molconvert conformer generation

User ac20ca0f01

09-08-2011 13:56:57

Hello,


 


I am using molconvert version 5.4.1.0 to generate 3D coordinated from isomeric SMILES. In the structure there are 2 chiral atoms and 3 sp3 Carbon atoms being part of a 6-membered ring. Therefore, this 6-membered ring can have different conformations. I have used the option -3:{fast} and -3:{fine} and both generated the same conformer. I have also used CORINA to generate the lowest energy connformer. It resulted in another structure which is supported by 1H-NMR coupling constants. I have attached the SMILES files and the generated MOL files.


Could you please give me a reference on the method used for the 3D structure generation? I would like to know  why the two options in molconvert resulted in the same conformer which is not the lowest energy conformer for the structure.


I would highly appreciate any help for the generation of the correct conformer using molconvert. 


 


kind regards,


Villy

ChemAxon 8b644e6bf4

11-08-2011 12:01:40

Dear Villy,


 


A brief description of the used method available at http://www.chemaxon.com/library/advanced-automatic-generation-of-3d-molecular-structures/ (see downloadable pdf file: http://www.chemaxon.com/conf/Advanced_automatic_generation_of_3D_molecular_structures.pdf ).


Generaly, fine build ( molconvert -3:S{fine} ) should be used instead of the "fast" ( molconvert -3:S{fast} ). As far as i know currently the fine build is the default selection ( molconvert -h thus probably wrong, it needs checkup ). For more option see help available by "molconvert -H3D"


To find low(est) energy conformer generating multiple conformers, using strict optimization limit and optionally using "hyperfine" filtering is advised.


The conformational analysis of the attached smiles structure seems to be incomplete: our code places the OH group only in axial position when generating up to 200 conformers; equatorial placement is missing. This needs further checkup from us.


 


Regards,


Gabor