Methods and apparatus for classifying or predicting the classes for samples based on gene expression are described. Also described are methods and apparatus for ascertaining or discovering new, previously unknown classes based on gene expression. Methods, computer systems and apparatus for classifying or predicting whether a sample is treatment sensitive (e.g., chemosensitive) or treatment resistant (e.g., chemoresistant) are also provided. Classification occurs based on analysis of gene expression data from samples that have been subjected to one or more compounds.