High-throughput biodiversity pipelines

Metabarcoding and metagenomic processing pipelines for high-throughput biodiversity surveys. <b>Top row</b>: Point locations across a landscape are surveyed for biodiversity, and DNA is separately extracted from each sample. Three common sample types are (i) bulk samples of arthropods (depicted here), (ii) environmental DNA (eDNA) from soil, water, and air, and (iii) invertebrate collectors of vertebrate DNA (iDNA), such as mosquitoes, leeches, flies, dung beetles, and ticks. <b>Left column</b>: Metabarcoding – Each sample’s DNA is amplified via PCR (polymerase chain reaction) for a particular marker gene that is taxonomically informative, the samples are pooled and sequenced on a high-throughput sequencer, and then sorted back to sample by the sample-specific tags added during PCR. The sequences are then clustered into OTUs (operational taxonomic units), which are species hypotheses, and assigned taxonomies by matching against online databases. <b>Right column</b>: Meta/mitogenomics – Each sample’s total DNA is sequenced, and the output DNA reads are matched to reference genomes, which are often mitochondrial genomes. <b>Bottom row</b>: The output of both processing pipelines is a ‘sample X species’ table. Metabarcoding pipelines are better for general biodiversity discovery and surveys because online barcode databases are more taxonomically complete, and even without taxonomic assignment, it is possible to calculate community metrics from OTUs only. Metagenomic pipelines are more costly, but advantageous when it is important to reliably identify particular sets of species and to preserve relative biomass information