[gmx-users] Assigning mass to a virtual site

[David van der Spoel]

On 2014-10-29 13:19, Justin Lemkul wrote:
>
>
> On 10/29/14 6:32 AM, Kester Wong wrote:
>> Dear all,
>>
>>
>> I am creating a topology file for hydroxide, closely following the
>> parameters
>> from the paper by Wolf et al:
>>
>> http://dx.doi.org/10.1016/j.bpj.2014.04.062
>>
>>
>> This is somewhat similar to the polarisable hydroxide topology that
>> was included
>> in the recent SWM4-NDP force field.
>>
>>
>> In this non-polarisable topology, I would like to know what atomic
>> number should
>> one label for the OH-OV2 type atom, as it was defined as a virtual
>> site (but
>> with 1 a.m.u assigned) in the paper's supporting material. I have
>> labeled the
>> atomic number with "X" as below:
>>
>>
>> [ atomtypes ]
>>
>> ;name   at.num  mass    charge  ptype   sigma   epsilon
>>
>> OH-O    8       11.9980         -1.32   A       0.00            0.00
>>
>> OH-H    1       1.00800          0.00   A       0.14            0.2150576
>>
>> OH-OV1  0       0.00000          0.00   D       0.2916          0.2150576
>>
>> OH-OV2  X       1.00000          0.00   A       0.25            0.2150576
>>
>> OH-qH   0       0.00000          0.32   D       0.00            0.00
>>
>>
>> Can an atom (ptype=A) have zero at.num?
>>
>
> I would suggest you simply ask the authors to provide you with their
> topology. There are several details that are unclear to me, and I do not
> see anything that says the V1 or V2 sites carry any mass.  Maybe I
> missed it due to a quick read-through, but this is in any case a very
> tricky setup.  Don't reinvent the wheel; get the information straight
> from the people that did it.
>
> -Justin
>
The topology is correct as downloaded from virtualchemistry.org.


-- 
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