All Competitors

PROTAC degradation prediction from molecular graphs of the target, linker, and E3 ligase, combining cross-attention with contrastive learning.

Bacterial protein-compound binding affinity prediction from amino acid sequence and SMILES, evaluated zero-shot on two species held out of training.

Structure-based 3D molecule generation that denoises a ligand at atom level and motif level at once, conditioned on the target protein's pocket.

Goal-oriented de novo molecule design with an instruction-tuned LLM that honors property targets, substructure constraints, and numeric values.

Ternary complex structure predictor for PROTACs and molecular glues, placing E3 ligase, degrader, and target protein in one SE(3)-equivariant pass.

Cell Painting image encoder that turns whole-slide multi-channel microscopy into morphological profiles in one pass, with no cell segmentation step.

Structure-based drug design generator pairing continuous flow matching with discrete Markov bridges to sample 3D ligands inside a protein pocket.

Inverse molecular design conditioned on transcriptomics, generating small molecules intended to revert a diseased cell to a healthy expression state.

Protein-ligand affinity foundation model that embeds pockets and ligands in one space, unifying virtual screening with hit-to-lead optimization.

Broad-spectrum antiviral screening framework pairing a pretrained molecular encoder with ESM-2 embeddings for phenotype- and target-based prediction.

Instruction-tuned LLMs for multi-property molecule optimization, rewriting a hit compound to improve three or more drug properties at once.

3D molecule generation that writes a valid 1D SELFIES string with a pretrained language model, then predicts its conformer with a diffusion module.

Multimodal small-molecule foundation model contrastively pretrained over SMILES, molecular graphs, and fingerprints for antibiotic screening.

Enzyme kcat and KM prediction from sequence and substrate SMILES, binned by order of magnitude so catalytic-site mutations shift the prediction.

Unified science foundation model treating molecules, proteins, RNA, DNA, and materials as one sequence language, in 1B, 8B, and 46.7B sizes.

Kinase-inhibitor binding affinity prediction fusing a contrastively pretrained molecular graph encoder with structure-informed kinase embeddings.

Structure-based drug discovery transformer that handles protein-ligand docking and pocket-aware 3D molecule design in one pretrained model.

Perturbation representation model embedding CRISPR gene targets and small molecules in one space, transferring genetic screen models to drug response.

Autoregressive graph generator that flattens molecules into token sequences, letting a decoder-only transformer sample valid structures in one pass.

Attention architecture fusing docking scores with protein language model embeddings, so an enzyme gets a different representation per substrate.

SMILES generative model for de novo drug design, pretrained on 200 million ZINC20 compounds with a tokenizer built from frequent substructures.

Structure-based drug design model generating 3D ligands inside a protein pocket, aligned by Best-of-K fine-tuning on drug-likeness and docking.

Molecular docking framework that poses several ligands sharing one protein pocket at once, using their consistency to sharpen each prediction.