I am trying to integrate over 15 datasets into one RGChannelSet in order to normalise it using the package funtoonorm. I used GEOquery to get the datasets from GEO. However, since minfi only reads IDAT files, I could not get an RGset from GEO datasets that do not have IDAT files. Hence, I had to create a methylumi object from the series matrix text file from GEO. However, funtoonorm only works with minfi object. Would any of you know if there is a way of converting methylumi object into RGset?
Thank you,
Poojitha
Thank you James:) I created a methylumiset object from series matrix in GEO and tried using the coercing function from methylumi vignette, but got the following error message:
> lumiMethy <- methylumiR("GSE50192_GPL13534_Matrix_raw_data.txt", sep="\t")
> rgSet450k <- as(mset450k, 'RGChannelSet')
Error in methylumiToMinfi(lumiMethy) :
Cannot construct an RGChannelSet without full (OOB) intensities
Would you know how to resolve this? Is there a way of retaining OOB intensities while creating the methylumi object?
It depends on what data the submitter sent to GEO. In the case of this particular GSE, it appears they just sent the processed beta values, and you can't convert those data to an
RGChannelSet
. AnRGChannelSet
implies that you have the original Red and Green channel data, and you can't get those values from the betas alone.