Robust feature selection with minimum redundancy for binary gene expression classification in machine learning

Abstract

High-dimensional gene expression data often contain noisy, irrelevant, and redundant features that can reduce classification accuracy, increase computational burden, and weaken model stability. Therefore, effective feature selection is essential for improving predictive performance and interpretability in biomedical machine learning. This paper proposes a robust feature selection framework with minimum redundancy, called Robust Feature Selection with Minimum Redundancy, for binary classification of gene expression data. The proposed method first constructs a robust class-location estimate for each gene by combining trimmed means, winsorized means, and tri-means. It then computes a normalized robust discriminative score to rank genes by class-separation ability while accounting for within-class variability. To obtain a compact, informative subset of genes, Robust Feature Selection with Minimum Redundancy applies a two-stage redundancy-control strategy: Spearman rank correlation reduces strong monotonic dependencies, while normalized mutual information controls nonlinear dependence among selected genes. We evaluate the proposed method on several publicly available microarray datasets using random forest, support vector machine, and k-nearest-neighbor classifiers. The experimental results show that RFS-MR selects a smaller and less redundant set of informative genes and achieves competitive or improved classification performance compared with existing feature selection methods. These findings suggest that the proposed framework is a robust, interpretable, and redundancy-aware approach for high-dimensional gene expression classification.

Keywords

• Feature selection
• gene expression binary classification
• high-dimensional data
• minimum redundancy
• normalized mutual information
• robust statistics
• Spearman rank correlation

Supporting Institution

This work is supported by United Arab Emirates University, Grant numbers (CARP 2024). Author Samir K Safi has received research support from Statistics and Business Analytics Department, College of Business and Economics.

Thanks

The authors would like to acknowledge the financial support received from the United Arab Emirates University, College of Business and Economics, the CBE Annual Research Program (CARP 2024). We also thank Enago (Crimson Interactive Inc.) for assistance with English language editing and proofreading. Professional editors at Enago reviewed this manuscript to ensure it is clear, grammatically accurate, and faithfully represents the author’s original intent.

References

1. [1] T. R. Golub, D. K. Slonim, P. Tamayo, C. Huard, M. Gaasenbeek, J. P. Mesirov et al., Molecular classification of cancer: Class discovery and class prediction by gene expression monitoring, Science 286 (5439), 531–537, 1999.
2. [2] J. Huang, H. Fang and X. Fan, Decision forest for classification of gene expression data, Comput. Biol. Med. 40 (8), 698–704, 2010.
3. [3] M. Zare, N. Azizizadeh and A. Kazemipour, Supervised feature selection on gene expression microarray datasets using manifold learning, Chemom. Intell. Lab. Syst. 237, 104828, 2023.
4. [4] A. Pino Angulo, Gene selection for microarray cancer data classification by a novel rule-based algorithm, Information 9 (1), 6, 2018.
5. [5] T. Des Touches, M. Munda, T. Cornet, P. Gerkens and T. Hellepute, Feature selection with prior knowledge improves interpretability of chemometrics models, Chemom. Intell. Lab. Syst. 240, 104905, 2023.
6. [6] H. Liu, Y. Yuan, G. Wang, W. Xu, L. Zhao and J. Li, An advanced variable selection method based on information gain and Fisher criterion reselection iteration for multivariate calibration, Chemom. Intell. Lab. Syst. 235, 104796, 2023.
7. [7] J. Zhu, X. Jiang, Q. Wang, J. Wu, S. Wu, X. Chen and Q. Chen, A novel hybrid variable selection strategy with application to molecular spectroscopic analysis, Chemom. Intell. Lab. Syst. 236, 104795, 2023.
8. [8] R. Alanni, J. Hou, H. Azzawi and Y. Xiang, A novel gene selection algorithm for cancer classification using microarray datasets, BMC Med. Genomics 12 (1), 1–12, 2019.

