Skip to main content

Table 6 Comparison of DEBT with other turn/non-turn prediction methods on three different datasets.

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

Dataset β-turn predictor MCC Q total (%) Q pred (%) Q obs (%)
GR426 DEBT 0.48 79.2 54.8 70.1
  Zheng and Kurgan [34] 0.47 80.9 62.7 55.6
  Hu and Li [37] 0.47 79.8 55.6 68.9
  Zhang et al. [35] 0.45 77.3 53.1 67.0
  BTSVM [36] 0.45 78.7 56.0 62.0
  MOLEBRNN [32] 0.45 77.9 53.9 66.0
  BETAPRED2 [31] 0.43 75.5 49.8 72.3
  COUDES [28] 0.42 74.8 48.8 69.9
  Kim [38] 0.40 75.0 46.5 66.7
  BTPRED [30] 0.35 74.4 48.3 57.3
FA547 DEBT 0.49 80.0 55.9 68.7
  Zheng and Kurgan [34] 0.45 80.5 61.6 54.2
  COUDES [28] 0.42 74.6 48.7 70.4
  Hu and Li [37] 0.43 76.6 47.6 70.2
FA823 DEBT 0.48 80.9 55.9 66.1
  Zheng and Kurgan [34] 0.45 80.6 60.8 54.6
  COUDES [28] 0.41 74.2 47.5 69.6
  Hu and Li [37] 0.45 76.8 53.0 72.3
  1. The methods are sorted by their reported MCC score. DEBT achieves the highest value on all datasets.