iRUNNER: A Baseline Mutation Burden Regression for Identifying Gene Interaction Between Rare Variants for Diseases

Genetic interactions play a crucial role in elucidating the susceptibility and etiology of complex multifactorial diseases. Despite significant efforts to identify disease-associated nonlinear effects in genome-wide association studies, efficient methods for detecting the epistatic impact of rare variants remain lacking. In this study, we proposed iRUNNER, a novel and powerful mutation burden test focused on analyzing the interaction effects of rare variants on a binary trait. Different from conventional association tests comparing cases with controls, iRUNNER evaluates the relative enrichment of rare variant interaction burden of pairwise genes in patients against its baseline, estimated by a recursive truncated negative-binomial regression model that leverages multiple genomic features from public databases. Extensive simulations demonstrated that iRUNNER outperforms existing epistasis tests in statistical power and maintains reasonable type I error rates even when population stratification exists in control samples. Applied to real datasets of five complex diseases, iRUNNER yielded substantial gains in gene-gene interaction detections. Notably, the majority of these signals were missed by alternative methods, especially in small to medium-sized samples. Furthermore, we found that these identified gene pairs of each trait can form interconnected networks, which may provide valuable insights into the underlying molecular mechanisms. We have implemented iRUNNER as a module in our integrative platform KGGSeq (http://pmglab.top/kggseq/) that enables rapid testing of pairwise interactions among all possible non-synonymous rare coding variants within hours.