Work at SourceForge, help us to make it a better place! We have an immediate need for a Support Technician in our San Francisco or Denver office.


#116 Detect Chiral Centers using 3D coordinates

Sebastian Rohrer

Here is a SD-File of compounds with Chiral Centers that are not detected by

The compounds originate from PubChem, explicit Hs were added and 3D-structures generated with CORINA.

Thanks for having a look at it.



  • Stefan Kuhn
    Stefan Kuhn

    The method only works on molecules containing wedge bonds. It does not take 3d coordinates into account. I put this into the javadoc comment. I make this a bug request, but do not expect it to be done soon (3d handling is weak in cdk overall, so the method would not be of much use _given the current cdk - e. g. smiles generation also works on 2d/wedges as well, to make this 3d more methods would need to be adopted etc.)

  • I could be wrong, but I don't think that coordinates are necessary for detection of these chiral centers. I'll look into this and see if it can be fixed with a minimum of hassle.


  • Stefan Kuhn
    Stefan Kuhn

    Well, depends on what you want to do, I suppose. Detecting where a potential chiral centre is, is probably possible (at least for tetrahedral chiralities) without the coordinates. The current method (which is not well named, as I realize) doesn't just detect this, but it detects fully specified chiral centres, i. e. such, where the positions of the atoms are fully given. This is the case in the example only using 3D coordinates, obviously.