It should be noted that the observed ZNF804A allelic imbalance is small; therefore it's crucial to use a large sample size. However, since ASE assays can only be performed on heterozygous subjects for a coding SNP, the sample sizes in published allelic imbalance studies using human brain are typically modest (Bray et al., 2003; Fukuda et al., 2006; Sun et al., 2010). Because small sample sizes reduce the statistical power, and since no effect of diagnosis on rs12476147 ASE was observed, we analyzed the ASE on the rs12476147 heterozygotes from four diagnostic groups merged together, thus reaching a sample size of 46 subjects. Furthermore, the majority of the approaches that have been used to detect ASE are end-point readings of PCR, and therefore they may not be accurate enough in quantifying the ratio between the two alleles. In order to have a higher sensitivity, in our ASE assay we used real time PCR with Taqman probes to distinguish the two alleles (Chen et al., 2008).
