apply edgeR TMM normalization without converting data format into 'DGEList'
2
0
Entering edit mode
wt215 • 0
@wt215-12405
Last seen 19 months ago
United Kingdom

Hi, I would like to apply edgeR TMM normalization without converting data format into "DGEList", in other words I would like to manipulate the raw count table. (By doing so I can get a feeling about the whole RNAseq analysis procedure).

May I ask that if I just do the following:

(Raw count table)*(norm.factor), where 
norm.factor=edgeR::calcNormFactors((Raw count table),method='TMM').

And then conduct downstream analysis. Does this procedure make sense? After downstream analysis, can I say that I have applied TMM method for normalisation?

If the above procedure does not make sense, how about the following procedure:

(Raw count table)/((norm.factor)*colSums((Raw count table))), where
colSums((Raw count table)) stands for library size (each column represents a sample or a cell)

Thank you very much!

edgeR tmm normalised values • 2.1k views
ADD COMMENT
5
Entering edit mode
Aaron Lun ★ 28k
@alun
Last seen 7 minutes ago
The city by the bay

Your second approach is closer to what edgeR does internally. Normalization is performed by conceptually dividing the counts by the effective library size. The effective library size is, in turn, defined as the product of the normalization factor and the library size for each sample, to account for composition biases.

In practice, normalization is achieved by scaling up the mean by the effective library size, rather than scaling down the observed count. This avoids distorting the mean-variance relationship from modifying the counts directly. I would not use the "normalized counts" for the DE analysis, only for stuff like clustering.

ADD COMMENT
0
Entering edit mode

Thank you very much! That's very helpful!

ADD REPLY
1
Entering edit mode
@gordon-smyth
Last seen 58 minutes ago
WEHI, Melbourne, Australia

After downstream analysis, can I say that I have applied TMM method for normalisation?

Well, it depends on what downstream analysis you do, which you haven't told us anything about. How can we comment on things you don't tell us?

ADD COMMENT
0
Entering edit mode

Sorry, I thought the downstream analysis involves DE analysis and clustering or any statistical methods. I thought that once we obtained a normalized count table, we can apply any statistical methods on it. 

ADD REPLY

Login before adding your answer.

Traffic: 528 users visited in the last hour
Help About
FAQ
Access RSS
API
Stats

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.

Powered by the version 2.3.6