Micha Livne

Reward-guided inference-time search over the Proteína-Complexa sequence-structure codesign model yields validated binders across 127 targets, kinases and a free carbohydrate, with codesigned sequences outperforming ProteinMPNN redesign.

The paper expands the AlphaFold Protein Structure Database to include proteome-scale quaternary structures, predicting 1.81M high-confidence protein complexes and revealing conserved structural clusters across domains of life.

ProFam is an open-source protein family language model that improves fitness prediction and design by leveraging evolutionary context through homologous sequences.

ProT-VAE designs functional proteins by learning sequence-function relationships using a transformer-based variational autoencoder, enabling conditional generation of novel sequences with high functionality.