EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Toward Robust Machine Learning Models for MALDI-TOF MS: Novel Approaches for Mycobacterium abscessus Subspecies Identification

Erica Padial-Fuillerat, Juan E Martínez-Manjón iD, Igor Zwir, Manuel J Arroyo, Mario Blázquez-Sánchez, David Rodríguez-Temporal, Belén Rodríguez, Luis Mancera, Coral Del Val iD

Journal of Proteome Research · 06.03.2026 · 25 (3) · 1305-1318

DOI: 10.1021/acs.jproteome.5c00534 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Proteome Research, online erschienen am 09.02.2026. Verfasst von Erica Padial-Fuillerat und 8 weiteren Autor:innen. Beteiligte Einrichtung: Clover Bioanalytical Software S.L, Granada 18016, Spain. Hauptthemen laut MEDLINE: Spectrometry, Mass, Matrix-Assisted Laser Desorption-Ionization, Mycobacterium abscessus, Machine Learning. Die Arbeit wurde bisher 4-mal zitiert (Crossref) und nennt 54 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Ministerio de Ciencia, Innovación y Universidades, NextGenerationEU, Ministerio de Asuntos Económicos y Transformación Digital, Gobierno de España > Agencia Estatal de Investigación.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Distinguishing Mycobacterium abscessus subspecies presents significant diagnostic challenges due to their genetic homogeneity and variability in analytical platforms. Our research combines matrix-assisted laser desorption/ionization time-of-flight (MALDI-TOF) mass spectrometry with machine learning (ML) approaches to enhance discrimination accuracy, utilizing 325 spectra profiles from diverse European hospitals. The analytical pipeline incorporates specialized techniques for geographical data harmonization, feature selection, and balancing class representation. The best model employs support vector machines (SVMs) with ComBat correction, Boruta feature selection, and centroid clustering for class imbalance, achieving a discrimination performance of 97% F1 score and 97.17% AUC-ROC on test samples. Noteworthily, most tested models improved their discrimination performance with the approach and demonstrated consistent performance metrics with high geometric mean (GEO) and index balanced accuracy (IBA) metrics (>0.90), ensuring consistent sensitivity and specificity across all subspecies. SHAP (SHapley Additive exPlanations) validated the biological relevance of selected spectral features, particularly improving discrimination of the diagnostically challenging M. abscessus subsp. bolletii. This work advances the state-of-the-art in M. abscessus classification, providing a scalable analytical framework for enhanced microbial diagnostics and targeted antimicrobial therapy selection.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Spectrometry, Mass, Matrix-Assisted Laser Desorption-IonizationMycobacterium abscessusMachine LearningSupport Vector MachineHumans

Stichwörter der Autor:innen

AMRMALDI-TOFbatch effectfeature selectionmachine learningmycobacteriumrobust biomarkers

Autor:innen und Institutionen

  • Clover Bioanalytical Software S.L, Granada 18016, Spain.
  • Department of Computer Science and Artificial Intelligence, Andalusian Research Institute in Data Science and Computational Intelligence (DaSCI), University of Granada, Granada 18071, Spain.
  • Instituto de Investigación Biosanitaria ibs.GRANADA, Complejo Hospitales Universitarios de Granada, Universidad de Granada, Granada 18012, Spain.
  • Clinical Microbiology and Infectious Diseases Department, Hospital General Universitario Gregorio Marañón, Madrid 28007, Spain.
  • Institute of Health Research Gregorio Marañón, Madrid 28007, Spain.

ORCID-Profile: Juan E Martínez-Manjón, Coral Del Val

Förderung

  • Ministerio de Ciencia, Innovación y Universidades
  • NextGenerationEU
  • Ministerio de Asuntos Económicos y Transformación Digital, Gobierno de España > Agencia Estatal de Investigación

Zitieren

APA

Padial-Fuillerat, E., Martínez-Manjón, J. E., Zwir, I., Arroyo, M. J., Blázquez-Sánchez, M., Rodríguez-Temporal, D., Rodríguez, B., Mancera, L., & Del Val, C. (2026). Toward Robust Machine Learning Models for MALDI-TOF MS: Novel Approaches for Mycobacterium abscessus Subspecies Identification. Journal of Proteome Research, 25(3), 1305-1318. https://doi.org/10.1021/acs.jproteome.5c00534

Vancouver

Padial-Fuillerat E, Martínez-Manjón JE, Zwir I, Arroyo MJ, Blázquez-Sánchez M, Rodríguez-Temporal D, et al. Toward Robust Machine Learning Models for MALDI-TOF MS: Novel Approaches for Mycobacterium abscessus Subspecies Identification. Journal of Proteome Research. 2026;25(3):1305-1318. doi:10.1021/acs.jproteome.5c00534

BibTeX
@article{PadialFuillerat2026,
  author = {Padial-Fuillerat, Erica and Martínez-Manjón, Juan E and Zwir, Igor and Arroyo, Manuel J and Blázquez-Sánchez, Mario and Rodríguez-Temporal, David and Rodríguez, Belén and Mancera, Luis and Del Val, Coral},
  title = {Toward Robust Machine Learning Models for MALDI-TOF MS: Novel Approaches for Mycobacterium abscessus Subspecies Identification},
  journal = {Journal of Proteome Research},
  year = {2026},
  volume = {25},
  number = {3},
  pages = {1305--1318},
  doi = {10.1021/acs.jproteome.5c00534}
}
Kostenlos starten →