Skip to main content

Advertisement

Table 1 Performance of structure-based classifiers on main dataset

From: HemeBIND: a novel method for heme binding residue prediction by combining structural and sequence information

Feature Recall (%) Precision (%) Accuracy (%) F1-score (%) MCC
DPX 73.28 22.81 62.97 34.77 0.239
RASA 70.52 28.64 72.36 40.70 0.313
CX 77.25 28.21 70.47 41.29 0.330
PSSM 74.55 29.88 72.98 42.64 0.342
RASA+DPX 76.69 27.82 70.03 40.79 0.323
PSSM+CX 75.66 31.11 74.10 44.04 0.361
PSSM+RASA 75.54 31.24 74.24 44.15 0.362
DPX+CX 80.56 30.40 72.53 44.11 0.370
PSSM+DPX 76.64 33.19 75.98 46.25 0.388
RASA+CX 75.97 33.68 76.64 46.66 0.391
PSSM+RASA+CX 76.11 31.90 74.80 44.89 0.371
RASA+DPX+CX 80.61 32.26 74.62 46.07 0.392
PSSM+RASA+DPX 76.90 33.87 76.54 46.94 0.396
PSSM+DPX+CX 78.81 33.64 76.15 47.10 0.401
PSSM+RASA+DPX+CX 79.08 34.07 76.49 47.56 0.407