Hello,

I recently upgraded to R 3.2.3 and re-installed my Bioconductor packages to the latest versions. I'm now seeing this error from arrayQualityMetrics:

Error: Unknown parameters: cex.axis

Below is how I got this; the data I'm using comes from https://www.ebi.ac.uk/arrayexpress/experiments/E-GEOD-29086/

> annotations <- data.frame( AssayName = c( "GSM720636", "GSM720635", "GSM720633", "GSM720624", "GSM720632", "GSM720631", "GSM720625", "GSM720627", "GSM720639", "GSM720637", "GSM720634", "GSM720630", "GSM720640", "GSM720641", "GSM720629", "GSM720638", "GSM720626", "GSM720628" ), FactorValue = c( "complete solution minus iron, Kas-1, 24 hour", "24 hour, Kas-1, complete solution", "24 hour, complete solution, Kas-1", "24 hour, complete solution, Tsu-1", "complete solution minus iron, Tsu-1, 48 hour", "complete solution minus iron, Tsu-1, 48 hour", "complete solution, Tsu-1, 24 hour", "Tsu-1, complete solution minus iron, 24 hour", "complete solution minus iron, Kas-1, 48 hour", "Kas-1, complete solution minus iron, 24 hour", "Kas-1, complete solution, 24 hour", "48 hour, complete solution minus iron, Tsu-1", "48 hour, Kas-1, complete solution minus iron", "48 hour, Kas-1, complete solution minus iron", "24 hour, complete solution minus iron, Tsu-1", "24 hour, complete solution minus iron, Tsu-1", "Tsu-1, complete solution, 24 hour", "24 hour, Tsu-1, complete solution minus iron" ), FileName = c( "GSM720636.CEL", "GSM720635.CEL", "GSM720633.CEL", "GSM720624.CEL", "GSM720632.CEL", "GSM720631.CEL", "GSM720625.CEL", "GSM720627.CEL", "GSM720639.CEL", "GSM720637.CEL", "GSM720634.CEL", "GSM720630.CEL", "GSM720640.CEL", "GSM720641.CEL", "GSM720629.CEL", "GSM720638.CEL", "GSM720626.CEL", "GSM720628.CEL") )

> rownames(annotations) <- annotations$AssayName

> factorColName <- "FactorValue"

> exptAcc <- "E-GEOD-29086"

> arrayDesign <- "A-AFFY-2"

> outDir <- "QC_TEST"

> reportTitle <- paste("Microarray quality metrics report for", exptAcc, "on array design", arrayDesign)

> library(Biobase)

> annotDF <- new("AnnotatedDataFrame", annotations)

> library(oligo)

> dataSet <- read.celfiles(annotations$FileName)

> phenoData(dataSet) <- annotDF

> library( arrayQualityMetrics )

> arrayQCdata <- arrayQualityMetrics(expressionset = dataSet, outdir = outDir, force = TRUE, do.logtransform = TRUE, intgroup = factorColName, spatial = FALSE, reporttitle = reportTitle )

I then see the following:

The report will be written into directory 'QC_TEST'.
Error: Unknown parameters: cex.axis
In addition: There were 19 warnings (use warnings() to see them)

> warnings()
Warning messages:
1: In maximumLevels(fac, n = length(colors)) :
  A factor was provided with 16 levels, but the colour map used here has only 9 colours. For the purpose of colouring, levels 9 ('complete solution minus iron, Kas-1, 24 hour') to 16 ('Tsu-1, complete solution, 24 hour') are being collapsed. Please consider grouping together some of the levels of your factor of interest to reduce the number of levels, this might improve the legibility of the plots.

2: In svgStyleAttributes(style, svgdev) :
  Removing non-SVG style attribute name(s): subscripts, group.number, group.value
3: In svgStyleAttributes(style, svgdev) :
  Removing non-SVG style attribute name(s): subscripts, group.number, group.value
4: In svgStyleAttributes(style, svgdev) :
  Removing non-SVG style attribute name(s): subscripts, group.number, group.value
5: In svgStyleAttributes(style, svgdev) :
  Removing non-SVG style attribute name(s): subscripts, group.number, group.value
6: In svgStyleAttributes(style, svgdev) :
  Removing non-SVG style attribute name(s): subscripts, group.number, group.value
7: In svgStyleAttributes(style, svgdev) :
  Removing non-SVG style attribute name(s): subscripts, group.number, group.value
8: In svgStyleAttributes(style, svgdev) :
  Removing non-SVG style attribute name(s): subscripts, group.number, group.value
9: In svgStyleAttributes(style, svgdev) :
  Removing non-SVG style attribute name(s): subscripts, group.number, group.value
10: In svgStyleAttributes(style, svgdev) :
  Removing non-SVG style attribute name(s): subscripts, group.number, group.value
11: In svgStyleAttributes(style, svgdev) :
  Removing non-SVG style attribute name(s): subscripts, group.number, group.value
12: In svgStyleAttributes(style, svgdev) :
  Removing non-SVG style attribute name(s): subscripts, group.number, group.value
13: In svgStyleAttributes(style, svgdev) :
  Removing non-SVG style attribute name(s): subscripts, group.number, group.value
14: In svgStyleAttributes(style, svgdev) :
  Removing non-SVG style attribute name(s): subscripts, group.number, group.value
15: In svgStyleAttributes(style, svgdev) :
  Removing non-SVG style attribute name(s): subscripts, group.number, group.value
16: In svgStyleAttributes(style, svgdev) :
  Removing non-SVG style attribute name(s): subscripts, group.number, group.value
17: In svgStyleAttributes(style, svgdev) :
  Removing non-SVG style attribute name(s): subscripts, group.number, group.value
18: In svgStyleAttributes(style, svgdev) :
  Removing non-SVG style attribute name(s): subscripts, group.number, group.value
19: In svgStyleAttributes(style, svgdev) :
  Removing non-SVG style attribute name(s): subscripts, group.number, group.value

