Investigators at UCLA used the UCLA ATLAS Community Health Initiative biobank to map genotype–phenotype associations across ancestries and identify pharmacogenomic signals tied to semaglutide response. The work, published in Cell, combined EHR data for 92,164 participants with genotype and exome-sequencing data from 61,797 individuals. The study reported ancestry-stratified associations across multiple disease phenotypes, including links to blood lipid levels and peripheral vascular disease, and signals involving cystic kidney disease and diabetes. In a pharmacology-focused analysis of 7,340 ATLAS participants with semaglutide prescriptions, the team found that a type 2 diabetes polygenic risk score was negatively associated with weight-loss response. For biotech and clinical development teams, the message is specific: diverse biobank sampling within a single health system can improve discovery of ancestry-informed risk and response patterns that are typically underpowered or missing in more homogenous datasets.