Table 4 The Performance of SVM model (Learning Parameter: g: 0.001 c: 4 j: 1) Using seven physiochemical properties.
From: Identification of ATP binding residues of a protein from its primary sequence
Threshold | Sensitivity | Specificity | Accuracy | MCC |
|---|---|---|---|---|
-0.9 | 93.68 | 18.71 | 56.19 | 0.19 |
-0.8 | 92.35 | 22.05 | 57.2 | 0.2 |
-0.7 | 90 | 26.85 | 58.43 | 0.22 |
-0.6 | 87.35 | 31.23 | 59.29 | 0.22 |
-0.5 | 84.3 | 36.72 | 60.51 | 0.24 |
-0.4 | 80.43 | 41.79 | 61.11 | 0.24 |
-0.3 | 76.32 | 47.35 | 61.84 | 0.25 |
-0.2 | 72.45 | 52.68 | 62.57 | 0.26 |
-0.1 | 68.25 | 57.78 | 63.01 | 0.26 |
0 | 63.18 | 62.95 | 63.06 | 0.26 |
0.1 | 57.95 | 68.08 | 63.01 | 0.26 |
0.2 | 52.55 | 72.75 | 62.65 | 0.26 |
0.3 | 47.15 | 77.05 | 62.1 | 0.25 |
0.4 | 41.56 | 80.86 | 61.21 | 0.24 |
0.5 | 36.56 | 84.77 | 60.66 | 0.24 |
0.6 | 31.79 | 87.42 | 59.6 | 0.23 |
0.7 | 26.99 | 89.57 | 58.28 | 0.21 |
0.8 | 23.15 | 92.55 | 57.85 | 0.22 |
0.9 | 19.44 | 94.34 | 56.89 | 0.21 |