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 8 Performance comparison of DTIOG variants on the IC dataset using ProtBERT

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	0.976	0.966	0.927	0.949
DTIOG-DT	0.965	0.983	0.976	0.948	0.963
DTIOG-MLP	0.926	0.971	0.917	0.809	0.869
DTIOG-KNeighbors	0.974	0.987	0.932	0.860	0.901
DTIOG-BaggingClassifier	0.981	0.993	0.941	0.878	0.913
DTIOG-GradientBoosting	0.776	0.877	0.823	0.569	0.686
DTIOG-GaussianNB	0.659	0.779	0.785	0.458	0.523
DTIOG-SGD	0.687	0.816	0.775	0.451	0.600
DTIOG-ETC	0.961	0.981	0.974	0.942	0.960
Euclidean distance
DTIOG-RF	0.956	0.978	0.970	0.934	0.954
DTIOG-DT	0.972	0.986	0.981	0.959	0.971
DTIOG-MLP	0.997	0.999	0.995	0.988	0.992
DTIOG-KNeighbors	0.971	0.986	0.971	0.936	0.955
DTIOG-BaggingClassifier	0.654	0.500	0.691	0.000	0.000
DTIOG-GradientBoosting	0.889	0.961	0.913	0.797	0.860
DTIOG-GaussianNB	0.713	0.817	0.784	0.492	0.647
DTIOG-SGD	0.745	0.877	0.816	0.566	0.696
DTIOG-ETC	0.966	0.983	0.977	0.949	0.964
Manhattan distance
DTIOG-RF	0.957	0.979	0.971	0.936	0.955
DTIOG-DT	0.973	0.987	0.982	0.961	0.972
DTIOG-MLP	0.997	0.999	0.995	0.989	0.992
DTIOG-KNeighbors	0.977	0.989	0.960	0.914	0.940
DTIOG-BaggingClassifier	0.981	0.993	0.965	0.924	0.947
DTIOG-GradientBoosting	0.809	0.916	0.859	0.664	0.762
DTIOG-GaussianNB	0.694	0.800	0.780	0.471	0.622
DTIOG-SGD	0.749	0.875	0.813	0.561	0.692
DTIOG-ETC	0.970	0.985	0.980	0.956	0.969
Jaccard similarity
DTIOG-RF	0.954	0.977	0.969	0.932	0.952
DTIOG-DT	0.969	0.985	0.979	0.954	0.968
DTIOG-MLP	0.988	0.995	0.976	0.945	0.962
DTIO-KNeighbors	0.979	0.990	0.948	0.889	0.922
DTIOGBaggingClassifier	0.991	0.996	0.960	0.913	0.939
DTIO-GradientBoosting	0.802	0.895	0.845	0.624	0.727
DTIO-GaussianNB	0.663	0.783	0.776	0.436	0.477
DTIO-SGD	0.722	0.846	0.800	0.518	0.648
DTIOG-ETC	0.965	0.983	0.976	0.948	0.963
Pearson correlation coefficient
DTIOG-RF	0.955	0.978	0.970	0.933	0.953
DTIOG-DT	0.968	0.985	0.979	0.953	0.967
DTIOG-MLP	0.992	0.997	0.991	0.980	0.986
DTIOG-KNeighbors	0.982	0.991	0.955	0.904	0.932
DTIOG-BaggingClassifier	0.991	0.996	0.966	0.926	0.948
DTIOG-GradientBoosting	0.806	0.900	0.850	0.639	0.741
DTIOG-GaussianNB	0.672	0.791	0.785	0.458	0.523
DTIOG-SGD	0.738	0.858	0.810	0.541	0.669
DTIOG-ETC	0.968	0.984	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