Quantifying direct genetic signal captured by principal component adjustment

Abstract

Contemporary genomic studies of complex traits, such as genome-wide association studies and polygenic index (PGI) analyses, frequently include the principal components of the genotype matrix (PCs) among their adjustments for population stratification. In this paper, we explore the extent to which we may be discounting direct genetic effects by adjusting for PCs. Using family-based models that control for parental genotype in the UK Biobank, we find that PCs capture direct genetic effects on all nine phenotypes we tested. These effects are generally small, and extremely similar to the population effects of the same PCs. This suggests that PC adjustment indeed diminishes, albeit very slightly, signals of direct genetic effects. Furthermore, we find that adjusting for PCs does not meaningfully alter estimates of the effect of the current PGIs within families. Our findings suggest that, for the phenotypes and populations studied here, direct genetic effects captured by PCs are modest, and adjusting for PCs does not seem to lead to meaningful attenuation in PGI estimates.