Table 7 The performance of data augmentation on the SPDS17 dataset
From: SigUNet: signal peptide recognition based on semantic segmentation
Comp | Eukaryotes | Gram-positive | Gram-negative | |||
|---|---|---|---|---|---|---|
Train | MCC (%) | FPRTM (%) | MCC (%) | FPRTM (%) | MCC (%) | FPRTM (%) |
SigUNet | ||||||
 As compa | 89.6 | 1.2 | 40.9 | 2.1 | 97.6 | 1.1 |
 All organismsb | 89.2 | 2.2 | 46.1 | 1.6 | 100.0 | 0.0 |
 Bacteriac | – | – | 49.5 | 1.1 | 97.6 | 1.1 |
SigUNet-light | ||||||
 As comp | 84.8 | 3.7 | 51.7 | 1.6 | 92.8 | 1.1 |
 All organisms | 89.1 | 2.2 | 43.3 | 1.6 | 100.0 | 0.0 |
 Bacteria | – | – | 49.5 | 1.1 | 100.0 | 0.0 |