AstraZeneca

Histopathology-to-molecular alignment model that queries H&E slides with gene-set signatures to predict pathway activity without sequencing.

Transformer that predicts protein-RNA binding affinity from Boltz-2 pre-structural embeddings via cross-modal attention, with no 3D structure step.

Flow-matching model that jointly samples 3D de novo molecules and several low-energy conformers, extending to pocket-conditioned ligand design.

Recombinant protein expression predictor scoring soluble E. coli production from construct sequence, with AUROC 0.83 on prospective constructs.

Drug-combination safety prediction that fuses molecular structure, pathway knowledge, cell viability, and transcriptomic response to perturbation.

Variant-origin classifier for cell-free DNA, separating clonal hematopoiesis from tumor-derived mutations without matched white blood cell sequencing.

Disease embeddings learned from human genetic evidence and phenotype ontologies, placing rare and common conditions in one mechanistic vector space.

Antibody language model that generates paired heavy and light variable domains, with a developability-conditioned variant for manufacturable designs.

Antibody language model that reads sequence and backbone coordinates together, so a masked CDR can be recovered from either or both modalities.

SMILES language model pretrained on 100M molecules, transferring to forward reaction prediction, retrosynthesis, molecular optimisation, and QSAR.

Chemical language model for small-molecule drug discovery, embedding SMILES for property prediction and sampling new molecules from its latent space.