EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC 4.0

ChemEmbed: a deep learning framework for metabolite identification using enhanced MS/MS data and multidimensional molecular embeddings

Muhammad Faizan-Khan, Roger Giné, Josep M Badia, Maribel Pérez-Ribera, Manuel Ruiz-Botella, Alexandra Junza, Jordi Capellades, Iván Pérez-López, Shipei Xing, Abubaker Patan, Laura Brugnara, Anna Novials und 6 weitere

Briefings in Bioinformatics · 07.01.2026 · 27 (1)

DOI: 10.1093/bib/bbag054 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Briefings in Bioinformatics, online erschienen am 13.02.2026. Verfasst von Muhammad Faizan-Khan und 17 weiteren Autor:innen. Beteiligte Einrichtung: Department of Electronic Engineering, Universitat Rovira i Virgili, Tarragona 43007, Spain. Hauptthemen laut MEDLINE: Metabolomics, Deep Learning, Tandem Mass Spectrometry, Software. Die Arbeit wurde bisher 9-mal zitiert (Crossref) und nennt 44 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0). Gefördert von: MCIN/AEI, FEDER, UE.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Machine learning offers a promising path to annotating the large number of unidentified MS/MS spectra in metabolomics, addressing the limited coverage of current reference spectral libraries. However, existing methods often struggle with the high dimensionality and sparsity of MS/MS spectra and metabolite structures. ChemEmbed tackles these challenges by integrating multidimensional, continuous vector representations of chemical structures with enhanced MS/MS spectra. This enhancement is achieved by merging spectra across multiple collision energies and incorporating calculated neutral losses from 38 472 distinct compounds, providing richer input for a convolutional neural network (CNN). ChemEmbed ranks the correct candidate first in over 42% of cases and within the top five in more than 76% of cases. In external benchmarks such as CASMI 2016 and 2022, ChemEmbed outperforms SIRIUS 6, the current state-of-the-art in computational metabolomics. We applied ChemEmbed to predict structures in the Annotated Recurrent Unidentified Spectra (ARUS) dataset and confirmed 25 previously unidentified compounds. These findings demonstrate ChemEmbed’s potential as a robust, scalable tool for accelerating metabolite identification in untargeted mass spectrometry workflows.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

MetabolomicsDeep LearningTandem Mass SpectrometrySoftwareConvolutional Neural NetworksHumans

Stichwörter der Autor:innen

deep learningmass spectrometrymetabolite identificationmolecular embeddingsuntargeted metabolomics

Autor:innen und Institutionen

  • Department of Electronic Engineering, Universitat Rovira i Virgili, Tarragona 43007, Spain.
  • CIBER de Diabetes y Enfermedades Metabólicas Asociadas (CIBERDEM), Instituto de Salud Carlos III, Madrid 28029, Spain.
  • Department of Chemical Engineering, Universitat Rovira i Virgili, Tarragona 43007, Spain.
  • Metabolomics Platform, Institut de Recerca Biomèdica Catalunya Sud, Hospital Universitari Sant Joan de Reus, Reus 43204, Spain.
  • Skaggs School of Pharmacy and Pharmaceutical Sciences, University of California, San Diego, La Jolla, CA 92093, United States.
  • Institut d'Investigacions Biomèdiques August Pi i Sunyer (IDIBAPS), Barcelona 08036, Spain.

ORCID-Profile: Marta Sales-Pardo, Roger Guimerà, Oscar Yanes

Förderung

  • MCIN/AEI
  • FEDER
  • UE
  • Government of Catalonia
  • European Union NextGenerationEU/PRTR
  • Instituto de Salud Carlos III
  • European Union
  • European Union's Horizon 2020 research and innovation programme

Zitieren

APA

Faizan-Khan, M., Giné, R., Badia, J. M., Pérez-Ribera, M., Ruiz-Botella, M., Junza, A., Capellades, J., Pérez-López, I., Xing, S., Patan, A., Brugnara, L., Novials, A., Servitja, J. M., Vinaixa, M., Dorrestein, P. C., Sales-Pardo, M., Guimerà, R., & Yanes, O. (2026). ChemEmbed: a deep learning framework for metabolite identification using enhanced MS/MS data and multidimensional molecular embeddings. Briefings in Bioinformatics, 27(1). https://doi.org/10.1093/bib/bbag054

Vancouver

Faizan-Khan M, Giné R, Badia JM, Pérez-Ribera M, Ruiz-Botella M, Junza A, et al. ChemEmbed: a deep learning framework for metabolite identification using enhanced MS/MS data and multidimensional molecular embeddings. Briefings in Bioinformatics. 2026;27(1). doi:10.1093/bib/bbag054

BibTeX
@article{FaizanKhan2026,
  author = {Faizan-Khan, Muhammad and Giné, Roger and Badia, Josep M and Pérez-Ribera, Maribel and Ruiz-Botella, Manuel and Junza, Alexandra and Capellades, Jordi and Pérez-López, Iván and Xing, Shipei and Patan, Abubaker and Brugnara, Laura and Novials, Anna and Servitja, Joan-Marc and Vinaixa, Maria and Dorrestein, Pieter C and Sales-Pardo, Marta and Guimerà, Roger and Yanes, Oscar},
  title = {ChemEmbed: a deep learning framework for metabolite identification using enhanced MS/MS data and multidimensional molecular embeddings},
  journal = {Briefings in Bioinformatics},
  year = {2026},
  volume = {27},
  number = {1},
  doi = {10.1093/bib/bbag054}
}
Kostenlos starten →