So I got the the DAS_tool contig2bin and my extend...
# mag
j
So I got the the DAS_tool contig2bin and my extended version of the same file implemented here: https://github.com/nf-core/mag/pull/965 When I run
nf-test
locally it runs fine with no error, but I'm getting snapshot differences: b ut I don't know why šŸ˜ž https://github.com/nf-core/mag/actions/runs/20854536246/job/59917663204?pr=965#step:4:944 It almost looks like to me like slight variation in the bins that are being output, but I don't know why that would be :\ Any suggestions?
šŸ‘€ 1
j
Do you get the same issue a second time if you re-trigger the CI tests?
j
I think so, as I pushed a commit and same issue but will re-run now just to double check (and then can compare the two)
j
There are definitely issues in general with the binners being stochastic in what they output. But I don't see why the changes you've made would materially change anything coming out the binners/going in to DAS Tool
j
Yeah precisely, that was what has surprised me
And annoying because I wanted to finish the aDNA PR today 😭
šŸ˜“ 1
Ugh there is varaibily...
shakefist 1
j
Are the various reproduciblity options all set? metabat_rng seed etc.?
j
Aha...
Because this is assembly input at least...
(but
metabat_rng
seed is not removed either)
I'll run on github code spaces while having lunch and see what happens there
šŸ‘ 1
Also works fine
image.png
shared a GIF
@Diego Alvarez Saravia any ideas ? šŸ˜…
Could you maybe try running my branch and
nf-test test --tag test_assembly_input --profile +<XYZ> --update-snapshot
and see if it passes for you too?
d
I’m running your branch now. The thing that seems more odd to me the different process count (294 vs. 296). Do you know what those processes are?
j
Nope...
I have added a few files though
So that could be expected
d
Fails for me
j
Whaaaat
And if you run on github code spaces?
What's going on
d
Weird. In what environment are you running the tests? docker or apptainer? linux?
j
docker ubuntu locally
And whatever github codespaces runs on with docker
Hrmmm wait something else is wrong... my local test is missing on codespaces is missing the contig2bin fle...
No wait no it's not
that loosk right
Thgat's what the
output
tree looks like for me...
d
Give me your bin_summary too
j
bin_summary(1).tsv
d
and the execution timeline too
j
execution_timeline_2026-01-12_11-47-12.html
d
It seems that it's semibin
j
It's not executing on me and codspaces right?
d
For you semibin produces 2+1 bins, for me 2+2
j
That's what the only thing I didn't really recognise but I don't know why
Hrm
SHould we skip it/
d
image.png,image.png
j
We still test it in
test_hybrid
d
Ok, let's skip it then.
j
OK I'll push that
Deactivating semibin2 worked! https://github.com/nf-core/mag/pull/965 Is it OK with you for me to merge @Diego Alvarez Saravia (as it was your identiifcation šŸ™‚ )
I have a āœ… from Jim already
d
Sure, merge it!
ā¤ļø 1
j
Thank you!
j
Hej! Is it something similar that you are fixing? It is not working for any of my samples (V5.3) Workflow execution completed unsuccessfully The exit status of the task that caused the workflow execution to fail was:
1
Copy code
Error executing process > 'NFCORE_MAG:MAG:BINNING_REFINEMENT:RENAME_PREDASTOOL (MEGAHIT-MetaBAT2-MMBO_02)'

Caused by:
  Process `NFCORE_MAG:MAG:BINNING_REFINEMENT:RENAME_PREDASTOOL (MEGAHIT-MetaBAT2-MMBO_02)` terminated with an error exit status (1)


Command executed:

  if [ -n "MEGAHIT-MetaBAT2-MMBO_02.1.fa MEGAHIT-MMBO_02.contigs.fa MEGAHIT-MetaBAT2-MMBO_02.10.fa MEGAHIT-MetaBAT2-MMBO_02.4.fa MEGAHIT-MetaBAT2-MMBO_02.2.fa MEGAHIT-MetaBAT2-MMBO_02.3.fa MEGAHIT-MetaBAT2-MMBO_02.11.fa MEGAHIT-MetaBAT2-MMBO_02.12.fa MEGAHIT-MetaBAT2-MMBO_02.5.fa MEGAHIT-MetaBAT2-MMBO_02.6.fa MEGAHIT-MetaBAT2-MMBO_02.7.fa MEGAHIT-MetaBAT2-MMBO_02.9.fa MEGAHIT-MetaBAT2-MMBO_02.8.fa" ]
  then
      for bin in MEGAHIT-MetaBAT2-MMBO_02.1.fa MEGAHIT-MMBO_02.contigs.fa MEGAHIT-MetaBAT2-MMBO_02.10.fa MEGAHIT-MetaBAT2-MMBO_02.4.fa MEGAHIT-MetaBAT2-MMBO_02.2.fa MEGAHIT-MetaBAT2-MMBO_02.3.fa MEGAHIT-MetaBAT2-MMBO_02.11.fa MEGAHIT-MetaBAT2-MMBO_02.12.fa MEGAHIT-MetaBAT2-MMBO_02.5.fa MEGAHIT-MetaBAT2-MMBO_02.6.fa MEGAHIT-MetaBAT2-MMBO_02.7.fa MEGAHIT-MetaBAT2-MMBO_02.9.fa MEGAHIT-MetaBAT2-MMBO_02.8.fa; do
          if [[ ${bin} =~ MEGAHIT-MetaBAT2-MMBO_02.([_[:alnum:]]+).fa ]]; then
              num=${BASH_REMATCH[1]}
              mv ${bin} MEGAHIT-MetaBAT2Refined-MMBO_02.${num}.fa
          else
              echo "ERROR: the bin filename ${bin} does not match the expected format 'MEGAHIT-MetaBAT2-MMBO_02.([_[:alnum:]]+).fa'!"
              exit 1
          fi
      done
  fi
  
  cat <<-END_VERSIONS > versions.yml
  "NFCORE_MAG:MAG:BINNING_REFINEMENT:RENAME_PREDASTOOL":
      coreutils: $(echo $(mv --version 2>&1) | sed 's/^.*(GNU coreutils) //; s/ Copyright.*$//')
  END_VERSIONS

Command exit status:
  1

Command output:
  ERROR: the bin filename MEGAHIT-MMBO_02.contigs.fa does not match the expected format 'MEGAHIT-MetaBAT2-MMBO_02.([_[:alnum:]]+).fa'!
j
@Juliana Assis no, your error is completely different... That's a new one!
āœ… 1
Could you send your input sample sheet? (Can be privatel by DM, if you wish)