SpecTUS: Spectral Translator for Unknown Structures Annotation from EI-MS Spectra
Analytical Chemistry · 28.07.2026 · 98 (29) · 21670-21682
DOI: 10.1021/acs.analchem.6c02423 ↗Kurz gefasst
Fachartikel in der Zeitschrift Analytical Chemistry, online erschienen am 13.07.2026. Verfasst von Adam Hájek und 5 weiteren Autor:innen. Beteiligte Einrichtung: Institute of Computer Science, Masaryk University, Šumavská 525/33, Brno602 00, Czech Republic. Die Arbeit nennt 32 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Horizon 2020 Framework Programme, Ministerstvo ?kolstv?, Ml?de?e a Telov?chovy.
Abstract (Originalsprache)
Compound identification and structure annotation from mass spectra are essential in drug detection, forensics, and small molecule discovery. Current approaches to compound identification from electron ionization mass spectra (EI-MS) are dependent on different forms of searching databases that are orders of magnitude smaller than the space of potential molecular structures they attempt to cover. We introduce SpecTUS: Spectral Translator for Unknown Structures, a deep learning model for de novo structural annotation, translating gas chromatography EI-MS spectra directly into molecular structures without requiring reference databases. This enables the identification of novel compounds absent from spectral libraries. In a rigorous evaluation, SpecTUS significantly outperformed standard database search techniques. On a held-out test set of 28,267 spectra from NIST 20, the model’s single suggestion perfectly reconstructed 43% of the subset’s compounds. On 76% of this test set, the single suggestion is strictly better, in terms of Tanimoto similarity of Morgan fingerprint, than the result of hybrid database search. With ten suggestions, SpecTUS achieved 65% perfect reconstructions, surpassing hybrid search on 84% of the test set.
Autor:innen und Institutionen
- Institute of Computer Science, Masaryk University, Šumavská 525/33, Brno602 00, Czech Republic.
- School of Computation, Information and Technology, Technical University Munich, Boltzmannstraße 3, Garching bei München85748, Germany.
- RECETOX, Masaryk University, Kamenice 753/5, Brno625 00, Czech Republic.
- Faculty of Informatics, Masaryk University, Botanická 554/68a, Brno602 00, Czech Republic.
ORCID-Profile: Helge Hecht, Aleš Křenek
Förderung
- Horizon 2020 Framework Programme
- Ministerstvo ?kolstv?, Ml?de?e a Telov?chovy
Zitieren
APA
Hájek, A., Starý, M., Price, E., Jozefov, F., Hecht, H., & Křenek, A. (2026). SpecTUS: Spectral Translator for Unknown Structures Annotation from EI-MS Spectra. Analytical Chemistry, 98(29), 21670-21682. https://doi.org/10.1021/acs.analchem.6c02423
Vancouver
Hájek A, Starý M, Price E, Jozefov F, Hecht H, Křenek A. SpecTUS: Spectral Translator for Unknown Structures Annotation from EI-MS Spectra. Analytical Chemistry. 2026;98(29):21670-21682. doi:10.1021/acs.analchem.6c02423
BibTeX
@article{Hjek2026,
author = {Hájek, Adam and Starý, Michal and Price, Elliott and Jozefov, Filip and Hecht, Helge and Křenek, Aleš},
title = {SpecTUS: Spectral Translator for Unknown Structures Annotation from EI-MS Spectra},
journal = {Analytical Chemistry},
year = {2026},
volume = {98},
number = {29},
pages = {21670--21682},
doi = {10.1021/acs.analchem.6c02423}
}
EUVIMED