Re: [Cdk-devel] [Cdk-user] Detect stereocenters of a molecule

[Christoph S. <ste...@eb...>]

gilleain torrance wrote:
> I don't think that morgan numbers (which as of last Friday, I now
> understand :) can determine the stereochemistry of a carbon center
> bonded to four atoms.

We are talking about the Morgan numbers of the neigbors, of course, and
not of the center.
You would need to put the element symbols into account, but I took that
for granted :-)

> Other cases, I am not so sure, but consider the two Br/Cl/F/I-carbons
> - they must necessarily both have the same number of bonds, so the
> same morgan numbers. At least with the implementation that we
> currently have, which only takes bonds into account.

See above.

Cheers,

Chris

-- 
Dr. Christoph Steinbeck
Head of Chemoinformatics and Metabolism
European Bioinformatics Institute (EBI)
Wellcome Trust Genome Campus
Hinxton, Cambridge CB10 1SD UK
Phone +44 1223 49 2640

What is man but that lofty spirit - that sense of enterprise.
... Kirk, "I, Mudd," stardate 4513.3..