[Rdkit-discuss] Axial chirality and Chirality in metallic complexes

[Lionel C. <li...@be...>]

Hi RDKitters,

I have a question not really specific to RDKit but I am sure that some 
of you already faced this problem.

Regarding some specific types of chiralities (not on carbons), I can 
manage chiralities from RS(=O)R' or R[N+]R'R''R''' compounds. But I do 
not know what is the best solution to manage compounds like the following :

1- Compounds with axial chirality e.g. binap compounds

(https://en.wikipedia.org/wiki/Axial_chirality)

2- Metallic complexes, in particular Pt complexes with 6 non equivalents 
bonds e.g. satraplatin

(https://en.wikipedia.org/wiki/Satraplatin)

My question is : what is the best format to store the structure and the 
chiral information of those structures? How do you manage them?

Of course, we should be able to regenerate the right structure from the 
chosen format.

In a step further, the file should be readable for (sub)structure 
searches [no failure when rdkit read the molecule].

At best the chirality should be questionable. Thus how to configure the 
HasSubstructureMatch in order to match the right chirality?

For satraplatin we can use chiral bonds that can be saved in .mol but 
not in smiles. For binap, the classical bold bond is saved as chiral 
bond in .mol and not saved in smiles.

Thanks for all the work done on RDKit,

Cheers,

Lionel