Academic Bibliography

 Search 200 years of publications by Ghent University researchers.
Advanced search
1 file | 2.23 MB
Add to list
  1. Search results
  2. Pre-trained MALDI transformers improv...

Pre-trained MALDI transformers improve MALDI-TOF MS-based prediction

Gaetan De Waele (UGent) , Gerben Menschaert (UGent) , Peter Vandamme (UGent) and Willem Waegeman (UGent)
(2025) COMPUTERS IN BIOLOGY AND MEDICINE. 186.
Author
Organization
Project
Abstract
For the last decade, matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS) has been the reference method for species identification in clinical microbiology. Hampered by a historical lack of open data, machine learning research towards models specifically adapted to MALDI-TOF MS remains in its infancy. Given the growing complexity of available datasets (such as large-scale antimicrobial resistance prediction), a need for models that (1) are specifically designed for MALDI-TOF MS data, and (2) have high representational capacity, presents itself. Here, we introduce Maldi Transformer, an adaptation of the state-of-the-art transformer architecture to the MALDI-TOF mass spectral domain. We propose the first self-supervised pre-training technique specifically designed for mass spectra. The technique is based on shuffling peaks across spectra, and pre-training the transformer as a peak discriminator. Extensive benchmarks confirm the efficacy of this novel design. The final result is a model exhibiting state-of-the-art (or competitive) performance on downstream prediction tasks. In addition, we show that Maldi Transformer’s identification of noisy spectra may be leveraged towards higher predictive performance. All code supporting this study is distributed on PyPI and is packaged under: https://github.com/gdewael/maldi-nn.

Downloads

  • Download
    published version.pdf
    • full text (Published version)
    • |
    • open access
    • |
    • PDF
    • |
    • 2.23 MB

Citation

Please use this url to cite or link to this publication:

MLA
De Waele, Gaetan, et al. “Pre-Trained MALDI Transformers Improve MALDI-TOF MS-Based Prediction.” COMPUTERS IN BIOLOGY AND MEDICINE, vol. 186, 2025, doi:10.1016/j.compbiomed.2025.109695.
APA
De Waele, G., Menschaert, G., Vandamme, P., & Waegeman, W. (2025). Pre-trained MALDI transformers improve MALDI-TOF MS-based prediction. COMPUTERS IN BIOLOGY AND MEDICINE, 186. https://doi.org/10.1016/j.compbiomed.2025.109695
Chicago author-date
De Waele, Gaetan, Gerben Menschaert, Peter Vandamme, and Willem Waegeman. 2025. “Pre-Trained MALDI Transformers Improve MALDI-TOF MS-Based Prediction.” COMPUTERS IN BIOLOGY AND MEDICINE 186. https://doi.org/10.1016/j.compbiomed.2025.109695.
Chicago author-date (all authors)
De Waele, Gaetan, Gerben Menschaert, Peter Vandamme, and Willem Waegeman. 2025. “Pre-Trained MALDI Transformers Improve MALDI-TOF MS-Based Prediction.” COMPUTERS IN BIOLOGY AND MEDICINE 186. doi:10.1016/j.compbiomed.2025.109695.
Vancouver
1.
De Waele G, Menschaert G, Vandamme P, Waegeman W. Pre-trained MALDI transformers improve MALDI-TOF MS-based prediction. COMPUTERS IN BIOLOGY AND MEDICINE. 2025;186.
IEEE
[1]
G. De Waele, G. Menschaert, P. Vandamme, and W. Waegeman, “Pre-trained MALDI transformers improve MALDI-TOF MS-based prediction,” COMPUTERS IN BIOLOGY AND MEDICINE, vol. 186, 2025.
@article{01JRA2MDJDMBMSFS9W74EQE83D,
  abstract     = {{For the last decade, matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS) has been the reference method for species identification in clinical microbiology. Hampered by a historical lack of open data, machine learning research towards models specifically adapted to MALDI-TOF MS remains in its infancy. Given the growing complexity of available datasets (such as large-scale antimicrobial resistance prediction), a need for models that (1) are specifically designed for MALDI-TOF MS data, and (2) have high representational capacity, presents itself.
Here, we introduce Maldi Transformer, an adaptation of the state-of-the-art transformer architecture to the MALDI-TOF mass spectral domain. We propose the first self-supervised pre-training technique specifically designed for mass spectra. The technique is based on shuffling peaks across spectra, and pre-training the transformer as a peak discriminator. Extensive benchmarks confirm the efficacy of this novel design. The final result is a model exhibiting state-of-the-art (or competitive) performance on downstream prediction tasks. In addition, we show that Maldi Transformer’s identification of noisy spectra may be leveraged towards higher predictive performance.
All code supporting this study is distributed on PyPI and is packaged under: https://github.com/gdewael/maldi-nn.}},
  articleno    = {{109695}},
  author       = {{De Waele, Gaetan and Menschaert, Gerben and Vandamme, Peter and Waegeman, Willem}},
  issn         = {{0010-4825}},
  journal      = {{COMPUTERS IN BIOLOGY AND MEDICINE}},
  language     = {{eng}},
  pages        = {{12}},
  title        = {{Pre-trained MALDI transformers improve MALDI-TOF MS-based prediction}},
  url          = {{http://doi.org/10.1016/j.compbiomed.2025.109695}},
  volume       = {{186}},
  year         = {{2025}},
}
Altmetric
View in Altmetric