[gmx-users] pdb2gmx with input topology

[Mark Abraham]

On 4/28/2011 10:25 PM, Justin A. Lemkul wrote:
>
>
> Ehud Schreiber wrote:
>> Dear Gromacs users,
>>
>>
>>
>> I wish to convert PDB files to gromacs format while specifying the 
>> topology; 

That is not the role of pdb2gmx. pdb2gmx generates a topology that 
matches a configuration (which could be in one of many file formats), 
and (as a side effect) produces a "cleaned up" version of that 
configuration (which again could be in one of many file formats).

If you already have your configuration and a .top file that has one or 
more [moleculetype] sections that match the molecules present in your 
configuration, you do not need pdb2gmx.

Perhaps this will help you reconsider whether your workflow is appropriate.

Mark

>> however, in the pdb2gmx program, a .top file can only be an output, 
>> not an input.
>>
>> The need arises as I have an NMR PDB structure (including hydrogens) 
>> which I rather violently perturb into multiple PDB structures using 
>> some non-gromacs means. When I convert those structures to .gro using 
>> pdb2gmx, sometimes the protonation state of Histidine is guessed 
>> incorrectly by the program (the hydrogen moves between HD1 and HE2).
>>
>
> If the atoms are properly named, I don't see how this is possible, 
> unless you're using the -ignh option.  If the atomic coordinates are 
> severely skewed, you should at least get a warning about long bonds.
>
>> Does anybody have a suggestion what can I do? I need the process to 
>> be automatic and so can’t use the interactive -his option.
>>
>>
>
> So is the .top the same in all cases, and you're just altering 
> coordinates?  If so, pdb2gmx is not necessary at all.  Just use the 
> .top you want in conjunction with any of the .pdb files.  But perhaps 
> I've missed the point.  There is definitely no way to input a .top 
> into pdb2gmx since its principal function is to produce such a file.
>
> -Justin
>
>>
>> Thanks,
>>
>> Ehud.
>>
>>
>>
>