scATAC Pseudobulk Pipeline

About

This Snakemake workflow is designed to split one or more scATAC-seq BAM files into pseudobulk replicates, each containing n cells.

Pipeline Steps

1. For each BAM file, count the number of unique occurences of each cell barcode
2. Assign cell barcodes a label corresponding to a pseudobulk replicate to create
3. Split the BAM file into pseudobulk BAM files using Sinto to separate the cell barcodes
4. Generate indexes for the pseudobulk BAM files
5. Generate bigWigs for the pseudobulks
6. Call peaks for the pseudobulks
7. Create a metadata file summarising the number of pseudobulks created from each input BAM file

Running the Pipeline

1. Set up the pseudobulk conda enviroment using workflow/envs/pseudobulk_env.yaml

   conda env create --name pseudobulk --file=pseudobulk_env.yml

2. Edit config/config.yaml to set the data and parameters

3. Then the pipeline can either be run manually via command line or scheduled to run on a cluster via Slurm.

Command Line

1. Open a terminal and navigate to the folder containing the workflow, e.g.

   cd user/path_to_projects/scATAC_Pseudobulk_Pipeline

2. Activate conda environment with dependencies

3. Run the workflow via the command below, setting n as the number of processors to use

snakemake --cores n

Running on a Cluster via Slurm

The pipeline can be scheduled to run on a cluster using the file submit.sh