A two-step feature selection approach for identifying SNPs associated with colorectal cancer

Abstract

Colorectal Cancer (CRC) continues to be a significant cause of cancer-related illness and deaths worldwide. Single Nucleotide Polymorphism (SNP) identification and analysis can serve as a potential biomarker for early detection and personalized treatment. This study contributes to this ongoing discourse by employing bioinformatics methods, focusing on feature selection for SNP analysis related to CRC. Utilizing metaheuristic algorithms, particularly the Genetic Algorithm (GA), we implement a two-step feature selection method using Spatially Uniform ReliefF (SURF) and GA to identify key SNPs correlated with CRC, utilizing a dataset obtained from a prior study. Our comprehensive experiment successfully identifies previously established genes associated with CRC, while also revealing novel SNPs that warrant further investigation for validation.­­­­