Normalised read counts from TMM

Question

[patel.anushi]

How can we obtain normalised read count matrix from TMM (EdgeR)? and how is it different from CPM?

Thanks!

Answer 1

[igor]

Normalized read counts are the raw counts that have been scaled so they can be compared across samples in your experiment. In edgeR, that's done by calcNormFactors: https://www.rdocumentation.org/packages/edgeR/versions/3.14.0/topics/calcNormFactors

CPM are "counts per million" that take into account the gene length and library size. EdgeR has a separate function for that: https://www.rdocumentation.org/packages/edgeR/versions/3.14.0/topics/cpm

There topic has been previously discussed in detail here:

• https://support.bioconductor.org/p/69433/
• RNA-seq normalization: How to use TMM and rpkm() in EdgeR
• How to TMM after HTSeq-count?