Hi everyone,

I have some doubts about the removeBatchEffect function from the Limma package.
I am working on association of expression data (RNAseq) with a continuous phenotype variation (instead of a case control experiment), and I am using this function to remove the batch effect.
Can I use the values for my continuous phenotype in the design matrix?

The model I used so far was:

design= model.matrix(~ as.numeric(phenotype))

mRNA.nobacth = removeBatchEffect(Expr.mRNA, batch = batch1, batch2 = batch2 covariates = cov,
                                 design= design)


Do you think this would introduce any bias to my analysis?
 

What do you mean by 'introduce any bias to my analysis'? If you are using removeBatchEffect prior to fitting a linear model, then yes, as noted in the help for that function:

Note:

     This function is not intended to be used prior to linear
     modelling. For linear modelling, it is better to include the batch
     factors in the linear model.