A fast and powerful W-test for pairwise epistasis testing.

Nucl. Acids Res. (08 July 2016) 44 (12): e115. doi:10.1093/nar/gkw347

The authors wish to make the following correction to their article:

In this paper the authors describe a novel method, the W-test, which provides an association test with data-set adaptive probability distributions that measures epistasis for both common and low frequency SNPs. The article compares W-test to three alternative methods, namely the logistic regression, Chi-squared test, and MDR. All methods are computed using existing R packages. The authors did not realize that the MDR package is different from the original version of MDR (1). The original version of MDR does not support the evaluation of exhaustive pair-wise for the one million replications reported in the study. The MDR simulation results reported in the article must therefore be removed from the article. The results of the W-test, logistic regression and Chi-square are not affected, and the conclusion of the article remains valid. A new manuscript that excludes the MDR analysis is available as supplementary material. The authors apologize to the readers for the inconvenience caused.

SUPPLEMENTARY DATA

Supplementary Data are available at NAR Online.