Data correction strategy for metabolomics analysis using gas chromatography–mass spectrometry

Gas chromatography–mass spectrometry metabolomics requires the original sample's derivatization. Therefore, systematic biases that might distort the one-to-one proportional relationship between the original metabolite concentration and derivative peak area profiles have to be considered. The first type of such biases change only the value of the proportionality constant between the two profiles among samples and are corrected by the use of an internal standard. The second type, however, might distort the one-to-one relationship and also change the proportionality constant between the two profiles among samples to a different fold-extent for each metabolite. Metabolomic profiles should be corrected from these biases, because changes due only to chemical kinetics could be assigned biological significance. This paper presents the first streamlined data correction and validation strategy that does not jeopardize the high-throughput nature of metabolomic analysis. This context allowed also for the chemical annotation of 15 currently unknown derivative peaks of (NH2)-group containing compounds.