Given the reported effect of chronic alcohol use on brain volume reduction (Fein et al., 2002; Pfefferbaum et al., 1998), we tested subject’s duration of drinking along with the candidate CNV regions to assess the genetic effect on brain volume variation using a linear regression model. As seen in Fig. 7 after controlling for duration of drinking, deletion carriers still have significantly, p < 0.05, higher loading coefficients than insertion carriers or subjects with no variation. In addition to this, there is still a significant negative slope associated with the loadings of brain volume component indicating a possible dosage effect of copy number on brain volume variation.
