Use multiple genomes and annotation for transcriptome mapping

[drpatelh]

Many thanks for contributing to nf-core/nanoseq!

Please fill in the appropriate checklist below (delete whatever is not relevant). These are the most common things requested on pull requests (PRs).

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 necessary, also make a PR on the nf-core/nanoseq branch on the nf-core/test-datasets repo
• Ensure the test suite passes (nextflow run . -profile test,docker).
• Make sure your code lints (nf-core lint .).
• Documentation in docs is updated
• CHANGELOG.md is updated
• README.md is updated

Learn more about contributing: https://github.com/nf-core/nanoseq/tree/master/.github/CONTRIBUTING.md

• Added transcriptome entry to samplesheet.csv input that can either be a gtf, fasta or will be resolved if iGenomes reference.
• Replaced if statements with when where possible to make it easier to read the pipeline
• Add gtf2bed conversion process and container
• Fixes FastQC fails when fastq file is not in gzip format #43
• Fixes Transcriptome alignment #32
• Fixes Transcriptome aware genome alignment #31

Im not going to have much time to work on this for the rest of the week because Ive spent the past few days pretty much glued to getting this solved. Some things left to do:

• There still may be some bugs in the logic so it will need extensive testing with different entries for genome and transcriptome, and by using the different --skip flags to see if the channels are all defined properly.
• Tests will also fail because the samplesheet.csv on test-datasets doesnt have a transcriptome entry
• Add documentation.

If someone fancies addressing the remaining points then its probably wise to update the samplesheet.csv on test-datasets, re-trigger the Travis tests here, merge after passing (hopefully will work first time, if not I can push a commit or two if required) and then some heavy testing 👍 😅

[lwratten]

Looking really good! 👍
Very good timing too! I just started implementing this yesterday - you really saved me a whole lot of headache with this one so thank you very much! 😁
Just a few things here and there from me that might make it a bit more robust/speed up certain processes.

Also was thinking it might be good to alter the output sub directories to reflect whether a sample was aligned to genome or txome - as this will not show up in the file name. What do you reckon?

Also, I'll go through and change the samplesheet now so we can get tests passing. Happy to implement some of the suggestions below too after a merge!

[lwratten]

Do we still need this?

[drpatelh]

It should have been removed 🤔

[drpatelh]

Ok. Ive tested all of the commands in .travis.yml and its working now.

Thanks for adding the transcriptome entry to the test-datasets @lwratten ! I was sitting here thinking the tests should be failing until I looked 😄 We still do need to resolve #37 too.

I dont think we should complicate things by having genome/transcriptome in the file/directory name. It seems tempting but everything should be stored in the samplesheet.csv. Ive tried to factor in as many possibilities as I can but its most likely that people will run the pipeline on a single reference genome/transcriptome. I think properly documenting the exact behaviour of the pipeline is probably more important. Also, please do test the pipeline by using separate genome/transcriptome entries in different formats to see if we are getting the expected behaviour.

[drpatelh]

Would be good if you can double-check the logic I have used for minimap2, graphmap2 for DNA, RNA etc.

[lwratten]

Logic for minimap2 and graphmap2 call look good!
Will have a solid test with lots of different references formats and combos once we're merged! 👍