2025-W9: Metabolomics Data Standardization and Harmonization for AI

Presenters: Rima Kaddurah-Daouk (Duke University), Session Overview and Goals Rick Dunn (University of Liverpool, UK), Metabolomics Standards Initiative (MSI) -I, 2007 and Launch of MSI -2 Jennifer Kirwan (Berlin Institute of Health at Charite Universitatsmedizin), mQACC perspective and Lessons learned from cohort studies Xianlin Han (UT Health – San Antonio), Lipidomics Standards Initiative and Outcomes Peter Meikle (Baker Heart and Diabetes Institute), Harmonization of Lipidomics Datasets and Novel approaches – from lipidomics to metabolomics Thomas Hankemeier (Leiden University) and the Dutch Team, FAIR Data Machine Readable and AI Ready Tuulia Hyötyläinen (Örebro University) and Team Members, Exposomics Data and Its Integration with Metabolomics and Lipdomics data Moderators, Speakers, and All, Discussion: Future concept, data visitation, and AI applications Euretos Moderators/Discussants: David Wishart, University of Alberta Matej Orešič, Örebro University Tuulia Hyötyläinen, Örebro University Susan Sumner, UNC Chapel-Hill Description: This workshop brings together leaders in large initiatives focused on integrating metabolomics and lipidomics data across studies, assays, laboratories, and repositories. Starting with updated recommendations from the Metabolomics Standards Initiative, the workshop will explore optimizing large-scale data use in open biorepositories, focus on standardizing lipid annotations and quantifications, and incorporate new approaches for data harmonization and FAIRification to coalesce different data types and existing knowledge. Finally, it will explore using digital twinning and AI to search across data resources. This workshop sets the stage for broader community discussions that include metabolomics and lipidomics societies, large initiatives, task groups, and interest groups to define the next steps in data reporting, integration, harmonization, data sharing, and visiting strategies. Workshop Objectives:  Metabolomics Lipidomics data standardization and harmonization complexities  getting data ready for AI  connecting biorepositories  future data visitation Learning Outcomes:  better practices in metabolomics data and its reporting  data harmonization approaches  machine readable data for AI applications