MasterOfPores - Dorado killed

[pcharoentong]

Hi,

I have trouble using the MasterOfPores pipeline. I'm trying to run mop_preprocess, and as soon as the downloadModel step finishes, the baseCall step starts and then fails with an error, saying that the dorado process has been killed.

Attached you'll find the output on the console as well as the nextflow.log and my params yaml file.

I made sure the work directory is on a drive with plenty of space. I also tried adding "process.memory = '32G'" to the nextflow.config.

I tried running the dorado command from within the work directory manually. This worked and wasn't killed:

$ singularity exec --nv -B /ag-jaeger:/ag-jaeger /tmp/singularity/cache/ontresearch-dorado-shaa2ceb44eb92c08f9a3a53f97077904d7e23e28ec.img dorado basecaller --models-directory ./ sup ./ > pod5---24.bam

But when it's run from the nextflow pipeline it is always killed immediately.

Would you have any idea why this might be happening? Thank you very much and looking forward to hearing from you.

Best,

Kay

params.test_044369.yaml
nextflow.log
MasterOfPoresError.txt

[yeroslaviz]

error 137 is usually a sign of lack of memory. Can you try providing it with more RAM?

what do you get when you change to the working directory /tmp/mop4_workdir/1c/783200c050dc0c48c777f02f9936fc and run the re-run the command bash .command.run ?