From: Axel Kohlmeyer (akohlmey_at_gmail.com)
Date: Wed Mar 12 2014 - 00:35:52 CDT

adam,

before speculating and guessing too much, i would recommend, you look
up the proper definition in a statistical mechanics or MD simulation
or simple liquids text book. in short, the radial distribution
function is the "relative" probability to find an atom of type b at a
given distance from an atom of type a. both types can be the same,
different or overlap. an ideal gas gives an g(r) that is 1.

in practice, you loop over all atoms in the first group and then
create a distance histogram of all atoms in the second group (which
can be identical). the GPU accelerated implementation in VMD has some
additional tricks and gimmicks and they are explained in this
publication:

 Fast Analysis of Molecular Dynamics Trajectories with Graphics
Processing Units: Radial Distribution Function Histogramming, B.G.
Levine, J.E. Stone, A. Kohlmeyer, J. Comput. Phys. 230(9), 3556-3569
(2011)

HTH,
     axel.

On Wed, Mar 12, 2014 at 12:55 AM, Adam Goler <adamgoler_at_gmail.com> wrote:
> Josh,
>
> So does the function ignore all atoms that are not part of the selection for
> the purposes of calculation?
>
> Is the radial pair distribution function determined by counting the number
> of atoms within a spherical shell as the radius increases from the location
> of selection 1 to selection 2?
>
> Thanks,
> Adam
>
>
> On Tue, Mar 11, 2014 at 8:00 PM, Josh Vermaas <vermaas2_at_illinois.edu> wrote:
>>
>> Hi Adam,
>>
>> The way I understand the gofr function to work is that there is one
>> selection that selects the particles that you will be measuring from, and
>> the second selects the particles you will be measuring to. So if the first
>> selection was "protein" and the second was "water", you'd be measuring the
>> gofr for water around the protein. Usually these sorts of selections are
>> hard to interpret, since lazy selections like I made above also include
>> buried atoms that may not be close to solution, and therefore don't
>> contribute until further out in the distribution, and lead to strange shapes
>> that are difficult to interpret.
>>
>> -Josh Vermaas
>>
>>
>> On 03/11/2014 08:03 PM, Adam Goler wrote:
>>
>> Hello,
>>
>> I'm curious about how this function works. If I understand correctly, a
>> simple radial distribution function loops over particles, and counts every
>> particle within r +/- dr for all r, where r is increasing radial distance
>> (then normalizes).
>>
>> What confuses me about the gofr gui plugin is that it allows one to choose
>> two selections. How does the recipe change to include the second function?
>> Does one merely perform the basic radial distribution function calculation
>> with relative displacements?
>>
>> Cheers,
>> Adam
>>
>> --
>> Adam Goler
>> Graduate Student
>> Fulmer 126
>> Dept. of Physics and Astronomy
>> Washington State University
>> PO Box 642814
>> Pullman, WA 99164-2814
>> Office Phone: (509) 335 7050
>>
>> adamgoler.com
>>
>>
>
>
>
> --
> Adam Goler
> Graduate Student
> Fulmer 126
> Dept. of Physics and Astronomy
> Washington State University
> PO Box 642814
> Pullman, WA 99164-2814
> Office Phone: (509) 335 7050
>
> adamgoler.com

-- 
Dr. Axel Kohlmeyer  akohlmey_at_gmail.com  http://goo.gl/1wk0
College of Science & Technology, Temple University, Philadelphia PA, USA
International Centre for Theoretical Physics, Trieste. Italy.