Defining the multivariate genomic signature of pubertal markers and impact on lifespan psychopathology

PROJECT SUMMARY/ABSTRACT Adolescent increases in psychopathology across multiple dimensions have a significant detrimental impact on morbidity, mortality and well-being during this period of life, and set the stage for adult physical and mental health difficulties. Prevalence rates of internalizing psychopathology, which is higher on average in females, and externalizing psychopathology, higher in males, diverge during adolescence, and this trajectory continues on subsequently over the lifespan, implicating sex-differentiated mechanisms. Consistent with this, a robust literature supports individual differences in pubertal timing (e.g., onset) and pubertal tempo (e.g., rate of change) as risk-factors for increased psychopathology. However, reliance on single sex-specific indicators, retrospective reports, and cross-sectional data to measure this dynamic period of maturation have made identification of mechanisms driving puberty and psychopathology links difficult. The current study aims to capitalize on measured genomics approaches integrated with longitudinal data in both sexes to inform the genomic signal of puberty across multiple physical and hormonal indicators, and examine genetic covariation between puberty and psychopathology across the lifespan. The first aim will leverage the novel method of genomic structural equation modeling (genomic SEM; training aim 1) to combine summary statistics from published genome-wise association studies (GWAS) of pubertal timing (i.e., age of menarche, relative age of voice break, relative age of first facial hair), pubertal growth spurt, pubertal maturation (i.e., Tanner staging), testosterone, estradiol, and sex hormone binding globulin (SHBG) to identify latent pubertal genomic factors both specific to and unified across sex. Polygenic scores (PGS; training aim 2) derived from the multivariate pubertal genomic factors will be validated by out-of-sample prediction of longitudinally measured pubertal timing and tempo characterized by multiple pubertal markers in the Adolescent Brain Cognitive Development (ABCD) Study. The second aim is to investigate measured genetic covariance of multivariate pubertal genomic signal with lifespan sex-differentiated psychopathology (training aim 3). This will be achieved through (a) examining correlations between the pubertal genomic factor model and previously established factor models of the genetic architecture of adult psychiatric traits using genomic SEM, and estimating pubertal PGS prediction of sex-specific lifetime psychiatric diagnoses in the UKBiobank; and (b) probing sex-specific and sex-unified pubertal PGS effects on longitudinally modeled adolescent symptoms of psychopathology in the ABCD Study, both directly and in conjunction with longitudinally measured pubertal timing and tempo. This research will yield a comprehensive model of measured genomic signal of puberty across multiple related phenotypes in both sexes, and provide improved tools for parsing genetic and non-genetic sources of covariation to disentangle multilayered mechanisms underlying pubertal risk for psychopathology.