Northeastern University

Enzyme-substrate specificity prediction by end-to-end co-folding, with no predefined binding pocket. AUROC 0.766 on unseen enzymes and substrates.

Molecule generation conditioned on single-cell transcriptomes, designing cell-type-specific compounds that revert diseased cell states.

De novo protein binder design that recasts structure-predictor confidence as an energy function, replacing ipTM as the hallucination objective.

Spatial transcriptomics prediction from H&E whole-slide images. One generative checkpoint covers 38,984 genes and 17 organs without fine-tuning.

Autoregressive graph transformer generating molecules as node and edge token sequences, fine-tunable for goal-directed design and property prediction.

Disease embeddings learned from human genetic evidence and phenotype ontologies, placing rare and common conditions in one mechanistic vector space.

De novo enzyme design conditioned on the reaction to be catalysed: substrate and product SMILES in, catalytic pocket, enzyme, and docked complex out.

Predicts binding free energy change (ΔΔG) at protein-protein interfaces by scoring bound and unbound states with an inverse folding model.