Ariane Mora

Designing enzymes for new-to-nature chemistry and non-natural substrates with AlphaProtein Novo

Reframing enzyme function prediction as conditional generation

Fluxion is presented, a generative deep learning framework that learns enzyme-dependent regioselectivity across cytochrome P450 enzymes by modeling dynamic electron flow trajectories across the enzyme’s catalytic residues by combining both synthetic chemistry and biochemical datasets with protein language model represe...

Squidly harnesses enzyme functional hierarchy and contrastive learning to efficiently predict catalytic residues from sequence

Squidly surpasses state-of-the-art ML annotation methods in catalytic residue prediction while remaining sufficiently fast to enable wide-scale screening of databases.

Discovery of a phenazine–thiol conjugase from sparse data using genome-informed machine learning

It is shown that integrating genome-informed data augmentation with contrastive learning in protein language space enables identification of phenazine-interacting proteins starting from only 14 known phenazine modifying sequences, and coupling comparative genomics with protein machine learning can convert “small data”...