Regression results for GC content across variant subtypes for rare variants, common variants, and substitutions. The relationship between local GC content and the observed conditional variant proportion for seven variant subtypes: (A) AT > GC, (B) AT > CG, (C) AT > TA, (D) CpG GC > AT, (E) GC > AT, (F) GC > TA, and (G) GC > CG. Filled points show the conditional variant proportions in each GC content bin, scaled by the intercept of the logistic regression, where is the intercept calculated in the regression, is the count of the given variant subtype, and is the number of ancestral invariant sites that could produce the given subtype in the th GC content bin. Symbol size represents the proportion of the given variant subtype falling into a given GC-content bin. The solid lines show the fitted logistic regression curve, where is the slope fitted in the logistic regression and is the GC content in the th bin. The gray dashed line represents the baseline of no effect, . Legends in each subplot show the regression slope calculated for each variant class and its significance. (***) P-value < 0.0001, (**) P-value < 0.001, (*) P-value < 0.01.