A community effort to collect a curated set of analysis pipelines built using Nextflow.

nf-core

I was hoping to use this to build a STAR index for a mouse genome.  I ran it and it didn't appear to actually do anything to build anything? Or am I running it wrong?

I realize that this appears to be a 0.1 version, so I have low expectations.

Hi <@U02J6NUJQSE> could you share what was your command line?

I have been doing some heavy refactoring since 0.1, getting towards a 0.2 soon, and hopefully a 1.0 later on

```~/bin/nextflow run ./references \
   -profile unc_lccc \
   --input asset.yml \
   --outdir $REF_DIR/GRCm39```
where `asset.yml` contains:

```- genome: GRCm39
  fasta: "<https://ftp.ebi.ac.uk/pub/databases/gencode/Gencode_mouse/release_M36/GRCm39.primary_assembly.genome.fa.gz>"
  gtf: "<https://ftp.ebi.ac.uk/pub/databases/gencode/Gencode_mouse/release_M36/gencode.vM36.primary_assembly.annotation.gtf.gz>"
  source_version: "GENCODE_M36"
  readme: "./README.md"
  source: "GENCODE_M36"
  source_vcf: "dbSNP"
  species: "mus_musculus"
  vcf: "<ftp://ftp.ncbi.nih.gov/snp/organisms/archive/mouse_10090/VCF/00-All.vcf.gz>"```

When I ran it, it gave me a couple warnings about not knowing what a couple of the options in the yml were, e.g. I think it didn't know about vcf.  going from memory here though.

I re-ran and the warnings  were:
```WARN: Found the following unidentified headers in /[full_path_hidden]/asset.yml:
	- source_vcf
	- vcf```

Hmm, I also got:
```WARN: No tools specified```

which makes me think I'm definitely running it wrong.  This makes me think that the documentation just isn't sufficiently mature to to tell me how to run it?

<https://nf-co.re/references/usage>

just gives largely generic nf-core boilerplate for sample-oriented runs and

<https://github.com/nf-core/references?tab=readme-ov-file>

doesn't say how to specify tools.

It stages the files, but the only nextflow processses it runs are a few ARCHIVE_EXTRACT processes and MULTIQC

How does one specify multiple tools?

Can I put that in the asset.yml or in a params file?

I have figured out how to put `tools: "comma,separated,list"` into a `params.yml`

sorry, missed your last message, but happy to see you figured it out

Thanks Maxime!  Now to figure out a way to make the FASTA staging happen faster...  :wink:

I have 1 recommendation, just downloading the files beforehand, and pointing to them locally

but yeah, I do think this is kinda the pain point at the moment

Yeah, I know I can do that.  I am enamored by the idea of a one-shot "do everything" reference download/build-indexes thing, so I'm trying it out.  Though,  after an hour and being only halfway through downloading a 740MB FASTA.gz is painful....

ok, I was thinking of adding a process and and option to force the download of the input, that way it could always be restarted from input that were already downloaded

I'm thinking this whole project as being quite flexible, especially in such an early stage

After the frustratingly slow download, most of the rest of the workflow worked great.  One process failed, but with `-resume` it seems to  be running fine on the second attempt.

Ok, so I'll note, improve the documentation and the warning message