Early in drug discovery the IC50 of compounds blocking the human ether-a-go-go-related gene (hERG)-encoded channel is used to assess drug-induced QT interval prolongation risk. We do this using the Caucasian wild-type (WT) variant & assume its pharmacology is the same as that of variants that exist in patient populations. Our objective was to test this assumption by defining the pharmacology of 9 hERG single nucleotide polymorphisms (SNPs) largely selected from Ackerman et al. (2003): R176W, R181Q, Del187-189, P347S, K897T, A915V, P917L, R1047L & A1116V. Each SNP was expressed in Tet-On CHO K1 cells. Tail current amplitudes were measured using automated electrophysiology (IonWorks (IW)) to estimate the potency of 48 hERG blockers, mostly those reviewed by Redfern et al. (2003). In phase 1 of the work each WT-SNP comparison was made on the same day but not in the same IW test-plate. IC50s were estimated by fitting to an 8-point, non-cumulative concentration-effect curve made up of 32-48 data points & expressed as a mean IC50; lower, upper 95% confidence limit (CL)). This gave 432 WT-SNP IC50 comparisons (48 compounds X 9 SNPs) expressed as △IC50 values. These values ranged from 4-fold (potency at SNP 〉WT) to 3-fold in the opposite direction. For 77 compound-SNP combinations the WT upper or lower CL did not overlap with the SNP’s. In phase 2 each of these 77 cases was re-examined in the same IW test-plate to minimise variability. Using this design, 62 no longer yielded IC50 values with non-overlapping CLs. For 7 of the remaining 15 cases there were again non-overlapping CLs but this time in the opposite direction. Thus for only 8 compound-SNP combinations were there IC50s with non-overlapping CLs in the same direction as for the first phase (△IC50 values were never 〉2-fold). The absence of large △IC50 values was unrelated to the IW-based assay system, since using a hERG drug-binding site mutant (Y652A) propafenone was 6-fold less potent (IC50 3.84; 3.46, 4.26 μM) than WT (IC50 0.65; 0.59, 0.75 μM), the same fold-shift reported by Witchel et al. (2004) using conventional electrophysiology. In summary, potencies defined using the Caucasian WT sequence, at least in this in vitro system, are representative of potencies for these relatively common SNPs. We cannot completely exclude the possibility that there are small differences in pharmacology for a few compound-SNP combinations, but such differences cannot confidently be distinguished from assay variability. However, there remains the possibility that these SNPs influence the pharmacology or other hERG channel characteristics in vivo.