[gmx-developers] ensemble code and restraints

[hessb at mpip-mainz.mpg.de]

Hi,

You can select any group in the molecule for determining the matrix R.
The orientation of molecules is never restrained.
I still don't see why you want to do that.
Once you have the matrix R, you can apply any restraints
on the rotated system, without rotating back the system.
That all there is.
You will have to code something yourself if you want to do more.
I guess you can do most things you want when you have access to R.

Berk

> OK I have almost found the thing that I want except there is one snag,
> it is not as flexible as I would like it to be. In the manual in the
> orientation description this appears:
>
> For reasons which are explained below, the D matrices are calculated
> which respect to a reference orientation of the molecule. The
> orientation is defined by a rotation matrix R which is needed to
> least-squares fit the current coordinates of a selected set of atoms
> onto a reference conformation. The reference conformation is the
> starting conformation of the simulation. In case of ensemble
> averaging, which will be treated later, the structure is taken from
> the first subsyste.
>
> would it be possible to define the RMS restraint in this way first
> calculating the rotation matirx R and then applying it? Then derriving
> forces based on the rmsd to the average structure? Also would it be
> possible to define the selection of atoms required for this
> least-squares fitting? I would like to just apply the restraint to the
> secondary structures of my system.
>
> Cheers,
>
>
> Bryn
>
>
>
>
> 2008/12/8 Robert Fenwick <robert.fenwick at irbbarcelona.org>:
>> I am not up to speed with optimizations of code, however I would have
>> thought that it was not so computationally intensive to calculate the
>> average coordinates at each step and then determine velocities based
>> on the atomic displacements? However I will look at the manual before
>> I ask more questions.
>>
>> 2008/12/8 Erik Lindahl <lindahl at cbr.su.se>:
>>> Hi,
>>>
>>> We don't have direct RMSD/cRMS restraints, partly because it would be
>>> costly
>>> to calculate on-the-fly in parallel runs, but there are some pretty
>>> advanced
>>> ways in which you can apply distance restraints in molecules (also
>>> ensemble-based ones) in order to maintain your cRMS. It's described in
>>> detail in the written manual!
>>>
>>> Cheers,
>>>
>>> Erik
>>>
>>>
>>>
>>>
>>> On Dec 8, 2008, at 9:13 AM, Robert Fenwick wrote:
>>>
>>>> OK that sounds great, I will wait for this and in the mean time work
>>>> on my second question. Can someone tell me, maybe Berk knows best, if
>>>> there is a simple RMSD restraint in GROMACS over an ensemble. I would
>>>> like to keep all the molecules aligned during the simulation.
>>>>
>>>> Also can one select sub samples of the ensemble over which distance
>>>> restraints are applied?
>>>>
>>>> Thanks for the help regarding force fields,
>>>>
>>>> Bryn
>>>>
>>>>
>>>> 2008/12/8 Erik Lindahl <lindahl at cbr.su.se>:
>>>>>
>>>>> Hi,
>>>>>
>>>>> Please contact Pär Bjelkmar (bjelkmar at cbr.su.se) if you need a charmm
>>>>> port;
>>>>> starting with gromacs 4.0.2 (or something like that, but it's in the
>>>>> release
>>>>> branch) there is code in  pdb2gmx and grompp to correctly work with
>>>>> the
>>>>> multiple torsions terms used in Charmm.
>>>>>
>>>>> The force field files will be added to CVS shortly too, but we'd love
>>>>> to
>>>>> have a "complete" version including CMAP before pushing it out into
>>>>> the
>>>>> world to avoid confusion  (that's also the only reason why we'd like
>>>>> Pär
>>>>> to
>>>>> have a list of people using the preliminary version)
>>>>>
>>>>> Cheers,
>>>>>
>>>>> Erik
>>>>>
>>>>>
>>>>> On Dec 8, 2008, at 3:54 AM, Mark Abraham wrote:
>>>>>
>>>>>> Daniel Larsson wrote:
>>>>>>>
>>>>>>> On Dec 7, 2008, at 11:15 , Robert Fenwick wrote:
>>>>>>>>
>>>>>>>> Hi,
>>>>>>>>
>>>>>>>> I have two questions.
>>>>>>>>
>>>>>>>> Firstly are the AMBER (99SB) and CHARMM (CMAP) force fields full
>>>>>>>> integrated in GROMACS 4.0?
>>>>>>>
>>>>>>> If you are interested to use the Amber force field with Gromacs,
>>>>>>> this
>>>>>>> page is a most valuable resource:
>>>>>>> http://chemistry.csulb.edu/ffamber/
>>>>>>> It is a port of the force field which originally was made for the
>>>>>>> Folding at Home-project.
>>>>>>> When it comes to the Charmm force field, I'm not aware of any good
>>>>>>> port.
>>>>>>
>>>>>> Well there is a CHARMM port or two (I wrote a partial one; it's in
>>>>>> the
>>>>>> contributions section of the GROMACS webpage) but invoking it is
>>>>>> trickier
>>>>>> than with other forcefields because of quirks in CHARMM and
>>>>>> limitations
>>>>>> of
>>>>>> grompp.
>>>>>>
>>>>>> Mark
>>>>>> _______________________________________________
>>>>>> gmx-developers mailing list
>>>>>> gmx-developers at gromacs.org
>>>>>> http://www.gromacs.org/mailman/listinfo/gmx-developers
>>>>>> Please don't post (un)subscribe requests to the list. Use the www
>>>>>> interface or send it to gmx-developers-request at gromacs.org.
>>>>>
>>>>> _______________________________________________
>>>>> gmx-developers mailing list
>>>>> gmx-developers at gromacs.org
>>>>> http://www.gromacs.org/mailman/listinfo/gmx-developers
>>>>> Please don't post (un)subscribe requests to the list. Use thewww
>>>>> interface
>>>>> or send it to gmx-developers-request at gromacs.org.
>>>>>
>>>> _______________________________________________
>>>> gmx-developers mailing list
>>>> gmx-developers at gromacs.org
>>>> http://www.gromacs.org/mailman/listinfo/gmx-developers
>>>> Please don't post (un)subscribe requests to the list. Use the
>>>> www interface or send it to gmx-developers-request at gromacs.org.
>>>
>>> _______________________________________________
>>> gmx-developers mailing list
>>> gmx-developers at gromacs.org
>>> http://www.gromacs.org/mailman/listinfo/gmx-developers
>>> Please don't post (un)subscribe requests to the list. Use thewww
>>> interface
>>> or send it to gmx-developers-request at gromacs.org.
>>>
>>
> _______________________________________________
> gmx-developers mailing list
> gmx-developers at gromacs.org
> http://www.gromacs.org/mailman/listinfo/gmx-developers
> Please don't post (un)subscribe requests to the list. Use the
> www interface or send it to gmx-developers-request at gromacs.org.
>