[BioC] More iterations for median polish using Expresso

Wolfgang Huber huber at ebi.ac.uk
Sat Jun 6 14:17:16 CEST 2009


Hi James

it may also be worthwhile to do some quality assessment (e.g. as 
provided by arrayQualityMetrics) in order to make sure that it is not 
some unexpected properties of the data that cause this behaviour of 
expresso.

Best wishes
      Wolfgang

------------------------------------------------
Wolfgang Huber, EMBL, http://www.ebi.ac.uk/huber


Martin Morgan ha scritto:
> Hi James, James,
> 
> I think the summary.param parameter does this, e.g., going in the other 
> direction...
> 
> example(expresso)
> 
> expresso(Dilution, bgcorrect.method="rma", normalize.method="constant", 
> pmcorrect.method="pmonly", summary.method="medianpolish", 
> summary.param=list(maxiter=1))
> 
> James W. MacDonald wrote:
>> Hi James,
>>
>> jchayes at mcb.umass.edu wrote:
>>> My name is James and I am having trouble normalizing some microarray 
>>> data.  I
>>> have three microarray sets that I am trying to normalize using 
>>> expresso.  Here
>>> is what I type and the error message which follows:
> 
> a 'warning', not an error.
> 
> Martin
> 
>>>
>>> cphy.prenorm <- expresso(cphy.data, bgcorrect.method="rma",
>>> + normalize=FALSE,pmcorrect.method="pmonly",
>>> + summary.method="medianpolish")
>>> background correction: rma
>>> PM/MM correction : pmonly
>>> expression values: medianpolish
>>> background correcting...done.
>>> 11092 ids to be processed
>>> |                    |
>>> |####################|
>>> Warning message:
>>> In medpolish(log2(x), trace.iter = FALSE, ...) :
>>>   medpolish() did not converge in 10 iterations
>>>
>>> Can I increase the number of iterations in expresso or do I have to 
>>> do each step
>>> one at a time?  Any help you can give me would be greatly appreciated.
>>
>> Not easily. If you really want to do this, you can download the 
>> sources and then modify the generateExprVal.method.medianpolish() 
>> function, adding a maxiter=30 or whatever, and then build and install 
>> the package.
>>
>> Best,
>>
>> Jim
>>
>>
>>
>>
>>>
>>> Best Regards,
>>>
>>> James
>>>
>>> _______________________________________________
>>> 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