Skip to main content

Table 18 Win/loss counts for ten random folds for extended similarites on DuD set ( α =0 . 3; approach B2)

From: Improving structural similarity based virtual screening using background knowledge

 

     ECFP ext

     MCS ext

 

     1%

     5%

     10%

     1%

     5%

     10%

 

Win

Loss

Win

Loss

Win

Loss

Win

Loss

Win

Loss

Win

Loss

HMGR

10

0

9

1

9

1

5

0

5

0

5

0

ER

8

2

9

1

7

3

10

0

9

1

7

3

PPAR γ

10

0

9

1

9

1

6

4

10

0

8

2

P38 MAP

9

1

10

0

9

1

9

1

9

1

10

0

TK

7

3

9

1

9

1

9

1

6

4

7

3

FXa

10

0

7

3

8

2

9

1

8

2

8

2

ADA

10

0

6

4

9

1

6

3

8

1

6

3

DHFR

7

3

10

0

10

0

10

0

10

0

10

0

ACHE

8

2

9

1

9

1

8

2

10

0

9

1

COX-2

8

2

9

1

7

3

7

3

8

2

8

2

Sum

87

13

87

13

86

14

79

15

83

11

78

16

  1. Win/loss counts for all ten random folds for extended similarites MCS ext and ECFP ext versus their base similarities MCS and ECFP for the receptor specific decoy sets DuD set at 1%, 5% and 10% of the database. The extension fingerprint is calculated from 10% (20% for HMGR, TK and ADA) of the ligands (approach B2).