Research Article
ACNNT3: Attention-CNN Framework for Prediction of Sequence-Based Bacterial Type III Secreted Effectors
Table 3
Comparison of ACNNT3 and DeepT3, Effective T3, BPBAac, and BEAN2 on a P. syringae dataset.
| Method | | | | | | | |
| ACNNT3-1 | 0.900 | 0.976 | 0.357 | 0.936 | 0.887 | 0.452 | 0.667 | ACNNT3-2 | 0.872 | 0.988 | 0.143 | 0.926 | 0.866 | 0.265 | 0.565 | DeepT3-1 | 0.905 | 0.962 | 0.429 | 0.932 | 0.884 | 0.472 | 0.838 | DeepT3-2 | 0.913 | 0.924 | 0.500 | 0.918 | 0.860 | 0.437 | 0.763 | Effective T3 | 0.906 | 0.906 | 0.428 | 0.906 | 0.838 | 0.334 | 0.810 | BPBAac | 0.875 | 0.494 | 0.571 | 0.631 | 0.505 | 0.046 | 0.562 | BEAN2 | 0.883 | 0.988 | 0.083 | 0.938 | 0.884 | 0.271 | 0.607 |
|
|
The bold values indicate the best prediction results.
|