[MODULE] velocyto.py module for nf-core/scrnaseq

Velocyto is a library for the analysis of RNA velocity.

I want to implement a module to run velocyto.py inside the scrnaseq pipeline.

Velocyto is tricky beacause it wants to link files and break a little the nextflow paradigms.

My actual solution is to stage copy input files instead of linking.

stageInMode 'copy'

This is due to the fact that velocyto.py command line use: Path(resolve_path=True) and breaks the nextflow logic of symbolic links.

Input file must be cellsorted.

To avoid running samtools sort FROM velocyto we need to create a file with:
samtools sort -t CB -O BAM -o cellsorted_possorted_genome_bam.bam possorted_genome_bam.bam

If velocyto.py in the work dir find a file named EXACTLY cellsorted_[ORIGINALBAMNAME] it will skip the samtools sort step.

THIS IS WHY I NEED TO HAVE IN work dir a file cellsorted_ in the same dir

reference: https://velocyto.org/velocyto.py/tutorial/cli.html#notes-on-first-runtime-and-parallelization