Date of Award

Fall 12-2021

Level of Access Assigned by Author

Open-Access Thesis

Degree Name

Master of Science (MS)


Marine Biology


Michael Kinnison

Second Committee Member

Kristina Cammen

Third Committee Member

Erin Grey


Environmental DNA (eDNA) metabarcoding is a tool that has been used to characterize biodiversity in a range of diverse systems. However, blind application of eDNA metabarcoding primer sets to new regions and species pools can result in poor taxon coverage and unaccounted detection biases. For the Maine-eDNA EPSCoR program, one of the main focuses is to understand and characterize community assemblages in the Gulf of Maine (GoM) using eDNA to further inform conservation, monitoring, and sustainability. In this study, I selected a subset of the best performing vertebrate and invertebrate metabarcoding assys to test against GoM species present in the New England Aquarium, Boston MA, USA. Each metabarcoding primer set was applied to the same set of replicate water samples taken from each of multiple aquarium displays with distinct and censused GoM assemblages. Using these known positive communities of fish and invertebrates I assessed the relative taxonomic specificity and overlap of the different assays, whether sequence counts can be applied to estimate relative species dominance within a sampling region, and what level of sample replication is needed to reliably and repeatedly account for dominant taxa. This study found that combining multiple metabarcoding assays for vertebrates can resolve a majority of GoM vertebrates, with the 12S MiFish-U assay and the 16S MarVer3 assay working best in combination for this goal. Additionally, it was found that rank species sequence counts are often approximately indicative of relative biomass, suggesting that eDNA metabarcoding may reveal more about GoM communities than just species occupancy. Finally, while there were always taxa missed by the vertebrate metabarcoding primer sets, rarefaction analysis suggested that as few as one or two samples were sufficient to detect most or all of the species that were ultimately detectable. For the invertebrate markers tested in this study, the18S set was unusable due to possible laboratory or sequencing errors. The COI assay used in this study provides promising results for broad invertebrate taxonomic coverage, even down to species level detections for GoM taxa. However, this wide taxonomic coverage came with a tradeoff of missing many known species within larger groups. Hence, while the COI invertebrate primer set might ultimately be a useful part of a metabarcoding toolset for resolving GoM invertebrates, it might often be best combined with other primer sets for GoM biodiversity questions requiring more comprehensive coverage of particular subgroups of invertebrates.