question about Chapter16.4 Pseudobulk making code

[dfc-1931475]

*Question
Could it be better to exchange the executing order between the following 2?

1. df_donor = df_donor.groupby(donor_key).agg(agg_dict)
2. df_donor[donor_key] = donor

These 2 commands are in "aggregate_and_filter" function**(Chapter16.4)** to make the pseudobulk data for certain cell_indentity per sample

I found if the obs_to_keep parameter was set to None. The program will crashed because there isn't a column named donor_key available in the df_donor, so that running df_donor = df_donor.groupby(donor_key).agg(agg_dict) will arise an error.

Since the donor_key will always be the same in each round of the df_donor, why not to run df_donor[donor_key] = donor first? even the obs_to_keep is None, the function will still be smooth.

[Zethson]

Thank you! We'll keep it in mind for our next big iterations in 2025. Please keep the feedback coming.