An Optimal Mean Based Block Robust Feature Extraction Method to Identify Colorectal Cancer Genes with Integrated Data

Area of Science:

• Genomics
• Bioinformatics
• Oncology

Background:

• Early diagnosis of colorectal cancer is critical for effective treatment.
• While some colorectal cancer genes are known, specific markers for advanced stages, especially in early diagnoses, remain under-explored.
• Understanding these genes is vital for improving patient outcomes.

Purpose of the Study:

• To develop and validate a novel feature extraction method for identifying genes associated with advanced colorectal cancer clinical stages.
• To analyze integrated genomic data for robust gene identification in early-stage colorectal cancer.

Main Methods:

• Proposed the Optimal Mean based Robust Feature Extraction method (OMRFE) using optimal mean and L2,1-norm.
• Developed the Optimal Mean based Block Robust Feature Extraction method (OMRFE) by incorporating block ideology.
• Applied OMBRFE to integrated colorectal cancer data, including copy number alterations, somatic mutations, methylation, and gene expression.

Main Results:

• The OMBRFE method demonstrated superior effectiveness in identifying feature genes compared to existing methods.
• Genes identified through OMBRFE were significantly correlated with advanced colorectal cancer clinical stages.
• The method successfully processed multi-omic data for comprehensive gene discovery.

Conclusions:

• OMBRFE is a powerful tool for identifying colorectal cancer-associated genes from integrated genomic data.
• The identified genes offer potential biomarkers for early detection of advanced colorectal cancer.
• This research contributes to a deeper understanding of colorectal cancer's molecular basis and progression.