Wrapshap a high dimension feature selection framework to explore biological systems through machine learning explanations
Abstract
Among the existing feature selection (FS) methods in machine learning (ML), those known as wrapper methods often produce the most effective subset of features. However, their high computational cost and tendency to overfit make them impractical for high-dimensional data, as is frequently encountered in biological research. The Wrapshap framework presents a novel FS method tailored for high-dimensional biological data, addressing challenges such as redundancy, multicollinearity, and interaction effects. By leveraging the local accuracy property of Additive Local Explanation methods of SHAP, Wrapshap translates complex biological data into interpretable insights. The framework, integrating a trained ML model, allows explanations to be viewed as a new exploratory data space containing a comprehensive understanding of the underlying complexity of the raw data structure. Benchmarks against 84 regression and 106 classification datasets demonstrate its superiority in computational speed, feature ranking quality, and predictive performance compared to state-of-the-art methods like Recursive Feature Elimination. The method's adaptability extends to various ML models, explanation methods and tasks. By requiring only a single training of the complex model, Wrapshap mitigates the computational inefficiencies of traditional wrapper methods while maintaining performance monitoring thus opening new possibilities for real-time, interactive FS with user involvement. This innovative approach addresses a notable gap in existing methods, and promotes informed biological hypothesis generation, enhancing the interpretability and usability of ML in biomedical research.
Related
- Explainable AI for Network Pharmacology: A Critical Framework for Interpreting Multi-Target Drug Discovery on Biological Graphs
- Extraction, Structure–Activity Relationships, and Modification of Plant-Derived Polysaccharides: An Integrated Framework for Evidence-Based Method Selection and Bioactivity Optimization—A Review
- Quantifying polymorphism and divergence from epigenetic data: a framework for inferring the action of selection
- Novel post-transcriptional dimension of glucocorticoid action through mRNA translation and P-body remodeling
- Biological Accessibility and Analytical Recovery in Myofibrillar Protein Loss During Skeletal Muscle Atrophy
BibTeX
@article{furger2026wrapshap,
title = {{Wrapshap a high dimension feature selection framework to explore biological systems through machine learning explanations}},
author = {Félix Furger and Miguel Thomas and Colas Foulon and Haomio Wang and Julien Aligon and Emmanuel Doumard and Chantal Soulé-Dupuy and Cyrille Delpierre and Louis Casteilla and Paul Monsarrat},
journal = {npj Systems Biology and Applications},
year = {2026},
month = jul,
doi = {10.1038/s41540-026-00779-z},
url = {https://hal.science/hal-05790254v1},
note = {CC BY-NC-ND 4.0},
}
RIS
TY - JOUR
TI - Wrapshap a high dimension feature selection framework to explore biological systems through machine learning explanations
AU - Félix Furger
AU - Miguel Thomas
AU - Colas Foulon
AU - Haomio Wang
AU - Julien Aligon
AU - Emmanuel Doumard
AU - Chantal Soulé-Dupuy
AU - Cyrille Delpierre
AU - Louis Casteilla
AU - Paul Monsarrat
PY - 2026
DA - 2026/07/21
JO - npj Systems Biology and Applications
DO - 10.1038/s41540-026-00779-z
UR - https://hal.science/hal-05790254v1
AB - Among the existing feature selection (FS) methods in machine learning (ML), those known as wrapper methods often produce the most effective subset of features. However, their high computational cost and tendency to overfit make them impractical for high-dimensional data, as is frequently encountered in biological research. The Wrapshap framework presents a novel FS method tailored for high-dimensional biological data, addressing challenges such as redundancy, multicollinearity, and interaction effects. By leveraging the local accuracy property of Additive Local Explanation methods of SHAP, Wrapshap translates complex biological data into interpretable insights. The framework, integrating a trained ML model, allows explanations to be viewed as a new exploratory data space containing a comprehensive understanding of the underlying complexity of the raw data structure. Benchmarks against 84 regression and 106 classification datasets demonstrate its superiority in computational speed, feature ranking quality, and predictive performance compared to state-of-the-art methods like Recursive Feature Elimination. The method's adaptability extends to various ML models, explanation methods and tasks. By requiring only a single training of the complex model, Wrapshap mitigates the computational inefficiencies of traditional wrapper methods while maintaining performance monitoring thus opening new possibilities for real-time, interactive FS with user involvement. This innovative approach addresses a notable gap in existing methods, and promotes informed biological hypothesis generation, enhancing the interpretability and usability of ML in biomedical research.
SN - 2056-7189
ER -