Spike ins normalization about chipseq HOT 11 OPEN

@chlazaris it should be a call that involves the key nf-core devs who have been working on this, for example @JoseEspinosa or @drpatelh . Maybe this is a conversation that could happen on the nf-core Slack channel initially to organize something?

from chipseq.

@cjfields this is a good idea. Please, feel free to start the discussion there but include @nservant in addition to @JoseEspinosa and @drpatelh, as he was the one who developed the version with spike-in normalization and IDR. Thank you!

from chipseq.

Hi all, just commenting in support of adding spike-in normalization similar to nf-core/cutandrun. It's an invaluable technique for ChIP-seq

from chipseq.

Hi guys, sorry for the delay. Happy to further discuss if needed

from chipseq.

This would be great to add as I've seen spike-ins being used in a few ChIP-Seq projects. I believe the only place spike-in normalization is included is with the bigWig file however, @EmanuelSoda is that your experience as well?

from chipseq.

Hi,
As far as I know it is also possible to normalize in some way the bam files in order to obtain a more reliable quantification afterwards. But I am not sure how.

from chipseq.

Spike-in normalization would be a very useful addition. @nservant has written a Nextflow pipeline which provides this feature in addition to others (https://github.com/bioinfo-pf-curie/ChIP-seq). I am testing it at the moment but it would be really useful to have the official nf-core ChIP-seq pipeline support this option.

from chipseq.

@chlazaris would it make sense to combine efforts here? It seems like there is considerable overlap, but the workflow from @nservant seems to tick a few more requests like spike-in norm and IDR, and nf-core has other options the above workflow is missing.

from chipseq.

Hello @cjfields, this is what I see too. We should probably have a Zoom call to see what the best options are. My time zone is EDT. What is the best way to reach you? Thanks

from chipseq.

Just a small note, there are different methods for spike-in norm that should be considered, for example:

https://github.com/hbctraining/Intro-to-ChIPseq-flipped/blob/main/lessons/spikeIn_normalization.md

from chipseq.

Hello @nservant and everyone else interested in spike-in normalization. Since, you have already worked on a version of the ChIP-seq pipeline with spike-ins @nservant, I am wondering what could be done to incorporate the spike-in normalization feature to the official nf-core pipeline, as the broader community would benefit a lot from this. There is an active thread on the nf-core Slack workspace. @cjfields started it and can be easily found by searching for "spike" on the #chipseq channel. Thank you and I will be happy to discuss further.

from chipseq.