I've been going through the various cuffdiff output threads, but I can't find the answer i'm looking for. Here goes - thanks in advance for your help
I ran cufflinks (RABT mode) -> cuffmerge with GTF -> cuffdiff
Now I want to know which isoforms are differentially expressed. As far as I can understand, there are 4 things relevant to this:
isoforms.fpkm_tracking
isoforms_counts_tracking
isoform_exp.diff file
splicing.diff
The tracking files have transcript names while the isoform_exp.diff file doesn't, so I map the transcript names over according to the TCONS (whatever that may be).
I presumed that each line in isoform_exp.diff would correspond to one transcript, but that's not the case. There are multiple lines with the same transcript name, which have different fold change values. So how do I get a single fold change and confidence per transcript? Isn't that was isoform-centric analysis is supposed to produce?
The splicing.diff file has fewer lines, but in it, all the lines say NOTEST (even ones that were OK in the isoform_exp.diff
I ran cufflinks (RABT mode) -> cuffmerge with GTF -> cuffdiff
Now I want to know which isoforms are differentially expressed. As far as I can understand, there are 4 things relevant to this:
isoforms.fpkm_tracking
isoforms_counts_tracking
isoform_exp.diff file
splicing.diff
The tracking files have transcript names while the isoform_exp.diff file doesn't, so I map the transcript names over according to the TCONS (whatever that may be).
I presumed that each line in isoform_exp.diff would correspond to one transcript, but that's not the case. There are multiple lines with the same transcript name, which have different fold change values. So how do I get a single fold change and confidence per transcript? Isn't that was isoform-centric analysis is supposed to produce?
The splicing.diff file has fewer lines, but in it, all the lines say NOTEST (even ones that were OK in the isoform_exp.diff
