Hi everyone,

I`m a bachelor student and for my bachelor thesis I need to build networks for gene expression analysis. I hardly get any support from anyone.

So I set up my DESeq2 dataset like this:

dds <- DESeqDataSetFromMatrix(countData = readcounts_bonobo,colData = coldata_Khrameeva, design = ~ brain_areal)

When I call on this I don't get any error.

This is what my coldata looks like:

Next I'm going to specify the factors for the comparison. I want to compare this 7 brain areas:

dds$brain_areal = factor(dds$brain_areal, levels = c("primary and secondary cortices","limbic and association cortices", "archicortex", "thalamus and hypothalamus", "white matter structures", "cerebellar gray matter", "striatum"))

This also works without any errors.

But when I want perform the differential expression analysis with:

dds = DESeq(dds)

I get this error:

Error in designAndArgChecker(object, betaPrior) : full model matrix is less than full rank

Is please anyone so kind to help me? I'm really desperate.

What I see immediately is that colData has corticies while your factor definition has cortices, so this mismatch (missing "i") will generate NAs in the factor levels. It is generally not necessary to define factor levels by hand or dds$brain_areal. DESeq2 will make character columns factorial internally. You can then use contrasts (see vignette) to define any comparison you need in results().