A top variant identification pipeline for protein engineering

Understanding the fitness of protein variants with combinatorial mutations is critical for effective protein engineering. In this issue of Cell Systems, Chu et al. present TopVIP, a top variant identification pipeline that enables accurate picking of the greatest number of best-performing protein variants with high-fitness leveraging zero-shot predictor and low-N iterative sampling. Understanding the fitness of protein variants with combinatorial mutations is critical for effective protein engineering. In this issue of Cell Systems, Chu et al. present TopVIP, a top variant identification pipeline that enables accurate picking of the greatest number of best-performing protein variants with high-fitness leveraging zero-shot predictor and low-N iterative sampling. Accurate top protein variant discovery via low-N pick-and-validate machine learningChu et al.Cell SystemsFebruary 9, 2024In BriefProtein engineering involves the screening of a vast combinatorial mutational landscape. We present a top-performing protein variant identification strategy that only requires experimentation on a few selected variants in large combinatorial mutant libraries. Full-Text PDF