# Polygenic scores capture genetic modification of the adiposity-cardiometabolic risk factor relationship

## TL;DR

This paper introduces a new method to compare different types of polygenic scores for detecting gene-environment interactions, showing that they can better capture how adiposity affects cardiometabolic risk factors.

## Contribution

The study presents a generalized pipeline for comparing interaction-based polygenic scores and demonstrates their improved performance in detecting genetic modification of adiposity-related risk factors.

## Key findings

- Interaction-based polygenic scores (iPGS) showed more consistent and replicable gene-environment interactions in the UK Biobank.
- The iPGS approach revealed a 72% stronger association between BMI and alanine aminotransferase in the All of Us dataset.
- PGS-by-adiposity interactions were significant for 16 out of 20 cardiometabolic risk factors analyzed.

## Abstract

Polygenic scores (PGSs) that can predict response to interventions can facilitate precision medicine and are detectable in observational datasets as PGS-by-exposure (PGS×E) interactions. PGSs based on interactions (iPGSs) or variance effects (vPGSs) may be more powerful than standard PGSs for detecting PGS×E, but these have yet to be systematically compared. We describe a generalized pipeline for developing and comparing these PGS types and apply it to detect genetic modification of the relationship between adiposity (measured by BMI) and a broad set of cardiometabolic risk factors. Our applied analysis in the UK Biobank identified significant PGS×BMI for 16/20 risk factors, most consistently for the iPGS approach. Many interactions replicated in All of Us (AoU); for example, we observed a 72% larger BMI-alanine aminotransferase association in the top iPGS decile in AoU. Our study provides a framework for the comparison of PGS×E strategies and informs efforts toward clinically useful response-focused PGSs.

•Our pipeline compares polygenic score (PGS) types for the detection of interactions•PGS-by-adiposity interactions impact cardiometabolic risk factors in the UK Biobank•PGS built from interaction effects show more consistent and replicable interactions•PGS particularly strongly modifies the adiposity-liver biomarker association

Our pipeline compares polygenic score (PGS) types for the detection of interactions

PGS-by-adiposity interactions impact cardiometabolic risk factors in the UK Biobank

PGS built from interaction effects show more consistent and replicable interactions

PGS particularly strongly modifies the adiposity-liver biomarker association

Polygenic scores can improve the power to detect gene-environment interactions, with implications for genome-wide interventions. Westerman et al. introduce a framework for comparing the performance of several types of polygenic scores, built from genetic main, interaction, and variance effects. They find a broad signal for interactions with adiposity impacting cardiometabolic biomarkers.