Decoding the role of transcriptomic clocks in the human prefrontal cortex.

Aging is characterized by interindividual variability quantified through aging clocks, informed by biological data and are linked to diverse health outcomes. In this study, we comprehensively profiled several transcriptomic clocks, characterize their stochastic components, compare their predictive accuracy, perform associations with mental disorders, and elucidate their functional implications in three human prefrontal cortex (PFC) datasets. Our analysis revealed substantial heterogeneity in transcriptomic age prediction across different clock signatures. Notably, deep learning models capture a greater proportion of stochastic variation compared to linear methods, suggesting they may be sensitive to non-deterministic components. We identified a consistent relationship between transcriptomic age and alcohol use. Using network analysis, we identified convergent biological mechanisms across different clocks despite limited gene-level overlaps, specifically, immune-related signaling and extracellular matrix remodeling. Our findings reveal associations between psychiatric traits and identify convergent biological pathways and networks in the human PFC that may influence aging and health risks.