Skip to main content
Fig. 4 | BMC Bioinformatics

Fig. 4

From: HGDTI: predicting drug–target interaction by using information aggregation based on heterogeneous graph neural network

Fig. 4

Comparison results of HGDTI with other state-of-the-art models in several exploratory experiments in terms of the AUPR scores. a A 10-fold cross-validation test in which the ratio between positive and negative samples is set to 1 : 10. b–e Ten-fold cross-validation with positive: negative ratios \(=1 : 10\) on several scenarios of removing redundancy in data. b Remove DTIs with similar drugs and proteins. c Remove DTIs with drugs sharing similar drug interactions. d Remove DTIs with drugs sharing similar side-effects. e Remove DTIs with drugs sharing similar disease. f Non-unique train set and unique test set. All results are summarized over 10 trials and expressed as mean ± standard deviation

Back to article page

BMC Bioinformatics

ISSN: 1471-2105

Contact us