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 7 Performance comparison of DTIOG variants on the ENZ 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	0.950	0.992	0.987	0.942	0.948
DTIOG-DT	0.963	0.994	0.990	0.957	0.961
DTIOG-MLP	0.701	0.928	0.924	0.590	0.621
DTIOG-KNeighbors	0.968	0.995	0.968	0.870	0.879
DTIOG-BaggingClassifier	0.980	0.997	0.978	0.906	0.914
DTIOG-GradientBoosting	0.455	0.808	0.898	0.384	0.402
DTIOG-GaussianNB	0.496	0.532	0.237	0.047	0.214
DTIOG-SGD	0.566	0.881	0.902	0.411	0.425
DTIOG-ETC	0.956	0.993	0.988	0.949	0.954
Euclidean distance
DTIOG-RF	0.958	0.994	0.989	0.952	0.956
DTIOG-DT	0.971	0.996	0.993	0.967	0.970
DTIOG-MLP	0.944	0.991	0.968	0.844	0.861
DTIOG-KNeighbors	0.972	0.996	0.971	0.882	0.891
DTIOG-BaggingClassifier	0.978	0.992	0.971	0.898	0.906
DTIOG-GradientBoosting	0.644	0.908	0.918	0.548	0.578
DTIOG-GaussianNB	0.425	0.711	0.494	0.158	0.264
DTIOG-SGD	0.652	0.920	0.916	0.533	0.562
DTIOG-ETC	0.963	0.994	0.990	0.957	0.962
Manhattan distance
DTIOG-RF	0.959	0.994	0.989	0.953	0.958
DTIOG-DT	0.973	0.996	0.993	0.969	0.972
DTIOG-MLP	0.964	0.995	0.975	0.882	0.893
DTIOG-KNeighbors	0.964	0.995	0.966	0.864	0.873
DTIOG-BaggingClassifier	0.976	0.997	0.973	0.889	0.898
DTIOG-GradientBoosting	0.380	0.698	0.895	0.311	0.278
DTIOG-GaussianNB	0.517	0.796	0.831	0.419	0.483
DTIOG-SGD	0.654	0.923	0.919	0.564	0.599
DTIOG-ETC	0.961	0.994	0.990	0.956	0.960
Jaccard similarity
DTIOG-RF	0.943	0.991	0.985	0.933	0.939
DTIOG-DT	0.962	0.994	0.990	0.956	0.961
DTIOG-MLP	0.701	0.903	0.925	0.574	0.586
DTIOG-KNeighbors	0.969	0.995	0.970	0.877	0.886
DTIOG-BaggingClassifier	0.980	0.997	0.977	0.902	0.910
DTIOG-GradientBoosting	0.585	0.869	0.910	0.473	0.489
DTIOG-GaussianNB	0.490	0.539	0.254	0.045	0.213
DTIOG-SGD	0.579	0.872	0.906	0.434	0.440
DTIOG-ETC	0.954	0.993	0.988	0.946	0.951
Pearson correlation coefficient
DTIOG-RF	0.950	0.992	0.987	0.942	0.947
DTIOG-DT	0.962	0.994	0.990	0.956	0.961
DTIOG-MLP	0.709	0.921	0.925	0.587	0.612
DTIOG-KNeighbors	0.967	0.995	0.968	0.869	0.879
DTIOG-BaggingClassifier	0.980	0.997	0.978	0.904	0.912
DTIOG-GradientBoosting	0.363	0.708	0.893	0.319	0.320
DTIOG-GaussianNB	0.493	0.528	0.235	0.038	0.211
DTIOG-SGD	0.567	0.880	0.901	0.422	0.446
DTIOG-ETC	0.954	0.993	0.988	0.947	0.952

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