[BioC] vsn2 dimension of the expression level matrix

[Wolfgang Huber]

Dear Sam

see the man page of vsn2 and in particular the argument 
minDataPointsPerStratum, whose default is 42.  The fitting algorithm of 
vsn2 becomes unstable when the number of data points is too small, and 
the purpose of having this parameter and error message is to get people 
thinking about this potential problem.

33 should be OK, just set minDataPointsPerStratum to a lower value as 
needed, and be sure to verify that the fit looks reasonable.

In the future, please remember to report the output of "sessionInfo()".

Best wishes
	Wolfgang

> Hi,
> 
> I am using vsn2 to normalize an expression. Here are my syntax:
> 
> controls<-vsn2(rawdata[idx.control,])
> genes.vsn2<-predict(controls,newdata=rawdata[-idx.control,])
> 
> I got the following error message:
> 
> Error in .local(x, reference, strata, ...) : 
>   One or more of the strata contain less than 42 elements.
> Please reduce the number of strata so that there is enough in each stratum.
> 
> The controls matrix only has 33 rows.  Does vsn2 require the number of rows for the data matrix to be a minimum of 42?
> 
> 
> Thanks,
> 
> Sam Wong
> 
> _______________________________________________
> 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

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