Method for Identifying Cancer-Related Genes Using Gene Similarity-Based Collaborative Filtering

Area of Science:

• Genomics
• Bioinformatics
• Cancer Research

Background:

• Gene expression data from RNA sequencing and microarrays are crucial for understanding cancer.
• Accurate diagnosis and prognosis prediction are vital for effective cancer treatment.
• Identifying specific biomarkers can significantly improve diagnostic and prognostic accuracy.

Purpose of the Study:

• To diagnose renal cell carcinoma stage and predict breast cancer prognosis.
• To identify reliable gene expression biomarkers for cancer prediction.
• To develop a machine learning model for improved cancer classification.

Main Methods:

• Utilized RNA sequencing and microarray data for gene expression analysis.
• Employed a collaborative filtering method with gene similarity coefficients to select top-N genes.
• Constructed a machine learning classification model using a selected feature set of identified genes.

Main Results:

• The proposed method achieved higher classification performance compared to models using all gene features.
• Feature selection using the proposed method outperformed existing correlation-based methods.
• Identified optimal genetic markers that maximized classification performance.

Conclusions:

• The developed gene selection approach effectively identifies biomarkers for cancer diagnosis and prognosis.
• Machine learning models incorporating these selected biomarkers demonstrate superior performance.
• This study offers a promising strategy for advancing cancer diagnostics and personalized medicine.