Breast Cancer Biomarker Discovery Using an Enhanced Quantum-Based Avian Navigation Optimizer and Ensemble Learning...

Area of Science:

• Bioinformatics
• Computational Biology
• Machine Learning

Background:

• Breast cancer early detection is critical but challenged by high false positives and complex gene datasets.
• Traditional differential evolution methods lack scalability for high-dimensional gene expression analysis.

Purpose of the Study:

• To introduce Ensemble-Based Logical Binary QANA (LBQANA_En), an enhanced differential evolution algorithm for breast cancer biomarker detection.
• To overcome limitations of small sample sizes and complex gene expression data through ensemble integration.

Main Methods:

• Developed LBQANA_En, inspired by quantum navigation and utilizing logical operators (XOR, OR).
• Integrated six gene expression datasets to enhance robustness and handle data complexity.
• Applied LBQANA_En to identify key breast cancer biomarkers.

Main Results:

• LBQANA_En demonstrated superior performance over other binary QANA variants in biomarker detection.
• Identified key biomarkers: LPL, LEP, CD36, CDC20, TOP2A, and EZH2.
• Achieved a high F1 score of 98.958%, significantly improving breast cancer detection accuracy and reducing false positives.

Conclusions:

• LBQANA_En offers a powerful new tool for large-scale global optimization in gene expression analysis.
• The identified biomarkers provide insights into critical pathways like AMPK and PPAR signaling.
• This research sets a new benchmark in computational biology, advancing diagnostic techniques for breast cancer.