EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

A Comprehensive Evaluation Framework for Benchmarking Multi-Objective Feature Selection in Omics-Based Biomarker Discovery

Luca Cattelani, Arindam Ghosh, Teemu J Rintala, Vittorio Fortino

IEEE/ACM Transactions on Computational Biology and Bioinformatics · 2024

DOI: 10.1109/tcbb.2024.3480150 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE/ACM Transactions on Computational Biology and Bioinformatics, 2024. Verfasst von Luca Cattelani und 3 weiteren Autor:innen. Hauptthemen laut MEDLINE: Biomarkers, Tumor, Computational Biology. Die Arbeit wurde bisher 5-mal zitiert (Crossref) und nennt 50 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Academy of Finland, Jane ja Aatos Erkon Säätiö.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Machine learning algorithms have been extensively used for accurate classification of cancer subtypes driven by gene expression-based biomarkers. However, biomarker models combining multiple gene expression signatures are often not reproducible in external validation datasets and their feature set size is often not optimized, jeopardizing their translatability into cost-effective clinical tools. We investigated how to solve the multi-objective problem of finding the best trade-offs between classification performance and set size applying seven algorithms for machine learning-driven feature subset selection and analyse how they perform in a benchmark with eight large-scale transcriptome datasets of cancer, covering both training and external validation sets. The benchmark includes evaluation metrics assessing the performance of the individual biomarkers and the solution sets, according to their accuracy, diversity, and stability of the composing genes. Moreover, a new evaluation metric for cross-validation studies is proposed that generalizes the hypervolume, which is commonly used to assess the performance of multi-objective optimization algorithms. Biomarkers exhibiting 0.8 of balanced accuracy on the external dataset for breast, kidney and ovarian cancer using respectively 4, 2 and 7 features, were obtained. Genetic algorithms often provided better performance than other considered algorithms, and the recently proposed NSGA2-CH and NSGA2-CHS were the best performing methods in most cases.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Biomarkers, TumorComputational BiologyHumansAlgorithmsGene Expression ProfilingBenchmarkingNeoplasmsTranscriptomeMachine LearningDatabases, Genetic

Substanzen

Biomarkers, Tumor

Förderung

  • Academy of Finland
  • Jane ja Aatos Erkon Säätiö

Zitieren

APA

Cattelani, L., Ghosh, A., Rintala, T. J., & Fortino, V. (2024). A Comprehensive Evaluation Framework for Benchmarking Multi-Objective Feature Selection in Omics-Based Biomarker Discovery. IEEE/ACM Transactions on Computational Biology and Bioinformatics. https://doi.org/10.1109/tcbb.2024.3480150

Vancouver

Cattelani L, Ghosh A, Rintala TJ, Fortino V. A Comprehensive Evaluation Framework for Benchmarking Multi-Objective Feature Selection in Omics-Based Biomarker Discovery. IEEE/ACM Transactions on Computational Biology and Bioinformatics. 2024. doi:10.1109/tcbb.2024.3480150

BibTeX
@article{Cattelani2024,
  author = {Cattelani, Luca and Ghosh, Arindam and Rintala, Teemu J and Fortino, Vittorio},
  title = {A Comprehensive Evaluation Framework for Benchmarking Multi-Objective Feature Selection in Omics-Based Biomarker Discovery},
  journal = {IEEE/ACM Transactions on Computational Biology and Bioinformatics},
  year = {2024},
  doi = {10.1109/tcbb.2024.3480150}
}
Kostenlos starten →