<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd">
<html xmlns="http://www.w3.org/1999/xhtml">
<head>
<meta content="text/html; charset=UTF-8" http-equiv="Content-Type" />
<title></title>
</head>
<body>
<p style="margin: 0px;"><span></span></p>
<div style="margin: 5px 0px 5px 0px;">
On September 9, 2010 at 10:04 PM "Justin A. Lemkul" <jalemkul@vt.edu> wrote:<br />
<br />
><br />
><br />
> TJ Mustard wrote:<br />
><br />
> <snip><br />
><br />
> > *Positional restraint mdp file:*<br />
> ><br />
> > <br />
> ><br />
> > define                  = -DPOSRES<br />
> > constraints             = all-bonds<br />
> > integrator              = md<br />
> > dt                      = 0.004 ; ps<br />
><br />
> Unless you're using virtual sites (which, per your commands below, you are not)<br />
> this time step is inappropriately large and is a likely source of instability.<br />
> With plain constraints, 2 fs is more appropriate.<br />
</div>
<p style="margin: 0px;"> </p>
<p style="margin: 0px;">Yes I usually run at 2 fs. I was actually running a huge batch of these at varying time steps and with or without -heavyh.</p>
<p style="margin: 0px;"> </p>
<div style="margin: 5px 0px 5px 0px;">
> > > nsteps                  = 2500 ; total ps = dt*nsteps<br />
> > nstcomm                 = 1<br />
> > nstxout                 = 200 ; output coordinates every ps = dt*nstxout<br />
> > nstvout                 = 1000 ; output velocities every ps = dt*nstvout<br />
> > nstfout                 = 0<br />
> > nstenergy               = 10<br />
> > nstlog                  = 10<br />
> > nstlist                 = 10<br />
> > ns_type                 = grid<br />
> > rlist                   = 0.9<br />
> > coulombtype             = PME<br />
> > rcoulomb                = 0.9<br />
> > rvdw                    = 1.0<br />
> > fourierspacing          = 0.12<br />
> > fourier_nx              = 0<br />
> > fourier_ny              = 0<br />
> > fourier_nz              = 0<br />
> > pme_order               = 6<br />
> > ewald_rtol              = 1e-5<br />
> > optimize_fft            = yes<br />
> > ; Berendsen temperature coupling is on<br />
> > Tcoupl                  = v-rescale<br />
> > tau_t                   = 0.1       0.1  0.1   0.1   0.1<br />
> > tc_grps                 = protein   RNA  SOL   NA    CL<br />
><br />
> Also a major concern - never couple solvent and ions separately!  See here:<br />
><br />
> http://www.gromacs.org/Documentation/Terminology/Thermostats<br />
>
</div>
<p style="margin: 0px;"> </p>
<p style="margin: 0px;">Yes I have since switched to a Protein Non-protein setting.</p>
<p style="margin: 0px;"> </p>
<div style="margin: 5px 0px 5px 0px;">
> > ref_t                   = 310       310  310   310   310<br />
> > ; Pressure coupling is on<br />
> > pcoupl                  = berendsen<br />
> > pcoupltype              = isotropic<br />
> > tau_p                   = 0.5<br />
> > compressibility         = 4.5e-5<br />
> > ref_p                   = 1.0<br />
> > ; Generate velocities is on at 310K (core body temp)<br />
> > gen_vel                 = yes<br />
> > gen_temp                = 310.0<br />
> > gen_seed                = 173529<br />
> ><br />
> > <br />
> ><br />
> > <br />
> ><br />
> > *main molecular dynamics mdp file:*<br />
> ><br />
> > <br />
> ><br />
><br />
> <snip><br />
><br />
> > define                  = -DPOSRES<br />
><br />
</div>
<p style="margin: 0px;">> Restraints during data collection, or did you paste the wrong file?</p>
<p style="margin: 0px;"> </p>
<p style="margin: 0px;">Yeah that will be fixed asap.</p>
<p style="margin: 0px;"> </p>
<div style="margin: 5px 0px 5px 0px;">
> > <snip><br />
><br />
> > ; Generate velocities is on at 310K (core body temp)<br />
> > gen_vel                 = yes<br />
><br />
> Re-generating velocities after equilibration defeats the whole purpose of<br />
> equilibrating.  Per your commands below, you are not preserving the velocities<br />
> (grompp -t with .trr or .cpt file) obtained during your PR equilibration.<br />
>
</div>
<p style="margin: 0px;"> </p>
<p style="margin: 0px;">Yeah that will be fixed asap.</p>
<p style="margin: 0px;"> </p>
