Error in DESeq2 analysis with dimnames

Question

I'm getting the following error in a DESeq2 analysis and I can't seem to figure out the issue:

Error in `assays<-`(`*tmp*`, withDimnames = withDimnames, ..., value = `*vtmp*`) : please use 'assay(x, withDimnames=FALSE)) <- value' or 'assays(x, withDimnames=FALSE)) <- value' when the dimnames on the supplied assay(s) are not identical to the dimnames on RangedSummarizedExperiment object 'x'

Here is my code:

library(DESeq2)
library(edgeR)

url <- "https://www.ncbi.nlm.nih.gov/geo/download/?acc=GSE63310&format=file"
utils::download.file(url, destfile="GSE63310_RAW.tar", mode="wb") 
utils::untar("GSE63310_RAW.tar", exdir = ".")
files <- c("GSM1545535_10_6_5_11.txt", "GSM1545536_9_6_5_11.txt", "GSM1545538_purep53.txt",
           "GSM1545539_JMS8-2.txt", "GSM1545540_JMS8-3.txt", "GSM1545541_JMS8-4.txt",
           "GSM1545542_JMS8-5.txt", "GSM1545544_JMS9-P7c.txt", "GSM1545545_JMS9-P8c.txt")
for(i in paste(files, ".gz", sep=""))
  R.utils::gunzip(i, overwrite=TRUE)

files <- c("GSM1545535_10_6_5_11.txt", "GSM1545536_9_6_5_11.txt", 
           "GSM1545538_purep53.txt", "GSM1545539_JMS8-2.txt", 
           "GSM1545540_JMS8-3.txt", "GSM1545541_JMS8-4.txt", 
           "GSM1545542_JMS8-5.txt", "GSM1545544_JMS9-P7c.txt", 
           "GSM1545545_JMS9-P8c.txt")

x <- readDGE(files, columns=c(1,3))

samplenames <- substring(colnames(x), 12, nchar(colnames(x)))

colnames(x) <- samplenames
group <- as.factor(c("LP", "ML", "Basal", "Basal", "ML", "LP", 
                     "Basal", "ML", "LP"))
x$samples$group <- group

#DESeq2 Analysis
phenodata <- new("AnnotatedDataFrame", data = data.frame(celltype = x$samples$group))
counts <- x$counts
ids <- c(1,3,4,6,7,9)

tmpcounts = counts[, ids]
tmpph = as(phenodata[ids, ], "data.frame")
rownames(tmpph) <- colnames(tmpcounts)

dds = DESeqDataSetFromMatrix(countData = tmpcounts, colData = tmpph, design = ~ tmpph$celltype)

Here are my package versions

> packageVersion("DESeq2")
[1] ‘1.30.1’

> packageVersion("SummarizedExperiment")
[1] ‘1.20.0’

Answer 1

What works is:

dds = DESeqDataSetFromMatrix(countData = data.frame(tmpcounts), colData = tmpph, design = ~celltype)

Mind the data.frame() around the count data object (which was a matrix) and removal of tmpph$ for the design because design should reference one or more columns of the colData. In newer versions of DESeq2 that by the way works without the data.frame() but I am not sure why. You might want to dig that out yourself, looking at the source code of how exactly the package (SummarizedExperiment) feeds data as an assay into a SummarizedExperiments object.