Entering edit mode
Hi!
I hope someone can help me with my problem.
My project aims to test the hypothesis that differentially expressed genes between germline tissues (ovaries and testes) may be at the origin of hybrid sterility that in most cases is only present in the male offspring. However, I don't know the best way to normalize different types of tissues (Do you know if I can use DESeq2 or do I need to use another one?)
If somebody can guide me in this matter I would be very grateful!
We have performed RNA total sequencing in ovaries and testes of fruit flies, in two biological replicates. I have aligned the transcripts into the reference genome and now I need to perform the differential expression analyses between these tissues. But, the problem is the high number of male-biased genes in testes. So, I would like to know if DESeq2 is a good tool to perform this kind of analyses or if I should use another method to analyzed these data.
Thanks.
If you want to specify a set of control genes to use for normalization you can use
controlGenesinestimateSizeFactors.e.g.
See
?estimateSizeFactorsfor more details.Hi. Sorry for delay, thank you for your answer.
I performed DESEQ2 analysis using the controlgenes argument, and for this I've selected 6237 genes with the same expression between the tissues. However, I've found these genes differentially expressed in the results. I thought that controlgenes shouldn't be detected as differentially expressed. Did I get it wrong or used the command in a wrong way?
Follow the command as I used:
library(DESeq2) dataset <- DESeqDataSetFromMatrix(countData = newbigtableedited, colData = colData, design = ~treatment) dataset datasetest = estimateSizeFactors(dataset,controlGenes=1:6237) dds = DESeq(dataset_est) head(dds)
Thanks
Think about it this way:
controlGenesis equal to all the genes by default (which is how most people typically run DE analyses). And in those cases, people still find DE.