all arrays for the samples to be combined and then imputed up to a common set of HapMap SNPs for analyses from a common set of genotyped SNPs. To test this hypothesis and correction strategy, we examined the degree to which each of the following occurs: (1) imputation across arrays based on the union of genotyped SNPs (i.e., SNPs available on one or more arrays) results in bias as evidenced by spurious associations (type 1 error) between imputed genotypes and arbitrarily assigned case/control status; (2) imputation across arrays based on the intersection of SNPs genotyped on all arrays does not evidence such bias; and (3) imputation quality varies by the size of the overlap of the intersection of genotyped SNPs across arrays. Finally, we examined the conditions under which using public controls adds sufficiently to a study’s power that the additional study complexity and administrative work to obtain public controls is worth the effort, considering the balance of sample size and imputation accuracy.
