Classifying different breast cancer subtypes is of great importance in breast cancer diagnosis and prevention, and Microarray data helps in cancer diagnosis and prognosis. However due to large number of features, its processing has become very complex. The current work has been done to i) find the smallest set of genes that can ensure accurate classification of breast cancer subtypes from microarray data with the help of novel gene selection method, ii) express similarity and dissimilarity among breast cancer samples based on ER and PR status using selected genes.
