Z. J. Yang

Since fall 2020, I started as an assistant professor of chemistry at Vanderbilt University. I was fascinated by how protein engineering expands what’s possible, creating enzymes with capacities absent from nature and inaccessible to small-molecule catalysis. Yet the deeper opportunity, a principled, systematic, mechanistically explainable path to design, remains untapped, as the field relies on screening-heavy protocols to identify variants that “work” without revealing why. I believe that the field lacks approaches to systematically derive predictive and generalizable principles linking sequence mutations to function via changes in free energy landscapes. [1] While AI models can accelerate variant discovery, AI alone cannot yield generalizable models given the inherently small, sparse datasets in biomolecular research.

Through combining computational and experimental approaches, my group develops first principle-based and AI-driven approaches to uncover fundamental molecular laws linking mutations to function, validates them with biochemical and biophysical assays, and leverages them to design synthetic, therapeutic, and industrial enzymes beyond the reach of screening-based engineering. To this end, we are developing Mutexa, a physics-informed, AI-assisted framework for “intelligent” protein engineering that generates testable hypotheses and mechanistic insight, thereby identifying function-enhancing variants beyond the reach of empirical high-throughput screening methods. [2]

I was born in Tianjin, China. I graduated from Nankai University with a B.S. degree in Chemistry (Po-Ling class) in 2013. I received my Ph.D. in the Department of Chemistry at the University of California, Los Angeles in 2017. Later, I was a postdoctoral scholar in the Department of Chemical Engineering at Massachusetts Institute of Technology until 2020.

My CV: CV