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 6 Performance comparison of DTIOG variants on the IC 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	0.960	0.980	0.973	0.941	0.958
DTIOG-DT	0.972	0.987	0.982	0.959	0.971
DTIOG-MLP	0.988	0.995	0.981	0.958	0.971
DTIOG-KNeighbors	0.987	0.994	0.955	0.904	0.932
DTIOG-BaggingClassifier	0.989	0.996	0.951	0.896	0.9272
DTIOG-GradientBoosting	0.834	0.926	0.865	0.681	0.776
DTIOG-GaussianNB	0.757	0.854	0.803	0.547	0.6914
DTIOG-SGD	0.767	0.895	0.828	0.597	0.715
DTIOG-ETC	0.970	0.985	0.980	0.955	0.969
Euclidean distance
DTIOG-RF	0.962	0.981	0.974	0.944	0.960
DTIOG-DT	0.977	0.989	0.985	0.967	0.977
DTIOG-MLP	0.962	0.984	0.952	0.887	0.920
DTIOG-KNeighbors	0.986	0.993	0.955	0.904	0.932
DTIOG-BaggingClassifier	0.988	0.995	0.953	0.900	0.929
DTIOG-GradientBoosting	0.831	0.927	0.872	0.697	0.787
DTIOG-GaussianNB	0.757	0.855	0.798	0.543	0.690
DTIOG-SGD	0.753	0.885	0.813	0.552	0.669
DTIOG-ETC	0.973	0.987	0.982	0.961	0.973
Manhattan distance
DTIOG-RF	0.962	0.982	0.975	0.944	0.961
DTIOG-DT	0.976	0.988	0.984	0.965	0.975
DTIOG-MLP	0.983	0.993	0.971	0.936	0.955
DTIOG-KNeighbors	0.985	0.993	0.957	0.908	0.935
DTIOG-BaggingClassifier	0.988	0.995	0.954	0.902	0.931
DTIOG-GradientBoosting	0.827	0.924	0.871	0.694	0.785
DTIOG-GaussianNB	0.771	0.863	0.801	0.555	0.699
DTIOG-SGD	0.752	0.883	0.815	0.560	0.683
DTIOG-ETC	0.973	0.987	0.982	0.960	0.972
Jaccard similarity
DTIOG-RF	0.958	0.979	0.972	0.938	0.956
DTIOG-DT	0.973	0.987	0.982	0.961	0.973
DTIOG-MLP	0.981	0.993	0.980	0.955	0.969
DTIOG-KNeighbors	0.983	0.992	0.940	0.874	0.911
DTIOG-BaggingClassifier	0.981	0.992	0.931	0.858	0.900
DTIOG-GradientBoosting	0.834	0.928	0.867	0.685	0.778
DTIOG-GaussianNB	0.754	0.848	0.809	0.555	0.693
DTIOG-SGD	0.779	0.899	0.828	0.592	0.705
DTIOG-ETC	0.967	0.984	0.978	0.951	0.966
Pearson correlation coefficient
DTIOG-RF	0.957	0.979	0.971	0.937	0.9560
DTIOG-DT	0.972	0.986	0.982	0.959	0.971
DTIOG-MLP	0.987	0.995	0.978	0.951	0.966
DTIOG-KNeighbors	0.984	0.992	0.945	0.883	0.918
DTIOG-BaggingClassifier	0.988	0.995	0.940	0.875	0.912
DTIOG-GradientBoosting	0.837	0.927	0.866	0.682	0.776
DTIOG-GaussianNB	0.757	0.849	0.799	0.542	0.689
DTIOG-SGD	0.770	0.894	0.824	0.601	0.724
DTIOG-ETC	0.968	0.985	0.979	0.953	0.967

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