Accuracy and applications of sequencing and genotyping approaches for CYP2A6 and homologous genes

OBJECTIVES: The aims of this study were: (1) to evaluate the genotype calling of several approaches in a high homology region of chromosome 19 (including CYP2A6, CYP2A7, CYP2A13, CYP2B6), and (2) to use this data to investigate associations of two common 3’-UTR CYP2A6 variants, CYP2A6*1B and rs8192733, with CYP2A6 activity in vivo. METHODS: (1) Individuals (n=1704) of European and African ancestry were phenotyped for the nicotine metabolite ratio (NMR), an index of CYP2A6 activity. Individuals were also genotyped/sequenced using various approaches (deep amplicon exon sequencing, SNP array, genotype imputation, targeted capture sequencing). Amplicon exon sequencing, after realignment to a reference chromosome 19 with CYP2A7 masked, was used as the gold standard. Genotype calls from each method were compared within-individual to those from the gold standard for the exons of CYP2A6, CYP2A7 (exons 1 and 2), CYP2A13, and CYP2B6. Individual data was combined to identify genomic positions with high discordance. (2) Linear regression models were used to evaluate the association of CYP2A6*1B and rs8192733 genotypes (coded additively) with the log-transformed NMR (logNMR). RESULTS: (1) Overall, all approaches were ≤2.6% discordant with the gold standard, with discordant calls concentrated at relatively few genomic positions. Fifteen genomic positions were discordant with the gold standard in >10% of individuals, with 12 appearing in regions of perfect or near-perfect identity between homologous genes (e.g. CYP2A6 and CYP2A7). A subset of positions (6/15) showed discrepancies between study major allele frequencies and those reported by online databases, suggesting similar errors in online sources. (2) In the European-ancestry group (n=935), both the CYP2A6*1B genotype and the rs8192733 genotype were associated with logNMR (p=