EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

HABiC: an algorithm based on the exact computation of the Kantorovich-Rubinstein optimizer for binary classification in transcriptomics

Chiara Cordier, Pascal Jézéquel iD, Mario Campone iD, Fabien Panloup, Agnes Basseville iD

Bioinformatics · 2025

DOI: 10.1093/bioinformatics/btaf310 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Bioinformatics, 2025. Verfasst von Chiara Cordier und 4 weiteren Autor:innen. Beteiligte Einrichtung: LAREMA, Univ Angers, CNRS, SFR MATHSTIC, Angers F-49000, France. Hauptthemen laut MEDLINE: Algorithms, Transcriptome, Gene Expression Profiling, Software. Die Arbeit nennt 41 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: French national league against cancer.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Motivation Machine learning analyses of molecular omics datasets largely drive the development of precision medicine in oncology, but mathematical challenges still hamper their application in the clinic. In particular, omics-based learning relies on high dimensional data with high degrees of freedom and multicollinearity issues, requiring more tailored algorithms. Here, we have developed a prediction algorithm that relies on the 1-Wasserstein distance to better capture complex relationships between variables, and that is built on a decision rule based on the exact computation of the Kantorovich-Rubinstein optimizer to increase the algorithm precision. We explored dimension reduction and aggregation methods to improve its robustness. The exact method was compared with a neural network-based approximate method, as well as with standard Euclidean distance-based classifiers. Results Experimental results on synthetic datasets with multiple scenarios of redundant/informative variables revealed that exact and approximate methods based on Wasserstein distance outperformed state-of-the-art algorithms when class information was spread across a large number of variables. When predicting clinical or biological outcomes from transcriptomics datasets, HABiC achieved consistently higher accuracy in most situations. Availability and implementation Python code for the HABiC classifier is available at https://github.com/chiaraco/HABiC.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

AlgorithmsTranscriptomeGene Expression ProfilingSoftwareComputational BiologyHumansMachine LearningNeural Networks, Computer

Autor:innen und Institutionen

  • LAREMA, Univ Angers, CNRS, SFR MATHSTIC, Angers F-49000, France.
  • Institut de Cancérologie de l'Ouest, Angers F-49000, France.
  • CRCI2NA, Nantes Université, Univ Angers, INSERM, CNRS, Nantes F-44000, France.
  • SIRIC ILIAD, Angers F-49000, France.

ORCID-Profile: Pascal Jézéquel, Mario Campone, Agnes Basseville

Förderung

  • French national league against cancer

Zitieren

APA

Cordier, C., Jézéquel, P., Campone, M., Panloup, F., & Basseville, A. (2025). HABiC: an algorithm based on the exact computation of the Kantorovich-Rubinstein optimizer for binary classification in transcriptomics. Bioinformatics. https://doi.org/10.1093/bioinformatics/btaf310

Vancouver

Cordier C, Jézéquel P, Campone M, Panloup F, Basseville A. HABiC: an algorithm based on the exact computation of the Kantorovich-Rubinstein optimizer for binary classification in transcriptomics. Bioinformatics. 2025. doi:10.1093/bioinformatics/btaf310

BibTeX
@article{Cordier2025,
  author = {Cordier, Chiara and Jézéquel, Pascal and Campone, Mario and Panloup, Fabien and Basseville, Agnes},
  title = {HABiC: an algorithm based on the exact computation of the Kantorovich-Rubinstein optimizer for binary classification in transcriptomics},
  journal = {Bioinformatics},
  year = {2025},
  doi = {10.1093/bioinformatics/btaf310}
}
Kostenlos starten →