CUT&Tag signal normalization for generating bigWig files

Question

[luminea]

Hi everyone,I have CUT&Tag data for histone marks and I want to generate bigWig files from my BAM files. I also have IgG control files. I’m wondering what would be the best approach:

• using bamCoverage with CPM or RPKM normalization?
• using HOMER?
• or generating bigWig files in CPM first, and then using the IgG files to normalize the signal with bigwigCompare?

Has anyone tried these approaches and could share their experience or recommendations? Thanks in advance for your advice!

[RD]

CUT&Tag shares protocol similarities with ATAC-seq, so I'd recommend using RPGC normalization via DeepTools. You might also consider testing RPKM normalization to see which method yields better results for your data.

[ATpoint]

I would ignore the IgGs and only use it during peak calling. For the IP samples I would scale bigwigs by a size factor that takes into account both library size and composition, via edgeR, see for code and discussion and motivation of using it: ATAC-seq sample normalization