[gmx-users] Problem with using distance restraints (disre) with PBC
TJ Piggot
t.piggot at bristol.ac.uk
Thu Mar 6 13:16:31 CET 2008
Hi i have modified the src to stop harmonic potentials being put into the
graph (as was suggested to me by Berk Hess, see below for the modification
you need to make). I would be imagine this is also possible to do for
distance/orientation restraints, however i only know a small amount of
programming so you will need someone else to help you here. Here is the
change i have made and it works fine for me:
A real fix would then be adding
i!=F_HARMONIC &&
to the if statement on line 272 of src/gmxlib/mshift.c
and recompiling Gromacs, such that the harmonic potentials
are not put into the graph.
Tom
--On Thursday, March 06, 2008 12:50:18 +1000 Mitchell Stanton-Cook
<s4026869 at student.uq.edu.au> wrote:
> Hi Tom,
>
> Have you modified the src? As far as I know this has not been done in 331
> or 332. I'm unsure of 333. I would be interested in the patch. As far as
> I know the problem is due to the the distance/orientation restraints
> being placed in the bonded graph. There was a bit of chatter on the
> developers list in 2007 about this but I haven't heard much else.
>
> For my system I was using a truncated dodecahedron box. The box vector
> 1/2d was my biggest problem. I simply changed to a triclinic box without
> too much cost. However I think I could grab an extra 300ps/day if I had
> an actual fix.
>
>
> Cheers
>
> Mitch
>
> TJ Piggot wrote:
>> I agree with increasing the size of the box (if seeing inconsistent
>> shifts, as i should have mentioned in my previous post) and in this
>> case it is practical to do as the system is quite small. However (just
>> for future reference) if the problem were in a much larger system (as
>> it is in my case) then it may be more suitable to modify the source
>> code to avoid increasing the size of the system simulated.
>>
>> Tom
>>
>> --On Thursday, March 06, 2008 11:52:01 +1000 Mitchell Stanton-Cook
>> <s4026869 at student.uq.edu.au> wrote:
>>
>>> I have had a similar problem with orientation restraints.
>>>
>>> Are the residues far is space? Are you getting inconsistent shifts?
>>> If so
>>> -
>>>
>>> If so re-define your box size so that the shortest box vector is greater
>>> than the distance restraint residue + some tolerance.
>>>
>>> I also noticed -
>>>
>>> [ distance_restraints ]
>>> ;ai aj type index type low up1 up2 fac
>>> 9 413 1 0 1 8.18 8.18 10.0
>>> 1.0
>>>
>>> Have you defined the low up and up2 in the correct units?
>>>
>>> Cheers
>>>
>>> Mitch
>>>
>>> TJ Piggot wrote:
>>>> Hi,
>>>>
>>>> Firstly I would strongly reiterate what Mark said. I would make sure
>>>> the distance restraints are causing your problem. Again as Mark said
>>>> make sure you can simulate your system without the distance
>>>> restraints, and then I would try to simulate the system with distance
>>>> restraints and with a much larger box of water surrounding your
>>>> peptide. Here you should also be able to overcome the problem (if it
>>>> is this problem) with a large enough distance between periodic images.
>>>>
>>>> If you can then confirm what you have suspected to be the problem you
>>>> will need to modify the source code slightly. If you search the list
>>>> (searching for "inconsistent shifts using multiple bonds type 6") you
>>>> can see the changes you need to make, as suggested by Berk Hess to me
>>>> when i had this problem with bond type 6.
>>>>
>>>>
>>>> Hope this helps
>>>>
>>>> Cheers
>>>>
>>>> Tom
>>>>
>>>> --On Wednesday, March 05, 2008 17:16:23 -0500 Grace Li
>>>> <fullofgrace88 at gmail.com> wrote:
>>>>
>>>>> Hello,
>>>>> I am having some trouble using distance restraints. My goal is to
>>>>> apply a
>>>>> harmonic potential to the distance between two atoms in a short
>>>>> peptide
>>>>> (35 residues). Doing simulations in vacuo, both of the following
>>>>> techniques worked to restrain the distance. Also, both techniques
>>>>> worked
>>>>> for simulations in water without periodic boundary conditions. As soon
>>>>> as I include pbc = xyz in my .mdp file and use PME for the
>>>>> electrostatics, my simulation crashes (producing step0.pdb and
>>>>> step-1.pdb). I am assuming this is because GROMACS is using the
>>>>> distance
>>>>> to the image of the other atom, not the actual distance in the box? Is
>>>>> there any way of getting around PBC for distance restraints?
>>>>>
>>>>> Specifically, I have used the following:
>>>>>
>>>>> 1. I have tried using bonds type 6 in my topology:
>>>>> [ bonds ]
>>>>> ; ai aj funct c0 c1 c2 c3
>>>>> 9 413 6 8.18 421.54
>>>>>
>>>>> 2. When using distance restraints, I added the following lines to my
>>>>> topology file:
>>>>> [ distance_restraints ]
>>>>> ;ai aj type index type low up1 up2 fac
>>>>> 9 413 1 0 1 8.18 8.18 10.0
>>>>> 1.0
>>>>>
>>>>> In the case of using [ distance_restraints ] in the topology file, I
>>>>> have
>>>>> added the following lines to my .mdp file:
>>>>> disre = simple
>>>>> disre_fc = 421.54
>>>>> disre_weighting = equal
>>>>> disre_tau = 0
>>>>> (and various other values of the force constant, disre_fc, but in all
>>>>> cases the run crashes).
>>>>>
>>>>> Thanks for any suggestions!
>>>>> _______________________________________________
>>>>> 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
>>>>
>>>>
>>>>
>>>> ----------------------
>>>> TJ Piggot
>>>> t.piggot at bristol.ac.uk
>>>> University of Bristol, UK.
>>>>
>>>> _______________________________________________
>>>> 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
>>>>
>>>
>>> _______________________________________________
>>> 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
>>
>>
>>
>> ----------------------
>> TJ Piggot
>> t.piggot at bristol.ac.uk
>> University of Bristol, UK.
>>
>> _______________________________________________
>> 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
>>
>
> _______________________________________________
> 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
----------------------
TJ Piggot
t.piggot at bristol.ac.uk
University of Bristol, UK.
More information about the gromacs.org_gmx-users
mailing list