Notes about normalization
2015-01-13azim58 - Notes about normalization
Some Normalization methods
- LOESS
- LOWESS
- Affymetrix normalization methods (Phil has a NAR paper in which he
o GCRMA (introduced to me by Hojoon)
o MAS5
paper on the topic:
Comparative analysis of microarray normalization procedures: effects on
reverse engineering gene networks
http://bioinformatics.oxfordjournals.org/content/23/13/i282.full
Note from Phil
Unless you have a certain manufacturing bias, like Affymetrix, or you
know there is a predictable dye bias, like Lowess or loess, then it's
usually not recommended to pre-process your data. Normalization methods
can be effective if you have measured the biases, but normalization
surfing usually results in wholescale altering of your data, resulting in
data that now has nothing to do with your observations. I know it can be
difficult to get arrays to work right, but optimizing the assay can be a
better place to start.
https://mail.google.com/mail/u/0/?ui=2&shva=1#inbox/137a8ea831a7d1d6
Combat Normalization
http://www.bu.edu/jlab/wp-assets/ComBat/Abstract.html
MAGE normalization