Is this something I'm getting wrong somewhere? Thanks in advance for any help/suggestions!

Maria

> sessionInfo()
R version 3.2.3 (2015-12-10)
Platform: x86_64-apple-darwin13.4.0 (64-bit)
Running under: OS X 10.11.2 (El Capitan)

locale:
[1] en_GB.UTF-8/en_GB.UTF-8/en_GB.UTF-8/C/en_GB.UTF-8/en_GB.UTF-8

attached base packages:
[1] stats4    parallel  stats     graphics  grDevices utils     datasets
[8] methods   base

other attached packages:
 [1] arrayQualityMetrics_3.26.0 pd.ath1.121501_3.12.0
 [3] RSQLite_1.0.0              DBI_0.3.1
 [5] oligo_1.34.0               Biostrings_2.38.3
 [7] XVector_0.10.0             IRanges_2.4.6
 [9] S4Vectors_0.8.6            oligoClasses_1.32.0
[11] Biobase_2.30.0             BiocGenerics_0.16.1
[13] BiocInstaller_1.20.1

loaded via a namespace (and not attached):
 [1] base64_1.1                 Rcpp_0.12.2
 [3] lattice_0.20-33            foreach_1.4.3
 [5] GenomeInfoDb_1.6.1         plyr_1.8.3
 [7] acepack_1.3-3.3            ggplot2_2.0.0
 [9] zlibbioc_1.16.0            annotate_1.48.0
[11] beadarray_2.20.0           gridSVG_1.5-0
[13] rpart_4.1-10               preprocessCore_1.32.0
[15] splines_3.2.3              gcrma_2.42.0
[17] stringr_1.0.0              foreign_0.8-66
[19] bit_1.1-12                 munsell_0.4.2
[21] nnet_7.3-11                SummarizedExperiment_1.0.2
[23] gridExtra_2.0.0            ff_2.2-13
[25] Hmisc_3.17-1               codetools_0.2-14
[27] XML_3.98-1.3               grid_3.2.3
[29] xtable_1.8-0               gtable_0.1.2
[31] affy_1.48.0                magrittr_1.5
[33] scales_0.3.0               stringi_1.0-1
[35] hwriter_1.3.2              reshape2_1.4.1
[37] genefilter_1.52.0          setRNG_2013.9-1
[39] affyio_1.40.0              limma_3.26.3
[41] latticeExtra_0.6-26        SVGAnnotation_0.93-1
[43] Formula_1.2-1              RColorBrewer_1.1-2
[45] BeadDataPackR_1.22.0       illuminaio_0.12.0
[47] iterators_1.0.8            tools_3.2.3
[49] RJSONIO_1.3-0              Cairo_1.5-9
[51] affyPLM_1.46.0             survival_2.38-3
[53] AnnotationDbi_1.32.3       colorspace_1.2-6
[55] vsn_3.38.0                 cluster_2.0.3
[57] GenomicRanges_1.22.3       affxparser_1.42.0

It looks like the cex.axis argument is defined by the meanSdPlot() function arrayQualityMetrics uses, but throws an error when this is passed to the underlying ggplot function.  I don't think you can specify for this not to be the case.  I suggest you email the maintainer of arrayQualityMetics to see if this can be changed.

In the mean time you can produce a version of the metrics workflow without the meanSdPlot using the following code:

preparedData <- prepdata(expressionset = dataSet, do.logtransform = TRUE, intgroup = factorColName)

bo = aqm.boxplot(preparedData)
de = aqm.density(preparedData)
hm = aqm.heatmap(preparedData)
pca = aqm.pca(preparedData)
ma = aqm.maplot(preparedData)
qm = list("Boxplot" = bo, "Density" = de, "Heatmap" = hm, "PCA" = pca, "MAplot" = ma)

res = aqm.writereport(modules = qm, arrayTable = preparedData$pData, 
                      reporttitle = reportTitle, outdir = outDir)

This should be fixed in arrayQualityMetrics 3.27.1 - available from subversion since revision 112370, should be on the Bioconductor website within a day or so. Thanks to Mike for figuring this out.