9. [9] M. Khorashadizade and S. Hosseini, An intelligent feature selection method using binary teaching-learning based optimization algorithm and ANN, Chemom. Intell. Lab. Syst. 240, 104880, 2023.
10. [10] G. T. Reddy, M. P. K. Reddy, K. Lakshmanna, R. Kaluri, D. S. Rajput, G. Srivastava and T. Baker, Analysis of dimensionality reduction techniques on big data, IEEE Access 8, 54776–54788, 2020.
11. [11] I. Ul Haq, D. M. Khan, M. Hamraz, N. Iqbal, A. Ali and Z. Khan, Optimal k-nearest neighbors based ensemble for classification and feature selection in chemometrics data, Chemom. Intell. Lab. Syst. 240, 104882, 2023.
12. [12] C. Balestra, C. Maj, E. Müller and A. Mayr, Redundancy-aware unsupervised ranking based on game theory: Ranking pathways in collections of gene sets, PLoS ONE 18 (3), e0282699, 2023.
13. [13] R. A. Carolina, A. P. Antonio, A. P. Diego and V. M. Elías, A hybrid simple filterwrapper feature selection approach for microarray classification, in International Congress of Telematics and Computing, Springer Nature Switzerland, 169–184, 2024.
14. [14] A. A. Joshi and R. M. Aziz, A two-phase cuckoo search based approach for gene selection and deep learning classification of cancer disease using gene expression data with a novel fitness function, Multimed. Tools Appl. 83 (28), 71721–71752, 2024.
15. [15] M. Hamraz, Z. Khan, D. M. Khan, N. Gul, A. Ali and S. Aldahmani, Gene selection in binary classification problems within functional genomics experiments via robust Fisher score, IEEE Access 10, 51682–51692, 2022.
16. [16] A. Esfandiari and N. Nasiri, Gene selection and cancer classification using interactionbased feature clustering and improved-binary Bat algorithm, Comput. Biol. Med. 181, 109071, 2024.
17. [17] S. Azadifar, M. Rostami, K. Berahmand, P. Moradi and M. Oussalah, Graph-based relevancy-redundancy gene selection method for cancer diagnosis, Comput. Biol. Med. 147, 105766, 2022.
18. [18] A. Rana, P. Thai, T. Dinh and P. Ghosh, Relevant and non-redundant feature selection for cancer classification and subtype detection, Cancers 13 (17), 4297, 2021.
19. [19] A. Tsanas, Relevance, redundancy, and complementarity trade-off (RRCT): A principled, generic, robust feature-selection tool, Patterns 3 (5), 2022.
20. [20] A. G. Karegowda, M. A. Jayaram and A. S. Manjunath, Feature subset selection problem using wrapper approach in supervised learning, Int. J. Comput. Appl. 1 (7), 13–17, 2010.
21. [21] A. Suwanwong, A. Harrison and O. Mahmoud, A gene selection method for classification with three classes using proportional overlapping scores, in 18th International Federation of Classification Societies, 2024.
22. [22] M. Hamraz, N. Gul, M. Raza, D. M. Khan, U. Khalil, S. Zubair and Z. Khan, Robust proportional overlapping analysis for feature selection in binary classification within functional genomic experiments, PeerJ Comput. Sci. 7, e562, 2021.
23. [23] Y. K. Saheed, B. F. Balogun, B. J. Odunayo and M. Abdulsalam, Microarray gene expression data classification via Wilcoxon sign rank sum and novel Grey Wolf optimized ensemble learning models, IEEE/ACM Trans. Comput. Biol. Bioinform. 20 (6), 3575–3587, 2023.
24. [24] G. Wang, F. Lauri and A. H. El Hassani, Feature selection by mRMR method for heart disease diagnosis, IEEE Access 10, 100786–100796, 2022.
25. [25] P. N. V. Mary and R. Nagarajan, Feature selection in microarray using proposed hybrid minimum redundancy–maximum relevance (MRMR) and modified genetic algorithm (MGA), Int. J. Exp. Res. Rev. 39, 82–91, 2024.
26. [26] M. Abd-Elnaby, M. Alfonse and M. Roushdy, Classification of breast cancer using microarray gene expression data: A survey, J. Biomed. Inform. 117, 103764, 2021.
27. [27] J. Liu, R. Su, J. Zhang and L. Wei, Classification and gene selection of triple-negative breast cancer subtype embedding gene connectivity matrix in deep neural network, Brief. Bioinform. 22 (5), bbaa395, 2021.
28. [28] Y. Xiong, Q. Li, P.Wang and M. Ye, Informative gene selection based on cost-sensitive fast correlation-based filter feature selection, Curr. Bioinform. 16 (8), 1060–1068, 2021.
29. [29] J. Leach, N. Yi, I. Aban and Alzheimer’s Disease Neuroimaging Initiative, The spikeand- slab lasso and scalable algorithm to accommodate multinomial outcomes in variable selection problems, J. Appl. Stat. 51 (11), 2039–2061, 2024.
30. [30] A. Rauschenberger and E. Glaab, Predicting dichotomised outcomes from highdimensional data in biomedicine, J. Appl. Stat. 51 (9), 1756–1771, 2024.
31. [31] J. Jin, S. Liu and T. Ma, Robust and efficient subsampling algorithms for massive data logistic regression, J. Appl. Stat. 51 (8), 1427–1445, 2024.
32. [32] F. Xu, S. Ma and Q. Zhang, Integrative rank-based regression for multi-source highdimensional data with multi-type responses, J. Appl. Stat., 1–20, 2025.
33. [33] B. Zhang and P. Cao, Classification of high dimensional biomedical data based on feature selection using redundant removal, PLoS ONE 14 (4), e0214406, 2019.
34. [34] Y. Wang, X. Li and R. Ruiz, Feature selection with maximal relevance and minimal supervised redundancy, IEEE Trans. Cybern. 53 (2), 707–717, 2022.
35. [35] W. Cunha, A. Moreo Fernández, A. Esuli, F. Sebastiani, L. Rocha and M. A. Gonçalves, A noise-oriented and redundancy-aware instance selection framework, ACM Trans. Inf. Syst. 43 (2), 1–33, 2025.
36. [36] K. K. Pandey, A. Mishra and G. Jain, Machine learning and statistic-based feature selection and extraction approach for omics data, in Feature Selection and Feature Extraction on Omics Data, Chapman and Hall/CRC, 1–50, 2026.
37. [37] Z. Khan, M. Hamraz, D. M. Khan, N. Gul, A. Ali and S. Aldahmani, Feature selection via robust weighted score for high dimensional binary class-imbalanced gene expression data, IEEE Access 12, 2024.