Dear Matthias,
We do not have anything that allows tracking or assignment of
symmetry-related positions. I can see the need if you have a crystal or
fibril structure you want ot use, but I think this is the first time that
problem comes up. This wold have to be coded specifically, likely with a
model change. It definitely needs doing, as always the problem is when and
by whom.
If you do not have any structure, you can distinguish between intra-chain,
inter-chain, and intra-or-inter using only two chains. We did that in a
practical case for a symetrical pentamer. You create chains A and B as
identical, and use the convention that assignment A<->A means
intra-or-inter, assignment B<->B means intra-only, and A<->B means
inter-only. When you have finished assignment you then use a script to
expand this to a full, symmetric set of restraints for your entire
molecular system - in our case the pentamer, and pass that to youyr
structure generation program. If you are interested I can easily find the
script (or recreate it from scratch).
Best we can do at the moment, I am afraid,
Rasmus
---------------------------------------------------------------------------
Dr. Rasmus H. Fogh Email: [log in to unmask]
Dept. of Biochemistry, University of Cambridge,
80 Tennis Court Road, Cambridge CB2 1GA, UK. FAX (01223)766002
On Thu, 24 Nov 2011, Matthias Huber wrote:
> Dear all,
>
> I like that ccpn can import a pdb structure and that I can use this when
> analyzing peak assignments (i.e. after structure calculations). Would it
> be possible to divide the "Dist" column in the assignment panel to show
> the shortest intramolecular contact and also the shortest intermolecular
> contact?
>
> I noticed that I can only assign intermolecular peaks if I make two
> chains and copy all the shifts. Can I specify that chains have identical
> shifts? For a symmetric dimer it is somewhat inconvenient if all
> residues appear twice in the resonance list and the shift averaging is
> done independently for chain A and B. For fibrils or crystals with more
> than two symmetric monomers this gets even more messy.
>
> Do I have to define several chains to look at intermolecular contacts or
> could that be done directly from one molecule and the symmetry rules
> specified in the pdb file? (i.e. can I avoid doing symmetry expansion in
> PyMol, importing the expanded pdb into ccpn and copying the shifts to
> all chains?)
>
>
> Can somebody share some advice for a good workflow or could some features to handle symmetric multimers more easily be added to (maybe the next version of) ccpn?
>
> Thanks!
> Matthias
>
> ---
> Matthias Huber
> ETH Zurich
> Physical Chemistry
> HCI D220.2
> Wolfgang Pauli-Str. 10
> 8093 Zürich, Switzerland
>
|