Associate Professor of Pathology & Immunology
S. Joshua Swamidass, MD PhD
I am a physician and computational scientist, and I lead a research group at Washington University School of Medicine. We use deep learning and other computational approaches to solve problems at the intersection of chemistry, biology and medicine: how the body metabolizes drugs, why some drugs become toxic, which mutations drive cancer, and how to guide clinical decisions from pathology images.
Research themes
All researchDrug metabolism and toxicity
Deep learning models that predict how the liver transforms drugs, and when those metabolites become reactive and toxic.
Deep learning of molecular structure
Neural networks that learn chemistry and quantum mechanics directly from molecular structure.
Chemical informatics
Foundational methods for representing, searching and comparing small molecules.
Screening and drug repurposing
Economic and statistical methods that make high-throughput screens more efficient and informative.
Genomics and bioinformatics
Statistical and deep learning methods for finding the mutations that drive cancer, and other genomic analyses.
AI in pathology
Deep learning that quantifies tissue in pathology images and supports time-critical clinical decisions.
AI in healthcare
Machine learning on electronic health records and clinical data, built to be interpretable, calibrated and fair.
Free for academic use
XenoSite
Web tools that predict how small molecules are metabolized by the liver and whether their metabolites are likely to be reactive. Each model comes from a peer-reviewed study by our group.
Open XenoSite
Recent publications
All publications- Domain Knowledge Inclusive Monotonic Neural Network Guides Patient-Specific Induction of General Anesthesia DosingA A Pract 2025;19(8):e02034DOIPubMed
- Superior transplant recipient outcome prediction and pathology assessment using rapid deep learning applied to procurement kidney biopsiesSci Rep 2025;16(1):1921DOIPubMedFree full text
- Development of a deep learning algorithm for Paneth cell density quantification for inflammatory bowel diseaseEBioMedicine 2024;110:105440DOIPubMedFree full text
- GAiN: An integrative tool utilizing generative adversarial neural networks for augmented gene expression analysisPatterns (N Y) 2024;5(2):100910DOIPubMedFree full text
- Role of Artificial Intelligence in Kidney Pathology: Promises and PitfallsKidney360 2024;5(7):1044-1046DOIPubMedFree full text