Selection of Robust Feature Selection Methods Used for Gene Expression Analysis of Microarray Data

Experimental genomics involves a series of techniques and methodologies aimed at analysing the organization and behaviour of genes and genomes. Microarrays have a large number of nucleic acids of the DNA bound to a single surface. It quantifies the comparative composition of nucleic acid sequences in a sample through hybridization. Microarrays are versatile tools used in genomics research to study gene expression, DNA sequencing, protein-protein interactions, and genotyping. There are many types of microarrays, each designed for specific applications. Feature selection on microarray data is a vital step in analysing high-dimensional gene expression datasets. It aims to identify a subgroup of informative genes with the strongest association with the outcome of interest, such as disease classification or phenotype prediction. Microarray data typically contains many gene features, but every one of them is not relevant to the research question. This research study discusses and reviews the most recent feature-selection methods used in analyzing gene expression data. This study aims to provide a good introduction to microarray data and a systematic comparison of recent feature selection techniques to assist in choosing a robust technique. This aids the researchers in making a well-informed decision to work with microarrays, thereby improving the quality of gene expression analysis.