<div style="margin: 5px 0px 5px 0px;">
> > gen_temp                = 310.0<br />
> > gen_seed                = 173529<br />
> ><br />
> > <br />
> ><br />
> > <br />
> ><br />
> > *And my script for running the whole process:*<br />
> ><br />
> ><br />
> ><br />
> > pdb2gmx -f receptor.pdb -o receptor.gro -p receptor.top<br />
> ><br />
> > editconf -bt cubic -f receptor.gro -o receptor.gro -c -d 1.5<br />
> ><br />
> > genbox -cp receptor.gro -cs spc216.gro -o receptor_b4ion.gro -p receptor.top<br />
> ><br />
> > grompp -f em.mdp -c receptor_b4ion.gro -p receptor.top -o receptor_b4ion.tpr<br />
> ><br />
> > genion -s receptor_b4ion.tpr -o receptor_b4em.gro -neutral -conc 0.0001<br />
> > -pname NA -nname CL -g receptor_ion.log -p receptor.top<br />
> ><br />
> > <br />
> ><br />
> > *Here I select the SOL for ions...*<br />
> ><br />
> > <br />
> ><br />
> > grompp -f em.mdp -c receptor_b4em.gro -p receptor.top -o receptor_em.tpr<br />
> ><br />
> > mdrun -v -s receptor_em.tpr -c "$base"_after_em.gro -g emlog.log<br />
> ><br />
> > grompp -f pr.mdp -c receptor_after_em.gro -p receptor.top -o receptor_pr.tpr<br />
> ><br />
> > mdrun -v -s receptor_pr.tpr -o receptor_pr.trr -e pr.edr -c<br />
> > receptor_after_pr.gro -g prlog.log -cpi state_pr.cpt -cpo state_pr.cpt<br />
> ><br />
> > grompp -f md.mdp -c receptor_after_pr.gro -p receptor.top -o receptor_md.tpr<br />
> ><br />
> > mdrun -s receptor_md.tpr -o receptor_md.trr -c receptor_after_pr.gro -g<br />
> > md.log -e md.edr -cpi state_md.cpt -cpo state_md.cpt<br />
> ><br />
> > *Where receptor is of course my protein/RNA pdb name*<br />
> ><br />
> > <br />
> ><br />
> > I always make sure that the pdb doesn't give me notes or warnings or<br />
> > errors of course in the pdb2gmx step. Most of the time I minimize to my<br />
> > computers "machine precision."<br />
> ><br />
> > **<br />
> ><br />
> > *This is the return I get from the "EM" step:*<br />
> ><br />
> > <br />
> ><br />
> > Stepsize too small, or no change in energy.<br />
> > Converged to machine precision,<br />
> > but not to the requested precision Fmax < 200<br />
> ><br />
> > Double precision normally gives you higher accuracy.<br />
> > You might need to increase your constraint accuracy, or turn<br />
> > off constraints alltogether (set constraints = none in mdp file)<br />
> ><br />
> > writing lowest energy coordinates.<br />
> ><br />
> > Steepest Descents converged to machine precision in 324 steps,<br />
> > but did not reach the requested Fmax < 200.<br />
> > Potential Energy  = -1.9868888e+07<br />
> > Maximum force     =  8.7276396e+03 on atom 19080<br />
><br />
> This is a very high Fmax, indicating that you have bad clashes in your system<br />
> that, when combined with the parameters above, surely leads to an unstable<br />
> system.  The fact that Gromos96 works is a bit curious, but you have too many<br />
> potential pitfalls listed here to specifically diagnose the difference based on<br />
> this information alone.<br />
><br />
> -Justin<br />
</div>
<p style="margin: 0px;"> </p>
<p style="margin: 0px;">Thank you. I will be making these changes and running the batch again very soon. If I hit the same issues I will post again.</p>
<p style="margin: 0px;"> </p>
<div style="margin: 5px 0px 5px 0px;">
> > > Norm of force     =  3.8592850e+01<br />
> ><br />
> > <br />
> ><br />
> > <br />
> ><br />
> > *I will get this error sometime in the "PR" step of my script:*<br />
> ><br />
> > <br />
> ><br />
> > step 0<br />
> > Step 11  Warning: pressure scaling more than 1%, mu: 1.03087 1.03087 1.03087<br />
> ><br />
> > Step 11  Warning: pressure scaling more than 1%, mu: 1.03087 1.03087 1.03087<br />
> ><br />
> > Step 21  Warning: pressure scaling more than 1%, mu: 0.849053 0.849053<br />
> > 0.849053<br />
> ><br />
> > Step 21  Warning: pressure scaling more than 1%, mu: 0.849053 0.849053<br />
> > 0.849053<br />
> ><br />
> > Step 22, time 0.088 (ps)  LINCS WARNING<br />
> > relative constraint deviation after LINCS:<br />
> > rms 0.005723, max 0.026407 (between atoms 6545 and 6542)<br />
> > bonds that rotated more than 30 degrees:<br />
