880 0.000 0.092 0.000
#> GSM97034 2 0.5837 0.4632 0.004 0.524 0.004 0.000 0.296 0.172
#> GSM97020 3 0.2201 0.7573 0.000 0.048 0.900 0.000 0.052 0.000
#> GSM97026 2 0.5439 0.5259 0.116 0.648 0.020 0.000 0.208 0.008
#> GSM97012 2 0.2303 0.6451 0.000 0.904 0.020 0.052 0.024 0.000
#> GSM97015 6 0.5614 0.4637 0.032 0.232 0.004 0.000 0.108 0.624
#> GSM97016 3 0.0363 0.7919 0.000 0.012 0.988 0.000 0.000 0.000
#> GSM97017 5 0.4413 0.1108 0.484 0.012 0.000 0.008 0.496 0.000
#> GSM97019 2 0.4130 0.6085 0.000 0.740 0.028 0.016 0.212 0.004
#> GSM97022 2 0.4495 0.6079 0.000 0.724 0.056 0.016 0.200 0.004
#> GSM97035 2 0.4225 0.6252 0.000 0.764 0.116 0.016 0.104 0.000
#> GSM97036 1 0.6384 0.0959 0.492 0.308 0.000 0.048 0.152 0.000
#> GSM97039 3 0.0622 0.7914 0.000 0.012 0.980 0.000 0.008 0.000
#> GSM97046 3 0.0363 0.7908 0.000 0.012 0.988 0.000 0.000 0.000
#> GSM97023 1 0.4261 0.5386 0.732 0.000 0.000 0.112 0.156 0.000
#> GSM97029 5 0.5214 0.4452 0.224 0.148 0.000 0.004 0.624 0.000
#> GSM97043 2 0.4430 0.6133 0.024 0.748 0.028 0.000 0.180 0.020
#> GSM97013 1 0.5434 0.4848 0.664 0.000 0.112 0.172 0.052 0.000
#> GSM96956 3 0.0820 0.7870 0.000 0.012 0.972 0.000 0.000 0.016
#> GSM97024 2 0.5848 0.4137 0.000 0.512 0.044 0.008 0.380 0.056
#> GSM97032 2 0.5317 0.4921 0.004 0.608 0.004 0.000 0.120 0.264
#> GSM97044 6 0.2954 0.7216 0.000 0.108 0.000 0.000 0.048 0.844
#> GSM97049 3 0.0146 0.7911 0.000 0.004 0.996 0.000 0.000 0.000
#> GSM96968 6 0.3397 0.7381 0.020 0.024 0.092 0.000 0.020 0.844
#> GSM96971 4 0.4707 0.2368 0.004 0.000 0.000 0.580 0.044 0.372
#> GSM96986 6 0.1977 0.7606 0.008 0.000 0.000 0.040 0.032 0.920
#> GSM97003 6 0.6260 0.3037 0.072 0.000 0.000 0.284 0.108 0.536
#> GSM96957 3 0.4494 0.5146 0.224 0.000 0.696 0.000 0.076 0.004
#> GSM96960 1 0.6081 0.3150 0.532 0.000 0.004 0.308 0.124 0.032
#> GSM96975 1 0.6216 0.1736 0.400 0.000 0.004 0.224 0.368 0.004
#> GSM96998 1 0.4012 0.4358 0.640 0.000 0.000 0.344 0.016 0.000
#> GSM96999 1 0.5103 0.4507 0.664 0.000 0.016 0.120 0.200 0.000
#> GSM97001 5 0.5920 0.2300 0.380 0.000 0.148 0.012 0.460 0.000
#> GSM97005 5 0.5094 0.3483 0.336 0.000 0.004 0.044 0.596 0.020
#> GSM97006 1 0.5157 0.3481 0.548 0.000 0.000 0.384 0.044 0.024
#> GSM97021 5 0.3652 0.5103 0.196 0.032 0.000 0.000 0.768 0.004
#> GSM97028 2 0.4868 0.4939 0.012 0.704 0.000 0.012 0.080 0.192
#> GSM97031 5 0.6249 0.0633 0.112 0.000 0.000 0.048 0.420 0.420
#> GSM97037 3 0.3593 0.6764 0.004 0.064 0.800 0.000 0.000 0.132
#> GSM97018 2 0.3511 0.6339 0.000 0.808 0.004 0.000 0.064 0.124
#> GSM97014 3 0.4428 0.4760 0.012 0.008 0.644 0.012 0.324 0.000
#> GSM97042 2 0.2812 0.6497 0.000 0.876 0.016 0.032 0.072 0.004
#> GSM97040 5 0.4146 0.5284 0.152 0.052 0.028 0.000 0.768 0.000
#> GSM97041 5 0.4704 0.3993 0.344 0.028 0.012 0.004 0.612 0.000
#> GSM96955 2 0.7638 0.1246 0.060 0.416 0.056 0.272 0.196 0.000
#> GSM96990 6 0.4786 0.4615 0.024 0.312 0.020 0.000 0.008 0.636
#> GSM96991 2 0.2812 0.6158 0.000 0.860 0.008 0.104 0.028 0.000
#> GSM97048 3 0.0146 0.7911 0.000 0.004 0.996 0.000 0.000 0.000
#> GSM96963 2 0.3965 0.5631 0.000 0.764 0.004 0.160 0.072 0.000
#> GSM96953 2 0.5858 0.3692 0.000 0.540 0.332 0.072 0.056 0.000
#> GSM96966 4 0.2573 0.6882 0.104 0.012 0.000 0.872 0.012 0.000
#> GSM96979 6 0.3230 0.6373 0.000 0.000 0.000 0.212 0.012 0.776
#> GSM96983 6 0.5864 0.2921 0.000 0.352 0.000 0.028 0.108 0.512
#> GSM96984 6 0.1010 0.7749 0.000 0.000 0.000 0.036 0.004 0.960
#> GSM96994 6 0.1223 0.7783 0.004 0.008 0.000 0.016 0.012 0.960
#> GSM96996 1 0.4463 0.4206 0.616 0.004 0.000 0.352 0.024 0.004
#> GSM96997 6 0.1970 0.7647 0.008 0.000 0.000 0.044 0.028 0.920
#> GSM97007 6 0.0798 0.7768 0.004 0.004 0.000 0.012 0.004 0.976
#> GSM96954 6 0.1471 0.7662 0.000 0.004 0.000 0.000 0.064 0.932
#> GSM96962 6 0.0363 0.7759 0.000 0.000 0.000 0.012 0.000 0.988
#> GSM96969 4 0.1594 0.7093 0.052 0.000 0.000 0.932 0.016 0.000
#> GSM96970 4 0.1801 0.7090 0.056 0.004 0.000 0.924 0.016 0.000
#> GSM96973 4 0.1409 0.7097 0.032 0.000 0.000 0.948 0.008 0.012
#> GSM96976 4 0.3142 0.6502 0.000 0.092 0.000 0.848 0.016 0.044
#> GSM96977 1 0.6555 0.0696 0.472 0.028 0.004 0.060 0.380 0.056
#> GSM96995 6 0.4842 0.6438 0.116 0.032 0.004 0.000 0.120 0.728
#> GSM97002 1 0.4553 0.2822 0.548 0.004 0.000 0.424 0.020 0.004
#> GSM97009 3 0.5515 0.2867 0.028 0.012 0.528 0.020 0.400 0.012
#> GSM97010 3 0.4678 0.3697 0.028 0.000 0.620 0.336 0.004 0.012
#> GSM96974 4 0.3905 0.6035 0.004 0.164 0.000 0.776 0.008 0.048
#> GSM96985 2 0.5898 0.2211 0.020 0.552 0.000 0.304 0.116 0.008
#> GSM96959 3 0.6873 0.0583 0.080 0.000 0.420 0.004 0.356 0.140
#> GSM96972 4 0.3829 0.5405 0.200 0.000 0.000 0.760 0.016 0.024
#> GSM96978 6 0.7007 0.2596 0.004 0.300 0.000 0.120 0.124 0.452
#> GSM96967 4 0.2213 0.6839 0.100 0.008 0.000 0.888 0.004 0.000
#> GSM96987 1 0.3002 0.5850 0.836 0.008 0.000 0.136 0.020 0.000
#> GSM97011 5 0.5264 0.3439 0.308 0.004 0.032 0.048 0.608 0.000
#> GSM96964 1 0.2750 0.5701 0.868 0.004 0.000 0.080 0.048 0.000
#> GSM96965 4 0.1887 0.7078 0.028 0.012 0.012 0.932 0.016 0.000
#> GSM96981 4 0.6572 -0.0404 0.316 0.016 0.004 0.380 0.284 0.000
#> GSM96982 4 0.6704 0.2167 0.272 0.064 0.000 0.472 0.192 0.000
#> GSM96988 2 0.5060 0.3334 0.004 0.628 0.000 0.020 0.052 0.296
#> GSM97000 5 0.5126 0.3774 0.096 0.000 0.000 0.008 0.616 0.280
#> GSM97004 1 0.4361 0.2512 0.544 0.004 0.000 0.436 0.016 0.000
#> GSM97008 5 0.5398 0.3716 0.316 0.000 0.020 0.008 0.592 0.064
#> GSM96950 1 0.3519 0.5827 0.804 0.008 0.000 0.144 0.044 0.000
#> GSM96980 4 0.4595 0.4364 0.264 0.020 0.000 0.676 0.040 0.000
#> GSM96989 1 0.2877 0.5890 0.848 0.008 0.000 0.124 0.020 0.000
#> GSM96992 1 0.4774 0.4885 0.672 0.000 0.000 0.136 0.192 0.000
#> GSM96993 1 0.5108 0.3078 0.688 0.152 0.000 0.020 0.136 0.004
#> GSM96958 1 0.4373 0.4479 0.720 0.000 0.004 0.084 0.192 0.000
#> GSM96951 1 0.5325 0.1666 0.568 0.000 0.004 0.052 0.352 0.024
#> GSM96952 1 0.4934 0.4566 0.660 0.000 0.004 0.124 0.212 0.000
#> GSM96961 1 0.2923 0.5344 0.848 0.000 0.000 0.052 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 disease.state(p) specimen(p) cell.type(p) other(p) k
#> CV:NMF 100 2.42e-05 0.251 1.12e-13 0.1801 2
#> CV:NMF 90 8.38e-06 0.106 9.92e-20 0.0238 3
#> CV:NMF 70 3.14e-04 0.469 2.72e-14 0.1386 4
#> CV:NMF 72 1.45e-02 0.612 1.16e-13 0.1118 5
#> CV:NMF 50 3.71e-04 0.251 9.45e-12 0.0525 6
If matrix rows can be associated to genes, consider to use functional_enrichment(res,
...) to perform function enrichment for the signature genes. See this vignette for more detailed explanations.
The object with results only for a single top-value method and a single partition method can be extracted as:
res = res_list["MAD", "hclust"]
# you can also extract it by
# res = res_list["MAD:hclust"]
A summary of res and all the functions that can be applied to it:
res
#> A 'ConsensusPartition' object with k = 2, 3, 4, 5, 6.
#> On a matrix with 21168 rows and 100 columns.
#> Top rows (1000, 2000, 3000, 4000, 5000) are extracted by 'MAD' method.
#> Subgroups are detected by 'hclust' method.
#> Performed in total 1250 partitions by row resampling.
#> Best k for subgroups seems to be 3.
#>
#> Following methods can be applied to this 'ConsensusPartition' object:
#> [1] "cola_report" "collect_classes" "collect_plots"
#> [4] "collect_stats" "colnames" "compare_signatures"
#> [7] "consensus_heatmap" "dimension_reduction" "functional_enrichment"
#> [10] "get_anno_col" "get_anno" "get_classes"
#> [13] "get_consensus" "get_matrix" "get_membership"
#> [16] "get_param" "get_signatures" "get_stats"
#> [19] "is_best_k" "is_stable_k" "membership_heatmap"
#> [22] "ncol" "nrow" "plot_ecdf"
#> [25] "rownames" "select_partition_number" "show"
#> [28] "suggest_best_k" "test_to_known_factors"
collect_plots() function collects all the plots made from res for all k (number of partitions)
into one single page to provide an easy and fast comparison between different k.
collect_plots(res)
The plots are:
k and the heatmap of
predicted classes for each k.k.k.k.All the plots in panels can be made by individual functions and they are plotted later in this section.
select_partition_number() produces several plots showing different
statistics for choosing “optimized” k. There are following statistics:
k;k, the area increased is defined as \(A_k - A_{k-1}\).The detailed explanations of these statistics can be found in the cola vignette.
Generally speaking, lower PAC score, higher mean silhouette score or higher
concordance corresponds to better partition. Rand index and Jaccard index
measure how similar the current partition is compared to partition with k-1.
If they are too similar, we won't accept k is better than k-1.
select_partition_number(res)
The numeric values for all these statistics can be obtained by get_stats().
get_stats(res)
#> k 1-PAC mean_silhouette concordance area_increased Rand Jaccard
#> 2 2 0.399 0.685 0.857 0.4291 0.602 0.602
#> 3 3 0.488 0.809 0.885 0.4433 0.743 0.588
#> 4 4 0.565 0.741 0.830 0.1299 0.920 0.793
#> 5 5 0.625 0.687 0.767 0.0862 0.909 0.704
#> 6 6 0.687 0.721 0.806 0.0549 0.962 0.827
suggest_best_k() suggests the best \(k\) based on these statistics. The rules are as follows:
suggest_best_k(res)
#> [1] 3
Following shows the table of the partitions (You need to click the show/hide
code output link to see it). The membership matrix (columns with name p*)
is inferred by
clue::cl_consensus()
function with the SE method. Basically the value in the membership matrix
represents the probability to belong to a certain group. The finall class
label for an item is determined with the group with highest probability it
belongs to.
In get_classes() function, the entropy is calculated from the membership
matrix and the silhouette score is calculated from the consensus matrix.
cbind(get_classes(res, k = 2), get_membership(res, k = 2))
#> class entropy silhouette p1 p2
#> GSM97038 2 0.6801 0.6883 0.180 0.820
#> GSM97045 2 0.0672 0.8442 0.008 0.992
#> GSM97047 1 0.8608 0.6251 0.716 0.284
#> GSM97025 2 0.0672 0.8442 0.008 0.992
#> GSM97030 2 0.9970 -0.0386 0.468 0.532
#> GSM97027 2 0.0672 0.8442 0.008 0.992
#> GSM97033 2 0.0000 0.8472 0.000 1.000
#> GSM97034 1 0.6973 0.7451 0.812 0.188
#> GSM97020 2 0.0376 0.8451 0.004 0.996
#> GSM97026 1 0.7219 0.7273 0.800 0.200
#> GSM97012 2 0.0000 0.8472 0.000 1.000
#> GSM97015 1 1.0000 0.1322 0.500 0.500
#> GSM97016 2 0.0000 0.8472 0.000 1.000
#> GSM97017 1 0.6531 0.7506 0.832 0.168
#> GSM97019 2 0.0000 0.8472 0.000 1.000
#> GSM97022 2 0.0000 0.8472 0.000 1.000
#> GSM97035 2 0.0000 0.8472 0.000 1.000
#> GSM97036 1 0.2423 0.8111 0.960 0.040
#> GSM97039 2 0.0000 0.8472 0.000 1.000
#> GSM97046 2 0.0000 0.8472 0.000 1.000
#> GSM97023 1 0.0000 0.8107 1.000 0.000
#> GSM97029 1 0.6973 0.7451 0.812 0.188
#> GSM97043 2 0.9881 0.1069 0.436 0.564
#> GSM97013 1 0.0938 0.8113 0.988 0.012
#> GSM96956 2 0.9710 0.1949 0.400 0.600
#> GSM97024 2 0.4161 0.7901 0.084 0.916
#> GSM97032 2 0.9963 -0.0141 0.464 0.536
#> GSM97044 1 0.9998 0.1549 0.508 0.492
#> GSM97049 2 0.0000 0.8472 0.000 1.000
#> GSM96968 1 0.9635 0.4819 0.612 0.388
#> GSM96971 1 0.9248 0.5179 0.660 0.340
#> GSM96986 1 0.9427 0.4887 0.640 0.360
#> GSM97003 1 0.0000 0.8107 1.000 0.000
#> GSM96957 1 0.4022 0.7993 0.920 0.080
#> GSM96960 1 0.0000 0.8107 1.000 0.000
#> GSM96975 1 0.2423 0.8122 0.960 0.040
#> GSM96998 1 0.0376 0.8113 0.996 0.004
#> GSM96999 1 0.4022 0.7993 0.920 0.080
#> GSM97001 1 0.4022 0.7993 0.920 0.080
#> GSM97005 1 0.1633 0.8128 0.976 0.024
#> GSM97006 1 0.0000 0.8107 1.000 0.000
#> GSM97021 1 0.6531 0.7504 0.832 0.168
#> GSM97028 1 0.8499 0.6245 0.724 0.276
#> GSM97031 1 0.5629 0.7575 0.868 0.132
#> GSM97037 2 0.9881 0.0717 0.436 0.564
#> GSM97018 1 0.9580 0.4816 0.620 0.380
#> GSM97014 1 0.8016 0.6799 0.756 0.244
#> GSM97042 2 0.0000 0.8472 0.000 1.000
#> GSM97040 1 0.7815 0.6910 0.768 0.232
#> GSM97041 1 0.6438 0.7534 0.836 0.164
#> GSM96955 2 0.7674 0.6240 0.224 0.776
#> GSM96990 1 0.9954 0.2779 0.540 0.460
#> GSM96991 2 0.0000 0.8472 0.000 1.000
#> GSM97048 2 0.0000 0.8472 0.000 1.000
#> GSM96963 2 0.0000 0.8472 0.000 1.000
#> GSM96953 2 0.0000 0.8472 0.000 1.000
#> GSM96966 1 0.0376 0.8109 0.996 0.004
#> GSM96979 1 0.9393 0.4963 0.644 0.356
#> GSM96983 1 0.9686 0.4100 0.604 0.396
#> GSM96984 1 0.9661 0.4190 0.608 0.392
#> GSM96994 1 0.9460 0.4806 0.636 0.364
#> GSM96996 1 0.0376 0.8109 0.996 0.004
#> GSM96997 1 0.9635 0.4283 0.612 0.388
#> GSM97007 1 0.9661 0.4190 0.608 0.392
#> GSM96954 1 0.9248 0.5179 0.660 0.340
#> GSM96962 1 0.9393 0.4963 0.644 0.356
#> GSM96969 1 0.0000 0.8107 1.000 0.000
#> GSM96970 1 0.0000 0.8107 1.000 0.000
#> GSM96973 1 0.1843 0.8104 0.972 0.028
#> GSM96976 1 0.4161 0.8019 0.916 0.084
#> GSM96977 1 0.3733 0.8053 0.928 0.072
#> GSM96995 1 0.9635 0.4819 0.612 0.388
#> GSM97002 1 0.0000 0.8107 1.000 0.000
#> GSM97009 1 0.8267 0.6607 0.740 0.260
#> GSM97010 1 0.3274 0.8064 0.940 0.060
#> GSM96974 1 0.3431 0.8056 0.936 0.064
#> GSM96985 1 0.9686 0.4100 0.604 0.396
#> GSM96959 2 0.9044 0.4409 0.320 0.680
#> GSM96972 1 0.0000 0.8107 1.000 0.000
#> GSM96978 1 0.9686 0.4100 0.604 0.396
#> GSM96967 1 0.0000 0.8107 1.000 0.000
#> GSM96987 1 0.0376 0.8113 0.996 0.004
#> GSM97011 1 0.8081 0.6738 0.752 0.248
#> GSM96964 1 0.0376 0.8115 0.996 0.004
#> GSM96965 1 0.4431 0.8006 0.908 0.092
#> GSM96981 1 0.0672 0.8121 0.992 0.008
#> GSM96982 1 0.0672 0.8121 0.992 0.008
#> GSM96988 1 0.8555 0.6190 0.720 0.280
#> GSM97000 1 0.6623 0.7427 0.828 0.172
#> GSM97004 1 0.0000 0.8107 1.000 0.000
#> GSM97008 1 0.3879 0.8037 0.924 0.076
#> GSM96950 1 0.2423 0.8101 0.960 0.040
#> GSM96980 1 0.0000 0.8107 1.000 0.000
#> GSM96989 1 0.0376 0.8113 0.996 0.004
#> GSM96992 1 0.0000 0.8107 1.000 0.000
#> GSM96993 1 0.2236 0.8115 0.964 0.036
#> GSM96958 1 0.2603 0.8121 0.956 0.044
#> GSM96951 1 0.0000 0.8107 1.000 0.000
#> GSM96952 1 0.0000 0.8107 1.000 0.000
#> GSM96961 1 0.0000 0.8107 1.000 0.000
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM97038 2 0.5966 0.726 0.104 0.792 0.104
#> GSM97045 2 0.0661 0.935 0.008 0.988 0.004
#> GSM97047 1 0.7880 0.630 0.636 0.268 0.096
#> GSM97025 2 0.0661 0.935 0.008 0.988 0.004
#> GSM97030 3 0.3879 0.785 0.000 0.152 0.848
#> GSM97027 2 0.0661 0.935 0.008 0.988 0.004
#> GSM97033 2 0.0000 0.942 0.000 1.000 0.000
#> GSM97034 1 0.6543 0.756 0.748 0.176 0.076
#> GSM97020 2 0.0475 0.938 0.004 0.992 0.004
#> GSM97026 1 0.5901 0.773 0.768 0.192 0.040
#> GSM97012 2 0.0000 0.942 0.000 1.000 0.000
#> GSM97015 3 0.3682 0.806 0.008 0.116 0.876
#> GSM97016 2 0.0000 0.942 0.000 1.000 0.000
#> GSM97017 1 0.5239 0.800 0.808 0.160 0.032
#> GSM97019 2 0.0000 0.942 0.000 1.000 0.000
#> GSM97022 2 0.0000 0.942 0.000 1.000 0.000
#> GSM97035 2 0.0000 0.942 0.000 1.000 0.000
#> GSM97036 1 0.1411 0.865 0.964 0.036 0.000
#> GSM97039 2 0.0000 0.942 0.000 1.000 0.000
#> GSM97046 2 0.0000 0.942 0.000 1.000 0.000
#> GSM97023 1 0.0237 0.865 0.996 0.000 0.004
#> GSM97029 1 0.6372 0.762 0.756 0.176 0.068
#> GSM97043 3 0.8198 0.544 0.100 0.304 0.596
#> GSM97013 1 0.0424 0.867 0.992 0.008 0.000
#> GSM96956 3 0.5216 0.678 0.000 0.260 0.740
#> GSM97024 2 0.4062 0.769 0.000 0.836 0.164
#> GSM97032 3 0.5559 0.751 0.028 0.192 0.780
#> GSM97044 3 0.3038 0.809 0.000 0.104 0.896
#> GSM97049 2 0.0000 0.942 0.000 1.000 0.000
#> GSM96968 3 0.8350 0.570 0.280 0.120 0.600
#> GSM96971 3 0.3267 0.804 0.116 0.000 0.884
#> GSM96986 3 0.1860 0.836 0.052 0.000 0.948
#> GSM97003 1 0.0237 0.865 0.996 0.000 0.004
#> GSM96957 1 0.3856 0.849 0.888 0.072 0.040
#> GSM96960 1 0.0237 0.865 0.996 0.000 0.004
#> GSM96975 1 0.2443 0.865 0.940 0.028 0.032
#> GSM96998 1 0.0000 0.865 1.000 0.000 0.000
#> GSM96999 1 0.3856 0.849 0.888 0.072 0.040
#> GSM97001 1 0.3856 0.849 0.888 0.072 0.040
#> GSM97005 1 0.1491 0.868 0.968 0.016 0.016
#> GSM97006 1 0.0237 0.865 0.996 0.000 0.004
#> GSM97021 1 0.5466 0.797 0.800 0.160 0.040
#> GSM97028 3 0.6506 0.680 0.236 0.044 0.720
#> GSM97031 1 0.5733 0.478 0.676 0.000 0.324
#> GSM97037 3 0.4883 0.739 0.004 0.208 0.788
#> GSM97018 3 0.7906 0.676 0.220 0.124 0.656
#> GSM97014 1 0.7187 0.701 0.692 0.232 0.076
#> GSM97042 2 0.0000 0.942 0.000 1.000 0.000
#> GSM97040 1 0.7222 0.704 0.696 0.220 0.084
#> GSM97041 1 0.5060 0.805 0.816 0.156 0.028
#> GSM96955 2 0.6561 0.670 0.144 0.756 0.100
#> GSM96990 3 0.4399 0.819 0.044 0.092 0.864
#> GSM96991 2 0.0000 0.942 0.000 1.000 0.000
#> GSM97048 2 0.0000 0.942 0.000 1.000 0.000
#> GSM96963 2 0.0000 0.942 0.000 1.000 0.000
#> GSM96953 2 0.0000 0.942 0.000 1.000 0.000
#> GSM96966 1 0.4178 0.777 0.828 0.000 0.172
#> GSM96979 3 0.1964 0.835 0.056 0.000 0.944
#> GSM96983 3 0.0237 0.830 0.000 0.004 0.996
#> GSM96984 3 0.0000 0.829 0.000 0.000 1.000
#> GSM96994 3 0.1753 0.836 0.048 0.000 0.952
#> GSM96996 1 0.0829 0.867 0.984 0.004 0.012
#> GSM96997 3 0.0237 0.830 0.004 0.000 0.996
#> GSM97007 3 0.0000 0.829 0.000 0.000 1.000
#> GSM96954 3 0.3267 0.804 0.116 0.000 0.884
#> GSM96962 3 0.1964 0.835 0.056 0.000 0.944
#> GSM96969 1 0.3482 0.805 0.872 0.000 0.128
#> GSM96970 1 0.3551 0.803 0.868 0.000 0.132
#> GSM96973 1 0.5024 0.731 0.776 0.004 0.220
#> GSM96976 1 0.7141 0.480 0.600 0.032 0.368
#> GSM96977 1 0.3993 0.849 0.884 0.052 0.064
#> GSM96995 3 0.8350 0.570 0.280 0.120 0.600
#> GSM97002 1 0.0237 0.865 0.996 0.000 0.004
#> GSM97009 1 0.7303 0.685 0.680 0.244 0.076
#> GSM97010 1 0.3993 0.850 0.884 0.052 0.064
#> GSM96974 1 0.6617 0.463 0.600 0.012 0.388
#> GSM96985 3 0.0237 0.830 0.000 0.004 0.996
#> GSM96959 2 0.7835 0.497 0.232 0.656 0.112
#> GSM96972 1 0.3340 0.810 0.880 0.000 0.120
#> GSM96978 3 0.0237 0.830 0.000 0.004 0.996
#> GSM96967 1 0.3482 0.805 0.872 0.000 0.128
#> GSM96987 1 0.0000 0.865 1.000 0.000 0.000
#> GSM97011 1 0.7226 0.694 0.688 0.236 0.076
#> GSM96964 1 0.0424 0.866 0.992 0.000 0.008
#> GSM96965 1 0.7170 0.500 0.612 0.036 0.352
#> GSM96981 1 0.0661 0.867 0.988 0.008 0.004
#> GSM96982 1 0.0661 0.867 0.988 0.008 0.004
#> GSM96988 3 0.6423 0.691 0.228 0.044 0.728
#> GSM97000 1 0.6488 0.763 0.756 0.160 0.084
#> GSM97004 1 0.0237 0.865 0.996 0.000 0.004
#> GSM97008 1 0.4194 0.844 0.876 0.064 0.060
#> GSM96950 1 0.2689 0.863 0.932 0.032 0.036
#> GSM96980 1 0.1031 0.861 0.976 0.000 0.024
#> GSM96989 1 0.0000 0.865 1.000 0.000 0.000
#> GSM96992 1 0.0237 0.865 0.996 0.000 0.004
#> GSM96993 1 0.1711 0.866 0.960 0.032 0.008
#> GSM96958 1 0.2564 0.865 0.936 0.028 0.036
#> GSM96951 1 0.0237 0.865 0.996 0.000 0.004
#> GSM96952 1 0.0237 0.865 0.996 0.000 0.004
#> GSM96961 1 0.0237 0.865 0.996 0.000 0.004
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM97038 2 0.5649 0.726 0.084 0.772 0.056 0.088
#> GSM97045 2 0.0779 0.930 0.004 0.980 0.000 0.016
#> GSM97047 1 0.7703 0.502 0.564 0.232 0.028 0.176
#> GSM97025 2 0.0779 0.930 0.004 0.980 0.000 0.016
#> GSM97030 3 0.4050 0.761 0.000 0.144 0.820 0.036
#> GSM97027 2 0.0779 0.930 0.004 0.980 0.000 0.016
#> GSM97033 2 0.0336 0.936 0.000 0.992 0.000 0.008
#> GSM97034 1 0.6616 0.626 0.680 0.144 0.024 0.152
#> GSM97020 2 0.0657 0.933 0.004 0.984 0.000 0.012
#> GSM97026 1 0.6128 0.650 0.692 0.152 0.004 0.152
#> GSM97012 2 0.0000 0.939 0.000 1.000 0.000 0.000
#> GSM97015 3 0.3974 0.773 0.008 0.108 0.844 0.040
#> GSM97016 2 0.0000 0.939 0.000 1.000 0.000 0.000
#> GSM97017 1 0.5528 0.683 0.732 0.124 0.000 0.144
#> GSM97019 2 0.0000 0.939 0.000 1.000 0.000 0.000
#> GSM97022 2 0.0000 0.939 0.000 1.000 0.000 0.000
#> GSM97035 2 0.0000 0.939 0.000 1.000 0.000 0.000
#> GSM97036 1 0.1510 0.774 0.956 0.028 0.000 0.016
#> GSM97039 2 0.0000 0.939 0.000 1.000 0.000 0.000
#> GSM97046 2 0.0000 0.939 0.000 1.000 0.000 0.000
#> GSM97023 1 0.1389 0.762 0.952 0.000 0.000 0.048
#> GSM97029 1 0.6473 0.633 0.688 0.144 0.020 0.148
#> GSM97043 3 0.7578 0.536 0.096 0.288 0.568 0.048
#> GSM97013 1 0.0779 0.770 0.980 0.004 0.000 0.016
#> GSM96956 3 0.5198 0.674 0.000 0.252 0.708 0.040
#> GSM97024 2 0.3647 0.768 0.000 0.832 0.152 0.016
#> GSM97032 3 0.5378 0.729 0.028 0.184 0.752 0.036
#> GSM97044 3 0.3463 0.776 0.000 0.096 0.864 0.040
#> GSM97049 2 0.0000 0.939 0.000 1.000 0.000 0.000
#> GSM96968 3 0.8181 0.488 0.248 0.096 0.552 0.104
#> GSM96971 3 0.4332 0.711 0.032 0.000 0.792 0.176
#> GSM96986 3 0.2399 0.784 0.032 0.000 0.920 0.048
#> GSM97003 1 0.2149 0.739 0.912 0.000 0.000 0.088
#> GSM96957 1 0.3764 0.749 0.844 0.040 0.000 0.116
#> GSM96960 1 0.2345 0.731 0.900 0.000 0.000 0.100
#> GSM96975 1 0.3853 0.760 0.848 0.020 0.016 0.116
#> GSM96998 1 0.1792 0.751 0.932 0.000 0.000 0.068
#> GSM96999 1 0.3764 0.749 0.844 0.040 0.000 0.116
#> GSM97001 1 0.3764 0.749 0.844 0.040 0.000 0.116
#> GSM97005 1 0.3102 0.770 0.872 0.004 0.008 0.116
#> GSM97006 1 0.2345 0.731 0.900 0.000 0.000 0.100
#> GSM97021 1 0.5747 0.679 0.724 0.120 0.004 0.152
#> GSM97028 3 0.5482 0.598 0.232 0.040 0.716 0.012
#> GSM97031 1 0.6654 0.215 0.588 0.000 0.296 0.116
#> GSM97037 3 0.4935 0.723 0.004 0.200 0.756 0.040
#> GSM97018 3 0.7443 0.580 0.212 0.112 0.620 0.056
#> GSM97014 1 0.6929 0.572 0.620 0.192 0.008 0.180
#> GSM97042 2 0.0000 0.939 0.000 1.000 0.000 0.000
#> GSM97040 1 0.7089 0.573 0.620 0.180 0.016 0.184
#> GSM97041 1 0.5428 0.687 0.740 0.120 0.000 0.140
#> GSM96955 2 0.6232 0.660 0.096 0.720 0.036 0.148
#> GSM96990 3 0.4909 0.767 0.036 0.076 0.812 0.076
#> GSM96991 2 0.0188 0.937 0.000 0.996 0.004 0.000
#> GSM97048 2 0.0000 0.939 0.000 1.000 0.000 0.000
#> GSM96963 2 0.0188 0.937 0.000 0.996 0.004 0.000
#> GSM96953 2 0.0000 0.939 0.000 1.000 0.000 0.000
#> GSM96966 4 0.5250 0.763 0.316 0.000 0.024 0.660
#> GSM96979 3 0.2644 0.783 0.032 0.000 0.908 0.060
#> GSM96983 3 0.0188 0.787 0.000 0.000 0.996 0.004
#> GSM96984 3 0.0817 0.786 0.000 0.000 0.976 0.024
#> GSM96994 3 0.2300 0.785 0.028 0.000 0.924 0.048
#> GSM96996 1 0.2216 0.751 0.908 0.000 0.000 0.092
#> GSM96997 3 0.0921 0.786 0.000 0.000 0.972 0.028
#> GSM97007 3 0.0817 0.786 0.000 0.000 0.976 0.024
#> GSM96954 3 0.4332 0.711 0.032 0.000 0.792 0.176
#> GSM96962 3 0.2644 0.783 0.032 0.000 0.908 0.060
#> GSM96969 4 0.4819 0.750 0.344 0.000 0.004 0.652
#> GSM96970 4 0.4781 0.756 0.336 0.000 0.004 0.660
#> GSM96973 4 0.4840 0.759 0.240 0.000 0.028 0.732
#> GSM96976 4 0.3975 0.620 0.064 0.016 0.064 0.856
#> GSM96977 1 0.4414 0.739 0.824 0.036 0.020 0.120
#> GSM96995 3 0.8181 0.488 0.248 0.096 0.552 0.104
#> GSM97002 1 0.2149 0.739 0.912 0.000 0.000 0.088
#> GSM97009 1 0.7146 0.562 0.612 0.204 0.016 0.168
#> GSM97010 1 0.4573 0.750 0.816 0.036 0.024 0.124
#> GSM96974 4 0.3687 0.612 0.064 0.000 0.080 0.856
#> GSM96985 3 0.0469 0.787 0.000 0.000 0.988 0.012
#> GSM96959 2 0.7486 0.481 0.180 0.616 0.044 0.160
#> GSM96972 4 0.4697 0.734 0.356 0.000 0.000 0.644
#> GSM96978 3 0.0469 0.787 0.000 0.000 0.988 0.012
#> GSM96967 4 0.4800 0.753 0.340 0.000 0.004 0.656
#> GSM96987 1 0.0336 0.769 0.992 0.000 0.000 0.008
#> GSM97011 1 0.7045 0.566 0.616 0.196 0.012 0.176
#> GSM96964 1 0.1022 0.769 0.968 0.000 0.000 0.032
#> GSM96965 4 0.4150 0.625 0.076 0.020 0.056 0.848
#> GSM96981 1 0.2345 0.745 0.900 0.000 0.000 0.100
#> GSM96982 1 0.2345 0.745 0.900 0.000 0.000 0.100
#> GSM96988 3 0.5636 0.604 0.224 0.040 0.716 0.020
#> GSM97000 1 0.6433 0.641 0.684 0.128 0.016 0.172
#> GSM97004 1 0.2345 0.731 0.900 0.000 0.000 0.100
#> GSM97008 1 0.4520 0.733 0.800 0.036 0.008 0.156
#> GSM96950 1 0.2662 0.766 0.900 0.016 0.000 0.084
#> GSM96980 1 0.4431 0.353 0.696 0.000 0.000 0.304
#> GSM96989 1 0.0336 0.769 0.992 0.000 0.000 0.008
#> GSM96992 1 0.2281 0.734 0.904 0.000 0.000 0.096
#> GSM96993 1 0.1733 0.774 0.948 0.028 0.000 0.024
#> GSM96958 1 0.3614 0.762 0.864 0.020 0.016 0.100
#> GSM96951 1 0.1867 0.758 0.928 0.000 0.000 0.072
#> GSM96952 1 0.2281 0.734 0.904 0.000 0.000 0.096
#> GSM96961 1 0.2281 0.734 0.904 0.000 0.000 0.096
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM97038 2 0.4649 0.705 0.000 0.740 0.044 0.016 0.200
#> GSM97045 2 0.1121 0.909 0.000 0.956 0.000 0.000 0.044
#> GSM97047 5 0.3842 0.574 0.004 0.132 0.016 0.028 0.820
#> GSM97025 2 0.1043 0.912 0.000 0.960 0.000 0.000 0.040
#> GSM97030 3 0.3952 0.718 0.000 0.132 0.812 0.024 0.032
#> GSM97027 2 0.1121 0.909 0.000 0.956 0.000 0.000 0.044
#> GSM97033 2 0.0703 0.919 0.000 0.976 0.000 0.000 0.024
#> GSM97034 5 0.4726 0.620 0.076 0.088 0.016 0.028 0.792
#> GSM97020 2 0.1043 0.912 0.000 0.960 0.000 0.000 0.040
#> GSM97026 5 0.2756 0.656 0.036 0.060 0.000 0.012 0.892
#> GSM97012 2 0.0000 0.926 0.000 1.000 0.000 0.000 0.000
#> GSM97015 3 0.3800 0.728 0.000 0.084 0.836 0.028 0.052
#> GSM97016 2 0.0162 0.925 0.004 0.996 0.000 0.000 0.000
#> GSM97017 5 0.2424 0.669 0.052 0.032 0.000 0.008 0.908
#> GSM97019 2 0.0000 0.926 0.000 1.000 0.000 0.000 0.000
#> GSM97022 2 0.0000 0.926 0.000 1.000 0.000 0.000 0.000
#> GSM97035 2 0.0000 0.926 0.000 1.000 0.000 0.000 0.000
#> GSM97036 5 0.3992 0.536 0.280 0.004 0.000 0.004 0.712
#> GSM97039 2 0.0162 0.925 0.004 0.996 0.000 0.000 0.000
#> GSM97046 2 0.0162 0.925 0.004 0.996 0.000 0.000 0.000
#> GSM97023 5 0.4114 0.321 0.376 0.000 0.000 0.000 0.624
#> GSM97029 5 0.4740 0.621 0.084 0.088 0.012 0.028 0.788
#> GSM97043 3 0.6550 0.537 0.000 0.240 0.560 0.020 0.180
#> GSM97013 5 0.3876 0.476 0.316 0.000 0.000 0.000 0.684
#> GSM96956 3 0.4934 0.633 0.000 0.244 0.700 0.028 0.028
#> GSM97024 2 0.3474 0.757 0.000 0.824 0.148 0.008 0.020
#> GSM97032 3 0.4922 0.697 0.000 0.160 0.744 0.024 0.072
#> GSM97044 3 0.3426 0.729 0.000 0.084 0.856 0.028 0.032
#> GSM97049 2 0.0162 0.925 0.004 0.996 0.000 0.000 0.000
#> GSM96968 3 0.7026 0.522 0.040 0.052 0.544 0.052 0.312
#> GSM96971 3 0.5814 0.606 0.180 0.000 0.612 0.208 0.000
#> GSM96986 3 0.4361 0.712 0.204 0.000 0.752 0.032 0.012
#> GSM97003 1 0.3586 0.829 0.736 0.000 0.000 0.000 0.264
#> GSM96957 5 0.3422 0.629 0.200 0.004 0.000 0.004 0.792
#> GSM96960 1 0.3461 0.824 0.772 0.000 0.000 0.004 0.224
#> GSM96975 5 0.5326 -0.215 0.464 0.000 0.012 0.028 0.496
#> GSM96998 1 0.4060 0.697 0.640 0.000 0.000 0.000 0.360
#> GSM96999 5 0.3422 0.629 0.200 0.004 0.000 0.004 0.792
#> GSM97001 5 0.3422 0.629 0.200 0.004 0.000 0.004 0.792
#> GSM97005 5 0.3730 0.514 0.288 0.000 0.000 0.000 0.712
#> GSM97006 1 0.3461 0.824 0.772 0.000 0.000 0.004 0.224
#> GSM97021 5 0.2228 0.666 0.040 0.028 0.000 0.012 0.920
#> GSM97028 3 0.4975 0.633 0.024 0.032 0.712 0.004 0.228
#> GSM97031 1 0.6000 0.236 0.636 0.000 0.140 0.020 0.204
#> GSM97037 3 0.4625 0.681 0.000 0.192 0.748 0.028 0.032
#> GSM97018 3 0.6390 0.601 0.016 0.084 0.608 0.028 0.264
#> GSM97014 5 0.3052 0.624 0.008 0.092 0.000 0.032 0.868
#> GSM97042 2 0.0000 0.926 0.000 1.000 0.000 0.000 0.000
#> GSM97040 5 0.3434 0.626 0.016 0.084 0.008 0.032 0.860
#> GSM97041 5 0.2139 0.669 0.052 0.032 0.000 0.000 0.916
#> GSM96955 2 0.5508 0.598 0.008 0.648 0.024 0.036 0.284
#> GSM96990 3 0.4344 0.724 0.000 0.056 0.804 0.044 0.096
#> GSM96991 2 0.0162 0.925 0.004 0.996 0.000 0.000 0.000
#> GSM97048 2 0.0162 0.925 0.004 0.996 0.000 0.000 0.000
#> GSM96963 2 0.0162 0.925 0.004 0.996 0.000 0.000 0.000
#> GSM96953 2 0.0000 0.926 0.000 1.000 0.000 0.000 0.000
#> GSM96966 4 0.4914 0.726 0.336 0.000 0.004 0.628 0.032
#> GSM96979 3 0.4703 0.712 0.204 0.000 0.736 0.040 0.020
#> GSM96983 3 0.0880 0.738 0.032 0.000 0.968 0.000 0.000
#> GSM96984 3 0.3419 0.724 0.180 0.000 0.804 0.016 0.000
#> GSM96994 3 0.4295 0.715 0.196 0.000 0.760 0.032 0.012
#> GSM96996 1 0.3876 0.779 0.684 0.000 0.000 0.000 0.316
#> GSM96997 3 0.3456 0.723 0.184 0.000 0.800 0.016 0.000
#> GSM97007 3 0.3419 0.724 0.180 0.000 0.804 0.016 0.000
#> GSM96954 3 0.5814 0.606 0.180 0.000 0.612 0.208 0.000
#> GSM96962 3 0.4703 0.712 0.204 0.000 0.736 0.040 0.020
#> GSM96969 4 0.4930 0.704 0.388 0.000 0.000 0.580 0.032
#> GSM96970 4 0.4846 0.713 0.384 0.000 0.000 0.588 0.028
#> GSM96973 4 0.4372 0.731 0.260 0.000 0.004 0.712 0.024
#> GSM96976 4 0.1074 0.634 0.000 0.016 0.012 0.968 0.004
#> GSM96977 5 0.4659 0.589 0.220 0.004 0.016 0.028 0.732
#> GSM96995 3 0.7026 0.522 0.040 0.052 0.544 0.052 0.312
#> GSM97002 1 0.3586 0.829 0.736 0.000 0.000 0.000 0.264
#> GSM97009 5 0.3612 0.617 0.016 0.100 0.004 0.036 0.844
#> GSM97010 1 0.5987 0.520 0.536 0.024 0.012 0.036 0.392
#> GSM96974 4 0.0865 0.630 0.000 0.000 0.024 0.972 0.004
#> GSM96985 3 0.1281 0.737 0.032 0.000 0.956 0.012 0.000
#> GSM96959 2 0.5938 0.368 0.004 0.528 0.036 0.032 0.400
#> GSM96972 4 0.4966 0.682 0.404 0.000 0.000 0.564 0.032
#> GSM96978 3 0.1281 0.737 0.032 0.000 0.956 0.012 0.000
#> GSM96967 4 0.4920 0.709 0.384 0.000 0.000 0.584 0.032
#> GSM96987 5 0.4045 0.387 0.356 0.000 0.000 0.000 0.644
#> GSM97011 5 0.3424 0.621 0.016 0.092 0.004 0.032 0.856
#> GSM96964 5 0.4101 0.350 0.372 0.000 0.000 0.000 0.628
#> GSM96965 4 0.1565 0.639 0.004 0.016 0.008 0.952 0.020
#> GSM96981 1 0.3814 0.820 0.720 0.000 0.000 0.004 0.276
#> GSM96982 1 0.3814 0.820 0.720 0.000 0.000 0.004 0.276
#> GSM96988 3 0.5143 0.637 0.024 0.032 0.712 0.012 0.220
#> GSM97000 5 0.4094 0.650 0.084 0.048 0.008 0.032 0.828
#> GSM97004 1 0.3461 0.824 0.772 0.000 0.000 0.004 0.224
#> GSM97008 5 0.3767 0.643 0.168 0.008 0.000 0.024 0.800
#> GSM96950 5 0.4070 0.571 0.256 0.004 0.000 0.012 0.728
#> GSM96980 1 0.5680 0.492 0.628 0.000 0.000 0.212 0.160
#> GSM96989 5 0.4045 0.387 0.356 0.000 0.000 0.000 0.644
#> GSM96992 1 0.3508 0.834 0.748 0.000 0.000 0.000 0.252
#> GSM96993 5 0.3766 0.550 0.268 0.004 0.000 0.000 0.728
#> GSM96958 5 0.4877 0.451 0.312 0.000 0.012 0.024 0.652
#> GSM96951 1 0.4101 0.670 0.628 0.000 0.000 0.000 0.372
#> GSM96952 1 0.3508 0.834 0.748 0.000 0.000 0.000 0.252
#> GSM96961 1 0.3534 0.833 0.744 0.000 0.000 0.000 0.256
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM97038 2 0.5206 0.624 0.000 0.696 0.092 0.004 0.160 0.048
#> GSM97045 2 0.1075 0.893 0.000 0.952 0.000 0.000 0.048 0.000
#> GSM97047 5 0.4181 0.652 0.000 0.064 0.076 0.012 0.800 0.048
#> GSM97025 2 0.1007 0.895 0.000 0.956 0.000 0.000 0.044 0.000
#> GSM97030 3 0.3466 0.684 0.000 0.096 0.816 0.000 0.004 0.084
#> GSM97027 2 0.1075 0.893 0.000 0.952 0.000 0.000 0.048 0.000
#> GSM97033 2 0.0632 0.905 0.000 0.976 0.000 0.000 0.024 0.000
#> GSM97034 5 0.4745 0.701 0.040 0.068 0.060 0.008 0.780 0.044
#> GSM97020 2 0.0937 0.898 0.000 0.960 0.000 0.000 0.040 0.000
#> GSM97026 5 0.2314 0.742 0.012 0.032 0.016 0.008 0.916 0.016
#> GSM97012 2 0.0000 0.911 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97015 3 0.3095 0.694 0.000 0.052 0.856 0.000 0.020 0.072
#> GSM97016 2 0.0551 0.908 0.000 0.984 0.004 0.000 0.004 0.008
#> GSM97017 5 0.1910 0.752 0.028 0.004 0.016 0.004 0.932 0.016
#> GSM97019 2 0.0146 0.911 0.000 0.996 0.000 0.000 0.004 0.000
#> GSM97022 2 0.0000 0.911 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97035 2 0.0000 0.911 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97036 5 0.4105 0.689 0.236 0.000 0.008 0.000 0.720 0.036
#> GSM97039 2 0.0551 0.908 0.000 0.984 0.004 0.000 0.004 0.008
#> GSM97046 2 0.0551 0.908 0.000 0.984 0.004 0.000 0.004 0.008
#> GSM97023 5 0.4614 0.581 0.336 0.000 0.004 0.004 0.620 0.036
#> GSM97029 5 0.4756 0.708 0.048 0.068 0.052 0.008 0.780 0.044
#> GSM97043 3 0.5323 0.588 0.000 0.204 0.632 0.000 0.152 0.012
#> GSM97013 5 0.4541 0.658 0.272 0.000 0.012 0.004 0.676 0.036
#> GSM96956 3 0.3924 0.617 0.000 0.208 0.740 0.000 0.000 0.052
#> GSM97024 2 0.2955 0.735 0.000 0.816 0.172 0.000 0.004 0.008
#> GSM97032 3 0.4255 0.687 0.000 0.128 0.772 0.000 0.048 0.052
#> GSM97044 3 0.2814 0.685 0.000 0.052 0.864 0.000 0.004 0.080
#> GSM97049 2 0.0551 0.908 0.000 0.984 0.004 0.000 0.004 0.008
#> GSM96968 3 0.6073 0.542 0.028 0.016 0.584 0.008 0.276 0.088
#> GSM96971 6 0.4691 0.777 0.000 0.000 0.124 0.196 0.000 0.680
#> GSM96986 6 0.2346 0.897 0.000 0.000 0.124 0.000 0.008 0.868
#> GSM97003 1 0.0937 0.805 0.960 0.000 0.000 0.000 0.040 0.000
#> GSM96957 5 0.3729 0.733 0.208 0.004 0.012 0.004 0.764 0.008
#> GSM96960 1 0.0291 0.781 0.992 0.000 0.004 0.000 0.000 0.004
#> GSM96975 1 0.5185 0.104 0.552 0.000 0.016 0.008 0.384 0.040
#> GSM96998 1 0.2773 0.733 0.828 0.000 0.004 0.000 0.164 0.004
#> GSM96999 5 0.3729 0.733 0.208 0.004 0.012 0.004 0.764 0.008
#> GSM97001 5 0.3729 0.733 0.208 0.004 0.012 0.004 0.764 0.008
#> GSM97005 5 0.4074 0.668 0.288 0.000 0.000 0.004 0.684 0.024
#> GSM97006 1 0.0291 0.781 0.992 0.000 0.004 0.000 0.000 0.004
#> GSM97021 5 0.1849 0.751 0.024 0.004 0.012 0.008 0.936 0.016
#> GSM97028 3 0.6048 0.537 0.004 0.020 0.572 0.004 0.220 0.180
#> GSM97031 1 0.6219 0.243 0.424 0.000 0.008 0.004 0.200 0.364
#> GSM97037 3 0.3707 0.660 0.000 0.156 0.784 0.000 0.004 0.056
#> GSM97018 3 0.5840 0.614 0.004 0.056 0.632 0.012 0.228 0.068
#> GSM97014 5 0.3365 0.699 0.000 0.044 0.040 0.016 0.856 0.044
#> GSM97042 2 0.0000 0.911 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97040 5 0.3630 0.701 0.012 0.028 0.052 0.016 0.848 0.044
#> GSM97041 5 0.1672 0.753 0.028 0.004 0.016 0.000 0.940 0.012
#> GSM96955 2 0.6160 0.478 0.000 0.588 0.080 0.016 0.252 0.064
#> GSM96990 3 0.3924 0.689 0.000 0.032 0.812 0.008 0.072 0.076
#> GSM96991 2 0.0146 0.911 0.000 0.996 0.004 0.000 0.000 0.000
#> GSM97048 2 0.0551 0.908 0.000 0.984 0.004 0.000 0.004 0.008
#> GSM96963 2 0.0146 0.911 0.000 0.996 0.004 0.000 0.000 0.000
#> GSM96953 2 0.0000 0.911 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM96966 4 0.3807 0.706 0.368 0.000 0.004 0.628 0.000 0.000
#> GSM96979 6 0.2806 0.892 0.000 0.000 0.136 0.004 0.016 0.844
#> GSM96983 3 0.2668 0.578 0.000 0.000 0.828 0.004 0.000 0.168
#> GSM96984 6 0.2491 0.890 0.000 0.000 0.164 0.000 0.000 0.836
#> GSM96994 6 0.2431 0.899 0.000 0.000 0.132 0.000 0.008 0.860
#> GSM96996 1 0.2146 0.774 0.880 0.000 0.000 0.000 0.116 0.004
#> GSM96997 6 0.2454 0.892 0.000 0.000 0.160 0.000 0.000 0.840
#> GSM97007 6 0.2491 0.890 0.000 0.000 0.164 0.000 0.000 0.836
#> GSM96954 6 0.4691 0.777 0.000 0.000 0.124 0.196 0.000 0.680
#> GSM96962 6 0.2806 0.892 0.000 0.000 0.136 0.004 0.016 0.844
#> GSM96969 4 0.3930 0.682 0.420 0.000 0.004 0.576 0.000 0.000
#> GSM96970 4 0.3915 0.692 0.412 0.000 0.004 0.584 0.000 0.000
#> GSM96973 4 0.3309 0.710 0.280 0.000 0.000 0.720 0.000 0.000
#> GSM96976 4 0.0810 0.597 0.000 0.008 0.008 0.976 0.004 0.004
#> GSM96977 5 0.4831 0.701 0.220 0.000 0.032 0.012 0.700 0.036
#> GSM96995 3 0.6073 0.542 0.028 0.016 0.584 0.008 0.276 0.088
#> GSM97002 1 0.0937 0.805 0.960 0.000 0.000 0.000 0.040 0.000
#> GSM97009 5 0.3837 0.700 0.012 0.036 0.056 0.016 0.836 0.044
#> GSM97010 1 0.4676 0.648 0.744 0.012 0.048 0.012 0.168 0.016
#> GSM96974 4 0.0692 0.594 0.000 0.000 0.020 0.976 0.000 0.004
#> GSM96985 3 0.2968 0.575 0.000 0.000 0.816 0.016 0.000 0.168
#> GSM96959 2 0.6537 0.249 0.000 0.464 0.092 0.016 0.372 0.056
#> GSM96972 4 0.3955 0.657 0.436 0.000 0.004 0.560 0.000 0.000
#> GSM96978 3 0.2968 0.575 0.000 0.000 0.816 0.016 0.000 0.168
#> GSM96967 4 0.3923 0.688 0.416 0.000 0.004 0.580 0.000 0.000
#> GSM96987 5 0.4450 0.587 0.336 0.000 0.008 0.000 0.628 0.028
#> GSM97011 5 0.3706 0.703 0.012 0.032 0.048 0.016 0.844 0.048
#> GSM96964 5 0.4353 0.558 0.360 0.000 0.004 0.000 0.612 0.024
#> GSM96965 4 0.1354 0.601 0.004 0.008 0.004 0.956 0.020 0.008
#> GSM96981 1 0.1625 0.801 0.928 0.000 0.000 0.000 0.060 0.012
#> GSM96982 1 0.1625 0.801 0.928 0.000 0.000 0.000 0.060 0.012
#> GSM96988 3 0.6248 0.533 0.004 0.020 0.564 0.012 0.212 0.188
#> GSM97000 5 0.4156 0.734 0.092 0.004 0.048 0.012 0.804 0.040
#> GSM97004 1 0.0291 0.781 0.992 0.000 0.004 0.000 0.000 0.004
#> GSM97008 5 0.4517 0.739 0.184 0.004 0.024 0.016 0.744 0.028
#> GSM96950 5 0.4362 0.698 0.256 0.000 0.016 0.008 0.700 0.020
#> GSM96980 1 0.3134 0.449 0.784 0.000 0.004 0.208 0.004 0.000
#> GSM96989 5 0.4450 0.587 0.336 0.000 0.008 0.000 0.628 0.028
#> GSM96992 1 0.1010 0.804 0.960 0.000 0.000 0.000 0.036 0.004
#> GSM96993 5 0.3960 0.701 0.224 0.000 0.008 0.000 0.736 0.032
#> GSM96958 5 0.5006 0.607 0.316 0.000 0.020 0.008 0.620 0.036
#> GSM96951 1 0.3121 0.713 0.804 0.000 0.000 0.004 0.180 0.012
#> GSM96952 1 0.1010 0.804 0.960 0.000 0.000 0.000 0.036 0.004
#> GSM96961 1 0.1082 0.805 0.956 0.000 0.000 0.000 0.040 0.004
Heatmaps for the consensus matrix. It visualizes the probability of two samples to be in a same group.
consensus_heatmap(res, k = 2)
consensus_heatmap(res, k = 3)
consensus_heatmap(res, k = 4)
consensus_heatmap(res, k = 5)
consensus_heatmap(res, k = 6)
Heatmaps for the membership of samples in all partitions to see how consistent they are:
membership_heatmap(res, k = 2)
membership_heatmap(res, k = 3)
membership_heatmap(res, k = 4)
membership_heatmap(res, k = 5)
membership_heatmap(res, k = 6)
As soon as we have had the classes for columns, we can look for signatures which are significantly different between classes which can be candidate marks for certain classes. Following are the heatmaps for signatures.
Signature heatmaps where rows are scaled:
get_signatures(res, k = 2)
get_signatures(res, k = 3)
get_signatures(res, k = 4)
get_signatures(res, k = 5)
get_signatures(res, k = 6)
Signature heatmaps where rows are not scaled:
get_signatures(res, k = 2, scale_rows = FALSE)
get_signatures(res, k = 3, scale_rows = FALSE)
get_signatures(res, k = 4, scale_rows = FALSE)
get_signatures(res, k = 5, scale_rows = FALSE)
get_signatures(res, k = 6, scale_rows = FALSE)
Compare the overlap of signatures from different k:
compare_signatures(res)
get_signature() returns a data frame invisibly. TO get the list of signatures, the function
call should be assigned to a variable explicitly. In following code, if plot argument is set
to FALSE, no heatmap is plotted while only the differential analysis is performed.
# code only for demonstration
tb = get_signature(res, k = ..., plot = FALSE)
An example of the output of tb is:
#> which_row fdr mean_1 mean_2 scaled_mean_1 scaled_mean_2 km
#> 1 38 0.042760348 8.373488 9.131774 -0.5533452 0.5164555 1
#> 2 40 0.018707592 7.106213 8.469186 -0.6173731 0.5762149 1
#> 3 55 0.019134737 10.221463 11.207825 -0.6159697 0.5749050 1
#> 4 59 0.006059896 5.921854 7.869574 -0.6899429 0.6439467 1
#> 5 60 0.018055526 8.928898 10.211722 -0.6204761 0.5791110 1
#> 6 98 0.009384629 15.714769 14.887706 0.6635654 -0.6193277 2
...
The columns in tb are:
which_row: row indices corresponding to the input matrix.fdr: FDR for the differential test. mean_x: The mean value in group x.scaled_mean_x: The mean value in group x after rows are scaled.km: Row groups if k-means clustering is applied to rows.UMAP plot which shows how samples are separated.
dimension_reduction(res, k = 2, method = "UMAP")
dimension_reduction(res, k = 3, method = "UMAP")
dimension_reduction(res, k = 4, method = "UMAP")
dimension_reduction(res, k = 5, method = "UMAP")
dimension_reduction(res, k = 6, method = "UMAP")
Following heatmap shows how subgroups are split when increasing k:
collect_classes(res)
Test correlation between subgroups and known annotations. If the known annotation is numeric, one-way ANOVA test is applied, and if the known annotation is discrete, chi-squared contingency table test is applied.
test_to_known_factors(res)
#> n disease.state(p) specimen(p) cell.type(p) other(p) k
#> MAD:hclust 78 0.000412 0.1991 7.19e-12 0.00845 2
#> MAD:hclust 95 0.002696 0.1882 1.66e-13 0.03007 3
#> MAD:hclust 95 0.001093 0.0924 8.74e-15 0.03654 4
#> MAD:hclust 90 0.000191 0.1509 8.12e-13 0.09718 5
#> MAD:hclust 95 0.000267 0.2186 3.97e-16 0.02322 6
If matrix rows can be associated to genes, consider to use functional_enrichment(res,
...) to perform function enrichment for the signature genes. See this vignette for more detailed explanations.
The object with results only for a single top-value method and a single partition method can be extracted as:
res = res_list["MAD", "kmeans"]
# you can also extract it by
# res = res_list["MAD:kmeans"]
A summary of res and all the functions that can be applied to it:
res
#> A 'ConsensusPartition' object with k = 2, 3, 4, 5, 6.
#> On a matrix with 21168 rows and 100 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 1.000 0.963 0.985 0.4866 0.519 0.519
#> 3 3 0.727 0.827 0.904 0.3351 0.759 0.566
#> 4 4 0.801 0.744 0.864 0.1391 0.827 0.549
#> 5 5 0.717 0.774 0.828 0.0624 0.909 0.667
#> 6 6 0.747 0.731 0.787 0.0447 0.967 0.844
suggest_best_k() suggests the best \(k\) based on these statistics. The rules are as follows:
suggest_best_k(res)
#> [1] 2
Following shows the table of the partitions (You need to click the show/hide
code output link to see it). The membership matrix (columns with name p*)
is inferred by
clue::cl_consensus()
function with the SE method. Basically the value in the membership matrix
represents the probability to belong to a certain group. The finall class
label for an item is determined with the group with highest probability it
belongs to.
In get_classes() function, the entropy is calculated from the membership
matrix and the silhouette score is calculated from the consensus matrix.
