inferring radicals from aromatic smiles

Greetings, I'm having problem getting JChem (3.1.12) to parse the following smiles properly:


Oc1c(-c2ccc(Cl)cc2)[n](=O)c3ccccc3[n]1=O


In particular, it's not clear to me why there are radicals attached to the two N atoms because the Kekule form


OC1=C(C2=CC=C(Cl)C=C2)N(=O)=C3C=CC=CC3=N1=O


seems to parse properly. Interestingly, PubChem's standardizer (http://pubchem.ncbi.nlm.nih.gov/standardize/standardize.cgi) can handle the former but chokes on the latter.





Thanks for any pointers,


Trung

Hi,





I have just checked the latest Marvin 5.0.6 version and the bug is there.


So I have checked the new Marvin 5.1 version which comes out soon and the bug has been fixed in it already. No radicals in any case.





Anyway thank you for the report.





Andras