Averaging with technical and biological replicates

Question

[Duckula]

Dear all,

I have a data-set consisting of biological and technical replicates and I am interested to use the mean expression of them by initially having the mean of the technical replicates and then the biological replicates before analyzing my data-set using DESEQ2. I wonder if any of you have any suggestion that how can I achieve it within R. I am already aware that there is a collapseReplicates in DESEQ2 but it only deals with technical replicates.

The technical and biological replicated would look like something like below

Day-1_1
Day-1_1
Day-1_2
Day-1_2

[Nicolas Rosewick]

Why would you merge your biological replicates ? Good practive is to merge technical replicates only.

Answer 1

[Benn]

It is not wise to calculate means of technical or biological replicates before doing statistics (the statistical power is in the biological replicates). My advice is to use limma (voom), and use duplicateCorrelation function for the technical replicates.