Marwin Segler, Machine Learning Researcher at BenevolentAI will be giving a talk on using machine learning to address challenges in Computer-Assisted Automated Synthesis Planning (CASP) at RICT 2019.
Computer-Assisted Automated Synthesis Planning (CASP) is one of the grand challenges in Chemistry, and under investigation since E.J. Corey’s pioneering work in the 1960ies [1,2]. CASP would be a highly valuable tool to find better synthetic routes and to determine the synthesizability of virtual de-novo designed compounds. However, despite several waves of research, CASP was never widely accepted by chemists, because the systems were slow, cumbersome to maintain, and the results were considered to be of unsatisfactory quality [3,4].
In this talk, we present how to address these limitations using machine learning and modern search algorithms to perform CASP [5,6]. First, we show that deep neural networks can be trained over night on very large reaction datasets (here, the complete Reaxys database), to predict and rank the most suitable retrosynthetic transforms to apply to a molecule . This way of training also allows the machine to learn the tolerated and conflicting functional groups of a transform implicitly . In earlier approaches, this information had to be entered manually by experts. Second, we combined deep neural networks with Monte Carlo Tree Search (MCTS) to perform search. Quantitatively, this allows our system to solve twice as many molecules while being two orders of magnitude faster faster than the state of the art . To assess the quality of the predicted routes, we conducted double blind tests. Here, we found, for the first time, that organic chemists could not distinguish between real routes taken from the literature and predicted routes . The talk will close by discussing the limitations and potential future lines of research.
Marwin Segler is Machine Learning Researcher at BenevolentAI. He studied chemistry at Westfälische Wilhelms-Universität Münster, Germany. Duringhis MSc thesis in synthetic organic chemistry in the group of Prof. Olga Garcia-Mancheño, he worked at the bench on C-H functionalization. For his PhD, still in Münster, Marwin joined the group of Prof. Mark Waller to study organic chemistry from a computational perspective. He investigated approaches for the computer-aided invention of novel chemical reaction types, introduced neural generative models to de-novo drug design, and introduced machine-learning to computer-aided retrosynthesis. Marwin now works at BenevolentAI in the Chemistry group.
 G. Vleduts, Information Storage and Retrieval, 1963, 117
 E.J. Corey, W.T. Wipke. Science, 1969, 166, 178
 W.D. Ihlenfeldt, J. Gasteiger, Angew. Chem. Int. Ed., 1996, 34, 2613
 S. Szymkuc et al., Angew. Chem. Int. Ed., 2016, 55, 5904
 M. Segler, M. P. Waller, Chem. Eur. J. 2017, 23, 5966