gms | German Medical Science

The focus of metabolomics studies are the end products of cellular processes that provide a snapshot of the physiological state in a

cell, tissue, or organism. Direct infusion methods coupled with high resolution mass analyzers avoid chromatography based separation of samples.

However, preprocessing and annotation of the resulting mass spectrometry data pose additional challenges since the spectra contain only information on mass-to-charge ratio (mz values) and intensities.

A new pipeline is proposed to extract intensities of relevant features and to provide annotation to the most probable metabolite. First, metabolites are identified that would be theoretically measurable in the conducted experiment, e.g. based on Lipid Maps or the Human Metabolome Database. Possible adducts and isotopes are calculated using the R package enviPat, defining the set of features of interest. Second, intensities are extracted for each feature based on local maxima as implemented in the R package FTICRMS and combined across samples. Third, features are filtered, e.g. based on detection in quality control and blank samples. Fourth, features are annotated using a Bayesian approach that integrates information about isotope patterns as well as adduct relationships and is implemented in the R package IPA.

Data from a study available in the MetaboLights database is used to show that the new pipeline identifies a similar number of metabolites and that thedetection of metabolite differences between the two disease groups and healthy controls can be reproduced.

The pipeline is implemented as an R package and will be made publicly available.

The authors declare that they have no competing interests.

The authors declare that an ethics committee vote is not required.