[gmx-users] g_mindist after trjconv -fit give different results

Tsjerk Wassenaar tsjerkw at gmail.com
Mon Dec 7 09:42:27 CET 2009

Hi Elad,

You can't count on operations that involve PBC after fitting. In your
case, you can use

trjconv -fit translation

to remove translational degrees of freedom only, keeping the system
consistent with PBC.

Hope it helps,


On Sun, Dec 6, 2009 at 8:50 PM,  <eladp at post.tau.ac.il> wrote:
> Hi Tsjerk,
> thanks for the reply.
> Will the MSD of the molecule also be affected by the fitting.
> I'm trying to calculate the relative diffusion coefficient of an ion near a
> protein. For this I thought to first fix the system according to the protein
> and then calculate the msd of the ion.
> I noticed in the g_msd code that periodicity is taken into account (as
> expected). However, in g_msd the interest of the periodicity handling
> regards regards the atom position relative to itself in previous frames.
> Will this relative movement (of the atom in relation to itself in the
> previous frame) be kept after doing "trjconv -fit".
> Thanks.
>> Hi Elad,
>> This has been mentioned several times before, in a different context.
>> But fitting only affects coordinates, and not the PBC. So after
>> fitting the coordinates and the PBC don't match any more and any
>> analysis requiring PBC, such as calculating minimal distances, will
>> give garbled results. Always do such things on untransformed
>> trajectories.
>> Cheers,
>> Tsjerk
>> On Fri, Dec 4, 2009 at 7:43 PM,  <eladp at post.tau.ac.il> wrote:
>>> Hi all,
>>> I have a system of protein+water+ions.
>>> I measured the minimum distance between one of the ions to the protein.
>>> Afterwards I fitted the trajectory using g_trjconv -fit rot+trans
>>> Did the same mindist again and the graphs are not always identical.
>>> Any suggestions?
>>> I use gromacs 4.0.5 with an octahedron box.
>>> same problem happens when I do trjconv -fit rot but does not happen when
>>> -fit trans.
>>> In VMD I do see a difference in the position of the ion relative to the
>>> protein.
>>> Thanks.
>>> Elad P.
> --
> gmx-users mailing list    gmx-users at gromacs.org
> http://lists.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 thewww 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.

Computational Chemist
Medicinal Chemist

More information about the gromacs.org_gmx-users mailing list