Tweet
Dear all,
I am using RNA-seq with Illumina data and I want to assess genes that are differentially expressed in response to my treatment.
My design is the following:
I have two treatment levels: control and heat treatment and for each two biological replicates taken from plants originating from two natural populations.
I want to know which genes are responding in the same way to the heat treatment in both populations. For this I was planning to use DESeq taking data of both populations as biological replicates. My problem, however, is that due to the long separation of the two populations the base expression (expression at control conditions) has already changed for a considerable part of the genes. So I believe using those replicates for the calculation of the variance function (used for the overdispersion) will yield a high biological variance not allowing me any more to detect genes e.g. that increase their expression upon treatment in both populations but already have a changes base expression.
To solve this issue I was wondering if some kind of tests exist that take into account paired controls, if this would be reasonable approach and whether anybody has some experience with this?
Thank you a lot for your help!
I am using RNA-seq with Illumina data and I want to assess genes that are differentially expressed in response to my treatment.
My design is the following:
I have two treatment levels: control and heat treatment and for each two biological replicates taken from plants originating from two natural populations.
I want to know which genes are responding in the same way to the heat treatment in both populations. For this I was planning to use DESeq taking data of both populations as biological replicates. My problem, however, is that due to the long separation of the two populations the base expression (expression at control conditions) has already changed for a considerable part of the genes. So I believe using those replicates for the calculation of the variance function (used for the overdispersion) will yield a high biological variance not allowing me any more to detect genes e.g. that increase their expression upon treatment in both populations but already have a changes base expression.
To solve this issue I was wondering if some kind of tests exist that take into account paired controls, if this would be reasonable approach and whether anybody has some experience with this?
Thank you a lot for your help!