Cuffdiff p- and q-value of zero

[Lien]

Hi all,

I've just performed my first RNA-seq analysis on 2 conditions with 3 replicates each. We followed the Tophat-Cufflinks-Cummerbund protocol as described in the Trapnell paper (Nature protocol 2012).
When I filter my list of differentially expressed genes using the parameters q-value < 0.001 and log2 fold change > plus of minus 2, I end up with a list of 703 differentially expressed genes. However, of these, I find that 324 genes have a p-value and a q-value of zero. Could anyone explain me how you end up with a value of zero?
I didn't include the genes that are not expressed at all in any of the 2 conditions (FPKM of zero), so that is not the cause of a zero q-value.

Any thoughts would be appreciated!
I include some examples below:

gene_id gene_short_name locus sample_1 sample_2 status value_1 value_2 log2_fold_change test_stat p_value q_value significant
XLOC_014955 ACSS1 20:24986872-25039616 C4_5_6 C10_11_12 OK 0.0351045 6.08325 7.43704 -9.95201 0 0 yes
XLOC_015108 BMP7 20:55743808-55841707 C4_5_6 C10_11_12 OK 0.0266919 5.32807 7.64107 -8.59536 0 0 yes
XLOC_023650 STOM 9:124101352-124132545 C4_5_6 C10_11_12 OK 0.0353902 7.89868 7.80212 -8.53728 0 0 yes
XLOC_017798 GUCY1B3 4:156680125-156728783 C4_5_6 C10_11_12 OK 0.0379018 10.7677 8.15023 -10.0701 0 0 yes
XLOC_020293 KCNK5 6:39156746-39197251 C4_5_6 C10_11_12 OK 0.28591 3.36844 3.55845 -11.9519 0 0 yes
XLOC_000544 EPHX4 1:92495532-92529093 C4_5_6 C10_11_12 OK 0.401335 4.86434 3.59937 -8.75222 0 0 yes
XLOC_012801 ZNF702P 19:53471503-53496784 C4_5_6 C10_11_12 OK 5.56173 1.25631 -2.14634 12.7103 0 0 yes
XLOC_012861 ZNF667 19:56950692-56988770 C4_5_6 C10_11_12 OK 3.83043 0.87479 -2.1305 9.37418 0 0 yes
XLOC_024898 RPS4Y1 Y:2709622-2734997 C4_5_6 C10_11_12 OK 356.118 0.337537 -10.0431 16.7393 0 0 yes
XLOC_024921 USP9Y Y:14813159-14972768 C4_5_6 C10_11_12 OK 24.2597 0.0245576 -9.94817 16.6818 0 0 yes

Cheers,
Lien

[jmwhitha]

I have the opposite problem; my p and q-values are 1 and so instead of getting many significant calls, I am getting genes that have a log 2 fold change of over 12 not being called significant. Do you know how to fix this?