The rising incidence of pediatric obesity and associated comorbidities has created an urgent need to identify effective therapeutic approaches. To address this problem, source-driven metabolite predictions (based on diet, microbiome, lifestyle and psychological parameters) are proposed, whereby links between metabolite sources and their biomarker signatures can be obtained as a basis towards individualization of treatments (4P principle). To build this framework, source data and metabolome signatures from 3 pediatric cohorts (MetaBEAse, FAME and ENVIRONAGE) compromising 1817 samples, will be included in a machine learning-based prediction model. To demonstrate the feasibility of moving 4P medicine to routine clinical practice, a patented sampler for optimal gut metabolome coverage (i.e. MetaSAMP®) will be further developed into a kit design and integrated into a rapid metabolomics workflow. This workflow will be cross-correlated with the conventional metabolomics workflow and a selection of source-relevant metabolites will be monitored during a 12-week individualized intervention, comprising dietary and lifestyle counselling, pro-, pre- and/or synbiotic supplementation and/or psychological therapy on a representative selection of children with overweight. Alterations in predicted metabolic signatures of source-relevant metabolites and clinical data will be used to assess the improvement of metabolic status, paving the way towards effective routinely applicable 4P medicine.