Features
ExpressionVAE discrete-token encoding for single-cell data
Clinical response prediction from baseline biopsies
Virtual cell models encoding patient biology
Pre-clinical to clinical data integration
Trial design optimization (e.g., power analysis)
Drug rescue identification for failed compounds
Foundation models trained on multi-omics data
Presentation at ICLR '26, CSHL '26, ICML '26
Public research blog with technical deep dives
Proprietary Recursion OS platform for drug discovery
Automated wet lab generating millions of cell experiments per week
Over 50 petabytes of biological and chemical data
NVIDIA BioHive-2 supercomputer for AI model training
Hit identification to IND-enabling studies with improved speed and cost
Potential first-in-class treatments in oncology and rare diseases
Advanced pipeline with phase 1/2/3 clinical trials (REC-4881, REC-3565)
Target identification and validation using machine learning
Generative AI for molecule design and optimization
De-identified patient data integration for precision medicine
Strategic partnerships with pharma and tech leaders
End-to-end drug development capability from preclinical to pivotal studies
Fit-for-purpose dataset generation including ADME and safety profiling
Real-world evidence generation through clinical trials
Continuous feedback loop from experiments into AI models