cbind(get_classes(res, k = 2), get_membership(res, k = 2))
#> class entropy silhouette p1 p2
#> GSM97038 2 0.0000 0.998 0.000 1.000
#> GSM97045 2 0.0000 0.998 0.000 1.000
#> GSM97047 2 0.0000 0.998 0.000 1.000
#> GSM97025 2 0.0000 0.998 0.000 1.000
#> GSM97030 2 0.0000 0.998 0.000 1.000
#> GSM97027 2 0.0000 0.998 0.000 1.000
#> GSM97033 2 0.0000 0.998 0.000 1.000
#> GSM97034 2 0.0000 0.998 0.000 1.000
#> GSM97020 2 0.0000 0.998 0.000 1.000
#> GSM97026 2 0.0000 0.998 0.000 1.000
#> GSM97012 2 0.0000 0.998 0.000 1.000
#> GSM97015 2 0.0000 0.998 0.000 1.000
#> GSM97016 2 0.0000 0.998 0.000 1.000
#> GSM97017 1 0.0000 0.975 1.000 0.000
#> GSM97019 2 0.0000 0.998 0.000 1.000
#> GSM97022 2 0.0000 0.998 0.000 1.000
#> GSM97035 2 0.0000 0.998 0.000 1.000
#> GSM97036 1 0.0000 0.975 1.000 0.000
#> GSM97039 2 0.0000 0.998 0.000 1.000
#> GSM97046 2 0.0000 0.998 0.000 1.000
#> GSM97023 1 0.0000 0.975 1.000 0.000
#> GSM97029 1 0.0000 0.975 1.000 0.000
#> GSM97043 2 0.0000 0.998 0.000 1.000
#> GSM97013 1 0.0000 0.975 1.000 0.000
#> GSM96956 2 0.0000 0.998 0.000 1.000
#> GSM97024 2 0.0000 0.998 0.000 1.000
#> GSM97032 2 0.0000 0.998 0.000 1.000
#> GSM97044 2 0.0000 0.998 0.000 1.000
#> GSM97049 2 0.0000 0.998 0.000 1.000
#> GSM96968 1 0.0000 0.975 1.000 0.000
#> GSM96971 1 0.0000 0.975 1.000 0.000
#> GSM96986 1 0.0000 0.975 1.000 0.000
#> GSM97003 1 0.0000 0.975 1.000 0.000
#> GSM96957 1 0.0000 0.975 1.000 0.000
#> GSM96960 1 0.0000 0.975 1.000 0.000
#> GSM96975 1 0.0000 0.975 1.000 0.000
#> GSM96998 1 0.0000 0.975 1.000 0.000
#> GSM96999 1 0.0000 0.975 1.000 0.000
#> GSM97001 1 0.0000 0.975 1.000 0.000
#> GSM97005 1 0.0000 0.975 1.000 0.000
#> GSM97006 1 0.0000 0.975 1.000 0.000
#> GSM97021 1 0.0000 0.975 1.000 0.000
#> GSM97028 1 0.9087 0.533 0.676 0.324
#> GSM97031 1 0.0000 0.975 1.000 0.000
#> GSM97037 2 0.0000 0.998 0.000 1.000
#> GSM97018 2 0.0000 0.998 0.000 1.000
#> GSM97014 2 0.0376 0.995 0.004 0.996
#> GSM97042 2 0.0000 0.998 0.000 1.000
#> GSM97040 1 0.9944 0.190 0.544 0.456
#> GSM97041 1 0.0000 0.975 1.000 0.000
#> GSM96955 2 0.0000 0.998 0.000 1.000
#> GSM96990 2 0.0000 0.998 0.000 1.000
#> GSM96991 2 0.0000 0.998 0.000 1.000
#> GSM97048 2 0.0000 0.998 0.000 1.000
#> GSM96963 2 0.0000 0.998 0.000 1.000
#> GSM96953 2 0.0000 0.998 0.000 1.000
#> GSM96966 1 0.0000 0.975 1.000 0.000
#> GSM96979 1 0.0000 0.975 1.000 0.000
#> GSM96983 2 0.0000 0.998 0.000 1.000
#> GSM96984 1 0.6801 0.780 0.820 0.180
#> GSM96994 2 0.0672 0.991 0.008 0.992
#> GSM96996 1 0.0000 0.975 1.000 0.000
#> GSM96997 1 0.0000 0.975 1.000 0.000
#> GSM97007 2 0.0672 0.991 0.008 0.992
#> GSM96954 1 0.0000 0.975 1.000 0.000
#> GSM96962 1 0.0000 0.975 1.000 0.000
#> GSM96969 1 0.0000 0.975 1.000 0.000
#> GSM96970 1 0.0000 0.975 1.000 0.000
#> GSM96973 1 0.0000 0.975 1.000 0.000
#> GSM96976 1 0.7674 0.719 0.776 0.224
#> GSM96977 1 0.0000 0.975 1.000 0.000
#> GSM96995 1 0.8661 0.612 0.712 0.288
#> GSM97002 1 0.0000 0.975 1.000 0.000
#> GSM97009 2 0.2948 0.944 0.052 0.948
#> GSM97010 1 0.0000 0.975 1.000 0.000
#> GSM96974 1 0.0000 0.975 1.000 0.000
#> GSM96985 1 0.0000 0.975 1.000 0.000
#> GSM96959 2 0.0000 0.998 0.000 1.000
#> GSM96972 1 0.0000 0.975 1.000 0.000
#> GSM96978 1 0.0000 0.975 1.000 0.000
#> GSM96967 1 0.0000 0.975 1.000 0.000
#> GSM96987 1 0.0000 0.975 1.000 0.000
#> GSM97011 1 0.0000 0.975 1.000 0.000
#> GSM96964 1 0.0000 0.975 1.000 0.000
#> GSM96965 1 0.0000 0.975 1.000 0.000
#> GSM96981 1 0.0000 0.975 1.000 0.000
#> GSM96982 1 0.0000 0.975 1.000 0.000
#> GSM96988 1 0.0000 0.975 1.000 0.000
#> GSM97000 1 0.0000 0.975 1.000 0.000
#> GSM97004 1 0.0000 0.975 1.000 0.000
#> GSM97008 1 0.0000 0.975 1.000 0.000
#> GSM96950 1 0.0000 0.975 1.000 0.000
#> GSM96980 1 0.0000 0.975 1.000 0.000
#> GSM96989 1 0.0000 0.975 1.000 0.000
#> GSM96992 1 0.0000 0.975 1.000 0.000
#> GSM96993 1 0.0000 0.975 1.000 0.000
#> GSM96958 1 0.0000 0.975 1.000 0.000
#> GSM96951 1 0.0000 0.975 1.000 0.000
#> GSM96952 1 0.0000 0.975 1.000 0.000
#> GSM96961 1 0.0000 0.975 1.000 0.000
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM97038 2 0.0000 0.9386 0.000 1.000 0.000
#> GSM97045 2 0.0237 0.9395 0.000 0.996 0.004
#> GSM97047 2 0.4733 0.6897 0.004 0.800 0.196
#> GSM97025 2 0.0237 0.9395 0.000 0.996 0.004
#> GSM97030 3 0.5733 0.6319 0.000 0.324 0.676
#> GSM97027 2 0.0237 0.9395 0.000 0.996 0.004
#> GSM97033 2 0.0000 0.9386 0.000 1.000 0.000
#> GSM97034 3 0.5706 0.6377 0.000 0.320 0.680
#> GSM97020 2 0.0000 0.9386 0.000 1.000 0.000
#> GSM97026 2 0.2590 0.8712 0.004 0.924 0.072
#> GSM97012 2 0.0237 0.9395 0.000 0.996 0.004
#> GSM97015 3 0.5706 0.6377 0.000 0.320 0.680
#> GSM97016 2 0.0000 0.9386 0.000 1.000 0.000
#> GSM97017 1 0.0237 0.9159 0.996 0.000 0.004
#> GSM97019 2 0.0237 0.9395 0.000 0.996 0.004
#> GSM97022 2 0.0237 0.9395 0.000 0.996 0.004
#> GSM97035 2 0.0237 0.9395 0.000 0.996 0.004
#> GSM97036 1 0.0237 0.9159 0.996 0.000 0.004
#> GSM97039 2 0.0000 0.9386 0.000 1.000 0.000
#> GSM97046 2 0.0000 0.9386 0.000 1.000 0.000
#> GSM97023 1 0.0237 0.9160 0.996 0.000 0.004
#> GSM97029 1 0.0237 0.9159 0.996 0.000 0.004
#> GSM97043 2 0.0237 0.9395 0.000 0.996 0.004
#> GSM97013 1 0.0237 0.9159 0.996 0.000 0.004
#> GSM96956 2 0.6244 -0.0575 0.000 0.560 0.440
#> GSM97024 2 0.0237 0.9395 0.000 0.996 0.004
#> GSM97032 3 0.5706 0.6377 0.000 0.320 0.680
#> GSM97044 3 0.5706 0.6377 0.000 0.320 0.680
#> GSM97049 2 0.0000 0.9386 0.000 1.000 0.000
#> GSM96968 3 0.3412 0.8014 0.124 0.000 0.876
#> GSM96971 3 0.0237 0.7995 0.004 0.000 0.996
#> GSM96986 3 0.2625 0.8181 0.084 0.000 0.916
#> GSM97003 1 0.2878 0.8804 0.904 0.000 0.096
#> GSM96957 1 0.0237 0.9159 0.996 0.000 0.004
#> GSM96960 1 0.3038 0.8750 0.896 0.000 0.104
#> GSM96975 1 0.0237 0.9159 0.996 0.000 0.004
#> GSM96998 1 0.0237 0.9160 0.996 0.000 0.004
#> GSM96999 1 0.0237 0.9159 0.996 0.000 0.004
#> GSM97001 1 0.0237 0.9159 0.996 0.000 0.004
#> GSM97005 1 0.0237 0.9159 0.996 0.000 0.004
#> GSM97006 1 0.2625 0.8860 0.916 0.000 0.084
#> GSM97021 1 0.0747 0.9113 0.984 0.000 0.016
#> GSM97028 3 0.3009 0.8205 0.052 0.028 0.920
#> GSM97031 1 0.1860 0.9022 0.948 0.000 0.052
#> GSM97037 3 0.6267 0.3534 0.000 0.452 0.548
#> GSM97018 3 0.5706 0.6377 0.000 0.320 0.680
#> GSM97014 2 0.3805 0.8223 0.092 0.884 0.024
#> GSM97042 2 0.0237 0.9395 0.000 0.996 0.004
#> GSM97040 1 0.4842 0.6795 0.776 0.000 0.224
#> GSM97041 1 0.0237 0.9159 0.996 0.000 0.004
#> GSM96955 2 0.1399 0.9132 0.004 0.968 0.028
#> GSM96990 3 0.5706 0.6377 0.000 0.320 0.680
#> GSM96991 2 0.0237 0.9395 0.000 0.996 0.004
#> GSM97048 2 0.0000 0.9386 0.000 1.000 0.000
#> GSM96963 2 0.0237 0.9395 0.000 0.996 0.004
#> GSM96953 2 0.0237 0.9395 0.000 0.996 0.004
#> GSM96966 1 0.5706 0.6874 0.680 0.000 0.320
#> GSM96979 3 0.2625 0.8181 0.084 0.000 0.916
#> GSM96983 3 0.2711 0.8017 0.000 0.088 0.912
#> GSM96984 3 0.2625 0.8181 0.084 0.000 0.916
#> GSM96994 3 0.2774 0.8092 0.008 0.072 0.920
#> GSM96996 1 0.0592 0.9151 0.988 0.000 0.012
#> GSM96997 3 0.2625 0.8181 0.084 0.000 0.916
#> GSM97007 3 0.2774 0.8092 0.008 0.072 0.920
#> GSM96954 3 0.2711 0.8168 0.088 0.000 0.912
#> GSM96962 3 0.2625 0.8181 0.084 0.000 0.916
#> GSM96969 1 0.5706 0.6874 0.680 0.000 0.320
#> GSM96970 1 0.5706 0.6874 0.680 0.000 0.320
#> GSM96973 1 0.5706 0.6874 0.680 0.000 0.320
#> GSM96976 3 0.0424 0.8001 0.008 0.000 0.992
#> GSM96977 1 0.4291 0.7490 0.820 0.000 0.180
#> GSM96995 3 0.4953 0.7624 0.176 0.016 0.808
#> GSM97002 1 0.2356 0.8934 0.928 0.000 0.072
#> GSM97009 2 0.8765 0.3523 0.252 0.580 0.168
#> GSM97010 1 0.2796 0.8649 0.908 0.000 0.092
#> GSM96974 3 0.0424 0.7991 0.008 0.000 0.992
#> GSM96985 3 0.4750 0.5908 0.216 0.000 0.784
#> GSM96959 3 0.6835 0.6594 0.040 0.284 0.676
#> GSM96972 1 0.5706 0.6874 0.680 0.000 0.320
#> GSM96978 3 0.1163 0.8090 0.028 0.000 0.972
#> GSM96967 1 0.5706 0.6874 0.680 0.000 0.320
#> GSM96987 1 0.0237 0.9160 0.996 0.000 0.004
#> GSM97011 1 0.1031 0.9071 0.976 0.000 0.024
#> GSM96964 1 0.0237 0.9160 0.996 0.000 0.004
#> GSM96965 1 0.5178 0.7472 0.744 0.000 0.256
#> GSM96981 1 0.0424 0.9144 0.992 0.000 0.008
#> GSM96982 1 0.1289 0.9098 0.968 0.000 0.032
#> GSM96988 3 0.2625 0.8181 0.084 0.000 0.916
#> GSM97000 1 0.4796 0.6861 0.780 0.000 0.220
#> GSM97004 1 0.2625 0.8868 0.916 0.000 0.084
#> GSM97008 1 0.1163 0.9052 0.972 0.000 0.028
#> GSM96950 1 0.0237 0.9159 0.996 0.000 0.004
#> GSM96980 1 0.4002 0.8412 0.840 0.000 0.160
#> GSM96989 1 0.0237 0.9160 0.996 0.000 0.004
#> GSM96992 1 0.0747 0.9143 0.984 0.000 0.016
#> GSM96993 1 0.0237 0.9159 0.996 0.000 0.004
#> GSM96958 1 0.0237 0.9160 0.996 0.000 0.004
#> GSM96951 1 0.0237 0.9160 0.996 0.000 0.004
#> GSM96952 1 0.0424 0.9153 0.992 0.000 0.008
#> GSM96961 1 0.0237 0.9160 0.996 0.000 0.004
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM97038 2 0.1118 0.965 0.000 0.964 0.000 0.036
#> GSM97045 2 0.0000 0.971 0.000 1.000 0.000 0.000
#> GSM97047 1 0.6265 0.404 0.644 0.284 0.056 0.016
#> GSM97025 2 0.0000 0.971 0.000 1.000 0.000 0.000
#> GSM97030 3 0.1510 0.919 0.016 0.028 0.956 0.000
#> GSM97027 2 0.0000 0.971 0.000 1.000 0.000 0.000
#> GSM97033 2 0.1118 0.965 0.000 0.964 0.000 0.036
#> GSM97034 3 0.1520 0.920 0.024 0.020 0.956 0.000
#> GSM97020 2 0.1118 0.965 0.000 0.964 0.000 0.036
#> GSM97026 1 0.5668 0.370 0.636 0.328 0.032 0.004
#> GSM97012 2 0.0000 0.971 0.000 1.000 0.000 0.000
#> GSM97015 3 0.1520 0.920 0.024 0.020 0.956 0.000
#> GSM97016 2 0.1118 0.965 0.000 0.964 0.000 0.036
#> GSM97017 1 0.0592 0.799 0.984 0.000 0.000 0.016
#> GSM97019 2 0.0000 0.971 0.000 1.000 0.000 0.000
#> GSM97022 2 0.0000 0.971 0.000 1.000 0.000 0.000
#> GSM97035 2 0.0000 0.971 0.000 1.000 0.000 0.000
#> GSM97036 1 0.1211 0.790 0.960 0.000 0.000 0.040
#> GSM97039 2 0.1118 0.965 0.000 0.964 0.000 0.036
#> GSM97046 2 0.1118 0.965 0.000 0.964 0.000 0.036
#> GSM97023 1 0.4855 0.153 0.644 0.000 0.004 0.352
#> GSM97029 1 0.0921 0.797 0.972 0.000 0.000 0.028
#> GSM97043 2 0.0657 0.962 0.012 0.984 0.004 0.000
#> GSM97013 1 0.1398 0.788 0.956 0.000 0.004 0.040
#> GSM96956 3 0.5430 0.629 0.008 0.252 0.704 0.036
#> GSM97024 2 0.0000 0.971 0.000 1.000 0.000 0.000
#> GSM97032 3 0.1510 0.919 0.016 0.028 0.956 0.000
#> GSM97044 3 0.1510 0.919 0.016 0.028 0.956 0.000
#> GSM97049 2 0.1118 0.965 0.000 0.964 0.000 0.036
#> GSM96968 3 0.0921 0.920 0.028 0.000 0.972 0.000
#> GSM96971 3 0.0921 0.919 0.000 0.000 0.972 0.028
#> GSM96986 3 0.0707 0.922 0.000 0.000 0.980 0.020
#> GSM97003 4 0.5127 0.653 0.356 0.000 0.012 0.632
#> GSM96957 1 0.0817 0.798 0.976 0.000 0.000 0.024
#> GSM96960 4 0.4819 0.661 0.344 0.000 0.004 0.652
#> GSM96975 1 0.0921 0.797 0.972 0.000 0.000 0.028
#> GSM96998 4 0.5088 0.579 0.424 0.000 0.004 0.572
#> GSM96999 1 0.1004 0.797 0.972 0.000 0.004 0.024
#> GSM97001 1 0.0592 0.799 0.984 0.000 0.000 0.016
#> GSM97005 1 0.0592 0.799 0.984 0.000 0.000 0.016
#> GSM97006 4 0.4872 0.656 0.356 0.000 0.004 0.640
#> GSM97021 1 0.0000 0.796 1.000 0.000 0.000 0.000
#> GSM97028 3 0.0657 0.924 0.012 0.004 0.984 0.000
#> GSM97031 1 0.4792 0.251 0.680 0.000 0.008 0.312
#> GSM97037 3 0.4282 0.800 0.016 0.140 0.820 0.024
#> GSM97018 3 0.1520 0.920 0.024 0.020 0.956 0.000
#> GSM97014 1 0.5254 0.418 0.672 0.300 0.000 0.028
#> GSM97042 2 0.0000 0.971 0.000 1.000 0.000 0.000
#> GSM97040 1 0.1661 0.755 0.944 0.000 0.052 0.004
#> GSM97041 1 0.0336 0.798 0.992 0.000 0.000 0.008
#> GSM96955 2 0.5460 0.481 0.340 0.632 0.000 0.028
#> GSM96990 3 0.1520 0.920 0.024 0.020 0.956 0.000
#> GSM96991 2 0.0000 0.971 0.000 1.000 0.000 0.000
#> GSM97048 2 0.1118 0.965 0.000 0.964 0.000 0.036
#> GSM96963 2 0.0000 0.971 0.000 1.000 0.000 0.000
#> GSM96953 2 0.0000 0.971 0.000 1.000 0.000 0.000
#> GSM96966 4 0.2363 0.637 0.056 0.000 0.024 0.920
#> GSM96979 3 0.0707 0.922 0.000 0.000 0.980 0.020
#> GSM96983 3 0.0376 0.923 0.000 0.004 0.992 0.004
#> GSM96984 3 0.0707 0.922 0.000 0.000 0.980 0.020
#> GSM96994 3 0.0895 0.922 0.000 0.004 0.976 0.020
#> GSM96996 4 0.5060 0.598 0.412 0.000 0.004 0.584
#> GSM96997 3 0.0707 0.922 0.000 0.000 0.980 0.020
#> GSM97007 3 0.0895 0.922 0.000 0.004 0.976 0.020
#> GSM96954 3 0.0895 0.922 0.020 0.000 0.976 0.004
#> GSM96962 3 0.0707 0.922 0.000 0.000 0.980 0.020
#> GSM96969 4 0.2363 0.637 0.056 0.000 0.024 0.920
#> GSM96970 4 0.2363 0.637 0.056 0.000 0.024 0.920
#> GSM96973 4 0.2363 0.637 0.056 0.000 0.024 0.920
#> GSM96976 3 0.4888 0.450 0.000 0.000 0.588 0.412
#> GSM96977 1 0.1389 0.761 0.952 0.000 0.048 0.000
#> GSM96995 3 0.1637 0.904 0.060 0.000 0.940 0.000
#> GSM97002 4 0.4837 0.660 0.348 0.000 0.004 0.648
#> GSM97009 1 0.3674 0.687 0.868 0.084 0.028 0.020
#> GSM97010 1 0.1388 0.795 0.960 0.000 0.012 0.028
#> GSM96974 4 0.4925 -0.101 0.000 0.000 0.428 0.572
#> GSM96985 4 0.4722 0.312 0.008 0.000 0.300 0.692
#> GSM96959 3 0.5396 0.214 0.464 0.000 0.524 0.012
#> GSM96972 4 0.2363 0.637 0.056 0.000 0.024 0.920
#> GSM96978 3 0.0469 0.922 0.000 0.000 0.988 0.012
#> GSM96967 4 0.2363 0.637 0.056 0.000 0.024 0.920
#> GSM96987 4 0.5105 0.564 0.432 0.000 0.004 0.564
#> GSM97011 1 0.0000 0.796 1.000 0.000 0.000 0.000
#> GSM96964 1 0.4905 0.108 0.632 0.000 0.004 0.364
#> GSM96965 4 0.5476 0.170 0.396 0.000 0.020 0.584
#> GSM96981 4 0.4907 0.594 0.420 0.000 0.000 0.580
#> GSM96982 4 0.4872 0.656 0.356 0.000 0.004 0.640
#> GSM96988 3 0.0188 0.923 0.000 0.000 0.996 0.004
#> GSM97000 1 0.1211 0.768 0.960 0.000 0.040 0.000
#> GSM97004 4 0.4781 0.662 0.336 0.000 0.004 0.660
#> GSM97008 1 0.0000 0.796 1.000 0.000 0.000 0.000
#> GSM96950 1 0.1489 0.785 0.952 0.000 0.004 0.044
#> GSM96980 4 0.1557 0.638 0.056 0.000 0.000 0.944
#> GSM96989 4 0.5105 0.564 0.432 0.000 0.004 0.564
#> GSM96992 4 0.4964 0.637 0.380 0.000 0.004 0.616
#> GSM96993 1 0.1022 0.795 0.968 0.000 0.000 0.032
#> GSM96958 1 0.4837 0.166 0.648 0.000 0.004 0.348
#> GSM96951 1 0.4889 0.124 0.636 0.000 0.004 0.360
#> GSM96952 4 0.4964 0.637 0.380 0.000 0.004 0.616
#> GSM96961 4 0.5112 0.555 0.436 0.000 0.004 0.560
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM97038 2 0.3474 0.901 0.000 0.836 0.004 0.116 0.044
#> GSM97045 2 0.0579 0.933 0.000 0.984 0.000 0.008 0.008
#> GSM97047 5 0.3433 0.690 0.000 0.032 0.132 0.004 0.832
#> GSM97025 2 0.0451 0.933 0.000 0.988 0.000 0.004 0.008
#> GSM97030 3 0.1442 0.845 0.000 0.004 0.952 0.012 0.032
#> GSM97027 2 0.0579 0.933 0.000 0.984 0.000 0.008 0.008
#> GSM97033 2 0.3165 0.906 0.000 0.848 0.000 0.116 0.036
#> GSM97034 3 0.1618 0.839 0.000 0.008 0.944 0.008 0.040
#> GSM97020 2 0.3291 0.904 0.000 0.840 0.000 0.120 0.040
#> GSM97026 5 0.4429 0.711 0.012 0.076 0.080 0.024 0.808
#> GSM97012 2 0.0451 0.934 0.000 0.988 0.008 0.004 0.000
#> GSM97015 3 0.1285 0.841 0.000 0.004 0.956 0.004 0.036
#> GSM97016 2 0.3242 0.905 0.000 0.844 0.000 0.116 0.040
#> GSM97017 5 0.2997 0.821 0.148 0.000 0.000 0.012 0.840
#> GSM97019 2 0.0579 0.933 0.000 0.984 0.008 0.008 0.000
#> GSM97022 2 0.0451 0.934 0.000 0.988 0.008 0.004 0.000
#> GSM97035 2 0.0451 0.934 0.000 0.988 0.008 0.004 0.000
#> GSM97036 5 0.5124 0.684 0.316 0.000 0.012 0.036 0.636
#> GSM97039 2 0.3165 0.906 0.000 0.848 0.000 0.116 0.036
#> GSM97046 2 0.3242 0.905 0.000 0.844 0.000 0.116 0.040
#> GSM97023 1 0.3123 0.740 0.828 0.000 0.000 0.012 0.160
#> GSM97029 5 0.4649 0.776 0.232 0.000 0.012 0.036 0.720
#> GSM97043 2 0.2868 0.868 0.000 0.884 0.072 0.012 0.032
#> GSM97013 5 0.4679 0.692 0.316 0.000 0.000 0.032 0.652
#> GSM96956 3 0.5859 0.570 0.000 0.168 0.676 0.116 0.040
#> GSM97024 2 0.0579 0.933 0.000 0.984 0.008 0.008 0.000
#> GSM97032 3 0.1538 0.839 0.000 0.008 0.948 0.008 0.036
#> GSM97044 3 0.1173 0.847 0.000 0.004 0.964 0.012 0.020
#> GSM97049 2 0.3242 0.905 0.000 0.844 0.000 0.116 0.040
#> GSM96968 3 0.1197 0.842 0.000 0.000 0.952 0.000 0.048
#> GSM96971 3 0.5312 0.693 0.008 0.000 0.628 0.308 0.056
#> GSM96986 3 0.4792 0.783 0.008 0.000 0.712 0.228 0.052
#> GSM97003 1 0.2787 0.708 0.856 0.000 0.004 0.136 0.004
#> GSM96957 5 0.3727 0.796 0.216 0.000 0.000 0.016 0.768
#> GSM96960 1 0.1121 0.761 0.956 0.000 0.000 0.044 0.000
#> GSM96975 5 0.3759 0.795 0.220 0.000 0.000 0.016 0.764
#> GSM96998 1 0.2193 0.800 0.912 0.000 0.000 0.028 0.060
#> GSM96999 5 0.4249 0.729 0.296 0.000 0.000 0.016 0.688
#> GSM97001 5 0.3039 0.820 0.152 0.000 0.000 0.012 0.836
#> GSM97005 5 0.2930 0.818 0.164 0.000 0.000 0.004 0.832
#> GSM97006 1 0.1121 0.761 0.956 0.000 0.000 0.044 0.000
#> GSM97021 5 0.2976 0.823 0.132 0.000 0.004 0.012 0.852
#> GSM97028 3 0.1364 0.846 0.000 0.000 0.952 0.012 0.036
#> GSM97031 1 0.4873 0.548 0.688 0.000 0.000 0.068 0.244
#> GSM97037 3 0.3937 0.752 0.000 0.072 0.832 0.052 0.044
#> GSM97018 3 0.1695 0.838 0.000 0.008 0.940 0.008 0.044
#> GSM97014 5 0.2337 0.746 0.004 0.080 0.004 0.008 0.904
#> GSM97042 2 0.0451 0.934 0.000 0.988 0.008 0.004 0.000
#> GSM97040 5 0.2813 0.796 0.064 0.000 0.048 0.004 0.884
#> GSM97041 5 0.2997 0.821 0.148 0.000 0.000 0.012 0.840
#> GSM96955 5 0.5192 0.214 0.000 0.388 0.032 0.008 0.572
#> GSM96990 3 0.1285 0.841 0.000 0.004 0.956 0.004 0.036
#> GSM96991 2 0.0451 0.934 0.000 0.988 0.008 0.004 0.000
#> GSM97048 2 0.3242 0.905 0.000 0.844 0.000 0.116 0.040
#> GSM96963 2 0.0613 0.934 0.000 0.984 0.008 0.004 0.004
#> GSM96953 2 0.0613 0.934 0.000 0.984 0.008 0.004 0.004
#> GSM96966 4 0.4138 0.797 0.384 0.000 0.000 0.616 0.000
#> GSM96979 3 0.4792 0.783 0.008 0.000 0.712 0.228 0.052
#> GSM96983 3 0.1331 0.845 0.000 0.000 0.952 0.040 0.008
#> GSM96984 3 0.4792 0.783 0.008 0.000 0.712 0.228 0.052
#> GSM96994 3 0.4764 0.784 0.008 0.000 0.716 0.224 0.052
#> GSM96996 1 0.1872 0.803 0.928 0.000 0.000 0.020 0.052
#> GSM96997 3 0.4898 0.781 0.012 0.000 0.708 0.228 0.052
#> GSM97007 3 0.4764 0.784 0.008 0.000 0.716 0.224 0.052
#> GSM96954 3 0.4045 0.813 0.004 0.000 0.796 0.136 0.064
#> GSM96962 3 0.4792 0.783 0.008 0.000 0.712 0.228 0.052
#> GSM96969 4 0.4138 0.797 0.384 0.000 0.000 0.616 0.000
#> GSM96970 4 0.4138 0.797 0.384 0.000 0.000 0.616 0.000
#> GSM96973 4 0.4138 0.797 0.384 0.000 0.000 0.616 0.000
#> GSM96976 4 0.4459 0.504 0.052 0.000 0.200 0.744 0.004
#> GSM96977 5 0.3405 0.815 0.104 0.000 0.036 0.012 0.848
#> GSM96995 3 0.1410 0.837 0.000 0.000 0.940 0.000 0.060
#> GSM97002 1 0.1408 0.766 0.948 0.000 0.000 0.044 0.008
#> GSM97009 5 0.2833 0.807 0.084 0.008 0.020 0.004 0.884
#> GSM97010 5 0.4315 0.730 0.276 0.000 0.000 0.024 0.700
#> GSM96974 4 0.5083 0.616 0.120 0.000 0.184 0.696 0.000
#> GSM96985 1 0.7043 -0.307 0.416 0.000 0.296 0.276 0.012
#> GSM96959 5 0.3895 0.470 0.000 0.000 0.320 0.000 0.680
#> GSM96972 4 0.4126 0.794 0.380 0.000 0.000 0.620 0.000
#> GSM96978 3 0.1740 0.842 0.000 0.000 0.932 0.056 0.012
#> GSM96967 4 0.4138 0.797 0.384 0.000 0.000 0.616 0.000
#> GSM96987 1 0.2423 0.794 0.896 0.000 0.000 0.024 0.080
#> GSM97011 5 0.2583 0.820 0.132 0.000 0.004 0.000 0.864
#> GSM96964 1 0.3602 0.719 0.796 0.000 0.000 0.024 0.180
#> GSM96965 4 0.5725 0.564 0.156 0.000 0.000 0.620 0.224
#> GSM96981 1 0.2270 0.800 0.904 0.000 0.000 0.020 0.076
#> GSM96982 1 0.1310 0.785 0.956 0.000 0.000 0.024 0.020
#> GSM96988 3 0.1872 0.843 0.000 0.000 0.928 0.052 0.020
#> GSM97000 5 0.2835 0.815 0.112 0.000 0.016 0.004 0.868
#> GSM97004 1 0.1341 0.752 0.944 0.000 0.000 0.056 0.000
#> GSM97008 5 0.2719 0.820 0.144 0.000 0.004 0.000 0.852
#> GSM96950 5 0.4975 0.688 0.316 0.000 0.012 0.028 0.644
#> GSM96980 1 0.4015 -0.132 0.652 0.000 0.000 0.348 0.000
#> GSM96989 1 0.2482 0.792 0.892 0.000 0.000 0.024 0.084
#> GSM96992 1 0.1469 0.799 0.948 0.000 0.000 0.016 0.036
#> GSM96993 5 0.5033 0.690 0.312 0.000 0.012 0.032 0.644
#> GSM96958 1 0.3419 0.722 0.804 0.000 0.000 0.016 0.180
#> GSM96951 1 0.2970 0.738 0.828 0.000 0.000 0.004 0.168
#> GSM96952 1 0.1251 0.801 0.956 0.000 0.000 0.008 0.036
#> GSM96961 1 0.1341 0.804 0.944 0.000 0.000 0.000 0.056
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM97038 2 0.4853 0.821 0.008 0.716 0.012 0.060 0.012 0.192
#> GSM97045 2 0.0260 0.891 0.000 0.992 0.000 0.008 0.000 0.000
#> GSM97047 5 0.2062 0.736 0.000 0.004 0.088 0.008 0.900 0.000
#> GSM97025 2 0.0260 0.891 0.000 0.992 0.000 0.008 0.000 0.000
#> GSM97030 3 0.0862 0.730 0.000 0.004 0.972 0.000 0.008 0.016
#> GSM97027 2 0.0260 0.891 0.000 0.992 0.000 0.008 0.000 0.000
#> GSM97033 2 0.4441 0.830 0.008 0.732 0.000 0.064 0.008 0.188
#> GSM97034 3 0.1406 0.739 0.000 0.004 0.952 0.020 0.008 0.016
#> GSM97020 2 0.4441 0.830 0.008 0.732 0.000 0.064 0.008 0.188
#> GSM97026 5 0.5108 0.716 0.012 0.008 0.112 0.036 0.732 0.100
#> GSM97012 2 0.0000 0.892 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97015 3 0.0551 0.739 0.000 0.004 0.984 0.004 0.008 0.000
#> GSM97016 2 0.4610 0.827 0.008 0.724 0.004 0.064 0.008 0.192
#> GSM97017 5 0.2862 0.764 0.052 0.000 0.000 0.020 0.872 0.056
#> GSM97019 2 0.0000 0.892 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97022 2 0.0000 0.892 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97035 2 0.0000 0.892 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97036 5 0.6906 0.450 0.300 0.000 0.016 0.056 0.468 0.160
#> GSM97039 2 0.4471 0.828 0.008 0.728 0.000 0.064 0.008 0.192
#> GSM97046 2 0.4610 0.827 0.008 0.724 0.004 0.064 0.008 0.192
#> GSM97023 1 0.3959 0.757 0.796 0.000 0.000 0.028 0.084 0.092
#> GSM97029 5 0.6563 0.569 0.228 0.000 0.012 0.056 0.544 0.160
#> GSM97043 2 0.2100 0.809 0.000 0.884 0.112 0.004 0.000 0.000
#> GSM97013 5 0.6508 0.494 0.288 0.000 0.004 0.048 0.500 0.160
#> GSM96956 3 0.5182 0.369 0.000 0.104 0.676 0.024 0.004 0.192
#> GSM97024 2 0.0146 0.890 0.000 0.996 0.004 0.000 0.000 0.000
#> GSM97032 3 0.0665 0.739 0.000 0.004 0.980 0.008 0.008 0.000
#> GSM97044 3 0.0858 0.720 0.000 0.004 0.968 0.000 0.000 0.028
#> GSM97049 2 0.4610 0.827 0.008 0.724 0.004 0.064 0.008 0.192
#> GSM96968 3 0.1369 0.731 0.000 0.000 0.952 0.016 0.016 0.016
#> GSM96971 6 0.5373 0.779 0.000 0.000 0.384 0.100 0.004 0.512
#> GSM96986 6 0.3810 0.927 0.000 0.000 0.428 0.000 0.000 0.572
#> GSM97003 1 0.3964 0.646 0.724 0.000 0.000 0.044 0.000 0.232
#> GSM96957 5 0.4870 0.692 0.200 0.000 0.004 0.024 0.700 0.072
#> GSM96960 1 0.2826 0.757 0.856 0.000 0.000 0.092 0.000 0.052
#> GSM96975 5 0.4639 0.697 0.216 0.000 0.008 0.028 0.712 0.036
#> GSM96998 1 0.4014 0.741 0.784 0.000 0.000 0.036 0.044 0.136
#> GSM96999 5 0.5249 0.624 0.264 0.000 0.004 0.024 0.636 0.072
#> GSM97001 5 0.2001 0.776 0.068 0.000 0.000 0.008 0.912 0.012
#> GSM97005 5 0.2186 0.773 0.056 0.000 0.000 0.012 0.908 0.024
#> GSM97006 1 0.2837 0.758 0.856 0.000 0.000 0.088 0.000 0.056
#> GSM97021 5 0.2115 0.776 0.032 0.000 0.000 0.020 0.916 0.032
#> GSM97028 3 0.2190 0.705 0.000 0.000 0.908 0.044 0.008 0.040
#> GSM97031 1 0.6017 0.411 0.504 0.000 0.000 0.020 0.156 0.320
#> GSM97037 3 0.3161 0.612 0.000 0.028 0.852 0.016 0.008 0.096
#> GSM97018 3 0.1579 0.737 0.000 0.004 0.944 0.024 0.008 0.020
#> GSM97014 5 0.1368 0.769 0.004 0.016 0.012 0.008 0.956 0.004
#> GSM97042 2 0.0000 0.892 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97040 5 0.1007 0.764 0.000 0.000 0.044 0.000 0.956 0.000
#> GSM97041 5 0.2981 0.764 0.052 0.000 0.000 0.020 0.864 0.064
#> GSM96955 5 0.4897 0.552 0.000 0.208 0.044 0.028 0.704 0.016
#> GSM96990 3 0.0551 0.739 0.000 0.004 0.984 0.004 0.008 0.000
#> GSM96991 2 0.0146 0.891 0.000 0.996 0.000 0.004 0.000 0.000
#> GSM97048 2 0.4610 0.827 0.008 0.724 0.004 0.064 0.008 0.192
#> GSM96963 2 0.0146 0.891 0.000 0.996 0.000 0.004 0.000 0.000
#> GSM96953 2 0.0146 0.891 0.000 0.996 0.000 0.004 0.000 0.000
#> GSM96966 4 0.2730 0.873 0.192 0.000 0.000 0.808 0.000 0.000
#> GSM96979 6 0.3817 0.930 0.000 0.000 0.432 0.000 0.000 0.568
#> GSM96983 3 0.2854 0.650 0.000 0.000 0.860 0.048 0.004 0.088
#> GSM96984 6 0.3975 0.936 0.000 0.000 0.452 0.004 0.000 0.544
#> GSM96994 6 0.3975 0.936 0.000 0.000 0.452 0.004 0.000 0.544
#> GSM96996 1 0.2782 0.792 0.876 0.000 0.000 0.024 0.032 0.068
#> GSM96997 6 0.3756 0.886 0.000 0.000 0.400 0.000 0.000 0.600
#> GSM97007 6 0.3975 0.936 0.000 0.000 0.452 0.004 0.000 0.544
#> GSM96954 3 0.4442 -0.720 0.000 0.000 0.536 0.020 0.004 0.440
#> GSM96962 6 0.3975 0.936 0.000 0.000 0.452 0.004 0.000 0.544
#> GSM96969 4 0.2730 0.873 0.192 0.000 0.000 0.808 0.000 0.000
#> GSM96970 4 0.2730 0.873 0.192 0.000 0.000 0.808 0.000 0.000
#> GSM96973 4 0.2730 0.873 0.192 0.000 0.000 0.808 0.000 0.000
#> GSM96976 4 0.3865 0.686 0.004 0.000 0.088 0.800 0.012 0.096
#> GSM96977 5 0.4015 0.768 0.044 0.000 0.052 0.036 0.820 0.048
#> GSM96995 3 0.1873 0.700 0.000 0.000 0.924 0.020 0.048 0.008
#> GSM97002 1 0.3128 0.761 0.848 0.000 0.000 0.088 0.012 0.052
#> GSM97009 5 0.1866 0.764 0.016 0.004 0.036 0.008 0.932 0.004
#> GSM97010 5 0.5925 0.633 0.212 0.000 0.020 0.044 0.628 0.096
#> GSM96974 4 0.3897 0.711 0.028 0.000 0.100 0.800 0.000 0.072
#> GSM96985 3 0.7423 -0.035 0.284 0.000 0.356 0.252 0.004 0.104
#> GSM96959 5 0.4071 0.556 0.000 0.000 0.248 0.020 0.716 0.016
#> GSM96972 4 0.2793 0.867 0.200 0.000 0.000 0.800 0.000 0.000
#> GSM96978 3 0.3248 0.592 0.000 0.000 0.828 0.052 0.004 0.116
#> GSM96967 4 0.2730 0.873 0.192 0.000 0.000 0.808 0.000 0.000
#> GSM96987 1 0.4232 0.717 0.772 0.000 0.000 0.044 0.052 0.132
#> GSM97011 5 0.1598 0.771 0.040 0.000 0.008 0.008 0.940 0.004
#> GSM96964 1 0.4850 0.684 0.732 0.000 0.004 0.044 0.088 0.132
#> GSM96965 4 0.3837 0.713 0.068 0.000 0.008 0.784 0.140 0.000
#> GSM96981 1 0.2113 0.794 0.916 0.000 0.004 0.028 0.044 0.008
#> GSM96982 1 0.1900 0.787 0.916 0.000 0.000 0.068 0.008 0.008
#> GSM96988 3 0.3017 0.621 0.000 0.000 0.848 0.052 0.004 0.096
#> GSM97000 5 0.1965 0.772 0.040 0.000 0.004 0.008 0.924 0.024
#> GSM97004 1 0.2747 0.757 0.860 0.000 0.000 0.096 0.000 0.044
#> GSM97008 5 0.1768 0.772 0.040 0.000 0.004 0.004 0.932 0.020
#> GSM96950 5 0.6594 0.470 0.300 0.000 0.004 0.052 0.484 0.160
#> GSM96980 1 0.3717 0.214 0.616 0.000 0.000 0.384 0.000 0.000
#> GSM96989 1 0.4372 0.714 0.768 0.000 0.004 0.044 0.052 0.132
#> GSM96992 1 0.1820 0.791 0.924 0.000 0.000 0.056 0.012 0.008
#> GSM96993 5 0.6881 0.459 0.300 0.000 0.016 0.056 0.472 0.156
#> GSM96958 1 0.4134 0.718 0.784 0.000 0.000 0.040 0.112 0.064
#> GSM96951 1 0.3097 0.779 0.856 0.000 0.000 0.028 0.080 0.036
#> GSM96952 1 0.1434 0.793 0.940 0.000 0.000 0.048 0.012 0.000
#> GSM96961 1 0.1874 0.796 0.928 0.000 0.000 0.028 0.028 0.016
Heatmaps for the consensus matrix. It visualizes the probability of two samples to be in a same group.
consensus_heatmap(res, k = 2)
consensus_heatmap(res, k = 3)
consensus_heatmap(res, k = 4)
consensus_heatmap(res, k = 5)
consensus_heatmap(res, k = 6)
Heatmaps for the membership of samples in all partitions to see how consistent they are:
membership_heatmap(res, k = 2)
membership_heatmap(res, k = 3)
membership_heatmap(res, k = 4)
membership_heatmap(res, k = 5)
membership_heatmap(res, k = 6)
As soon as we have had the classes for columns, we can look for signatures which are significantly different between classes which can be candidate marks for certain classes. Following are the heatmaps for signatures.
Signature heatmaps where rows are scaled:
get_signatures(res, k = 2)
get_signatures(res, k = 3)
get_signatures(res, k = 4)
get_signatures(res, k = 5)
get_signatures(res, k = 6)
Signature heatmaps where rows are not scaled:
get_signatures(res, k = 2, scale_rows = FALSE)
get_signatures(res, k = 3, scale_rows = FALSE)
get_signatures(res, k = 4, scale_rows = FALSE)
get_signatures(res, k = 5, scale_rows = FALSE)
get_signatures(res, k = 6, scale_rows = FALSE)
Compare the overlap of signatures from different k:
compare_signatures(res)
get_signature() returns a data frame invisibly. TO get the list of signatures, the function
call should be assigned to a variable explicitly. In following code, if plot argument is set
to FALSE, no heatmap is plotted while only the differential analysis is performed.
# code only for demonstration
tb = get_signature(res, k = ..., plot = FALSE)
An example of the output of tb is:
#> which_row fdr mean_1 mean_2 scaled_mean_1 scaled_mean_2 km
#> 1 38 0.042760348 8.373488 9.131774 -0.5533452 0.5164555 1
#> 2 40 0.018707592 7.106213 8.469186 -0.6173731 0.5762149 1
#> 3 55 0.019134737 10.221463 11.207825 -0.6159697 0.5749050 1
#> 4 59 0.006059896 5.921854 7.869574 -0.6899429 0.6439467 1
#> 5 60 0.018055526 8.928898 10.211722 -0.6204761 0.5791110 1
#> 6 98 0.009384629 15.714769 14.887706 0.6635654 -0.6193277 2
...
The columns in tb are:
which_row: row indices corresponding to the input matrix.fdr: FDR for the differential test. mean_x: The mean value in group x.scaled_mean_x: The mean value in group x after rows are scaled.km: Row groups if k-means clustering is applied to rows.UMAP plot which shows how samples are separated.
dimension_reduction(res, k = 2, method = "UMAP")
dimension_reduction(res, k = 3, method = "UMAP")
dimension_reduction(res, k = 4, method = "UMAP")
dimension_reduction(res, k = 5, method = "UMAP")
dimension_reduction(res, k = 6, method = "UMAP")
Following heatmap shows how subgroups are split when increasing k:
collect_classes(res)
Test correlation between subgroups and known annotations. If the known annotation is numeric, one-way ANOVA test is applied, and if the known annotation is discrete, chi-squared contingency table test is applied.
test_to_known_factors(res)
#> n disease.state(p) specimen(p) cell.type(p) other(p) k
#> MAD:kmeans 99 4.79e-06 0.212 5.56e-16 0.0326 2
#> MAD:kmeans 97 3.86e-05 0.294 5.38e-19 0.0178 3
#> MAD:kmeans 86 4.63e-05 0.238 4.32e-13 0.0220 4
#> MAD:kmeans 96 5.29e-05 0.214 2.61e-15 0.1319 5
#> MAD:kmeans 91 7.04e-05 0.388 1.34e-15 0.0362 6
If matrix rows can be associated to genes, consider to use functional_enrichment(res,
...) to perform function enrichment for the signature genes. See this vignette for more detailed explanations.
The object with results only for a single top-value method and a single partition method can be extracted as:
res = res_list["MAD", "skmeans"]
# you can also extract it by
# res = res_list["MAD:skmeans"]
A summary of res and all the functions that can be applied to it:
res
#> A 'ConsensusPartition' object with k = 2, 3, 4, 5, 6.
#> On a matrix with 21168 rows and 100 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 3.
#>
#> Following methods can be applied to this 'ConsensusPartition' object:
#> [1] "cola_report" "collect_classes" "collect_plots"
#> [4] "collect_stats" "colnames" "compare_signatures"
#> [7] "consensus_heatmap" "dimension_reduction" "functional_enrichment"
#> [10] "get_anno_col" "get_anno" "get_classes"
#> [13] "get_consensus" "get_matrix" "get_membership"
#> [16] "get_param" "get_signatures" "get_stats"
#> [19] "is_best_k" "is_stable_k" "membership_heatmap"
#> [22] "ncol" "nrow" "plot_ecdf"
#> [25] "rownames" "select_partition_number" "show"
#> [28] "suggest_best_k" "test_to_known_factors"
collect_plots() function collects all the plots made from res for all k (number of partitions)
into one single page to provide an easy and fast comparison between different k.
collect_plots(res)
The plots are:
k and the heatmap of
predicted classes for each k.k.k.k.All the plots in panels can be made by individual functions and they are plotted later in this section.
select_partition_number() produces several plots showing different
statistics for choosing “optimized” k. There are following statistics:
k;k, the area increased is defined as \(A_k - A_{k-1}\).The detailed explanations of these statistics can be found in the cola vignette.
Generally speaking, lower PAC score, higher mean silhouette score or higher
concordance corresponds to better partition. Rand index and Jaccard index
measure how similar the current partition is compared to partition with k-1.
If they are too similar, we won't accept k is better than k-1.
select_partition_number(res)
The numeric values for all these statistics can be obtained by get_stats().
get_stats(res)
#> k 1-PAC mean_silhouette concordance area_increased Rand Jaccard
#> 2 2 1.000 0.986 0.993 0.4998 0.500 0.500
#> 3 3 0.974 0.953 0.976 0.3071 0.788 0.600
#> 4 4 0.761 0.672 0.848 0.1403 0.844 0.585
#> 5 5 0.714 0.715 0.826 0.0631 0.879 0.584
#> 6 6 0.705 0.548 0.720 0.0399 0.977 0.894
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
#> GSM97038 2 0.000 0.989 0.000 1.000
#> GSM97045 2 0.000 0.989 0.000 1.000
#> GSM97047 2 0.000 0.989 0.000 1.000
#> GSM97025 2 0.000 0.989 0.000 1.000
#> GSM97030 2 0.000 0.989 0.000 1.000
#> GSM97027 2 0.000 0.989 0.000 1.000
#> GSM97033 2 0.000 0.989 0.000 1.000
#> GSM97034 2 0.000 0.989 0.000 1.000
#> GSM97020 2 0.000 0.989 0.000 1.000
#> GSM97026 2 0.000 0.989 0.000 1.000
#> GSM97012 2 0.000 0.989 0.000 1.000
#> GSM97015 2 0.000 0.989 0.000 1.000
#> GSM97016 2 0.000 0.989 0.000 1.000
#> GSM97017 1 0.000 0.997 1.000 0.000
#> GSM97019 2 0.000 0.989 0.000 1.000
#> GSM97022 2 0.000 0.989 0.000 1.000
#> GSM97035 2 0.000 0.989 0.000 1.000
#> GSM97036 1 0.000 0.997 1.000 0.000
#> GSM97039 2 0.000 0.989 0.000 1.000
#> GSM97046 2 0.000 0.989 0.000 1.000
#> GSM97023 1 0.000 0.997 1.000 0.000
#> GSM97029 1 0.000 0.997 1.000 0.000
#> GSM97043 2 0.000 0.989 0.000 1.000
#> GSM97013 1 0.000 0.997 1.000 0.000
#> GSM96956 2 0.000 0.989 0.000 1.000
#> GSM97024 2 0.000 0.989 0.000 1.000
#> GSM97032 2 0.000 0.989 0.000 1.000
#> GSM97044 2 0.000 0.989 0.000 1.000
#> GSM97049 2 0.000 0.989 0.000 1.000
#> GSM96968 1 0.644 0.801 0.836 0.164
#> GSM96971 1 0.000 0.997 1.000 0.000
#> GSM96986 1 0.000 0.997 1.000 0.000
#> GSM97003 1 0.000 0.997 1.000 0.000
#> GSM96957 1 0.000 0.997 1.000 0.000
#> GSM96960 1 0.000 0.997 1.000 0.000
#> GSM96975 1 0.000 0.997 1.000 0.000
#> GSM96998 1 0.000 0.997 1.000 0.000
#> GSM96999 1 0.000 0.997 1.000 0.000
#> GSM97001 1 0.000 0.997 1.000 0.000
#> GSM97005 1 0.000 0.997 1.000 0.000
#> GSM97006 1 0.000 0.997 1.000 0.000
#> GSM97021 1 0.000 0.997 1.000 0.000
#> GSM97028 2 0.518 0.870 0.116 0.884
#> GSM97031 1 0.000 0.997 1.000 0.000
#> GSM97037 2 0.000 0.989 0.000 1.000
#> GSM97018 2 0.000 0.989 0.000 1.000
#> GSM97014 2 0.000 0.989 0.000 1.000
#> GSM97042 2 0.000 0.989 0.000 1.000
#> GSM97040 2 0.000 0.989 0.000 1.000
#> GSM97041 1 0.000 0.997 1.000 0.000
#> GSM96955 2 0.000 0.989 0.000 1.000
#> GSM96990 2 0.000 0.989 0.000 1.000
#> GSM96991 2 0.000 0.989 0.000 1.000
#> GSM97048 2 0.000 0.989 0.000 1.000
#> GSM96963 2 0.000 0.989 0.000 1.000
#> GSM96953 2 0.000 0.989 0.000 1.000
#> GSM96966 1 0.000 0.997 1.000 0.000
#> GSM96979 1 0.000 0.997 1.000 0.000
#> GSM96983 2 0.000 0.989 0.000 1.000
#> GSM96984 2 0.469 0.888 0.100 0.900
#> GSM96994 2 0.000 0.989 0.000 1.000
#> GSM96996 1 0.000 0.997 1.000 0.000
#> GSM96997 1 0.000 0.997 1.000 0.000
#> GSM97007 2 0.000 0.989 0.000 1.000
#> GSM96954 1 0.000 0.997 1.000 0.000
#> GSM96962 1 0.000 0.997 1.000 0.000
#> GSM96969 1 0.000 0.997 1.000 0.000
#> GSM96970 1 0.000 0.997 1.000 0.000
#> GSM96973 1 0.000 0.997 1.000 0.000
#> GSM96976 2 0.000 0.989 0.000 1.000
#> GSM96977 1 0.000 0.997 1.000 0.000
#> GSM96995 2 0.000 0.989 0.000 1.000
#> GSM97002 1 0.000 0.997 1.000 0.000
#> GSM97009 2 0.000 0.989 0.000 1.000
#> GSM97010 1 0.000 0.997 1.000 0.000
#> GSM96974 1 0.000 0.997 1.000 0.000
#> GSM96985 1 0.000 0.997 1.000 0.000
#> GSM96959 2 0.000 0.989 0.000 1.000
#> GSM96972 1 0.000 0.997 1.000 0.000
#> GSM96978 2 0.821 0.662 0.256 0.744
#> GSM96967 1 0.000 0.997 1.000 0.000
#> GSM96987 1 0.000 0.997 1.000 0.000
#> GSM97011 1 0.118 0.981 0.984 0.016
#> GSM96964 1 0.000 0.997 1.000 0.000
#> GSM96965 1 0.000 0.997 1.000 0.000
#> GSM96981 1 0.000 0.997 1.000 0.000
#> GSM96982 1 0.000 0.997 1.000 0.000
#> GSM96988 1 0.000 0.997 1.000 0.000
#> GSM97000 1 0.000 0.997 1.000 0.000
#> GSM97004 1 0.000 0.997 1.000 0.000
#> GSM97008 1 0.000 0.997 1.000 0.000
#> GSM96950 1 0.000 0.997 1.000 0.000
#> GSM96980 1 0.000 0.997 1.000 0.000
#> GSM96989 1 0.000 0.997 1.000 0.000
#> GSM96992 1 0.000 0.997 1.000 0.000
#> GSM96993 1 0.000 0.997 1.000 0.000
#> GSM96958 1 0.000 0.997 1.000 0.000
#> GSM96951 1 0.000 0.997 1.000 0.000
#> GSM96952 1 0.000 0.997 1.000 0.000
#> GSM96961 1 0.000 0.997 1.000 0.000
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM97038 2 0.0000 0.975 0.000 1.000 0.000
#> GSM97045 2 0.0000 0.975 0.000 1.000 0.000
#> GSM97047 2 0.0000 0.975 0.000 1.000 0.000
#> GSM97025 2 0.0000 0.975 0.000 1.000 0.000
#> GSM97030 3 0.2165 0.928 0.000 0.064 0.936
#> GSM97027 2 0.0000 0.975 0.000 1.000 0.000
#> GSM97033 2 0.0000 0.975 0.000 1.000 0.000
#> GSM97034 3 0.1964 0.934 0.000 0.056 0.944
#> GSM97020 2 0.0000 0.975 0.000 1.000 0.000
#> GSM97026 2 0.0000 0.975 0.000 1.000 0.000
#> GSM97012 2 0.0000 0.975 0.000 1.000 0.000
#> GSM97015 3 0.1964 0.934 0.000 0.056 0.944
#> GSM97016 2 0.0000 0.975 0.000 1.000 0.000
#> GSM97017 1 0.0000 0.981 1.000 0.000 0.000
#> GSM97019 2 0.0000 0.975 0.000 1.000 0.000
#> GSM97022 2 0.0000 0.975 0.000 1.000 0.000
#> GSM97035 2 0.0000 0.975 0.000 1.000 0.000
#> GSM97036 1 0.0000 0.981 1.000 0.000 0.000
#> GSM97039 2 0.0000 0.975 0.000 1.000 0.000
#> GSM97046 2 0.0000 0.975 0.000 1.000 0.000
#> GSM97023 1 0.0000 0.981 1.000 0.000 0.000
#> GSM97029 1 0.0424 0.977 0.992 0.008 0.000
#> GSM97043 2 0.0000 0.975 0.000 1.000 0.000
#> GSM97013 1 0.0000 0.981 1.000 0.000 0.000
#> GSM96956 2 0.4235 0.787 0.000 0.824 0.176
#> GSM97024 2 0.0000 0.975 0.000 1.000 0.000
#> GSM97032 3 0.4702 0.754 0.000 0.212 0.788
#> GSM97044 3 0.1753 0.938 0.000 0.048 0.952
#> GSM97049 2 0.0000 0.975 0.000 1.000 0.000
#> GSM96968 3 0.0237 0.959 0.004 0.000 0.996
#> GSM96971 3 0.0000 0.961 0.000 0.000 1.000
#> GSM96986 3 0.0000 0.961 0.000 0.000 1.000
#> GSM97003 1 0.0592 0.978 0.988 0.000 0.012
#> GSM96957 1 0.0000 0.981 1.000 0.000 0.000
#> GSM96960 1 0.0592 0.978 0.988 0.000 0.012
#> GSM96975 1 0.0000 0.981 1.000 0.000 0.000
#> GSM96998 1 0.0000 0.981 1.000 0.000 0.000
#> GSM96999 1 0.0000 0.981 1.000 0.000 0.000
#> GSM97001 1 0.0000 0.981 1.000 0.000 0.000
#> GSM97005 1 0.0000 0.981 1.000 0.000 0.000
#> GSM97006 1 0.0424 0.979 0.992 0.000 0.008
#> GSM97021 1 0.0000 0.981 1.000 0.000 0.000
#> GSM97028 3 0.0000 0.961 0.000 0.000 1.000
#> GSM97031 1 0.0592 0.978 0.988 0.000 0.012
#> GSM97037 2 0.4504 0.759 0.000 0.804 0.196
#> GSM97018 3 0.3941 0.836 0.000 0.156 0.844
#> GSM97014 2 0.0000 0.975 0.000 1.000 0.000
#> GSM97042 2 0.0000 0.975 0.000 1.000 0.000
#> GSM97040 2 0.1411 0.938 0.036 0.964 0.000
#> GSM97041 1 0.0000 0.981 1.000 0.000 0.000
#> GSM96955 2 0.0000 0.975 0.000 1.000 0.000
#> GSM96990 3 0.2165 0.928 0.000 0.064 0.936
#> GSM96991 2 0.0000 0.975 0.000 1.000 0.000
#> GSM97048 2 0.0000 0.975 0.000 1.000 0.000
#> GSM96963 2 0.0000 0.975 0.000 1.000 0.000
#> GSM96953 2 0.0000 0.975 0.000 1.000 0.000
#> GSM96966 1 0.1753 0.957 0.952 0.000 0.048
#> GSM96979 3 0.0000 0.961 0.000 0.000 1.000
#> GSM96983 3 0.0000 0.961 0.000 0.000 1.000
#> GSM96984 3 0.0000 0.961 0.000 0.000 1.000
#> GSM96994 3 0.0000 0.961 0.000 0.000 1.000
#> GSM96996 1 0.0237 0.980 0.996 0.000 0.004
#> GSM96997 3 0.0000 0.961 0.000 0.000 1.000
#> GSM97007 3 0.0000 0.961 0.000 0.000 1.000
#> GSM96954 3 0.0000 0.961 0.000 0.000 1.000
#> GSM96962 3 0.0000 0.961 0.000 0.000 1.000
#> GSM96969 1 0.1753 0.957 0.952 0.000 0.048
#> GSM96970 1 0.1753 0.957 0.952 0.000 0.048
#> GSM96973 1 0.1753 0.957 0.952 0.000 0.048
#> GSM96976 3 0.1289 0.946 0.000 0.032 0.968
#> GSM96977 1 0.2448 0.922 0.924 0.000 0.076
#> GSM96995 3 0.0747 0.955 0.000 0.016 0.984
#> GSM97002 1 0.0424 0.979 0.992 0.000 0.008
#> GSM97009 2 0.0000 0.975 0.000 1.000 0.000
#> GSM97010 1 0.1765 0.961 0.956 0.004 0.040
#> GSM96974 3 0.0000 0.961 0.000 0.000 1.000
#> GSM96985 3 0.4178 0.778 0.172 0.000 0.828
#> GSM96959 2 0.5291 0.639 0.000 0.732 0.268
#> GSM96972 1 0.1753 0.957 0.952 0.000 0.048
#> GSM96978 3 0.0000 0.961 0.000 0.000 1.000
#> GSM96967 1 0.1753 0.957 0.952 0.000 0.048
#> GSM96987 1 0.0000 0.981 1.000 0.000 0.000
#> GSM97011 1 0.2448 0.914 0.924 0.076 0.000
#> GSM96964 1 0.0000 0.981 1.000 0.000 0.000
#> GSM96965 1 0.1753 0.957 0.952 0.000 0.048
#> GSM96981 1 0.0000 0.981 1.000 0.000 0.000
#> GSM96982 1 0.0237 0.980 0.996 0.000 0.004
#> GSM96988 3 0.0000 0.961 0.000 0.000 1.000
#> GSM97000 1 0.4974 0.693 0.764 0.000 0.236
#> GSM97004 1 0.0424 0.979 0.992 0.000 0.008
#> GSM97008 1 0.0000 0.981 1.000 0.000 0.000
#> GSM96950 1 0.0000 0.981 1.000 0.000 0.000
#> GSM96980 1 0.0592 0.978 0.988 0.000 0.012
#> GSM96989 1 0.0000 0.981 1.000 0.000 0.000
#> GSM96992 1 0.0000 0.981 1.000 0.000 0.000
#> GSM96993 1 0.0000 0.981 1.000 0.000 0.000
#> GSM96958 1 0.0000 0.981 1.000 0.000 0.000
#> GSM96951 1 0.0000 0.981 1.000 0.000 0.000
#> GSM96952 1 0.0000 0.981 1.000 0.000 0.000
#> GSM96961 1 0.0000 0.981 1.000 0.000 0.000
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM97038 2 0.0000 0.943 0.000 1.000 0.000 0.000
#> GSM97045 2 0.0000 0.943 0.000 1.000 0.000 0.000
#> GSM97047 2 0.4804 0.486 0.384 0.616 0.000 0.000
#> GSM97025 2 0.0000 0.943 0.000 1.000 0.000 0.000
#> GSM97030 3 0.1743 0.903 0.004 0.056 0.940 0.000
#> GSM97027 2 0.0000 0.943 0.000 1.000 0.000 0.000
#> GSM97033 2 0.0000 0.943 0.000 1.000 0.000 0.000
#> GSM97034 3 0.1576 0.909 0.004 0.048 0.948 0.000
#> GSM97020 2 0.0000 0.943 0.000 1.000 0.000 0.000
#> GSM97026 2 0.0921 0.924 0.028 0.972 0.000 0.000
#> GSM97012 2 0.0000 0.943 0.000 1.000 0.000 0.000
#> GSM97015 3 0.1004 0.920 0.004 0.024 0.972 0.000
#> GSM97016 2 0.0000 0.943 0.000 1.000 0.000 0.000
#> GSM97017 1 0.0188 0.692 0.996 0.000 0.000 0.004
#> GSM97019 2 0.0000 0.943 0.000 1.000 0.000 0.000
#> GSM97022 2 0.0000 0.943 0.000 1.000 0.000 0.000
#> GSM97035 2 0.0000 0.943 0.000 1.000 0.000 0.000
#> GSM97036 1 0.4989 0.036 0.528 0.000 0.000 0.472
#> GSM97039 2 0.0000 0.943 0.000 1.000 0.000 0.000
#> GSM97046 2 0.0000 0.943 0.000 1.000 0.000 0.000
#> GSM97023 1 0.4941 0.169 0.564 0.000 0.000 0.436
#> GSM97029 1 0.4936 0.478 0.700 0.020 0.000 0.280
#> GSM97043 2 0.0000 0.943 0.000 1.000 0.000 0.000
#> GSM97013 1 0.4406 0.453 0.700 0.000 0.000 0.300
#> GSM96956 2 0.4331 0.570 0.000 0.712 0.288 0.000
#> GSM97024 2 0.0000 0.943 0.000 1.000 0.000 0.000
#> GSM97032 3 0.4584 0.589 0.004 0.300 0.696 0.000
#> GSM97044 3 0.0376 0.927 0.004 0.004 0.992 0.000
#> GSM97049 2 0.0000 0.943 0.000 1.000 0.000 0.000
#> GSM96968 3 0.0188 0.927 0.004 0.000 0.996 0.000
#> GSM96971 3 0.2081 0.877 0.000 0.000 0.916 0.084
#> GSM96986 3 0.0469 0.924 0.000 0.000 0.988 0.012
#> GSM97003 4 0.2675 0.681 0.100 0.000 0.008 0.892
#> GSM96957 1 0.0817 0.689 0.976 0.000 0.000 0.024
#> GSM96960 4 0.2011 0.688 0.080 0.000 0.000 0.920
#> GSM96975 4 0.4941 0.179 0.436 0.000 0.000 0.564
#> GSM96998 4 0.4643 0.420 0.344 0.000 0.000 0.656
#> GSM96999 1 0.4925 0.155 0.572 0.000 0.000 0.428
#> GSM97001 1 0.0188 0.692 0.996 0.000 0.000 0.004
#> GSM97005 1 0.0469 0.693 0.988 0.000 0.000 0.012
#> GSM97006 4 0.2647 0.669 0.120 0.000 0.000 0.880
#> GSM97021 1 0.0469 0.693 0.988 0.000 0.000 0.012
#> GSM97028 3 0.0376 0.927 0.004 0.004 0.992 0.000
#> GSM97031 1 0.5414 0.319 0.604 0.000 0.020 0.376
#> GSM97037 2 0.4800 0.451 0.004 0.656 0.340 0.000
#> GSM97018 3 0.4188 0.687 0.004 0.244 0.752 0.000
#> GSM97014 1 0.4977 -0.192 0.540 0.460 0.000 0.000
#> GSM97042 2 0.0000 0.943 0.000 1.000 0.000 0.000
#> GSM97040 1 0.1118 0.667 0.964 0.036 0.000 0.000
#> GSM97041 1 0.0188 0.692 0.996 0.000 0.000 0.004
#> GSM96955 2 0.1792 0.890 0.068 0.932 0.000 0.000
#> GSM96990 3 0.1978 0.894 0.004 0.068 0.928 0.000
#> GSM96991 2 0.0000 0.943 0.000 1.000 0.000 0.000
#> GSM97048 2 0.0000 0.943 0.000 1.000 0.000 0.000
#> GSM96963 2 0.0000 0.943 0.000 1.000 0.000 0.000
#> GSM96953 2 0.0000 0.943 0.000 1.000 0.000 0.000
#> GSM96966 4 0.0188 0.688 0.000 0.000 0.004 0.996
#> GSM96979 3 0.2345 0.855 0.000 0.000 0.900 0.100
#> GSM96983 3 0.0000 0.928 0.000 0.000 1.000 0.000
#> GSM96984 3 0.0000 0.928 0.000 0.000 1.000 0.000
#> GSM96994 3 0.0000 0.928 0.000 0.000 1.000 0.000
#> GSM96996 4 0.2345 0.680 0.100 0.000 0.000 0.900
#> GSM96997 3 0.0336 0.926 0.000 0.000 0.992 0.008
#> GSM97007 3 0.0000 0.928 0.000 0.000 1.000 0.000
#> GSM96954 3 0.0188 0.927 0.004 0.000 0.996 0.000
#> GSM96962 3 0.0000 0.928 0.000 0.000 1.000 0.000
#> GSM96969 4 0.0188 0.688 0.000 0.000 0.004 0.996
#> GSM96970 4 0.0188 0.688 0.000 0.000 0.004 0.996
#> GSM96973 4 0.0188 0.688 0.000 0.000 0.004 0.996
#> GSM96976 4 0.7171 -0.148 0.000 0.136 0.400 0.464
#> GSM96977 1 0.5025 0.518 0.716 0.000 0.032 0.252
#> GSM96995 3 0.0921 0.919 0.028 0.000 0.972 0.000
#> GSM97002 4 0.2011 0.687 0.080 0.000 0.000 0.920
#> GSM97009 2 0.4431 0.614 0.304 0.696 0.000 0.000
#> GSM97010 4 0.2353 0.654 0.056 0.012 0.008 0.924
#> GSM96974 4 0.4855 0.112 0.000 0.000 0.400 0.600
#> GSM96985 4 0.2647 0.590 0.000 0.000 0.120 0.880
#> GSM96959 3 0.7344 0.286 0.380 0.160 0.460 0.000
#> GSM96972 4 0.0188 0.688 0.000 0.000 0.004 0.996
#> GSM96978 3 0.0188 0.927 0.000 0.000 0.996 0.004
#> GSM96967 4 0.0188 0.688 0.000 0.000 0.004 0.996
#> GSM96987 4 0.4948 0.227 0.440 0.000 0.000 0.560
#> GSM97011 1 0.1174 0.685 0.968 0.012 0.000 0.020
#> GSM96964 4 0.4972 0.181 0.456 0.000 0.000 0.544
#> GSM96965 4 0.0895 0.682 0.020 0.000 0.004 0.976
#> GSM96981 4 0.3528 0.604 0.192 0.000 0.000 0.808
#> GSM96982 4 0.1302 0.691 0.044 0.000 0.000 0.956
#> GSM96988 3 0.0592 0.923 0.000 0.000 0.984 0.016
#> GSM97000 1 0.1767 0.665 0.944 0.000 0.044 0.012
#> GSM97004 4 0.1867 0.689 0.072 0.000 0.000 0.928
#> GSM97008 1 0.0469 0.693 0.988 0.000 0.000 0.012
#> GSM96950 1 0.4955 0.129 0.556 0.000 0.000 0.444
#> GSM96980 4 0.0000 0.688 0.000 0.000 0.000 1.000
#> GSM96989 4 0.4948 0.227 0.440 0.000 0.000 0.560
#> GSM96992 4 0.4776 0.354 0.376 0.000 0.000 0.624
#> GSM96993 1 0.4933 0.165 0.568 0.000 0.000 0.432
#> GSM96958 4 0.4972 0.166 0.456 0.000 0.000 0.544
#> GSM96951 4 0.4992 0.104 0.476 0.000 0.000 0.524
#> GSM96952 4 0.4888 0.282 0.412 0.000 0.000 0.588
#> GSM96961 4 0.4955 0.204 0.444 0.000 0.000 0.556
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM97038 2 0.0000 0.9439 0.000 1.000 0.000 0.000 0.000
#> GSM97045 2 0.0000 0.9439 0.000 1.000 0.000 0.000 0.000
#> GSM97047 5 0.4232 0.5622 0.000 0.312 0.012 0.000 0.676
#> GSM97025 2 0.0000 0.9439 0.000 1.000 0.000 0.000 0.000
#> GSM97030 3 0.2077 0.7974 0.000 0.084 0.908 0.000 0.008
#> GSM97027 2 0.0000 0.9439 0.000 1.000 0.000 0.000 0.000
#> GSM97033 2 0.0000 0.9439 0.000 1.000 0.000 0.000 0.000
#> GSM97034 3 0.3163 0.7590 0.004 0.124 0.852 0.008 0.012
#> GSM97020 2 0.0000 0.9439 0.000 1.000 0.000 0.000 0.000
#> GSM97026 2 0.3639 0.7975 0.044 0.848 0.020 0.004 0.084
#> GSM97012 2 0.0000 0.9439 0.000 1.000 0.000 0.000 0.000
#> GSM97015 3 0.1173 0.8276 0.000 0.020 0.964 0.004 0.012
#> GSM97016 2 0.0000 0.9439 0.000 1.000 0.000 0.000 0.000
#> GSM97017 5 0.3048 0.7546 0.176 0.000 0.000 0.004 0.820
#> GSM97019 2 0.0000 0.9439 0.000 1.000 0.000 0.000 0.000
#> GSM97022 2 0.0000 0.9439 0.000 1.000 0.000 0.000 0.000
#> GSM97035 2 0.0000 0.9439 0.000 1.000 0.000 0.000 0.000
#> GSM97036 1 0.3229 0.6500 0.840 0.000 0.000 0.032 0.128
#> GSM97039 2 0.0000 0.9439 0.000 1.000 0.000 0.000 0.000
#> GSM97046 2 0.0000 0.9439 0.000 1.000 0.000 0.000 0.000
#> GSM97023 1 0.3226 0.6880 0.852 0.000 0.000 0.060 0.088
#> GSM97029 1 0.4265 0.4884 0.712 0.012 0.000 0.008 0.268
#> GSM97043 2 0.0290 0.9381 0.000 0.992 0.008 0.000 0.000
#> GSM97013 1 0.3388 0.5655 0.792 0.000 0.000 0.008 0.200
#> GSM96956 2 0.4225 0.4052 0.000 0.632 0.364 0.000 0.004
#> GSM97024 2 0.0324 0.9377 0.000 0.992 0.004 0.000 0.004
#> GSM97032 3 0.4403 0.4475 0.000 0.340 0.648 0.004 0.008
#> GSM97044 3 0.0290 0.8317 0.000 0.000 0.992 0.000 0.008
#> GSM97049 2 0.0000 0.9439 0.000 1.000 0.000 0.000 0.000
#> GSM96968 3 0.1372 0.8353 0.004 0.000 0.956 0.016 0.024
#> GSM96971 3 0.4953 0.4462 0.000 0.000 0.532 0.440 0.028
#> GSM96986 3 0.3844 0.8170 0.000 0.000 0.792 0.164 0.044
#> GSM97003 1 0.6102 0.2965 0.488 0.000 0.020 0.420 0.072
#> GSM96957 1 0.4310 0.2622 0.604 0.000 0.000 0.004 0.392
#> GSM96960 1 0.4851 0.4569 0.624 0.000 0.000 0.340 0.036
#> GSM96975 1 0.6227 0.4263 0.536 0.000 0.000 0.280 0.184
#> GSM96998 1 0.2079 0.6801 0.916 0.000 0.000 0.064 0.020
#> GSM96999 1 0.4350 0.6633 0.764 0.000 0.000 0.084 0.152
#> GSM97001 5 0.3160 0.7401 0.188 0.000 0.000 0.004 0.808
#> GSM97005 5 0.2179 0.7919 0.112 0.000 0.000 0.000 0.888
#> GSM97006 1 0.4822 0.5289 0.664 0.000 0.000 0.288 0.048
#> GSM97021 5 0.2424 0.7844 0.132 0.000 0.000 0.000 0.868
#> GSM97028 3 0.1168 0.8387 0.000 0.000 0.960 0.032 0.008
#> GSM97031 1 0.7208 0.4013 0.452 0.000 0.036 0.192 0.320
#> GSM97037 2 0.4542 0.1416 0.000 0.536 0.456 0.000 0.008
#> GSM97018 3 0.4893 0.3930 0.000 0.360 0.612 0.012 0.016
#> GSM97014 5 0.2930 0.7376 0.000 0.164 0.000 0.004 0.832
#> GSM97042 2 0.0000 0.9439 0.000 1.000 0.000 0.000 0.000
#> GSM97040 5 0.1764 0.8021 0.036 0.012 0.012 0.000 0.940
#> GSM97041 5 0.3461 0.7124 0.224 0.000 0.000 0.004 0.772
#> GSM96955 2 0.2813 0.7463 0.000 0.832 0.000 0.000 0.168
#> GSM96990 3 0.1894 0.8050 0.000 0.072 0.920 0.000 0.008
#> GSM96991 2 0.0000 0.9439 0.000 1.000 0.000 0.000 0.000
#> GSM97048 2 0.0000 0.9439 0.000 1.000 0.000 0.000 0.000
#> GSM96963 2 0.0000 0.9439 0.000 1.000 0.000 0.000 0.000
#> GSM96953 2 0.0000 0.9439 0.000 1.000 0.000 0.000 0.000
#> GSM96966 4 0.2773 0.8285 0.164 0.000 0.000 0.836 0.000
#> GSM96979 3 0.4477 0.7498 0.000 0.000 0.708 0.252 0.040
#> GSM96983 3 0.1124 0.8400 0.000 0.000 0.960 0.036 0.004
#> GSM96984 3 0.3409 0.8305 0.000 0.000 0.824 0.144 0.032
#> GSM96994 3 0.3283 0.8323 0.000 0.000 0.832 0.140 0.028
#> GSM96996 1 0.4522 0.5017 0.660 0.000 0.000 0.316 0.024
#> GSM96997 3 0.3973 0.8158 0.008 0.000 0.792 0.164 0.036
#> GSM97007 3 0.3409 0.8305 0.000 0.000 0.824 0.144 0.032
#> GSM96954 3 0.2359 0.8423 0.000 0.000 0.904 0.060 0.036
#> GSM96962 3 0.3409 0.8305 0.000 0.000 0.824 0.144 0.032
#> GSM96969 4 0.2773 0.8285 0.164 0.000 0.000 0.836 0.000
#> GSM96970 4 0.2732 0.8300 0.160 0.000 0.000 0.840 0.000
#> GSM96973 4 0.2732 0.8300 0.160 0.000 0.000 0.840 0.000
#> GSM96976 4 0.2678 0.6372 0.000 0.016 0.100 0.880 0.004
#> GSM96977 1 0.7265 -0.0745 0.408 0.000 0.056 0.140 0.396
#> GSM96995 3 0.2237 0.8051 0.000 0.004 0.904 0.008 0.084
#> GSM97002 1 0.4768 0.3768 0.592 0.000 0.000 0.384 0.024
#> GSM97009 5 0.4613 0.3582 0.004 0.408 0.000 0.008 0.580
#> GSM97010 4 0.5152 0.4625 0.344 0.004 0.000 0.608 0.044
#> GSM96974 4 0.2574 0.6788 0.012 0.000 0.112 0.876 0.000
#> GSM96985 4 0.4022 0.7339 0.100 0.000 0.092 0.804 0.004
#> GSM96959 5 0.5155 0.5357 0.000 0.056 0.276 0.008 0.660
#> GSM96972 4 0.2773 0.8285 0.164 0.000 0.000 0.836 0.000
#> GSM96978 3 0.2753 0.8314 0.000 0.000 0.856 0.136 0.008
#> GSM96967 4 0.2732 0.8300 0.160 0.000 0.000 0.840 0.000
#> GSM96987 1 0.1399 0.6810 0.952 0.000 0.000 0.020 0.028
#> GSM97011 5 0.1877 0.7979 0.064 0.000 0.000 0.012 0.924
#> GSM96964 1 0.1251 0.6817 0.956 0.000 0.000 0.008 0.036
#> GSM96965 4 0.3280 0.8042 0.176 0.000 0.000 0.812 0.012
#> GSM96981 1 0.5232 0.4409 0.600 0.000 0.000 0.340 0.060
#> GSM96982 1 0.4961 0.1892 0.524 0.000 0.000 0.448 0.028
#> GSM96988 3 0.3552 0.8141 0.012 0.000 0.812 0.164 0.012
#> GSM97000 5 0.1924 0.7923 0.064 0.000 0.008 0.004 0.924
#> GSM97004 1 0.4524 0.4756 0.644 0.000 0.000 0.336 0.020
#> GSM97008 5 0.1732 0.7962 0.080 0.000 0.000 0.000 0.920
#> GSM96950 1 0.2674 0.6554 0.868 0.000 0.000 0.012 0.120
#> GSM96980 4 0.3876 0.5641 0.316 0.000 0.000 0.684 0.000
#> GSM96989 1 0.1403 0.6806 0.952 0.000 0.000 0.024 0.024
#> GSM96992 1 0.3929 0.6232 0.764 0.000 0.000 0.208 0.028
#> GSM96993 1 0.2286 0.6552 0.888 0.000 0.000 0.004 0.108
#> GSM96958 1 0.3702 0.6825 0.820 0.000 0.000 0.084 0.096
#> GSM96951 1 0.3644 0.6875 0.824 0.000 0.000 0.080 0.096
#> GSM96952 1 0.3452 0.6594 0.820 0.000 0.000 0.148 0.032
#> GSM96961 1 0.2491 0.6864 0.896 0.000 0.000 0.068 0.036
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM97038 2 0.2144 0.90283 0.000 0.908 0.012 0.004 0.008 0.068
#> GSM97045 2 0.0603 0.91922 0.000 0.980 0.000 0.000 0.004 0.016
#> GSM97047 5 0.4268 0.65054 0.000 0.220 0.020 0.004 0.728 0.028
#> GSM97025 2 0.0146 0.91916 0.000 0.996 0.000 0.000 0.000 0.004
#> GSM97030 3 0.1333 0.48828 0.000 0.048 0.944 0.000 0.000 0.008
#> GSM97027 2 0.0363 0.91964 0.000 0.988 0.000 0.000 0.000 0.012
#> GSM97033 2 0.1621 0.91242 0.000 0.936 0.008 0.004 0.004 0.048
#> GSM97034 3 0.3636 0.45197 0.000 0.128 0.804 0.004 0.004 0.060
#> GSM97020 2 0.1863 0.90943 0.000 0.924 0.008 0.004 0.008 0.056
#> GSM97026 2 0.5823 0.59429 0.056 0.688 0.044 0.004 0.080 0.128
#> GSM97012 2 0.0000 0.91940 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97015 3 0.1092 0.48429 0.000 0.020 0.960 0.000 0.000 0.020
#> GSM97016 2 0.2144 0.90478 0.000 0.908 0.012 0.004 0.008 0.068
#> GSM97017 5 0.4279 0.67130 0.140 0.000 0.000 0.000 0.732 0.128
#> GSM97019 2 0.0000 0.91940 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97022 2 0.0000 0.91940 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97035 2 0.0260 0.92051 0.000 0.992 0.000 0.000 0.000 0.008
#> GSM97036 1 0.5393 0.46505 0.656 0.004 0.004 0.024 0.096 0.216
#> GSM97039 2 0.1985 0.90688 0.000 0.916 0.008 0.004 0.008 0.064
#> GSM97046 2 0.2202 0.90306 0.000 0.904 0.012 0.004 0.008 0.072
#> GSM97023 1 0.4592 0.62260 0.756 0.000 0.000 0.084 0.088 0.072
#> GSM97029 1 0.6039 0.31976 0.572 0.028 0.004 0.000 0.184 0.212
#> GSM97043 2 0.0806 0.91269 0.000 0.972 0.020 0.000 0.000 0.008
#> GSM97013 1 0.5096 0.41145 0.652 0.000 0.000 0.008 0.136 0.204
#> GSM96956 2 0.5223 0.21656 0.000 0.508 0.416 0.004 0.004 0.068
#> GSM97024 2 0.0692 0.91255 0.000 0.976 0.020 0.000 0.000 0.004
#> GSM97032 3 0.3608 0.36177 0.000 0.272 0.716 0.000 0.000 0.012
#> GSM97044 3 0.1151 0.46573 0.000 0.012 0.956 0.000 0.000 0.032
#> GSM97049 2 0.2101 0.90495 0.000 0.908 0.008 0.004 0.008 0.072
#> GSM96968 3 0.3657 0.36890 0.012 0.000 0.788 0.024 0.004 0.172
#> GSM96971 6 0.6205 0.31595 0.000 0.000 0.276 0.340 0.004 0.380
#> GSM96986 6 0.4211 0.58418 0.000 0.000 0.456 0.004 0.008 0.532
#> GSM97003 1 0.6827 0.29464 0.412 0.000 0.008 0.264 0.032 0.284
#> GSM96957 1 0.5914 0.29320 0.556 0.000 0.000 0.028 0.272 0.144
#> GSM96960 1 0.5211 0.45565 0.580 0.000 0.000 0.340 0.024 0.056
#> GSM96975 1 0.7027 0.37078 0.412 0.000 0.000 0.288 0.220 0.080
#> GSM96998 1 0.3857 0.61656 0.788 0.000 0.000 0.112 0.008 0.092
#> GSM96999 1 0.5693 0.57692 0.656 0.000 0.000 0.100 0.124 0.120
#> GSM97001 5 0.4389 0.66687 0.168 0.000 0.000 0.012 0.736 0.084
#> GSM97005 5 0.2866 0.74418 0.084 0.000 0.000 0.004 0.860 0.052
#> GSM97006 1 0.5419 0.48491 0.588 0.000 0.000 0.308 0.028 0.076
#> GSM97021 5 0.3472 0.73041 0.092 0.000 0.000 0.000 0.808 0.100
#> GSM97028 3 0.2362 0.42165 0.000 0.000 0.860 0.004 0.000 0.136
#> GSM97031 1 0.7904 0.29290 0.324 0.000 0.016 0.164 0.220 0.276
#> GSM97037 3 0.4756 0.26349 0.000 0.332 0.608 0.000 0.004 0.056
#> GSM97018 3 0.4460 0.39852 0.000 0.200 0.728 0.004 0.020 0.048
#> GSM97014 5 0.3780 0.73587 0.020 0.096 0.000 0.004 0.812 0.068
#> GSM97042 2 0.0000 0.91940 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97040 5 0.2421 0.76526 0.028 0.004 0.032 0.000 0.904 0.032
#> GSM97041 5 0.5198 0.53592 0.240 0.000 0.000 0.000 0.608 0.152
#> GSM96955 2 0.5224 0.58674 0.000 0.668 0.020 0.012 0.220 0.080
#> GSM96990 3 0.2003 0.48587 0.000 0.044 0.912 0.000 0.000 0.044
#> GSM96991 2 0.0000 0.91940 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97048 2 0.2101 0.90495 0.000 0.908 0.008 0.004 0.008 0.072
#> GSM96963 2 0.0363 0.92036 0.000 0.988 0.000 0.000 0.000 0.012
#> GSM96953 2 0.0458 0.92034 0.000 0.984 0.000 0.000 0.000 0.016
#> GSM96966 4 0.1010 0.81879 0.036 0.000 0.000 0.960 0.000 0.004
#> GSM96979 6 0.4857 0.60891 0.000 0.000 0.408 0.060 0.000 0.532
#> GSM96983 3 0.2491 0.34454 0.000 0.000 0.836 0.000 0.000 0.164
#> GSM96984 3 0.3868 -0.60291 0.000 0.000 0.504 0.000 0.000 0.496
#> GSM96994 3 0.3867 -0.58770 0.000 0.000 0.512 0.000 0.000 0.488
#> GSM96996 1 0.5225 0.50398 0.608 0.000 0.000 0.292 0.016 0.084
#> GSM96997 6 0.4306 0.57317 0.012 0.000 0.464 0.004 0.000 0.520
#> GSM97007 3 0.3857 -0.55316 0.000 0.000 0.532 0.000 0.000 0.468
#> GSM96954 3 0.4261 -0.20894 0.004 0.000 0.620 0.008 0.008 0.360
#> GSM96962 3 0.3868 -0.59679 0.000 0.000 0.508 0.000 0.000 0.492
#> GSM96969 4 0.1219 0.80792 0.048 0.000 0.000 0.948 0.000 0.004
#> GSM96970 4 0.1010 0.81879 0.036 0.000 0.000 0.960 0.000 0.004
#> GSM96973 4 0.1010 0.81879 0.036 0.000 0.000 0.960 0.000 0.004
#> GSM96976 4 0.3737 0.60978 0.000 0.008 0.036 0.772 0.000 0.184
#> GSM96977 1 0.7878 -0.00192 0.384 0.000 0.052 0.100 0.296 0.168
#> GSM96995 3 0.3864 0.40139 0.000 0.000 0.796 0.016 0.096 0.092
#> GSM97002 1 0.4992 0.41397 0.564 0.000 0.000 0.376 0.016 0.044
#> GSM97009 5 0.5762 0.40123 0.004 0.320 0.008 0.008 0.552 0.108
#> GSM97010 4 0.6663 0.12895 0.296 0.000 0.004 0.400 0.024 0.276
#> GSM96974 4 0.3295 0.67808 0.000 0.000 0.056 0.816 0.000 0.128
#> GSM96985 4 0.4584 0.70550 0.080 0.000 0.036 0.752 0.004 0.128
#> GSM96959 5 0.5766 0.49469 0.000 0.008 0.256 0.016 0.588 0.132
#> GSM96972 4 0.1082 0.81360 0.040 0.000 0.000 0.956 0.000 0.004
#> GSM96978 3 0.4593 -0.08619 0.000 0.000 0.620 0.056 0.000 0.324
#> GSM96967 4 0.1010 0.81879 0.036 0.000 0.000 0.960 0.000 0.004
#> GSM96987 1 0.3507 0.59743 0.816 0.000 0.000 0.044 0.016 0.124
#> GSM97011 5 0.1649 0.76012 0.036 0.000 0.000 0.000 0.932 0.032
#> GSM96964 1 0.3337 0.60111 0.832 0.000 0.000 0.032 0.024 0.112
#> GSM96965 4 0.1887 0.79305 0.048 0.000 0.000 0.924 0.012 0.016
#> GSM96981 1 0.5768 0.42537 0.528 0.000 0.000 0.356 0.068 0.048
#> GSM96982 1 0.5063 0.28897 0.496 0.000 0.000 0.448 0.024 0.032
#> GSM96988 3 0.5125 -0.02934 0.008 0.000 0.612 0.076 0.004 0.300
#> GSM97000 5 0.2468 0.75035 0.016 0.000 0.008 0.000 0.880 0.096
#> GSM97004 1 0.4864 0.42356 0.576 0.000 0.000 0.372 0.016 0.036
#> GSM97008 5 0.2471 0.74174 0.052 0.000 0.000 0.004 0.888 0.056
#> GSM96950 1 0.4331 0.49844 0.728 0.000 0.000 0.008 0.072 0.192
#> GSM96980 4 0.3915 0.37498 0.288 0.000 0.000 0.692 0.004 0.016
#> GSM96989 1 0.3310 0.60068 0.832 0.000 0.000 0.040 0.016 0.112
#> GSM96992 1 0.4920 0.55063 0.668 0.000 0.000 0.248 0.044 0.040
#> GSM96993 1 0.4898 0.47643 0.692 0.000 0.004 0.020 0.076 0.208
#> GSM96958 1 0.5052 0.61055 0.716 0.000 0.000 0.116 0.084 0.084
#> GSM96951 1 0.4962 0.61307 0.720 0.000 0.000 0.124 0.096 0.060
#> GSM96952 1 0.4377 0.58214 0.728 0.000 0.000 0.204 0.040 0.028
#> GSM96961 1 0.3258 0.61753 0.832 0.000 0.000 0.120 0.032 0.016
Heatmaps for the consensus matrix. It visualizes the probability of two samples to be in a same group.
consensus_heatmap(res, k = 2)
consensus_heatmap(res, k = 3)
consensus_heatmap(res, k = 4)
consensus_heatmap(res, k = 5)
consensus_heatmap(res, k = 6)
Heatmaps for the membership of samples in all partitions to see how consistent they are:
membership_heatmap(res, k = 2)
membership_heatmap(res, k = 3)
membership_heatmap(res, k = 4)
membership_heatmap(res, k = 5)
membership_heatmap(res, k = 6)
As soon as we have had the classes for columns, we can look for signatures which are significantly different between classes which can be candidate marks for certain classes. Following are the heatmaps for signatures.
