ROOFS: RObust biOmarker Feature Selection
Abstract
Background: Feature selection (FS) is essential for biomarker discovery and clinical predictive modeling. Over the past decades, the methodological literature on FS has become rich and mature, offering a wide spectrum of algorithmic approaches. However, much of this methodological progress has not been fully adopted in applied biomedical research. Results: To bridge this gap, we propose ROOFS (RObust biOmarker Feature Selection), a Python package designed to help researchers in the choice of the FS method most adapted to their problem. Currently supporting classification tasks, ROOFS benchmarks multiple FS methods on the user's data and generates reports summarizing a comprehensive set of evaluation metrics, including downstream predictive performance estimated using optimism correction, stability, robustness of individual features, and true positive and false positive rates assessed on semi-synthetic data with a simulated outcome. We demonstrate the utility of ROOFS on data from the PIONeeR biomarker study, aimed at identifying predictors of resistance to anti-PD-(L)1 immunotherapy in lung cancer. Conclusion: Comprehensive benchmarking with ROOFS has the potential to improve the reproducibility of FS discoveries and increase the translational value of clinical models. Availability and implementation: ROOFS is available at https://gitlab.inria.fr/compo/roofs.
Related
- A multimodal deep learning for rice crop health monitoring using vegetation index feature selection
- ssGSEA3.0: Feature-Level Bootstrap Resampling for Robust Assessment of Statistical Significance
- Phylogeny-Guided Global-Local Shrinkage for Microbiome Feature Selection
- Photoplethysmography-based glucose estimation with feature selection
- An integrated instance segmentation and deep feature fusion model for robust leaf species classification in natural environments
BibTeX
@article{bakhmach2026roofs,
title = {{ROOFS: RObust biOmarker Feature Selection}},
author = {Anastasiia Bakhmach and Paul Dufossé and Simon Charpigny and Florence Monville and Laurent Greillier and Fabrice Barlési and Sébastien Benzekry},
journal = {BMC Bioinformatics},
year = {2026},
month = oct,
doi = {10.1186/s12859-026-06667-x},
url = {https://inria.hal.science/hal-05241230v2},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - ROOFS: RObust biOmarker Feature Selection
AU - Anastasiia Bakhmach
AU - Paul Dufossé
AU - Simon Charpigny
AU - Florence Monville
AU - Laurent Greillier
AU - Fabrice Barlési
AU - Sébastien Benzekry
PY - 2026
DA - 2026/10/15
JO - BMC Bioinformatics
DO - 10.1186/s12859-026-06667-x
UR - https://inria.hal.science/hal-05241230v2
AB - Background: Feature selection (FS) is essential for biomarker discovery and clinical predictive modeling. Over the past decades, the methodological literature on FS has become rich and mature, offering a wide spectrum of algorithmic approaches. However, much of this methodological progress has not been fully adopted in applied biomedical research. Results: To bridge this gap, we propose ROOFS (RObust biOmarker Feature Selection), a Python package designed to help researchers in the choice of the FS method most adapted to their problem. Currently supporting classification tasks, ROOFS benchmarks multiple FS methods on the user's data and generates reports summarizing a comprehensive set of evaluation metrics, including downstream predictive performance estimated using optimism correction, stability, robustness of individual features, and true positive and false positive rates assessed on semi-synthetic data with a simulated outcome. We demonstrate the utility of ROOFS on data from the PIONeeR biomarker study, aimed at identifying predictors of resistance to anti-PD-(L)1 immunotherapy in lung cancer. Conclusion: Comprehensive benchmarking with ROOFS has the potential to improve the reproducibility of FS discoveries and increase the translational value of clinical models. Availability and implementation: ROOFS is available at https://gitlab.inria.fr/compo/roofs.
SN - 1471-2105
ER -