Sammendrag
Background: Fishmeal is a key component in a variety of animal feed, and the origin and composition is of importance for quality and mar- ket price. DNA metabarcoding can be used to identify the species and authenticate the content of fishmeal since commercially harvested fish species and potential by-catch are well represented in COI refer- ence libraries. While DNA metabarcoding is of limited value in the identification of populations and origins within a species, it is well suited to identify and separate species. We tested the power of DNA metabarcoding for identification of species in mixed DNA samples of known fish species relevant to fishmeal production, and of a selection of commercially available fishmeals. Results: DNA metabarcoding of mixed DNA samples detected all species, even in highly complex sam- ples containing up to 30 species of equal DNA concentration. How- ever, the percentage reads did not correspond well to the known DNA concentration in each sample, and was heavily biased towards a hand- ful of species, especially Salmo salar and Sardina pilchardus. Other species, such as Sardinella aurita and Brevoortia patronus, were under- represented. DNA metabarcoding of mixed fishmeals performed bet- ter, and the number of reads corresponded closely to the known mass composition. In addition, DNA metabarcoding proved useful in reveal- ing undocumented species in presumably species-pure fishmeals, and in detecting traces of pests. Significance: Because fishmeals of differ- ent species trade at different prices in a competitive market, there is a risk of fraud. We conclude that while DNA metabarcoding using COI is subjected to primer and other biases, it is still effective in detecting unwanted species in mixed samples, even at low concentrations. DNA metabarcoding using general COI primers is a useful tool in quality control of fishmeal and detection of fraud.
Vis fullstendig beskrivelse
