Gene Extraction and Identification Tumor/Cancer for Microarray Data of Ovarian Cancer

A typical microarray data of ovarian cancer consists of the expressions of tens of thousands of genes on a genomic scale. To avoid higher computational complexity, we want to find the most likely differentially expressed gene that best explain the effects of tumor/cancer for ovarian cancer. In this paper, we derive a hybrid approach for extracting and evaluating informative genes from microarray data of ovarian cancer. Furthermore, the expression patterns of extracted genes are used to identify tumor/cancer for ovarian cancer. In the proposed approach, the method analysis of variance (ANOVA) will, using P-value, test gene expressions that are significantly different in microarray data. The genetic algorithm is applied to extract genes and then the support vector machine is processed to identify tumor/cancer for ovarian cancer. We show that this extracted set of genes can be used to significantly identify ovarian tumors (OVT) and ovarian cancers (OVCA)