Hi Robert, I've gotten the figure down to 120.
69 of the remaining problems (as far as I can tell) are due to differences between how the InChI code and Open Babel handle the situation where you have a tetrahedral center where two of the three non-stereo bonds are very close (in angle) to each other. If the angle is within 4 degrees (or so), then the InChI binary will mark it as ambiguous stereochemistry. Open Babel does not do this - it just works out the angle and ploughs ahead. I don't regard this as a bug in Open Babel. An example of this is 12040015. I guess the source of the problem is with the depiction or structure generation - the two bonds should not have been placed so close together (GIGO). Note that this is reasonable behaviour by the InChI code, but it would have been useful if there were an option to turn this off. I still have some work to do on the bonds marked "4", but fixing this will have little effect on your test set (as you pointed out). - Noel On 23 August 2011 11:29, Noel O'Boyle <baoille...@gmail.com> wrote: > I note that Chris has fixed some problems with metals, and we're down > to 146 failures. Of these 57 have a particular substructure that > appears to confuse OB. > > - Noel > ------------------------------------------------------------------------------ EMC VNX: the world's simplest storage, starting under $10K The only unified storage solution that offers unified management Up to 160% more powerful than alternatives and 25% more efficient. Guaranteed. http://p.sf.net/sfu/emc-vnx-dev2dev _______________________________________________ OpenBabel-discuss mailing list OpenBabel-discuss@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/openbabel-discuss
