Phenome-wide association study of population-differentiating genetic variants around gene ACSL1

Abstract

Abstract Background and objectives Demographic dynamics and natural selection during human evolution shaped the present-day patterns of genetic variations, and geographically varying genetic factors contribute to different disease prevalences across human populations. This study aims to evaluate the presence of positive selection on the gene encoding long-chain fatty acyl-CoA synthetase 1 (ACSL1) and the phenotypic impacts of population-differentiating genetic variants around this gene. Methodology Three types of statistical tests for positive selection, based on site frequency spectrum, extended haplotype homozygosity, and population differentiation, were applied to the whole-genome sequencing data from the 1000 Genomes Project. A phenome-wide association study of ACSL1 was performed with published genome-wide association studies (GWAS) and transcriptome-wide association studies (TWAS), including phenome-wide studies in biobanks. Results Genetic variants associated with ACSL1 expression in various tissues exhibit geographically varying allele frequencies. Three types of statistical tests consistently supported the presence of positive selection on the coding and regulatory regions of ACSL1 in African, European, South Asian, and East Asian populations. A phenome-wide association study of ACSL1 revealed associations with type 2 diabetes, blood glucose, age at menopause, mean platelet volume, and mean reticulocyte volume. The top allele associated with lower diabetes risk has the highest frequency in European populations, while the top allele associated with later menopause has the highest frequency in African populations. Conclusions and implications Positive selection on ACSL1 resulted in geographically varying genetic variants, which may contribute to differential phenotypes across human populations, including type 2 diabetes and age at menopause.