Genomic-Relatedness Matching Expands Population Coverage, Improves Power, and Reduces Bias in Genetic Association...

Area of Science:

• Genetics
• Bioinformatics
• Population Genetics

Background:

• Genome-wide association studies (GWAS) are limited to homogeneous, unrelated samples and can be biased by population structure.
• Existing GWAS methods struggle with gene-environment correlation and assortative mating, impacting results in diverse populations.

Purpose of the Study:

• Introduce Genomic-Relatedness-Matched Association (GRMA) studies as a novel, bias-reducing alternative to traditional GWAS.
• Enable genetic association studies in ancestrally diverse populations without requiring ancestry labels.
• Reduce bias and improve precision compared to standard GWAS methods.

Main Methods:

• GRMA matches individuals to control groups based on user-defined pairwise relatedness thresholds.
• Generates single nucleotide polymorphism (SNP)-level summary statistics from within-group associations.
• Utilizes open-source software for computational efficiency in large-scale studies.

Main Results:

• GRMA demonstrated favorable performance over GWAS in bias, precision, and population coverage using UK Biobank and All of Us data.
• GRMA analysis suggests "genetic nurture" is not a significant source of genome-wide bias for common complex traits like BMI, height, and educational attainment.
• The method proved computationally efficient and scalable for large datasets.

Conclusions:

• GRMA provides a robust and inclusive framework for genetic association studies, particularly in diverse populations.
• The findings challenge previous assumptions about "genetic nurture" bias in large-scale GWAS.
• GRMA facilitates broader application of genetic association studies in biomedical research.