Table 1 Comparative analysis on our RF model using different encoding schemes under 5-fold cross-validation on different development datasets
From: iPseU-NCP: Identifying RNA pseudouridine sites using random forest and NCP-encoded features
Dataset | Encoding Scheme | ACC (%) | SN (%) | SP(%) | MCC |
|---|---|---|---|---|---|
H_990 | PseKNC | 59.39 | 69.49 | 49.29 | 0.19 |
CKSNAP | 60.00 | 83.84 | 36.16 | 0.23 | |
NCP | 62.92 | 58.79 | 65.05 | 0.24 | |
S_628 | PseKNC | 58.76 | 51.91 | 65.61 | 0.18 |
CKSNAP | 60.03 | 56.37 | 63.69 | 0.20 | |
NCP | 69.59 | 77.07 | 62.10 | 0.40 | |
M_944 | PseKNC | 56.57 | 44.49 | 68.64 | 0.14 |
CKSNAP | 57.52 | 52.54 | 62.50 | 0.15 | |
NCP | 71.82 | 67.37 | 76.27 | 0.44 |