loading page

Exploring the data that explores the oceans: working towards robust eDNA workflows for ocean wildlife monitoring
  • +11
  • Jessica Pearce,
  • Philipp Bayer,
  • Adam Bennett,
  • Eric Raes,
  • Marcelle Ayad,
  • Shannon Corrigan,
  • Matthew Fraser,
  • Madalyn Cooper,
  • Denise Anderson,
  • Priscila Goncalves,
  • Benjamin Callahan,
  • Michael Bunce,
  • Stephen Burnell,
  • Sebastian Rauschert
Jessica Pearce
Minderoo Foundation
Author Profile
Philipp Bayer
Minderoo Foundation
Author Profile
Adam Bennett
Minderoo Foundation
Author Profile
Eric Raes
Minderoo Foundation
Author Profile
Marcelle Ayad
Minderoo Foundation
Author Profile
Shannon Corrigan
Minderoo Foundation
Author Profile
Matthew Fraser
Minderoo Foundation
Author Profile
Madalyn Cooper
Minderoo Foundation
Denise Anderson
INSiGENe Pty Ltd
Author Profile
Priscila Goncalves
Minderoo Foundation
Author Profile
Benjamin Callahan
North Carolina State University at Raleigh
Author Profile
Michael Bunce
New Zealand Department of Conservation
Author Profile
Stephen Burnell
Minderoo Foundation
Author Profile
Sebastian Rauschert
Minderoo Foundation

Corresponding Author:[email protected]

Author Profile

Abstract

In the face of a looming ecological crisis, effective and efficient biomonitoring methods will be critical to the conservation of marine environments. Environmental DNA (eDNA) is now established for detecting species presence within ecosystems and is increasingly being recognised as a tool for marine management. However, concerns surrounding reproducibility and standardisation has led to a lack of confidence and uptake by some end-users as the method transitions from the research domain into biomonitoring practice. Here we incorporate DADA2 into our automatable, containerised, and fully reproducible amplicon (metabarcoding) workflow and highlight how changing analytical steps and parameters may impact inferences from eDNA-derived biodiversity data. By pooling all samples for amplicon sequence variant (ASV) inference, compared to the DADA2 default of independent sample inference, we were able to recover significantly higher ASV counts per sample and directly increased detection of ASVs across replicates and sites. This led to a higher number of species detections and revealed known species as ecological drivers only when ASV inference was pooled across samples. Although, for the purpose of this study, we have focused on the manipulation of a single parameter for a single analytical tool, our findings reiterate the importance of ensuring both are selected in a manner appropriate to the research questions being addressed, and that suitability for comparisons to previously generated datasets is considered. Finally, we provide guidance for robust data processing with the aim to make eDNA more effective, transparent, and useful for management.