EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Pretrained Spectrum Representations for Diverse Tandem Mass Spectrometry Proteomics Tasks

Justin Sanders, Melih Yilmaz, Jacob H Russell, Wout Bittremieux, William E Fondrie, Nicholas M Riley, Sewoong Oh, William S Noble

Molecular & Cellular Proteomics · 2026-11 · 25 (11) · 101673

DOI: 10.1016/j.mcpro.2026.101673 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Molecular & Cellular Proteomics, 2026-11. Verfasst von Justin Sanders und 7 weiteren Autor:innen. Beteiligte Einrichtung: Paul G. Allen School of Computer Science and Engineering, University of Washington, Seattle, Washington, USA. Die Arbeit nennt 50 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Institute of General Medical Sciences, Fonds Wetenschappelijk Onderzoek, National Institutes of Health.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Mass spectrometry is the dominant technology in the field of proteomics, enabling high-throughput analysis of the protein content of complex biological samples. Due to the complexity of the instrumentation and resulting data, sophisticated computational methods are required for the processing and interpretation of acquired mass spectra. Machine learning has shown great promise to improve the analysis of MS data, with numerous purpose-built methods for improving specific steps in the data acquisition and analysis pipeline reaching widespread adoption. Each of these tasks relies on a shared fundamental understanding of the information present in a mass spectrum. Here, we propose a transfer learning framework, CasanovoTL, which utilizes pretrained spectrum representations, derived from a de novo sequencing model trained on a massive proteomics dataset, rather than training a spectrum encoder from scratch for each task. We show that using these pre-trained spectrum representations significantly improves our performance on the four downstream tasks of spectrum quality prediction, chimericity prediction, phosphorylation prediction, and glycosylation status prediction. Finally, we perform end-to-end finetuning, as well as multi-task training, and find that finetuned spectrum representations further improve performance on each individual task. Overall, our work demonstrates that a pretrained spectrum encoder for tandem mass spectrometry proteomics trained on de novo sequencing learns generalizable representations of spectra, improves performance on downstream tasks where training data is limited, and can ultimately enhance data acquisition and analysis in proteomics experiments.

Schlagwörter

Stichwörter der Autor:innen

de novo sequencingdeep learningfoundation modemass spectrometry

Autor:innen und Institutionen

  • Paul G. Allen School of Computer Science and Engineering, University of Washington, Seattle, Washington, USA.
  • Department of Chemistry, University of Washington, Seattle, Washington, USA.
  • Department of Computer Science, University of Antwerp, Antwerp, Belgium.
  • Talus Bioscience, Seattle, Washington, USA.
  • Paul G. Allen School of Computer Science and Engineering, University of Washington, Seattle, Washington, USA; Department of Genome Sciences, University of Washington, Seattle, Washington, USA. Electronic address: willia…

Förderung

  • National Institute of General Medical Sciences
  • Fonds Wetenschappelijk Onderzoek
  • National Institutes of Health
  • Kinship Foundation
  • National Science Foundation

Zitieren

APA

Sanders, J., Yilmaz, M., Russell, J. H., Bittremieux, W., Fondrie, W. E., Riley, N. M., Oh, S., & Noble, W. S. (2026). Pretrained Spectrum Representations for Diverse Tandem Mass Spectrometry Proteomics Tasks. Molecular & Cellular Proteomics, 25(11), 101673. https://doi.org/10.1016/j.mcpro.2026.101673

Vancouver

Sanders J, Yilmaz M, Russell JH, Bittremieux W, Fondrie WE, Riley NM, et al. Pretrained Spectrum Representations for Diverse Tandem Mass Spectrometry Proteomics Tasks. Molecular & Cellular Proteomics. 2026;25(11):101673. doi:10.1016/j.mcpro.2026.101673

BibTeX
@article{Sanders2026,
  author = {Sanders, Justin and Yilmaz, Melih and Russell, Jacob H and Bittremieux, Wout and Fondrie, William E and Riley, Nicholas M and Oh, Sewoong and Noble, William S},
  title = {Pretrained Spectrum Representations for Diverse Tandem Mass Spectrometry Proteomics Tasks},
  journal = {Molecular & Cellular Proteomics},
  year = {2026},
  volume = {25},
  number = {11},
  pages = {101673},
  doi = {10.1016/j.mcpro.2026.101673}
}
Kostenlos starten →