Hyebin Song

The paper investigates whether proteins with native entanglements are more prone to misfolding and associated with disease, finding significant statistical associations.

The paper introduces EntDetect, a tool to detect non-covalent lasso entanglements in protein structures, simulations, and mass spectrometry data, and demonstrates its application in identifying misfolded states.

The paper examines how modifier variants affect clinical outcomes in individuals with primary disease variants, showing context-dependent phenotypic associations and synergistic interactions.

The paper identifies a widespread protein misfolding mechanism involving non-covalent lasso entanglement (NCLE) and shows that chaperones rescue it more effectively for essential proteins.

This paper develops a statistical learning framework to infer protein fitness landscapes from laboratory evolution data and applies it to DHFR evolution to understand epistasis and design future proteins.

This paper develops a positive-unlabeled learning framework to infer sequence-function relationships from DMS data and applies it to design stabilized enzymes.