Computational Approaches to Advance Genomic Medicine

The integration of electronic health record (EHR)-linked biobanks with large-scale clinical, genotyping, sequencing, and multi-omics datasets have created new opportunities for advancing genomic medicine. These data resources enable the development of computational approaches aimed at improving disease prediction and phenotyping, genetic discovery, population-based penetrance estimation and drug target prioritization. In this talk, I will highlight recent research focusing on four key areas: i) Using machine learning and EHR-derived clinical data to improve coronary artery disease risk estimation; ii) Demonstrating how digital markers of disease can enhance the discovery of rare coding variants; iii) Developing a machine learning-based approach to estimate population-based penetrance of genetic variants;  and iv) Developing genetics-based prioritization frameworks to predict drug outcomes. These efforts demonstrate how data-driven approaches can generate biological and clinical insights from large-scale population datasets to advance genomic medicine.