Table 3 Statistical metrics for all four models of each dataset
From: Using Jupyter Notebooks for re-training machine learning models
Models | LR | SVM | RF | k-NN | ||||
|---|---|---|---|---|---|---|---|---|
Train | Test | Train | Test | Train | Test | Train | Test | |
BCRP | ||||||||
Accuracy | 0.73 | 0.74 | 0.76 | 0.67 | 0.80 | 0.70 | 0.76 | 0.70 |
Sensitivity | 0.75 | 0.81 | 0.75 | 0.69 | 0.79 | 0.71 | 0.79 | 0.74 |
Specificity | 0.69 | 0.50 | 0.77 | 0.61 | 0.83 | 0.63 | 0.73 | 0.53 |
Balanced accuracy | 0.72 | 0.65 | 0.76 | 0.65 | 0.80 | 0.67 | 0.76 | 0.63 |
F1-score | 0.74 | 0.83 | 0.77 | 0.77 | 0.80 | 0.79 | 0.78 | 0.79 |
AUC | 0.80 | 0.65 | 0.83 | 0.65 | 0.88 | 0.67 | 0.80 | 0.63 |
Precision | 0.74 | 0.86 | 0.79 | 0.87 | 0.85 | 0.88 | 0.77 | 0.86 |
MCC | 0.46 | 0.28 | 0.53 | 0.25 | 0.61 | 0.28 | 0.52 | 0.23 |
BSEP | ||||||||
Accuracy | 0.84 | 0.72 | – | 0.78 | – | 0.83 | - | |
Sensitivity | 0.22 | – | 0.84 | – | 0.79 | – | 0.52 | - |
Specificity | 0.96 | – | 0.69 | – | 0.77 | – | 0.89 | - |
Balanced accuracy | 0.59 | – | 0.77 | – | 0.77 | – | 0.71 | - |
F1-score | 0.30 | – | 0.54 | – | 0.57 | – | 0.53 | - |
AUC | 0.73 | – | 0.85 | – | 0.87 | – | 0.79 | - |
Precision | 0.59 | – | 0.42 | – | 0.50 | – | 0.60 | - |
MCC | 0.28 | – | 0.44 | – | 0.49 | – | 0.45 | - |
OATP1B1 | ||||||||
Accuracy | 0.86 | 0.38 | 0.80 | 0.76 | 0.85 | 0.71 | 0.87 | 0.71 |
Sensitivity | 0.20 | 0.33 | 0.74 | 0.83 | 0.63 | 0.72 | 0.35 | 0.67 |
Specificity | 0.97 | 0.67 | 0.81 | 0.33 | 0.89 | 0.67 | 0.96 | 1 |
Balanced accuracy | 0.59 | 0.50 | 0.77 | 0.58 | 0.74 | 0.69 | 0.65 | 0.83 |
F1-score | 0.27 | 0.48 | 0.52 | 0.86 | 0.55 | 0.81 | 0.43 | 0.80 |
AUC | 0.77 | 0.50 | 0.83 | 0.58 | 0.84 | 0.69 | 0.81 | 0.83 |
Precision | 0.47 | 0.86 | 0.40 | 0.88 | 0.49 | 0.93 | 0.58 | 1 |
MCC | 0.24 | - | 0.44 | 0.15 | 0.47 | 0.29 | 0.34 | 0.47 |
OATP1B3 | ||||||||
Accuracy | 0.91 | 0.35 | 0.84 | 0.71 | 0.86 | 0.59 | 0.92 | 0.65 |
Sensitivity | 0.14 | 0.23 | 0.81 | 0.77 | 0.77 | 0.69 | 0.36 | 0.62 |
Specificity | 0.98 | 0.75 | 0.84 | 0.50 | 0.87 | 0.25 | 0.97 | 0.75 |
Balanced accuracy | 0.56 | 0.49 | 0.83 | 0.64 | 0.82 | 0.47 | 0.67 | 0.68 |
F1-score | 0.20 | 0.35 | 0.46 | 0.80 | 0.48 | 0.72 | 0.41 | 0.73 |
AUC | 0.79 | 0.49 | 0.88 | 0.64 | 0.89 | 0.47 | 0.80 | 0.68 |
Precision | 0.45 | 0.75 | 0.32 | 0.83 | 0.35 | 0.75 | 0.50 | 0.89 |
MCC | 0.20 | − 0.02 | 0.44 | 0.25 | 0.46 | − 0.05 | 0.38 | 0.31 |
MRP3 | ||||||||
Accuracy | 0.88 | – | 0.60 | – | 0.59 | – | 0.78 | – |
Sensitivity | 0 | – | 0.77 | – | 0.68 | – | 0.20 | – |
Specificity | 0.99 | – | 0.58 | – | 0.59 | – | 0.86 | – |
Balanced accuracy | 0.5 | – | 0.67 | – | 0.62 | – | 0.53 | – |
F1-score | 0 | – | 0.43 | – | 0.37 | – | 0.21 | – |
AUC | 0.44 | – | 0.67 | – | 0.63 | – | 0.57 | – |
Precision | 0.1 | – | 0.35 | – | 0.32 | – | 0.34 | – |
MCC | 0 | – | 0.30 | – | 0.20 | – | 0.12 | – |
P-gp | ||||||||
Accuracy | 0.74 | 0.65 | 0.72 | 0.68 | 0.76 | 0.68 | 0.71 | 0.64 |
Sensitivity | 0.81 | 0.92 | 0.72 | 0.81 | 0.81 | 0.92 | 0.76 | 0.88 |
Specificity | 0.64 | 0.28 | 0.71 | 0.50 | 0.70 | 0.35 | 0.64 | 0.32 |
Balanced accuracy | 0.73 | 0.60 | 0.71 | 0.65 | 0.76 | 0.64 | 0.70 | 0.60 |
F1-score | 0.78 | 0.75 | 0.73 | 0.74 | 0.79 | 0.77 | 0.75 | 0.74 |
AUC | 0.80 | 0.60 | 0.77 | 0.65 | 0.80 | 0.64 | 0.76 | 0.60 |
Precision | 0.77 | 0.64 | 0.78 | 0.69 | 0.80 | 0.66 | 0.75 | 0.64 |
MCC | 0.46 | 0.27 | 0.44 | 0.33 | 0.53 | 0.34 | 0.43 | 0.25 |