Table 7 The performance evaluation of the proposed features based on PyFeat with AB compared with other techniques: five numerical representations, RFF, Pse-in-One2.0, iLearn, and SubFeat with 10-fold cross-validation based on the lncRNA dataset. Significant values are in bold.
From: Predicting Parkinson disease related genes based on PyFeat and gradient boosted decision tree
Metric | ACC (%) | AUC (%) | AUPR (%) | F1-score (%) | MCC | SEN (%) | SPC (%) |
|---|---|---|---|---|---|---|---|
Binary | 60.4 | 64.9 | 66.2 | 60.1 | 0.209 | 60.1 | 60.7 |
Integer | 61.8 | 64.4 | 67.9 | 59.6 | 0.237 | 57.9 | 65.6 |
Real | 59.3 | 61.4 | 65.3 | 59.3 | 0.190 | 60.1 | 58.5 |
Z-curve | 60.4 | 63.8 | 66.8 | 59.2 | 0.214 | 58.5 | 62.3 |
EIIP | 60.1 | 63.1 | 66.7 | 60.2 | 0.206 | 60.7 | 59.6 |
RFF | 67.5 | 67.4 | 64.0 | 66.5 | 0355 | 65.4 | 69.2 |
Pse-in-One2.0 | 65.5 | 64.2 | 63.7 | 65.6 | 0.316 | 66.5 | 64.7 |
iLearn | 59.4 | 64.8 | 67.5 | 61.2 | 0.193 | 64.4 | 54.2 |
SubFeat | 63.6 | 66.7 | 67.7 | 63.5 | 0.278 | 64.5 | 62.6 |
Proposed features | 77.8 | 84.1 | 84.5 | 77.4 | 0.560 | 77.3 | 78.3 |
