focused on subjects ascertained as daily smokers to find out the relationship between CHRNA5-CHRNA3-CHRNB4 variants and onset of daily/habitual smoking behaviors, believing that this selected group can explain more genetic variance than smoking initiation or exposure. We found that the rs1996371 and correlated variants located upstream of CHRNB4 gene were associated with age at onset of daily smoking. Since our sample is multi-ethnic we also restricted our analyses to the largest subgroup (European Americans) to ensure that the association was not a false positive due to heterogeneity in our dataset. The association remained in this subgroup. The association between age at onset of daily smoking and rs1996371 and correlated SNPs also remained statistically significant after correcting for multiple testing proposed by Nyholt et al [25], (rs1996371, p = 0.0058). The daily smokers who had at least one copy of the minor allele of rs1996371 had a mean AAO of daily smoking 1 year younger than individuals without the minor allele. The non-synonymous SNP, rs16969968 was not associated with age at onset of daily smoking. We further wanted to explore whether daily smokers with these variants are at increased risk of becoming habitual smokers in future. Our Cox-proportional hazard analysis indeed
