Yiquan Wang

This paper shows that adding biological evidence to large language models improves zero-shot protein variant prediction without retraining the model.

MuseDrift generates protein variants with specified sequence identity to a wild-type sequence, achieving higher structural prediction confidence and better mutation-effect scores than existing methods.

This study investigates the constraints on the breadth evolution of the CR9114 broadly neutralizing antibody by comparing binding affinity landscapes across different influenza HA subtypes.

This study introduces oPool+ display, a high-throughput platform for synthesizing and characterizing natively paired antibodies, demonstrating its application in analyzing influenza HA antibodies.

This study uses deep mutational scanning to analyze the functional impact of NEP mutations in influenza A virus, revealing pleiotropic effects on viral replication and host adaptation.

This study identifies sequence constraints for IGHV1-69 antibodies targeting influenza HA stem, showing light chain and CDR H3 sequences coordinate HA stem specificity.