Fig. 6: Drug candidate identification and drug repurposing for breast cancer based on ColdstartCPI’s prediction. | Nature Communications

Fig. 6: Drug candidate identification and drug repurposing for breast cancer based on ColdstartCPI’s prediction.

From: ColdstartCPI: Induced-fit theory-guided DTI predictive model with improved generalization performance

Fig. 6

a The distribution of pocket docking affinities of the top 100 compound candidates with Receptor tyrosine-protein kinase erbB-2 (number of data points n = 98, 23, 75, 98 in each group; center line, median; box limits, upper and lower quartiles; whiskers, maximum and minimum values; white circles, mean values; dots, outliers). b The docking pose and non-covalent interactions of DB00878 with ERBB2 (UniProt ID: P04626, PDB ID: 3RCD). c The distribution of blind docking affinities of the top 100 compound candidates with ERBB2 (number of data points n = 98, 23, 75, 98 in each group). d The docking pose and non-covalent interactions of CNP0266780 with ERBB2. e The distribution of docking affinities of the top 50 protein candidates with Paclitaxel (number of data points n = 50, 39, 11, 50 in each group). f The docking pose and non-covalent interactions of Paclitaxel (DrugBank ID: DB01229) with Afamin (UniProt ID: P43652, PDB ID: 5OKL). In b, d, and f, the legends show the types of protein-ligand interactions, which have been introduced in detail in Supplementary Note 4. Source data are provided as a Source Data file.

Back to article page