Let ’s call it the effect allele: a suggestion for GWAS naming conventions

In recent years, the amount of publicly available summary data from genome-wide association studies (GWAS) has rapidly increased. So too has the number of researchers accessing and utilizing these data. These summary data can be used in many subsequent analyses, including Mendelian randomization, genetic correlation and polygenic score analysis. It is therefore vital that we can ensure consistency across these datasets to minimize the risk of analytical mistakes due to user error. One such inconsistency is the naming of the effect allele in these datasets. This is of particular concern given the increasing availability of automated software packages for downstream analyses (e.g. MR-Base,1 LDpred,2 LD Hub3) which make complex analyses easy to run, even with little understanding of the underlying processes. Data harmonization is a crucial step for many of these analyses. Guidelines for harmonizing datasets to use in two-sample Mendelian randomization have been suggested by Hartwiget al.4 In order to effectively implement these steps, it is critical to know which is the effect allele. This starting point is assumed to be known in the Hartwiget al. paper, however, in many datasets this is unclear and could lead to substantial errors. This issue could be easily avoided if consistent naming conventions were used.

Source: International Journal of Epidemiology - September 2, 2020 Category: Epidemiology Source Type: research