[gmx-users] Simulation of protein at specific pH

[David van der Spoel]

On 2011-04-24 20.47, Justin A. Lemkul wrote:
>
>
> prerna bhardwaj wrote:
>> Dear Gromacs User,
>>
>> I want to simulate my protein at a specific pH. Using H++ server I
>> have generated my .pdb file at pH 5. But pdb2gmx is showing all the
>> usual fatal error "like "Atom HB3 in residue MET 1 not found in rtp
>> entry with 17 atoms". But I can't use -ignh option, because it is
>> bringing back at the same normal stage. So what should I do I want to
>> carry out simulation at pH 5.
>>
>
> It sounds like you should indeed be using -ignh because you have a
> malformed input. Under no conditions should the beta-carbon of
> methionine ever have three protons. Carbon atoms can't have five bonds.
It doesn't have three protons but the official (IUPAC/IUPAB) side chain 
atom naming is: CG (counts as 1) HB2 HB3, so the H++ server is correct, 
and the braindead force fields are wrong.

> Either the atom naming in your .pdb file is wrong (which will cause
> pdb2gmx to fail) or the protonation state is wrong (which will cause
> pdb2gmx to fail in the absence of -ignh).
>
> You can set the protonation state of any titratable residues and termini
> using pdb2gmx -inter, in concert with -ignh, which appears to be
> necessary in your case.
>
> -Justin
>
>> Thanks
>> Prerna
>>
>


-- 
David van der Spoel, Ph.D., Professor of Biology
Dept. of Cell & Molec. Biol., Uppsala University.
Box 596, 75124 Uppsala, Sweden. Phone:	+46184714205.
spoel at xray.bmc.uu.se    http://folding.bmc.uu.se