Walss-Bass C, et al. - Researchers, in this study, examined if specific immune mediators play an early role in the etiopathogenesis of depression and anxiety in adolescents. They used component-wise gradient boosting in a longitudinal design to build models of depression and anxiety in adolescents from a large set of candidate predictors, including sex, race, 39 inflammatory proteins, and the interactions between those proteins and time. They recognized transforming growth factor alpha (TGF-alpha) as a potential novel biomarker in the development of adolescent depression. Machine learning afforded optimized modeling of adolescent depression and anxiety. Predictors of adolescent depression included female sex, growth-related oncogene, and TGF-alpha. Female sex was observed to be the predictor of adolescent anxiety.