Learning to Predict Chemical Reactions

By Matthew A. Kayala, Chloé-Agathe Azencott, Jonathan H. Chen, Pierre Baldi. 2011

175 citations recorded by the source corpus.

Being able to predict the course of arbitrary chemical reactions is essential to the theory and applications of organic chemistry. Approaches to the reaction prediction problems can be organized around three poles corresponding to: (1) physical laws; (2) rule-based expert systems; and (3) inductive machine learning. Previous approaches at these poles respectively are not high-throughput, are not generalizable or scalable, or lack sufficient data and structure to be implemented. We propose a new approach to reaction prediction utilizing elements from each pole. Using a physically inspired conceptualization, we describe single mechanistic reactions as interactions between coarse approximations of molecular orbitals (MOs) and use topological and physicochemical attributes as descriptors. Using an existing rule-based system (Reaction Explorer), we derive a restricted chemistry dataset consisting of 1630 full multi-step reactions with 2358 distinct starting materials and intermediates, associated with 2989 productive mechanistic steps and 6.14 million unproductive mechanistic steps. And from machine learning, we pose identifying productive mechanistic steps as a statistical ranking, information retrieval, problem: given a set of reactants and a description of conditions, learn a ranking model over potential filled-to-unfilled MO interactions such that the top ranked mechanistic steps yield the major products. The machine learning implementation follows a two-stage approach, in which we first train atom level reactivity filters to prune 94.00% of non-productive reactions with a 0.01% error rate. Then, we train an ensemble of ranking models on pairs of interacting MOs to learn a relative productivity function over mechanistic steps in a given system. Without the use of explicit transformation patterns, the ensemble perfectly ranks the productive mechanism at the top 89.05% of the time, rising to 99.86% of the time when the top four are considered. Furthermore, the system is generalizable, making reasonable predictions over reactants and conditions which the rule-based expert does not handle. A web interface to the machine learning based mechanistic reaction predictor is accessible through our chemoinformatics portal (http://cdb.ics.uci.edu) under the Toolkits section.

Subfields: Computational Theory and Mathematics, Control and Systems Engineering, Physical and Theoretical Chemistry, Computational Chemistry & Drug Discovery

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computational Theory and Mathematics

In Control and Systems Engineering

In Physical and Theoretical Chemistry

In Computational Chemistry & Drug Discovery

Citations in the corpus

Predicting Organic Reaction Outcomes with Weisfeiler-Lehman Network (2017, methodology): “[8] also presented a template-free approach to predict reaction outcomes.”
Automating drug discovery (2017, methodology): “ReactionPredictor is such a method and automatically identifies and ranks electron-transfer steps by use of a simplified molecular orbital description 157 .”
Neural Networks for the Prediction of Organic Chemistry Reactions (2016, unknown): “…similar enough to a molecule already in its database of 7 million reactions, identifying both one-pot reactions that do not require time-consuming puri ﬁ cation of intermediate products 18 and full multistep reactions that account for the cost of the materials, labor, and safety of the…”
Detecting Cardiovascular Disease from Mammograms with Deep Learning (2017, unknown): “have been used to successfully tackle a variety of problems in engineering, ranging from computer vision [24]–[27] to speech recognition [28], as well as in the natural sciences, in areas ranging from high energy physics [29], [30], to chemistry [31]–[33], and to biology [34]–[36].”
Learning a Local-Variable Model of Aromatic and Conjugated Systems (2018, background): “15 Consequently, most e ﬀ ort has focused on di ﬀ erent types of convolutional networks, which are e ﬃ cient because they locally aggregate information in molecular graphs 14,29 (Figures 1A and 5A − D).”