genetic variation [65] among self-identified White participants using 45 ancestry informative markers genotyped across all individuals. Genotypes were imputed with IMPUTE v2.1.2 [66] using 1000 Genomes CEU (Utah residents with ancestry from northern and western Europe) August 2010 haplotype data at CHRNB2 and chr15q25.1 (chr1:154476304-154616304 and chr15:78747906-79045112 [NCBI build 37], respectively). Imputed dosage was converted to genotypes with a 0.90 dosage probability cutoff using GTOOL v0.6.5. (http://www.well.ox.ac.uk/~cfreeman/software/gwas/gtool.html). rs2072661 and rs1051730 genotype data was extracted from GoldenGate genotyping data, and rs588765 genotype data was imputed for all RCTs. rs578776 genotype data was extracted from GoldenGate genotyping data for RCTs 3A and 3B, and imputed for the remaining RCTs. Among the expected 10,532 genotypes from four SNPs at 2,633 individuals tested for association, the overall missing genotype rate was 1.3%, while 57.0% and 41.6% were extracted from GoldenGate genotyping data or imputed, respectively. 97.7% of rs588765 and 98.8% of rs578776 imputed genotype dosage probabilities were within 10% of modal values. nAChR SNP minor allele frequencies did not differ significantly across the 26 Arms. We evaluated rs2072661 and rs1051730 genotype distributions by randomization arm and observed two arm-by-SNP strata with Hardy-Weinberg equilibrium p-values<0.05, versus 2.5 expected by chance (See Table, Supplemental Digital Content