Signature heatmaps where rows are scaled:
get_signatures(res, k = 2)
get_signatures(res, k = 3)
get_signatures(res, k = 4)
get_signatures(res, k = 5)
get_signatures(res, k = 6)
Signature heatmaps where rows are not scaled:
get_signatures(res, k = 2, scale_rows = FALSE)
get_signatures(res, k = 3, scale_rows = FALSE)
get_signatures(res, k = 4, scale_rows = FALSE)
get_signatures(res, k = 5, scale_rows = FALSE)
get_signatures(res, k = 6, scale_rows = FALSE)
Compare the overlap of signatures from different k:
compare_signatures(res)
get_signature() returns a data frame invisibly. TO get the list of signatures, the function
call should be assigned to a variable explicitly. In following code, if plot argument is set
to FALSE, no heatmap is plotted while only the differential analysis is performed.
# code only for demonstration
tb = get_signature(res, k = ..., plot = FALSE)
An example of the output of tb is:
#> which_row fdr mean_1 mean_2 scaled_mean_1 scaled_mean_2 km
#> 1 38 0.042760348 8.373488 9.131774 -0.5533452 0.5164555 1
#> 2 40 0.018707592 7.106213 8.469186 -0.6173731 0.5762149 1
#> 3 55 0.019134737 10.221463 11.207825 -0.6159697 0.5749050 1
#> 4 59 0.006059896 5.921854 7.869574 -0.6899429 0.6439467 1
#> 5 60 0.018055526 8.928898 10.211722 -0.6204761 0.5791110 1
#> 6 98 0.009384629 15.714769 14.887706 0.6635654 -0.6193277 2
...
The columns in tb are:
which_row: row indices corresponding to the input matrix.fdr: FDR for the differential test. mean_x: The mean value in group x.scaled_mean_x: The mean value in group x after rows are scaled.km: Row groups if k-means clustering is applied to rows.UMAP plot which shows how samples are separated.
dimension_reduction(res, k = 2, method = "UMAP")
dimension_reduction(res, k = 3, method = "UMAP")
dimension_reduction(res, k = 4, method = "UMAP")
dimension_reduction(res, k = 5, method = "UMAP")
dimension_reduction(res, k = 6, method = "UMAP")
Following heatmap shows how subgroups are split when increasing k:
collect_classes(res)
Test correlation between subgroups and known annotations. If the known annotation is numeric, one-way ANOVA test is applied, and if the known annotation is discrete, chi-squared contingency table test is applied.
test_to_known_factors(res)
#> n disease.state(p) specimen(p) cell.type(p) other(p) k
#> MAD:skmeans 100 9.36e-05 0.173 5.21e-13 0.0903 2
#> MAD:skmeans 100 1.14e-04 0.290 6.36e-17 0.0685 3
#> MAD:skmeans 76 7.71e-05 0.481 5.77e-13 0.0246 4
#> MAD:skmeans 82 2.08e-05 0.437 4.16e-16 0.0648 5
#> MAD:skmeans 59 6.82e-05 0.421 1.51e-15 0.1276 6
If matrix rows can be associated to genes, consider to use functional_enrichment(res,
...) to perform function enrichment for the signature genes. See this vignette for more detailed explanations.
The object with results only for a single top-value method and a single partition method can be extracted as:
res = res_list["MAD", "pam"]
# you can also extract it by
# res = res_list["MAD:pam"]
A summary of res and all the functions that can be applied to it:
res
#> A 'ConsensusPartition' object with k = 2, 3, 4, 5, 6.
#> On a matrix with 21168 rows and 100 columns.
#> Top rows (1000, 2000, 3000, 4000, 5000) are extracted by 'MAD' method.
#> Subgroups are detected by 'pam' method.
#> Performed in total 1250 partitions by row resampling.
#> Best k for subgroups seems to be 2.
#>
#> Following methods can be applied to this 'ConsensusPartition' object:
#> [1] "cola_report" "collect_classes" "collect_plots"
#> [4] "collect_stats" "colnames" "compare_signatures"
#> [7] "consensus_heatmap" "dimension_reduction" "functional_enrichment"
#> [10] "get_anno_col" "get_anno" "get_classes"
#> [13] "get_consensus" "get_matrix" "get_membership"
#> [16] "get_param" "get_signatures" "get_stats"
#> [19] "is_best_k" "is_stable_k" "membership_heatmap"
#> [22] "ncol" "nrow" "plot_ecdf"
#> [25] "rownames" "select_partition_number" "show"
#> [28] "suggest_best_k" "test_to_known_factors"
collect_plots() function collects all the plots made from res for all k (number of partitions)
into one single page to provide an easy and fast comparison between different k.
collect_plots(res)
The plots are:
k and the heatmap of
predicted classes for each k.k.k.k.All the plots in panels can be made by individual functions and they are plotted later in this section.
select_partition_number() produces several plots showing different
statistics for choosing “optimized” k. There are following statistics:
k;k, the area increased is defined as \(A_k - A_{k-1}\).The detailed explanations of these statistics can be found in the cola vignette.
Generally speaking, lower PAC score, higher mean silhouette score or higher
concordance corresponds to better partition. Rand index and Jaccard index
measure how similar the current partition is compared to partition with k-1.
If they are too similar, we won't accept k is better than k-1.
select_partition_number(res)
The numeric values for all these statistics can be obtained by get_stats().
get_stats(res)
#> k 1-PAC mean_silhouette concordance area_increased Rand Jaccard
#> 2 2 0.898 0.946 0.971 0.4625 0.540 0.540
#> 3 3 0.473 0.503 0.752 0.3681 0.757 0.569
#> 4 4 0.580 0.654 0.809 0.1546 0.812 0.529
#> 5 5 0.699 0.774 0.837 0.0753 0.874 0.583
#> 6 6 0.723 0.598 0.750 0.0516 0.950 0.767
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
#> GSM97038 2 0.0000 0.971 0.000 1.000
#> GSM97045 2 0.0000 0.971 0.000 1.000
#> GSM97047 2 0.4690 0.876 0.100 0.900
#> GSM97025 2 0.0000 0.971 0.000 1.000
#> GSM97030 2 0.0000 0.971 0.000 1.000
#> GSM97027 2 0.0000 0.971 0.000 1.000
#> GSM97033 2 0.0000 0.971 0.000 1.000
#> GSM97034 2 0.0000 0.971 0.000 1.000
#> GSM97020 2 0.0000 0.971 0.000 1.000
#> GSM97026 2 0.0000 0.971 0.000 1.000
#> GSM97012 2 0.0000 0.971 0.000 1.000
#> GSM97015 2 0.0672 0.966 0.008 0.992
#> GSM97016 2 0.0000 0.971 0.000 1.000
#> GSM97017 1 0.0672 0.968 0.992 0.008
#> GSM97019 2 0.0000 0.971 0.000 1.000
#> GSM97022 2 0.0000 0.971 0.000 1.000
#> GSM97035 2 0.0000 0.971 0.000 1.000
#> GSM97036 1 0.7453 0.763 0.788 0.212
#> GSM97039 2 0.0000 0.971 0.000 1.000
#> GSM97046 2 0.0000 0.971 0.000 1.000
#> GSM97023 1 0.0000 0.970 1.000 0.000
#> GSM97029 1 0.1843 0.960 0.972 0.028
#> GSM97043 2 0.0000 0.971 0.000 1.000
#> GSM97013 1 0.8081 0.686 0.752 0.248
#> GSM96956 2 0.0000 0.971 0.000 1.000
#> GSM97024 2 0.0000 0.971 0.000 1.000
#> GSM97032 2 0.0000 0.971 0.000 1.000
#> GSM97044 2 0.0000 0.971 0.000 1.000
#> GSM97049 2 0.0000 0.971 0.000 1.000
#> GSM96968 1 0.2423 0.954 0.960 0.040
#> GSM96971 1 0.1184 0.966 0.984 0.016
#> GSM96986 1 0.0000 0.970 1.000 0.000
#> GSM97003 1 0.0000 0.970 1.000 0.000
#> GSM96957 1 0.0000 0.970 1.000 0.000
#> GSM96960 1 0.0000 0.970 1.000 0.000
#> GSM96975 1 0.0000 0.970 1.000 0.000
#> GSM96998 1 0.0000 0.970 1.000 0.000
#> GSM96999 1 0.0000 0.970 1.000 0.000
#> GSM97001 1 0.0000 0.970 1.000 0.000
#> GSM97005 1 0.0000 0.970 1.000 0.000
#> GSM97006 1 0.0000 0.970 1.000 0.000
#> GSM97021 1 0.0376 0.969 0.996 0.004
#> GSM97028 1 0.1843 0.962 0.972 0.028
#> GSM97031 1 0.0000 0.970 1.000 0.000
#> GSM97037 2 0.0000 0.971 0.000 1.000
#> GSM97018 2 0.2948 0.927 0.052 0.948
#> GSM97014 1 0.7745 0.749 0.772 0.228
#> GSM97042 2 0.0000 0.971 0.000 1.000
#> GSM97040 1 0.3431 0.938 0.936 0.064
#> GSM97041 1 0.5178 0.876 0.884 0.116
#> GSM96955 1 0.5519 0.876 0.872 0.128
#> GSM96990 2 0.0376 0.969 0.004 0.996
#> GSM96991 2 0.0000 0.971 0.000 1.000
#> GSM97048 2 0.0000 0.971 0.000 1.000
#> GSM96963 2 0.0000 0.971 0.000 1.000
#> GSM96953 2 0.0000 0.971 0.000 1.000
#> GSM96966 1 0.0000 0.970 1.000 0.000
#> GSM96979 1 0.1184 0.966 0.984 0.016
#> GSM96983 2 0.8207 0.668 0.256 0.744
#> GSM96984 2 0.9044 0.547 0.320 0.680
#> GSM96994 1 0.3584 0.934 0.932 0.068
#> GSM96996 1 0.0000 0.970 1.000 0.000
#> GSM96997 1 0.0938 0.967 0.988 0.012
#> GSM97007 2 0.7139 0.765 0.196 0.804
#> GSM96954 1 0.1184 0.966 0.984 0.016
#> GSM96962 1 0.1184 0.966 0.984 0.016
#> GSM96969 1 0.0000 0.970 1.000 0.000
#> GSM96970 1 0.0000 0.970 1.000 0.000
#> GSM96973 1 0.0000 0.970 1.000 0.000
#> GSM96976 1 0.3733 0.931 0.928 0.072
#> GSM96977 1 0.1184 0.966 0.984 0.016
#> GSM96995 1 0.3114 0.943 0.944 0.056
#> GSM97002 1 0.0000 0.970 1.000 0.000
#> GSM97009 1 0.7219 0.789 0.800 0.200
#> GSM97010 1 0.2948 0.947 0.948 0.052
#> GSM96974 1 0.2423 0.954 0.960 0.040
#> GSM96985 1 0.1184 0.966 0.984 0.016
#> GSM96959 1 0.3431 0.938 0.936 0.064
#> GSM96972 1 0.0000 0.970 1.000 0.000
#> GSM96978 1 0.1633 0.963 0.976 0.024
#> GSM96967 1 0.0000 0.970 1.000 0.000
#> GSM96987 1 0.0000 0.970 1.000 0.000
#> GSM97011 1 0.2043 0.959 0.968 0.032
#> GSM96964 1 0.0000 0.970 1.000 0.000
#> GSM96965 1 0.3274 0.940 0.940 0.060
#> GSM96981 1 0.0000 0.970 1.000 0.000
#> GSM96982 1 0.0000 0.970 1.000 0.000
#> GSM96988 1 0.1184 0.966 0.984 0.016
#> GSM97000 1 0.0000 0.970 1.000 0.000
#> GSM97004 1 0.0000 0.970 1.000 0.000
#> GSM97008 1 0.0000 0.970 1.000 0.000
#> GSM96950 1 0.0376 0.969 0.996 0.004
#> GSM96980 1 0.0000 0.970 1.000 0.000
#> GSM96989 1 0.0000 0.970 1.000 0.000
#> GSM96992 1 0.0000 0.970 1.000 0.000
#> GSM96993 1 0.2236 0.958 0.964 0.036
#> GSM96958 1 0.0000 0.970 1.000 0.000
#> GSM96951 1 0.0000 0.970 1.000 0.000
#> GSM96952 1 0.0000 0.970 1.000 0.000
#> GSM96961 1 0.0000 0.970 1.000 0.000
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM97038 2 0.0000 0.8884 0.000 1.000 0.000
#> GSM97045 2 0.0000 0.8884 0.000 1.000 0.000
#> GSM97047 2 0.5178 0.7562 0.000 0.744 0.256
#> GSM97025 2 0.0000 0.8884 0.000 1.000 0.000
#> GSM97030 2 0.5363 0.7533 0.000 0.724 0.276
#> GSM97027 2 0.0000 0.8884 0.000 1.000 0.000
#> GSM97033 2 0.0000 0.8884 0.000 1.000 0.000
#> GSM97034 2 0.5138 0.7727 0.000 0.748 0.252
#> GSM97020 2 0.0000 0.8884 0.000 1.000 0.000
#> GSM97026 2 0.0747 0.8838 0.000 0.984 0.016
#> GSM97012 2 0.0000 0.8884 0.000 1.000 0.000
#> GSM97015 2 0.5733 0.7075 0.000 0.676 0.324
#> GSM97016 2 0.0000 0.8884 0.000 1.000 0.000
#> GSM97017 3 0.6527 0.3825 0.320 0.020 0.660
#> GSM97019 2 0.0000 0.8884 0.000 1.000 0.000
#> GSM97022 2 0.0000 0.8884 0.000 1.000 0.000
#> GSM97035 2 0.0000 0.8884 0.000 1.000 0.000
#> GSM97036 3 0.7699 0.3762 0.116 0.212 0.672
#> GSM97039 2 0.0000 0.8884 0.000 1.000 0.000
#> GSM97046 2 0.0000 0.8884 0.000 1.000 0.000
#> GSM97023 1 0.5560 0.5263 0.700 0.000 0.300
#> GSM97029 3 0.7140 0.3689 0.328 0.040 0.632
#> GSM97043 2 0.2261 0.8673 0.000 0.932 0.068
#> GSM97013 1 0.9857 0.2090 0.416 0.276 0.308
#> GSM96956 2 0.4605 0.8011 0.000 0.796 0.204
#> GSM97024 2 0.2261 0.8682 0.000 0.932 0.068
#> GSM97032 2 0.4605 0.8011 0.000 0.796 0.204
#> GSM97044 2 0.5859 0.6947 0.000 0.656 0.344
#> GSM97049 2 0.0000 0.8884 0.000 1.000 0.000
#> GSM96968 3 0.4062 0.5341 0.164 0.000 0.836
#> GSM96971 3 0.5733 0.2187 0.324 0.000 0.676
#> GSM96986 3 0.2261 0.4919 0.068 0.000 0.932
#> GSM97003 3 0.5621 0.4352 0.308 0.000 0.692
#> GSM96957 3 0.5178 0.4890 0.256 0.000 0.744
#> GSM96960 1 0.5591 0.5235 0.696 0.000 0.304
#> GSM96975 3 0.4796 0.5162 0.220 0.000 0.780
#> GSM96998 1 0.5560 0.5263 0.700 0.000 0.300
#> GSM96999 3 0.5650 0.4189 0.312 0.000 0.688
#> GSM97001 3 0.5363 0.4670 0.276 0.000 0.724
#> GSM97005 1 0.5706 0.5024 0.680 0.000 0.320
#> GSM97006 1 0.6225 0.3845 0.568 0.000 0.432
#> GSM97021 3 0.5678 0.4239 0.316 0.000 0.684
#> GSM97028 3 0.3028 0.4659 0.032 0.048 0.920
#> GSM97031 1 0.6235 0.3810 0.564 0.000 0.436
#> GSM97037 2 0.4605 0.8009 0.000 0.796 0.204
#> GSM97018 2 0.6096 0.7673 0.040 0.752 0.208
#> GSM97014 2 0.7841 0.1104 0.064 0.576 0.360
#> GSM97042 2 0.0000 0.8884 0.000 1.000 0.000
#> GSM97040 3 0.3941 0.5344 0.156 0.000 0.844
#> GSM97041 3 0.9617 -0.0263 0.280 0.248 0.472
#> GSM96955 3 0.9217 0.2356 0.164 0.344 0.492
#> GSM96990 2 0.5835 0.6978 0.000 0.660 0.340
#> GSM96991 2 0.0000 0.8884 0.000 1.000 0.000
#> GSM97048 2 0.0000 0.8884 0.000 1.000 0.000
#> GSM96963 2 0.0000 0.8884 0.000 1.000 0.000
#> GSM96953 2 0.0000 0.8884 0.000 1.000 0.000
#> GSM96966 1 0.5882 0.0390 0.652 0.000 0.348
#> GSM96979 3 0.1860 0.4310 0.052 0.000 0.948
#> GSM96983 2 0.6126 0.6171 0.000 0.600 0.400
#> GSM96984 3 0.5706 0.1641 0.000 0.320 0.680
#> GSM96994 3 0.0000 0.4671 0.000 0.000 1.000
#> GSM96996 3 0.5859 0.3597 0.344 0.000 0.656
#> GSM96997 3 0.5058 0.1987 0.244 0.000 0.756
#> GSM97007 3 0.6381 0.0947 0.012 0.340 0.648
#> GSM96954 3 0.5926 -0.0815 0.356 0.000 0.644
#> GSM96962 3 0.5733 -0.0288 0.324 0.000 0.676
#> GSM96969 1 0.5706 0.0855 0.680 0.000 0.320
#> GSM96970 1 0.5948 0.0212 0.640 0.000 0.360
#> GSM96973 1 0.5948 0.0212 0.640 0.000 0.360
#> GSM96976 3 0.6286 0.2170 0.464 0.000 0.536
#> GSM96977 3 0.4178 0.5337 0.172 0.000 0.828
#> GSM96995 3 0.4002 0.5348 0.160 0.000 0.840
#> GSM97002 3 0.5926 0.3408 0.356 0.000 0.644
#> GSM97009 3 0.7366 0.3346 0.072 0.260 0.668
#> GSM97010 3 0.6544 0.5036 0.164 0.084 0.752
#> GSM96974 3 0.5785 0.2218 0.332 0.000 0.668
#> GSM96985 3 0.6302 0.2021 0.480 0.000 0.520
#> GSM96959 3 0.3941 0.5344 0.156 0.000 0.844
#> GSM96972 1 0.0000 0.3755 1.000 0.000 0.000
#> GSM96978 3 0.4235 0.3912 0.176 0.000 0.824
#> GSM96967 1 0.5591 0.1093 0.696 0.000 0.304
#> GSM96987 1 0.5560 0.5263 0.700 0.000 0.300
#> GSM97011 3 0.5631 0.5276 0.164 0.044 0.792
#> GSM96964 1 0.5591 0.5224 0.696 0.000 0.304
#> GSM96965 1 0.9319 -0.1159 0.464 0.168 0.368
#> GSM96981 3 0.5859 0.3597 0.344 0.000 0.656
#> GSM96982 3 0.6235 0.1504 0.436 0.000 0.564
#> GSM96988 3 0.1529 0.4745 0.040 0.000 0.960
#> GSM97000 3 0.4555 0.5247 0.200 0.000 0.800
#> GSM97004 1 0.0000 0.3755 1.000 0.000 0.000
#> GSM97008 3 0.5016 0.5020 0.240 0.000 0.760
#> GSM96950 1 0.6476 0.2085 0.548 0.004 0.448
#> GSM96980 1 0.5058 0.1854 0.756 0.000 0.244
#> GSM96989 1 0.5560 0.5263 0.700 0.000 0.300
#> GSM96992 1 0.5560 0.5263 0.700 0.000 0.300
#> GSM96993 3 0.6799 -0.1233 0.456 0.012 0.532
#> GSM96958 3 0.5988 0.3201 0.368 0.000 0.632
#> GSM96951 1 0.5591 0.5224 0.696 0.000 0.304
#> GSM96952 1 0.5560 0.5263 0.700 0.000 0.300
#> GSM96961 1 0.5560 0.5263 0.700 0.000 0.300
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM97038 2 0.0000 0.836 0.000 1.000 0.000 0.000
#> GSM97045 2 0.0000 0.836 0.000 1.000 0.000 0.000
#> GSM97047 2 0.6678 0.382 0.016 0.564 0.360 0.060
#> GSM97025 2 0.0000 0.836 0.000 1.000 0.000 0.000
#> GSM97030 2 0.4925 0.425 0.000 0.572 0.428 0.000
#> GSM97027 2 0.0000 0.836 0.000 1.000 0.000 0.000
#> GSM97033 2 0.0000 0.836 0.000 1.000 0.000 0.000
#> GSM97034 2 0.5353 0.408 0.000 0.556 0.432 0.012
#> GSM97020 2 0.0000 0.836 0.000 1.000 0.000 0.000
#> GSM97026 2 0.1635 0.812 0.000 0.948 0.044 0.008
#> GSM97012 2 0.0000 0.836 0.000 1.000 0.000 0.000
#> GSM97015 2 0.6008 0.284 0.020 0.504 0.464 0.012
#> GSM97016 2 0.0000 0.836 0.000 1.000 0.000 0.000
#> GSM97017 4 0.6526 0.703 0.204 0.032 0.084 0.680
#> GSM97019 2 0.0000 0.836 0.000 1.000 0.000 0.000
#> GSM97022 2 0.0000 0.836 0.000 1.000 0.000 0.000
#> GSM97035 2 0.0000 0.836 0.000 1.000 0.000 0.000
#> GSM97036 4 0.9427 0.281 0.188 0.332 0.124 0.356
#> GSM97039 2 0.0000 0.836 0.000 1.000 0.000 0.000
#> GSM97046 2 0.0000 0.836 0.000 1.000 0.000 0.000
#> GSM97023 1 0.0000 0.817 1.000 0.000 0.000 0.000
#> GSM97029 4 0.5442 0.649 0.288 0.040 0.000 0.672
#> GSM97043 2 0.3626 0.720 0.000 0.812 0.184 0.004
#> GSM97013 1 0.4746 0.439 0.632 0.368 0.000 0.000
#> GSM96956 2 0.4746 0.528 0.000 0.632 0.368 0.000
#> GSM97024 2 0.3764 0.696 0.000 0.784 0.216 0.000
#> GSM97032 2 0.5085 0.509 0.000 0.616 0.376 0.008
#> GSM97044 3 0.2408 0.745 0.000 0.104 0.896 0.000
#> GSM97049 2 0.0000 0.836 0.000 1.000 0.000 0.000
#> GSM96968 4 0.5968 0.690 0.092 0.000 0.236 0.672
#> GSM96971 3 0.3539 0.694 0.004 0.000 0.820 0.176
#> GSM96986 3 0.4134 0.565 0.000 0.000 0.740 0.260
#> GSM97003 4 0.6570 0.712 0.204 0.000 0.164 0.632
#> GSM96957 4 0.5732 0.682 0.264 0.000 0.064 0.672
#> GSM96960 1 0.0000 0.817 1.000 0.000 0.000 0.000
#> GSM96975 4 0.6118 0.706 0.208 0.000 0.120 0.672
#> GSM96998 1 0.0000 0.817 1.000 0.000 0.000 0.000
#> GSM96999 4 0.5344 0.658 0.300 0.000 0.032 0.668
#> GSM97001 4 0.6078 0.715 0.152 0.000 0.164 0.684
#> GSM97005 1 0.3616 0.707 0.852 0.000 0.112 0.036
#> GSM97006 1 0.0000 0.817 1.000 0.000 0.000 0.000
#> GSM97021 4 0.6650 0.695 0.200 0.000 0.176 0.624
#> GSM97028 3 0.5430 0.595 0.012 0.036 0.716 0.236
#> GSM97031 1 0.2469 0.732 0.892 0.000 0.108 0.000
#> GSM97037 2 0.4761 0.523 0.000 0.628 0.372 0.000
#> GSM97018 2 0.5414 0.494 0.000 0.604 0.376 0.020
#> GSM97014 2 0.6965 -0.246 0.000 0.460 0.112 0.428
#> GSM97042 2 0.0000 0.836 0.000 1.000 0.000 0.000
#> GSM97040 4 0.6133 0.665 0.088 0.000 0.268 0.644
#> GSM97041 1 0.7889 0.217 0.460 0.336 0.012 0.192
#> GSM96955 4 0.6033 0.557 0.000 0.204 0.116 0.680
#> GSM96990 3 0.1452 0.808 0.000 0.036 0.956 0.008
#> GSM96991 2 0.0188 0.835 0.000 0.996 0.004 0.000
#> GSM97048 2 0.0000 0.836 0.000 1.000 0.000 0.000
#> GSM96963 2 0.0000 0.836 0.000 1.000 0.000 0.000
#> GSM96953 2 0.0000 0.836 0.000 1.000 0.000 0.000
#> GSM96966 4 0.0592 0.655 0.016 0.000 0.000 0.984
#> GSM96979 3 0.2737 0.784 0.008 0.000 0.888 0.104
#> GSM96983 3 0.2530 0.732 0.000 0.112 0.888 0.000
#> GSM96984 3 0.0336 0.817 0.000 0.008 0.992 0.000
#> GSM96994 3 0.0000 0.817 0.000 0.000 1.000 0.000
#> GSM96996 4 0.4564 0.630 0.328 0.000 0.000 0.672
#> GSM96997 3 0.4059 0.683 0.200 0.000 0.788 0.012
#> GSM97007 3 0.0000 0.817 0.000 0.000 1.000 0.000
#> GSM96954 3 0.2329 0.801 0.072 0.000 0.916 0.012
#> GSM96962 3 0.0469 0.818 0.012 0.000 0.988 0.000
#> GSM96969 4 0.2216 0.617 0.092 0.000 0.000 0.908
#> GSM96970 4 0.0592 0.655 0.016 0.000 0.000 0.984
#> GSM96973 4 0.0592 0.655 0.016 0.000 0.000 0.984
#> GSM96976 4 0.0469 0.650 0.000 0.000 0.012 0.988
#> GSM96977 4 0.5964 0.691 0.096 0.000 0.228 0.676
#> GSM96995 4 0.6295 0.641 0.088 0.000 0.296 0.616
#> GSM97002 4 0.4730 0.593 0.364 0.000 0.000 0.636
#> GSM97009 4 0.8660 0.391 0.088 0.372 0.120 0.420
#> GSM97010 4 0.7046 0.697 0.092 0.060 0.188 0.660
#> GSM96974 4 0.4994 -0.363 0.000 0.000 0.480 0.520
#> GSM96985 4 0.0657 0.650 0.004 0.000 0.012 0.984
#> GSM96959 4 0.6508 0.574 0.088 0.000 0.344 0.568
#> GSM96972 1 0.4500 0.529 0.684 0.000 0.000 0.316
#> GSM96978 3 0.4697 0.546 0.000 0.000 0.644 0.356
#> GSM96967 4 0.2921 0.570 0.140 0.000 0.000 0.860
#> GSM96987 1 0.0000 0.817 1.000 0.000 0.000 0.000
#> GSM97011 4 0.5848 0.689 0.088 0.000 0.228 0.684
#> GSM96964 1 0.0336 0.813 0.992 0.000 0.000 0.008
#> GSM96965 4 0.0469 0.651 0.000 0.012 0.000 0.988
#> GSM96981 4 0.4585 0.627 0.332 0.000 0.000 0.668
#> GSM96982 1 0.4992 -0.322 0.524 0.000 0.000 0.476
#> GSM96988 3 0.6576 0.554 0.152 0.000 0.628 0.220
#> GSM97000 4 0.6081 0.670 0.088 0.000 0.260 0.652
#> GSM97004 1 0.2216 0.746 0.908 0.000 0.000 0.092
#> GSM97008 4 0.5944 0.699 0.104 0.000 0.212 0.684
#> GSM96950 1 0.4431 0.338 0.696 0.000 0.000 0.304
#> GSM96980 4 0.4072 0.388 0.252 0.000 0.000 0.748
#> GSM96989 1 0.0000 0.817 1.000 0.000 0.000 0.000
#> GSM96992 1 0.0000 0.817 1.000 0.000 0.000 0.000
#> GSM96993 1 0.6451 0.485 0.656 0.004 0.204 0.136
#> GSM96958 4 0.4804 0.568 0.384 0.000 0.000 0.616
#> GSM96951 1 0.0592 0.808 0.984 0.000 0.000 0.016
#> GSM96952 1 0.0000 0.817 1.000 0.000 0.000 0.000
#> GSM96961 1 0.0000 0.817 1.000 0.000 0.000 0.000
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM97038 2 0.0000 0.8619 0.000 1.000 0.000 0.000 0.000
#> GSM97045 2 0.0510 0.8601 0.000 0.984 0.000 0.016 0.000
#> GSM97047 5 0.5237 0.6067 0.000 0.160 0.140 0.004 0.696
#> GSM97025 2 0.0162 0.8617 0.000 0.996 0.000 0.004 0.000
#> GSM97030 2 0.4331 0.5243 0.000 0.596 0.400 0.004 0.000
#> GSM97027 2 0.0162 0.8617 0.000 0.996 0.000 0.004 0.000
#> GSM97033 2 0.0000 0.8619 0.000 1.000 0.000 0.000 0.000
#> GSM97034 2 0.6093 0.6690 0.000 0.612 0.204 0.012 0.172
#> GSM97020 2 0.0000 0.8619 0.000 1.000 0.000 0.000 0.000
#> GSM97026 2 0.1560 0.8480 0.000 0.948 0.020 0.004 0.028
#> GSM97012 2 0.2793 0.8540 0.000 0.876 0.000 0.036 0.088
#> GSM97015 2 0.5949 0.4353 0.000 0.528 0.368 0.004 0.100
#> GSM97016 2 0.0000 0.8619 0.000 1.000 0.000 0.000 0.000
#> GSM97017 5 0.2879 0.7550 0.100 0.032 0.000 0.000 0.868
#> GSM97019 2 0.2793 0.8540 0.000 0.876 0.000 0.036 0.088
#> GSM97022 2 0.2952 0.8536 0.000 0.872 0.004 0.036 0.088
#> GSM97035 2 0.2952 0.8536 0.000 0.872 0.004 0.036 0.088
#> GSM97036 5 0.8336 0.1734 0.248 0.312 0.112 0.004 0.324
#> GSM97039 2 0.0000 0.8619 0.000 1.000 0.000 0.000 0.000
#> GSM97046 2 0.0000 0.8619 0.000 1.000 0.000 0.000 0.000
#> GSM97023 1 0.0000 0.8650 1.000 0.000 0.000 0.000 0.000
#> GSM97029 5 0.6586 0.0318 0.384 0.208 0.000 0.000 0.408
#> GSM97043 2 0.2694 0.8137 0.000 0.864 0.128 0.004 0.004
#> GSM97013 1 0.4060 0.4771 0.640 0.360 0.000 0.000 0.000
#> GSM96956 2 0.3790 0.7116 0.000 0.724 0.272 0.004 0.000
#> GSM97024 2 0.5143 0.7877 0.000 0.740 0.136 0.036 0.088
#> GSM97032 2 0.5213 0.6436 0.000 0.652 0.276 0.004 0.068
#> GSM97044 3 0.0324 0.8589 0.000 0.000 0.992 0.004 0.004
#> GSM97049 2 0.0000 0.8619 0.000 1.000 0.000 0.000 0.000
#> GSM96968 3 0.3980 0.6078 0.000 0.008 0.708 0.000 0.284
#> GSM96971 3 0.3098 0.8212 0.000 0.000 0.836 0.016 0.148
#> GSM96986 3 0.3003 0.7911 0.000 0.000 0.812 0.000 0.188
#> GSM97003 5 0.4149 0.7211 0.128 0.000 0.088 0.000 0.784
#> GSM96957 1 0.4863 0.5639 0.672 0.000 0.056 0.000 0.272
#> GSM96960 1 0.0000 0.8650 1.000 0.000 0.000 0.000 0.000
#> GSM96975 5 0.2685 0.7662 0.028 0.000 0.092 0.000 0.880
#> GSM96998 1 0.0000 0.8650 1.000 0.000 0.000 0.000 0.000
#> GSM96999 1 0.4464 0.5828 0.684 0.000 0.028 0.000 0.288
#> GSM97001 5 0.2228 0.7706 0.040 0.000 0.048 0.000 0.912
#> GSM97005 5 0.4015 0.4849 0.348 0.000 0.000 0.000 0.652
#> GSM97006 1 0.0000 0.8650 1.000 0.000 0.000 0.000 0.000
#> GSM97021 5 0.2853 0.7656 0.072 0.000 0.052 0.000 0.876
#> GSM97028 3 0.3561 0.6537 0.000 0.000 0.740 0.000 0.260
#> GSM97031 1 0.0510 0.8564 0.984 0.000 0.000 0.000 0.016
#> GSM97037 2 0.3766 0.7156 0.000 0.728 0.268 0.004 0.000
#> GSM97018 2 0.5706 0.5920 0.000 0.612 0.276 0.004 0.108
#> GSM97014 5 0.3895 0.6148 0.000 0.320 0.000 0.000 0.680
#> GSM97042 2 0.2793 0.8540 0.000 0.876 0.000 0.036 0.088
#> GSM97040 5 0.1965 0.7595 0.000 0.000 0.096 0.000 0.904
#> GSM97041 5 0.5354 0.6313 0.108 0.240 0.000 0.000 0.652
#> GSM96955 5 0.3532 0.7544 0.000 0.076 0.092 0.000 0.832
#> GSM96990 3 0.2928 0.7886 0.000 0.032 0.872 0.004 0.092
#> GSM96991 2 0.2793 0.8540 0.000 0.876 0.000 0.036 0.088
#> GSM97048 2 0.0000 0.8619 0.000 1.000 0.000 0.000 0.000
#> GSM96963 2 0.2793 0.8540 0.000 0.876 0.000 0.036 0.088
#> GSM96953 2 0.2793 0.8540 0.000 0.876 0.000 0.036 0.088
#> GSM96966 4 0.1043 0.9611 0.000 0.000 0.000 0.960 0.040
#> GSM96979 3 0.3967 0.7854 0.000 0.000 0.800 0.108 0.092
#> GSM96983 3 0.0162 0.8593 0.000 0.000 0.996 0.004 0.000
#> GSM96984 3 0.0000 0.8603 0.000 0.000 1.000 0.000 0.000
#> GSM96994 3 0.0162 0.8611 0.000 0.000 0.996 0.000 0.004
#> GSM96996 5 0.3992 0.5894 0.268 0.000 0.000 0.012 0.720
#> GSM96997 3 0.3812 0.8012 0.096 0.000 0.812 0.000 0.092
#> GSM97007 3 0.0000 0.8603 0.000 0.000 1.000 0.000 0.000
#> GSM96954 3 0.2820 0.8500 0.056 0.000 0.884 0.004 0.056
#> GSM96962 3 0.1597 0.8612 0.012 0.000 0.940 0.000 0.048
#> GSM96969 4 0.1195 0.9551 0.028 0.000 0.000 0.960 0.012
#> GSM96970 4 0.1043 0.9611 0.000 0.000 0.000 0.960 0.040
#> GSM96973 4 0.1043 0.9611 0.000 0.000 0.000 0.960 0.040
#> GSM96976 4 0.1043 0.9611 0.000 0.000 0.000 0.960 0.040
#> GSM96977 5 0.1952 0.7630 0.004 0.000 0.084 0.000 0.912
#> GSM96995 5 0.2561 0.7396 0.000 0.000 0.144 0.000 0.856
#> GSM97002 1 0.3756 0.6668 0.744 0.000 0.000 0.008 0.248
#> GSM97009 5 0.3999 0.5870 0.000 0.344 0.000 0.000 0.656
#> GSM97010 5 0.3861 0.7265 0.000 0.128 0.068 0.000 0.804
#> GSM96974 4 0.1082 0.9439 0.000 0.000 0.028 0.964 0.008
#> GSM96985 4 0.1282 0.9570 0.000 0.000 0.004 0.952 0.044
#> GSM96959 5 0.2280 0.7509 0.000 0.000 0.120 0.000 0.880
#> GSM96972 4 0.1043 0.9486 0.040 0.000 0.000 0.960 0.000
#> GSM96978 3 0.2389 0.8315 0.000 0.000 0.880 0.004 0.116
#> GSM96967 4 0.1043 0.9486 0.040 0.000 0.000 0.960 0.000
#> GSM96987 1 0.0000 0.8650 1.000 0.000 0.000 0.000 0.000
#> GSM97011 5 0.1851 0.7606 0.000 0.000 0.088 0.000 0.912
#> GSM96964 1 0.0000 0.8650 1.000 0.000 0.000 0.000 0.000
#> GSM96965 4 0.2424 0.8586 0.000 0.000 0.000 0.868 0.132
#> GSM96981 5 0.3885 0.5919 0.268 0.000 0.000 0.008 0.724
#> GSM96982 1 0.3109 0.7231 0.800 0.000 0.000 0.000 0.200
#> GSM96988 3 0.3967 0.8039 0.000 0.000 0.800 0.092 0.108
#> GSM97000 5 0.2230 0.7496 0.000 0.000 0.116 0.000 0.884
#> GSM97004 1 0.0510 0.8564 0.984 0.000 0.000 0.016 0.000
#> GSM97008 5 0.2171 0.7696 0.024 0.000 0.064 0.000 0.912
#> GSM96950 1 0.3003 0.7491 0.812 0.000 0.000 0.000 0.188
#> GSM96980 4 0.1043 0.9486 0.040 0.000 0.000 0.960 0.000
#> GSM96989 1 0.0000 0.8650 1.000 0.000 0.000 0.000 0.000
#> GSM96992 1 0.0000 0.8650 1.000 0.000 0.000 0.000 0.000
#> GSM96993 1 0.5869 0.5462 0.656 0.012 0.160 0.004 0.168
#> GSM96958 1 0.3534 0.6601 0.744 0.000 0.000 0.000 0.256
#> GSM96951 1 0.0000 0.8650 1.000 0.000 0.000 0.000 0.000
#> GSM96952 1 0.0000 0.8650 1.000 0.000 0.000 0.000 0.000
#> GSM96961 1 0.0000 0.8650 1.000 0.000 0.000 0.000 0.000
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM97038 2 0.0713 0.6085 0.000 0.972 0.028 0.000 0.000 0.000
#> GSM97045 2 0.3607 0.0847 0.000 0.652 0.348 0.000 0.000 0.000
#> GSM97047 5 0.5279 0.5438 0.000 0.048 0.312 0.004 0.604 0.032
#> GSM97025 2 0.2378 0.5184 0.000 0.848 0.152 0.000 0.000 0.000
#> GSM97030 2 0.6546 0.2008 0.000 0.464 0.260 0.004 0.028 0.244
#> GSM97027 2 0.2300 0.5286 0.000 0.856 0.144 0.000 0.000 0.000
#> GSM97033 2 0.1610 0.5809 0.000 0.916 0.084 0.000 0.000 0.000
#> GSM97034 3 0.6310 -0.0124 0.000 0.204 0.584 0.004 0.112 0.096
#> GSM97020 2 0.1765 0.5724 0.000 0.904 0.096 0.000 0.000 0.000
#> GSM97026 2 0.3841 0.5190 0.000 0.780 0.164 0.000 0.036 0.020
#> GSM97012 3 0.3797 0.3883 0.000 0.420 0.580 0.000 0.000 0.000
#> GSM97015 3 0.7118 -0.0935 0.000 0.320 0.420 0.004 0.144 0.112
#> GSM97016 2 0.0000 0.6167 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97017 5 0.3771 0.7299 0.056 0.000 0.180 0.000 0.764 0.000
#> GSM97019 3 0.3797 0.3883 0.000 0.420 0.580 0.000 0.000 0.000
#> GSM97022 3 0.3607 0.3906 0.000 0.348 0.652 0.000 0.000 0.000
#> GSM97035 3 0.3756 0.3305 0.000 0.400 0.600 0.000 0.000 0.000
#> GSM97036 3 0.8648 -0.0207 0.152 0.200 0.320 0.004 0.236 0.088
#> GSM97039 2 0.0363 0.6130 0.000 0.988 0.012 0.000 0.000 0.000
#> GSM97046 2 0.0000 0.6167 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97023 1 0.0000 0.8199 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM97029 5 0.6552 0.0521 0.320 0.100 0.084 0.004 0.492 0.000
#> GSM97043 2 0.4676 0.4508 0.000 0.684 0.216 0.004 0.000 0.096
#> GSM97013 1 0.3823 0.2241 0.564 0.436 0.000 0.000 0.000 0.000
#> GSM96956 2 0.5329 0.4011 0.000 0.656 0.216 0.004 0.028 0.096
#> GSM97024 3 0.2969 0.3294 0.000 0.224 0.776 0.000 0.000 0.000
#> GSM97032 2 0.6512 0.2397 0.000 0.484 0.336 0.004 0.080 0.096
#> GSM97044 6 0.3608 0.6872 0.000 0.000 0.248 0.004 0.012 0.736
#> GSM97049 2 0.0000 0.6167 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM96968 6 0.5594 0.1756 0.004 0.008 0.080 0.004 0.440 0.464
#> GSM96971 6 0.2527 0.7894 0.000 0.000 0.000 0.024 0.108 0.868
#> GSM96986 6 0.1765 0.7904 0.000 0.000 0.000 0.000 0.096 0.904
#> GSM97003 5 0.3308 0.6610 0.096 0.000 0.000 0.004 0.828 0.072
#> GSM96957 1 0.3944 0.4162 0.568 0.000 0.000 0.004 0.428 0.000
#> GSM96960 1 0.0146 0.8188 0.996 0.000 0.000 0.004 0.000 0.000
#> GSM96975 5 0.1049 0.7324 0.032 0.000 0.000 0.008 0.960 0.000
#> GSM96998 1 0.0000 0.8199 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM96999 1 0.4062 0.3912 0.552 0.000 0.000 0.008 0.440 0.000
#> GSM97001 5 0.0603 0.7394 0.016 0.000 0.000 0.004 0.980 0.000
#> GSM97005 5 0.5998 0.5490 0.240 0.000 0.176 0.000 0.556 0.028
#> GSM97006 1 0.0146 0.8188 0.996 0.000 0.000 0.004 0.000 0.000
#> GSM97021 5 0.4196 0.7154 0.040 0.000 0.180 0.000 0.752 0.028
#> GSM97028 3 0.6095 -0.3308 0.000 0.000 0.392 0.004 0.224 0.380
#> GSM97031 1 0.1036 0.8017 0.964 0.000 0.000 0.004 0.024 0.008
#> GSM97037 2 0.5225 0.4113 0.000 0.672 0.200 0.004 0.028 0.096
#> GSM97018 2 0.6996 0.0956 0.000 0.388 0.376 0.004 0.136 0.096
#> GSM97014 5 0.5475 0.4803 0.000 0.316 0.148 0.000 0.536 0.000
#> GSM97042 3 0.3797 0.3883 0.000 0.420 0.580 0.000 0.000 0.000
#> GSM97040 5 0.3345 0.7197 0.000 0.000 0.184 0.000 0.788 0.028
#> GSM97041 5 0.6409 0.5385 0.064 0.216 0.180 0.000 0.540 0.000
#> GSM96955 5 0.1908 0.7143 0.000 0.096 0.000 0.004 0.900 0.000
#> GSM96990 6 0.5518 0.4740 0.000 0.000 0.316 0.004 0.136 0.544
#> GSM96991 3 0.3774 0.3904 0.000 0.408 0.592 0.000 0.000 0.000
#> GSM97048 2 0.0000 0.6167 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM96963 3 0.3797 0.3883 0.000 0.420 0.580 0.000 0.000 0.000
#> GSM96953 2 0.3864 -0.3298 0.000 0.520 0.480 0.000 0.000 0.000
#> GSM96966 4 0.0146 0.9722 0.000 0.000 0.000 0.996 0.004 0.000
#> GSM96979 6 0.1765 0.7904 0.000 0.000 0.000 0.000 0.096 0.904
#> GSM96983 6 0.3517 0.7237 0.000 0.000 0.188 0.004 0.028 0.780
#> GSM96984 6 0.0713 0.8122 0.000 0.000 0.028 0.000 0.000 0.972
#> GSM96994 6 0.0547 0.8144 0.000 0.000 0.020 0.000 0.000 0.980
#> GSM96996 5 0.3740 0.5180 0.228 0.000 0.000 0.032 0.740 0.000
#> GSM96997 6 0.1765 0.7904 0.000 0.000 0.000 0.000 0.096 0.904
#> GSM97007 6 0.0713 0.8122 0.000 0.000 0.028 0.000 0.000 0.972
#> GSM96954 6 0.2683 0.7961 0.004 0.000 0.020 0.004 0.104 0.868
#> GSM96962 6 0.0547 0.8148 0.000 0.000 0.000 0.000 0.020 0.980
#> GSM96969 4 0.0260 0.9709 0.008 0.000 0.000 0.992 0.000 0.000
#> GSM96970 4 0.0146 0.9722 0.000 0.000 0.000 0.996 0.004 0.000
#> GSM96973 4 0.0146 0.9722 0.000 0.000 0.000 0.996 0.004 0.000
#> GSM96976 4 0.0363 0.9677 0.000 0.000 0.000 0.988 0.012 0.000
#> GSM96977 5 0.0551 0.7385 0.008 0.000 0.000 0.004 0.984 0.004
#> GSM96995 5 0.3952 0.6879 0.000 0.000 0.108 0.008 0.780 0.104
#> GSM97002 1 0.4301 0.4579 0.584 0.000 0.000 0.024 0.392 0.000
#> GSM97009 5 0.3619 0.5309 0.000 0.316 0.004 0.000 0.680 0.000
#> GSM97010 5 0.2951 0.6932 0.004 0.092 0.008 0.004 0.864 0.028
#> GSM96974 4 0.0000 0.9710 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM96985 4 0.0717 0.9641 0.000 0.000 0.000 0.976 0.016 0.008
#> GSM96959 5 0.3377 0.7184 0.000 0.000 0.188 0.000 0.784 0.028
#> GSM96972 4 0.0260 0.9709 0.008 0.000 0.000 0.992 0.000 0.000
#> GSM96978 6 0.1285 0.8100 0.000 0.000 0.004 0.000 0.052 0.944
#> GSM96967 4 0.0260 0.9709 0.008 0.000 0.000 0.992 0.000 0.000
#> GSM96987 1 0.0000 0.8199 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM97011 5 0.0291 0.7379 0.000 0.000 0.000 0.004 0.992 0.004
#> GSM96964 1 0.0000 0.8199 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM96965 4 0.2703 0.7837 0.000 0.004 0.000 0.824 0.172 0.000
#> GSM96981 5 0.3566 0.5302 0.224 0.000 0.000 0.024 0.752 0.000
#> GSM96982 1 0.3619 0.5716 0.680 0.000 0.000 0.004 0.316 0.000
#> GSM96988 6 0.4734 0.7317 0.000 0.000 0.152 0.080 0.040 0.728
#> GSM97000 5 0.3712 0.7117 0.000 0.000 0.180 0.000 0.768 0.052
#> GSM97004 1 0.0547 0.8117 0.980 0.000 0.000 0.020 0.000 0.000
#> GSM97008 5 0.3562 0.7233 0.008 0.000 0.180 0.000 0.784 0.028
#> GSM96950 1 0.3445 0.6542 0.744 0.000 0.012 0.000 0.244 0.000
#> GSM96980 4 0.0363 0.9687 0.012 0.000 0.000 0.988 0.000 0.000
#> GSM96989 1 0.0000 0.8199 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM96992 1 0.0000 0.8199 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM96993 1 0.6388 0.3964 0.580 0.000 0.132 0.004 0.188 0.096
#> GSM96958 1 0.3841 0.4869 0.616 0.000 0.000 0.004 0.380 0.000
#> GSM96951 1 0.0000 0.8199 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM96952 1 0.0146 0.8188 0.996 0.000 0.000 0.004 0.000 0.000
#> GSM96961 1 0.0000 0.8199 1.000 0.000 0.000 0.000 0.000 0.000
Heatmaps for the consensus matrix. It visualizes the probability of two samples to be in a same group.
consensus_heatmap(res, k = 2)
consensus_heatmap(res, k = 3)
consensus_heatmap(res, k = 4)
consensus_heatmap(res, k = 5)
consensus_heatmap(res, k = 6)
Heatmaps for the membership of samples in all partitions to see how consistent they are:
membership_heatmap(res, k = 2)
membership_heatmap(res, k = 3)
membership_heatmap(res, k = 4)
membership_heatmap(res, k = 5)
membership_heatmap(res, k = 6)
As soon as we have had the classes for columns, we can look for signatures which are significantly different between classes which can be candidate marks for certain classes. Following are the heatmaps for signatures.
Signature heatmaps where rows are scaled:
get_signatures(res, k = 2)
get_signatures(res, k = 3)
get_signatures(res, k = 4)
get_signatures(res, k = 5)
get_signatures(res, k = 6)
Signature heatmaps where rows are not scaled:
get_signatures(res, k = 2, scale_rows = FALSE)
get_signatures(res, k = 3, scale_rows = FALSE)
get_signatures(res, k = 4, scale_rows = FALSE)
get_signatures(res, k = 5, scale_rows = FALSE)
get_signatures(res, k = 6, scale_rows = FALSE)
Compare the overlap of signatures from different k:
compare_signatures(res)
get_signature() returns a data frame invisibly. TO get the list of signatures, the function
call should be assigned to a variable explicitly. In following code, if plot argument is set
to FALSE, no heatmap is plotted while only the differential analysis is performed.
# code only for demonstration
tb = get_signature(res, k = ..., plot = FALSE)
An example of the output of tb is:
#> which_row fdr mean_1 mean_2 scaled_mean_1 scaled_mean_2 km
#> 1 38 0.042760348 8.373488 9.131774 -0.5533452 0.5164555 1
#> 2 40 0.018707592 7.106213 8.469186 -0.6173731 0.5762149 1
#> 3 55 0.019134737 10.221463 11.207825 -0.6159697 0.5749050 1
#> 4 59 0.006059896 5.921854 7.869574 -0.6899429 0.6439467 1
#> 5 60 0.018055526 8.928898 10.211722 -0.6204761 0.5791110 1
#> 6 98 0.009384629 15.714769 14.887706 0.6635654 -0.6193277 2
...
The columns in tb are:
which_row: row indices corresponding to the input matrix.fdr: FDR for the differential test. mean_x: The mean value in group x.scaled_mean_x: The mean value in group x after rows are scaled.km: Row groups if k-means clustering is applied to rows.UMAP plot which shows how samples are separated.
dimension_reduction(res, k = 2, method = "UMAP")
dimension_reduction(res, k = 3, method = "UMAP")
dimension_reduction(res, k = 4, method = "UMAP")
dimension_reduction(res, k = 5, method = "UMAP")
dimension_reduction(res, k = 6, method = "UMAP")
Following heatmap shows how subgroups are split when increasing k:
collect_classes(res)
Test correlation between subgroups and known annotations. If the known annotation is numeric, one-way ANOVA test is applied, and if the known annotation is discrete, chi-squared contingency table test is applied.
test_to_known_factors(res)
#> n disease.state(p) specimen(p) cell.type(p) other(p) k
#> MAD:pam 100 1.39e-06 0.440 9.32e-16 0.05269 2
#> MAD:pam 54 3.15e-03 0.810 1.13e-13 0.12815 3
#> MAD:pam 85 1.60e-05 0.826 1.90e-17 0.00887 4
#> MAD:pam 95 1.81e-07 0.279 1.77e-20 0.01760 5
#> MAD:pam 70 1.72e-06 0.339 2.38e-14 0.01998 6
If matrix rows can be associated to genes, consider to use functional_enrichment(res,
...) to perform function enrichment for the signature genes. See this vignette for more detailed explanations.
The object with results only for a single top-value method and a single partition method can be extracted as:
res = res_list["MAD", "mclust"]
# you can also extract it by
# res = res_list["MAD:mclust"]
A summary of res and all the functions that can be applied to it:
res
#> A 'ConsensusPartition' object with k = 2, 3, 4, 5, 6.
