The development of effective strategies to prevent childhood obesity and its comorbidities requires new, reliable early biomarkers. Here, we aimed to identify in peripheral blood cells potential transcript-based biomarkers of unhealthy metabolic profile associated to overweight/obesity in children. We performed a whole-genome microarray analysis in blood cells to identify genes differentially expressed between overweight and normal weight children to obtain novel transcript-based biomarkers predictive of metabolic complications. The most significant enriched pathway of differentially expressed genes was related to oxidative phosphorylation, for which most of genes were downregulated in overweight versus normal weight children. Other genes were involved in carbohydrate metabolism/glucose homoeostasis or in lipid metabolism (for example, TCF7L2, ADRB3, LIPE, GIPR), revealing plausible mechanisms according to existing biological knowledge. A set of differentially expressed genes was identified to discriminate in overweight children those with high or low triglyceride levels. Functional microarray analysis has revealed a set of potential blood-cell transcript-based biomarkers that may be a useful approach for early identification of children with higher predisposition to obesity-related metabolic alterations.