[BioC] nsFilter() errors on ExpressionSetIllumina data object

Elliot Kleiman kleiman at rohan.sdsu.edu
Mon May 18 11:33:00 CEST 2009

Hi BioC,

When calling function nsFilter() from package 'genefilter' on an
"ExpressionSetIllumina"data object from package, 'beadarray',
I get the following errors, and I am not sure why?:

 > nsFilter(eSet.illumina)
Error in function (classes, fdef, mtable)  :
  unable to find an inherited method for function "nsFilter", for 
signature "ExpressionSetIllumina"

RE: I am trying to filter-out low varying probes from my expression data 
to make my
"gene universe" a minimal set prior to performing hypergeometric testing 
of GO terms.

Here is some object info regarding the session data object I am working 
with, 'eSet.illumina':
(Note: I am using function readBeadSummaryData() from package 
'beadarray' to read-in
my raw bead summary data file, 'sample_probe_profile.txt'; details of 
session object
construction shown at bottom)

# print object class
 > class(eSet.illumina)
[1] "ExpressionSetIllumina"
[1] "beadarray"

# print object dimensions
 > dim(eSet.illumina)
Features  Samples
   22517       48

# subset object
 > eSet.illumina[1:2,2:5]
Error: subscript out of bounds

..., I found that creating an ExpressionSet object from scratch,
somehow fixes the subsetting error,

 > exprs.illumina <- exprs(eSet.illumina)
 > minimalSet <- new("ExpressionSet", exprs=exprs.illumina, 
 > minimalSet[1:2,2:5]
ExpressionSet (storageMode: lockedEnvironment)
assayData: 2 features, 4 samples
  element names: exprs
  sampleNames: 1677718209_B, 1677718209_C, 1677718209_D, 1677718209_E
  varLabels and varMetadata description: none
  featureNames: 4560377, 5860059
  fvarLabels and fvarMetadata description: none
experimentData: use 'experimentData(object)'
Annotation: illuminaRatv1BeadID.db

..., but when I call nsFilter() on my new minimal ExpressionSet object, 
I get
a different error:

 > nsFilter(minimalSet)
Error in .checkKeys(value, Lkeys(x), x at ifnotfound) :
  value for "4010324" not found

You can replicate the nsFilter() .checkKeys() error by using the data from,
and using the annotation package, illuminaHumanv1BeadID.db

And finally, I constructed my local session ExpressionSetIllumina data 
object, 'eSet.illumina'
by doing,

# library setup
 > library(beadarray)
 > library(illuminaRatv1BeadID.db)
 > library(genefilter)

# ---
# show BeadStudio version for file, 'sample_probe_profile.txt'
# ---
# Illumina Inc. BeadStudio version 2.3.47
# Normalization = none
# Array Content = RatRef_12_v1_11222119_A.xml.xml
# Error Model = none

# read-in bead summary data file using function
# readBeadSummaryData() from package 'beadarray':
 > dataFile <- "./sample_probe_profile.txt"
 > eSet.illumina <- readBeadSummaryData( dataFile,
+                                skip = 7,
+                                columns = list( exprs = "AVG_Signal",
+                                                se.exprs = "BEAD_STDEV",
+                                                NoBeads = "Avg_NBEADS",
+                                                Detection = "Detection"
+                                              ),
+                                annoPkg = "illuminaRatv1BeadID.db"
+                               )

* What am I missing here?

Thank you,

Elliot Kleiman

# print SessionInfo()
 > sessionInfo()
R version 2.9.0 (2009-04-17)


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

other attached packages:
[1] genefilter_1.24.0            illuminaRatv1BeadID.db_1.2.0
[3] RSQLite_0.7-1                DBI_0.2-4
[5] AnnotationDbi_1.6.0          beadarray_1.12.0
[7] sma_0.5.15                   hwriter_1.1
[9] Biobase_2.4.1

loaded via a namespace (and not attached):
[1] annotate_1.22.0 limma_2.18.0    splines_2.9.0   survival_2.35-4
[5] tools_2.9.0     xtable_1.5-5

MS graduate student 
Computational Science Program 
San Diego State University

More information about the Bioconductor mailing list