#> On a matrix with 21168 rows and 100 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 1.000 0.974 0.980 0.3429 0.665 0.665
#> 3 3 0.862 0.879 0.946 0.8111 0.659 0.507
#> 4 4 0.853 0.868 0.936 0.1777 0.876 0.678
#> 5 5 0.791 0.860 0.877 0.0764 0.898 0.649
#> 6 6 0.827 0.833 0.838 0.0429 0.948 0.756
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
#> GSM97038 2 0.1184 0.995 0.016 0.984
#> GSM97045 2 0.1184 0.995 0.016 0.984
#> GSM97047 1 0.2423 0.965 0.960 0.040
#> GSM97025 2 0.1184 0.995 0.016 0.984
#> GSM97030 1 0.2603 0.962 0.956 0.044
#> GSM97027 2 0.1184 0.995 0.016 0.984
#> GSM97033 2 0.1184 0.995 0.016 0.984
#> GSM97034 1 0.2236 0.968 0.964 0.036
#> GSM97020 2 0.1184 0.995 0.016 0.984
#> GSM97026 1 0.2423 0.965 0.960 0.040
#> GSM97012 2 0.1184 0.995 0.016 0.984
#> GSM97015 1 0.1414 0.978 0.980 0.020
#> GSM97016 2 0.1184 0.995 0.016 0.984
#> GSM97017 1 0.0000 0.981 1.000 0.000
#> GSM97019 2 0.1184 0.995 0.016 0.984
#> GSM97022 2 0.1184 0.995 0.016 0.984
#> GSM97035 2 0.1184 0.995 0.016 0.984
#> GSM97036 1 0.0000 0.981 1.000 0.000
#> GSM97039 2 0.1184 0.995 0.016 0.984
#> GSM97046 2 0.1184 0.995 0.016 0.984
#> GSM97023 1 0.0000 0.981 1.000 0.000
#> GSM97029 1 0.0000 0.981 1.000 0.000
#> GSM97043 2 0.1184 0.995 0.016 0.984
#> GSM97013 1 0.0000 0.981 1.000 0.000
#> GSM96956 1 0.8386 0.658 0.732 0.268
#> GSM97024 2 0.1184 0.995 0.016 0.984
#> GSM97032 1 0.2423 0.965 0.960 0.040
#> GSM97044 1 0.2423 0.965 0.960 0.040
#> GSM97049 2 0.1184 0.995 0.016 0.984
#> GSM96968 1 0.1184 0.980 0.984 0.016
#> GSM96971 1 0.1184 0.980 0.984 0.016
#> GSM96986 1 0.1184 0.980 0.984 0.016
#> GSM97003 1 0.0672 0.981 0.992 0.008
#> GSM96957 1 0.0376 0.982 0.996 0.004
#> GSM96960 1 0.0000 0.981 1.000 0.000
#> GSM96975 1 0.0000 0.981 1.000 0.000
#> GSM96998 1 0.0000 0.981 1.000 0.000
#> GSM96999 1 0.0376 0.982 0.996 0.004
#> GSM97001 1 0.0000 0.981 1.000 0.000
#> GSM97005 1 0.0672 0.981 0.992 0.008
#> GSM97006 1 0.0000 0.981 1.000 0.000
#> GSM97021 1 0.0672 0.981 0.992 0.008
#> GSM97028 1 0.1184 0.980 0.984 0.016
#> GSM97031 1 0.0672 0.981 0.992 0.008
#> GSM97037 1 0.8327 0.665 0.736 0.264
#> GSM97018 1 0.2423 0.965 0.960 0.040
#> GSM97014 1 0.1633 0.975 0.976 0.024
#> GSM97042 2 0.1184 0.995 0.016 0.984
#> GSM97040 1 0.0672 0.981 0.992 0.008
#> GSM97041 1 0.0376 0.982 0.996 0.004
#> GSM96955 1 0.4298 0.919 0.912 0.088
#> GSM96990 1 0.2423 0.965 0.960 0.040
#> GSM96991 2 0.4690 0.911 0.100 0.900
#> GSM97048 2 0.1184 0.995 0.016 0.984
#> GSM96963 2 0.1633 0.988 0.024 0.976
#> GSM96953 2 0.1184 0.995 0.016 0.984
#> GSM96966 1 0.1184 0.972 0.984 0.016
#> GSM96979 1 0.1184 0.980 0.984 0.016
#> GSM96983 1 0.1414 0.978 0.980 0.020
#> GSM96984 1 0.1184 0.980 0.984 0.016
#> GSM96994 1 0.1184 0.980 0.984 0.016
#> GSM96996 1 0.0000 0.981 1.000 0.000
#> GSM96997 1 0.1184 0.980 0.984 0.016
#> GSM97007 1 0.1184 0.980 0.984 0.016
#> GSM96954 1 0.1184 0.980 0.984 0.016
#> GSM96962 1 0.1184 0.980 0.984 0.016
#> GSM96969 1 0.1184 0.972 0.984 0.016
#> GSM96970 1 0.1184 0.972 0.984 0.016
#> GSM96973 1 0.1184 0.972 0.984 0.016
#> GSM96976 1 0.0672 0.981 0.992 0.008
#> GSM96977 1 0.0672 0.981 0.992 0.008
#> GSM96995 1 0.1184 0.980 0.984 0.016
#> GSM97002 1 0.0938 0.975 0.988 0.012
#> GSM97009 1 0.1633 0.976 0.976 0.024
#> GSM97010 1 0.0000 0.981 1.000 0.000
#> GSM96974 1 0.0672 0.981 0.992 0.008
#> GSM96985 1 0.0672 0.981 0.992 0.008
#> GSM96959 1 0.1414 0.978 0.980 0.020
#> GSM96972 1 0.1184 0.972 0.984 0.016
#> GSM96978 1 0.1184 0.980 0.984 0.016
#> GSM96967 1 0.1184 0.972 0.984 0.016
#> GSM96987 1 0.0000 0.981 1.000 0.000
#> GSM97011 1 0.0672 0.981 0.992 0.008
#> GSM96964 1 0.0000 0.981 1.000 0.000
#> GSM96965 1 0.0000 0.981 1.000 0.000
#> GSM96981 1 0.0000 0.981 1.000 0.000
#> GSM96982 1 0.0376 0.980 0.996 0.004
#> GSM96988 1 0.1184 0.980 0.984 0.016
#> GSM97000 1 0.0672 0.981 0.992 0.008
#> GSM97004 1 0.1184 0.972 0.984 0.016
#> GSM97008 1 0.0672 0.981 0.992 0.008
#> GSM96950 1 0.0000 0.981 1.000 0.000
#> GSM96980 1 0.1184 0.972 0.984 0.016
#> GSM96989 1 0.0000 0.981 1.000 0.000
#> GSM96992 1 0.0000 0.981 1.000 0.000
#> GSM96993 1 0.0000 0.981 1.000 0.000
#> GSM96958 1 0.0000 0.981 1.000 0.000
#> GSM96951 1 0.0672 0.981 0.992 0.008
#> GSM96952 1 0.0000 0.981 1.000 0.000
#> GSM96961 1 0.0000 0.981 1.000 0.000
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM97038 2 0.0000 0.9077 0.000 1.000 0.000
#> GSM97045 2 0.0000 0.9077 0.000 1.000 0.000
#> GSM97047 2 0.8331 0.5012 0.208 0.628 0.164
#> GSM97025 2 0.0000 0.9077 0.000 1.000 0.000
#> GSM97030 3 0.0237 0.9240 0.000 0.004 0.996
#> GSM97027 2 0.0000 0.9077 0.000 1.000 0.000
#> GSM97033 2 0.0000 0.9077 0.000 1.000 0.000
#> GSM97034 3 0.0475 0.9253 0.004 0.004 0.992
#> GSM97020 2 0.0000 0.9077 0.000 1.000 0.000
#> GSM97026 2 0.7001 0.4539 0.340 0.628 0.032
#> GSM97012 2 0.0000 0.9077 0.000 1.000 0.000
#> GSM97015 3 0.0237 0.9267 0.004 0.000 0.996
#> GSM97016 2 0.0000 0.9077 0.000 1.000 0.000
#> GSM97017 1 0.0747 0.9631 0.984 0.000 0.016
#> GSM97019 2 0.0000 0.9077 0.000 1.000 0.000
#> GSM97022 2 0.0000 0.9077 0.000 1.000 0.000
#> GSM97035 2 0.0000 0.9077 0.000 1.000 0.000
#> GSM97036 1 0.0747 0.9631 0.984 0.000 0.016
#> GSM97039 2 0.0000 0.9077 0.000 1.000 0.000
#> GSM97046 2 0.0000 0.9077 0.000 1.000 0.000
#> GSM97023 1 0.0000 0.9604 1.000 0.000 0.000
#> GSM97029 1 0.0747 0.9631 0.984 0.000 0.016
#> GSM97043 2 0.0000 0.9077 0.000 1.000 0.000
#> GSM97013 1 0.0747 0.9631 0.984 0.000 0.016
#> GSM96956 2 0.5760 0.4952 0.000 0.672 0.328
#> GSM97024 2 0.0237 0.9042 0.000 0.996 0.004
#> GSM97032 3 0.6026 0.3667 0.000 0.376 0.624
#> GSM97044 3 0.0237 0.9267 0.004 0.000 0.996
#> GSM97049 2 0.0000 0.9077 0.000 1.000 0.000
#> GSM96968 3 0.3941 0.7826 0.156 0.000 0.844
#> GSM96971 3 0.0237 0.9267 0.004 0.000 0.996
#> GSM96986 3 0.0237 0.9267 0.004 0.000 0.996
#> GSM97003 1 0.1289 0.9577 0.968 0.000 0.032
#> GSM96957 1 0.0747 0.9631 0.984 0.000 0.016
#> GSM96960 1 0.0747 0.9600 0.984 0.000 0.016
#> GSM96975 1 0.0747 0.9631 0.984 0.000 0.016
#> GSM96998 1 0.0000 0.9604 1.000 0.000 0.000
#> GSM96999 1 0.0747 0.9631 0.984 0.000 0.016
#> GSM97001 1 0.0747 0.9631 0.984 0.000 0.016
#> GSM97005 1 0.0747 0.9631 0.984 0.000 0.016
#> GSM97006 1 0.0747 0.9600 0.984 0.000 0.016
#> GSM97021 1 0.0747 0.9631 0.984 0.000 0.016
#> GSM97028 3 0.0237 0.9267 0.004 0.000 0.996
#> GSM97031 1 0.1289 0.9577 0.968 0.000 0.032
#> GSM97037 2 0.5882 0.4523 0.000 0.652 0.348
#> GSM97018 3 0.6704 0.3535 0.016 0.376 0.608
#> GSM97014 2 0.7074 0.0979 0.480 0.500 0.020
#> GSM97042 2 0.0000 0.9077 0.000 1.000 0.000
#> GSM97040 1 0.0747 0.9631 0.984 0.000 0.016
#> GSM97041 1 0.0747 0.9631 0.984 0.000 0.016
#> GSM96955 2 0.3267 0.8348 0.044 0.912 0.044
#> GSM96990 3 0.0661 0.9232 0.004 0.008 0.988
#> GSM96991 2 0.0000 0.9077 0.000 1.000 0.000
#> GSM97048 2 0.0000 0.9077 0.000 1.000 0.000
#> GSM96963 2 0.0000 0.9077 0.000 1.000 0.000
#> GSM96953 2 0.0000 0.9077 0.000 1.000 0.000
#> GSM96966 1 0.0892 0.9591 0.980 0.000 0.020
#> GSM96979 3 0.4291 0.7390 0.180 0.000 0.820
#> GSM96983 3 0.0237 0.9240 0.000 0.004 0.996
#> GSM96984 3 0.0237 0.9267 0.004 0.000 0.996
#> GSM96994 3 0.0237 0.9267 0.004 0.000 0.996
#> GSM96996 1 0.0000 0.9604 1.000 0.000 0.000
#> GSM96997 3 0.0237 0.9267 0.004 0.000 0.996
#> GSM97007 3 0.0237 0.9267 0.004 0.000 0.996
#> GSM96954 3 0.0592 0.9218 0.012 0.000 0.988
#> GSM96962 3 0.0237 0.9267 0.004 0.000 0.996
#> GSM96969 1 0.0892 0.9591 0.980 0.000 0.020
#> GSM96970 1 0.0892 0.9591 0.980 0.000 0.020
#> GSM96973 1 0.0892 0.9591 0.980 0.000 0.020
#> GSM96976 1 0.6309 0.0566 0.500 0.000 0.500
#> GSM96977 1 0.0747 0.9631 0.984 0.000 0.016
#> GSM96995 3 0.3941 0.7823 0.156 0.000 0.844
#> GSM97002 1 0.0592 0.9601 0.988 0.000 0.012
#> GSM97009 1 0.5519 0.8029 0.812 0.120 0.068
#> GSM97010 1 0.1163 0.9597 0.972 0.000 0.028
#> GSM96974 1 0.4291 0.8003 0.820 0.000 0.180
#> GSM96985 1 0.2625 0.9164 0.916 0.000 0.084
#> GSM96959 3 0.1015 0.9193 0.008 0.012 0.980
#> GSM96972 1 0.0892 0.9591 0.980 0.000 0.020
#> GSM96978 3 0.0237 0.9267 0.004 0.000 0.996
#> GSM96967 1 0.0892 0.9591 0.980 0.000 0.020
#> GSM96987 1 0.0000 0.9604 1.000 0.000 0.000
#> GSM97011 1 0.0747 0.9631 0.984 0.000 0.016
#> GSM96964 1 0.0000 0.9604 1.000 0.000 0.000
#> GSM96965 1 0.1411 0.9578 0.964 0.000 0.036
#> GSM96981 1 0.0237 0.9589 0.996 0.000 0.004
#> GSM96982 1 0.0892 0.9591 0.980 0.000 0.020
#> GSM96988 1 0.5968 0.4561 0.636 0.000 0.364
#> GSM97000 1 0.1289 0.9577 0.968 0.000 0.032
#> GSM97004 1 0.0892 0.9591 0.980 0.000 0.020
#> GSM97008 1 0.0747 0.9631 0.984 0.000 0.016
#> GSM96950 1 0.0747 0.9631 0.984 0.000 0.016
#> GSM96980 1 0.0892 0.9591 0.980 0.000 0.020
#> GSM96989 1 0.0000 0.9604 1.000 0.000 0.000
#> GSM96992 1 0.0000 0.9604 1.000 0.000 0.000
#> GSM96993 1 0.0747 0.9631 0.984 0.000 0.016
#> GSM96958 1 0.0592 0.9630 0.988 0.000 0.012
#> GSM96951 1 0.0592 0.9630 0.988 0.000 0.012
#> GSM96952 1 0.0000 0.9604 1.000 0.000 0.000
#> GSM96961 1 0.0000 0.9604 1.000 0.000 0.000
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM97038 2 0.0000 0.9745 0.000 1.000 0.000 0.000
#> GSM97045 2 0.0000 0.9745 0.000 1.000 0.000 0.000
#> GSM97047 2 0.3024 0.8129 0.148 0.852 0.000 0.000
#> GSM97025 2 0.0000 0.9745 0.000 1.000 0.000 0.000
#> GSM97030 3 0.0000 0.9510 0.000 0.000 1.000 0.000
#> GSM97027 2 0.0000 0.9745 0.000 1.000 0.000 0.000
#> GSM97033 2 0.0000 0.9745 0.000 1.000 0.000 0.000
#> GSM97034 3 0.0000 0.9510 0.000 0.000 1.000 0.000
#> GSM97020 2 0.0000 0.9745 0.000 1.000 0.000 0.000
#> GSM97026 2 0.1940 0.9022 0.076 0.924 0.000 0.000
#> GSM97012 2 0.0000 0.9745 0.000 1.000 0.000 0.000
#> GSM97015 3 0.0000 0.9510 0.000 0.000 1.000 0.000
#> GSM97016 2 0.0000 0.9745 0.000 1.000 0.000 0.000
#> GSM97017 1 0.0000 0.8901 1.000 0.000 0.000 0.000
#> GSM97019 2 0.0000 0.9745 0.000 1.000 0.000 0.000
#> GSM97022 2 0.0000 0.9745 0.000 1.000 0.000 0.000
#> GSM97035 2 0.0000 0.9745 0.000 1.000 0.000 0.000
#> GSM97036 1 0.0188 0.8896 0.996 0.000 0.000 0.004
#> GSM97039 2 0.0000 0.9745 0.000 1.000 0.000 0.000
#> GSM97046 2 0.0000 0.9745 0.000 1.000 0.000 0.000
#> GSM97023 1 0.3266 0.8334 0.832 0.000 0.000 0.168
#> GSM97029 1 0.0000 0.8901 1.000 0.000 0.000 0.000
#> GSM97043 2 0.0000 0.9745 0.000 1.000 0.000 0.000
#> GSM97013 1 0.0000 0.8901 1.000 0.000 0.000 0.000
#> GSM96956 2 0.2081 0.8939 0.000 0.916 0.084 0.000
#> GSM97024 2 0.0000 0.9745 0.000 1.000 0.000 0.000
#> GSM97032 3 0.4776 0.4007 0.000 0.376 0.624 0.000
#> GSM97044 3 0.0000 0.9510 0.000 0.000 1.000 0.000
#> GSM97049 2 0.0000 0.9745 0.000 1.000 0.000 0.000
#> GSM96968 3 0.0000 0.9510 0.000 0.000 1.000 0.000
#> GSM96971 3 0.0000 0.9510 0.000 0.000 1.000 0.000
#> GSM96986 3 0.0000 0.9510 0.000 0.000 1.000 0.000
#> GSM97003 1 0.4565 0.8075 0.796 0.000 0.064 0.140
#> GSM96957 1 0.0000 0.8901 1.000 0.000 0.000 0.000
#> GSM96960 1 0.4790 0.5621 0.620 0.000 0.000 0.380
#> GSM96975 1 0.0000 0.8901 1.000 0.000 0.000 0.000
#> GSM96998 1 0.3649 0.8107 0.796 0.000 0.000 0.204
#> GSM96999 1 0.0000 0.8901 1.000 0.000 0.000 0.000
#> GSM97001 1 0.0000 0.8901 1.000 0.000 0.000 0.000
#> GSM97005 1 0.0000 0.8901 1.000 0.000 0.000 0.000
#> GSM97006 1 0.3907 0.7850 0.768 0.000 0.000 0.232
#> GSM97021 1 0.0000 0.8901 1.000 0.000 0.000 0.000
#> GSM97028 3 0.0000 0.9510 0.000 0.000 1.000 0.000
#> GSM97031 1 0.1557 0.8772 0.944 0.000 0.000 0.056
#> GSM97037 2 0.3837 0.7037 0.000 0.776 0.224 0.000
#> GSM97018 3 0.4713 0.4396 0.000 0.360 0.640 0.000
#> GSM97014 1 0.4356 0.5030 0.708 0.292 0.000 0.000
#> GSM97042 2 0.0000 0.9745 0.000 1.000 0.000 0.000
#> GSM97040 1 0.0000 0.8901 1.000 0.000 0.000 0.000
#> GSM97041 1 0.0000 0.8901 1.000 0.000 0.000 0.000
#> GSM96955 2 0.0817 0.9544 0.024 0.976 0.000 0.000
#> GSM96990 3 0.0000 0.9510 0.000 0.000 1.000 0.000
#> GSM96991 2 0.0000 0.9745 0.000 1.000 0.000 0.000
#> GSM97048 2 0.0000 0.9745 0.000 1.000 0.000 0.000
#> GSM96963 2 0.0000 0.9745 0.000 1.000 0.000 0.000
#> GSM96953 2 0.0000 0.9745 0.000 1.000 0.000 0.000
#> GSM96966 4 0.0000 0.8728 0.000 0.000 0.000 1.000
#> GSM96979 3 0.0000 0.9510 0.000 0.000 1.000 0.000
#> GSM96983 3 0.0000 0.9510 0.000 0.000 1.000 0.000
#> GSM96984 3 0.0000 0.9510 0.000 0.000 1.000 0.000
#> GSM96994 3 0.0000 0.9510 0.000 0.000 1.000 0.000
#> GSM96996 1 0.3688 0.8076 0.792 0.000 0.000 0.208
#> GSM96997 3 0.0000 0.9510 0.000 0.000 1.000 0.000
#> GSM97007 3 0.0000 0.9510 0.000 0.000 1.000 0.000
#> GSM96954 3 0.0000 0.9510 0.000 0.000 1.000 0.000
#> GSM96962 3 0.0000 0.9510 0.000 0.000 1.000 0.000
#> GSM96969 4 0.0000 0.8728 0.000 0.000 0.000 1.000
#> GSM96970 4 0.0000 0.8728 0.000 0.000 0.000 1.000
#> GSM96973 4 0.0000 0.8728 0.000 0.000 0.000 1.000
#> GSM96976 4 0.4356 0.5563 0.000 0.000 0.292 0.708
#> GSM96977 1 0.0000 0.8901 1.000 0.000 0.000 0.000
#> GSM96995 3 0.0469 0.9390 0.012 0.000 0.988 0.000
#> GSM97002 1 0.4955 0.4120 0.556 0.000 0.000 0.444
#> GSM97009 1 0.0336 0.8867 0.992 0.008 0.000 0.000
#> GSM97010 1 0.0000 0.8901 1.000 0.000 0.000 0.000
#> GSM96974 4 0.3610 0.6936 0.000 0.000 0.200 0.800
#> GSM96985 4 0.2868 0.7698 0.000 0.000 0.136 0.864
#> GSM96959 3 0.2216 0.8450 0.092 0.000 0.908 0.000
#> GSM96972 4 0.0000 0.8728 0.000 0.000 0.000 1.000
#> GSM96978 3 0.0000 0.9510 0.000 0.000 1.000 0.000
#> GSM96967 4 0.0000 0.8728 0.000 0.000 0.000 1.000
#> GSM96987 1 0.3726 0.8045 0.788 0.000 0.000 0.212
#> GSM97011 1 0.0000 0.8901 1.000 0.000 0.000 0.000
#> GSM96964 1 0.2973 0.8450 0.856 0.000 0.000 0.144
#> GSM96965 4 0.4103 0.6608 0.256 0.000 0.000 0.744
#> GSM96981 1 0.3444 0.8249 0.816 0.000 0.000 0.184
#> GSM96982 4 0.4898 0.0166 0.416 0.000 0.000 0.584
#> GSM96988 3 0.0000 0.9510 0.000 0.000 1.000 0.000
#> GSM97000 1 0.0000 0.8901 1.000 0.000 0.000 0.000
#> GSM97004 4 0.0000 0.8728 0.000 0.000 0.000 1.000
#> GSM97008 1 0.0000 0.8901 1.000 0.000 0.000 0.000
#> GSM96950 1 0.0000 0.8901 1.000 0.000 0.000 0.000
#> GSM96980 4 0.0000 0.8728 0.000 0.000 0.000 1.000
#> GSM96989 1 0.3486 0.8218 0.812 0.000 0.000 0.188
#> GSM96992 1 0.3726 0.8045 0.788 0.000 0.000 0.212
#> GSM96993 1 0.0000 0.8901 1.000 0.000 0.000 0.000
#> GSM96958 1 0.0707 0.8867 0.980 0.000 0.000 0.020
#> GSM96951 1 0.2408 0.8613 0.896 0.000 0.000 0.104
#> GSM96952 1 0.3726 0.8045 0.788 0.000 0.000 0.212
#> GSM96961 1 0.3610 0.8138 0.800 0.000 0.000 0.200
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM97038 2 0.0510 0.939 0.000 0.984 0.000 0.016 0.000
#> GSM97045 2 0.0162 0.940 0.000 0.996 0.000 0.004 0.000
#> GSM97047 5 0.5552 0.624 0.176 0.052 0.016 0.040 0.716
#> GSM97025 2 0.0000 0.940 0.000 1.000 0.000 0.000 0.000
#> GSM97030 3 0.3885 0.840 0.176 0.000 0.784 0.040 0.000
#> GSM97027 2 0.0162 0.940 0.000 0.996 0.000 0.004 0.000
#> GSM97033 2 0.0609 0.939 0.000 0.980 0.000 0.020 0.000
#> GSM97034 3 0.3885 0.840 0.176 0.000 0.784 0.040 0.000
#> GSM97020 2 0.0609 0.939 0.000 0.980 0.000 0.020 0.000
#> GSM97026 5 0.4489 0.563 0.004 0.280 0.008 0.012 0.696
#> GSM97012 2 0.0898 0.936 0.008 0.972 0.000 0.020 0.000
#> GSM97015 3 0.3885 0.840 0.176 0.000 0.784 0.040 0.000
#> GSM97016 2 0.0000 0.940 0.000 1.000 0.000 0.000 0.000
#> GSM97017 5 0.0290 0.914 0.000 0.000 0.000 0.008 0.992
#> GSM97019 2 0.0609 0.938 0.000 0.980 0.000 0.020 0.000
#> GSM97022 2 0.0703 0.938 0.000 0.976 0.000 0.024 0.000
#> GSM97035 2 0.0404 0.939 0.000 0.988 0.000 0.012 0.000
#> GSM97036 5 0.1907 0.884 0.028 0.000 0.000 0.044 0.928
#> GSM97039 2 0.0609 0.939 0.000 0.980 0.000 0.020 0.000
#> GSM97046 2 0.0609 0.939 0.000 0.980 0.000 0.020 0.000
#> GSM97023 1 0.3210 0.922 0.788 0.000 0.000 0.000 0.212
#> GSM97029 5 0.0963 0.904 0.000 0.000 0.000 0.036 0.964
#> GSM97043 2 0.0000 0.940 0.000 1.000 0.000 0.000 0.000
#> GSM97013 5 0.0510 0.912 0.000 0.000 0.000 0.016 0.984
#> GSM96956 2 0.5126 0.684 0.176 0.724 0.076 0.024 0.000
#> GSM97024 2 0.3531 0.789 0.148 0.816 0.000 0.036 0.000
#> GSM97032 3 0.6479 0.670 0.176 0.176 0.608 0.040 0.000
#> GSM97044 3 0.3885 0.840 0.176 0.000 0.784 0.040 0.000
#> GSM97049 2 0.0609 0.939 0.000 0.980 0.000 0.020 0.000
#> GSM96968 3 0.2395 0.867 0.040 0.000 0.912 0.012 0.036
#> GSM96971 3 0.0000 0.886 0.000 0.000 1.000 0.000 0.000
#> GSM96986 3 0.0000 0.886 0.000 0.000 1.000 0.000 0.000
#> GSM97003 1 0.4492 0.888 0.744 0.000 0.056 0.004 0.196
#> GSM96957 5 0.0000 0.915 0.000 0.000 0.000 0.000 1.000
#> GSM96960 1 0.3921 0.905 0.784 0.000 0.000 0.044 0.172
#> GSM96975 5 0.4101 0.086 0.372 0.000 0.000 0.000 0.628
#> GSM96998 1 0.3596 0.924 0.784 0.000 0.000 0.016 0.200
#> GSM96999 5 0.0510 0.908 0.016 0.000 0.000 0.000 0.984
#> GSM97001 5 0.0000 0.915 0.000 0.000 0.000 0.000 1.000
#> GSM97005 5 0.1121 0.884 0.044 0.000 0.000 0.000 0.956
#> GSM97006 1 0.3710 0.920 0.784 0.000 0.000 0.024 0.192
#> GSM97021 5 0.0000 0.915 0.000 0.000 0.000 0.000 1.000
#> GSM97028 3 0.2280 0.867 0.120 0.000 0.880 0.000 0.000
#> GSM97031 1 0.3752 0.848 0.708 0.000 0.000 0.000 0.292
#> GSM97037 2 0.7060 0.233 0.176 0.516 0.264 0.044 0.000
#> GSM97018 3 0.6416 0.680 0.176 0.168 0.616 0.040 0.000
#> GSM97014 5 0.0880 0.892 0.000 0.032 0.000 0.000 0.968
#> GSM97042 2 0.1251 0.930 0.008 0.956 0.000 0.036 0.000
#> GSM97040 5 0.0000 0.915 0.000 0.000 0.000 0.000 1.000
#> GSM97041 5 0.0000 0.915 0.000 0.000 0.000 0.000 1.000
#> GSM96955 2 0.2474 0.887 0.000 0.908 0.012 0.040 0.040
#> GSM96990 3 0.3885 0.840 0.176 0.000 0.784 0.040 0.000
#> GSM96991 2 0.1444 0.927 0.012 0.948 0.000 0.040 0.000
#> GSM97048 2 0.0609 0.939 0.000 0.980 0.000 0.020 0.000
#> GSM96963 2 0.1106 0.934 0.012 0.964 0.000 0.024 0.000
#> GSM96953 2 0.0771 0.938 0.004 0.976 0.000 0.020 0.000
#> GSM96966 4 0.2674 0.896 0.140 0.000 0.000 0.856 0.004
#> GSM96979 3 0.0000 0.886 0.000 0.000 1.000 0.000 0.000
#> GSM96983 3 0.0000 0.886 0.000 0.000 1.000 0.000 0.000
#> GSM96984 3 0.0000 0.886 0.000 0.000 1.000 0.000 0.000
#> GSM96994 3 0.0000 0.886 0.000 0.000 1.000 0.000 0.000
#> GSM96996 1 0.3863 0.921 0.772 0.000 0.000 0.028 0.200
#> GSM96997 3 0.0000 0.886 0.000 0.000 1.000 0.000 0.000
#> GSM97007 3 0.0000 0.886 0.000 0.000 1.000 0.000 0.000
#> GSM96954 3 0.0000 0.886 0.000 0.000 1.000 0.000 0.000
#> GSM96962 3 0.0000 0.886 0.000 0.000 1.000 0.000 0.000
#> GSM96969 4 0.2516 0.899 0.140 0.000 0.000 0.860 0.000
#> GSM96970 4 0.2516 0.899 0.140 0.000 0.000 0.860 0.000
#> GSM96973 4 0.2516 0.899 0.140 0.000 0.000 0.860 0.000
#> GSM96976 4 0.3177 0.751 0.000 0.000 0.208 0.792 0.000
#> GSM96977 5 0.0404 0.911 0.012 0.000 0.000 0.000 0.988
#> GSM96995 3 0.4464 0.829 0.176 0.000 0.764 0.040 0.020
#> GSM97002 1 0.4417 0.866 0.760 0.000 0.000 0.092 0.148
#> GSM97009 5 0.3089 0.802 0.076 0.000 0.012 0.040 0.872
#> GSM97010 5 0.0290 0.913 0.008 0.000 0.000 0.000 0.992
#> GSM96974 4 0.2891 0.790 0.000 0.000 0.176 0.824 0.000
#> GSM96985 4 0.3151 0.824 0.020 0.000 0.144 0.836 0.000
#> GSM96959 3 0.5970 0.739 0.176 0.000 0.664 0.040 0.120
#> GSM96972 4 0.2516 0.899 0.140 0.000 0.000 0.860 0.000
#> GSM96978 3 0.0000 0.886 0.000 0.000 1.000 0.000 0.000
#> GSM96967 4 0.2516 0.899 0.140 0.000 0.000 0.860 0.000
#> GSM96987 1 0.3690 0.923 0.780 0.000 0.000 0.020 0.200
#> GSM97011 5 0.0000 0.915 0.000 0.000 0.000 0.000 1.000
#> GSM96964 1 0.3519 0.919 0.776 0.000 0.000 0.008 0.216
#> GSM96965 4 0.3194 0.754 0.020 0.000 0.000 0.832 0.148
#> GSM96981 1 0.3663 0.919 0.776 0.000 0.000 0.016 0.208
#> GSM96982 1 0.4504 0.743 0.748 0.000 0.000 0.168 0.084
#> GSM96988 3 0.0000 0.886 0.000 0.000 1.000 0.000 0.000
#> GSM97000 5 0.0000 0.915 0.000 0.000 0.000 0.000 1.000
#> GSM97004 1 0.3966 0.393 0.664 0.000 0.000 0.336 0.000
#> GSM97008 5 0.0000 0.915 0.000 0.000 0.000 0.000 1.000
#> GSM96950 5 0.2409 0.847 0.068 0.000 0.000 0.032 0.900
#> GSM96980 4 0.2561 0.897 0.144 0.000 0.000 0.856 0.000
#> GSM96989 1 0.3724 0.922 0.776 0.000 0.000 0.020 0.204
#> GSM96992 1 0.3496 0.924 0.788 0.000 0.000 0.012 0.200
#> GSM96993 5 0.1041 0.905 0.004 0.000 0.000 0.032 0.964
#> GSM96958 1 0.3837 0.821 0.692 0.000 0.000 0.000 0.308
#> GSM96951 1 0.3534 0.889 0.744 0.000 0.000 0.000 0.256
#> GSM96952 1 0.3496 0.924 0.788 0.000 0.000 0.012 0.200
#> GSM96961 1 0.3496 0.924 0.788 0.000 0.000 0.012 0.200
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM97038 2 0.1297 0.9314 0.012 0.948 0.040 0.000 0.000 0.000
#> GSM97045 2 0.1500 0.9270 0.012 0.936 0.052 0.000 0.000 0.000
#> GSM97047 5 0.5183 0.0746 0.000 0.024 0.456 0.000 0.480 0.040
#> GSM97025 2 0.0993 0.9326 0.012 0.964 0.024 0.000 0.000 0.000
#> GSM97030 3 0.3996 0.7096 0.000 0.004 0.512 0.000 0.000 0.484
#> GSM97027 2 0.1434 0.9279 0.012 0.940 0.048 0.000 0.000 0.000
#> GSM97033 2 0.1563 0.9262 0.012 0.932 0.056 0.000 0.000 0.000
#> GSM97034 3 0.3867 0.7048 0.000 0.000 0.512 0.000 0.000 0.488
#> GSM97020 2 0.1745 0.9218 0.012 0.920 0.068 0.000 0.000 0.000
#> GSM97026 5 0.3794 0.8164 0.012 0.092 0.036 0.012 0.828 0.020
#> GSM97012 2 0.2165 0.9117 0.000 0.884 0.108 0.008 0.000 0.000
#> GSM97015 3 0.3867 0.7048 0.000 0.000 0.512 0.000 0.000 0.488
#> GSM97016 2 0.0291 0.9332 0.004 0.992 0.004 0.000 0.000 0.000
#> GSM97017 5 0.1078 0.9174 0.008 0.000 0.012 0.016 0.964 0.000
#> GSM97019 2 0.1866 0.9209 0.000 0.908 0.084 0.008 0.000 0.000
#> GSM97022 2 0.1918 0.9194 0.000 0.904 0.088 0.008 0.000 0.000
#> GSM97035 2 0.1812 0.9216 0.000 0.912 0.080 0.008 0.000 0.000
#> GSM97036 5 0.3768 0.8408 0.044 0.000 0.076 0.064 0.816 0.000
#> GSM97039 2 0.0713 0.9336 0.000 0.972 0.028 0.000 0.000 0.000
#> GSM97046 2 0.1007 0.9330 0.000 0.956 0.044 0.000 0.000 0.000
#> GSM97023 1 0.1411 0.8170 0.936 0.000 0.000 0.004 0.060 0.000
#> GSM97029 5 0.2763 0.8831 0.040 0.000 0.028 0.052 0.880 0.000
#> GSM97043 2 0.1151 0.9318 0.012 0.956 0.032 0.000 0.000 0.000
#> GSM97013 5 0.1622 0.9104 0.016 0.000 0.016 0.028 0.940 0.000
#> GSM96956 3 0.5418 0.4366 0.000 0.352 0.520 0.000 0.000 0.128
#> GSM97024 3 0.4531 0.1170 0.000 0.464 0.504 0.000 0.000 0.032
#> GSM97032 3 0.4882 0.7099 0.000 0.060 0.512 0.000 0.000 0.428
#> GSM97044 3 0.3867 0.7048 0.000 0.000 0.512 0.000 0.000 0.488
#> GSM97049 2 0.1686 0.9236 0.012 0.924 0.064 0.000 0.000 0.000
#> GSM96968 6 0.1588 0.8401 0.000 0.000 0.072 0.000 0.004 0.924
#> GSM96971 6 0.0000 0.9622 0.000 0.000 0.000 0.000 0.000 1.000
#> GSM96986 6 0.0000 0.9622 0.000 0.000 0.000 0.000 0.000 1.000
#> GSM97003 1 0.2812 0.7774 0.868 0.000 0.004 0.004 0.040 0.084
#> GSM96957 5 0.0000 0.9221 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM96960 1 0.2000 0.8142 0.916 0.000 0.004 0.048 0.032 0.000
#> GSM96975 1 0.5990 0.4131 0.400 0.000 0.232 0.000 0.368 0.000
#> GSM96998 1 0.4442 0.7926 0.696 0.000 0.248 0.020 0.036 0.000
#> GSM96999 5 0.0405 0.9200 0.008 0.000 0.004 0.000 0.988 0.000
#> GSM97001 5 0.0000 0.9221 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM97005 5 0.2234 0.8278 0.124 0.000 0.004 0.000 0.872 0.000
#> GSM97006 1 0.1924 0.8133 0.920 0.000 0.004 0.048 0.028 0.000
#> GSM97021 5 0.0000 0.9221 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM97028 6 0.2762 0.5191 0.000 0.000 0.196 0.000 0.000 0.804
#> GSM97031 1 0.2562 0.7856 0.860 0.000 0.004 0.004 0.128 0.004
#> GSM97037 3 0.5708 0.5886 0.000 0.216 0.520 0.000 0.000 0.264
#> GSM97018 3 0.4837 0.7096 0.000 0.056 0.512 0.000 0.000 0.432
#> GSM97014 5 0.0363 0.9180 0.012 0.000 0.000 0.000 0.988 0.000
#> GSM97042 2 0.2212 0.9100 0.000 0.880 0.112 0.008 0.000 0.000
#> GSM97040 5 0.0000 0.9221 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM97041 5 0.0000 0.9221 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM96955 2 0.3650 0.8225 0.004 0.820 0.108 0.000 0.040 0.028
#> GSM96990 3 0.3996 0.7096 0.000 0.004 0.512 0.000 0.000 0.484
#> GSM96991 2 0.2257 0.9082 0.000 0.876 0.116 0.008 0.000 0.000
#> GSM97048 2 0.1563 0.9262 0.012 0.932 0.056 0.000 0.000 0.000
#> GSM96963 2 0.2257 0.9082 0.000 0.876 0.116 0.008 0.000 0.000
#> GSM96953 2 0.1866 0.9206 0.000 0.908 0.084 0.008 0.000 0.000
#> GSM96966 4 0.1765 0.8932 0.096 0.000 0.000 0.904 0.000 0.000
#> GSM96979 6 0.0000 0.9622 0.000 0.000 0.000 0.000 0.000 1.000
#> GSM96983 6 0.0000 0.9622 0.000 0.000 0.000 0.000 0.000 1.000
#> GSM96984 6 0.0000 0.9622 0.000 0.000 0.000 0.000 0.000 1.000
#> GSM96994 6 0.0000 0.9622 0.000 0.000 0.000 0.000 0.000 1.000
#> GSM96996 1 0.4815 0.7853 0.676 0.000 0.244 0.052 0.028 0.000
#> GSM96997 6 0.0000 0.9622 0.000 0.000 0.000 0.000 0.000 1.000
#> GSM97007 6 0.0000 0.9622 0.000 0.000 0.000 0.000 0.000 1.000
#> GSM96954 6 0.0000 0.9622 0.000 0.000 0.000 0.000 0.000 1.000
#> GSM96962 6 0.0000 0.9622 0.000 0.000 0.000 0.000 0.000 1.000
#> GSM96969 4 0.1610 0.8998 0.084 0.000 0.000 0.916 0.000 0.000
#> GSM96970 4 0.1610 0.8998 0.084 0.000 0.000 0.916 0.000 0.000
#> GSM96973 4 0.1610 0.8998 0.084 0.000 0.000 0.916 0.000 0.000
#> GSM96976 4 0.3345 0.7620 0.000 0.000 0.028 0.788 0.000 0.184
#> GSM96977 5 0.0551 0.9209 0.008 0.000 0.004 0.000 0.984 0.004
#> GSM96995 3 0.4260 0.7068 0.000 0.000 0.512 0.000 0.016 0.472
#> GSM97002 1 0.4868 0.7627 0.672 0.000 0.216 0.104 0.008 0.000
#> GSM97009 5 0.2145 0.8604 0.000 0.000 0.072 0.000 0.900 0.028
#> GSM97010 5 0.1426 0.9131 0.028 0.000 0.008 0.016 0.948 0.000
#> GSM96974 4 0.3206 0.7990 0.004 0.000 0.028 0.816 0.000 0.152
#> GSM96985 4 0.3161 0.8136 0.008 0.000 0.028 0.828 0.000 0.136
#> GSM96959 3 0.4967 0.6705 0.000 0.000 0.512 0.000 0.068 0.420
#> GSM96972 4 0.1610 0.8998 0.084 0.000 0.000 0.916 0.000 0.000
#> GSM96978 6 0.0000 0.9622 0.000 0.000 0.000 0.000 0.000 1.000
#> GSM96967 4 0.1610 0.8998 0.084 0.000 0.000 0.916 0.000 0.000
#> GSM96987 1 0.4531 0.7872 0.680 0.000 0.264 0.020 0.036 0.000
#> GSM97011 5 0.0000 0.9221 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM96964 1 0.2750 0.8291 0.868 0.000 0.080 0.004 0.048 0.000
#> GSM96965 4 0.3932 0.7498 0.032 0.000 0.080 0.800 0.088 0.000
#> GSM96981 1 0.4492 0.7898 0.684 0.000 0.260 0.016 0.040 0.000
#> GSM96982 1 0.5277 0.7383 0.636 0.000 0.228 0.120 0.016 0.000
#> GSM96988 6 0.0260 0.9513 0.000 0.000 0.000 0.008 0.000 0.992
#> GSM97000 5 0.0622 0.9167 0.008 0.000 0.000 0.000 0.980 0.012
#> GSM97004 1 0.5438 0.5820 0.560 0.000 0.160 0.280 0.000 0.000
#> GSM97008 5 0.0000 0.9221 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM96950 5 0.3842 0.8206 0.096 0.000 0.052 0.044 0.808 0.000
#> GSM96980 4 0.1806 0.8961 0.088 0.000 0.004 0.908 0.000 0.000
#> GSM96989 1 0.4531 0.7872 0.680 0.000 0.264 0.020 0.036 0.000
#> GSM96992 1 0.1572 0.8179 0.936 0.000 0.000 0.028 0.036 0.000
#> GSM96993 5 0.2252 0.8970 0.028 0.000 0.020 0.044 0.908 0.000
#> GSM96958 1 0.2178 0.7882 0.868 0.000 0.000 0.000 0.132 0.000
#> GSM96951 1 0.2196 0.7995 0.884 0.000 0.004 0.004 0.108 0.000
#> GSM96952 1 0.2883 0.8271 0.868 0.000 0.076 0.020 0.036 0.000
#> GSM96961 1 0.1616 0.8169 0.932 0.000 0.000 0.020 0.048 0.000
Heatmaps for the consensus matrix. It visualizes the probability of two samples to be in a same group.
consensus_heatmap(res, k = 2)
consensus_heatmap(res, k = 3)
consensus_heatmap(res, k = 4)
consensus_heatmap(res, k = 5)
consensus_heatmap(res, k = 6)
Heatmaps for the membership of samples in all partitions to see how consistent they are:
membership_heatmap(res, k = 2)
membership_heatmap(res, k = 3)
membership_heatmap(res, k = 4)
membership_heatmap(res, k = 5)
membership_heatmap(res, k = 6)
As soon as we have had the classes for columns, we can look for signatures which are significantly different between classes which can be candidate marks for certain classes. Following are the heatmaps for signatures.
Signature heatmaps where rows are scaled:
get_signatures(res, k = 2)
get_signatures(res, k = 3)
get_signatures(res, k = 4)
get_signatures(res, k = 5)
get_signatures(res, k = 6)
Signature heatmaps where rows are not scaled:
get_signatures(res, k = 2, scale_rows = FALSE)
get_signatures(res, k = 3, scale_rows = FALSE)
get_signatures(res, k = 4, scale_rows = FALSE)
get_signatures(res, k = 5, scale_rows = FALSE)
get_signatures(res, k = 6, scale_rows = FALSE)
Compare the overlap of signatures from different k:
compare_signatures(res)
get_signature() returns a data frame invisibly. TO get the list of signatures, the function
call should be assigned to a variable explicitly. In following code, if plot argument is set
to FALSE, no heatmap is plotted while only the differential analysis is performed.
# code only for demonstration
tb = get_signature(res, k = ..., plot = FALSE)
An example of the output of tb is:
#> which_row fdr mean_1 mean_2 scaled_mean_1 scaled_mean_2 km
#> 1 38 0.042760348 8.373488 9.131774 -0.5533452 0.5164555 1
#> 2 40 0.018707592 7.106213 8.469186 -0.6173731 0.5762149 1
#> 3 55 0.019134737 10.221463 11.207825 -0.6159697 0.5749050 1
#> 4 59 0.006059896 5.921854 7.869574 -0.6899429 0.6439467 1
#> 5 60 0.018055526 8.928898 10.211722 -0.6204761 0.5791110 1
#> 6 98 0.009384629 15.714769 14.887706 0.6635654 -0.6193277 2
...
The columns in tb are:
which_row: row indices corresponding to the input matrix.fdr: FDR for the differential test. mean_x: The mean value in group x.scaled_mean_x: The mean value in group x after rows are scaled.km: Row groups if k-means clustering is applied to rows.UMAP plot which shows how samples are separated.
dimension_reduction(res, k = 2, method = "UMAP")
dimension_reduction(res, k = 3, method = "UMAP")
dimension_reduction(res, k = 4, method = "UMAP")
dimension_reduction(res, k = 5, method = "UMAP")
dimension_reduction(res, k = 6, method = "UMAP")
Following heatmap shows how subgroups are split when increasing k:
collect_classes(res)
Test correlation between subgroups and known annotations. If the known annotation is numeric, one-way ANOVA test is applied, and if the known annotation is discrete, chi-squared contingency table test is applied.
test_to_known_factors(res)
#> n disease.state(p) specimen(p) cell.type(p) other(p) k
#> MAD:mclust 100 6.54e-06 0.4378 1.48e-10 0.00517 2
#> MAD:mclust 92 8.69e-05 0.4307 9.63e-18 0.03403 3
#> MAD:mclust 96 1.04e-05 0.0566 5.29e-19 0.03038 4
#> MAD:mclust 97 1.30e-04 0.2812 4.26e-16 0.24324 5
#> MAD:mclust 96 8.80e-06 0.1944 1.18e-18 0.04965 6
If matrix rows can be associated to genes, consider to use functional_enrichment(res,
...) to perform function enrichment for the signature genes. See this vignette for more detailed explanations.
The object with results only for a single top-value method and a single partition method can be extracted as:
res = res_list["MAD", "NMF"]
# you can also extract it by
# res = res_list["MAD:NMF"]
A summary of res and all the functions that can be applied to it:
res
#> A 'ConsensusPartition' object with k = 2, 3, 4, 5, 6.
#> On a matrix with 21168 rows and 100 columns.
#> Top rows (1000, 2000, 3000, 4000, 5000) are extracted by 'MAD' method.
#> Subgroups are detected by 'NMF' method.
#> Performed in total 1250 partitions by row resampling.
#> Best k for subgroups seems to be 2.
#>
#> Following methods can be applied to this 'ConsensusPartition' object:
#> [1] "cola_report" "collect_classes" "collect_plots"
#> [4] "collect_stats" "colnames" "compare_signatures"
#> [7] "consensus_heatmap" "dimension_reduction" "functional_enrichment"
#> [10] "get_anno_col" "get_anno" "get_classes"
#> [13] "get_consensus" "get_matrix" "get_membership"
#> [16] "get_param" "get_signatures" "get_stats"
#> [19] "is_best_k" "is_stable_k" "membership_heatmap"
#> [22] "ncol" "nrow" "plot_ecdf"
#> [25] "rownames" "select_partition_number" "show"
#> [28] "suggest_best_k" "test_to_known_factors"
collect_plots() function collects all the plots made from res for all k (number of partitions)
into one single page to provide an easy and fast comparison between different k.
collect_plots(res)
The plots are:
k and the heatmap of
predicted classes for each k.k.k.k.All the plots in panels can be made by individual functions and they are plotted later in this section.
select_partition_number() produces several plots showing different
statistics for choosing “optimized” k. There are following statistics:
k;k, the area increased is defined as \(A_k - A_{k-1}\).The detailed explanations of these statistics can be found in the cola vignette.
Generally speaking, lower PAC score, higher mean silhouette score or higher
concordance corresponds to better partition. Rand index and Jaccard index
measure how similar the current partition is compared to partition with k-1.
If they are too similar, we won't accept k is better than k-1.
select_partition_number(res)
The numeric values for all these statistics can be obtained by get_stats().
get_stats(res)
#> k 1-PAC mean_silhouette concordance area_increased Rand Jaccard
#> 2 2 1.000 0.966 0.986 0.4937 0.508 0.508
#> 3 3 0.889 0.900 0.952 0.3218 0.791 0.609
#> 4 4 0.626 0.628 0.753 0.1367 0.844 0.595
#> 5 5 0.600 0.549 0.737 0.0727 0.844 0.495
#> 6 6 0.622 0.460 0.709 0.0441 0.852 0.417
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
#> GSM97038 2 0.0000 0.98907 0.000 1.000
#> GSM97045 2 0.0000 0.98907 0.000 1.000
#> GSM97047 2 0.0000 0.98907 0.000 1.000
#> GSM97025 2 0.0000 0.98907 0.000 1.000
#> GSM97030 2 0.0000 0.98907 0.000 1.000
#> GSM97027 2 0.0000 0.98907 0.000 1.000
#> GSM97033 2 0.0000 0.98907 0.000 1.000
#> GSM97034 2 0.0000 0.98907 0.000 1.000
#> GSM97020 2 0.0000 0.98907 0.000 1.000
#> GSM97026 2 0.0000 0.98907 0.000 1.000
#> GSM97012 2 0.0000 0.98907 0.000 1.000
#> GSM97015 2 0.0000 0.98907 0.000 1.000
#> GSM97016 2 0.0000 0.98907 0.000 1.000
#> GSM97017 1 0.0000 0.98304 1.000 0.000
#> GSM97019 2 0.0000 0.98907 0.000 1.000
#> GSM97022 2 0.0000 0.98907 0.000 1.000
#> GSM97035 2 0.0000 0.98907 0.000 1.000
#> GSM97036 1 0.0938 0.97438 0.988 0.012
#> GSM97039 2 0.0000 0.98907 0.000 1.000
#> GSM97046 2 0.0000 0.98907 0.000 1.000
#> GSM97023 1 0.0000 0.98304 1.000 0.000
#> GSM97029 1 0.2043 0.95731 0.968 0.032
#> GSM97043 2 0.0000 0.98907 0.000 1.000
#> GSM97013 1 0.0000 0.98304 1.000 0.000
#> GSM96956 2 0.0000 0.98907 0.000 1.000
#> GSM97024 2 0.0000 0.98907 0.000 1.000
#> GSM97032 2 0.0000 0.98907 0.000 1.000
#> GSM97044 2 0.0000 0.98907 0.000 1.000
#> GSM97049 2 0.0000 0.98907 0.000 1.000
#> GSM96968 1 0.4562 0.89062 0.904 0.096
#> GSM96971 1 0.0000 0.98304 1.000 0.000
#> GSM96986 1 0.0000 0.98304 1.000 0.000
#> GSM97003 1 0.0000 0.98304 1.000 0.000
#> GSM96957 1 0.0376 0.98041 0.996 0.004
#> GSM96960 1 0.0000 0.98304 1.000 0.000
#> GSM96975 1 0.0000 0.98304 1.000 0.000
#> GSM96998 1 0.0000 0.98304 1.000 0.000
#> GSM96999 1 0.0000 0.98304 1.000 0.000
#> GSM97001 1 0.0000 0.98304 1.000 0.000
#> GSM97005 1 0.0000 0.98304 1.000 0.000
#> GSM97006 1 0.0000 0.98304 1.000 0.000
#> GSM97021 1 0.0000 0.98304 1.000 0.000
#> GSM97028 1 1.0000 0.00385 0.504 0.496
#> GSM97031 1 0.0000 0.98304 1.000 0.000
#> GSM97037 2 0.0000 0.98907 0.000 1.000
#> GSM97018 2 0.0000 0.98907 0.000 1.000
#> GSM97014 2 0.0000 0.98907 0.000 1.000
#> GSM97042 2 0.0000 0.98907 0.000 1.000
#> GSM97040 2 0.1184 0.97420 0.016 0.984
#> GSM97041 1 0.3584 0.92260 0.932 0.068
#> GSM96955 2 0.0000 0.98907 0.000 1.000
#> GSM96990 2 0.0000 0.98907 0.000 1.000
#> GSM96991 2 0.0000 0.98907 0.000 1.000
#> GSM97048 2 0.0000 0.98907 0.000 1.000
#> GSM96963 2 0.0000 0.98907 0.000 1.000
#> GSM96953 2 0.0000 0.98907 0.000 1.000
#> GSM96966 1 0.0000 0.98304 1.000 0.000
#> GSM96979 1 0.0000 0.98304 1.000 0.000
#> GSM96983 2 0.0000 0.98907 0.000 1.000
#> GSM96984 1 0.6343 0.80969 0.840 0.160
#> GSM96994 2 0.0000 0.98907 0.000 1.000
#> GSM96996 1 0.0000 0.98304 1.000 0.000
#> GSM96997 1 0.0000 0.98304 1.000 0.000
#> GSM97007 2 0.0000 0.98907 0.000 1.000
#> GSM96954 1 0.0000 0.98304 1.000 0.000
#> GSM96962 1 0.0000 0.98304 1.000 0.000
#> GSM96969 1 0.0000 0.98304 1.000 0.000
#> GSM96970 1 0.0000 0.98304 1.000 0.000
#> GSM96973 1 0.0000 0.98304 1.000 0.000
#> GSM96976 2 0.6048 0.82213 0.148 0.852
#> GSM96977 1 0.0000 0.98304 1.000 0.000
#> GSM96995 2 0.8443 0.62121 0.272 0.728
#> GSM97002 1 0.0000 0.98304 1.000 0.000
#> GSM97009 2 0.0000 0.98907 0.000 1.000
#> GSM97010 1 0.1184 0.97140 0.984 0.016
#> GSM96974 1 0.0000 0.98304 1.000 0.000
#> GSM96985 1 0.0000 0.98304 1.000 0.000
#> GSM96959 2 0.0000 0.98907 0.000 1.000
#> GSM96972 1 0.0000 0.98304 1.000 0.000
#> GSM96978 1 0.2423 0.95021 0.960 0.040
#> GSM96967 1 0.0000 0.98304 1.000 0.000
#> GSM96987 1 0.0000 0.98304 1.000 0.000
#> GSM97011 1 0.1184 0.97131 0.984 0.016
#> GSM96964 1 0.0000 0.98304 1.000 0.000
#> GSM96965 1 0.0376 0.98041 0.996 0.004
#> GSM96981 1 0.0000 0.98304 1.000 0.000
#> GSM96982 1 0.0000 0.98304 1.000 0.000
#> GSM96988 1 0.0000 0.98304 1.000 0.000
#> GSM97000 1 0.0000 0.98304 1.000 0.000
#> GSM97004 1 0.0000 0.98304 1.000 0.000
#> GSM97008 1 0.0000 0.98304 1.000 0.000
#> GSM96950 1 0.0000 0.98304 1.000 0.000
#> GSM96980 1 0.0000 0.98304 1.000 0.000
#> GSM96989 1 0.0000 0.98304 1.000 0.000
#> GSM96992 1 0.0000 0.98304 1.000 0.000
#> GSM96993 1 0.0672 0.97757 0.992 0.008
#> GSM96958 1 0.0000 0.98304 1.000 0.000
#> GSM96951 1 0.0000 0.98304 1.000 0.000
#> GSM96952 1 0.0000 0.98304 1.000 0.000
#> GSM96961 1 0.0000 0.98304 1.000 0.000
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM97038 2 0.0237 0.9311 0.000 0.996 0.004
#> GSM97045 2 0.0424 0.9278 0.008 0.992 0.000
#> GSM97047 2 0.0424 0.9278 0.008 0.992 0.000
#> GSM97025 2 0.0000 0.9305 0.000 1.000 0.000
#> GSM97030 3 0.3551 0.8338 0.000 0.132 0.868
#> GSM97027 2 0.0424 0.9278 0.008 0.992 0.000
#> GSM97033 2 0.0000 0.9305 0.000 1.000 0.000
#> GSM97034 3 0.5098 0.6748 0.000 0.248 0.752
#> GSM97020 2 0.0424 0.9278 0.008 0.992 0.000
#> GSM97026 2 0.0424 0.9278 0.008 0.992 0.000
#> GSM97012 2 0.0424 0.9305 0.000 0.992 0.008
#> GSM97015 3 0.4555 0.7498 0.000 0.200 0.800
#> GSM97016 2 0.0237 0.9311 0.000 0.996 0.004
#> GSM97017 1 0.2356 0.9213 0.928 0.072 0.000
#> GSM97019 2 0.0424 0.9305 0.000 0.992 0.008
#> GSM97022 2 0.0592 0.9286 0.000 0.988 0.012
#> GSM97035 2 0.0424 0.9305 0.000 0.992 0.008
#> GSM97036 1 0.3412 0.8713 0.876 0.124 0.000
#> GSM97039 2 0.0237 0.9311 0.000 0.996 0.004
#> GSM97046 2 0.0237 0.9311 0.000 0.996 0.004
#> GSM97023 1 0.0000 0.9652 1.000 0.000 0.000
#> GSM97029 1 0.3340 0.8752 0.880 0.120 0.000
#> GSM97043 2 0.0237 0.9311 0.000 0.996 0.004
#> GSM97013 1 0.1411 0.9503 0.964 0.036 0.000
#> GSM96956 3 0.6192 0.2667 0.000 0.420 0.580
#> GSM97024 2 0.0892 0.9230 0.000 0.980 0.020
#> GSM97032 2 0.5591 0.5617 0.000 0.696 0.304
#> GSM97044 3 0.2356 0.8921 0.000 0.072 0.928
#> GSM97049 2 0.0424 0.9278 0.008 0.992 0.000
#> GSM96968 3 0.0237 0.9378 0.004 0.000 0.996
#> GSM96971 3 0.0237 0.9378 0.004 0.000 0.996
#> GSM96986 3 0.0237 0.9378 0.004 0.000 0.996
#> GSM97003 1 0.5254 0.6913 0.736 0.000 0.264
#> GSM96957 1 0.0747 0.9607 0.984 0.016 0.000
#> GSM96960 1 0.2625 0.9251 0.916 0.000 0.084
#> GSM96975 1 0.0424 0.9655 0.992 0.000 0.008
#> GSM96998 1 0.0000 0.9652 1.000 0.000 0.000
#> GSM96999 1 0.0237 0.9643 0.996 0.004 0.000
#> GSM97001 1 0.1411 0.9501 0.964 0.036 0.000
#> GSM97005 1 0.0237 0.9643 0.996 0.004 0.000
#> GSM97006 1 0.1411 0.9570 0.964 0.000 0.036
#> GSM97021 1 0.0592 0.9621 0.988 0.012 0.000
#> GSM97028 3 0.0424 0.9372 0.000 0.008 0.992
#> GSM97031 1 0.2878 0.9141 0.904 0.000 0.096
#> GSM97037 2 0.4452 0.7421 0.000 0.808 0.192
#> GSM97018 2 0.6260 0.1874 0.000 0.552 0.448
#> GSM97014 2 0.1529 0.8998 0.040 0.960 0.000
#> GSM97042 2 0.0424 0.9305 0.000 0.992 0.008
#> GSM97040 2 0.2448 0.8619 0.076 0.924 0.000
#> GSM97041 1 0.3482 0.8672 0.872 0.128 0.000
#> GSM96955 2 0.0237 0.9311 0.000 0.996 0.004
#> GSM96990 2 0.6299 0.0767 0.000 0.524 0.476
#> GSM96991 2 0.0424 0.9305 0.000 0.992 0.008
#> GSM97048 2 0.0000 0.9305 0.000 1.000 0.000
#> GSM96963 2 0.0424 0.9305 0.000 0.992 0.008
#> GSM96953 2 0.0424 0.9305 0.000 0.992 0.008
#> GSM96966 1 0.1031 0.9623 0.976 0.000 0.024
#> GSM96979 3 0.0592 0.9324 0.012 0.000 0.988
#> GSM96983 3 0.0424 0.9372 0.000 0.008 0.992
#> GSM96984 3 0.0237 0.9378 0.000 0.004 0.996
#> GSM96994 3 0.0424 0.9372 0.000 0.008 0.992
#> GSM96996 1 0.0424 0.9655 0.992 0.000 0.008
#> GSM96997 3 0.0237 0.9378 0.004 0.000 0.996
#> GSM97007 3 0.0424 0.9372 0.000 0.008 0.992
#> GSM96954 3 0.0592 0.9324 0.012 0.000 0.988
#> GSM96962 3 0.0237 0.9378 0.004 0.000 0.996
#> GSM96969 1 0.2165 0.9403 0.936 0.000 0.064
#> GSM96970 1 0.1163 0.9607 0.972 0.000 0.028
#> GSM96973 1 0.2261 0.9380 0.932 0.000 0.068
#> GSM96976 3 0.0424 0.9372 0.000 0.008 0.992
#> GSM96977 1 0.0892 0.9638 0.980 0.000 0.020
#> GSM96995 3 0.1860 0.9099 0.000 0.052 0.948
#> GSM97002 1 0.0592 0.9652 0.988 0.000 0.012
#> GSM97009 2 0.0424 0.9278 0.008 0.992 0.000
#> GSM97010 1 0.0848 0.9659 0.984 0.008 0.008
#> GSM96974 3 0.0237 0.9378 0.004 0.000 0.996
#> GSM96985 3 0.0424 0.9351 0.008 0.000 0.992
#> GSM96959 2 0.5138 0.6556 0.000 0.748 0.252
#> GSM96972 1 0.2165 0.9403 0.936 0.000 0.064
#> GSM96978 3 0.0237 0.9378 0.000 0.004 0.996
#> GSM96967 1 0.2165 0.9405 0.936 0.000 0.064
#> GSM96987 1 0.0237 0.9643 0.996 0.004 0.000
#> GSM97011 1 0.1031 0.9575 0.976 0.024 0.000
#> GSM96964 1 0.0000 0.9652 1.000 0.000 0.000
#> GSM96965 1 0.0848 0.9658 0.984 0.008 0.008
#> GSM96981 1 0.0237 0.9654 0.996 0.000 0.004
#> GSM96982 1 0.0892 0.9636 0.980 0.000 0.020
#> GSM96988 3 0.0237 0.9378 0.004 0.000 0.996
#> GSM97000 1 0.1529 0.9550 0.960 0.000 0.040
#> GSM97004 1 0.0747 0.9645 0.984 0.000 0.016
#> GSM97008 1 0.0237 0.9643 0.996 0.004 0.000
#> GSM96950 1 0.0237 0.9643 0.996 0.004 0.000
#> GSM96980 1 0.0892 0.9636 0.980 0.000 0.020
#> GSM96989 1 0.0000 0.9652 1.000 0.000 0.000
#> GSM96992 1 0.0592 0.9652 0.988 0.000 0.012
#> GSM96993 1 0.1031 0.9573 0.976 0.024 0.000
#> GSM96958 1 0.0424 0.9655 0.992 0.000 0.008
#> GSM96951 1 0.0592 0.9652 0.988 0.000 0.012
#> GSM96952 1 0.0424 0.9655 0.992 0.000 0.008
#> GSM96961 1 0.0424 0.9655 0.992 0.000 0.008
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM97038 2 0.0817 0.8288 0.000 0.976 0.000 0.024
#> GSM97045 2 0.0927 0.8179 0.016 0.976 0.000 0.008
#> GSM97047 2 0.3725 0.6522 0.180 0.812 0.000 0.008
#> GSM97025 2 0.1211 0.8311 0.000 0.960 0.000 0.040
#> GSM97030 3 0.0657 0.8799 0.000 0.012 0.984 0.004
#> GSM97027 2 0.0657 0.8168 0.012 0.984 0.000 0.004
#> GSM97033 2 0.0000 0.8230 0.000 1.000 0.000 0.000
#> GSM97034 3 0.4389 0.7713 0.000 0.072 0.812 0.116
#> GSM97020 2 0.0927 0.8179 0.016 0.976 0.000 0.008
#> GSM97026 2 0.2521 0.8284 0.024 0.912 0.000 0.064
#> GSM97012 2 0.4585 0.7560 0.000 0.668 0.000 0.332
#> GSM97015 3 0.1082 0.8762 0.004 0.020 0.972 0.004
#> GSM97016 2 0.1637 0.8321 0.000 0.940 0.000 0.060
#> GSM97017 1 0.3626 0.5948 0.812 0.184 0.000 0.004
#> GSM97019 2 0.4072 0.7944 0.000 0.748 0.000 0.252
#> GSM97022 2 0.4164 0.7896 0.000 0.736 0.000 0.264
#> GSM97035 2 0.4356 0.7767 0.000 0.708 0.000 0.292
#> GSM97036 1 0.5440 0.4151 0.736 0.104 0.000 0.160
#> GSM97039 2 0.1022 0.8298 0.000 0.968 0.000 0.032
#> GSM97046 2 0.2469 0.8281 0.000 0.892 0.000 0.108
#> GSM97023 1 0.0469 0.6330 0.988 0.000 0.000 0.012
#> GSM97029 1 0.4152 0.6019 0.808 0.160 0.000 0.032
#> GSM97043 2 0.1940 0.8307 0.000 0.924 0.000 0.076
#> GSM97013 1 0.3610 0.5893 0.800 0.200 0.000 0.000
#> GSM96956 2 0.7650 0.2442 0.000 0.424 0.364 0.212
#> GSM97024 2 0.2197 0.8309 0.000 0.916 0.004 0.080
#> GSM97032 3 0.6800 -0.0942 0.000 0.444 0.460 0.096
#> GSM97044 3 0.0336 0.8803 0.000 0.008 0.992 0.000
#> GSM97049 2 0.0804 0.8197 0.012 0.980 0.000 0.008
#> GSM96968 3 0.0336 0.8801 0.008 0.000 0.992 0.000
#> GSM96971 3 0.3975 0.6874 0.000 0.000 0.760 0.240
#> GSM96986 3 0.0188 0.8808 0.004 0.000 0.996 0.000
#> GSM97003 3 0.6626 0.1898 0.364 0.000 0.544 0.092
#> GSM96957 1 0.4792 0.5155 0.680 0.312 0.000 0.008
#> GSM96960 1 0.5372 -0.4225 0.544 0.000 0.012 0.444
#> GSM96975 1 0.4072 0.3191 0.748 0.000 0.000 0.252
#> GSM96998 1 0.3444 0.4755 0.816 0.000 0.000 0.184
#> GSM96999 1 0.0524 0.6355 0.988 0.008 0.000 0.004
#> GSM97001 1 0.4697 0.5280 0.696 0.296 0.000 0.008
#> GSM97005 1 0.1004 0.6352 0.972 0.024 0.000 0.004
#> GSM97006 1 0.3975 0.3545 0.760 0.000 0.000 0.240
#> GSM97021 1 0.4673 0.5307 0.700 0.292 0.000 0.008
#> GSM97028 3 0.0469 0.8802 0.000 0.000 0.988 0.012
#> GSM97031 1 0.5296 -0.0184 0.500 0.000 0.492 0.008
#> GSM97037 2 0.6359 0.6265 0.000 0.648 0.220 0.132
#> GSM97018 2 0.7085 0.6210 0.000 0.544 0.156 0.300
#> GSM97014 2 0.4328 0.5484 0.244 0.748 0.000 0.008
#> GSM97042 2 0.4585 0.7562 0.000 0.668 0.000 0.332
#> GSM97040 1 0.5203 0.3387 0.576 0.416 0.000 0.008
#> GSM97041 1 0.4973 0.4788 0.644 0.348 0.000 0.008
#> GSM96955 2 0.4250 0.7753 0.000 0.724 0.000 0.276
#> GSM96990 3 0.4235 0.7800 0.000 0.092 0.824 0.084
#> GSM96991 2 0.4817 0.7163 0.000 0.612 0.000 0.388
#> GSM97048 2 0.0188 0.8219 0.004 0.996 0.000 0.000
#> GSM96963 2 0.4761 0.7289 0.000 0.628 0.000 0.372
#> GSM96953 2 0.4500 0.7648 0.000 0.684 0.000 0.316
#> GSM96966 4 0.4973 0.7584 0.348 0.000 0.008 0.644
#> GSM96979 3 0.0707 0.8744 0.000 0.000 0.980 0.020
#> GSM96983 3 0.1302 0.8700 0.000 0.000 0.956 0.044
#> GSM96984 3 0.0000 0.8810 0.000 0.000 1.000 0.000
#> GSM96994 3 0.0188 0.8808 0.000 0.000 0.996 0.004
#> GSM96996 1 0.4877 -0.2518 0.592 0.000 0.000 0.408
#> GSM96997 3 0.0000 0.8810 0.000 0.000 1.000 0.000
#> GSM97007 3 0.0000 0.8810 0.000 0.000 1.000 0.000
#> GSM96954 3 0.0469 0.8793 0.012 0.000 0.988 0.000
#> GSM96962 3 0.0000 0.8810 0.000 0.000 1.000 0.000
#> GSM96969 4 0.5040 0.7518 0.364 0.000 0.008 0.628
#> GSM96970 4 0.4917 0.7602 0.336 0.000 0.008 0.656
#> GSM96973 4 0.4792 0.7539 0.312 0.000 0.008 0.680
#> GSM96976 4 0.1938 0.4626 0.000 0.052 0.012 0.936
#> GSM96977 1 0.0844 0.6357 0.980 0.004 0.004 0.012
#> GSM96995 3 0.0712 0.8799 0.004 0.008 0.984 0.004
#> GSM97002 4 0.4985 0.5967 0.468 0.000 0.000 0.532
#> GSM97009 2 0.2714 0.7366 0.112 0.884 0.000 0.004
#> GSM97010 1 0.5669 -0.4881 0.516 0.016 0.004 0.464
#> GSM96974 4 0.1362 0.5492 0.012 0.004 0.020 0.964
#> GSM96985 4 0.1674 0.5670 0.032 0.004 0.012 0.952
#> GSM96959 3 0.4867 0.7261 0.048 0.164 0.780 0.008
#> GSM96972 4 0.5085 0.7429 0.376 0.000 0.008 0.616
#> GSM96978 3 0.4679 0.6091 0.000 0.000 0.648 0.352
#> GSM96967 4 0.4877 0.7594 0.328 0.000 0.008 0.664
#> GSM96987 1 0.3074 0.5334 0.848 0.000 0.000 0.152
#> GSM97011 1 0.4228 0.5694 0.760 0.232 0.000 0.008
#> GSM96964 1 0.1211 0.6259 0.960 0.000 0.000 0.040
#> GSM96965 4 0.4049 0.6881 0.212 0.008 0.000 0.780
#> GSM96981 1 0.4925 -0.3312 0.572 0.000 0.000 0.428
#> GSM96982 4 0.4955 0.6498 0.444 0.000 0.000 0.556
#> GSM96988 3 0.4252 0.7152 0.004 0.000 0.744 0.252
#> GSM97000 1 0.5595 0.2009 0.576 0.012 0.404 0.008
#> GSM97004 4 0.4972 0.6256 0.456 0.000 0.000 0.544
#> GSM97008 1 0.4975 0.5690 0.752 0.208 0.032 0.008
#> GSM96950 1 0.1792 0.6103 0.932 0.000 0.000 0.068
#> GSM96980 4 0.4991 0.7315 0.388 0.000 0.004 0.608
#> GSM96989 1 0.3123 0.5265 0.844 0.000 0.000 0.156
#> GSM96992 1 0.2921 0.5382 0.860 0.000 0.000 0.140
#> GSM96993 1 0.1913 0.6365 0.940 0.040 0.000 0.020
#> GSM96958 1 0.1022 0.6270 0.968 0.000 0.000 0.032
#> GSM96951 1 0.1151 0.6312 0.968 0.000 0.008 0.024
#> GSM96952 1 0.2814 0.5480 0.868 0.000 0.000 0.132
#> GSM96961 1 0.1211 0.6232 0.960 0.000 0.000 0.040
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM97038 5 0.3355 0.6301 0.000 0.184 0.000 0.012 0.804
#> GSM97045 5 0.4855 0.2396 0.024 0.424 0.000 0.000 0.552
#> GSM97047 5 0.3651 0.5597 0.108 0.060 0.000 0.004 0.828
#> GSM97025 2 0.4371 0.4549 0.012 0.644 0.000 0.000 0.344
#> GSM97030 3 0.1768 0.7338 0.000 0.072 0.924 0.000 0.004
#> GSM97027 5 0.4978 0.0592 0.028 0.476 0.000 0.000 0.496
#> GSM97033 5 0.3395 0.6135 0.000 0.236 0.000 0.000 0.764
#> GSM97034 2 0.5186 0.3316 0.040 0.612 0.340 0.000 0.008
#> GSM97020 5 0.3700 0.6082 0.008 0.240 0.000 0.000 0.752
#> GSM97026 2 0.5278 0.5345 0.156 0.692 0.000 0.004 0.148
#> GSM97012 2 0.3596 0.6287 0.000 0.776 0.000 0.012 0.212
#> GSM97015 3 0.3381 0.6676 0.016 0.160 0.820 0.000 0.004
#> GSM97016 5 0.4060 0.4278 0.000 0.360 0.000 0.000 0.640
#> GSM97017 1 0.4197 0.6946 0.760 0.024 0.000 0.012 0.204
#> GSM97019 2 0.3300 0.6529 0.000 0.792 0.000 0.004 0.204
#> GSM97022 2 0.3607 0.6246 0.000 0.752 0.000 0.004 0.244
#> GSM97035 2 0.4130 0.5474 0.000 0.696 0.000 0.012 0.292
#> GSM97036 1 0.5505 0.4007 0.604 0.304 0.000 0.092 0.000
#> GSM97039 5 0.3452 0.6065 0.000 0.244 0.000 0.000 0.756
#> GSM97046 5 0.3700 0.6068 0.000 0.240 0.000 0.008 0.752
#> GSM97023 1 0.2409 0.7146 0.900 0.000 0.000 0.068 0.032
#> GSM97029 1 0.4106 0.6587 0.800 0.140 0.000 0.020 0.040
#> GSM97043 2 0.4031 0.6323 0.048 0.788 0.004 0.000 0.160
#> GSM97013 1 0.4001 0.6909 0.820 0.048 0.000 0.028 0.104
#> GSM96956 3 0.7158 -0.0909 0.000 0.236 0.404 0.020 0.340
#> GSM97024 2 0.4088 0.5913 0.004 0.712 0.008 0.000 0.276
#> GSM97032 2 0.5061 0.4300 0.020 0.644 0.312 0.000 0.024
#> GSM97044 3 0.1792 0.7314 0.000 0.084 0.916 0.000 0.000
#> GSM97049 5 0.3231 0.6318 0.004 0.196 0.000 0.000 0.800
#> GSM96968 3 0.0798 0.7499 0.008 0.016 0.976 0.000 0.000
#> GSM96971 4 0.4559 -0.0735 0.000 0.008 0.480 0.512 0.000
#> GSM96986 3 0.2424 0.7292 0.008 0.000 0.908 0.032 0.052
#> GSM97003 3 0.7455 0.1687 0.124 0.000 0.484 0.292 0.100
#> GSM96957 1 0.4517 0.4556 0.556 0.000 0.008 0.000 0.436
#> GSM96960 4 0.6029 0.4894 0.324 0.004 0.060 0.584 0.028
#> GSM96975 4 0.6614 0.1771 0.340 0.008 0.000 0.476 0.176
#> GSM96998 1 0.3920 0.4889 0.724 0.004 0.000 0.268 0.004
#> GSM96999 1 0.4998 0.6463 0.716 0.000 0.004 0.172 0.108
#> GSM97001 5 0.4655 -0.1082 0.384 0.000 0.004 0.012 0.600
#> GSM97005 1 0.5839 0.6327 0.648 0.000 0.028 0.092 0.232
#> GSM97006 1 0.5842 0.1668 0.536 0.000 0.032 0.392 0.040
#> GSM97021 1 0.3918 0.6794 0.752 0.008 0.008 0.000 0.232
#> GSM97028 3 0.4723 0.2645 0.016 0.448 0.536 0.000 0.000
#> GSM97031 3 0.6659 0.4052 0.208 0.000 0.596 0.056 0.140
#> GSM97037 2 0.6036 0.3767 0.004 0.540 0.340 0.000 0.116
#> GSM97018 2 0.3352 0.5531 0.004 0.800 0.192 0.000 0.004
#> GSM97014 5 0.2875 0.5953 0.056 0.052 0.000 0.008 0.884
#> GSM97042 2 0.3621 0.6458 0.000 0.788 0.000 0.020 0.192
#> GSM97040 1 0.5103 0.5325 0.616 0.024 0.016 0.000 0.344
#> GSM97041 1 0.4490 0.6565 0.724 0.052 0.000 0.000 0.224
#> GSM96955 5 0.5309 0.4992 0.012 0.236 0.000 0.076 0.676
#> GSM96990 3 0.4779 0.1349 0.012 0.448 0.536 0.000 0.004
#> GSM96991 2 0.2719 0.6217 0.000 0.884 0.000 0.048 0.068
#> GSM97048 5 0.3143 0.6302 0.000 0.204 0.000 0.000 0.796
#> GSM96963 2 0.4219 0.5700 0.000 0.772 0.000 0.072 0.156
#> GSM96953 5 0.5083 0.2481 0.000 0.432 0.000 0.036 0.532
#> GSM96966 4 0.1831 0.7405 0.076 0.004 0.000 0.920 0.000
#> GSM96979 3 0.2011 0.7311 0.004 0.000 0.908 0.088 0.000
#> GSM96983 3 0.3123 0.6968 0.000 0.160 0.828 0.012 0.000
#> GSM96984 3 0.0880 0.7491 0.000 0.000 0.968 0.032 0.000
#> GSM96994 3 0.0955 0.7500 0.000 0.000 0.968 0.028 0.004
#> GSM96996 4 0.4796 0.2014 0.468 0.004 0.000 0.516 0.012
#> GSM96997 3 0.1579 0.7444 0.000 0.000 0.944 0.024 0.032
#> GSM97007 3 0.0807 0.7505 0.000 0.012 0.976 0.012 0.000
#> GSM96954 3 0.1267 0.7504 0.024 0.012 0.960 0.000 0.004
#> GSM96962 3 0.0510 0.7495 0.000 0.000 0.984 0.016 0.000
#> GSM96969 4 0.1792 0.7400 0.084 0.000 0.000 0.916 0.000
#> GSM96970 4 0.1430 0.7398 0.052 0.000 0.000 0.944 0.004
#> GSM96973 4 0.0955 0.7354 0.028 0.000 0.004 0.968 0.000
#> GSM96976 4 0.3047 0.6747 0.000 0.096 0.012 0.868 0.024
#> GSM96977 1 0.5895 0.6695 0.704 0.016 0.040 0.104 0.136
#> GSM96995 3 0.1653 0.7499 0.028 0.024 0.944 0.000 0.004
#> GSM97002 4 0.4201 0.5558 0.328 0.008 0.000 0.664 0.000
#> GSM97009 5 0.2625 0.5799 0.056 0.028 0.000 0.016 0.900
#> GSM97010 4 0.5304 0.6501 0.108 0.000 0.020 0.712 0.160
#> GSM96974 4 0.2930 0.6661 0.004 0.164 0.000 0.832 0.000
#> GSM96985 4 0.4758 0.6253 0.040 0.248 0.004 0.704 0.004
#> GSM96959 5 0.5342 0.2995 0.072 0.000 0.268 0.008 0.652
#> GSM96972 4 0.2604 0.7337 0.108 0.004 0.004 0.880 0.004
#> GSM96978 3 0.6328 0.4353 0.000 0.228 0.528 0.244 0.000
#> GSM96967 4 0.1942 0.7412 0.068 0.012 0.000 0.920 0.000
#> GSM96987 1 0.3321 0.6528 0.832 0.032 0.000 0.136 0.000
#> GSM97011 5 0.4409 0.3712 0.220 0.004 0.000 0.040 0.736
#> GSM96964 1 0.2610 0.6944 0.892 0.028 0.000 0.076 0.004
#> GSM96965 4 0.2006 0.7245 0.024 0.020 0.000 0.932 0.024
#> GSM96981 4 0.6216 0.4284 0.280 0.012 0.000 0.572 0.136
#> GSM96982 4 0.5291 0.6471 0.220 0.040 0.000 0.696 0.044
#> GSM96988 3 0.5807 0.2496 0.020 0.444 0.488 0.048 0.000
#> GSM97000 3 0.7214 0.1740 0.252 0.000 0.464 0.032 0.252
#> GSM97004 4 0.4299 0.4654 0.388 0.004 0.000 0.608 0.000
#> GSM97008 1 0.6460 0.3810 0.472 0.000 0.092 0.028 0.408
#> GSM96950 1 0.3171 0.6893 0.864 0.044 0.000 0.084 0.008
#> GSM96980 4 0.2971 0.7173 0.156 0.008 0.000 0.836 0.000
#> GSM96989 1 0.3165 0.6681 0.848 0.036 0.000 0.116 0.000
#> GSM96992 1 0.4599 0.6224 0.752 0.004 0.004 0.176 0.064
#> GSM96993 1 0.3497 0.6732 0.840 0.108 0.000 0.044 0.008
#> GSM96958 1 0.4877 0.6703 0.732 0.000 0.004 0.136 0.128
#> GSM96951 1 0.4267 0.7035 0.800 0.000 0.028 0.052 0.120
#> GSM96952 1 0.3567 0.6589 0.820 0.004 0.000 0.144 0.032
#> GSM96961 1 0.2069 0.7051 0.912 0.000 0.000 0.076 0.012
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM97038 3 0.2152 0.6871 0.004 0.024 0.904 0.000 0.068 0.000
#> GSM97045 2 0.6247 0.2835 0.020 0.416 0.384 0.000 0.180 0.000
#> GSM97047 5 0.4800 0.2808 0.012 0.040 0.316 0.004 0.628 0.000
#> GSM97025 2 0.5404 0.5193 0.024 0.576 0.324 0.000 0.076 0.000
#> GSM97030 6 0.2009 0.7430 0.000 0.084 0.004 0.000 0.008 0.904
#> GSM97027 2 0.6237 0.2711 0.028 0.412 0.408 0.000 0.152 0.000
#> GSM97033 3 0.3784 0.5368 0.000 0.144 0.776 0.000 0.080 0.000
#> GSM97034 2 0.5347 0.4741 0.060 0.648 0.008 0.004 0.028 0.252
#> GSM97020 3 0.2432 0.6514 0.008 0.080 0.888 0.000 0.024 0.000
#> GSM97026 2 0.5898 0.4497 0.328 0.536 0.092 0.000 0.044 0.000
#> GSM97012 2 0.3338 0.6327 0.004 0.800 0.176 0.012 0.008 0.000
#> GSM97015 6 0.4021 0.6967 0.048 0.120 0.008 0.000 0.028 0.796
#> GSM97016 3 0.1411 0.6869 0.000 0.060 0.936 0.000 0.004 0.000
#> GSM97017 5 0.4736 0.2597 0.432 0.008 0.032 0.000 0.528 0.000
#> GSM97019 2 0.3088 0.6433 0.000 0.808 0.172 0.000 0.020 0.000
#> GSM97022 2 0.3791 0.6210 0.000 0.732 0.236 0.000 0.032 0.000
#> GSM97035 2 0.3985 0.5818 0.004 0.688 0.292 0.004 0.012 0.000
#> GSM97036 1 0.3635 0.5304 0.788 0.176 0.008 0.016 0.012 0.000
#> GSM97039 3 0.1080 0.7048 0.004 0.032 0.960 0.000 0.004 0.000
#> GSM97046 3 0.0692 0.7099 0.000 0.020 0.976 0.000 0.004 0.000
#> GSM97023 1 0.3130 0.6033 0.828 0.000 0.000 0.048 0.124 0.000
#> GSM97029 1 0.4642 0.4106 0.688 0.240 0.020 0.000 0.052 0.000
#> GSM97043 2 0.4901 0.6314 0.072 0.728 0.156 0.000 0.024 0.020
#> GSM97013 1 0.3951 0.5637 0.792 0.012 0.140 0.040 0.016 0.000
#> GSM96956 3 0.4779 0.4178 0.000 0.052 0.684 0.004 0.020 0.240
#> GSM97024 2 0.4976 0.6134 0.004 0.676 0.228 0.000 0.072 0.020
#> GSM97032 2 0.5306 0.2594 0.028 0.556 0.020 0.000 0.020 0.376
#> GSM97044 6 0.2451 0.7354 0.004 0.108 0.000 0.004 0.008 0.876
#> GSM97049 3 0.0260 0.7122 0.000 0.000 0.992 0.000 0.008 0.000
#> GSM96968 6 0.2432 0.7547 0.008 0.036 0.020 0.004 0.024 0.908
#> GSM96971 4 0.4586 0.3263 0.000 0.012 0.000 0.640 0.036 0.312
#> GSM96986 6 0.3067 0.7303 0.000 0.012 0.004 0.064 0.060 0.860
#> GSM97003 6 0.6655 -0.0743 0.012 0.012 0.004 0.376 0.212 0.384
#> GSM96957 3 0.5863 -0.1351 0.164 0.008 0.492 0.000 0.336 0.000
#> GSM96960 4 0.7324 0.0465 0.308 0.016 0.004 0.396 0.220 0.056
#> GSM96975 5 0.5082 0.4196 0.116 0.000 0.012 0.216 0.656 0.000
#> GSM96998 1 0.3455 0.5615 0.776 0.004 0.000 0.200 0.020 0.000
#> GSM96999 1 0.6197 0.2943 0.496 0.008 0.016 0.160 0.320 0.000
#> GSM97001 5 0.4899 0.5499 0.100 0.000 0.228 0.008 0.664 0.000
#> GSM97005 5 0.4409 0.5315 0.192 0.000 0.000 0.064 0.728 0.016
#> GSM97006 4 0.6897 -0.0403 0.396 0.012 0.004 0.408 0.112 0.068
#> GSM97021 5 0.4534 0.4494 0.296 0.016 0.024 0.000 0.660 0.004
#> GSM97028 2 0.5448 -0.1413 0.020 0.464 0.004 0.000 0.056 0.456
#> GSM97031 5 0.5892 0.1541 0.028 0.008 0.004 0.064 0.464 0.432
#> GSM97037 6 0.6264 0.1585 0.004 0.172 0.384 0.000 0.016 0.424
#> GSM97018 2 0.3904 0.5168 0.012 0.768 0.008 0.004 0.016 0.192
#> GSM97014 5 0.4235 0.1117 0.004 0.004 0.448 0.004 0.540 0.000
#> GSM97042 2 0.2837 0.6404 0.004 0.840 0.144 0.008 0.004 0.000
#> GSM97040 5 0.4440 0.5356 0.212 0.016 0.056 0.000 0.716 0.000
#> GSM97041 1 0.5561 -0.1800 0.484 0.020 0.080 0.000 0.416 0.000
#> GSM96955 5 0.7410 0.1915 0.012 0.192 0.192 0.148 0.456 0.000
#> GSM96990 6 0.5049 0.5751 0.020 0.208 0.072 0.000 0.012 0.688
#> GSM96991 2 0.3074 0.5991 0.008 0.872 0.056 0.036 0.024 0.004
#> GSM97048 3 0.0260 0.7122 0.000 0.000 0.992 0.000 0.008 0.000
#> GSM96963 2 0.4277 0.5584 0.012 0.788 0.092 0.076 0.032 0.000
#> GSM96953 2 0.5492 0.3799 0.008 0.520 0.400 0.032 0.040 0.000
#> GSM96966 4 0.2519 0.7025 0.072 0.020 0.000 0.888 0.020 0.000
#> GSM96979 6 0.4127 0.5619 0.000 0.008 0.004 0.252 0.024 0.712
#> GSM96983 6 0.4811 0.5645 0.004 0.248 0.004 0.004 0.068 0.672
#> GSM96984 6 0.1965 0.7575 0.000 0.008 0.004 0.040 0.024 0.924
#> GSM96994 6 0.1780 0.7600 0.000 0.004 0.004 0.024 0.036 0.932
#> GSM96996 1 0.5639 0.0601 0.492 0.012 0.004 0.416 0.068 0.008
#> GSM96997 6 0.2826 0.7379 0.000 0.012 0.004 0.056 0.052 0.876
#> GSM97007 6 0.0964 0.7629 0.000 0.000 0.004 0.012 0.016 0.968
#> GSM96954 6 0.1478 0.7614 0.000 0.020 0.000 0.004 0.032 0.944
#> GSM96962 6 0.1495 0.7615 0.000 0.008 0.004 0.020 0.020 0.948
#> GSM96969 4 0.1951 0.7036 0.060 0.004 0.000 0.916 0.020 0.000
#> GSM96970 4 0.1616 0.7049 0.028 0.012 0.000 0.940 0.020 0.000
#> GSM96973 4 0.0653 0.7033 0.012 0.004 0.000 0.980 0.004 0.000
#> GSM96976 4 0.2737 0.6717 0.008 0.072 0.012 0.884 0.020 0.004
#> GSM96977 5 0.6044 0.3461 0.324 0.012 0.004 0.060 0.552 0.048
#> GSM96995 6 0.3638 0.6971 0.008 0.036 0.004 0.000 0.156 0.796
#> GSM97002 4 0.5339 0.1807 0.396 0.012 0.004 0.524 0.064 0.000
#> GSM97009 3 0.5439 0.0901 0.000 0.020 0.496 0.036 0.432 0.016
#> GSM97010 3 0.5791 -0.0852 0.032 0.012 0.456 0.456 0.008 0.036
#> GSM96974 4 0.3675 0.6215 0.012 0.192 0.000 0.776 0.012 0.008
#> GSM96985 4 0.7293 0.2551 0.084 0.380 0.004 0.384 0.128 0.020
#> GSM96959 5 0.5567 0.4416 0.000 0.000 0.272 0.016 0.584 0.128
#> GSM96972 4 0.2854 0.6715 0.108 0.004 0.000 0.860 0.012 0.016
#> GSM96978 6 0.7249 0.2888 0.008 0.292 0.004 0.148 0.104 0.444
#> GSM96967 4 0.2222 0.6943 0.084 0.012 0.000 0.896 0.008 0.000
#> GSM96987 1 0.2113 0.6461 0.908 0.004 0.000 0.060 0.028 0.000
#> GSM97011 5 0.3889 0.5414 0.016 0.004 0.180 0.028 0.772 0.000
#> GSM96964 1 0.1464 0.6380 0.944 0.004 0.000 0.016 0.036 0.000
#> GSM96965 4 0.1635 0.7013 0.016 0.016 0.012 0.944 0.012 0.000
#> GSM96981 5 0.5016 0.2949 0.096 0.000 0.000 0.312 0.592 0.000
#> GSM96982 4 0.5776 0.0775 0.112 0.016 0.000 0.448 0.424 0.000
#> GSM96988 2 0.6288 -0.1003 0.036 0.456 0.004 0.012 0.084 0.408
#> GSM97000 5 0.4383 0.5068 0.016 0.000 0.000 0.036 0.696 0.252
#> GSM97004 1 0.5244 0.0318 0.496 0.008 0.000 0.424 0.072 0.000
#> GSM97008 5 0.3149 0.5749 0.080 0.000 0.028 0.016 0.860 0.016
#> GSM96950 1 0.1511 0.6405 0.944 0.012 0.000 0.032 0.012 0.000
#> GSM96980 4 0.4256 0.6207 0.176 0.020 0.000 0.748 0.056 0.000
#> GSM96989 1 0.1410 0.6467 0.944 0.004 0.000 0.044 0.008 0.000
#> GSM96992 1 0.5339 0.0952 0.464 0.008 0.000 0.080 0.448 0.000
#> GSM96993 1 0.1524 0.6123 0.932 0.060 0.000 0.000 0.008 0.000
#> GSM96958 5 0.4945 -0.0299 0.452 0.000 0.000 0.064 0.484 0.000
#> GSM96951 5 0.4512 0.1515 0.436 0.008 0.000 0.004 0.540 0.012
#> GSM96952 1 0.5218 0.2812 0.540 0.008 0.000 0.076 0.376 0.000
#> GSM96961 1 0.3698 0.5397 0.756 0.004 0.000 0.028 0.212 0.000
Heatmaps for the consensus matrix. It visualizes the probability of two samples to be in a same group.
consensus_heatmap(res, k = 2)
consensus_heatmap(res, k = 3)
consensus_heatmap(res, k = 4)
consensus_heatmap(res, k = 5)
consensus_heatmap(res, k = 6)
Heatmaps for the membership of samples in all partitions to see how consistent they are:
membership_heatmap(res, k = 2)
membership_heatmap(res, k = 3)
membership_heatmap(res, k = 4)
membership_heatmap(res, k = 5)
membership_heatmap(res, k = 6)
As soon as we have had the classes for columns, we can look for signatures which are significantly different between classes which can be candidate marks for certain classes. Following are the heatmaps for signatures.
