[gmx-users] Position restrain of structure

[Justin Lemkul]

On 6/1/16 11:22 PM, Md. Imrul Reza Shishir wrote:
> Dear all
> I have a structure file of cellulose nano fibril of 36 chain (each have 40
> residue). I want to simulate it as a single compound. I try to run the
> simulation with OPLS-AA force field. After energy minimization when i run
> the equilibrium step then the fibril structure broken and all the chain
> aligned in a single chain like structure. I want to restrain the position

If you can upload before and after images (to some file sharing service, and 
provide us a URL on the list) of what's going on, it would help.  This could 
either be an issue with (1) PBC, (2) setting up the box incorrectly, (3) 
incorrect topology, or (4) some combination of all of the above.

> of chain in there original sheet like structure. But I am  not able to
> define.
>
> I start with this pdb2gmx command .
>
> gmx_mpi pdb2gmx -f (filename).pdb -o (filename).gro -water spce
> -missing -merge all
>
> How can i define *chainsep* option in the pdb2gmx command.
>
> ------------------
> ATOM    842  OB  UN1    40      29.218  50.526  48.874  1.00  0.00
> ATOM    843  HOB UN1    40      28.409  50.951  48.646  1.00  0.00
>
> ATOM    844  HOA UN1    41      15.236  41.916 250.988  1.00  0.00
> ATOM    845  OA  UN1    41      15.877  41.231 251.063  1.00  0.00
> ------------------
>
> Residue 40 and 41 is the terminating and initial next of next chain.
> How i can define id_or_ter for *chainsep* option.
>

-chainsep creates distinct molecules based on TER delimiters or chain identifier 
characters.  See the PDB format documentation for more.

> In nvt, npt or md file i use
>
>
> -----
> define			= -DPOSRES	;
> refcoord_scaling    	= no		;
>
> --------
>
> for position restrain this is the right way. Or I have to do something
> else. Actually my main target to retrain the fibril structure and 36 chain
> are formed 7 sheet not a single chain like structure.
>

That should restrain the atoms, although the choice of refcoord_scaling may not 
be appropriate.  But two lines of an .mdp file is insufficient to judge whether 
or not you're doing things correctly.

-Justin

-- 
==================================================

Justin A. Lemkul, Ph.D.
Ruth L. Kirschstein NRSA Postdoctoral Fellow

Department of Pharmaceutical Sciences
School of Pharmacy
Health Sciences Facility II, Room 629
University of Maryland, Baltimore
20 Penn St.
Baltimore, MD 21201

jalemkul at outerbanks.umaryland.edu | (410) 706-7441
http://mackerell.umaryland.edu/~jalemkul

==================================================