Hello, I am trying to analyze some RNASeq data using the WCGNA package. For the most part it works fine but I can't seem to connect my trait data with my expression data.

I have a normalized expression matrix with 39 samples and a lot of contigs. My data is a developmental series with 3-16 samples at each stage of development. The "trait" data is just the developmental stage. So I have a trait file with "Stage" (numbered 0-5) and "Individual". I added X's before the individual names in both files, because they were just numbers. When I try and connect the traits with the expression data I get a problem with the rownames.

The code I am using and error are:

>allTraits = traitData[, c(1:2) ]
>    dim(allTraits) 
[1] 39  2
>    names(allTraits)
[1] "Individual"   "Stage_number"
>   Eep_samples=rownames(datExpr)
>   traitRows = match(Eep_samples, allTraits$Individual)
>   datTraits = allTraits[traitRows, -1]
>   rownames(datTraits) = allTraits[traitRows, 1]
Error in `rownames<-`(`*tmp*`, value = c("X221_quant", "X267_quant", "X320_quant",  : 
  attempt to set 'rownames' on an object with no dimensions

I am just learning R. Am missunderstanding what the code is doing and am actually deleting all the trait data? I assume the -1, 1 are defining where the row names are...

Any help would be appreciated! Thank you!

> sessionInfo() 

R version 3.4.1 (2017-06-30)
Platform: x86_64-apple-darwin15.6.0 (64-bit)
Running under: macOS Sierra 10.12.6

Add a "drop = FALSE" to this line: datTraits = allTraits[traitRows, -1]

datTraits = allTraits[traitRows, -1, drop = FALSE]

Your allTraits contains only 2 columns; when you remove the first one, you are left with just one column and R converts that into a single vector unless you add the drop = FALSE argument.

HTH

Peter