John, Siva, all, I would not worry too much about these questions (see also [0]).
1. in the end you need to check individual molecules, study the 3D structure, bond lengths, etc. That's how you determine if they are "really" aromatic. Some of this you can predict with QM approaches, but even then hard 2. literature has no clear definition of aromaticity, and different domains may have different opinions. As a trained organic chemist, I learned that a ring current must take place. But in John's last example, there are still different ring current possible. Again, only experimental methods (e.g. NMR) can give the important info you determine what is real 2. there are no generally working heuristic rules, but there are models. The CDK had two, and since John recoded them, a few more. This involves heuristics like what atoms can be involved, how many electrons do they contribute, how large the ring systems are that you still want to count the number of "ring electrons" for, etc, etc Practically, there is no single cheminformatics "aromaticity" model that fits all. There are always exceptions. Something would then be a bug in the code, when the calculated aromaticity does not match the followed *model*, not reality. I am also not aware of studies that really compare computational (cheminformatics) models for aromaticity with reality. Also, I am not really aware (nor have recently looked for) data sets of experimental data (like NMR) that can be used to estimate how well an aromaticity model reproduces reality. Where we should matter is how "aromaticity" is used in our cheminformatics. In many cases the downstream cheminformatics does not really need to know if an atom is "aromatic" (or bond) and in many cases it suffices to define the atom as sp2 or whether that nitrogen ring atom has a planar structure or not (see e.g. [1]). (this seems to be the case in the example, because if not, it would break the aromaticty). Greetings, Egon 0.http://chem-bla-ics.blogspot.nl/2007/11/evidence-of-aromaticity.html 1.http://chem-bla-ics.blogspot.nl/2010/07/new-cdk-default-fingerprinter.html On Fri, Aug 7, 2015 at 11:06 AM, John M <john.wilkinson...@gmail.com> wrote: > Just to put everyone on the same page the issue is when to mark a bond > between two aromatic atoms as aromatic? Consider the following where the > fused bond is always single and not in an aromatic ring (only the outer > ring is 4n+2). > [image: Inline images 1][image: Inline images 2] > [image: Inline images 3] > > Regards, > John W May > john.wilkinson...@gmail.com > > On 7 August 2015 at 10:01, John M <john.wilkinson...@gmail.com> wrote: > >> I'm still unsure on this but I believe it is correct. Consider biphenyl - >> a bond between aromatic atoms is not always aromatic. We could perhaps say >> then a bond between aromatic atoms that is in a ring is aromatic but then >> consider biphenylene? >> >> In your example it is the outer envelope ring that is aromatic and that >> fused bond is always single. Another example is azulene. I'm send also to >> the mailing list as this probably needs some discussion. >> >> John >> >> > > > > ------------------------------------------------------------------------------ > > _______________________________________________ > Cdk-user mailing list > Cdk-user@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/cdk-user > > -- E.L. Willighagen 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 ORCID: 0000-0001-7542-0286 ImpactStory: https://impactstory.org/EgonWillighagen
------------------------------------------------------------------------------
_______________________________________________ Cdk-user mailing list Cdk-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/cdk-user
