RNA-Seq analysis edgeR

Question

[eb0906]

We have a time course infection experiment with 12 samples (2 technical replicates and 2 biological replicates per sample). The samples were sequenced on a HiSeq over 2 lanes, 1 x 50 bp reads.

1. 24hr_infection1_lane3
2. 24hr_infection1_ lane4
3. 24hr_infection2_lane3
4. 24hr_infection2_lane4
5. 24hr_control1_lane3
6. 24hr_control1_lane4
7. 24hr_control2_lane3
8. 24hr_control2_lane4
9. 48hr_infection1_lane3
10. 48hr_infection1_lane4
11. 48hr_infection2_lane3
12. 48hr_infection2_lane4

I aligned the samples using Tophat and then converted the .bam files to .sam files and created a counts file using HTSeq for each sample. My questions now are:

1. Should the technical replicates be collapsed (add gene counts for technical replicates)?
2. How should the design matrix be setup in edgeR?

Answer 1

[Devon Ryan]

1. Yes, add the technical replicates together.
2. ~time+treatment, where time is a factor of 24hr and 48hr and treatment is control and infection. You obviously can't distinguish between an effect of time and an interaction between time and treatment.

[eb0906]

Thank you for your help!