Hi all,

Following some problems getting agreement with the InChI DLL as
described at 
http://baoilleach.blogspot.com/2010/10/measuring-information-loss-in-file_07.html
(specifally the CID15550 example), I've rewritten the handling of
cis/trans stereochem a little bit. There's a little bit more to go, so
I won't be checking it in today, but here's the before and after
picture:

Before:
(1) The current code does not perceive cis/trans stereo if a bond is
in a ring (and removes it if specified in the input).

After:
(1) Double bonds in rings are now treated as cis/trans stereo centers.
(2) If stereo is not specified for these bonds, they are considered
implicitly *cis* if in rings of size <= 12
(3) When writing SMILES containing double bonds in rings, the
cis/trans symbol is not written if in rings of size <= 12
(4) When writing InChIs containing double bonds in rings, the
cis/trans stereodescriptor will not be written for rings of size <= 7

The value of 12 is set by inspection of Daylight Depict output. The
assumption is that rings of size <= 12 cannot have a trans bond and
thus are implicitly cis. The value of 7 is from the InChI technical
manual.

Any thoughts before I check it in? Should I hold off on merging this
until after the release?

It is worth noting the the SMILES on the PubChem page for the compound
(http://pubchem.ncbi.nlm.nih.gov/summary/summary.cgi?cid=15550) does
not show any stereo symbol - and neither will Open Babel. However, the
InChI on that page is incorrect - it should show the stereochemistry
at the double bonds as the ring is of size 8.

- Noel

------------------------------------------------------------------------------
Download new Adobe(R) Flash(R) Builder(TM) 4
The new Adobe(R) Flex(R) 4 and Flash(R) Builder(TM) 4 (formerly 
Flex(R) Builder(TM)) enable the development of rich applications that run
across multiple browsers and platforms. Download your free trials today!
http://p.sf.net/sfu/adobe-dev2dev
_______________________________________________
OpenBabel-Devel mailing list
OpenBabel-Devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/openbabel-devel

Reply via email to