Assembly a multi-domain protein from available mono-domain?

[Sidamo]

Hello

I am new to SAXS, this is the first time of my atomic modeling, so maybe the questions I asked are really naive .
I'm trying to assembly a two domain proteins, which have available structure of each two domains, and I've collected the SAXS data of full-length protein.
But now, I'm not really sure which software to use.
According to https://bmcstructbiol.biomedcentral.com ... 6807-12-17
the BUNCH program is designed specifically for the multi-domain assembly task, so I tried to use it but it didn't work. ( I'm not sure the reason, but I've put the .pdb or .seq file in the bin folder, but the pre-bunch answered me "not exist" and crash.

In the other hand, I tried to use FoXSDock and fitted two mono-domain in SAXS data to build full-length atomic model, but the SAXS X score is too high....(higher than 30...)

Please give me some ideas, or is there any protocol or practicing about assembling a multi-domain proteins from available mono-domain?

Thank you very much.

[Sidamo]

This is the product of FoXSDock, and I'm wondering why the X score is so high....and how to fix it, and also it seems normal in c1 c2.



this is a two domain protein, and I got full-length SAXS experimental data.
One of these domains is available crystal structure from PDB, the other one is from swiss homologous modeling and I use UCSF chimera to build missing residues.
So the molecular weight is equal to the SAXS experimental sample which I used.

[AL]

Try SASREF online.

Step 1
Experimental data: SAXS (X-ray) only, standard
Number of curves: 1
Number of subunits: 2
Overall symmetry: P1

Step 2
Provide your experimental SAXS data and the two subunits.
Check 'Provide contact conditions'.

Step 3
Provide one contact condition by choosing the last residue of the first subunit and the first residue of the second subunit. It is possible to provide a range of residues, to specify one residue you have to choose the same residue twice.

[Sidamo]

It's embarrassed about the result.
I've tried several times, but the fitting is still terrible...


the distance constrain seems like it doesn't work or something wrong.
or maybe something wrong in my setting of parameter.

[SaxsMax]

It seems that the experimental data are in inverse angstroems,
but 1/nm is given as the input. Try to rerun with proper units.

[Sidamo]

Thank you for your reply!
but, I fit the units into volumetric map,
their fitting are not really bad like that.