Moderating heritability with genomic data

Abstract

Environmental moderators may amplify or suppress the heritability (i.e., the proportion of genetic variation) of a phenotype. This genetic sensitivity to the environment is called gene-environment interaction (GxE). Existing GxE methods struggle to identify replicable interactions because they focus on the interaction coefficients. We propose a novel method for estimating GxE heritability using genetic marginal effects from GxE genome-wide analyses and LD Score Regression (LDSC). We demonstrate the effectiveness of our method for body mass index (BMI) treating biological sex (binary) and age (continuous) as moderators. We find robust, interpretable evidence for GxE that is not detected by existing methods.