Report and Quality Control

MultiQC

Many tools report their output statistics via MultiQC. We store the resulting html file in qc/multiqc.html for each run.

See the Rule Call Graph for an overview of the data flow into MultiQC, and see the multiqc rule for full details.

Also, there might be cases where you only want the quality control without the actual variant calling, for example, to get an overview of the quality of some sequencing run. In that case, you can also just run all the pipeline steps needed to obtain the MultiQC report by calling

snakemake [other options] --until multiqc

or

snakemake [other options] multiqc

In this case, those two are equivalent, see the Snakemake CLI for details.

Important remark: Note however that by default, SnpEff is an input to MultiQC, which itself depends on the variant calls, meaning that by default, the variant calling is still being run. So, if SnpEff is not needed in your MultiQC report, make sure to deactivate it in your config.yaml first!

Snakemake Report

We also offer to automatically generate a Snakemake report for a run of the pipeline

snakemake --report my-analysis/report.html

This needs to be amended by the --directory option as needed.

For this to work, the Python packages networkx and pygraphviz must be installed:

sudo apt-get install python3-dev graphviz libgraphviz-dev pkg-config python-pip
sudo pip install networkx pygraphviz