We explored genetic ancestry in two different ways. Note that the sample of N=1,888 individuals with clean genotypes includes two MZ twin pairs. For the purposes of estimating genetic ancestry, we removed one individual (randomly) from each of the two MZ twin pairs resulting in a final analysis sample of N=1,886. For our first approach to estimating genetic ancestry, we used KING (Manichaikul et al., 2010) to identify clusters of individuals based upon genetic similarity. KING uses multidimensional scaling (MDS) with Euclidean distance to generate principal coordinates (PCs) that can be used to identify population substructure. For the KING procedure, we used the same set of 231,649autosomal SNP markers in approximate linkage equilibrium that was used for the estimation of genetic relatedness.
