GDCL-NcDA: identifying non-coding RNA-disease associations via contrastive learning between deep graph learning and deep matrix factorization

Table 5 Performance of GDCL-NcDA and its variants on miRNA-disease MHN

Methods	AUC	AUPRC	F1-score	Recall	Precision
GDCL-NcDA\(\_\)GCNII	0.8761	0.8810	0.8096	0.8508	0.7736
GDCL-NcDA\(\_\)GATGCNII	0.8838	0.8940	0.8173	0.8477	0.7906
GDCL-NcDA\(\_\)DMF	0.8556	0.8661	0.8096	0.8508	0.7736
GDCL-NcDA\(\_\)GCNII+DMF	0.9720	0.9628	0.9247	0.9153	0.9347
GDCL-NcDA\(\_\)GCNII+DMF+CL	0.9741	0.9783	0.9328	0.9382	0.9278
GDCL-NcDA	0.9761	0.9806	0.9394	0.9352	0.9439

ISSN: 1471-2164