CAT-DTI: cross-attention and Transformer network with domain adaptation for drug-target interaction prediction

Zeng, Xiaoting; Chen, Weilin; Lei, Baiying

doi:10.1186/s12859-024-05753-2

BMC Bioinformatics

Table 2 Comparison results of CAT-DTI and baselines on three datasets

From: CAT-DTI: cross-attention and Transformer network with domain adaptation for drug-target interaction prediction

Datasets	Methods	AUROC	AUPRC	F1	Sensitivity	Specificity	Accuracy
BindingDB	SVM [10]	0.904 ± 0.000	0.865 ± 0.001	0.785 ± 0.000	0.776 ± 0.000	0.857 ± 0.002	0.824 ± 0.001
	RF [11]	0.942 ± 0.001	0.923 ± 0.001	0.844 ± 0.002	0.840 ± 0.002	0.893 ± 0.002	0.871 ± 0.001
	GraphDTA [23]	0.944 ± 0.004	0.923 ± 0.006	0.880 ± 0.005	0.858 ± 0.026	0.897 ± 0.014	0.874 ± 0.010
	TransformerCPI [29]	0.947 ± 0.003	0.932 ± 0.004	0.888 ± 0.005	0.886 ± 0.016	0.890 ± 0.008	0.888 ± 0.007
	MolTrans [30]	0.947 ± 0.004	0.927 ± 0.006	0.886 ± 0.005	0.877 ± 0.018	0.894 ± 0.014	0.884 ± 0.007
	DrugBAN [31]	0.961 ± 0.001	0.948 ± 0.001	0.903 ± 0.001	0.894 ± 0.011	0.908 ± 0.009	0.901 ± 0.003
	CAT-DTI	0.960 ± 0.001	0.947 ± 0.001	0.900 ± 0.001	0.884 ± 0.010	0.913 ± 0.009	0.896 ± 0.002
BioSNAP	SVM [10]	0.819 ± 0.045	0.839 ± 0.038	0.827 ± 0.053	0.665 ± 0.046	0.835 ± 0.054	0.750 ± 0.050
	RF [11]	0.857 ± 0.001	0.872 ± 0.001	0.787 ± 0.001	0.763 ± 0.002	0.823 ± 0.001	0.793 ± 0.001
	GraphDTA [23]	0.871 ± 0.001	0.870 ± 0.005	0.807 ± 0.005	0.761 ± 0.015	0.838 ± 0.011	0.800 ± 0.005
	TransformerCPI [29]	0.876 ± 0.004	0.881 ± 0.007	0.803 ± 0.006	0.768 ± 0.024	0.827 ± 0.012	0.797 ± 0.008
	MolTrans [30]	0.895 ± 0.006	0.899 ± 0.006	0.825 ± 0.007	0.791 ± 0.032	0.848 ± 0.014	0.820 ± 0.011
	DrugBAN [31]	0.902 ± 0.001	0.905 ± 0.002	0.838 ± 0.003	0.825 ± 0.014	0.847 ± 0.006	0.836 ± 0.004
	CAT-DTI	0.909 ± 0.002	0.907 ± 0.004	0.840 ± 0.004	0.816 ± 0.012	0.857 ± 0.006	0.836 ± 0.005
Human	SVM [10]	0.913 ± 0.000	0.905 ± 0.000	0.811 ± 0.000	0.782 ± 0.000	0.830 ± 0.000	0.838 ± 0.000
	RF [11]	0.939 ± 0.002	0.927 ± 0.001	0.848 ± 0.005	0.833 ± 0.006	0.893 ± 0.007	0.866 ± 0.006
	GraphDTA [23]	0.965 ± 0.003	0.955 ± 0.003	0.907 ± 0.008	0.912 ± 0.017	0.904 ± 0.016	0.908 ± 0.008
	TransformerCPI [29]	0.954 ± 0.002	0.941 ± 0.002	0.891 ± 0.005	0.831 ± 0.023	0.939 ± 0.018	0.879 ± 0.007
	MolTrans[30]	0.981 ± 0.002	0.976 ± 0.002	0.943 ± 0.005	0.949 ± 0.011	0.939 ± 0.017	0.941 ± 0.004
	DrugBAN [31]	0.981 ± 0.001	0.969 ± 0.005	0.940 ± 0.004	0.938 ± 0.010	0.941 ± 0.013	0.940 ± 0.003
	CAT-DTI	0.983 ± 0.001	0.976 ± 0.003	0.944 ± 0.002	0.929 ± 0.007	0.957 ± 0.008	0.942 ± 0.002

Bold values indicate the best results achieved by all these competitive methods

Back to article page

ISSN: 1471-2105

Contact us

General enquiries: journalsubmissions@springernature.com