> >  atom 1 atom 2  angle  previous, current, constraint length<br />
> ><br />
> > <br />
> ><br />
> > <br />
> ><br />
> > <br />
> ><br />
> > I hope this is enough information. Any help would be much appreciated.<br />
> ><br />
> > <br />
> ><br />
> > TJ Mustard<br />
> > Email: mustardt@onid.orst.edu<br />
> > Cell: 509-879-4173<br />
> ><br />
> > On September 9, 2010 at 9:11 PM "Justin A. Lemkul" <jalemkul@vt.edu> wrote:<br />
> ><br />
> >  ><br />
> >  ><br />
> >  > TJ Mustard wrote:<br />
> >  > ><br />
> >  > > First off I am using gromacs 4.5. I will also post all of my files and<br />
> >  > > errors if they help.<br />
> >  > ><br />
> >  > ><br />
> >  > ><br />
> >  > > If I run a protein in GROMOS96 all my md runs complete succesfully. But<br />
> >  > > if I change to any of the AMBER force fields I get LINCS errors in my<br />
> >  > > positional restraint md run. I have tried using shake, 1 fs step sizes,<br />
> >  > > -heavyh, and many more. Does anyone know what is going on here?<br />
> >  > ><br />
> >  ><br />
> >  > A complete (but not overly lengthy) post will save everyone a lot of<br />
> > time.<br />
> >  > Based on the information you've provided here, I see now way to<br />
> > diagnose the<br />
> >  > problem.  The most important information to post would be your .mdp file.<br />
> >  > Certain settings can influence stability.  A description of the hardware,<br />
> >  > compilers used, etc. can also be useful.<br />
> >  ><br />
> >  > -Justin<br />
> >  ><br />
> >  > ><br />
> >  > ><br />
> >  > > The reason I want to use AMBER is the fact that I want to run md on the<br />
> >  > > 30s rybosome and amber converts RNA much easier than GROMOS force<br />
> > fields.<br />
> >  > ><br />
> >  > ><br />
> >  > ><br />
> >  > ><br />
> >  > ><br />
> >  > > Thank you in advance,<br />
> >  > ><br />
> >  > ><br />
> >  > ><br />
> >  > > TJ Mustard Email: mustardt@onid.orst.edu<br />
> >  > > Cell: 509-879-4173<br />
> >  > ><br />
> >  > ><br />
> >  > ><br />
> >  > ><br />
> >  ><br />
> >  > --<br />
> >  > ========================================<br />
> >  ><br />
> >  > Justin A. Lemkul<br />
> >  > Ph.D. Candidate<br />
> >  > ICTAS Doctoral Scholar<br />
> >  > MILES-IGERT Trainee<br />
> >  > Department of Biochemistry<br />
> >  > Virginia Tech<br />
> >  > Blacksburg, VA<br />
> >  > jalemkul[at]vt.edu | (540) 231-9080<br />
> >  > http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin<br />
> >  ><br />
> >  > ========================================<br />
> >  > --<br />
> >  > gmx-users mailing list    gmx-users@gromacs.org<br />
> >  > http://lists.gromacs.org/mailman/listinfo/gmx-users<br />
> >  > Please search the archive at<br />
> > http://www.gromacs.org/Support/Mailing_Lists/Search before posting!<br />
> >  > Please don't post (un)subscribe requests to the list. Use the<br />
> >  > www interface or send it to gmx-users-request@gromacs.org.<br />
> >  > Can't post? Read http://www.gromacs.org/Support/Mailing_Lists<br />
> >  ><br />
> ><br />
> > <br />
> ><br />
> > <br />
> ><br />
><br />
> --<br />
> ========================================<br />
><br />
> Justin A. Lemkul<br />
> Ph.D. Candidate<br />
> ICTAS Doctoral Scholar<br />
> MILES-IGERT Trainee<br />
> Department of Biochemistry<br />
> Virginia Tech<br />
> Blacksburg, VA<br />
> jalemkul[at]vt.edu | (540) 231-9080<br />
> http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin<br />
><br />
> ========================================<br />
> --<br />
> gmx-users mailing list    gmx-users@gromacs.org<br />
> http://lists.gromacs.org/mailman/listinfo/gmx-users<br />
> Please search the archive at http://www.gromacs.org/Support/Mailing_Lists/Search before posting!<br />
> Please don't post (un)subscribe requests to the list. Use the<br />
> www interface or send it to gmx-users-request@gromacs.org.<br />
> Can't post? Read http://www.gromacs.org/Support/Mailing_Lists<br />
><br />
</div>
<p style="margin: 0px;"> </p>
<p style="font-family: monospace; white-space: nowrap; margin: 5px 0px 5px 0px;">TJ Mustard<br />
Email: mustardt@onid.orst.edu</p>
</body>
</html>