Combining Feature-Based Molecular Networking and Contextual Mass Spectral Libraries to Decipher Nutrimetabolomics...

Area of Science:

• Metabolomics
• Bioinformatics
• Nutritional Science

Background:

• Untargeted metabolomics often struggles with complex data, limiting the structural identification of unknown metabolites.
• Feature-Based Molecular Networking (FBMN) is a powerful tool, but its annotation coverage can be expanded.
• Nutritional studies require robust methods for identifying plant-derived and endogenous metabolites in biofluids.

Purpose of the Study:

• To evaluate the metabolite discovery capacity of FBMN by incorporating novel contextual mass spectral libraries.
• To compare the annotation performance of the enhanced FBMN approach with existing open-source protocols.
• To analyze the postprandial metabolic behavior of annotated compounds using chemometric methods.

Main Methods:

• Created and integrated two new contextual mass spectral libraries (~300 molecules each) for plant-based nutrikinetic studies into the GNPS platform.
• Applied the FBMN approach to postprandial urinary metabolome data from a *Vaccinium* supplement intervention.
• Utilized Pearson product-moment correlation to analyze the quantitative data and metabolic pathways.

Main Results:

• The enhanced FBMN approach successfully annotated 67 berry-related and human endogenous metabolites.
• Achieved structural annotation coverage comparable to or exceeding existing non-commercial workflows.
• Identified correlations between molecular families and phase I/II metabolic processes.

Conclusions:

• Integrating novel contextual mass spectral libraries significantly boosts the annotation power of FBMN for metabolomics.
• This strategy effectively reduces the chemical space of unknowns, aiding the characterization of biochemically relevant metabolites.
• The approach is highly valuable for longitudinal studies and understanding metabolite behavior in human biofluids.