# Three Loci Affecting Variance of Body Mass Index in African Americans and Sub‐Saharan Africans

## TL;DR

This study identifies genetic loci that affect the variability of BMI in African Americans and Sub-Saharan Africans, highlighting the role of vQTLs in broad-sense heritability.

## Contribution

The study introduces vQTL analysis to uncover loci affecting BMI variance, contributing to understanding broad-sense heritability.

## Key findings

- Eight vQTL loci for BMI were identified in African Americans.
- Three loci were replicated in Sub-Saharan Africans.
- Two loci showed evidence of gene interactions and ancestry effects.

## Abstract

Conventional genome‐wide association studies (GWAS) are designed to assess the effect of a genetic locus on phenotypic mean by genotype. Such loci explain a proportion of phenotypic variance known as narrow‐sense heritability. In contrast, variance quantitative trait loci (vQTL) are associated with the phenotypic variance by genotype. These loci explain an additional proportion of phenotypic variance and contribute to broad‐sense heritability but not to narrow‐sense heritability. Here, a genome‐wide vQTL analysis in 22,805 African Americans yielded eight loci for body mass index (BMI). Of these loci, three were replicated in 6002 sub‐Saharan Africans. No locus reached genome‐wide significance using the standard additive model. Furthermore, no locus showed evidence for natural selection, haplotype effects, or gene × sex or gene × study interactions. Two loci showed evidence for an effect of locus‐specific ancestry resulting from admixture and for a gene × gene interaction. One locus showed evidence for interaction with diastolic blood pressure, consistent with this vQTL capturing an unmodeled gene × covariate interaction. These analyses demonstrate that relevant BMI loci can be detected by evaluating vQTL and that these loci contribute to the underexplored broad‐sense heritability for this trait.