Retrosynthetic reaction prediction using neural sequence-to-sequence models

TL;DR

This paper introduces a neural sequence-to-sequence model for retrosynthetic reaction prediction, trained on patent data, achieving comparable results to rule-based systems and advancing computational retrosynthesis.

Contribution

The paper presents a fully data-driven neural model for retrosynthesis that overcomes limitations of rule-based systems, marking progress in computational chemical analysis.

Findings

Model performs comparably with rule-based expert systems.

Overcomes limitations of rule-based and hybrid machine learning approaches.

Trained on 50,000 reaction examples from patent literature.

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 towards solving the challenging problem of computational retrosynthetic analysis.