[gmx-users] changing protonation state of a residue during a simulation

[Justin Lemkul]

On 5/15/13 7:14 PM, Payman Pirzadeh wrote:
> Hello,
> I am simulating a protein using side chain pKa values estimated from
> PROPKA based on crystal structure of the protein. I re-estimated the pKa
> of side chains after running 10 independent replica of 50ns (500ns in
> total) of NPT simulations. It turned out that the average pKa of a
> surface Glu was overestimated in the crystal structure (protonated), and
> needs to be reconsidered (not protonated). I was wondering how I can
> re-assign the protonation state of this single residue without going
> through pdb2gmx (or may be I have to) to keep the current conformation
> of the protein and currently equilibrated solvent molecules? Is there a
> way to continue the current simulations I have and only change the
> charge of this single surface Glu?

No, because doing so (1) isn't possible and (2) doesn't make sense.  You're not 
just changing charges, you're deleting an atom.  You have to generate a new 
topology with pdb2gmx, simple as that.  You can easily just run the whole system 
through pdb2gmx to preserve the configuration (provided you don't have any other 
weird molecules involved that pdb2gmx won't like, but you can also work around 
that easily).  Preserving the coordinates is relatively insignificant compared 
to the fact that there is no way to preserve the previous ensemble.  You can't 
change the topology and then try to continue from a previous state (i.e., 
checkpoint file), so you have to re-equilibrate anyway.

-Justin

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

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

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