search

GuangyuWangLab2021 / cellDancer

Predict RNA velocity through deep learning
https://guangyuwanglab2021.github.io/cellDancer_website/
BSD 3-Clause "New" or "Revised" License
60 stars 11 forks source link

Save output of adata_to_df_with_embed in environment #16

Closed rrydbirk closed 9 months ago

rrydbirk commented 10 months ago

Thank you for this nice tool!

I don't understand why you force users to save data for pp to CSV. CSV is extremely slow and inefficient. Furthermore, it's not necessary to save the data to disk, we could just keep working with the output in the current environment.

I highly recommend that if save_path is None, the output is stored in a variable, e.g:

pp_dat = cdutil.adata_to_df_with_embed(adata,
                                       us_para=['Mu','Ms'],
                                       cell_type_para='clusters',
                                       embed_para='X_umap',
                                       save_path=None)

will output the Pandas DataFrame to pp_dat.