Fig. 1: Schematic representation of our conceptual framework.

In the ABCD sample, we set out to test the causal sequence underpinning the relationship between polygenic risk (panel A) and differential vulnerability to internalizing vs externalizing spectrum disorders (panel D) via neurobiological maturation processes: sexual differentiation of brain function (panel B), and systemic aging (panel C), specifically, physiological wear-and-tear (i.e., immune and metabolic dysregulation, assessed via the PhenoAge algorithm, panel C-i) and reproductive maturation (specifically, pubertal hormone levels and self-reported adrenarche vs gonadarche, panel C-ii). The PhenoAge algorithm (panel C–i) uses an exponential function to predict mortality from a set of biomarkers in a reference group. An individual’s PhenoAge biological age prediction corresponds to the chronological age at which their mortality risk would be normal in the reference group. A PhenoAge estimate higher/lower than an individual’s chronological age indicates advanced/delayed physiological aging, respectively. The longitudinal analyses from the ABCD sample (as indicated through the dark blue arrows) were supplemented by an examination of the cross-sectional relationships among the same variables (apart from genetic risk, PhenoAge, and pubertal hormones) in the HCP-D sample (as indicated through the light blue arrows). ADHD attention deficit hyperactivity disorder, BMI body mass index, CD conduct disorder, OCD obsessive-compulsive disorder, ODD Oppositional Defiant Disorder, PTSD post-traumatic stress disorder, SCT slow cognitive tempo.