Hi Egon, thanks for your answer. Glad to hear that this is a known issue, because it troubles me a lot.
> The CDK algorithm for aromaticity defines rings with double bonds > pointing out of the ring as non-aromatic; for this compound to be > aromatic, it would require a charge separation on the double bonds, > which you can reflect in the chemical graph. Doing that, should make > the compound aromatic. I there a way to do that automatically? > BTW, the CDK is very open to alternative algorithmic definitions of > 'aromaticity'. We currently only have one 'standard' implemented, and > for long I have wanted to implement other standards too. Not that I am > aware of any algorithm being standardized on, or even just published. Yes, it would be great to have some "compatibility mode" towards OpenBabel. I am working on a piece of software that uses OpenBabel, CDK and Jmol, and aromaticity causes a lot of problems. Best regards, Martin On Tue, Feb 21, 2012 at 12:09 PM, Egon Willighagen <[email protected]> wrote: > Hi Martin, > > On Tue, Feb 21, 2012 at 11:39 AM, Martin Guetlein > <[email protected]> wrote: >> I found another example ( O=C1C=CNC(=O)N1 ) that is not aromatic >> according to CDK, but according to daylight >> (http://www.daylight.com/daycgi/depict?4f3d4331433d434e43283d4f294e31) >> and OpenBabel. Is this a bug? > > No, it is not a bug. It is a difference in the algorithmic definition > of aromaticity: > > http://apps.ideaconsult.net:8080/ambit2/depict?search=O=C1C=CNC(=O)N1 > > And, incidentally, exactly what I talked about in my talk for NBIC > last Thursday [0]. > > The CDK algorithm for aromaticity defines rings with double bonds > pointing out of the ring as non-aromatic; for this compound to be > aromatic, it would require a charge separation on the double bonds, > which you can reflect in the chemical graph. Doing that, should make > the compound aromatic. > > Compare the above depictions with these: > > http://apps.ideaconsult.net:8080/ambit2/depict?search=%5BO-%5D%5BC%2B%5D1C%3DCN%5BC%2B%5D%28%5BO-%5D%29N1 > > The fact that the CDK did not mark that representation as aromatic, > would be a bug (and it should be filed). > > Nina, if you are reading this, can I request a similar depiction > service, visualizing the atom types labels for each atom? > > BTW, the CDK is very open to alternative algorithmic definitions of > 'aromaticity'. We currently only have one 'standard' implemented, and > for long I have wanted to implement other standards too. Not that I am > aware of any algorithm being standardized on, or even just published. > > Aromaticity is one of the most misunderstood concepts in > cheminformatics :) (Not claiming I understand all aspects of it...) > > Egon > > 0.http://chem-bla-ics.blogspot.com/2012/02/chemical-interoperability.html > > -- > 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 -- Dipl-Inf. Martin Gütlein Phone: +49 (0)761 203 7633 (office) +49 (0)177 623 9499 (mobile) Email: [email protected] ------------------------------------------------------------------------------ Keep Your Developer Skills Current with LearnDevNow! The most comprehensive online learning library for Microsoft developers is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3, Metro Style Apps, more. Free future releases when you subscribe now! http://p.sf.net/sfu/learndevnow-d2d _______________________________________________ Cdk-user mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/cdk-user
