Skip to main content
Fig. 3 | Journal of Cheminformatics

Fig. 3

From: Prediction of activity and selectivity profiles of human Carbonic Anhydrase inhibitors using machine learning classification models

Fig. 3

Building models to predict activity classes: representation of variations of TP, TN, FP, and FN percentages as group size (N), the algorithm used, and the investigated isoform vary for trained models in the testing phase. All models were built applying a Pearson Correlation Coefficient threshold of 0.95 to filter the most correlated features. The black line represents overall accuracy

Back to article page