Dear Christoph,
---------------------------------------------------------------------------
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 Wed, 12 Apr 2006, Christoph Brockmann wrote:
> > If the structure calculation engines could pass their resolutions back
> > then the problem would be solved. As Rasmus points, out a consistent
> > mapping of Hga* to Hg2 etc in creating constraints would allow you to
> > check where stereochemistries (but not atom names) have been swapped -
> > i.e. the IUPAC stereochemistry is wrong. However this wouldn't work for
> > all calculation engines, and is not exactly future-proof.
> >
> I my hands, XPLOR (CNS), ARIA and CYANA treat these cases by swapping
> the stereochemistry. Leaving you with wrong stereochemistry but
> consistent naming of atoms...
True, but the data model is not happy with that option - non-NMR users
can assume that stereochemistry is correct. We can swap the constraint
assignment instead. This is more correct (and the newest ARIA version does
it that way too, says Wolfgang). Remains the problem of how to get hold of
(and store) the information that a given prochiral has been swopped and
that it is intended rather than a program error.
Yours,
Rasmus
>
> Christoph
> > Is there any other way to treat unresolved prochiral violations? Take the
> > closest? Something else?
> >
> > I am happy to go with the community in any event...
> >
> > T.
> >
> --------------------------------------------------------------------
> Christoph Brockmann
> Leibnitz-Institut fuer Molekulare Pharmakologie (FMP)
> Robert Roessle Str. 10
> D-13125 BERLIN, Germany
> phone: +49-30-94793-223
> email: [log in to unmask]
> --------------------------------------------------------------------
>
|