Twenty SNPs in CHRNA4 and CHRNB2 were originally genotyped for this study. Polymorphisms were selected on a number of criteria. SNPs were chosen based on the work of Conti and others (2008) who examined a number of SNPs in these genes and reported low linkage disequilibrium (LD) in their sample to test statistically independent loci (HapMap data were also examined, where available). Second, SNPs were selected based on previous evidence for association with smoking behaviors in other studies (Breitling et al. 2009; Conti et al. 2008; Ehringer et al. 2007; Ehringer et al. 2010; Feng et al. 2004b; Greenbaum et al. 2006; Han et al. 2011; Hoft et al. 2011; Hutchison et al. 2007; Li et al. 2005; Saccone et al. 2009; Saccone et al. 2010). Finally, only common SNPs (minor allele frequency of > 0.01 in dbSNP HapMap CEU population (http://www.ncbi.nlm.nih.gov/SNP/) or had a minor allele frequency of > 0.01 in the CADD sample in our previous work (Ehringer et al. 2007)) were included. Genomic DNA was extracted from buccal or blood cells and amplified with primer extension preamplification
