Hi Lauren,
SMARTS doesn't have a direct way of saying an atom is non-racemic, but you
can express that idea using recursive SMARTS. For example,
In [46]: racemic =
Chem.MolFromSmiles('c12ccccc1cncc2NC(=O)C(CCO2)c1cc(Cl)ccc12')
In [47]: chiral1 = Chem.MolFromSmiles('c12ccccc1cncc2NC(=O)[C@H
](CCO2)c1cc(Cl)ccc12')
In [48]: chiral2 = Chem.MolFromSmiles('c12ccccc1cncc2NC(=O)[C@
@H](CCO2)c1cc(Cl)ccc12')
In [49]: [m.HasSubstructMatch(Chem.MolFromSmarts('c12ccccc1cncc2NC(=O)
[CH;!$([@])](CC)c1cc(Cl)ccc1'), useChirality=True) for m in [racemic,
chiral1, chiral2]]
Out[49]: [True, False, False]
Where the highlighted atom [CH;!$([@])] means "a carbon with a hydrogen AND
not a chiral atom".
Hope this helps,
Ivan
On Wed, Mar 17, 2021 at 6:18 AM Lauren Reid <lauren.r...@medchemica.com>
wrote:
> Hi,
>
> I would like to perform a substructure search in which a racemic chiral
> SMARTS finds only racemic compounds and not those that have specified
> stereochemistry, e.g these compounds from the COVID moonshot project:
>
> Does anyone know if there’s a way to specify this distinction in an rdkit
> substructure search?
>
> Thanks,
>
> Lauren
>
> Dr Lauren Reid
> Computational Chemist / Developer
> lauren.r...@medchemica.com
> www.medchemica.com
>
> Medchemica Ltd is a company registered in England and Wales with company
> number 8162245.
>
> _______________________________________________
> Rdkit-discuss mailing list
> Rdkit-discuss@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/rdkit-discuss
>
_______________________________________________
Rdkit-discuss mailing list
Rdkit-discuss@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/rdkit-discuss
