[gmx-users] parameter on Energy Minimization

[Justin A. Lemkul]

Quoting Liu Shiyong <liushiyong at gmail.com>:

> Thanks Mark and  Justin.
>
>  I did experiment with the  following  editconf.    It works.  Thanks.
>
>   editconf -f ${file}.gro -o ${file}_box.gro -d 0.75 -bt cubic
>
>     By the way,  how to select box size automatically ?  I just gave an
> arbitrary number: 0.75

editconf -box

You can always check the positioning of your solute within the box with ngmx
prior to running your process, just to be sure.

>
>    Another question is how to get the structure with lowest energy  from
> *.trr ?
>
>   When I used the following command:
>    trjconv  -f ${file}.minim_traj.trr -s ${file}.input.tpr -o
> ${file}.minim_traj.pdb
>
>   It outputs 15 MODELS with   t= 125.00000 t= 252.00000, t= 379.00000 ,
> ...t= 2001.00000
>
>  I guess  t= 2001.00000 has the lowest energy.

Yep, that's what minimization does.  When the process converges, you should be
left with the structure of lowest potential energy, which should also be
produced as confout.gro, so you shouldn't have to extract the structure from
the .trr file.

-Justin

>
>   Best
>
>
>
> On Tue, Mar 11, 2008 at 5:58 PM, Mark Abraham <mark.abraham at anu.edu.au>
> wrote:
>
> > > Hi,
> > >
> > > I used a script given below for energy minimization of  a native
> > > protein-protein complex 1akj.
> > >
> > > I removed HETATMs and water molecules from the structure, and the
> > > minimized
> > > structure  (1akj_oplsaa.minim_traj.pdb) with epsilon_r=80.
> > > Then I used rasmol to check the result.  The ligand is ran away from
> > > receptor.
> > >  rasmol 1akj_oplsaa.minim_traj.pdb
> >
> > As Justin said, you might well be observing a PBC artefact. Having not
> > turned off PBC in your .mdp file, or used editconf to change the size of
> > the box that pdb2gmx chooses, then your EM is probably using defaults that
> > you'd rather not use.
> >
> > > Then I tried the larger epsilon_r           =  800000.0, and got very
> > > similar result. I mean that for two substantially different values of
> > the
> > > epsilon_r minimization gives absolutely the same final positions of the
> > > ligand.
> >
> > Being good scientists, we deduce that the approximate location of the
> > local energy minimum from this starting point is not dependent on the
> > dielectic of the medium. :-)
> >
> > > I also tried difference forcefield encads, encadv, G43a1, G43a2, G43b1,
> > > G45a3, G53a5, G53a6,  gmx, oplsaa.
> > > I also got similar result.
> >
> > Likewise.
> >
> > Mark
> >
> > _______________________________________________
> > gmx-users mailing list    gmx-users at gromacs.org
> > http://www.gromacs.org/mailman/listinfo/gmx-users
> > Please search the archive at http://www.gromacs.org/search before posting!
> > Please don't post (un)subscribe requests to the list. Use the
> > www interface or send it to gmx-users-request at gromacs.org.
> > Can't post? Read http://www.gromacs.org/mailing_lists/users.php
> >
>
>
>
> --
> Shiyong Liu
> Research Assistant
> center for bioinformatics in the university of kansas
> Lab: (785)864-1962
> Email: syliu at ku.edu (shiyongliu at ku.edu or liushiyong at ku.edu)
> Homepage: http://www.people.ku.edu/~syliu
> Lab: http://vakser.bioinformatics.ku.edu/people
> Phone: (785) 864-1962
>



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

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

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