Roman A. Melnyk

This study shows that reducing the catalytic activity of immunotoxins while balancing receptor engagement improves their selectivity and therapeutic window.

This study shows that secondary bile acid metabolism, not nutrient competition, reduces C. difficile toxin-mediated disease in vivo, independent of colonization resistance.

This paper designs antibodies de novo with atomic precision using RFdiffusion and yeast display, validated by cryo-EM and biophysical methods.

This paper designs potent small protein inhibitors of C. difficile toxin B using deep learning and Rosetta, showing high affinity and stability for potential clinical use.

The paper designs a potent inhibitor against P. sordellii lethal toxin using deep learning, showing protective effects in mice against toxic shock.

This paper identifies a diphtheria toxin homolog from Austwickia chelonae that evades pre-existing anti-DT antibodies and shows improved cargo delivery compared to diphtheria toxin.

The paper identifies SEMA6A and SEMA6B as receptors for C. sordellii lethal toxin and shows how mutations can switch receptor specificity between related toxins.