Figure 4. Consensus clustering analysis results displaying the robustness of sample classification using multiple iterations of k-means clustering.

In this heatmap of the clustered consensus matrix, rows and columns of the matrix both indicate samples. The darkness of color in the heatmap represents how frequent 2 samples are clustered together. The darker, the more frequent. This heatmap clearly shows a pattern of 5 compact groups of samples frequently clustered together along the diagonal. The color bar on top indicates 5 clusters of samples.