Hi guys!
Thank you very much for the pipeline :)
...
# denovotranscriptr
Roma Kurilov
12/15/2024, 3:05 PMHi guys!
Thank you very much for the pipeline :)
I would like to ask @Avani Bhojwani a question. I am currently trying to run denovotranscript on my institute's cluster, and I am getting an error during the BUSCO - metaeuk step. I am attaching nextflow.log and busco_config.ini . I am not really sure how to fix this error (maybe I need to specify metaeuk path differently in the config file? 🤔). I wonder if you could have a quick look, maybe it would be easier for you to spot the problem.
a
Avani Bhojwani
12/16/2024, 3:46 AMHi Roma,
I haven't tried using the busco module with a config file before, but I can make a guess. Maybe the ini file should point to the
metaeuk/usr/local/bin/r
Roma Kurilov
12/16/2024, 9:10 AMHi Avani, thank you very much for having a look, and for the suggestion, I'll try that.
Roma Kurilov
12/18/2024, 10:43 AM@Avani Bhojwani it worked! thank you very much for your help :)
a
Avani Bhojwani
12/18/2024, 10:53 AMGood to hear! No worries :)
🙂 1
r
Roma Kurilov
12/21/2024, 11:18 AMHi @Avani Bhojwani, I would like to ask one more question.
After successful pipeline run with one sample I launched the job with 27 samples and it failed during SPAdes step ("mmap(2) failed. Reason: Cannot allocate memory. Error code: 12", I attach the full log).
As far as I understand from reading issues on denovotranscript github, I need to make changes to the config file (https://nf-co.re/docs/usage/configuration#tuning-workflow-resources)
Can you please advise me which exact changes it would make sense to make in this case?
Each fastq.gz file is around 9Gb. When I run this job on cluster I request 320Gb memory in total.
a
Avani Bhojwani
12/22/2024, 7:28 AMHi Roma, are you trying to use the pipeline through the command line?
If so, you can create a config file (for example, you can call it custom.config) and add this to the file:
Copy code
process {
withName: SPADES {
cpus = { 20 * task.attempt }
memory = { 144.GB * task.attempt }
time = { 32.h * task.attempt }
}
}-c /path/to/custom.config.command.outspades.log/omics/groups/OE0436/data/kurilov/nikolas/de_novo_rna/work/93/3f41134417b204e9852823c85beb69r
Roma Kurilov
12/23/2024, 3:01 PMHi Avani, thank you very much, I'll try running it with that config file.
Just in case I attach the SPAdes log.
👍🏽 1
Roma Kurilov
02/21/2025, 9:36 AMHi Avani, I finally managed to run the pipeline on all my samples (27 samples i.e. 54 fastq files) successfully. Thank you for your help.
Just in case here is the custom_config.ini file that I used in the end (for SPADES it was critical to give more memory but not too many cpus, for BUSCO -- a bit more time, and TRANSRATE run took 10 days, i.e. around 240 hours):
process {
withName: SPADES {
cpus = { 12 * task.attempt }
memory = { 415.GB * task.attempt }
time = { 200.h * task.attempt }
}
withName: BUSCO_BUSCO {
cpus = { 12 * task.attempt }
memory = { 200.GB * task.attempt }
time = { 100.h * task.attempt }
}
withName: TRANSRATE {
cpus = { 15 * task.attempt }
memory = { 350.GB * task.attempt }
time = { 350.h * task.attempt }
}
}