[gmx-developers] Regarding commit Implementation of WAXS/SAXS refinement in MD!

[Anjaiah Nalaparaju]

Thank you for providing a detailed explanation.

Actually I should focus only on asking about all atom form factors table
which is " sfactor_atom_nods_Fourier.xml ". In this there is no solvent
displaced correction so we no need to use any polynomial correction. I
suppose the coefficients in these table should be close to the cromer-mann
coefficients (from quantum calculations) or to be in a form to use in the
following equation to calculate atomic form factor.
[image: Inline image 1]

As I can see the coefficients given in the .xml file are very different I
would like to know what is the equation used to calculate atomic form
factor. In principle, whatever the method or coefficients we use the atomic
factor at q=0 will be equal to atom number. I am not clear how this is
satisfied from your coefficients given in " sfactor_atom_nods_Fourier.xml
". For example for atom "Carbon" I expect sum of a1 to a5 should be equal
to 6.

Please correct me if I have mistaken any concept in your implementation.

Thanks for your time and kind help.
Best regards




On Mon, Oct 13, 2014 at 6:26 PM, Alexander Björling <alex.bjorling at gmail.com
> wrote:

>
> From: Anjaiah Nalaparaju <anjai.che at gmail.com>
>>
>> My question is related to the unmerged patch for module waxsdebye.
>>
>> 1. Can anybody tell the reference for the form factor
>> coefficients(a1-a5,b1-b5,p1-p3) in file "sfactor_atom_nods_Fourier.xml" or
>> all the new .xml files on sfactors. These number look very much different
>> from the sfactor.dat which are cromer-mann coefficients.
>>
> There is no reference (yet). The numbers are different because it's a
> different parametrization. Since solvent-corrected form factors have a more
> complex shape, we've used a short Fourier sum plus a polynomial, instead of
> gaussians. The form factor is calculated from these parameters in
> src/gromacs/waxsdebye/scattering_factors.cpp.
>
>
>>
>> 2. If we use the following equation to calculate atomic form factors, at
>> q=0  sum of co-eff a1 to a5 should be equal to the form factor value at
>> q=0
>> given in the publication "Niebling, Bj?rling and Westenhoff, J. Appl.
>> Cryst. 47 (2014) 1190"
>>
> No they shouldn't, there's also a polynomial there. Again, see
> scattering_factors.cpp. Have that code print out the scattering factors if
> you would like to test them.
>
>
>> [image: View the MathML source]
>>
>> But seems the coefficients in .xml files are not corresponding to this
>> equation.
>>
>> 3. How can the coefficients after displaced volume correction can be a
>> constant value ( sfactor_atom_ds_Fourier.xml), as the displaced volume
>> correction term also function of "q". "Svergun, Barberato and Koch, J
>> Appl.
>> Cryst. 27 (1995) 768."
>>
> I'm not sure what you mean by "constant value".
>
>
>>
>> 3. Where the form-factors with displaced volume corrections are
>> calculated
>> in the code.
>>
> The displaced volume corrections are not done in the code. The scattering
> factors with or without the correction (depending on which xml file is
> used) are calculated in scattering_factors.cpp.
>
>
> --
> Gromacs Developers mailing list
>
> * Please search the archive at
> http://www.gromacs.org/Support/Mailing_Lists/GMX-developers_List before
> posting!
>
> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
>
> * For (un)subscribe requests visit
> https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-developers
> or send a mail to gmx-developers-request at gromacs.org.
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-developers/attachments/20141013/1e3304fe/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: image.png
Type: image/png
Size: 3547 bytes
Desc: not available
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-developers/attachments/20141013/1e3304fe/attachment.png>