TABLE 1.

Performance Comparisons of DeepT3_4 with other three methods on the independent test dataset 2.

Model	TP	TN	FN	FP	ACC	F-value	Recall	PRE	MCC
					(%)	(%)	(%)	(%)
Bastion3	108	18	0	12	91.3	94.7	100.0	90.0	0.735
Bastion4	83	29	25	1	81.2	86.5	76.9	98.8	0.621
CNN-T4SE	59	29	49	1	63.8	70.2	54.6	98.3	0.427
DeepT3_4	101	26	7	4	92.0	94.8	93.5	96.2	0.775