To the editor

We thank Wang and colleagues for their interest in our article published in NeuroMolecular Medicine (Liu et al. 2013). In this article above, we evaluated the potential association of rs3851179 variant with Alzheimer’s disease (AD) in Asian population by a pooled analysis and a meta-analysis. The results show that rs3851179 variant is significantly associated with AD in Asian population by both methods. Here, Wang and colleagues discussed three issues.

First, Wang and colleagues described that an exhaustive search is necessary for meta-analysis. We searched the PubMed database to select all the possible studies in the original article (Liu et al. 2013). We accept this suggestion and will use multiple literature databases in further studies, including PubMed, AlzGene, Google Scholar and China Knowledge Resource Integrated Database (CNKI), as did in our recent studies (Liu et al. 2014; Shen et al. 2015; Li et al. 2015; Zhang et al. 2015).

Second, we thank Wang and colleagues again, as they pointed out a writing error about the significance level of Cochran’s Q test (0.01 and 0.05) in the “Evaluation of the Heterogeneity” section. We also accept the suggestion about the Cochran’s Q test significance level 0.1. In further, we will give an erratum to this article published in NeuroMolecular Medicine (Liu et al. 2013).

Third, we accept the suggestion that the Hardy–Weinberg equilibrium (HWE) test should be conducted only in controls but no in cases in case–control studies. We think it may not suitable for combining the data of China and Japan. In further study, we will calculate the p value for HWE, respectively.

In summary, we think that these three issues are important, and may provide supplementary information about our previous article (Liu et al. 2013).