Is DESeq2 normalisation method appropriate for datasets with multiple cell types?
1
1
Entering edit mode
thkapell ▴ 10
@tkapell-14647
Last seen 13 months ago
Helmholtz Center Munich, Germany

Hi all,

 a theoretical question about DESeq2. I have a dataset of sorted blood immune cell types from healthy and diseased humans. I want to calculate DE genes for all the immune cell types so that I can generate a pan-immune disease signature. My worry is that since the blood immune cell types have different amounts of RNA (e.g. neutrophils vs B cells, etc.), the normalisation method that DESeq2 uses internally might give me erroneous results. Do you think I can still use DESeq2 or should I use a different package/normalisation method?

Best, Theo

deseq2 normalization • 503 views
ADD COMMENT
0
Entering edit mode
@mikelove
Last seen 22 hours ago
United States

The best way to assess this is by looking at an MA plot. Is it plausible to identify the constant or lowly changing genes in the MA plot?

ADD COMMENT
0
Entering edit mode

Thank you very much for the prompt answer. By "identify the constant or lowly changing genes", do you mean to identify these at the MA plot and then test whether these are the same in each of the samples/cell types? Or did I miss your point?

ADD REPLY
0
Entering edit mode

I just mean that you can look at the MA plot to assess if and which genes are being categorized as having LFC near 0. If you have any prior information about genes which you expect should be constitutively expressed you could label those points as well, e.g.

https://bioconductor.org/packages/release/workflows/vignettes/rnaseqGene/inst/doc/rnaseqGene.html#ma-plot

ADD REPLY

Login before adding your answer.

Traffic: 719 users visited in the last hour
Help About
FAQ
Access RSS
API
Stats

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.

Powered by the version 2.3.6