Hi I just reported github.com/nf-core/mag/issues/1101 but it's a bit weird to me because, at least looking at the code, it should skip BUSCO just fine. So not sure if it has something to do with our setup or some weird interaction with

-resume

(was run with

--run_busco

before). So yeah, let me know if you need more info.

Hi! I think it could be because of this https://nf-co.re/blog/2026/parameter-types

Yeah I was thinking the same thing but I thought parameters were always typed if you have a schema...

Yeah, it was like that until the latest Nextflow version (26.04). From that version, converting the parameters from the string to the actual type requires defining the parameters using the new typing system. We still have not adopted that, but plan to do it on next release.

Yeah I think you got it. In your linked blog post it actaully mentions this

After validation,

<http://params.my|params.my>_int

is still the string “42”, not the integer 42. Code that does arithmetic on

<http://params.my|params.my>_int

, or relies on

<http://params.my|params.my>_bool

being a real boolean in a Groovy truthiness check, will still misbehave under syntax parser v2.

I thought it would be fine because it validated but clearly not. It also needs a typed parameter block as described in the blog. Well at least it makes sense now.