I've encountered a strange problem when doing constrained embedding that seems to be related to chirality. The example is here: https://gist.github.com/tdudgeon/c4604f3ee9124eeec60668b5eefe465e
The molecule being embedded has a single chiral carbon, and while that atom is tethered, only one of it's attached atoms is also tethered, so if I understand correctly, there should be no reason why the chiral atom cannot be embedded correctly. But when I use the enforceChirality=True option the embedding usually fails, and takes several seconds to do so. Very occasionally it succeeds and the chirality is preserved as expected. If however I use the enforceChirality=False option the embedding succeeds and is very quick. In all cases I've run so far the stereochemistry is preserved. Can someone explain this behaviour? Tim
_______________________________________________ Rdkit-discuss mailing list Rdkit-discuss@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/rdkit-discuss
