Hi I have been experimenting with Bowtie and Cufflinks and I now reach a point where it appears to work; but I think there is a subtle error.
I don't find ENOUGH gene hits.
So (1) I got the same dataset professionally analysed by GATC and they returned 20k genes, but when I do it I get 2k. (2) When I inspected the tabular output from Cufflinks, all my 2k hits are in the first 4000 genes in the list. The Tabular file displays 33000 rows, and all hits in the first 4000 rows? That's implausible. Putting two and two together it seems like I am only really properly mapping the first bit of the files? Any ideas anyone?
I think it is a cufflinks thing, because even when I used the BAM from the GATC service, I still only got 10% of the hits they returned. I guess (a) Is there a way that cufflinks would trip up mid way through a run and NOT return any errors? (b) Are there, like, some essential options I should have selected in my cufflinks run. I have tried a couple of different options, but the manual to me is not really idiot proof enough! Thanks Everyone Richard