Abstract We aimed to test the association between sleep-related polygenic scores (PGSs) and accelerometer-based sleep metrics among Brazilian adolescents, and to evaluate potential mechanisms underlying the association through the enrichment of obesity, and cortisol pathway-specific polygenic scores (PRSet). Utilizing data from The 2004 Pelotas (Brazil) Birth Cohort, sleep time window and sleep efficiency were measured at the 11-year-old follow-up using ActiGraph accelerometers. Three sleep-PGSs were developed based on the most recent genome-wide association study (GWAS) of accelerometer-based sleep measures. PRSet, calculated using variants linked to body mass index (BMI) and plasmatic cortisol concentration, aimed to assess pleiotropic effects. Linear regression models, adjusted for sex and the first 10 principal components of ancestry, were employed to explore the impact of sleep-PGS and specific-PRSet on sleep phenotypes. The number of nocturnal sleep episodes-PGS was positively associated with sleep time window (β =2.306, SE: 0.92, p=0.011). Nocturnal sleep episodes were also associated with sleep time window when restricted to BMI-PRSet (β=2.682, SE: 0.912, competitive-p=0.003). Both number of sleep episodes and sleep time window cortisol-PRSets were associated (β=0.002, SE: 0.001, p=0.013; β=0.003, SE: 0.001, p=0.003, respectively) and exhibited enrichment in molecular pathways (competitive-p=0.011; competitive-p=0.003, respectively) with sleep efficiency. Sleep polygenetic components observed in European adults may partially explain accelerometer-based sleep time window in Brazilian adolescents. Specific BMI molecular pathway strengthened the association between sleep-PGS and sleep time window, while cortisol concentration pathway had a significant impact on the genetic liability for sleep efficiency. Our results suggest genetic overlap as a potential etiological pathway for sleep-related comorbidities, emphasizing common genetic mechanisms.