Fig. 2: Characterization of gain-/loss-of-function missense variants. | Nature Communications

Fig. 2: Characterization of gain-/loss-of-function missense variants.

From: PreMode predicts mode-of-action of missense variants by deep graph representation learning of protein sequence and structural context

Fig. 2

a Genome wide comparison of gain- and loss-of-function variant sites’ pLDDT (AlphaFold2 prediction confidence), protein folding energy change, relevant solvent accessibility, and conservation in 2043 GoF, 7889 LoF variants across 1300 genes and 13704 benign variants from same set of genes, p values were calculated by two-sided Mann-Whitney test. b Protein family wise comparison of gain- and loss-of-function variants enriched secondary structures, statistical tests were performed with two-sided binomial test followed by Benjamin-Hochberg procedure to control the FDR, “*” indicates statistically significant. c Gene wise comparison of gain- and loss-of-function missense variants numbers. Red colored and labeled genes have more than 15 GoF and 12 LoF variants.

Back to article page