Premium
Neural‐Symbolic Machine Learning for Retrosynthesis and Reaction Prediction
Author(s) -
Segler Marwin H. S.,
Waller Mark P.
Publication year - 2017
Publication title -
chemistry – a european journal
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.687
H-Index - 242
eISSN - 1521-3765
pISSN - 0947-6539
DOI - 10.1002/chem.201605499
Subject(s) - retrosynthetic analysis , context (archaeology) , reactivity (psychology) , artificial intelligence , artificial neural network , set (abstract data type) , computer science , machine learning , training set , expert system , chemistry , programming language , geography , archaeology , organic chemistry , medicine , total synthesis , alternative medicine , pathology
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.