[BioC] loess plot

[Sean Davis]

On Fri, May 30, 2008 at 10:45 AM,  <shetumi at aim.com> wrote:
> I was trying to plot some data in R. I used the following code to draw a loess fit and got the output as
>
>>?lines(lowess(log(abs(t(res))), log(abs(t(synthesised)))), col="red")
> Error in lowess(log(abs(t(res))), log(abs(t(synthesised)))) :?? NA/NaN/Inf in foreign function call (arg 1)
> Then I thought to use your Limma package for background correction. Do you think it's a right choice ?
>
> However, I installed Limma using
>> source("http://www.bioconductor.org/biocLite.R")
>> biocLite("limma")
>> biocLite("statmod")
>> library(limma)
>
> and then used the commands which generated the error as follows --
>
>> system.time(fit <- loessFit( log(abs(t(res))), log(abs(t(synthesised)))))
> Error in loessFit(log(abs(t(res))), log(abs(t(synthesised)))) :
> ? binary operation on non-conformable arrays
> Timing stopped at: 0.57 0.06 0.64 NA NA
>
>> system.time(fit <- plotPrintTipLoess( log(abs(t(res))), log(abs(t(synthesised)))))
> Error in 1:layout$ngrid.c : NA/NaN argument
> In addition: Warning messages:
> 1: In object$printer :
> ? $ operator is invalid for atomic vectors, returning NULL
> 2: In object$M : $ operator is invalid for atomic vectors, returning NULL
> 3: In object$A : $ operator is invalid for atomic vectors, returning NULL
> 4: In object$M : $ operator is invalid for atomic vectors, returning NULL
> 5: In object$A : $ operator is invalid for atomic vectors, returning NULL
> 6: In layout$ngrid.c :
> ? $ operator is invalid for atomic vectors, returning NULL
> Timing stopped at: 0.66 0.01 0.67 NA NA
>
>
> Can someone please?tell me where the problem is and?what could be the right?command to use ?

You'll need to read the help for plotPrintTipLoess and loessFit.  For
the loessFit part, you will want to make sure that the matrices are
the right shape and size.  You didn't give a reproducible example, so
it is impossible to tell what exactly is wrong.  Finally, be sure to
include sessionInfo() in emails to the list so that people have at
least minimal information about your R setup.

Sean