A gentle guide to the analysis of metabolomic data.

Modern molecular biology crucially relies on computational tools to handle and interpret the large amounts of data that are generated by high-throughput measurements. To this end, much effort is dedicated to devise novel sophisticated methods that allow one to integrate, evaluate, and analyze biological data. However, prior to an application of specifically designed methods, simple and well-known statistical approaches often provide a more appropriate starting point for further analysis. This chapter seeks to describe several well-established approaches to data analysis, including various clustering techniques, discriminant function analysis, principal component analysis, multidimensional scaling, and classification trees. The chapter is accompanied by a webpage, describing the application of all algorithms in a ready-to-use format.