Feature Selection for Cancer Classification Using Microarray Gene Expression Data
Authored by Jingjing Wu
The DNA microarray technology enables us to measure the expression levels of thousands of genes simultaneously, providing great chance for cancer diagnosis and prognosis. The number of genes often exceeds tens of thousands, whereas the number of subjects available is often no more than a hundred. Therefore, it is necessary and important to perform gene selection for classification purpose. A good subset of discriminative genes can improve prediction accuracy of classifiers and save computational cost with reduced dimension of data. In this paper, we use data on microarray gene expression level to determine marker genes that are relevant to a type of cancer. We investigate a distance-based feature selection method for two-group classification problem. In order to select marker genes, the Bhattacharyya distance is implemented to measure the dissimilarity in gene expression levels between groups. We use the support vector machine to make classification with use of the selected marker genes. The performance of marker gene selection and classification are illustrated in both simulation studies and two real data analysis.
To Read More…. Full Text in Biostatistics and Biometrics Open Access Journal in Juniper Publishers