Genome-wide Association Studies of Type 2 Diabetes in Samoan Adults

Background: Type 2 diabetes (T2D) is a major public health concern for the nation of Samoa and the U.S. territory of American Samoa in the South Pacific. There have been no genome-wide studies of T2D or associated phenotypes in individuals of Polynesian ancestry, whose population history might result in allele frequencies allowing for the observation of associations undetectable in other populations. Here, we performed a genome-wide association study (GWAS) of T2D. Since BMI influences T2D, we performed two additional GWASs, one only including people above the Polynesian obesity threshold (BMI > 32 kg/m2) and another including only people below the obesity threshold. Methods: Participants in the study were 5,266 Samoan individuals from both independent and American Samoa. Genotypes came from two sources: first, up to 659,492 variants were genotyped using genotyping arrays, and then up to 9 million additional variants were imputed using a Samoan-specific haplotype reference panel. I performed association testing using logistic mixed models adjusting for the fixed effects of age, sex, study, and principal components of ancestry derived through PC‑AiR and for random effects of kinship derived from PC-Relate. Results: One unique locus was associated with T2D risk in the meta-analysis of T2D at p < 5 × 10−8, and an additional 72 unique loci were associated at p < 1 × 10−5. The genome-wide significant signal is in the transcript of two genes: RFFL, which encodes an E3 ubiquitin-protein ligase, and AC004223.3, a candidate for nonsense-mediated mRNA decay. The peak variant was associated with T2D among all individuals (p = 2.12 × 10−8). Conclusion: I observed several previously known and novel genetic loci suggestively associated with T2D in Samoans. This indicates that while some of the genetic architecture of T2D is shared across ancestries, there may be unique associations among Polynesians. Additional studies will be necessary to validate these associations and determine the biological underpinnings of these associations, which may point to previously unknown biological mechanisms or social determinants