LAUSR

Significance Genetic association studies have identified widespread DNA methylation (DNAm) quantitative trait loci (mQTLs) that may illuminate causal genetic variations underlying human complex traits. However, due to extensive linkage disequilibrium (LD) of the genome, it is challenging to identify causal genetic variations that drive DNAm levels. Here we present a deep learning model to predict effects of genetic variations on DNAm levels in the human brain. We demonstrate that deep learning-derived DNAm regulatory variants are not confounded by LD, are convergent with mQTL evidence from genetic association analysis, and underlie the genetic basis of brain-related traits. Our study shows the power of a deep learning approach to identify functional regulatory variants that may elucidate the genetic basis of complex traits.