Table 2 Performance of DTR model for twelve distinct fingerprints.

From: Drug repurposing targeting COVID-19 3CL protease using molecular docking and machine learning regression approaches

Sr. No.

Descriptors/Fingerprints

Q2

R2

Difference

MSE

RMSE

1

CDK

0.85

0.97

0.12

2.81

1.68

2

MACCS

0.87

0.97

0.10

2.46

1.57

3

Extended CDK

0.97

0.90

0.07

3.38

1.84

4

PubChem

0.74

0.93

0.19

3.50

1.87

5

2D Atom pair

0.78

0.82

0.04

2.91

1.71

6

E-state

0.51

0.70

0.19

2.70

1.64

7

Graph Only

0.76

0.81

0.05

2.87

1.69

8

Klekota Roth

0.63

0.79

0.16

3.27

1.81

9

Klekota Roth count

0.79

0.68

0.11

3.11

1.76

10

Substructure

0.59

0.82

0.23

2.85

1.69

11

Substructure count

0.77

0.72

0.05

2.81

1.68

12

2D Atom pair count

0.80

0.83

0.03

3.59

1.90

Back to article page