I'm trying to do some contrasting in DESeq2 and I keep getting an error that my object is empty.

Here's my metadata table:

Sample Data: [21 samples by 5 sample variables]: Sample Site Treatment Impacted AllParams Indv Cass2 Lake Uncontaminated No CLUn CLUn2 Cass1 Lake Uncontaminated No CLUn CLUn1 Cass16 Lake Oil Yes CLOil CLOil16 Cass17 Lake Oil Yes LOil CLOil17 Cass12 Lake Plume Yes LPlume LPlume12 Cass15 Lake Plume Yes LPlume LPlume15 Cass6 Lake Plume Yes LPlume LPlume6 B14 Bem Uncontaminated No BUncon BUn14 B15 Bem Uncontaminated No BUncon BUn15 B115 Bem Oil Yes BOil BOil115 B124 Bem Oil Yes BOil BOil124 Bem26 Bem Oil Yes BOil BOil26 B75 Bem Plume Yes BPlume BPlume75 B76 Bem Plume Yes BPlume BPlume76 B71 Bem Plume Yes BPlume B_Plume71

I subset it so that it now looks like this:

Sample Data: [4 samples by 5 sample variables]: Sample Site Treatment Impacted AllParams Indv B14 Bem Uncontaminated No BUn BUn14 B15 Bem Uncontaminated No BUn BUn15 Cass1 Lake Uncontaminated No LUn LUn1 Cass2 Lake Uncontaminated No LUn L_Un2

The name of my subset file is Uncontaminated_phyla

From the start: clbbacphyloseq <- import_mothur(shared file & constaxonomy file)

clbbacmetadata <- data.frame(read .csv)

clbbacmetadataphyloseq <- sampledata(data.frame(Site = clbbacmetadata["Site"], Treatment = clbbacmetadata["Treatment"], Impacted = clbbacmetadata["Impacted"], AllParams = clbbacmetadata["AllParams"], Indv = clbbacmetadata["Indv"]))

clbbacfullphyloseq <- mergephyloseq(clbbacphyloseq, clbbacmetadata_phyloseq)

clbbacfullphyla <- taxglom(clbbacfull_phyloseq, taxrank = "Rank2")

Uncontaminatedphyla <- subsetsamples(clbbacfullphyla, Treatment=="Uncontaminated") sampledata(Uncontaminated_phyla)

ddsUncontaminatedphyla <- phyloseqtodeseq2(Uncontaminatedphyla, ~Indv) sampledata(ddsUncontaminatedphyla)

When I run phyloseqtodeseq2, my object is empty: Error in access(object, "samdata", errorIfNULL) : samdata slot is empty.

This is a phyloseq issue, as you haven't gotten the data into DESeq2 yet.

Try making a new post (use backticks as Mike Smith suggests also for code) and tagging that package instead of DESeq2 to notify the appropriate package maintainers.