Could some body please guide me about how to calculate differential usage of splice junction between two samples. I have read count information for each junction detected from TopHat2 alignment. The samples are in replicates, so P-value and FDR will be needed to select significant differentially used splice junctions.
Could it be so simple that I normalize the splice junction read count by total mapped read and then further normalize with gene expression (FPKM value) (to rule out the altered usage of junction reads was due altered gene expression. Please correct me if i am wrong.