Because there is evidence to suggest a connection between AUD and brain structural variations (Harper & Kril, 1989; Jernigan et al., 1991; de la Monte, 1988; Pfefferbaum et al., 1995, 1998); we use brain volume as a putative intermediate phenotype to connect CNVs with the risk to AUD. The dual factorization framework of parallel independent component analysis (parallel ICA) presented in (Jingyu, Demirci, & Calhoun, 2008; Liu et al., 2009) is used to search for connected brain volume variation and hazardous drinking behavior and is explained below.
