Table 7 Performance comparison of predictors using fivefold cross-validation on benchmark dataset
From: iProL: identifying DNA promoters from sequence information based on Longformer pre-trained model
Predictor | Sn(%) | Sp(%) | Acc(%) | MCC | AUC |
|---|---|---|---|---|---|
iPSW(2L)-PseKNC | 81.37 | 84.89 | 83.13 | 0.6630 | 0.9054 |
iPromoter-ET | 84.23 | 86.04 | 85.14 | 0.7030 | 0.9193 |
dPromoter-XGBoost | 85.72 | 81.92 | 83.81 | 0.6770 | / |
BERT-Promoter | 84.34 | 86.56 | 85.45 | / | 0.9020 |
iProL (ours) | 84.62 | 86.61 | 85.62 | 0.7130 | 0.9211 |