Epsilon-Genic Effects Bridge the Gap Between Polygenic and Omnigenic Complex Traits
Wei Cheng, Sohini Ramachandran, and Lorin Crawford
Received Date: 9th April 19
Traditional univariate genome-wide association studies generate false positives and negatives due to difficulties distinguishing causal variants from "interactive" variants (i.e., variants correlated with causal variants without directly influencing the trait). Recent efforts have been directed at identifying gene or pathway associations, but these are often computationally costly and hampered by strict model assumptions. Here, we present gene-ϵ, a new approach for identifying statistical associations between sets of variants and quantitative traits. Our key innovation is a recalibration of the genome-wide null model to include small-yet-nonzero associations emitted by interactive variants, which we refer to as "epsilon-genic" effects. gene-ϵ efficiently identifies core genes under a variety of simulated genetic architectures, achieving up to ~90% true positive rate at 1% false positive rate for polygenic traits. Lastly, we apply gene-ϵ to summary statistics derived from six quantitative traits using European-ancestry individuals in the UK Biobank, and identify gene sets that are enriched in biological relevant pathways.
Read in full at bioRxiv.
This is an abstract of a preprint hosted on an independent third party site. It has not been peer reviewed but is currently under consideration at Nature Communications.