Signature heatmaps where rows are scaled:
get_signatures(res, k = 2)
get_signatures(res, k = 3)
get_signatures(res, k = 4)
get_signatures(res, k = 5)
get_signatures(res, k = 6)
Signature heatmaps where rows are not scaled:
get_signatures(res, k = 2, scale_rows = FALSE)
get_signatures(res, k = 3, scale_rows = FALSE)
get_signatures(res, k = 4, scale_rows = FALSE)
get_signatures(res, k = 5, scale_rows = FALSE)
get_signatures(res, k = 6, scale_rows = FALSE)
Compare the overlap of signatures from different k:
compare_signatures(res)
get_signature() returns a data frame invisibly. TO get the list of signatures, the function
call should be assigned to a variable explicitly. In following code, if plot argument is set
to FALSE, no heatmap is plotted while only the differential analysis is performed.
# code only for demonstration
tb = get_signature(res, k = ..., plot = FALSE)
An example of the output of tb is:
#> which_row fdr mean_1 mean_2 scaled_mean_1 scaled_mean_2 km
#> 1 38 0.042760348 8.373488 9.131774 -0.5533452 0.5164555 1
#> 2 40 0.018707592 7.106213 8.469186 -0.6173731 0.5762149 1
#> 3 55 0.019134737 10.221463 11.207825 -0.6159697 0.5749050 1
#> 4 59 0.006059896 5.921854 7.869574 -0.6899429 0.6439467 1
#> 5 60 0.018055526 8.928898 10.211722 -0.6204761 0.5791110 1
#> 6 98 0.009384629 15.714769 14.887706 0.6635654 -0.6193277 2
...
The columns in tb are:
which_row: row indices corresponding to the input matrix.fdr: FDR for the differential test. mean_x: The mean value in group x.scaled_mean_x: The mean value in group x after rows are scaled.km: Row groups if k-means clustering is applied to rows.UMAP plot which shows how samples are separated.
dimension_reduction(res, k = 2, method = "UMAP")
dimension_reduction(res, k = 3, method = "UMAP")
dimension_reduction(res, k = 4, method = "UMAP")
dimension_reduction(res, k = 5, method = "UMAP")
dimension_reduction(res, k = 6, method = "UMAP")
Following heatmap shows how subgroups are split when increasing k:
collect_classes(res)
Test correlation between subgroups and known annotations. If the known annotation is numeric, one-way ANOVA test is applied, and if the known annotation is discrete, chi-squared contingency table test is applied.
test_to_known_factors(res)
#> n disease.state(p) specimen(p) cell.type(p) other(p) k
#> MAD:NMF 99 1.61e-05 0.1658 3.04e-14 0.1381 2
#> MAD:NMF 97 7.29e-05 0.2888 5.42e-18 0.0784 3
#> MAD:NMF 84 1.19e-05 0.0649 5.06e-18 0.0322 4
#> MAD:NMF 69 2.76e-03 0.3427 4.61e-15 0.0176 5
#> MAD:NMF 60 1.92e-03 0.6206 9.68e-14 0.0968 6
If matrix rows can be associated to genes, consider to use functional_enrichment(res,
...) to perform function enrichment for the signature genes. See this vignette for more detailed explanations.
The object with results only for a single top-value method and a single partition method can be extracted as:
res = res_list["ATC", "hclust"]
# you can also extract it by
# res = res_list["ATC:hclust"]
A summary of res and all the functions that can be applied to it:
res
#> A 'ConsensusPartition' object with k = 2, 3, 4, 5, 6.
#> On a matrix with 21168 rows and 100 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 3.
#>
#> Following methods can be applied to this 'ConsensusPartition' object:
#> [1] "cola_report" "collect_classes" "collect_plots"
#> [4] "collect_stats" "colnames" "compare_signatures"
#> [7] "consensus_heatmap" "dimension_reduction" "functional_enrichment"
#> [10] "get_anno_col" "get_anno" "get_classes"
#> [13] "get_consensus" "get_matrix" "get_membership"
#> [16] "get_param" "get_signatures" "get_stats"
#> [19] "is_best_k" "is_stable_k" "membership_heatmap"
#> [22] "ncol" "nrow" "plot_ecdf"
#> [25] "rownames" "select_partition_number" "show"
#> [28] "suggest_best_k" "test_to_known_factors"
collect_plots() function collects all the plots made from res for all k (number of partitions)
into one single page to provide an easy and fast comparison between different k.
collect_plots(res)
The plots are:
k and the heatmap of
predicted classes for each k.k.k.k.All the plots in panels can be made by individual functions and they are plotted later in this section.
select_partition_number() produces several plots showing different
statistics for choosing “optimized” k. There are following statistics:
k;k, the area increased is defined as \(A_k - A_{k-1}\).The detailed explanations of these statistics can be found in the cola vignette.
Generally speaking, lower PAC score, higher mean silhouette score or higher
concordance corresponds to better partition. Rand index and Jaccard index
measure how similar the current partition is compared to partition with k-1.
If they are too similar, we won't accept k is better than k-1.
select_partition_number(res)
The numeric values for all these statistics can be obtained by get_stats().
get_stats(res)
#> k 1-PAC mean_silhouette concordance area_increased Rand Jaccard
#> 2 2 0.529 0.885 0.923 0.2501 0.818 0.818
#> 3 3 0.538 0.735 0.872 1.2495 0.596 0.506
#> 4 4 0.557 0.701 0.839 0.1975 0.903 0.765
#> 5 5 0.545 0.539 0.743 0.0892 0.939 0.808
#> 6 6 0.589 0.608 0.725 0.0537 0.834 0.478
suggest_best_k() suggests the best \(k\) based on these statistics. The rules are as follows:
suggest_best_k(res)
#> [1] 3
Following shows the table of the partitions (You need to click the show/hide
code output link to see it). The membership matrix (columns with name p*)
is inferred by
clue::cl_consensus()
function with the SE method. Basically the value in the membership matrix
represents the probability to belong to a certain group. The finall class
label for an item is determined with the group with highest probability it
belongs to.
In get_classes() function, the entropy is calculated from the membership
matrix and the silhouette score is calculated from the consensus matrix.
cbind(get_classes(res, k = 2), get_membership(res, k = 2))
#> class entropy silhouette p1 p2
#> GSM97038 2 0.0000 0.910 0.000 1.000
#> GSM97045 2 0.0000 0.910 0.000 1.000
#> GSM97047 2 0.0000 0.910 0.000 1.000
#> GSM97025 2 0.0000 0.910 0.000 1.000
#> GSM97030 2 0.0000 0.910 0.000 1.000
#> GSM97027 2 0.0000 0.910 0.000 1.000
#> GSM97033 2 0.0000 0.910 0.000 1.000
#> GSM97034 2 0.0000 0.910 0.000 1.000
#> GSM97020 2 0.0000 0.910 0.000 1.000
#> GSM97026 2 0.0000 0.910 0.000 1.000
#> GSM97012 2 0.0000 0.910 0.000 1.000
#> GSM97015 2 0.0000 0.910 0.000 1.000
#> GSM97016 2 0.0000 0.910 0.000 1.000
#> GSM97017 2 0.0000 0.910 0.000 1.000
#> GSM97019 2 0.0000 0.910 0.000 1.000
#> GSM97022 2 0.0000 0.910 0.000 1.000
#> GSM97035 2 0.0000 0.910 0.000 1.000
#> GSM97036 2 0.0000 0.910 0.000 1.000
#> GSM97039 2 0.0000 0.910 0.000 1.000
#> GSM97046 2 0.0000 0.910 0.000 1.000
#> GSM97023 2 0.5842 0.872 0.140 0.860
#> GSM97029 2 0.0000 0.910 0.000 1.000
#> GSM97043 2 0.0000 0.910 0.000 1.000
#> GSM97013 2 0.6973 0.847 0.188 0.812
#> GSM96956 2 0.0000 0.910 0.000 1.000
#> GSM97024 2 0.0000 0.910 0.000 1.000
#> GSM97032 2 0.0000 0.910 0.000 1.000
#> GSM97044 2 0.0000 0.910 0.000 1.000
#> GSM97049 2 0.0000 0.910 0.000 1.000
#> GSM96968 2 0.7602 0.823 0.220 0.780
#> GSM96971 1 0.7950 0.600 0.760 0.240
#> GSM96986 2 0.8327 0.776 0.264 0.736
#> GSM97003 2 0.7674 0.820 0.224 0.776
#> GSM96957 2 0.3584 0.896 0.068 0.932
#> GSM96960 2 0.7745 0.816 0.228 0.772
#> GSM96975 2 0.6801 0.854 0.180 0.820
#> GSM96998 2 0.6343 0.863 0.160 0.840
#> GSM96999 2 0.6801 0.854 0.180 0.820
#> GSM97001 2 0.3584 0.896 0.068 0.932
#> GSM97005 2 0.6973 0.847 0.188 0.812
#> GSM97006 2 0.7815 0.812 0.232 0.768
#> GSM97021 2 0.0000 0.910 0.000 1.000
#> GSM97028 2 0.0000 0.910 0.000 1.000
#> GSM97031 2 0.6973 0.847 0.188 0.812
#> GSM97037 2 0.0000 0.910 0.000 1.000
#> GSM97018 2 0.0000 0.910 0.000 1.000
#> GSM97014 2 0.0000 0.910 0.000 1.000
#> GSM97042 2 0.0000 0.910 0.000 1.000
#> GSM97040 2 0.0000 0.910 0.000 1.000
#> GSM97041 2 0.0000 0.910 0.000 1.000
#> GSM96955 2 0.0000 0.910 0.000 1.000
#> GSM96990 2 0.0000 0.910 0.000 1.000
#> GSM96991 2 0.0000 0.910 0.000 1.000
#> GSM97048 2 0.0000 0.910 0.000 1.000
#> GSM96963 2 0.0000 0.910 0.000 1.000
#> GSM96953 2 0.0000 0.910 0.000 1.000
#> GSM96966 1 0.0000 0.968 1.000 0.000
#> GSM96979 2 0.8327 0.776 0.264 0.736
#> GSM96983 2 0.0000 0.910 0.000 1.000
#> GSM96984 2 0.8327 0.776 0.264 0.736
#> GSM96994 2 0.0672 0.910 0.008 0.992
#> GSM96996 2 0.2043 0.905 0.032 0.968
#> GSM96997 2 0.8327 0.776 0.264 0.736
#> GSM97007 2 0.2043 0.905 0.032 0.968
#> GSM96954 2 0.8016 0.800 0.244 0.756
#> GSM96962 2 0.8327 0.776 0.264 0.736
#> GSM96969 1 0.0000 0.968 1.000 0.000
#> GSM96970 1 0.0000 0.968 1.000 0.000
#> GSM96973 1 0.0000 0.968 1.000 0.000
#> GSM96976 1 0.0000 0.968 1.000 0.000
#> GSM96977 2 0.7528 0.827 0.216 0.784
#> GSM96995 2 0.0000 0.910 0.000 1.000
#> GSM97002 2 0.7674 0.820 0.224 0.776
#> GSM97009 2 0.0376 0.910 0.004 0.996
#> GSM97010 2 0.7528 0.827 0.216 0.784
#> GSM96974 1 0.0000 0.968 1.000 0.000
#> GSM96985 2 0.6148 0.868 0.152 0.848
#> GSM96959 2 0.0000 0.910 0.000 1.000
#> GSM96972 1 0.0000 0.968 1.000 0.000
#> GSM96978 2 0.7674 0.820 0.224 0.776
#> GSM96967 1 0.0000 0.968 1.000 0.000
#> GSM96987 2 0.2236 0.904 0.036 0.964
#> GSM97011 2 0.0376 0.910 0.004 0.996
#> GSM96964 2 0.6343 0.864 0.160 0.840
#> GSM96965 1 0.0000 0.968 1.000 0.000
#> GSM96981 2 0.6801 0.854 0.180 0.820
#> GSM96982 2 0.6801 0.854 0.180 0.820
#> GSM96988 2 0.0938 0.909 0.012 0.988
#> GSM97000 2 0.6887 0.850 0.184 0.816
#> GSM97004 2 0.7745 0.816 0.228 0.772
#> GSM97008 2 0.6148 0.868 0.152 0.848
#> GSM96950 2 0.6973 0.847 0.188 0.812
#> GSM96980 2 0.7815 0.813 0.232 0.768
#> GSM96989 2 0.4562 0.888 0.096 0.904
#> GSM96992 2 0.6343 0.863 0.160 0.840
#> GSM96993 2 0.0000 0.910 0.000 1.000
#> GSM96958 2 0.7219 0.839 0.200 0.800
#> GSM96951 2 0.6973 0.847 0.188 0.812
#> GSM96952 2 0.6343 0.863 0.160 0.840
#> GSM96961 2 0.6343 0.863 0.160 0.840
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM97038 2 0.1163 0.7976 0.028 0.972 0.000
#> GSM97045 2 0.0000 0.7968 0.000 1.000 0.000
#> GSM97047 2 0.5465 0.6684 0.288 0.712 0.000
#> GSM97025 2 0.0000 0.7968 0.000 1.000 0.000
#> GSM97030 1 0.6189 0.4129 0.632 0.364 0.004
#> GSM97027 2 0.0000 0.7968 0.000 1.000 0.000
#> GSM97033 2 0.0000 0.7968 0.000 1.000 0.000
#> GSM97034 2 0.5810 0.6012 0.336 0.664 0.000
#> GSM97020 2 0.0000 0.7968 0.000 1.000 0.000
#> GSM97026 2 0.3412 0.7711 0.124 0.876 0.000
#> GSM97012 2 0.0000 0.7968 0.000 1.000 0.000
#> GSM97015 2 0.5948 0.5556 0.360 0.640 0.000
#> GSM97016 2 0.0000 0.7968 0.000 1.000 0.000
#> GSM97017 2 0.5497 0.6645 0.292 0.708 0.000
#> GSM97019 2 0.0000 0.7968 0.000 1.000 0.000
#> GSM97022 2 0.0000 0.7968 0.000 1.000 0.000
#> GSM97035 2 0.0000 0.7968 0.000 1.000 0.000
#> GSM97036 1 0.6095 0.3156 0.608 0.392 0.000
#> GSM97039 2 0.0000 0.7968 0.000 1.000 0.000
#> GSM97046 2 0.0000 0.7968 0.000 1.000 0.000
#> GSM97023 1 0.1753 0.8364 0.952 0.048 0.000
#> GSM97029 2 0.5835 0.5953 0.340 0.660 0.000
#> GSM97043 2 0.1964 0.7946 0.056 0.944 0.000
#> GSM97013 1 0.0237 0.8445 0.996 0.000 0.004
#> GSM96956 2 0.5706 0.4922 0.320 0.680 0.000
#> GSM97024 2 0.1163 0.7966 0.028 0.972 0.000
#> GSM97032 2 0.5560 0.6481 0.300 0.700 0.000
#> GSM97044 1 0.6189 0.4129 0.632 0.364 0.004
#> GSM97049 2 0.0000 0.7968 0.000 1.000 0.000
#> GSM96968 1 0.1999 0.8448 0.952 0.012 0.036
#> GSM96971 3 0.5058 0.6303 0.244 0.000 0.756
#> GSM96986 1 0.2878 0.8148 0.904 0.000 0.096
#> GSM97003 1 0.1643 0.8394 0.956 0.000 0.044
#> GSM96957 1 0.3752 0.7634 0.856 0.144 0.000
#> GSM96960 1 0.1643 0.8388 0.956 0.000 0.044
#> GSM96975 1 0.2793 0.8437 0.928 0.044 0.028
#> GSM96998 1 0.1525 0.8432 0.964 0.032 0.004
#> GSM96999 1 0.2176 0.8468 0.948 0.032 0.020
#> GSM97001 1 0.3752 0.7634 0.856 0.144 0.000
#> GSM97005 1 0.0237 0.8445 0.996 0.000 0.004
#> GSM97006 1 0.1860 0.8365 0.948 0.000 0.052
#> GSM97021 2 0.5497 0.6645 0.292 0.708 0.000
#> GSM97028 2 0.6215 0.3800 0.428 0.572 0.000
#> GSM97031 1 0.0237 0.8445 0.996 0.000 0.004
#> GSM97037 1 0.6189 0.4129 0.632 0.364 0.004
#> GSM97018 2 0.5733 0.6171 0.324 0.676 0.000
#> GSM97014 2 0.5465 0.6684 0.288 0.712 0.000
#> GSM97042 2 0.0000 0.7968 0.000 1.000 0.000
#> GSM97040 2 0.5497 0.6645 0.292 0.708 0.000
#> GSM97041 2 0.5497 0.6645 0.292 0.708 0.000
#> GSM96955 2 0.0000 0.7968 0.000 1.000 0.000
#> GSM96990 2 0.5882 0.5779 0.348 0.652 0.000
#> GSM96991 2 0.0000 0.7968 0.000 1.000 0.000
#> GSM97048 2 0.0000 0.7968 0.000 1.000 0.000
#> GSM96963 2 0.0000 0.7968 0.000 1.000 0.000
#> GSM96953 2 0.1031 0.7964 0.024 0.976 0.000
#> GSM96966 3 0.0237 0.9684 0.004 0.000 0.996
#> GSM96979 1 0.2878 0.8148 0.904 0.000 0.096
#> GSM96983 1 0.5623 0.5998 0.716 0.280 0.004
#> GSM96984 1 0.2878 0.8148 0.904 0.000 0.096
#> GSM96994 1 0.5815 0.5526 0.692 0.304 0.004
#> GSM96996 1 0.5560 0.5319 0.700 0.300 0.000
#> GSM96997 1 0.2878 0.8148 0.904 0.000 0.096
#> GSM97007 1 0.5171 0.7083 0.784 0.204 0.012
#> GSM96954 1 0.2680 0.8315 0.924 0.008 0.068
#> GSM96962 1 0.2878 0.8148 0.904 0.000 0.096
#> GSM96969 3 0.0237 0.9684 0.004 0.000 0.996
#> GSM96970 3 0.0237 0.9684 0.004 0.000 0.996
#> GSM96973 3 0.0237 0.9684 0.004 0.000 0.996
#> GSM96976 3 0.0237 0.9684 0.004 0.000 0.996
#> GSM96977 1 0.1711 0.8448 0.960 0.008 0.032
#> GSM96995 2 0.5859 0.5860 0.344 0.656 0.000
#> GSM97002 1 0.1529 0.8397 0.960 0.000 0.040
#> GSM97009 1 0.6295 -0.0383 0.528 0.472 0.000
#> GSM97010 1 0.1525 0.8439 0.964 0.004 0.032
#> GSM96974 3 0.0237 0.9684 0.004 0.000 0.996
#> GSM96985 1 0.3765 0.8207 0.888 0.084 0.028
#> GSM96959 2 0.5859 0.5860 0.344 0.656 0.000
#> GSM96972 3 0.0237 0.9684 0.004 0.000 0.996
#> GSM96978 1 0.2903 0.8382 0.924 0.028 0.048
#> GSM96967 3 0.0237 0.9684 0.004 0.000 0.996
#> GSM96987 1 0.5905 0.4247 0.648 0.352 0.000
#> GSM97011 1 0.6295 -0.0383 0.528 0.472 0.000
#> GSM96964 1 0.1525 0.8447 0.964 0.032 0.004
#> GSM96965 3 0.0237 0.9684 0.004 0.000 0.996
#> GSM96981 1 0.2793 0.8437 0.928 0.044 0.028
#> GSM96982 1 0.2793 0.8437 0.928 0.044 0.028
#> GSM96988 2 0.6309 0.1390 0.496 0.504 0.000
#> GSM97000 1 0.0475 0.8454 0.992 0.004 0.004
#> GSM97004 1 0.1643 0.8388 0.956 0.000 0.044
#> GSM97008 1 0.1411 0.8413 0.964 0.036 0.000
#> GSM96950 1 0.0237 0.8445 0.996 0.000 0.004
#> GSM96980 1 0.1860 0.8377 0.948 0.000 0.052
#> GSM96989 1 0.5553 0.5914 0.724 0.272 0.004
#> GSM96992 1 0.1525 0.8432 0.964 0.032 0.004
#> GSM96993 1 0.6095 0.3156 0.608 0.392 0.000
#> GSM96958 1 0.0747 0.8444 0.984 0.000 0.016
#> GSM96951 1 0.0237 0.8445 0.996 0.000 0.004
#> GSM96952 1 0.1525 0.8432 0.964 0.032 0.004
#> GSM96961 1 0.1525 0.8432 0.964 0.032 0.004
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM97038 2 0.1356 0.7586 0.032 0.960 0.008 0.000
#> GSM97045 2 0.0000 0.7629 0.000 1.000 0.000 0.000
#> GSM97047 2 0.5814 0.6171 0.300 0.644 0.056 0.000
#> GSM97025 2 0.0000 0.7629 0.000 1.000 0.000 0.000
#> GSM97030 3 0.4638 0.6602 0.044 0.180 0.776 0.000
#> GSM97027 2 0.0000 0.7629 0.000 1.000 0.000 0.000
#> GSM97033 2 0.0000 0.7629 0.000 1.000 0.000 0.000
#> GSM97034 2 0.6407 0.5561 0.332 0.584 0.084 0.000
#> GSM97020 2 0.0000 0.7629 0.000 1.000 0.000 0.000
#> GSM97026 2 0.4286 0.7215 0.136 0.812 0.052 0.000
#> GSM97012 2 0.0000 0.7629 0.000 1.000 0.000 0.000
#> GSM97015 2 0.7117 0.5739 0.264 0.556 0.180 0.000
#> GSM97016 2 0.0000 0.7629 0.000 1.000 0.000 0.000
#> GSM97017 2 0.5905 0.6111 0.304 0.636 0.060 0.000
#> GSM97019 2 0.0000 0.7629 0.000 1.000 0.000 0.000
#> GSM97022 2 0.0000 0.7629 0.000 1.000 0.000 0.000
#> GSM97035 2 0.0000 0.7629 0.000 1.000 0.000 0.000
#> GSM97036 1 0.6141 0.3814 0.624 0.300 0.076 0.000
#> GSM97039 2 0.0000 0.7629 0.000 1.000 0.000 0.000
#> GSM97046 2 0.0000 0.7629 0.000 1.000 0.000 0.000
#> GSM97023 1 0.1109 0.8065 0.968 0.004 0.028 0.000
#> GSM97029 2 0.6382 0.5454 0.340 0.580 0.080 0.000
#> GSM97043 2 0.2996 0.7439 0.064 0.892 0.044 0.000
#> GSM97013 1 0.1557 0.8082 0.944 0.000 0.056 0.000
#> GSM96956 2 0.5284 0.2465 0.016 0.616 0.368 0.000
#> GSM97024 2 0.1824 0.7397 0.004 0.936 0.060 0.000
#> GSM97032 2 0.6452 0.6250 0.268 0.620 0.112 0.000
#> GSM97044 3 0.4638 0.6602 0.044 0.180 0.776 0.000
#> GSM97049 2 0.0000 0.7629 0.000 1.000 0.000 0.000
#> GSM96968 1 0.5523 0.3528 0.628 0.012 0.348 0.012
#> GSM96971 4 0.4072 0.6379 0.000 0.000 0.252 0.748
#> GSM96986 3 0.4356 0.7289 0.148 0.000 0.804 0.048
#> GSM97003 1 0.3166 0.7656 0.868 0.000 0.116 0.016
#> GSM96957 1 0.3370 0.7455 0.872 0.080 0.048 0.000
#> GSM96960 1 0.1743 0.8004 0.940 0.000 0.056 0.004
#> GSM96975 1 0.1452 0.8110 0.956 0.008 0.036 0.000
#> GSM96998 1 0.0524 0.8122 0.988 0.004 0.008 0.000
#> GSM96999 1 0.2384 0.7929 0.916 0.008 0.072 0.004
#> GSM97001 1 0.3370 0.7455 0.872 0.080 0.048 0.000
#> GSM97005 1 0.1557 0.8082 0.944 0.000 0.056 0.000
#> GSM97006 1 0.3224 0.7627 0.864 0.000 0.120 0.016
#> GSM97021 2 0.5905 0.6111 0.304 0.636 0.060 0.000
#> GSM97028 2 0.7489 0.4873 0.296 0.492 0.212 0.000
#> GSM97031 1 0.1637 0.8077 0.940 0.000 0.060 0.000
#> GSM97037 3 0.4638 0.6602 0.044 0.180 0.776 0.000
#> GSM97018 2 0.6613 0.6000 0.288 0.596 0.116 0.000
#> GSM97014 2 0.5814 0.6171 0.300 0.644 0.056 0.000
#> GSM97042 2 0.0000 0.7629 0.000 1.000 0.000 0.000
#> GSM97040 2 0.5905 0.6111 0.304 0.636 0.060 0.000
#> GSM97041 2 0.5905 0.6111 0.304 0.636 0.060 0.000
#> GSM96955 2 0.0000 0.7629 0.000 1.000 0.000 0.000
#> GSM96990 2 0.6790 0.5770 0.296 0.576 0.128 0.000
#> GSM96991 2 0.0000 0.7629 0.000 1.000 0.000 0.000
#> GSM97048 2 0.0000 0.7629 0.000 1.000 0.000 0.000
#> GSM96963 2 0.0000 0.7629 0.000 1.000 0.000 0.000
#> GSM96953 2 0.1389 0.7451 0.000 0.952 0.048 0.000
#> GSM96966 4 0.0000 0.9692 0.000 0.000 0.000 1.000
#> GSM96979 3 0.4356 0.7289 0.148 0.000 0.804 0.048
#> GSM96983 3 0.3652 0.7010 0.052 0.092 0.856 0.000
#> GSM96984 3 0.4356 0.7289 0.148 0.000 0.804 0.048
#> GSM96994 3 0.4673 0.6907 0.076 0.132 0.792 0.000
#> GSM96996 1 0.5494 0.5673 0.716 0.208 0.076 0.000
#> GSM96997 3 0.4356 0.7289 0.148 0.000 0.804 0.048
#> GSM97007 3 0.2256 0.7006 0.056 0.020 0.924 0.000
#> GSM96954 3 0.5966 0.4801 0.368 0.008 0.592 0.032
#> GSM96962 3 0.4356 0.7289 0.148 0.000 0.804 0.048
#> GSM96969 4 0.0000 0.9692 0.000 0.000 0.000 1.000
#> GSM96970 4 0.0000 0.9692 0.000 0.000 0.000 1.000
#> GSM96973 4 0.0000 0.9692 0.000 0.000 0.000 1.000
#> GSM96976 4 0.0000 0.9692 0.000 0.000 0.000 1.000
#> GSM96977 1 0.5270 0.4293 0.660 0.008 0.320 0.012
#> GSM96995 2 0.6739 0.5746 0.304 0.576 0.120 0.000
#> GSM97002 1 0.1661 0.8016 0.944 0.000 0.052 0.004
#> GSM97009 1 0.6357 0.0653 0.544 0.388 0.068 0.000
#> GSM97010 1 0.5112 0.4427 0.668 0.004 0.316 0.012
#> GSM96974 4 0.0000 0.9692 0.000 0.000 0.000 1.000
#> GSM96985 1 0.3015 0.7932 0.884 0.024 0.092 0.000
#> GSM96959 2 0.6739 0.5746 0.304 0.576 0.120 0.000
#> GSM96972 4 0.0000 0.9692 0.000 0.000 0.000 1.000
#> GSM96978 3 0.5633 0.5127 0.348 0.012 0.624 0.016
#> GSM96967 4 0.0000 0.9692 0.000 0.000 0.000 1.000
#> GSM96987 1 0.5900 0.4797 0.664 0.260 0.076 0.000
#> GSM97011 1 0.6357 0.0653 0.544 0.388 0.068 0.000
#> GSM96964 1 0.1767 0.8143 0.944 0.012 0.044 0.000
#> GSM96965 4 0.0000 0.9692 0.000 0.000 0.000 1.000
#> GSM96981 1 0.1452 0.8110 0.956 0.008 0.036 0.000
#> GSM96982 1 0.1452 0.8110 0.956 0.008 0.036 0.000
#> GSM96988 2 0.7910 0.2212 0.308 0.360 0.332 0.000
#> GSM97000 1 0.1792 0.8081 0.932 0.000 0.068 0.000
#> GSM97004 1 0.1743 0.8004 0.940 0.000 0.056 0.004
#> GSM97008 1 0.2413 0.8098 0.916 0.020 0.064 0.000
#> GSM96950 1 0.1557 0.8082 0.944 0.000 0.056 0.000
#> GSM96980 1 0.2021 0.7997 0.932 0.000 0.056 0.012
#> GSM96989 1 0.4914 0.6174 0.748 0.208 0.044 0.000
#> GSM96992 1 0.0524 0.8122 0.988 0.004 0.008 0.000
#> GSM96993 1 0.6141 0.3814 0.624 0.300 0.076 0.000
#> GSM96958 1 0.2266 0.7965 0.912 0.000 0.084 0.004
#> GSM96951 1 0.1557 0.8082 0.944 0.000 0.056 0.000
#> GSM96952 1 0.0524 0.8122 0.988 0.004 0.008 0.000
#> GSM96961 1 0.0524 0.8122 0.988 0.004 0.008 0.000
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM97038 2 0.2873 0.6796 0.020 0.860 0.000 0.000 0.120
#> GSM97045 2 0.0324 0.7026 0.004 0.992 0.000 0.000 0.004
#> GSM97047 2 0.6459 0.4834 0.244 0.500 0.000 0.000 0.256
#> GSM97025 2 0.0324 0.7026 0.004 0.992 0.000 0.000 0.004
#> GSM97030 3 0.3911 0.6347 0.000 0.144 0.796 0.000 0.060
#> GSM97027 2 0.0324 0.7026 0.004 0.992 0.000 0.000 0.004
#> GSM97033 2 0.0324 0.7026 0.004 0.992 0.000 0.000 0.004
#> GSM97034 2 0.7135 0.4270 0.276 0.436 0.020 0.000 0.268
#> GSM97020 2 0.0324 0.7026 0.004 0.992 0.000 0.000 0.004
#> GSM97026 2 0.5190 0.6030 0.096 0.668 0.000 0.000 0.236
#> GSM97012 2 0.0000 0.7004 0.000 1.000 0.000 0.000 0.000
#> GSM97015 2 0.8023 0.4361 0.212 0.424 0.120 0.000 0.244
#> GSM97016 2 0.0324 0.7026 0.004 0.992 0.000 0.000 0.004
#> GSM97017 2 0.6491 0.4750 0.244 0.492 0.000 0.000 0.264
#> GSM97019 2 0.0000 0.7004 0.000 1.000 0.000 0.000 0.000
#> GSM97022 2 0.0000 0.7004 0.000 1.000 0.000 0.000 0.000
#> GSM97035 2 0.0000 0.7004 0.000 1.000 0.000 0.000 0.000
#> GSM97036 5 0.6075 0.1635 0.356 0.132 0.000 0.000 0.512
#> GSM97039 2 0.0000 0.7004 0.000 1.000 0.000 0.000 0.000
#> GSM97046 2 0.0000 0.7004 0.000 1.000 0.000 0.000 0.000
#> GSM97023 1 0.2074 0.5701 0.896 0.000 0.000 0.000 0.104
#> GSM97029 2 0.7078 0.4182 0.284 0.432 0.016 0.000 0.268
#> GSM97043 2 0.3944 0.6523 0.032 0.768 0.000 0.000 0.200
#> GSM97013 1 0.0162 0.6120 0.996 0.000 0.004 0.000 0.000
#> GSM96956 2 0.4862 0.1613 0.000 0.604 0.364 0.000 0.032
#> GSM97024 2 0.1800 0.6747 0.000 0.932 0.048 0.000 0.020
#> GSM97032 2 0.7385 0.4841 0.212 0.472 0.052 0.000 0.264
#> GSM97044 3 0.3911 0.6347 0.000 0.144 0.796 0.000 0.060
#> GSM97049 2 0.0290 0.7020 0.000 0.992 0.000 0.000 0.008
#> GSM96968 1 0.4775 0.2801 0.660 0.004 0.304 0.000 0.032
#> GSM96971 4 0.4141 0.6211 0.000 0.000 0.248 0.728 0.024
#> GSM96986 3 0.4587 0.6674 0.160 0.000 0.744 0.000 0.096
#> GSM97003 1 0.2726 0.5425 0.884 0.000 0.064 0.000 0.052
#> GSM96957 1 0.3681 0.4572 0.808 0.044 0.000 0.000 0.148
#> GSM96960 1 0.4268 -0.3766 0.556 0.000 0.000 0.000 0.444
#> GSM96975 5 0.4249 0.4827 0.432 0.000 0.000 0.000 0.568
#> GSM96998 1 0.2020 0.5787 0.900 0.000 0.000 0.000 0.100
#> GSM96999 1 0.3390 0.5696 0.840 0.000 0.060 0.000 0.100
#> GSM97001 1 0.3681 0.4572 0.808 0.044 0.000 0.000 0.148
#> GSM97005 1 0.0162 0.6120 0.996 0.000 0.004 0.000 0.000
#> GSM97006 1 0.2863 0.5340 0.876 0.000 0.060 0.000 0.064
#> GSM97021 2 0.6491 0.4750 0.244 0.492 0.000 0.000 0.264
#> GSM97028 2 0.8340 0.3389 0.236 0.360 0.152 0.000 0.252
#> GSM97031 1 0.0290 0.6119 0.992 0.000 0.008 0.000 0.000
#> GSM97037 3 0.3911 0.6347 0.000 0.144 0.796 0.000 0.060
#> GSM97018 2 0.7529 0.4605 0.232 0.448 0.056 0.000 0.264
#> GSM97014 2 0.6459 0.4834 0.244 0.500 0.000 0.000 0.256
#> GSM97042 2 0.0000 0.7004 0.000 1.000 0.000 0.000 0.000
#> GSM97040 2 0.6491 0.4750 0.244 0.492 0.000 0.000 0.264
#> GSM97041 2 0.6491 0.4750 0.244 0.492 0.000 0.000 0.264
#> GSM96955 2 0.0880 0.6929 0.000 0.968 0.000 0.000 0.032
#> GSM96990 2 0.7693 0.4433 0.240 0.432 0.068 0.000 0.260
#> GSM96991 2 0.0880 0.6929 0.000 0.968 0.000 0.000 0.032
#> GSM97048 2 0.0000 0.7004 0.000 1.000 0.000 0.000 0.000
#> GSM96963 2 0.0880 0.6929 0.000 0.968 0.000 0.000 0.032
#> GSM96953 2 0.1485 0.6840 0.000 0.948 0.032 0.000 0.020
#> GSM96966 4 0.0000 0.9679 0.000 0.000 0.000 1.000 0.000
#> GSM96979 3 0.4587 0.6674 0.160 0.000 0.744 0.000 0.096
#> GSM96983 3 0.3323 0.6419 0.000 0.056 0.844 0.000 0.100
#> GSM96984 3 0.4587 0.6674 0.160 0.000 0.744 0.000 0.096
#> GSM96994 3 0.4444 0.6485 0.020 0.088 0.788 0.000 0.104
#> GSM96996 1 0.5708 -0.0481 0.504 0.084 0.000 0.000 0.412
#> GSM96997 3 0.4587 0.6674 0.160 0.000 0.744 0.000 0.096
#> GSM97007 3 0.1608 0.6286 0.000 0.000 0.928 0.000 0.072
#> GSM96954 3 0.5559 0.4045 0.380 0.000 0.544 0.000 0.076
#> GSM96962 3 0.4587 0.6674 0.160 0.000 0.744 0.000 0.096
#> GSM96969 4 0.0000 0.9679 0.000 0.000 0.000 1.000 0.000
#> GSM96970 4 0.0000 0.9679 0.000 0.000 0.000 1.000 0.000
#> GSM96973 4 0.0000 0.9679 0.000 0.000 0.000 1.000 0.000
#> GSM96976 4 0.0000 0.9679 0.000 0.000 0.000 1.000 0.000
#> GSM96977 1 0.4397 0.3221 0.696 0.000 0.276 0.000 0.028
#> GSM96995 2 0.7625 0.4409 0.244 0.432 0.060 0.000 0.264
#> GSM97002 1 0.4305 -0.4564 0.512 0.000 0.000 0.000 0.488
#> GSM97009 1 0.6765 -0.0312 0.448 0.248 0.004 0.000 0.300
#> GSM97010 1 0.4350 0.3341 0.704 0.000 0.268 0.000 0.028
#> GSM96974 4 0.0000 0.9679 0.000 0.000 0.000 1.000 0.000
#> GSM96985 5 0.4088 0.4695 0.368 0.000 0.000 0.000 0.632
#> GSM96959 2 0.7625 0.4409 0.244 0.432 0.060 0.000 0.264
#> GSM96972 4 0.0000 0.9679 0.000 0.000 0.000 1.000 0.000
#> GSM96978 3 0.5554 0.4425 0.360 0.004 0.568 0.000 0.068
#> GSM96967 4 0.0000 0.9679 0.000 0.000 0.000 1.000 0.000
#> GSM96987 5 0.5915 0.1534 0.384 0.108 0.000 0.000 0.508
#> GSM97011 1 0.6765 -0.0312 0.448 0.248 0.004 0.000 0.300
#> GSM96964 1 0.1430 0.6061 0.944 0.000 0.004 0.000 0.052
#> GSM96965 4 0.0000 0.9679 0.000 0.000 0.000 1.000 0.000
#> GSM96981 5 0.4249 0.4827 0.432 0.000 0.000 0.000 0.568
#> GSM96982 5 0.4249 0.4827 0.432 0.000 0.000 0.000 0.568
#> GSM96988 3 0.8554 -0.1495 0.228 0.256 0.304 0.000 0.212
#> GSM97000 1 0.0671 0.6113 0.980 0.000 0.016 0.000 0.004
#> GSM97004 5 0.4307 0.3889 0.496 0.000 0.000 0.000 0.504
#> GSM97008 1 0.1365 0.6043 0.952 0.004 0.004 0.000 0.040
#> GSM96950 1 0.0162 0.6120 0.996 0.000 0.004 0.000 0.000
#> GSM96980 5 0.4559 0.4016 0.480 0.000 0.000 0.008 0.512
#> GSM96989 1 0.5884 -0.1544 0.480 0.100 0.000 0.000 0.420
#> GSM96992 1 0.2127 0.5724 0.892 0.000 0.000 0.000 0.108
#> GSM96993 5 0.6075 0.1635 0.356 0.132 0.000 0.000 0.512
#> GSM96958 1 0.1386 0.5992 0.952 0.000 0.032 0.000 0.016
#> GSM96951 1 0.0162 0.6120 0.996 0.000 0.004 0.000 0.000
#> GSM96952 1 0.2127 0.5724 0.892 0.000 0.000 0.000 0.108
#> GSM96961 1 0.2127 0.5724 0.892 0.000 0.000 0.000 0.108
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM97038 2 0.3003 0.68828 0.016 0.812 0.000 0.000 0.172 0.000
#> GSM97045 2 0.0508 0.88884 0.004 0.984 0.000 0.000 0.012 0.000
#> GSM97047 5 0.6379 0.39498 0.196 0.376 0.024 0.000 0.404 0.000
#> GSM97025 2 0.0508 0.88884 0.004 0.984 0.000 0.000 0.012 0.000
#> GSM97030 3 0.5160 0.73437 0.000 0.084 0.680 0.000 0.044 0.192
#> GSM97027 2 0.0508 0.88884 0.004 0.984 0.000 0.000 0.012 0.000
#> GSM97033 2 0.0508 0.88884 0.004 0.984 0.000 0.000 0.012 0.000
#> GSM97034 5 0.6932 0.45750 0.240 0.312 0.060 0.000 0.388 0.000
#> GSM97020 2 0.0508 0.88884 0.004 0.984 0.000 0.000 0.012 0.000
#> GSM97026 2 0.5350 0.05481 0.068 0.548 0.020 0.000 0.364 0.000
#> GSM97012 2 0.0000 0.88988 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97015 5 0.7861 0.39206 0.184 0.300 0.144 0.000 0.348 0.024
#> GSM97016 2 0.0508 0.88884 0.004 0.984 0.000 0.000 0.012 0.000
#> GSM97017 5 0.6311 0.40135 0.196 0.372 0.020 0.000 0.412 0.000
#> GSM97019 2 0.0000 0.88988 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97022 2 0.0000 0.88988 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97035 2 0.0000 0.88988 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97036 5 0.5892 0.24503 0.292 0.024 0.112 0.000 0.564 0.008
#> GSM97039 2 0.0000 0.88988 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97046 2 0.0000 0.88988 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97023 1 0.1843 0.80751 0.912 0.000 0.004 0.000 0.080 0.004
#> GSM97029 5 0.6856 0.45832 0.244 0.312 0.052 0.000 0.392 0.000
#> GSM97043 2 0.4643 0.38766 0.028 0.660 0.028 0.000 0.284 0.000
#> GSM97013 1 0.0632 0.84055 0.976 0.000 0.000 0.000 0.000 0.024
#> GSM96956 2 0.5771 0.14436 0.000 0.544 0.332 0.000 0.040 0.084
#> GSM97024 2 0.2060 0.83074 0.000 0.900 0.084 0.000 0.016 0.000
#> GSM97032 5 0.7128 0.39338 0.184 0.352 0.100 0.000 0.364 0.000
#> GSM97044 3 0.5160 0.73437 0.000 0.084 0.680 0.000 0.044 0.192
#> GSM97049 2 0.1082 0.86884 0.000 0.956 0.004 0.000 0.040 0.000
#> GSM96968 1 0.4264 0.33113 0.604 0.000 0.008 0.000 0.012 0.376
#> GSM96971 4 0.3266 0.63630 0.000 0.000 0.000 0.728 0.000 0.272
#> GSM96986 6 0.0865 0.80604 0.036 0.000 0.000 0.000 0.000 0.964
#> GSM97003 1 0.2886 0.76788 0.836 0.000 0.004 0.000 0.016 0.144
#> GSM96957 1 0.2879 0.69073 0.816 0.004 0.004 0.000 0.176 0.000
#> GSM96960 5 0.6118 -0.12685 0.404 0.000 0.116 0.000 0.444 0.036
#> GSM96975 5 0.5625 0.01880 0.284 0.000 0.120 0.000 0.576 0.020
#> GSM96998 1 0.1806 0.81018 0.908 0.000 0.004 0.000 0.088 0.000
#> GSM96999 1 0.3208 0.80357 0.844 0.000 0.012 0.000 0.076 0.068
#> GSM97001 1 0.2879 0.69073 0.816 0.004 0.004 0.000 0.176 0.000
#> GSM97005 1 0.0632 0.84055 0.976 0.000 0.000 0.000 0.000 0.024
#> GSM97006 1 0.3035 0.76054 0.828 0.000 0.008 0.000 0.016 0.148
#> GSM97021 5 0.6311 0.40135 0.196 0.372 0.020 0.000 0.412 0.000
#> GSM97028 5 0.8172 0.34544 0.200 0.232 0.156 0.000 0.364 0.048
#> GSM97031 1 0.0713 0.84099 0.972 0.000 0.000 0.000 0.000 0.028
#> GSM97037 3 0.5160 0.73437 0.000 0.084 0.680 0.000 0.044 0.192
#> GSM97018 5 0.7201 0.43071 0.200 0.324 0.104 0.000 0.372 0.000
#> GSM97014 5 0.6379 0.39498 0.196 0.376 0.024 0.000 0.404 0.000
#> GSM97042 2 0.0000 0.88988 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97040 5 0.6311 0.40135 0.196 0.372 0.020 0.000 0.412 0.000
#> GSM97041 5 0.6311 0.40135 0.196 0.372 0.020 0.000 0.412 0.000
#> GSM96955 2 0.1895 0.83767 0.000 0.912 0.072 0.000 0.016 0.000
#> GSM96990 5 0.7388 0.44076 0.208 0.304 0.112 0.000 0.372 0.004
#> GSM96991 2 0.1895 0.83767 0.000 0.912 0.072 0.000 0.016 0.000
#> GSM97048 2 0.0000 0.88988 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM96963 2 0.1895 0.83767 0.000 0.912 0.072 0.000 0.016 0.000
#> GSM96953 2 0.1970 0.84098 0.000 0.912 0.060 0.000 0.028 0.000
#> GSM96966 4 0.0000 0.96912 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM96979 6 0.0865 0.80604 0.036 0.000 0.000 0.000 0.000 0.964
#> GSM96983 3 0.3917 0.66625 0.000 0.012 0.752 0.000 0.032 0.204
#> GSM96984 6 0.0865 0.80604 0.036 0.000 0.000 0.000 0.000 0.964
#> GSM96994 3 0.5452 0.69295 0.020 0.036 0.668 0.000 0.068 0.208
#> GSM96996 5 0.5483 0.01917 0.444 0.008 0.064 0.000 0.472 0.012
#> GSM96997 6 0.0865 0.80604 0.036 0.000 0.000 0.000 0.000 0.964
#> GSM97007 3 0.3428 0.50688 0.000 0.000 0.696 0.000 0.000 0.304
#> GSM96954 6 0.4273 0.52189 0.324 0.000 0.012 0.000 0.016 0.648
#> GSM96962 6 0.0865 0.80604 0.036 0.000 0.000 0.000 0.000 0.964
#> GSM96969 4 0.0000 0.96912 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM96970 4 0.0000 0.96912 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM96973 4 0.0000 0.96912 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM96976 4 0.0000 0.96912 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM96977 1 0.4088 0.41954 0.636 0.000 0.008 0.000 0.008 0.348
#> GSM96995 5 0.7350 0.44593 0.212 0.308 0.104 0.000 0.372 0.004
#> GSM97002 5 0.6025 -0.03917 0.336 0.000 0.124 0.000 0.508 0.032
#> GSM97009 5 0.6121 0.27183 0.412 0.116 0.036 0.000 0.436 0.000
#> GSM97010 1 0.4031 0.45741 0.652 0.000 0.008 0.000 0.008 0.332
#> GSM96974 4 0.0000 0.96912 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM96985 5 0.5457 0.03754 0.164 0.000 0.148 0.000 0.652 0.036
#> GSM96959 5 0.7350 0.44593 0.212 0.308 0.104 0.000 0.372 0.004
#> GSM96972 4 0.0000 0.96912 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM96978 6 0.6080 0.36755 0.300 0.000 0.188 0.000 0.016 0.496
#> GSM96967 4 0.0000 0.96912 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM96987 5 0.6011 0.20359 0.304 0.016 0.128 0.000 0.540 0.012
#> GSM97011 5 0.6121 0.27183 0.412 0.116 0.036 0.000 0.436 0.000
#> GSM96964 1 0.1873 0.83622 0.924 0.000 0.008 0.000 0.048 0.020
#> GSM96965 4 0.0000 0.96912 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM96981 5 0.5625 0.01880 0.284 0.000 0.120 0.000 0.576 0.020
#> GSM96982 5 0.5625 0.01880 0.284 0.000 0.120 0.000 0.576 0.020
#> GSM96988 3 0.8438 -0.19149 0.196 0.160 0.300 0.000 0.276 0.068
#> GSM97000 1 0.1152 0.83944 0.952 0.000 0.000 0.000 0.004 0.044
#> GSM97004 5 0.6077 -0.02266 0.320 0.000 0.128 0.000 0.516 0.036
#> GSM97008 1 0.1536 0.83247 0.940 0.000 0.004 0.000 0.040 0.016
#> GSM96950 1 0.0790 0.84094 0.968 0.000 0.000 0.000 0.000 0.032
#> GSM96980 5 0.6311 -0.01456 0.300 0.000 0.136 0.008 0.520 0.036
#> GSM96989 5 0.5961 -0.00118 0.408 0.016 0.108 0.000 0.460 0.008
#> GSM96992 1 0.2001 0.80586 0.900 0.000 0.004 0.000 0.092 0.004
#> GSM96993 5 0.5892 0.24503 0.292 0.024 0.112 0.000 0.564 0.008
#> GSM96958 1 0.1812 0.82614 0.912 0.000 0.000 0.000 0.008 0.080
#> GSM96951 1 0.0713 0.84088 0.972 0.000 0.000 0.000 0.000 0.028
#> GSM96952 1 0.2001 0.80586 0.900 0.000 0.004 0.000 0.092 0.004
#> GSM96961 1 0.2001 0.80586 0.900 0.000 0.004 0.000 0.092 0.004
Heatmaps for the consensus matrix. It visualizes the probability of two samples to be in a same group.
consensus_heatmap(res, k = 2)
consensus_heatmap(res, k = 3)
consensus_heatmap(res, k = 4)
consensus_heatmap(res, k = 5)
consensus_heatmap(res, k = 6)
Heatmaps for the membership of samples in all partitions to see how consistent they are:
membership_heatmap(res, k = 2)
membership_heatmap(res, k = 3)
membership_heatmap(res, k = 4)
membership_heatmap(res, k = 5)
membership_heatmap(res, k = 6)
As soon as we have had the classes for columns, we can look for signatures which are significantly different between classes which can be candidate marks for certain classes. Following are the heatmaps for signatures.
Signature heatmaps where rows are scaled:
get_signatures(res, k = 2)
get_signatures(res, k = 3)
get_signatures(res, k = 4)
get_signatures(res, k = 5)
get_signatures(res, k = 6)
Signature heatmaps where rows are not scaled:
get_signatures(res, k = 2, scale_rows = FALSE)
get_signatures(res, k = 3, scale_rows = FALSE)
get_signatures(res, k = 4, scale_rows = FALSE)
get_signatures(res, k = 5, scale_rows = FALSE)
get_signatures(res, k = 6, scale_rows = FALSE)
Compare the overlap of signatures from different k:
compare_signatures(res)
get_signature() returns a data frame invisibly. TO get the list of signatures, the function
call should be assigned to a variable explicitly. In following code, if plot argument is set
to FALSE, no heatmap is plotted while only the differential analysis is performed.
# code only for demonstration
tb = get_signature(res, k = ..., plot = FALSE)
An example of the output of tb is:
#> which_row fdr mean_1 mean_2 scaled_mean_1 scaled_mean_2 km
#> 1 38 0.042760348 8.373488 9.131774 -0.5533452 0.5164555 1
#> 2 40 0.018707592 7.106213 8.469186 -0.6173731 0.5762149 1
#> 3 55 0.019134737 10.221463 11.207825 -0.6159697 0.5749050 1
#> 4 59 0.006059896 5.921854 7.869574 -0.6899429 0.6439467 1
#> 5 60 0.018055526 8.928898 10.211722 -0.6204761 0.5791110 1
#> 6 98 0.009384629 15.714769 14.887706 0.6635654 -0.6193277 2
...
The columns in tb are:
which_row: row indices corresponding to the input matrix.fdr: FDR for the differential test. mean_x: The mean value in group x.scaled_mean_x: The mean value in group x after rows are scaled.km: Row groups if k-means clustering is applied to rows.UMAP plot which shows how samples are separated.
dimension_reduction(res, k = 2, method = "UMAP")
dimension_reduction(res, k = 3, method = "UMAP")
dimension_reduction(res, k = 4, method = "UMAP")
dimension_reduction(res, k = 5, method = "UMAP")
dimension_reduction(res, k = 6, method = "UMAP")
Following heatmap shows how subgroups are split when increasing k:
collect_classes(res)
Test correlation between subgroups and known annotations. If the known annotation is numeric, one-way ANOVA test is applied, and if the known annotation is discrete, chi-squared contingency table test is applied.
test_to_known_factors(res)
#> n disease.state(p) specimen(p) cell.type(p) other(p) k
#> ATC:hclust 100 5.90e-02 0.5263 4.19e-04 0.45353 2
#> ATC:hclust 89 6.61e-08 0.1775 4.27e-16 0.06201 3
#> ATC:hclust 88 6.64e-06 0.1174 1.27e-19 0.00812 4
#> ATC:hclust 61 4.28e-04 0.1236 1.61e-17 0.00533 5
#> ATC:hclust 62 5.10e-04 0.0427 9.55e-18 0.01490 6
If matrix rows can be associated to genes, consider to use functional_enrichment(res,
...) to perform function enrichment for the signature genes. See this vignette for more detailed explanations.
The object with results only for a single top-value method and a single partition method can be extracted as:
res = res_list["ATC", "kmeans"]
# you can also extract it by
# res = res_list["ATC:kmeans"]
A summary of res and all the functions that can be applied to it:
res
#> A 'ConsensusPartition' object with k = 2, 3, 4, 5, 6.
#> On a matrix with 21168 rows and 100 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 3.
#>
#> Following methods can be applied to this 'ConsensusPartition' object:
#> [1] "cola_report" "collect_classes" "collect_plots"
#> [4] "collect_stats" "colnames" "compare_signatures"
#> [7] "consensus_heatmap" "dimension_reduction" "functional_enrichment"
#> [10] "get_anno_col" "get_anno" "get_classes"
#> [13] "get_consensus" "get_matrix" "get_membership"
#> [16] "get_param" "get_signatures" "get_stats"
#> [19] "is_best_k" "is_stable_k" "membership_heatmap"
#> [22] "ncol" "nrow" "plot_ecdf"
#> [25] "rownames" "select_partition_number" "show"
#> [28] "suggest_best_k" "test_to_known_factors"
collect_plots() function collects all the plots made from res for all k (number of partitions)
into one single page to provide an easy and fast comparison between different k.
collect_plots(res)
The plots are:
k and the heatmap of
predicted classes for each k.k.k.k.All the plots in panels can be made by individual functions and they are plotted later in this section.
select_partition_number() produces several plots showing different
statistics for choosing “optimized” k. There are following statistics:
k;k, the area increased is defined as \(A_k - A_{k-1}\).The detailed explanations of these statistics can be found in the cola vignette.
Generally speaking, lower PAC score, higher mean silhouette score or higher
concordance corresponds to better partition. Rand index and Jaccard index
measure how similar the current partition is compared to partition with k-1.
