GallVp
October 26, 2024, 11:22pm
1
We are trying to add fq2bam_meth module to nf-core/modules. Please see the PR for full details: add fq2bammeth tests + snap by sateeshperi · Pull Request #6857 · nf-core/modules · GitHub
However, fq2bam_meth is failing with the exisiting paired end sample data.
The nf-test code is available on the PR: modules/modules/nf-core/parabricks/fq2bammeth/tests/main.nf.test at f590541134c86619f4c0ba5b624e979372bfa686 · nf-core/modules · GitHub
The full error log is also available: add fq2bammeth tests + snap by sateeshperi · Pull Request #6857 · nf-core/modules · GitHub
OS: cat /etc/os-release
Machine: g4dn.xlarge
Thank you!
Hi @GallVp , thank you for reaching out.
What GPU are you running this on? Based on the error log, it looks like the GPU could be running out of memory, which can be fixed by adding the --low-memory flag to your run command.
GallVp
November 4, 2024, 9:32pm
3
Thank you @gburnett
I have been using g4dn.xlarge on AWS which has a NVIDIA T4 Tensor Core GPUs.
Yes, some other failing tests were successful after we added the --low-memory flag. Not this one. This one fails despite the --low-memory flag.
The full command and the error log can be seen in this message: add fq2bammeth tests + snap by sateeshperi · Pull Request #6857 · nf-core/modules · GitHub
Thank you for sharing. Is this data publicly available so I can reproduce this?
GallVp
November 15, 2024, 8:18pm
5
Yes, the data is publicly available. It is listed here:
process {
"""
input[0] = Channel.of([
[ id:'test', single_end:false ], // meta map
[
file('https://github.com/nf-core/test-datasets/raw/methylseq/testdata/Ecoli_10K_methylated_R1.fastq.gz', checkIfExists: true),
file('https://github.com/nf-core/test-datasets/raw/methylseq/testdata/Ecoli_10K_methylated_R2.fastq.gz', checkIfExists: true),
]
])
input[1] = Channel.of([
[ id:'test' ], // meta map
file('https://github.com/nf-core/test-datasets/raw/methylseq/reference/genome.fa', checkIfExists: true)
])
input[2] = BWAMETH_INDEX.out.index
input[3] = []