mvMAPIT: Leveraging the genetic correlation between traits improves the detection of epistasis in genome-wide association studies.

Date:

Download the slides as PDF.

Multivariate MAPIT is a tool for the analysis of shared genetic architecture of traits. We can detect causal genetic variants by leveraging correlations of epistatic variance components that are due to a shared genetic architecture in complex traits. This shared architecture may be hidden when only looking at trait covariance. By leveraging correlations between the pairwise interaction variance components of multiple traits mvMAPIT gains sensitivity for the detection of epistasis in genome-wide association studies. Marginalizing over all pairwise interactions of a given variant with any other variant accumulates weak effects into a strong signal and avoids the problem of a large multiple testing burden.

Download the poster for the ISMB 2023 poster session as PDF.