[gmx-users] Re: which charge is used in gromacs program

[Mark Abraham]

On 1/09/2011 10:34 AM, Dr. Vitaly V. Chaban wrote:
> On Wed, Aug 31, 2011 at 8:23 PM, Justin A. Lemkul<jalemkul at vt.edu>  wrote:
>>
>> Dr. Vitaly V. Chaban wrote:
>>>> Hi, ALL
>>>>
>>>>      I have a question about the charge defined in the topology file.
>>>>      It seems to me that the charge for each atom apprears twice in
>>>> different sections of the topology files.
>>>>
>>>> ------------------------------------------------------------------------------
>>>>      One is
>>>> [ atomtypes ]
>>>> ;name     mass      charge   ptype   c6            c12
>>>>
>>>>      The other is
>>>> [ atoms ]
>>>> ; id    at type res nr  residu name at name cg nr   charge
>>>>
>>>> ------------------------------------------------------------------------------
>>>>       And the charge number in [ atomtypes ] are always 0.0.
>>>>       Which one is actually read by gromacs program ?
>>>
>>> That one which is mentioned the last
>>>
>> That is true in the case where grompp finds multiply defined types for
>> bonded interactions and atom types; the charges in ffnonbonded.itp are never
>> used by any program.  The reason for their existence is not clear, but
>> perhaps they were used by older versions or were intended for some
>> streamlined force field organization that never came to fruition.
>>
>
> I believe, this column is intended for aminoacids for which there
> should be a consistent set of parameters including charges. This is
> what we have in AMBER.

However that only works if the charge is the same for that atom type in 
each possible residue. That's not a very effective constraint.

Mark