DNA barcoding uses a short, standardized segment of an organism’s genome for identification. Much like a fingerprint which is unique to an individual, selected fragment(s) of DNA are unique to each species but require comparison to reference libraries for identification.
The reference libraries are built by sequencing these DNA fragments from expertly identified specimens, vouchered in public collections.
Several genes are targeted for DNA barcoding, depending on the group:
Animal studies primarily use a region in the mitochondrial cytochrome c oxidase 1 gene (aka. CO1, COX1, COI), though other genes such as 12S are now routinely used for vertebrates.
Plant studies typically use one or a few plastid regions (e.g. rbcL, matK, trnH-psbA) and the internal transcribed spacer (ITS) region of nuclear ribosomal DNA.
Fungal studies use the internal transcribed spacer (ITS) region of nuclear ribosomal DNA.