Is there a program to average read counts from technical replicates?

Question

[Tota]

I have cell lines where RNA was extracted and was split across two lanes giving me two technical replicates per sample. I have my read counts for each technical replicate and I am going to use edgeR to get a DEG list. Is there a program which will average the reads from the technical replicates? Thank you.

Answer 1

[lieven.sterck]

so it's the same biological sample but just run on two different lanes (where the combinations is your requested #reads) ? If so, simply concatenate the two technical repeats into a single file and work with that one.

[Tota]

Yes same biological sample, wouldn't averaging be more appropriate? Concatenating the two technical seems to be a lot easier.

[Friederike]

You don't have to do anything manually, if you tell edgeR which samples represent the same underlying biological sample in the design matrix, you should be set.

[Tota]

Thank you. I will try that!

[lieven.sterck]

on top of what Friederike said, no, no need for averaging. That sample could just have well be run in a single lane, summing both will be enough (especially when each lane only contributes 1/2 of the amount of sequence you ordered)

[Tota]

That makes sense, thank you. I will sum the reads in that case.

Answer 2

[Gordon Smyth]

edgeR povides the function sumTechReps() for this purpose.