�鶹APP

The Department of Electrical and Computer Engineering presents its fall seminar series featuring guest speaker Jun Wen, a postdoctoral research fellow in the Department of Biomedical Informatics at Harvard Medical School, who will give a presentation on “Enhancing Computational Precision Medicine with Electronic Health Records.” This seminar will take place on Friday, October 10, from 12:45–1:45 p.m. over Zoom.

Abstract

Electronic Health Records (EHRs) have transformed a wide range of translational research. In this talk, I will first introduce PheMART, an EHR-enhanced machine learning framework designed to predict the phenotypic effects of missense variants (MVs), which play a critical role in numerous clinical phenotypes. Unlike existing computational approaches that primarily assess MV pathogenicity without considering phenotypic heterogeneity, PheMART maps millions of MVs, along with 4,179 phenotypes, into a unified low-dimensional metric space, where proximity reflects biological relevance. This approach integrates diverse data sources, including protein language models, protein-protein interactions, protein domains, medical knowledge graphs, and EHRs, to establish a more comprehensive understanding of MV’s phenotypic effects. Besides substantially outperforming existing models, PheMART aids in diagnosing individuals with rare diseases by effectively pinpointing clinical diagnoses and causative MVs. In the second part of my talk, I will present INTERLACE, a novel EHR-enhanced drug repurposing model that predicts drug-disease associations while pinpointing causal genes. INTERLACE achieves this by integrating biomedical knowledge graphs with large-scale EHR data, leveraging structured and real-world evidence to uncover potential therapeutic applications. Together, these approaches highlight the power of EHR-driven computational models in advancing precision medicine.

Biography

Jun Wen is a postdoctoral research fellow in the Department of Biomedical Informatics at Harvard Medical School and will join the Department of Computational Biology at the Mohamed bin Zayed University of Artificial Intelligence (MBZUAI) as an assistant professor in January 2026. He received his Ph.D. in computer science from Zhejiang University in 2020. Wen’s research lies in the intersection of artificial intelligence and computational precision medicine. His work integrates multimodal biomedical data, including electronic health records, knowledge graphs, and foundational models, to predict the phenotypic effects of genetic variants and uncover novel therapeutic opportunities through drug repurposing. Broadly, Wen aims to develop AI-driven computational frameworks that leverage large-scale observational cohorts, precision-medicine knowledge graphs, and multimodal omics data to elucidate complex molecular and clinical relationships among drugs, genetic variants, and diseases, advancing the next generation of precision medicine.