[BioC] Beadarray expression data

[Simon Luo]

Ina Hoeschele <inah at ...> writes:

> 
> Hi all,
>    I am completely new to the analysis of Illumina Beadarray gene 
> expression data (but quite familiar with Affy data).  Such data (with 
> samples run on 3 chips)  was sent to me by a collaborator. From the 
> vignettes for Bioconductor packages 'beadarray' and 'BeadExplorer', it 
> looks like these accommodate  Beadstudio version 1 and 2 data, while our 
> data was created by BeadStudio version  3.0.14.. Would it still work? My 
> collaborator sent two types of files to me: (1) a number of .csv files 
> (read by excel, 2 for each sample) with columns Illumicode, N, mean GRN, 
> Dev GRN. (2) a summary .txt file with columns TargetID    
> MIN_Signal-1814647013_A    AVG_Signal-1814647013_A    
> MAX_Signal-1814647013_A    NARRAYS-1814647013_A    
> ARRAY_STDEV-1814647013_A    BEAD_STDEV-1814647013_A    
> Avg_NBEADS-1814647013_A    Detection-1814647013_A, with the 
> MIN_Signal-... to Detection-... columns repeated for the remaining samples.
> It looks like beadarray expects to read the summary data from a .csv 
> rather than a .txt file? From the example given on page 2 of the 
> beadarray vignette, the summary data are read in by
> BSData <- readBeadSummaryData(datafile, ..., columns = list(exprs = 
> AVG_Signal", BeadStDev = "BEAD_STDEV", NoBeads = "Avg_NBEADS", ...)
> 
> Since I have AVG_Signal-1814647013_A, AVG_Signal-1814647013_B etc. and 
> not just one AVG_Signal, how do I specify the columns to be read?
> 
> Sorry for these ignorant questions, but this is my first try at these 
> data. Thanks for any advice!
> Ina
> 
> _______________________________________________
> Bioconductor mailing list
> Bioconductor at ...
> https://stat.ethz.ch/mailman/listinfo/bioconductor
> Search the archives:
http://news.gmane.org/gmane.science.biology.informatics.conductor
> 
> 

Hi Ina, 

I used beadarray version 1.2.2 and my bead summary data were from BeadStudio V2.
I don't think the BeadStudio version should not be a concern when extracting the
bead summary data from the raw data file (either .txt or csv file). Your data
are summary data because of the column "AVG_Signal-", readBeadSummaryData()
function reads in all the columns begin with "AVG_Signal-" from the summary data
file and create a BSData object for further preprocessing steps. If you still
concern about the version compatibility, you can use something like
read.table/read.csv to extract the raw summary data (all columns start with
"AVG_Signal-") and compare the result to the BSData object that you obtained
from using readBeadSummaryData() function.

Regarding how to use the readBeadSummaryData() function in beadarray, if you set
everything correct in the target file, in which you need to name the path of
your raw data file and other information for the data set (refer to beadarray
vignette for details), you don't need to pass anything for the parameter
columns. The function knows what to do with the columns in the data file.

Best,

Simon