Dear Bioconductor Community,

i would like to ask an "introductory" but important question regarding the appropriate interpretation/explanation of background correction for Illumina Bead Arrays. Recently, i have acquired an Illumina Human HT-12 v4 beadchip microarray dataset. My main question is mostly general but in conjuction to limma normalization function for illumina microarrays:

In detail, as my "sample probe profile" txt file which includes the probe summary file, from which a very small subset for the first sample looks like this:

    PROBE_ID SYMBOL 1A.AVG_Signal 1A.Detection.Pval 1A.BEAD_STDERR 1A.Avg_NBEADS
ILMN_1762337    7A5      113.6922        0.46233770       3.659430            23
ILMN_2055271  A 1BG      162.7333        0.02077922       6.885669            23
ILMN_1736007  A 1BG      114.4497        0.44545450       3.599054            31
ILMN_2383229  A 1CF      122.7193        0.25974030       8.953144            15
ILMN_1806310  A 1CF      134.9052        0.14285710       8.729850            18
ILMN_1779670  A 1CF      130.3925        0.18441560       8.197718            19

Even though i had pre-processed in the past illumina microarrays, my question is the following:

from the above file i can see(and after importing with read.ilmn) that these are raw summarized intensities. However, can i also assume that also are not  background substracted ? In other words, the default output from BeadStudio/GenomeStudio does perform some kind of background correction ? Or im mistaken and only is an option(as i dont see any negative values in the above file )?

Finally, even if is the case , this should not be a consern as neqc includes a default offset ??

Please excuse me for my naive(even "silly") question, but because i have aqcuired the above file with also two more files(phenotype info and controls info) without any other information, this matter troubles me !! 

Any suggestion would be essential !!

There is no problem -- this the sort of data that neqc() expects.

The Illumina proprietary software does two types of background correction.

The first is always done at the image analysis level. For each feature on each array, Illumina subtracts the average intensity of the lowest pixels from that of the remaining pixels. This ensures that the raw intensity reported for each feature is always positive. This behaviour is not documented at the user level and you cannot turn it off.

The second is global background correction whereby the average intensity of negative control probes is subtracted from that of the other probes for each array. This is the option called "background correction" in the GenomeStudio documentation.

Normally neqc() expects to get intensities that are not global background corrected, and that appears to be true of your data. However neqc() should work fine in any case.