cuffdiff & significant diffex genes

[secda1]

Hey guys,

I've been using cuffdiff (the latest version) to find differentially expressed genes.
I have 2 conditions and 3 biological replicates for each.
First i was surprised to see how cuffdiff seems to be, with very little diffex genes identified.

Then I looked more in details and saw that a lot of genes were not significant though there fold change was really high (and the test status being "OK"), despite having really similar number of reads per condition .
I am putting an example of 1 gene which is almost not expressed in condition 1 and increase to 13 FPKM in condition 2, and it's still no signicant...

tracking_id condition replicate raw_frags internal_scaled_frags external_scaled_frags FPKM effective_length status
XLOC_035496 plus_Pi 0 1.00024 1.18994 1.31993 0.0618412 - OK
XLOC_035496 plus_Pi 1 0 0 0 0 - OK
XLOC_035496 plus_Pi 2 0 0 0 0 - OK
XLOC_035496 minus_Pi 0 473.807 538.184 489.933 14.5036 - OK
XLOC_035496 minus_Pi 2 676.988 546.757 497.737 14.7058 - OK
XLOC_035496 minus_Pi 1 420.967 436.014 396.923 12.3208 - OK


Any suggestions ? or option i should play around ?

thanks a lot

David