Skip to main content

Table 3 Evaluation of free energy estimators (native trajectory included). Performance of SEEFs and PEEFs on the set including native simulation. Performance is evaluated according to different methods (see Methods section). The lower part of the table refers to scores averaged over a moving window of 50 molecular dynamics snapshots.

From: A decoy set for the thermostable subdomain from chicken villin headpiece, comparison of different free energy estimators

NATIVE+F1+F3+F4+F7 – no average

Score funct.

logP(B1)

logP(B10)

F.E.

corr. coeff.

RMSD (Ã…) best score

contact + pc

-1.08

-1.08

13.8

0.62

3.03

FRST

-1.38

-1.94

23.2

0.48

2.61

MM

-0.25

-1.39

10.6

0.21

7.45

MM/GBSA

-1.71

-2.02

29.6

0.66

2.40

MM/PBSA

-1.79

-2.02

23.2

0.58

2.35

NATIVE+F1+F3+F4+F7 – with average

Score funct.

logP(B1)

logP(B10)

F.E.

corr. coeff.

RMSD (Ã…) best score

contact + pc

-0.91

-0.91

16.3

0.65

3.77

FRST

-1.79

-2.78

43.2

0.54

2.39

MM

-0.70

-1.61

14.8

0.25

4.25

MM/GBSA

-1.14

-1.67

36.1

0.72

2.88

MM/PBSA

-1.54

-2.12

28.6

0.67

2.52