Null diffusion-based enrichment for metabolomics data

• Published in: PloS one Vol. 12; no. 12; p. e0189012
• Main Authors: Picart-Armada, Sergio, Fernández-Albert, Francesc, Vinaixa, Maria, Rodríguez, Miguel A., Aivio, Suvi, Stracker, Travis H., Yanes, Oscar, Perera-Lluna, Alexandre
• Format: Journal Article Publication
• Language: English
• Published: United States Public Library of Science 06.12.2017; Public Library of Science (PLoS)
• Subjects: Analysis; Aplicacions de la informàtica; Bioengineering; Bioinformatics; Bioinformàtica; Biologia de sistemes; Biology and Life Sciences; Biomedical materials; Biomedical research; Cell metabolism; Chromatography; Computer simulation; Crosstalk; Data bases; Data processing; Datasets; Diabetes; Enrichment; Experiments; Gene expression; Genetic engineering; Genomes; Graphical representations; Informàtica; Investigació operativa; Knowledge base; Knowledge bases (artificial intelligence); Labeling; Labelling; Matemàtica aplicada a les ciències; Matemàtiques i estadística; Medicine and Health Sciences; Metabolic disorders; Metabolites; Metabolomics; Metabòlits; Methods; Mitochondria; Network analysis; NMR; Nuclear magnetic resonance; Physical Sciences; Physiological aspects; Research and Analysis Methods; Simulació; Simulació per ordinador; Statistical hypothesis testing; Studies; Systems biology; Topology; Àrees temàtiques de la UPC; Spain
• ISSN: 1932-6203; 1932-6203
• DOI: 10.1371/journal.pone.0189012

Metabolomics experiments identify metabolites whose abundance varies as the conditions under study change. Pathway enrichment tools help in the identification of key metabolic processes and in building a plausible biological explanation for these variations. Although several methods are available for pathway enrichment using experimental evidence, metabolomics does not yet have a comprehensive overview in a network layout at multiple molecular levels. We propose a novel pathway enrichment procedure for analysing summary metabolomics data based on sub-network analysis in a graph representation of a reference database. Relevant entries are extracted from the database according to statistical measures over a null diffusive process that accounts for network topology and pathway crosstalk. Entries are reported as a sub-pathway network, including not only pathways, but also modules, enzymes, reactions and possibly other compound candidates for further analyses. This provides a richer biological context, suitable for generating new study hypotheses and potential enzymatic targets. Using this method, we report results from cells depleted for an uncharacterised mitochondrial gene using GC and LC-MS data and employing KEGG as a knowledge base. Partial validation is provided with NMR-based tracking of 13C glucose labelling of these cells.