Transformer-Based Approach for Automated Functional Group Replacement in Chemical Compounds

2026-01-12

AI Summary

Traditional rule-based functional group replacement methods struggle to simultaneously achieve precise structural modifications and generate molecular diversity. To address this limitation, this work proposes a two-stage Transformer model that learns chemical transformation rules from matched molecular pairs (MMPs) in the ChEMBL dataset using SMIRKS notation. In the first stage, the model predicts the substructure to be removed; in the second stage, it generates a corresponding replacement moiety. By moving beyond single-step generation paradigms, this approach enables accurate, controllable, substructure-level editing of molecules. The method maintains high chemical validity while significantly enhancing both structural diversity and model scalability in de novo molecular design.

Technology Category

• Machine Learning: Structured Learning
• Cognitive Modeling & Cognitive Systems: Computational Creativity
• Natural Language Processing: Generation

Application Category

• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Semantics and Knowledge: Scalable techniques for the creation, curation, publication, maintenance, and consumption of large, Web-based, structured, reusable, knowledge graphs and ontologies

Abstract

Functional group replacement is a pivotal approach in cheminformatics to enable the design of novel chemical compounds with tailored properties. Traditional methods for functional group removal and replacement often rely on rule-based heuristics, which can be limited in their ability to generate diverse and novel chemical structures. Recently, transformer-based models have shown promise in improving the accuracy and efficiency of molecular transformations, but existing approaches typically focus on single-step modeling, lacking the guarantee of structural similarity. In this work, we seek to advance the state of the art by developing a novel two-stage transformer model for functional group removal and replacement. Unlike one-shot approaches that generate entire molecules in a single pass, our method generates the functional group to be removed and appended sequentially, ensuring strict substructure-level modifications. Using a matched molecular pairs (MMPs) dataset derived from ChEMBL, we trained an encoder-decoder transformer model with SMIRKS-based representations to capture transformation rules effectively. Extensive evaluations demonstrate our method's ability to generate chemically valid transformations, explore diverse chemical spaces, and maintain scalability across varying search sizes.

Problem

• functional group replacement
• structural similarity
• molecular transformation
• cheminformatics
• diverse chemical structures

Innovation

• transformer
• functional group replacement
• matched molecular pairs
• SMIRKS
• two-stage modeling