RNA-seq normalization methods between samples

Question

[chansik]

I'm a beginner in RNA-seq. I'm trying to learn RNA-seq analysis with a practical and simple analysis with public RNA-seq data.

I downloaded 9 RNA-seq files (same sample prepared on 9 different days) I've done mapping them with Hisat2. so far it seems pretty fine as average 95% of reads were mapped.

For analysis practice, if I want to identify housekeeping genes with those samples, what normalization method should I use? (e.g. read counts, FPKM or TPM? or something else?) Because I read some posts that netiher FPKM or TPM is useful to normalize between samples.

And can you recommend tools (or pipeline) to get the values?

I'm planning to compare coefficient of variation (CV) and fold change with normalized value between samples.

Many thanks in advance.

Answer 1

[i.sudbery]

I would recommend VST or rLog from the DESeq2 package. In particular, because you are going to be comparing coefficients of variation, you will want something that detrends the variance.