Metadata & the Decomposition Microbiome¶
The workshop dataset comes from the Metcalf Lab's vertebrate decomposition research: postmortem microbial communities sampled from cadavers and the surrounding soil across the active stages of decomposition. This block frames that biological context before you touch the code, and walks through the metadata file that encodes it.
What this lecture covers¶
Decomposition microbiome biology¶
- The stages of vertebrate decomposition (fresh, bloat, active decay, advanced decay, dry/skeletal) and what each stage means for the microbial community
- Postmortem microbial succession: how host-, soil-, and arthropod-derived taxa change in relative abundance across a decomposition timeline
- Accumulated degree days (ADD) as the temporal axis used in decomposition studies: time since death corrected for ambient temperature, which captures biological progression more reliably than calendar time alone
- The role of soil "necrobiome" communities and how they differ between sample types and facilities
- Why decomposition microbiomes have forensic and ecological significance (postmortem interval estimation, nutrient cycling)
The metadata file that encodes it¶
- What QIIME2 expects in a metadata file: TSV layout, the sample-id column, column types, special markers
- Categorical vs numeric variables, and why mistakes here cause silent analysis errors
- The columns specific to this dataset:
sample_type,facility,host_subject_id,add_0c(ADD), decomposition stage labels, and the controls - Common pitfalls: spaces in IDs, inconsistent NA encoding, hidden Excel autoformatting that mangles dates and identifiers
Hands-on follow-along¶
Open metadata_q2_workshop.txt in a text editor (or run qiime metadata tabulate once you have QIIME2 active in Activating QIIME2 & Importing Metadata) and skim the columns. Note which are categorical (sample type, facility, stage) and which are continuous (ADD, read counts), and which sample IDs correspond to controls vs experimental cadaver samples.