I’m a postdoc at Columbia University in the Shen Lab, where I focus on applying AI for biology. My work centers on developing and interpreting deep learning models, with a particular emphasis on protein language models (pLMs). I’m interested in exploring how protein structural dynamics can be incorporated into deep learning models and how to accurately predict protein fitness landscapes. I developed SeqDance and ESMDance, two pLMs trained on protein dynamics data. I explained the scaling behaviour of pLMs on fitness prediction. I also developed MotifAE for unsupervised discovery of functional motifs from pLMs.
I earned my Ph.D. in Bioinformatics from Peking University in the Li Lab, where I built computational tools for studying phase separation and protein degradation.
I was born in Zhucheng, China, my Chinese name is 侯超. I like basketball, photography, and skiing.
Updated: Nov 2025