MarkushGlyph and OCSRGlyph: Improved Chemical Structure Recognition

2026-07-30

AI Summary

This study addresses the challenging problem of accurately converting chemical structure images—including both single-molecule and Markush structures—into linear textual representations. It pioneers an end-to-end vision-language approach by formulating Markush structure recognition as an image-to-text translation task, unifying optical chemical structure recognition (OCSR) and Markush parsing under a common framework. The authors propose two models, OCSRGlyph and MarkushGlyph, which jointly model visual features and chemical semantics through a vision–language architecture and incorporate fine-grained handling of stereochemical information. By circumventing conventional multi-stage pipelines, the method introduces novel evaluation metrics that more accurately assess Markush translation quality. Experiments demonstrate that OCSRGlyph achieves state-of-the-art performance on single-molecule recognition, while MarkushGlyph substantially improves parsing accuracy, confirming the effectiveness and robustness of the proposed approach.

Abstract

Chemical structures appear in patents and the scientific literature as images. For programmatic usage, such as indexing in databases or constructing machine learning model training sets, they must be transformed into line notations. The two common forms of this task are translating an image of a single molecule (optical chemical structure recognition - OCSR) and translating a Markush structure that represents a family of molecules. While prior work in the former case is quite mature, Markush structure parsing remains a challenging task. In this work, we treat both tasks as an image-to-text translation problem. We then propose OCSRGlyph, a state-of-the-art OCSR model, improving performance over prior methods by carefully considering stereochemistry. For the Markush task, we introduce MarkushGlyph, a vision-language model that reads the entire Markush structure as an image. This contrasts with prior systems, which often use multiple stages to separately process visual and text input content. Finally, we introduce a new metric for determining the accuracy of Markush structure translations, handling failure modes present in prior metrics.

Problem

• Markush structure
• optical chemical structure recognition
• image-to-text translation
• chemical structure recognition
• line notation

Innovation

• OCSRGlyph
• MarkushGlyph
• image-to-text translation
• stereochemistry
• vision-language model