[gmx-users] BOND ... ANGLE and TORSION energy discrepancy between gromacs and cp2k

[David van der Spoel]

On 2013-08-08 01:16, Golshan Hejazi wrote:
> Hello,
>
> I have performed quit a lot of tests and I have some question:
> I have an amber.crd and amber.top files ... I perform a single step energy calculation with amber , gromacs and cp2k. in all of them I am using amber force field but the point is that in cp2k and amber, I am reading crd and top file directly while in gromacs, i need to convert them to gromacs format.
>
> I performed this test for a simple ace-ala-nme system. Lets talk only about the BOND, ANGLE and TORSION energies which are the same in all of these packages.  these energy components output are identical in amber and cp2k. However, there is a big difference between them and gromacs:
>
> cp2k, amber ... BOND= 0.0206  (kcal/mol)   ANGLE=0.3620 (kcal/mol)   TORSION=8.1071 (kcal/mol)
> gromacs    ....    BOND=0.14044 (kcal/mol)   ANGLE=0.3780 (kcal/mol)   TORSION=9.74190 (kcal/mol)
>
> The other terms are also different. But lets focus on these because they are the same in all these packages.
>
> Now, how did i convert the crd and top file to gromacs fromat?
>
> 1- I tried amb2gmx perl script
> 2- I used ambpdb of amber to generate a pdb file and then pdb2gmx of gromacs to generate the top and gro file for gromacs
>
> BOTH of these ways gives different results from each other and also from cp2k and amber!
> I would really appreciate any help about this problem ... it seems to me there is a bug in amber to gmx file convertors?!
>
it looks like some kind of rounding error, pdb file have limited 
precision. the force constants you can check yourself.

> Thanks
> Golshan
>


-- 
David van der Spoel, Ph.D., Professor of Biology
Dept. of Cell & Molec. Biol., Uppsala University.
Box 596, 75124 Uppsala, Sweden. Phone:	+46184714205.
spoel at xray.bmc.uu.se    http://folding.bmc.uu.se