[BioC] GSE normalization

Stephen Taylor staylor at molbiol.ox.ac.uk
Fri Jun 22 18:19:10 CEST 2007


> I posted a similiar question as you yesterday and my situation is a
> little bit different from you:
> I have two ideas and one of them is to evaluae the variance across
> sites so normalization should not be done for that purpose; but my
> second idea needs the normalization, and my way of doing that is pool
> the CEL files in one directory and run justGCRMA or whatever on them.
> I am working on this approach now. Maybe we (or anyone interested in
> this) could share the experience.

Someone has just advised me (off list) that the SOFT files can't be 
analysed using rma since the data is at the gene level. So I guess 
downloading the CEL files is the best approach...

Ultimately I would like to incorporate other platform(s) as well, though 
I realise this has many problems. Another set of data I have is from 
Functional ID v2.0 Array 1 from Rosetta, which I am not familiar with. 
Does anyone know what is the best normalisation procedure for this type 
of data?

I agree with Weiwei though it would be useful to find out what is an 
appropriate methodology. I have had a brief look around and seen a 
bioconductor package called MergeMaid. Has anyone had any experience 
with this or can recommend something else?

Thanks again,


> HTH,
> Weiwei
> On 6/22/07, Steve Taylor <staylor at molbiol.ox.ac.uk> wrote:
>> Hi,
>> I have created 3 separate GSE Objects, read from SOFT format files 
>> using GeoQuery. I would like to normalise across experiments using 
>> rma. What's the best way to merge the data and then normalise?
>> Ultimately I want to analyse these using limma.
>> Thanks for any help,
>> Steve
>> _______________________________________________
>> Bioconductor mailing list
>> Bioconductor at stat.math.ethz.ch
>> https://stat.ethz.ch/mailman/listinfo/bioconductor
>> Search the archives: 
>> http://news.gmane.org/gmane.science.biology.informatics.conductor

More information about the Bioconductor mailing list