Fig. 3

Performance of MIENTURNET in detecting miRNA activity. ROC curves resulting by the MIENTURNET application on simulated data by considering both predicted interactions from TargetScan (blue curve) and validated interactions from miRTarBase (orange curve). The data simulated the activity of 10 miRNAs at different levels (α∈ [0.3-1] with 0.05 steps). For each level of α, we computed the true positive rate (i.e. sensitivity) placed on Y-axis, and the false positive rate (i.e. 1 - specificity) placed on X-axis. Sensitivity is the rate of truly active miRNAs identified by MIENTURNET on the total number of active miRNAs; specificity is the rate of truly inactive miRNAs identified by MIENTURNET on the total number of inactive miRNAs. We run MIENTURNET under default parameters. Diagonal grey line represents the line of no-discrimination