[gmx-users] using position restraints
Ken Rotondi
ksr at chemistry.umass.edu
Mon Nov 1 17:27:06 CET 2004
Found my problem and simultaneously identified a typo in the GROMACS
3.2 manual.
On page 129 under "Preprocessing" default define options the option to
include posre.itp is shown as -DPOSRE. However, the #ifdef statement in
the .top file is POSRES. As a result the proper include option is
-DPOSRES. Yes, I should have seen this sooner looking at both the .top
and .mdp files, but sometimes you just can't see the nose on your face.
I tried to post this awhile back, but it was when the list server was
acting funky.
Thanks to all for the help,
Ken
On Oct 19, 2004, at 5:03 PM, Oliver Beckstein wrote:
> Hi Ken,
>
> haven't got a magic answer. You could run 'grompp -debug ...' and
> check that the output contains something like
>
> converting bonded parameters...
> ...
> # LJ14: 7539
> # POSRES: 1898 <---------
> # CONSTR: 3840
> ...
>
> and then look at grompp.log if your posre.itp gets picked up. Perhaps
> it
> is some problem with the path. I assume you have posre.itp in the same
> directory as the top file?
>
>> Thanks for the quick response. Not at all trivial to me. My .top file
>> does indeed have the following lines:
>>
>> #ifdef POSRES
>> #include "posre.itp"
>> #endif
>
> looks good to me (if the flag is -DPOSRES)
>
>>
>> I'm running on a Mac, so I'm not sure if the missing "./" prior to
>> "posre.itp" is significant.
>
> Neither do I, I'm running on Linux.
>
>>
>> However, I do not have a line in the gmxdump of my .tpr file of the
>> form
>>
>> functype[N]=POSRES
>>
>> Changed the line in my .top file to
>>
>> #include "./posre.itp"
>
> Odd, I'd thought that would do it. Does Mac cpp encode paths
> differently? Or are you using m4 or something else instead as
> preprocessor (I am absolutely clueless when it come to Macs)?
>
>> and re-ran grompp. Still no line of type
>>
>> functype[N]=POSRES
>>
>> So I clearly don't have any positional restraints. Any other ideas?
>
> Not really. As an example, my top [1] (see below) contains itp files
> for
> separate chains, and each itp [2] itself contains a statement for
> restraining the backbone atoms only [3] (which is activated with
> -DPOSRES_BACKBONE in the mdp, see below to see what I mean).
>
> Not sure if this helps you in any way, but I can't think of why your
> setup is not working.
>
> oli
>
> ============================================================
> [1] top
> ============================================================
>
> ; Include forcefield parameters
> #include "ffG43a1.itp"
> #include "methaneslab.itp"
>
> ; Include chain topologies
> #include "protX_P1_A.itp"
> #include "protX_P1_B.itp"
> #include "protX_P1_C.itp"
> #include "protX_P1_D.itp"
> #include "protX_P1_E.itp"
>
> ; Include water topology
> #include "spc.itp"
>
> #ifdef POSRES_WATER
> ; Position restraint for each water oxygen
> [ position_restraints ]
> ; i funct fcx fcy fcz
> 1 1 1000 1000 1000
> #endif
>
> #ifdef WITH_IONS
> #include "ions.itp"
> #endif
>
> [ system ]
> ; Name
> Protein
>
> [ molecules ]
> ; Compound #mols
> Protein_A 1
> Protein_B 1
> Protein_C 1
> Protein_D 1
> Protein_E 1
> LAY 484
> SOL 2837
> ------------------------------------------------------------
>
> ============================================================
> [2] excerpt from protX_P1_A.itp
> ============================================================
> [ moleculetype ]
> ; Name nrexcl
> Protein_A 3
> ....
> [ atoms ]
> ...
> [ bonds ]
> ...
>
> ; Include Position restraint file
> #ifdef POSRES
> #include "posre_A.itp"
> #endif
>
> ; restraining backbone atoms only: (N|CA|C)
> #ifdef POSRES_BACKBONE
> #include "posres_backbone_protX_P1_A.itp"
> #endif
> ------------------------------------------------------------
>
> ============================================================
> [3] posres_backbone_protX_P1_A.itp
> ============================================================
> ; restraining backbone atoms only: (N|CA|C)
> ; generated from protX_P1_A.itp by genprchain.pl
> [ position_restraints ]
> 3 1 1000 1000 1000
> 6 1 1000 1000 1000
> 12 1 1000 1000 1000
> 14 1 1000 1000 1000
> 16 1 1000 1000 1000
> ...
> 277 1 1000 1000 1000
> 281 1 1000 1000 1000
> ------------------------------------------------------------
>
>
>
>
>>
>> Ken
>>
>> On Oct 19, 2004, at 3:46 PM, Oliver Beckstein wrote:
>>
>>> Hi Ken,
>>>
>>>> simulating. The restraint file, posre.itp is generated from pdb2gmx.
>>>> Upon viewing this file I found that the force constants at all
>>>> positions were set to x,y,z = 1000,1000,1000. I changed this such
>>>> that
>>>> a single atom had a restraint force constant of 1000,1000, 1000 and
>>>> the
>>>> rest were 0,0,0. I modified my .mdp file to include define = -DPOSRE
>>>> and ran a short test simulation.
>>>
>>> Sorry to ask a trivial question but have you also got lines in your
>>> topology like
>>>
>>> #ifdef POSRE
>>> #include "./posre.itp"
>>> #endif
>>>
>>> The -DPOSRE in the mdp file only instructs the preprocessor to set
>>> the
>>> flag POSRE, which in turn switches on the if branch in the top file.
>>> (The
>>> topology is processed by cpp so anything that goes for cpp goes for
>>> top.)
>>>
>>> The final test is analysing your tpr file, using gmxdump.
>>> gmxdump -s md.tpr > md.txt
>>>
>>> and search in md.txt for
>>>
>>> functype[N]=POSRES
>>>
>>> where N is the internal number of your restrained atom (might require
>>> some
>>> detective work top find it). It tends to be in the last quarter of
>>> the
>>> rather huge output. If it's not there then you don't have position
>>> restraints.
>>>
>>> hth
>>> oli
>>>
>>> --
>>> Oliver Beckstein * oliver at biop.ox.ac.uk
>>> http://sansom.biop.ox.ac.uk/oliver/
>>>
>>> _______________________________________________
>>> gmx-users mailing list
>>> gmx-users at gromacs.org
>>> http://www.gromacs.org/mailman/listinfo/gmx-users
>>> Please don't post (un)subscribe requests to the list. Use the
>>> www interface or send it to gmx-users-request at gromacs.org.
>>
>> _______________________________________________
>> gmx-users mailing list
>> gmx-users at gromacs.org
>> http://www.gromacs.org/mailman/listinfo/gmx-users
>> Please don't post (un)subscribe requests to the list. Use the
>> www interface or send it to gmx-users-request at gromacs.org.
>>
>>
>
> --
> Oliver Beckstein * oliver at biop.ox.ac.uk
> http://sansom.biop.ox.ac.uk/oliver/
>
> _______________________________________________
> gmx-users mailing list
> gmx-users at gromacs.org
> http://www.gromacs.org/mailman/listinfo/gmx-users
> Please don't post (un)subscribe requests to the list. Use the
> www interface or send it to gmx-users-request at gromacs.org.
More information about the gromacs.org_gmx-users
mailing list