Differential Expression for leukemia?

[ymc]

I noticed that when doing differential expression for leukemia, usually researchers take cancerous bone marrow mononuclear cells as tumor and compare it to purified normal bone marrow mononuclear cells from another normal individual.

My question is, are there anywhere I can download a set of RNA-Seq fastqs from normal bone marrow cells?

Thanks a lot in advance!

[ymc]

http://www.ncbi.nlm.nih.gov/pubmed/23722552

How come this leukemia research is able to do tumor-normal pair analysis? How does that work?

[sBeier]

I have no access to the paper, but guessing from the fact that they studied an acute form of leukemia in children, they might have gotten healthy bone marrow from them after they recovered. Or, in some study setting, maybe before they got sick.

[ymc]

I have no access to the paper, but guessing from the fact that they studied an acute form of leukemia in children, they might have gotten healthy bone marrow from them after they recovered. Or, in some study setting, maybe before they got sick.

Thank you for your reply. I find their entry at GEO. It looks like the so-called "normal" is called "remission" there, so I suppose your theory of "bone marrow after they recovered" should be correct.

[ymc]

Another question I have is how to do differential expression analysis if we use a set of RNA-Seq data of normal bone marrow cells as control as described in the first post. I know how to do it with the typical tumor-normal case with EdgeR. But what about this case?

Suppose I have 10 tumors and 10 normals. Should I match each tumor with all 10 normals such that there are total 100 matching comparisions in my EdgeR analysis? Does that make any sense? Or is there a right way to do it?

Thanks a lot!