[gmx-users] mdrun writes gro file in two different formats and failed visualization of trajectories

[Justin Lemkul]

On 8/19/15 7:22 AM, anu chandra wrote:
> Dear Gromacs users,
>
> I just noticed that during simulation 'mdrun'  writes gro file in two
> formats - one with broken molecules at the boundaries and other one where
> surface molecules are perfectly wrapped. Interestingly, this difference in
> format depends on where I run the simulation - either in a cluster or in my
> local 8-core desktop machine. Using 'mdrun' in cluster gives gro file with
> broken molecules at the boundaries and my  desktop workstation gives the
> one with complete molecules at the surface. Below is the mdp file I used
> for simulation in both case
>
> *************************************************************************************
>
> define                  = -DPOSRES_MD
> ;
> integrator              = md
> dt                      = 0.002
> nsteps                  = 50000
> ;
> nstlog                  = 1000
> nstxout                 = 1000
> nstvout                 = 1000
> nstfout                 = 1000
> nstcalcenergy           = 100
> nstenergy               = 1000
> ;
> cutoff-scheme           = Verlet
> nstlist                 = 20
> rlist                   = 1.2
> coulombtype             = pme
> rcoulomb                = 1.2
> vdwtype                 = Cut-off
> vdw-modifier            = Force-switch
> rvdw_switch             = 1.0
> rvdw                    = 1.2
> ;
> pbc                     = xyz
> ;
> tcoupl                  = Nose-Hoover
> tc_grps                 = Protein POPC CL_SOL
> tau_t                   = 0.5    0.5     0.5
> ref_t                   = 305.0    305.0    305.0
> ;
> pcoupl                  = Parrinello-Rahman
> pcoupltype              = semiisotropic
> tau_p                   = 5.0
> compressibility         = 4.5e-5   4.5e-5
> ref_p                   = 1.0     1.0
> ;
> constraints             = h-bonds
> constraint_algorithm    = LINCS
> continuation            = yes
> ;
> nstcomm                 = 100
> comm_mode               = linear
> comm_grps               = Protein_POPC CL_SOL
> ;
> refcoord_scaling        = com
> ***********************************************************************
>
> May I know what is making this difference? Due to the presence of broken
> molecules, I failed to visualizing the trajectories (.trr file) using VMD.
> Though I could make a gro file with complete molecules at the boundaries
> using  trjconv command,
>
>
> *trjconv -f input.gro -o ouput.gro -s input.tpr -pbc mol*
> the visualization of trr file with output.gro also shows long bond between
> atoms across the boundaries.  Please help me to rectify this issue with
> visualizing the trr file.
>

You need to process the .trr file the same way using trjconv.  Accounting for 
periodicity effects is a routine first step after a run finishes.

> In server, following grommp and mdrun commands are used,
>
>
>
> *gmx_mpi grompp -f md.mdp -o MD5.tpr -c MD4.gro -t MD4.cpt -r ref.pdb -n
> index.ndx -p topol.topmpirun -ppn 16 -np 64 gmx_mpi mdrun -deffnm MD5*
>

Here, you're using domain decomposition (anything > 8 cores).  Broken molecules 
get written.

> and, in my desktop workstation following commands are used
>
>
>
> *gmx grompp -f md.mdp -o MD5.tpr -c MD4.gro -t MD4.cpt -r ref.pdb -n
> index.ndx -p topol.topgmx mdrun -nt 8 -deffnm MD5*
>

Here, there is no DD.  Molecules are whole.

-Justin

-- 
==================================================

Justin A. Lemkul, Ph.D.
Ruth L. Kirschstein NRSA Postdoctoral Fellow

Department of Pharmaceutical Sciences
School of Pharmacy
Health Sciences Facility II, Room 629
University of Maryland, Baltimore
20 Penn St.
Baltimore, MD 21201

jalemkul at outerbanks.umaryland.edu | (410) 706-7441
http://mackerell.umaryland.edu/~jalemkul

==================================================