The well defined genetic architecture and metabolic network of Saccharomyces cerevisiae make this organism a cornerstone for metabolomics research. Recent efforts have focused on robust sample preparation techniques, analytical tools to quantitatively identify hundreds of metabolites at the same time, and elegant approaches for analyzing and interpreting the data. While equally important, we focus here on approaches for extracting useful information from the data itself. We outline several statistical and mathematical methods that can be used to digest and validate the most important features in the data. These multivariate approaches are from either the well established standard portfolio of statistical methods, or can be adapted from other areas where similar problems can be identified and where statistical and mathematical methods exist. Looking forward, we also describe approaches for fusing metabolome data with other cellular measurements and network structure to elucidate biosynthetic control mechanisms.