Repository navigation
Arabidopsis thaliana rnaseq_collapse Segmentation fault #245
Description
Hello, I am annotating a new Arabidopsis genome with a size of less than 150 Mb using EGAPx 5.2. However, I encountered a segmentation fault during rnaseq_collapse. Interestingly, the same approach worked successfully for a 10 Gb genome, with the only difference being that I replaced the short and long reads. Could you explain why this might be happening?
Below are my YAML file and log file.
Activity
kingbin97 commentedon May 12, 2026
Below are my YAML file and log file.
command.err.log
egapx2.yaml
nextflow.log
If I use the following annotation YAML file, it will work normally:
genome: /04_Anno/05_RepeatMasker/Ath/Ath.genome.fa.masked
taxid: 3700
annotation_provider: Kingbin
locus_tag_prefix: Ath
trnascan:
enabled: true
cmsearch:
enabled: true
pstrope commentedon May 12, 2026
Can you please retry with this change?
In your egapx_config directory in the working directory, there is a process_resources.config file. Change the value of small_mem to 16GB.
kingbin97 commentedon May 14, 2026
Thank you for your suggestion. My RNA-seq data was downloaded from EBI. Later, I changed the ID of each read so that it no longer starts with SRR or ERR, and also renamed the files to not begin with SRR, ERR, etc., before using them as input. Surprisingly, I successfully obtained the annotation results.
pstrope commentedon May 14, 2026
Thank you for letting us know about this. We'll look into it.
murphyte commentedon May 15, 2026
@kingbin97 - please see my comment on #246. I think there may be a slight advantage to formatting the seq-ids like
gnl|SRA|ERR4470940.3, specifically related to handling for stranded data. We'll have to do more testing to be sure, but you might want to try your run like that instead of using theS01prefix.kingbin97 commentedon May 15, 2026
Thank you. I will follow your advice later. Thank you!