Other Frequently Asked Questions about ATSAS

Frequently Asked Questions about ATSAS small-angle scattering analysis program package
Post Reply
Message
Author
User avatar
AL
Administrator
Posts: 686
Joined: 2007.08.03 18:55
Location: EMBL Hamburg, Germany
Contact:

Other Frequently Asked Questions about ATSAS

#1 Post by AL » 2007.08.17 22:09

Please post a new question only if you can not find it on the forum.
Here are some frequently asked questions:

Q: Could you take a fast look to the two curves I'm sending to you. I measured them two weeks ago in Grenoble. I was trying to calculate the P(r) function. However when I get a nice fitting of the profile, the P(r) goes to negative values. In fact I can only get reasonable fittings when I cut the last part of the curve. Could it be that the small number of points could cause this problem?
A: Compare it yourself to the perfect p(r) functions. One can get nice fitting and P(r) by manually changing the Alpha parameter in GNOM.

Q: Can I specify the random seed in Dammin?
A: Yes, if you run an ab initio program (Dammin/Gasbor) in a batch mode you can use "/SD" key to input the desired seed.

Q: I was trying to run GASBOR on my data, and just realised that there is a limit of total residues being 8000 on a PC. Can you confirm if this is the total residues in the assymetric unit or is it total residues in the system itself. So does the program not support anything beyond 8000 residues?
A: 8000 is a limit of total residues + dummy waters. For systems bigger than 2000 AA it is much better to use Dammin or Dammif

Q: I thought that in Bunch I would need two rigid bodies or more and then add a missing part to that.
A: Although Bunch is aimed to rigid body modelling of multidomain proteins, it will also work to add one (or two) loop(s) to a single domain.

Q: Am I right to think that there is no way to have sasref-like contacts in Bunch ?... I have an example where two cysteines (modelled as dummy residues in Bunch) are known to form a dissulfide bridge ...
A: A Bunch version with contact restraints will be released soon (in ATSAS 2.2)

Q: I am trying to use Credo and Chadd for adding a missing part (approx 100 AA) to my model. I cannot really get anything meaningful out of it. When I run it I get with Credo that the missing part is flying around separated from the protein, and with Chadd, I get at long linker instead of the globular addition in one end. I have tried to change some of the penalty values in the expert mode, but not with any luck.
A: Given very anisometric shape of the protein it is better to use larger search volume. This can be done by running CREDO in the Expert mode and giving increased Dmax (e.g. by 10%) while using the same Gnom file.

Q: Do you have a linux compiled version of ASSA? I use a linux workstation and generally run MASSHA using the "WINE" emulation program but would prefer to run ASSA natively if possible.
A: Unfortunately, there is no Linux version of ASSA.

Q: I checked back and it seems that we have some protein in the freezer. How much would you need for a measurement at X33 and in what buffer?
A: Normally, samples are measure at at least three concentrations (e.g. 1, 2 and 5 mg/ml), the volume for one measurement is about 80 mkl so that about 1mg is required. The buffer should be the same as the dialysis one.

Q: What does ATSAS stand for?
A: ATSAS stands for All That Small Angle Scattering.

Post Reply