[gmx-users] trjconv -pbc

[Tsjerk Wassenaar]

Hi Mauricio,

That this seems to work best for you is, sorry to say, "just between
the ears". I'm not completely sure, but I think the implementation was
done correctly, so as to first deal with pbc related options, before
fitting. In that case, the results would be the same as running
trjconv twice, once with -pbc nojump and once with -fit rot+trans.
Now I may actually be wrong, in which case fitting is done first (I
should check the code to be sure). In that case, your procedure will
be equal to running trjconv -fit rot+trans first, followed by trjconv
-pbc nojump. But this is very very bad practice! Rotations only apply
to the coordinates and not to the lattice. That means that any option
regarding periodic boundaries will give incorrect results after
fitting with rotations (or rotations per se). Rotation  of coordinates
causes the coordinates and the lattice to be detached from each other.
Translation is okay, since this does not change the orientation with
respect to the lattice.

I hope I have made myself clear. Otherwise, you may dig up some other
post of mine from the archive in which I have given the same story in
a bit different wording. Also, give some thoughts to periodic boundary
conditions, lattice vectors, and particle coordinates within periodic
systems.

Cheers,

Tsjerk


On 6/21/07, Mauricio Pablo Sica <msica at unq.edu.ar> wrote:
> > Hello all,
> >
> > I have tried using trjconv -pbc(all options) to get a continuous
> > trjactoary visulations on VMD.  I am modeling an 11 residue receptor
> > antiparellel to itself and one of the receptor strand jumps outside the
> > box sometimes.
> >
> > I have used search to solve the problem but everthing i tried didnt
> > work
> > and the -pbc just makes it worse.
> >
> > I tried using the -pbc nojump alone and this made my two receptor
> > strands
> > away from each other all the way (one inside and one outside the box).
> > However i want them to be close together since they are hydrogen
> > bonded.
> >
> > I have also tried: nojump then whole; whole then nojump and they make
> > the
> > receptor just fly around or pieces of it stretch.
> >
> > anyhelp is appreciated.
>
> I have best resutls when using try '-fit rot+trans' in combination with
> '-pbc nojump'. If this is useless, I generate a new starting topol.tpr
> where the monomers are in the same cell and rerun the simulation. Then,
> trjcovn with '-fit rot+transand -pbc nojump'.
>
> Good luck
>
> >
> > Belquis
> >
> ________________________________________________________________
>
> Dr. Mauricio P. Sica LEPP
> Phone: (54011)4365-7100 ext. 169   Laboratorio de Expresión
> R. Saenz Peña 354 y Plegado Proteico
> (B1876BXF) Bernal, Buenos AiresUniversidad de Quilmes
> Argentina
> ________________________________________________________________
>
>
>
>
> _______________________________________________
> 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
>


-- 
Tsjerk A. Wassenaar, Ph.D.
Junior UD (post-doc)
Biomolecular NMR, Bijvoet Center
Utrecht University
Padualaan 8
3584 CH Utrecht
The Netherlands
P: +31-30-2539931
F: +31-30-2537623