0. Edit config.yaml, and set scratch_dir to a scratch directory where you have write access.

1. First, install and activate the conda env. Note that this pipeline uses R within conda, rather than from CEDAR, to better facilitate R-python interop via reticulate.

conda env create -f environment.yaml
conda activate ngs5388_snakemake

2. Within the conda R, manually install DropletQC:

devtools::install_github("powellgenomicslab/DropletQC")

3. Run snakemake.

4. After successfully completing the pipeline, proceed onto the folder 003_initial_plotting_dsdb_upload. (Note the conda environment is no longer needed for this -- the standard CEDAR R is used for the downstream plotting and analyses after the initial Snakemake pipeline).