Table 3.
Performance comparison of Porpoise and seven state-of-the-art methods on the same benchmark training datasets
| Species (dataset) | Method | ACC (%) | Sn (%) | Sp (%) | MCC (%) |
|---|---|---|---|---|---|
| H. sapiens (H_990) | Porpoise (10-fold) | 78.53 | 89.11 | 67.94 | 58.45 |
| iRNA-PseU | 60.40 | 61.01 | 59.80 | 21.00 | |
| PseUI | 64.24 | 64.85 | 63.64 | 28.00 | |
| iPseU-CNN | 66.68 | 65.00 | 68.78 | 34.00 | |
| XG-PseU | 65.44 | 63.64 | 67.24 | 31.00 | |
| EnsemPseU | 66.28 | 63.46 | 69.09 | 33.00 | |
| RF-PseU (10-fold) | 64.30 | 66.10 | 62.60 | 29.00 | |
| RF-PseU (LOO) | 64.00 | 65.90 | 62.60 | 29.00 | |
| MU-PseUDeep | 72.60 | 70.90 | 81.00 | 52.40 | |
| S. cerevisiae (S_628) | Porpoise (10-fold) | 81.69 | 81.21 | 82.17 | 63.38 |
| iRNA-PseU | 64.49 | 64.65 | 64.33 | 29.00 | |
| PseUI | 65.13 | 62.74 | 67.52 | 30.00 | |
| iPseU-CNN | 68.15 | 66.36 | 70.45 | 37.00 | |
| XG-PseU | 68.15 | 66.84 | 69.45 | 37.00 | |
| EnsemPseU | 74.16 | 73.88 | 74.45 | 49.00 | |
| RF-PseU (10-fold) | 74.80 | 77.20 | 72.40 | 49.00 | |
| RF-PseU (LOO) | 75.80 | 78.20 | 73.40 | 52.00 | |
| MU-PseUDeep | 76.80 | 74.20 | 79.80 | 54.60 | |
| M. musculus (M_944) | Porpoise (10-fold) | 77.75 | 77.83 | 77.67 | 55.55 |
| iRNA-PseU | 69.07 | 73.31 | 64.83 | 38.00 | |
| PseUI | 70.44 | 74.58 | 66.31 | 41.00 | |
| iPseU-CNN | 71.81 | 74.79 | 69.11 | 44.00 | |
| XG-PseU | 72.03 | 76.48 | 67.57 | 45.00 | |
| EnsemPseU | 73.85 | 75.43 | 72.25 | 48.00 | |
| RF-PseU (10-fold) | 74.80 | 73.10 | 76.50 | 50.00 | |
| RF-PseU (LOO) | 74.50 | 72.70 | 75.20 | 48.00 | |
| MU-PseUDeep | 76.00 | 80.00 | 73.00 | 53.70 |
Notes: 10-fold—10-fold cross-validation; LOO—leave-one-out cross-validation. Bold values indicate the best performance in terms of the corresponding measure.