Table 6.

Comparison of the spEnhancer model performances using different dimensions of word vectors on the independent test dataset. The first column gives the binary classification problem. The column “WV” is the dimension of the word vector. The other five columns give the prediction performances Acc, Sn, Sp, MCC, and AUC.

	Word Vector Dimension	Acc	Sn	Sp	MCC	AUC
enhancers vs. non-enhancers	12	0.7085	0.8550	0.5606	0.4943	0.8177
	24	0.6658	0.9150	0.4141	0.4655	0.8094
	48	0.7538	0.8150	0.6919	0.5408	0.8167
	96	0.7060	0.8650	0.5455	0.4971	0.7359
	192	0.7186	0.7650	0.6717	0.4580	0.8078
	394	0.7337	0.8400	0.6263	0.5253	0.8172
	768 (this study)	0.7725	0.8300	0.7150	0.5793	0.8235
	1536	0.7764	0.7550	0.7980	0.5408	0.8281
strong enhancers vs. weak enhancers	12	0.5550	0.4300	0.6800	0.0984	0.6351
	24	0.5000	1.0000	0.0000	0.0000	0.6324
	48	0.6000	0.7100	0.4900	0.2265	0.6342
	96	0.6250	0.8000	0.4500	0.3101	0.6275
	192	0.5700	0.6700	0.4700	0.1565	0.6279
	394	0.5950	0.7100	0.4800	0.2165	0.5987
	768 (this study)	0.6200	0.9100	0.3300	0.3703	0.6253
	1536	0.5800	0.8700	0.2900	0.2469	0.5972