[BioC] merged affy 430 A&B

James W. MacDonald jmacdon at med.umich.edu
Wed Apr 6 20:41:55 CEST 2011


Hi Yongde,

On 4/6/2011 12:24 PM, YBao wrote:
> Hi Jim,
>
> Thanks much for sharing your thought. For where I stand, I thought it
> would be easier to handle in the downstream analysis if chip A and
> chip B could merge into one set of metadata. At least for comparing
> that to data from later chip type such as MG430_2.

I suppose, but combining data becomes easier after each step of the 
process. In other words, combining all the data before normalizing and 
computing expression values is really difficult, and I don't see how you 
will gain anything to offset the pain of doing so.

Combining all data just prior to analyzing with limma is really easy, 
but annotating the combined data will be harder because it takes two 
different annotation packages to do so. And you don't gain much by 
combining there anyway.

Combining all the data after running limma and annotating is really 
easy; you can just cat the two files together. Since the data are 
annotated, you can easily compare results from the A/B chips to the 
mg4302 chip (based on overlapping probesets if you like, or genes if you 
want to go that way).

Best,

Jim


>
> Best,
>
> Yongde
>
> On Wed, Apr 6, 2011 at 9:50 AM, James W. MacDonald
> <jmacdon at med.umich.edu>  wrote:
>> Hi Yongde,
>>
>> On 4/5/2011 11:53 AM, YBao wrote:
>>>
>>> Hi all,
>>>
>>> I wonder if someone has done this -- merging the A set and B set of
>>> Affy Mouse chip 430 into one data set so as to facilitate further
>>> analysis such as lima. Some pointers will be appreciated.
>>
>> I don't see much to gain here by combining. There is AFAIK not much overlap
>> between the two (and you certainly don't want to get into the business of
>> combining probesets from different chips). You would gain some small
>> precision in the overall variability estimate used in limma, but otherwise
>> just running separate analyses seems easiest.
>>
>> Best,
>>
>> Jim
>>
>>
>>>
>>> Thanks
>>>
>>> Yongde Bao
>>> University of Virginia
>>>
>>> _______________________________________________
>>> Bioconductor mailing list
>>> Bioconductor at r-project.org
>>> https://stat.ethz.ch/mailman/listinfo/bioconductor
>>> Search the archives:
>>> http://news.gmane.org/gmane.science.biology.informatics.conductor
>>
>> --
>> James W. MacDonald, M.S.
>> Biostatistician
>> Douglas Lab
>> University of Michigan
>> Department of Human Genetics
>> 5912 Buhl
>> 1241 E. Catherine St.
>> Ann Arbor MI 48109-5618
>> 734-615-7826
>> **********************************************************
>> Electronic Mail is not secure, may not be read every day, and should not be
>> used for urgent or sensitive issues
>>

-- 
James W. MacDonald, M.S.
Biostatistician
Douglas Lab
University of Michigan
Department of Human Genetics
5912 Buhl
1241 E. Catherine St.
Ann Arbor MI 48109-5618
734-615-7826
**********************************************************
Electronic Mail is not secure, may not be read every day, and should not be used for urgent or sensitive issues 



More information about the Bioconductor mailing list