Fig. 8From: Constraining PERMANOVA and LDM to within-set comparisons by projection improves the efficiency of analyses of matched sets of microbiome dataComparing the matched-pair design (solid lines) with the independent case-control design (dashed lines) over varying sample-level heterogeneity θ2. The effect size β was set to 0.1 (S1, power), 0.25 (S2, power), 0.8 (S1, sensitivity), and 0.6 (S2, sensitivity). All simulations shown here use between-set heterogeneity parameter θ1=0.02Back to article page