All Articles

Joint Mathematics-TADS Colloquium

Author: Lona

Speaker: Dr. S. Joshua Swamidass, Washington University in St. Louis

TITLE: Deep Learning in Biology, Medicine, and Chemistry

ABSTRACT: Deep learning, an approach to machine learning, is coming of age, establishing itself as a transformative approach to understanding data, enabling advances in several scientific domains. Our group has been using deep learning for more than a decade to understand key questions in drug metabolism and healthcare. For example, drugs often become “bioactivated” by metabolism into toxic metabolites, and we showed this complex process can be accurately modelled by graph neural networks. Likewise, we showed that convolutional neural networks can increase the supply of kidneys for transplantation, by more precisely and reproducibly measuring key predictors of kidney graft survival. In these problems, and many others, deep learning is showing how mathematical and computational advances are enabling important advances in science and healthcare.

BIO: S. Joshua Swamidass MD PhD is a physician and a scientist, associate professor of Laboratory and Genomic Medicine at Washington University in St Louis. HIs group uses machine learning to answer questions at the intersection of medicine, biology, and chemistry.