[gmx-users] why the energy items in gromacs is different from charmm?

[Mark Abraham]

On Fri, Jul 4, 2014 at 6:44 AM, wliu <wliu at itcs.ecnu.edu.cn> wrote:

> Hi,
>
> I have a following issue:
>     I use the force field CHARMM22-CMAP (that is charmm27-CMAP) in gromacs
> to calculate the chignolin.pdb, I want to know the energy items of just
> this one conformation. I choose to run a very simple md, just one step. my
> input file writes as follows:
>
> md.mdp
> emtol                    = 10
> emstep                   = 0.001
> niter                    = 1
> fcstep                   = 0
> nstcgsteep               = 1000
> nbfgscorr                = 10
> rtpi                     = 0.05
> nstxout                  = 0
> nstvout                  = 0
> nstfout                  = 0
> nstlog                   = 100
> nstcalcenergy            = -1
> nstenergy                = 10
> nstxtcout                = 100
> xtc-precision            = 1000
> energygrps               = System
>

That will mean GROMACS uses its default settings for non-bonded and bonded
energy and force evaluations. These have nothing to do with how the CHARMM
force field should be used, or is used in the Perl script you reporting
using, so large differences are expected. The shift implementations differ
slightly, so you will not be able to get perfect agreement for all terms,
but at the moment there's not the slightest reason to expect agreement.


> And I also use CHARMM with the same force field CHARMM22-CMAP, but the
> enerCHARMM.pl command to calculate chignolin's energy items. But I found
> that their results are quite different. I'd like to list below:
>
>                               charmm22     gromacs/md22
> chignolin(kcal/mol)
> angles                         20.9848           /
> bonds                          22.7760        195.141
> cmap                           -0.0595         -0.0684
> dihedrals                      46.0533           /
> elec                           19.5690           /
> impropers                       0.0002          0.014
> total                      453415.3682     201520.813
> ureyb                           2.7128         38.779
> vdwaals                    453303.3316           /
> proper dihedral                  /             46.478
> LJ-14                            /             82.409
> Coulomb-14                       /            650.335
> LJ (SR)                          /         201106.699
> Coulomb (SR)                     /           -598.990
>
>
> torsion                        45.994          46.4236
>                             =  -0.0595     =   -0.0684
>                              + 46.0533      +  46.478
>                              +  0.0002      +   0.014               ( I
> have Unified their unit to kcal/mol)
>
> I want to know why their results shown such a difference? I also use
> step=0 and minimization method to try, but nearly the same result.


Don't use either of those methods. Follow
http://www.gromacs.org/Documentation/How-tos/Single-Point_Energy


> Theoretically，Two different MD programs calculating one same structure
> should give out the results which differ no more than 0.01kcal/mol, but,
> this makes me confused. So could you please help me find where the problem
> is or the reasons?
> Thank you!
>
> Looking forward to your kind reply！
>
>
> Sincerely yours,
> Wei Liu，
>
> wliu at itcs.ecnu.edu.cn
> East China Normal University
> ShangHai，china，200062
>
>
> --
> Gromacs Users mailing list
>
> * Please search the archive at http://www.gromacs.org/
> Support/Mailing_Lists/GMX-Users_List before posting!
>
> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
>
> * For (un)subscribe requests visit
> https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or
> send a mail to gmx-users-request at gromacs.org.
>