RMA normalization for bgx file

Question

[sahar.taheri91]

how I can normalize this dataset GSE14078 which has a .bgx format for the raw data in R. I would appreciate if you could help me from scratch.

Answer 1

[Kevin Blighe]

The data must be an Illumina BeadArray. The BGX file does not contain expression data - it is merely a gzipped annotation text file that has a BGX extension. Please search the GEO record for the actual expression data. If the authors uploaded the raw data files, then these will have the 'idat' extension; otherwise, it may have been uploaded as an Excel, TSV, or CSV file of expression values.

If you can obtain the idat files and the BGX file, then you can read these into R via limma:

project <- limma::read.idat(
  idatfiles = my_idat_files, # character vector of file paths
  bgxfile = bgxfile,
  dateinfo = TRUE,
  annotation = c('Source','Symbol','Transcript','ILMN_Gene','RefSeq_ID',
    'Entrez_Gene_ID','Symbol','Protein_Product','Probe_Id','Probe_Type',
    'Probe_Start','Chromosome','Probe_Chr_Orientation','Probe_Coordinates',
    'Cytoband', 'Definition', 'Ontology_Component', 'Ontology_Process',
    'Ontology_Function', 'Synonyms'))

----------------------------------

You can also download the expression data directly via:

library(Biobase)
library(GEOquery)

# load series and platform data from GEO
gset <- getGEO("GSE14078", GSEMatrix =TRUE, getGPL=FALSE)
if (length(gset) > 1) idx <- grep("GPL6104", attr(gset, "names")) else idx <- 1
gset <- gset[[idx]]

Edit: I checked the GEO record and, worryingly, the authors for the study state the following (key point in bold):

The data were analyzed with Illumina Bead Studio. Raw data was used without further normalization due to the high inherent variability in sperm RNA contributions.

Kevin