Table 2 shows smoking-status associations for the 25 top SNPs from the study by Beirut et al.[13], comparing cases and controls in the present study. As noted earlier, we decided to first test rs16969968 to avoid the need for multiple testing corrections, and then test the other 24 SNPs. Although the other associations did not achieve significance after correction for multiple testing [β = 0.40 to detect an odds ratio (OR) of 1.5 for a SNP with a minor allele frequency (MAF) of 25% at α = 0.05 in the combined, race-adjusted sample], P-values for all the SNPs tested are listed in Table 2 as a guide for further research. The key finding is that, after adjustment for gender and race, rs16969968 was associated significantly with smoking status in Caucasians (OR = 1.51, P = 0.01) and in Caucasians plus African Americans (combined sample OR = 1.48, P = 0.01). Although Beirut et al. presented evidence suggesting that rs16969968 ‘A’ (smoking-risk) alleles act recessively, significant association was based on genetic effects modeled as the additive effect of minor alleles in the present study.
