Epistatic contributions to human traits via transcription factor mechanisms Epistasis causes an individual’s genetic background to modulate a DNA variant’s effect on trait. Epistatic interactions among different loci in human complex traits are expected to be widespread but have not been found. This could be due to small interaction effect sizes and susceptibility to model-misspecification bias, the statistical complexity of estimating interactions that is higher than marginal variant effects, and a substantial multiple testing burden in a genome-wide scan. Targeting interacting variants that contribute to the same biological pathway could lighten this burden. In this talk, I will present a novel approach rooted in semi-parametric efficient estimation theory, integrating population genetics, functional genomics and targeted machine learning (TarGene), to quantify epistatic contributions to human traits via transcription factor mechanisms. By taking experimentally verified differentially binding variants across 9 nuclear hormone receptors as candidates and using UK Biobank data across 768 traits, we reveal, for the first time, hundreds of epistatic interactions involving these transcription factor mechanisms. The material of this talk is from our paper in Biostatistics, preprint on medRxiv, Julia package TMLE.jl, and population genetics nextflow pipeline TarGene. This article was published on 2025-11-27
