The ‘Vitruvian Man’ by Leonardo da Vinci shows that the highly conserved relationships between different anatomical measures in man have been at the forefront of research since Roman times. Metabolomics, the analysis of the metabolic system underlying an organism as a whole, allows the observation of such relationships occurring between metabolites. However, the current ‘standard’ data analysis methods used in metabolomics do not analyse these relations with the canonical view on all metabolites—familiar from for example Principal Component Analysis (PCA). We therefore propose the concept of Between Metabolite Relationships (BMRs): common changes in the covariance (or correlation) between all metabolites in an organism. Such structural changes may indicate a metabolic change brought about by experimental manipulation, which would be lost with standard data analysis methods. These Between Metabolite Relationships can be analysed by INdividual Differences SCALing (INDSCAL). In this project the quantification of BMRs using the INDSCAL method is performed on several data sets (among others a nutritional metabolomics dataset used to study green tea metabolism, a metabolic profiling dataset on induced plant defence and a metabolomics dataset on side-effects of Prednisolone). INDSCAL is thereby compared to standard metabolomics data analysis tools.