Hi I have run ampliseq with dada2 with the database sbdi gtd nf-core #ampliseq

Hi! I have run ampliseq with dada2 with the databa...

Karin Garefelt

03/20/2026, 8:40 AM

Hi! I have run ampliseq with dada2 with the database sbdi-gtdb, and now i would like to re-annotate the same ASVs with silva. What is the best way to set this up? I created new params files with a changed entry for

"dada_ref_taxonomy": "silva=138.2"

, but I am not confident that it is using the same ASVs from the last run. I am running on dardel with the pdc_kth profile. How can I know with parts of the pipline outputs were used form the cached run?

Daniel Lundin

03/20/2026, 9:44 AM

Easiest is to just run the exact same command as before, only adding

-resume --dada_ref_taxonomy silva=138.2

on the command line. Parameters given on the command line take precedence over params in files.

Daniel Lundin

03/20/2026, 9:44 AM

Nextflow will show you which steps are cached.

Karin Garefelt

03/20/2026, 9:55 AM

Thank you! Just to confirm, does "cached" here mean that it is using the files from a previous run?

Daniel Lundin

03/20/2026, 9:55 AM

Correct.

Daniel Lundin

03/20/2026, 9:56 AM

Everything upstream of DADA2 taxonomy assignment should not run. Several steps downstream of DADA2 taxonomy might run, e.g. QIIME2. If you don't want that, set

--skip_qiime

and perhaps other skip params.

Karin Garefelt

03/20/2026, 9:57 AM

Super, thank you!

👍 1

Karin Garefelt

03/20/2026, 10:10 AM

I think it looks like it used 95 cached samples, so not all samples were retrieved from the cache. I suspect this could be from me interrupting some runs before they finished during testing. Can I choose to specifically used cached inputs from a certain run (intergalactic_fermat) ?

Karin Garefelt

03/20/2026, 10:15 AM

I read in the nextflow documentation that you can use the session ID! Seemed to work like i intended 🙂

👍 1

5 Views

Open in Slack

Previous Next