I am willing to filter the highly transcribed genes of my experiment based on their
fpkm values. First I need a confirmation whether that is a correct strategy?
If so I am filtering from
genes.fpkm_tracking of the
cuffdiff outputs not the rest (cds, isoforms or tss). I hope that is correct also.
Then I am wondering why several different genes show up with exact
fpkm value in the
genes.fpkm_tracking file? Can some body explain that please? And let's assume initially I select the
500 highest fpkms, then I split those genes and I end up with
1000 genes. Now my aim was to filter the top 500 highly transcribed genes, that 1000 gene result will ruin it so I have to select for less genes for a second time! Is that how it works?
Thanks in advance!