Current Status: Genetics and Predicting Disease Risk
The majority of participants in genome wide association studies are of European genetic ancestry (~86%) with the rest ~10% of Asians, ~2% African-American or Afro-Caribbean, ~1.3% Hispanic/Latino (https://lnkd.in/gbYtTzN).
The Population Architecture Genomics and Epidemiology (PAGE) team analyzed ~50,000 non-Europeans for just 26 phenotypes and demonstrated that at least 132 SNPs had differences in effect size across ancestral populations (https://lnkd.in/gAFg8VE). Scaling these numbers to the GWAS catalog suggest a very large number of associations would be affected by ancestry. The effect size of at least some (if not many) variants may also be differentially affected by diet and environment further complicating simplistic use of gene-variant associations.
The inescapable conclusions are (i) that single SNP or polygenic risk scores can currently be used for individuals of European ancestry, (ii) ignoring ancestry in direct-to-consumer or clinical decisions may have serious health consequences (example in paper), and (iii) the continued lack of diversity of participants in GWAS (and other genetic studies) exacerbates existing health disparities and prevents equitable development of precision medicine and nutrition. #precisionmedicine #personalizednutrition