It is advantageous to incorporate internal controls in the genotyping pipeline to estimate genotyping reproducibility rate and for selecting which markers to eliminate based on poor reproducibility. Many studies routinely genotype DNA samples from the HapMap cell lines [39,40]. In addition to providing samples of known ancestry to anchor the Structure analysis discussed in the Population Substructure section above, genotype calls on HapMap samples can be compared to the corresponding publicly available reference genotypes to estimate the degree of concordance. Genotyping for the Marshfield PMRP and Group Health was performed by CIDR, which considered any SNP having more than one replicate error on HapMap samples run with the study samples to be a technical failure, and only intensity data were released for these markers. CIDR also considered SNPs technical failures if the SNP had a call rate <85%, if the absolute difference in call rate between sexes is greater than 2.5%, if the absolute difference in heterozygosity between sexes is greater than 7%, or if cluster separation <0.20. Vanderbilt BioVU, Mayo, and Northwestern NUGene samples were genotyped at the Broad
