Invited Lecture – Connor W. Coley

Artificial intelligence and machine learning have become important components of the computational toolbox that can be used to advance chemical research and discovery. In this talk, I will discuss our group’s work advancing AI/ML as it applies to the broad subfields of synthetic organic chemistry and analytical chemistry. I will describe several approaches to facilitate decision-making during synthesis planning and reaction development, including the long-standing task of computer-aided retrosynthetic analysis. Though most research in “predictive chemistry” focuses on applying known reactivity to new substrates, ongoing work has also started to show promise for reaction discovery. I will also describe our recent work in analytical chemistry, specifically using tandem mass spectrometry data for structure elucidation of unknown small molecule metabolites. A pervasive theme of our research is the use of domain expertise to inform modeling, from formulating chemistry challenges as statistical learning problems to designing new neural network architectures uniquely suited to chemistry data.