Saved in:
Bibliographic Details
Main Authors: Zhuang, Jiaxi, Zhang, Yu, Zhou, Aimin, Qian, Ying
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2510.16588
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908601663094784
author Zhuang, Jiaxi
Zhang, Yu
Zhou, Aimin
Qian, Ying
author_facet Zhuang, Jiaxi
Zhang, Yu
Zhou, Aimin
Qian, Ying
contents Retrosynthesis prediction is fundamental to drug discovery and chemical synthesis, requiring the identification of reactants that can produce a target molecule. Current template-free methods struggle to capture the structural invariance inherent in chemical reactions, where substantial molecular scaffolds remain unchanged, leading to unnecessarily large search spaces and reduced prediction accuracy. We introduce C-SMILES, a novel molecular representation that decomposes traditional SMILES into element-token pairs with five special tokens, effectively minimizing editing distance between reactants and products. Building upon this representation, we incorporate a copy-augmented mechanism that dynamically determines whether to generate new tokens or preserve unchanged molecular fragments from the product. Our approach integrates SMILES alignment guidance to enhance attention consistency with ground-truth atom mappings, enabling more chemically coherent predictions. Comprehensive evaluation on USPTO-50K and large-scale USPTO-FULL datasets demonstrates significant improvements: 67.2% top-1 accuracy on USPTO-50K and 50.8% on USPTO-FULL, with 99.9% validity in generated molecules. This work establishes a new paradigm for structure-aware molecular generation with direct applications in computational drug discovery.
format Preprint
id arxiv_https___arxiv_org_abs_2510_16588
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Copy-Augmented Representation for Structure Invariant Template-Free Retrosynthesis
Zhuang, Jiaxi
Zhang, Yu
Zhou, Aimin
Qian, Ying
Machine Learning
Computation and Language
Retrosynthesis prediction is fundamental to drug discovery and chemical synthesis, requiring the identification of reactants that can produce a target molecule. Current template-free methods struggle to capture the structural invariance inherent in chemical reactions, where substantial molecular scaffolds remain unchanged, leading to unnecessarily large search spaces and reduced prediction accuracy. We introduce C-SMILES, a novel molecular representation that decomposes traditional SMILES into element-token pairs with five special tokens, effectively minimizing editing distance between reactants and products. Building upon this representation, we incorporate a copy-augmented mechanism that dynamically determines whether to generate new tokens or preserve unchanged molecular fragments from the product. Our approach integrates SMILES alignment guidance to enhance attention consistency with ground-truth atom mappings, enabling more chemically coherent predictions. Comprehensive evaluation on USPTO-50K and large-scale USPTO-FULL datasets demonstrates significant improvements: 67.2% top-1 accuracy on USPTO-50K and 50.8% on USPTO-FULL, with 99.9% validity in generated molecules. This work establishes a new paradigm for structure-aware molecular generation with direct applications in computational drug discovery.
title Copy-Augmented Representation for Structure Invariant Template-Free Retrosynthesis
topic Machine Learning
Computation and Language
url https://arxiv.org/abs/2510.16588