Metabolite Pathway Enrichment Analysis (MPEA)

MPEA is a rapid tool for functional analysis and biological interpretation of metabolic profiling data. In particular, MPEA is designed to be used with data generated by gas chromatography–mass spectrometry (GC–MS); one of the most prominent analytical methods for metabolic studies and able to quantify hundreds of small molecules from biological extracts in a single run.

The concept of MPEA is the same as that of the widely-accepted GSEA (gene set enrichment analysis). MPEA accepts a ranked list of mass spectra and tests whether metabolites belonging to some KEGG-pathway tend to occur toward the top (or bottom) of this list. More specifically, the identity of the query analytes is solved by using ms-analysis tool at GMD. The mass spectrum of the query analyte is compared against the spectral library and the query analyte is marked to belong to the KEGG pathways to which its matching reference analytes belong. Ambiguous analyte identifications are solved and for each KEGG-pathway hypergeometic distribution is then applied to calculate the statistical enrichment of KEGG-compounds of the pathways within the data (MPEA tests if there are more analytes belonging to the pathway in the list than could be expected by change.) The statistical test is repetitively performed by walking down the analyte list and the most significant rank position is stored for each pathway. To reliably estimate the random chance to obtain a result as significant as the one obtained, the entire statistical process is done multiple times after randomizing the order of analytes and by calculating permutation based enrichment values.

Compressed file that contains the tool and script that can be used to create necessary KEGG-files. Please check out the original license from KEGG web-page to find out whether or not you are allowed to retrieve the all required files.

Sample output from MPEA. The result page shows results of an analysis carried out to 14 twin-pairs with discordant for body weight described in the manuscript. The result page highlights that metabolites belonging to some KEGG-pathways are systematically more abundant or diminished between obese and non-obese humans. The results should also correspond to the results one should obtain when analyzing the second example dataset (KEGG-compound example) using default parameters.

This tool was developed by Matti Kankainen, Liisa Holm and Matej Oresic
University of Helsinki and Technical Research Center of Finland
Contact the Webmaster.
© 2010 University of Helsinki, Technical Research Center of Finland