Hi

So I am doing some clustering of array data and I want to remove variation associated with variables of non interest (gender, race, and age) prior to. I have used combat in the past to remove categorical variables, but I am wondering how to remove variation associated with continuous variables.

I have found the limma function:

x <- removeBatchEffect(data, covariate)  

Is it OK to remove the variation in my dataset using the previous line of code and the age, gender, and race variables? Note we do not have 'regular treatment' groups to add in the design matrix for the function in all cases.

Thanks,

Chris