A Two-step Feature Selection Approach for Identifying SNPs Associated With Colorectal Cancer
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.
Authors:
Jason Sebastian Sulistyawan, Kelvin Julian, Gregorius Natanael Elwirehardja, Kuncahyo Setyo Nugroho, Bens Pardamean
Communications in Mathematical Biology and Neuroscience (CMBN)