Skip to main content
Fig. 8 | Journal of Cheminformatics

Fig. 8

From: Exploration and augmentation of pharmacological space via adversarial auto-encoder model for facilitating kinase-centric drug development

Fig. 8

a Phylogenetic tree to display the performance of EPA on kinases from the Metz’s set. Circles represent the kinases included in training set. Squares represent the kinases absent from training set. b ROC analysis of EPA and ENB predictions to DUD-E dataset. c Distribution of BEDROC (alpha = 80.5) covering 26 kinase targets in DUD-E for EPA and docking programs. Statistical significance of the difference between the performance of EPA and other methods was determined with paired t-test. ns: p > 0.05; *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001

Back to article page