RNAseq and Differential Expression

[Marianna85]

Dear all,
I have to find the best way to measure the differential expression in samples subjected to RNA sequencing.
I briefly summary the experiment in order to help you in understanding the question:
I have the RNAseq data of 6 developmental stages, there are no biological replicates but each library consists of a pool of several animals. (I know that it would be better to have replicates, but that's the way!)
I'd like to find differential expressed genes between two stages.
I know that, without replicates, I can only use NOIseq and DEseq. And which normalization method should I adopt?

Which software should be better?
Anyone has a similar experimental plan?
Suggestions are well accepted.

Many thanks in advance to everyone will help me!

Marianna

[mbblack]

(I know that it would be better to have replicates, but that's the way!)
Marianna

Actually, not "better" to have replicates, but essential to have them. Seriously, without replicates, all you can do is rank the pairs by simple difference. You have no power to discriminate significance between those differences, and any p-values or FDR values you compute will essentially be meaningless.

Without replicates, you have no way to account for biological variation within your groups, so you have no way of assessing how significant any observed change between them is. There is no algorithmic magical cure for that.

In your case, the best you can do is look at raw differences, pick the largest ones and confirm them by some other experimental method (qPCR for example).