I'm trying to perform RNASeq analysis with DESeq2 in RStudio, and as one of the quality control steps I am trying to perform Principal Component Analysis on my DESeq2 object. I've included the code chunk below - I am able to create the DESeq2 object just fine, but run into problems plotPCA(). I think there's an issue with the packages talking to each other/masking each other, but I'm not sure how to resolve it. I've also tried DESeq2::plotPCA() and still received the same error message. Any help would be greatly appreciated - thanks!

#code from my script

library(DESeq2)
dataset = "./data/merged.csv"
dataset = read.table(dataset,
                     header = T,
                     sep = ",",
                     row.names = 1,
                     stringsAsFactors = F)


metadata = data.frame(condition = (c("CTL", "TS21", "TS21", "CTL")),
                      row.names = c("NDC", "DSIPS", "THO3B", "THO3F"))

dds_DS <- DESeqDataSetFromMatrix(countData = dataset,
                                 colData = metadata,
                                 design = ~condition)

library(pheatmap)
library(RColorBrewer)

plotPCA(vst_dds_DS, intgroup="condition")

#Error Message!
Error in (function (classes, fdef, mtable)  : 
  unable to find an inherited method for function ‘plotPCA’ for signature ‘"matrix"’

sessionInfo()
R version 4.0.0 (2020-04-24)
Platform: x86_64-apple-darwin17.0 (64-bit)
Running under: macOS Catalina 10.15.4

Matrix products: default
BLAS:   /System/Library/Frameworks/Accelerate.framework/Versions/A/Frameworks/vecLib.framework/Versions/A/libBLAS.dylib
LAPACK: /Library/Frameworks/R.framework/Versions/4.0/Resources/lib/libRlapack.dylib

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

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

other attached packages:
 [1] RColorBrewer_1.1-2          pheatmap_1.0.12            
 [3] DESeq2_1.28.1               SummarizedExperiment_1.18.1
 [5] DelayedArray_0.14.0         matrixStats_0.56.0         
 [7] Biobase_2.48.0              GenomicRanges_1.40.0       
 [9] GenomeInfoDb_1.24.0         IRanges_2.22.2             
[11] S4Vectors_0.26.1            BiocGenerics_0.34.0        

loaded via a namespace (and not attached):
 [1] Rcpp_1.0.4.6           locfit_1.5-9.4        
 [3] lattice_0.20-41        digest_0.6.25         
 [5] R6_2.4.1               RSQLite_2.2.0         
 [7] ggplot2_3.3.1          pillar_1.4.4          
 [9] zlibbioc_1.34.0        rlang_0.4.6           
[11] rstudioapi_0.11        annotate_1.66.0       
[13] blob_1.2.1             Matrix_1.2-18         
[15] splines_4.0.0          BiocParallel_1.22.0   
[17] geneplotter_1.66.0     RCurl_1.98-1.2        
[19] bit_1.1-15.2           munsell_0.5.0         
[21] tinytex_0.23           compiler_4.0.0        
[23] xfun_0.14              pkgconfig_2.0.3       
[25] tidyselect_1.1.0       tibble_3.0.1          
[27] GenomeInfoDbData_1.2.3 XML_3.99-0.3          
[29] crayon_1.3.4           dplyr_1.0.0           
[31] bitops_1.0-6           grid_4.0.0            
[33] xtable_1.8-4           gtable_0.3.0          
[35] lifecycle_0.2.0        DBI_1.1.0             
[37] magrittr_1.5           scales_1.1.1          
[39] farver_2.0.3           XVector_0.28.0        
[41] genefilter_1.70.0      ellipsis_0.3.1        
[43] vctrs_0.3.1            generics_0.0.2        
[45] tools_4.0.0            bit64_0.9-7           
[47] glue_1.4.1             purrr_0.3.4           
[49] survival_3.1-12        yaml_2.2.1            
[51] AnnotationDbi_1.50.0   colorspace_1.4-1      
[53] memoise_1.1.0          knitr_1.28            

Any help would be greatly appreciated! Thanks!

I'll give the answer but also how you could find it with some sleuthing.

Let's say I try to call plotPCA on this object x:

> x <- 5
> plotPCA(x)
Error in (function (classes, fdef, mtable)  :
  unable to find an inherited method for function ‘plotPCA’ for signature ‘"numeric"’
Calls: plotPCA -> <Anonymous>
> class(x)
[1] "numeric"

So R is telling me that plotPCA is not designed for objects like x. So what class is it designed for?

> ?plotPCA

After selecting the DESeq2 method, you get:

Arguments:

  object: a ‘DESeqTransform’ object, with data in ‘assay(x)’, produced
          for example by either ‘rlog’ or
          ‘varianceStabilizingTransformation’.

So you should provide plotPCA with an object that is a DESeqTransform object as output by rlog or vst or varianceStabilizingTransformation.

What you've likely done is taken out just the assay data and provided this to plotPCA? But that won't work because then how can the function know which samples belong to which group?

Here is an example of use:

http://bioconductor.org/packages/devel/bioc/vignettes/DESeq2/inst/doc/DESeq2.html#principal-component-plot-of-the-samples