A Novel Hybrid Runge Kutta Optimizer with Support Vector Machine on Gene Expression Data for Cancer Classification.

Area of Science:

• Bioinformatics
• Computational Biology
• Genomics

Background:

• Accurate cancer categorization from microarray data is essential for effective diagnosis and treatment.
• Gene expression analysis is a key tool in cancer research, but high dimensionality poses challenges.
• Identifying informative genes is critical for improving cancer classification accuracy.

Purpose of the Study:

• To develop and evaluate a novel computational approach for selecting significant genes for cancer classification using microarray data.
• To enhance the accuracy of cancer detection by effectively handling high-dimensional gene expression datasets.
• To assess the performance of the proposed method against existing algorithms on various cancer datasets.

Main Methods:

• A hybrid approach combining the Runge Kutta optimizer (RUN) for gene selection and Support Vector Machine (SVM) for classification.
• Implementation of the ReliefF method for preprocessing high-dimensional microarray data to reduce noise and redundancy.
• Testing the RUN-SVM approach on diverse binary-class (Breast2, Prostate) and multi-class (Brain Tumor1, Brain Tumor2, Breast3, Lung Cancer) cancer gene expression datasets.

Main Results:

• The proposed RUN-SVM approach demonstrated superior performance in identifying significant genes for cancer classification.
• Statistical analysis confirmed that RUN-SVM significantly outperformed competing algorithms across six different cancer gene expression datasets.
• The method effectively addressed the challenge of high dimensionality inherent in microarray data.

Conclusions:

• The novel RUN-SVM approach offers a robust and effective solution for gene selection in cancer classification.
• This method provides a statistically significant improvement over existing algorithms for cancer detection using gene expression data.
• The innovative search technique employed by RUN-SVM contributes to its enhanced performance in analyzing complex biological datasets.