eDNA metabarcoding, explained without the jargon

If you have looked into eDNA at all, you have probably seen the word metabarcoding. It sounds like buzzword filler, but it points to a specific and important trick: reading many species out of one sample at the same time. Understanding metabarcoding is the difference between thinking eDNA is magic and understanding what it actually can and cannot do.

DNA barcoding: one species at a time

The original idea of DNA barcoding, from the early 2000s, was simple: pick a short region of the genome that varies enough between species to identify them, and use it as a molecular fingerprint. COI for animals, ITS for fungi, rbcL and matK for plants — each became a standard barcode.

For a long time, barcoding was applied one specimen at a time. You had a beetle, you sequenced its COI, you matched it to a reference and got a species name.

Metabarcoding: many species from one soup

Metabarcoding applies the same barcode idea to a mixed sample. Instead of DNA from one beetle, you have DNA from a pond, a soil core, or a stomach sample — hundreds of species mixed together.

The workflow uses primers that anchor to conserved regions flanking a variable barcode, so a single PCR reaction copies the diagnostic region from every species carrying that anchor. High-throughput sequencing then reads the whole mixture, and bioinformatics untangles which reads came from which species.

Primer choice controls what you see

The primer set is the lens. A vertebrate 12S primer will show you fish, amphibians, reptiles, birds, and mammals but ignore insects and plants. A COI primer will pick up a wide sweep of animals but is not ideal for fungi or plants. ITS is standard for fungi. rbcL and ITS2 are common for plants.

Serious surveys often use several primer sets in parallel to cover the full community. Consumer-facing tests usually pick primers matched to the target audience: pollinators, backyard vertebrates, freshwater fish, and so on.

The reference database problem

Metabarcoding is only as good as the reference sequences it compares to. A species that has never been sequenced simply cannot be identified by name — it shows up as an unknown OTU. Databases are strong for North American vertebrates and getting better for insects, but coverage of tropical invertebrates and many fungi is still patchy.

This is why some hits come back as a genus rather than a species, or as 'unknown Diptera' — the barcode region for that group is either variable enough only at higher taxonomic ranks, or the exact species has not been sequenced yet.

Frequently asked questions

Why do some hits come back only to genus?

Either the barcode region does not have enough variation to separate closely related species, or the specific species has not yet been added to the reference database.

Can one primer set cover everything?

No. Broad primers exist, but every primer set has bias. Comprehensive surveys combine primers targeted at different groups.