Issue with Chip-seq pipeline: jobid: 27: one of the commands exited with non-zero exit code; note that snakemake uses bash strict mode! about seq2science HOT 8 CLOSED

from seq2science.

from seq2science.

Thanks for the issue! I think there's actually a problem on "our" side here. Seems like we haven't defined the environment properly and we are making use of the shuf command which isn't by default on all systems (e.g. on a mac)...

Let me take a look!

For now you can turn off the deeptools figures for the QC report as this is purely used for visualization only. In the config.yaml add:

deeptools_qc: false

from seq2science.

I don't have a mac, so you will have to check for me if this solution fixes it 😄

This file should already exists:

/opt/miniconda3/envs/seq2science/lib/python3.8/site-packages/seq2science/envs/bedtools.yaml

Can you change it so that it has coreutils added:

name: bedtools
channels:
  - conda-forge
  - bioconda
  - defaults
dependencies:
  - bioconda::bedtools=2.29.2
  - conda-forge::coreutils=9.1
  - conda-forge::conda-ecosystem-user-package-isolation=1.0

And then try seq2science again? Seq2science should automatically detect the environment definition has been changed and will re-install the environment.

from seq2science.

Have you had the time to test if this works?

from seq2science.

from seq2science.

Not sure, 16GB isn't too much indeed for preprocessing.. Which step(s) is the one crashing? Still the random_subset_peaks rule? Most rules make a logfile, perhaps there is a clue about what goes wrong?

https://vanheeringen-lab.github.io/seq2science/content/faq.html#one-of-the-rules-failed-and-the-log-turns-red

from seq2science.

Your issue(s) should be solved with the latest 0.9.7 release. Let me know if you encounter anything else!!!

from seq2science.