Main content area

MetaboQC: A tool for correcting untargeted metabolomics data with mass spectrometry detection using quality controls

Calderón-Santiago, Mónica, López-Bascón, María A., Peralbo-Molina, Ángela, Priego-Capote, Feliciano
Talanta 2017 v.174 pp. 29-37
algorithms, data collection, filtration, ionization, mass spectrometry, metabolites, metabolomics, principal component analysis, quality control
Nowadays most metabolomic studies involve the analysis of large sets of samples to find a representative metabolite pattern associated to the factor under study. During a sequence of analyses the instrument signals can be subjected to the influence of experimental variability sources. Implementation of quality control (QC) samples to check the contribution of experimental variability is the most common approach in metabolomics. This practice is based on the filtration of molecular entities experiencing a variation coefficient higher than that measured in the QC data set. Although other robust correction algorithms have been proposed, none of them has provided an easy-to-use and easy-to-install tool capable of correcting experimental variability sources. In this research an R-package –the MetaboQC– has been developed to correct intra-day and inter-days variability using QCs analyzed within a pre-set sequence of experiments. MetaboQC has been tested in two data sets to assess the correction effects by comparing the metabolites variability before and after application of the proposed tool. As a result, the number of entities in QCs significantly different between days was reduced from 86% to 19% in the negative ionization mode and from 100% to 13% in the positive ionization mode. Furthermore, principal component analysis allowed detecting the filtration of instrumental variability associated to the injection order.