Enabling pan-repository reanalysis for big data science of public metabolomics data

Public untargeted metabolomics data is a growing resource for metabolite and phenotype discovery; however, accessing and utilizing these data across repositories pose significant challenges. Therefore, here we develop pan-repository universal identifiers and harmonized cross-repository metadata. This ecosystem facilitates discovery by integrating diverse data sources from public repositories including MetaboLights, Metabolomics Workbench, and GNPS/MassIVE. Our approach simplified data handling and unlocks previously inaccessible reanalysis workflows, fostering unmatched research opportunities.