High-Quality Predicted Pathway Annotations Greatly Improve Pathway Enrichment Analysis of Metabolomics Datasets.

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 knowledgebases 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. In this work, we used a 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 Metabolomics Workbench deposited datasets. Finally, we demonstrate over a 10-fold increase in the median number of enriched pathways detected across these datasets compared to using only knowledgebase-derived annotations, substantially improving their biological and biomedical interpretability.