Fig. 2From: Multi-omics integration with weighted affinity and self-diffusion applied for cancer subtypes identificationSimulation on evaluation study. Clustering performance, measured by averaged NMIs are shown for selecting the appropriate local scale neighbors(k = 5) (A) and self-diffusion iterations(t = 3) (B). C Cell-to-cell similarities matrices represented by different iterations (t = 1, 3, 7, 15). Clusters are arranged according to the ground-truth labelsBack to article page