Skip to main content

Table 2 Overall performance of each model

From: Utilizing machine learning-based QSAR model to overcome standalone consensus docking limitation in beta-lactamase inhibitors screening: a proof-of-concept study

 

RF wt docking

LR wt docking

RF w/o docking

LR w/o docking

Training

Testing

Training

Testing

Training

Testing

Training

Testing

True Positive

27.00

5.00*

27.00

4.00

27.00

5.00*

27.00

5.00*

True Negative

39.00

10.00

39.00

8.00

39.00

11.00*

39.00

8.00

False Positive

0.00

4.00

0.00

6.00

0.00

3.00*

0.00

6.00

False Negative

0.00

4.00*

0.00

5.00

0.00

4.00*

0.00

4.00*

% Sensitivity (% True Positive Rate)

100.00

55.56*

100.00

44.44

100.00

55.56*

100.00

55.56*

% Specificity

100.00

71.43

100.00

57.14

100.00

78.57*

100.00

57.14

% False Positive Rate

0.00

28.57

0.00

42.86

0.00

21.43*

0.00

42.86

% Accuracy

100.00

65.22

100.00

52.17

100.00

69.57*

100.00

56.52

ROC_AUC

1.00

0.63

1.00

0.51

1.00

0.67*

1.00

0.56

  1. Bold and asterisk indicate the best value of the testing sets of each model