Script to add ref_mito and ref_fasta_index to scpca-meta.json

[allyhawkins]

Closes #173

Before reprocessing any projects, we need to make sure the scpca-meta.json files from previous versions of scpca-nf match the scpca-meta.json file needed for the latest version (v0.5.2). The difference between these two versions is the addition of ref_mito and ref_fasta_index to the json files.

This script runs through all of the libraries in the existing metadata file and does the following:

• Grabs the corresponding scpca-meta.json file, using the technology for that run ID to define the path to where the scpca-meta.json file should exist. If the file does not exist (as in that sample hasn't been processed), then that sample is skipped.
• If ref_mito and/or ref_fasta_index do not exist in the metadata file, those are added to the dictionary holding the json contents.
• The updated json is uploaded to the same location on S3, overwriting the original file.

I used the arguments found in update_scpca_checkpoints.py for inspiration. Because of that, I have an argument for the checkpoints_prefix, which is where the files live that should be modified. I have been testing using scpca/processed/checkpoints, but when we run this for real, we will want to use scpca_prod/checkpoints.

[allyhawkins]

Thanks for the comments @sjspielman. I added some information about usage at the top of the script and a readme to explain how we use/used both of these scripts.

[allyhawkins]

Looking to next steps, let me know when you've run this through (if not already?) and I can get started on re-processing CITEseq projects.

This script has now been run so all scpca-meta.json files are up to date and you should be good to go to re-process CITE-seq projects.