A learning-based method for drug-target interaction prediction based on feature representation learning and deep neural network

Table 3 The AUROC, AUPR scores of sequentially strip out a network from the original heterogeneous network as new input data of DTI-CNN on drug-target interactions prediction task

Networks	AUROC	AUPR
Without drug-drug	0.9299	0.9370
Without drug-disease	0.9343	0.9416
Without drug side-effect	0.9344	0.9444
Without drug similarities	0.9345	0.9425
Without protein-protein	0.9418	0.9499
Without protein-disease	0.9364	0.9452
Without protein similarities	0.9327	0.9411

ISSN: 1471-2105