New module: Kraken2/Bracken on Unaligned Sequences for Contamination Detection

[egreenberg7]

Closes #271. This contribution adds Kraken2/Bracken as an optional quality control step to the rnaseq pipeline for the HISAT2 and STAR/Salmon aligners. Contamination is a widely reported issue in rna-sequencing data, and the use of metagenomics tools can be used to address this. Kraken2 is particularly strong at detecting low levels of pathogens, which makes it appropriate for this task. This PR adds the option of providing a Kraken2 database to perform taxonomic classifications on unaligned reads.

Note: If the --bracken-precision parameter is set to something other than 'S', the current MultiQC version does not work properly. In future versions of MultiQC, this will not be an issue (see this MultiQC bug fix).

PR checklist

• This comment contains a description of changes (with reason).
• If you've fixed a bug or added code that should be tested, add tests!
• If you've added a new tool - have you followed the pipeline conventions in the contribution docs
• If necessary, also make a PR on the nf-core/rnaseq branch on the nf-core/test-datasets repository.
• Make sure your code lints (nf-core lint).
• Ensure the test suite passes (nextflow run . -profile test,docker --outdir <OUTDIR>).
• Check for unexpected warnings in debug mode (nextflow run . -profile debug,test,docker --outdir <OUTDIR>).
• Usage Documentation in docs/usage.md is updated. [I do not think this is needed for my PR.]
• Output Documentation in docs/output.md is updated.
• CHANGELOG.md is updated.
• README.md is updated (including new tool citations and authors/contributors).

[MatthiasZepper]

LGTM (Let's get this merged!), provided that you rebase before to the latest dev (or merge in those changes).

@maxulysse ideally should help with the subway map, though.

[egreenberg7]

The subway map has been updated, though not the animated one. I had an issue with the merge request, so I unintentionally now have all the 3.15 commits showing up here. Unsure if we can do something about that here or if I should open a new, clean PR from my fork. I'm guessing opening a new, clean PR may be easiest. Otherwise, the linting failure seems to be due to the HISAT2 patch that allows for using contaminant screening to have the unaligned reads saved for further processing downstream.

[Shaun-Regenbaum]

Congrats @egreenberg7 :) Truly awesome work!

[egreenberg7]

New PR will be opened to show more clear version changes due to the merge issues. I'll also copy some of the comments over for the record

[maxulysse]

No need to copy over the comments, just mention this PR, that'll be enough