[BioC] medpolish Warning

[James W. MacDonald]

Nayeem Quayum wrote:
> Hello everybody,
> I was trying to normalize my data using expresso and found this warning
> message. Can please explain what went wrong?
> Data <- ReadAffy()
> eset <- expresso(Data,bgcorrect.method="mas",
> normalize.method="quantiles",
> pmcorrect.method="mas",
> summary.method="medianpolish")
> The wornging I got :"medpolish() not converged in 10 iterations in medpolish
> (log2(x),..."

You have to know exactly what medpolish() does in order to completely 
understand the error, but basically it means that the summary.method is 
not finishing its job in 10 iterations. I believe you should be able to 
pass a maxiter = 20 as part of your call to expresso(), which should 
help. However, I sort of wonder what your data look like if medpolish() 
isn't converging in 10 iterations - it usually converges in two or three 
iterations for most data I have ever looked at.

If that doesn't work, I think you may be out of luck. You would have to 
know quite a bit about S4 methods to figure out how to hack the code to 
increase the number of iterations. Might be best in that case to either 
change your call to expresso(), or use one of the 'canned' methods (rma, 
gcrma, plier, mas, etc.).

Best,

Jim


> Thanks.
> 
> 	[[alternative HTML version deleted]]
> 
> _______________________________________________
> Bioconductor mailing list
> Bioconductor at stat.math.ethz.ch
> https://stat.ethz.ch/mailman/listinfo/bioconductor


-- 
James W. MacDonald
Affymetrix and cDNA Microarray Core
University of Michigan Cancer Center
1500 E. Medical Center Drive
7410 CCGC
Ann Arbor MI 48109
734-647-5623