A novel joint location-scale testing framework for improved detection of variants with main or interaction effects

TL;DR

This paper introduces a new joint location-scale testing framework that enhances detection of genetic variants with main or interaction effects, suitable for large-scale genomic studies without explicit interaction modeling.

Contribution

It presents a simple, effective method combining location and scale tests using Fisher's method for improved genetic association analysis.

Findings

Effective in detecting variants with complex effects

Applicable to single-variant, gene-set, and pathway analyses

Suitable for large-scale genome scans and meta-analyses

Abstract

We propose a novel and easy-to-implement joint location-scale association testing procedure that can account for complex genetic architecture without explicitly modeling interaction effects, and is suitable for large-scale whole-genome scans and meta-analyses. We focus on Fisher's method and use it to combine evidence from the standard location test and the more recent scale test, and we describe its use for single-variant, gene-set and pathway association analyses.