[bgibb]

Regarding cuffcompare output files, some of them appear to be deposited in the path from which you call cuffcompare (even if specifying <outprefix> using the -o option). The affected files include:
cuffcmp.stats
cuffcmp.loci
cuffcmp.combined.gtf
cuffcmp.tracking
This could lead to some confusion if you happen to be in datapath1 when you call cuffcompare on datapath2. A simple work around is to add the appropriate cd command to your scripts just prior to calling cuffcompare.

Using Cufflinks 1.0.3

[ecofriendly]

Quick question about Cuffdiff: Can it handle a time course in which each time point is compared back to the same reference point?

I am doing a complicated time series of various treatments in various genetic backgrounds, but thus far have not relied upon the [-T] option in Cuffdiff, because of this issue. (When I tried -T back in July, it compared each additional time point to the previous time point, not the time 0 untreated control). Any info/ suggestion would be appreciated.

Thanks.
Elena