Fig. 7From: Synergy conformal prediction applied to large-scale bioactivity datasets and in federated learningDistribution of efficiency for the individual models compared to the federated for Set 2. RDKit descriptors on top row and FP bottom, active left and inactive right. Statistically significant differences are indicated (Wilcoxon signed-rank test with Bonferroni correction for multiple testing, 0.05 significance level)Back to article page