I was essentially doing the same things posted previously.

1: https://support.bioconductor.org/p/70064/

2: https://support.bioconductor.org/p/102005/

Following the note posted on 1, I have successfully downloaded the data of my interest.

library(GEOquery)
data <- getGEO("GSE32894")[[1]]

Unfortunately, I got stuck when I was trying to read "GSE32894" with limma.

idata <- read.ilmn("GSE32894_non-normalized_308UCsamples.txt",probeid = "PROBE_ID",expr="SKBR")

The error shows as follows;

Error in readGenericHeader(fname, columns = expr, sep = sep) : 
  Specified column headings not found in file

I looked into the documentation (https://www.rdocumentation.org/packages/limma/versions/3.28.14/topics/read.ilmn), none of them worked out.

It should be great if you can give me any kind of suggestion to fix this problem.

Thank you.

In your first code chunk you are reading in the wrong GSE (GSE32849), which is a CHiP-Chip experiment, rather than the one you want. In the second case, you have a text file containing something that isn't what you think it is:

sed -n '5p' GSE32894_non-normalized_308UCsamples.txt | sed 's/\t/\n/g' | head
ID_REF
UC_0001_1
UC_0001_1.detection.p.value
UC_0002_1
UC_0002_1.detection.p.value
UC_0003_1
UC_0003_1.detection.p.value
UC_0006_2
UC_0006_2.detection.p.value
UC_0007_1

You are probably better off just using the data you get from getGEO:

 z <- getGEO("GSE32894")[[1]]
> z
ExpressionSet (storageMode: lockedEnvironment)
assayData: 24402 features, 308 samples 
  element names: exprs 
protocolData: none
phenoData
  sampleNames: GSM814052 GSM814053 ... GSM814359 (308 total)
  varLabels: title geo_accession ... tumor_stage:ch1 (55 total)
  varMetadata: labelDescription
featureData
  featureNames: ILMN_1343291 ILMN_1343295 ... ILMN_2415979 (24402
    total)
  fvarLabels: ID nuID ... GB_ACC (30 total)
  fvarMetadata: Column Description labelDescription
experimentData: use 'experimentData(object)'
  pubMedIds: 22553347 
Annotation: GPL6947 
> pData(z)[1:5,54:55]
          tumor_grade:ch1 tumor_stage:ch1
GSM814052              G3              T2
GSM814053              G2              T2
GSM814054              G2              T1
GSM814055              G2              T1
GSM814056              G3             T3b

> table(pData(z)[,54:55])
               tumor_stage:ch1
tumor_grade:ch1 T1 T2 T2a T2b T3 T3b T4a Ta Tx
             G1  0  0   0   0  0   0   0 48  0
             G2 35  9   1   0  1   0   0 56  1
             G3 61 73   0   2  0   5   1 11  1
             G4  1  0   0   0  0   0   0  0  0
             Gx  0  0   0   0  0   1   0  1  0

You can just use limma directly on that ExpressionSet, based on whatever phenotypic groups you care to compare.