Table 5.

Accuracy with which the Domain Boundaries are Identified by Various Prediction Methods

Method	Pc_c	Delta_b	Delta_e
baker	56.2	-1.2(0.3)	2.2(0.5)
chop	26.1	-2.9(1.0)	1.9(0.7)
chophomo	25.0	-2.6(1.0)	2.9(1.0)
distill	33.6	-1.5(0.6)	3.2(0.8)
domfold	38.0	-1.9(0.6)	2.9(0.7)
domssea	42.9	-1.7(0.6)	2.5(0.7)
dps	38.7	-2.2(0.8)	1.6(0.9)
foldpro	62.8	-1.3(0.4)	2.0(0.4)
hhpred1	43.3	-2.1(0.5)	2.6(0.5)
hhpred3	43.4	-2.1(0.5)	2.7(0.5)
maopus	54.2	-1.4(0.6)	3.0(0.8)
metadp	39.8	-1.3(0.7)	3.3(0.7)
NNput	30.8	-1.9(0.7)	2.4(0.8)
robetta	57.9	-1.0(0.3)	1.5(0.5)

The following data are shown: the percentage of domains that are correctly predicted (see text for details) PC_C, the average deviation between the real and the predicted beginning of the domain Delta_b, and the average difference between the real and the predicted end of the domain Delta_e (standard deviations of the mean in parentheses).