Table 3 Detailed performance metrics of various models on the test split of the dataset.
From: Protein feature engineering framework for AMPylation site prediction
Model | Representations | Feat types | Offsets | Nfeat | Accuracy | Precision | Recall | F1 score | AUC ROC | MCC |
|---|---|---|---|---|---|---|---|---|---|---|
ANN | (‘conform’, ‘no_reduction’) | (‘mat’) | (1, 3) | 898 | 0.815789 | 0.724138 | 0.777778 | 0.750000 | 0.900605 | 0.605428 |
SVM | (‘no_reduction’) | (‘counts’, ‘mat’) | (1, 2) | 820 | 0.802632 | 0.772727 | 0.629630 | 0.693878 | 0.913076 | 0.556759 |
XGB | (‘hydro’, ‘no_reduction’) | (‘counts’, ‘mat’) | (1, 3) | 875 | 0.776316 | 0.678571 | 0.703704 | 0.690909 | 0.867725 | 0.515952 |
LGBM | (‘conform’, ‘no_reduction’) | (‘mat’) | (1, 2, 3) | 1347 | 0.750000 | 0.642857 | 0.666667 | 0.654545 | 0.879819 | 0.458957 |
RF | (‘conform’, ‘no_reduction’, ‘hydro’) | (‘counts’, ‘mat’) | (1, 3) | 980 | 0.789474 | 0.823529 | 0.518519 | 0.636364 | 0.908541 | 0.525200 |
