Evaluating the usefulness of next-generation sequencing for herb authentication

Food authentication is a rapidly growing field driven by increasing public awareness of food quality and safety. Foods containing herbs are particularly prone to industrial fraud and adulteration. Several methodologies are currently used to evaluate food authenticity, but recently increased focus has been on DNA technologies, with DNA barcoding the most widely used. DNA barcoding is based on the sequencing and comparison of orthologous DNA regions from all species in a sample, but the approach is limited by its low resolution to distinguish closely-related species. Here we developed a customised database and bioinformatic pipeline to identify herbal ingredients implemented as a metagenomics approach for plant-derived product authenticity testing and evaluated the accuracy of the method by using publicly available as well as 47 de novo partly sequenced whole plant genomes or barcodes as query sequences. Our results show that using our mapping algorithm with the customised barcode database correctly identifies the main components of a wide range of plant-derived samples, albeit with variable additional false-positive matches across samples depending on the tested samples and barcodes. Our result also show that at the current stage the usefulness of metagenomics is limited by the availability of reference sequences and the needed sequencing depth. However, this method shows promise for evaluating the authenticity of different herbal products provided that the method is further refined to increase the qualitative and quantitative accuracy.