All Competitors

Relative protein-ligand binding affinity prediction from docked complexes, matching Schrodinger FEP+ ranking accuracy zero-shot on the FEP benchmark.

Domain-level language model treating Pfam protein domains as tokens to predict and design bacterial and fungal biosynthetic gene clusters.

RNA-protein contact prediction from sequence, built on ERNIE-RNA and ESM-2 embeddings. Reaches 0.77 auROC where AlphaFold 3 reaches 0.61.

Spatial proteomics foundation model, marker-aware and panel-agnostic, pretrained on 47 million multiplexed tissue-imaging patches from 175 markers.

Codon optimization model for heterologous expression in E. coli, fine-tuning ProtBert to label each residue with an expression-weighted codon.

DNA foundation model pretrained by supervised genomic profile prediction, using mixture-of-experts routing across species and assay types.

Antibody-antigen binding prediction from heavy chain, light chain, and antigen sequence, scoring 0.946 AUROC on a SARS-CoV-2 benchmark.

Open-source framework for building RNA and DNA foundation models, featuring WCED pretraining for transcriptomics and SNP-aware encoding for genomics.

Open model that jointly predicts biomolecular structure and small-molecule binding affinity, approaching FEP+ accuracy in seconds on a single GPU.

Long-context protein language model that reads whole viral genomes, using interaction-guided sparse attention over contexts of 61,000 amino acids.

Cross-modal co-embedding of biosynthetic gene clusters and natural products, enabling bidirectional retrieval between gene cluster and compound.

Multimodal clinical foundation model reasoning jointly over 2D and 3D medical images, ECG time-series, and text reports across nine clinical domains.

Protein conformation ensemble generation aligned to force-field energies, calibrating an AlphaFold 3-style diffusion model against MD thermodynamics.

Molecular embedding model that turns SMILES into SE(3)-invariant vectors for property prediction, similarity search, and compound clustering.

Plant DNA-binding protein prediction that averages a ProtT5 sequence-embedding classifier with a SaProt structure-aware one at the score level.

Antibody caninisation model generating canine framework regions around given CDRs, released with a dataset of 430,000 canine antibody sequences.

mRNA optimization model that raises codon adaptation, tRNA adaptation, and folding stability at once while preserving the encoded protein sequence.

Instruction-tuned LLM series for multi-property molecule optimization that improves named drug properties while preserving those already in range.

Vision foundation model for the tree of life, trained on 214 million organism images across 952,000 taxa for zero-shot species classification.

Protein-ligand docking model for high-throughput virtual screening, predicting binding poses with graph neural networks at low computational cost.

DNA-LLM reasoning model fusing genome foundation model embeddings with an LLM to produce step-by-step pathway and variant effect explanations.

Diffusion protein language model for de novo design conditioned jointly on GO terms, InterPro domains, EC numbers, motifs, and backbone structure.

Missense variant effect predictor that scores mutations from Δ-embeddings — wild-type minus mutant protein language model representations.