Hi, I did RNA-seq for 20 samples and I am interested in the differential expression of normal transcripts and novel transcripts within just 2 genes using galaxy. Do I have to map my reads to the whole reference genome and what tools can I use beside cuffdiff. in the end, i want to compare 10 of my normal samples transcripts with each of the 10 mutated samples (each has different mutations) thank you in advanced,
If you only care about 2 genes it would have been vastly cheaper and faster to do qPCR.
Anyway, yes, you must align the data to the entire genome (or transcriptome). You're better off with featureCounts and DEseq2 rather than cuff-anything. Analyse all of the genes but if you only care about 2 genes, then just take the unadjusted p-values for them.