Metabolite identification: are you sure? And how do your peers gauge your confidence?

• DOI: 10.1007/s11306-014-0656-8
• Publication Date: 2014-04-07T13:45:36Z
• Authors: Darren J. Creek; Warwick B. Dunn; Oliver Fiehn; Julian L. Griffin; Robert D. Hall; Zhentian Lei; Robert Mistrik; Steffen Neumann; Emma L. Schymanski; Lloyd W. Sumner; Robert Trengove; Jean-Luc Wolfender
• Keywords: 0301 basic medicine; ddc:615; 03 medical and health sciences; info:eu-repo/classification/ddc/615; spectrometry data

Metabolomics is still faced with several significant challenges which currently limit its full scientific potential. The identification of metabolites is essential to convert analytical data into meaningful biological knowledge. However, identification confidence can vary widely because the process of identification is complex and dependent on the analytical platform and robustness of the methods applied, as well as the databases and resources used. Confident and unequivocal structure identification requires significant effort, which is multiplied dramatically in non-targeted metabolomics studies where 10–100s of metabolites can be deemed as biologically important and require identification. Mass spectrometry (MS), nuclear magnetic resonance spectroscopy (NMR) or integrated MS–NMR strategies (Dunn et al. 2013; Kind and Fiehn 2010; van der Hooft et al. 2011) provide much information for the identification of metabolites (e.g. 1D/2D-NMR and MS/MS).