Amplicon & microbiome sequencing
PCR of one or two 16S variable regions (often V4 or V3–V4) with barcoded primers, paired-end reads merged and denoised into exact variants, classified against a 16S reference database.
Our ruleSimulations prepare you for the bench but don't replace hands-on practice.
16S rRNA
PCR of one or two 16S variable regions (often V4 or V3–V4) with barcoded primers, paired-end reads merged and denoised into exact variants, classified against a 16S reference database.
How it differs from the standard method
- Resolution usually reaches genus; species and strains only where the region differs.
- 16S copy number varies between taxa, so read shares are not cell shares.
- Needs a diversity spike-in (PhiX) or phased primers on the run.
- Mock community and extraction blank are part of every run.
1Counting who is there
A microbial community can be profiled in two ways. Marker-gene (amplicon) sequencing amplifies one stretch of DNA that every member carries but that differs between taxa: a variable region of the 16S rRNA gene for bacteria and archaea, or the internal transcribed spacer (ITS) for fungi.
Exact sequence variants
Denoising separates real sequences from sequencing errors down to a single base, so strains that differ by one base stay apart.
Mock community
Known strains in known proportions, processed with the samples: it shows what lysis, PCR and the pipeline do to a true composition.
Extraction blank
Reagents with no sample, carried through every step: it shows the contaminant DNA that every kit and water batch carries.
2Where the profile is distorted
Every step between cells and counts has its own bias. Cells with tough walls (Gram-positive bacteria, moulds) release DNA only with mechanical bead beating.
3Depth, saturation and contamination
- Amplicon libraries have very low base diversity: every read starts with the same primer. A diverse spike-in such as 10–20% PhiX keeps the run's quality up.
- Richness depends on depth. A rarefaction curve that is still climbing means rare members are still being found, so comparing richness between samples compares depth.
- Low-biomass samples (skin, tissue, air, clean rooms) can be dominated by reagent contaminants. Blanks are the only way to recognise them.
- In shotgun data from host-rich samples, most reads are human unless host DNA is depleted. A rare pathogen then gets a tiny share of an already small microbial share.
- Read classification only finds what the database contains; a missing genome means unclassified reads, not absent organisms.
Key terms
Practise these as flashcards →The small-subunit ribosomal RNA gene of bacteria and archaea; conserved regions anchor primers, variable regions (V1–V9) distinguish taxa.
Spacer regions between fungal rRNA genes (ITS1, ITS2); the standard fungal barcode, highly variable in sequence and length.
Exact sequence variants from denoising, versus operational taxonomic units clustered at a fixed identity such as 97%.
A PCR artefact joining parts of two templates, formed when an incomplete extension primes on another template in a later cycle.
Features observed against reads sampled; a flattening curve means the sample has been sequenced deeply enough.
Diversity within one sample (richness, Shannon) and difference between samples (for example Bray–Curtis dissimilarity).
Contaminating microbial DNA from extraction kits, reagents and water, prominent in low-biomass samples.
Metagenome-assembled genome: a genome reconstructed from shotgun data by assembly and binning, usually needing ≥ 10× coverage.
When to use it
- Comparing bacterial or fungal composition across many samples cheaply (amplicons).
- Low-biomass or host-rich samples where only a targeted PCR gives enough microbial signal (amplicons, with strict controls).
- Species- and strain-level profiles, functional genes such as antimicrobial resistance, and genome recovery (shotgun).
- Communities that include viruses or organisms without a reliable marker gene (shotgun).
Compared with alternatives
For a single cultured isolate: complete genome, typing and resistance genes without community background.
The run behind both approaches; low-diversity amplicon libraries need its PhiX and loading lessons.
Identifies one cultured organism quickly and cheaply when culture is possible.
Absolute quantity of a known taxon or gene; sequencing profiles are relative.
Sub-techniques
Choosing one here switches the whole page, including the protocol and the lab, to it.
PCR of one or two 16S variable regions (often V4 or V3–V4) with barcoded primers, paired-end reads merged and denoised into exact variants, classified against a 16S reference database.
- Resolution usually reaches genus; species and strains only where the region differs.
- 16S copy number varies between taxa, so read shares are not cell shares.
- Needs a diversity spike-in (PhiX) or phased primers on the run.
- Mock community and extraction blank are part of every run.