Oxidative stress is implicated in the development of a wide range of chronic human diseases,ranging from cardiovascular to neurodegenerative and inflammatory disorders. As oxidative stressresults from a complex cascade of biochemical reactions, its quantitative prediction remainsincomplete. Here, we describe a machine-learning approach to the prediction of levels of oxidativestress in human subjects. From a database of biochemical analyses of oxidative stress biomarkersin blood, plasma and urine, non-linear models have been designed, with a statistical methodologythat includes variable selection, model training and model selection. Our data demonstrate that,despite a large inter- and intra-individual variability, levels of biomarkers of oxidative damage inbiological fluids can be predicted quantitatively from measured concentrations of a limited numberof exogenous and endogenous antioxidants.
