The influence of the strength of residual correlation, i.e. the relative amount of shared genetics, and sign of genetic correlation, i.e. difference in sign of QTL effect, on power varied across the different methods. For MV-PLINK, MV-SNPTEST, MultiPhen and MV-BIMBAM, higher power was observed with increasing residual correlation in case of a single QTL trait and when two or all three traits were associated with the QTL with a negative genetic correlation. The latter is due to the resulting opposite sign of genetic and residual correlation. Indeed, we observed a similar increase in power when simulating a positive genetic correlation and negative residual correlation (data not shown). This effect has been described before for these and other methods [7], [9], [12], [14] and was demonstrated analytically by Evans for bivariate linkage analysis [26]. In contrast, when residual and genetic correlation were in the same direction, power of these four methods decreased with increasing residual correlation between the traits, which also corroborates previous findings [7], [12]. PCHAT and TATES were relatively independent of the underlying (genetic) correlations of the traits. For
