Loading…
Sequence-only interaction model scoring protein-protein and protein-ligand pairs plus functional annotation across whole proteomes in seconds.
Sequence-only protein-protein interaction prediction at proteome scale. Each protein is embedded once, so a pair score compares two stored vectors.
Protein language model trained with masked diffusion, unifying representation learning and generative design in one 650M-parameter model.
Long-context protein language model that reads whole viral genomes, using interaction-guided sparse attention over contexts of 61,000 amino acids.
Protein binding affinity prediction from sequence alone, returning pKd and per-residue interface labels instead of a yes-or-no interaction call.
Reshapes a frozen ESM-2 latent space by contrasting it against EC, GO, InterPro and Gene3D ontology tokens for function-aware protein embeddings.
Pretrains on EC, GO, InterPro and Gene3D ontology tokens with no amino acids, building a protein feature space from curated function alone.
Generates protein sequences from EC, GO, InterPro and Gene3D prompts by cross-attending an ESM-2 decoder onto an annotation transformer encoder.
Extends an ESM-2 token embedding matrix with EC, GO, InterPro and Gene3D tokens so one transformer reads residues and ontology terms together.
Codon-vocabulary protein language model that converts ProtBERT to 64 codon tokens via embedding seeding, masked pretraining, and distillation.
Sequence-pair interaction classifier fine-tuned from ProtBERT-BFD, trained against synthetic negatives generated by BLOSUM62-guided mutation.