Skip to main content

Table 5 Performance of DEBT for the prediction of β-turn location on three datasets.

From: Predicting β-turns and their types using predicted backbone dihedral angles and secondary structures

Dataset

MCC

Q total (%)

Q pred (%)

Q obs (%)

S

AUC

GR426

0.48 (0.43)

79.2 (78.6)

54.8 (53.9)

70.1 (61.6)

47.5 (43.2)

0.84 (0.83)

FA547

0.49 (0.44)

80.0 (79.2)

55.9 (54.5)

68.7 (60.5)

48.3 (43.6)

0.85 (0.83)

FA823

0.48 (0.42)

80.9 (79.9)

55.9 (54.1)

66.1 (56.5)

48.0 (42.3)

0.84 (0.82)

  1. In the parentheses are the predictions using PSSM-only input. There is significant improvement on all measures when the input vector is augmented by predicted dihedral angles and secondary structures.