[DOCKTESTERS] Sanger pipeline taking a very long on DO50311 and issue with BWA-Mem

[Denis Yuen]

Hi,

re: bwa-mem on HCC1143

I have not tried that. Unfortunately, according to the table ( https://wiki.oicr.on.ca/display/PANCANCER/Workflow+Testing+Data ) I believe you're the first.

re: Sanger on DO50311
Yes, unfortunately it did take a bit over 8 days for me to run that donor on a pretty beefy machine (15 cpu, 125 GB ram).



________________________________
From: mikisvaz at gmail.com [mikisvaz at gmail.com] on behalf of Miguel Vazquez [miguel.vazquez at cnio.es]
Sent: November 29, 2016 6:06 AM
To: Denis Yuen
Cc: Francis Ouellette; docktesters at lists.icgc.org
Subject: [DOCKTESTERS] Sanger pipeline taking a very long on DO50311 and issue with BWA-Mem

Hi all,

I just wanted to give a partial update on my side and ask about two issues.

1) I've ran BAW-Mem on the test data HCC1143 and the process has failed, should it work for that dataset or is there a reason it shouldn't. About unaligning the BAM files I've checked the instructions on

https://wiki.oicr.on.ca/pages/viewpage.action?spaceKey=PANCANCER&title=Preparing+paired-end+data+for+upload<redir.aspx?REF=vzZnpHgQUQ-TmnK2WmjdtthDxHvxJtPVpEqrB_qN7UMgNBmdaxjUCAFodHRwczovL3dpa2kub2ljci5vbi5jYS9wYWdlcy92aWV3cGFnZS5hY3Rpb24_c3BhY2VLZXk9UEFOQ0FOQ0VSJnRpdGxlPVByZXBhcmluZytwYWlyZWQtZW5kK2RhdGErZm9yK3VwbG9hZA..>

and I ended up just doing

cat initial.bam | bamreset exclude=QCFAIL,SECONDARY,SUPPLEMENTARY  > cleaned.bam

I skipped all the stuff about the SAM header, which I didn't fully understood but it seemed like housekeeping stuff that should affect the variant calling and didn't quite apply to this test data. Could that be the issue?

I tried to run it on a normal donor but it turns out I had first to finish the test on Sanger. When I do I can save the container, thanks to Denis tip on a previous thread, and help debug it if need be. I'm sorry that I cannot give more details on the error but at the time I could not find the log files that where supposed to be there; when the next issue is resolved I'll come back to this.

2) I'm running Sanger on donor DO50311. I put the job on November 22, about a week ago, and it is still running. It seems to be doing caveman for the last few days. It going full throttle using 100% of the CPU and with running jobs not older than 15 hours, so I guess its not stuck or anything. But just in case, is taking this long normal?

Best

Miguel
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.icgc.org/mailman/private/docktesters/attachments/20161129/c1b3017b/attachment.html>