We had available to us data from a second sample, the TRacking Adolescents’ Individual Lives Survey (TRAILS). This sample consists of 1,290 Dutch adolescents with DNA available; at three waves of assessment between the ages 10 and 18, CBCL and YSR data were collected, as was a measure of parental monitoring parallel to the CDP measure. Thirty-five SNPs were genotyped across CHRM2 in that sample, and we conducted analyses parallel to those run in the CDP. One SNP (rs10271552) yielded a p value less than .05 for the interaction effect (p = .035, β = 0.26), and results for 3 additional SNPs were suggestive (ps =.057–.098). However, none of the other SNPs genotyped showed interaction effects. The empirical p value taking into account the multiple SNPs analyzed across the gene was not significant either at the gene level (p = .51) or in a set-based analysis limited to the 6 SNPs genotyped in the linkage disequilibrium block where interaction was detected in the CDP (p = .25). The 1 SNP yielding a significant interaction effect was located in the same
