[gmx-users] enegry minimisation

[Mark Abraham]

----- Original Message -----
From: sonali dhindwal <sonali11dhindwal at yahoo.co.in>
Date: Friday, May 21, 2010 15:38
Subject: Re: [gmx-users] enegry minimisation
To: jalemkul at vt.edu, Discussion list for GROMACS users <gmx-users at gromacs.org>

-----------------------------------------------------------
| > hello Jusitn,
> Thanks for your reply,,
> I am sending you the link,
so that you can see the changes in the protein, I have specifically
shown that part of the protein, where I am seeing changes,
> 
> http://picasaweb.google.co.in/sonali11dhindwal/Protein?feat=directlink
> 
> 
> Yello
beeta sheets are of the protein after EM, and magenta are that of the
refrence structure, you can see how this time,I am surprised myself
that sheets have become longer than the refrence structure.
> I have corrected that       define      =  -DPOSRES also.
> and have taken that posres file generated by pdb2gmx.is this the problem of the visualiser I m using, I am using pymol for it

I wouldn't describe what I can see there as significant change. The criterion any visualization program uses for assigning beta-strand vs not-strand is a reasonably arbitrary choice, and might differ between programs. It's quite plausible that EM could see a residue here and there cross a boundary if they were originally close.

Looking at all-atom or heavy-atom RMSD before and after EM (and maybe RMSD also over the subset of atoms of direct interest) is a much more reliable indicator that nothing has gone wrong. Off the top of my head, for a protein with no disordered regions, I'd expect EM on an experimental structure and normal force field to move about or less than 0.02nm on heavy-atom RMSD. All the EM is really expected to do is (e.g.) make sensible the hydrogen atoms whose positions were inferred or generated.

Mark