Table 10.
Overall performance comparison of our method with the existing web tools for predicting plastid proteins.
| Tools | Sensitivity (%) |
Specificity (%) |
Accuracy (%) |
MCC | Precision (%) | RFP (%) |
|---|---|---|---|---|---|---|
| WoLF PSORT | 56.96 | 74.76 | 65.82 | 0.3223 | 69.50 | 30.50 |
| TargetP | 55.70 | 85.89 | 65.97 | 0.3998 | 88.44 | 11.56 |
| iLoc-PLant | 36.39 | 98.42 | 67.41 | 0.4438 | 95.83 | 4.17 |
| YLoc (HighRes) | 34.81 | 97.47 | 66.14 | 0.4142 | 93.22 | 6.78 |
| PLpred (DIPEP) | 60.44 | 92.72 | 76.58 | 0.56 | 89.25 | 10.75 |
| PLpred (NCC) | 65.82 | 87.97 | 76.90 | 0.55 | 84.55 | 15.45 |
Performance comparison done on an 'independent dataset' that contains 316 plastid and 316 non-plastid proteins. MCC = Matthews Correlation Coefficient, RFP = Rate of False Predictions, DIPEP = Dipeptide composition-based classifier, NCC = Nterminal-Center-Cterminal composition-based classifier.