Hi, On Sat, Feb 20, 2010 at 6:57 PM, Geoffrey Hutchison <ge...@geoffhutchison.net> wrote: > > On Feb 20, 2010, at 12:18 PM, Tim Vandermeersch wrote: > >> best option for nitrogen. Detecting chiral (bridge-head) nitrogen >> atoms seems complex... > > Well, the starting place would be to check that the nitrogen was connected to > three ring atoms. If not, it's not chiral. Then I'd check the actual geometry > of the four atoms -- if they're approximately co-planar (e.g., bridge > nitrogen in an aromatic system), it's definitely non-chiral. > > At that point, I think you'd treat it like a CH from a chiral symmetry > perspective.
We can't really use coordinates to determine if the atoms around the nitrogen are planar since we might be dealing with 0D/2D. However, three neighboring ring atoms should be good for now. In the nci.smi file used by smilesmatch test, I removed the chiral nitrogens from the following smiles: C([C@@H](CC(=O)c1ccccc1)[N@@]1[C@@H](C)CCCC1)(=O)c1ccccc1 c...@]1[c@@H](CCC1)[C@@H]1CCCNC1 They have no chiral N and would cause the unit test to fail.There is one smiles left with a chiral N: S(=O)(=O)(O)O.O=C1N2[C@@H]3[C@@]4([C@@H]5[N@@](CC=6[C@@H]([C@@H]3[C@@H](C1)OCC6)C5)CC4)c1c2cc(OC)c(c1)OC This is correctly identified now and can be left as it is. Cheers, Tim > That's my $0.02, > -Geoff > > P.S. For now, why don't you submit the change you suggested. At the worst, it > will mark a few heteroatoms as chiral when they aren't. ------------------------------------------------------------------------------ Download Intel® Parallel Studio Eval Try the new software tools for yourself. Speed compiling, find bugs proactively, and fine-tune applications for parallel performance. See why Intel Parallel Studio got high marks during beta. http://p.sf.net/sfu/intel-sw-dev _______________________________________________ OpenBabel-Devel mailing list OpenBabel-Devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/openbabel-devel
