Skip to main content
Fig. 4 | BMC Bioinformatics

Fig. 4

From: Utilizing random Forest QSAR models with optimized parameters for target identification and its application to target-fishing server

Fig. 4

ROC curves and the area under curves computed by internal cross validation. (Left) ROC curves for each target and overall ROC curve. Blue dotted line indicates ROC curve for random selection with AUC = 0.5. Red lines are ROC curves for each target and black line is overall ROC curve built using all the screening data throughout targets. (Right) Box plot of AUC values for targets. Red line indicates the median value of AUC, which is 0.97

Back to article page