Genetic Insights into Coronary Artery Disease in Underrepresented Populations: Assessing Two-Sample Mendelian Randomization across Diverse Ancestry Populations
Listed in
This article is not in any list yet, why not save it to one of your lists.Abstract
Understanding the causal effect of modifiable risk factors on a disease is crucial for aiding and shaping public health policies, identifying targets of interventions and advancing our global understanding of health and diseases. With this however, the disproportionate representation of non-European ancestries in research has raised important questions regarding the transferability and reliability of genetic findings on a diverse global scale. In this study, we investigated the feasibility of conducting a two-sample Mendelian randomization (MR) analysis in populations of diverse ancestries, focusing on both methodological challenges and biologically differences when data from European, East Asian, South Asian and African ancestry populations were compared against each other. Employing data generated from large-scale genome-wise association studies (GWAS), we chose to compare the causal effects of lipid traits, blood pressure, body-mass index, type-2 diabetes and kidney function on coronary artery disease. Due to an insufficient number of identified single-nucleotide polymorphisms in non-European data when strict alpha thresholds were employed, we were not able to conduct MR analyses across all ancestry populations until the threshold was relaxed. We found that allowing for a lenient inclusion threshold and extending the MR methodology to be more inclusive of non-European data resulted in an increase in weak instrument bias, resulting in imprecise estimates and a reduced ability to detect true causal effects. Notably, our results showed causal associations known to be inconsistent with established findings, specifically between lipid traits in South Asian populations compared to European ancestry populations. These findings reiterate the urgent need for independent large-scale GWAS in non-European populations, to improve the power and reliability of MR studies as well as to develop methods which take into account population-specific effects.