A Hybrid Ensemble Equilibrium Optimizer Gene Selection Algorithm for Microarray Data.

Area of Science:

• Genomics
• Bioinformatics
• Computational Biology

Background:

• Gene expression data is crucial for cancer diagnosis and prognosis.
• High-dimensional gene expression data presents challenges like the curse of dimensionality and overfitting.
• Effective gene selection is vital for accurate analysis and reliable outcomes.

Purpose of the Study:

• To introduce a novel hybrid ensemble equilibrium optimizer gene selection algorithm.
• To address challenges posed by high dimensionality, redundancy, and noise in gene expression data.
• To improve the accuracy and efficiency of gene selection for cancer-related applications.

Main Methods:

• A two-stage hybrid approach combining ensemble filtering and an improved equilibrium optimizer.
• Stage 1: Hybrid filter and gene correlation methods for initial gene subset selection.
• Stage 2: Equilibrium optimizer with Gaussian Barebone and gene pruning for optimal subset identification.

Main Results:

• The ensemble filtering method demonstrated stability and effectively reduced the search space.
• The improved equilibrium optimizer enhanced prediction accuracy.
• The proposed algorithm significantly reduced the number of selected features compared to existing methods.
• Validation on 15 datasets showed superiority over nine other feature selection techniques.

Conclusions:

• The proposed hybrid ensemble equilibrium optimizer is an effective gene selection strategy.
• The method successfully mitigates issues associated with high-dimensional gene expression data.
• This approach offers a valuable tool for cancer diagnosis and prognosis monitoring through optimized gene selection.