[gmx-users] question about NVT run

[Justin Lemkul]

On 10/12/12 9:39 AM, reisingere at rostlab.informatik.tu-muenchen.de wrote:
> Hi everybody,
> I want to do a md run for my membrane protein. For this I first did the
> energy minimization run.
> This finished with no errors. Here is the output of it:
>
> Steepest Descents did not converge to Fmax < 10 in 5001 steps.
> Potential Energy  = -8.7463000e+05
> Maximum force     =  4.5984982e+02 on atom 1107
> Norm of force     =  5.0847759e+00
>
>
> The mdp file contains:
>
> define                  = -DPOSRES
> integrator              = steep
> emtol                   = 10
> nsteps                  = 5000
> nstenergy               = 1
> energygrps              = System
> coulombtype             = PME
> rcoulomb                = 0.9
> rvdw                    = 0.9
> rlist                   = 0.9
> fourierspacing          = 0.12
> pme_order               = 4
> ewald_rtol              = 1e-5
> pbc                     = xyz
>
>
>
> Now I want to do a NVT to run to prepare the MD run. But here I get the
> error:
>
> Segmentation fault
> mdrun -nice 0 -c protein.pdb -compact -deffnm protein -s protein.tpr -v
> 2>> nvtLogErr >> nvtLogOut
>
>
> And there are is no protein.pdb file as output but only step.pdb files
> like step43b_n6.pdb
>
> The mdp file looks like this:
>
> define               = -DPOSRES
> integrator           = md
> dt                   = 0.005
> nsteps               = 10000
> nstxout              = 0
> nstvout              = 0
> nstfout              = 0
> nstlog               = 1000
> nstxtcout            = 0
> nstenergy            = 5
> energygrps           = Protein Non-Protein
> nstcalcenergy        = 5
> nstlist              = 10
> ns-type              = Grid
> pbc                  = xyz
> rlist                = 0.9
> coulombtype          = PME
> rcoulomb             = 0.9
> rvdw                 = 0.9
> fourierspacing       = 0.12
> pme_order            = 4
> ewald_rtol           = 1e-5
> gen_vel              = yes
> gen_temp             = 200.0
> gen_seed             = 9999
> constraints          = all-bonds
> tcoupl               = V-rescale
> tc-grps              = Protein  Non-Protein
> tau_t                = 0.1      0.1
> ref_t                = 298      298
> pcoupl               = no
>
>
> I already did this but the last time I fixed all atoms except of the
> hydrogen atoms because I wanted to do a MD run with them and keep the rest
> of the  structure.
> But know I want to do the MD run with all the side chains. And now the
> error occurs.
>
> Can you please tell me what a segmentation fault means? Is this because of
> the mdp file or because of the protein?
>

Your time step (5 fs) is much too high unless you are using virtual sites. 
Reduce it to 2 fs.

-Justin

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

Justin A. Lemkul, Ph.D.
Research Scientist
Department of Biochemistry
Virginia Tech
Blacksburg, VA
jalemkul[at]vt.edu | (540) 231-9080
http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin

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