To investigate whether the association between CPD and the CYP2A6 locus can be explained completely by the observed CNP, we carried out tests of association for all CNPs and SNPs spanning the CYP2A6 locus conditional upon this CNP (Figure 2b). Surprisingly, we found that a SNP (rs11878604; ) 30 kb downstream of the CYP2A6 gene was a secondary marker, which explains the additional effect of CYP2A6 locus on CPD. Further conditioning on rs11878604 after conditioning on the CNP (rs8102683) did not show any obvious signal of association in the region (Figure 2c), suggesting that the major effect of the CYP2A6 locus on CPD can be explained by the combination of the CNP and rs11878604.
