This is a test version of Biostars. For the public version, visit https://www.biostars.org.
Could the batch effect be a set of numeric values in DEseq2?

Hello!

I am using Deseq2 to analyze a RNA-seq dataset. One of the potential batch effect contains numeric data, which indicates the wasp load in a host. The coldata looks like this:

X,Time,Treatment,batch
sampe1,Treat1,1
sample2,Treat1,7
sample3,Treat1,23
sampe4,Treat2,4
sample5,Treat2,59
sample6,Treat2,10
sample7,Control,0
sample8,Control,0
sample9,Control,0

Does Deseq2 also handle numeric conditions or one must convert the wasp load into different levels of factors? It will be very much appreciated if anyone can help!

Cheers, C

rna-seq deseq2 design

1 answer

DESeq2 and similar programs are perfectly happy with covariates (i.e., numeric values associated with colData). Just note that if you do this with a batch effect then the value associated is expected to have a linear effect...which is probably not the case for actual batches.

Thanks Devon. Your comments are always helpful!

Log in to answer this question.