Jason Yim

RFdiffusion2 scaffolds unindexed atom-level catalytic active sites via flow matching without inverse rotamer enumeration, solving all 41 in silico benchmark cases and yielding active de novo enzymes across four reactions.

This paper designs highly active metallohydrolases using a new generative AI method, achieving kcat/KM values up to 53,000 M-1 s-1 without experimental optimization.

This paper introduces RFD2-MI, a deep learning framework for designing phosphotyrosine peptide binders, showing high specificity and affinity for phosphorylated targets.

BioEmu, a diffusion model trained on structures, molecular dynamics trajectories, and stability assays, samples protein equilibrium ensembles and folding free energies within ~1 kcal/mol in GPU-hours.

Fine-tuning RoseTTAFold as a diffusion denoiser gives RFdiffusion, a general protein backbone generator; a designed influenza haemagglutinin binder matched its model by cryo-EM.

AlphaFold trained on complexes, with cross-chain paired alignments and symmetry-aware losses, gets 70% of recent heteromeric interfaces acceptable or better (DockQ ≥ 0.23), 27 points above linker-joined AlphaFold.