Hi,

I have a DESeq2 script that used to work until recently, I'm not sure if I updated a package which suddenly caused it to stop function properly, and I'm hoping to get some insight on what the issue might be stemming from.

colData has 6 factors (A,B,C,D,E,F)

dds <- DESeqDataSetFromMatrix(countData = countData, colData = colData, design = ~ Experiment)
dds <- DESeq(dds)
resultsNames(dds)
[1] "Intercept"   "Experiment1" "Experiment2" "Experiment3" "Experiment4" "Experiment5"
res <- results(dds, contrast=c("Experiment","B","A"))

This error pops up for when results() is called: "Error in cleanContrast(object, contrast, expanded = isExpanded, listValues = listValues, : as A is the reference level, was expecting Experiment_B_vs_A to be present in 'resultsNames(object)'"

The resultsNames(dds) output is not what I was expecting and I assume it is a good clue to the source of the issue, but I'm still not sure why this suddenly stopped working from a year ago.

Thanks for your time.

The resultsNames(dds) output is not what I was expecting and I assume it is a good clue to the source of the issue, but I'm still not sure why this suddenly stopped working from a year ago.

Likely because R changed how it deals with strings:

https://developer.r-project.org/Blog/public/2020/02/16/stringsasfactors/index.html