Bunch of modification from Jacques

[Juke34]

Hi guys,

As discussed here a first bunch of modifications. It should simplify main workflow, improve usability and reproducibility.
With the original version I was not able to test the pipeline, you need extra prior step to prepare the genome.dict and it is not clear how to get the vep db.

What might be improve:

• allow URL for params.ref in order to retrieve automatically the ref in fasta.gz (at least for test. I think ensembl might provide such link)
• deal with fasta, fasta.gz
• Add profiles for ressource local/hpc to centralise computing ressources
• Add profiles for ressource docker/singu/conda
• Add a test profile
• Define software within the pipe for reproducibility
• Make a help in main.nf
• Make a single MultiQC adding extra steps (e.g. fastp, bwamem...)
• Add a CI test (would be possible if we have a minimal dataset. I'm thinking we work only on a part of one of the chromosome listed In the bed...)
• automate the creation of genome.dict
• Add parameter section in the README
• Improve README to explain (or give example) where to retrieve genome.fa (ref), genome.dict (dict) and vep db ().
• make .dict automatically
• make bwa index automatically
• be able to run vep in offline and online mode
• Add authors in the manifest (nexflow.config)
• When test is working - make a release

I guess the full nf-core version would be a version2 of this pipeline ( for a second paper ;) ).

Cheers

[Juke34]

Oh, I just realise that this PR cover a bunch of change addressed by the PR #6 (at least the split into modules)
I will wait we clarify what are your plans before going further @Char-Al
I do not want to disturb, just to help ^^

cheers

[Juke34]

I finally polished some stuff to be able to test it. This is a working version.

[SimCab-CHU]

Super!
Merci Jacques. Je regarde ça attentivement et je reviens vers toi.

Aplus

Simon