Skip to main content

Table 2 Statistical results of the seven classification models based on 144 molecular features for the training (random fivefold cross-validation) and test sets

From: ADMET evaluation in drug discovery. 20. Prediction of breast cancer resistance protein inhibition through machine learning

 Training set (random fivefold cross-validation)Test set
GABAMCCAUCGABAMCCAUC
SVM0.9020.8930.7930.9470.9110.9050.8120.958
DNN0.8940.8920.7800.9500.9070.9040.8060.960
XGBoost0.9020.8940.7930.9560.8910.8830.7700.957
SGB0.9010.8940.7920.9520.8860.8790.7590.958
RLR0.8750.8720.7400.9320.8730.8670.7340.936
k-NN0.8630.8620.7170.9130.8570.8560.7050.917
NB0.8260.8340.6540.8980.7800.7930.5720.888
Consensus10.9020.8930.793NA0.9030.8970.797NA
Consensus20.9010.8950.7930.9560.9090.9030.8080.963
  1. NA not available