Question: Normalization in edgeR
0
gravatar for myprogramming2016
3.6 years ago by
myprogramming20160 wrote:

Hi,

I am not quite clear about the normalization in edgeR. I am not seeing any change in the actual read counts.

They are same numbers except filtered for >10 reads, library size is reduced and norm.factor is 1.

Could you please comments on this?

Codes:

x<-read.delim("counts.txt",header=T,sep="\t")

y<-DGEList(counts=x[,2:151])
y <- calcNormFactors(y)
str(y)
keep<-rowSums(cpm(y)>10)>=3  
y<-y[keep,,keep.lib.sizes=FALSE]
str(y)
head(y$counts)
y$counts
# recaclulate the library size
y<-DGEList(counts=y)
y$samples  

rnaseq edger bioconductor • 1.1k views
ADD COMMENTlink modified 3.6 years ago by Gordon Smyth38k • written 3.6 years ago by myprogramming20160
Answer: Normalization in edgeR
2
gravatar for Gordon Smyth
3.6 years ago by
Gordon Smyth38k
Walter and Eliza Hall Institute of Medical Research, Melbourne, Australia
Gordon Smyth38k wrote:

Well, you reset all the normalization factors back to 1 when you ran DGEList() again at the end. Why did you do that? The library sizes had been recalculated already.

Have you read the section on normalization in the edgeR User's Guide? Amongst other things, that tells you why the read counts themselves are never changed.

Minor points:

I think it is usually better to run calcNormFactors() after filtering rather than before.

There is no need for str(y). Just type show(y) or just y by itself. It's far more informative.

ADD COMMENTlink modified 3.5 years ago • written 3.6 years ago by Gordon Smyth38k

Thanks for your help. I have re-written the codes. 

I would like to subset the read counts without converting them into CPM. I mean, subsetting for raw read counts. I am planning to use median method of normalization over TMM. 

I want to subset for >10 counts and it should be present in atleast 3 libraries.

Could you please suggest a code?

Thanks

 

 

ADD REPLYlink written 3.5 years ago by myprogramming20160
Please log in to add an answer.

Help
Access

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
Powered by Biostar version 16.09
Traffic: 178 users visited in the last hour