Downloading bioactivities for cell lines about chembl_webresource_client

[spadavec]

The issue here is that the target_chembl_id needs to be changed to cell_chembl_id, and the rest will work out.

Closing.

[spadavec]

It appears I spoke too soon--

I'm using the following code block to (attempt) to download the bioactivities against the cell line CHEMBL3308372:

activities = new_client.activity  

for x in chembl_ids:
    print("Checking cell id {}".format(x))         
    res = activities.filter(cell_chembl_id='{}'.format(x),assay_type='F',standard_type="EC50", standard_value__isnull=False)
    activities_df = pandas.DataFrame.from_dict(list(res))
    print(activities_df)

This returns approximately ~20,0000 molecules, of which all show EC50 values against cell lines that aren't CHEMBL3308372. For example, one of the compounds reported is CHEMBL174698, which shows EC50 against a completely different cell line. Any ideas on how to fix this?

[mnowotka]

Please note, that the activitiy enpoint (https://www.ebi.ac.uk/chembl/api/data/activity.json) doens't have a cell_chembl_id field. This is why this filter gets (silently) ignored so it's equivalent to the following:

res = activities.filter(assay_type='F',standard_type="EC50", standard_value__isnull=False)

[spadavec]

Sorry if I'm not understanding properly, but is it then not possible to download activities against a cell line? If you try to use target_chembl_id and filter for functional assays and a specific standard type, it will return an empty list.

[mnowotka]

Yes, it's not possible. You can filter the assay endpoint using cell_cehmbl_id filter and then retrive all activities with the corresponding assays.