Hello rna-seq developers, I have implemented Sylph as an alternative for contaminant screening (see PR here). This was mentioned as an alternative to Kraken/Bracken when that was first added, and it would be a much less memory intensive alternative tool. The PR is pretty much all ready for review and has worked locally on my device (I tested it using the minimal testset reads and a sylph database from the minimal dataset genome). I also made a PR on MultiQC to add support for the Sylph-tax output, so it may be worth waiting until that is dealt with and then MultiQC is updated in the rna-seq pipeline, but I thought this would probably be the right place to make a “ready for review except for the MultiQC” statement. Thanks!

Looks like you have some unit test failures on the MultiQC PR - if you can try to get them passing I'll get to a review ASAP.

Thanks! Locally, my module is passing the tests. Are you sure the PR fails are not because the sylphtax data files haven’t yet been added to the test-data directory?

Possible, there's an easy way to check!

Thank you very much Ezra for tackling this and doing such a fantastic job! 🤩 I think, your work will soon be adapted to #C06PHJM0Y3D as well!

Good to hear, thanks!

Thanks! I'm not sure how much time I'll have for MultiQC this week so it'll likely be the new year before I get to this now sorry

Clarification: the above was an update on the original nf-core/rnaseq PR, not the MultiQC stuff. I didn’t really intend to do more with MultiQC currently though I guess a Sylph rather than Sylph-tax module could be something for the future

Even better news! 😆