demonstrated a critical role of multiethnic studies in genome research of complex disease(Pulit et al., 2010; Masunuru et al., 2012; Carlson et al., 2013; Gong et al., 2013; Manichaikul et al., 2012; Manku et al., 2013; Marigorta et al., 2013; Sabater-Lleal et al., 2013). Often, these multiethnic studies will conduct genome-wide analyses within a relatively homogenous European descent sample and simultaneously conduct a genome-wide analysis among a more genetically diverse sample such as African-Americans before combining the association signals using meta-analysis. An alternative approach, and one that was chosen for the present study, conducts the genome-wide analysis on the entire sample across multiple ethnic backgrounds. This approach has been successfully conducted in other studies of complex disease (for examples, see Kurreeman et al., 2012 and Xu et al., 2013). However, rather than use the Add Health sample as a discovery sample (often requiring very large sample sizes) we adopted a weighting scheme based upon the GIANT consortium (Speliotes et al., 2010) that is comprised of a series of European descent samples. Therefore, the approach taken for the present study is one that explores the extent to which the variants discovered in European descent GIANT consortium may also be of relevance
