-
Notifications
You must be signed in to change notification settings - Fork 583
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Picard MarkDuplicates 1 out many thousands of runs broke #2063
Comments
@adthrasher did some investigation and discovered the problem. More of a bug in Picard than MultiQC. The BAM in question has a read group in the header that doesn't have any reads associated with it. Which would be why I've only seen this bug once out of thousands of runs. Look at
The last row has a dangling tab literal. That leaves the empty string as the value for |
Looks like a duplicate of #2061 MultiQC should never have an unhandled exception, so something to fix here either way. |
@a-frantz, would you be able to attach the bad output, or even better - create a PR in https://github.com/ewels/MultiQC_TestData? The block you copied unfortunately has all the space characters messed up, and I can't guess what was incorrect in the original file, this can't reproduce the issue. It's generally better to attach examples as files :) |
@vladsavelyev sorry about the bad code snippet. But it comes directly from the file I attached in the original bug report. That should suit your needs. |
@a-frantz, thank you for the example! Managed to reproduce the issue now, and implemented a proper fix. |
Description of bug
Very odd. I've run this version of Picard and this version of MultiQC literally thousands of times without an issue. I took a look at the offending MarkDups file, and it seems well-formed to me. Also looked over the stderr logs from my
picard MarkDuplicates
call, and nothing seems off (although with how verbose Picard logs are, I might've missed something).I'm going to clear the
miniwdl
cache for this run and see if it happens again. Either way, thought you'd want to see this. Based on the traceback looks like it should be an easy fix.File that triggers the error
SJMB027_D.WholeGenome.MarkDuplicates.metrics.txt
MultiQC Error log
Before submitting
The text was updated successfully, but these errors were encountered: