"roundtrip" retrieval not working

[shortydutchie]

Hi

I tried a 'roundtrip' retrieval:

• US20210277002 gives me a list of smiles.

• Selecting one of these smiles as query, eg. "CC1COCCN1C1=CC(C2=C(S(=O)(=O)NC3CC3)C=CC=C2)=C2C=NN(C3=NNC=C3)C2=N1" from page 114 does not seem to give me back the patent.

My script works fine with the example query_smiles "CC1(C)C2(C)CC2CC1CC1C(CO)(C)C1" from the examples directory.

I further note that all chirality has been stripped, and that smiles need to be specifically kekulised, as per following, is that correct? It's worth stating that explicitly somewhere, think it should be as follows?

    molecule = Chem.MolFromSmiles(query_smiles)
    Chem.RemoveStereochemistry(molecule)           #NB chirality not present in db
    Chem.RemoveHs(molecule)                       
    Chem.Kekulize(molecule)
    query_smiles = Chem.MolToSmiles(molecule,kekuleSmiles=True)

Thank you!

[lucas-morin]

Hi @shortydutchie,

Sorry for the late reply, and thanks for pointing out this problem.

It should now be fixed. You can try the latest version of the notebook molecule_query.ipynb, with your query_smiles:
"CC1COCCN1C1=CC(C2=C(S(=O)(=O)NC3CC3)C=CC=C2)=C2C=NN(C3=NNC=C3)C2=N1".

For information, the issue was that when using grep to search for the string "CC1COCCN1c1cc(-c2ccccc2S(=O)(=O)NC2CC2)c2cnn(-c3cc[nH]n3)c2n1", I forgot to escape the characters [ and ].

The query smiles should not be kekulized, and the stereo-chemistry must be removed.
As you suggested, I added this in the notebook.

Best,
Lucas

[shortydutchie]

Hi Lucas, Many thanks for this, it is much appreciated. I'll have a play hopefully soon. Best, W

…

On Tue, 26 Nov 2024, 15:47 Lucas Morin, ***@***.***> wrote: Hi @shortydutchie <https://github.com/shortydutchie>, Sorry for the late reply, and thanks for pointing out this problem. It should now be fixed. You can try the latest version of the notebook molecule_query.ipynb <https://github.com/DS4SD/PatCID/blob/main/examples/molecule_query.ipynb>, with your query_smiles: "CC1COCCN1C1=CC(C2=C(S(=O)(=O)NC3CC3)C=CC=C2)=C2C=NN(C3=NNC=C3)C2=N1". For information, the issue was that when using grep to search for the string "CC1COCCN1c1cc(-c2ccccc2S(=O)(=O)NC2CC2)c2cnn(-c3cc[nH]n3)c2n1", I forgot to escape the characters [ and ]. The query smiles should not be kekulized, and the stereo-chemistry must be removed. As you suggested, I added this in the notebook. Best, Lucas — Reply to this email directly, view it on GitHub <#2 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AO3HKETZUUDUCPXAHKQLNQT2CSJ2ZAVCNFSM6AAAAABOSP5XQGVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDKMBRGIYTINZSG4> . You are receiving this because you were mentioned.Message ID: ***@***.***>