I am trying to understand what this error means: Error in nbinomLRT(object, full = full, reduced = reduced, quiet = quiet, : less than one degree of freedom, perhaps full and reduced models are not in the correct order

I am following https://www.bioconductor.org/packages/devel/workflows/vignettes/rnaseqGene/inst/doc/rnaseqGene.html#time-course-experiments

I have a design where I am measuring cell types, alterations to the cell and time.

ddsTC <- DESeqDataSet(dds, ~ change + time + cell + change:time) ddsTC

class: DESeqDataSet dim: 36122 8 metadata(1): version assays(4): counts mu H cooks rownames(36122): ENSG00000000003 ENSG00000000419 ... ENSG00000283122 ENSG00000283125 rowData names(30): baseMean baseVar ... deviance maxCooks colnames: NULL colData names(5): sample_id change time cell sizeFactor

But when I run the next line I get the error mentioned above:

ddsTC <- DESeq(ddsTC, test="LRT", reduced = ~ change + time + cell + change:time)

How do you determine what the "correct order" would be?

Check the vignette section on the LRT. The reduced design should have fewer terms than the full design. Above you have the identical design for both.