-
Notifications
You must be signed in to change notification settings - Fork 1
Troubleshooting Cheatsheet
avialee edited this page Mar 23, 2021
·
4 revisions
- Instruction to reboot the LIMS: https://github.com/epigen-UCSD/epigen_ucsd_django/wiki/Deploy-instruction
- Make sure there is only one blank or one tab between Sequencing_ID and i7index. Or LIMS will recognize blank as an index id and throw the KeyError.
- The current pipeline does not support a set with only one library.
- The current pipeline does not support a set including libraries with different genomes.
- Try to avoid using the special character ' in set name. You are able to add it in LIMS without throwing an error, but the pipeline would not work after you click to submit the setQC job.
- TSCC is probably not in good condition. Most of time, TSCC will be in good condition. After you click to submit job through LIMS, the status will quickly change to 'JobSubmitted' when you first run this job or give you a pop-up warning to let you confirm you want re-run this job. However, sometimes TSCC will run very slow. In this situation, you would not get response when you click to submit the job. And the setQC detail page will load slowly and even fail to load.
- If the error message is one of the following or other weird messages, please try change status to 'ClickToSubmit' and submit the job again. If TSCC is in good condition, it usually works.
PBS: job killed: walltime ### exceeded limit ###
ssh: Could not resolve hostname epigenomics.sdsc.edu: Name or service not known
- Instruction to install a new specie: https://github.com/epigen-UCSD/epigen_ucsd_django/wiki/Add-a-new-species-or-a-new-genome
- Apart from the above instruction written by Frank, a few things should be taken care of:
- Then ENCODE Pipeline uses 'chr1','chr2',...'chrM' naming, so if the genome and annotation file is not in this convention (like ENSEMBLE uses 1,2,..MT naming), they should be replaced to 'chr*' first (only for main chromosome).
- When calling peaks using MACS2, there is a step to extract main chromosome (https://github.com/epigen-UCSD/atac_seq_pipeline/blob/efd8811b9e3518a6784decdaea5ae4beafc41e74/modules/callpeak_macs2_atac.bds line 47). Please modify it accordingly.
RM_CHR_ := "zcat $tag | grep -Ei '(chr[0-9xylrwz]+[[:space:]])|(chrm[t]*[[:space:]])' | gzip -nc > $prefix.rm_chr.tmp.gz; [[ $(zcat $prefix.rm_chr.tmp.gz|wc -l) -gt 0 ]] && mv $prefix.rm_chr.tmp.gz $tag"
- Go to admin interface http://epigenomics.sdsc.edu:8000/admin/ and log in (only when the user's staff status is active)
- Add the user
- Go to the detail page of this user and change the permission (whether activate the user's staff status and superuser status) and the group (wet-lab or bioinformatics or manager)