Figure 2 illustrates how the strength of the association (depicted as standardized betas) between life events and TRF externalizing varied as a function of genotype. In the model that included parental antisocial behavior (Panel A, right) life events predicted subsequent TRF externalizing for those with no copies of the MA (β0MA = 0.19, 95% CI [0.03, 0.35]), but not for those with one or two copies of the minor allele (β1MA = 0.07, 95% CI [−0.05, 0.20]); β2MA = −0.16, 95% CI [−0.39, 0.06]). Similarly, in the model that included parental alcohol problems (Panel B, right), life events predicted subsequent TRF externalizing for those with no copies of the MA (β0MA = 0.17, 95% CI [0.03, 0.32]), but not for those with one or two copies of the MA (β1MA = 0.08, 95% CI [−0.05, 0.21]); β2MA = −0.17, 95% CI [−0.38, 0.05]). In supplementary analyses that included all covariate × environment and covariate × genotype interaction terms, the interaction between life events and GABRA2 continued to be significant (p values ≤ 0.02). These results are summarized as Model 3 in Tables 5 and 6.
