Hi,

We tried to plot a heatmap of the 1000 most variable CpGs in our data set of 12 samples using QC.GUI, but found out that the column labels shown are not in the right order.

This is reproducible:

tt = matrix(rnorm(6*1000), ncol=6, nrow=1000)
ttt = cbind(tt, tt+3)
rownames(ttt) = paste("R",1:1000, sep="_")
colnames(ttt) = paste("C",1:12, sep="_")

pheatmap(ttt) # heatmap as expected with C1-C6 and C7-C12 clustering together

QC.GUI(beta = ttt, pheno = c(rep("A",6),rep("B",6)), arraytype = "EPIC") # wrong column name order in heatmap

Best regards,

Hilmar

> sessionInfo()
R version 3.3.0 (2016-05-03)
Platform: x86_64-pc-linux-gnu (64-bit)
Running under: Ubuntu 14.04.5 LTS

locale:
 [1] LC_CTYPE=en_US.UTF-8       LC_NUMERIC=C               LC_TIME=en_US.UTF-8        LC_COLLATE=en_US.UTF-8    
 [5] LC_MONETARY=en_US.UTF-8    LC_MESSAGES=en_US.UTF-8    LC_PAPER=en_US.UTF-8       LC_NAME=C                 
 [9] LC_ADDRESS=C               LC_TELEPHONE=C             LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C       

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

other attached packages:
 [1] plotly_4.5.6                               ggplot2_2.2.1                             
 [3] shiny_0.14.2                               pheatmap_1.0.8                            
 [5] matrixStats_0.51.0                         ChAMP_2.4.1                               
 [7] IlluminaHumanMethylationEPICmanifest_0.3.0 Illumina450ProbeVariants.db_1.10.0        
 [9] DMRcate_1.10.1                             DMRcatedata_1.10.1                        
[11] DSS_2.14.0                                 bsseq_1.10.0                              
[13] FEM_3.2.0                                  graph_1.52.0                              
[15] org.Hs.eg.db_3.4.0                         impute_1.48.0                             
[17] igraph_1.0.1                               corrplot_0.77                             
[19] marray_1.52.0                              Matrix_1.2-7.1                            
[21] AnnotationDbi_1.36.0                       ChAMPdata_2.2.0                           
[23] minfi_1.20.2                               bumphunter_1.14.0                         
[25] locfit_1.5-9.1                             iterators_1.0.8                           
[27] foreach_1.4.3                              Biostrings_2.42.1                         
[29] XVector_0.14.0                             SummarizedExperiment_1.4.0                
[31] GenomicRanges_1.26.2                       GenomeInfoDb_1.10.2                       
[33] IRanges_2.8.1                              S4Vectors_0.12.1                          
[35] Biobase_2.34.0                             BiocGenerics_0.20.0                       
[37] limma_3.30.7                              

Hello Hilmar:

You may check Plotly uprading issue in ChAMP package to see if your error is this, I checked you sessionInfo and which shows you are using plotly 4.5.6. Which is not supported by ChAMP for now.

Actually...I thought you can not plot most figures, not only heatmap on your current computer. As I right?

Best

Yuan Tian

Hi Yuan,

Thanks a lot. I could well be that the plotly version plays a role here, however, looking at your code it might as well be a problem in Plotly itself. In theory this kind of problems should be caught by appropriate unit tests, which is not trivial to do for interactive GUI code, though. It is definitely hard to do thorough testing of GUI interfaces that then will be used by biologist that tend to rely on the output presented - which is why probably most R packages do not have them.

And, yes, only 2 out of tabs of the QC.GUI showed any plot.

Thanks again and best regards,

Hilmar