Hello Charles, hello Ryan,
I am also working with a protein-ligand complex and also had some
trouble, but I was using Xplor-NIH and PRODRG. The protein structure in
presence of the ligand had been calculated before by ARIA/CNS, but my
structures kept diverging from that quite a lot. In addition, if I added
the ligand the planar parts of it significantly deviated from planarity.
I found solutions for both problems. First I found out that Xplor-NIH
didn't get the NOEs calculated by ARIA/CNS right. The problem is that
ambiguous or even unambiguous restraints, that are separated by an "or"
statement aren't recognized. Xplor-NIH kept reading only the first line
of such a statement and ignored the rest without a single sign of a
warning or something like that. If the input is e. g.:
assign (segid " " and resid 103 and name HA) (segid " " and
resid 32 and name HD22) 2.666 0.853 0.853 ! spec=13C NOESY, no=0,
id=1648
or (segid " " and resid 103 and name HA) (segid " " and
resid 32 and name HD21)
or (segid " " and resid 103 and name HA) (segid " " and
resid 32 and name HD23)
Xplor-NIH only read the first line (ignoring the two lines beginning
with "or") and went on to the next NOE without ANY sign of this behavior
except from a resulting structure that was quite different from the one
using ARIA/CNS.
The deviation from planarity could be stopped by simply adding
potList.add(XplorPot("DIHE"))
to the other other invocations
potList.add(XplorPot("ANGL"))
potList.add(XplorPot("BOND"))
potList.add(XplorPot("IMPR"))
and adding the DIHE-potential to the potlists. Maybe this is interfering
with the IVM and the torsion angle dynamics used, but the resulting
structures look much better to me. Perhaps someone can comment on that?
I hope this helps!
Andreas
--
Andreas Ceymann
Experimentelle Physik V, Physikalisches Institut
Universit?t W?rzburg, Am Hubland, 97074 W?rzburg
asceyman at physik.uni-wuerzburg.de
0049/931/888 5182
Am Montag, den 07.05.2007, 15:55 -0400 schrieb Charles at schwieters.org:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
>
> Hello Ryan--
>
> > I was working with a protein-ligand complex using CNS, way back when,
> > and have moved my calculations over to Xplor-NIH, in part to make use
> > of its facilities for multiple conformer interpretation of NOEs and
> > other refinement goodies.
> >
> > I generated my ligand parameters using Xplo2D. When implemented in
> > CNS I had no problems with the ligand's naphthalene ring deviating
> > from planarity. After scrutinizing the parameter file (and consulting
> > with Xplo2D's author, Gerard Kleywegt) I am quite certain the
> > planarity ought to be maintained in Xplor-NIH. Nonetheless, the
> > naphthalene ring is puckering in my current calculations.
>
> How is planarity maintained? Is it using the DIHE potential term? This
> term is unused in the Python protocols -- the IMPR term should be used
> instead.
>
> To maintain strict planarity, a call to the IVM can be used, assuming
> planarity is correct initiallly:
> ivm.group("planar atom selection")
>
> where "planar atom selection" selects all atoms which should be
> planar. Note that this should be called for each IVM object used
> [e.g. for the dyn and minc objects in eginput.gb1_rdc/anneal.py]
>
> hope this helps--
> Charles
> -----BEGIN PGP SIGNATURE-----
> Version: GnuPG v1.2.1 (GNU/Linux)
> Comment: Processed by Mailcrypt 3.5.8+ <http://mailcrypt.sourceforge.net/>
>
> iD8DBQFGP4QhPK2zrJwS/lYRAmM4AJ96TE2a97aMWWp2mmLjzU9q90BinACeM3Lc
> aChBbZWETRhytS//1O/YXio=
> =paXI
> -----END PGP SIGNATURE-----
> _______________________________________________
> Xplor-nih mailing list
> Xplor-nih at nmr.cit.nih.gov
> http://dcb.cit.nih.gov/mailman/listinfo/xplor-nih
