Take a look at eginput/rna_refi. The scripts in there are all
classic xplor scripts, rather than python. But the techniques are
the same, regardless of the scripting language. My suggestions are:
1. Add basepair planarity restraints, using the "restraints plane"
command, as shown in eginput/rna_refi/plane.inp
2. Add the DELPHIC torsion potential. This makes a very large
difference in the precision of nucleic acid structures (which is
quite different from its behavior with protein structures). Nucleic
acid backbones are extremely flexible and usually poorly restrained
experimentally, so any torsion angle information can make a very big
difference.
3. Add the DELPHIC position potential, as shown in eginput/rna_refi/
rna_database_refine.inp. This uses a technique that Marius and I
published in JACS 125, 1518-1525. It can help to fill in base
stacking and so forth in a reasonable way.
A quick note of caution, though, is in order. The database of
nucleic acid xray structures is much smaller than that of proteins,
and so these potentials are therefore less reliable than those for
proteins. In particular, you shouldn't apply them to the residues in
the hairpin itself, since there are rather few hairpin residues in
our databases. We haven't found any cases in our work where the
database potentials for nucleic acids have created errors, but it is
possible. They should certainly help to get the stem of your
structure looking reasonable, though.
Hope this helps.
JK
On Dec 10, 2007, at 12:49 PM, Brendan Duggan wrote:
> Greetings
>
> I am new to Xplor and have been having difficulties trying to
> generate an RNA stem-loop as a starting structure for further
> refinement. I have been using the Python interface and have been
> able to generate a random extended structure using
>
> seqToPSF('rna.seq',seqType='rna',startResid=135)
> protocol.genExtendedStructure("rna_extended.pdb",verbose=1)
>
> However, when I apply distance restraints for the hydrogen bonds
> that define the base pairing, I get a globular structure rather
> than a stem-loop. Does anyone have any tips or tricks on how to
> get the bases to pair up nicely?
>
> Brendan
> --
> Brendan Duggan <bmduggan at musc.edu>
> Medical University of South Carolina
> _______________________________________________
> Xplor-nih mailing list
> Xplor-nih at nmr.cit.nih.gov
> http://dcb.cit.nih.gov/mailman/listinfo/xplor-nih
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
http://dcb.cit.nih.gov/pipermail/xplor-nih/attachments/20071210/e35e33fe/attachment.html
