chiral information on the phosphorus atom

Hi, We found we can not input two structures with different chiral information on the phosphorus atom into the database of jchem base(5.3.7). The error showed that these two structures were duplicate

Could you help to how to avoid this error?

I have checked and the chirality information is different indeed.

volfi wrote:

I have checked and the chirality information is different indeed.

Yes, These two structures can be marked with different chirality information in Marvin, but these two structures can not be input into the same structure database with jchem base. The database is just set with duplicate filtering on, if the different chirality information on carbon atom, the database can allow the structure to be input, but forbid for phosphorus atom

We want to know how to fix it or avoid it?

If you need anything else, please let us know

 Hi,

 

The stereo information for atoms over 4 valence electrons is not considered during searching. This was introduced for allowing substructure matches on pentavalent phosphorus, that have 5 ligands by tetrahedral query. Because of having 5 ligands on the target side no tetrahedral stereo information can be interpreted.

We will consider changing this behaviour to check the number of ligands instead of the binding electrons. In this case however a separate condition will be needed for Cis/trans stereo checking.

We will notify you when this is implemented.

Bye,

Robert

Thank you for your reply.

We wonder if you had made a different setting about stereochemistry on "P" atom. By using the old jchem of (version 5.1.4), we can input the above structure of "R" and "S" into our database at the same time. but we can not do that by using the new jchem(5.3.7) with the same setting of duplicate filtering. Now we had many pairs structures need to be input into database. Could you give us a suggestion to distinguish between the above two class structures NOW?

Thank in abvance

Hi,

We will change the behaviour in case of such multivalent atoms. The criterion for stereo checking will be based on the number of bonds. Structures with the normal amount of bonds (e.g. your structures) will be checked for stereo equivalence, above this stereo checking will be skipped. This will be released probably in version 5.4.1.

As a workaround currently you can use data sgroups to distinguish between the two stereo configurations. See http://www.chemaxon.com/jchem/doc/user/query_features.html#attached_data_duplicate

Bye,

Robert

Hi,

The previously written change has been released with version 5.4.1.

Regards,

Robert