On 06/04/16 21:58, Geoffrey Hutchison wrote: >> A bit off topic but it would be great if a new kekulizer had an option to >> preserve existing hydrogens. This is the only blocker to applying the >> respective valence models on reading for smiles and molfile. It also of >> course makes the problem a lot easier. > > That’s a given, actually. I’ve sat down and thought out the aromaticity > detection and Kekule code and how to explain it to CS undergrads. While I’m > sure there will be some initial bugs, I think the results will be strong. > > The base concept will be to keep atoms as-is when specified and apply bit > masks as constraints. That is, there will be a pass to remove atoms/nodes and > edges from the graph for Kekule assignment (e.g., aromatic S in thiophene or > -NH in pyrrole). > > I also have the beginnings of an aromaticity detection code that does not use > SSSR or LSSR assignments - simply doing a BFS path to count electrons, > inspired by recent discussion on OpenSMILES. > > The students also seem eager to attempt a more efficient, likely OpenMP > enabled ConnectTheDots implementation. > > Again, I’m looking for a good test set. I have 200 molecules on which OB fails to find the double bonds (at least when we tested 2 years ago: tested by looking at them one by on Avogadro). The files are derived from Gaussian which in addition to the atoms gives the total charge and multiplicity. It would be great if some/most of these can be fixed!
http://folding.bmc.uu.se/OB-BondOrderProblems.zip David. > > -Geoff > ------------------------------------------------------------------------------ > _______________________________________________ > OpenBabel-Devel mailing list > OpenBabel-Devel@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/openbabel-devel > -- David van der Spoel, Ph.D., Professor of Biology Dept. of Cell & Molec. Biol., Uppsala University. Box 596, 75124 Uppsala, Sweden. Phone: +46184714205. sp...@xray.bmc.uu.se http://folding.bmc.uu.se ------------------------------------------------------------------------------ _______________________________________________ OpenBabel-Devel mailing list OpenBabel-Devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/openbabel-devel
