STOUT V2.0: SMILES to IUPAC name conversion using transformer models – Cheminformatics and Computational Metabolomics

Our article, “STOUT V2.0: SMILES to IUPAC name conversion using transformer models” has been published in the Journal of Cheminformatics. In this work, we introduce an updated model, STOUT (SMILES-TO-IUPAC-name Translator) V2, which tackles the challenge of accurately translating chemical compounds from SMILES notation into their corresponding IUPAC names. Using a transformer-based architecture, STOUT V2 translates string representations of chemical structures into IUPAC names. Trained on a dataset of nearly 1 billion SMILES strings and their corresponding IUPAC names, STOUT V2 achieves remarkable accuracy, even when dealing with complex chemical structures. While established deterministic algorithms remain the gold standard for systematic chemical naming, our work, enabled by access to OpenEye’s Lexichem software through an academic license, demonstrates the potential of neural approaches for machine translation in chemical nomenclature.

Code availability: https://github.com/Kohulan/Smiles-TO-iUpac-Translator
Web app: https://stout.decimer.ai
Web app code: https://github.com/Kohulan/STOUT_WebApp

Rajan, K., Zielesny, A. & Steinbeck, C. STOUT V2.0: SMILES to IUPAC name conversion using transformer models. J Cheminform 16, 146 (2024). https://doi.org/10.1186/s13321-024-00941-x