If they are too similar, we won't accept k is better than k-1.
select_partition_number(res)
The numeric values for all these statistics can be obtained by get_stats().
get_stats(res)
#> k 1-PAC mean_silhouette concordance area_increased Rand Jaccard
#> 2 2 1.000 0.965 0.976 0.4972 0.495 0.495
#> 3 3 0.976 0.966 0.977 0.1960 0.899 0.800
#> 4 4 0.597 0.605 0.763 0.1931 0.829 0.604
#> 5 5 0.763 0.692 0.832 0.1012 0.805 0.429
#> 6 6 0.837 0.866 0.882 0.0536 0.923 0.658
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
#> GSM97038 2 0.0000 0.999 0.000 1.000
#> GSM97045 2 0.0000 0.999 0.000 1.000
#> GSM97047 2 0.0000 0.999 0.000 1.000
#> GSM97025 2 0.0000 0.999 0.000 1.000
#> GSM97030 2 0.0000 0.999 0.000 1.000
#> GSM97027 2 0.0000 0.999 0.000 1.000
#> GSM97033 2 0.0000 0.999 0.000 1.000
#> GSM97034 2 0.0000 0.999 0.000 1.000
#> GSM97020 2 0.0000 0.999 0.000 1.000
#> GSM97026 2 0.0000 0.999 0.000 1.000
#> GSM97012 2 0.0000 0.999 0.000 1.000
#> GSM97015 2 0.0000 0.999 0.000 1.000
#> GSM97016 2 0.0000 0.999 0.000 1.000
#> GSM97017 2 0.0000 0.999 0.000 1.000
#> GSM97019 2 0.0000 0.999 0.000 1.000
#> GSM97022 2 0.0000 0.999 0.000 1.000
#> GSM97035 2 0.0000 0.999 0.000 1.000
#> GSM97036 2 0.0000 0.999 0.000 1.000
#> GSM97039 2 0.0000 0.999 0.000 1.000
#> GSM97046 2 0.0000 0.999 0.000 1.000
#> GSM97023 1 0.2236 0.964 0.964 0.036
#> GSM97029 2 0.0000 0.999 0.000 1.000
#> GSM97043 2 0.0000 0.999 0.000 1.000
#> GSM97013 1 0.2236 0.964 0.964 0.036
#> GSM96956 2 0.0000 0.999 0.000 1.000
#> GSM97024 2 0.0000 0.999 0.000 1.000
#> GSM97032 2 0.0000 0.999 0.000 1.000
#> GSM97044 2 0.0000 0.999 0.000 1.000
#> GSM97049 2 0.0000 0.999 0.000 1.000
#> GSM96968 1 0.2236 0.964 0.964 0.036
#> GSM96971 1 0.0000 0.954 1.000 0.000
#> GSM96986 1 0.0938 0.960 0.988 0.012
#> GSM97003 1 0.0938 0.960 0.988 0.012
#> GSM96957 1 0.8499 0.675 0.724 0.276
#> GSM96960 1 0.0938 0.960 0.988 0.012
#> GSM96975 1 0.2236 0.964 0.964 0.036
#> GSM96998 1 0.2236 0.964 0.964 0.036
#> GSM96999 1 0.2236 0.964 0.964 0.036
#> GSM97001 2 0.2423 0.955 0.040 0.960
#> GSM97005 1 0.2236 0.964 0.964 0.036
#> GSM97006 1 0.0938 0.960 0.988 0.012
#> GSM97021 2 0.0000 0.999 0.000 1.000
#> GSM97028 2 0.0000 0.999 0.000 1.000
#> GSM97031 1 0.2236 0.964 0.964 0.036
#> GSM97037 2 0.0000 0.999 0.000 1.000
#> GSM97018 2 0.0000 0.999 0.000 1.000
#> GSM97014 2 0.0000 0.999 0.000 1.000
#> GSM97042 2 0.0000 0.999 0.000 1.000
#> GSM97040 2 0.0000 0.999 0.000 1.000
#> GSM97041 2 0.0000 0.999 0.000 1.000
#> GSM96955 2 0.0000 0.999 0.000 1.000
#> GSM96990 2 0.0000 0.999 0.000 1.000
#> GSM96991 2 0.0000 0.999 0.000 1.000
#> GSM97048 2 0.0000 0.999 0.000 1.000
#> GSM96963 2 0.0000 0.999 0.000 1.000
#> GSM96953 2 0.0000 0.999 0.000 1.000
#> GSM96966 1 0.0000 0.954 1.000 0.000
#> GSM96979 1 0.0938 0.960 0.988 0.012
#> GSM96983 2 0.0000 0.999 0.000 1.000
#> GSM96984 1 0.0938 0.960 0.988 0.012
#> GSM96994 2 0.0000 0.999 0.000 1.000
#> GSM96996 1 0.2423 0.961 0.960 0.040
#> GSM96997 1 0.0938 0.960 0.988 0.012
#> GSM97007 1 0.2423 0.961 0.960 0.040
#> GSM96954 1 0.2236 0.964 0.964 0.036
#> GSM96962 1 0.0938 0.960 0.988 0.012
#> GSM96969 1 0.0000 0.954 1.000 0.000
#> GSM96970 1 0.0000 0.954 1.000 0.000
#> GSM96973 1 0.0000 0.954 1.000 0.000
#> GSM96976 1 0.0000 0.954 1.000 0.000
#> GSM96977 1 0.2236 0.964 0.964 0.036
#> GSM96995 2 0.0000 0.999 0.000 1.000
#> GSM97002 1 0.2236 0.964 0.964 0.036
#> GSM97009 2 0.0000 0.999 0.000 1.000
#> GSM97010 1 0.2236 0.964 0.964 0.036
#> GSM96974 1 0.0000 0.954 1.000 0.000
#> GSM96985 1 0.2236 0.964 0.964 0.036
#> GSM96959 2 0.0000 0.999 0.000 1.000
#> GSM96972 1 0.0000 0.954 1.000 0.000
#> GSM96978 1 0.0938 0.960 0.988 0.012
#> GSM96967 1 0.0000 0.954 1.000 0.000
#> GSM96987 1 0.9087 0.590 0.676 0.324
#> GSM97011 2 0.0000 0.999 0.000 1.000
#> GSM96964 1 0.2236 0.964 0.964 0.036
#> GSM96965 1 0.0000 0.954 1.000 0.000
#> GSM96981 1 0.2236 0.964 0.964 0.036
#> GSM96982 1 0.2236 0.964 0.964 0.036
#> GSM96988 1 0.9286 0.549 0.656 0.344
#> GSM97000 1 0.2236 0.964 0.964 0.036
#> GSM97004 1 0.2236 0.964 0.964 0.036
#> GSM97008 1 0.9000 0.606 0.684 0.316
#> GSM96950 1 0.2236 0.964 0.964 0.036
#> GSM96980 1 0.0000 0.954 1.000 0.000
#> GSM96989 1 0.2236 0.964 0.964 0.036
#> GSM96992 1 0.2236 0.964 0.964 0.036
#> GSM96993 2 0.0000 0.999 0.000 1.000
#> GSM96958 1 0.2236 0.964 0.964 0.036
#> GSM96951 1 0.2236 0.964 0.964 0.036
#> GSM96952 1 0.2236 0.964 0.964 0.036
#> GSM96961 1 0.2236 0.964 0.964 0.036
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM97038 2 0.0000 0.974 0.000 1.000 0.000
#> GSM97045 2 0.0000 0.974 0.000 1.000 0.000
#> GSM97047 2 0.0000 0.974 0.000 1.000 0.000
#> GSM97025 2 0.0000 0.974 0.000 1.000 0.000
#> GSM97030 2 0.1399 0.964 0.004 0.968 0.028
#> GSM97027 2 0.0000 0.974 0.000 1.000 0.000
#> GSM97033 2 0.0000 0.974 0.000 1.000 0.000
#> GSM97034 2 0.1399 0.964 0.004 0.968 0.028
#> GSM97020 2 0.0000 0.974 0.000 1.000 0.000
#> GSM97026 2 0.0000 0.974 0.000 1.000 0.000
#> GSM97012 2 0.0000 0.974 0.000 1.000 0.000
#> GSM97015 2 0.1399 0.964 0.004 0.968 0.028
#> GSM97016 2 0.0000 0.974 0.000 1.000 0.000
#> GSM97017 2 0.2356 0.923 0.072 0.928 0.000
#> GSM97019 2 0.0000 0.974 0.000 1.000 0.000
#> GSM97022 2 0.0000 0.974 0.000 1.000 0.000
#> GSM97035 2 0.0000 0.974 0.000 1.000 0.000
#> GSM97036 2 0.2356 0.923 0.072 0.928 0.000
#> GSM97039 2 0.0000 0.974 0.000 1.000 0.000
#> GSM97046 2 0.0000 0.974 0.000 1.000 0.000
#> GSM97023 1 0.0237 0.977 0.996 0.000 0.004
#> GSM97029 2 0.2356 0.923 0.072 0.928 0.000
#> GSM97043 2 0.0000 0.974 0.000 1.000 0.000
#> GSM97013 1 0.0237 0.977 0.996 0.000 0.004
#> GSM96956 2 0.1399 0.964 0.004 0.968 0.028
#> GSM97024 2 0.0000 0.974 0.000 1.000 0.000
#> GSM97032 2 0.0892 0.969 0.000 0.980 0.020
#> GSM97044 2 0.1399 0.964 0.004 0.968 0.028
#> GSM97049 2 0.0000 0.974 0.000 1.000 0.000
#> GSM96968 1 0.1163 0.960 0.972 0.000 0.028
#> GSM96971 3 0.0892 0.991 0.020 0.000 0.980
#> GSM96986 1 0.2537 0.923 0.920 0.000 0.080
#> GSM97003 1 0.0000 0.976 1.000 0.000 0.000
#> GSM96957 1 0.0592 0.969 0.988 0.000 0.012
#> GSM96960 1 0.0237 0.977 0.996 0.000 0.004
#> GSM96975 1 0.0237 0.977 0.996 0.000 0.004
#> GSM96998 1 0.0237 0.977 0.996 0.000 0.004
#> GSM96999 1 0.0237 0.977 0.996 0.000 0.004
#> GSM97001 1 0.1163 0.955 0.972 0.028 0.000
#> GSM97005 1 0.0000 0.976 1.000 0.000 0.000
#> GSM97006 1 0.0237 0.977 0.996 0.000 0.004
#> GSM97021 2 0.2448 0.919 0.076 0.924 0.000
#> GSM97028 2 0.1399 0.964 0.004 0.968 0.028
#> GSM97031 1 0.0424 0.972 0.992 0.000 0.008
#> GSM97037 2 0.1399 0.964 0.004 0.968 0.028
#> GSM97018 2 0.0892 0.969 0.000 0.980 0.020
#> GSM97014 2 0.0000 0.974 0.000 1.000 0.000
#> GSM97042 2 0.0000 0.974 0.000 1.000 0.000
#> GSM97040 2 0.0000 0.974 0.000 1.000 0.000
#> GSM97041 2 0.2448 0.919 0.076 0.924 0.000
#> GSM96955 2 0.0000 0.974 0.000 1.000 0.000
#> GSM96990 2 0.1267 0.966 0.004 0.972 0.024
#> GSM96991 2 0.0000 0.974 0.000 1.000 0.000
#> GSM97048 2 0.0000 0.974 0.000 1.000 0.000
#> GSM96963 2 0.0000 0.974 0.000 1.000 0.000
#> GSM96953 2 0.0000 0.974 0.000 1.000 0.000
#> GSM96966 3 0.1163 0.999 0.028 0.000 0.972
#> GSM96979 1 0.2537 0.923 0.920 0.000 0.080
#> GSM96983 2 0.1399 0.964 0.004 0.968 0.028
#> GSM96984 1 0.2537 0.923 0.920 0.000 0.080
#> GSM96994 2 0.1751 0.960 0.012 0.960 0.028
#> GSM96996 1 0.0237 0.977 0.996 0.000 0.004
#> GSM96997 1 0.2537 0.923 0.920 0.000 0.080
#> GSM97007 1 0.3499 0.885 0.900 0.072 0.028
#> GSM96954 1 0.1163 0.960 0.972 0.000 0.028
#> GSM96962 1 0.2537 0.923 0.920 0.000 0.080
#> GSM96969 3 0.1163 0.999 0.028 0.000 0.972
#> GSM96970 3 0.1163 0.999 0.028 0.000 0.972
#> GSM96973 3 0.1163 0.999 0.028 0.000 0.972
#> GSM96976 3 0.1163 0.999 0.028 0.000 0.972
#> GSM96977 1 0.0000 0.976 1.000 0.000 0.000
#> GSM96995 2 0.3678 0.908 0.080 0.892 0.028
#> GSM97002 1 0.0237 0.977 0.996 0.000 0.004
#> GSM97009 2 0.3272 0.913 0.080 0.904 0.016
#> GSM97010 1 0.0000 0.976 1.000 0.000 0.000
#> GSM96974 3 0.1163 0.999 0.028 0.000 0.972
#> GSM96985 1 0.0237 0.977 0.996 0.000 0.004
#> GSM96959 2 0.1267 0.966 0.004 0.972 0.024
#> GSM96972 3 0.1163 0.999 0.028 0.000 0.972
#> GSM96978 1 0.2537 0.923 0.920 0.000 0.080
#> GSM96967 3 0.1163 0.999 0.028 0.000 0.972
#> GSM96987 1 0.0237 0.977 0.996 0.000 0.004
#> GSM97011 2 0.2448 0.919 0.076 0.924 0.000
#> GSM96964 1 0.0237 0.977 0.996 0.000 0.004
#> GSM96965 3 0.1163 0.999 0.028 0.000 0.972
#> GSM96981 1 0.0237 0.977 0.996 0.000 0.004
#> GSM96982 1 0.0237 0.977 0.996 0.000 0.004
#> GSM96988 1 0.1163 0.960 0.972 0.000 0.028
#> GSM97000 1 0.0892 0.966 0.980 0.000 0.020
#> GSM97004 1 0.0237 0.977 0.996 0.000 0.004
#> GSM97008 1 0.1163 0.960 0.972 0.000 0.028
#> GSM96950 1 0.0237 0.977 0.996 0.000 0.004
#> GSM96980 1 0.2448 0.923 0.924 0.000 0.076
#> GSM96989 1 0.0237 0.977 0.996 0.000 0.004
#> GSM96992 1 0.0237 0.977 0.996 0.000 0.004
#> GSM96993 2 0.2448 0.919 0.076 0.924 0.000
#> GSM96958 1 0.0237 0.977 0.996 0.000 0.004
#> GSM96951 1 0.0000 0.976 1.000 0.000 0.000
#> GSM96952 1 0.0237 0.977 0.996 0.000 0.004
#> GSM96961 1 0.0237 0.977 0.996 0.000 0.004
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM97038 2 0.0000 0.83366 0.000 1.000 0.000 0.000
#> GSM97045 2 0.0000 0.83366 0.000 1.000 0.000 0.000
#> GSM97047 2 0.4941 0.17792 0.000 0.564 0.436 0.000
#> GSM97025 2 0.0000 0.83366 0.000 1.000 0.000 0.000
#> GSM97030 2 0.4585 0.48905 0.000 0.668 0.332 0.000
#> GSM97027 2 0.0000 0.83366 0.000 1.000 0.000 0.000
#> GSM97033 2 0.0000 0.83366 0.000 1.000 0.000 0.000
#> GSM97034 3 0.5256 0.14974 0.012 0.392 0.596 0.000
#> GSM97020 2 0.0000 0.83366 0.000 1.000 0.000 0.000
#> GSM97026 2 0.4933 0.18263 0.000 0.568 0.432 0.000
#> GSM97012 2 0.0000 0.83366 0.000 1.000 0.000 0.000
#> GSM97015 3 0.4898 0.10633 0.000 0.416 0.584 0.000
#> GSM97016 2 0.0000 0.83366 0.000 1.000 0.000 0.000
#> GSM97017 3 0.6013 0.45222 0.064 0.312 0.624 0.000
#> GSM97019 2 0.0000 0.83366 0.000 1.000 0.000 0.000
#> GSM97022 2 0.0000 0.83366 0.000 1.000 0.000 0.000
#> GSM97035 2 0.0000 0.83366 0.000 1.000 0.000 0.000
#> GSM97036 3 0.6058 0.58209 0.180 0.136 0.684 0.000
#> GSM97039 2 0.0000 0.83366 0.000 1.000 0.000 0.000
#> GSM97046 2 0.0000 0.83366 0.000 1.000 0.000 0.000
#> GSM97023 1 0.4454 0.63007 0.692 0.000 0.308 0.000
#> GSM97029 3 0.6231 0.59139 0.184 0.148 0.668 0.000
#> GSM97043 2 0.0000 0.83366 0.000 1.000 0.000 0.000
#> GSM97013 1 0.1389 0.71012 0.952 0.000 0.048 0.000
#> GSM96956 2 0.4585 0.48926 0.000 0.668 0.332 0.000
#> GSM97024 2 0.0000 0.83366 0.000 1.000 0.000 0.000
#> GSM97032 2 0.4925 0.24986 0.000 0.572 0.428 0.000
#> GSM97044 3 0.5256 0.08319 0.012 0.392 0.596 0.000
#> GSM97049 2 0.0000 0.83366 0.000 1.000 0.000 0.000
#> GSM96968 1 0.3907 0.60460 0.768 0.000 0.232 0.000
#> GSM96971 4 0.0000 1.00000 0.000 0.000 0.000 1.000
#> GSM96986 1 0.3791 0.62693 0.796 0.000 0.200 0.004
#> GSM97003 1 0.2216 0.68050 0.908 0.000 0.092 0.000
#> GSM96957 3 0.4933 -0.00567 0.432 0.000 0.568 0.000
#> GSM96960 1 0.4040 0.66939 0.752 0.000 0.248 0.000
#> GSM96975 1 0.4877 0.47179 0.592 0.000 0.408 0.000
#> GSM96998 1 0.4250 0.65808 0.724 0.000 0.276 0.000
#> GSM96999 1 0.4304 0.65428 0.716 0.000 0.284 0.000
#> GSM97001 3 0.4477 0.34651 0.312 0.000 0.688 0.000
#> GSM97005 1 0.0336 0.70978 0.992 0.000 0.008 0.000
#> GSM97006 1 0.0469 0.71048 0.988 0.000 0.012 0.000
#> GSM97021 3 0.6224 0.59020 0.188 0.144 0.668 0.000
#> GSM97028 3 0.4888 0.11739 0.000 0.412 0.588 0.000
#> GSM97031 1 0.2011 0.68492 0.920 0.000 0.080 0.000
#> GSM97037 2 0.4989 0.19134 0.000 0.528 0.472 0.000
#> GSM97018 2 0.4992 0.11506 0.000 0.524 0.476 0.000
#> GSM97014 2 0.4925 0.19359 0.000 0.572 0.428 0.000
#> GSM97042 2 0.0000 0.83366 0.000 1.000 0.000 0.000
#> GSM97040 2 0.5586 0.06186 0.020 0.528 0.452 0.000
#> GSM97041 3 0.6013 0.57211 0.196 0.120 0.684 0.000
#> GSM96955 2 0.0469 0.82317 0.000 0.988 0.012 0.000
#> GSM96990 3 0.4888 0.11739 0.000 0.412 0.588 0.000
#> GSM96991 2 0.0000 0.83366 0.000 1.000 0.000 0.000
#> GSM97048 2 0.0000 0.83366 0.000 1.000 0.000 0.000
#> GSM96963 2 0.0000 0.83366 0.000 1.000 0.000 0.000
#> GSM96953 2 0.0000 0.83366 0.000 1.000 0.000 0.000
#> GSM96966 4 0.0000 1.00000 0.000 0.000 0.000 1.000
#> GSM96979 1 0.3791 0.62693 0.796 0.000 0.200 0.004
#> GSM96983 3 0.4877 0.11971 0.000 0.408 0.592 0.000
#> GSM96984 1 0.4053 0.60550 0.768 0.000 0.228 0.004
#> GSM96994 3 0.4933 0.25446 0.016 0.296 0.688 0.000
#> GSM96996 3 0.4907 0.04011 0.420 0.000 0.580 0.000
#> GSM96997 1 0.3791 0.62693 0.796 0.000 0.200 0.004
#> GSM97007 1 0.4948 0.29477 0.560 0.000 0.440 0.000
#> GSM96954 1 0.3907 0.60460 0.768 0.000 0.232 0.000
#> GSM96962 1 0.3982 0.61214 0.776 0.000 0.220 0.004
#> GSM96969 4 0.0000 1.00000 0.000 0.000 0.000 1.000
#> GSM96970 4 0.0000 1.00000 0.000 0.000 0.000 1.000
#> GSM96973 4 0.0000 1.00000 0.000 0.000 0.000 1.000
#> GSM96976 4 0.0000 1.00000 0.000 0.000 0.000 1.000
#> GSM96977 1 0.0000 0.70851 1.000 0.000 0.000 0.000
#> GSM96995 3 0.4638 0.49603 0.044 0.180 0.776 0.000
#> GSM97002 1 0.4304 0.65428 0.716 0.000 0.284 0.000
#> GSM97009 3 0.5932 0.59518 0.172 0.132 0.696 0.000
#> GSM97010 1 0.0469 0.71025 0.988 0.000 0.012 0.000
#> GSM96974 4 0.0000 1.00000 0.000 0.000 0.000 1.000
#> GSM96985 1 0.4898 0.45476 0.584 0.000 0.416 0.000
#> GSM96959 3 0.4790 0.17877 0.000 0.380 0.620 0.000
#> GSM96972 4 0.0000 1.00000 0.000 0.000 0.000 1.000
#> GSM96978 1 0.3907 0.60460 0.768 0.000 0.232 0.000
#> GSM96967 4 0.0000 1.00000 0.000 0.000 0.000 1.000
#> GSM96987 3 0.4907 0.04011 0.420 0.000 0.580 0.000
#> GSM97011 3 0.6224 0.59020 0.188 0.144 0.668 0.000
#> GSM96964 1 0.4331 0.65180 0.712 0.000 0.288 0.000
#> GSM96965 4 0.0000 1.00000 0.000 0.000 0.000 1.000
#> GSM96981 1 0.4877 0.47179 0.592 0.000 0.408 0.000
#> GSM96982 1 0.4331 0.65180 0.712 0.000 0.288 0.000
#> GSM96988 3 0.3610 0.48940 0.200 0.000 0.800 0.000
#> GSM97000 1 0.3610 0.62858 0.800 0.000 0.200 0.000
#> GSM97004 1 0.4331 0.65180 0.712 0.000 0.288 0.000
#> GSM97008 3 0.4454 0.35475 0.308 0.000 0.692 0.000
#> GSM96950 1 0.1389 0.71012 0.952 0.000 0.048 0.000
#> GSM96980 1 0.5452 0.66082 0.736 0.000 0.156 0.108
#> GSM96989 1 0.4877 0.47179 0.592 0.000 0.408 0.000
#> GSM96992 1 0.4331 0.65180 0.712 0.000 0.288 0.000
#> GSM96993 3 0.6013 0.57211 0.196 0.120 0.684 0.000
#> GSM96958 1 0.3219 0.69405 0.836 0.000 0.164 0.000
#> GSM96951 1 0.0000 0.70851 1.000 0.000 0.000 0.000
#> GSM96952 1 0.4331 0.65180 0.712 0.000 0.288 0.000
#> GSM96961 1 0.4331 0.65180 0.712 0.000 0.288 0.000
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM97038 2 0.0000 0.9909 0.000 1.000 0.000 0.000 0.000
#> GSM97045 2 0.0000 0.9909 0.000 1.000 0.000 0.000 0.000
#> GSM97047 3 0.5040 0.6333 0.192 0.080 0.716 0.000 0.012
#> GSM97025 2 0.0000 0.9909 0.000 1.000 0.000 0.000 0.000
#> GSM97030 3 0.4352 0.6388 0.000 0.244 0.720 0.000 0.036
#> GSM97027 2 0.0000 0.9909 0.000 1.000 0.000 0.000 0.000
#> GSM97033 2 0.0000 0.9909 0.000 1.000 0.000 0.000 0.000
#> GSM97034 3 0.2782 0.7604 0.000 0.072 0.880 0.000 0.048
#> GSM97020 2 0.0000 0.9909 0.000 1.000 0.000 0.000 0.000
#> GSM97026 3 0.6590 0.4304 0.228 0.320 0.452 0.000 0.000
#> GSM97012 2 0.0992 0.9744 0.024 0.968 0.008 0.000 0.000
#> GSM97015 3 0.2850 0.7612 0.000 0.092 0.872 0.000 0.036
#> GSM97016 2 0.0000 0.9909 0.000 1.000 0.000 0.000 0.000
#> GSM97017 1 0.5951 -0.2343 0.464 0.072 0.452 0.000 0.012
#> GSM97019 2 0.0000 0.9909 0.000 1.000 0.000 0.000 0.000
#> GSM97022 2 0.0000 0.9909 0.000 1.000 0.000 0.000 0.000
#> GSM97035 2 0.0000 0.9909 0.000 1.000 0.000 0.000 0.000
#> GSM97036 1 0.5283 -0.0668 0.540 0.028 0.420 0.000 0.012
#> GSM97039 2 0.0162 0.9894 0.004 0.996 0.000 0.000 0.000
#> GSM97046 2 0.1082 0.9734 0.028 0.964 0.008 0.000 0.000
#> GSM97023 1 0.3016 0.5877 0.848 0.000 0.020 0.000 0.132
#> GSM97029 1 0.5328 -0.1526 0.492 0.028 0.468 0.000 0.012
#> GSM97043 2 0.0000 0.9909 0.000 1.000 0.000 0.000 0.000
#> GSM97013 5 0.4054 0.7017 0.224 0.000 0.028 0.000 0.748
#> GSM96956 3 0.3810 0.7120 0.000 0.176 0.788 0.000 0.036
#> GSM97024 2 0.0000 0.9909 0.000 1.000 0.000 0.000 0.000
#> GSM97032 3 0.2605 0.7469 0.000 0.148 0.852 0.000 0.000
#> GSM97044 3 0.3051 0.7544 0.000 0.076 0.864 0.000 0.060
#> GSM97049 2 0.0162 0.9894 0.004 0.996 0.000 0.000 0.000
#> GSM96968 5 0.0992 0.8375 0.008 0.000 0.024 0.000 0.968
#> GSM96971 4 0.0771 0.9896 0.000 0.000 0.020 0.976 0.004
#> GSM96986 5 0.1043 0.8358 0.000 0.000 0.040 0.000 0.960
#> GSM97003 5 0.1800 0.8360 0.048 0.000 0.020 0.000 0.932
#> GSM96957 1 0.4254 0.4467 0.740 0.000 0.220 0.000 0.040
#> GSM96960 1 0.4009 0.4329 0.684 0.000 0.004 0.000 0.312
#> GSM96975 1 0.1168 0.6113 0.960 0.000 0.008 0.000 0.032
#> GSM96998 1 0.4184 0.4832 0.700 0.000 0.016 0.000 0.284
#> GSM96999 1 0.4384 0.4283 0.660 0.000 0.016 0.000 0.324
#> GSM97001 1 0.4026 0.4133 0.736 0.000 0.244 0.000 0.020
#> GSM97005 5 0.3321 0.7907 0.136 0.000 0.032 0.000 0.832
#> GSM97006 5 0.4835 0.4371 0.380 0.000 0.028 0.000 0.592
#> GSM97021 3 0.5330 0.1150 0.480 0.028 0.480 0.000 0.012
#> GSM97028 3 0.2676 0.7622 0.000 0.080 0.884 0.000 0.036
#> GSM97031 5 0.1725 0.8299 0.044 0.000 0.020 0.000 0.936
#> GSM97037 3 0.2959 0.7587 0.000 0.100 0.864 0.000 0.036
#> GSM97018 3 0.2280 0.7550 0.000 0.120 0.880 0.000 0.000
#> GSM97014 3 0.6898 0.4305 0.228 0.304 0.456 0.000 0.012
#> GSM97042 2 0.0162 0.9893 0.004 0.996 0.000 0.000 0.000
#> GSM97040 3 0.5409 0.5791 0.252 0.076 0.660 0.000 0.012
#> GSM97041 1 0.5230 -0.0528 0.528 0.024 0.436 0.000 0.012
#> GSM96955 2 0.1393 0.9671 0.024 0.956 0.012 0.000 0.008
#> GSM96990 3 0.2293 0.7622 0.000 0.084 0.900 0.000 0.016
#> GSM96991 2 0.1393 0.9671 0.024 0.956 0.012 0.000 0.008
#> GSM97048 2 0.0162 0.9894 0.004 0.996 0.000 0.000 0.000
#> GSM96963 2 0.1393 0.9671 0.024 0.956 0.012 0.000 0.008
#> GSM96953 2 0.0000 0.9909 0.000 1.000 0.000 0.000 0.000
#> GSM96966 4 0.0290 0.9945 0.000 0.000 0.008 0.992 0.000
#> GSM96979 5 0.1043 0.8358 0.000 0.000 0.040 0.000 0.960
#> GSM96983 3 0.2616 0.7617 0.000 0.076 0.888 0.000 0.036
#> GSM96984 5 0.1043 0.8358 0.000 0.000 0.040 0.000 0.960
#> GSM96994 3 0.2795 0.7522 0.000 0.056 0.880 0.000 0.064
#> GSM96996 1 0.0898 0.6105 0.972 0.000 0.020 0.000 0.008
#> GSM96997 5 0.1043 0.8358 0.000 0.000 0.040 0.000 0.960
#> GSM97007 3 0.4306 0.0095 0.000 0.000 0.508 0.000 0.492
#> GSM96954 5 0.0794 0.8371 0.000 0.000 0.028 0.000 0.972
#> GSM96962 5 0.1043 0.8358 0.000 0.000 0.040 0.000 0.960
#> GSM96969 4 0.0000 0.9951 0.000 0.000 0.000 1.000 0.000
#> GSM96970 4 0.0000 0.9951 0.000 0.000 0.000 1.000 0.000
#> GSM96973 4 0.0000 0.9951 0.000 0.000 0.000 1.000 0.000
#> GSM96976 4 0.0609 0.9912 0.000 0.000 0.020 0.980 0.000
#> GSM96977 5 0.3151 0.7923 0.144 0.000 0.020 0.000 0.836
#> GSM96995 3 0.2333 0.7451 0.028 0.040 0.916 0.000 0.016
#> GSM97002 1 0.3814 0.4959 0.720 0.000 0.004 0.000 0.276
#> GSM97009 3 0.5149 0.4041 0.356 0.020 0.604 0.000 0.020
#> GSM97010 5 0.3343 0.7742 0.172 0.000 0.016 0.000 0.812
#> GSM96974 4 0.0510 0.9921 0.000 0.000 0.016 0.984 0.000
#> GSM96985 1 0.2813 0.5940 0.868 0.000 0.024 0.000 0.108
#> GSM96959 3 0.2162 0.7500 0.008 0.064 0.916 0.000 0.012
#> GSM96972 4 0.0000 0.9951 0.000 0.000 0.000 1.000 0.000
#> GSM96978 5 0.1041 0.8365 0.004 0.000 0.032 0.000 0.964
#> GSM96967 4 0.0000 0.9951 0.000 0.000 0.000 1.000 0.000
#> GSM96987 1 0.0898 0.6105 0.972 0.000 0.020 0.000 0.008
#> GSM97011 3 0.5330 0.1150 0.480 0.028 0.480 0.000 0.012
#> GSM96964 1 0.4227 0.4846 0.692 0.000 0.016 0.000 0.292
#> GSM96965 4 0.0290 0.9945 0.000 0.000 0.008 0.992 0.000
#> GSM96981 1 0.0898 0.6111 0.972 0.000 0.008 0.000 0.020
#> GSM96982 1 0.3814 0.4959 0.720 0.000 0.004 0.000 0.276
#> GSM96988 3 0.3949 0.5029 0.300 0.000 0.696 0.000 0.004
#> GSM97000 5 0.0955 0.8346 0.004 0.000 0.028 0.000 0.968
#> GSM97004 1 0.3814 0.4959 0.720 0.000 0.004 0.000 0.276
#> GSM97008 1 0.5047 -0.1144 0.496 0.000 0.472 0.000 0.032
#> GSM96950 5 0.3942 0.6977 0.232 0.000 0.020 0.000 0.748
#> GSM96980 1 0.5456 0.3257 0.608 0.000 0.004 0.072 0.316
#> GSM96989 1 0.1484 0.6098 0.944 0.000 0.008 0.000 0.048
#> GSM96992 1 0.3661 0.4959 0.724 0.000 0.000 0.000 0.276
#> GSM96993 1 0.5283 -0.0593 0.540 0.028 0.420 0.000 0.012
#> GSM96958 5 0.4723 0.2316 0.448 0.000 0.016 0.000 0.536
#> GSM96951 5 0.3106 0.7943 0.140 0.000 0.020 0.000 0.840
#> GSM96952 1 0.3661 0.4959 0.724 0.000 0.000 0.000 0.276
#> GSM96961 1 0.4138 0.4894 0.708 0.000 0.016 0.000 0.276
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM97038 2 0.0520 0.972 0.008 0.984 0.008 0.000 0.000 0.000
#> GSM97045 2 0.0520 0.972 0.008 0.984 0.008 0.000 0.000 0.000
#> GSM97047 5 0.4287 0.804 0.008 0.024 0.312 0.000 0.656 0.000
#> GSM97025 2 0.0520 0.972 0.008 0.984 0.008 0.000 0.000 0.000
#> GSM97030 3 0.1765 0.829 0.000 0.096 0.904 0.000 0.000 0.000
#> GSM97027 2 0.0520 0.972 0.008 0.984 0.008 0.000 0.000 0.000
#> GSM97033 2 0.0520 0.972 0.008 0.984 0.008 0.000 0.000 0.000
#> GSM97034 3 0.0508 0.884 0.000 0.012 0.984 0.000 0.000 0.004
#> GSM97020 2 0.0520 0.972 0.008 0.984 0.008 0.000 0.000 0.000
#> GSM97026 5 0.4743 0.824 0.008 0.076 0.248 0.000 0.668 0.000
#> GSM97012 2 0.1382 0.947 0.008 0.948 0.008 0.000 0.036 0.000
#> GSM97015 3 0.0508 0.884 0.000 0.012 0.984 0.000 0.004 0.000
#> GSM97016 2 0.0405 0.972 0.004 0.988 0.008 0.000 0.000 0.000
#> GSM97017 5 0.4354 0.886 0.052 0.008 0.236 0.000 0.704 0.000
#> GSM97019 2 0.0260 0.972 0.000 0.992 0.008 0.000 0.000 0.000
#> GSM97022 2 0.0260 0.972 0.000 0.992 0.008 0.000 0.000 0.000
#> GSM97035 2 0.0405 0.971 0.000 0.988 0.008 0.000 0.004 0.000
#> GSM97036 5 0.4506 0.871 0.088 0.000 0.204 0.000 0.704 0.004
#> GSM97039 2 0.1121 0.965 0.008 0.964 0.008 0.000 0.016 0.004
#> GSM97046 2 0.2100 0.935 0.024 0.916 0.008 0.000 0.048 0.004
#> GSM97023 1 0.2531 0.861 0.856 0.000 0.000 0.000 0.132 0.012
#> GSM97029 5 0.4248 0.887 0.052 0.004 0.236 0.000 0.708 0.000
#> GSM97043 2 0.0520 0.972 0.008 0.984 0.008 0.000 0.000 0.000
#> GSM97013 6 0.4821 0.743 0.184 0.000 0.000 0.000 0.148 0.668
#> GSM96956 3 0.1765 0.830 0.000 0.096 0.904 0.000 0.000 0.000
#> GSM97024 2 0.0520 0.972 0.008 0.984 0.008 0.000 0.000 0.000
#> GSM97032 3 0.0858 0.881 0.000 0.028 0.968 0.000 0.004 0.000
#> GSM97044 3 0.1333 0.871 0.000 0.008 0.944 0.000 0.000 0.048
#> GSM97049 2 0.1223 0.965 0.016 0.960 0.008 0.000 0.012 0.004
#> GSM96968 6 0.2262 0.851 0.008 0.000 0.016 0.000 0.080 0.896
#> GSM96971 4 0.1340 0.978 0.000 0.000 0.008 0.948 0.040 0.004
#> GSM96986 6 0.0508 0.847 0.000 0.000 0.012 0.000 0.004 0.984
#> GSM97003 6 0.1313 0.852 0.028 0.000 0.004 0.000 0.016 0.952
#> GSM96957 5 0.3050 0.690 0.136 0.000 0.028 0.000 0.832 0.004
#> GSM96960 1 0.1471 0.881 0.932 0.000 0.000 0.000 0.004 0.064
#> GSM96975 1 0.2416 0.849 0.844 0.000 0.000 0.000 0.156 0.000
#> GSM96998 1 0.2794 0.857 0.860 0.000 0.000 0.000 0.080 0.060
#> GSM96999 1 0.3700 0.805 0.780 0.000 0.000 0.000 0.152 0.068
#> GSM97001 5 0.3017 0.750 0.108 0.000 0.052 0.000 0.840 0.000
#> GSM97005 6 0.4634 0.767 0.164 0.000 0.000 0.000 0.144 0.692
#> GSM97006 1 0.4200 0.697 0.720 0.000 0.000 0.000 0.072 0.208
#> GSM97021 5 0.4134 0.886 0.052 0.000 0.240 0.000 0.708 0.000
#> GSM97028 3 0.0767 0.884 0.004 0.012 0.976 0.000 0.008 0.000
#> GSM97031 6 0.3293 0.831 0.048 0.000 0.000 0.000 0.140 0.812
#> GSM97037 3 0.0790 0.881 0.000 0.032 0.968 0.000 0.000 0.000
#> GSM97018 3 0.0951 0.880 0.004 0.020 0.968 0.000 0.008 0.000
#> GSM97014 5 0.4671 0.832 0.008 0.072 0.244 0.000 0.676 0.000
#> GSM97042 2 0.0405 0.966 0.000 0.988 0.008 0.000 0.004 0.000
#> GSM97040 5 0.4029 0.845 0.012 0.012 0.288 0.000 0.688 0.000
#> GSM97041 5 0.4066 0.876 0.064 0.000 0.204 0.000 0.732 0.000
#> GSM96955 2 0.2804 0.903 0.036 0.876 0.008 0.000 0.072 0.008
#> GSM96990 3 0.0622 0.882 0.000 0.012 0.980 0.000 0.008 0.000
#> GSM96991 2 0.2804 0.903 0.036 0.876 0.008 0.000 0.072 0.008
#> GSM97048 2 0.1121 0.965 0.008 0.964 0.008 0.000 0.016 0.004
#> GSM96963 2 0.2804 0.903 0.036 0.876 0.008 0.000 0.072 0.008
#> GSM96953 2 0.0260 0.972 0.000 0.992 0.008 0.000 0.000 0.000
#> GSM96966 4 0.0937 0.982 0.000 0.000 0.000 0.960 0.040 0.000
#> GSM96979 6 0.0508 0.847 0.000 0.000 0.012 0.000 0.004 0.984
#> GSM96983 3 0.1579 0.877 0.008 0.004 0.944 0.000 0.024 0.020
#> GSM96984 6 0.0508 0.847 0.000 0.000 0.012 0.000 0.004 0.984
#> GSM96994 3 0.1836 0.866 0.008 0.004 0.928 0.000 0.012 0.048
#> GSM96996 1 0.2048 0.843 0.880 0.000 0.000 0.000 0.120 0.000
#> GSM96997 6 0.0508 0.847 0.000 0.000 0.012 0.000 0.004 0.984
#> GSM97007 3 0.4100 0.440 0.004 0.000 0.600 0.000 0.008 0.388
#> GSM96954 6 0.0820 0.850 0.000 0.000 0.012 0.000 0.016 0.972
#> GSM96962 6 0.0508 0.847 0.000 0.000 0.012 0.000 0.004 0.984
#> GSM96969 4 0.0713 0.982 0.000 0.000 0.000 0.972 0.028 0.000
#> GSM96970 4 0.0000 0.989 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM96973 4 0.0000 0.989 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM96976 4 0.0622 0.987 0.000 0.000 0.008 0.980 0.012 0.000
#> GSM96977 6 0.4638 0.768 0.156 0.000 0.000 0.000 0.152 0.692
#> GSM96995 3 0.0858 0.867 0.004 0.000 0.968 0.000 0.028 0.000
#> GSM97002 1 0.1625 0.883 0.928 0.000 0.000 0.000 0.012 0.060
#> GSM97009 5 0.3855 0.867 0.024 0.000 0.272 0.000 0.704 0.000
#> GSM97010 6 0.4736 0.756 0.164 0.000 0.000 0.000 0.156 0.680
#> GSM96974 4 0.0622 0.987 0.000 0.000 0.008 0.980 0.012 0.000
#> GSM96985 1 0.2445 0.839 0.872 0.000 0.000 0.000 0.108 0.020
#> GSM96959 3 0.3384 0.522 0.004 0.008 0.760 0.000 0.228 0.000
#> GSM96972 4 0.0000 0.989 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM96978 6 0.1605 0.847 0.012 0.000 0.016 0.000 0.032 0.940
#> GSM96967 4 0.0000 0.989 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM96987 1 0.2003 0.844 0.884 0.000 0.000 0.000 0.116 0.000
#> GSM97011 5 0.4167 0.887 0.056 0.000 0.236 0.000 0.708 0.000
#> GSM96964 1 0.3475 0.835 0.800 0.000 0.000 0.000 0.140 0.060
#> GSM96965 4 0.0363 0.988 0.000 0.000 0.000 0.988 0.012 0.000
#> GSM96981 1 0.1910 0.857 0.892 0.000 0.000 0.000 0.108 0.000
#> GSM96982 1 0.2046 0.882 0.908 0.000 0.000 0.000 0.032 0.060
#> GSM96988 3 0.3992 0.649 0.180 0.000 0.748 0.000 0.072 0.000
#> GSM97000 6 0.2766 0.841 0.008 0.000 0.008 0.000 0.140 0.844
#> GSM97004 1 0.1524 0.883 0.932 0.000 0.000 0.000 0.008 0.060
#> GSM97008 5 0.3089 0.777 0.060 0.000 0.092 0.000 0.844 0.004
#> GSM96950 6 0.4825 0.744 0.180 0.000 0.000 0.000 0.152 0.668
#> GSM96980 1 0.2627 0.866 0.884 0.000 0.000 0.016 0.036 0.064
#> GSM96989 1 0.1714 0.861 0.908 0.000 0.000 0.000 0.092 0.000
#> GSM96992 1 0.1267 0.883 0.940 0.000 0.000 0.000 0.000 0.060
#> GSM96993 5 0.4281 0.884 0.068 0.000 0.228 0.000 0.704 0.000
#> GSM96958 1 0.5574 0.176 0.504 0.000 0.000 0.000 0.152 0.344
#> GSM96951 6 0.4602 0.770 0.160 0.000 0.000 0.000 0.144 0.696
#> GSM96952 1 0.1267 0.883 0.940 0.000 0.000 0.000 0.000 0.060
#> GSM96961 1 0.2571 0.865 0.876 0.000 0.000 0.000 0.064 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 disease.state(p) specimen(p) cell.type(p) other(p) k
#> ATC:kmeans 100 1.24e-07 0.714 5.25e-14 0.102 2
#> ATC:kmeans 100 1.01e-06 0.230 3.56e-15 0.224 3
#> ATC:kmeans 70 6.91e-05 0.298 8.78e-14 0.251 4
#> ATC:kmeans 73 5.03e-04 0.427 5.55e-11 0.106 5
#> ATC:kmeans 98 8.19e-05 0.582 5.47e-16 0.476 6
If matrix rows can be associated to genes, consider to use functional_enrichment(res,
...) to perform function enrichment for the signature genes. See this vignette for more detailed explanations.
The object with results only for a single top-value method and a single partition method can be extracted as:
res = res_list["ATC", "skmeans"]
# you can also extract it by
# res = res_list["ATC:skmeans"]
A summary of res and all the functions that can be applied to it:
res
#> A 'ConsensusPartition' object with k = 2, 3, 4, 5, 6.
#> On a matrix with 21168 rows and 100 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 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 0.983 0.993 0.5053 0.495 0.495
#> 3 3 0.950 0.912 0.964 0.2883 0.822 0.652
#> 4 4 0.719 0.764 0.788 0.1213 0.862 0.632
#> 5 5 0.783 0.773 0.816 0.0712 0.937 0.769
#> 6 6 0.962 0.925 0.947 0.0600 0.924 0.672
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
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
#> GSM97038 2 0.0000 0.993 0.000 1.000
#> GSM97045 2 0.0000 0.993 0.000 1.000
#> GSM97047 2 0.0000 0.993 0.000 1.000
#> GSM97025 2 0.0000 0.993 0.000 1.000
#> GSM97030 2 0.0000 0.993 0.000 1.000
#> GSM97027 2 0.0000 0.993 0.000 1.000
#> GSM97033 2 0.0000 0.993 0.000 1.000
#> GSM97034 2 0.0000 0.993 0.000 1.000
#> GSM97020 2 0.0000 0.993 0.000 1.000
#> GSM97026 2 0.0000 0.993 0.000 1.000
#> GSM97012 2 0.0000 0.993 0.000 1.000
#> GSM97015 2 0.0000 0.993 0.000 1.000
#> GSM97016 2 0.0000 0.993 0.000 1.000
#> GSM97017 2 0.0000 0.993 0.000 1.000
#> GSM97019 2 0.0000 0.993 0.000 1.000
#> GSM97022 2 0.0000 0.993 0.000 1.000
#> GSM97035 2 0.0000 0.993 0.000 1.000
#> GSM97036 2 0.0000 0.993 0.000 1.000
#> GSM97039 2 0.0000 0.993 0.000 1.000
#> GSM97046 2 0.0000 0.993 0.000 1.000
#> GSM97023 1 0.0000 0.992 1.000 0.000
#> GSM97029 2 0.0000 0.993 0.000 1.000
#> GSM97043 2 0.0000 0.993 0.000 1.000
#> GSM97013 1 0.0000 0.992 1.000 0.000
#> GSM96956 2 0.0000 0.993 0.000 1.000
#> GSM97024 2 0.0000 0.993 0.000 1.000
#> GSM97032 2 0.0000 0.993 0.000 1.000
#> GSM97044 2 0.0000 0.993 0.000 1.000
#> GSM97049 2 0.0000 0.993 0.000 1.000
#> GSM96968 1 0.0000 0.992 1.000 0.000
#> GSM96971 1 0.0000 0.992 1.000 0.000
#> GSM96986 1 0.0000 0.992 1.000 0.000
#> GSM97003 1 0.0000 0.992 1.000 0.000
#> GSM96957 1 0.0000 0.992 1.000 0.000
#> GSM96960 1 0.0000 0.992 1.000 0.000
#> GSM96975 1 0.0000 0.992 1.000 0.000
#> GSM96998 1 0.0000 0.992 1.000 0.000
#> GSM96999 1 0.0000 0.992 1.000 0.000
#> GSM97001 2 0.9044 0.527 0.320 0.680
#> GSM97005 1 0.0000 0.992 1.000 0.000
#> GSM97006 1 0.0000 0.992 1.000 0.000
#> GSM97021 2 0.0000 0.993 0.000 1.000
#> GSM97028 2 0.0000 0.993 0.000 1.000
#> GSM97031 1 0.0000 0.992 1.000 0.000
#> GSM97037 2 0.0000 0.993 0.000 1.000
#> GSM97018 2 0.0000 0.993 0.000 1.000
#> GSM97014 2 0.0000 0.993 0.000 1.000
#> GSM97042 2 0.0000 0.993 0.000 1.000
#> GSM97040 2 0.0000 0.993 0.000 1.000
#> GSM97041 2 0.0000 0.993 0.000 1.000
#> GSM96955 2 0.0000 0.993 0.000 1.000
#> GSM96990 2 0.0000 0.993 0.000 1.000
#> GSM96991 2 0.0000 0.993 0.000 1.000
#> GSM97048 2 0.0000 0.993 0.000 1.000
#> GSM96963 2 0.0000 0.993 0.000 1.000
#> GSM96953 2 0.0000 0.993 0.000 1.000
#> GSM96966 1 0.0000 0.992 1.000 0.000
#> GSM96979 1 0.0000 0.992 1.000 0.000
#> GSM96983 2 0.0000 0.993 0.000 1.000
#> GSM96984 1 0.0000 0.992 1.000 0.000
#> GSM96994 2 0.0000 0.993 0.000 1.000
#> GSM96996 1 0.0000 0.992 1.000 0.000
#> GSM96997 1 0.0000 0.992 1.000 0.000
#> GSM97007 1 0.3114 0.936 0.944 0.056
#> GSM96954 1 0.0000 0.992 1.000 0.000
#> GSM96962 1 0.0000 0.992 1.000 0.000
#> GSM96969 1 0.0000 0.992 1.000 0.000
#> GSM96970 1 0.0000 0.992 1.000 0.000
#> GSM96973 1 0.0000 0.992 1.000 0.000
#> GSM96976 1 0.0000 0.992 1.000 0.000
#> GSM96977 1 0.0000 0.992 1.000 0.000
#> GSM96995 2 0.0000 0.993 0.000 1.000
#> GSM97002 1 0.0000 0.992 1.000 0.000
#> GSM97009 2 0.0000 0.993 0.000 1.000
#> GSM97010 1 0.0000 0.992 1.000 0.000
#> GSM96974 1 0.0000 0.992 1.000 0.000
#> GSM96985 1 0.0000 0.992 1.000 0.000
#> GSM96959 2 0.0000 0.993 0.000 1.000
#> GSM96972 1 0.0000 0.992 1.000 0.000
#> GSM96978 1 0.0000 0.992 1.000 0.000
#> GSM96967 1 0.0000 0.992 1.000 0.000
#> GSM96987 1 0.0938 0.981 0.988 0.012
#> GSM97011 2 0.0000 0.993 0.000 1.000
#> GSM96964 1 0.0000 0.992 1.000 0.000
#> GSM96965 1 0.0000 0.992 1.000 0.000
#> GSM96981 1 0.0000 0.992 1.000 0.000
#> GSM96982 1 0.0000 0.992 1.000 0.000
#> GSM96988 1 0.9000 0.537 0.684 0.316
#> GSM97000 1 0.0000 0.992 1.000 0.000
#> GSM97004 1 0.0000 0.992 1.000 0.000
#> GSM97008 1 0.0000 0.992 1.000 0.000
#> GSM96950 1 0.0000 0.992 1.000 0.000
#> GSM96980 1 0.0000 0.992 1.000 0.000
#> GSM96989 1 0.0000 0.992 1.000 0.000
#> GSM96992 1 0.0000 0.992 1.000 0.000
#> GSM96993 2 0.0000 0.993 0.000 1.000
#> GSM96958 1 0.0000 0.992 1.000 0.000
#> GSM96951 1 0.0000 0.992 1.000 0.000
#> GSM96952 1 0.0000 0.992 1.000 0.000
#> GSM96961 1 0.0000 0.992 1.000 0.000
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM97038 2 0.0000 0.962 0.000 1.000 0.000
#> GSM97045 2 0.0000 0.962 0.000 1.000 0.000
#> GSM97047 2 0.0000 0.962 0.000 1.000 0.000
#> GSM97025 2 0.0000 0.962 0.000 1.000 0.000
#> GSM97030 2 0.0000 0.962 0.000 1.000 0.000
#> GSM97027 2 0.0000 0.962 0.000 1.000 0.000
#> GSM97033 2 0.0000 0.962 0.000 1.000 0.000
#> GSM97034 2 0.0000 0.962 0.000 1.000 0.000
#> GSM97020 2 0.0000 0.962 0.000 1.000 0.000
#> GSM97026 2 0.0000 0.962 0.000 1.000 0.000
#> GSM97012 2 0.0000 0.962 0.000 1.000 0.000
#> GSM97015 2 0.0000 0.962 0.000 1.000 0.000
#> GSM97016 2 0.0000 0.962 0.000 1.000 0.000
#> GSM97017 2 0.6008 0.396 0.372 0.628 0.000
#> GSM97019 2 0.0000 0.962 0.000 1.000 0.000
#> GSM97022 2 0.0000 0.962 0.000 1.000 0.000
#> GSM97035 2 0.0000 0.962 0.000 1.000 0.000
#> GSM97036 1 0.5948 0.436 0.640 0.360 0.000
#> GSM97039 2 0.0000 0.962 0.000 1.000 0.000
#> GSM97046 2 0.0000 0.962 0.000 1.000 0.000
#> GSM97023 1 0.0237 0.930 0.996 0.000 0.004
#> GSM97029 2 0.6008 0.396 0.372 0.628 0.000
#> GSM97043 2 0.0000 0.962 0.000 1.000 0.000
#> GSM97013 3 0.1860 0.946 0.052 0.000 0.948
#> GSM96956 2 0.0000 0.962 0.000 1.000 0.000
#> GSM97024 2 0.0000 0.962 0.000 1.000 0.000
#> GSM97032 2 0.0000 0.962 0.000 1.000 0.000
#> GSM97044 2 0.0000 0.962 0.000 1.000 0.000
#> GSM97049 2 0.0000 0.962 0.000 1.000 0.000
#> GSM96968 3 0.0000 0.977 0.000 0.000 1.000
#> GSM96971 3 0.0237 0.976 0.004 0.000 0.996
#> GSM96986 3 0.0000 0.977 0.000 0.000 1.000
#> GSM97003 3 0.0000 0.977 0.000 0.000 1.000
#> GSM96957 1 0.0237 0.930 0.996 0.000 0.004
#> GSM96960 1 0.0237 0.930 0.996 0.000 0.004
#> GSM96975 1 0.0000 0.930 1.000 0.000 0.000
#> GSM96998 1 0.0237 0.930 0.996 0.000 0.004
#> GSM96999 1 0.5178 0.597 0.744 0.000 0.256
#> GSM97001 1 0.0237 0.930 0.996 0.000 0.004
#> GSM97005 3 0.1860 0.946 0.052 0.000 0.948
#> GSM97006 3 0.1860 0.946 0.052 0.000 0.948
#> GSM97021 2 0.6045 0.375 0.380 0.620 0.000
#> GSM97028 2 0.0000 0.962 0.000 1.000 0.000
#> GSM97031 3 0.2448 0.924 0.076 0.000 0.924
#> GSM97037 2 0.0000 0.962 0.000 1.000 0.000
#> GSM97018 2 0.0000 0.962 0.000 1.000 0.000
#> GSM97014 2 0.0000 0.962 0.000 1.000 0.000
#> GSM97042 2 0.0000 0.962 0.000 1.000 0.000
#> GSM97040 2 0.0000 0.962 0.000 1.000 0.000
#> GSM97041 1 0.5926 0.446 0.644 0.356 0.000
#> GSM96955 2 0.0000 0.962 0.000 1.000 0.000
#> GSM96990 2 0.0000 0.962 0.000 1.000 0.000
#> GSM96991 2 0.0000 0.962 0.000 1.000 0.000
#> GSM97048 2 0.0000 0.962 0.000 1.000 0.000
#> GSM96963 2 0.0000 0.962 0.000 1.000 0.000
#> GSM96953 2 0.0000 0.962 0.000 1.000 0.000
#> GSM96966 3 0.0237 0.976 0.004 0.000 0.996
#> GSM96979 3 0.0000 0.977 0.000 0.000 1.000
#> GSM96983 2 0.0000 0.962 0.000 1.000 0.000
#> GSM96984 3 0.0000 0.977 0.000 0.000 1.000
#> GSM96994 2 0.0592 0.951 0.000 0.988 0.012
#> GSM96996 1 0.0000 0.930 1.000 0.000 0.000
#> GSM96997 3 0.0000 0.977 0.000 0.000 1.000
#> GSM97007 3 0.0000 0.977 0.000 0.000 1.000
#> GSM96954 3 0.0000 0.977 0.000 0.000 1.000
#> GSM96962 3 0.0000 0.977 0.000 0.000 1.000
#> GSM96969 3 0.0237 0.976 0.004 0.000 0.996
#> GSM96970 3 0.0237 0.976 0.004 0.000 0.996
#> GSM96973 3 0.0237 0.976 0.004 0.000 0.996
#> GSM96976 3 0.0237 0.976 0.004 0.000 0.996
#> GSM96977 3 0.0000 0.977 0.000 0.000 1.000
#> GSM96995 2 0.0000 0.962 0.000 1.000 0.000
#> GSM97002 1 0.0000 0.930 1.000 0.000 0.000
#> GSM97009 2 0.0424 0.955 0.008 0.992 0.000
#> GSM97010 3 0.0000 0.977 0.000 0.000 1.000
#> GSM96974 3 0.0237 0.976 0.004 0.000 0.996
#> GSM96985 1 0.0424 0.925 0.992 0.000 0.008
#> GSM96959 2 0.0000 0.962 0.000 1.000 0.000
#> GSM96972 3 0.0237 0.976 0.004 0.000 0.996
#> GSM96978 3 0.0237 0.976 0.004 0.000 0.996
#> GSM96967 3 0.0237 0.976 0.004 0.000 0.996
#> GSM96987 1 0.0000 0.930 1.000 0.000 0.000
#> GSM97011 2 0.6008 0.396 0.372 0.628 0.000
#> GSM96964 1 0.0237 0.930 0.996 0.000 0.004
#> GSM96965 3 0.0237 0.976 0.004 0.000 0.996
#> GSM96981 1 0.0000 0.930 1.000 0.000 0.000
#> GSM96982 1 0.0000 0.930 1.000 0.000 0.000
#> GSM96988 1 0.0237 0.928 0.996 0.004 0.000
#> GSM97000 3 0.0000 0.977 0.000 0.000 1.000
#> GSM97004 1 0.0000 0.930 1.000 0.000 0.000
#> GSM97008 1 0.0237 0.930 0.996 0.000 0.004
#> GSM96950 3 0.1860 0.946 0.052 0.000 0.948
#> GSM96980 3 0.0592 0.973 0.012 0.000 0.988
#> GSM96989 1 0.0000 0.930 1.000 0.000 0.000
#> GSM96992 1 0.0237 0.930 0.996 0.000 0.004
#> GSM96993 1 0.5926 0.445 0.644 0.356 0.000
#> GSM96958 3 0.5621 0.588 0.308 0.000 0.692
#> GSM96951 3 0.1860 0.946 0.052 0.000 0.948
#> GSM96952 1 0.0237 0.930 0.996 0.000 0.004
#> GSM96961 1 0.0237 0.930 0.996 0.000 0.004
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM97038 2 0.3837 0.8043 0.000 0.776 0.224 0.000
#> GSM97045 2 0.3837 0.8043 0.000 0.776 0.224 0.000
#> GSM97047 2 0.2469 0.7313 0.000 0.892 0.108 0.000
#> GSM97025 2 0.3837 0.8043 0.000 0.776 0.224 0.000
#> GSM97030 3 0.3688 0.9372 0.000 0.208 0.792 0.000
#> GSM97027 2 0.3837 0.8043 0.000 0.776 0.224 0.000
#> GSM97033 2 0.3837 0.8043 0.000 0.776 0.224 0.000
#> GSM97034 3 0.3688 0.9372 0.000 0.208 0.792 0.000
#> GSM97020 2 0.3837 0.8043 0.000 0.776 0.224 0.000
#> GSM97026 2 0.3764 0.7997 0.000 0.784 0.216 0.000
#> GSM97012 2 0.3837 0.8043 0.000 0.776 0.224 0.000
#> GSM97015 3 0.3688 0.9372 0.000 0.208 0.792 0.000
#> GSM97016 2 0.3837 0.8043 0.000 0.776 0.224 0.000
#> GSM97017 2 0.3123 0.5484 0.156 0.844 0.000 0.000
#> GSM97019 2 0.3837 0.8043 0.000 0.776 0.224 0.000
#> GSM97022 2 0.3837 0.8043 0.000 0.776 0.224 0.000
#> GSM97035 2 0.3837 0.8043 0.000 0.776 0.224 0.000
#> GSM97036 2 0.4103 0.4314 0.256 0.744 0.000 0.000
#> GSM97039 2 0.3837 0.8043 0.000 0.776 0.224 0.000
#> GSM97046 2 0.3837 0.8043 0.000 0.776 0.224 0.000
#> GSM97023 1 0.0188 0.8888 0.996 0.004 0.000 0.000
#> GSM97029 2 0.3123 0.5484 0.156 0.844 0.000 0.000
#> GSM97043 2 0.3837 0.8043 0.000 0.776 0.224 0.000
#> GSM97013 4 0.7568 0.5755 0.280 0.004 0.208 0.508
#> GSM96956 3 0.3688 0.9372 0.000 0.208 0.792 0.000
#> GSM97024 2 0.3837 0.8043 0.000 0.776 0.224 0.000
#> GSM97032 3 0.3688 0.9372 0.000 0.208 0.792 0.000
#> GSM97044 3 0.3688 0.9372 0.000 0.208 0.792 0.000
#> GSM97049 2 0.3837 0.8043 0.000 0.776 0.224 0.000
#> GSM96968 4 0.4250 0.7931 0.000 0.000 0.276 0.724
#> GSM96971 4 0.0000 0.8102 0.000 0.000 0.000 1.000
#> GSM96986 4 0.3688 0.8203 0.000 0.000 0.208 0.792
#> GSM97003 4 0.3688 0.8203 0.000 0.000 0.208 0.792
#> GSM96957 1 0.3400 0.7779 0.820 0.180 0.000 0.000
#> GSM96960 1 0.0000 0.8904 1.000 0.000 0.000 0.000
#> GSM96975 1 0.0000 0.8904 1.000 0.000 0.000 0.000
#> GSM96998 1 0.0000 0.8904 1.000 0.000 0.000 0.000
#> GSM96999 1 0.4055 0.7368 0.832 0.000 0.108 0.060
#> GSM97001 1 0.3837 0.7439 0.776 0.224 0.000 0.000
#> GSM97005 4 0.7568 0.5755 0.280 0.004 0.208 0.508
#> GSM97006 4 0.7625 0.4326 0.360 0.000 0.208 0.432
#> GSM97021 2 0.3172 0.5452 0.160 0.840 0.000 0.000
#> GSM97028 3 0.3688 0.9372 0.000 0.208 0.792 0.000
#> GSM97031 4 0.7613 0.4425 0.352 0.000 0.208 0.440
#> GSM97037 3 0.3688 0.9372 0.000 0.208 0.792 0.000
#> GSM97018 3 0.3688 0.9372 0.000 0.208 0.792 0.000
#> GSM97014 2 0.0188 0.6656 0.000 0.996 0.004 0.000
#> GSM97042 2 0.3837 0.8043 0.000 0.776 0.224 0.000
#> GSM97040 2 0.0000 0.6629 0.000 1.000 0.000 0.000
#> GSM97041 2 0.4382 0.3499 0.296 0.704 0.000 0.000
#> GSM96955 2 0.3837 0.8043 0.000 0.776 0.224 0.000
#> GSM96990 3 0.3688 0.9372 0.000 0.208 0.792 0.000
#> GSM96991 2 0.3837 0.8043 0.000 0.776 0.224 0.000
#> GSM97048 2 0.3837 0.8043 0.000 0.776 0.224 0.000
#> GSM96963 2 0.3837 0.8043 0.000 0.776 0.224 0.000
#> GSM96953 2 0.3837 0.8043 0.000 0.776 0.224 0.000
#> GSM96966 4 0.0000 0.8102 0.000 0.000 0.000 1.000
#> GSM96979 4 0.3688 0.8203 0.000 0.000 0.208 0.792
#> GSM96983 3 0.3688 0.9372 0.000 0.208 0.792 0.000
#> GSM96984 4 0.3688 0.8203 0.000 0.000 0.208 0.792
#> GSM96994 3 0.3688 0.9372 0.000 0.208 0.792 0.000
#> GSM96996 1 0.0188 0.8888 0.996 0.004 0.000 0.000
#> GSM96997 4 0.3688 0.8203 0.000 0.000 0.208 0.792
#> GSM97007 3 0.3837 0.0573 0.000 0.000 0.776 0.224
#> GSM96954 4 0.4250 0.7931 0.000 0.000 0.276 0.724
#> GSM96962 4 0.3688 0.8203 0.000 0.000 0.208 0.792
#> GSM96969 4 0.0000 0.8102 0.000 0.000 0.000 1.000
#> GSM96970 4 0.0000 0.8102 0.000 0.000 0.000 1.000
#> GSM96973 4 0.0000 0.8102 0.000 0.000 0.000 1.000
#> GSM96976 4 0.0000 0.8102 0.000 0.000 0.000 1.000
#> GSM96977 4 0.3688 0.8203 0.000 0.000 0.208 0.792
#> GSM96995 3 0.3873 0.9131 0.000 0.228 0.772 0.000
#> GSM97002 1 0.0000 0.8904 1.000 0.000 0.000 0.000
#> GSM97009 2 0.0469 0.6552 0.012 0.988 0.000 0.000
#> GSM97010 4 0.3688 0.8203 0.000 0.000 0.208 0.792
#> GSM96974 4 0.0000 0.8102 0.000 0.000 0.000 1.000
#> GSM96985 1 0.4776 0.4285 0.624 0.000 0.000 0.376
#> GSM96959 3 0.3942 0.9055 0.000 0.236 0.764 0.000
#> GSM96972 4 0.0000 0.8102 0.000 0.000 0.000 1.000
#> GSM96978 4 0.0188 0.8107 0.000 0.000 0.004 0.996
#> GSM96967 4 0.0000 0.8102 0.000 0.000 0.000 1.000
#> GSM96987 1 0.0188 0.8888 0.996 0.004 0.000 0.000
#> GSM97011 2 0.3172 0.5452 0.160 0.840 0.000 0.000
#> GSM96964 1 0.0000 0.8904 1.000 0.000 0.000 0.000
#> GSM96965 4 0.0000 0.8102 0.000 0.000 0.000 1.000
#> GSM96981 1 0.0000 0.8904 1.000 0.000 0.000 0.000
#> GSM96982 1 0.0000 0.8904 1.000 0.000 0.000 0.000
#> GSM96988 1 0.4989 0.1995 0.528 0.000 0.472 0.000
#> GSM97000 4 0.6678 0.7022 0.172 0.000 0.208 0.620
#> GSM97004 1 0.0000 0.8904 1.000 0.000 0.000 0.000
#> GSM97008 1 0.6296 0.6568 0.652 0.224 0.124 0.000
#> GSM96950 4 0.7399 0.5790 0.280 0.000 0.208 0.512
#> GSM96980 4 0.1792 0.7714 0.068 0.000 0.000 0.932
#> GSM96989 1 0.0000 0.8904 1.000 0.000 0.000 0.000
#> GSM96992 1 0.0000 0.8904 1.000 0.000 0.000 0.000
#> GSM96993 2 0.4431 0.3320 0.304 0.696 0.000 0.000
#> GSM96958 1 0.7080 0.1882 0.568 0.000 0.196 0.236
#> GSM96951 4 0.7399 0.5790 0.280 0.000 0.208 0.512
#> GSM96952 1 0.0000 0.8904 1.000 0.000 0.000 0.000
#> GSM96961 1 0.0000 0.8904 1.000 0.000 0.000 0.000
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM97038 2 0.3684 0.76538 0.000 0.720 0.280 0.000 0.000
#> GSM97045 2 0.3684 0.76538 0.000 0.720 0.280 0.000 0.000
#> GSM97047 2 0.4117 0.60496 0.000 0.788 0.096 0.116 0.000
#> GSM97025 2 0.3684 0.76538 0.000 0.720 0.280 0.000 0.000
#> GSM97030 3 0.0404 0.90733 0.000 0.012 0.988 0.000 0.000
#> GSM97027 2 0.3684 0.76538 0.000 0.720 0.280 0.000 0.000
#> GSM97033 2 0.3684 0.76538 0.000 0.720 0.280 0.000 0.000
#> GSM97034 3 0.0000 0.91075 0.000 0.000 1.000 0.000 0.000
#> GSM97020 2 0.3684 0.76538 0.000 0.720 0.280 0.000 0.000
#> GSM97026 2 0.3534 0.75292 0.000 0.744 0.256 0.000 0.000
#> GSM97012 2 0.3684 0.76538 0.000 0.720 0.280 0.000 0.000
#> GSM97015 3 0.0000 0.91075 0.000 0.000 1.000 0.000 0.000
#> GSM97016 2 0.3684 0.76538 0.000 0.720 0.280 0.000 0.000
#> GSM97017 2 0.4270 0.48637 0.048 0.748 0.000 0.204 0.000
#> GSM97019 2 0.3684 0.76538 0.000 0.720 0.280 0.000 0.000
#> GSM97022 2 0.3684 0.76538 0.000 0.720 0.280 0.000 0.000
#> GSM97035 2 0.3684 0.76538 0.000 0.720 0.280 0.000 0.000
#> GSM97036 2 0.6250 0.13022 0.256 0.540 0.000 0.204 0.000
#> GSM97039 2 0.3684 0.76538 0.000 0.720 0.280 0.000 0.000
#> GSM97046 2 0.3684 0.76538 0.000 0.720 0.280 0.000 0.000
#> GSM97023 1 0.0609 0.90259 0.980 0.000 0.000 0.000 0.020
#> GSM97029 2 0.4129 0.49184 0.040 0.756 0.000 0.204 0.000
#> GSM97043 2 0.3684 0.76538 0.000 0.720 0.280 0.000 0.000
#> GSM97013 5 0.1121 0.86096 0.044 0.000 0.000 0.000 0.956
#> GSM96956 3 0.0404 0.90723 0.000 0.012 0.988 0.000 0.000
#> GSM97024 2 0.3684 0.76538 0.000 0.720 0.280 0.000 0.000
#> GSM97032 3 0.0609 0.90040 0.000 0.020 0.980 0.000 0.000
#> GSM97044 3 0.0000 0.91075 0.000 0.000 1.000 0.000 0.000
#> GSM97049 2 0.3684 0.76538 0.000 0.720 0.280 0.000 0.000
#> GSM96968 5 0.0324 0.87722 0.000 0.000 0.004 0.004 0.992
#> GSM96971 4 0.3143 0.98356 0.000 0.000 0.000 0.796 0.204
#> GSM96986 5 0.0290 0.87772 0.000 0.000 0.000 0.008 0.992
#> GSM97003 5 0.0290 0.87772 0.000 0.000 0.000 0.008 0.992
#> GSM96957 1 0.6626 0.52347 0.572 0.228 0.000 0.168 0.032
#> GSM96960 1 0.1043 0.88540 0.960 0.000 0.000 0.000 0.040
#> GSM96975 1 0.0609 0.89994 0.980 0.000 0.000 0.020 0.000
#> GSM96998 1 0.0703 0.90043 0.976 0.000 0.000 0.000 0.024
#> GSM96999 1 0.2329 0.80378 0.876 0.000 0.000 0.000 0.124
#> GSM97001 1 0.6690 0.44912 0.508 0.276 0.000 0.204 0.012
#> GSM97005 5 0.0880 0.86720 0.032 0.000 0.000 0.000 0.968
#> GSM97006 5 0.4537 0.36705 0.396 0.000 0.000 0.012 0.592
#> GSM97021 2 0.4681 0.46959 0.064 0.728 0.000 0.204 0.004
#> GSM97028 3 0.0000 0.91075 0.000 0.000 1.000 0.000 0.000
#> GSM97031 5 0.1121 0.85947 0.044 0.000 0.000 0.000 0.956
#> GSM97037 3 0.0404 0.90733 0.000 0.012 0.988 0.000 0.000
#> GSM97018 3 0.0404 0.90688 0.000 0.012 0.988 0.000 0.000
#> GSM97014 2 0.2513 0.56491 0.000 0.876 0.008 0.116 0.000
#> GSM97042 2 0.3684 0.76538 0.000 0.720 0.280 0.000 0.000
#> GSM97040 2 0.3812 0.50156 0.024 0.772 0.000 0.204 0.000
#> GSM97041 2 0.6396 0.12035 0.256 0.536 0.000 0.204 0.004
#> GSM96955 2 0.3684 0.76538 0.000 0.720 0.280 0.000 0.000
#> GSM96990 3 0.0510 0.90421 0.000 0.016 0.984 0.000 0.000
#> GSM96991 2 0.3684 0.76538 0.000 0.720 0.280 0.000 0.000
#> GSM97048 2 0.3684 0.76538 0.000 0.720 0.280 0.000 0.000
#> GSM96963 2 0.3684 0.76538 0.000 0.720 0.280 0.000 0.000
#> GSM96953 2 0.3684 0.76538 0.000 0.720 0.280 0.000 0.000
#> GSM96966 4 0.3143 0.98356 0.000 0.000 0.000 0.796 0.204
#> GSM96979 5 0.0404 0.87521 0.000 0.000 0.000 0.012 0.988
#> GSM96983 3 0.0000 0.91075 0.000 0.000 1.000 0.000 0.000
#> GSM96984 5 0.0404 0.87521 0.000 0.000 0.000 0.012 0.988
#> GSM96994 3 0.0000 0.91075 0.000 0.000 1.000 0.000 0.000
#> GSM96996 1 0.0290 0.90028 0.992 0.000 0.000 0.008 0.000
#> GSM96997 5 0.0290 0.87772 0.000 0.000 0.000 0.008 0.992
#> GSM97007 3 0.3884 0.47601 0.000 0.000 0.708 0.004 0.288
#> GSM96954 5 0.0162 0.87686 0.000 0.000 0.004 0.000 0.996
#> GSM96962 5 0.0290 0.87772 0.000 0.000 0.000 0.008 0.992
#> GSM96969 4 0.3143 0.98356 0.000 0.000 0.000 0.796 0.204
#> GSM96970 4 0.3143 0.98356 0.000 0.000 0.000 0.796 0.204
#> GSM96973 4 0.3143 0.98356 0.000 0.000 0.000 0.796 0.204
#> GSM96976 4 0.3143 0.98356 0.000 0.000 0.000 0.796 0.204
#> GSM96977 5 0.0290 0.87772 0.000 0.000 0.000 0.008 0.992
#> GSM96995 3 0.0794 0.88898 0.000 0.028 0.972 0.000 0.000
#> GSM97002 1 0.0865 0.89751 0.972 0.000 0.000 0.024 0.004
#> GSM97009 2 0.3630 0.50599 0.016 0.780 0.000 0.204 0.000
#> GSM97010 5 0.0510 0.87230 0.000 0.000 0.000 0.016 0.984
#> GSM96974 4 0.3143 0.98356 0.000 0.000 0.000 0.796 0.204
#> GSM96985 1 0.4219 0.26920 0.584 0.000 0.000 0.416 0.000
#> GSM96959 3 0.2969 0.78174 0.000 0.128 0.852 0.020 0.000
#> GSM96972 4 0.3143 0.98356 0.000 0.000 0.000 0.796 0.204
#> GSM96978 4 0.3561 0.91690 0.000 0.000 0.000 0.740 0.260
#> GSM96967 4 0.3143 0.98356 0.000 0.000 0.000 0.796 0.204
#> GSM96987 1 0.0162 0.90218 0.996 0.000 0.000 0.004 0.000
#> GSM97011 2 0.4270 0.48637 0.048 0.748 0.000 0.204 0.000
#> GSM96964 1 0.0510 0.90344 0.984 0.000 0.000 0.000 0.016
#> GSM96965 4 0.3143 0.98356 0.000 0.000 0.000 0.796 0.204
#> GSM96981 1 0.0000 0.90325 1.000 0.000 0.000 0.000 0.000
#> GSM96982 1 0.0771 0.89968 0.976 0.000 0.000 0.020 0.004
#> GSM96988 3 0.4375 0.18737 0.420 0.000 0.576 0.004 0.000
#> GSM97000 5 0.0000 0.87697 0.000 0.000 0.000 0.000 1.000
#> GSM97004 1 0.0771 0.89968 0.976 0.000 0.000 0.020 0.004
#> GSM97008 5 0.8323 0.13584 0.160 0.272 0.000 0.204 0.364
#> GSM96950 5 0.1205 0.86470 0.040 0.000 0.000 0.004 0.956
#> GSM96980 4 0.4010 0.88874 0.072 0.000 0.000 0.792 0.136
#> GSM96989 1 0.0162 0.90440 0.996 0.000 0.000 0.000 0.004
#> GSM96992 1 0.0290 0.90467 0.992 0.000 0.000 0.000 0.008
#> GSM96993 2 0.6420 0.00781 0.300 0.496 0.000 0.204 0.000
#> GSM96958 5 0.4045 0.46128 0.356 0.000 0.000 0.000 0.644
#> GSM96951 5 0.0963 0.86522 0.036 0.000 0.000 0.000 0.964
#> GSM96952 1 0.0290 0.90467 0.992 0.000 0.000 0.000 0.008
#> GSM96961 1 0.0404 0.90443 0.988 0.000 0.000 0.000 0.012
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM97038 2 0.0000 0.998 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97045 2 0.0000 0.998 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97047 5 0.3464 0.614 0.000 0.312 0.000 0.000 0.688 0.000
#> GSM97025 2 0.0000 0.998 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97030 3 0.1501 0.941 0.000 0.076 0.924 0.000 0.000 0.000
#> GSM97027 2 0.0000 0.998 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97033 2 0.0000 0.998 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97034 3 0.1411 0.942 0.000 0.060 0.936 0.000 0.000 0.004
#> GSM97020 2 0.0000 0.998 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97026 2 0.0865 0.955 0.000 0.964 0.000 0.000 0.036 0.000
#> GSM97012 2 0.0000 0.998 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97015 3 0.1327 0.942 0.000 0.064 0.936 0.000 0.000 0.000
#> GSM97016 2 0.0000 0.998 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97017 5 0.0713 0.910 0.000 0.028 0.000 0.000 0.972 0.000
#> GSM97019 2 0.0000 0.998 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97022 2 0.0000 0.998 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97035 2 0.0000 0.998 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97036 5 0.1148 0.907 0.016 0.020 0.004 0.000 0.960 0.000
#> GSM97039 2 0.0000 0.998 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97046 2 0.0000 0.998 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97023 1 0.1780 0.925 0.932 0.000 0.012 0.000 0.028 0.028
#> GSM97029 5 0.0713 0.910 0.000 0.028 0.000 0.000 0.972 0.000
#> GSM97043 2 0.0000 0.998 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97013 6 0.1989 0.917 0.024 0.000 0.016 0.012 0.020 0.928
#> GSM96956 3 0.1501 0.941 0.000 0.076 0.924 0.000 0.000 0.000
#> GSM97024 2 0.0000 0.998 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97032 3 0.1610 0.936 0.000 0.084 0.916 0.000 0.000 0.000
#> GSM97044 3 0.1349 0.941 0.000 0.056 0.940 0.000 0.000 0.004
#> GSM97049 2 0.0000 0.998 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM96968 6 0.0972 0.946 0.000 0.000 0.008 0.028 0.000 0.964
#> GSM96971 4 0.0260 0.977 0.000 0.000 0.000 0.992 0.000 0.008
#> GSM96986 6 0.1049 0.946 0.000 0.000 0.008 0.032 0.000 0.960
#> GSM97003 6 0.0972 0.947 0.000 0.000 0.008 0.028 0.000 0.964
#> GSM96957 5 0.4083 0.743 0.140 0.000 0.024 0.008 0.784 0.044
#> GSM96960 1 0.0405 0.944 0.988 0.000 0.000 0.000 0.008 0.004
#> GSM96975 1 0.1065 0.940 0.964 0.000 0.020 0.008 0.008 0.000
#> GSM96998 1 0.1448 0.931 0.948 0.000 0.012 0.000 0.016 0.024
#> GSM96999 1 0.2308 0.893 0.896 0.000 0.012 0.000 0.016 0.076
#> GSM97001 5 0.1026 0.890 0.004 0.000 0.012 0.008 0.968 0.008
#> GSM97005 6 0.1514 0.925 0.016 0.000 0.016 0.004 0.016 0.948
#> GSM97006 1 0.4387 0.579 0.684 0.000 0.008 0.016 0.016 0.276
#> GSM97021 5 0.0458 0.908 0.000 0.016 0.000 0.000 0.984 0.000
#> GSM97028 3 0.1267 0.942 0.000 0.060 0.940 0.000 0.000 0.000
#> GSM97031 6 0.1251 0.925 0.024 0.000 0.008 0.000 0.012 0.956
#> GSM97037 3 0.1501 0.941 0.000 0.076 0.924 0.000 0.000 0.000
#> GSM97018 3 0.1556 0.939 0.000 0.080 0.920 0.000 0.000 0.000
#> GSM97014 5 0.3515 0.589 0.000 0.324 0.000 0.000 0.676 0.000
#> GSM97042 2 0.0000 0.998 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97040 5 0.0713 0.910 0.000 0.028 0.000 0.000 0.972 0.000
#> GSM97041 5 0.0458 0.908 0.000 0.016 0.000 0.000 0.984 0.000
#> GSM96955 2 0.0000 0.998 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM96990 3 0.1501 0.941 0.000 0.076 0.924 0.000 0.000 0.000
#> GSM96991 2 0.0000 0.998 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97048 2 0.0000 0.998 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM96963 2 0.0000 0.998 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM96953 2 0.0000 0.998 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM96966 4 0.0260 0.977 0.000 0.000 0.000 0.992 0.000 0.008
#> GSM96979 6 0.1124 0.945 0.000 0.000 0.008 0.036 0.000 0.956
#> GSM96983 3 0.1141 0.939 0.000 0.052 0.948 0.000 0.000 0.000
#> GSM96984 6 0.1124 0.945 0.000 0.000 0.008 0.036 0.000 0.956
#> GSM96994 3 0.1401 0.919 0.000 0.028 0.948 0.000 0.004 0.020
#> GSM96996 1 0.1176 0.930 0.956 0.000 0.020 0.000 0.024 0.000
#> GSM96997 6 0.1049 0.946 0.000 0.000 0.008 0.032 0.000 0.960
#> GSM97007 3 0.2100 0.840 0.000 0.000 0.884 0.000 0.004 0.112
#> GSM96954 6 0.0972 0.946 0.000 0.000 0.008 0.028 0.000 0.964
#> GSM96962 6 0.1124 0.945 0.000 0.000 0.008 0.036 0.000 0.956
#> GSM96969 4 0.0260 0.977 0.000 0.000 0.000 0.992 0.000 0.008
#> GSM96970 4 0.0260 0.977 0.000 0.000 0.000 0.992 0.000 0.008
#> GSM96973 4 0.0260 0.977 0.000 0.000 0.000 0.992 0.000 0.008
#> GSM96976 4 0.0260 0.977 0.000 0.000 0.000 0.992 0.000 0.008
#> GSM96977 6 0.0935 0.946 0.000 0.000 0.004 0.032 0.000 0.964
#> GSM96995 3 0.1508 0.936 0.000 0.048 0.940 0.004 0.004 0.004
#> GSM97002 1 0.0508 0.943 0.984 0.000 0.012 0.000 0.000 0.004
#> GSM97009 5 0.1440 0.901 0.004 0.044 0.004 0.004 0.944 0.000
#> GSM97010 6 0.1643 0.925 0.000 0.000 0.008 0.068 0.000 0.924
#> GSM96974 4 0.0260 0.977 0.000 0.000 0.000 0.992 0.000 0.008
#> GSM96985 1 0.3062 0.781 0.816 0.000 0.024 0.160 0.000 0.000
#> GSM96959 3 0.3642 0.824 0.000 0.080 0.800 0.004 0.116 0.000
#> GSM96972 4 0.0260 0.977 0.000 0.000 0.000 0.992 0.000 0.008
#> GSM96978 4 0.2653 0.828 0.000 0.000 0.012 0.844 0.000 0.144
#> GSM96967 4 0.0260 0.977 0.000 0.000 0.000 0.992 0.000 0.008
#> GSM96987 1 0.0909 0.938 0.968 0.000 0.020 0.000 0.012 0.000
#> GSM97011 5 0.1067 0.909 0.004 0.024 0.004 0.004 0.964 0.000
#> GSM96964 1 0.1364 0.933 0.952 0.000 0.012 0.000 0.016 0.020
#> GSM96965 4 0.0260 0.977 0.000 0.000 0.000 0.992 0.000 0.008
#> GSM96981 1 0.0951 0.939 0.968 0.000 0.020 0.004 0.008 0.000
#> GSM96982 1 0.0603 0.943 0.980 0.000 0.016 0.000 0.000 0.004
#> GSM96988 3 0.3668 0.481 0.328 0.000 0.668 0.000 0.004 0.000
#> GSM97000 6 0.0146 0.939 0.000 0.000 0.004 0.000 0.000 0.996
#> GSM97004 1 0.0508 0.943 0.984 0.000 0.012 0.000 0.000 0.004
#> GSM97008 5 0.2231 0.860 0.012 0.000 0.020 0.008 0.912 0.048
#> GSM96950 6 0.2072 0.925 0.024 0.000 0.012 0.024 0.016 0.924
#> GSM96980 4 0.1866 0.897 0.084 0.000 0.000 0.908 0.000 0.008
#> GSM96989 1 0.0458 0.942 0.984 0.000 0.016 0.000 0.000 0.000
#> GSM96992 1 0.0405 0.944 0.988 0.000 0.000 0.000 0.008 0.004
#> GSM96993 5 0.1148 0.907 0.016 0.020 0.004 0.000 0.960 0.000
#> GSM96958 6 0.4378 0.546 0.292 0.000 0.012 0.008 0.016 0.672
#> GSM96951 6 0.1325 0.929 0.016 0.000 0.012 0.004 0.012 0.956
#> GSM96952 1 0.0405 0.944 0.988 0.000 0.000 0.000 0.008 0.004
#> GSM96961 1 0.1078 0.937 0.964 0.000 0.008 0.000 0.016 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 disease.state(p) specimen(p) cell.type(p) other(p) k
#> ATC:skmeans 100 1.24e-07 0.714 5.25e-14 0.102 2
#> ATC:skmeans 93 4.35e-06 0.515 2.47e-18 0.392 3
#> ATC:skmeans 91 6.42e-06 0.601 1.86e-17 0.250 4
#> ATC:skmeans 86 3.28e-05 0.179 7.68e-16 0.311 5
#> ATC:skmeans 99 1.82e-05 0.447 5.89e-16 0.420 6
If matrix rows can be associated to genes, consider to use functional_enrichment(res,
...) to perform function enrichment for the signature genes. See this vignette for more detailed explanations.
