[gmx-users] Discrepancy in a single point energy simulation between Gromacs and Namd.

[Mark Abraham]

Hi,

Best practice for single-points is here
http://www.gromacs.org/Documentation/How-tos/Single-Point_Energy, but you
are fortunately avoiding most of the issues by not using constraints.

Last time someone reported a discrepancy like this (AMBER, IIRC) there was
a difference in the way the two codes were shifting the non-bonded
potential to be zero at the cutoff. Your .mdp forces Gromacs to use no
shifts, but I don't know whether you have done so for NAMD, if it is needed
and/or possible. You can try coulomb_modifier = potential-shift.

Mark


On Fri, Nov 7, 2014 at 5:45 PM, IÑIGO SAENZ <inigo.saenz01 at estudiant.upf.edu
> wrote:

> Hi all,
>
> I've got a system of a protein inside a cube of water parametrized with the
> amber ff99SB force-field. I have converted the .prmtop and .pdb files of
> this system into a .gro and .top by means of ACPYPE. The problem is that
> when I perform the single point energy calculation of these system with
> Namd and Gromacs I obtain very opposing results in the Electrostatic energy
> results and I suspect that I'm overlooking some details in the .mdp file
> or  in the namd input file. I paste you the results that I obtain with NAMD
> and gromacs and also the configuration files.
>
>
> BOND:     38088.4829     38081.7399
> ANGLE:    1573.6867      1573.7189
> DIHEDRAL: 9873.2782      9872.7127
> COULOMB:  -179655.6187   141241 (1-4 interactions)  -1.84836e+07 (short
> range)
> VDW:      21796.8418     21797.2514
>
> First column corresponds to NAMD results and the second one to Gromacs
> results,
> Now, the *SPE.mdp file*
>
> define                   = -DFLEXIBLE
> integrator               = md
> nsteps                   = 0
> dt                       = 0.001
> constraints              = none
> nstcomm                  = 1
> ns_type                  = simple
> nstlist                  = 1
> coulombtype              = cut-off
> vdw_modifier             = none
> coulomb_modifier         = none
> rlist                    = 0.9
> rcoulomb                 = 0.9
> rvdw                     = 0.9
> Tcoupl                   = no
> Pcoupl                   = no
> gen_vel                  = no
> nstxout                  = 1
> pbc                      = xyz
> nstlog                   = 1
> nstenergy                = 1
> nstvout                  = 1
> nstfout                  = 1
> nstxtcout                = 1
> comm_mode                = ANGULAR
> continuation             = yes
>
> *NAMD.conf file:*
>
> rigidBonds     none
> timestep       2
> temperature    0
> cutoff         9
> switching      off
> PME            off
> cellBasisVector1   117.210 0  0
> cellBasisVector2   0  94.119  0
> cellBasisVector3   0  0  78.923
> # NAMD doesn't force neutralisation of charge
> amber          on  # Specify this is AMBER force field
> parmfile       parameters  # Input PARM file
> ambercoor      s.inpcrd  # Input coordinate file
> outputname     run_namd  # Prefix of output files
> exclude        scaled1-4
> 1-4scaling     0.8333  # =1/1.2, default is 1.0
> run            0 ;# 2ps
>
> I should mention that in .gro I put the box size correctly, the same values
> that the box-size values that appear in the NAMD.conf file divided by 10 to
> convert them to nanometers.
>
> I'm desperate with this error, please, any help would be highly
> appreciated! If you need more information about this problem, feel free to
> ask me.
>
> Thank you very much in advance!
> --
> 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.
>