Table 5 Performance comparison of deepstacked-AVPs method with existing models
Dataset | Predictor | Acc (%) | Sn (%) | Sp (%) | MCC |
|---|---|---|---|---|---|
Training dataset | AVPpred [8] | 85.00 | 82.20 | 88.20 | 0.70 |
Chang et al. [9] | 85.10 | 86.60 | 83.00 | 0.70 | |
Meta-iAVP [16] | 88.20 | 89.20 | 86.90 | 0.76 | |
Chowdhury et al. [54] | 92.40 | 93.30 | 91.10 | 0.84 | |
Deepstacked-AVPs | 96.60 | 94.85 | 97.35 | 0.92 | |
Independent dataset | AVPpred [8] | 92.50 | 93.30 | 91.70 | 0.85 |
Chang et al. [9] | 93.30 | 91.70 | 95.00 | 0.87 | |
Meta-iAVP [16] | 94.90 | 91.70 | 98.30 | 0.90 | |
AntiVPP 1.0 [12] | 93.00 | 87.00 | 97.00 | 0.87 | |
Deepstacked-AVPs | 95.15 | 96.27 | 94.91 | 0.90 |