Clinical genetic variation across Hispanic populations in the Mexican Biobank

Genetic testing for specific alleles is often recommended based on an individual's ancestry. However, the frequency of pathogenic and pharmacogenomic alleles across different Hispanic groups has not been well characterized, and existing guidelines often fail to recognize the geographic and ancestral diversity within these populations. Here analyzing data from 6,011 individuals from the nationwide Mexican Biobank, we show that Mexican individuals have striking regional differences in biomedically relevant allele frequencies, shaped both by their overall admixture proportions, but also by the local Indigenous ancestral groups contributing to their genome (for example, Nahua in central Mexico, Zapotec in the South or Maya in the Yucatan peninsula). We found ancestry-specific patterns with clinical implications that could not have been detected without a local ancestry-informed approach, including variants affecting fentanyl (rs2242480) and statin (rs4149056) metabolism, examples particularly relevant to the epidemiology of Hispanic populations. This analysis framework could inform genetic testing guidelines across the Americas. We are making available the results for 42,769 biomedically relevant genotyped variants through MexVar, a user-friendly platform designed to improve access to genomic data for the scientific community and support genetic analyses for populations of Mexican descent worldwide.