Refining dummy atom positions

Linear (OLIGOMER), and non-linear (MIXTURE) analysis, singular value decomposition (SVDPLOT), addition of missing fragments (BUNCH, CORAL), analysis of flexible systems (EOM/RANCH & GAJOE), flexible refinement of high-resolution models (SREFLEX)
Post Reply
Message
Author
sdw
Guest
Posts: 2
Joined: 2010.08.18 16:34
Location: K.U. leuven

Refining dummy atom positions

#1 Post by sdw » 2010.09.01 14:28

Hi, I have recently collected scattering curves of a protein that I believe is forming homo-tetramers. I believe this because we have collected SAXS data for a truncated version of the protein, for which have tetrameric crystal structure, which gives a excellent chi*2 value in crysol. This same model fits the SAXS data for the full-length protein very poorly in part, I assume, because the individual monomers are missing approximately 50 residues off their N-termini and 10 residues of the C-termini. The caveat here is that we expect these missing sequences to be flexible and unstructured.
Obviously I would like to use EOM to build an ensemble of chains that would best fit the data, but if I understand correctly EOM needs (assumes) that you have a monomer of linked structured subunits, rather than an oligomer. As an alternative I have crudely used Ranch to generate a pdb file with random coil extensions for each chain. (Taking the eomtmp.pdb file generated), and then stitched these models together to generate a very crude model. This dummy atom extended model gives a better chi*2 fit to my SAXS data (much better than the truncated crystallographic model as well as a CHADD build on this), but could clearly be improved. Therefore I'm wondering if this model can be fed into any ATSAS programme to refine the dummy atom positions against the scattering curve (or more likely the P(r) plot) whilst maintaining correct connectivity of said dummy atoms and avoiding obvious steric clashes?

(Apologies for the long winded question)

Alex
Active member
Posts: 697
Joined: 2007.08.09 21:10
Location: Planet Earth

#2 Post by Alex » 2010.11.24 12:50

Hi

Do you also have a data for a monomer?

Cheers
Alex

sdw
Guest
Posts: 2
Joined: 2010.08.18 16:34
Location: K.U. leuven

monomer

#3 Post by sdw » 2010.11.24 14:28

Nothing for the monomer, as it likely does not exist in solution. I do have SAXS curve for the truncated tetramer (minus most of the floopy regions) as well as a further truncated dimer.

Alex
Active member
Posts: 697
Joined: 2007.08.09 21:10
Location: Planet Earth

#4 Post by Alex » 2010.11.24 16:14

Ok. I see. I wonder whether unstructured parts can be on oligomer interface?
Alex

User avatar
Hayds
Active member
Posts: 103
Joined: 2008.05.21 19:01
Location: EMBL, Hamburg

#5 Post by Hayds » 2010.11.25 23:09

Hi,

why not use BUNCH and symmetry to generate the oligomer? You can also use the more recently developed program CORAL (in the latest ATSAS package).

I'm not sure why you need to use EOM? EOM should be used to assess flexibility and the degree of compaction of the construct, the models are only part of the tool such that Rg distributions can be created.

I would use your defined monomer as a rigid body in CORAL, position the initial structure such that when the appropriate symmetry operation is applied you reproduce the expected tetramer. The missing portions of the structure will then be modelled using dummy residue chains.

Hope it helps!

Haydyn

Post Reply