I have a methylation file with raw beta values that I am trying to normalize. There are some probes excluded from the data due to not passing QC, and those excluded probes have been denoted as "NA".
When I normalize the beta values with minfi, it gives normalized beta values even for those cpgs that were denoted as "NA" for a particular sample.
It shouldn't be possible for an "NA" value to be normalized to an actual value, unless there is something I am missing? Is it inputting a default value? And if so, how is that determined?