Figure 2

(A) The non-uniformity of distributions of mutations across annotated genes-tumor suppressors increases with a value of the predicted functional impact for missenese mutations discovered in each of six TCGA projects [3–6, 10, 11]; missense mutations are separated into three groups by the predicted functional impact. The non-uniformity, μ, is defined as a ratio of the total number of affected genes in a dataset to the "effective" number of genes that carry the majority of mutations (Eq.2); μ~1 means that mutations are distributed fairly uniformly across genes; μ≫1 means that majority of mutations are selected in a small fraction of all mutated genes. The non-uniformities of "Silent" and truncating mutations ("Trunc") affecting the same groups of genes are given for comparison. (B). The non-uniformities of mutation distributions computed for different gene groups in glioblastoma (brain) cancer [6]. "All" stands for mutations affecting all genes; TS, OG, CG, nCG, stand for mutations affecting, respectively, annotated tumor-suppressors, annotated oncogenes, annotated cancer genes, genes with no cancer annotations; the non-uniformities of truncating ("Trunc") and "Silent" mutations are computed for truncating and silent mutations affecting all genes.