On 22/03/2013 01:37, Craig James wrote: > Am I confused, or is this just wrong? These two SMILES should be the > identical molecule: > > c1ncccc1 > c1[n]cccc1 > > The first uses valence rules to deduce that there is no H on the > nitrogen. The second specifically says the same thing. Both should > thus be the identical molecule. However... > > $ echo "[n]1ccccc1" | obabel -i smi -o can > c1ccc[n]c1 > > $ echo "n1ccccc1" | obabel -i smi -o can > c1cccnc1 > > If you print out the SDF, you can see the problem: the first form of the > SMILES with the [n] is marked as a radical: > > $ echo "[n]1ccccc1" | obabel -i smi -o sdf > ... > M RAD 1 1 2 > M END > $$$$ > > Am I missing something, or should this be filed as a bug?
I think this is a bug and it is caused by the call to OBAtomTyper::CorrectAromaticNitrogens() in line 259 of typer.cpp. This ancient function is labelled //! \deprecated Currently unused for anything significant. but it erroneously changes the ImplicitValence of the [n] from 2 to 3. Removing the call fixes the bug. I will commit this change and hope it doesn't mess up anything else. Maybe a proposed simplification of the valence system http://forums.openbabel.org/ImplicitH-failures-td2990439.html should be looked at again. Chris ------------------------------------------------------------------------------ Everyone hates slow websites. So do we. Make your web apps faster with AppDynamics Download AppDynamics Lite for free today: http://p.sf.net/sfu/appdyn_d2d_mar _______________________________________________ OpenBabel-Devel mailing list OpenBabel-Devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/openbabel-devel
