Use of multivariate GWAS can be recommended even when genetic correlations between traits are expected to be weak. Indeed, even when only one of the traits was associated with the QTL and thus in the absence of genetic correlation and pleiotropy, MV analyses resulted in higher power than UV analyses. This was described before by Liu et al. for bivariate analyses and is due to the differences in the penalty for multiple testing [9]. Note that this penalty is commonly not applied in multiple UV analyses of real data.
