spike-in normalization of bigwigs fails miserably - Githubissues

snystrom / cutNrun-pipeline

Analysis pipeline for CUT&RUN Data. Currently configured for use in the McKay Lab at UNC, but extensible elsewhere by changing configuration files.

1 stars 7 forks source link

spike-in normalization of bigwigs fails miserably #44

Closed snystrom closed 2 years ago

snystrom commented 3 years ago

Found by Alex.

Things to fix:

add samtoolsVer to envmodules loading command
get fixed rule from Alex

half-adder commented 2 years ago

Is this still a problem?

snystrom commented 2 years ago

Alex found a fix but I never got the patch, and it's not merged on McKaylab either, so I assume so.

snystrom commented 2 years ago

there's a thread on the slack somewhere about it, too if you can find it.

snystrom commented 2 years ago

By the way if you do get the fix, please send a PR or just shoot me the .patch file, I'd love to know the issue.

half-adder commented 2 years ago

Is this the thread? https://mckaylabunc.slack.com/archives/CSXV0JLFP/p1625601293097400

snystrom commented 2 years ago

Ah whoops never replied here. I am not sure if that's the thread, I am not on the Slack anymore.

mniederhuber commented 2 years ago

I think I found the error that caused the spike normalization to fail.

in rule makeSpikeNormFragmentBedGraphs there is an errant forward slash that I believe escapes the end quote. The spikeScale variable is not assigned any value. Removing that backslash seems to fix things, but I haven't tested with more than one spike-in genome.

snystrom commented 2 years ago

Oh my god. Thank you. I think about this issue multiple times a month.