Skip to main content

Table 2 Results of different sets of features performed on both single and two-layer models

From: Improving the performance of β-turn prediction using predicted shape strings and a two-layer support vector machine model

Features

Predictor

Q total

MCC

Q observed

Q predicted

PSS (PHD, Jpred, PROTEUS)

PSSMs

Single

78.3

0.52

79.7

54.5

 

Two-layer

81.1

0.51

64.8

61.7

PSS (PROTEUS)

PSSMs

Single

80.8

0.58

84.1

58.0

 

Two-layer

82.7

0.55

69.5

64.2

PSS (PROTEUS)

PSSMs

Predicted Shape Strings

Single

87.3

0.69

86.5

69.8

 

Two-layer

87.2

0.66

75.9

73.8

  1. Note: PSS refers to predicted secondary structure.