BamQC-pipeline 🌻

Snakemake workflow for automated BAM alignment statistics intended to run on the BIH cluster using SLURM scheduler for parallelization. This pipeline will be appended by more utilities soon and for now employs Alfred, an efficient and versatile command-line application that computes multi-sample quality control metrics in a read-group aware manner.

1. 📕 Installation

Clone the repository into your personal work (e.g. /fast/users/$USER/work/)

git clone https://github.com/pweidner/BamQC-pipeline.git

The pipeline currently looks like this:

BamQC-pipeline/            # Run the pipeline from here
│
├── config/
│   └── config.yaml        # Config parameters for user input
└── workflow/
    ├── envs/
    │   └── qc_env.yaml    # Conda environment housing all tools (automated installation upon first execution)
    ├── errors/            # Error files by rules (generated upon first execution)
    ├── logs/              # Log files by rules (generated upon first execution)
    ├── profiles/
    │   └── config.yaml    # BIH specific SLURM config
    ├── Snakefile
    └── utils.py    
    

To run the pipeline, you need to set up a conda environment to launch snakemake from

mamaba create -n snakemake snakemake=7.32.0

2. 🛑 Run the pipeline

To launch the pipeline, simply cd into the folder containing the pipeline and activate the snakemake env:

cd /path/to/BamQC-pipeline
conda activate snakemake

Once inside the repo, you can launch any number of runs from here on I/O folders (e.g. from within a dedicated screen session):

snakemake --config ref=/path/to/hg38 data_location=/path/to/bamdir output_location=/path/to/outputdir --profile workflow/profiles

When first executed, the pipeline automatically installs all tools from a conda qc_env.yaml, which can be reused for any other runs.

ref -> Currently specifies the path to my personal reference files I use for alignments, should be updated to group references soon (without the .fa ending).

Find a summary of the current output generated here.