[BioC] beadarray: problem with reading in Bead Summary Data

[P.D. Moerland]

Dear Mark and other beadarray users,

I have some problems when trying to read in 24 samples of BeadChip data
using the readBeadSummaryData function (beadarray v1.2.2, see below for
complete sessionInfo). Everything works fine if I have only file
containing the raw bead summary values for 24 arrays. However, when the
bead summary values are in separate files for each array, calling
readBeadSummaryData fails with:

Error in names(a$exprs) : object "a" not found

> traceback()
3: names(a$exprs)
2: gsub("AVG_Signal.(.+)", "\\1", names(a$exprs))
1: readBeadSummaryData(targets, skip = 0)

Looking into the code of readBeadSummary, variable "a" is indeed never
initialized if readAcross = FALSE (as is the case here). 

Similarly, if the bead summary values are in three separate files for
each group of eight arrays (columnwise from a Human_RefSeq-8 chip)
readBeadSummaryData fails. This time variable "a" in readBeadSummary is
defined since readAcross = TRUE but variable "a" is assigned to only
once, namely for the first batch of eight arrays. The final
BeadSummaryList object that is returned by readBeadSummaryData,
therefore contains only information from the first eight arrays and
ignores the remaining 16 arrays.

Does anybody have any suggestions how to correct this?

Best,
Perry

PS: Mark, I can make scripts and data available off the BioC discussion
list if you want to reproduce the problem

> sessionInfo()
R version 2.4.0 (2006-10-03)
i386-pc-solaris2.10

locale:
C

attached base packages:
[1] "tools"     "methods"   "stats"     "graphics"  "grDevices" "utils"
[7] "datasets"  "base"

other attached packages:
beadarray      affy    affyio   Biobase     limma
  "1.2.2"  "1.12.1"   "1.2.0"  "1.12.2"   "2.9.1"


---
Perry Moerland, PhD
Room J1B-206, Bioinformatics Laboratory
Department of Clinical Epidemiology, Biostatistics and Bioinformatics
Academic Medical Centre, University of Amsterdam 
Postbus 22660, 1100 DD Amsterdam, The Netherlands