Plenary Speaker

Guowei Wei, Department of Mathematics, Michigan State University

Topology-based Deep Learning for Drug Discovery

Designing efficient drugs for curing diseases is of essential importance for the 21st century's life science. It involves an extremely complicated procedure, including disease identification, target hypothesis, virtual screening, drug structural optimization, preclinical in vitro and in vivo tests, clinical trials and finally optimizing drug's efficacy, toxicity, and pharmacokinetics properties. We integrate algebraic topology and deep learning algorithms for high throughput drug screening and design, including the predictions of drug binding poses, binding affinity, solubility, partition coefficient, and toxicity. We demonstrate that the proposed mathematical strategies outperform other conventional methods.