DNA metabarcoding marker choice skews perception of marine eukaryotic biodiversity
Casey, Jordan M et al. (2021), DNA metabarcoding marker choice skews perception of marine eukaryotic biodiversity, Dryad, Dataset, https://doi.org/10.5061/dryad.rr4xgxd8r
DNA metabarcoding is an increasingly popular technique to investigate biodiversity; however, many methodological unknowns remain, especially concerning the biases resulting from marker choice. Regions of the cytochrome c oxidase subunit I (COI) and 18S rDNA (18S) genes are commonly employed “universal” markers for eukaryotes, but the extent of taxonomic biases introduced by these markers and how such biases may impact metabarcoding performance is not well quantified. Here, focusing on macro-eukaryotes, we use standardized sampling from autonomous reef monitoring structures (ARMS) deployed in the world’s most biodiverse marine ecosystem, the Coral Triangle, to compare the performance of COI and 18S markers. We then compared metabarcoding data to image-based annotations of ARMS plates. Although both markers provided similar estimates of taxonomic richness and total sequence reads, marker choice skewed estimates of eukaryotic diversity. The COI marker recovered relative abundances of the dominant sessile phyla consistent with image annotations. Both COI and the image annotations provided higher relative abundance estimates of Bryozoa and Porifera and lower estimates of Chordata as compared to 18S, but 18S recovered 25% more phyla than COI. Thus, while COI more reliably reflects the occurrence of dominant sessile phyla, 18S provides a more holistic representation of overall taxonomic diversity. Ideal marker choice is, therefore, contingent on study system and research question, especially in relation to desired taxonomic resolution, and a multi-marker approach provides the greatest application across a broad range of research objectives. As metabarcoding becomes an essential tool to monitor biodiversity in our changing world, it is critical to evaluate biases associated with marker choice.
All methods are available in the associated manuscript in Environmental DNA.
This submission is connected to the manuscript entitled, "DNA metabarcoding marker choice skews perception of marine eukaryotic biodiversity" published in Environmental DNA. Files include the metadata ("Bali_Metadata.csv"), final COI OTU table ("COI_OTU_Taxonomy.csv"), final 18S data OTU table ("18S_OTU_Taxonomy.csv"), CoralNet image analysis data ("coralnet.csv"), the R script (COIv18S.R) used for all analyses in the manuscript, and a README file that defines variables and describes data files (CASEY_README.txt).
National Science Foundation, Award: OISE 1243541