Skip to content

Shotgun Metagenomics

Domain key: metagenomics

Shotgun metagenomics sequences all DNA in a sample without prior amplification, enabling taxonomic profiling and de novo assembly of microbial communities. Host DNA removal is the first mandatory step.

Step Tool Notes
QC + trimming fastp Fast combined QC and trimming
Host removal Bowtie2 Align to host genome, keep unmapped
Taxonomy (k-mer) Kraken2 + Bracken Fast classification + abundance re-estimation
Taxonomy (marker) MetaPhlAn4 High-specificity species profiling
Assembly MEGAHIT Standard for complex communities
Annotation Prokka Rapid prokaryotic genome annotation

Best Practices Enforced

  • Host DNA removal (human or other host) is always Step 1 — reads mapping to the host genome are discarded before any microbial analysis
  • Both Kraken2 and MetaPhlAn4 are run by default for orthogonal taxonomy estimates
  • Bracken is run after Kraken2 to re-estimate species-level abundances from genus-level k-mer hits
  • MEGAHIT is used for assembly; SPAdes is available as an alternative for lower-complexity samples

Example Prompts

Build a shotgun metagenomics pipeline for human gut samples.
Remove human host reads first, then run Kraken2 and MetaPhlAn4.

codaris > I have soil metagenomics samples — no host to remove.
          Run MEGAHIT assembly and Prokka annotation.

Build a metagenomics pipeline with both taxonomic profiling
and de novo assembly. Samples are from mouse cecum — remove
mouse genome reads first.