DNA was extracted from whole blood and EBV transformed cells and was aliquoted and stored frozen at −80°C until distributed to the genotyping labs. Details of the genotyping and quality control for the 226 SNPs have been previously described (Saccone et al., in press). Initial genotyping of the case-control sample was carried out by Perlegen Sciences using custom arrays as previously described by the Nicotine Single Nucleotide Polymorphism (NICSNP) study (Bierut et al., 2007; Saccone et al., 2007). Additional genotyping was then carried out by the Center for Inherited Disease Research using Illumina Golden Gate technology. SNPs in the CHRN genes were selected with these goals. First, SNPs with poor or marginal genotype call rates (<98%), or failed quality control measures described in Bierut et al. (2007) and Saccone et al. (2007) were re-genotyped. Second, r2 bin tag SNPs (Carlson et al., 2004) were selected to improve coverage of the CHRN genes. Third, because of the strong associations with case-control status in the CHRNA5/CHRNA3/CHRNB4 and CHRNB3/CHRNA6 gene clusters, we densely covered these regions for fine-mapping. All SNPs were required to
