Skip to main content

Table 2 Prediction results of different ensemble models on the stratified tenfold CV

From: In-silico target prediction by ensemble chemogenomic model based on multi-scale information of chemical structures and protein sequences

Methods

Pair-split

Compound-split

ACC

SE

SP

AUC

ACC

SE

SP

AUC

Mean

84.83 ± 0.16

86.96 ± 0.10

82.44 ± 0.24

92.84

84.41 ± 0.03

86.5 ± 0.04

82.13 ± 0.04

92.41

Maximum

80.73 ± 0.19

94.53 ± 0.08

65.39 ± 0.32

92.46

79.97 ± 0.05

94.63 ± 0.05

63.71 ± 0.08

91.98

Stacked

83.80 ± 0.23

85.07 ± 0.20

82.37 ± 0.30

91.47

82.93 ± 0.08

84.33 ± 0.08

81.40 ± 0.11

91.47