Metabolites, the chemical entities that are transformed during metabolism, provide a functional readout of cellular biochemistry that offers the best prediction of the phenotype and the nature of a disease. Mass spectrometry now allows thousands of metabolites to be quantitated. The targeted or untargeted data from metabolic profiling can be combined with either supervised or unsupervised approaches to improve interpretation. These sophisticated statistical techniques are computationally intensive. This chapter reviews techniques applicable to metabolomics approaches to disease.