Dear All,

I want to analyze two color .GPR files from Agilent 4x44K array downloaded from GEO. This data set is a nucleosomal ChIP-ChIP experiment where channel 1 and 2 are labeled with two different antibodies. I am a newbie to ChIP-ChIP data analysis, though I could manage to create the RGlist and ExpressionSet using limma and Ringo package. Further I tried using CMARRT algorithm from Starr package to find Chip-enriched regions. But it throws error "Error in density.default(logR) : 'x' contains missing values". Please point out the mistake in the script I am using for analysis.

targets <- readTargets("targets.txt", row.names="Name")
RG <- read.maimages(targets,columns=list(R="F635 Mean",G="F532 Mean",Rb="B635 Median",Gb="B532 Median"), wt.fun=wtflags(weight=0,cutoff=-50),
          annotation=c("Block","Row","Column","ID","Name")) 
eSetB <- preprocess(RG, method="loess", ChIPChannel="R", inputChannel="G",
                    returnMAList=FALSE, idColumn="ID", backgroundCorrect(RG, method="subtract"))

peaks = cmarrt.ma(eSetB, GPL_probeAnno, chr = NULL, M= NULL,
                   frag.length = 300)

"Error in density.default(logR) : 'x' contains missing values"

I have noticed that if i avoid mentioning Rb="B635 Median",Gb="B532 Median" in the limma package's read.maimages command, The script does not throw any error, and i get the peaks as list object. It would be great If someone could point out the mistake and help in troubleshooting.

sessionInfo()   

R version 3.2.0 (2015-04-16)
Platform: x86_64-w64-mingw32/x64 (64-bit)
Running under: Windows 8 x64 (build 9200)

locale:
[1] LC_COLLATE=English_United States.1252  LC_CTYPE=English_United States.1252   
[3] LC_MONETARY=English_United States.1252 LC_NUMERIC=C                          
[5] LC_TIME=English_United States.1252    

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

other attached packages:
 [1] Starr_1.24.0        affxparser_1.40.0   affy_1.46.1         Ringo_1.32.0        lattice_0.20-31    
 [6] Matrix_1.2-1        limma_3.24.12       RColorBrewer_1.1-2  Biobase_2.28.0      BiocGenerics_0.14.0

loaded via a namespace (and not attached):
 [1] AnnotationDbi_1.30.1  MASS_7.3-42           splines_3.2.0         zlibbioc_1.14.0      
 [5] IRanges_2.2.5         pspline_1.0-17        xtable_1.7-4          GenomeInfoDb_1.4.1   
 [9] tools_3.2.0           DBI_0.3.1             genefilter_1.50.0     survival_2.38-3      
[13] preprocessCore_1.30.0 affyio_1.36.0         S4Vectors_0.6.1       RSQLite_1.0.0        
[17] BiocInstaller_1.18.3  stats4_3.2.0          XML_3.98-1.3          annotate_1.46.0      
[21] vsn_3.36.0 

The hint comes from ?backgroundCorrect:

Details:

     This function implements the background correction methods
     reviewed or developed in Ritchie et al (2007) and Silver at al
     (2009). Ritchie et al (2007) recommend  method="normexp"  whenever
      RG  contains local background estimates. Silver et al (2009)
     shows that either  normexp.method="mle"  or
      normexp.method="saddle"  are excellent options for normexp. If
      RG  contains morphological background estimates instead
     (available from SPOT or GenePix image analysis software), then
      method="subtract"  performs well.

     If  method="none"  then no correction is done, i.e., the
     background intensities are treated as zero. If  method="subtract"
     then the background intensities are subtracted from the foreground
     intensities. This is the traditional background correction method,
     but is not necessarily recommended. If  method="movingmin"  then
     the background estimates are replaced with the minimums of the
     backgrounds of the spot and its eight neighbors, i.e., the
     background is replaced by a moving minimum of 3x3 grids of spots.

     The remaining methods are all designed to produce positive
     corrected intensities. If  method="half"  then any intensity which
     is less than 0.5 after background subtraction is reset to be equal
     to 0.5. If  method="minimum"  then any intensity which is zero or
     negative after background subtraction is set equal to half the
     minimum of the positive corrected intensities for that array. If
      method="edwards"  a log-linear interpolation method is used to
     adjust lower intensities as in Edwards (2003). If
      method="normexp"  a convolution of normal and exponential
     distributions is fitted to the foreground intensities using the
     background intensities as a covariate, and the expected signal
     given the observed foreground becomes the corrected intensity.
     This results in a smooth monotonic transformation of the
     background subtracted intensities such that all the corrected
     intensities are positive.

If you use a background correction method that can result in negative values, then when you take logs you get a NaN, which will be interpreted as a missing value.