Abstract
Biological research is currently undergoing a revolution. With the advent of microarray technology the behavior of thousands of genes can be measured simultaneously. This capability opens a wide range of research opportunities in biology, but the technology generates a vast amount of data that cannot be handled manually. Computational analysis is thus a prerequisite for the success of this technology, and research and development of computational tools for microarray analysis are of great importance. One application of microarray technology is cancer studies where supervised learning may be used for predicting tumor subtypes and clinical parameters. We present a general Rough Set approach for classification of tumor samples analyzed with microarrays. This approach is tested on a data set of gastric tumors, and we develop classifiers for six clinical parameters. One major obstacle in training classifiers from microarray data is that the number of objects is much smaller that the number of attributes. We therefore introduce a feature selection method based on bootstrapping for selecting genes that discriminate significantly between the classes, and study the performance of this method. Moreover, the efficacy of several learning and discretization methods implemented in the ROSETTA system [18] is examined. Their performance is compared to that of linear and quadratic discrimination analysis. The classifiers are also biologically validated. One of the best classifiers is selected for each clinical parameter, and the connection between the genes used in these classifiers and the parameters are compared to the establish knowledge in the biomedical literature.