How to contrast gene expression level using like control some HouseKeeping Genes after a DEG using DESEQ2

Question: How to contrast gene expression level using like control some HouseKeeping Genes after a DEG using DESEQ2

4 months ago by

m986 • 0

m986 • 0 wrote:

In my lab we have a question about this (we are not experts in bioinformatics by the way, we still learning):

We are doing rna-seq of resistant vs susceptible peppers. I did a DEG analysis using DESEQ2, I got up and down regulated genes using a FDR:0.01 and LFC:0.5. Now I have my respective list of DE genes, but a questions springed to us: with our information is possible to do a "contrast" of genes expression levels using as "control" some house keeping genes ? Like a RT-PCR.

dge deseq-2 rna-seq • 205 views

ADD COMMENT • link •

modified 4 months ago by Asaf ♦ 7.0k • written 4 months ago by m986 • 0

4 months ago by

Asaf ♦ 7.0k

Israel

Asaf ♦ 7.0k wrote:

Yes, it can be easily done. If you have a list of genes that you assume have the same expression between the two conditions you can use them as controlGenes in the estimateSizeFactors function call. See https://www.rdocumentation.org/packages/DESeq2/versions/1.12.3/topics/estimateSizeFactors:

estimateSizeFactors"(object, type = c("ratio", "iterate"), locfunc = stats::median, geoMeans, **controlGenes**, normMatrix)

ADD COMMENT • link written 4 months ago by Asaf ♦ 7.0k

awesome! Thank you Asaf!! I'll try .

ADD REPLY • link written 4 months ago by m986 • 0

If you have a list of genes that you assume have the same expression between the two conditions

I am really interested in this. Can you do such assumptions on non model organisms (peppers in this case) without an experimental validation on the same samples used for the DEG analysis?

ADD REPLY • link written 4 months ago by andres.firrincieli • 420

You have to make an assumption eventually, either implicit or explicit. The implicit assumption is that most of the genes have the same expression level. Well, if it doesn't hold then the conditions you are comparing might be too distinct. An explicit assumption using a list of genes might be beneficial especially if you have a lot of experience and know what to expect for.

ADD REPLY • link written 4 months ago by Asaf ♦ 7.0k

Thank you for the answer!

ADD REPLY • link written 4 months ago by andres.firrincieli • 420

Similar posts • Search »