https://nf-co.re logo
Join Slack
Powered by
I can also add: the gtdbtk_summary.tsv file is emp...
# mag
h
I can also add: the gtdbtk_summary.tsv file is empty, and I am using mag version 5.0.0
j
Did any of your GTDB jobs pass successfully? I think that's one of the modules it unusefully gives exit code 1
h
The db_prep process passed at least. And yea I got an exit code 1 on the summary process
j
Sorry, I mean the GTDBTK_CLASSIFYWF job (I think it's called)
h
that one did not run at all, or wasn't called i mean
i did find that weird
j
OK that will be why it failed
Did you get any any warnings about 'Did not reach contamination or completeness thresholds' in your log?
h
Not that I can see, no
The classifywf process doesn't seem to be triggered. I assume they run in parallell with BUSCO etc?
j
That's odd...
Could you grep 
.nextflow.log
for 
No contigs passed GTDB-TK min
h
Here it is: 
grep: No contigs passed GTDB-TK min: No such file or directory
j
try 
grep 'No contigs passed' .nextflow.log
h
nothing
as in, the same as above
j
Hrmmm
Can you send the command you used?
h
Copy code
nextflow_25.04.7 run /cluster/projects/nn10070k/pipelines/nf-core-mag_5.0.0/5_0_0/main.nf --input mag_samplesheet.csv --outdir mag_16102025 --reads_minlength 50 --fastp_qualified_quality 25 --fastp_cut_mean_quality 25 --fastp_trim_polyg --host_fasta /cluster/projects/nn9305k/active/hkaspersen/references/shotgun_ref/broiler_turkey_human_genome.fna --host_fasta_bowtie2index /cluster/projects/nn9305k/active/hkaspersen/references/shotgun_ref --skip_prodigal --skip_prokka --skip_metaeuk -profile apptainer -work-dir $USERWORK/mag -c /cluster/projects/nn10070k/projects/pahw_SOA8_ST2.5/spiking_experiment_2025/results/mag/saga_mag.config --skip_spades --skip_concoct --binning_map_mode own --min_contig_size 1500 --bin_domain_classification --binqc_tool busco --busco_db_lineage bacteria_odb12 --busco_db /cluster/projects/nn10070k/databases/bacteria_odb12.2025-05-14.tar.gz --busco_clean --gtdb_db /cluster/projects/nn10070k/databases/gtdbtk_r220_data.tar.gz -resume
j
And the full 
.nextflow.log
file
?
h
.nextflow.log
got another error for this run though, but the gtdbtk didn't run here either
j
Sorry, still half looking after kids
Looks like to me from that log, that the pipeline just didn't get to GTDBTK_CLASSIFY yet because of the error
You have the error here:
Copy code
Nov-25 12:17:52.411 [PublishDir-2] ERROR nextflow.processor.PublishDir - Failed to publish file: /cluster/work/users/hkaspersen/mag/fc/bb2e5dce5a4a7c1c0827dcd5c1f656/Sample1-1-ecoli-plateau-bacteria_odb12-busco; to: /cluster/projects/nn10070k/projects/pahw_SOA8_ST2.5/spiking_experiment_2025/results/mag/mag_16102025/GenomeBinning/QC/BUSCO/Sample1-1-ecoli-plateau/Sample1-1-ecoli-plateau-bacteria_odb12-busco [copy] -- See log file for details
dev.failsafe.FailsafeException: java.nio.file.DirectoryNotEmptyException: /cluster/projects/nn10070k/projects/pahw_SOA8_ST2.5/spiking_experiment_2025/results/mag/mag_16102025/GenomeBinning/QC/BUSCO/Sample1-1-ecoli-plateau/Sample1-1-ecoli-plateau-bacteria_odb12-busco/MEGAHIT-MetaBAT2-Sample1-1-ecoli-plateau.57.fa/run_bacteria_odb12/busco_sequences
And then a few lines later:
Copy code
[process] NFCORE_MAG:MAG:GTDBTK:GTDBTK_CLASSIFYWF
  status=ACTIVE
  port 0: (queue) OPEN  ; channel: -
  port 1: (value) bound ; channel: -
  port 2: (value) bound ; channel: use_pplacer_scratch_dir
  port 3: (cntrl) -     ; channel: $
I think this BUSCO publishing error may have already been fixed by @Diego Alvarez Saravia in the latest reelase, maybe? 🤔
d
Yes, busco publishing should be fixed on 5.2.0
h
aha! Thanks for digging for me!
Question then: can i resume the pipeline with a new version of it?
because the assembly steps take a ridiculous amount of time and resources, so i'd like to not have to run them again...
d
In theory yes, because we have not changed anything on assembly or earlier steps. But I'm not sure if template update could mean breaking changes, do you know @James Fellows Yates?
j
Resume remains a bit of a mystery 😕
d
If not, you can you the pipeline in assembly_input mode, so you don't have to do the assembly again (but binning will have to)
party this 1
j
I think the template changes may result in major changes
h
ah right, inputting assemblies make sense
thanks!
Open in Slack
PreviousNext
Powered by