[gmx-users] problem fitting trajectory
Daniel Adriano Silva M
dadriano at gmail.com
Thu Jun 11 02:16:59 CEST 2009
Hi,
This is a recurrent issue. I have figured a bash script that had
worked for every system which I had tested (i.e. protein,
protein-ligand, protein-DNA, protein-protein), but this is not a
warranty. It centers the protein on the box, and at the same time puts
waters inside the box ; however the apparent effect is that the water
box moves around "a center".
Test it, and please advice if it does the work and/or comment (maybe
developers could integrate a similar filter in a future gmx_trjconv
release, or just publish something similar to this in the wiki).
Best.
Daniel
***********
Usage notes:
- Just copy/and/paste the code on a textfile and make it executable
with "chmod 700", !!!take care of broken/split lines by email/post
artifacts!!!
- I use mdrun with -deffnm for the outputs; therefore this script is
made to work when all the output files have the same name.
- Do not forget to replace the *** field for the GROBIN variable on the code.
- On the code 0 and 1 are the index groups for system and protein
respectively. However, for any complex the trick to have a nice
representation is to center the system only on one of the proteins,
for example, for me:
echo -e "del 1-14\n r 1-238\n r 1-239\n q\n" |
$GROBIN/make_ndx -f pre1_$NAME.gro -o pre1_$NAME.ndx
makes the trick, since it erases all the groups from my index, except
system, and then creates the group 1 only for one of my proteins
(residues 1-238) and additionally the group 2 used to output my
complex in the absence of waters (residues 1-239).
To run use:
./scriptname
PATH-TO-DINAMIC/MDOUTPUT-NAME-WITHOUT-DOT-OR-EXTENSIONS OUTPUT-NAME
Example:
./filterscript /home/dsilva/dynamic/production/md filteredmd
***********
CODE:
***********
#!/bin/bash
#Daniel Silva, script to filter trajectories....
PATH=$1
NAME=$2
GROBIN=/usr/local/gromacs/bin #MODIFY THIS LINE TO POINT TO
YOUR GROMACS BINARIES PATH
/bin/cp $PATH.tpr ./$NAME-filtered.tpr
echo -e "0\n" | $GROBIN/trjconv -f $PATH -s
$NAME-filtered.tpr -pbc whole -o pre1_$NAME
echo -e "0\n" | $GROBIN/trjconv -f $PATH -s
$NAME-filtered.tpr -pbc whole -o pre1_$NAME.gro -e 0
### Modify the next line if you have any kind of complex, in order to
make group 1 pointing only to one of proteins in the complex.
(Example: echo -e "del 1-14\n r 1-238\n r 1-239\n q\n" |
$GROBIN/make_ndx -f pre1_$NAME.gro -o pre1_$NAME.ndx)
echo -e "q\n" | $GROBIN/make_ndx -f pre1_$NAME.gro -o pre1_$NAME.ndx
echo -e "1\n0\n" | $GROBIN/trjconv -f pre1_$NAME -s
$NAME-filtered.tpr -pbc atom -ur compact -center -n pre1_$NAME.ndx -o
pre2_$NAME
echo -e "1\n0\n" | $GROBIN/trjconv -f pre2_$NAME -s
$NAME-filtered.tpr -fit rot+trans -n pre1_$NAME.ndx -o $NAME-filtered
echo -e "0\n" | $GROBIN/trjconv -f $NAME-filtered -s
$NAME-filtered.tpr -e 0 -n pre1_$NAME.ndx -o $NAME-filtered.gro
### If you also want to output trajectories without waters uncomment
the next two lines. Also, if you have complexes take care of the group
that you are sending to the output (Example, for me it needs to be
group 2 not 1; echo -e "1\n2\n" and echo -e "2\n", respectively for
the next two lines)
# echo -e "1\n1\n" | $GROBIN/trjconv -f pre2_$NAME -s
$NAME-filtered.tpr -fit rot+trans -n pre1_$NAME.ndx -o
$NAME-filtered-NoHOH
# echo -e "1\n" | $GROBIN/trjconv -f $NAME-filtered -s
$NAME-filtered.tpr -e 0 -n pre1_$NAME.ndx -o $NAME-filtered-NoHOH.gro
/bin/rm ./pre*_$NAME.*
************
###CODE END
************
2009/6/10 Tsjerk Wassenaar <tsjerkw at gmail.com>:
> Hi,
>
> It is better to do PBC options first and fitting options after, with
> separate calls to trjconv. Fitting and PBC don't go well together,
> This has been elaborately discussed on the mailing list before.
>
> Cheers,
>
> Tsjerk
>
> On Wed, Jun 10, 2009 at 8:59 PM, Justin A. Lemkul<jalemkul at vt.edu> wrote:
>>
>> I have found that the combination of -pbc mol -ur compact always looks
>> pretty :) So in addition to -fit and perhaps -center, try those two options
>> together.
>>
>> -Justin
>>
>> Rebeca García Fandiño wrote:
>>>
>>> Hello,
>>> I have done a simulation using Gromacs 4 (4.0.2) and I would like to have
>>> a trajectory were the protein is fitted to the first structure (to mantain
>>> the exact orientation).
>>>
>>> When I have tried
>>>
>>> echo 1 1 0 | trjconv -f production_1_3.xtc -s production1.tpr -center -fit
>>> rot+trans -pbc mol -o production_1_3_fit.xtc
>>>
>>> visualizing production_1_3_fit.xtc I can see that the protein is fitted,
>>> however there are a lot of "holes" into the water box. It seems like the pbc
>>> were not correctly applied. I have tried changing a lot of options (-pbc
>>> res, -boxcenter rect,...) but nothing works.
>>>
>>> The options
>>>
>>> trjconv -f production_1_3.xtc -s production1.tpr -center -boxcenter rect
>>> -fit rot+trans -pbc mol -o production_1_3_fit.xtc
>>>
>>> do not produce any hole in the box of water, but the box entirely moves
>>> around the protein, and sometimes it is outside of it and not solvated.
>>>
>>> I have also tried using Gromacs 4.0.4, but the problem is still the same.
>>>
>>> Could anybody give an idea of how solving it, please?
>>>
>>> Thank you very much for your help.
>>>
>>> Best wishes,
>>>
>>> Rebeca Garcia
>>> Academic Visitor
>>> Oxford University
>>>
>>> ------------------------------------------------------------------------
>>> Charlas más divertidas con el nuevo Windows Live Messenger
>>> <http://download.live.com>
>>>
>>>
>>> ------------------------------------------------------------------------
>>>
>>> _______________________________________________
>>> 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 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
>>
>> --
>> ========================================
>>
>> Justin A. Lemkul
>> Ph.D. Candidate
>> ICTAS Doctoral Scholar
>> Department of Biochemistry
>> Virginia Tech
>> Blacksburg, VA
>> jalemkul[at]vt.edu | (540) 231-9080
>> http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin
>>
>> ========================================
>> _______________________________________________
>> 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 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
> _______________________________________________
> 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 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
>
More information about the gromacs.org_gmx-users
mailing list