Treatment of carboxylate groups

Hi,

Starting with a 3D structure containing a carboxylate group, I calculated the dominant tautomer distribution using:

./cxcalc dominanttautomerdistribution -H 7.4 original_ligand.sdf > tautomers.sdf

I then kept only the top tautomer, saving it as top_tautomer.sdf

I then regenerated the 3D conformation with all explicit hydrogens shown using:

./cxcalc conformers -m 1 -s true top_tautomer.sdf > top_tautomer_3D.sdf

I am happy with most of the resulting molecule except the carboxylate group, which has changed from a normal carboxylate group (-COO) to a weird group (-CHOHOH), as seen in top_tautomer_3D.sdf

Is there a way to avoid this treatment of carboxylate groups?

---

I have thought about getting the SMILES string for the resulting compounds and modifying it such that the weird group is turned back into a carboxylate group. From all examples I have so far encountered, this group is always at the very beginning of the SMILES string, as [H]OC([H])(O[H]). I have thought to simply modify this to read: O=C(O).

Do you know if such a group (and thus a carboxylate group) would always come first in a SMILES string when using the molconvert utility?

Thank you for any thoughts on this!

Juliette

Hi,

I have opened the "original_ligand.sdf" file in Marvin GUI and there is no carboxyl group at all in  the molecule. See the attached picture. The   "geminal-diol" group can be recognized among others. My suspect is that your "original_ligand.sdf " is misstyped.   

Jozsi

Hi,

I think that's because the Marvin GUI automatically edits my carboxylate group. When I open original_ligand.sdf in our Viewer (Hermes), which does not add hydrogens, I definitely have a carboxylate group, although the bond orders for both C-O bonds are single - perhaps this is what tells Marvin to make it a gemdiol?

If I were to start with a carboxylic acid group, as seen in new_ligand.sdf, and apply the same steps, could you tell me if I would get a carboxylic acid at the end of the process (dominant tautomer and then single conformer)? I'm sorry I can't check immediately, my Marvin licence has just expired this morning!

Thanks,

Juliette

Hi,

If I were to start with a carboxylic acid group, as seen in new_ligand.sdf, and apply the same steps, could you tell me if I would get a carboxylic acid at the end of the process (dominant tautomer and then single conformer)? I'm sorry I can't check immediately, my Marvin licence has just expired this morning! 

Yes ,you would get carboxyl group.

You can use Marvin, freely, at this site here: http://www.chemaxon.com/marvin/sketch/index.php

I think that your sdf file is corrupted and one of the "C-O" should be double bonded.  

Jozsi

Hi Jozsi,

Thank you for your help. I have a licence again and was able to check that the carboxylate is indeed preserved.

The issue was that I started from a mol2 file of my ligand where the carboxylate C-O bonds were defined as both bieng aromatic, rather than a single and a double. Converting this to sdf file and the information about bond types was lost and thus Marvin understood two single bonds and thus a gem-diol.

Juliette