Background: Blood levels of amino acids are important biomarkers of disease and are influenced by synthesis, protein degradation, and gene-environment interactions. Whole genome sequence analysis of amino acid levels may establish a paradigm for analyzing quantitative risk factors.
Results: In a discovery cohort of 1872 African Americans and a replication cohort of 1552 European Americans we sequenced exons and whole genomes and measured serum levels of 70 amino acids. Rare and low-frequency variants (minor allele frequency ≤5%) were analyzed by three types of aggregating motifs defined by gene exons, regulatory regions, or genome-wide sliding windows. Common variants (minor allele frequency >5%) were analyzed individually. Over all four analysis strategies, 14 gene-amino acid associations were identified and replicated. The 14 loci accounted for an average of 1.8% of the variance in amino acid levels, which ranged from 0.4 to 9.7%. Among the identified locus-amino acid pairs, four are novel and six have been reported to underlie known Mendelian conditions. These results suggest that there may be substantial genetic effects on amino acid levels in the general population that may underlie inborn errors of metabolism. We also identify a predicted promoter variant in AGA (the gene that encodes aspartylglucosaminidase) that is significantly associated with asparagine levels, with an effect that is independent of any observed coding variants.
Conclusions: These data provide insights into genetic influences on circulating amino acid levels by integrating -omic technologies in a multi-ethnic population. The results also help establish a paradigm for whole genome sequence analysis of quantitative traits.
Keywords: Amino acids; Metabolomics; Multi-ethnic; Rare variants; Whole genome sequence.