Is it possible to do DE analysis on single tumor-normal pair with EdgeR?

[ymc]

I heard that if you have multiple tumor-normal pairs, you can estimate the BCV. Then you can use this BCV to find DE genes for each individual tumor-normal pair.

Is this true? If so, how to use BCV to find DE genes for individual tumor-normal pairs. I couldn't find that in the EdgeR manual.

Thanks very much in advance!

[dpryan]

Regardless of the software you use, a 1 vs. 1 comparison will always yield close to useless results.

[ymc]

Thanks for your reply.

I have another question. I currently have six pairs of lung tumor-normal pairs RNA-seq data. Suppose I later obtain another six pairs of breast tumor-control RNA-seq data. Can I mix them up for a combined analysis?

Can you tell me why this is not ok? It seems to me even for the "same" lung tumor, there might be heterogeneity among them. Therefore it appears to me "different" cancer can be mixed. Is that correct?

[dpryan]

I suppose you could, if you're looking for commonalities between the tumors. It all depends on the biological question you want to ask.