CODEWITHSUNDEEP
r
Stacey John
Stacey John

Reputation: 151

Normalization in R

ID_REF           GSM133971          GSM133972               GSM13397
244901_at   5.95461158752441E+001   1.88836498260498E+001   1.9821
244902_at   7.25400314331055E+001   2.65250205993652E+001   1.9532
244903_at   2.23291015625000E+002   1.78133636474609E+002   5.3816
244904_at   3.65505447387695E+001   3.04016351699829E+000   6.8046
244905_at   4.07523269653320E+001   2.58869285583496E+001   1.1808
244906_at   1.50184860229492E+002   5.39840278625488E+001   4.1491
244907_at   2.65190029144287E+001   1.39808797836304E+000   2.0866
244908_at   3.50156021118164E+000   1.05003893375397E+000   4.5765

In the above dataset the first two columns are MAS5.0 normalized and the last column is RMA normalized. Would it be be possible to denormalize the dataset so that I can perform RMA normaliztion throughout ? The last column was from a different experiment which I needed to append. SO the normaliztions are different.

Upvotes: 0

Views: 403

Answers (1)

csgillespie
csgillespie

Reputation: 60452

The short answer is no. Take a simple normalisation procedure where we have a mean of zero:

#Two data sets 
 c(1, 2, 3)
 c(4, 5, 6)

The normalised data for both data sets is:

 c(-1, 0, 1)

I realise that MAS5.0 is a lot more complicated, but the same concept applies.

Upvotes: 1

Related Questions