GCMM: graph convolution network based on multimodal attention mechanism for drug repurposing

BMC Bioinformatics

Table 5 Performance of multi-source information

	Auc	Aupr	F1	ACC	Recall	Precision	Specificity
\(G^C + G^T + G^M + G^A\)	0.9013	0.9131	0.8160	0.8155	0.8167	0.8139	0.8105
\(G^C + G^M\)	0.8734	0.8891	0.7890	0.7951	0.7665	0.8129	0.8236
\(G^C + G^A\)	0.8685	0.8771	0.7918	0.7844	0.8198	0.7656	0.7490
\(G^T + G^M\)	0.8432	0.8444	0.7744	0.7452	0.8750	0.6946	0.6153
\(G^T + G^A\)	0.8657	0.8758	0.7913	0.7912	0.7917	0.7909	0.7907
\(G^C + G^T + G^M\)	0.8698	0.8781	0.7926	0.7776	0.8491	0.7426	0.7054
\(G^C + G^T + G^A\)	0.8651	0.8757	0.7900	0.7917	0.7839	0.7963	0.7994
\(G^C + G^M + G^A\)	0.8827	0.8928	0.8000	0.7955	0.8178	0.7829	0.7733
\(G^T + G^M + G^A\)	0.8800	0.8908	0.8051	0.8018	0.8188	0.7919	0.7849

ISSN: 1471-2105