[BioC] logical subscript too long in normalizeRobustSpline

[Gordon K Smyth]

I don't know what the problem is.  If I was debugging this with my data I would:

1. Identify which array in a multi-array experiment is causing the error (by subsetting the RG
object given to normalizeWithinArrays())

2. Check that the data for the array is ok, e.g., using summary(RG$R) etc

3. Try debug(normalizeRobustSpline), then re-run the normalization and step through the execution
normlizeRobustSpline() line by line (by typing ENTER) to see where the error occurs.

Gordon

> Tue Nov 2 23:14:50 CET 2004
> Hi again,
>
> Getting this error using robustspline normalization in limma:
> Error in normalizeRobustSpline(object$M[, j], object$A[, j], layout, df = df,  :
>         (subscript) logical subscript too long
>
> I'm on:
> Win2k SP2
> R 2.0
> limma 1.8.8
>
> This isn't a general problem with robustspline, as I seem to be able to get it
> to work on other datasets.  Any suggestions on how to debug?
>
> Paul