Table 7 Performance comparison of DPI-CDF predictor with existing state-of-the-art methods on independent dataset \(DP_{ind}\)
From: DPI_CDF: druggable protein identifier using cascade deep forest
Predictor | Algorithm | ACC (%) | SEN (%) | SPE (%) | MCC | F-score |
|---|---|---|---|---|---|---|
Yu’s method | CNN-RNN | 89.8 | 84.8 | 89.5 | 0.799 | 0.889 |
SPIDER | SVM | 90.7 | 85.7 | 95.4 | 0.816 | 0.899 |
DPI_CDF (our method) | CDF | 95.01 | 96.87 | 93.24 | 0.90 | 0.949 |