basic aromaticity method and chemaxon reactor

Hello,

We found that basic aromatized SMILES strings could trigger reactions which
they should not trigger.

Here is an example,

SMILES for pyridin-2(1H)-one:
O=C1NC=CC=C1 - generated by Molecule.aromatize(MoleculeGraph.AROM_BASIC);
O=c1cccc[nH]1 - generated by Molecule.aromatize();

One representation "O=C1NC=CC=C1" doesn't have an aromatic ring while the
other "O=c1cccc[nH]1" has one aromatic ring.

we have a reaction (bt0005_3939.mrv) requires substrates having an aromatic
ring to be triggered. We did a simple test by running,

react -v -r bt0005_3939.mrv "O=C1NC=CC=C1"

Surprisingly, the reaction was triggered and gave a product
"OC1=C(O)C=CC(=O)N1". It looked like reactor automatically transferred the
substrate based on general aromaticity method and triggered the
reaction.

We checked chemaxon reactor documents and didn't find any parameters
related to the aromaticity settings.

Do you have any suggestions regarding to this problem? Thanks!

btw, our software version is 5.3.4.

- Jeff

Hi,

There is a MolSearch object inside the Reactor engine which matches the input reactants with the reactants in the reaction scheme. This MolSearch object performs general aromatization during the search, and currently there is no way to change this aromatization method. So for now reactor cannot handle these kind of aromatic molecules as non-aromatic.  We will consider implementing a reactor option to switch off the automatic aromaticity handling.

Zsolt

Zsolt wrote:

Hi, There is a MolSearch object inside the Reactor engine which matches the input reactants with the reactants in the reaction scheme. This MolSearch object performs general aromatization during the search, and currently there is no way to change this aromatization method. So for now reactor cannot handle these kind of aromatic molecules as non-aromatic.  We will consider implementing a reactor option to switch off the automatic aromaticity handling. Zsolt

Zsolt,

Thank you for letting us know this. Using basic aromaticity method in reactor is very important to us (and possibly other users). Please let us know when you decide to implement this feature and whtat version will it be in.

Jeff

Jeff,

This option will be available only in API; we will add a "setAutoAromatize(boolean)" method to Reactor class. If the auto aromatization is switched off, then users should take care of the aromatization of molecules. It will be released in JChem 5.4.n (where 0 <= n <= 4).

Zsolt

Zsolt wrote:

Jeff, This option will be available only in API; we will add a "setAutoAromatize(boolean)" method to Reactor class. If the auto aromatization is switched off, then users should take care of the aromatization of molecules. It will be released in JChem 5.4.n (where 0 <= n <= 4). Zsolt

Thank you for letting us know Zsolt!

- Jeff

I'm taking over for Jeff.  I see that JChem 5.4.0 and 5.4.1 have been released but I didn't see any mention of adding setAutoAromatize(boolean) to Reactor.  Could you give us an updated timeline on this bugfix?

 

 - Nick

Hi Nick,

It will be implemented in JChem 5.4.3 or 5.4.4.

Zsolt

Thanks!

I see that JChem 5.5 has recently been released.  I didn't notice anything in the changelogs related to this bug.  Has it been fixed in 5.5?  If it's not fixed, do you have an updated timeline of when it may be fixed?

Sorry, it was not implemented in 5.5, it will be available in 5.7.

Zsolt

Hi, I'm taking over for Nick.  Is this issue still planned to be fixed in version 5.7?  Do you have a tentative idea of when 5.7 will be released?

Shawn

Hi,

Unfortunately we switched to more frequent releases, so this feature will not be included in JChem 5.7.

5.6 was released this week, 5.7 will be released in few weeks. API support to switch off auto aromatization can be expected in few months (not sure in which release).

Zsolt

Just checking in again to see if you now have a better idea of when/which version this will be addressed.

-Shawn

Shawn has left the project, so please let me know when to expect basic aromatization in Reactor.

 

- Lynda Ellis

We will let you know soon.

Zsolt

Any new information yet on timeline for this change? - Lynda Ellis

Please let us know when to expect this change.  

 

Thank you, Lynda Ellis

The honor of a response would be appreciated.

 

- Lynda Ellis

Hi Lynda,

Unfortunately, meanwhile a lot of other features we have to implement got higher priority, so this feature will not be implemented in the near future. Sorry for the late answer.

Zsolt