Comparative assessment of eDNA metabarcoding and longline deployments for elasmobranch surveying across a large tropical marine park network

Full description

This eDNA sequence data was collected as part of a collaborative project between WA DPIRD, Curtin University and CSIRO examining the use of eDNA metabarcoding and longline deployments for elasmobranch surveying in the Kimberley and Roebuck Marine Parks, Western Australia.
Lineage: Seawater samples for eDNA analyses were collected from 31 sites (coinciding with the longline surveys) across the Kimberley Marine Park and the Roebuck Marine Park between August and September 2020. Seawater samples were also collected at an additional five sites where longline surveys were not conducted, surmising 180 seawater samples in total. At each site, we used a Niskin bottle to collect five replicate one-litre seawater samples at similar depths to the longline survey (i.e., close to the sea floor). Replicates were immediately collected post-longline deployment along the same interval transect; with the exception of six sites where samples were collected approximately 12 hrs prior to the longline survey and five site where samples were collected approximately two weeks prior (see manuscript supplementary information). Water samples were stored at 5°C in bleach-sterilized 1 L Nalgene bottles and then individually filtered across Pall 0.45 µm Supor® polyethersulphone membranes using a Pall Sentino® Microbiology pump (Pall Corporation) within five hours of collection.

DNA was extracted from half of each filter membrane within two months of collection using the DNeasy Blood and Tissue Kit (QIAGEN) with modifications. Elasmobranch DNA was specifically amplified using two previously published PCR assays (COI Elasmobranch and 16S Fish) from our mixed seawater samples (see manuscript for more information).

Libraries were sequenced on 300 cycle (for unidirectional sequencing) MiSeq® V2 Standard Flow Cells on an Illumina MiSeq platform (Illumina, San Diego, USA), housed in the TrEnD Laboratory at Curtin University, Western Australia. Sequencing reads were demultiplexed and quality filtered in OBITools (v1.2.9; Boyer et al., 2014) and in R (v3.5.3; RStudio Team, 2015) using the DADA2 (v1.10.1) bioinformatics package (Callahan et al., 2016).

We have uploaded demultiplexed (unfiltered) eDNA data for public use. Each sample file is in a fastq format; with sample names corresponding to site/replicate numbers (see eDNA_fastq_READ_ME.txt in Supporting Documentation).