Nichole Reisdorph, Corey D Broeckling, Rafael Brüschweiler, Clary B Clish, Valérie Copié, Timothy Md Ebbels, Arthur S Edison, Facundo M Fernández, Ulrich L Günther, Ewy A Mathé, Vanessa V Phelan, Robert Powers, Daniel Raftery, Tracey B Schock, Dajana Vuckovic, David S Wishart, Min Zhang, Elizabeth Want
Metabolomics has become a mainstream approach that examines the metabolic composition of biological systems in relation to their physiological and disease states. Using analytical techniques such as nuclear magnetic resonance spectroscopy and mass spectrometry, metabolomics enables detailed profiling of small molecules in cells, tissues, and biofluids. Recent methodological advances, along with the integration of big data analytics and expanding metabolite databases, have significantly enhanced the accuracy and scope of metabolomics research. However, despite over twenty years of development, metabolomics still encounters challenges in annotation, automation, and standardization, which are crucial for fully leveraging recent advances in machine learning and applying metabolomics in biological and clinical studies. Recently, a group of experts assembled to discuss future strategies for standardization and automation and their potential in clinical research. This review underscores critical and unresolved issues in current metabolomics research and suggests pathways toward solutions.