With the advent of high throughput and high dimensional metabolomics data the question of the appropriate interpretation and integration with other omics data became pivotal. Here we focus on the human model system since it is the best characterized model system and thereby can be seen as a benchmark for the direction of developments for other species.
In addition the basics on which the selection and the mode how metabolomics data can be combined with other omics data are reviewed. Genome, transcriptome and proteome data are discussed in respect to their inherent characteristics in terms of measurement, data features, coverage and their relatedness to metabolome data. Due to its recently gained importance we also review the specific features of serum metabolomics. On the metabolome side we discuss the emerging relevance of flux and the often neglected tight interconnectedness of small molecules with signaling pathways. The contribution of metabolome data to integrated pathway analyses is so far based on either only the combination, intrinsic correlation or detection of overrepresentation. In order to overcome the shortcomings of these approaches data interpretation should be performed by making use of a detailed pathway topology. This functional integration is crucial for the intended comprehensive phenotyping and demands for specific precautionary means of data assessment and data evaluation.
Keywords: Data integration, genomics, metabolomics, proteomic, transcriptomics.