[BioC] Re: NormalizeForPrintorder

[Gordon Smyth]

At 03:24 AM 18/06/2003, muehleisen wrote:
>Dear Dr. Smyth,
>
>if I use the function "normalizeForPrintorder" like the following > 
>Af.slide.25 <- normalizeForPrintorder (RG.slides.25.28.29.30.flagged, 
>layout=list(ngrid.r=12, ngrid.c=4, nspot.r=21, nspot.c=21), plot=FALSE, 
>method="loess", span=0.4), I get the error message Error in .local(...) : 
>formal argument "R" matched by multiple actual arguments. Even I split the 
>RG list object (containing slides 25, 28, 29 and 30) to a R and a G vector 
>(each containing only slide 25) like the following > Af.slide.25 <- 
>normalizeForPrintorder (Rraw.slide.25, Graw.slide.25, 
>layout=list(ngrid.r=12, ngrid.c=4, nspot.r=21, nspot.c=21)) I got the same 
>error message.

This error arose because I tried to get clever with object orientated 
programming in R, and it doesn't work as I expected. I have revised the 
functions by going back to an older but simpler style and they should all 
work again.

>However, if I use the function "plotPrintorder" like > 
>plotPrintorder(Rraw.slide.25, Graw.slide.25, layout=list(ngrid.r=12, 
>ngrid.c=4, nspot.r=21, nspot.c=21), method="loess", 
>separate.channels=FALSE, span=0.4), I surprisingly get the plot. In 
>general, what span do you recommend for "normalizeForPrintorder" and 
>"normalizeWithinArrays"?

The default values are what I would recommend.

>Would you please send me an example for "normalizeForPrintorder"?

The help document now has an example using data from the SMA package, see 
limma version 1.1.2.

Best wishes
Gordon

>Thank you in advance!!!
>
>Sincerely,
>
>Thomas W. Muehleisen
>
>
>Max Planck Institute for Brain Research
>Deutschordenstrasse 46
>D-60528 Frankfurt/Main