File Entry: Genetic Algorithms Wrapper Approach to Select Informative Genes for Gene Expression Microarray Classification using Support Vector Machines

 Constantly improving gene expression technology offer the ability to measure the expression levels of thousand of genes in parallel. Gene expression data is expected to significantly aid in the development of efficient cancer diagnosis and classification platforms. Key issues that need to be addressed under such circumstances are the efficient selection of minimum number of genes that contribute to a disease from the thousands of genes measured on microarrays that are inherently noisy. This work deals with finding the minimum number of informative genes from gene expression microarray data which maximum the classification accuracy. In this work, we apply genetic algorithm wrapper to search out and identify the minimum number of potential informative genes combinations for classification and then use the classification accuracy from the support vector machine classifier to determine the fitness in genetic algorithm for each of the combinations. Experimental results using benchmark dataset produced the proposed approach achieves better classification accuracies by using minimum informative genes than other published methods on the same datasets. The genes from the outcomes are explored for biological plausibility.