[gmx-users] About frozen groups

[Justin A. Lemkul]

Mohammed Kamal wrote:
>  
> Many thanks for Justin for your helpful comments, I think I still need 
> your help for the first question... I have included more details about 
> my system below
>  
>  
>  
>  > Mohammed Kamal wrote:
>  >
>  > > Tcoupl = berendsen
>  > >
>  > > tau_t = 0.1 0.1 0.1 0.1
>  > >
>  > > tc_grps = protein T27 SOL Cl ; T27 is the ligand
>  >
>  > General advice: do not couple solvent and ions separately. I believe 
> there is a
>  > newer version of the tutorial that fixes this information. See here:
>  >
>  > http://wiki.gromacs.org/index.php/thermostats
>  
> Thanks much for that advice
> 

Another thing that I just noticed as I scrolled through.  "Cl" corresponds to a 
chloride ion under ffgmx.  This force field is long since deprecated, and should 
not be used for new simulations.  Go with a newer Gromos96 force field.

>  >
>  > >
>  > > ref_t = 300 300 300 300
>  > >
>  > > *energygrps = protein T27 SOL *
>  > >
>  > > *energygrp_excl = IRS IRS IRS SOL*
>  > >
>  > > *freezegrps = IRS*
>  > >
>  > > *freezedim = Y Y Y*
>  > >
>  > > Pcoupl = berendsen
>  > >
>  > > pcoupltype = isotropic
>  > >
>  > > tau_p = 0.5
>  > >
>  > > compressibility = 4.5e-5
>  > >
>  > > ref_p = 1.0
>  > >
>  > > gen_vel = yes
>  > >
>  > > gen_temp = 300
>  > >
>  > > gen_seed = 173529
>  > >
>  > >
>  > >
>  > > and when I used the grommp line of
>  > >
>  > > *grompp* -f md.mdp -c 2zd1_eq.gro -p 2zd1.top -n IRS.ndx -o md.tpr
>  > >
>  > > I had (as expected) an error massage states that IRS is not an energy
>  > > group however if I have added IRS as an energy group I had an error 
> that
>  > > some atoms are in multiple energy Mon. groups.
>  > >
>  >
>  > Well, what is IRS? Is it a protein, another molecule? You've got very
>  > different groups labeled for energy monitoring and energygrp-excl, so 
> something
>  > needs to be fixed. Without any clue what exactly you're dealing with, 
> it's hard
>  > to give any useful information.
>  >
>  
>  
> 
> IRS is a group of residues in the protein, suppose that my protein has 
> 50 residues, IRS is a group that contains residues 2, 7, 9, 15, 45 for 
> example. So the IRS group is actually part of my protein
> 
> Now I have four groups the SOL, ligand (T27), protein and IRS which 
> overlapped with the protein (this is the problem) ...
> 
> I need to fix the position of IRS residues and exclude the non-bonding 
> interactions between these residues and the solvent. Gromacs will not 
> accept IRS as an energy group because there will be some residues that 
> are common with the protein and so I can't state it in the energy 
> exclusions??
> 
>  

No, since IRS and your Protein have common (bonded) interactions, you cannot 
simply specify an exclusion to separate IRS from the protein.

The following is just thinking out loud, others please feel free to comment 
since I have never tried anything like this: It may be possible to specify a 
modified "Protein" index group that contains all residues NOT including your IRS 
residues (type "help" at the make_ndx prompt to see examples of clever syntax). 
  These groups could be merged for the purpose of T-coupling (i.e., 
Protein_IRS), but then specified differently for energygrps and energygrp_excl.

Worth a shot, but as I said, I've never done anything like that.

> 
> That's fine... so if I just add unconditional #include "POSRE_IRS.itp" 
> to the .top file Gromacs will apply postion restraints even without 
> adding the define line to the .mdp file?

Correct, just be careful that artificial restraints do not influence the 
dynamics of interest.

-Justin

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

Justin A. Lemkul
Graduate Research Assistant
Department of Biochemistry
Virginia Tech
Blacksburg, VA
jalemkul[at]vt.edu | (540) 231-9080
http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin

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