MET-COFEA: A Liquid Chromatography/Mass Spectrometry Data Processing Platform for Metabolite Compound Feature Extraction and Annotation

• Published in: Analytical chemistry (Washington) Vol. 86; no. 13; pp. 6245 - 6253
• Main Authors: Zhang, Wenchao, Chang, Junil, Lei, Zhentian, Huhman, David, Sumner, Lloyd W, Zhao, Patrick X
• Format: Journal Article
• Language: English
• Published: United States American Chemical Society 01.07.2014
• Subjects: Algorithms; Alignment; Annotations; Chromatography; Chromatography, Liquid - methods; Extraction processes; Feature extraction; Liquid chromatography; Mass spectrometry; Mass Spectrometry - methods; Metabolites; Metabolomics - methods; Software
• ISSN: 0003-2700; 1520-6882
• DOI: 10.1021/ac501162k

In this paper, we present a novel liquid chromatography/mass spectrometry (LC/MS) data processing and analysis platform, MET-COFEA (METabolite COmpound Feature Extraction and Annotation). MET-COFEA detects and clusters chromatographic peak features for each metabolite compound by first comprehensively evaluating retention time and peak shape criteria and then annotating the associations between each peak’s observed m/z value with the corresponding metabolite compound’s molecular mass. MET-COFEA integrates a series of innovative approaches, including novel mass trace based extracted-ion chromatogram (EIC) extraction, continuous wavelet transform (CWT)-based peak detection, and compound-associated peak clustering and peak annotation algorithms. On the basis of the deduced neutral molecular mass and retention time, we have also developed a new alignment algorithm that uses compound-associated peak groups instead of individual peaks to align the same metabolite compound across samples from different electrospray ionization (ESI) modes, different instruments, even different experimental conditions. MET-COFEA has been systematically tested on a series of LC/MS profiles of mixed standards at different concentrations as well as real untargeted LC/MS plant metabolomics data. We compared the performances of MET-COFEA with the existing publicly available tools at LC/MS peak analysis level and demonstrated its excellent performance in this arena. MET-COFEA is freely available at http://bioinfo.noble.org/manuscript-support/met-cofea/.