Table 2.

AUC performance metric of DeepD2V with other deep learning model.

Cell Line	TF	DeepBind	DanQ	WSCNNLSTM	DeepD2V	Cell Line	TF	DeepBind	DanQ	WSCNNLSTM	DeepD2V
Gm12878	Batf	0.887	0.904	0.906	0.939	H1hesc	Rad	0.973	0.978	0.981	0.984
Gm12878	Bcl1	0.823	0.831	0.844	0.907	H1hesc	Sin3	0.894	0.898	0.912	0.927
Gm12878	Bcl3	0.858	0.884	0.892	0.927	H1hesc	Sp1	0.874	0.874	0.885	0.930
Gm12878	Bclaf	0.843	0.852	0.866	0.898	H1hesc	Srf	0.949	0.956	0.964	0.966
Gm12878	Ebf	0.861	0.877	0.882	0.910	H1hesc	Taf1	0.908	0.911	0.912	0.927
Gm12878	Egr1	0.934	0.942	0.949	0.960	H1hesc	Tcf12	0.850	0.863	0.874	0.922
Gm12878	Elf1	0.905	0.914	0.912	0.927	H1hesc	Usf1	0.973	0.978	0.981	0.984
Gm12878	Ets1	0.912	0.868	0.929	0.951	H1hesc	Yy1	0.923	0.929	0.939	0.955
Gm12878	Irf4	0.833	0.815	0.847	0.891	K562	Atf3	0.931	0.945	0.952	0.954
Gm12878	Mef2a	0.852	0.837	0.876	0.927	K562	E2f6	0.935	0.943	0.945	0.958
Gm12878	Nrsf	0.899	0.905	0.915	0.937	K562	Egr1	0.947	0.954	0.960	0.967
Gm12878	Pax5c20	0.850	0.861	0.866	0.905	K562	Elf1	0.943	0.941	0.947	0.952
Gm12878	Pax5n19	0.845	0.845	0.872	0.928	K562	Ets1	0.0.883	0.893	0.891	0.918
Gm12878	Pbx3	0.855	0.879	0.879	0.931	K562	Fosl1	0.935	0.945	0.949	0.960
Gm12878	Pou2	0.819	0.835	0.843	0.894	K562	Gabp	0.932	0.927	0.943	0.961
Gm12878	Pu1	0.949	0.962	0.967	0.977	K562	Gata2	0.827	0.844	0.847	0.885
Gm12878	Rad21	0.978	0.985	0.988	0.988	K562	Hey1	0.875	0.879	0.875	0.891
Gm12878	Sp1	0.800	0.810	0.821	0.883	K562	Max	0.905	0.914	0.926	0.944
Gm12878	Srf	0.883	0.890	0.922	0.949	K562	Nrsf	0.880	0.883	0.901	0.943
Gm12878	Taf1	0.887	0.904	0.906	0.939	K562	Pu1	0.971	0.981	0.983	0.987
Gm12878	Tcf12	0.871	0.879	0.890	0.917	K562	Rad21	0.982	0.989	0.991	0.992
Gm12878	Usf1	0.918	0.948	0.953	0.961	K562	Srf	0.878	0.855	0.898	0.939
Gm12878	Yy1	0.888	0.891	0.901	0.927	K562	Taf1	0.898	0.909	0.909	0.925
H1hesc	Gabp	0.905	0.913	0.916	0.937	K562	Usf1	0.973	0.978	0.981	0.984
H1hesc	Nrsf	0.945	0.950	0.959	0.969	K562	Yy1	0.912	0.914	0.920	0.940
							Average ROC AUC	0.899	0.905	0.915	0.939