Prefer wedging ring bonds around atropisomers #7373
Conversation
|
@tadhurst-cdd your feedback on this would be very welcome |
wedging bonds in 5 rings make later attempts at improving kekulization very difficult update expected results
There was a problem hiding this comment.
Look good. I see that you added a preference for larger rings over smaller ones too. That may cause a problem if one end of the atrop bond is in a macrocycle, which are now (or soon to be) allowed. If that occurs, the wedge bond in the macrocycle could generate a second atropisomer bond inside the macrocycle.
Thanks for that feedback @tadhurst-cdd. My intent there was to try and avoid putting the wedged bond into 5 rings - this is connected to trying to avoid making wedged bonds double when kekulizing, which is often impossible with 5 rings. I can redo that logic to just explicitlly avoid 5 rings and thus not cause the problem with macrocycles. |
|
looks good |
The code change is relatively simple - switching the preference from wedging non-ring bonds to wedging ring bonds.
There's also a change to prefer wedging bonds in larger non-macrocyclic rings. This is relevant in a coming PR where the kekulization code is modified to try and avoid making wedged bonds double bonds.
The rest of this is updating expected results.
Fixes #7371