Table 5 Effect of the BiLSTM module on the performance of the model
From: iProL: identifying DNA promoters from sequence information based on Longformer pre-trained model
Model | Sn(%) | Sp(%) | Acc(%) | MCC | AUC |
|---|---|---|---|---|---|
BiLSTM ×0 | 82.55 | 87.14 | 84.85 | 0.6981 | 0.9140 |
BiLSTM ×1 (iProL) | 84.62 | 86.61 | 85.62 | 0.7130 | 0.9211 |