All Competitors

Structure-based virtual screening model that jointly predicts protein-ligand complex structures and binding fitness from sequence and SMILES.

Protein-ligand interaction model pretrained on solvent-aware conformer ensembles, reaching 97.1% AUC on DUD-E virtual screening.

MSA-based protein language model for unsupervised contact prediction, outperforming ESM2-15B with 111M parameters and leading on interface contacts.

Binding free energy change (ΔΔG) predictor for protein-protein interfaces, decomposing mutational effects into inverse-folding and energy-model terms.

Protein surface tokenizer that turns surface-exposed residues into codebook tokens, lifting SKEMPI binding affinity change prediction to r = 0.600.

Host-pathogen protein interaction predictor scoring bacterial effector and human protein pairs from frozen ESM-2 embeddings with a transformer.

Proteome-scale protein language model whose representations enable zero-shot protein-protein interaction and gene essentiality prediction.

B-cell epitope predictor fusing ESM-2 embeddings with residue contact and protrusion features to score linear and conformational epitopes.

Flow matching model that builds any non-canonical amino acid into a protein pocket from its SMILES string, at 1.43 Å mean RMSD on held-out ncAAs.

Sequence-based protein-protein interaction predictor over ProtT5 embeddings that reaches 0.70 AUROC on the leakage-free gold standard benchmark.

Protein-protein interface prediction from 3D structure using face-centered surface fingerprints and geometric graph attention, at ROC AUC 0.89.

Bacteriophage gene function prediction from genomic synteny, pairing protein language model embeddings with circular attention. AUC above 0.84.

Multi-state protein inverse folding model that designs one sequence for two conformations, improving sequence recovery 12% over ProteinMPNN.

Enzyme screening framework pairing a sequence-structure CNN classifier with CLIP-style protein-reaction retrieval to link orphan enzymes to genes.

Peptide toxicity prediction that fuses frozen ProtT5 residue embeddings with ESMFold-predicted structure in an E(3)-equivariant graph neural network.

Structure-based drug design framework pairing pharmacophore-guided latent diffusion with training-free, pocket-aware evolutionary optimization.

Scientific multimodal foundation model, a 241B-parameter MoE with a tokenizer that reads molecular formulas and protein sequences natively.

Enzyme thermostability prediction from sequence, using segment-level attention over ESM-2 embeddings to rank mutation sites for protein engineering.

Multi-modal protein language model using the MSA evolutionary profile as a reasoning step between structure and sequence. 650M outperforms ESM-3 1.4B.

Protein-ligand binding affinity and mutation ΔΔG predictor fusing residue, ligand, and interaction graphs, evaluated on leak-proof LP-PDBBind splits.

Tandem mass spectrum prediction for intact N- and O-glycopeptides that localizes O-glycosylation sites from HCD spectra alone, without ETD.

Protein language models trained on billions of natural and synthetic sequences for de novo design and zero-shot mutation-effect prediction.

Bacterial genomics foundation model reading whole genomes as ordered protein sequences. Predicts operons, gene essentiality, and phenotypic traits.