Minfi preprocessNoob vs preprocessQuantile normalization for Illumina EPIC array

Question

[jeni]

Hi,

I would be very grateful if someone could explain me which method is better to use for differencial methylation analysis of 2 groups of samples from the same tissue (the difference is that one group respond and the other don't to a treatment). Would it be better to apply Noob or Quantile normalization? and what is the reason for that?

Thanks in advance.

[igor]

Any reason why you don't think other preprocessing normalization options in minfi (SWAN or Functional normalization) could be even better?

[jeni]

Thanks for your response. Actually I am not sure about what to use; but what I undestood from what I read is that Noob and Quantile are the more appropiate methods for closer groups, while the others are better for globally different groups.