EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Exploring the role of normalization and feature selection in microbiome disease classification pipelines

Ignacio Garach Vélez iD, Francisco Manuel Ortuño Guzmán iD, Ignacio Rojas Ruiz iD, Luis Javier Herrera Maldonado iD

GigaScience · 2025

DOI: 10.1093/gigascience/giaf096 ↗

Kurz gefasst

Fachartikel in der Zeitschrift GigaScience, 2025. Verfasst von Ignacio Garach Vélez und 3 weiteren Autor:innen. Beteiligte Einrichtung: Department of Computer Engineering, Automation and Robotics (ICAR), University of Granada, 18071 Granada, Spain. Hauptthemen laut MEDLINE: Microbiota, Computational Biology. Die Arbeit wurde bisher 11-mal zitiert (Crossref) und nennt 76 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Ministerio de Ciencia e Innovación, Consejería de Conocimiento, Investigación y Universidad, Junta de Andalucía, European Union.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Background

Disease classification using 16S rRNA microbiome data faces challenges of high dimensionality, compositionality, and sparsity, compounded by the inherent small sample sizes in many studies. Machine learning and feature selection techniques offer potential to identify robust biomarkers and improve classification performance; however, their comparative effectiveness across diverse methods and datasets has been insufficiently explored. This study evaluates multiple feature selection techniques alongside normalization strategies, focusing on their interplay with classifier performance.

Results

Our analyses revealed that centered log-ratio normalization improves the performance of logistic regression and support vector machine models and facilitates feature selection, whereas random forest models yield strong results using relative abundances. Interestingly, presence–absence normalization was able to achieve similar performance compared to abundance-based transformations across classifiers. Among feature selection methods, minimum redundancy maximum relevancy (mRMR) surpassed most methods in identifying compact feature sets and demonstrated performance comparable to least absolute shrinkage and selection operator (LASSO), which obtained top results requiring lower computation times. Autoencoders needed larger latent spaces to perform well and lacked interpretability, Mutual Information suffered from redundancy, and ReliefF struggled with data sparsity.

Conclusions

Overall, feature selection pipelines improved model focus and robustness via a massive reduction of the feature space. mRMR and LASSO emerged as the most effective methods across datasets.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

MicrobiotaComputational BiologyHumansRNA, Ribosomal, 16SSupport Vector MachineMachine Learning

Stichwörter der Autor:innen

classificationfeature selectionmachine learningmicrobiomenormalization

Substanzen

RNA, Ribosomal, 16S

Autor:innen und Institutionen

  • Department of Computer Engineering, Automation and Robotics (ICAR), University of Granada, 18071 Granada, Spain.

ORCID-Profile: Ignacio Garach Vélez, Francisco Manuel Ortuño Guzmán, Ignacio Rojas Ruiz, Luis Javier Herrera Maldonado

Förderung

  • Ministerio de Ciencia e Innovación
  • Consejería de Conocimiento, Investigación y Universidad, Junta de Andalucía
  • European Union
  • Universidad de Granada

Zitieren

APA

Garach Vélez, I., Ortuño Guzmán, F. M., Rojas Ruiz, I., & Herrera Maldonado, L. J. (2025). Exploring the role of normalization and feature selection in microbiome disease classification pipelines. GigaScience. https://doi.org/10.1093/gigascience/giaf096

Vancouver

Garach Vélez I, Ortuño Guzmán FM, Rojas Ruiz I, Herrera Maldonado LJ. Exploring the role of normalization and feature selection in microbiome disease classification pipelines. GigaScience. 2025. doi:10.1093/gigascience/giaf096

BibTeX
@article{GarachVlez2025,
  author = {Garach Vélez, Ignacio and Ortuño Guzmán, Francisco Manuel and Rojas Ruiz, Ignacio and Herrera Maldonado, Luis Javier},
  title = {Exploring the role of normalization and feature selection in microbiome disease classification pipelines},
  journal = {GigaScience},
  year = {2025},
  doi = {10.1093/gigascience/giaf096}
}
Kostenlos starten →