Foundation models for genomics
I build AI/ML models that learn regulatory information from DNA sequence and genetic variation, with an emphasis on scalable and biologically meaningful representations for human genomics.
Harvard Medical School · Department of Biomedical Informatics
Boyang Fu (付伯阳)
AI/ML researcher in human genetics and genomics
I develop machine learning methods for human genetics, functional genomics, and perturbation biology. My research focuses on building models that connect DNA sequence, genetic variation, and regulatory intervention to molecular and cellular phenotypes, and eventually to understand human complex diseases.
I am a postdoctoral research fellow at Harvard Medical School, advised by Prof. Marinka Zitnik. I received my Ph.D. in Computer Science from UCLA, where I was advised by Prof. Sriram Sankararaman.
Research
Broadly, I am interested in using AI/ML to understand how human genomes encode regulation, how perturbations reshape cellular state, and how genetic variation talks to each other to influence complex human traits and diseases.
Perturbation biology
I study how sequence, regulatory context, and intervention shape transcriptomic responses, aiming to connect DNA-level regulation to cellular state changes under perturbation.
Statistical genetics and disease modeling
I develop methods to characterize nonlinear variant effects, integrate multi-omic data, and improve causal interpretation of complex disease biology.