[BioC] SVA: Removing batch effects without regards to differential expression

[Mete Civelek]

Hi All,
I have microarray data from more than 1,000 people. I am interested in removing some "batch effects/adjustment variables and then correlating the expression of the genes with multiple phenotypes measured in these people. The phenotypes are continuous variables such as body mass index and cholesterol levels. The batch effects are both categorical, such as microarray plate, smoking status, or continuous, such as amount of exercise. When I looked at the SVA package, it seems to me that there is a requirement for inputting the variable of interest in the model matrix before adjusting for the batch effects. However, in my case I have multiple phenotypes that are variable of interest. Is there a way or another approach to remove the adjustment variables regardless of the phenotypes (variables of interest)?
Thanks in advance for your help
Mete

________________________________

IMPORTANT WARNING: This email (and any attachments) is o...{{dropped:9}}