Recursion Pharmaceuticals

Protein-ligand binding affinity prediction from sequence and SMILES, without MSAs. Coarse-grained cofolding runs over 10x faster than Boltz-2.

Transcriptomics foundation model from Recursion that masks and reconstructs RNA-seq gene expression counts to learn reusable sample embeddings.

Conditional chemical language model prompted with a protein target and mechanism of action to score and design molecules without structural input.

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

Transcriptomic perturbation prediction across unseen single and double gene knockdowns and unseen cell lines, driven by gene-gene knowledge graphs.

De novo small molecule generation that assembles drug-like graphs atom by atom, pretrained on cheap property proxies and finetuned per objective.

Cell Painting microscopy foundation model, a channel-agnostic masked autoencoder producing morphological embeddings for zero-shot phenotypic analysis.

Cell microscopy foundation model with a 1.9-billion-parameter masked autoencoder producing embeddings that stay consistent across screening batches.

Microscopy foundation model for high-content screening, embedding genome-scale CRISPR knockout and compound perturbations from Cell Painting images.

Bulk tumor transcriptome model ensembling hundreds of variational autoencoders into interpretable cancer-specific latent spaces for 18 cancers.