DragMap fails if --trim_fastq flag is enabled

[Ilaria-Pirona]

Description of the bug

Good evening,
I am experiencing problems with Sarek's pipeline. When I try to remove the adapters from my WES data, the dragmap tool runs into "Segmentation Fault".
Do you have any idea what the problem might be?

Command line:
nextflow run main.nf
--input sample_sheet_3.csv
--genome GATK.GRCh38
--aligner dragmap
--save_output_as_bam
--step mapping
--outdir './results'
--save_output_as_bam
--wes false
-resume
-profile aws
--trim_fastq

Command used and terminal output

No response

Relevant files

No response

System information

No response

[FriederikeHanssen]

What version are you running? Can you send the error message? there are several issues with dragmap that were never resolved: Illumina/DRAGMAP#47

[Ilaria-Pirona]

The error is:

line 8: 115 Segmentation fault (core dumped) dragen-os -r dragmap --num-threads 12 -1 0002.11718FA_1.fastp.fastq.gz -2 0002.11718FA_2.fastp.fastq.gz --output-directory output --output-file-prefix 11718FA 2> 11718FA.dragmap.log

and I am using dragmap version:
root@f1bc5cc9b074:/# dragen-os --version
1.2.1

pulled from
quay.io/biocontainers/mulled-v2-580d344d9d4a496cd403932da8765f9e0187774d:5ebebbc128cd624282eaa37d2c7fe01505a91a69-0

[edmundmiller]

Can you confirm it works without the --trim_fastq param?

[Ilaria-Pirona]

Yes I tried multiple times, and when i added --trim_fastq it failed.

Sarek by default put the arguments
"--disable_adapter_trimming --split_by_lines 200000000 " in the process.

However, "--disable_adapter_trimming" to me is apparently in conflicts with "--detect_adapter_for_pe " which is written inside the module.

[FPGro]

This issue persists as of version 3.4.3. Related: Illumina/DRAGMAP#11

In my case, specifying any of --trim_fastq and/or --trim_nextseq 10 results in dragmap segfaulting. I also tried --three_prime_clip_r1 2 and --three_prime_clip_r2 2 and these don't seem to cause issues.

Since dragmap does not seem to be actively maintained at the moment, but there is an issue in place to replace it with a better maintained fork #1358, I suggest putting a warning into the parameter docstrings of --aligner that fastp is currently buggy with dragmap.