Selecting differentially expressed genes using support vector machines

DNA microarrays provide a new technique of measuring gene expression that attracted a lot of research interest in recent years. It has been suggested that gene expression data from microarrays (biochips) can be utilized in many biomedical areas, for example in cancer classification. Whereas several, new and existing, methods of classification has been tested, a selection of proper (optimal) set of genes, which expression can serve during classification, is still an open problem. It was shown that support sector machine (SVM) technique is very efficient tool for classification based on gene expression levels. Moreover, two gene selection methods Recurrent Feature Elimination (RFE) and Recurrent Feature Replacement (RFR) also use SVM methodology. In this paper we combine these two methods in order to obtain less cross-validation error. RFE method is used for finding starting gene subsets in RFR method. We illustrate effectiveness of proposed approach on papillary thyroid data set.