Hi channel, I was having a look over in https://github.com/nf-core/test-datasets/tree/modules and noticed that there is a scarcity of datasets suitable for metagenomics. This is all I could find, from the README: • fasta ◦ haemophilus_influenzae.fna.gz: Multi-chromosomed reference genome file (NZ_LS483480.1 from Haemophilus influenzae strain NCTC13377) ◦ SARS-sarscov2.fasta: Reference genome file for SARS-CoV-2 (MT192765.1 Severe acute respiratory syndrome coronavirus 2 isolate SARS-CoV-2/human/USA/PC00101P/2020) Would you welcome a PR adding a new file that has a bunch of short read sequences to it, suitable to benchmark metagenomic profilers? I see eg. kraken uses the sars-cov2 files for testing, but I wonder whether a reference genome is most appropriate for this sort of application. cc @Joon Klaps @James Fellows Yates as I found you on the git blame for the README

Yes, would be very welcome!

I think in general a nice test suite of metagenomic data would be helpful: • 2-3 (reduced) reference bacterial genomes, an archaeal genome • A "metagenome" assembly that is the concatenation of the above • Some small readsets that map to those genomes - long and short • BAM files for those reads + coverage/depth files • CheckM/CheckM2 TSV results for the genomes • Maybe some kraken2 reports/etc.?

@James Fellows Yates @Jim Downie https://github.com/nf-core/test-datasets/pull/1863 would appreciate your input! 🙂