Table 4.

Performance comparison of the independent test on the same independent test dataset

Layer	Method	ACC(%)	MCC	SN(%)	SP(%)	AUC(%)
First Layer (Enhancer Identification)	iEnhancer-2L [11]	73.00	0.460	71.00	75.00	80.62
	EnhancerPred [12]	74.00	0.480	73.50	74.50	80.13
	iEnhancer-EL [6]	74.75	0.496	71.00	78.50	81.73
	iEnhancer-ECNN [16]	76.90	0.537	78.50	75.20	83.20
	iEnhancer-XG [13]	75.75	0.515	74.00	77.50	-
	BERT-Enhancer [17]	75.60	0.514	80.00	71.20	-
	iEnhancer-EBLSTM [20]	77.20	0.534	75.50	79.50	83.50
	iEnhancer-RF [14]	$\underset{\_}{79.75}$	$\underset{\_}{0.595}$	78.50	$\underset{\_}{81.00}$	86.00
	iEnhancer-RD [18]	78.80	0.576	$\underset{\_}{81.00}$	76.50	84.40
	spEnhancer [21]	77.25	0.579	83.00	71.50	82.35
	iEnhancer-DCSA (Ours)	82.50	0.651	79.50	85.50	$\underset{\_}{85.58}$
Second Layer (Enhancer Classification)	iEnhancer-2L [11]	60.50	0.218	47.00	74.00	66.78
	EnhancerPred [12]	55.00	0.102	45.00	65.00	57.90
	iEnhancer-EL [6]	61.00	0.222	54.00	68.00	68.01
	iEnhancer-ECNN [16]	67.80	0.368	79.10	56.40	74.80
	iEnhancer-XG [13]	63.50	0.272	70.00	57.00	-
	BERT-Enhancer [17]	-	-	-	-	-
	iEnhancer-EBLSTM [20]	65.80	0.324	81.20	53.60	68.80
	iEnhancer-RF [14]	$\underset{\_}{85.00}$	$\underset{\_}{0.709}$	$\underset{\_}{93.00}$	$\underset{\_}{77.00}$	97.00
	iEnhancer-RD [18]	70.50	0.426	84.00	57.00	79.20
	spEnhancer [21]	62.00	0.370	91.00	33.00	62.53
	iEnhancer-DCSA (Ours)	91.50	0.837	98.00	85.00	$\underset{\_}{96.60}$

Note: ‘-’ indicates no result in the paper, and the best performance is highlighted in bold while the second-best performance is underlined