MIDcor, an R-program for deciphering mass interferences in mass spectra of metabolites enriched in stable isotopes

BMC Bioinformatics

Methodology Article

Anibal Miranda¹  Adrián Benito¹  Silvia Marin¹  Marta Cascante¹  Ibrahim Halil Polat¹  Vitaly A. Selivanov¹  Josep J. Centelles¹  Esther Aguilar¹  Anusha Jayaraman¹  Paul W. N. Lee² 

[1] Department of Biochemistry and Molecular Biology, Faculty of Biology, Universitat de Barcelona, 08028, Barcelona, Spain;Institute of Biomedicine of the Universitat de Barcelona (IBUB) and Associated Unit to CSIC, Barcelona, Spain;Department of Pediatrics, Harbor-UCLA Medical Center, Research and Education Institute, 90502, Torrance, CA, USA;

关键词: C labeling of metabolites;    Gas chromatography/mass spectrometry;    Correction of peaks overlapping;    Isotopic effect;    Computational analysis;    R-program;   

DOI  :  10.1186/s12859-017-1513-3

 received in 2016-09-01, accepted in 2017-01-31,  发布年份 2017

来源: Springer

【 摘 要 】

BackgroundTracing stable isotopes, such as 13C using various mass spectrometry (MS) methods provides a valuable information necessary for the study of biochemical processes in cells. However, extracting such information requires special care, such as a correction for naturally occurring isotopes, or overlapping mass spectra of various components of the cell culture medium. Developing a method for a correction of overlapping peaks is the primary objective of this study.ResultsOur computer program-MIDcor (free at https://github.com/seliv55/mid_correct) written in the R programming language, corrects the raw MS spectra both for the naturally occurring isotopes and for the overlapping of peaks corresponding to various substances. To this end, the mass spectra of unlabeled metabolites measured in two media are necessary: in a minimal medium containing only derivatized metabolites and chemicals for derivatization, and in a complete cell incubated medium. The MIDcor program calculates the difference (D) between the theoretical and experimentally measured spectra of metabolites containing only the naturally occurring isotopes. The result of comparison of D in the two media determines a way of deciphering the true spectra. (1) If D in the complete medium is greater than that in the minimal medium in at least one peak, then unchanged D is subtracted from the raw spectra of the labeled metabolite. (2) If D does not depend on the medium, then the spectrum probably overlaps with a derivatized fragment of the same metabolite, and D is modified proportionally to the metabolite labeling. The program automatically reaches a decision regarding the way of correction. For some metabolites/fragments in the case (2) D was found to decrease when the tested substance was 13C labeled, and this isotopic effect also can be corrected automatically, if the user provides a measured spectrum of the substance in which the 13C labeling is known a priori.ConclusionUsing the developed program improves the reliability of stable isotope tracer data analysis.

【 授权许可】

CC BY   
© The Author(s). 2017