conda install python==3.10
pip install cellxgene_census
pip install joblib

• output: tissue_list.tsv
• output: 00data/<tissue>.obs_id.tsv
• output: 00data/<tissue>.var.tsv

This script lists the target files to be downloaded in the above files.

• output: 01data/<tissue>-<start>-<end>.jbl

This script performs the download. The download is divided into N samples and saved. This makes it easier to perform parallel processing and allows you to resume from where you left off if the download is interrupted.

• output: 02data/<tissue>-<start>-<end>.h5ac

All downloaded files are converted to standard scanpy files. This is not done simultaneously with the download to prevent conversion errors from stopping the download.