Neural‐Symbolic Machine Learning for Retrosynthesis and Reaction Prediction

• Published in: Chemistry : a European journal Vol. 23; no. 25; pp. 5966 - 5971
• Main Authors: Segler, Marwin H. S., Waller, Mark P.
• Format: Journal Article
• Language: English
• Published: Germany Wiley Subscription Services, Inc 02-05-2017
• Subjects: artificial intelligence; Artificial neural networks; Chemistry; Expert systems; Machine learning; Neural networks; Organic chemistry; retrosynthesis; synthesis design; total synthesis; total synthesis; machine learning; artificial intelligence; retrosynthesis; synthesis design
• ISSN: 0947-6539; 1521-3765
• DOI: 10.1002/chem.201605499

Reaction prediction and retrosynthesis are the cornerstones of organic chemistry. Rule‐based expert systems have been the most widespread approach to computationally solve these two related challenges to date. However, reaction rules often fail because they ignore the molecular context, which leads to reactivity conflicts. Herein, we report that deep neural networks can learn to resolve reactivity conflicts and to prioritize the most suitable transformation rules. We show that by training our model on 3.5 million reactions taken from the collective published knowledge of the entire discipline of chemistry, our model exhibits a top10‐accuracy of 95 % in retrosynthesis and 97 % for reaction prediction on a validation set of almost 1 million reactions.
Smart synthesis: Computers can be taught to perform retrosynthesis and reaction prediction with deep neural networks. The machine learns to prioritize rules and identify reactivity conflicts. This solves one of the main limitations of current systems for computer‐aided synthesis design.