Regarding GxE, previous studies reported interactive effects between genetic and peer influences on the development of alcohol use behaviors in adolescents (Mrug and Windle, 2014; Salvatore et al., 2014a). The current study found no evidence of interaction between close friend substance use and PRS in predicting either initial levels of heavy episodic drinking (intercepts) or the rate of change of heavy episodic drinking from adolescence to early adulthood (slope 1) and from early adulthood to later adulthood (slope 2). It is possible that study-related differences between the current study and others may partially account for the absence of GxE effect in the current study. For instance, Mrug and Windle (2014) used a candidate gene approach to testing GxE and focused on a single (albeit well-characterized) DRD4 polymorphism in relation to peer substance use effects on trajectories of alcohol use. Salvatore et al. (2014a) found a significant GxE effect involving PRS and low parental knowledge and high peer deviance on age 14 alcohol problems, but they used cross-sectional moderation analyses on a twin sample (FinnTwin12). The failure to replicate previous findings
