Spike ins normalization

[EmanuelSoda]

Description of feature

Hi nf-core team,
first of all thanks for the amazing work!
I was wondering if the pipeline already takes in consideration the spike ins and in that case how to use it.
If this function in not implemented yet, it is possible to implement it?

Thanks in advance

Emanuel

[cjfields]

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?

[EmanuelSoda]

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.

[chlazaris]

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.

[cjfields]

@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.

[chlazaris]

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

[cjfields]

@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?

[chlazaris]

@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!

[cjfields]

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

[ahepperla]

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

[nservant]

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

[chlazaris]

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.