Skip to main content

Table 1 Performance comparison between the three methods of catalytic residue prediction when evaluation was carried out by chain, family, superfamily and fold. Methods were evaluated on the same data set using 10-fold cross-validation. sn means the sensitivity when specificity is 0.95.

From: Structure-based kernels for the prediction of catalytic residues and their involvement in human inherited disease

 

FEATURE

GBT

Structure Kernel

 

AUC

sn

AUC

sn

AUC

sn

Fold

76.4

0.20

80.4

0.34

86.1

0.40

Superfamily

76.5

0.20

80.8

0.34

86.1

0.40

Family

76.7

0.21

80.7

0.34

86.8

0.42

Chain

76.7

0.21

81.1

0.34

87.3

0.45