Using information from chromosomes 1-22 (919,509 SNP markers) on the clean set of 1,888individual samples, we estimated Identity by State (IBS) and Identity by Descent (IBD) using PLINK (Purcell et al., 2007) as well asthe Kinship Coefficient using KING (Manichaikul et al., 2010). These measures are used to test duplicate concordance, confirm expected biological relationships, identify unknown or cryptic relatedness in the sample and provide the information necessary to assess genetic ancestry. The relationship measures are calculated pairwise for all individuals in the dataset. As generally recommended, we pruned autosomal SNPs to establish an approximately independent set of SNP markers to be used for IBS, IBD and Kinship Coefficient estimation. We used a linkage disequilibrium threshold (r2) of 0.20 with a SNP window size of 50 and number of SNPs to shift window at each step of 5 (PLINK command: --indep-pairwise 50 5 0.20). After pruning, a set of 231,649 autosomal SNP markers in approximate linkage equilibrium was used to estimate the relationship measures. Pairwise mean IBD was estimated using PLINK (“PI_HAT”). However, PLINK’s estimates of IBD may be biased
