ASSET Seminar: How to Design Molecules that Dock Well but Can’t Exist (Jake Gardner, University of Pennsylvania)

Machine learning has become an indispensable aid to researchers developing the next generation of novel therapeutics. In this talk, I will discuss how some of the most important problems in virtual screening for new potential drug molecules can be cast as black-box optimization problems, where the goal is to find molecules maximizing some desired property — for example, the binding affinity to a known drug target. By leveraging recent work on representation learning for molecules and high dimensional black-box optimization, we are able to achieve up to a 20x performance improvement over state of the art on several of the most widely used benchmarks for molecule design. I will then show how this powerful new approach reveals flaws in tools commonly used for computational molecule design. Even the most widely used docking simulators can be fooled by a sufficiently powerful optimizer producing molecules that could not plausibly exist in nature — a challenge reminiscent of adversarial image generation in computer vision. These findings can be mitigated to a degree through the use of constrained optimization, but also motivate adapting lessons from robust machine learning to the docking simulators themselves.