Workflow exits after TrimGalore with “successful” status but no alignments produced

[sansense]

Description of the bug

The workflow terminates after the TrimGalore step with a successful exit status, but no alignment files are generated. TrimGalore completes normally and its output files are present in the expected work directory (verified manually). However, no downstream alignment process is executed. In the .nextflow.log file, there is an unusual debug message stating: Skipping output binding because one or more optional files are missing.

Dec-09 14:34:27.550 [TaskFinalizer-3] DEBUG nextflow.processor.TaskProcessor - Process NFCORE_METHYLSEQ:METHYLSEQ:TRIMGALORE > Skipping output binding because one or more optional files are missing: fileoutparam<2:1>

After TrimGalore, the pipeline proceeds directly to MultiQC and then exits with a successful completion message:

-[nf-core/methylseq] Pipeline completed successfully- Completed at: 09-Dec-2025 14:35:42 Duration : 19m 15s CPU hours : 3.6 Succeeded : 4

Despite this, no alignment output is present, indicating that the alignment step is being skipped without an error being raised.

Command used and terminal output

~/bin/nextflow run nf-core/methylseq \
-r 4.1.0 -resume -profile singularity \
  --input samplesheet.csv \
  --comprehensive \
  --outdir results \
  --bismark_index /nas/longleaf/home/sant/data/genomes/medaka

Relevant files

samplesheet.csv slurm_bismark_410_test_23922691.console.log

nextflow.log

System information

Nextflow version: version 25.04.8 Hardware: HPC Executor: Slurm Container engine: Singularity OS: RHEL9 Version of nf-core/methylseq: 4.1.0

[sateeshperi]

hi @sansense could you kindly try with the current -r dev branch and see if the issue still persists. Do report back as soon as possible as we are about to release the next version. Thanks

[sansense]

hi @sateeshperi Thanks for a quick reply. I tried with the current -r dev branch, but the issue remains the same

Log files are attached

nextflow.log slurm_bismark_410_dev_24032924.log

[sateeshperi]

this is strange....what do you have in this local config bismark_410_dev/nextflow.config ?

[sansense]

nextflow.config content

process.executor = 'slurm'

// increase process qualimap memory as it is getting killed due to less JVM memory
process {
  withName: qualimap {
    memory = 50.GB
  }
}

[sateeshperi]

I am not able to reproduce the error....could try running this locally instead of on slurm and see if you are able to replicate. Also please check and provide the contents of your bismark index folder once.

[sateeshperi]

oh and how are you providing reference --fasta ?

[sansense]

I found a way to replicate the error.

1. First run the pipeline using test profile, but saving the reference using --save_reference

~/bin/nextflow run nf-core/methylseq \
-r 4.1.0 -resume -profile test,singularity --save_reference --outdir results 

2. Use that saved reference to run the bismark pipeline with test data

~/bin/nextflow run nf-core/methylseq \
        -r 4.1.0 -resume -profile singularity \
        --outdir results \
        --input samplesheet.csv \
        --comprehensive \
        --bismark_index /users/s/a/sant/projects/wgbs/sourav_wgbs_liver_bpa_f0f1f2_2025/0nextflow/profile_test_exec-slurm/results/bismark/reference_genome

The same error (no alignmnet) is produced whether I use bismark/reference_genome or bismark/reference_genome/BismarkIndex or
bismark/reference_genome/BismarkIndex/Bisulfite_Genome/ as path of bismark_index

Nextflow and console logs are included nextflow.log nf_test_reference_genome.log samplesheet.csv

[sansense]

hi @sateeshperi are you able to reproduce the error?

I've now run the above test locally on Ubuntu 22.04 LTS, without using SLURM, and I am getting the same erroneous behavior. Logs included below

nextflow.log nf_test.log nf_test.sh samplesheet.csv

[sateeshperi]

hi @sansense

It is mandatory to provide --fasta along with --bismark_index /--bwameth_index parameters see usage docs here

I will make a change in the pipeline to not start if fasta not provided in the next release.