FOXS: experimental curve does not fit predicted curve

Small angle scattering software (except ATSAS)
Post Reply
Message
Author
FeiCai_JGI
Guest
Posts: 2
Joined: 2011.11.15 02:36

FOXS: experimental curve does not fit predicted curve

#1 Post by FeiCai_JGI » 2011.11.15 02:42

Hi all, I'm new to SAXS and I had this question really bothers me. Let's say that I got SAXS data of a protein, and I do have pdb file for the same protein as well. When I generated predicted curve using FOXS server, it does not fit experimental SAXS curve! See in the attached image, at around 0.1 q-1 there is a deep in both experimental and predicted. But it's way deeper in predicted. Why is that? Can I still consider them a fit? Chi is around 5.

BTW, I did build in missing residues on both terminus in the pdb file I feed into FOXS.
Attachments
hisCcmP_formateNC_051211_hisP.png
hisCcmP_formateNC_051211_hisP.png (5.58 KiB) Viewed 6903 times

franke
Administrator
Posts: 408
Joined: 2007.08.10 11:09
Contact:

Re: experimental curve does not fit predicted curve

#2 Post by franke » 2011.11.15 15:21

FeiCai_JGI wrote:Can I still consider them a fit?
No.

FeiCai_JGI
Guest
Posts: 2
Joined: 2011.11.15 02:36

#3 Post by FeiCai_JGI » 2011.11.15 19:32

Then what could possibly cause that deeper drop?

User avatar
AL
Administrator
Posts: 686
Joined: 2007.08.03 18:55
Location: EMBL Hamburg, Germany
Contact:

#4 Post by AL » 2011.11.25 15:48

FeiCai_JGI wrote:When I generated predicted curve using FOXS server, it does not fit experimental SAXS curve!
Does CRYSOL give a better fit?
FeiCai_JGI wrote:Then what could possibly cause that deeper drop?
If we assume that you have a good sample (a monodisperse homogeneous solution) then the difference in the shape of your protein in solution and your model could be the reason for the bad fit.
FeiCai_JGI wrote:BTW, I did build in missing residues on both terminus in the pdb file I feed into FOXS.
If you've built many missing residues in a wrong way - the overall shape of the model might be wrong.

Post Reply