Towards the standardisation of bioinformatic eDNA workflows for ocean conservation: Considerations and best practice for robust amplicon data analysis

Sebastian Rauschert¹, Jessica Pearce¹, Philipp Bayer¹, Shannon Corrigan¹, Eric Raes¹, Marcelle Ayad¹, Priscila Goncalves¹, Madalyn Cooper¹, Julie Robidart¹, Steve Burnell¹

¹Minderoo Foundation, Perth, Australia

Given the rapid changes and increasing challenges facing our oceans, there is an ever-growing need for the development of robust and standardised tools and approaches for monitoring of diversity and distribution of marine wildlife. Marine eDNA metabarcoding pipelines vary across studies, but most use either the tools USEARCH/VSEARCH or DADA2 for processing of sequencing reads, and usually applying their default settings, regardless of the underlying targeted assays used or the number of sites sampled. Here we present a systematic assessment of amplicon data analyses targeting 16S and 12S rRNA mitochondrial genes for fish species identification. We derived a pipeline based on the most widely used software tools in the literature and used NCBI-nt, a custom Western Australian 16S, and the MitoFish (12S) database for taxonomic assignment. We tested the effect of changing the DADA2 specific default settings on the retention of rarer and site-specific species. We filtered reads by length, varied the number of bases to overlap for read merging and analysed the eDNA data separately for each sampling site, as compared to analysing all samples in bulk together. Our results indicate that the default settings are not always appropriate, as we managed to retain more taxa with the site-specific analysis.  We believe understanding the implications of adjusting the settings of analytical tools on the outcome of eDNA-based surveys will help create trust and reliability in biodiversity assessments and will contribute to strengthen the field as a whole.

Biography:

Biography to come