On Wed, Jun 10, 2009 at 5:27 PM, Igor Filippov<[email protected]> wrote: > >> Igor, as you said, you spend a lot of time thinking about this stuff: >> what would you say the general rule is? >> >> -greg > > Oh, I'm certainly not getting into a flame war about the correct way to > depict stereochemistry!!! :) I've seen people far more knowledgeable > than me having arguments about this until these days at the InChI > steering committee meeting. This is probably an argument that has been > going on for decades with no resolution.
Ah, you're no fun. ;-) > input drawing as closely as possible. So for me the best resolution > would be to allow the programmer (in python or c++) to create the > molecular object the way he sees fit and not to force him to adhere to > an arbitrary limitation. Agreed. And this is how the RDKit handles it: if you use WedgeMolBonds() it does things "my" way (=one wedged bond per chiral center), but you are always free to add additional wedging in your own code. -greg

