Dear All,

I noticed the error in geNetClassifier when using my own dataset which halts the execution. The same error appears when executing data and the code from the vignette [I ommited only translation of Ensembl gene IDs into Gene Symbols which is optional].

I will appreciate your help in resolving this issue.

Kind regards,

Pawel

> data(leukemiasEset)

> leukemiasEset

ExpressionSet (storageMode: lockedEnvironment)

assayData: 20172 features, 60 samples

  element names: exprs, se.exprs

protocolData

  sampleNames: GSM330151.CEL GSM330153.CEL ... GSM331677.CEL (60 total)

  varLabels: ScanDate

  varMetadata: labelDescription

phenoData

  sampleNames: GSM330151.CEL GSM330153.CEL ... GSM331677.CEL (60 total)

  varLabels: Project Tissue ... Subtype (5 total)

  varMetadata: labelDescription

featureData: none

experimentData: use 'experimentData(object)'

Annotation: genemapperhgu133plus2

> leukemiasClassifier <- geNetClassifier(leukemiasEset, sampleLabels="LeukemiaType", plotsName="leukemiasClassifier", estimateGError=TRUE)

11:10:40 - Filtering data and calculating the genes ranking...

11:11:19 - Calculating correlations between genes...

Network calculated without interactions

Gene associations (network) calculated between genes with posterior probability over 0.95

11:11:20 - All required parameters have been calculated and checked. Building classifier...

11:12:02 - 1 out of 5 cross-validation loops finished.

11:12:40 - 2 out of 5 cross-validation loops finished.

11:13:30 - 3 out of 5 cross-validation loops finished.

11:14:14 - 4 out of 5 cross-validation loops finished.

11:14:55 - 5 out of 5 cross-validation loops finished.

Error in (function (cl, name, valueClass)  :

  assignment of an object of class “table” is not valid for @‘confMatrix’ in an object of class “GeneralizationError”; is(value, "matrix") is not TRUE

In addition: Warning message:

In geNetClassifier(leukemiasEset, sampleLabels = "LeukemiaType",  :

  Since the number of samples is not multiple of 5, some samples might be used as test in several cross-validation loops when estimating the generalization error of the classifier.

> sessionInfo(package = NULL)

R version 3.3.0 (2016-05-03)

Platform: x86_64-w64-mingw32/x64 (64-bit)

Running under: Windows 7 x64 (build 7601) Service Pack 1

[1] parallel  stats     graphics  grDevices utils     datasets  methods   base    

other attached packages:

 [1] cluster_2.0.4          geNetClassifier_1.12.0 minet_3.30.0           EBarrays_2.36.0        lattice_0.20-34      

 [6] leukemiasEset_1.8.0    randomForest_4.6-12    CMA_1.30.0             Biobase_2.32.0         BiocGenerics_0.18.0  
[11] VSURF_1.0.3          

loaded via a namespace (and not attached):

 [1] codetools_0.2-14  class_7.3-14      foreach_1.4.3     grid_3.3.0        e1071_1.6-7       doParallel_1.0.10 rpart_4.1-10    

 [8] iterators_1.0.8   tools_3.3.0       compiler_3.3.0