[BioC] Error messages from eBayes()(limma)

[Wolfgang Huber]

Hi Juan, James

it does seem strange that this warning should occur for such an 
experimental design. I would advise to spend a bit of time on data 
quality control (the arrayQualityMetrics package is a good start), just 
to make sure that there are no problems such as duplicate files, 
inappropriate preprocessing, etc.

Supposing that your 'intensity' object is a matrix, how does this look like?

   library("vsn")
   meanSdPlot(intensity)

   library("genefilter")
   rt = rowttests(intensity, fac=factor(TS))
   hist(rt$p.value, breaks=50)

(I couldn't test this code on your data object, so adaptations might be 
needed.)

	Best wishes
	Wolfgang


James W. MacDonald scripsit 01/15/2010 03:15 PM:
> Hi Juan,
> 
> Juan Lin wrote:
>> Dear All,
>>
>> 1.       I am working on a microarray dataset. The array platform is
>> GeneChip Mouse Gene 1.0 ST Array. The experiment design is very straight
>> forward: two groups comparison--control vs. treatment, with three 
>> chips in
>> each group.
>>
>> 2.       
>> 3.       Below is my code:
>>
>>> library(limma)
>>
>>> TS<-c("Control","Control","Control","Trt","Trt","Trt")
>>
>>> TS<-factor(TS,levels=c("Control","Trt"))
>>
>>> design<-model.matrix(~0+TS)
>>
>>> colnames(design)<-levels(TS)
>>
>>> fit<-lmFit(intensity,design)
>>
>>> contrast.matrix<-makeContrasts(TrtvsControl=Trt-Control,levels=design)
>>
>>> fit2<-contrasts.fit(fit,contrast.matrix)
>>
>>  
>>
>> Everything was going well before I tried to use the eBayes function. I 
>> got
>> the following error:
>>
>>  
>>
>>> fit2<-eBayes(fit2)
>>
>> Warning message:
>>
>> In ebayes(fit = fit, proportion = proportion, stdev.coef.lim =
>> stdev.coef.lim) :
>>
>>   Estimation of var.prior failed - set to default value
> 
> That's not an error. It's a warning stating that the prior variance 
> estimate couldn't be estimated, so you are using the default value.
> 
> In other words, the function ran, and gave you results, but one of the 
> parameters that is usually estimated from the data was not estimable, so 
> you used a default value.
> 
> In contrast, an error would result in the function halting without 
> returning anything, possibly followed by a big grinding noise, smoke and 
> sparks. Or is that just my computer? ;-D
> 
> Best,
> 
> Jim
> 
> 
>>
>>  
>>
>> Anybody has idea about the reason for the error? I tried looking up the
>> bioconductor archives, but couldn't find any reply that could help 
>> solve my
>> problem.
>>  
>> I am using R.2.10.1.  
>>> sessionInfo()
>> R version 2.10.1 (2009-12-14) i386-pc-mingw32  
>> locale:
>> [1] LC_COLLATE=English_United States.1252 [2] LC_CTYPE=English_United 
>> States.1252   [3] LC_MONETARY=English_United States.1252
>> [4] LC_NUMERIC=C                          [5] LC_TIME=English_United 
>> States.1252     
>> attached base packages:
>> [1] stats     graphics  grDevices utils     datasets  methods   
>> base      
>> other attached packages:
>> [1] limma_3.2.1
>>  
>> loaded via a namespace (and not attached):
>> [1] tools_2.10.1
>>
>>  
>>
>> Many thanks!
>>
>> Juan
>>
>>  
>>
>> -----------------------------
>>
>> Juan Lin
>>
>> Faculty Associate
>>
>> Department of Epidemiology and Population Health
>>
>> Albert Einstein College of Medicine
>>
>>  
>>
>>  
>>
>>
>>     [[alternative HTML version deleted]]
>>
>> _______________________________________________
>> 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
> 


-- 

Best wishes
      Wolfgang


--
Wolfgang Huber
EMBL
http://www.embl.de/research/units/genome_biology/huber/contact