Dr. Pat Walters

Dr. Pat Walters is Chief Data Officer at Relay Therapeutics in Cambridge, MA. Prior to joining Relay, he spent more than 20 years at Vertex Pharmaceuticals where he was Global Head of Modeling & Informatics. Pat is the 2023 recipient of the Herman Skolnik Award for Chemical Information Science from the American Chemical Society. He is a member of the editorial advisory boards for the Journal of Chemical Information and Modeling and Artificial Intelligence in the Life Sciences, and previously held a similar role with the Journal of Medicinal Chemistry. Pat is co-author of the book “Deep Learning for the Life Sciences”, published in 2019 by O’Reilly and Associates. He received his Ph.D. in Organic Chemistry from the University of Arizona where he studied the application of artificial intelligence in conformational analysis. Prior to obtaining his Ph.D., Pat worked at Varian Instruments as both a chemist and a software developer. He received his B.S. in Chemistry from the University of California, Santa Barbara.

Combining Active Learning, Synthesis on Demand Libraries, and Fragment Screening in Early Drug Discovery

The introduction of ultra-large screening libraries has presented both opportunities and challenges for virtual screening. With incredibly vast collections like the Enamine REAL and WuXi GalaXi, brute-force screening is no longer a practical solution. To address this, computational groups are actively developing innovative approaches. One such approach is Thompson Sampling (TS), an active learning method that utilizes machine learning models as surrogates for more computationally intensive calculations. TS offers a rapid and accurate approach for the virtual screening of multi-billion molecule libraries. It seamlessly integrates with docking, similarity search, and other machine learning models. This presentation will showcase a real-world application of TS and highlight how it complements experimental data.