Hi Yamol, this is not an answer. Please make this a comment so that you can get clarification and the post can be updated. That will help people answer the question.
Hi all,
I am quite sure that column normalization can be used to make all chips vary in the same range (correct me if I am wrong). Anyway I cannot understand row wise normalization (probe wise). Why is it used and for which purpose?
Thanks
2 answers
The main example I can think of is for visualization: when creating heatmaps, most software will give you an option to "standardize" your data (set mean to 0, sd =1 for each gene / probe / isoform). This way, genes in the up-regulated group are generally colored red (or whatever color you choose), regardless of absolute intensity / expression.
In general, I would say you are right: most "normalization" steps focus on making expression values between chips / samples most comparable.
Which paper shows the need?
Log in to answer this question.