The results of this study suggest that specific CNVs play a role in brain volume variations and drinking behavior, however there are several limitations. First, the subject sample size included in this study is relatively small given CNV’s infrequency. Other exploratory studies which have found statistically significant results using CNVs (Pinto et al., 2010) have included significantly larger sample sizes (approximately 1000). We partially mitigate this effect by introducing brain volume as an intermediate phenotype to gain statistical power. However, we have conducted a genome wide search for CNVs and by including such a large number of CNVs in the study, the statistical power is further limited. Finally, the results presented here are only statistical relations between variables. They need to be further validated by cellular studies of the genetic function at the identified CNV region.
