Canonicalisation algorithm changes and the JChem Cartridge

Hi,

I have just got the new JChem Java API ( V 5.2.3_1) as there is a couple of fixes in there, previously I was running V 5.1.3_2.  I notice that the Canonicalisation algorithm has changed eg:

input SMILES: CN1CC[C@]23[C@H]4Oc5c3c(C[C@@H]1[C@@H]2C=C[C@@H]4O)ccc5O

Canonicalised with V 5.2.3_1: CN1CC[C@]23[C@H]4Oc5c(O)ccc(C[C@@H]1[C@@H]2C=C[C@@H]4O)c35                 

Canonicalised with V 5.1.3_2: CN1CC[C@@]23[C@H]4Oc5c2c(C[C@@H]1[C@@H]3C=C[C@@H]4O)ccc5O

 

Obviously all the canonical forms are equivalent but my question is do we need to upgrade the SMILES stored in our database  running the JChem Cartidge ?  As we use the Java API to create SQL queries is it possible that any of the Chemaxon functions (eg: jc_compare, jc_equals, jc_compare, jc_tanimoto etc) will not work in exactly the same way beacause we will be including SMILES strings in a different canonical form (generated by the new API) ?

 

Hope this makes sense.

 

Thanks in advance.

Hi Derek,

This change of smiles output does not effect the cartridge operators and functions.

(At each release we carefully check what is changing, and we force regeneration if the changes have any effect.)

 

However, if you run direct string comparisons of the generated canonical smiles strings,
then yes, those have to be regenerated.

 

Best regards,

Szabolcs

ok thats great, just thought I'd check.

 

All the best

 

Derek