Fig. 2.

Comparison of KS-test P-values for in-class vs out-of-class target annotation in FUSION, CP, and SNF. Dot plots of −log₁₀ KS-test P-values in FUSION, CP and SNF datasets, for each target annotation class with at least five members, using (A) Euclidean distance or (B) Pearson correlation as the similarity metric. Significance threshold is represented by the horizontal line (P = 0.01). (C) KS-test P-values for every target class in SNF-Pearson vs. SNF-Euclidean. (D) Venn diagrams illustrating the overlap in target classes scoring as significantly self-associated be KS-test (P < 0.01) using Euclidean distance or Pearson correlation.