Skip to main content

Table 4 Prediction performance on different sliding windows for encoding input vectors on the CASP8 data set.

From: LigandRFs: random forest ensemble to identify ligand-binding residues from sequence information alone

 

5

7

9

11

17

27

37

47

57

All

Sen(%)

48.00

51.06

63.30

51.43

53.74

45.27

49.24

58.53

51.25

52.11

Spe(%)

96.86

96.27

93.15

96.10

95.02

97.93

96.35

92.89

95.52

97.11

Acc(%)

93.93

93.46

90.97

93.31

92.37

94.65

93.41

90.59

92.73

94.19

MCC

0.42

0.43

0.42

0.42

0.39

0.43

0.39

0.37

0.38

0.44

Prec(%)

48.95

48.80

39.94

46.42

38.04

53.58

43.04

33.94

39.75

50.49

F1(%)

40.97

42.02

41.17

41.00

38.66

41.16

38.97

37.17

38.54

43.02

  1. The ensemble of all sliding windows is shown at the last column of the table.
  2. The italic numbers denote the best performance among different sliding windows on the measure of MCC, while the italic number in the last column is for the combination of all sliding windows.
Back to article page

BMC Bioinformatics

ISSN: 1471-2105

Contact us