Abstract—Microarray based gene expression profiling has become an important and promising dataset for cancer classification that are used for diagnosis and prognosis purposes. It is important to determine the informative genes that cause the cancer to improve early cancer diagnosis and to give effective chemotherapy treatment. Furthermore, find accurate gene selection method that reduce the dimensionality and select informative genes is very significant issue in cancer classification area. In literature, there are several gene selection methods for cancer classification using microarray dataset. However, most of them did not concern on identifying minimum number of informative genes with high classification accuracy. Therefore, in our research study we discuss the performance of Bio-Inspired evolutionary gene selection method in cancer classification using microarray dataset. And, we prove that the Bio- Inspired evolutionary gene selection methods have superior classification accuracy with minimum number of selected genes. Index Terms—Bio-inspired evolutionary methods, cancer classification, microarray, gene selection, gene expression. I. INTRODUCTION Gene expression profiling or DNA microarray dataset has enabled the measurement of thousands of genes in a single RNA sample by hybridized to a labeled unknown molecular extracted from a particular tissue of interest. It offers an efficient method of gathering data that can be used to determine the patterns of gene expression of all the genes in an organism in a single experiment [1], [2]. DNA microarrays can be used to determine which genes are being expressed in a given cell type at a particular time and under particular conditions, to compare the gene expression in two different cell types or tissue samples, and then we can determine the more informative genes that are responsible to cause specific disease or cancer [3]. Recently, microarray technologies have opened up many windows of opportunities to investigate cancer diseases using gene expressions. In our research, we focus on microarray data classification and cancer microarray dataset classification in particular. The primary task of microarray data classification is to determine a computational model from a given microarray data that determines the class of unknown samples. Accuracy, quality, and robustness are important elements of microarray classification models. The Manuscript received December 25, 2013; revised March 18, 2014. The authors are with Computer Science Department, King Saud University, Saudi Arabia (tel.: +96612309954; e-mail: halshamlan@ksu.edu.sa, badrghada@hotmail.com, yousef@ksu.edu.sa). accuracy of microarray dataset classification depends on both the quality of the provided microarray data and the utilized classification method. However, microarray dataset suffers from the curse of dimensionality, the small number of samples, and the level of irrelevant and noise genes, makes the classification task for a given sample more challenging [4] [5]. Those irrelevant genes not only introduce some unnecessary noise to gene expression data analysis, but also increase the dimensionality of the gene expression matrix, which results in the increase of the computational complexity in various consequent researches such as classification and clustering [6]. As a consequence, it is significant to eliminate those irrelevant genes and identify the informative genes, which is a feature (genes) selection problem crucial in microarray data analysis. Therefore, the first step of classifying the microarray data is to identify a small subset of genes that are primarily more predictive for the cancer [5]. In literature, there are several gene selection methods for cancer classification using microarray dataset. However, most of them did not concern on identifying minimum number of informative genes with high classification accuracy. In literature, up to our knowledge, there are several gene selection methods that are based on Bio-Inspired evolutionary algorithm such as Genetic Algorithm (GA), Particle Swarm Optimization (PSO), or Ant Colony Optimization (ACO). These gene selection methods are capable of searching for optimal or near-optimal solutions on complex and large spaces of possible solutions. Thus, in this paper we will evaluate the performance of Bio-Inspired evolutionary gene selection methods in cancer classification using microarray gene expression profile. And, we prove that the Bio-Inspired evolutionary gene selection methods have superior classification accuracy with minimum number of selected genes. The rest of this paper is organized as follow: In Section II, first we define what is the gene selection process in cancer classification using microarray gene expression profile, followed by description of the various kinds of gene selection methods. The performance of Bio-Inspired evolutionary gene selection method is discussed in section III. In Section IV, we present our contribution. We conclude the paper in Section V. II. GENE SELECTION METHODS IN CANCER CLASSIFICATION Gene selection is the process of selecting the smallest subset of informative genes that are most predictive to its relative class using a classification model. This maximizes The Performance of Bio-Inspired Evolutionary Gene Selection Methods for Cancer Classification Using Microarray Dataset Hala M. Alshamlan, Ghada H. Badr, and Yousef A. Alohali International Journal of Bioscience, Biochemistry and Bioinformatics, Vol. 4, No. 3, May 2014 166 DOI: 10.7763/IJBBB.2014.V4.332