Merge two runs of RNA-seq data

[diego diaz]

Hi all

I have two runs of RNA-seq data from the same experiment, 4 WT and 4 KO mice. The second run was made because the first one yielded very few reads. Strictly speaking, the second run is a technical replicate of the first.

My question is this,

Can i merge this technical replicates to get more coverage in the DE analysis?

if i merge this replicates, am i introducing some biases in the proportions of the transcripts?

Thanks in advance!!

[crazyhottommy]

I also want to know some insights in this issue.

[crazyhottommy]

I merged the sequences reads for ChIP-seq, and it is fine. I am not sure with RNA-seq

I also want to know some insights in this issue.

[mikep]

There are some differential engines which can distinguish between technical and biological replicates, or you could merge them. However, the larger issue is "the first one yielded very few reads"

Do you know why this is? If there was some kind of technical error it could be an issue. Also, how few is few? Is merging even worth it?

You could map them separately and look at the distributions, see if they are different.

Finally, the plural of mouse is mice, not mouses.

[diego diaz]

Thanks for your answers

Finally, I mapped each group of technical replicates separately and I made some PCA plots, heatmaps, etc, then I discarded samples that looked weird.