3D generation and stereochemistry

Hello,





I would like to know how stereochemistry is handled in case of using the clean method. In case of specified enantiomers, is the stereochemistry kept in the generated conformers ?


If stereochemistry is not specified, is a racemic mix generated ?


In that case how is handled the case there a a high number of chiral centers ?


In my case, if there is less than 5 centers, i would need 16 conformers each of different stereochemistry. Is it possible to select the correct conformers ? ( I guess I can detect the stereochemistry in all generated conformers and choosing the ones having the lowest energy, but maybe there is an easiest way ?)





Thanks, and have a nice week end !





Vincent

Hi Vincent,





> In case of specified enantiomers, is the stereochemistry kept in the generated conformers ?





Yes, Clean3D tries keeping the specified stereochemistry. If this fails, then if {nofaulty) has been specified then it gives back the result, otherwise it gives an error report.





> If stereochemistry is not specified, is a racemic mix generated ?





Yes, but regular Clean3D gives back only one structure.





> In that case how is handled the case there a a high number of chiral centers ?





The same way as conformers, up to the limit the lowest energy structures are kept during the process.





> In my case, if there is less than 5 centers, i would need 16 conformers each of different stereochemistry. Is it possible to select the correct conformers ?





For an automatic process, based simply on Clean3D, you can convert the reulting conformers back to smiles strings, and, since the conformers are sorted by energy do not process a conformer if its smiles string has appeared already. This might not always ensure that you get each stereoisomer but could be faster than cleaning the stereoisomers separately.





Best wishes,





Ödön

Hi,





I would have two more questionsabout this topic :





- Can relative and absolute stereochemistry be specified in 2d molfiles ? How does the clean 3D handle these structures ?





- How does clean 3d handles chiral centers in cycle, or in cycle junctions ?





greetings,





Vincent

Hi Vincent,





First of all, we moved the topic under category "Structure editing, viewing and file formats", as it better fits there.

vincentberenz wrote:

- Can relative and absolute stereochemistry be specified in 2d molfiles ?

Yes, you can use the chiral flag or the enhanced stereo representation.


Please visit the following links for explanation:





http://www.jchem.com/doc/user/Query.html#chirality


http://www.chemaxon.com/marvin/doc/user/stereo-doc.html





The use of enhanced stereo representation is straightforward, but the use of the chiral flag is a bit complicated:





The first attached image shows how you can set and unset the chiral flag, and the second shows where to set the chiral flag display option. In this display mode, (R) and (S) labels are displayed when the chiral flag is set.


In the future we plan to show the label "Abs" next to the molecule when the chiral flag is set.





Odon will address your other questions.





All the best,


Szabolcs

vincentberenz wrote:

Hi, I would have two more questionsabout this topic : - Can relative and absolute stereochemistry be specified in 2d molfiles ? How does the clean 3D handle these structures ? - How does clean 3d handles chiral centers in cycle, or in cycle junctions ? greetings, Vincent

Clean3D handles the chiral centers in cycles normally, if in certain cases it is impossible to generate a structure for the chrality/parity requirements then by default (fine mode) it leaves the original 2D structure in the GUI, prints out "Clean failed" and the smiles string for the molecule. In fast mode it might change the problematic parity and gives a 3D structure. The batch mode (molconvert) it has the[acceptfault] flag to accept structures with incorrect parity.





Best wishes,





Ödön

vincentberenz wrote:

Hi, I would have two more questionsabout this topic : - Can relative and absolute stereochemistry be specified in 2d molfiles ? How does the clean 3D handle these structures ? - How does clean 3d handles chiral centers in cycle, or in cycle junctions ? greetings, Vincent

I've forgot to address the absolute/relative stereo issue. Clean/3D treats the stereo information as absolute stereo. Since changing to the mirror image and changing each atom's parity is easy, we will very likely solve the problem with a menu option for "Flip/Mirror" transformation. On the other hand the mirror transformation is just changing the sign for one (X, Y or Z) but the same coordinate of each atom.





Best wishes,





Ödön