Root-aligned SMILES: A Tight Representation for Chemical Reaction Prediction

Abstract

Chemical reaction prediction, involving forward synthesis and retrosynthesis prediction, is a fundamental problem in organic synthesis. A popular computational paradigm formulates synthesis prediction as a sequence-to-sequence translation problem, where the typical SMILES is adopted for molecule representations. However, the general-purpose SMILES neglects the characteristics of chemical reactions, where the molecular graph topology is largely unaltered from reactants to products, resulting in the suboptimal performance of SMILES if straightforwardly applied. In this article, we propose the root-aligned SMILES (R-SMILES), which specifies a tightly aligned one-to-one mapping between the product and the reactant SMILES for more efficient synthesis prediction. Due to the strict one-to-one mapping and reduced edit distance, the computational model is largely relieved from learning the complex syntax and dedicated to learning the chemical knowledge for reactions. We compare the proposed R-SMILES with various state-of-the-art baselines and show that it significantly outperforms them all, demonstrating the superiority of the proposed method.

Chemical Science 2022. Main paper: 16 pages, 5 figures, and 6 tables; supplementary information: 8 pages, 5 figures and 3 tables. Code repository: https://github.com/otori-bird/retrosynthesis

References in corpus (4)

• Masked Autoencoders Are Scalable Vision Learners
• Predicting Organic Reaction Outcomes with Weisfeiler-Lehman Network
• Linking the Neural Machine Translation and the Prediction of Organic Chemistry Reactions
• RetCL: A Selection-based Approach for Retrosynthesis via Contrastive Learning

Cited by in corpus (8)

• MechRetro is a chemical-mechanism-driven graph learning framework for interpretable retrosynthesis prediction and pathway planning
• ${\mathsf{G}}^2\mathsf{Retro}$ as a Two-Step Graph Generative Models for Retrosynthesis Prediction
• Re-evaluating Retrosynthesis Algorithms with Syntheseus
• Developing ChemDFM as a large language foundation model for chemistry
• Node-Aligned Graph-to-Graph (NAG2G): Elevating Template-Free Deep Learning Approaches in Single-Step Retrosynthesis
• UAlign: Pushing the Limit of Template-free Retrosynthesis Prediction with Unsupervised SMILES Alignment
• DirectMultiStep: Direct Route Generation for Multistep Retrosynthesis
• RetroMPA: A Molecular Property-Aware Auxiliary Framework for Enhancing Retrosynthesis Prediction