Link to the Google Doc version of the pipeline: https://docs.google.com/document/d/1o5iAcs4Bk6hNrprGu31-JPdNrSWn1hSetQbGrz1v_SM/edit?usp=sharing

Moving this to "Needs Review" alongside the linked ticket (IGBF-3708) so that the pipeline documentation can be reviewed at the same time. Please note that I've kept track of encountered errors and their fixes on the linked ticket, not in the pipeline documentation.

Thanks for the draft!

Change requests:

• Please move Table 1 to the end of the document.
• Make a new section "Introduction" and add some filler text "explain what this document is for" in the Introduction. We'll add more later!
• Please remove version-controlled code from the document. Instead, link to the file in bitbucket, e.g., https://bitbucket.org/lorainelab/tardigrade/src/main/src/prefetch.sh
• Include some text explaining what the user will see in their file system after the "prefetch.sh" command has run. What should they observe after running it?
• Instead of running "prefetch" in "fastq" run it inside another directory called "sra" (you make it). This will ensure that the "sra" files get downloaded into a single location where no other important files will reside. Then, after the ".sra" files get downloaded and used to create the "fastq" files, we can delete the "sra" directory and all its big files in a single command.
• Run the "fasterq-dump" step inside the "sra" directory after runing the "prefetch" step.
• After that, we use "gzip.sh" to compress all the fastq files.
• After the "fasterq-dump" and "gzip" steps finish, we'll move the compressed files to a new directory called "fastq" using unix "mv" command.

Also, a note:

The "fasterq-dump" script is going to break when it encounters non-paired-end data. That is, if there is no read 2 file, it will fail. We need to develop a new version that is more able to handle non-paired-end data.

See linked ticket IGBF-3790 for the most recent trial run of the tardigrade RNA-Seq pipeline in Loraine Lab git repository called "tardigrade."