[BioC] Lumi package: control data for background correction

[merja matilainen]

Hi!

I have Illumina data for analysis that is produced with BeadStudio v3. It is not background corrected and also not normalized. I wanted to try the lumi package and I found a previous posting to the mailing list describing how to read in the control data from a separate file http://article.gmane.org/gmane.science.biology.informatics.conductor/15391/match=gordon+background+correction+lumi

I did the same (as far as I can tell)
> rawX=lumiR("raw_data.txt")
> controlgp=lumiR("qcinfo.txt")
> rawX at controlData=as.data.frame(exprs(controlgp))
> rawXbgc=lumiB(rawX,method="bgAdjust")

Here is what I have in rawX
> exprs(rawX[1:5,1:8])
        1881436004_A 1881436004_B 1881436004_C 1881436009_A 1881436009_B 1881436009_C 1881436049_A 1881436049_B
20605      270.30210    252.15860    271.25070    277.39090    249.06970    221.26020     223.3545    231.69390
3450747    483.33750    423.05290    460.44330    547.50350    462.51610    439.66790     408.1301    433.04090
3060450    620.23050    621.86890    626.29790    703.47220    562.79860    483.12060     523.9857    571.69780
870131      78.45721     71.16602     67.74918     75.95979     75.14289     74.56358      80.1152     74.99714
5310368     90.98277     79.14721     82.17856     85.82687     89.79446     81.63494      80.4482     74.59328


Here is part of my @controlData
> rawX at controlData
                    1881436004_A 1881436004_B 1881436004_C 1881436004_D 1881436004_E 1881436004_F 1881436009_A
BIOTIN                6683.65000   6280.45700   5796.20100   6058.32500   5540.00200   6039.07700   6612.78700
CY3_HYB              13238.52000  12945.04000  12741.06000  12797.29000  11272.87000  12798.53000  12288.96000
HIGH_STRINGENCY_HYB  31363.13000  33693.30000  31494.96000  32456.38000  29554.45000  31253.40000  29861.95000
HOUSEKEEPING         20446.56000  19445.20000  19390.52000  19781.67000  18907.65000  19454.81000  20602.37000
LABELING                85.73669     78.42913     82.58459     79.09451     80.69878     80.78196     90.47139
LOW_STRINGENCY_HYB   13349.75000  12996.49000  12820.47000  12872.48000  11340.68000  12900.72000  12363.55000
NEGATIVE                84.07156     80.37030     83.57912     79.34184     80.80109     78.84014 

So why does lumiB then subtract CY3_HYB from the original values because that seems to have happened here?????
> exprs(rawXbgc[1:5,1:8])
        1881436004_A 1881436004_B 1881436004_C 1881436009_A 1881436009_B 1881436009_C 1881436049_A 1881436049_B
20605      -12968.22    -12692.88    -12469.81    -12011.57    -13004.72    -12574.24    -11326.32    -10950.92
3450747    -12755.18    -12521.99    -12280.62    -11741.46    -12791.27    -12355.83    -11141.54    -10749.57
3060450    -12618.29    -12323.17    -12114.76    -11585.49    -12690.99    -12312.38    -11025.68    -10610.91
870131     -13160.06    -12873.87    -12673.31    -12213.00    -13178.65    -12720.94    -11469.55    -11107.61
5310368    -13147.54    -12865.89    -12658.88    -12203.13    -13164.00    -12713.87    -11469.22    -11108.02

If you have trouble reading the columns, here an example with the first datapoint: 270.30210-13238.52000 = -12968.22

What should have happened is subtraction of average value of negative control values so sth like 80.

Here is the session info if that is useful:

R version 2.6.1 (2007-11-26) 
i386-apple-darwin8.10.1 

locale:
C

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

other attached packages:
 [1] lumi_1.4.0           annotate_1.16.1      xtable_1.5-2         AnnotationDbi_1.0.6  RSQLite_0.6-4       
 [6] DBI_0.2-4            mgcv_1.3-29          affy_1.16.0          preprocessCore_1.0.0 affyio_1.6.1        
[11] Biobase_1.16.2


Could this be sth wrong with my file? Anyone able to point me to the problem?

Thanks!

Merja
###########################################

This message has been scanned by F-Secure Anti-Virus for...{{dropped:4}}