Just curious if anyone has been working on plasmid identification as part of the workflow, particularly on the long read assembly side, or if that would perhaps be better suited to a module in funcscan to work with the assembly/unbinned outputs from nf-core/MAG?

That's a very good question! I don't think anyone has been working on this in terms of code at least. My personal opinion would be that it would be an interesting thing to support within the community though! I think if it's simply plasmid detection with a single tool (or a couple that do the same thing), it would fit in the scipe of mag - as it would sort of be equivalent to e.g. genomad or tiara for 'identifying' viral or eukaryotic sequences. However if it would be more involved, it would be better in a downstream pipeline, e.g. similar to what #C05E9C2B2BG was trying to do. I'm not sure if it would fit in #C02K5GX2W93 because that is meant to be quite specifically gene centric, not for other genomic features (so to say)

I think this is technically already done in GeNomad? Though obviously nothing happens with that information downstream of just running the tool

I agree. If this would mean a few tools, I see its place in mag. But if it require some complex logic, I guess it would be outside scope.

Ahh perfect, I did not catch that GeNomad was an optional module incorporated and that the plasmid files are included (not just viruses), so that might be sufficient enough to add that flag!