On Sat, Mar 17, 2012 at 3:04 AM,  <[email protected]> wrote:
> "O=C1Oc6ccccc6(C(O)C1C5c2ccccc2CC(c3ccc(cc3)c4ccccc4)C5)"

OK, I updated the unit test for that one.

> Btw - fun to try your code (either with the modified algorithm or the 
> original CDK version) on the Smiles for
> Acridine: "c1c2ccccc2nc3c1cccc3".  It usually returns 7 double bonds but, 
> occasionally (1 in 30+ occasions),
> it only returns 6 as discussed previously.

Yeah, that is not a good sign... I hate unpredictable outcomes... :/

Also, this is the SMILES Ola and I were discussing earlier, in
relation to Bioclipse:

COc1ccc2[C@@H]3[C@H](COc2c1)C(C)(C)OC4=C3C(=O)C(=O)C5=C4OC(C)(C)[C@@H]6COc7cc(OC)ccc7[C@H]56

This SMILES works in neither the original code, nor in your modified code...

This is a good example of the advantage of just looking at that parts
of the molecule where something should happen, which are here two
isolated phenyl rings... since this still one big ring system, your
patch does not help a bit in this example...

I have pushed a new branch: https://github.com/egonw/cdk/commits/371-14x-dbst

Kevin, it has your patch, and two unit tests.

While it does not solve all limitations of the code base, it should at
least be a step forward.

Rajarshi (or someone else), can you review the patches for cdk-1.4.x?
Then I can release CDK 1.4.8 this weekend with this patch, and a few
more already in the cdk-1.4.x branch.

Egon

-- 
Dr E.L. Willighagen
Postdoctoral Researcher
Department of Bioinformatics - BiGCaT
Maastricht University (http://www.bigcat.unimaas.nl/)
Homepage: http://egonw.github.com/
LinkedIn: http://se.linkedin.com/in/egonw
Blog: http://chem-bla-ics.blogspot.com/
PubList: http://www.citeulike.org/user/egonw/tag/papers

------------------------------------------------------------------------------
This SF email is sponsosred by:
Try Windows Azure free for 90 days Click Here 
http://p.sf.net/sfu/sfd2d-msazure
_______________________________________________
Cdk-user mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/cdk-user

Reply via email to