Table 4.

The comparison of our prediction quality with Yu's method by independent dataset test on the test set.

Type	Reference	T1SP	T2SP	T3SP	T4SP	T5SP	T7SP	Total
Number of sequences		25	29	28	22	35	33	172
The “one-to-one” algorithm
Correct hit		22	23	28	18	35	29	155
Sensitivity (%)		88.00	79.31	100.00	81.82	100.00	87.88	90.12
The “one-to-the-rest” algorithm
Correct hit	[23]	20	22	28	17	34	27	148
Sensitivity (%)	[23]	80.00	75.86	100.00	77.27	97.14	81.82	86.05
Correct hit	Our method	25	29	28	22	35	33	172
Sensitivity (%)	Our method	100.0	100.0	100.0	100.0	100.0	100.0	100.0