As additional validation of MiXeR results, we tested the hypothesis that MiXeR- and LAVA-derived measures of mixed effect directions were correlated, we calculated the Pearson correlation coefficient for 1) the proportion of shared “causal” variants with concordant effects (i.e causal variants affecting both traits in the same direction) estimated using MiXeR and 2) the proportion of significantly correlated genetic loci with positive correlation estimated using LAVA91(see Supplementary Figure S9-3).
