[am@i]

hi Simon
Thanks for the last reply !!!
I have another query?
I am getting all the p- values and FDR values as non-significant while performing DE analysis of miRNA's in DEseq. Whereas almost all are significant with chi-square test.What is the reason for such discrepancy? How should I use DEseq to get meaningful result of DE analysis for miRNA?

[Simon Anders]

What do you mean by chi^2 test?

[am@i]

Ok forget about other tests, my problem is DEseq gives all the p- values and FDR values as non-significant while performing DE analysis of miRNA seq from illumina (which should not be the case looking at he count data). How should I use DEseq to get meaningful result of DE analysis for miRNA?

[Simon Anders]

To get a meaningful answer you will need to give us much more details on what exactly you did and why you think that DESeq's result is wrong. Also, please consider using DESeq2.

[Simon Anders]

Just to be clear: The result that all p values are non-significant means that your experiment failed. This conclusion may be wrong, but it is certainly not "meaningless". Quite the contrary...

[am@i]

please find the attached count file of miRNA. many counts seem to be vastly different but still i am getting non-significant DE.
kindly suggest!

[Simon Anders]

I see no replicates. A standard DESeq analysis won't work here. To be honest, it sounds a bit as if you haven't read the vignette yet (as you have made no mention on whether you used blind dispersion estimation or not.)

Please look through the many older threads on the topic on this forum to see why a chi^2 test is incorrect.

[am@i]

I used this command :-
cds2 <- estimateDispersions( cds2, method="blind", fitType="local", sharingMode="fit-only" )
res2 = nbinomTest( cds2, "control", "treatment" )