Trouble in data collection or analysis

Calculation of SAXS and SANS profiles (CRYSOL, CRYSON), superposition of models (SUPCOMB, DAMAVER, DAMCLUST), database DARA
Post Reply
Message
Author
kermani
Guest
Posts: 1
Joined: 2018.04.10 19:56

Trouble in data collection or analysis

#1 Post by kermani » 2018.04.10 22:19

I am quite new to SAXS. A colleague of mine collected the scattering data on my protein using 2 different protein concentrations. After initial data analysis it appeared to me something is not quite right. The bead model generated using DAMMIF and DAMMIN is strange and the scattering curve generated using the pdb file from a very close homologue does not fit at all.

At this stage I am not sure whether I made a mistake in my data analysis or the protein is aggregated or what?
I would highly appreciate your help and expertise.
Attachments
3b5d00.fit
The Crysol fit data
(66.58 KiB) Downloaded 12 times

nassos
Member
Posts: 5
Joined: 2017.05.27 10:34

Re: Trouble in data collection or analysis

#2 Post by nassos » 2018.04.11 15:34

Hi,

If your dataset is good then you need to consider if your protein adopts a different conformation in solution than the crystal structure of the homologue. Also did you collect SEC-SAXS data?
Maybe you can consider running DAMMIN for 10-20 times and average the output models so you will have a better idea of your protein model.

n.

User avatar
Hayds
Active member
Posts: 99
Joined: 2008.05.21 19:01
Location: EMBL, Hamburg

Re: Trouble in data collection or analysis

#3 Post by Hayds » 2018.04.20 12:44

Dear Kermani,

the issue could be the system itself. The data looks like it comes from a membrane protein in detergent, is this the case? If so, one thing you can do is extract only the "shape" scattering by using the data up to the first minimum in the curve as input for DAMMIN/DAMMIF. You need to consider, however, that the accuracy of such a model from a multi-contrast system is limited (but will give you an indicator of overall shape).

For calculation of the best fit model, if detergent molecules are part of the calculation and you are using CRYSOL, you will most likely need to make sure the PDB files are fully protonated (add Hydrogens) and run with the "explicit hydrogens" option.

If this is not a membrane protein system, I'm not sure what the problem is :)

Cheers,

Haydyn

Post Reply