Starting from normalized values in RNAseq

[Akis]

Hello,

I have received a matrix with normalized values from an RNA sequencing experiment; the normalization was done as following: The raw counts were converted to cpm and then to RLE by using edgeR. I have no other information, only the matrix with these values.

How would you continue in order to do a DE analysis?
Would it make sense to try to convert them to raw counts?

Some help would be appreciated!

[neavemj]

Both edgeR and DESeq require un-normalized raw counts for differential expression analysis. I suppose the easiest thing would be to email the person who analysed the data originally and ask for raw counts or the reads?

If you don't even have access to the sequence reads then it may not be worth going any further as these should be provided along with any paper that you may write..