The object with results only for a single top-value method and a single partition method can be extracted as:
res = res_list["ATC", "pam"]
# you can also extract it by
# res = res_list["ATC:pam"]
A summary of res and all the functions that can be applied to it:
res
#> A 'ConsensusPartition' object with k = 2, 3, 4, 5, 6.
#> On a matrix with 21168 rows and 100 columns.
#> Top rows (1000, 2000, 3000, 4000, 5000) are extracted by 'ATC' method.
#> Subgroups are detected by 'pam' method.
#> Performed in total 1250 partitions by row resampling.
#> Best k for subgroups seems to be 3.
#>
#> Following methods can be applied to this 'ConsensusPartition' object:
#> [1] "cola_report" "collect_classes" "collect_plots"
#> [4] "collect_stats" "colnames" "compare_signatures"
#> [7] "consensus_heatmap" "dimension_reduction" "functional_enrichment"
#> [10] "get_anno_col" "get_anno" "get_classes"
#> [13] "get_consensus" "get_matrix" "get_membership"
#> [16] "get_param" "get_signatures" "get_stats"
#> [19] "is_best_k" "is_stable_k" "membership_heatmap"
#> [22] "ncol" "nrow" "plot_ecdf"
#> [25] "rownames" "select_partition_number" "show"
#> [28] "suggest_best_k" "test_to_known_factors"
collect_plots() function collects all the plots made from res for all k (number of partitions)
into one single page to provide an easy and fast comparison between different k.
collect_plots(res)
The plots are:
k and the heatmap of
predicted classes for each k.k.k.k.All the plots in panels can be made by individual functions and they are plotted later in this section.
select_partition_number() produces several plots showing different
statistics for choosing “optimized” k. There are following statistics:
k;k, the area increased is defined as \(A_k - A_{k-1}\).The detailed explanations of these statistics can be found in the cola vignette.
Generally speaking, lower PAC score, higher mean silhouette score or higher
concordance corresponds to better partition. Rand index and Jaccard index
measure how similar the current partition is compared to partition with k-1.
If they are too similar, we won't accept k is better than k-1.
select_partition_number(res)
The numeric values for all these statistics can be obtained by get_stats().
get_stats(res)
#> k 1-PAC mean_silhouette concordance area_increased Rand Jaccard
#> 2 2 0.958 0.933 0.971 0.4912 0.515 0.515
#> 3 3 0.908 0.948 0.978 0.1942 0.899 0.804
#> 4 4 0.649 0.763 0.837 0.2083 0.855 0.664
#> 5 5 0.860 0.792 0.915 0.1006 0.872 0.608
#> 6 6 0.854 0.754 0.887 0.0452 0.928 0.696
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
#> GSM97038 2 0.0000 0.997 0.000 1.000
#> GSM97045 2 0.0000 0.997 0.000 1.000
#> GSM97047 2 0.0000 0.997 0.000 1.000
#> GSM97025 2 0.0000 0.997 0.000 1.000
#> GSM97030 2 0.0000 0.997 0.000 1.000
#> GSM97027 2 0.0000 0.997 0.000 1.000
#> GSM97033 2 0.0000 0.997 0.000 1.000
#> GSM97034 2 0.0672 0.989 0.008 0.992
#> GSM97020 2 0.0000 0.997 0.000 1.000
#> GSM97026 2 0.0000 0.997 0.000 1.000
#> GSM97012 2 0.0000 0.997 0.000 1.000
#> GSM97015 2 0.0000 0.997 0.000 1.000
#> GSM97016 2 0.0000 0.997 0.000 1.000
#> GSM97017 2 0.0000 0.997 0.000 1.000
#> GSM97019 2 0.0000 0.997 0.000 1.000
#> GSM97022 2 0.0000 0.997 0.000 1.000
#> GSM97035 2 0.0000 0.997 0.000 1.000
#> GSM97036 2 0.0000 0.997 0.000 1.000
#> GSM97039 2 0.0000 0.997 0.000 1.000
#> GSM97046 2 0.0000 0.997 0.000 1.000
#> GSM97023 1 0.0376 0.954 0.996 0.004
#> GSM97029 1 0.9993 0.150 0.516 0.484
#> GSM97043 2 0.0000 0.997 0.000 1.000
#> GSM97013 1 0.0376 0.954 0.996 0.004
#> GSM96956 2 0.0000 0.997 0.000 1.000
#> GSM97024 2 0.0000 0.997 0.000 1.000
#> GSM97032 2 0.0000 0.997 0.000 1.000
#> GSM97044 2 0.0000 0.997 0.000 1.000
#> GSM97049 2 0.0000 0.997 0.000 1.000
#> GSM96968 1 0.0376 0.954 0.996 0.004
#> GSM96971 1 0.0000 0.952 1.000 0.000
#> GSM96986 1 0.0376 0.954 0.996 0.004
#> GSM97003 1 0.0376 0.954 0.996 0.004
#> GSM96957 1 0.0376 0.954 0.996 0.004
#> GSM96960 1 0.0376 0.954 0.996 0.004
#> GSM96975 1 0.0376 0.954 0.996 0.004
#> GSM96998 1 0.0376 0.954 0.996 0.004
#> GSM96999 1 0.0376 0.954 0.996 0.004
#> GSM97001 1 0.0376 0.954 0.996 0.004
#> GSM97005 1 0.0376 0.954 0.996 0.004
#> GSM97006 1 0.0376 0.954 0.996 0.004
#> GSM97021 1 0.9686 0.406 0.604 0.396
#> GSM97028 2 0.0000 0.997 0.000 1.000
#> GSM97031 1 0.0376 0.954 0.996 0.004
#> GSM97037 2 0.0000 0.997 0.000 1.000
#> GSM97018 2 0.0000 0.997 0.000 1.000
#> GSM97014 2 0.0000 0.997 0.000 1.000
#> GSM97042 2 0.0000 0.997 0.000 1.000
#> GSM97040 2 0.0000 0.997 0.000 1.000
#> GSM97041 1 0.9661 0.416 0.608 0.392
#> GSM96955 2 0.0000 0.997 0.000 1.000
#> GSM96990 2 0.0000 0.997 0.000 1.000
#> GSM96991 2 0.0000 0.997 0.000 1.000
#> GSM97048 2 0.0000 0.997 0.000 1.000
#> GSM96963 2 0.0000 0.997 0.000 1.000
#> GSM96953 2 0.0000 0.997 0.000 1.000
#> GSM96966 1 0.0000 0.952 1.000 0.000
#> GSM96979 1 0.0376 0.954 0.996 0.004
#> GSM96983 2 0.2423 0.955 0.040 0.960
#> GSM96984 1 0.0376 0.954 0.996 0.004
#> GSM96994 1 0.0672 0.951 0.992 0.008
#> GSM96996 1 0.0376 0.954 0.996 0.004
#> GSM96997 1 0.0376 0.954 0.996 0.004
#> GSM97007 1 0.0376 0.954 0.996 0.004
#> GSM96954 1 0.0376 0.954 0.996 0.004
#> GSM96962 1 0.0376 0.954 0.996 0.004
#> GSM96969 1 0.0000 0.952 1.000 0.000
#> GSM96970 1 0.0000 0.952 1.000 0.000
#> GSM96973 1 0.0000 0.952 1.000 0.000
#> GSM96976 1 0.0000 0.952 1.000 0.000
#> GSM96977 1 0.0376 0.954 0.996 0.004
#> GSM96995 1 0.6623 0.784 0.828 0.172
#> GSM97002 1 0.0376 0.954 0.996 0.004
#> GSM97009 1 0.9552 0.451 0.624 0.376
#> GSM97010 1 0.0376 0.954 0.996 0.004
#> GSM96974 1 0.0000 0.952 1.000 0.000
#> GSM96985 1 0.0376 0.954 0.996 0.004
#> GSM96959 2 0.3879 0.912 0.076 0.924
#> GSM96972 1 0.0000 0.952 1.000 0.000
#> GSM96978 1 0.0376 0.954 0.996 0.004
#> GSM96967 1 0.0000 0.952 1.000 0.000
#> GSM96987 1 0.0376 0.954 0.996 0.004
#> GSM97011 1 0.9635 0.425 0.612 0.388
#> GSM96964 1 0.0376 0.954 0.996 0.004
#> GSM96965 1 0.0000 0.952 1.000 0.000
#> GSM96981 1 0.0376 0.954 0.996 0.004
#> GSM96982 1 0.0376 0.954 0.996 0.004
#> GSM96988 1 0.0376 0.954 0.996 0.004
#> GSM97000 1 0.0376 0.954 0.996 0.004
#> GSM97004 1 0.0376 0.954 0.996 0.004
#> GSM97008 1 0.0376 0.954 0.996 0.004
#> GSM96950 1 0.0376 0.954 0.996 0.004
#> GSM96980 1 0.0000 0.952 1.000 0.000
#> GSM96989 1 0.0376 0.954 0.996 0.004
#> GSM96992 1 0.0376 0.954 0.996 0.004
#> GSM96993 1 0.9661 0.416 0.608 0.392
#> GSM96958 1 0.0376 0.954 0.996 0.004
#> GSM96951 1 0.0376 0.954 0.996 0.004
#> GSM96952 1 0.0376 0.954 0.996 0.004
#> GSM96961 1 0.0376 0.954 0.996 0.004
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM97038 2 0.0000 0.978 0.000 1.000 0.000
#> GSM97045 2 0.0000 0.978 0.000 1.000 0.000
#> GSM97047 2 0.0000 0.978 0.000 1.000 0.000
#> GSM97025 2 0.0000 0.978 0.000 1.000 0.000
#> GSM97030 2 0.0000 0.978 0.000 1.000 0.000
#> GSM97027 2 0.0000 0.978 0.000 1.000 0.000
#> GSM97033 2 0.0000 0.978 0.000 1.000 0.000
#> GSM97034 2 0.3038 0.856 0.104 0.896 0.000
#> GSM97020 2 0.0000 0.978 0.000 1.000 0.000
#> GSM97026 2 0.0000 0.978 0.000 1.000 0.000
#> GSM97012 2 0.0000 0.978 0.000 1.000 0.000
#> GSM97015 2 0.0000 0.978 0.000 1.000 0.000
#> GSM97016 2 0.0000 0.978 0.000 1.000 0.000
#> GSM97017 2 0.0000 0.978 0.000 1.000 0.000
#> GSM97019 2 0.0000 0.978 0.000 1.000 0.000
#> GSM97022 2 0.0000 0.978 0.000 1.000 0.000
#> GSM97035 2 0.0000 0.978 0.000 1.000 0.000
#> GSM97036 2 0.4121 0.766 0.168 0.832 0.000
#> GSM97039 2 0.0000 0.978 0.000 1.000 0.000
#> GSM97046 2 0.0000 0.978 0.000 1.000 0.000
#> GSM97023 1 0.0000 0.962 1.000 0.000 0.000
#> GSM97029 1 0.5560 0.604 0.700 0.300 0.000
#> GSM97043 2 0.0000 0.978 0.000 1.000 0.000
#> GSM97013 1 0.0000 0.962 1.000 0.000 0.000
#> GSM96956 2 0.0000 0.978 0.000 1.000 0.000
#> GSM97024 2 0.0000 0.978 0.000 1.000 0.000
#> GSM97032 2 0.0000 0.978 0.000 1.000 0.000
#> GSM97044 2 0.0000 0.978 0.000 1.000 0.000
#> GSM97049 2 0.0000 0.978 0.000 1.000 0.000
#> GSM96968 1 0.0000 0.962 1.000 0.000 0.000
#> GSM96971 3 0.0000 1.000 0.000 0.000 1.000
#> GSM96986 1 0.0000 0.962 1.000 0.000 0.000
#> GSM97003 1 0.0000 0.962 1.000 0.000 0.000
#> GSM96957 1 0.0000 0.962 1.000 0.000 0.000
#> GSM96960 1 0.0000 0.962 1.000 0.000 0.000
#> GSM96975 1 0.0000 0.962 1.000 0.000 0.000
#> GSM96998 1 0.0000 0.962 1.000 0.000 0.000
#> GSM96999 1 0.0000 0.962 1.000 0.000 0.000
#> GSM97001 1 0.0000 0.962 1.000 0.000 0.000
#> GSM97005 1 0.0000 0.962 1.000 0.000 0.000
#> GSM97006 1 0.0000 0.962 1.000 0.000 0.000
#> GSM97021 1 0.4887 0.715 0.772 0.228 0.000
#> GSM97028 2 0.4931 0.673 0.232 0.768 0.000
#> GSM97031 1 0.0000 0.962 1.000 0.000 0.000
#> GSM97037 2 0.0000 0.978 0.000 1.000 0.000
#> GSM97018 2 0.0000 0.978 0.000 1.000 0.000
#> GSM97014 2 0.0000 0.978 0.000 1.000 0.000
#> GSM97042 2 0.0000 0.978 0.000 1.000 0.000
#> GSM97040 2 0.0000 0.978 0.000 1.000 0.000
#> GSM97041 1 0.4235 0.782 0.824 0.176 0.000
#> GSM96955 2 0.0000 0.978 0.000 1.000 0.000
#> GSM96990 2 0.0000 0.978 0.000 1.000 0.000
#> GSM96991 2 0.0000 0.978 0.000 1.000 0.000
#> GSM97048 2 0.0000 0.978 0.000 1.000 0.000
#> GSM96963 2 0.0000 0.978 0.000 1.000 0.000
#> GSM96953 2 0.0000 0.978 0.000 1.000 0.000
#> GSM96966 3 0.0000 1.000 0.000 0.000 1.000
#> GSM96979 1 0.0000 0.962 1.000 0.000 0.000
#> GSM96983 2 0.2448 0.893 0.076 0.924 0.000
#> GSM96984 1 0.0000 0.962 1.000 0.000 0.000
#> GSM96994 1 0.3551 0.832 0.868 0.132 0.000
#> GSM96996 1 0.0000 0.962 1.000 0.000 0.000
#> GSM96997 1 0.0000 0.962 1.000 0.000 0.000
#> GSM97007 1 0.2356 0.893 0.928 0.072 0.000
#> GSM96954 1 0.0000 0.962 1.000 0.000 0.000
#> GSM96962 1 0.0000 0.962 1.000 0.000 0.000
#> GSM96969 3 0.0000 1.000 0.000 0.000 1.000
#> GSM96970 3 0.0000 1.000 0.000 0.000 1.000
#> GSM96973 3 0.0000 1.000 0.000 0.000 1.000
#> GSM96976 3 0.0000 1.000 0.000 0.000 1.000
#> GSM96977 1 0.0000 0.962 1.000 0.000 0.000
#> GSM96995 1 0.1964 0.914 0.944 0.056 0.000
#> GSM97002 1 0.0000 0.962 1.000 0.000 0.000
#> GSM97009 1 0.4504 0.759 0.804 0.196 0.000
#> GSM97010 1 0.0000 0.962 1.000 0.000 0.000
#> GSM96974 3 0.0000 1.000 0.000 0.000 1.000
#> GSM96985 1 0.0000 0.962 1.000 0.000 0.000
#> GSM96959 2 0.2448 0.890 0.076 0.924 0.000
#> GSM96972 3 0.0000 1.000 0.000 0.000 1.000
#> GSM96978 1 0.0000 0.962 1.000 0.000 0.000
#> GSM96967 3 0.0000 1.000 0.000 0.000 1.000
#> GSM96987 1 0.0000 0.962 1.000 0.000 0.000
#> GSM97011 1 0.4654 0.742 0.792 0.208 0.000
#> GSM96964 1 0.0000 0.962 1.000 0.000 0.000
#> GSM96965 3 0.0000 1.000 0.000 0.000 1.000
#> GSM96981 1 0.0000 0.962 1.000 0.000 0.000
#> GSM96982 1 0.0000 0.962 1.000 0.000 0.000
#> GSM96988 1 0.0000 0.962 1.000 0.000 0.000
#> GSM97000 1 0.0000 0.962 1.000 0.000 0.000
#> GSM97004 1 0.0000 0.962 1.000 0.000 0.000
#> GSM97008 1 0.0000 0.962 1.000 0.000 0.000
#> GSM96950 1 0.0000 0.962 1.000 0.000 0.000
#> GSM96980 1 0.0592 0.953 0.988 0.000 0.012
#> GSM96989 1 0.0000 0.962 1.000 0.000 0.000
#> GSM96992 1 0.0000 0.962 1.000 0.000 0.000
#> GSM96993 1 0.3879 0.809 0.848 0.152 0.000
#> GSM96958 1 0.0000 0.962 1.000 0.000 0.000
#> GSM96951 1 0.0000 0.962 1.000 0.000 0.000
#> GSM96952 1 0.0000 0.962 1.000 0.000 0.000
#> GSM96961 1 0.0000 0.962 1.000 0.000 0.000
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM97038 2 0.0000 0.9032 0.000 1.000 0.000 0.000
#> GSM97045 2 0.0000 0.9032 0.000 1.000 0.000 0.000
#> GSM97047 3 0.4713 0.6131 0.000 0.360 0.640 0.000
#> GSM97025 2 0.0000 0.9032 0.000 1.000 0.000 0.000
#> GSM97030 2 0.0188 0.8997 0.000 0.996 0.004 0.000
#> GSM97027 2 0.0000 0.9032 0.000 1.000 0.000 0.000
#> GSM97033 2 0.0000 0.9032 0.000 1.000 0.000 0.000
#> GSM97034 3 0.0707 0.5561 0.000 0.020 0.980 0.000
#> GSM97020 2 0.0000 0.9032 0.000 1.000 0.000 0.000
#> GSM97026 2 0.4843 -0.0161 0.000 0.604 0.396 0.000
#> GSM97012 2 0.0000 0.9032 0.000 1.000 0.000 0.000
#> GSM97015 3 0.4304 0.6401 0.000 0.284 0.716 0.000
#> GSM97016 2 0.0000 0.9032 0.000 1.000 0.000 0.000
#> GSM97017 3 0.6147 0.6109 0.056 0.380 0.564 0.000
#> GSM97019 2 0.0000 0.9032 0.000 1.000 0.000 0.000
#> GSM97022 2 0.0000 0.9032 0.000 1.000 0.000 0.000
#> GSM97035 2 0.0000 0.9032 0.000 1.000 0.000 0.000
#> GSM97036 3 0.7110 0.7245 0.200 0.236 0.564 0.000
#> GSM97039 2 0.0000 0.9032 0.000 1.000 0.000 0.000
#> GSM97046 2 0.0000 0.9032 0.000 1.000 0.000 0.000
#> GSM97023 1 0.1211 0.7954 0.960 0.000 0.040 0.000
#> GSM97029 3 0.7114 0.7259 0.204 0.232 0.564 0.000
#> GSM97043 2 0.0000 0.9032 0.000 1.000 0.000 0.000
#> GSM97013 1 0.2760 0.7988 0.872 0.000 0.128 0.000
#> GSM96956 2 0.2868 0.7335 0.000 0.864 0.136 0.000
#> GSM97024 2 0.0000 0.9032 0.000 1.000 0.000 0.000
#> GSM97032 2 0.3649 0.6025 0.000 0.796 0.204 0.000
#> GSM97044 2 0.4193 0.5663 0.000 0.732 0.268 0.000
#> GSM97049 2 0.0000 0.9032 0.000 1.000 0.000 0.000
#> GSM96968 1 0.4907 0.7140 0.580 0.000 0.420 0.000
#> GSM96971 4 0.0000 1.0000 0.000 0.000 0.000 1.000
#> GSM96986 1 0.4933 0.7075 0.568 0.000 0.432 0.000
#> GSM97003 1 0.4907 0.7134 0.580 0.000 0.420 0.000
#> GSM96957 1 0.0469 0.7924 0.988 0.000 0.012 0.000
#> GSM96960 1 0.0000 0.7897 1.000 0.000 0.000 0.000
#> GSM96975 1 0.3024 0.7787 0.852 0.000 0.148 0.000
#> GSM96998 1 0.0469 0.7925 0.988 0.000 0.012 0.000
#> GSM96999 1 0.1557 0.7988 0.944 0.000 0.056 0.000
#> GSM97001 1 0.3907 0.4656 0.768 0.000 0.232 0.000
#> GSM97005 1 0.4040 0.7707 0.752 0.000 0.248 0.000
#> GSM97006 1 0.3649 0.7752 0.796 0.000 0.204 0.000
#> GSM97021 3 0.7054 0.7274 0.196 0.232 0.572 0.000
#> GSM97028 3 0.5180 0.6834 0.064 0.196 0.740 0.000
#> GSM97031 1 0.3907 0.7711 0.768 0.000 0.232 0.000
#> GSM97037 2 0.2760 0.7443 0.000 0.872 0.128 0.000
#> GSM97018 2 0.4866 -0.0437 0.000 0.596 0.404 0.000
#> GSM97014 3 0.4941 0.5094 0.000 0.436 0.564 0.000
#> GSM97042 2 0.0000 0.9032 0.000 1.000 0.000 0.000
#> GSM97040 3 0.4907 0.5349 0.000 0.420 0.580 0.000
#> GSM97041 3 0.7082 0.7118 0.252 0.184 0.564 0.000
#> GSM96955 2 0.0000 0.9032 0.000 1.000 0.000 0.000
#> GSM96990 3 0.4222 0.6467 0.000 0.272 0.728 0.000
#> GSM96991 2 0.0000 0.9032 0.000 1.000 0.000 0.000
#> GSM97048 2 0.0000 0.9032 0.000 1.000 0.000 0.000
#> GSM96963 2 0.0000 0.9032 0.000 1.000 0.000 0.000
#> GSM96953 2 0.0000 0.9032 0.000 1.000 0.000 0.000
#> GSM96966 4 0.0000 1.0000 0.000 0.000 0.000 1.000
#> GSM96979 1 0.4933 0.7075 0.568 0.000 0.432 0.000
#> GSM96983 2 0.6187 0.0491 0.052 0.516 0.432 0.000
#> GSM96984 1 0.4933 0.7075 0.568 0.000 0.432 0.000
#> GSM96994 1 0.6000 0.5345 0.508 0.040 0.452 0.000
#> GSM96996 1 0.0707 0.7790 0.980 0.000 0.020 0.000
#> GSM96997 1 0.4933 0.7075 0.568 0.000 0.432 0.000
#> GSM97007 1 0.4981 0.6809 0.536 0.000 0.464 0.000
#> GSM96954 1 0.4933 0.7075 0.568 0.000 0.432 0.000
#> GSM96962 1 0.4933 0.7075 0.568 0.000 0.432 0.000
#> GSM96969 4 0.0000 1.0000 0.000 0.000 0.000 1.000
#> GSM96970 4 0.0000 1.0000 0.000 0.000 0.000 1.000
#> GSM96973 4 0.0000 1.0000 0.000 0.000 0.000 1.000
#> GSM96976 4 0.0000 1.0000 0.000 0.000 0.000 1.000
#> GSM96977 1 0.4304 0.7631 0.716 0.000 0.284 0.000
#> GSM96995 3 0.4487 0.6777 0.100 0.092 0.808 0.000
#> GSM97002 1 0.0000 0.7897 1.000 0.000 0.000 0.000
#> GSM97009 3 0.5410 0.7158 0.080 0.192 0.728 0.000
#> GSM97010 1 0.3873 0.7749 0.772 0.000 0.228 0.000
#> GSM96974 4 0.0000 1.0000 0.000 0.000 0.000 1.000
#> GSM96985 1 0.1792 0.7962 0.932 0.000 0.068 0.000
#> GSM96959 3 0.3308 0.6669 0.036 0.092 0.872 0.000
#> GSM96972 4 0.0000 1.0000 0.000 0.000 0.000 1.000
#> GSM96978 1 0.4697 0.7308 0.644 0.000 0.356 0.000
#> GSM96967 4 0.0000 1.0000 0.000 0.000 0.000 1.000
#> GSM96987 3 0.4999 0.3800 0.492 0.000 0.508 0.000
#> GSM97011 3 0.7058 0.7292 0.200 0.228 0.572 0.000
#> GSM96964 1 0.0000 0.7897 1.000 0.000 0.000 0.000
#> GSM96965 4 0.0000 1.0000 0.000 0.000 0.000 1.000
#> GSM96981 1 0.0592 0.7935 0.984 0.000 0.016 0.000
#> GSM96982 1 0.0188 0.7909 0.996 0.000 0.004 0.000
#> GSM96988 1 0.2408 0.7474 0.896 0.000 0.104 0.000
#> GSM97000 1 0.4933 0.7073 0.568 0.000 0.432 0.000
#> GSM97004 1 0.0000 0.7897 1.000 0.000 0.000 0.000
#> GSM97008 1 0.4454 0.7474 0.692 0.000 0.308 0.000
#> GSM96950 1 0.3486 0.7740 0.812 0.000 0.188 0.000
#> GSM96980 1 0.3554 0.7496 0.844 0.000 0.020 0.136
#> GSM96989 1 0.0000 0.7897 1.000 0.000 0.000 0.000
#> GSM96992 1 0.0000 0.7897 1.000 0.000 0.000 0.000
#> GSM96993 3 0.6201 0.5894 0.376 0.060 0.564 0.000
#> GSM96958 1 0.1792 0.7970 0.932 0.000 0.068 0.000
#> GSM96951 1 0.4134 0.7694 0.740 0.000 0.260 0.000
#> GSM96952 1 0.0000 0.7897 1.000 0.000 0.000 0.000
#> GSM96961 1 0.0000 0.7897 1.000 0.000 0.000 0.000
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM97038 2 0.0000 0.9560 0.000 1.000 0.000 0.000 0.000
#> GSM97045 2 0.0000 0.9560 0.000 1.000 0.000 0.000 0.000
#> GSM97047 5 0.0162 0.8757 0.000 0.004 0.000 0.000 0.996
#> GSM97025 2 0.0000 0.9560 0.000 1.000 0.000 0.000 0.000
#> GSM97030 2 0.0000 0.9560 0.000 1.000 0.000 0.000 0.000
#> GSM97027 2 0.0000 0.9560 0.000 1.000 0.000 0.000 0.000
#> GSM97033 2 0.0000 0.9560 0.000 1.000 0.000 0.000 0.000
#> GSM97034 5 0.3741 0.5662 0.000 0.004 0.264 0.000 0.732
#> GSM97020 2 0.0000 0.9560 0.000 1.000 0.000 0.000 0.000
#> GSM97026 5 0.4227 0.2615 0.000 0.420 0.000 0.000 0.580
#> GSM97012 2 0.0000 0.9560 0.000 1.000 0.000 0.000 0.000
#> GSM97015 5 0.0162 0.8757 0.000 0.004 0.000 0.000 0.996
#> GSM97016 2 0.0000 0.9560 0.000 1.000 0.000 0.000 0.000
#> GSM97017 5 0.0000 0.8759 0.000 0.000 0.000 0.000 1.000
#> GSM97019 2 0.0000 0.9560 0.000 1.000 0.000 0.000 0.000
#> GSM97022 2 0.0000 0.9560 0.000 1.000 0.000 0.000 0.000
#> GSM97035 2 0.0000 0.9560 0.000 1.000 0.000 0.000 0.000
#> GSM97036 5 0.0000 0.8759 0.000 0.000 0.000 0.000 1.000
#> GSM97039 2 0.0000 0.9560 0.000 1.000 0.000 0.000 0.000
#> GSM97046 2 0.0000 0.9560 0.000 1.000 0.000 0.000 0.000
#> GSM97023 1 0.0000 0.8392 1.000 0.000 0.000 0.000 0.000
#> GSM97029 5 0.0000 0.8759 0.000 0.000 0.000 0.000 1.000
#> GSM97043 2 0.0000 0.9560 0.000 1.000 0.000 0.000 0.000
#> GSM97013 1 0.3895 0.5414 0.680 0.000 0.320 0.000 0.000
#> GSM96956 2 0.0000 0.9560 0.000 1.000 0.000 0.000 0.000
#> GSM97024 2 0.0000 0.9560 0.000 1.000 0.000 0.000 0.000
#> GSM97032 2 0.3837 0.5336 0.000 0.692 0.000 0.000 0.308
#> GSM97044 2 0.5190 0.5942 0.000 0.688 0.172 0.000 0.140
#> GSM97049 2 0.0000 0.9560 0.000 1.000 0.000 0.000 0.000
#> GSM96968 3 0.4451 -0.2093 0.492 0.000 0.504 0.000 0.004
#> GSM96971 4 0.0290 0.9923 0.000 0.000 0.008 0.992 0.000
#> GSM96986 3 0.0000 0.8277 0.000 0.000 1.000 0.000 0.000
#> GSM97003 3 0.1270 0.7961 0.052 0.000 0.948 0.000 0.000
#> GSM96957 1 0.2230 0.8017 0.912 0.000 0.044 0.000 0.044
#> GSM96960 1 0.0000 0.8392 1.000 0.000 0.000 0.000 0.000
#> GSM96975 1 0.2890 0.7363 0.836 0.000 0.160 0.000 0.004
#> GSM96998 1 0.0000 0.8392 1.000 0.000 0.000 0.000 0.000
#> GSM96999 1 0.0566 0.8364 0.984 0.000 0.012 0.000 0.004
#> GSM97001 5 0.0880 0.8559 0.032 0.000 0.000 0.000 0.968
#> GSM97005 1 0.3949 0.5189 0.668 0.000 0.332 0.000 0.000
#> GSM97006 1 0.2852 0.7040 0.828 0.000 0.172 0.000 0.000
#> GSM97021 5 0.0162 0.8750 0.004 0.000 0.000 0.000 0.996
#> GSM97028 5 0.0162 0.8757 0.000 0.004 0.000 0.000 0.996
#> GSM97031 1 0.3816 0.5531 0.696 0.000 0.304 0.000 0.000
#> GSM97037 2 0.0000 0.9560 0.000 1.000 0.000 0.000 0.000
#> GSM97018 5 0.4182 0.3166 0.000 0.400 0.000 0.000 0.600
#> GSM97014 5 0.0162 0.8757 0.000 0.004 0.000 0.000 0.996
#> GSM97042 2 0.0000 0.9560 0.000 1.000 0.000 0.000 0.000
#> GSM97040 5 0.0162 0.8757 0.000 0.004 0.000 0.000 0.996
#> GSM97041 5 0.0000 0.8759 0.000 0.000 0.000 0.000 1.000
#> GSM96955 2 0.0000 0.9560 0.000 1.000 0.000 0.000 0.000
#> GSM96990 5 0.0162 0.8757 0.000 0.004 0.000 0.000 0.996
#> GSM96991 2 0.0000 0.9560 0.000 1.000 0.000 0.000 0.000
#> GSM97048 2 0.0000 0.9560 0.000 1.000 0.000 0.000 0.000
#> GSM96963 2 0.0000 0.9560 0.000 1.000 0.000 0.000 0.000
#> GSM96953 2 0.0000 0.9560 0.000 1.000 0.000 0.000 0.000
#> GSM96966 4 0.0000 0.9992 0.000 0.000 0.000 1.000 0.000
#> GSM96979 3 0.0000 0.8277 0.000 0.000 1.000 0.000 0.000
#> GSM96983 2 0.6394 0.1465 0.000 0.476 0.180 0.000 0.344
#> GSM96984 3 0.0000 0.8277 0.000 0.000 1.000 0.000 0.000
#> GSM96994 3 0.6783 0.2143 0.012 0.200 0.476 0.000 0.312
#> GSM96996 1 0.0162 0.8385 0.996 0.000 0.000 0.000 0.004
#> GSM96997 3 0.0000 0.8277 0.000 0.000 1.000 0.000 0.000
#> GSM97007 3 0.0000 0.8277 0.000 0.000 1.000 0.000 0.000
#> GSM96954 3 0.0510 0.8229 0.016 0.000 0.984 0.000 0.000
#> GSM96962 3 0.0000 0.8277 0.000 0.000 1.000 0.000 0.000
#> GSM96969 4 0.0000 0.9992 0.000 0.000 0.000 1.000 0.000
#> GSM96970 4 0.0000 0.9992 0.000 0.000 0.000 1.000 0.000
#> GSM96973 4 0.0000 0.9992 0.000 0.000 0.000 1.000 0.000
#> GSM96976 4 0.0000 0.9992 0.000 0.000 0.000 1.000 0.000
#> GSM96977 1 0.4425 0.2688 0.544 0.000 0.452 0.000 0.004
#> GSM96995 5 0.2852 0.7089 0.000 0.000 0.172 0.000 0.828
#> GSM97002 1 0.0162 0.8385 0.996 0.000 0.000 0.000 0.004
#> GSM97009 5 0.0162 0.8739 0.004 0.000 0.000 0.000 0.996
#> GSM97010 1 0.4425 0.2688 0.544 0.000 0.452 0.000 0.004
#> GSM96974 4 0.0000 0.9992 0.000 0.000 0.000 1.000 0.000
#> GSM96985 1 0.2068 0.7982 0.904 0.000 0.092 0.000 0.004
#> GSM96959 5 0.0000 0.8759 0.000 0.000 0.000 0.000 1.000
#> GSM96972 4 0.0000 0.9992 0.000 0.000 0.000 1.000 0.000
#> GSM96978 3 0.0771 0.8211 0.020 0.000 0.976 0.000 0.004
#> GSM96967 4 0.0000 0.9992 0.000 0.000 0.000 1.000 0.000
#> GSM96987 1 0.0404 0.8330 0.988 0.000 0.000 0.000 0.012
#> GSM97011 5 0.0000 0.8759 0.000 0.000 0.000 0.000 1.000
#> GSM96964 1 0.0000 0.8392 1.000 0.000 0.000 0.000 0.000
#> GSM96965 4 0.0000 0.9992 0.000 0.000 0.000 1.000 0.000
#> GSM96981 1 0.0771 0.8341 0.976 0.000 0.020 0.000 0.004
#> GSM96982 1 0.0324 0.8385 0.992 0.000 0.004 0.000 0.004
#> GSM96988 5 0.4273 0.2427 0.448 0.000 0.000 0.000 0.552
#> GSM97000 3 0.5915 0.0889 0.384 0.000 0.508 0.000 0.108
#> GSM97004 1 0.0000 0.8392 1.000 0.000 0.000 0.000 0.000
#> GSM97008 5 0.4329 0.4254 0.016 0.000 0.312 0.000 0.672
#> GSM96950 1 0.4425 0.2688 0.544 0.000 0.452 0.000 0.004
#> GSM96980 1 0.3807 0.7054 0.792 0.000 0.028 0.176 0.004
#> GSM96989 1 0.0000 0.8392 1.000 0.000 0.000 0.000 0.000
#> GSM96992 1 0.0000 0.8392 1.000 0.000 0.000 0.000 0.000
#> GSM96993 5 0.0162 0.8750 0.004 0.000 0.000 0.000 0.996
#> GSM96958 1 0.2286 0.7914 0.888 0.000 0.108 0.000 0.004
#> GSM96951 1 0.4101 0.4515 0.628 0.000 0.372 0.000 0.000
#> GSM96952 1 0.0000 0.8392 1.000 0.000 0.000 0.000 0.000
#> GSM96961 1 0.0000 0.8392 1.000 0.000 0.000 0.000 0.000
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM97038 2 0.0000 0.96400 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97045 2 0.0000 0.96400 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97047 5 0.0363 0.83940 0.000 0.012 0.000 0.000 0.988 0.000
#> GSM97025 2 0.0000 0.96400 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97030 2 0.0922 0.94347 0.000 0.968 0.024 0.000 0.004 0.004
#> GSM97027 2 0.0000 0.96400 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97033 2 0.0000 0.96400 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97034 5 0.5383 0.29307 0.000 0.000 0.232 0.000 0.584 0.184
#> GSM97020 2 0.0000 0.96400 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97026 5 0.3823 0.19890 0.000 0.436 0.000 0.000 0.564 0.000
#> GSM97012 2 0.0000 0.96400 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97015 5 0.0508 0.83932 0.000 0.000 0.012 0.000 0.984 0.004
#> GSM97016 2 0.0000 0.96400 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97017 5 0.0146 0.84121 0.000 0.004 0.000 0.000 0.996 0.000
#> GSM97019 2 0.0000 0.96400 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97022 2 0.0000 0.96400 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97035 2 0.0000 0.96400 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97036 5 0.0146 0.84121 0.000 0.004 0.000 0.000 0.996 0.000
#> GSM97039 2 0.0000 0.96400 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97046 2 0.0000 0.96400 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97023 1 0.0000 0.89544 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM97029 5 0.3817 0.20423 0.000 0.000 0.432 0.000 0.568 0.000
#> GSM97043 2 0.0000 0.96400 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97013 6 0.6120 0.40982 0.316 0.000 0.320 0.000 0.000 0.364
#> GSM96956 2 0.0653 0.95060 0.000 0.980 0.012 0.000 0.004 0.004
#> GSM97024 2 0.0000 0.96400 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97032 2 0.3848 0.56152 0.000 0.692 0.012 0.000 0.292 0.004
#> GSM97044 2 0.2932 0.81013 0.000 0.840 0.024 0.000 0.132 0.004
#> GSM97049 2 0.0000 0.96400 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM96968 3 0.3911 0.00114 0.008 0.000 0.624 0.000 0.000 0.368
#> GSM96971 4 0.0363 0.98789 0.000 0.000 0.000 0.988 0.000 0.012
#> GSM96986 6 0.0146 0.70607 0.000 0.000 0.004 0.000 0.000 0.996
#> GSM97003 6 0.3578 0.57053 0.000 0.000 0.340 0.000 0.000 0.660
#> GSM96957 3 0.3727 0.39771 0.388 0.000 0.612 0.000 0.000 0.000
#> GSM96960 1 0.0000 0.89544 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM96975 3 0.1814 0.70891 0.100 0.000 0.900 0.000 0.000 0.000
#> GSM96998 1 0.0363 0.88915 0.988 0.000 0.012 0.000 0.000 0.000
#> GSM96999 3 0.3592 0.53043 0.344 0.000 0.656 0.000 0.000 0.000
#> GSM97001 5 0.3490 0.55664 0.008 0.000 0.268 0.000 0.724 0.000
#> GSM97005 6 0.6116 0.41586 0.312 0.000 0.320 0.000 0.000 0.368
#> GSM97006 1 0.0508 0.88698 0.984 0.000 0.012 0.000 0.000 0.004
#> GSM97021 5 0.0000 0.84127 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM97028 5 0.0508 0.83932 0.000 0.000 0.012 0.000 0.984 0.004
#> GSM97031 1 0.4004 0.22596 0.620 0.000 0.012 0.000 0.000 0.368
#> GSM97037 2 0.0922 0.94347 0.000 0.968 0.024 0.000 0.004 0.004
#> GSM97018 5 0.4218 0.27664 0.000 0.400 0.012 0.000 0.584 0.004
#> GSM97014 5 0.0363 0.83940 0.000 0.012 0.000 0.000 0.988 0.000
#> GSM97042 2 0.0000 0.96400 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97040 5 0.0000 0.84127 0.000 0.000 0.000 0.000 1.000 0.000
#> GSM97041 5 0.0146 0.84103 0.000 0.000 0.004 0.000 0.996 0.000
#> GSM96955 2 0.0000 0.96400 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM96990 5 0.0508 0.83932 0.000 0.000 0.012 0.000 0.984 0.004
#> GSM96991 2 0.0000 0.96400 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97048 2 0.0000 0.96400 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM96963 2 0.0000 0.96400 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM96953 2 0.0000 0.96400 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM96966 4 0.0000 0.99867 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM96979 6 0.0146 0.70607 0.000 0.000 0.004 0.000 0.000 0.996
#> GSM96983 2 0.4498 0.45353 0.000 0.632 0.040 0.000 0.324 0.004
#> GSM96984 6 0.0146 0.70607 0.000 0.000 0.004 0.000 0.000 0.996
#> GSM96994 3 0.2201 0.64330 0.000 0.036 0.904 0.000 0.056 0.004
#> GSM96996 3 0.3883 0.53953 0.332 0.000 0.656 0.000 0.012 0.000
#> GSM96997 6 0.0146 0.70607 0.000 0.000 0.004 0.000 0.000 0.996
#> GSM97007 6 0.0363 0.69889 0.000 0.000 0.012 0.000 0.000 0.988
#> GSM96954 6 0.3499 0.58512 0.000 0.000 0.320 0.000 0.000 0.680
#> GSM96962 6 0.0146 0.70607 0.000 0.000 0.004 0.000 0.000 0.996
#> GSM96969 4 0.0000 0.99867 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM96970 4 0.0000 0.99867 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM96973 4 0.0000 0.99867 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM96976 4 0.0000 0.99867 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM96977 3 0.4230 0.00786 0.024 0.000 0.612 0.000 0.000 0.364
#> GSM96995 5 0.0603 0.84062 0.000 0.000 0.016 0.000 0.980 0.004
#> GSM97002 1 0.3804 -0.08318 0.576 0.000 0.424 0.000 0.000 0.000
#> GSM97009 5 0.0547 0.83551 0.000 0.000 0.020 0.000 0.980 0.000
#> GSM97010 3 0.3134 0.49839 0.024 0.000 0.808 0.000 0.000 0.168
#> GSM96974 4 0.0000 0.99867 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM96985 3 0.2092 0.71224 0.124 0.000 0.876 0.000 0.000 0.000
#> GSM96959 5 0.0405 0.84014 0.000 0.000 0.008 0.000 0.988 0.004
#> GSM96972 4 0.0000 0.99867 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM96978 3 0.0632 0.66096 0.000 0.000 0.976 0.000 0.000 0.024
#> GSM96967 4 0.0000 0.99867 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM96987 1 0.0000 0.89544 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM97011 5 0.1204 0.81256 0.000 0.000 0.056 0.000 0.944 0.000
#> GSM96964 1 0.1501 0.80894 0.924 0.000 0.076 0.000 0.000 0.000
#> GSM96965 4 0.0000 0.99867 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM96981 3 0.2793 0.68986 0.200 0.000 0.800 0.000 0.000 0.000
#> GSM96982 3 0.3531 0.55999 0.328 0.000 0.672 0.000 0.000 0.000
#> GSM96988 3 0.4795 0.50396 0.324 0.000 0.604 0.000 0.072 0.000
#> GSM97000 6 0.6230 0.42036 0.016 0.000 0.332 0.000 0.204 0.448
#> GSM97004 1 0.0000 0.89544 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM97008 5 0.5583 0.06144 0.000 0.000 0.152 0.000 0.500 0.348
#> GSM96950 3 0.2662 0.57364 0.024 0.000 0.856 0.000 0.000 0.120
#> GSM96980 3 0.1957 0.70867 0.112 0.000 0.888 0.000 0.000 0.000
#> GSM96989 1 0.0000 0.89544 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM96992 1 0.0363 0.88915 0.988 0.000 0.012 0.000 0.000 0.000
#> GSM96993 5 0.0146 0.84103 0.000 0.000 0.004 0.000 0.996 0.000
#> GSM96958 3 0.1075 0.68252 0.048 0.000 0.952 0.000 0.000 0.000
#> GSM96951 6 0.6116 0.41586 0.312 0.000 0.320 0.000 0.000 0.368
#> GSM96952 1 0.0000 0.89544 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM96961 1 0.0000 0.89544 1.000 0.000 0.000 0.000 0.000 0.000
Heatmaps for the consensus matrix. It visualizes the probability of two samples to be in a same group.
consensus_heatmap(res, k = 2)
consensus_heatmap(res, k = 3)
consensus_heatmap(res, k = 4)
consensus_heatmap(res, k = 5)
consensus_heatmap(res, k = 6)
Heatmaps for the membership of samples in all partitions to see how consistent they are:
membership_heatmap(res, k = 2)
membership_heatmap(res, k = 3)
membership_heatmap(res, k = 4)
membership_heatmap(res, k = 5)
membership_heatmap(res, k = 6)
As soon as we have had the classes for columns, we can look for signatures which are significantly different between classes which can be candidate marks for certain classes. Following are the heatmaps for signatures.
Signature heatmaps where rows are scaled:
get_signatures(res, k = 2)
get_signatures(res, k = 3)
get_signatures(res, k = 4)
get_signatures(res, k = 5)
get_signatures(res, k = 6)
Signature heatmaps where rows are not scaled:
get_signatures(res, k = 2, scale_rows = FALSE)
get_signatures(res, k = 3, scale_rows = FALSE)
get_signatures(res, k = 4, scale_rows = FALSE)
get_signatures(res, k = 5, scale_rows = FALSE)
get_signatures(res, k = 6, scale_rows = FALSE)
Compare the overlap of signatures from different k:
compare_signatures(res)
get_signature() returns a data frame invisibly. TO get the list of signatures, the function
call should be assigned to a variable explicitly. In following code, if plot argument is set
to FALSE, no heatmap is plotted while only the differential analysis is performed.
# code only for demonstration
tb = get_signature(res, k = ..., plot = FALSE)
An example of the output of tb is:
#> which_row fdr mean_1 mean_2 scaled_mean_1 scaled_mean_2 km
#> 1 38 0.042760348 8.373488 9.131774 -0.5533452 0.5164555 1
#> 2 40 0.018707592 7.106213 8.469186 -0.6173731 0.5762149 1
#> 3 55 0.019134737 10.221463 11.207825 -0.6159697 0.5749050 1
#> 4 59 0.006059896 5.921854 7.869574 -0.6899429 0.6439467 1
#> 5 60 0.018055526 8.928898 10.211722 -0.6204761 0.5791110 1
#> 6 98 0.009384629 15.714769 14.887706 0.6635654 -0.6193277 2
...
The columns in tb are:
which_row: row indices corresponding to the input matrix.fdr: FDR for the differential test. mean_x: The mean value in group x.scaled_mean_x: The mean value in group x after rows are scaled.km: Row groups if k-means clustering is applied to rows.UMAP plot which shows how samples are separated.
dimension_reduction(res, k = 2, method = "UMAP")
dimension_reduction(res, k = 3, method = "UMAP")
dimension_reduction(res, k = 4, method = "UMAP")
dimension_reduction(res, k = 5, method = "UMAP")
dimension_reduction(res, k = 6, method = "UMAP")
Following heatmap shows how subgroups are split when increasing k:
collect_classes(res)
Test correlation between subgroups and known annotations. If the known annotation is numeric, one-way ANOVA test is applied, and if the known annotation is discrete, chi-squared contingency table test is applied.
test_to_known_factors(res)
#> n disease.state(p) specimen(p) cell.type(p) other(p) k
#> ATC:pam 94 3.02e-09 0.472 7.85e-18 0.0110 2
#> ATC:pam 100 1.61e-07 0.221 1.13e-18 0.0321 3
#> ATC:pam 95 2.77e-06 0.545 2.06e-16 0.4184 4
#> ATC:pam 88 1.57e-04 0.490 5.32e-21 0.3015 5
#> ATC:pam 84 4.04e-04 0.937 7.19e-19 0.4925 6
If matrix rows can be associated to genes, consider to use functional_enrichment(res,
...) to perform function enrichment for the signature genes. See this vignette for more detailed explanations.
The object with results only for a single top-value method and a single partition method can be extracted as:
res = res_list["ATC", "mclust"]
# you can also extract it by
# res = res_list["ATC:mclust"]
A summary of res and all the functions that can be applied to it:
res
#> A 'ConsensusPartition' object with k = 2, 3, 4, 5, 6.
#> On a matrix with 21168 rows and 100 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 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 0.975 0.990 0.216 0.787 0.787
#> 3 3 0.587 0.791 0.903 1.614 0.600 0.498
#> 4 4 0.983 0.955 0.983 0.179 0.901 0.766
#> 5 5 0.736 0.785 0.882 0.112 0.881 0.673
#> 6 6 0.941 0.887 0.951 0.112 0.878 0.572
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 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
#> GSM97038 2 0.000 0.993 0.000 1.000
#> GSM97045 2 0.000 0.993 0.000 1.000
#> GSM97047 2 0.000 0.993 0.000 1.000
#> GSM97025 2 0.000 0.993 0.000 1.000
#> GSM97030 2 0.000 0.993 0.000 1.000
#> GSM97027 2 0.000 0.993 0.000 1.000
#> GSM97033 2 0.000 0.993 0.000 1.000
#> GSM97034 2 0.000 0.993 0.000 1.000
#> GSM97020 2 0.000 0.993 0.000 1.000
#> GSM97026 2 0.000 0.993 0.000 1.000
#> GSM97012 2 0.000 0.993 0.000 1.000
#> GSM97015 2 0.000 0.993 0.000 1.000
#> GSM97016 2 0.000 0.993 0.000 1.000
#> GSM97017 2 0.000 0.993 0.000 1.000
#> GSM97019 2 0.000 0.993 0.000 1.000
#> GSM97022 2 0.000 0.993 0.000 1.000
#> GSM97035 2 0.000 0.993 0.000 1.000
#> GSM97036 2 0.000 0.993 0.000 1.000
#> GSM97039 2 0.000 0.993 0.000 1.000
#> GSM97046 2 0.000 0.993 0.000 1.000
#> GSM97023 2 0.000 0.993 0.000 1.000
#> GSM97029 2 0.000 0.993 0.000 1.000
#> GSM97043 2 0.000 0.993 0.000 1.000
#> GSM97013 2 0.000 0.993 0.000 1.000
#> GSM96956 2 0.000 0.993 0.000 1.000
#> GSM97024 2 0.000 0.993 0.000 1.000
#> GSM97032 2 0.000 0.993 0.000 1.000
#> GSM97044 2 0.000 0.993 0.000 1.000
#> GSM97049 2 0.000 0.993 0.000 1.000
#> GSM96968 2 0.000 0.993 0.000 1.000
#> GSM96971 1 0.000 0.959 1.000 0.000
#> GSM96986 2 0.295 0.945 0.052 0.948
#> GSM97003 2 0.000 0.993 0.000 1.000
#> GSM96957 2 0.000 0.993 0.000 1.000
#> GSM96960 2 0.000 0.993 0.000 1.000
#> GSM96975 2 0.000 0.993 0.000 1.000
#> GSM96998 2 0.000 0.993 0.000 1.000
#> GSM96999 2 0.000 0.993 0.000 1.000
#> GSM97001 2 0.000 0.993 0.000 1.000
#> GSM97005 2 0.000 0.993 0.000 1.000
#> GSM97006 2 0.000 0.993 0.000 1.000
#> GSM97021 2 0.000 0.993 0.000 1.000
#> GSM97028 2 0.000 0.993 0.000 1.000
#> GSM97031 2 0.000 0.993 0.000 1.000
#> GSM97037 2 0.000 0.993 0.000 1.000
#> GSM97018 2 0.000 0.993 0.000 1.000
#> GSM97014 2 0.000 0.993 0.000 1.000
#> GSM97042 2 0.000 0.993 0.000 1.000
#> GSM97040 2 0.000 0.993 0.000 1.000
#> GSM97041 2 0.000 0.993 0.000 1.000
#> GSM96955 2 0.000 0.993 0.000 1.000
#> GSM96990 2 0.000 0.993 0.000 1.000
#> GSM96991 2 0.000 0.993 0.000 1.000
#> GSM97048 2 0.000 0.993 0.000 1.000
#> GSM96963 2 0.000 0.993 0.000 1.000
#> GSM96953 2 0.000 0.993 0.000 1.000
#> GSM96966 1 0.000 0.959 1.000 0.000
#> GSM96979 2 0.295 0.945 0.052 0.948
#> GSM96983 2 0.000 0.993 0.000 1.000
#> GSM96984 2 0.295 0.945 0.052 0.948
#> GSM96994 2 0.000 0.993 0.000 1.000
#> GSM96996 2 0.000 0.993 0.000 1.000
#> GSM96997 2 0.295 0.945 0.052 0.948
#> GSM97007 2 0.204 0.964 0.032 0.968
#> GSM96954 2 0.000 0.993 0.000 1.000
#> GSM96962 2 0.295 0.945 0.052 0.948
#> GSM96969 1 0.000 0.959 1.000 0.000
#> GSM96970 1 0.000 0.959 1.000 0.000
#> GSM96973 1 0.000 0.959 1.000 0.000
#> GSM96976 1 0.000 0.959 1.000 0.000
#> GSM96977 2 0.000 0.993 0.000 1.000
#> GSM96995 2 0.000 0.993 0.000 1.000
#> GSM97002 2 0.295 0.943 0.052 0.948
#> GSM97009 2 0.000 0.993 0.000 1.000
#> GSM97010 2 0.000 0.993 0.000 1.000
#> GSM96974 1 0.000 0.959 1.000 0.000
#> GSM96985 1 0.991 0.198 0.556 0.444
#> GSM96959 2 0.000 0.993 0.000 1.000
#> GSM96972 1 0.000 0.959 1.000 0.000
#> GSM96978 2 0.141 0.976 0.020 0.980
#> GSM96967 1 0.000 0.959 1.000 0.000
#> GSM96987 2 0.000 0.993 0.000 1.000
#> GSM97011 2 0.000 0.993 0.000 1.000
#> GSM96964 2 0.000 0.993 0.000 1.000
#> GSM96965 1 0.000 0.959 1.000 0.000
#> GSM96981 2 0.000 0.993 0.000 1.000
#> GSM96982 2 0.000 0.993 0.000 1.000
#> GSM96988 2 0.000 0.993 0.000 1.000
#> GSM97000 2 0.000 0.993 0.000 1.000
#> GSM97004 2 0.745 0.720 0.212 0.788
#> GSM97008 2 0.000 0.993 0.000 1.000
#> GSM96950 2 0.000 0.993 0.000 1.000
#> GSM96980 1 0.000 0.959 1.000 0.000
#> GSM96989 2 0.000 0.993 0.000 1.000
#> GSM96992 2 0.000 0.993 0.000 1.000
#> GSM96993 2 0.000 0.993 0.000 1.000
#> GSM96958 2 0.000 0.993 0.000 1.000
#> GSM96951 2 0.000 0.993 0.000 1.000
#> GSM96952 2 0.000 0.993 0.000 1.000
#> GSM96961 2 0.000 0.993 0.000 1.000
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM97038 2 0.0000 0.8279 0.000 1.000 0.000
#> GSM97045 2 0.0000 0.8279 0.000 1.000 0.000
#> GSM97047 2 0.5138 0.7517 0.252 0.748 0.000
#> GSM97025 2 0.0000 0.8279 0.000 1.000 0.000
#> GSM97030 2 0.5098 0.7551 0.248 0.752 0.000
#> GSM97027 2 0.0000 0.8279 0.000 1.000 0.000
#> GSM97033 2 0.0000 0.8279 0.000 1.000 0.000
#> GSM97034 2 0.5138 0.7517 0.252 0.748 0.000
#> GSM97020 2 0.0000 0.8279 0.000 1.000 0.000
#> GSM97026 2 0.5138 0.7517 0.252 0.748 0.000
#> GSM97012 2 0.0000 0.8279 0.000 1.000 0.000
#> GSM97015 2 0.5138 0.7517 0.252 0.748 0.000
#> GSM97016 2 0.0000 0.8279 0.000 1.000 0.000
#> GSM97017 1 0.6008 0.4203 0.628 0.372 0.000
#> GSM97019 2 0.0000 0.8279 0.000 1.000 0.000
#> GSM97022 2 0.0000 0.8279 0.000 1.000 0.000
#> GSM97035 2 0.0000 0.8279 0.000 1.000 0.000
#> GSM97036 1 0.5621 0.5739 0.692 0.308 0.000
#> GSM97039 2 0.0000 0.8279 0.000 1.000 0.000
#> GSM97046 2 0.0000 0.8279 0.000 1.000 0.000
#> GSM97023 1 0.4974 0.6753 0.764 0.236 0.000
#> GSM97029 1 0.5621 0.5739 0.692 0.308 0.000
#> GSM97043 2 0.0000 0.8279 0.000 1.000 0.000
#> GSM97013 1 0.0000 0.8731 1.000 0.000 0.000
#> GSM96956 2 0.5098 0.7551 0.248 0.752 0.000
#> GSM97024 2 0.0000 0.8279 0.000 1.000 0.000
#> GSM97032 2 0.4796 0.7688 0.220 0.780 0.000
#> GSM97044 2 0.5138 0.7517 0.252 0.748 0.000
#> GSM97049 2 0.0000 0.8279 0.000 1.000 0.000
#> GSM96968 1 0.0000 0.8731 1.000 0.000 0.000
#> GSM96971 3 0.0000 0.9659 0.000 0.000 1.000
#> GSM96986 1 0.0000 0.8731 1.000 0.000 0.000
#> GSM97003 1 0.0000 0.8731 1.000 0.000 0.000
#> GSM96957 1 0.5621 0.5739 0.692 0.308 0.000
#> GSM96960 1 0.0000 0.8731 1.000 0.000 0.000
#> GSM96975 1 0.0000 0.8731 1.000 0.000 0.000
#> GSM96998 1 0.0000 0.8731 1.000 0.000 0.000
#> GSM96999 1 0.0000 0.8731 1.000 0.000 0.000
#> GSM97001 1 0.5621 0.5739 0.692 0.308 0.000
#> GSM97005 1 0.0000 0.8731 1.000 0.000 0.000
#> GSM97006 1 0.0000 0.8731 1.000 0.000 0.000
#> GSM97021 1 0.5650 0.5658 0.688 0.312 0.000
#> GSM97028 2 0.5138 0.7517 0.252 0.748 0.000
#> GSM97031 1 0.0000 0.8731 1.000 0.000 0.000
#> GSM97037 2 0.5098 0.7551 0.248 0.752 0.000
#> GSM97018 2 0.5058 0.7574 0.244 0.756 0.000
#> GSM97014 2 0.5138 0.7517 0.252 0.748 0.000
#> GSM97042 2 0.0000 0.8279 0.000 1.000 0.000
#> GSM97040 2 0.5138 0.7517 0.252 0.748 0.000
#> GSM97041 1 0.5621 0.5739 0.692 0.308 0.000
#> GSM96955 2 0.0000 0.8279 0.000 1.000 0.000
#> GSM96990 2 0.5138 0.7517 0.252 0.748 0.000
#> GSM96991 2 0.0000 0.8279 0.000 1.000 0.000
#> GSM97048 2 0.0000 0.8279 0.000 1.000 0.000
#> GSM96963 2 0.0000 0.8279 0.000 1.000 0.000
#> GSM96953 2 0.0000 0.8279 0.000 1.000 0.000
#> GSM96966 3 0.0000 0.9659 0.000 0.000 1.000
#> GSM96979 1 0.0000 0.8731 1.000 0.000 0.000
#> GSM96983 2 0.5098 0.7551 0.248 0.752 0.000
#> GSM96984 1 0.0000 0.8731 1.000 0.000 0.000
#> GSM96994 2 0.5138 0.7517 0.252 0.748 0.000
#> GSM96996 1 0.0237 0.8714 0.996 0.004 0.000
#> GSM96997 1 0.0000 0.8731 1.000 0.000 0.000
#> GSM97007 1 0.6244 -0.0444 0.560 0.440 0.000
#> GSM96954 1 0.0000 0.8731 1.000 0.000 0.000
#> GSM96962 1 0.0000 0.8731 1.000 0.000 0.000
#> GSM96969 3 0.0000 0.9659 0.000 0.000 1.000
#> GSM96970 3 0.0000 0.9659 0.000 0.000 1.000
#> GSM96973 3 0.0000 0.9659 0.000 0.000 1.000
#> GSM96976 3 0.0000 0.9659 0.000 0.000 1.000
#> GSM96977 1 0.0000 0.8731 1.000 0.000 0.000
#> GSM96995 2 0.6305 0.1134 0.484 0.516 0.000
#> GSM97002 1 0.0000 0.8731 1.000 0.000 0.000
#> GSM97009 1 0.5926 0.4638 0.644 0.356 0.000
#> GSM97010 1 0.0000 0.8731 1.000 0.000 0.000
#> GSM96974 3 0.0000 0.9659 0.000 0.000 1.000
#> GSM96985 1 0.0000 0.8731 1.000 0.000 0.000
#> GSM96959 2 0.5138 0.7517 0.252 0.748 0.000
#> GSM96972 3 0.0000 0.9659 0.000 0.000 1.000
#> GSM96978 1 0.0000 0.8731 1.000 0.000 0.000
#> GSM96967 3 0.0000 0.9659 0.000 0.000 1.000
#> GSM96987 1 0.4452 0.7262 0.808 0.192 0.000
#> GSM97011 1 0.5621 0.5739 0.692 0.308 0.000
#> GSM96964 1 0.1860 0.8448 0.948 0.052 0.000
#> GSM96965 3 0.0000 0.9659 0.000 0.000 1.000
#> GSM96981 1 0.0000 0.8731 1.000 0.000 0.000
#> GSM96982 1 0.0000 0.8731 1.000 0.000 0.000
#> GSM96988 1 0.4346 0.7345 0.816 0.184 0.000
#> GSM97000 1 0.0000 0.8731 1.000 0.000 0.000
#> GSM97004 1 0.0000 0.8731 1.000 0.000 0.000
#> GSM97008 1 0.5621 0.5739 0.692 0.308 0.000
#> GSM96950 1 0.0000 0.8731 1.000 0.000 0.000
#> GSM96980 3 0.5254 0.5476 0.264 0.000 0.736
#> GSM96989 1 0.0000 0.8731 1.000 0.000 0.000
#> GSM96992 1 0.0000 0.8731 1.000 0.000 0.000
#> GSM96993 1 0.5621 0.5739 0.692 0.308 0.000
#> GSM96958 1 0.0000 0.8731 1.000 0.000 0.000
#> GSM96951 1 0.0000 0.8731 1.000 0.000 0.000
#> GSM96952 1 0.0000 0.8731 1.000 0.000 0.000
#> GSM96961 1 0.2165 0.8367 0.936 0.064 0.000
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM97038 2 0.000 0.987 0.000 1.000 0.000 0
#> GSM97045 2 0.000 0.987 0.000 1.000 0.000 0
#> GSM97047 2 0.000 0.987 0.000 1.000 0.000 0
#> GSM97025 2 0.000 0.987 0.000 1.000 0.000 0
#> GSM97030 2 0.000 0.987 0.000 1.000 0.000 0
#> GSM97027 2 0.000 0.987 0.000 1.000 0.000 0
#> GSM97033 2 0.000 0.987 0.000 1.000 0.000 0
#> GSM97034 2 0.000 0.987 0.000 1.000 0.000 0
#> GSM97020 2 0.000 0.987 0.000 1.000 0.000 0
#> GSM97026 2 0.000 0.987 0.000 1.000 0.000 0
#> GSM97012 2 0.000 0.987 0.000 1.000 0.000 0
#> GSM97015 2 0.000 0.987 0.000 1.000 0.000 0
#> GSM97016 2 0.000 0.987 0.000 1.000 0.000 0
#> GSM97017 1 0.000 0.962 1.000 0.000 0.000 0
#> GSM97019 2 0.000 0.987 0.000 1.000 0.000 0
#> GSM97022 2 0.000 0.987 0.000 1.000 0.000 0
#> GSM97035 2 0.000 0.987 0.000 1.000 0.000 0
#> GSM97036 1 0.000 0.962 1.000 0.000 0.000 0
#> GSM97039 2 0.000 0.987 0.000 1.000 0.000 0
#> GSM97046 2 0.000 0.987 0.000 1.000 0.000 0
#> GSM97023 1 0.000 0.962 1.000 0.000 0.000 0
#> GSM97029 1 0.000 0.962 1.000 0.000 0.000 0
#> GSM97043 2 0.000 0.987 0.000 1.000 0.000 0
#> GSM97013 1 0.000 0.962 1.000 0.000 0.000 0
#> GSM96956 2 0.000 0.987 0.000 1.000 0.000 0
#> GSM97024 2 0.000 0.987 0.000 1.000 0.000 0
#> GSM97032 2 0.000 0.987 0.000 1.000 0.000 0
#> GSM97044 2 0.000 0.987 0.000 1.000 0.000 0
#> GSM97049 2 0.000 0.987 0.000 1.000 0.000 0
#> GSM96968 3 0.000 1.000 0.000 0.000 1.000 0
#> GSM96971 4 0.000 1.000 0.000 0.000 0.000 1
#> GSM96986 3 0.000 1.000 0.000 0.000 1.000 0
#> GSM97003 1 0.450 0.547 0.684 0.000 0.316 0
#> GSM96957 1 0.000 0.962 1.000 0.000 0.000 0
#> GSM96960 1 0.000 0.962 1.000 0.000 0.000 0
#> GSM96975 1 0.000 0.962 1.000 0.000 0.000 0
#> GSM96998 1 0.000 0.962 1.000 0.000 0.000 0
#> GSM96999 1 0.000 0.962 1.000 0.000 0.000 0
#> GSM97001 1 0.000 0.962 1.000 0.000 0.000 0
#> GSM97005 1 0.000 0.962 1.000 0.000 0.000 0
#> GSM97006 1 0.000 0.962 1.000 0.000 0.000 0
#> GSM97021 1 0.000 0.962 1.000 0.000 0.000 0
#> GSM97028 2 0.000 0.987 0.000 1.000 0.000 0
#> GSM97031 1 0.000 0.962 1.000 0.000 0.000 0
#> GSM97037 2 0.000 0.987 0.000 1.000 0.000 0
#> GSM97018 2 0.000 0.987 0.000 1.000 0.000 0
#> GSM97014 2 0.000 0.987 0.000 1.000 0.000 0
#> GSM97042 2 0.000 0.987 0.000 1.000 0.000 0
#> GSM97040 1 0.473 0.436 0.636 0.364 0.000 0
#> GSM97041 1 0.000 0.962 1.000 0.000 0.000 0
#> GSM96955 2 0.000 0.987 0.000 1.000 0.000 0
#> GSM96990 2 0.000 0.987 0.000 1.000 0.000 0
#> GSM96991 2 0.000 0.987 0.000 1.000 0.000 0
#> GSM97048 2 0.000 0.987 0.000 1.000 0.000 0
#> GSM96963 2 0.000 0.987 0.000 1.000 0.000 0
#> GSM96953 2 0.000 0.987 0.000 1.000 0.000 0
#> GSM96966 4 0.000 1.000 0.000 0.000 0.000 1
#> GSM96979 3 0.000 1.000 0.000 0.000 1.000 0
#> GSM96983 2 0.000 0.987 0.000 1.000 0.000 0
#> GSM96984 3 0.000 1.000 0.000 0.000 1.000 0
#> GSM96994 2 0.000 0.987 0.000 1.000 0.000 0
#> GSM96996 1 0.000 0.962 1.000 0.000 0.000 0
#> GSM96997 3 0.000 1.000 0.000 0.000 1.000 0
#> GSM97007 3 0.000 1.000 0.000 0.000 1.000 0
#> GSM96954 3 0.000 1.000 0.000 0.000 1.000 0
#> GSM96962 3 0.000 1.000 0.000 0.000 1.000 0
#> GSM96969 4 0.000 1.000 0.000 0.000 0.000 1
#> GSM96970 4 0.000 1.000 0.000 0.000 0.000 1
#> GSM96973 4 0.000 1.000 0.000 0.000 0.000 1
#> GSM96976 4 0.000 1.000 0.000 0.000 0.000 1
#> GSM96977 1 0.000 0.962 1.000 0.000 0.000 0
#> GSM96995 2 0.391 0.661 0.232 0.768 0.000 0
#> GSM97002 1 0.000 0.962 1.000 0.000 0.000 0
#> GSM97009 1 0.454 0.507 0.676 0.324 0.000 0
#> GSM97010 1 0.000 0.962 1.000 0.000 0.000 0
#> GSM96974 4 0.000 1.000 0.000 0.000 0.000 1
#> GSM96985 1 0.000 0.962 1.000 0.000 0.000 0
#> GSM96959 2 0.302 0.790 0.148 0.852 0.000 0
#> GSM96972 4 0.000 1.000 0.000 0.000 0.000 1
#> GSM96978 3 0.000 1.000 0.000 0.000 1.000 0
#> GSM96967 4 0.000 1.000 0.000 0.000 0.000 1
#> GSM96987 1 0.000 0.962 1.000 0.000 0.000 0
#> GSM97011 1 0.000 0.962 1.000 0.000 0.000 0
#> GSM96964 1 0.000 0.962 1.000 0.000 0.000 0
#> GSM96965 4 0.000 1.000 0.000 0.000 0.000 1
#> GSM96981 1 0.000 0.962 1.000 0.000 0.000 0
#> GSM96982 1 0.000 0.962 1.000 0.000 0.000 0
#> GSM96988 1 0.000 0.962 1.000 0.000 0.000 0
#> GSM97000 1 0.466 0.483 0.652 0.000 0.348 0
#> GSM97004 1 0.000 0.962 1.000 0.000 0.000 0
#> GSM97008 1 0.000 0.962 1.000 0.000 0.000 0
#> GSM96950 1 0.000 0.962 1.000 0.000 0.000 0
#> GSM96980 1 0.000 0.962 1.000 0.000 0.000 0
#> GSM96989 1 0.000 0.962 1.000 0.000 0.000 0
#> GSM96992 1 0.000 0.962 1.000 0.000 0.000 0
#> GSM96993 1 0.000 0.962 1.000 0.000 0.000 0
#> GSM96958 1 0.000 0.962 1.000 0.000 0.000 0
#> GSM96951 1 0.000 0.962 1.000 0.000 0.000 0
#> GSM96952 1 0.000 0.962 1.000 0.000 0.000 0
#> GSM96961 1 0.000 0.962 1.000 0.000 0.000 0
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM97038 2 0.0000 0.9741 0.000 1.000 0.000 0.000 0.000
#> GSM97045 2 0.0000 0.9741 0.000 1.000 0.000 0.000 0.000
#> GSM97047 5 0.2424 0.8328 0.000 0.132 0.000 0.000 0.868
#> GSM97025 2 0.0000 0.9741 0.000 1.000 0.000 0.000 0.000
#> GSM97030 5 0.2773 0.8047 0.000 0.164 0.000 0.000 0.836
#> GSM97027 2 0.0000 0.9741 0.000 1.000 0.000 0.000 0.000
#> GSM97033 2 0.0000 0.9741 0.000 1.000 0.000 0.000 0.000
#> GSM97034 5 0.2424 0.8328 0.000 0.132 0.000 0.000 0.868
#> GSM97020 2 0.0000 0.9741 0.000 1.000 0.000 0.000 0.000
#> GSM97026 5 0.6122 0.3737 0.348 0.140 0.000 0.000 0.512
#> GSM97012 2 0.0000 0.9741 0.000 1.000 0.000 0.000 0.000
#> GSM97015 5 0.2424 0.8328 0.000 0.132 0.000 0.000 0.868
#> GSM97016 2 0.0000 0.9741 0.000 1.000 0.000 0.000 0.000
#> GSM97017 1 0.4297 0.3690 0.528 0.000 0.000 0.000 0.472
#> GSM97019 2 0.0000 0.9741 0.000 1.000 0.000 0.000 0.000
#> GSM97022 2 0.0000 0.9741 0.000 1.000 0.000 0.000 0.000
#> GSM97035 2 0.0000 0.9741 0.000 1.000 0.000 0.000 0.000
#> GSM97036 1 0.4219 0.4798 0.584 0.000 0.000 0.000 0.416
#> GSM97039 2 0.0000 0.9741 0.000 1.000 0.000 0.000 0.000
#> GSM97046 2 0.0000 0.9741 0.000 1.000 0.000 0.000 0.000
#> GSM97023 1 0.2127 0.8022 0.892 0.000 0.000 0.000 0.108
#> GSM97029 1 0.4297 0.3690 0.528 0.000 0.000 0.000 0.472
#> GSM97043 2 0.0000 0.9741 0.000 1.000 0.000 0.000 0.000
#> GSM97013 1 0.0000 0.8329 1.000 0.000 0.000 0.000 0.000
#> GSM96956 5 0.2516 0.8268 0.000 0.140 0.000 0.000 0.860
#> GSM97024 2 0.4283 0.0704 0.000 0.544 0.000 0.000 0.456
#> GSM97032 5 0.2424 0.8328 0.000 0.132 0.000 0.000 0.868
#> GSM97044 5 0.2424 0.8328 0.000 0.132 0.000 0.000 0.868
#> GSM97049 5 0.4210 0.4142 0.000 0.412 0.000 0.000 0.588
#> GSM96968 1 0.4302 0.2264 0.520 0.000 0.480 0.000 0.000
#> GSM96971 4 0.1270 0.9633 0.000 0.000 0.052 0.948 0.000
#> GSM96986 3 0.0000 0.9638 0.000 0.000 1.000 0.000 0.000
#> GSM97003 1 0.4182 0.3778 0.600 0.000 0.400 0.000 0.000
#> GSM96957 1 0.2852 0.7693 0.828 0.000 0.000 0.000 0.172
#> GSM96960 1 0.0000 0.8329 1.000 0.000 0.000 0.000 0.000
#> GSM96975 1 0.0000 0.8329 1.000 0.000 0.000 0.000 0.000
#> GSM96998 1 0.0000 0.8329 1.000 0.000 0.000 0.000 0.000
#> GSM96999 1 0.0000 0.8329 1.000 0.000 0.000 0.000 0.000
#> GSM97001 1 0.3143 0.7451 0.796 0.000 0.000 0.000 0.204
#> GSM97005 1 0.0162 0.8324 0.996 0.000 0.000 0.000 0.004
#> GSM97006 1 0.1121 0.8180 0.956 0.000 0.044 0.000 0.000
#> GSM97021 1 0.4305 0.3282 0.512 0.000 0.000 0.000 0.488
#> GSM97028 5 0.2424 0.8328 0.000 0.132 0.000 0.000 0.868
#> GSM97031 1 0.0579 0.8305 0.984 0.000 0.008 0.000 0.008
#> GSM97037 5 0.2424 0.8328 0.000 0.132 0.000 0.000 0.868
#> GSM97018 5 0.2424 0.8328 0.000 0.132 0.000 0.000 0.868
#> GSM97014 5 0.5974 0.4450 0.320 0.132 0.000 0.000 0.548
#> GSM97042 2 0.0000 0.9741 0.000 1.000 0.000 0.000 0.000
#> GSM97040 5 0.4273 -0.1904 0.448 0.000 0.000 0.000 0.552
#> GSM97041 1 0.4256 0.4435 0.564 0.000 0.000 0.000 0.436
#> GSM96955 2 0.0000 0.9741 0.000 1.000 0.000 0.000 0.000
#> GSM96990 5 0.2424 0.8328 0.000 0.132 0.000 0.000 0.868
#> GSM96991 2 0.0000 0.9741 0.000 1.000 0.000 0.000 0.000
#> GSM97048 2 0.0000 0.9741 0.000 1.000 0.000 0.000 0.000
#> GSM96963 2 0.0000 0.9741 0.000 1.000 0.000 0.000 0.000
#> GSM96953 2 0.0000 0.9741 0.000 1.000 0.000 0.000 0.000
#> GSM96966 4 0.1121 0.9690 0.000 0.000 0.044 0.956 0.000
#> GSM96979 3 0.0000 0.9638 0.000 0.000 1.000 0.000 0.000
#> GSM96983 5 0.2424 0.8328 0.000 0.132 0.000 0.000 0.868
#> GSM96984 3 0.0000 0.9638 0.000 0.000 1.000 0.000 0.000
#> GSM96994 5 0.2727 0.8181 0.000 0.116 0.016 0.000 0.868
#> GSM96996 1 0.1121 0.8232 0.956 0.000 0.000 0.000 0.044
#> GSM96997 3 0.0000 0.9638 0.000 0.000 1.000 0.000 0.000
#> GSM97007 3 0.1121 0.9180 0.000 0.000 0.956 0.000 0.044
#> GSM96954 3 0.1965 0.8556 0.096 0.000 0.904 0.000 0.000
#> GSM96962 3 0.0000 0.9638 0.000 0.000 1.000 0.000 0.000
#> GSM96969 4 0.0000 0.9798 0.000 0.000 0.000 1.000 0.000
#> GSM96970 4 0.0000 0.9798 0.000 0.000 0.000 1.000 0.000
#> GSM96973 4 0.0000 0.9798 0.000 0.000 0.000 1.000 0.000
#> GSM96976 4 0.1121 0.9690 0.000 0.000 0.044 0.956 0.000
#> GSM96977 1 0.0000 0.8329 1.000 0.000 0.000 0.000 0.000
#> GSM96995 5 0.5074 0.6949 0.168 0.132 0.000 0.000 0.700
#> GSM97002 1 0.0000 0.8329 1.000 0.000 0.000 0.000 0.000
#> GSM97009 5 0.5889 0.1057 0.428 0.100 0.000 0.000 0.472
#> GSM97010 1 0.0000 0.8329 1.000 0.000 0.000 0.000 0.000
#> GSM96974 4 0.0000 0.9798 0.000 0.000 0.000 1.000 0.000
#> GSM96985 1 0.0703 0.8228 0.976 0.000 0.000 0.000 0.024
#> GSM96959 5 0.2424 0.8328 0.000 0.132 0.000 0.000 0.868
#> GSM96972 4 0.0000 0.9798 0.000 0.000 0.000 1.000 0.000
#> GSM96978 1 0.4306 0.1953 0.508 0.000 0.492 0.000 0.000
#> GSM96967 4 0.0000 0.9798 0.000 0.000 0.000 1.000 0.000
#> GSM96987 1 0.2648 0.7825 0.848 0.000 0.000 0.000 0.152
#> GSM97011 1 0.4294 0.3783 0.532 0.000 0.000 0.000 0.468
#> GSM96964 1 0.2127 0.8022 0.892 0.000 0.000 0.000 0.108
#> GSM96965 4 0.1121 0.9690 0.000 0.000 0.044 0.956 0.000
#> GSM96981 1 0.0000 0.8329 1.000 0.000 0.000 0.000 0.000
#> GSM96982 1 0.0000 0.8329 1.000 0.000 0.000 0.000 0.000
#> GSM96988 1 0.3366 0.6599 0.768 0.000 0.000 0.000 0.232
#> GSM97000 1 0.4341 0.4446 0.628 0.000 0.364 0.000 0.008
#> GSM97004 1 0.0000 0.8329 1.000 0.000 0.000 0.000 0.000
#> GSM97008 1 0.3508 0.7044 0.748 0.000 0.000 0.000 0.252
#> GSM96950 1 0.0000 0.8329 1.000 0.000 0.000 0.000 0.000
#> GSM96980 1 0.1818 0.8072 0.932 0.000 0.044 0.000 0.024
#> GSM96989 1 0.0000 0.8329 1.000 0.000 0.000 0.000 0.000
#> GSM96992 1 0.0000 0.8329 1.000 0.000 0.000 0.000 0.000
#> GSM96993 1 0.4294 0.3783 0.532 0.000 0.000 0.000 0.468
#> GSM96958 1 0.0000 0.8329 1.000 0.000 0.000 0.000 0.000
#> GSM96951 1 0.0000 0.8329 1.000 0.000 0.000 0.000 0.000
#> GSM96952 1 0.0000 0.8329 1.000 0.000 0.000 0.000 0.000
#> GSM96961 1 0.2127 0.8022 0.892 0.000 0.000 0.000 0.108
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM97038 2 0.0000 0.96951 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97045 2 0.0000 0.96951 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97047 3 0.0000 0.97444 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM97025 2 0.0000 0.96951 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97030 3 0.0000 0.97444 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM97027 2 0.0000 0.96951 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97033 2 0.0000 0.96951 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97034 3 0.0000 0.97444 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM97020 2 0.0000 0.96951 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97026 5 0.2213 0.83068 0.004 0.008 0.100 0.000 0.888 0.000
#> GSM97012 2 0.0000 0.96951 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97015 3 0.0000 0.97444 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM97016 2 0.0000 0.96951 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97017 5 0.0777 0.88432 0.004 0.000 0.024 0.000 0.972 0.000
#> GSM97019 2 0.0000 0.96951 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97022 2 0.0000 0.96951 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97035 2 0.0000 0.96951 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97036 5 0.0777 0.88432 0.004 0.000 0.024 0.000 0.972 0.000
#> GSM97039 2 0.0000 0.96951 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97046 2 0.0000 0.96951 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97023 1 0.0632 0.90968 0.976 0.000 0.000 0.000 0.024 0.000
#> GSM97029 5 0.0777 0.88432 0.004 0.000 0.024 0.000 0.972 0.000
#> GSM97043 2 0.0000 0.96951 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97013 1 0.0146 0.91462 0.996 0.000 0.000 0.000 0.004 0.000
#> GSM96956 3 0.0632 0.94388 0.000 0.024 0.976 0.000 0.000 0.000
#> GSM97024 2 0.2854 0.73530 0.000 0.792 0.208 0.000 0.000 0.000
#> GSM97032 3 0.0000 0.97444 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM97044 3 0.0000 0.97444 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM97049 2 0.3695 0.41371 0.000 0.624 0.376 0.000 0.000 0.000
#> GSM96968 6 0.1141 0.96056 0.052 0.000 0.000 0.000 0.000 0.948
#> GSM96971 4 0.0632 0.98387 0.000 0.000 0.000 0.976 0.000 0.024
#> GSM96986 6 0.0000 0.96981 0.000 0.000 0.000 0.000 0.000 1.000
#> GSM97003 1 0.3737 0.32693 0.608 0.000 0.000 0.000 0.000 0.392
#> GSM96957 1 0.4405 -0.00718 0.504 0.000 0.024 0.000 0.472 0.000
#> GSM96960 1 0.0632 0.90968 0.976 0.000 0.000 0.000 0.024 0.000
#> GSM96975 1 0.2883 0.70013 0.788 0.000 0.000 0.000 0.212 0.000
#> GSM96998 1 0.0000 0.91569 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM96999 1 0.0146 0.91462 0.996 0.000 0.000 0.000 0.004 0.000
#> GSM97001 5 0.0405 0.87449 0.004 0.000 0.008 0.000 0.988 0.000
#> GSM97005 1 0.0000 0.91569 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM97006 1 0.0692 0.90875 0.976 0.000 0.000 0.000 0.004 0.020
#> GSM97021 5 0.0777 0.88432 0.004 0.000 0.024 0.000 0.972 0.000
#> GSM97028 3 0.0000 0.97444 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM97031 1 0.0000 0.91569 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM97037 3 0.0000 0.97444 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM97018 3 0.0000 0.97444 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM97014 5 0.2772 0.74763 0.004 0.000 0.180 0.000 0.816 0.000
#> GSM97042 2 0.0000 0.96951 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97040 5 0.0777 0.88432 0.004 0.000 0.024 0.000 0.972 0.000
#> GSM97041 5 0.0777 0.88432 0.004 0.000 0.024 0.000 0.972 0.000
#> GSM96955 2 0.0000 0.96951 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM96990 3 0.0000 0.97444 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM96991 2 0.0000 0.96951 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM97048 2 0.0000 0.96951 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM96963 2 0.0000 0.96951 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM96953 2 0.0000 0.96951 0.000 1.000 0.000 0.000 0.000 0.000
#> GSM96966 4 0.0458 0.98857 0.000 0.000 0.000 0.984 0.000 0.016
#> GSM96979 6 0.0000 0.96981 0.000 0.000 0.000 0.000 0.000 1.000
#> GSM96983 3 0.0000 0.97444 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM96984 6 0.0000 0.96981 0.000 0.000 0.000 0.000 0.000 1.000
#> GSM96994 3 0.0000 0.97444 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM96996 5 0.1556 0.82393 0.080 0.000 0.000 0.000 0.920 0.000
#> GSM96997 6 0.0000 0.96981 0.000 0.000 0.000 0.000 0.000 1.000
#> GSM97007 6 0.0632 0.96695 0.024 0.000 0.000 0.000 0.000 0.976
#> GSM96954 6 0.1075 0.96228 0.048 0.000 0.000 0.000 0.000 0.952
#> GSM96962 6 0.0000 0.96981 0.000 0.000 0.000 0.000 0.000 1.000
#> GSM96969 4 0.0000 0.99293 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM96970 4 0.0000 0.99293 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM96973 4 0.0000 0.99293 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM96976 4 0.0547 0.98670 0.000 0.000 0.000 0.980 0.000 0.020
#> GSM96977 1 0.0000 0.91569 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM96995 3 0.0000 0.97444 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM97002 1 0.0363 0.91406 0.988 0.000 0.000 0.000 0.012 0.000
#> GSM97009 3 0.3601 0.50395 0.004 0.000 0.684 0.000 0.312 0.000
#> GSM97010 1 0.0000 0.91569 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM96974 4 0.0000 0.99293 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM96985 5 0.3864 0.05109 0.480 0.000 0.000 0.000 0.520 0.000
#> GSM96959 3 0.0000 0.97444 0.000 0.000 1.000 0.000 0.000 0.000
#> GSM96972 4 0.0000 0.99293 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM96978 6 0.1141 0.96056 0.052 0.000 0.000 0.000 0.000 0.948
#> GSM96967 4 0.0000 0.99293 0.000 0.000 0.000 1.000 0.000 0.000
#> GSM96987 5 0.0777 0.88432 0.004 0.000 0.024 0.000 0.972 0.000
#> GSM97011 5 0.0777 0.88432 0.004 0.000 0.024 0.000 0.972 0.000
#> GSM96964 1 0.0632 0.90968 0.976 0.000 0.000 0.000 0.024 0.000
#> GSM96965 4 0.0363 0.99013 0.000 0.000 0.000 0.988 0.000 0.012
#> GSM96981 5 0.3789 0.25948 0.416 0.000 0.000 0.000 0.584 0.000
#> GSM96982 1 0.0363 0.91411 0.988 0.000 0.000 0.000 0.012 0.000
#> GSM96988 5 0.2950 0.78223 0.148 0.000 0.024 0.000 0.828 0.000
#> GSM97000 6 0.1204 0.95654 0.056 0.000 0.000 0.000 0.000 0.944
#> GSM97004 1 0.0000 0.91569 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM97008 1 0.3876 0.53482 0.700 0.000 0.024 0.000 0.276 0.000
#> GSM96950 1 0.0146 0.91462 0.996 0.000 0.000 0.000 0.004 0.000
#> GSM96980 1 0.0692 0.90875 0.976 0.000 0.000 0.000 0.004 0.020
#> GSM96989 1 0.3563 0.46204 0.664 0.000 0.000 0.000 0.336 0.000
#> GSM96992 1 0.0260 0.91500 0.992 0.000 0.000 0.000 0.008 0.000
#> GSM96993 5 0.0777 0.88432 0.004 0.000 0.024 0.000 0.972 0.000
#> GSM96958 1 0.0146 0.91462 0.996 0.000 0.000 0.000 0.004 0.000
#> GSM96951 1 0.0000 0.91569 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM96952 1 0.0000 0.91569 1.000 0.000 0.000 0.000 0.000 0.000
#> GSM96961 1 0.0632 0.90968 0.976 0.000 0.000 0.000 0.024 0.000
Heatmaps for the consensus matrix. It visualizes the probability of two samples to be in a same group.
