All Competitors

Cross-modal continued pretraining on curated mass-spectrometry proteomes lifts a 70M single-cell model past RNA-only checkpoints far larger.

World model that simulates a human cell as one persistent state, propagating drug and gene perturbations from DNA through to whole-cell morphology.

RNA–small-molecule interaction model that scores binding and localizes the pocket to individual nucleotides in one pass, from sequence and SMILES.

Olfactory receptor-odorant interaction prediction from sequence and SMILES, pairing protein and chemical language models through cross-attention.

Drug-target affinity model scoring small molecules across the whole human proteome to yield tissue-weighted off-target binding profiles.

Molecular glue degrader activity prediction from SMILES and protein sequence, with two-stage cross-attention that follows E3-then-substrate binding.

Infrared spectroscopy foundation model pretrained on 60 million simulated spectra, then adapted to real FTIR measurements of molecules and mixtures.

Single-cell perturbation response prediction by conditional latent diffusion, trained on the Tahoe-100M atlas of 100 million drug-treated cells.

Molecular property prediction model pretrained jointly on SMILES strings and 2D graphs, fusing the two views through bidirectional cross-attention.

Antisense oligonucleotide activity prediction from sequence, position-specific chemistry, dose, and cell context. Spearman 0.5970 on ASO Atlas.

Tandem mass spectrum prediction that builds explicit fragmentation pathways, mapping unknown spectra onto 800 million predicted PubChem spectra.

Cryo-EM density enhancement for protein-ligand binding sites, sharpening weak ligand maps with a 3D Swin-Conv UNet trained on 6,511 complexes.

Chemistry embedding model placing SMILES strings and scientific prose in one vector space, top-tier on 43 of 48 molecular and text benchmarks.

Hybrid diffusion-and-physics docking model that predicts protein-ligand binding poses and generalizes out-of-distribution for virtual screening.

Protein-ligand binding affinity scoring model for virtual screening that generalizes to novel pockets and ligands under strict train-test splits.

Enzyme-substrate specificity prediction by end-to-end co-folding, with no predefined binding pocket. AUROC 0.766 on unseen enzymes and substrates.

Physics-guided distillation that transfers 3D molecular dynamics knowledge into SMILES language models, improving MoleculeNet property prediction.

Dual-target structure-based drug design that fuses two pocket-conditioned Bayesian flow distributions to generate 3D ligands binding both proteins.

Cellular foundation models predicting how human cells respond to genetic and pharmacological perturbation, trained on petascale multi-omic data.

Structure-free virtual screening model co-embedding protein residues and small molecules from sequence and 2D chemistry, scoring a compound in 10 ms.

Peptide-protein co-folding and small-molecule docking with an all-atom diffusion transformer. Recovers 59.1% of peptide interfaces to sub-2 Å RMSD.

Preparation-free protein-ligand binding affinity prediction from a protein sequence and a ligand SMILES, using a cascaded mixture-of-experts fusion.

Mass spectrometry foundation model for untargeted metabolomics and lipidomics, naming and quantifying molecules with no reference library.