Dear all,

I'm recalculating the RPKM value of a RNASeq data on Rsubread through featureCounts function, and I'd like to know if should I use just the "assigned" reads or the total reads, including "unassigned ambiguity, multimapping..." (see below), in the RPKM formula. Looking for the answer in forums and in the Mortazavi et al. (2008), I've just find out that " N is the total number of mappable reads in the experiment". So, could anybody please help in this regards?

RPKM = N/(L*T) 

where: 

N: number of reads assigned to a gene

L: lenght of the gene (kb)

T: total mapped reads (Millions)

Thanks in advance! 

You should use the assigned reads only. For your purposes, the library size consists of all the reads that you will be using to infer transcript abundance, not the total number of reads that you generated.