Advancing drug–target interaction prediction: a comprehensive graph-based approach integrating knowledge graph embedding and ProtBert pretraining

Djeddi, Warith Eddine; Hermi, Khalil; Ben Yahia, Sadok; Diallo, Gayo

doi:10.1186/s12859-023-05593-6

Table 5 Performance comparison of DTIOG variants on the GPCR dataset using KGE

From: Advancing drug–target interaction prediction: a comprehensive graph-based approach integrating knowledge graph embedding and ProtBert pretraining

Approaches	AUPR	AUC	ACC	MCC	F1
Cosine similarity
DTIOG-RF	0.951	0.980	0.971	0.931	0.949
DTIOG-DT	0.968	0.987	0.981	0.955	0.967
DTIOG-MLP	0.970	0.992	0.977	0.944	0.959
DTIOG-KNeighbors	0.949	0.979	0.899	0.792	0.843
DTIOG-BaggingClassifier	0.951	0.984	0.896	0.787	0.839
DTIOG-GradientBoosting	0.755	0.908	0.855	0.625	0.720
DTIOG-GaussianNB	0.655	0.784	0.716	0.405	0.586
DTIOG-SGD	0.617	0.822	0.775	0.438	0.574
DTIOG-ETC	0.962	0.984	0.977	0.947	0.960
Euclidean distance
DTIOG-RF	0.956	0.982	0.974	0.938	0.954
DTIOG-DT	0.967	0.987	0.981	0.954	0.966
DTIOG-MLP	0.960	0.990	0.979	0.950	0.963
DTIOG-KNeighbors	0.946	0.977	0.901	0.797	0.846
DTIOG-BaggingClassifier	0.947	0.983	0.898	0.790	0.841
DTIOG-GradientBoosting	0.736	0.888	0.847	0.601	0.700
DTIOG-GaussianNB	0.653	0.778	0.709	0.388	0.575
DTIOG-SGD	0.589	0.811	0.765	0.417	0.555
DTIOG-ETC	0.964	0.985	0.979	0.949	0.962
Manhattan distance
DTIOG-RF	0.953	0.980	0.972	0.933	0.951
DTIOG-DT	0.966	0.986	0.980	0.952	0.965
DTIOG-MLP	0.969	0.992	0.980	0.952	0.964
DTIOG-KNeighbors	0.963	0.985	0.908	0.808	0.855
DTIOG-BaggingClassifier	0.958	0.986	0.898	0.790	0.841
DTIOG-GradientBoosting	0.762	0.909	0.862	0.644	0.736
DTIOG-GaussianNB	0.646	0.781	0.719	0.418	0.594
DTIOG-SGD	0.599	0.817	0.777	0.435	0.570
DTIOG-ETC	0.966	0.986	0.980	0.952	0.965
Jaccard similarity
DTIOG-RF	0.952	0.980	0.971	0.932	0.950
DTIOG-DT	0.966	0.986	0.980	0.953	0.965
DTIOG-MLP	0.936	0.982	0.960	0.906	0.931
DTIOG-KNeighbors	0.957	0.983	0.907	0.801	0.844
DTIOG-BaggingClassifier	0.971	0.990	0.911	0.813	0.859
DTIOG-GradientBoosting	0.779	0.920	0.868	0.660	0.748
DTIOG-GaussianNB	0.654	0.782	0.726	0.423	0.597
DTIOG-SGD	0.627	0.830	0.783	0.443	0.553
DTIOG-ETC	0.961	0.984	0.977	0.945	0.959
Pearson correlation coefficient
DTIOG-RF	0.952	0.980	0.971	0.933	0.950
DTIOG-DT	0.965	0.986	0.980	0.951	0.964
DTIOG-MLP	0.965	0.991	0.980	0.951	0.964
DTIOG-KNeighbors	0.958	0.983	0.902	0.797	0.847
DTIOG-BaggingClassifier	0.957	0.986	0.892	0.780	0.834
DTIOG-GradientBoosting	0.754	0.907	0.856	0.628	0.723
DTIOG-GaussianNB	0.648	0.783	0.725	0.414	0.592
DTIOG-SGD	0.597	0.815	0.775	0.424	0.556
DTIOG-ETC	0.962	0.985	0.978	0.947	0.961

The best results are indicated in bold concerning each row

Back to article page

ISSN: 1471-2105

Contact us

General enquiries: journalsubmissions@springernature.com

BMC Bioinformatics

Contact us