[BioC] 2 color Nimblegen data and limma

[Tobias Straub]

Hi Tom,

are you doing cgh, expression tiling or ChIP-chip?

As for ChIP-chip I am not sure why limma should do a particular good  
or the best job.

At least there are many other t-statistics that perform well and do  
not require your data to be in RGList format. Example: sam ('samr'  
package). if you, however, want moderated-t on a signal matrix, you  
can also use a wrapper to limma (modt.stat function in the 'st'  
package).

for normalization you don't need ndf files. only the raw signals from  
the pairs files are needed. the choice of normalization will depend on  
your application. i personally think that 'vsn' does a good job on  
nimblegen arrays. you also might want to try quantile normalization  
(as provided by the normalize.quantiles function in the 'affy' package).

have a look to http://www.epigenome-noe.net/researchtools/protocol.php?protid=43 
  if you are doing ChIP chip. It's not to advertise my own analysis  
procedure, rather to suggest an alternative to forcing your data into  
data structures you don't really need for a basic analysis.

best
Tobias


On Jul 14, 2009, at 4:18 AM, Thomas Hampton wrote:

> Benilton,
>
> Thanks very much.
>
> Which normalization do you suggest, given that I retrieve the
> NDF and XYS files?
>
> Tom
>
>
> Am I on the right page, and, do you think
> On Jul 13, 2009, at 5:56 PM, Carvalho, Benilton wrote:
>
>> Get the NDF and XYS files from those who generated the pair files.
>> Then use pdInfoBuilder as described in the vignette to create an
>> annotation package. Later, you use the oligo pkg to read in the data
>> and limma as you want. b
>>
>> On 13/07/2009, at 18:45, "Thomas Hampton" <Thomas.H.Hampton at dartmouth.edu
>>> wrote:
>>
>>> From what I can tell, the way to process Nimblegen data is use
>>> limma.  The latter does not read Nimblegen ".pair" files, so
>>> you sort of glue together your own limma-friendly data structures
>>> and then let limma take over.
>>>
>>> Is that about it?
>>>
>>> Thanks,
>>>
>>> Tom
>>>
>>> _______________________________________________
>>> 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
>
> _______________________________________________
> 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

----------------------------------------------------------------------
Tobias Straub   ++4989218075439   Adolf-Butenandt-Institute, München D