I am a tenure-track Assistant Professor in the Department of Biostatistics and Bioinformatics at Duke University, where I develop AI-driven methods for biomedical data science. My research focuses on building intelligent, generalizable systems that integrate machine learning, large language models, and multimodal biological data to advance scientific discovery.
A major direction of my work is the use of large language models in biomedical research. I study how models such as GPT-4 can perform tasks including cell-type annotation, code generation, and automated data analysis, and how they compare to human experts. My goal is to transform biomedical data analysis from manual, fragmented pipelines into adaptive, AI-guided systems that improve accessibility, efficiency, and reproducibility.
In parallel, I develop AI and deep learning methods for spatial transcriptomics and biomedical imaging, enabling the integration of molecular and spatial information to better understand tissue organization. I also apply these approaches to study cellular senescence and aging at single-cell resolution.
We contribute to large-scale consortia, including the Cellular Senescence Network (SenNet) and the Encyclopedia of DNA Elements (ENCODE).
My lab is primarily supported by the National Institute of General Medical Sciences (NIGMS). I also serve as an Editorial Board Member for Genome Biology.