Evolutionary Mechanism Based Conserved Gene Expression Biclustering Module Analysis for Breast Cancer Genomics

Area of Science:

• Bioinformatics and Computational Biology
• Genomics and Transcriptomics
• Cancer Research

Background:

• RNA sequencing generates vast gene expression data, necessitating methods to identify significant gene biclusters and functionally related genes.
• Existing algorithms for gene expression module discovery face challenges in efficiency and accuracy when analyzing large datasets.

Purpose of the Study:

• To propose a novel algorithm, Conserved Gene Expression Module based on Genetic Algorithm (CGEMGA), for identifying significant gene biclusters.
• To evaluate the performance of CGEMGA against existing methods using breast cancer data from The Cancer Genome Atlas (TCGA) database.
• To assess the biological relevance of identified gene modules through enrichment analysis of driver genes and cancer-related pathways.

Main Methods:

• Development and implementation of the Conserved Gene Expression Module based on Genetic Algorithm (CGEMGA).
• Application of CGEMGA to breast cancer RNA-sequencing data from the TCGA database.
• Statistical evaluation using Fisher's exact test (p-values) and F-test to compare CGEMGA with other algorithms (e.g., Cheng and Church, CGEM).
• Analysis of computational cost by measuring algorithm running times.
• Enrichment analysis to identify driver genes and cancer-related pathways.

Main Results:

• CGEMGA achieved a superior average p-value of 1.54 × 10^-4 ± 3.06 × 10^-5 across 10 independent runs, indicating higher significance.
• The F-test showed a significant difference between CGEMGA and the CGEM algorithm (p-value = 0.039).
• CGEMGA demonstrated significantly shorter average runtime (5.22 × 10⁰ ± 1.65 × 10^-1 s), highlighting its computational efficiency.
• Enrichment analysis confirmed that genes identified by CGEMGA are significantly enriched for driver genes.

Conclusions:

• CGEMGA is a fast, robust, and efficient algorithm for extracting co-expressed genes and associated biclusters from RNA-seq data.
• The algorithm demonstrates superior performance in identifying significant gene modules compared to existing methods.
• CGEMGA provides a valuable tool for cancer research by effectively identifying biologically relevant gene sets.