One problem with discriminant analysis of microarray data is representation of each sample by a large number of genes that are possibly irrelevant, insignificant, or redundant. Methods of variable selection are, therefore, of great significance in microarray data analysis. A new method for key gene selection has been proposed on the basis of interval segmentation purity that is defined as the purity of samples belonging to a certain class in intervals segmented by a mode search algorithm. This method identifies key variables most discriminative for each class, which offers possibility of unraveling the biological implication of selected genes. A salient advantage of the new strategy over existing methods is the capability of selecting genes that, though possibly exhibit a multimodal distribution, are the most discriminative for the classes of interest, considering that the expression levels of some genes may reflect systematic difference in within-class samples derived from different pathogenic mechanisms. On the basis of the key genes selected for individual classes, a support vector machine with block-wise kernel transform is developed for the classification of different classes. The combination of the proposed gene mining approach with support vector machine is demonstrated in cancer classification using two public data sets. The results reveal that significant genes have been identified for each class, and the classification model shows satisfactory performance in training and prediction for both data sets.