is it necessary to normalize between a Paired-End and Single-End library?

Question

[Zee_S]

Hello everyone!

I would like to look at the correlation between an RNAseq dataset and a Chip-seq dataset. I m using a CPM score per bin method but I'm not sure how the two libararies need to be normalised to e/o first.

the RNA seq library is PE. and the ChIP seq library is SE. how to go about it please?

Many thanks for your help!

[GouthamAtla]

What correlation are you looking at ? Its not necessary to normalize across ChIP and RNA-Seq. If you are mixing RNA-Seq data from SE and PE, it needs to be accounted.

[Zee_S]

Hello, Thank you for your reply! my RNA seq library is PE only. and the chIP seq library is SE. I want to look at the correlation between my expression data and protein occupancy data.

I made bins of the genome and I want to use bedtools to compute coverage across bins for the read libraries. but I'm not sure how the libraries need to first be normalized to compute CPM per bin.

[WouterDeCoster]

The result of Chip-Seq and RNA-seq is already quite different, so there is no need to perform additional normalization for the sequencing method.