High-Quality Predicted Metabolite-Pathway Annotations Expand Pathway Coverage and Increase Enriched Pathway Detection Across Metabolomics Datasets

Background/Objectives: Metabolism-level interpretation of metabolomics datasets requires aggregation analyses across metabolites. One highly used aggregation analysis is pathway enrichment analysis (PEA), which involves detecting pathways enriched with metabolites that are differential between experimental groups. Annotating metabolites with pathway associations is a prerequisite for PEA. While several knowledge bases define pathways and include metabolite-pathway annotations, these definitions are often partially or even grossly incomplete due to limitations in current metabolic knowledge and its curation, which greatly limits the effectiveness of PEA. Methods: In this work, we used our novel multitask classification, graph convolutional-like neural network to generate high-quality metabolite-pathway annotations for pathways defined across KEGG, MetaCyc, and Reactome. We then included these predicted metabolite-pathway annotations when performing PEA on 990 datasets deposited in Metabolomics Workbench. Results: We demonstrate an 8-fold increase in the median number of enriched pathways detected across these datasets compared to using only knowledge base-derived annotations. Conclusions: The significant increase in enriched pathways has the potential to substantially improve the biological and biomedical interpretability of metabolomics datasets.