Characterization of poplar metabotypes via mass difference enrichment analysis

Franco Moritz, Moritz Kaling, Jörg-Peter Schnitzler, Philippe Schmitt-Kopplin
2017 Plant, Cell and Environment  
Instrumentation technology for metabolomics has advanced drastically in recent years in terms of sensitivity and specificity. Despite these technical advances, data analytical strategies are still in their infancy in comparison with other "omics". Plants are known to possess an immense diversity of secondary metabolites. Typically, more than 70% of metabolomics data are not amenable to systems biological interpretation due to poor database coverage. Here, we propose a new general strategy for
more » ... ss spectrometry-based metabolomics that incorporates all exact mass features with known sum formulae into the evaluation and interpretation of metabolomics studies. We extend the use of mass differences, commonly used for feature annotation, by re-defining them as variables that reflect the remaining "omic" domains. The strategy uses exact mass difference network analyses exemplified for the metabolomic description of two gray poplar (Populus x canescens) genotypes that differ in their capability to emit isoprene. This strategy established a direct connection between the metabotype and the non-isoprene emitting phenotype, as mass differences pertaining to prenylation reactions were over-represented in non-isoprene emitting poplars. The analysis of mass differences was not only able to grasp the known chemical biology of poplar but it also improved the interpretability of yet unknown biochemical relationships. A major part of mass spectrometric data is not amenable to data interpretation as metabolite databases are far from being complete. This work presents the concept and rules on how Mass Difference Enrichment Analysis (MDEA) enables data driven analysis and interpretation of metabolomics data. This new metabolomics approach is presented vis-à-vis the biochemically well-characterized gray poplar isoprene emitting and non-emitting mutants, and yields results that are in perfect accordance with prior metabolite and physiological knowledge. MDEA is shown to extend prior knowledge supporting the formulation of new, testable biochemical working hypotheses. This article is protected by copyright. All rights reserved.
doi:10.1111/pce.12878 pmid:27943315 fatcat:jvllcuue5bfsnoosgw6ck3am4u