We used genotype data of the 89 Japanese subjects from the 1000 Genomes Project (phase I; 2011-11-23) as a reference panel for the imputation study to perform fine mapping of the CYP2A6 locus in relation to CPD. We selected SNPs within a region of 1 cM across the most significant CNP (Figure S4). For SNPs within the commonly deleted region, the reference genotype data were updated to reflect the inferred copy number (Table S7). For individuals with zero copies of the CYP2A6 gene, genotypes within the region were represented as homozygote pairs of the deletion allele (referred to as “O”). For individuals with one copy of the CYP2A6 gene, genotypes were represented as heterozygous pairs of a normal SNP allele (referred to as “A” or “B”) and the deletion allele. Heterozygous genotypes of normal SNP alleles were represented as having an unknown genotype because we were unable to infer which of the two normal alleles were inherited by each individual (these unknown genotypes in the reference panel were also imputed during the imputation).
