Add final rule for multiqc

Add final rule for multiqc

################################################################################
### Multiqc
################################################################################

rule multiqc:
    input:
        fastqc_1 = expand(os.path.join(config["output_dir"], "{sample}", "fastqc_1"), sample=get_samples()),
        fastqc_2 = expand(os.path.join(config["output_dir"], "{sample}", "fastqc_2"), sample=get_samples()),
        STAR = expand(STAR)
    output:
        multiqc_dir = directory(os.path.join(config["output_dir"], "summary", "qc"))
    log:
        os.path.join(config["local_log"], "multiqc_fastqc.log")
    singularity:
        "docker://zavolab/multiqc:1.8"
    shell:
        "(multiqc --outdir {output.multiqc_dir} results logs) &> {log}"

assigned to @gypas

Is multiqc an alternative to fastp? Probably not because fastp also subsumes cutadapt.

No multiqc collects logs files from other tools and generates general reports at the end (e.g. parses the logs from STAR, fastqc and creates an html file). fastp generates quality statistics plots (similar to fastqc.

OK, so as I just asked Mihaela - I will work on integrating results from distinct steps into MultiQC report at the end of the workflow.

assigned to @bakma and unassigned @gypas

assigned to @gypas and unassigned @bakma

assigned to @bakma and unassigned @gypas

added Doing label

• Prepare dummy 1-rule MultiQC workflow
• Branch from master and add dummy MultiQC rule to the workflow
• Test execution: local/slurm with singularity
• Construct the initial version of the report: adjust the YAML config

added 1 deleted label

mentioned in issue #6 (closed)

mentioned in issue #51 (closed)

changed milestone to %v0.1.0 release

added To Do label and removed Doing label

added Doing label and removed To Do label

I have branched the master and added two rules at the end of the workflow:

1. gathers logs, re-formats (if necessary) and creates config for MultiQC
2. MultiQC execution (runs in the author's docker container)

Dummy runs finish successfully.

It seems that this aggregator is highly customizable:
https://multiqc.info/docs/#configuring-multiqc

I believe that if we ensure a proper output/log directory structure and prepare a proper YAML config for the tool we can easily have very appealing reports. I will have to really dive into how to create this configuration file for our pipeline now.

@bakma Does this mean that we can add reports for custom tools and scripts? (e.g. output of https://git.scicore.unibas.ch/zavolan_group/pipelines/rnaseqpipeline/blob/master/scripts/perform_PCA.py)

Yes, but it is not advised and significantly limited:
https://multiqc.info/docs/#custom-content

Well, guess we should write modules for our own tools then. Well, first we should package our tools properly, in their own repos :)