UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Epigenetic signatures of developmental psychopathology : time-varying associations of general psychopathology across childhood and adolescence with late adolescent DNA methylation Separovic, Lea

Abstract

Psychopathology is known to fluctuate throughout childhood and adolescence, in association with a variety of intrinsic (e.g., genetic) and extrinsic (e.g., environmental) factors. DNA methylation (DNAm) is a commonly studied epigenetic modification that can reflect the interplay between genetics and environments. DNAm is a stable, reversible mark added onto DNA that can regulate gene expression and may serve as a biomarker of psychopathology. Many studies have identified links between DNAm and childhood and adolescent psychopathology, but most have been limited by assessing narrow timeframes. The present study harnessed a wealth of longitudinal data from a community-based cohort of children in Wisconsin (n=170) to examine the General Psychopathology Factor (GPF) – a broad construct capturing shared variance between psychiatric symptoms – across childhood and adolescence, in relation to DNAm from buccal epithelial cells collected at age 18. To determine which GPF timepoint (of 7 timepoints across age 9-18) or GPF trajectory (of 5 derived trajectory measures) provided the best explanation of age 18 DNAm patterns, a two-phase comparative statistical approach was employed, consisting of screening measures for high-dimensional data and the Structured Life-Course Modelling Approach (SLCMA). The SLCMA identified which GPF measure explained the most DNAm variation at 298,415 methylation sites (CpGs). At 12,910 CpGs, the selected GPF measure explained more DNAm variation than expected by chance (R² > 3.7%). After multiple test correction, 486 medium-confidence (Padjusted < 0.20), and 10 high-confidence (Padjusted < 0.05) associations remained. Age 16 GPF was the most frequently selected timepoint; worsening GPF was the most frequently selected trajectory. This suggested that adolescent psychopathology was potentially particularly reflective of adolescent biological patterns, highlighting a timeframe for investigation if longitudinal data are unavailable. Additionally, worsening mental health across childhood and adolescence, regardless of the degree of worsening, seemed particularly salient for biological embedding. Overall, this study demonstrated the application of a two-phase approach for comparative longitudinal analyses, and these findings highlighted time-varying associations between developmental psychopathology and DNAm, illustrating the degree to which trajectory- and age-specific GPF was reflected biologically. Future research should explore this in larger and more diverse samples, and examine potential windows for intervention and prevention.

Item Citations and Data

Rights

Attribution-NonCommercial-NoDerivatives 4.0 International