Journal article
Retrosynthetic Reaction Prediction Using Neural Sequence-to-Sequence Models
ACS central science, Vol.3(10), pp.1103-1113
10/25/2017
DOI: 10.1021/acscentsci.7b00303
PMCID: PMC5658761
PMID: 29104927
Abstract
We describe a fully data driven model that learns to perform a retrosynthetic reaction prediction task, which is treated as a sequence-to-sequence mapping problem. The end-to-end trained model has an encoder–decoder architecture that consists of two recurrent neural networks, which has previously shown great success in solving other sequence-to-sequence prediction tasks such as machine translation. The model is trained on 50,000 experimental reaction examples from the United States patent literature, which span 10 broad reaction types that are commonly used by medicinal chemists. We find that our model performs comparably with a rule-based expert system baseline model, and also overcomes certain limitations associated with rule-based expert systems and with any machine learning approach that contains a rule-based expert system component. Our model provides an important first step toward solving the challenging problem of computational retrosynthetic analysis.
Details
- Title: Subtitle
- Retrosynthetic Reaction Prediction Using Neural Sequence-to-Sequence Models
- Creators
- Bowen Liu - Stanford UniversityBharath Ramsundar - Stanford UniversityPrasad Kawthekar - Stanford UniversityJade Shi - Stanford UniversityJoseph Gomes - Stanford UniversityQuang Luu Nguyen - Department of ChemistryStephen Ho - Stanford UniversityJack Sloane - Stanford UniversityPaul Wender - Stanford UniversityVijay Pande - Stanford University
- Resource Type
- Journal article
- Publication Details
- ACS central science, Vol.3(10), pp.1103-1113
- DOI
- 10.1021/acscentsci.7b00303
- PMID
- 29104927
- PMCID
- PMC5658761
- NLM abbreviation
- ACS Cent Sci
- ISSN
- 2374-7943
- eISSN
- 2374-7951
- Publisher
- American Chemical Society
- Grant note
- DOI: 10.13039/100000165, name: Division of Chemistry, award: CHE1265956; DOI: 10.13039/100000060, name: National Institute of Allergy and Infectious Diseases, award: U19 AI109662; DOI: 10.13039/100000048, name: American Cancer Society, award: PF-15-007-01-CDD; DOI: 10.13039/100005883, name: Hertz Foundation; DOI: 10.13039/100000057, name: National Institute of General Medical Sciences, award: R01 GM062868
- Language
- English
- Date published
- 10/25/2017
- Academic Unit
- Chemical and Biochemical Engineering
- Record Identifier
- 9984197161702771
Metrics
20 Record Views