A hybrid of Information gain and a Coati Optimization Algorithm for gene selection in microarray gene expression data classification.

Document Type : Original Article


Computer Science Department, Faculty of Computers and Information, Minia University, Minia, Egypt


Gene expression data has become an essential tool for cancer classification because it provides substantial insights into the underlying mechanisms of cancer progression. However, the high-dimensional nature of microarray gene expression data presents a significant challenge. This paper introduces a new method called IG-COA, which combines Information Gain (IG) approach and Coati Optimization Algorithm (COA), to identify the biomarkers genes. COA is a recent algorithm that has not been previously examined for feature or gene selection, to the best of our knowledge. Firstly, the IG method is used because using COA directly on microarray datasets is ineffective and can make it challenging to train a classifier accurately. Secondly, the COA algorithm is utilized to select the optimal subset of genes from the previously selected ones. The effectiveness of the suggested IG-COA method with a Support Vector Machine is tested on several microarray gene expression datasets, and it exceeds other state-of-the-art methods.


Main Subjects