On biclustering with feature selection for microarray data sets

Let a set of training and test samples be given, and the samples from the training set be partitioned into a number of classes, while classification of the test samples is unknown. The classification problem consists in determining classes of the test samples utilizing the information provided by the training set. Supervised biclustering is a specific type of classification problems, where we simultaneously partition both the set of samples and the set of their features. Samples and features classified together are supposed to have a high relevance to each other which can be observed by intensity of their expressions. Moreover, not all features of the data set are informative, and we need to find a subset of features relevant to the classification of interest. This task is called the feature selection. In this paper we applied supervised biclustering with feature selection to several microarray data sets. Computational results indicate that the obtained solution provides a reliable feature selection and the test set classification based on it.