Following publication of this article [1], we recently noticed that two subjects in this study were not assigned to the correct genetic groups. One of the AA homozygotes was incorrectly entered as a C allele carrier and we had the opposite situation for a second subject.

We have re-run the primary statistical analysis with the subjects assigned to their correct groups. The corrected analysis produced very similar results to our initial, published findings as we observed a significant treatment effect and a significant genotype x treatment interaction due to a higher degree of response in the AA homozygotes.

Correcting our dataset did result in very slight differences in the mean values for 40 k time. The changes are presented in Table 1.

Table 1 .