metaXCMS: second-order analysis of untargeted metabolomics data

Area of Science:

• Metabolomics
• Bioinformatics
• Computational Biology

Background:

• Untargeted metabolomics identifies numerous features, complicating causal inference.
• Distinguishing direct effects from downstream alterations is a key challenge.

Purpose of the Study:

• Introduce metaXCMS, a novel software for second-order meta-analysis of untargeted metabolomics data.
• Enable analysis across multiple sample groups to prioritize causal metabolites.
• Enhance the identification of metabolites related to specific phenotypes.

Main Methods:

• metaXCMS software performs meta-analysis on XCMS results from multiple sample classes.
• Data filtering, realignment, and comparison identify shared up- or down-regulated metabolites.
• Application to three pain models to identify commonly altered metabolites.

Main Results:

• metaXCMS facilitates robust meta-analysis of untargeted metabolomics data.
• Identified histamine as a consistently altered metabolite across three pain models.
• Demonstrates increased probability of identifying causally related metabolites.

Conclusions:

• metaXCMS enhances the interpretation of complex metabolomics data.
• The software improves the identification of phenotype-associated metabolites.
• Facilitates discovery of biomarkers across diverse biological models.