consensus_heatmap(res, k = 2)
consensus_heatmap(res, k = 3)
consensus_heatmap(res, k = 4)
consensus_heatmap(res, k = 5)
consensus_heatmap(res, k = 6)
Heatmaps for the membership of samples in all partitions to see how consistent they are:
membership_heatmap(res, k = 2)
membership_heatmap(res, k = 3)
membership_heatmap(res, k = 4)
membership_heatmap(res, k = 5)
membership_heatmap(res, k = 6)
As soon as we have had the classes for columns, we can look for signatures which are significantly different between classes which can be candidate marks for certain classes. Following are the heatmaps for signatures.
Signature heatmaps where rows are scaled:
get_signatures(res, k = 2)
get_signatures(res, k = 3)
get_signatures(res, k = 4)
get_signatures(res, k = 5)
get_signatures(res, k = 6)
Signature heatmaps where rows are not scaled:
get_signatures(res, k = 2, scale_rows = FALSE)
get_signatures(res, k = 3, scale_rows = FALSE)
get_signatures(res, k = 4, scale_rows = FALSE)
get_signatures(res, k = 5, scale_rows = FALSE)
get_signatures(res, k = 6, scale_rows = FALSE)
Compare the overlap of signatures from different k:
compare_signatures(res)
get_signature() returns a data frame invisibly. TO get the list of signatures, the function
call should be assigned to a variable explicitly. In following code, if plot argument is set
to FALSE, no heatmap is plotted while only the differential analysis is performed.
# code only for demonstration
tb = get_signature(res, k = ..., plot = FALSE)
An example of the output of tb is:
#> which_row fdr mean_1 mean_2 scaled_mean_1 scaled_mean_2 km
#> 1 38 0.042760348 8.373488 9.131774 -0.5533452 0.5164555 1
#> 2 40 0.018707592 7.106213 8.469186 -0.6173731 0.5762149 1
#> 3 55 0.019134737 10.221463 11.207825 -0.6159697 0.5749050 1
#> 4 59 0.006059896 5.921854 7.869574 -0.6899429 0.6439467 1
#> 5 60 0.018055526 8.928898 10.211722 -0.6204761 0.5791110 1
#> 6 98 0.009384629 15.714769 14.887706 0.6635654 -0.6193277 2
...
The columns in tb are:
which_row: row indices corresponding to the input matrix.fdr: FDR for the differential test. mean_x: The mean value in group x.scaled_mean_x: The mean value in group x after rows are scaled.km: Row groups if k-means clustering is applied to rows.UMAP plot which shows how samples are separated.
dimension_reduction(res, k = 2, method = "UMAP")
dimension_reduction(res, k = 3, method = "UMAP")
dimension_reduction(res, k = 4, method = "UMAP")
dimension_reduction(res, k = 5, method = "UMAP")
dimension_reduction(res, k = 6, method = "UMAP")
Following heatmap shows how subgroups are split when increasing k:
collect_classes(res)
Test correlation between subgroups and known annotations. If the known annotation is numeric, one-way ANOVA test is applied, and if the known annotation is discrete, chi-squared contingency table test is applied.
test_to_known_factors(res)
#> n disease.state(p) specimen(p) cell.type(p) other(p) k
#> ATC:mclust 99 3.22e-02 0.424 8.13e-04 0.463 2
#> ATC:mclust 96 4.68e-06 0.136 2.76e-16 0.124 3
#> ATC:mclust 98 2.37e-04 0.321 3.62e-21 0.124 4
#> ATC:mclust 83 5.25e-06 0.148 1.34e-19 0.022 5
#> ATC:mclust 94 8.30e-05 0.399 1.53e-18 0.256 6
If matrix rows can be associated to genes, consider to use functional_enrichment(res,
...) to perform function enrichment for the signature genes. See this vignette for more detailed explanations.
The object with results only for a single top-value method and a single partition method can be extracted as:
res = res_list["ATC", "NMF"]
# you can also extract it by
# res = res_list["ATC:NMF"]
A summary of res and all the functions that can be applied to it:
res
#> A 'ConsensusPartition' object with k = 2, 3, 4, 5, 6.
#> On a matrix with 21168 rows and 100 columns.
#> Top rows (1000, 2000, 3000, 4000, 5000) are extracted by 'ATC' method.
#> Subgroups are detected by 'NMF' method.
#> Performed in total 1250 partitions by row resampling.
#> Best k for subgroups seems to be 2.
#>
#> Following methods can be applied to this 'ConsensusPartition' object:
#> [1] "cola_report" "collect_classes" "collect_plots"
#> [4] "collect_stats" "colnames" "compare_signatures"
#> [7] "consensus_heatmap" "dimension_reduction" "functional_enrichment"
#> [10] "get_anno_col" "get_anno" "get_classes"
#> [13] "get_consensus" "get_matrix" "get_membership"
#> [16] "get_param" "get_signatures" "get_stats"
#> [19] "is_best_k" "is_stable_k" "membership_heatmap"
#> [22] "ncol" "nrow" "plot_ecdf"
#> [25] "rownames" "select_partition_number" "show"
#> [28] "suggest_best_k" "test_to_known_factors"
collect_plots() function collects all the plots made from res for all k (number of partitions)
into one single page to provide an easy and fast comparison between different k.
collect_plots(res)
The plots are:
k and the heatmap of
predicted classes for each k.k.k.k.All the plots in panels can be made by individual functions and they are plotted later in this section.
select_partition_number() produces several plots showing different
statistics for choosing “optimized” k. There are following statistics:
k;k, the area increased is defined as \(A_k - A_{k-1}\).The detailed explanations of these statistics can be found in the cola vignette.
Generally speaking, lower PAC score, higher mean silhouette score or higher
concordance corresponds to better partition. Rand index and Jaccard index
measure how similar the current partition is compared to partition with k-1.
If they are too similar, we won't accept k is better than k-1.
select_partition_number(res)
The numeric values for all these statistics can be obtained by get_stats().
get_stats(res)
#> k 1-PAC mean_silhouette concordance area_increased Rand Jaccard
#> 2 2 1.000 0.979 0.991 0.5016 0.500 0.500
#> 3 3 0.846 0.862 0.933 0.2843 0.699 0.478
#> 4 4 0.828 0.865 0.930 0.1331 0.836 0.582
#> 5 5 0.686 0.651 0.825 0.0634 0.911 0.699
#> 6 6 0.652 0.538 0.752 0.0414 0.916 0.672
suggest_best_k() suggests the best \(k\) based on these statistics. The rules are as follows:
suggest_best_k(res)
#> [1] 2
Following shows the table of the partitions (You need to click the show/hide
code output link to see it). The membership matrix (columns with name p*)
is inferred by
clue::cl_consensus()
function with the SE method. Basically the value in the membership matrix
represents the probability to belong to a certain group. The finall class
label for an item is determined with the group with highest probability it
belongs to.
In get_classes() function, the entropy is calculated from the membership
matrix and the silhouette score is calculated from the consensus matrix.
cbind(get_classes(res, k = 2), get_membership(res, k = 2))
#> class entropy silhouette p1 p2
#> GSM97038 2 0.0000 0.985 0.000 1.000
#> GSM97045 2 0.0000 0.985 0.000 1.000
#> GSM97047 2 0.0000 0.985 0.000 1.000
#> GSM97025 2 0.0000 0.985 0.000 1.000
#> GSM97030 2 0.0000 0.985 0.000 1.000
#> GSM97027 2 0.0000 0.985 0.000 1.000
#> GSM97033 2 0.0000 0.985 0.000 1.000
#> GSM97034 2 0.0000 0.985 0.000 1.000
#> GSM97020 2 0.0000 0.985 0.000 1.000
#> GSM97026 2 0.0000 0.985 0.000 1.000
#> GSM97012 2 0.0000 0.985 0.000 1.000
#> GSM97015 2 0.0000 0.985 0.000 1.000
#> GSM97016 2 0.0000 0.985 0.000 1.000
#> GSM97017 2 0.0000 0.985 0.000 1.000
#> GSM97019 2 0.0000 0.985 0.000 1.000
#> GSM97022 2 0.0000 0.985 0.000 1.000
#> GSM97035 2 0.0000 0.985 0.000 1.000
#> GSM97036 2 0.0000 0.985 0.000 1.000
#> GSM97039 2 0.0000 0.985 0.000 1.000
#> GSM97046 2 0.0000 0.985 0.000 1.000
#> GSM97023 1 0.0000 0.999 1.000 0.000
#> GSM97029 2 0.0000 0.985 0.000 1.000
#> GSM97043 2 0.0000 0.985 0.000 1.000
#> GSM97013 1 0.0000 0.999 1.000 0.000
#> GSM96956 2 0.0000 0.985 0.000 1.000
#> GSM97024 2 0.0000 0.985 0.000 1.000
#> GSM97032 2 0.0000 0.985 0.000 1.000
#> GSM97044 2 0.0000 0.985 0.000 1.000
#> GSM97049 2 0.0000 0.985 0.000 1.000
#> GSM96968 1 0.0376 0.995 0.996 0.004
#> GSM96971 1 0.0000 0.999 1.000 0.000
#> GSM96986 1 0.0000 0.999 1.000 0.000
#> GSM97003 1 0.0000 0.999 1.000 0.000
#> GSM96957 2 0.1184 0.971 0.016 0.984
#> GSM96960 1 0.0000 0.999 1.000 0.000
#> GSM96975 1 0.0938 0.988 0.988 0.012
#> GSM96998 1 0.0000 0.999 1.000 0.000
#> GSM96999 1 0.0000 0.999 1.000 0.000
#> GSM97001 2 0.0000 0.985 0.000 1.000
#> GSM97005 1 0.0000 0.999 1.000 0.000
#> GSM97006 1 0.0000 0.999 1.000 0.000
#> GSM97021 2 0.0000 0.985 0.000 1.000
#> GSM97028 2 0.0000 0.985 0.000 1.000
#> GSM97031 1 0.0000 0.999 1.000 0.000
#> GSM97037 2 0.0000 0.985 0.000 1.000
#> GSM97018 2 0.0000 0.985 0.000 1.000
#> GSM97014 2 0.0000 0.985 0.000 1.000
#> GSM97042 2 0.0000 0.985 0.000 1.000
#> GSM97040 2 0.0000 0.985 0.000 1.000
#> GSM97041 2 0.0000 0.985 0.000 1.000
#> GSM96955 2 0.0000 0.985 0.000 1.000
#> GSM96990 2 0.0000 0.985 0.000 1.000
#> GSM96991 2 0.0000 0.985 0.000 1.000
#> GSM97048 2 0.0000 0.985 0.000 1.000
#> GSM96963 2 0.0000 0.985 0.000 1.000
#> GSM96953 2 0.0000 0.985 0.000 1.000
#> GSM96966 1 0.0000 0.999 1.000 0.000
#> GSM96979 1 0.0000 0.999 1.000 0.000
#> GSM96983 2 0.0000 0.985 0.000 1.000
#> GSM96984 1 0.0000 0.999 1.000 0.000
#> GSM96994 2 0.0000 0.985 0.000 1.000
#> GSM96996 2 0.8955 0.556 0.312 0.688
#> GSM96997 1 0.0000 0.999 1.000 0.000
#> GSM97007 2 0.9686 0.359 0.396 0.604
#> GSM96954 1 0.0000 0.999 1.000 0.000
#> GSM96962 1 0.0000 0.999 1.000 0.000
#> GSM96969 1 0.0000 0.999 1.000 0.000
#> GSM96970 1 0.0000 0.999 1.000 0.000
#> GSM96973 1 0.0000 0.999 1.000 0.000
#> GSM96976 1 0.0000 0.999 1.000 0.000
#> GSM96977 1 0.0000 0.999 1.000 0.000
#> GSM96995 2 0.0000 0.985 0.000 1.000
#> GSM97002 1 0.0000 0.999 1.000 0.000
#> GSM97009 2 0.0000 0.985 0.000 1.000
#> GSM97010 1 0.0000 0.999 1.000 0.000
#> GSM96974 1 0.0000 0.999 1.000 0.000
#> GSM96985 1 0.0000 0.999 1.000 0.000
#> GSM96959 2 0.0000 0.985 0.000 1.000
#> GSM96972 1 0.0000 0.999 1.000 0.000
#> GSM96978 1 0.0000 0.999 1.000 0.000
#> GSM96967 1 0.0000 0.999 1.000 0.000
#> GSM96987 2 0.0938 0.975 0.012 0.988
#> GSM97011 2 0.0000 0.985 0.000 1.000
#> GSM96964 1 0.0000 0.999 1.000 0.000
#> GSM96965 1 0.0000 0.999 1.000 0.000
#> GSM96981 1 0.2043 0.967 0.968 0.032
#> GSM96982 1 0.0000 0.999 1.000 0.000
#> GSM96988 2 0.3274 0.928 0.060 0.940
#> GSM97000 1 0.0000 0.999 1.000 0.000
#> GSM97004 1 0.0000 0.999 1.000 0.000
#> GSM97008 2 0.1184 0.971 0.016 0.984
#> GSM96950 1 0.0000 0.999 1.000 0.000
#> GSM96980 1 0.0000 0.999 1.000 0.000
#> GSM96989 1 0.0376 0.995 0.996 0.004
#> GSM96992 1 0.0000 0.999 1.000 0.000
#> GSM96993 2 0.0000 0.985 0.000 1.000
#> GSM96958 1 0.0000 0.999 1.000 0.000
#> GSM96951 1 0.0000 0.999 1.000 0.000
#> GSM96952 1 0.0000 0.999 1.000 0.000
#> GSM96961 1 0.0000 0.999 1.000 0.000
cbind(get_classes(res, k = 3), get_membership(res, k = 3))
#> class entropy silhouette p1 p2 p3
#> GSM97038 2 0.0592 0.9719 0.012 0.988 0.000
#> GSM97045 2 0.0592 0.9719 0.012 0.988 0.000
#> GSM97047 2 0.1643 0.9559 0.044 0.956 0.000
#> GSM97025 2 0.0424 0.9721 0.008 0.992 0.000
#> GSM97030 2 0.1182 0.9649 0.012 0.976 0.012
#> GSM97027 2 0.0892 0.9695 0.020 0.980 0.000
#> GSM97033 2 0.0424 0.9721 0.008 0.992 0.000
#> GSM97034 2 0.1751 0.9580 0.028 0.960 0.012
#> GSM97020 2 0.1031 0.9676 0.024 0.976 0.000
#> GSM97026 2 0.1163 0.9656 0.028 0.972 0.000
#> GSM97012 2 0.0000 0.9713 0.000 1.000 0.000
#> GSM97015 2 0.1878 0.9603 0.044 0.952 0.004
#> GSM97016 2 0.0000 0.9713 0.000 1.000 0.000
#> GSM97017 1 0.2625 0.8580 0.916 0.084 0.000
#> GSM97019 2 0.0000 0.9713 0.000 1.000 0.000
#> GSM97022 2 0.0237 0.9701 0.000 0.996 0.004
#> GSM97035 2 0.0237 0.9701 0.000 0.996 0.004
#> GSM97036 1 0.2356 0.8685 0.928 0.072 0.000
#> GSM97039 2 0.0424 0.9721 0.008 0.992 0.000
#> GSM97046 2 0.0237 0.9719 0.004 0.996 0.000
#> GSM97023 1 0.0000 0.9005 1.000 0.000 0.000
#> GSM97029 1 0.4235 0.7607 0.824 0.176 0.000
#> GSM97043 2 0.0592 0.9719 0.012 0.988 0.000
#> GSM97013 1 0.1163 0.9013 0.972 0.000 0.028
#> GSM96956 2 0.1647 0.9492 0.004 0.960 0.036
#> GSM97024 2 0.0000 0.9713 0.000 1.000 0.000
#> GSM97032 2 0.0424 0.9723 0.008 0.992 0.000
#> GSM97044 2 0.2434 0.9357 0.024 0.940 0.036
#> GSM97049 2 0.0592 0.9719 0.012 0.988 0.000
#> GSM96968 3 0.6244 0.2658 0.440 0.000 0.560
#> GSM96971 3 0.0237 0.8832 0.004 0.000 0.996
#> GSM96986 3 0.3816 0.8070 0.148 0.000 0.852
#> GSM97003 1 0.2625 0.8562 0.916 0.000 0.084
#> GSM96957 1 0.1163 0.8947 0.972 0.028 0.000
#> GSM96960 1 0.1163 0.8962 0.972 0.000 0.028
#> GSM96975 1 0.1163 0.9006 0.972 0.000 0.028
#> GSM96998 1 0.1031 0.9013 0.976 0.000 0.024
#> GSM96999 1 0.1860 0.8932 0.948 0.000 0.052
#> GSM97001 1 0.1529 0.8889 0.960 0.040 0.000
#> GSM97005 1 0.0747 0.8998 0.984 0.000 0.016
#> GSM97006 1 0.1289 0.8946 0.968 0.000 0.032
#> GSM97021 1 0.1860 0.8821 0.948 0.052 0.000
#> GSM97028 2 0.1031 0.9699 0.024 0.976 0.000
#> GSM97031 1 0.0892 0.8993 0.980 0.000 0.020
#> GSM97037 2 0.1315 0.9620 0.020 0.972 0.008
#> GSM97018 2 0.0747 0.9709 0.016 0.984 0.000
#> GSM97014 2 0.1289 0.9631 0.032 0.968 0.000
#> GSM97042 2 0.0237 0.9701 0.000 0.996 0.004
#> GSM97040 1 0.5327 0.6287 0.728 0.272 0.000
#> GSM97041 1 0.2066 0.8779 0.940 0.060 0.000
#> GSM96955 2 0.0592 0.9719 0.012 0.988 0.000
#> GSM96990 2 0.1163 0.9671 0.028 0.972 0.000
#> GSM96991 2 0.0000 0.9713 0.000 1.000 0.000
#> GSM97048 2 0.0592 0.9719 0.012 0.988 0.000
#> GSM96963 2 0.0000 0.9713 0.000 1.000 0.000
#> GSM96953 2 0.0592 0.9667 0.000 0.988 0.012
#> GSM96966 3 0.0892 0.8830 0.020 0.000 0.980
#> GSM96979 3 0.1860 0.8764 0.052 0.000 0.948
#> GSM96983 2 0.2414 0.9347 0.020 0.940 0.040
#> GSM96984 3 0.1529 0.8781 0.040 0.000 0.960
#> GSM96994 2 0.3370 0.9031 0.024 0.904 0.072
#> GSM96996 1 0.1170 0.8995 0.976 0.016 0.008
#> GSM96997 1 0.6215 0.1618 0.572 0.000 0.428
#> GSM97007 3 0.6066 0.6093 0.024 0.248 0.728
#> GSM96954 1 0.6302 -0.0409 0.520 0.000 0.480
#> GSM96962 3 0.1964 0.8750 0.056 0.000 0.944
#> GSM96969 3 0.1289 0.8803 0.032 0.000 0.968
#> GSM96970 3 0.0237 0.8832 0.004 0.000 0.996
#> GSM96973 3 0.0237 0.8832 0.004 0.000 0.996
#> GSM96976 3 0.0661 0.8790 0.004 0.008 0.988
#> GSM96977 1 0.6192 0.1999 0.580 0.000 0.420
#> GSM96995 2 0.2625 0.9182 0.084 0.916 0.000
#> GSM97002 1 0.1753 0.8947 0.952 0.000 0.048
#> GSM97009 2 0.4654 0.7487 0.208 0.792 0.000
#> GSM97010 3 0.6302 0.0802 0.480 0.000 0.520
#> GSM96974 3 0.0237 0.8832 0.004 0.000 0.996
#> GSM96985 3 0.5948 0.4312 0.360 0.000 0.640
#> GSM96959 2 0.3116 0.8870 0.108 0.892 0.000
#> GSM96972 3 0.1289 0.8803 0.032 0.000 0.968
#> GSM96978 3 0.0592 0.8789 0.012 0.000 0.988
#> GSM96967 3 0.0592 0.8836 0.012 0.000 0.988
#> GSM96987 1 0.1643 0.8879 0.956 0.044 0.000
#> GSM97011 1 0.4750 0.7090 0.784 0.216 0.000
#> GSM96964 1 0.0424 0.9016 0.992 0.000 0.008
#> GSM96965 3 0.0237 0.8832 0.004 0.000 0.996
#> GSM96981 1 0.1031 0.9005 0.976 0.000 0.024
#> GSM96982 1 0.1643 0.8966 0.956 0.000 0.044
#> GSM96988 1 0.4062 0.7675 0.836 0.164 0.000
#> GSM97000 1 0.0747 0.9002 0.984 0.000 0.016
#> GSM97004 1 0.1753 0.8947 0.952 0.000 0.048
#> GSM97008 1 0.0983 0.8965 0.980 0.016 0.004
#> GSM96950 1 0.1964 0.8913 0.944 0.000 0.056
#> GSM96980 3 0.3879 0.7928 0.152 0.000 0.848
#> GSM96989 1 0.1031 0.9005 0.976 0.000 0.024
#> GSM96992 1 0.1529 0.8975 0.960 0.000 0.040
#> GSM96993 1 0.2066 0.8779 0.940 0.060 0.000
#> GSM96958 1 0.1411 0.8985 0.964 0.000 0.036
#> GSM96951 1 0.1163 0.8971 0.972 0.000 0.028
#> GSM96952 1 0.1163 0.9009 0.972 0.000 0.028
#> GSM96961 1 0.0424 0.9010 0.992 0.000 0.008
cbind(get_classes(res, k = 4), get_membership(res, k = 4))
#> class entropy silhouette p1 p2 p3 p4
#> GSM97038 2 0.0376 0.9622 0.004 0.992 0.004 0.000
#> GSM97045 2 0.0188 0.9622 0.004 0.996 0.000 0.000
#> GSM97047 2 0.4857 0.4752 0.008 0.668 0.324 0.000
#> GSM97025 2 0.0376 0.9622 0.004 0.992 0.004 0.000
#> GSM97030 3 0.4164 0.6830 0.000 0.264 0.736 0.000
#> GSM97027 2 0.0188 0.9622 0.004 0.996 0.000 0.000
#> GSM97033 2 0.0657 0.9597 0.004 0.984 0.012 0.000
#> GSM97034 3 0.1557 0.8641 0.000 0.056 0.944 0.000
#> GSM97020 2 0.0188 0.9622 0.004 0.996 0.000 0.000
#> GSM97026 2 0.0188 0.9622 0.004 0.996 0.000 0.000
#> GSM97012 2 0.0188 0.9607 0.000 0.996 0.004 0.000
#> GSM97015 3 0.1637 0.8635 0.000 0.060 0.940 0.000
#> GSM97016 2 0.0188 0.9622 0.000 0.996 0.004 0.000
#> GSM97017 1 0.1474 0.8937 0.948 0.052 0.000 0.000
#> GSM97019 2 0.0188 0.9622 0.000 0.996 0.004 0.000
#> GSM97022 2 0.0336 0.9614 0.000 0.992 0.008 0.000
#> GSM97035 2 0.0188 0.9607 0.000 0.996 0.004 0.000
#> GSM97036 1 0.1302 0.8995 0.956 0.044 0.000 0.000
#> GSM97039 2 0.0188 0.9622 0.000 0.996 0.004 0.000
#> GSM97046 2 0.0188 0.9607 0.000 0.996 0.004 0.000
#> GSM97023 1 0.0469 0.9152 0.988 0.000 0.012 0.000
#> GSM97029 1 0.3873 0.7121 0.772 0.228 0.000 0.000
#> GSM97043 2 0.0188 0.9622 0.004 0.996 0.000 0.000
#> GSM97013 1 0.0779 0.9143 0.980 0.000 0.016 0.004
#> GSM96956 2 0.3161 0.8433 0.000 0.864 0.124 0.012
#> GSM97024 2 0.0592 0.9576 0.000 0.984 0.016 0.000
#> GSM97032 2 0.0592 0.9576 0.000 0.984 0.016 0.000
#> GSM97044 3 0.0921 0.8711 0.000 0.028 0.972 0.000
#> GSM97049 2 0.0657 0.9597 0.004 0.984 0.012 0.000
#> GSM96968 3 0.0992 0.8715 0.012 0.004 0.976 0.008
#> GSM96971 4 0.0921 0.9405 0.000 0.000 0.028 0.972
#> GSM96986 3 0.1109 0.8675 0.004 0.000 0.968 0.028
#> GSM97003 3 0.1929 0.8632 0.024 0.000 0.940 0.036
#> GSM96957 1 0.1059 0.9127 0.972 0.012 0.016 0.000
#> GSM96960 1 0.0657 0.9147 0.984 0.000 0.012 0.004
#> GSM96975 1 0.1305 0.8990 0.960 0.000 0.004 0.036
#> GSM96998 1 0.0524 0.9151 0.988 0.000 0.008 0.004
#> GSM96999 1 0.0524 0.9152 0.988 0.000 0.008 0.004
#> GSM97001 1 0.0937 0.9136 0.976 0.012 0.012 0.000
#> GSM97005 1 0.5167 0.0286 0.508 0.000 0.488 0.004
#> GSM97006 1 0.1042 0.9117 0.972 0.000 0.020 0.008
#> GSM97021 1 0.1182 0.9113 0.968 0.016 0.016 0.000
#> GSM97028 2 0.3266 0.7903 0.000 0.832 0.168 0.000
#> GSM97031 3 0.2593 0.8210 0.104 0.000 0.892 0.004
#> GSM97037 3 0.2921 0.8153 0.000 0.140 0.860 0.000
#> GSM97018 2 0.0657 0.9597 0.004 0.984 0.012 0.000
#> GSM97014 2 0.0376 0.9620 0.004 0.992 0.004 0.000
#> GSM97042 2 0.0188 0.9607 0.000 0.996 0.004 0.000
#> GSM97040 1 0.5050 0.3486 0.588 0.408 0.004 0.000
#> GSM97041 1 0.0921 0.9085 0.972 0.028 0.000 0.000
#> GSM96955 2 0.0844 0.9539 0.004 0.980 0.012 0.004
#> GSM96990 3 0.3907 0.7319 0.000 0.232 0.768 0.000
#> GSM96991 2 0.0336 0.9589 0.000 0.992 0.008 0.000
#> GSM97048 2 0.0376 0.9622 0.004 0.992 0.004 0.000
#> GSM96963 2 0.0336 0.9589 0.000 0.992 0.008 0.000
#> GSM96953 2 0.0188 0.9622 0.000 0.996 0.004 0.000
#> GSM96966 4 0.0376 0.9529 0.004 0.000 0.004 0.992
#> GSM96979 3 0.1557 0.8553 0.000 0.000 0.944 0.056
#> GSM96983 3 0.2973 0.8087 0.000 0.144 0.856 0.000
#> GSM96984 3 0.1211 0.8622 0.000 0.000 0.960 0.040
#> GSM96994 3 0.1229 0.8718 0.004 0.020 0.968 0.008
#> GSM96996 1 0.0524 0.9127 0.988 0.000 0.004 0.008
#> GSM96997 3 0.1151 0.8679 0.008 0.000 0.968 0.024
#> GSM97007 3 0.0657 0.8718 0.000 0.012 0.984 0.004
#> GSM96954 3 0.0804 0.8707 0.008 0.000 0.980 0.012
#> GSM96962 3 0.0921 0.8666 0.000 0.000 0.972 0.028
#> GSM96969 4 0.0469 0.9513 0.012 0.000 0.000 0.988
#> GSM96970 4 0.0188 0.9537 0.004 0.000 0.000 0.996
#> GSM96973 4 0.0376 0.9534 0.004 0.000 0.004 0.992
#> GSM96976 4 0.0188 0.9523 0.000 0.000 0.004 0.996
#> GSM96977 3 0.2844 0.8472 0.048 0.000 0.900 0.052
#> GSM96995 3 0.3962 0.8004 0.028 0.152 0.820 0.000
#> GSM97002 1 0.0592 0.9114 0.984 0.000 0.000 0.016
#> GSM97009 2 0.5160 0.7163 0.136 0.760 0.104 0.000
#> GSM97010 3 0.7764 0.0953 0.240 0.000 0.404 0.356
#> GSM96974 4 0.0188 0.9523 0.000 0.000 0.004 0.996
#> GSM96985 4 0.4546 0.7438 0.204 0.012 0.012 0.772
#> GSM96959 3 0.5404 0.5400 0.028 0.328 0.644 0.000
#> GSM96972 4 0.0804 0.9513 0.012 0.000 0.008 0.980
#> GSM96978 4 0.3528 0.7595 0.000 0.000 0.192 0.808
#> GSM96967 4 0.0188 0.9537 0.004 0.000 0.000 0.996
#> GSM96987 1 0.0564 0.9142 0.988 0.004 0.004 0.004
#> GSM97011 1 0.4188 0.6897 0.752 0.244 0.004 0.000
#> GSM96964 1 0.0469 0.9152 0.988 0.000 0.012 0.000
#> GSM96965 4 0.0000 0.9529 0.000 0.000 0.000 1.000
#> GSM96981 1 0.0657 0.9113 0.984 0.000 0.004 0.012
#> GSM96982 1 0.0707 0.9100 0.980 0.000 0.000 0.020
#> GSM96988 1 0.4544 0.7357 0.780 0.192 0.016 0.012
#> GSM97000 3 0.1229 0.8711 0.020 0.004 0.968 0.008
#> GSM97004 1 0.0779 0.9100 0.980 0.000 0.004 0.016
#> GSM97008 1 0.4343 0.6243 0.732 0.004 0.264 0.000
#> GSM96950 1 0.1975 0.8910 0.936 0.000 0.016 0.048
#> GSM96980 4 0.1661 0.9237 0.052 0.000 0.004 0.944
#> GSM96989 1 0.0524 0.9127 0.988 0.000 0.004 0.008
#> GSM96992 1 0.0188 0.9144 0.996 0.000 0.000 0.004
#> GSM96993 1 0.0921 0.9086 0.972 0.028 0.000 0.000
#> GSM96958 1 0.0779 0.9140 0.980 0.000 0.016 0.004
#> GSM96951 3 0.3933 0.7183 0.200 0.000 0.792 0.008
#> GSM96952 1 0.0188 0.9144 0.996 0.000 0.000 0.004
#> GSM96961 1 0.0469 0.9152 0.988 0.000 0.012 0.000
cbind(get_classes(res, k = 5), get_membership(res, k = 5))
#> class entropy silhouette p1 p2 p3 p4 p5
#> GSM97038 2 0.0703 0.8458 0.000 0.976 0.000 0.000 0.024
#> GSM97045 2 0.0703 0.8424 0.000 0.976 0.000 0.000 0.024
#> GSM97047 2 0.4352 0.6794 0.148 0.772 0.004 0.000 0.076
#> GSM97025 2 0.0000 0.8469 0.000 1.000 0.000 0.000 0.000
#> GSM97030 3 0.3586 0.6689 0.000 0.188 0.792 0.000 0.020
#> GSM97027 2 0.0703 0.8424 0.000 0.976 0.000 0.000 0.024
#> GSM97033 2 0.0963 0.8376 0.000 0.964 0.000 0.000 0.036
#> GSM97034 3 0.4295 0.6421 0.008 0.200 0.760 0.004 0.028
#> GSM97020 2 0.1408 0.8289 0.008 0.948 0.000 0.000 0.044
#> GSM97026 2 0.0566 0.8451 0.004 0.984 0.000 0.000 0.012
#> GSM97012 2 0.1197 0.8343 0.000 0.952 0.000 0.000 0.048
#> GSM97015 3 0.1670 0.7874 0.000 0.052 0.936 0.000 0.012
#> GSM97016 2 0.0510 0.8470 0.000 0.984 0.000 0.000 0.016
#> GSM97017 1 0.5168 0.3212 0.592 0.356 0.000 0.000 0.052
#> GSM97019 2 0.0794 0.8444 0.000 0.972 0.000 0.000 0.028
#> GSM97022 2 0.0609 0.8464 0.000 0.980 0.000 0.000 0.020
#> GSM97035 2 0.0880 0.8424 0.000 0.968 0.000 0.000 0.032
#> GSM97036 1 0.4302 0.6349 0.744 0.048 0.000 0.000 0.208
#> GSM97039 2 0.0162 0.8465 0.000 0.996 0.000 0.000 0.004
#> GSM97046 2 0.0703 0.8458 0.000 0.976 0.000 0.000 0.024
#> GSM97023 1 0.1195 0.6953 0.960 0.000 0.012 0.000 0.028
#> GSM97029 2 0.5368 0.4123 0.332 0.596 0.000 0.000 0.072
#> GSM97043 2 0.0703 0.8458 0.000 0.976 0.000 0.000 0.024
#> GSM97013 1 0.4534 0.6084 0.796 0.032 0.004 0.080 0.088
#> GSM96956 3 0.5648 0.1495 0.000 0.448 0.476 0.000 0.076
#> GSM97024 2 0.0510 0.8471 0.000 0.984 0.000 0.000 0.016
#> GSM97032 2 0.3921 0.6770 0.000 0.784 0.044 0.000 0.172
#> GSM97044 3 0.0404 0.7981 0.000 0.000 0.988 0.000 0.012
#> GSM97049 2 0.2504 0.7955 0.040 0.896 0.000 0.000 0.064
#> GSM96968 3 0.0162 0.7996 0.004 0.000 0.996 0.000 0.000
#> GSM96971 4 0.1485 0.8910 0.000 0.000 0.020 0.948 0.032
#> GSM96986 3 0.1087 0.7971 0.008 0.000 0.968 0.016 0.008
#> GSM97003 3 0.1082 0.7932 0.028 0.000 0.964 0.008 0.000
#> GSM96957 1 0.3174 0.6518 0.868 0.036 0.016 0.000 0.080
#> GSM96960 1 0.4165 0.5564 0.672 0.000 0.008 0.000 0.320
#> GSM96975 1 0.4169 0.6234 0.732 0.000 0.000 0.028 0.240
#> GSM96998 1 0.0703 0.6933 0.976 0.000 0.000 0.000 0.024
#> GSM96999 1 0.1522 0.6881 0.944 0.000 0.000 0.044 0.012
#> GSM97001 1 0.2208 0.6629 0.908 0.020 0.000 0.000 0.072
#> GSM97005 1 0.6075 0.4946 0.680 0.012 0.172 0.048 0.088
#> GSM97006 1 0.3047 0.6896 0.868 0.000 0.024 0.012 0.096
#> GSM97021 1 0.4219 0.5772 0.772 0.156 0.000 0.000 0.072
#> GSM97028 5 0.5915 0.1792 0.000 0.108 0.384 0.000 0.508
#> GSM97031 3 0.4517 0.2364 0.436 0.000 0.556 0.000 0.008
#> GSM97037 3 0.2248 0.7688 0.000 0.088 0.900 0.000 0.012
#> GSM97018 2 0.4950 0.2415 0.008 0.552 0.016 0.000 0.424
#> GSM97014 2 0.4139 0.6934 0.132 0.784 0.000 0.000 0.084
#> GSM97042 2 0.1341 0.8292 0.000 0.944 0.000 0.000 0.056
#> GSM97040 2 0.5331 0.3322 0.372 0.568 0.000 0.000 0.060
#> GSM97041 1 0.3950 0.5981 0.796 0.136 0.000 0.000 0.068
#> GSM96955 5 0.4310 0.0985 0.004 0.392 0.000 0.000 0.604
#> GSM96990 3 0.3353 0.6730 0.000 0.196 0.796 0.000 0.008
#> GSM96991 2 0.4283 0.2199 0.000 0.544 0.000 0.000 0.456
#> GSM97048 2 0.0510 0.8447 0.000 0.984 0.000 0.000 0.016
#> GSM96963 2 0.3816 0.5433 0.000 0.696 0.000 0.000 0.304
#> GSM96953 2 0.0609 0.8464 0.000 0.980 0.000 0.000 0.020
#> GSM96966 4 0.0703 0.9100 0.000 0.000 0.000 0.976 0.024
#> GSM96979 3 0.1430 0.7872 0.000 0.000 0.944 0.052 0.004
#> GSM96983 5 0.4574 0.1721 0.000 0.012 0.412 0.000 0.576
#> GSM96984 3 0.0566 0.7991 0.000 0.000 0.984 0.012 0.004
#> GSM96994 3 0.2339 0.7547 0.000 0.004 0.892 0.004 0.100
#> GSM96996 1 0.4440 0.3169 0.528 0.000 0.000 0.004 0.468
#> GSM96997 3 0.0162 0.7996 0.000 0.000 0.996 0.004 0.000
#> GSM97007 3 0.0510 0.7971 0.000 0.000 0.984 0.000 0.016
#> GSM96954 3 0.0000 0.7992 0.000 0.000 1.000 0.000 0.000
#> GSM96962 3 0.0324 0.7996 0.000 0.000 0.992 0.004 0.004
#> GSM96969 4 0.0963 0.9073 0.000 0.000 0.000 0.964 0.036
#> GSM96970 4 0.0609 0.9106 0.000 0.000 0.000 0.980 0.020
#> GSM96973 4 0.0000 0.9098 0.000 0.000 0.000 1.000 0.000
#> GSM96976 4 0.0880 0.9024 0.000 0.000 0.000 0.968 0.032
#> GSM96977 3 0.6201 0.4251 0.272 0.000 0.596 0.104 0.028
#> GSM96995 3 0.2956 0.7499 0.012 0.020 0.872 0.000 0.096
#> GSM97002 1 0.4738 0.2994 0.520 0.000 0.000 0.016 0.464
#> GSM97009 2 0.4690 0.5922 0.240 0.708 0.004 0.000 0.048
#> GSM97010 4 0.6508 0.2361 0.188 0.000 0.312 0.496 0.004
#> GSM96974 4 0.0703 0.9102 0.000 0.000 0.000 0.976 0.024
#> GSM96985 5 0.2932 0.4775 0.104 0.000 0.000 0.032 0.864
#> GSM96959 3 0.3992 0.5625 0.000 0.268 0.720 0.000 0.012
#> GSM96972 4 0.0324 0.9084 0.004 0.000 0.000 0.992 0.004
#> GSM96978 3 0.4467 0.6193 0.000 0.000 0.752 0.084 0.164
#> GSM96967 4 0.0794 0.9092 0.000 0.000 0.000 0.972 0.028
#> GSM96987 1 0.4114 0.4934 0.624 0.000 0.000 0.000 0.376
#> GSM97011 1 0.5014 0.3082 0.592 0.368 0.000 0.000 0.040
#> GSM96964 1 0.1831 0.6948 0.920 0.000 0.004 0.000 0.076
#> GSM96965 4 0.0162 0.9092 0.000 0.000 0.000 0.996 0.004
#> GSM96981 1 0.4130 0.5840 0.696 0.000 0.000 0.012 0.292
#> GSM96982 5 0.4682 -0.1754 0.420 0.000 0.000 0.016 0.564
#> GSM96988 5 0.3477 0.4739 0.140 0.008 0.024 0.000 0.828
#> GSM97000 3 0.3368 0.6801 0.156 0.000 0.820 0.000 0.024
#> GSM97004 1 0.4708 0.3622 0.548 0.000 0.000 0.016 0.436
#> GSM97008 1 0.4411 0.6016 0.788 0.020 0.120 0.000 0.072
#> GSM96950 1 0.4436 0.5954 0.784 0.008 0.004 0.120 0.084
#> GSM96980 4 0.4054 0.7074 0.072 0.000 0.000 0.788 0.140
#> GSM96989 1 0.3816 0.5846 0.696 0.000 0.000 0.000 0.304
#> GSM96992 1 0.3336 0.6440 0.772 0.000 0.000 0.000 0.228
#> GSM96993 1 0.2971 0.6795 0.836 0.008 0.000 0.000 0.156
#> GSM96958 1 0.1365 0.6961 0.952 0.000 0.004 0.004 0.040
#> GSM96951 1 0.5390 0.3267 0.608 0.000 0.332 0.012 0.048
#> GSM96952 1 0.3336 0.6447 0.772 0.000 0.000 0.000 0.228
#> GSM96961 1 0.2338 0.6893 0.884 0.000 0.004 0.000 0.112
cbind(get_classes(res, k = 6), get_membership(res, k = 6))
#> class entropy silhouette p1 p2 p3 p4 p5 p6
#> GSM97038 2 0.1176 0.82799 0.000 0.956 0.024 0.000 0.020 0.000
#> GSM97045 2 0.1477 0.82106 0.004 0.940 0.008 0.000 0.048 0.000
#> GSM97047 2 0.4587 0.65678 0.044 0.712 0.024 0.000 0.216 0.004
#> GSM97025 2 0.1334 0.82520 0.000 0.948 0.020 0.000 0.032 0.000
#> GSM97030 6 0.3542 0.56271 0.000 0.156 0.016 0.000 0.028 0.800
#> GSM97027 2 0.1624 0.82110 0.004 0.936 0.020 0.000 0.040 0.000
#> GSM97033 2 0.1511 0.82048 0.004 0.940 0.012 0.000 0.044 0.000
#> GSM97034 6 0.6071 0.12286 0.004 0.424 0.020 0.004 0.108 0.440
#> GSM97020 2 0.1226 0.82185 0.004 0.952 0.004 0.000 0.040 0.000
#> GSM97026 2 0.1218 0.82862 0.004 0.956 0.012 0.000 0.028 0.000
#> GSM97012 2 0.2058 0.81261 0.000 0.908 0.056 0.000 0.036 0.000
#> GSM97015 6 0.2554 0.59648 0.000 0.088 0.012 0.000 0.020 0.880
#> GSM97016 2 0.1088 0.82582 0.000 0.960 0.024 0.000 0.016 0.000
#> GSM97017 1 0.4875 0.44491 0.660 0.264 0.032 0.000 0.044 0.000
#> GSM97019 2 0.1930 0.81589 0.000 0.916 0.048 0.000 0.036 0.000
#> GSM97022 2 0.1341 0.82375 0.000 0.948 0.028 0.000 0.024 0.000
#> GSM97035 2 0.2001 0.81688 0.000 0.912 0.048 0.000 0.040 0.000
#> GSM97036 1 0.2113 0.66592 0.912 0.032 0.048 0.000 0.008 0.000
#> GSM97039 2 0.0891 0.82528 0.000 0.968 0.008 0.000 0.024 0.000
#> GSM97046 2 0.2846 0.79059 0.000 0.856 0.084 0.000 0.060 0.000
#> GSM97023 1 0.1367 0.66805 0.944 0.000 0.012 0.000 0.044 0.000
#> GSM97029 2 0.5331 0.38356 0.316 0.588 0.024 0.000 0.072 0.000
#> GSM97043 2 0.1565 0.82248 0.000 0.940 0.028 0.000 0.028 0.004
#> GSM97013 5 0.5389 0.45009 0.320 0.008 0.008 0.072 0.588 0.004
#> GSM96956 6 0.5916 0.21939 0.000 0.400 0.092 0.000 0.036 0.472
#> GSM97024 2 0.2231 0.81765 0.000 0.908 0.028 0.000 0.048 0.016
#> GSM97032 2 0.5574 0.47972 0.000 0.628 0.088 0.000 0.052 0.232
#> GSM97044 6 0.0436 0.59949 0.000 0.004 0.004 0.000 0.004 0.988
#> GSM97049 2 0.2544 0.78673 0.004 0.864 0.012 0.000 0.120 0.000
#> GSM96968 6 0.4109 0.13387 0.004 0.000 0.008 0.000 0.392 0.596
#> GSM96971 4 0.2240 0.90533 0.000 0.000 0.032 0.908 0.044 0.016
#> GSM96986 6 0.4129 -0.16240 0.004 0.000 0.000 0.004 0.496 0.496
#> GSM97003 6 0.4233 0.46102 0.024 0.000 0.016 0.016 0.196 0.748
#> GSM96957 1 0.4513 0.23407 0.572 0.028 0.004 0.000 0.396 0.000
#> GSM96960 1 0.3915 0.49186 0.704 0.000 0.272 0.020 0.004 0.000
#> GSM96975 1 0.5550 0.39430 0.592 0.000 0.292 0.076 0.040 0.000
#> GSM96998 1 0.1625 0.66902 0.928 0.000 0.012 0.000 0.060 0.000
#> GSM96999 1 0.3469 0.64818 0.824 0.000 0.012 0.092 0.072 0.000
#> GSM97001 1 0.4030 0.58662 0.776 0.040 0.032 0.000 0.152 0.000
#> GSM97005 5 0.5162 0.30404 0.384 0.004 0.016 0.020 0.560 0.016
#> GSM97006 1 0.5477 0.56957 0.712 0.000 0.040 0.104 0.088 0.056
#> GSM97021 1 0.5072 0.51721 0.700 0.132 0.040 0.000 0.128 0.000
#> GSM97028 3 0.6118 0.15782 0.000 0.124 0.488 0.000 0.036 0.352
#> GSM97031 6 0.6408 -0.34900 0.308 0.000 0.012 0.000 0.320 0.360
#> GSM97037 6 0.5189 0.45933 0.000 0.280 0.036 0.000 0.056 0.628
#> GSM97018 2 0.6735 0.30422 0.008 0.520 0.188 0.000 0.068 0.216
#> GSM97014 2 0.3880 0.71314 0.024 0.772 0.028 0.000 0.176 0.000
#> GSM97042 2 0.2420 0.80293 0.000 0.884 0.076 0.000 0.040 0.000
#> GSM97040 1 0.6047 0.00961 0.444 0.432 0.036 0.000 0.080 0.008
#> GSM97041 1 0.4660 0.55286 0.736 0.108 0.032 0.000 0.124 0.000
#> GSM96955 3 0.3869 0.43473 0.008 0.236 0.736 0.004 0.016 0.000
#> GSM96990 6 0.5036 0.41675 0.000 0.332 0.036 0.000 0.032 0.600
#> GSM96991 3 0.4791 0.10192 0.000 0.384 0.564 0.000 0.048 0.004
#> GSM97048 2 0.1398 0.82211 0.000 0.940 0.008 0.000 0.052 0.000
#> GSM96963 2 0.4385 0.22997 0.000 0.532 0.444 0.000 0.024 0.000
#> GSM96953 2 0.1168 0.82780 0.000 0.956 0.028 0.000 0.016 0.000
#> GSM96966 4 0.1036 0.92898 0.004 0.000 0.008 0.964 0.024 0.000
#> GSM96979 5 0.5067 0.14887 0.000 0.000 0.000 0.076 0.488 0.436
#> GSM96983 3 0.4572 0.07248 0.000 0.012 0.512 0.000 0.016 0.460
#> GSM96984 6 0.2730 0.50917 0.000 0.000 0.000 0.000 0.192 0.808
#> GSM96994 6 0.1845 0.58988 0.000 0.004 0.072 0.000 0.008 0.916
#> GSM96996 1 0.4049 0.26644 0.580 0.000 0.412 0.004 0.004 0.000
#> GSM96997 6 0.2845 0.51210 0.000 0.000 0.004 0.004 0.172 0.820
#> GSM97007 6 0.0405 0.59796 0.000 0.000 0.008 0.000 0.004 0.988
#> GSM96954 6 0.1285 0.59037 0.000 0.000 0.004 0.000 0.052 0.944
#> GSM96962 6 0.3309 0.39068 0.000 0.000 0.000 0.000 0.280 0.720
#> GSM96969 4 0.1194 0.92829 0.000 0.000 0.008 0.956 0.032 0.004
#> GSM96970 4 0.0146 0.93205 0.000 0.000 0.000 0.996 0.004 0.000
#> GSM96973 4 0.0458 0.93130 0.000 0.000 0.000 0.984 0.016 0.000
#> GSM96976 4 0.1908 0.91492 0.000 0.000 0.028 0.916 0.056 0.000
#> GSM96977 5 0.5159 0.54802 0.028 0.000 0.000 0.108 0.672 0.192
#> GSM96995 6 0.2302 0.58053 0.032 0.000 0.060 0.000 0.008 0.900
#> GSM97002 3 0.4111 0.00565 0.456 0.000 0.536 0.004 0.004 0.000
#> GSM97009 2 0.5797 0.48730 0.100 0.576 0.032 0.000 0.288 0.004
#> GSM97010 5 0.5876 0.52390 0.012 0.000 0.040 0.176 0.632 0.140
#> GSM96974 4 0.1578 0.92454 0.000 0.000 0.012 0.936 0.048 0.004
#> GSM96985 3 0.3544 0.47662 0.108 0.000 0.828 0.032 0.024 0.008
#> GSM96959 6 0.5040 0.51491 0.020 0.188 0.044 0.000 0.040 0.708
#> GSM96972 4 0.1364 0.92107 0.000 0.000 0.004 0.944 0.048 0.004
#> GSM96978 3 0.6481 -0.05332 0.000 0.000 0.432 0.076 0.104 0.388
#> GSM96967 4 0.0603 0.93096 0.000 0.000 0.004 0.980 0.016 0.000
#> GSM96987 1 0.2838 0.59075 0.808 0.000 0.188 0.000 0.004 0.000
#> GSM97011 2 0.6754 0.02650 0.348 0.412 0.060 0.000 0.180 0.000
#> GSM96964 1 0.4476 0.38131 0.640 0.000 0.052 0.000 0.308 0.000
#> GSM96965 4 0.1075 0.92303 0.000 0.000 0.000 0.952 0.048 0.000
#> GSM96981 3 0.4631 0.02610 0.428 0.000 0.536 0.004 0.032 0.000
#> GSM96982 3 0.3802 0.29701 0.312 0.000 0.676 0.000 0.012 0.000
#> GSM96988 3 0.6643 0.43978 0.212 0.012 0.560 0.012 0.044 0.160
#> GSM97000 5 0.4811 0.17488 0.036 0.000 0.008 0.000 0.508 0.448
#> GSM97004 1 0.4831 0.26963 0.580 0.000 0.368 0.040 0.012 0.000
#> GSM97008 1 0.6063 0.26067 0.532 0.028 0.048 0.000 0.348 0.044
#> GSM96950 5 0.5122 0.58070 0.188 0.000 0.004 0.148 0.656 0.004
#> GSM96980 4 0.5290 0.56352 0.104 0.000 0.188 0.668 0.040 0.000
#> GSM96989 1 0.2520 0.64231 0.872 0.000 0.108 0.012 0.008 0.000
#> GSM96992 1 0.2631 0.64072 0.860 0.000 0.124 0.004 0.008 0.004
#> GSM96993 1 0.1088 0.67203 0.960 0.016 0.024 0.000 0.000 0.000
#> GSM96958 1 0.5450 0.19553 0.544 0.000 0.040 0.032 0.376 0.008
#> GSM96951 5 0.5759 0.58498 0.252 0.000 0.000 0.024 0.580 0.144
#> GSM96952 1 0.2466 0.64611 0.872 0.000 0.112 0.008 0.008 0.000
#> GSM96961 1 0.0935 0.67010 0.964 0.000 0.032 0.000 0.004 0.000
Heatmaps for the consensus matrix. It visualizes the probability of two samples to be in a same group.
consensus_heatmap(res, k = 2)
consensus_heatmap(res, k = 3)
consensus_heatmap(res, k = 4)
consensus_heatmap(res, k = 5)
consensus_heatmap(res, k = 6)
Heatmaps for the membership of samples in all partitions to see how consistent they are:
membership_heatmap(res, k = 2)
membership_heatmap(res, k = 3)
membership_heatmap(res, k = 4)
membership_heatmap(res, k = 5)
membership_heatmap(res, k = 6)
As soon as we have had the classes for columns, we can look for signatures which are significantly different between classes which can be candidate marks for certain classes. Following are the heatmaps for signatures.
Signature heatmaps where rows are scaled:
get_signatures(res, k = 2)
get_signatures(res, k = 3)
get_signatures(res, k = 4)
get_signatures(res, k = 5)
get_signatures(res, k = 6)
Signature heatmaps where rows are not scaled:
get_signatures(res, k = 2, scale_rows = FALSE)
get_signatures(res, k = 3, scale_rows = FALSE)
get_signatures(res, k = 4, scale_rows = FALSE)
get_signatures(res, k = 5, scale_rows = FALSE)
get_signatures(res, k = 6, scale_rows = FALSE)
Compare the overlap of signatures from different k:
compare_signatures(res)
get_signature() returns a data frame invisibly. TO get the list of signatures, the function
call should be assigned to a variable explicitly. In following code, if plot argument is set
to FALSE, no heatmap is plotted while only the differential analysis is performed.
# code only for demonstration
tb = get_signature(res, k = ..., plot = FALSE)
An example of the output of tb is:
#> which_row fdr mean_1 mean_2 scaled_mean_1 scaled_mean_2 km
#> 1 38 0.042760348 8.373488 9.131774 -0.5533452 0.5164555 1
#> 2 40 0.018707592 7.106213 8.469186 -0.6173731 0.5762149 1
#> 3 55 0.019134737 10.221463 11.207825 -0.6159697 0.5749050 1
#> 4 59 0.006059896 5.921854 7.869574 -0.6899429 0.6439467 1
#> 5 60 0.018055526 8.928898 10.211722 -0.6204761 0.5791110 1
#> 6 98 0.009384629 15.714769 14.887706 0.6635654 -0.6193277 2
...
The columns in tb are:
which_row: row indices corresponding to the input matrix.fdr: FDR for the differential test. mean_x: The mean value in group x.scaled_mean_x: The mean value in group x after rows are scaled.km: Row groups if k-means clustering is applied to rows.UMAP plot which shows how samples are separated.
dimension_reduction(res, k = 2, method = "UMAP")
dimension_reduction(res, k = 3, method = "UMAP")
dimension_reduction(res, k = 4, method = "UMAP")
dimension_reduction(res, k = 5, method = "UMAP")
dimension_reduction(res, k = 6, method = "UMAP")
Following heatmap shows how subgroups are split when increasing k:
collect_classes(res)
Test correlation between subgroups and known annotations. If the known annotation is numeric, one-way ANOVA test is applied, and if the known annotation is discrete, chi-squared contingency table test is applied.
test_to_known_factors(res)
#> n disease.state(p) specimen(p) cell.type(p) other(p) k
#> ATC:NMF 99 6.29e-06 0.6173 1.57e-11 0.2857 2
#> ATC:NMF 94 9.01e-05 0.0862 1.22e-21 0.1605 3
#> ATC:NMF 96 7.69e-04 0.3426 1.63e-19 0.1058 4
#> ATC:NMF 78 1.26e-04 0.1133 1.89e-19 0.0431 5
#> ATC:NMF 61 7.20e-05 0.2446 1.92e-14 0.1045 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